2nd R. N.

Raikar Memorial International Conference

& Banthia-Basheer International Symposium on
ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE

Technical Papers
Volume I, II & III

18 - 19 December 2015  Hotel The Lalit, Mumbai, India

Organised by

INDIA CHAPTER OF ACI

 2nd R.N. Raikar Memorial International Conference
& Banthia-Basheer International Symposium on
ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
18th & 19th December 2015 t The Lalit, Mumbai

TECHNICAL PAPERS
Volume I

Organised by

INDIA
India Chapter CHAPTER OF
of American ACI
Concrete Institute

SUPPORTED BY

Federation of Indian Chambers of

Commerce and Industry, New Delhi

INTERNATIONAL PARTNERS

Asian Concrete Federation, Thailand

Instituto Mexicano del Cemento y (Australia, India, Indonesia, Japan, The Concrete Society, UK
del Concreto, Mexico Korea, Mongolia, Singapore, Taiwan,
Thailand, Vietnam)

Japan Concrete Institute LEEDS University, UK The International Union of Laboratories

Institute of Concrete Technology
and Experts in Construction Materials,
London, UK
Systems and Structures, France

NATIONAL PARTNERS

Institute for Research,

Indian Concrete Institute, Development and Training
Builders Association of India of Construction Trade and
Chennai
Management, Bangalore

India Chapter of American Concrete Institute

2-3, Nagree Terraces, Soonawala Agiary Road, Mahim (West), Mumbai 400 016.
Tel.: +91 (0) 22 2446 9175 t Telefax: +91 (0) 22 2446 0760 t Email: infocaci@gmail.com t Web: www.icaci.com

ISBN No. 81-86876-16-2

Organised by
India Chapter of American Concrete Institute i
R. N. Raikar (1939 – 2008)

It was his intense desire to share knowledge that urged

him to become a visiting lecturer at the J. J. School of
Architecture – a duty he carried on till 1975 when he had
to grudgingly discontinue due to increased professional
commitments.Mr. Raikar’s first technical contribution,
‘Technology of Building Repairs’ was published in 1974.
After four re-prints over four decades, the book continues
to serve as a bible for engineers – a testimony to his
profound and timeless knowledge on the subject.

A recipient of countless national and international

accolades in the field of Structural Engineering, and
Rehabilitation and Restoration, Mr. Raikar was appointed
as an advisor by State and Central Governments on
almost all advisory panels for collapses of structures.
His experience of Collapse investigations of more than
100 structures was documented in his second technical
endeavor, ‘Learning from Failures’ (1986) and in
subsequent books ‘Diagnosis and Treatment of Structures
in Distress’ (1994) and ‘Durable Structures through
Mr. R. N. Raikar was a man of unparalleled virtues: an Planning for Preventive Maintenance’ (1994). He lived by
engineer par excellence, a thorough professional, an the American Concrete Institute adage, ‘Progress through
earnest teacher, and most importantly a pious human Knowledge’.
being. As the first Civil Engineer in the family, Mr. R.
N. Raikar or RNR as he was fondly called graduated Mr. Raikar and few other like-minded professionals
in the year 1961 from Pune Engineering College. Such launched the India Chapter of American Concrete Institute
was his charisma and encouragement that most of his (ICACI) in 1979. The Chapter is a proud recipient of the
family members followed in his foot-steps and pursued ‘Excellent Chapter’ award for the past consecutive 14 years
civil engineering. Today, his first family boasts 11 Civil – an insurmountable feat that could only be accomplished
Engineers and five Architects. by a towering personality like Mr. Raikar.

After gaining experience in Military Engineering Services Mr. Raikar’s contribution to the growth of the Chapter is
and Bombay Port Trust he decided to form his own unmatched. He was instrumental in organizing more
company, Structwel, in 1967. Started initially as a structural than 35 seminars on concrete and construction related
engineering organization, he immediately diversified into topics during his stint at the Chapter. It was his brain
Forensic Engineering in construction and repaired the child to start a construction supervisor’s course which
first building in the inaugural year of his company. has recently completed its 19th installment. His initiative
to bring Technician’s training courses to India has given
Mr. Raikar’s outstanding flair for building repairs, Indian engineers and technicians an opportunity to avail of
rehabilitation and restoration prompted the State these initiatives at affordable prices.
Government to invite him to the advisory panel of the
Repairs Board in 1968 – a unique recognition for an His passion, commitment and zeal towards knowledge
engineer with only seven years of professional experience. advancement was recognized by the American Concrete
Institute when he was awarded the celebrated ‘Honorary
He became a Member of the prestigious IStructE, UK in Membership’ in 2004, becoming the first Asian to receive
the year 1969 – a membership highly coveted and attained this honour and the first person to get it outside the United
only after a grueling seven-hour examination. Needless States of America.
to say Mr. Raikar cleared it in his first attempt, became
a member, was subsequently awarded Fellowship of Mr. Raikar’s organization, Structwel, is at the forefront of
IStructE, UK, and finally made it as the organization’s India structural engineering. Its uniqueness is the presence of
representative. a structural design arm, a material testing laboratory, a

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

ii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Dr. Surendra P. Shah
Walter P. Murphy Emeritus Professor of Civil Engg,
Northwestern University, Evanston, IL, USA and
Distinguished Professor of Civil Engg., IIT Madras,
Chennai, India
Research and Development centre, and an army of trained Prof. Surendra P. Shah was
engineers in the field of rehabilitation and restoration, born in Gujarat. He received
with each department aiming for excellence. Integrity his Bachelor degree in Civil
and professionalism - virtues of Mr. Raikar - are today Engineering from the B.V.M.
displayed by every employee of Structwel, currently Engineering College at
spearheaded by his able sons, Chetan and Kaustubh. Vallabh Vidyabnagar - the
oldest Engineering College
Mr. Raikar breathed his last on 8th of March, 2008 after in Gujarat. One among the
being in coma for three months. He suffered from a brain few Indians at that time to
stroke while delivering a key-note address at an ICACI pursue higher studies in the
seminar on Forensic Engineering on 6th December 2007. USA, Prof. Shah received
He walked into the seminar fully aware of the aneurism his MS and Ph.D. degrees
in his brain and the dangers it presented. But it was his in Civil Engineering from
destiny to be remembered by the fraternity as a brave Lehigh University and Cornell University, respectively.
soldier who departed this life with his shoes on. Then, he joined the Civil Engineering Faculty at the
University of Illinois at Chicago. After a few years, he
The board of India Chapter of American Concrete Institute moved to the Northwestern University in USA, where he
and the entire engineering fraternity salutes the invaluable continues to serve as a faculty member.
contribution of Mr. R. N. Raikar - a legend like none other.
Prof. Shah is arguably one of the finest concrete
technologists in the world today. He is wellknown for his
visionary leadership skills to visualize and pursue frontier
research and development in the field of concrete science
and technology. One of his most significant contributions
is the establishment of the Center for Advanced Cement
Based Materials (ACBM) at the Northwestern University.
The ACBM is one of the finest research centers of its kind
in the world.
Prof. Shah continues to guide the concrete research
and development activities at various universities and
industrial organizations worldwide. He has co-authored
about 400 research articles and three text books, and
delivered more than 35 keynote lectures in international
conferences. He is the Editor-in-Chief of Concrete Science
and Engineering Journal and the Editorial Board member
of several other peer-reviewed journals.
Many agencies and universities worldwide have
recognized Prof. Shah for his contributions towards the
teaching, research, and development of concrete science
and technology. He is an elected member/fellow of the
National Academies of Engineering in the USA, China,
and India. In 2006, the Concrete Construction magazine
has included Prof. Shah in the “Top Ten Most Influential
Persons in the Concrete Industry.” Also, he is designated
as the UNESCO Expert to India and the UNIDO Consultant
to China. The American Concrete Institute (ACI) and
the International Union of laboratories and experts in
Construction Materials Systems and Structures (RILEM)
have bestowed him with honorary member status.
Recently, Prof. Shah has served as honorary/visiting
faculty member in IIT Bombay and IIT Madras

Organised by
India Chapter of American Concrete Institute iii
Professor (Dr.) Nemkumar Professor (Dr.) P. A.
(Nemy) Banthia Muhammed Basheer
Professor, Distinguished University Scholar and Senior Chair in Structural Engineering and Head of School of Civil
Canada Research Chair in Infrastructure Rehabilitation Engineering, University of Leeds
University of British Columbia, Canada
Nemkumar (Nemy) Banthia P.A. Muhammed Basheer
is a Professor of Civil Engg, is Chair in Structural
Distinguished University Engineering and Head of
Scholar and a Senior School of Civil Engineering
Canada Research Chair at University of Leeds,
at the University of British Leeds, England. He has
Columbia, Vancouver, been an educationalist
Canada. He obtained his and researcher in the
Ph.D. from the University field of civil (structural)
of British Columbia in engineering for nearly 35
1987, and has since then years. He is distinguished
actively pursued research for his seminal research on
in the areas of cement- concrete materials that have
based and polymer-based fibre reinforced composites, led to new understandings
with particular emphasis on testing and standardization, of fluid transport in concrete and its linkage to durability
fracture behaviour, strain-rate effects, durability and of concrete structures. He has developed new tests and
development of sustainable materials. sensors for measuring the fluid transport and predicting
the residual life of concrete structures. His patents have
He pioneered the process of structural repair and been successfully utilised worldwide by two spin-out
strengthening using sprayed fibre reinforced polymers. companies.
He is credited for his contributions to the fundamental
Professor Basheer has introduced innovation to teaching
understanding of sprayed concrete including particle
by way of developing computer assisted self-learning
and fibre kinematics, rebound modelling, in-situ quality
tutorials in analysis, design and durability of concrete
control and performance characterization. In the area structures at UG and PG levels. These contributions
of fibre reinforced cementitious composites, he has resulted in him being the first research professor to
developed a number of novel fibres for concrete and receive a University Teaching Award in 2003.
shotcrete reinforcement, and these fibres are now being
used in numerous large projects around the world. Professor Basheer has taught at University of Leeds,
Regional Engineering College, Calicut, India and Queen’s
Dr. Banthia has edited/co-edited fifteen volumes, published University Belfast, Northern Ireland, and served as a
over 250 technical papers, and holds three international visiting professor at Chonqing University and Zhejiang
patents. He serves on the technical committees of various University, China. He has authored over 400 refereed
professional societies including the ACI where he chairs journal and conference papers.
the committee on fibre reinforced concrete, RILEM where
Professor Basheer has received the ACI/James
he chairs the Technical Committee on FRP-Concrete Bond,
Instruments Award for the best non-destructive test
and the CSA where he chairs the Durability Committee of method in 1991 and 1999. He also received the Young
the Highway Bridge Design Code. He also serves on the Scientist Award from the Department of Science and
Editorial Boards of four international journals. Technology, Kerala, in 1991 for his research work on
Dr. Banthia has received numerous national and permeability of concrete, a special award from the
international awards in recognition of his research Committee of International Conferences (formerly
accomplishments. Of particular note, he was inducted CANMET/ACI) for his contributions to concrete technology
as a Distinguished University Scholar at UBC in 2003 research in 2012 and a lifetime achievement award from
and was appointed as a Senior Canada Research Chair in Civil Engineering Research Association of Ireland in 2014
Infrastructure Rehabilitation and Sustainability in 2006. for his contributions to education, research and technology
Other prestigious awards include the WG Hislop Award transfer in the field of concrete and concrete structures.
of the ACI-BC Chapter, four Best Paper Awards, and the He is an elected Fellow of the American Concrete Institute,
Wason Medal from the American Concrete Institute. the Irish Academy of Engineering, Royal Academy of
Dr. Banthia was named Scientific Director of IC-IMPACTS Engineering, Institution of Civil Engineers and several
in 2012, which is an innovative partnership between other notable organisations.
Canadian and Indian institutions to facilitate the effective Professor Basheer is an editor of the international
and rapid mobilization of new technologies to improve journal of Construction and Building Materials and was
water quality, increase the safety and sustainability of an associate editor of the International Journal of Civil
critical civil infrastructure, and improve health across Structural Health Monitoring. He is an assessor for grant
both nations. applications in USA, Finland, Poland, and Saudi Arabia.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

iv ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
American Concrete Institute Message
Sharon L. Wood
President, American Concrete Institute
Dean, Cockrell School of Engineering
The University of Texas at Austin

The American Concrete Institute was founded in 1904 as

a non-profit membership organization dedicated to public
service and representing the user interest in the field of
concrete. ACI gathers and distributes information on the
improvement of design, construction and maintenance
of concrete products and structures. The work of ACI On behalf of the members of the American Concrete
is conducted by individual ACI members and through Institute, I extend well wishes to the attendees and
volunteer committees composed of both members and participants of the Second R.N. Raikar International
non-members. Conference and Banthia-Basheer International
Symposium – 2015.
The committees, as well as ACI as a whole, operate under
a consensus format, which assures all participants the ACI is pleased to provide co-sponsorship to the conference
right to have their views considered. and symposium. We recognize and applaud the significant,
ongoing efforts of the India Chapter – ACI to further expand
Committee activities include the development of building
cooperation and collaboration within the worldwide
codes and specifications; analysis of research and
concrete community. The presentations planned for the
development results; presentation of construction and
conference and symposium will advance the international
repair techniques and education.
database of concrete technology knowledge.
Individuals interested in the activities of ACI are
It is fitting that the influence of R.N. Raikar, Nemy Banthia,
encouraged to become a member. There are no
and Muhammed Basheer—individuals who have all
educational or employment requirements. ACI’s
contributed greatly to the work of ACI—has inspired
membership is composed of engineers, architects,
the conference’s technical program. We commend the
scientists, contractors, educators and representatives
members of the India Chapter – ACI for organizing this
from a variety of companies and organizations.
event focused on “Advances in the Science and Technology
Members are encouraged to participate in committee of Concrete.”
activities that relate to their specific areas of interest. For
more details, visit www.concrete.org
Sharon L. Wood

Organised by
India Chapter of American Concrete Institute v
India Chapter of Message
American Concrete Institute
Pankaj Shah
President, India Chapter of American Concrete Institute

The India Chapter of

American Concrete Institute
(ICACI), now in its 37th
year, was founded by Mr.
R. N. Raikar - an engineer
par excellence. His vision,
INDIA CHAPTER OF ACI untiring efforts and sheer
determination catapulted
The American Concrete Institute (ACI) is the premier ICACI into a national body of
professional institution in the sphere of concrete for over prominence.
100 years. Its motto is “Progress through knowledge”. Mr. Raikar was passionate
towards knowledge
Indian Chapter is in its 35th year. Technical dissemination is advancement and
a most appropriate method for enhancing our Continuous technology transfer in the field of concrete. In his honour,
Professional Development (CPD). We have around 2,000 ICACI launched the Biennial R. N. Raikar Memorial
members spread all over India, who actively participate in International Conference series, with a view to celebrate
the Chapter Program. the contribution of some of the most outstanding concrete
professionals of Indian origin. The first conference
India Chapter, the largest ACI Chapter, is in it’s 35th was held in 2013 and it felicitated the world famous Dr.
year and is privileged to receive “Excellent Chapter” Surendra P. Shah from Northwestern University - one of
Award consecutively for the last 15 years. The Chapter is the finest concrete technologists in the world today.
committed to train and propagate good concrete making After its blockbuster debut, the second conference
practices through seminars, demonstrations, workshops has returned to Mumbai with an ever-expanding list
and competitions for the construction industry. It believes of distinguished concrete technocrats from India and
in Continuous Progress and Development in “Knowledge overseas. This conference celebrates the achievements of
Dissemination” as an ongoing activity. This conference is two towering personalities, Dr. Nemkumar (Nemy) Banthia
a sequel to it. and Prof. Dr. P. A. Muhammed Basheer by convening an
international symposium in their honor. Our aim is to
In 2009, India Chapter successfully launched the ACI introduce participants to cutting edge technologies in the
Certification Programme of “Concrete Field Testing field of concrete and to facilitate technology transfer in
Technician Grade – I”. In a short period of a year, the Chapter appropriate areas.
has trained and examined 200 Concrete Professionals We are privileged that this conference is co-sponsored by
and Technicians. the ACI and supported by a host of international technical
bodies such as the ICT, London; ACF; IMCYC, Mexico;
Leeds University UK amongst several others. Celebrated
Indian bodies such as FICCI; Builders Association of India;
Indian Concrete Institute; INSTRUCT have also pledged
their support – a testament of this conference’s robust
technical content.
The endeavour of ICACI continues to be the Development
and Advancement of Good Practice of Concrete
Technology’ in India. Towards this end, we are constantly
organizing seminars, symposiums, technical lectures
and workshops with the participation of field experts and
professional bodies.
The driving force for thousands of our members,
comprising concrete professionals, engineers, concrete
practitioners, academicians, researchers, consultants,
material scientists, constructors, students etc., was, is
and will always be ‘Progress Through Knowledge’.

Pankaj Shah

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

vi ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
In conversation with Dr. Surendra Manjrekar
Conference Chair

My fellow concrete practitioners from the global delegates are essentially ‘technocrat ambassadors’ of
community and India, the most significant part of that their respective countries and together we are creating
global community. Namaskar! Welcome to this sequel of an “information pool” to be handed over to our respective
R N Raikar International Conference! government bodies and private nation builders as a ready
The first RNR conference in December 2013 was reckoner to assist in their policy making of international
considered to be unprecedented by the fraternity and was techno commercial collaborative acts. Proactive
a record in terms of international participation, quality of government bodies and industry will surely be helped by
papers and full house attendance on both days until the our efforts and these RNR conferences besides being of
very last session. And there were many others. great academic importance will definitely help change
and touch lives of the people of India as well as in other
Records particularly when they concern “Knowledge participating countries. This effect is sure to grow by each
Transfer Initiative” are made to be broken by the such conference.
subsequent events. I am writing this piece almost one
week in advance and as of tonight the delegate count, Dr. A. P. J. Kalam – Past President of India had a
papers count and international participation count has wonderful vision for beyond 2020. He proposed a ‘Societal
far exceeded our 2013 event which abundantly shows Development Radar’ to monitor and review sustainable
that R N Raikar conference has become an ‘International growth. His user connectivity pyramid was built on “natural
brand’. I am convinced that this overwhelming response is resources at the top, information and communication
because of underlying mission of “Knowledge should find next, the convergence of technology, societal business
its own level like water does”. models, applications in the next tier, and the end users at
the bottom.
I am truly honoured and delighted to be Chair of this
conference yet again. The passion and the dream to On more micro basis, we conduct these conferences to
connect the global concrete fraternity to India is shared give tribute to Indian minds, by Indian minds. Undoubtedly
by many of my colleagues at the India Chapter of ACI and Indian DNA is provenly outstanding, may it be contributing
their collective efforts of over two decades has paved the in development within India or outside India. These
way for the success of this conference. We are grateful and RNR conferences celebrate the contributions by Indian
acknowledge the presence of the galaxy of international minds who ventured to be Columbus and go to other
speakers and well-wishers who have contributed to this shores, put a stamp of their creativity and in the process
process. developed and contributed to “concrete science”. One
of these acclaimed scintillating minds is Dr. Surendra
India is the second largest manufacturer of concrete and is Shah whose contributions to the industry and science are
today “the” destination of global investment. About 21% of worth consideration even for Nobel Prize. We celebrated
foreign investment into India will be in housing, real estate his achievements in last episode of this R N Raikar
and other construction activities. In absolute terms it is Conference series. This time we are fortunate to have him
projected to be whopping 250 billion US $ by 2020. This is as a ‘mentor’ of this conference. Such is his affection and
over and above our own national spending of 1 trillion US $ great sense of obligation to India. Friends, personally I am
for infrastructure allocated in the ongoing five year plan. so indebted and grateful to him.
Indian leadership in cement and construction industry is This time we are twice as fortunate to celebrate the
crucial for global sustainable development and friends contributions of two outstanding and comparatively
thus the creation of this ‘ExpressWay’ of knowledge younger scientists who are shining stars on the horizon of
transfer assumes great importance. All our international global concrete science. Yes, I am referring to one and only

Organised by
India Chapter of American Concrete Institute vii
Dr. Nemkumar Banthia from Canada and yet again one and Sharon Wood has repeatedly asked me to convey her best
only Dr. PAM Basheer from United Kingdom. Both of them wishes to all the delegates for an excellent conference and
have made significant contributions to steer the course of for the New Year 2016. Several Past Presidents including
concrete science in right direction and will be responsible Anne Ellis, Bill Rushing, Jim White, Jim Pierce, José Pepe
for igniting many young engineering minds during this , Tony Fiorato and Jim Cagley have wished best for this
conference to follow their suit in this thriving and vibrant outstanding conference.
‘promised land’ of India. My colleagues and I are indeed
Another first is that we have been able to get the support
thankful to you Dr. Surendra Shah, Dr. Nemkumar Banthia
of National Federation of Industry and Chamber of
and Dr. PAM Basheer for motivating and guiding Indian
Commerce and Industries (FICCI) which is the apex body
engineers so that due justice is given to their inherent
in India for the major industries of India which totals to
genes & DNA and will perform commensurately.
250000 in number. This again bears testimony to the
My dear young professionals, I have an explicit assurance prominent position of this conference on the gird of this
from all three of these luminaries to handhold and guide huge country.
you whenever you need that casting vote and guidance in
We have the active support of highly reputed international
your careers. After the last conference, I am informed
organizations viz: American Concrete Institute (ACI-USA),
that many young engineers have been able to pursue
The International Union of Laboratories and Experts in  
international research and professional careers. Thank
Construction Materials, Systems and Structures (RILEM–
you Suru, Nemy and Basheer on behalf of our aspiring
France), Asian Concrete Federation (ACF–Thailand),
young engineers and students who are the future of India.
Japan Concrete Institute (JCI-Japan), Korea Concrete
India is a huge country of 1.27 billion. To do something Institute (KCI-South Korea), The Concrete Society (UK),
meaningful for the first time in this vastness needs University of Leeds (UK), Institute of Concrete Technology
innovation and imagination. There are many ‘firsts’ in (ICT –London), Instituto Mexicano del Cemento y del
this R N Raikar Conference. For the first time selected Concreto (IMCYC- South America, Mexico), University of
sets of conference papers will be published in reputed British Columbia, Canada.
international journals like Construction and Building
Finally, I take this opportunity to wish each one of you a
Materials (UK) & Cement and Concrete Research (USA). At
very prosperous 2016 and welcome you to the 2nd R N
the ACI Denver Convention the Editor in Chief of ‘Concrete
Raikar International Conference and Banthia-Basheer
International’ expressed a possibility of publishing select
International Symposium. Have an excellent conference!
papers in ACI Journals during the ‘Global Sustainability
Forum 8’ where I was an invited speaker. This publication
of papers in international journals is a big impetus to Indian
Dr. S. K. Manjrekar
concrete practitioners to generate quality research and
Chair - 2nd R N Raikar Intl. Conf. and Banthia-Basheer
field case studies and be part of the global mainstream.
Intl. Symposium – 2015
For the first time, the “International Organising Committee Past President - India Chapter of ACI [1995-1998] [1998-
(IOC)” meeting of the 2nd RNR Conference was a part of the 2001] [2005-2008]
official ACI Convention program due to the extraordinary
international interest and participation in this conference. I
must mention that the hall was full during the IOC meeting
and many ACI officials including Dr. Ron Burg, Executive
Vice President were in attendance. President of ACI Dr.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

viii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Message Message
Vinay Mathur Dr. Nicolas Roussel
Deputy Secretary General RILEM Technical Activities Committee Chair
Federation of Indian Chambers of Commerce and
industry

India is witnessing large scale urbanization which is a It is for the RILEM association a pleasure to co-sponsor
direct result of high economic growth of its economy. the 2nd R N Raikar Memorial International Conference &
Migration of people from rural areas to towns and cities ‘Banthia – Basheer International Symposium’.
is continuing unabated. In 2001, India had an urban
The mission of this conference is to introduce the
population of about 285 million, living in approximately
participants to the cutting edge technologies in the field
5200 urban agglomerations. In 2011, it has increased to
of concrete and concrete construction and to facilitate
almost 400 million living in as many as 7935 towns.
technology transfer in appropriate areas. This event
For coping with this challenge, and building a sound infrast therefore strongly contributes to the mission of the RILEM
ruct ure, Advances in Science and Technology of Concrete association, which is to advance scientific knowledge
and Construction Chemicals have a very important role related to construction materials, systems and structures
to play. The benefits of usage of Construction chemicals and to encourage transfer and application of this
are multi-fold. Every small increase in initial input cost knowledge world-wide.
by addition of these products means increased life of the
I wish you all a successful conference
end products with improved performance and longer life.
It also makes the application easier, thus reducing the Dr. Nicolas Roussel
labour cost and minimizing the wastage, adding further
value for customers.
I am very happy to note that the India Chapter of the
American Concrete Institute jointly with FICCI is
organising the ‘2nd R N Raikar International Conference’
and ‘Banthia-Basheer International Symposium’ on 18th
and 19th of December 2015 in Mumbai, India. This indeed
is a very timely initiative.
I am sure the deliberations in the conference will facilitate
growth of this important segment of national economy.

Vinay Mathur

Organised by
India Chapter of American Concrete Institute ix
Message Message
Dr. Toru Kawai Hyunmock Shin
Executive Director, Japan Concrete Institute President. Korea Concrete Institute

JCI is delighted to become a co-sponsor to 2nd R N Raikar On the behalf of Korea Concrete Institute, I am very
Memorial International Conference & Dr. Basheer and Dr. pleased to co-sponsor the 2nd R N Raikar International
Banthia International Symposium on advances in science Conference and Banthia-Basheer International
and technology concrete on 18th and 19th December 2015. Symposium on Advances in Science and Technology of
Concrete, which will be organized by India Chapter of the
On behalf of JCI members, I sincerely hope that the newly American Concrete on 18th and 19th of December 2015 in
published papers will be featured and the conference will Mumbai, India.
be extremely successful.
Congratulating the success of the 1st R N Raikar
Dr. Toru Kawai International Conference, which was held 2 years ago, I
and KCI members sincerely wish you another success
with the 2nd R N Raikar International Conference and
Banthia-Basheer International Symposium where more
than 30 countries participate in presenting more than 100
papers.

Concrete has become one of most important construction

materials for a long time, and will be further widely used
in our future. I am certain that new concrete technology
will be continuously developed as one of the most
important tasks all over the countries, especially for
the development of new environment- friendly concrete
considering conservation of natural resources. From the
point of view, international collaboration became very
important to achieve sustainable development of concrete
technology.

I hope the conference make contribution to sustainable

development in the field of cement concrete throughout
the world, and all participants get fruitful results.

Hyunmock Shin

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

x ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Message Message
Kathy Calverley Professor John Fisher, CBE
Managing Director, The Concrete Society Deputy Vice-Chancellor, University of Leeds

The Concrete Society is delighted to support the Banthia- The University of Leeds is delighted to learn that one of its
Basheer symposium. Dr Basheer has been an active professors, P A Muhammed Basheer, is honoured at the
member and supporter since 1992 and has served as Vice 2nd R.N. Raikar International conference by holding the
Chairman of our Northern Ireland Branch and currently Banthia-Basheer International Symposium in Mumbai on
an active committee member of the Yorkshire and 18th and 19th of December 2015.
Humberside Region. During that time he has worked with
The Society to further improve educational programmes Professor Basheer joined the University in September
for the construction and concrete industry in the UK. 2014 as Chair in Structural Engineering and was
appointed as the Head of School of Civil Engineering
Founded in 1966, The Society is an independent in 2015. The history of the School is synonymous with
membership organisation offering a comprehensive research on structures. Professor Roy Evans started this
portfolio of products and services including publications, tradition during his headship (1946-1968). This was further
magazines and information services within the industry. strengthened by Professor Adam Neville and continued
by Professor Tony Cousens. In more recent times this has
For further information regarding The Society, including continued in both structural engineering (by Professor
membership, please contact: Andrew Beeby) and concrete materials by Professor Joe
Cabrera and Professor Chris Page. Professor Basheer,
Email: communications@concrete.org.uk or go to our
with his international reputation in performance of
Website: www.concrete.org.uk
concrete in structures and developing novel methods of
assessing them, is recognised by numerous prizes and
Kathy Calverley fellowships of professional institutions. He is a Fellow of
the Royal Academy of Engineering, which is the highest
level of recognition for engineers in the United Kingdom.

The University of Leeds is very pleased to be part of

honouring Professor Basheer and I believe he is an
outstanding ambassador for the University of Leeds
in India.

Professor John Fisher, CBE

Organised by
India Chapter of American Concrete Institute xi
Message Message
Hare Ram Shrestha Tuk Lal Adhikari
President President, Nepal Geotechnical Society
Society of Consulting Architectural and Engineering
Firms, Nepal

We have pleasure to note that ACI Indian Chapter is We have pleasure to note that ACI Indian Chapter is
organizing 2nd R N Raikar International Conference and organizing 2nd R N Raikar International Conference and
Basheer-Banthia International Symposium - 2015 on Basheer-Banthia International Symposium - 2015 at The
18th-19th December. Lalit Mumbai 18th-19th December.

We wish to offer our best wishes on behalf of Society of We take this opportunity to offer our best wishes on behalf
Consulting Architectural and Engineering Firms (SCAEF) of Nepal Geotechnical Society (NGS) for grand success
for grand success of the event. of this august event and we believe its proceeding will
provide valuable reference to engineers globally.
We believe its proceeding will provide valuable reference
to practicing engineers and consultants globally. We also wish to express our commitment to join hands
with the ACI Indian Chapter in its future endeavors for
We also wish to express our commitment to partner with advances in science and technology of concrete and also
the ACI Indian Chapter in its future endeavors for advances wish to partner with your esteemed institution.
in science and technology of concrete.
NGS is committed to promote the field of geotechnical
SCAEF is committed to highest level ethics, code of engineering both in scholarly researches as well as
ethics and professionalism in the consulting industry practical applications in Nepal and to enhance the
in Nepal and is duly recognized by the Government of knowledge and expertise of Nepalese geotechnical
Nepal as national body of consultants. The society is engineers. The society is affiliated to the International
affiliated with International Federation of Consulting Society of Soil Mechanics and Geotechnical Engineering
Engineers (Fédération Internationale Des Ingénieurs- (ISSMGE).
Conseils) abbreviated as FIDIC and is committed to quality
enhancement in consulting engineers profession. With best wishes for the grand success of the event, we
remain
With best wishes for the grand success of the event, we
remain. Tuk Lal Adhikari

Hare Ram Shrestha

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Organised by
India Chapter of American Concrete Institute xiii
 WE ALWAYS FIND
A WAY FOR YOU

H.O : 71-C, New Avadi Road, Kilpauk, Chennai 600 010, INDIA,
Phones: + 91-44-42183033, 2644 1678, Fax : +91-44-2644-3621 ,
Email: padmaja@pincgroup.com, Web : www.pincgroup.com

20 YEARS OF SERVICE
OVER 1000 CLIENTABLE | 50 PRODUCT VARIETIES

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xiv ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Organised by
India Chapter of American Concrete Institute xv
 Ensure long-term performance
of your concrete with

STEEL FIBRES
TUNNEL INDUSTRIAL FLOORING

ROAD BRIDGE

Availibility :
0.45 mm to 1.00 mm. Diameter
25 mm to 60 mm. Length
Metaphors

Hooked-end, Round Crimped and

Flat steel fibres can be custom-made

AN ISO 9001:2008 CERTIFIED COMPANY

5-8B, Nagpur Industrial Estate, Kamptee Road, Uppalwadi, Nagpur - 440 026. India
Tel.: +91-712-2641040, 2640613, 3291281 | Fax :+91-712-2641760 | Grams: Steelwool
e-mail : sales@stewols.com

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xvi ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Organised by
India Chapter of American Concrete Institute xvii
Technical Papers

CONTENTS
VOLUME I
INAUGURAL SESSION
A Right Concrete, Right Way .. Spreading ACI Concrete Field Certification Course in India “Train the
Trainer” initiative continues …
Ishita Manjrekar
PLENARY SESSION
1 Evolution of Unbonded Post-tensioned Concrete in the United States, and its Implications for Lifetime 1
Performance and Restoration Philosophies
Kyle Stanish and Daniel E. Moser
2 1 Kilometer Tall Kingdom Tower: Concrete Reaches New Heights 8
Robert C. Sinn and Kaushik Dutta
3 The importance of Indian Leadership in Cement and Construction Industry for Global Sustainable 10
Development
Dr. S. K. Manjrekar
4 Insitu Strength of Concrete – The Engineer’s Holy Grail! 14
Dr. Michael G. Grantham and Richard Rogerson
5 Replacement of Steel with GFRP for Sustainable Reinforced Concrete Structures 21
Dr. S. A. Sheikh and Z. Kharal
SESSION 1 A
6 Stability of Basalt Fibers in Concrete Medium 29
Indubhushan Patnaikuni, Himabindu Myadaraboina and Erick Saputra Atmaja
7 Concrete using Siliceous Fly Ash at very High Levels of Cement Replacement: Influence of Lime 36
Content and Temperature
G. V. P. Bhagath Singh and Prof. Kolluru V. L. Subramaniam
8 Do crystalline water proofing admixtures affect restrained plastic shrinkage behavior of concrete? 47
Rishi Gupta, Alireza Biparva
9 Condition Evaluation & Repair of 100+ Years Old Buildings 52
Ashok Kakade
10 Geopolymer Cement Concrete - An Emerging Technology for the Delivery of Resilient Highway 59
Infrastructure Solutions
Dr. B. J. Magee, A. Wilkinson, D. Woodward, S. Tretsiakova-McNally and Patrick Lemoine
SESSION 1 B
11 Sustainable Concrete as a Platform for Outreach 67
Roger P. West, Ahmed Alawais, Naveen Kwatra
12 Life Cycle Assessment and Durability of Concrete Containing Limestone 72
Prof. D. K. Panesar, K. Seto and M. Aqel
13 Ageing of old and modern concrete structures - Observations and Research 78
Prof. Dr. Klaas van Breugel
14 Influence of Supplementary Cementitious Materials on the Properties of Ultra - high Performance 90
Concrete
Zhengqi Li and Prasada Rao Rangaraju
15 Fiber Reinforced Concrete Pavement Bangalore-Mysore Corridor-NICE Project - India’s Longest 99
Whitetopped Road
K. R. S. Narayan

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xviii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Contents

SESSION 1 C
16 Repair of jetty damaged by cyclone – Case Study 105
Manish Mokal, Dr. N. V. Nayak
17 Repair of core wall in high rise building – Case Study 112
Manish Mokal, Dr. N. V. Nayak, Amit Datta
18 An Experimental Investigation on the Behaviour of High Strength Restrained one Way GPC/TVC Slabs 117
R. Mourougane, C. Sashidhar, C. G. Puttappa, K. U. Muthu
19 Behaviour of reinforced concrete beams strengthened with Glass Fibre Reinforced Polymer 125
laminates subjected to corrosion damage
Kolluru Hemanth Kumar, Sanjay Rajpal, Shashank Jha and Dr. P.K. Jain
20 Behaviour of Synthetic Fibre Reinforced Prestressed Hollowcore Slabs under Flexure-Shear 130
Sameer K. Sarma Pachalla, Pradeep K. and Dr. Suriya Prakash S.
SESSION 2 A
21 Shrinkage Cracking of Self Compacting Concrete (SCC) with Supplementary Cementitious Materials 135
Prof. Salah Al Toubat , Moussa Leblouba, Deena Badran
22 Concrete cracking in two marine micro-climates 146
P. Castro-Borges, J.A. Cabrera-Madrid, M.G. Balancán-Zapata and A. A. Torres-Acosta
23 Improving segmented Functionally Graded Concrete concept by using SCC technology 157
Dr. Olga Río, Khanh Nguyen, Xabier Turrillas, Judith Oró-Solé, Ana E. Carrillo
24 Controlling plastic shrinkage cracking in concrete using polypropylene microfiber 167
Dr. M. Sappakittipakorn, P. Sukontasukkul and N. Banthia
SESSION 2 B
25 Fiber/Textile Reinforced Permanent Formwork that can provide Shear Capacity to Concrete Beams 173
Changli Yu and Dr. Christopher KY Leung
26 Fire Resistance of Fibre Reinforced Concrete Beam 182
Piti Sukontasukkul, Sunisa Sukchoo, Manote Suppakittipakorn
27 Enhancement of substrate-repair bond strength and durability in concrete structures 187
Prof. C. Zanotti
28 Mechanical and durability properties of recycled aggregate concrete (RAC) made with different 197
replacement levels of recycled coarse aggregate (RCA)
Sumaiya Binte Huda and Dr. M. Shahria Alam
SESSION 2 C
29 Compression Behavior of Synthetic Fiber Reinforced Cellular Concrete Masonry Prisms 207
Abdur Rasheed M, Dr. Suriya Prakash S
30 Flexural Behavior of Synthetic Fiber Reinforced Cellular Light Weight Concrete 213
Abdur Rasheed M, Dr. Suriya Prakash S
31 Development of Self-Compacting Concrete using Potable Water Treatment Sludge as a Costless 220
Self-Curing Agent
Rampradheep G. S., Dr.Sivaraja M., Arunkumar N., Gopinath K., Dhanasekaran M., Saranya R.
32 Partial replacement of natural sand with recycled waste materials in concrete for sustainable 228
construction practices
Ram Prasad V S, S Ashwin Bharathwaj, A. V. Marckson

Organised by
India Chapter of American Concrete Institute xix
Technical Papers

VOLUME II
SESSION 3 A
33 Strength and Permeability of Porous Concrete with Polypropylene Fibres 241
Ominda Nanayakkara and Puyue Gong
34 Dynamic Shear Resistance of RC Beams based on Modified Field Compression Theory 248
K. Fujikake and A. Somraj
35 Role of Advanced Non Destructive Testing in Health Assessment of Cooling Tower Structures 254
Vinayak Samal and Chetan R. Raikar
36 Study of the Post-Cracking Behaviour of Steel and Polymer Fibre Reinforced Concretes 258
Sujatha Jose, Stefie J. Stephen, Ravindra Gettu
37 Validation Needs for Concrete Modeling 264
Prof. John E. Bolander
38 Alkali Activated Slag: Reaction Kinetics and Hydration Products 272
Akash Dakhane, Zihui Peng, Robert Marzke, Narayanan Neithalath, D.Ramachandran, Vinita Vishwakarma,
N.Anbarasan, K. Viswanathan, R.P. George, Kalpana Kumari
SESSION 3 B
39 Studies of Strength, Durability and Microstructural Properties of Cow Dung Ash Modified Concrete 283
V. Venkatachalapathy and D. Ramachandran
40 The Shear Strengthening of (RC) Beams with Textile-Reinforced Mortar 292
Mouza Abdullah Al-Salmi
41 Thermal Properties of Lightweight Dry-Mix Shotcrete for Mine Insulation 307
Vivek Bindiganavile, Victor Liu and Derek Apel
42 Mechanical Properties of Cementitious Composites containing Phase Change Materials: 316
Experimental Data and Effective Medium Approximations
G. Falzone, G. Puerta Falla, Z. Wei, M. Zhao, A. Kumar, M. Bauchy, and G. Sant, N. Neithalath, L. Pilon

43 Advanced Experiments and Models to Predict the Effective Elastic Properties of Cementitious 324
Systems: Application to a Geopolymer
Sumanta Das, Pu Yang, Sudhanshu S. Singh, Nikhilesh Chawla, Narayanan Neithalath, James C.E. Mertens,
Xianghui Xiao
44 Evolving Acceptance Criteria for Concrete Durability Tests In Construction Projects 334
Manu Santhanam, Sarath Kumar, Ravindra Gettu and Radhakrishna Pillai
SESSION 3 C
45 Some Aspects of Concrete Science and Concrete Technology 343
Dr. S. C.Maiti, Raj K. Agarwal
46 Partial Replacement of Cement in Mortar with Red Mud and Ultrafines 349
M. P. Deshmukh, D. D. Sarode, S. S. Pendhari, I. Alam
47 Production of M60 Grade of Concrete in Difficult & Underdeveloped Conditions 354
Durga P Shrestha
48 Role of Supplementary Cementitious Materials on Chloride Induced Corrosion - An overview 356
Ms. Anita N. Borade and Dr. B. Kondraivendhan
49 Pervious Concrete - A Value added material 365
S B Kulkarni and Clinton Pereira

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xx ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Contents

50 Preparation of an ultra-high performance concrete with superior mechanical properties using 370
corundum sand as fine aggregate under normal heat treatment
Fangyu Han, Jianzhong Liu, Qianqian Zhang, Jianfang Sha, Jiaping Liu
SESSION 4 A
51 Sustainability Evaluation of Two Iconic Bridge Corridors under Construction using Fuzzy Vikor 377
Technique: A Case Study
Shishir Bansal, Amandeep Singh & S.K. Singh
52 Design and Construction of Wirewound Circular Precast, Prestressed Concrete Tanks with Sliding 387
Base
Sanjay Mehta
53 Damage Assessment in Concrete Structures using PZT patches 396
Arun Narayanan and Prof. Kolluru V. L. Subramaniam
54 Chemical Prestress of Reinforced ASR-Expansive SHCC Beams and Possibility of other Expansive 401
Materials
Hiroo Takada, Keitetsu Rokugo, Koichi Kobayashi, Masashi Kawamura, Yukio Asano And Hyundo Yun
55 Factors Influencing the Bonding between Steel Fiber and Magnesium Phosphate Cement Mortars 409
Caijun Shi, Nan Yang, Zemei Wu, Yuan Chang, Linlin Chong, Jianming Yang,
SESSION 4 B
56 Preserving the life of infrastructure through effective monitoring and intervention 417
Professor Peter Robery
57 Rebuilding Nepal for Next earthquake 427
Badan L Nyachhyon
58 Building Durable Concrete Infrastructure using Fibre- Reinforced Polymer (FRP) Bars 438
Hamdy M. Mohamed and Brahim Benmokrane
59 Experimental Investigations on Use of Rubber Concrete in Railway Sleepers 446
A.P. Shashikala, Anilkumar P. M., George Joseph, Jestin John and Lijith K. P.
60 Chloride induced corrosion of steel bars in alkali activated slag concretes 453
Qianmin Ma, Sreejith V. Nanukuttan, P. A. M. Basheer, Yun Bai and Changhui Yang
SESSION 4 C
61 Electro-Mechanical Impedance Technique based Monitoring of Concrete Health using Piezo 461
Transducers
Sarvesh Goel and Sumedha Moharana
62 Chemical Attack on Foundation Concrete and Modern Preventive Measures 467
Dr. R. A. Hegde and Dr. Shirish Vichare
63 Strengthening using Active Prestressed CFRP in RCC Structures 475
Dr. Gopal L. Rai
64 Effect of Fiber and Silica Fume on High Performance Concrete 480
Sushil Kumar Swar, Dr. Sanjay Kumar Sharma, Paaras Gupta, Dr. Hari Krishan Sharma
65 Performance Based Concrete for Residential and Commercial Structures 490
Kalahasti Srikanth, Raajesh Ladhad, Siddappa A Hasbi
66 Features and Effects Gorkha Earthquake 495
Tuk Lal Adhikari
67 Selecting Concrete Smartly for making Sustainable Buildings in Smart Cities 501
Prof. Vinod Vanvari, Dr. Sumedh Mhaske, Neelam S. Varpe and Zeeral Jadav

Organised by
India Chapter of American Concrete Institute xxi
Technical Papers

SESSION 5 A
68 Outcomes of a Major Research on Full Scale Testing of RC Frames in Post Earthquake Fire 507
Asif H. Shah, Umesh K. Sharma, Pradeep Bhargava, G. R. Reddy, Tarvinder Singh
69 Using calorimetry to understand fly ash reactivity in high volume fly ash concretes 518
Dr. Paul J Sandberg and Pratik Bhayani
70 Tensile Performance of SHCC exposed to Low and High Temperatures 521
Keitetsu Rokugo, Koichi Kobayashi, Daichi Hayashi And Yukio Asano, Mitsuo Ozawa, Hyundo Yun
71 Effect of Slab on Strength and Behaviour of Exterior RC Beam Column Joint 528
N. Ganesan, Nidhi M. and P.V. Indira
SESSION 5 B
72 Nano Technology of Tomorrow made useful today - Effective and Cost Effective Stabilization of 535
various Soils by using Subnano Molecules
Sourabh Manjrekar & Ishita Manjrekar
73 Precast Industry Contribution toward Green Construction 542
Dr. Ekasit Limsuwan
74 Behaviour of Steel Reinforcement in Chloride and Combined Chloride - Sulfate Contaminated 547
Concrete Powder Solution Extracts
Fouzia Shaheena, Bulu Pradhan
75 Use of Paraffin in Drywalls as Sensible Heat Storage Material for Temperature Moderation 553
Rampradheep G. S., Dr. Sivaraja M., Namratha, Ariram Prasath M., Surya K., Saranya R., Arunkumar N.,
Lekha G., Manissa N.
SESSION 5 C
76 Assessment of fresh properties of Cementitious grouts used for Post-tensioning applications in 563
India
Suruthi Kamalakkannan, Ramya Thirunavukkarasu, Radhakrishna G. Pillai and Manu Santhanam
77 Study on Mechnical and Durability Properties of Recycled Aggregate Concrete Incorporated with 570
Silica Fume and Mineral Quartz
Anand K. Darji, Dr. Indrajit N Patel, Mrs. Jagruti Shah
78 Importance and Comparison of various Micro materials in high performance Concretes 576
Yatin Joshi
79 Comparative Evaluation of Constitutive Models for Concrete under Cyclic Compression 580
Mayank Tripathi, A. Kanchana Devi, Saptarshi Sasmal and K. Ramanjaneyulu
80 Use of waste marble powder as partial replacement in cement sand mix 585
Nitisha Sharma and Ravi Kumar

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xxii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Contents

VOLUME III
SUPPLEMENTARY PAPERS
81 Optimizing the Structural and Foundation Systems of the 151 Story Inchon Tower : The Development 589
of New Generation of Tall Building System
Ahmad Abdelrazaq, Moonsook-Jeong and Soogon-Lee, Taeyoung-Kim
82 Effect of Fly Ash Utilization for the Treatment of Entrapped Air in Copper Slag Concrete 596
Koji Sakai, Yuya Tanaka, Kunpei Watanabe
83 Self-Compacting Concretes Using Fly Ash and Fine Limestone Powder: Shrinkage and Surface 604
Electrical Resistivity of Equivalent Mortars
A. Durán-Herrera, J. De-León-Esquivel, D.P. Bentz, Pedro Valdez-Tamez
84 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced 613
design tools
J.A.O. Barros, R.M. Lameiras, A. Abrishambaf, C.M.V. Frazão, V.M.C.F. Cunha, M.A.D. Azenha, I.M.B.
Valente, D.M.F. Gonçalves, L.A.P. Lourenço
85 A methodology to quantify the self-healing capacity of HPFRCCs 626
L. Ferrara and V. Krelani
86 Health & Life Cycle Monitoring of Infrastructure Asset Management- An Emerging Industry and the 637
Role of Long Term Planning
Janvi Shah
87 Disclosing the creep mechanisms of cement paste by micro- indentation at different relative humidity 644
L. Sorelli, J. Frech-Baronet, Z. Chen
88 Influences of w/c ratio on the properties of lightweight concrete 653
Pengkun Hou, Xin Cheng, Zhaoheng Guo, Maoqiang Fu, Zonghui Zhou, Peng Du, Xiuzhi Zhang, Lina Zhang
89 Thermal Strain and Strength Development of Cement-based Materials at Cryogenic Temperatures 660
JIANG Zhengwu, DENG Zilong, LI Wenting
90 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating 667
Different Mineral Admixtures
Yueyi Gao, Chuanlin Hu, Yamei Zhang, Zongjin Li, Jinlong Pan
91 Sustainable Production of Fiber Reinforced Cementitious Composites 678
B.Y. Pekmezci
92 Effects of anionic and non-ionic anti-washout admixtures on the performance of underwater 684
concrete
Bo Pang, Zonghui Zhou, Xin Cheng
93 Basics of Corrosion in Reinforced Concrete 691
F. R. Goodwin
94 Bi-Directional Electro-Migration Rehabilitation for RC durability improving and its application 699
Weiliang Jin, You Zuo, Jiayun Chen, Jianghong Mao and Chen Xu, Jin Xia
95 Assessment of Minimum Flexural Steel Reinforcement Ratio for Concrete Beams 708
Azad A. Mohammed
96 Influence of limestone powder on hydration properties of Portland cement 715
Xiuzhi Zhang, Mingle Liu, Guodong Zhang, Xin Cheng, Zonghui Zhou
97 Imaging the pH profiles in cement based materials 722
Engui Liu, Masoud Ghandehai, Weihua Jin, Alexey Sidelev, Christian Bruckner, Gamal Khalil
98 Damage detection of reinforced concrete beams based on the embedded acceleration sensor of 727
concrete structure
Hongda Geng, Suhui Zhai, Shifeng Huang, Fan Lu, Xiao Yuan, Xin Cheng*

Organised by
India Chapter of American Concrete Institute xxiii
Inaugural Session

Right Concrete, Right Way ….

Spreading ACI Concrete Field Certification Course in India
“Train The Trainer” Initiative Continues….
Ishita Manjrekar
Hon. Secretary, India Chapter of American Concrete Institute

Introduction Contents of the Course and its Technical

“If you are thinking a year ahead, sow a seed, Relevance to Concrete
If you are thinking ten years ahead, plant a tree, ll Temperature of Freshly Mixed Hydraulic –Cement
If you are thinking hundred years ahead, educate people” Concrete.
India is second largest manufacturer of cement in the ll Sampling Freshly Mixed Concrete.
world. Due to large population and wide gap between ll Slump of Hydraulic-Cement Concrete.
the projected demand and supply, the production will be ll Density (Unit Weight), Yield, and Air Content
beyond 400m tons by 2017 at CAGR of 10% plus. (Gravimetric) of Concrete.
ll Air Content of Freshly Mixed Concrete by the Pressure
Government has allocated in current plan one trillion Method.
US $ to construction sector where major beneficiaries ll Air Content of Freshly Mixed Concrete by the Volumetric
are sectors like housing, infrastructure, Industrial Method
and commercial structures. In the current alarming
ll Making and Curing Concrete Test Specimens in the
international fiscal scene India is the next hub and this will
Field.
trigger foreign investment and in turn the construction
rate. Continuous Growth in cement and construction are The above seven parameters influence the ultimate
imperatives for the projected growth. Hence, for durability performance of set concrete. Hence, when these are
and desired usability of these structures for a given measured properly as per uniform methodology, they
service life and beyond, we need to look at the following can give very important information about the behavior
aspects of concrete: of the set concrete. For example concrete temperature
a) General understanding about concrete as a material. is one of the most important factors influencing the
b) Good understanding about the mechanism of hydration quality, time of set and strength of the concrete. Without
of cement. control of the concrete temperature, predicting the
concrete’s performance is very difficult, if not impossible.
c) Awareness about good concrete practices
Temperature can give us following indications about
d) Quality assurance and control measures of making properties of set concrete.
good concrete.
1) A concrete with a high initial temperature will probably
e) Testing of concrete in green state. have higher than normal early strength and lower than
f) Testing of concrete in hardened state for required normal later strength. The ultimate overall quality of
performance. the concrete will also probably be lowered.
2) Conversely concrete placed and cured at low
temperatures will develop strength at a slower rate
Concrete Field Testing Course of ACI
but ultimately will have higher strength and be of a
In practice, none of the above requirements are considered higher quality.
in any of the contemporary engineering degree or diploma
3) The temperature of concrete is used to indicate the
programs in India. As a result, engineers who join the
type of curing and protection that will be needed, as
industry have to acquire concrete related experience
well as the length of time the curing and protection
on site and through self motivation to excel in concrete
should be maintained.
related knowledge. Thus, there is a conspicuous need of a
structured program which can train and certify concrete 4) By controlling the concrete temperatures within
practitioners in various aspects of concrete and related acceptable limits, immediate and future problems may
activities. Such a structured program was devised by be avoided.
the American Concrete Institute (ACI,www.concrete.org) 5) In today’s context concrete temperature affects the
about 35 years ago in the form of the ‘ACI Certification performance of chemical admixtures, air entraining
program’. admixtures, pozzolanic materials and other additives

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xxiv ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Right Concrete, Right Way …. Spreading ACI Concrete Field Certification Course in India “Train The Trainer” Initiative Continues….

and admixtures: The detailed explanations about the Vinicio who continuously supported the motion of taking
measurement of temperature of concrete gives the certification to India. At this juncture, Dr. Manjrekar has to
concrete practitioners insight into importance of other also acknowledge the exceptional interest taken in India’s
remaining six parameters dealt with in ACI Certification concrete Industry by ACI International Past Presidents
Course. Indeed these parameters give ample notably, Anthony M. Fiorato, James R. Cagley, Thomas D.
information about the probable behavior of set concrete. Verti, David Darwin, Luis E. Garcia and Florian G. Barth.
Some of them have also visited India. Executive Vice
ACI has been administering the “Concrete Field Testing
President, Bill Tolley was the friend, philosopher and guide
Technician Grade-I Course” through local ACI Chapters
of the India Chapter for the last three decades. They first
and other organizations. About 5,00,000 concrete
met in 1990 when, Dr. Manjrekar was honorary secretary
practitioners are working all over the world who now are
of the Chapter and he was Senior Managing Director of
certified by ACI and taking care of concrete knowledgeably.
ACI, looking after international chapters. Dr. Manjrekar
Amongst all the certifications, ACI Concrete Field Testing distinctly remembers him, telling Dr. Manjrekar on 2nd
program has become so successful that it has virtually Dec. 1990, in presence of then ACI President Dr. John
become the industry’s standard entry-level qualification. Hanson, during the reception hours of the conference,
The tremendous response to this course has been that for India’s concrete related problems the ‘certification
largely due to the cooperation received from industry program’ would be most ideal and not only ‘India Chapter’
stakeholders, in recognition of the fact that proper but the government authorities also should look into it.
handling of concrete leads to fewer disputes and results As a result of the collective efforts over the years, ACI
in higher quality and on-schedule projects. international passed a proposal at their Director Board
The India Chapter of ACI organizes this knowledge meeting to send two ACI experts to India to propagate
dissemination program as a local sponsoring group and ‘Field Testing Certification’ course. It was going to be a
currently the 24th batch is in progress. As of date, there are modal run to asses whether such initiatives could be taken
more than 500 certified professional engineers in India. in other countries too. Professor Luke Snell and Mr. John
Conn were deputed to India by ACI and their meetings
History of Aci Certification in India were firmed up with engineers from Mumbai, Bangalore
‘Twenty first century belongs to the Knowledge age, where and Delhi by the ACI India Chapter.
acquisition, possession and application of knowledge is The historic turn, came from M/s. Larsen & Toubro (L &
the most important resource.’ T), the engineering majors of India takes great measures
Often, we have come across disasters that could have to improve the skills of their engineers on a continual
been avoided, if a structured program of quality had basis. As a part of the skills improvement initiative, 23 of
been implemented at site. To meet this need, late Mr. R. their senior managers and regional heads were chosen to
N. Raikar and Dr. Manjrekar had been interacting with attend the certification training and undergo the exams.
ACI International since the year 2003 to work out the These engineers were from all over the country and some
possibility of implementing certification programs in the of them were from the Middle East as well. Mr. Vivek B.
Indian sub-continent through the India Chapter of ACI as a Gadgil – Sr. Vice President of L & T took a very learned
local sponsoring group (LSG). interest in this path breaking initiative for L&T. He even
attended part of training session held by Prof. Snell after
The India Chapter of ACI, in consultation with ACI
inaugurating the same, at L&T’s Manapakkam Complex
International decided that Dr. Manjrekar should undergo
in Chennai. L&T treated this course as ‘Train The Trainer’
the course. Eng. Khaled who is now incoming President of
initiative, thereby meaning that the successful certified
ACI, used to conduct this course in Qatar, where he invited senior Managers would in turn train their juniors who in
Dr.Manjrekar to attend it, as an observer or to undertake turn would train their subordinates. This was a privileged
the certification. Hence, Dr. Manjrekar decided to undergo batch to get the opportunity to take the training from Dr.
the ordeals of taking the course and the field concrete Snell and also to get the distinction of being the maiden
testing certification exam in hot scorching weather of certified batch in India. All the engineers, though at very
50° Celsius. The examiner, Eng. Abdul Kader deserves senior level did the course in total seriousness and all
compliments for not showing any sympathy towards his of them got through the examination with flying colours
grey hair, old age of 58 years and a fractured index finger. and joined international community of ACI certified
He made Dr.Manjrekar do all the practicals with total personnel.
seriousness and integrity. Dr. Manjrekar greatly indebted
to him, as his skills were honed due to this approach Progress of Aci Certification in India
and the ‘no nonsense’ well defined ACI exam pattern.
'We make a living by what we get, but we make life by
The India Chapter of ACI is extremely thankful to then
what we give.’
International certification committee chair, Khaled Awad
as well as other committee members viz. Mario, Francois, Enthused and energized by the initial success of L&T, India
Alejandro, Donato, Roberto, Jorge, Wally, Raul, Mostapha, Chapter of ACI decided to take this initiative to benefit

Organised by
India Chapter of American Concrete Institute xxv
Inaugural Session

other organizations throughout the country. In the initial vii) Substantial penalties and the possibility of required
stages, the resources in terms of trained and devoted removal and replacement of concrete with low
expert faculty to train the candidates on a sustained strength results make it essential that the fresh
basis, as well as the laboratory facility becomes the key concrete tests be properly performed, and that the
to popularize the course. In order to complete the long technicians employed possess the confidence of the
cherished mission of the chapter and ACI international, Dr. project managers.
Manjrekar took up the task of training the candidates as viii) This initiative can avoid the errors and mishaps by
a social responsibility and in a structured manner taught introducing “Field Concrete Testing” culture in or as a
concrete to the candidates from all over the country. The “third party quality implementation and monitoring.”
sessions are conducted in his office conference room with
proper audiovisual arrangements.
Train the Trainer Initiative
Importantly, M/s. Structwel Designers, offered laboratory ‘When learning is purposeful, creativity blossoms; when
facilities both for a full day practice session as well creativity blossoms, thinking emanates; When thinking
as for written and practical examination for which the emanates, it transforms into knowledge; And when
question papers were sent and assessed directly by ACI knowledge is shared, economy flourishes.’
International, Detroit USA. Availability of free lab facility,
along with trained manpower, has resulted in offering the Initially to spread this light of knowledge and than to
certification course at a much subsidised cost, compared share it further, we need as much support, in terms
to contemporary international costing of the course. of dedicated trained brainpower. Who else could be
this team of dedicated trained brainpower but the
trained certified personnel from the last 23 batches?
This course is taught and conducted as “TRAIN THE
TRAINER” initiative. Trained and qualified engineers
alone can spread this light when they are certified
themselves. Hence, extraordinary importance is given
to teaching about cement concrete hydration and good
concrete practices along with inbuilt ‘Field Testing’. This
value addition, not only gets the candidates enchanted,
enthralled and enlightened during the course, but
they definitely get inspired to train and educate their
peers, juniors and fellow engineers by disseminating
Fig. 1: the knowledge that they acquired. There have been
several such heartfelt feedbacks from the successful
Direct advantages of ACI Certification certified engineers. This indeed is ‘TRAIN THE TRAINER’
The first success story of L&T in Chennai, along with initiative as envisaged and it proved to be successful.
subsequent passing out of 23 batches brought out the Several private, public and government organizations
following conclusions which are accepted universally and have participated in this program benefitting the cause
also applicable to the Indian concrete scenario: of quality for their organization and thus serving the
i) Everyone involved in a construction project benefits national cause of producing durable structures. All these
organizations and heads of the institutions have a vision
from the use of certified technicians to perform the
to create a better tomorrow. Hence, they participated in
field tests on freshly mixed concrete.
the program with total conviction.
ii) Proper performance of the tests improves the
reliability of the test results.
Vision and Leadership is Key to Quality
iii) It aids in quality control of the concrete and can
I must cite here, an outstanding example of training and
minimize costly delays resulting from a lack of certification of a contingent of HUNDRED senior engineers
confidence in the test results. of Maharashtra Housing and Area Development Authority
iv) Most importantly, proper field testing assures (MHADA). All of us know MHADA as the only organization
accuracy in the identification of good quality concrete which deals with construction of houses in masses, for
and sub-standard concrete. masses, throughout Maharashtra.
v) Complete and accurate records filed by the certified It all began in 2008 on ‘Engineer’s day’ celebration of
technician are essential in the event of a dispute. MHADA where Dr. Manjrekar had an opportunity to be the
vi) The increasing use of ‘end result specifications’ keynote speaker. Dr. Manjrekar initiated a dialogue about
is another reason for having trained certified this certification course with the dynamic Vice-President
technicians on construction projects. & Chief Executive Officer, Mr. Gautam Chatterjee (IAS).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xxvi ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Right Concrete, Right Way …. Spreading ACI Concrete Field Certification Course in India “Train The Trainer” Initiative Continues….

always convinced in his mind that if the learning comes

to any educated personnel in the form which is extremely
interesting, connected with day to day operations and if the
teaching is, involved and full of concrete expertise, at any
age engineers find themselves absorbing involved. Even
at the end of a tiring day, which is full of administrative
hassles, engineers get reinvigorated and tuned to their
channel of knowledge. His experience was absolutely
reassuring. Some of the engineers, typically seniors came
Fig. 2: ACI International then President & then Vice President up to the course with lots of apprehension. However, as the
for giving certifications. (3rd & 4th Sept. 2010) Ref.: International concrete got more and more demystified and simplified
Conference on Sustainable Technologies for Concrete in front of them, it was a pleasant metamorphosis into a
Construction – 3rd & 4th Sept. 2010 serious and sincere student wanting to know more about
concrete. Their zeal and enthusiasm to excel in the course
Mr. Mohan Jacob, Past President of India Chapter who enhanced by every learning session was a gratifying
has been closely associated in the efforts of bringing ACI experience. For him, an equally tired man at the end of
certification to India, further deliberated with Mr. Gautam the day, to teach every new lecture with added information
Chatterjee about how this initiative can bring about “skills and enthusiasm, session after session was possible, only
improvement” and enhancement of concrete related due to the matching willingness from the candidates to
knowledge in MHADA engineers. Mr. Chatterjee was so learn. In all the students, the discipline and eagerness to
deeply involved in the betterment of skills of his engineers absorb more and more, even at a mature stage in life and
and the qualitative output of MHADA as an organization, higher organizational position in engineering career was
that he readily nominated a block of 50 Engineers for this exemplary. He gives the entire credit to bring out these
training. These engineers were divided into five batches qualities of the department’s engineers to progressive
and attended the course with other engineers from the outlook to visionaries like Mr. Gautam Chatterjee. MHADA
diversified background of organizations in the Industry. structures are in the safe hands of, learned and trained
This created a healthy blend of engineers and the sharing concrete practitioners and posterity will remember the
of knowledge became still more valuable. For example likes of Mr. Gautam Chatterjee.
along with MHADA engineers, others from reputed MHADA and other organizations have has acknowledged
organizations like Maharashtra Industrial Development the fact that ‘Tomorrow’s world would be one which would
Corporation (MIDC), Hiranandani Constructions Pvt. Ltd., recognize knowledge in its most comprehensive form
Untiy Infraproject Pvt. Ltd., B. G. Shirke Construction and add further value to products through innovative
Technology Ltd., Larsen & Toubro Pvt. Ltd., Lafarge knowledge and these knowledge products would largely
Aggregates Concrete India Pvt. Ltd., Godrej Properties contribute to the growth of nations’. And in order to make
Ltd., Structwel Designers and Consultants., Hirco,
this growth all pervading and sustainable in India more
Sunanda Speciality Coatings Pvt. Ltd. , Reliance Industries,
and more organizations and Government departments
RMC Ready Mix., Shashank Mehendale & Assoc.,
like PWD, CPWD, Irrigation, Railways, MES, etc will have
B.E. Billimoria., Ultratech., Sunshine Group., Kraheja
to undertake this training. Dr. Manjrekar is confident that
Corp., Nina Concrete., Shirish Patel & Asso., Shashank
the example of MHADA will become a leading light to all
Mehendale & Associates., SP Consultants., Epicons
other government departments to emulate and spread
Consultants., Roads & Builds. Dept.(Vadodara), Thornton
the movement of creating healthy and durable concrete
Tomasetti, Bhagubhai College of Polytechnic and many
structures through certified quality control personnel.
more, also have taken a keen interest by deputing their
engineers for this ‘Train the Trainer’ Initiative. Another example is of the doyen of engineering fraternity,
Padmashree Dr. E. Sreedharan who is now leading Delhi
‘Concrete Field Testing Technician Grade – I encouraging
Metro Rail Corporation (DMRC). He has seen the potential
popularity of recently launched training and certification
in ACI training and requested Dr. Manjrekar to take the
course by India Chapter of ACI, as local sponsoring group’
introductory sessions to ACI certification in Delhi. This
course was taken by about 55 engineers in Delhi on 21st &
Gratifying Experiences and Impressions of a 22nd August 2009.
Teacher Some of the engineers have already stopped the wrong
Now, I must share with you Dr. Manjrekar’s experiences practices which were up to now considered normal (due
with training of the candidates in last 23 batches of concrete to ignorance) like using reinforcement bar for rodding
engineers. Dr. Manjrekar has had the pleasant privilege the cubes, taking wrong samples from the first portion of
to impart the learning to all the cadres of engineers concrete as it comes to the site, not taking temperature
from Chief Engineers, Deputy Chief Engineers, Executive at all or taking it wrongly, rotating the slump cone etc.
Engineers and their junior colleagues. Dr. Manjrekar was to name a few. Even if these small things are taught to

Organised by
India Chapter of American Concrete Institute xxvii
Inaugural Session

on 14th March 2015.

4. For 23rd batch candidates were from all over
Maharashtra like Pune, Latur, Nanded, Warora,
Butibnori, Amravati, Ambernath, Taloja, Roha,
Mahape, Thane, Dombivli and Andheri attended the
course. Candidates from Reliance Industries Limited -
Jamnagar, R & B Department – Gujarat - Gandhinagar
and M/s. Thornton Tomasetti - Mumbai attended the
Fig. 4: Dr. Surendra Manjrekar teaching the engineers of the course.
Delhi Metro Rail Corporation
Certification “Train The Trainer” Initiative
rank and files of the department by the 200 plus certified
Spreads at four Corners of the Country
trainers, it would in a year reach to 2000 concrete related
manpower all over India at the rate of 20 men per person ll Chennai Maiden Batch held in CHENNAI – 6th August
trained by these TRAINERS. One can only imagine the 2008
cascading effects of this “TRAIN THE TRAINER” initiative ll (Sr. Engineers from All over India & UAE)
in subsequent years reaching to safety of 10000 structures ll Mumbai - 18 Batches within the geographical on radius
at the conservative estimate of 10 structures/ year. 500k.m.
In conclusion, today Indian nation has more than 500
ll 20th Batch held in Bengaluru –28th February 2014
‘Concrete Field Testing’ ACI certified personnel. These
trainers have been trained in last two years. Another 200 ll 22nd Batch held in New Delhi - 14th March 2015
engineers are lined up for training. Together these 600 plus
The diplomas were given in a grand ceremony in the
trainers would take this knowledge about good concrete
presence of Principal Secretary PWD - Shri Arun Baroka
practices to say 60,000 practitioners / field technicians.
and Ron Burg - Executive Vice President of ACI, in the
But is this number enough to reach the concrete of entire
country? presence of Dr. S. K. Manjrekar and Shri Sarvagya Kumar
Srivastava, ADG, PWD Secretariat, who was the main
Obviously not. Hence this movement of “Concrete Field
motivator of the initiative in capital city.
Testing” should percolate to nooks and corners and at the
India Chapter of ACI, we are all determined to endeavor
for the same.
The Scope of Aci Certification in India
Thus, we come to the scope of concrete field Testing
initiative. Like mentioned earlier construction sector
Some Salient Features of the Journey of ACI
in India is the second largest economic activity next to
Certification throughout India agriculture and employs 33 million people. This lack of
1. 20th batch was held in Bangalore on March 2014. skilled personnel / technicians substantially lowers the
2. For the 21st batch candidates had come from Gurgaon, productivity and losses due to unforced errors. If the
New Delhi, Gujarat, Panvel (Raigad), Navi Mumbai skilling efforts are not taken seriously, then in year 2022
(Badlapur, Khairane), Mumbai (Vile Parle, Powai, when construction sector is expected to employ 83 million
Vikhroli, Ghatkopar) & Thane. people situation could be alarming as substantial portion
3. 22nd batch was held in New Delhi for PWD engineers of 83 million will be in concrete and related fields.

Fig. 3: Dr. S.K. Manjrekar, Principal Convenor, IC-ACI with participants of Delhi Metro Rail Corporation (DMRC) at the introductory
course on “Concrete Field Testing Technician Grade – I” held at New Delhi

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xxviii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Right Concrete, Right Way …. Spreading ACI Concrete Field Certification Course in India “Train The Trainer” Initiative Continues….

As of today there is no established competency standard or

training module at the national level. Hence, government
of India has established National Skill Development
Corporation Ltd., in 2008 as not for Profit Company
licensed under section 25 of companies Act. NSDC has
targeted to train 150 million people by 2022 out of an
incremental demand of 240 million as estimated. Above
data on amply underlines the need of several agencies /
programs to cater to the requirement.
Fig. 6: Dr. Manjrkar and Ron Burg Executive Vice President of
Can Aci Concrete Field Testing become an ACI explained the initiative of certification to Director General
Option? of CPWD – Shri Divakar Garg and others in Delhi

practitioners / consultants / Federal Departments…etc.

B) The course is in English language and in India at
technician’s level only local languages are medium of
communication. Hence, the course cannot be taught
and administered to technicians or field operators in
English. Hence, Train the Trainer or certify the trainer
imitative needs to be adopted.
C) As far as the adoptability of the course throughout the
targeted manpower of 2022 is concern strong efforts
Fig. 5: 22nd Batch of ‘Train the Trainer’ certification course are required. ACI, India Chapter and local sponsoring
– PWD Delhi - 13th May 2015 at New Delhi - Certificates group / s, need to collaborate and convince federal
distribution ceremony authorities. It will be good idea to discuss the initiative
with NSDC. This is important.
There are positives and also challenges in obtaining a D) Subsequent to the limited success (due to qualitative
leadership. Positives have already been discussed earlier nature of ACI course) organizations like NRMCA
in this paper. Now the factors which are challenging may and others have started the similar initiative in India
be thought about as under: - with the help of parallel concrete bodies. In fact this
is a needed approach, for this huge country as many
This would allow us to find the solution for the greater
training, skilling on certification efforts are required.
interest of the concrete. Some of the challenging points
to be pondered over (which may increase in time to come) As we travel along there will be more solutions in the
are as follows: skills improvement process to better the concrete in India.
A) ACI certification is based on ASTM standards. In India
throughout the country Indian code BIS 456 is followed. References
ASTM has no acceptance or cross reference in Indian 1. www.concrete.org
2. Indian Construction, Vol. 45, Nov. 2012
standard. This lack of choice confuses many specifiers /

Ms. Ishita Manjrekar

Ishita Manjrekar is a Director at Sunanda Speciality Coatings Pvt. Ltd., and oversees Sunanda’ s R
& D, with a specific focus on developing and marketing Sunanda’s line of sustainable construction
chemicals.
In this role, Ishita draws on 4 years of rich experience in sustainability at Primary Global Res
earch in San Francisco and New York. While at Primary Global Research Ishita led the “Cleantech
and Green Technologies” business unit. She has been invited to feature on Bloomberg TV as well
as Bloomberg Radio numerous times as a subject matter expert on sustainability and green
technologies. Ishita expertise has also been sought by multiple print media including Reuters,
Financial Times, Bloomberg, Forb es, BBC News and Marketwatch in USA.
Ishita is member of ACI International and works actively on various international board appointed as well as technical
committees. She currently serves on the ACI International Board Committees for International Advisory Committee,
ACI Membership Committee, ACI Marketing Committee, International Project Awards Committee, IPAC Judging
Subcommittee and Student & Young Professionals Activities Committee.
Ishita serves as Honorary Secretary on the Board of Direction of the India Chapter of American Concrete Institute.

Organised by
India Chapter of American Concrete Institute xxix
Inaugural Session

Head Office:
C/8, Sunita Bldg., Varsha Society, L.B.S. Road, Naupada, Thane (W)-400 602.
Tel.: +91-22 25303692 / 25437709 / 25423539
Cel: +91-9892466282
Email: rehabartifact@gmail.com
www: rehabartifact.com

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xxx ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
PLENARY SESSION
 With Best Compliments
from

SANJAY CONSTRUCTION
Residential & Industrial Construction & Rehabilitation of RCC
Admn. Off.: Shop No.3, Amrapali CHS, Ltd., Pirojshanagar,
Next to Railway Crossing, Vikroli (East), Mumbai - 400079.
E-Mail: sanjayconsl@yahoo.com
Web.: www.sanjayconstruction.co.in
Tel.: 022·257411 55
Fax: 022·25741166

SANJAY RATHOD
Mob.:+91-9 820332265

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xxxii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on
viii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 With Best Compliments
from

Phoenix Contracts & Services (P) Ltd. is a Civil Engineering and

Construction Company carrying out various work in the field of
Civil, Structural, Rehabilitation of old structures, Special Painting
of structure and associated jobs, Construction of Building and
associated jobs including architectural services.

Phoenix Contracts & Services (P) Ltd.

3B/702, Anand CHS. Ltd., Natwarnagar, Road No.5,
Jogeshwari- East, Mumbai -400 060.
Contact : Mr. V. Seshadri
Mobile : 98694 39072
Landline : 022 28383456
Email : phoenixc@vsnl.com / phoenixc09@gmail.com

Organised by
India Chapter of American Concrete Institute ix
 With Best Compliments
from

ROYAL CONSTRUCTION LIMITED

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

x ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 With Best Compliments
from

MOHITE INFRAPROJECTS PVT. LTD.

Organised by
India Chapter of American Concrete Institute xi
 With Best Compliments
from

SMC INFRASTRUCTURE P. LTD.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 With Best Compliments
from

SUHEB ZUBER DESHMUKH

Organised by
India Chapter of American Concrete Institute xiii
 With Best Compliments
from

GOVARDHANI CONSTRUCTION CO.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xiv ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 With Best Compliments
from

TRITECH ENGINEERING ENTERPRISES

Organised by
India Chapter of American Concrete Institute xv
 With Best Compliments
from

ANUKUL CONSTRUCTION

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

xvi ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Technical Papers

With Best Compliments

from

Mahendra & Co.

Engineers Contractors & Interor Decorators
An ISO 9001 :2008 Certified Company
A-703, Koldongri Society, Paraswada,
Sahar Road, Andheri (E), Mumbai - 99
Tel.: +91 22 28374868
Telefax : +91 22 28269238
Email: mahendra_co@rediffmail.com / mahendraco2012@gmail.com
Website : www.mahendraco.com

Mangesh Sharma
Mobile : 9821802062

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

viii ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolution of Unbonded Post-tensioned Concrete in the United States, and its Implications for Lifetime Performance and Restoration Philosophies

Evolution of Unbonded Post-tensioned Concrete in the United States,

and its Implications for Lifetime Performance and Restoration
Philosophies
Kyle Stanish, Ph.D., S.E., P.E. and Daniel E. Moser, S.E., P.E. Walker Restoration Consultants
Chicago, Illinois, U.S.A.

Abstract Unbonded post-tensioning construction and materials

have developed over the years. A parking structure
Practical unbonded post-tensioned concrete for use in
constructed with post-tensioning in the 1950’s is not
buildings has been around since approximately the 1950’s.
the same as one that would be constructed now. As
Unbonded post-tensioned reinforced concrete structures
they evolved, they have been improved in a number of
have become a more popular structural system as they
ways. (Post-tensioning Manual, 2006) This ranges from
have evolved, and are currently used in a wide variety of
improvement of the quality of the concrete to improving
building types. The long-term performance and behavior
design details such as improving drainage. This paper
of these structures is very different than for conventionally
concentrates on the changes to the hardware used in
reinforced structures. Evolving technology and building
unbonded post-tensioning and what impact this has had on
code requirements have also impacted the expected long
its long-term performance. Also discussed are different
term performance and possible restoration approaches.
ways of repairing unbonded post-tensioned concrete
Of particular importance is the development of fully
structures that have compromised post-tensioning
encapsulated tendons and the requirement to include a
tendons.
minimum bonded reinforcement, usually conventional
reinforcing steel. Post-tensioning is used as structural reinforcement in all
of the flexural members, including both beams and slabs.
Experiences with the long term performance of unbonded
However, as the slabs are the most directly exposed to
post-tensioned concrete parking structures in aggressive
the environment, they are where distress normally arises
environments are discussed. Typical failure modes are
sooner. This paper thus focuses on the performance of the
reviewed, including how the impact differs for buildings
slab post-tensioning, and not beams.
of different eras as the construction practices changed
in response to changing understanding, technology and
code requirements. Paper-Wrapped Button Head Post-Tensioning
Keywords: Concrete; Unbonded Post-tensioned Systems (1950’S and 60’S)
Concrete; Repair; Service Life Some of the oldest practical post-tensioning systems
are button-head systems. They consist of a steel wire,
typically 0.25 in (6 mm) diameter. Multiple wires are
Introduction
used in parallel in a bundle, ranging from typically 4 to 12
Unbonded post-tensioned concrete construction began wires for slabs. They are anchored by threading the ends
to be developed for widespread construction in the through a plate and then creating a cold-formed “button”
1950’s in the United States. Post-tensioned construction on the end of the wire. (Photo 1) They were then coated
has become widely adopted as it has some advantages with a lubricant and spiral wrapped in heavy paper (kraft
over conventional construction. It can have longer spans paper) prior to casting of the concrete. This allowed the
with thinner sections. It also may provide “crack free” wires to move and stretch when stressed. Stressing was
concrete. One building type where unbonded post- performed after the concrete had hardened and gained
tensioning is currently widely used is standalone parking a minimum strength, and was done by separating the
structures, many of which are considered to be in an plates at an intermediate location using a jack. After initial
aggressive environment. The vast majority of standalone stressing, the wires were locked in place using wedges.
parking structures are currently constructed out of either (PTI DC80.3-12/ICRI 320.6)
unbonded post-tensioned concrete or precast concrete
(with bonded prestressing tendons) (ACI 362.1R-12). The Limitations of System
longer clear spans that post-tensioned construction
allows means cars can park more easily without columns The life of this system is limited by its resistance to
interfering. corrosion in some structures. For parking structures,
particularly those in cold climates, corrosion is initiated

Organised by
India Chapter of American Concrete Institute 1
Plenary Session - Paper 1

strength, a small section is left out of the initial concrete

placement. This area is used to provide access to the
anchorage plates. Once the wires have been stressed and
locked in, the remaining concrete is placed. This inevitably
creates a joint between the two different placements
of concrete. This joint is typically initially protected by
sealant, but this joint sealant needs to be maintained. It
typically needs to be replaced every 5-7 years, depending
on exposure. After this time, the sealant fails and chloride-
contaminated water can enter the joint and penetrate to
the level of the steel, leading to corrosion.
Button-head post-tensioning is typically a paper-wrapped
system. The paper, while greased to be able to allow the
wires to move during stressing, does not provide long term
Photo 1: Button-head post tensioning
protection. Chloride-contaminated water will be absorbed
by the paper, and pass through to the level of the steel. In
by the presence of chloride ions, typically from deicing
addition, at the anchors, the wires typically are splayed,
salts that are either applied directly to the structure, or
as each wire is passed through a separate hole arranged
are brought in by vehicles from the deicing salts applied
around the anchorage plate. This separation makes it
to roads (ACI 362.1R-10). Chloride-contaminated water
difficult to evenly wrap the paper around the wires in this
then penetrates either through the pore structure of the
area, and thus provides even less protection. The result of
concrete, or much more rapidly through open cracks or
this is that once the chloride ions have reached the level of
joints. Once the chloride levels have built up to a threshold
the steel, they can begin to corrode immediately.
concentration at the level of the steel, corrosion begins,
compromising the performance of the structure (Photo 2).
Repair Philosophy
Once corrosion has begun, this results in a number
of effects on the structure. Just as for conventional
reinforced concrete structures, the corrosion product
exerts an expansive force on the concrete, leading to
spalling and delamination. The spall can be repaired in
a similar manner to those encountered in conventional
reinforced concrete structures (Newman, 2001; ACI RAP
Bulletin 7).
As corrosion progresses in the wires, however, they are
still under tension. As the corrosion leads to section loss,
the force remains the same, but the stress increases until
it reaches a point when the remaining portion of the wire
or wires break (Photo 3). Some structures can experience
significant wire failure yet still not exhibit cracking or
Photo 2: Corrosion started on a Button Head system

Properly designed post-tensioned concrete structures

tend to have a greater resistance to chloride penetration
than conventional reinforced concrete structures.
Whereas conventional concrete structures will typically
crack in service, a post-tensioned structure is designed
not to. This eliminates a potentially direct path for rapid
chloride ingress to the level of the steel. Chloride-
contaminated water can only reach the level of the steel
by travelling relatively slowly through the concrete pore
structure or rapidly by travelling through joints or cracks.
For button-head post-tensioned structures, joints are a
necessity however. They will be present around all of the
stressing pockets that are needed to stress the wires.
As the wires are stressed after the concrete has gained Photo 3: Broken Button-head post-tensioning wires

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

2 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolution of Unbonded Post-tensioned Concrete in the United States, and its Implications for Lifetime Performance and Restoration Philosophies

other signs of distress. A broken wire is frequently the Development of Seven-Wire Strand and
first indication that there is corrosion occurring, and this Plastic Sheathing (1960’S and 70’S)
is more often expressed indirectly, by flexural cracks
developing the structural slab, although occasionally wires To overcome some of the drawbacks with button head
will break out through the slab surface. It is recommended systems, seven-wire strand and loose plastic sheathing
when evaluating button-head post-tensioned structures was developed.
that the condition of the wires be evaluated by carefully Seven-wire strand is a form of reinforcement where
exposing the wires and visually examining their condition the high strength steel wires (now smaller in diameter)
and checking if they are under tension. Extreme caution are wrapped together to form something more similar
must be used when opening up the concrete near end to a rope (Figure 1). There is one central wire and six
anchors as they are potentially still under tension. wrapped around it. They are typically used as 0.5 in (13
When the corrosion has progressed to the level where mm) diameter strands, but other sizes are possible. They
the tendons are breaking, the repair is more complicated typically are either used singly, or as pairs at one location
than for conventional reinforced concrete. If a mild steel in the slab. The advantage of this type of reinforcing is that
bar has significant corrosion, it can be supplemented by it can be stressed at the ends of the strand. The strand is
addition of a steel bar adjacent to the corroded one, as passed through a plate that has a conical wedge anchor.
long as it is developed on either side of the location where (Photo 5) The strand can be pulled from the end using a
the corrosion is occurring (Newman, 2001). For post- jack, and when the load is released the strand is kept from
tensioned concrete, not only does the damaged tendon retracting by the wedge anchors.
need to be replaced, but it needs to be tensioned.
In order to repair the corroded tendons and restress
them, a new section of tendon needs to be spliced in.
Typically seven-wire strand (discussed below) is used
to replace the section of corroded wire that is removed.
The size and number of strands that are used will depend
on the number of wires that are present, and should be
determined by an experienced engineer. The splicing
is accomplished by exposing the wires at the site of the
break, removing the corroded steel wires, attaching the
clean ends of the wires to a plate by forming new button
heads, and then using the plates to anchor the new section
of strand. A jack is then used to stress the new strand, Fig. 1: Composition of Seven-Wire Strand and Sheathing (PTI
transferring the force into the plates with wedge anchors. DC80.3-12)
This also serves to stress the wires over their length
(Photo 4). The force that should be applied should be
established by an experienced engineer. During the repair,
the need for shoring of the area should also be evaluated,
depending upon the number of locations where breaks
have occurred.

Photo 5: Typical Monostrand Anchorage Hardware

When strand was initially developed, it was often paper-

wrapped, just as button-head strand was. This was
replaced by a loose fitting plastic sheathing that either
had the steel pushed through, or an open plastic strip
was wrapped around and heat sealed. These systems
did typically have a rust inhibiting and lubricating coating,
although it frequently did not completely fill the space
between the sheathing and the strands. (PTI DC80.3-12/
ICRI 320.6)
Photo 4: Repair and restressing of Button head post-tensioning
wires

Organised by
India Chapter of American Concrete Institute 3
Plenary Session - Paper 1

Limitations of System
The use of seven-wire strand has a number of advantages
over button-head for durability. The ability to stress
the strands at the ends rather than in the middle of the
wire means that an intermediate stressing pocket is not
necessary. It is simpler to protect the ends of the tendons
on the outside of the structure than at a mid-point in the
slab, where button-head was most commonly stressed.
Stressing button head tendons at their ends was possible,
but rarely performed in practice. Sometimes intermediate
anchors are necessary for seven-wire strand, such as at
cold joints between different pours, so the need to protect
the joints is still present just as for button head strands,
but the configuration was more compact so easier to
Photo 7: Tendon lowpoint with water pouring out when cut. (PTI
protect.
DC80.3-12)
Stressing the strands at the ends did have a limitation,
however. As the stands are stressed, they slide through the gap around the steel within the sheath, there is less
the anchor plate, due both to elongation of the strand and likelihood of spalls in the concrete. The corrosion product
shortening of the concrete. The plastic sheath did not has room to expand within the sheathing before significant
have any allowance for movement, however, and it was pressure is developed in the concrete. It is not uncommon
necessary to strip a section of sheathing from in front of to find a large number of broken strands in a slab that has
the anchors. This would leave a section of the tendon steel relatively little spalling.
unprotected. (Photo 6) This was particularly problematic
Without widespread spalling as an initial warning sign,
at intermediate anchors, where the exposed tendon steel
the first signs of distress are different for this type of
is at a joint that can allow the penetration of chloride-
structure. Most obviously, broken strands can be seen
contaminated water.
protruding from the surface of the slab. (Photo 8) When
the strands break, they retract with sufficient force that
they break the concrete cover and “loop” out if the cover
is relatively thin. The other warning sign is when flexural
cracks begun to develop either on the underside of the
span (at midspan) or over the beams on the topside of
the slab. While these are normal cracks for conventional
reinforced concrete structures, they should not develop
for post-tensioned structures due to the precompression
force that is provided by the tendons. The presence of
these cracks indicates that this precompression force is
being lost in the concrete, likely due to broken strands.
For post- tensioned structures of this type, due to the lack
of conventional steel the cracks are frequently wider than
for conventional structures when they do develop. Both of
Photo 6: Tendon stripped of sheathing. (PTI DC 80.3-12) these signs do not show up, however, until at least some of
the strands are broken, which is at a more advanced stage
Although the plastic sheathing did a better job of resisting of deterioration than the first warning signs show up for
chloride-ingress than a paper wrap, it does not provide conventional concrete structures.
a continuous barrier. There were frequently gaps in
the heat sealing (if relevant) as well as damage during
construction, in addition to the aforementioned locations
where the sheathing was purposefully stripped. The loose
sheathing with its gap around the strands then allowed
contaminants that have penetrated to travel along the
length of the strand, frequently gathering at lowpoints or
pinches in the sheathing. (Photo 7) The strands will then
corrode starting at these locations.
As corrosion occurs, there is section loss of the strands,
just as for button-head. However, due to the presence of Photo 8: Broken strand looping out of underside of slab

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

4 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolution of Unbonded Post-tensioned Concrete in the United States, and its Implications for Lifetime Performance and Restoration Philosophies

The level of bonded steel that is required is typically low,

with the current ACI 318 (2014) requiring a minimum of
bonded steel area equal to 0.4 % of the concrete area
that can go into tension. The presence of this steel has an
impact on the long-term performance and the implications
of deterioration.

Impact on Long-term Performance

Photo 9: "Dogbone” tendon splice hardware
As previously discussed, unbonded post-tensioning
reinforcement, whether wires or strands, can break due
Once broken, these tendons can be spliced similar to to corrosion with little or no visible distress shown on the
buttonhead post-tensioning, although the hardware is surface of the concrete. The first visible warning is often
different. A similar strand is used as the replacement, so a when tendons begin to break. Unfortunately, the tendons
different splice anchor is used, frequently referred to as a are the main structural reinforcement for the slab, and
“dogbone” anchor due to its shape. (Photo 9) A new splice once broken can go to zero capacity along their entire
would be necessary at every break within a tendon run. length. This lack of redundancy of these earlier systems
As an alternative to splicing, depending on the number results in a much more hazardous situation than when
breaks that are present, it may be more cost- effective to corrosion occurs in conventionally reinforced concrete
install a new tendon. Both end anchors are exposed, the slabs (Post-tensioning Manual 2006).
wedges removed and a jack used to extract the tendon. If there is only a few broken tendons in a region, then the
A new tendon can then be threaded through the same structure may be able to take service level loads, albeit with
opening. (Photo 10) Sometimes a smaller diameter tendon increased deflection and cracking in some instances. The
is used as the replacement, though the feasibility of this hazard develops when the deterioration progresses and
should be evaluated by a structural engineer. the majority of tendons have failed, at least within an area.
The structure may become unable to take even service level
loads, and theoretically could no longer be self-supporting.
Although rare, it is possible for unbonded post-tensioned
structures to collapse with little or no warning solely due
to corrosion. The authors have not been involved with any
structures that have collapsed due to corrosion of the
unbonded post-tensioning, but they have been involved in
a number of structures where the level of deterioration has
reached the point that they have recommended closing all
or a portion of the structure and installing shoring.
The inclusion of bonded reinforcement, while not capable
of resisting the entire design load, does provide a level of
redundancy and ductility to the slab. The bonded steel can
participate in alternative load paths for at least a portion
Photo 10: New tendon being threaded through existing tendon of the design loads. This will decrease the likelihood of
opening collapse without warning.

Requirement for Minimum Bonded Steel Current Post-Tensioning Technology (Post-1985)

Requirements The current construction practice for post-tensioning
As more familiarity was developed with post- systems date to approximately 1985, although it was
tensioning systems, different codes developed different gradually adopted. Post-tensioning tendons are used
requirements for its use. One important requirement that which are fully encapsulated in plastic sheathing.
was introduced in different codes in the United States at Extruded plastic sheathing is used that is tight to the steel
different times was the requirement for bonded steel – tendons, and the remaining area is completely filled with
most commonly fulfilled in slabs with unbonded PT by the an anti-corrosive coating, leaving no room for water or
addition of some conventional mild steel reinforcement. other contaminants to enter into the sheathing. In addition,
Some codes initially incorporated bonded steel to address the anchors are also encapsulated and are designed to
seismic concerns and others required bonded steel allow the movement of the tendons during stressing. Once
when a certain tensile stress threshold was reached. the stressing has been completed, a grease filled cap is
The concept has spread to all post-tensioned structures. applied, providing complete protection. (Post-tensioning
(Post- tensioning Manual, 2006) Manual, 2006). (Photo 11)

Organised by
India Chapter of American Concrete Institute 5
Plenary Session - Paper 1

Conclusions
Unbonded post-tensioned concrete structures have
evolved as our increased experience and understanding
of the structural system has led to improvements in
technology and changing code requirements. The early
button head post-tensioning systems when used in
aggressive environments experienced corrosion as the
strands were not well protected. The protection was
Photo 11: Encapsulated Tendon Hardware, showing cap improved with the next generation of seven-wire strand,
encased in a loose sheath, but corrosion still occurred
Impact on Long-term Performance in aggressive environments. Over time, the sheathing,
Structures built with fully encapsulated systems are coatings and anchorage materials have significantly
now reaching approximately 20 to 30 years of age. Some improved which has led a fully encapsulated unbonded
structures with button-head and loosely encapsulated post-tensioning tendon system. The current technology
tendons have by this age begun to show signs of of encapsulated unbonded post-tensioning strands has
deterioration. This has not proven to be the case for proven to provide significantly improved protection from
structures with fully encapsulated systems. The authors corrosion, with examples of excellent performance with
have evaluated a number of fully encapsulated structures, no corrosion after over 20 years, even in aggressive
including exposing the tendons. In all cases, the tendons environments.
were in good condition, fully stressed and without any
significant visible corrosion. (Photo 12) These structures Code requirements have also evolved. Initially, there was
ranged in age, but were up to 21 years old at the time of no code-requirement for bonded steel in certain post-
evaluation. tensioned concrete slabs. This meant that if the slab
had unbonded tendons and the tendons failed, which
they could do with little warning in certain aggressive
environments, there was little residual strength to ensure
ductility and flexural performance. Current codes require
a minimum level of bonded steel reinforcement, however,
to help address this concern by providing some ductility.
Repair approaches to address broken tendons have been
developed, both for button-head post- tensioning and for
seven-wire strand. These involve either splicing in a new
tendon section or (for seven- wire strand) replacing the
strands completely.

Acknowledgement
The authors would like to acknowledge the contribution of
the American Concrete Institute for their support.

References
1. ACI 318-14: Building Code Requirements for Structural Concrete,
American Concrete Institute, 2014, 520 pp.
2. ACI 362.1R-12: Design of Durable Parking Structures, American
Concrete Institute, 2012, 33 pp.
3. ACI RAP Bulletin 7: Spall Repair of Horizontal Surfaces, American
Concrete Institute, 2010, 7 pp.
4. Newman, A., Structural Renovation of Buildings: Methods, Details
and Design Examples, McGraw-Hill, 2001, 867 pp.
5. Post-tensioning Manual , Post-tensioning Institute, 6th ed., 2006,
355 pp.
6. PTI DC80.3/ICRI 320.6: Guide for Evaluation and Repair of Unbonded
Photo 12: 21-year-old Encapsulated Tendon. Sheath in good Post-tensioned Concrete Structures , Post-tensioning Institute/
condition (top) and Uncorroded Steel (bottom) International Concrete Repair Institute, 2012, 54 pp.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

6 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolution of Unbonded Post-tensioned Concrete in the United States, and its Implications for Lifetime Performance and Restoration Philosophies

Kyle Stanish
Kyle Stanish, Ph.D., S.E., P.E., ACI member, is a Senior Restoration Consultant at Walker Restoration
Consultants in Chicago, IL. He completed his Ph.D. in 2002 at the University of Toronto, Toronto, Canada
on the migration of chloride ions in concrete under an electrical potential gradient. He currently evaluates
deteriorated concrete structures and designs appropriate rehabilitation approaches. He is currently
the Secretary of ACI 365: Service Life Prediction, and a member of ACI 562: Evaluation, Repair and
Rehabilitation of Concrete and ACI 563: Specifications for Repair of Structural Concrete in Buildings.

Daniel E. Moser
Daniel E. Moser, S.E., P.E. is a Principal and a Restoration Department Head at Walker Restoration
Consultants, a division of Walker Parking Consultants in Elgin, Illinois. Dan has over 24 years of experience
with specialized expertise in structural evaluations and restoration design on a wide variety of structures
ranging from 800’ tall reinforced concrete chimneys to buildings and bridges, as well as building exteriors,
and parking structures. Dan’s experience also includes evaluation and repair design for numerous Post-
Tensioned structures. Dan is a member of PTI’s DC-80 Committee on Evaluation and Repair of Post-
tensioned Structures as well as CRT-60 PT Repair Certification Committee.

Organised by
India Chapter of American Concrete Institute 7
Plenary Session - Paper 2

1 Kilometer Tall Kingdom Tower: Concrete Reaches New Heights

Robert C. Sinn
Principal, Thornton Tomasetti, Chicago, USA

Abstract The superstructure frame is composed almost entirely

Construction of the 1000+ meter tall Kingdom Tower of reinforced concrete walls and coupling beams. The
in Jeddah, Kingdom of Saudi Arabia is currently well structural system was developed based on the need for
under way with all tower piling and raft foundation simplicity and repetition during the construction process.
complete. The presentation will focus on the significant The concrete bearing wall system chosen is unique for
technical engineering challenges of designing the ultratall tower schemes in that it relies on no outriggers
next world’s tallest building. A brief overview of the or belt walls, no column transfers, very little differential
architectural and master planning scheme for the tower shortening in vertical elements, and only 85 MPa concrete
and the surrounding developments is presented as an strength. The presentation will focus on the development
introduction to this unique, ground-breaking project. of the tower structural system including historical
Important aspects of the geotechnical site exploration precedents, the wind tunnel testing program and other
program, piled raft foundation design and significant unique aspects of the tower structural engineering design
foundation-tower interaction studies are presented; along and construction planning. Particularly critical technical
with long term settlement predictions, the completed issues such as the prediction of vertical shortening due to
pile load testing program, and ground seismicity studies. the long-term creep and shrinkage of the concrete frame,
and behavioral characteristics of the tower under lateral
and gravity loadings are also highlighted. The project is
scheduled for completion in late-2018.

Courtesy of Adrian Smith + Gordon Gill Architecture RWDI Wind Tunnel Laboratories, Ontario, Canada

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

8 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 1 Kilometer Tall Kingdom Tower: Concrete Reaches New Heights

Robert Sinn
Robert Sinn is a prolific and award-winning writer, having authored more than 60 papers and articles on his
work in structural engineering in the last three decades. He applies his 30+ years of experience to growing
the firm’s Structural Engineering practice and ensuring the quality and consistency of deliverables. He
has significant experience in medium- to large-scale projects, from long-span structures to high-rise
buildings, including Kingdom Tower, the next world’s tallest building and the first man-made structure to
reach a height of one kilometer.
He was named in Structural Engineering & Design magazine’s Power List in 2011.
Prior to joining Thornton Tomasetti, Bob worked at Skidmore, Owings & Merrill LLP for more than two
decades, designing and managing award-winning high-rise, supertall and other complex projects around
the world, including Exchange House at Broadgate in London, Hotel Arts at Vila Olimpica in Barcelona,
Guggenheim Museum Bilbao in Spain, and Trump International Hotel and Tower in Chicago, the tallest
all-concrete building in North America. He actively shares his knowledge through lectures and numerous
publications detailing innovations in building materials, structural engineering and system design.
His writing has been recognized by trade associations including the American Society of Civil Engineers,
the National Council of Structural Engineers Associations and The Institution of Structural Engineers in
the United Kingdom.

Kaushik Dutta
Kaushik Dutta has more than 14 years of experience in structural analysis and design for a variety of
building types, including high-rise commercial and residential, hospitality, education, research laboratory
and aviation, in the U.S., India and the Middle East.
After a decade in the U.S. – including three years in Thornton Tomasetti’s Newark, New Jersey, office
– Kaushik returned to India in 2010 to establish the firm’s office in Mumbai, which he co-manages. He
oversees the design and construction of complex, large-scale projects in the region. His responsibilities
include the preparation of structural plans, analysis and design; the preparation of drawings; and
coordination and communication with external consultants.

Organised by
India Chapter of American Concrete Institute 9
Plenary Session - Paper 3

The importance of Indian Leadership in Cement and Construction

Industry for Global Sustainable Development
Dr. S. K. Manjrekar
Chairman and Managing Director, M/s. Sunanda Speciality Coatings Pvt. Ltd.

Abstract growth of the nation as well as GHG emission along with

sustainability.
India is second largest manufacturer of cement in the
world. Due to large population and wide gap between India is the second largest manufacturer of cement in the
the projected demand and supply, the production will be world. All this cement is used for captive consumption,
beyond 400m tons by 2017 at CAGR of 10% plus. and due to large population and wide gap between the
projected demand and supply, the production of cement is
Government has allocated in current plan one trillion
slated to reach much beyond 400m tons by 2017 at CAGR
US $ to construction sector where major beneficiaries
of about 10% plus.
are sectors like housing, infrastructure, Industrial
and commercial structures. In the current alarming Government has allocated in its 12th five year plan
international fiscal scene, India is the next hub, and this one trillion US $ to construction sector where major
will trigger foreign investment and in turn the construction beneficiaries are housing and infrastructure sector
rate. Continuous Growth in cement and construction followed by Industrial and commercial construction.
are imperatives for the projected growth. However, for In the current alarming international scene of failing
reducing CO2 emission the increased use of fly ash and economies of so called strong as well as bankrupt
GGBS is necessary along with other measures to achieve nations, international business is looking at India as the
sustainable development. next hub to be, and this also will trigger flow of FDI and
in turn the construction rate. Hence, Continuous Growth
Adoption of Sustainability practices is a continuous
in the production of cement and further construction
process, and India is adopting total approach than a patch
related major activities are imperatives for the sustained
work “points systems” or a “grade based certification”
progress of the nation. However, it is also imperative to
system. Industry and the government is looking
not lose sight of endeavours to be made in the direction
systematically at 1) Planning, design and specifications
of sustainable development in terms of GHG emissions
based on performance and service life, 2) Construction
and global warming. In the recently concluded summit on
Practices, 3) Material Conservation and Selection, 4)
‘Global Warming’ in Paris on December 1st, Prime Minister
Demolition and recycling, and 5) Energy Conservation.
Modi assured that India will fulfil all its responsibilities
This presentation provides an inclusive approach which
with regards to climate change.
will allow India to grow and will emerge as an effective
leader for global sustainable development.
The Cement Industry in India can be
Vision For India Beyond 2020 1 Discussed as Under
Dr. Kalam was India’s Sustainability Man. He believed in ll Size of the Industry
India growing at 9 % to solve most of its problems. But he ll Growth Drivers / Road Ahead
didn’t mean growth was equivalent to more consumption.
He said: “For millions of years, humanity has been taking ll Cement Industry – Sustainability
more and more resources from Nature. Time has come to  Power Consumption in Manufacturing
take less and less from Nature to achieve sustainability.”
 Combating CO₂ Emissions
He proposed a ‘Societal Development Radar’ to monitor and
review sustainable growth. His user connectivity pyramid ll Use of Fly Ash/GGBS
was built on “natural resources, info- communication, ll Environment Impact Assessment
convergence of technology, societal business model,
applications and at the bottom end, the users.” ll Emission Standards in the Indian Cement Industry

In his firm belief in imperative 9% growth for India he ll Government Initiatives

was very aware about the rapid strides Indian cement / ll India as an Attractive Investment (FDI) Destination
Construction Industry is taking and the ‘trade off’ between
ll Conclusions

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

10 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The importance of Indian Leadership in Cement and Construction Industry for Global Sustainable Development

Size of Indian Cement Industry vii) In future, domestic cement companies could go for
global listings either through the Foreign Currency
The Indian cement industry is very large, second only
Convertible Bond (FCCB) route or the Global
to China in terms of installed capacity. India's current
Depositary Receipt (GDR) route.
installed capacity for cement production is approx 350
million tonnes per annum. This demand is expected to viii) Help from the government in terms of Friendlier
reach 550-600 million tonnes per annum (MTPA) by 2025. Laws, Lower Taxation and Increased infrastructure
spending
With above growth drivers, the sector will grow and take
India’s economy forward and consolidate its position as a
Leader in the region.

Government / Industry Initiatives

Government is increasingly giving attention to
I) SUSTAINABILITY
II) EMPLOYMENT
III) POTENTIAL / INVESTMENTS
Fig. 1: World cement production 2014 by region and main Global Green house gases emission would increase
countries, (Source: The European Cement Association) from 4.3 billion tones in 2014 to 6 - 7 billion tons of CO2
in 2030. Of this, more than half a billion ton will be from
Growth Drivers / Road Ahead India’s cement production alone. Hence, the responsibility
towards sustainability also has to be part of the Leadership
The housing sector is the biggest demand driver of cement,
role.
accounting for about 67 % of the total consumption in
India. The other major consumers of cement include The Indian cement industry, though large, is also one of the
Infrastructure at 13 %, Commercial construction at 11 % most energy efficient. This is evident from the reports from
and Industrial construction at 9 %.
i) World Business Council for Sustainable Development's
Some of the growth drivers are listed as under : (WBCSD)
i) The country's per capita consumption stands at ii) Cement Sustainability Initiative's (CSI)
around 190 kg as against global consumption average
iii) Getting the Numbers Right (GNR) data program.
of ~300kg/capita /yr.
ii) The world's largest democracy and third largest
economy in terms of purchasing power parity, India
Cement Industry - Sustainability
is the second most populous country after China, a Energy consumption in manufacturing
position also held by its incredible cement industry.
In latest data, India performed very favourably in terms of
iii) The eastern states of India are likely to be the newer specific energy consumption per ton of clinker produced,
and virgin markets for cement companies and could with an average 3130MJ/t across the 50% of cement
contribute to their bottom line in future. capacity. Brazil and China, which also have rapidly-
iv) By 2017-18, everyday construction of 25 kilometers developing large cement industries, performed slightly
of road is aimed at by Ministry of Surface Transport. less well. It is the recent expansion of the industry that
Today it is almost at 11 kilometers. Cement plants provides this thermal efficiency.
near the ports, for instance the plants in Gujarat and This is of course the consequence of modern plants being
Visakhapatnam, will have an added advantage for more efficient than older ones. It speaks well as it is in
exports. comparison with the EU27 group of countries (and the
v) In the next 10 years, India could become the major USA also to a greater extent). The use of expensive foreign
exporter of clinker and gray cement to the Middle coal as the dominant fuel for the cement industry acts as
East, Africa, and other developing nations in the a strong contributor towards efficiency.
ASEAN region.
vi) A large number of foreign players are also expected Combating CO2 Emissions
to enter the cement sector, owing to attractive profit When it comes to CO2 emissions per ton of clinker, India
margins and steady demand. performs less well, making 837kg/t of clinker. Though
this is close to the global average of ~300kg/capita/yr.

Organised by
India Chapter of American Concrete Institute 11
Plenary Session - Paper 3

However, global warming point of view cement industry Environment Impact Assessment
is behind those industries that have successfully Cement and Mining projects fall under Schedule of EIA
implemented alternative fuel. In conclusion very shortly Notification 2006 of MoEF, and are classified into category
the new slogan would be No Coal or less coal use of A and B. For category ‘A’ industry, EIA is mandatory and
alternative fuel produce less CO2. environmental clearance has to be obtained from MoEF.
For category ‘B’ industry, environmental clearance has to
Use of Fly Ash/Ggbs be obtained from State Pollution Control Boards.
Use of fly-ash as a partial replacement for Portland cement
is generally limited to CLASS F Fly-ash as this fly-ash Emission Standards In India
is pozzolanic in nature, and contains less than 20% lime Cement plants are equipped with air pollution control
CaO. The current utilization of fly-ash in cement industry equipments (APCEs) like Electrostatic Precipitators
is 48.13% in 2011-12 (Central Electricity Authority Annual (ESP), Baghouse with glass fibre membrane filters and
Report 2011-12)2. The fly ash utilization in the country has ESPs modified with bag filters, called hybrid filters. For
increased to 57.63% in the year 2013-14. (Govt. of India, new plants, including grinding units (plants commissioned
Ministry of Environment, Forests and Climate change).3 on or after 03-02-2006), the emission limit is 50 mg/Nm3.

Employment
Ever since it was deregulated in 1982, the Indian cement
industry has attracted huge investments, both from Indian
as well as foreign investors. India is the second largest
producer of cement in the world providing employment to
more than 32 million people, directly or indirectly.

Potential / Investments
India has a lot of potential for development in the
infrastructure and construction sector and the cement
sector is expected to largely benefit from it. Some of the
recent major government initiatives such as development
Fig. 2: Progressive Utilization of Fly Ash in Construction of
Roads/Embankments/Ashdyke Raising during the Period 1998- of 98 smart cities are expected to provide a major boost to
99 to 2012-13 (Source: Central Electricity Authority, New Delhi, the sector.
December, 2011) Aided by suitable government foreign policies, several
foreign players such as Lafarge-Holcim, Heidelberg
Soon 100% of cement production will be pozzolonic with Cement, and Vicat have invested in the country in the
Fly Ash, GGBS or any other supplementary cementitious recent past. A significant factor which aids the growth of
materials, with projected replacement volume of this sector is the ready availability of the raw materials for
almost 25%. making cement, such as limestone and coal.

Projected Use of Fly Ashs4

Expected Fly-ash absorption in cement (million tons per annum)
(Source: WBCSD/CSI/LOW Carbon technology road map for
Indian cement industry)
Serial Year Expected Fly-ash absorption in Indian
No. Cement Industry (million tons per annum)
1 2015 52.65
2 2020 63.01
3 2025 94.63
4 2030 120.50
5 2035 143.72
6 2040 158.02
7 2045 167.74
8 2050 177.45
Fig. 3: Expected Fly-Ash absorption in Cement (Million tons per
annum) (Source : WBCSD/CSI/LOW Carbon technology road Fig. 4: FDI Markets (Source: Estimates of FDI inflows for the
map for Indian cement Industry) first half of 2015 5)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

12 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The importance of Indian Leadership in Cement and Construction Industry for Global Sustainable Development

ll India is assuming leadership position in the global

scenario of cement and construction industry.
ll There are several growth drivers which includes the
initiatives of stable Government.
ll As a result of growth story, India has become an
attractive investment destination.
ll This leadership in manufacturing volumes brings
about the responsibility towards sustainability.
Note : 1) Reasonably Good Distribution ll Indian cement industry has to combat effectively CO2
2) More FDI Shall be attracted in Manufacturing sector
emissions and more effectively use fly ash, GGBS and
Fig. 5: India – Sector Wise FDI, (Source: Department of industrial other supplementary cementitious materials.
Policy and Promotion)
ll Indian Government is vigilant about environmental
Cement Industry – India is Clearly an impact assessment and emission standards
conserving Indian cement industry.
Attractive Investments Destination for FDI
According to data released by the Department of Industrial ll Industry is responsive to its obligations but continuous
Policy and Promotion (DIPP), cement and gypsum efforts are needed.
products attracted foreign direct investment (FDI) worth
US$ 3,099.80 million between April 2000 and June 2015. Acknowledgement
The paper was presented in ‘Global Sustainability Forum
Conclusions 8’ at ACI Convention, Denver held on 11th Nov.2015
ll The Indian cement industry is large, growing, however,
with consumption of just 185kg/capita/yr in 2013 References
(compared to global average of ~300kg/capita/yr) the 1. Sustainability Next, Vol 3, Issue 7, August 2015
country itself has the capacity to demand significantly 2. Central Electricity Authority Annual Report 2011-12
more cement as it develops. 3. Govt. of India, Ministry of Environment, Forests and Climate change

ll For example, using alternative fuels, it will be able to 4. World Business Council for Sustainable Development's (WBCSD)
and Cement Sustainability Initiative's (CSI)
take advantage of increasing demand while remaining
ahead of their competitors economically as well as 5. FDI Markets : Estimates are for the first half of 2015
Sustainability point of view.

Dr. S. K. Manjrekar
Dr. S. K. Manjrekar, FACI, is Chairman and Managing Director of M/s. Sunanda Speciality Coatings Pvt.
Ltd. He obtained his Ph.D. in 1977 from University of Bombay. He is a noted material scientist in the field of
corrosion, concrete, sustainability, waterproofing and construction chemicals. He is working on “Corrosion
Mission of India”. He has been conferred with the “CHAPTER ACTIVITIES AWARD” by American Concrete
Institute in 2003 in Vancouver Canada. He has been working on several technical and admin committees
of ACI viz ; 1) COMMITTEE 364 (for Repairs and Rehabilitation), 2) SUB COMMITTEE ON INTERNATIONAL
CERTIFICATION 3) INTERNATIONAL PARTNERS/PUBLICATIONS COMMITTEE 4) INTERNATIONAL
COMMITTEE (IC) 5) CHAPTER ACITIVITIES COMMITTEE. In March 2008 at San Francisco USA he was
invited to participate in ISO/TC 71 meeting which is a Joint initiative of ACI, ANSI & ISO. Recently he gave
invited speech in “Sustainability Forum 8 on India Focus” in Denver – Colorado (USA) on 11th Nov. 2015.
Earlier, he was invited as a keynote speaker for the international forum 5 on Sustainability with special focus
on BRIC Countries in Pittsburgh held on 23rd October 2010. He was invited to lecture to professors and
PhD students of ‘University of Leeds’ on 13th Nov. 2015.He teaches ACI Certification Concrete Field Testing
course in India and is the “Principal Convenor”. He has published more than 110 papers in various national
and international journals alongwith more than 200 lectures that have resulted in deeper understanding of
corrosion, concrete, waterproofing and repair science nationally and internationally.

Organised by
India Chapter of American Concrete Institute 13
Plenary Session - Paper 4

Insitu Strength of Concrete – The Engineer’s Holy Grail!

Michael G. Grantham and Richard Rogerson
Sandberg LLP, London, UK

Abstract in core samples also had an effect and the curing received
The estimation of the strength of concrete in an existing by the insitu concrete was different to the water curing
structure, or the assessment of compliance of concrete received by concrete cube samples.
supplied to a project in terms of compressive strength Unscrupulous concrete suppliers were quick to exploit
remain the Engineer’s Holy Grail. The procedures to these uncertainties to justify concrete which had been
undertake these assessments are complex and need to shown to have low cube strength, applying correction
be approached with caution. NDT methods are often used factors for curing, steel, air content etc, which, when
in conjunction with cores to minimise the damage to a applied one after the other, often meant that concrete
structure when conducting assessments and it is crucial of quite significantly low strength could be justified to
that these are carried out correctly and in the right places “comply” with the concrete specification. To be fair,
if meaningful results are to be obtained. BS6089 also sometimes the low cube strength resulted from poor
includes a method to confirm that two individual concretes preparation of the concrete cubes, or poor curing, or both.
can be considered to come from the same population, by
assessing strength on each by NDT and considering the With the introduction of European standards, the industry
standard deviation of the results. In this way, an area adopted a different approach, deeming the concrete
of suspect concrete can be compared with an area of from a quality assured supplier being compliant with the
concrete already accepted, without the need to resort to specification, but allowing for so called “identity tests” to
core samples at all. This paper reviews the methods and be performed on cubes taken from the concrete to confirm
techniques available to the Engineer and highlights the that the concrete supplied came from a conforming
problems and pitfalls in conducting such assessments. population.

Keywords: concrete, strength, NDT, SONREB, evaluation, Confusion still often existed with Engineers often failing to
assessment understand that a C32/40 concrete (with a characteristic
cylinder strength of 32 MPa and a characteristic cube
strength of 40 MPa) would usually have significantly higher
HISTORY strengths than those values when tested in the laboratory.
For many years, concrete strength in the UK was The reason for this is that compliance was based on a
established using either the guidance given in BS1881 normal, Gaussian, distribution of strength values when
Part 116 (BSI, 1983) for concrete cube samples, when samples were tested, with a spread of values which
testing fresh concrete or on concrete core samples depended on the quality control of the concrete produced
using the procedures in BS1881 Part 120:1983. It was by the supplier. With good quality control, a standard
recognized quite quickly that confusion existed in the deviation of some 4 MPa was to be expected and so, for
minds of Engineers, when attempting to assess concrete 95% confidence in achieving the required 40 MPa, the
strength because even if core test results were calculated supplier would typically aim for a target mean strength of
as “estimated insitu cube strength” the results did not some 48 MPa. In practice this meant that concrete cubes,
usually match those that would be expected from cubes. compliant with the specification would typically lie within
Typically, core tests would give some 75% of the values the range 40 to maybe as high as 52 MPa, and occasionally
that would have been obtained from cubes made from the could be lower than 40 MPa, perhaps as low as 37 MPa,
same concrete and tested at 28 days. This resulted in the and still be deemed compliant. This concept was plainly
issue of a report by the Concrete Society, Concrete Society not fully understood by Engineers in many cases. When
Technical Report No. 11 (CSTR11), originally in 1976, and training Engineers, I often ask them to anonymously write
amended in 1987 (Concrete Society, 1976 Amd 1987). the strength they would expect from a concrete cube
Interpretation was further complicated because insitu prepared and cured properly from a C32/40 concrete.
concrete is not usually perfectly compacted so air voids The variation in answers often beggared belief, varying
can reduce the strength. The presence of reinforcement from 32 MPa (frequently) to 40 MPa (also frequently)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

14 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Insitu Strength of Concrete – The Engineer’s Holy Grail

but occasionally quite unjustifiable numbers such as 20 Granulated Blastfurnace Slag and PFA – Pulverised Fuel
Mpa or 60 Mpa, for which the reasoning was completely Ash typically) . At the same time, the addendum sought to
unclear. Since the question was answered anonymously, I explore the effect of core sample location, type of member
was never able to identify the logic behind such answers. and age of concrete. The results were very interesting and
What was plain, however, was that Engineers often did not showed that the relationships assumed in CSTR 11 were
understand this topic fully. by no means simple and that the effect of cement type,
sample position and age could cause dramatic variation in
In 2004, the Concrete Society produced Project Report 4
the strength to be expected from concrete core samples.
(Concrete Society, 2004), an addendum report to Technical
Report No 11, to provide further information on the results From the data obtained from the Concrete Society
to be expected from blended cements (with GGBS – Ground Addendum report, it became obvious that the previously

Table 1
Typical behaviour of concrete core strength in relation to standard cube strength from Concrete Society Project Report 3: 2004 (0.80
indicates for example 80% of the strength of the equivalent cube was obtained from a core tested from a cast member)

 28 Days 42 Days 84 Days 365 Days


Mixes Element
Cement
Min Max Min Max Min Max Min Max
Type

Block 0.75 0.95 0.80 0.95 0.85 1.05 0.95 1.10

PC Lower Strength Mixes Slab 1.00 1.25 1.05 1.35 1.05 1.50 1.15 1.60

Wall 0.85 0.95 0.85 1.00 0.85 1.05 0.90 1.10

Block 0.60 0.70 0.70 0.75 0.70 0.80 0.75 0.90

Higher Strength Mixes Slab 0.85 1.00 0.95 1.10 0.95 1.15 1.05 1.25

Wall 0.85 0.95 0.80 0.95 0.85 1.00 0.95 1.05

Block 0.85 1.10 0.85 1.20 0.95 1.35 1.45 1.65

P/FA-B Lower Strength Mixes Slab 0.80 1.20 0.90 1.30 1.00 1.50 1.65 2.15

Wall 0.80 1.05 0.85 1.10 1.05 1.30 1.50 1.60

Block 0.75 1.00 0.80 1.00 0.95 1.05 1.00 1.35

Higher Strength Mixes Slab 0.75 0.95 1.00 1.15 1.15 1.25 1.20 1.70

Wall 0.80 1.00 0.90 1.05 1.05 1.25 1.35 1.40

Block 0.70 1.10 0.85 1.10 1.00 1.20 1.15 1.35

P/B Lower Strength Mixes Slab 0.65 1.15 0.85 1.30 1.20 1.40 1.35 1.70

Wall 0.70 1.00 0.85 1.15 1.05 1.25 1.15 1.35

Block 0.80 1.05 0.95 1.20 0.90 1.25 1.00 1.55

Higher Strength Mixes Slab 0.90 1.05 1.05 1.30 1.20 1.30 1.30 1.80

Wall 0.70 1.00 0.85 1.10 1.00 1.20 1.05 1.25

Block 0.80 0.95 0.85 1.00 0.95 1.10 1.00 1.15

PLC Lower Strength Mixes Slab 0.95 1.05 1.05 1.20 1.15 1.35 1.20 1.45

Wall 0.85 1.00 0.90 1.05 0.95 1.00 1.10 1.20

Block 0.60 0.80 0.70 0.90 0.70 1.00 0.80 0.95

Higher Strength Mixes Slab 0.80 0.90 0.95 1.00 1.00 1.05 1.05 1.20

Wall 0.80 0.95 0.90 1.00 0.90 1.05 1.00 1.15

PC = Portland Cement (CEM1), P/FA-B = ...etc.

Organised by
India Chapter of American Concrete Institute 15
Plenary Session - Paper 4

assumed relationship between core and cube strength The Current Situation
was no longer reliable with blended cements or when
For some while, the industry has been following the advice
subjected to a more rigorous analysis involving different
in BS EN 13791 and the complementary guidance given in
member types and ages. Concrete Society Technical
the UK code BS6089 (BSI, 2010). BS EN13791 suggests the
Report 11 was therefore withdrawn and is no longer used.
following situations where knowledge of concrete insitu
BS1881 Part 120 was subsequently withdrawn and strength may be required:-
replaced by BS EN12504-1 (BSI, 2009). Assessment of
ll When an existing structure is to be modified or
insitu strength in structures is described in BS EN13791
redesigned;
(BSI, 2007). This has caused some problems for those
attempting to assess strength as it recommends testing ll To assess structural adequacy when doubt arises
cores after 3 days stored in a laboratory atmosphere, about the compressive strength in the structure due to
whereas prior to this (in the UK at least) cores were defective workmanship, deterioration of concrete due
always tested in a saturated condition. There is typically a to fire or other causes;
difference in strength of around 10% with the laboratory air ll When an assessment of the in-situ concrete strength
dried cores giving higher strength. This again introduced is needed during construction;
confusion, especially where attempting to interpret
whether a questionable concrete actually complied with ll To assess structural adequacy in the case of non-
a specification, because some laboratories would test wet conformity of the compressive strength obtained from
(permitted in BS EN12504) and some dry. Then there is standard test specimens;
the question of what is “dry” – is a core that has stood in ll Assessment of conformity of the in-situ concrete
a laboratory atmosphere for 3 days “dry?” If the core is compressive strength when specified in a specification
made from a very dense concrete, the rate of drying may or product standard.
be quite slow. That leaves a sample with an unknown
moisture condition with a strength maybe somewhere BS 6089 adds further to this and suggests the following
between what might be expected from wet and dry additional reasons why insitu strength may be important.
cores. The reasoning behind this is complex. If a core is ll Deterioration of concrete due to: overloading;
immersed in water, the outer part swells and is restrained
by the dry interior. The result is that the outer part goes »» fatigue;
into compression and the strength is reduced. Similarly, »» chemical action; fire;
if a core is damp inside but dry outside, the outside goes
into tension, the inside into compression and again a lower »» explosion;
strength can occur. »» weathering;
The author does not claim to have a suitable answer for ll To ascertain whether the in-situ strength of concrete is
these quandaries, other than to say the previous practice acceptable for: the designed loading system;
of water saturation at least gave an even moisture
condition, even if it didn’t mimic the condition likely to be »» the actual loading system;
encountered in the structure, unless it was submerged »» a projected loading system for a new use;
concrete.
ll Doubt concerning the strength of concrete in the
The combination of all the uncertainties from type of structure due to:
cement, age, type of member, location within a member,
»» nonconformity of concrete;
moisture condition and so forth, means that any attempt to
try and define the insitu strength of the concrete is fraught »» differences between identity testing and conformity;
with difficulty. If one is attempting to show whether a
»» workmanship involved in placing, compacting or
concrete complies with some specified value, then that is
curing of concrete.
even more difficult!
In deciding whether to pursue an investigation to establish
With the closer co-operation occurring within Europe
the insitu strength of concrete it is important to know
and with Engineering Practices, building projects
what the goal is. The strength can be considered for
and construction often being pan-European, further
three different scales: the component scale (an individual
complications are now arising because in Europe,
member, say), a part of a structure (which can involve
concrete compliance tests have traditionally been carried
several components) and finally the whole structure.
out on cylinder shaped samples rather than cubes, with
Inevitably to be able to achieve a high confidence in testing,
a 2:1 height to diameter ratio, rather than the 1:1 test
a large investigation may be required and inevitably that
performed on a cube. This again introduces yet another
can be quite costly. Clients rarely want to pay for such a
variant into the mix.
detailed investigation but nevertheless expect unequivocal
answers with minimum outlay.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

16 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Insitu Strength of Concrete – The Engineer’s Holy Grail

The RILEM Committee on evaluation of insitu strength Where Xr = the mean value for the reference concrete
(TC245 ISC) discusses in their draft guidance on this topic, X s = the mean value for the sample concrete (the suspect
the influence of cost. In Eurocode 8, the estimation of concrete)
concrete strength contributes to the determination of the
Sr = The standard deviation for the reference tests
knowledge level (KL) of the structure under study. Three
KLs are defined (limited, normal and full knowledge) in Ss = the standard deviation for the sample concrete tests
order to choose the appropriate confidence factor (CF) (suspect concrete)
value to be adopted in the evaluation process. CF deals If the numerical value obtained lies within the range
with the incompleteness of knowledge that is always of ± 2.024 then the element under investigation can
present in the evaluation of existing structures. To this be considered as coming from concrete of the same
end, CF applies to the mean strength obtained from in-situ
population as the reference concrete and should be
tests and from the additional sources of information (e.g.
accepted. This technique is based upon the concept of the
original design specifications) to determine the design
95% confidence interval for the difference of the means
strength to be used in the calculation of the structural
and aims to show whether, when considering the spread
capacity. The recommended values of CF are 1.35, 1.2 and
of results obtained, there is a significant overlap at the
1 for limited, normal and full KL, respectively. Although a
certain KL regarding material properties can be achieved 95% confidence level.
on the basis of original design specifications or test Where methods are employed that rely on calibration of
reports, complementary in-situ tests must be provided. an NDT method against core testing, the advice currently
Tests are mandatory because concrete strength is difficult in BS EN 13791 uses two options. The first is to prepare a
to estimate and because it is a function of component correlation between the chosen NDT method using core
properties and mixture (cement, aggregates, water), tests and NDT tests at the same location, with a minimum
casting operations, various maturity phases, interactions of 18 core and NDT tests being used. The second method
between them and, finally, possible deterioration due to uses pre-defined basic curves based upon previous
environmental impact. research of the shape of the relationship between the two
In practice, with so many things to consider and now cost data sets, but with the values set artificially low. Using
added into the mix, any evaluation of concrete strength in 9 sets of tests ( 9 core tests and 9 NDT tests) a curve is
an existing structure, whether it be to assess compliance constructed and a value of the shift required to adjust the
or to assess physical load capacity in some way, for an pre-defined curve is calculated. The adjusted curve may
element, a group of elements or even a whole structure, then be used to assess further NDT results.
cannot be sensibly based only on core samples and the
From such testing the insitu characteristic strength
favoured approach and the most sensible one in terms
can be determined from the lower of the two following
of cost and time is to use a combination of core tests
values.
and non-destructive testing, with the former used as a
means of calibrating the latter in some way so that the Fck.is = Fm(n)is - 1.48 # s .................................................(2)
extent of the testing can be significantly extended without
Or
incurring huge time penalties (coring and core testing is a
slow process, whereas NDT techniques are usually very Fck.is = Fis, lowest + 4 ..........................................................(3)
much faster and have the added benefit of not damaging
Where
the structure). It may even be possible to assess whether
some suspect concrete in a structure, which has failed Fck.is is the insitu characteristic strength
identity testing, is actually compliant using a simple
Fm(n)is is the mean insitu compressive strength for n
comparison with some similar concrete that has shown
samples
compliance when tested by using, for example Rebound
Hammer methods or Ultrasound measurements and s is the standard deviation of the test results
comparing the data statistically to show whether the
The problem with both methods, especially the first, is
two concretes can both be considered to come from a
conforming population. the number of cores required to calibrate the technique.
Taking 18 core samples is very time consuming and
BS 6089 describes such a procedure, where 20 tests costly, whereas 9 cores may be more achievable, but is
are performed on each concrete (that which is suspect still a burden. A draft revision of BS EN 13791 is currently
compared to that which has passed compliance testing). being considered, which recognises this and has reduced
The mean value of the test results for each concrete is the minimum number of cores to 12 for the first method
determined, together with the sample standard deviation (although it says that it may be possible to establish a
in each case. A simple calculation is performed: correlation with fewer cores). Where less than 12 cores
t calc =
Xr - Xs ...........................................................(1) are being used it is recommended that NDT is used to
S 2r + S s2 establish a correlation.
20

Organised by
India Chapter of American Concrete Institute 17
Plenary Session - Paper 4

Using Combined Methods A series of correction coefficients can be applied in

order to improve the accuracy of prediction obtained
There can be significant problems when attempting to use
from the “nomogram”. These coefficients account for
a single NDT method to establish a correlation against
the type of cement, the cement content, the aggregate
cores. For example, using the Rebound Hammer method,
type, the aggregate fine fraction (less than 0.1 mm) and
both moisture content and carbonation of the concrete
the maximum size of aggregates. The accuracy of the
can significantly alter the concrete surface hardness.
estimated strength (the range comprising 90% of all the
For this reason, the use of combined methods can show
results) is considered to be 10 to 14% when the correlation
superior results, especially if the interfering parameter
relationship is developed with known strength values of
has a different effect on the chosen methods. For example,
cast specimens or cores and when the composition is
increased moisture content lowers Rebound Value but
known, and 15 to 20% when only the composition is known.
increases Ultrasonic Pulse Velocity, so a combination
of these two techniques can be very useful. In the same The normal SONREB equation is of the form Fck = aVbRc
way, surface carbonation would be unlikely to significantly Where Fck = insitu strength of concrete
affect pulse velocity when transmitted through a member
V = UPV in m/s
(though it might affect surface pulse velocity measured
using the indirect method). R= Rebound Value (or Q Value if using a SilverSchmidt)
And a, b and c are constants
This technique was first explored by RILEM Technical
Committees 7 NDT and 43 CND. The technique was first The solution to determining the SONREB coefficients a, b
developed in the early 1980’s and used a combination and c can be found using the SOLVER function in Microsoft
of Rebound Hammer and Ultrasonic Pulse Velocity Excel, or a dedicated Macro enabled spreadsheet available
measurements. These days, with the improved operation from Proceq.
and reliability of, for example, the Proceq “SilverSchmidt” There is good evidence that an improved reliability can be
rebound hammer and the PUNDIT Lab, developed since obtained from combined methods as compared to using a
Proceq took over the PUNDIT apparatus from CNS Farnell, single NDT method. (Grantham, 2012)
the technique is becoming more popular and being used
more widely.
Estimating “Potential” Strength
Figure 1 shows this relationship in the form of a
So far in this paper we have been looking at the strength
“nomogram”. By knowing the rebound hammer and pulse
of the concrete in terms of the values to be assumed for
velocity, the compressive strength is estimated. The form
assessing structural adequacy and, for example, designing
of the curve will however vary with the type of concrete
remedial works. However as mentioned earlier, using the
and an individual calibration must be carried out for a
insitu strength to assess the compliance or otherwise of
particular type of concrete on site. In practice, however,
the concrete originally supplied is notoriously difficult to
producing such a curve is a complex task.
achieve with confidence from the testing of cores at any
time, and especially when it is older than 28 days.
The compressive strength values obtained from testing
drilled cores cannot be compared directly with the
‘characteristic’ strength, as envisaged for the mix design
or stipulated in a specification. When a concrete cube
or core specimen is crushed, the failure load divided by
the cross-sectional area of the core gives a measured
compressive strength.
The measured core compressive strength can then be
converted into a ‘Corrected core strength’ expressed as
the strength of an equivalent cube. This is intended to
provide a measure of the actual strength of the concrete as
it presently exists in the structure with specific allowance
in calculation for the geometrical differences between
cores and standard cubes. Correction is made for shape,
the ratio of length to diameter of the specimen under test
and the influence of any embedded reinforcing steel. BS
1881: Part 120:1983 also corrected for direction of drilling
(i.e. horizontal, or perpendicular to casting, and vertical,
or parallel to casting). However the latest guidance in
BS EN 13791:2007 notes that the core test result should
Fig. 1: Nomogram for SONREB testing (Breysse, 2012)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

18 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Insitu Strength of Concrete – The Engineer’s Holy Grail

not be modified for drilling direction except in special strength of a single member? Is it the strength of some
circumstances. The ‘Corrected core strength’ is the value specific part of a member or structure which has been
that should be used for any assessment of structural represented by some failed identity tests? Or what?
adequacy. It is also referred to in BS EN 13791 as the insitu
ll The influence of such things as moisture and
compressive strength.
carbonation on the measurement of strength must
In terms of compliance assessment, BS EN 13791 notes be fully considered. The former affects both cores
that ‘The compressive strength of cores and the in-situ and NDT methods, the latter is most likely to affect
strength will generally be less than that measured on Rebound Hammer measurements. Ignoring these will
standard test specimens taken from the same batch of introduce serious errors in strength estimation.
concrete’. BS EN 13971 introduces a term ‘characteristic
ll Attempting to determine strength with cores
in-situ compressive strength’ which is defined as the value
alone is fraught with difficulty, due to the inherent
of in-situ compressive strength below which 5% of the
inhomogeneity of concrete and the differences in
population of all possible strength determinations of the
curing that are inevitably experienced between cubes
volume of concrete under consideration are expected to
or cylinders cast from the fresh concrete and core
fall. A ratio of characteristic in-situ compressive strength
samples removed from the structure.
(fck,is,cube) to characteristic strength of standard cube
specimens (fck) of 0.85 is given. ll Using “targeted” core samples, by first using e.g.
Rebound Hammer to establish the location of
However if the structure is shown to be inadequate
regions of low strength, high strength and average
with respect to strength, this does not necessarily
strength means that cores can be taken from more
prove that the supplied concrete failed to satisfy its
representative zones and a better correlation achieved.
specification. This should be a matter for the contractor
and concrete producer to resolve without the involvement ll Combining NDT methods can significantly improve the
of the Engineer although this is not always the case. The correlations attained and give more reliable strength
Concrete Society report CSTR11 introduced the concept estimates.
of ‘potential strength’, in an endeavour to systematise this
ll Using comparative procedures and NDT, rapid
assessment, but as mentioned earlier, more recent review
resolution of disputes involving concrete quality can be
has caused this approach to be discontinued and replaced
achieved without the need to resort to core sampling
by the guidance given in BS EN 13791 and BS 6089. In
and the inevitable additional questions that introduces.
practice, there are numerous materials, construction
Estimation of so called “Potential Strength” is fraught
and environmental factors that influence and complicate
with difficulty and best avoided if at all possible.
the relationship between current ‘actual’ strength and
hypothetical ‘potential’ cube test strength at 28 days, and
References
many or most of these factors cannot be reliably known or
1. BSI, BS 1881-120:1983 Testing concrete. Method for determination
assessed sometime after the event. of the compressive strength of concrete cores (Superceded,
withdrawn)
To show that the concrete failed to meet its specification,
the corrected core strengths need to be adjusted for 2. BSI, BS 1881-116:1983 Testing concrete. Method for determination of
compressive strength of concrete cubes (Superceded. Withdrawn).
excess voidage, curing and maturity. Some guidance
is given in Annex A of BS 6089 but the standard warns 3. BSI, BS EN 12504-1:2009, Testing concrete in structures. Cored
specimens. Taking, examining and testing in compression
that ‘The uncertainties associated with any estimated
potential strength are so large that this procedure is best 4. BSI, BS EN 13791:2007, Assessment of in-situ compressive strength
in structures and pre-cast concrete components
avoided’. Unfortunately, in some instances it is the only
5. BSI, BS 6089:2010. Assessment of in-situ compressive strength
way to resolve disputes over costs between producer and
in structures and precast concrete components. Complementary
contractor. guidance to that given in BS EN 13791
6. Concrete Society, Concrete Society Technical Report 11 (TR 11)
Conclusions Concrete core testing for strength 1976, amended 1987
7. Concrete Society, Concrete Society Project Report 3: 2004 In situ
ll Determining the strength of concrete in a structure
concrete strength. An investigation into the relationship between
remains the “Holy Grail” for many Engineers. Before core strength and standard cube strength
pursuing that question, the Engineer needs to consider 8. Grantham, M. Using the SONREB Method for Evaluation of Concrete
what it is he or she is seeking. Is it the worst credible Strength Using Rebound Hammer & Ultrasonic testing, Proceedings
strength for a structure? Is it the characteristic of Structural Faults & Repair, Edinburgh University, M. Forde Ed.
2012

Organised by
India Chapter of American Concrete Institute 19
Plenary Session - Paper 4

Michael Grantham
Michael Grantham is a specialist in concrete repair and non-destructive testing. He is co-author of the
book “Testing of Concrete in Structures,” with Bungey and Millard and also author of “Concrete Repair,
a Practical Guide,” together with numerous other publications. He is Chair and organiser of the well-
known “Concrete Solutions” series of International Conferences on Concrete Repair. He is Editor in Chief of
Elsevier’s “Case Studies in Construction Materials” and the current President of the Institute of Concrete
Technology. He is a retained Consultant to Sandberg LLP, Materials Engineers, in London.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

20 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Replacement of Steel with GFRP for Sustainable Reinforced Concrete

Replacement of Steel with GFRP for Sustainable Reinforced Concrete

S. A. Sheikh and Z. Kharal
Department of Civil Engineering, University of Toronto

Abstract advantage over stainless steel. While research into

internal GFRP bars has been ongoing over the last few
The corrosion of steel in reinforced concrete structures
decades, significant progress was made in the past two
has cost a significant amount of resources globally over
decades specifically with the work of pioneering research
the past few decades. Glass fibre reinforced polymer
carried out by groups such as ISIS Canada and resulted
(GFRP) bars present a feasible and cost effective solution
in several design recommendations and guidelines for
to the steel corrosion problem. Despite many advantages
of GFRP, most designers are still reluctant to use GFRP designing GFRP reinforced concrete members, including
as the main reinforcement in concrete members due to the ACI 440- 1R-06, CSA-S6-06 and CSA S806-12. GFRP
its different behavior than steel. The aim of this paper is bars have been used in bridge decks and barrier walls.
to let engineers gain a better understanding of the overall Hall’s Harbour Wharf Bridge in Nova Scotia, Joffre Bridge
behaviour of GFRP as internal reinforcement so that in Quebec and Crowchild Trail Bridge in Alberta are some
they have better confidence in using it as a sustainable examples of Canadian bridges in which GFRP was used
material. This paper provides a few significant outcomes as the main reinforcement (Mufti et al. 2007). Despite
from an extensive experimental program underway these advances in the field of the GFRP-RC over the last
at the University of Toronto. The part of the program few years, most designers are still reluctant to replace
discussed here involved testing of 24 GFRP reinforced steel with GFRP as the main reinforcement in reinforced
beams, 60 GFRP direct tension specimens and 17 GFRP concrete members. This is primarily due to a lack of data
confined columns in which the behavior of GFRP-RC in and analytical procedures in this field compared with that
flexure, shear, tension and compression was investigated. available for steel- reinforced structures.
A new tension-stiffening model is also presented here Although GFRP is a viable reinforcement alternative,
that significantly improved the prediction of deflection it presents its own challenges in terms of having very
and stiffness in GFRP-RC beams. Another significant different properties than steel. Other than its lack of
conclusion drawn from this research is that GFRP post-elastic behaviour, the major shortcoming of GFRP is
spirals not only can be used efficiently as primary lateral its relatively low stiffness in comparison with steel. The
reinforcement in columns but also confine the concrete reduced stiffness results in deflections and crack-widths
core of a column more effectively than steel spirals. in GFRP-RC flexural members that are much larger than
Keywords: GFRP bars; reinforced concrete; corrosion; conventional steel-RC members. This results in GFRP
deflection; tension stiffening; compression, energy reinforced concrete design being largely controlled by
dissipation, earthquakes. service conditions that makes it essential to predict the
deflection in GFRP reinforced members more accurately
than for steel-RC members. The lower modulus of
Introduction elasticity of GFRP can also adversely affect the shear
The engineers today are faced with a huge problem of response by widening cracks resulting potentially in a
deteriorating infrastructure world-wide. The main cause greater rotation of the diagonal principal compression
of deterioration among reinforced concrete structures is strut thus reducing the efficiency of shear reinforcement.
the corrosion of internal reinforcing steel. The total annual While the hesitation to use GFRP-RC in reinforced
cost of corrosion worldwide in 2010 was estimated at USD concrete beams and slabs that depict flexural behavior
$2.2 trillion which amounts to about 3% of the world’s has been addressed to some extent, the compressive
GDP (Hays, 2010). Glass fibre reinforced polymer, GFRP, response of GFRP bars is mostly unexplored. Therefore
bars have been introduced as a light-weight, corrosion many design codes in North America such as the ACI-440
resistant material as a viable replacement for traditional prevent designers from using GFRP bars in members
steel reinforcing bars. under compression while CSA-S6-06 does not have
GFRP also remains one of the more cost effective FRP any provisions regarding this application. This is mostly
products commercially available with significant cost due to the uncertainty in the response of GFRP bars in

Organised by
India Chapter of American Concrete Institute 21
Plenary Session - Paper 5

compression and a lack of sufficient data on the behavior The deformability of the GFRP-RC beams was also
of GFRP-RC columns. Use of GFRP spirals is also a investigated in a test program consisting of 15 GFRP-
solution to prevent the deterioration of the cover. The reinforced beams with section size one-half of that shown
expansion caused by corrosion of steel spirals results in in Figure 1 (Getzlaf 2012). Deformability and performance
the spalling of concrete cover which results in the drop of in flexure has been a critical and often contentious topic
load carrying capacity due to a smaller cross-section and regarding GFRP design. Ductility, as defined traditionally
severely damages the structural integrity of the column does not explain the behaviour of GFRP-reinforced test
due to loss of confinement. The repair and rehabilitation specimens appropriately. However from the load versus
of such structures is very costly. deflection responses of some beam tests (Figure 2), it
can be seen that GFRP-RC has the capacity to deform
The paper is part of the research effort that aims at
substantially which provides the ability to absorb energy.
creating a better understanding of the overall behaviour of
Hence, the lack of ductility in GFRP bars does not
GFRP as internal reinforcement. The research presented
necessarily directly translate to brittleness of GFRP-
here is part of a large research program carried out at the
reinforced members. All the beams in Figure 2 contained
University of Toronto investigating the behaviour of GFRP
concrete which had a compressive strength of 32 MPa
in a variety concrete structures. In one of the projects,
and bar Type C which had ribbed surface profile. NT, 22B,
Vint and Sheikh (2012) investigated a variety of bent,
40B and 50B in the names of the specimens indicate,
straight and anchored GFRP products for bond behaviour.
respectively, no transverse reinforcement, 0.22%, 0.40%
Getzlaf (2012) investigated the flexural behaviour of
and 0.50% transverse reinforcement provided by bent
GFRP reinforced concrete including the deformability of
bar stirrups. It was shown in the interpretation of test
GFRP-RC beams presenting new methods for evaluating
results (Johnson, 2014) that the choice of transverse
member deformability. Johnson and Sheikh (2013) studied
reinforcement has a significant impact on not only the
the behaviour of large concrete beams reinforced with
flexural failure load but also the ultimate deflection of
GFRP bars and stirrups under flexure and shear. Kharal
the beam. In fact, GFRP-RC beams with relatively lower
(2014) investigated the deformation and serviceability
shear reinforcement displayed visibly wider cracking
characteristics of GFRP-RC beams and investigated
prior to failure giving indication of failure. The research
the role of tension stiffening. This paper provides an
showed that GFRP-RC structures have the potential to be
overall view of the developments from this research. It
designed for a non-brittle failure.
also presents results on the behavior of columns for a
better understanding of the application of GFRP bars in The experimental beam results were simulated using
compression and GFRP spirals for confinement. nonlinear finite element analysis. The beam section was
modelled in VecTor2 which is a 2-Dimensional nonlinear
finite element analysis program (Vecchio 1990). The
Behaviour of Gfrp-Rc Beams
GFRP bond model developed by Vint and Sheikh (2012)
A systematic investigation was carried out to determine was adopted in VecTor2 to improve the accuracy of the
the influence of concrete strength, GFRP transverse predicted results. Figure 3 shows the comparison of
reinforcement and longitudinal reinforcement stiffness the analytical and experimental load versus deflection
on the behaviour of beams in flexure and shear. To this results of a normal and high strength concrete beam.
effect, a set of twenty-four large scale GFRP beams were When tension stiffening was not considered in the analysis
constructed and tested (Johnson and Sheikh, 2013 and (VT2 No Tension Stiffening), it resulted in very soft and
Johnson, 2014). All the beam specimens had the same unreasonable beam responses. Similar results were
overall dimensions of 400 mm x 650 mm x 3650 mm observed for other beams. Hence the study concluded that
and a longitudinal reinforcement ratio of 1%. The loading tension stiffening consideration is much more important
arrangement was a 3-point set-up with a constant shear in GFRP-RC members for deflection calculation than in
span of 1,680 mm. The bearing plates used were 150 mm steel-RC members due to the low stiffness of GFRP.
wide resulting in a clear shear span of 1530 mm. Testing
was conducted in a monotonic displacement controlled The beams were then analyzed using the VecTor2 default
manner. The typical specimen geometry and test set-up Benz 2003 tension stiffening model and Collins and
can be seen in Figure 1. Mitchell 1987 model (Figure 3). However, the difference

Fig. 1: Test set-up for specimen JSC32-NT (Johnson and Sheikh 2013)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

22 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Replacement of Steel with GFRP for Sustainable Reinforced Concrete

But even this model resulted in only minor improvements

to the peak capacity and did not change the post-cracking
stiffness. This is because this model does not take into
account the effects on reduced GFRP modulus that
affects the bond and crack widths which are critical to
tension stiffening. Irrespective of the tension stiffening
model used, the ultimate mid-span deflection for all the
GFRP-reinforced beams was underestimated. This under
prediction of deflection is undesirable, particularly at
service load levels.
In summary, the predictions from various analyses
were found to be inaccurate because either they did not
take tension stiffening into account or the deflection was
Fig. 2: Load versus deflection for JSC32 beams2013) derived based upon steel tension stiffening formulations
that showed limited accuracy for GFRP-reinforced
concrete. The accuracy of deflection prediction depends
on the accurate determination of effective moment of
inertia, which is in turn dependent on tension stiffening. It
was thus concluded that direct tension tests were needed
to evaluate the tension stiffening behaviour of GFRP-RC
in tension and develop a model for tension stiffening in
GFRP-reinforced concrete.

Tension Stiffening in GFRP-RC

As a necessary extension to the previous testing program,
an experimental program was carried out in which GFRP-
RC specimens were tested under uniaxial tension. Although
to a limited extent, tests like these had been performed
previously, none had been as comprehensive in terms
of the number of the specimens tested, the consistency
of materials used or the parameters considered. The
test program consisted of a total of 52 GFRP reinforced
concrete specimens and 8 steel reinforced concrete
control specimens. All the specimens were square in
cross section and 1000 mm in length, cast around a single
reinforcing bar. The square concrete cross sections varied
from 100 x 100 mm to 200 x 200 mm while the reinforcing
bar sizes included 10M and 15M for steel reinforcement
and 13 mm, 16 mm and 19 mm for GFRP reinforcement.
The parameters studied included reinforcement ratio,
concrete strength, bar diameter and bar type. GFRP
bars from three different manufacturers were used. The
difference mainly lied in the bars’ surface profiles. One bar
was helically wrapped with fibers, the second was heavily
sand-coated and the last one had ribbed bar surface. The
GFRP-RC specimen detail and instrumentation can be
seen in Figure 4.

Fig. 3: Predictions by VecTor2, VT2, for: (a) normal strength

concrete beam and (b) high strength concrete beam

between the measured and predicted deflections was

still found to be considerable. Among all the steel-RC
tension stiffening models investigated, the Collins and
Mitchell 1987 tension stiffening model was found to give
comparatively better results for the GFRP-RC specimens. Fig. 4: Direct tension specimen with instrumentation

Organised by
India Chapter of American Concrete Institute 23
Plenary Session - Paper 5

All the GFRP reinforced specimens followed a typical is the concrete tensile cracking strain, Eb is the modulus
behaviour. Initially the elastic stiffness of the specimens of elasticity of the GFRP bar (MPa), and constant β1 is fixed
was high till the development of the first crack. At this at 1400. The factor γ is 0.5 for steel reinforcement, 1.5 for
point the stiffness of the specimen dropped and cracks helically wrapped bars (Bar Type A), 1.0 for heavily sand-
started developing further. After a while, the cracking coated bars (Bar Type V) and 0.8 for bars with ribs (Bar
stabilized and the stiffness increased since the GFRP bar Type C).
started taking most of the load. Hence, three phases were
observed in a typical GFRP-RC specimen (Figure 5); the The proposed model’s validity was checked against
pre-cracking elastic phase, the cracking phase, and the the results obtained in the experimental program. The
post-cracking phase. results showed that the proposed equations simulated
tension stiffening quite well for all bar types, concrete
strengths and reinforcement ratios (Figure 6). The
specimens shown in Figure 6 can be explained as
follows: C30-16A-100(1) means concrete strength of
30 MPa, 16 mm bar size of Type A GFRP in a 100 mm
square prism 1 in a duplicate set of two specimens. The
predicted response for the GFRP- reinforced specimens
was found to be more accurate than that from any of the
currently existing models.

Fig. 5: Typical behaviour of GFRP-RC direct tension specimen

As discussed above, the predictions from the currently

available models (Figure 3) had clearly shown a need for
the development of a more accurate tension stiffening
model. Before an attempt could be made to develop such
a model, it was necessary to first identify the parameters
that affect tension stiffening. From the test results, it was
determined that GFRP members exhibited greater tension
stiffening than steel-RC. It was also determined that
reinforcement ratio, concrete strength and bar diameter
had no significant effect on tension stiffening if shrinkage
was appropriately taken into account. However, various
GFRP surface profiles were found to have some influence
on tension stiffening (Kharal, 2014). The proposed model
also took into account the main parameters that affected
the tension stiffening behaviour of GFRP-RC specimens
including the effect of surface profiles. The following
proposed equation is presented here to determine the
average concrete tensile strength for GFRP reinforced
specimens after cracking.
ftl
1 + #S 20000 $ b 1 $ Q f cf - f cr V&
Eb X
fc = c
..................................(1)

Where fc is the average concrete tensile strength after Fig. 6: Predicted response of tension stiffening by the proposed
cracking, f’t is the concrete tensile strength taken as model for ρb=2% and reinforced with: (a) GFRP bar with helically
0.38(f’c)0.5, γ is a factor based on surface profile of the wrapped surface profile (Type A) and (b) GFRP bar with ribbed
reinforcing bar, εcf is the average net concrete strain, εcr surface profile (Type C).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

24 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Replacement of Steel with GFRP for Sustainable Reinforced Concrete

Once the GFRP tension stiffening models validity had been excursions while subjected to constant axial load. Various
confirmed, it was adopted in VecTor2 to predict the GFRP- combinations of steel and GFRP reinforcements were
RC beam behavior again. All other parameters in the investigated.
analysis were kept the same. The only parameter varied
The test program on bars in compression consisted
was the tension stiffening model. Figure 7 shows that
of 34 specimens of 25M GFRP bars tested under direct
the predicted results for the load-deflection response of
compression. The lengths of the specimens were varied
beams utilizing the proposed tension stiffening model were
between 50 mm and 600 mm to establish the relationship
considerably more accurate. As expected, the analytical
between length and strength. An ultimate compressive
ultimate deflection was now closer to experimental value
strength of 730 MPa was observed in bars with nominal
and the accuracy of post-cracking stiffness prediction
un-braced length of about 230mm which is about 60% of
was significantly improved.
the tensile strength. The average compressive modulus
of elasticity was determined to be 60 GPa, based on
measurements provided by gauges during the initial
part of each test. This value is similar to the reported
nominal modulus in tension. Longer specimens, with a
nominal un-braced length larger than 315 mm, failed by
buckling with ultimate compressive strength being an
inverse function of the un-braced bar length. Between
230 mm and 315 mm bar length, the failure mode was
a combination of buckling and material failure. Figure 8
shows the compressive strength vs. unbraced length for
the 25 mm GFRP bars.

Fig. 8: Compressive strength of bars in compression as affected

by unbraced length

Seventeen columns with GFRP reinforcement were tested

under lateral displacement excursions while subjected
to constant axial load as shown in Figure 9. The format
of lateral displacement excursions is also shown in the
figure. Each column had a diameter of 356 mm and a
length of 1470 mm and was cast integrally with a large
concrete stub. Main variables compared were the axial
load level, type of reinforcement and the size, volumetric
ratio and spacing of transverse reinforcement. Column
size, concrete strength, and amount and distribution
Fig. 7: VecTor2 prediction by proposed tension stiffening model:
(a) normal strength concrete beam and (b) high strength con-
of longitudinal bars were kept constant in order to
crete beam make direct comparisons. The stubs had dimensions of
484×700×800 mm, were heavily reinforced and acted as
a fixed support for the column.
Behaviour of Gfrp-Rc Columns
An experimental program was undertaken to evaluate Behaviour of columns in the form of shear vs. tip deflection
the behavior of GFRP bars in compression. In addition to and moment vs. curvature was studied. Figure 10 shows
testing the bars under concentric compression, a number the behavior of a column in which six 25 mm diameter
of columns were also tested under lateral displacement GFRP bars were used as longitudinal reinforcement and

Organised by
India Chapter of American Concrete Institute 25
Plenary Session - Paper 5

(a)

(b)

Fig. 9: Column Specimen 8, 24th cycle: (a) Test Set-up and (b)
Cyclic lateral displacements

12 mm GFRP spirals with 50 mm pitch were used as

Fig. 10: Shear vs. deflection and moment vs. curvature relations
lateral reinforcement (Tavassoli et al., 2015). The figure of P28-C-12-50
shows a stable column behaviour with a large energy
dissipation capacity and high level of deformability. The
columns underwent several load cycles before failure.
Columns containing GFRP longitudinal bars and spirals
in general had lower shear and moment capacities in
comparison with conventional steel reinforced columns
due to the lower stiffness of longitudinal GFRP bars (Liu
and Sheikh 2013). They also displayed softer response.
Hence it was decided to test columns reinforced with steel
longitudinal bars and GFRP spirals. It was postulated that
the longitudinal steel bars would restore the required
stiffness and strength of the column and at the same time
the GFRP transverse reinforcement would provide the
required confinement and corrosion resistance. Figure 11
shows the behaviour of a column which is similar to the one
discussed in Figure 10. The only difference between the two
columns is the type of longitudinal reinforcement. Column
P28-LS-12-50 had six 25 mm steel longitudinal bars while
P28-C-12-50 contained six 25 mm GFRP bars. A comparison
of the behaviour of columns in Figures 10 and 11 shows
that columns with steel GFRP bars are more deformable
but have lower strength and stiffness. They are also able
to undergo larger number of displacement excursions. It
should be noted that the stiffness of GFRP bars is about
30% of that of steel. If the amount of GFRP longitudinal
reinforcement is adjusted upward to match the stiffness of
the steel reinforcement, the behaviour of the columns would Fig. 11: Shear vs. deflection and moment vs. curvature relations
improve significantly in terms of strength and stiffness. of P28-LS-12-50

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

26 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Replacement of Steel with GFRP for Sustainable Reinforced Concrete

tension stiffening models resulted in overestimation of

stiffness and underestimating the ultimate mid-span
deflection of beams. A new tension stiffening model was
proposed for GFRP reinforced concrete members that
addressed this parameter and predicted the experimental
results significantly better than the existing models.
Confinement with steel in columns deteriorates rapidly
after yielding of steel while GFRP spirals are capable of
providing effective confinement until a strain of about 2%
thereby delaying crushing of the column core. GFRP bars
were able to resist stress levels in excess of 700 MPa in
monotonic compression. Concrete columns reinforced
Fig. 12: Stress-strain relationship of GFRP vs Steel with GFRP longitudinal bars and spirals and tested under
simulated earthquake forces behaved in a robust manner
It was observed that failure in the columns was initiated displaying large deformability. Use of GFRP longitudinal
mostly due to the crushing of concrete core and/or yielding bars results in lower strength and stiffness of columns
of the longitudinal steel bars. Rupture of GFRP spirals compared with steel bars. The overall strength and
was observed in some columns only in the last stages of deformability of columns confined with GFRP was found
the column’s life. Hence, the major advantage of GFRP to be similar to or better than that of columns confined
over steel was thus observed for the strain range beyond with steel. Thus, GFRP spirals can be used as primary
yielding. Steel spirals are very efficient up to a strain of lateral reinforcement in columns designed for non-
0.002 when they yield. Beyond the yield strain, a major seismic and seismic regions. An optimum solution against
drop occurs in its modulus of elasticity and the steel spirals corrosion of reinforcement in columns appears to be
are not able to contain the concrete core effectively. On the the use of longitudinal steel bars and GFRP transverse
other hand, GFRP behaves linear elastically until rupture reinforcement.
at an approximate strain of 0.02. Therefore, GFRP lateral
reinforcement continued to provide resistance with the
same stiffness against the expansion of the concrete core Acknowledgement
beyond the strain at Point A (Figure 12). The authors would like to express their gratitude and
appreciation to the sponsors of the research summarized
here. The financial support was provided by Schoeck
Concluding Remarks Canada Inc., Facca Inc and the Natural Sciences and
Corrosion of steel in reinforced concrete structures Engineering Research Council of Canada (NSERC) through
presents a serious challenge to the society especially the IC-IMPACTS Network of Centers of Excellence and
engineering community. Selected results are presented through a CRD grant. The material and technical support
in this paper from an extensive research program provided by Pultrall Inc., Schoeck Canada Inc., Hughes
which involves investigation of GFRP bars as internal Brothers Inc., Vector Construction Group and Dufferin
reinforcement to replace steel in concrete structures. Concrete is also acknowledged. The experimental work
These structures include beams, slab and columns. Thus was carried out in the Structures Laboratories at the
the performance of GFRP bars in compression, tension, University of Toronto.
shear and cyclic loads is under investigation with the main
purpose being to develop better understanding of the References
behavior of GFRP reinforced specimens under a variety 1. ACI 440.1R-06, 2006. Guide for the Design and Construction of
of loads. This research should help engineers gain more Structural Concrete Reinforced with FRP Bars. 2006. ACI Committee
confidence in using GFRP-RC as a sustainable alternative 440. American Concrete Institute, Farmington Hills, Mich., 44p.
to steel-RC structures. The following conclusions can be 2. Canadian Standards Association, 2012, Design and Construction of
drawn based on the work presented here. Building Components with Fibre Reinforced Polymers. CSA S806-12
2012, CSA Mississauga.
Test results show that GFRP-reinforced members 3. Collins, M.P. and Mitchell, D., 1997. Prestressed Concrete Structures.
behave in an excellent manner in flexure and shear Response Publications, Toronto and Montreal, Canada, 766 p.
displaying large deformations before failure. Due to 4. Getzlaf, D., 2012. An Investigation into the Flexural Behaviour of
the lower stiffness of GFRP, the reinforced concrete GFRP Reinforced Concrete Beams. Master’s Thesis, University of
design is largely controlled by service limit conditions Toronto-Toronto.
that makes it essential to predict the deflection in GFRP 5. Hays, G.F., 2010. NACE-International, The corrosion society. http://
reinforced members more accurately than for steel-RC events.nace.org/euro/corrodia/Fall_2010/wco.asp (June. 09, 2015)
members. The lower modulus of elasticity of GFRP can 6. Johnson, D. T. C. and Sheikh, S. A. 2013. Performance of bent
also adversely affect the shear response by widening stirrups and headed bars in concrete structures. Canadian Journal
of Civil Engineering, Oct. 2013, Vol. 40, No. 1, pp. 1082-1090.
cracks and reduce the efficiency of shear reinforcement.
Nonlinear finite element analysis using the existing
Organised by
India Chapter of American Concrete Institute 27
Plenary Session - Paper 5

7. Johnson, D., 2014. Investigation of GFRP bars as Internal 11. Tavassoli, A., 2015. Behaviour of Circular Columns Reinforced with
Reinforcement for Concrete Structures. PhD Thesis, University of Steel and GFRP under Combined Axial Load and Bending. MASc
Toronto-Toronto. Thesis, University of Toronto-Toronto.
8. Mufti, A., Nemkumar, B., Benmokrane, B., Boulfiza, M., and 12. Tavassoli, A., Liu, J., Sheikh, S.A., 2015. Glass Fiber-Reinforced
Newhook, J. 2007. Durability of GFRP Composite Rods. Concrete Polymer-Reinforced Circular Columns under Simulated Seismic
International. Feb. 2007. pp 37-42. Loads. ACI Structural Journal, V. 112, No. 1, Jan-Feb, 2015, 103-114.
9. Kharal, Z., 2014. Tension Stiffening and Cracking Behavior of GFRP 13. Vecchio, F. J., 1990. VecTor2 nonlinear finite element analysis.
Reinforced Concrete. MASc Thesis, University of Toronto-Toronto. Copyright 1990-2013 F.J. Vecchio.
10. Liu, J. and Sheikh, S. A. 2013. “FRP-confined circular columns under 14. Vint, L.M., 2012. Investigation of Bond Properties of Glass Fibre
simulated seismic loads”, ACI Structural Journal, Vol. 110, No. 6, Reinforced Polymer Bars in Concrete under Direct Tension. M.ASc.
Nov-Dec 2013, pp 941-952. Dissertation, University of Toronto: Department of Civil Engineering,
506 p.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

28 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 1 A
Session 1 A - Paper 1

Stability of Basalt Fibers in Concrete Medium

Indubhushan Patnaikuni, Himabindu Myadaraboina and Erick Saputra Atmaja
School of Civil Environmental and Chemical Engineering, RMIT University, Melbourne, Australia

Abstract There has been little previous research on the chemical

stability of BF. Some short term studies reported that
Basalt Fibers in concrete is relatively new area and recently
the alkali resistance of BF is good (Militky et al, 2003 &
some studies have shown that Basalt Fibers are facing
Wang 2008). But, more recently there have been studies
durability problem in concrete. Hence, this study investigates
reporting the chemical degradation of BF in alkali
the durability of two types of basalt fibers with similar
environment (Bin et al, 2011, Rabinovich, 2001, Bo et al.
chemical composition from two different sources to check
2010 and Jung and Subramanian 1994)
and confirm their durability following immersion in a range of
chemical solutions representative of the concrete medium. Research in RMIT university, under guidance of Dr
The tests were conducted over a 62 day period for the first Patnaikuni shows that BF in concrete have disappeared
type and for second for over a 35 days period. The solutions just after 28 days of age (Solikin, 2012), which is the
were sodium hydroxide, sodium chloride, sodium sulphate basis for the present investigation of BF under alkaline
and combinations of the three. The other set of solutions environment which resembles concrete.
were calcium hydroxide, calcium chloride, calcium sulphate
The present study confirms the degradation of BF
and combinations with sodium hydroxide. In both cases,
in alkaline environment with more detailed study by
the weight loss was observed in all alkali solutions with the
comparatively long-term analysis with chemicals as well
worst being the combination of the sodium hydroxide with
as micro structural analysis using SEM.
the sodium sulphate. The impact of chloride was minimal
in the alkaline environment and reduced the impact of the
sulphate when both were present. Microstructure analysis Research Significance & Methodology
undertaken for type one using SEM/EDX found that the silica As there is no trace of BF after 28 days in concrete with BF
structure present in basalt fibers was degraded by the OH- (Solikin, 2012), whereas Steel Fibers have shown up after
from the alkali medium. The study suggests the need for the 28days, the concern over durability of BF has increased.
modification or treatment of basalt fibers before using as Figure 1 shows the difference.
fiber reinforcement in concrete.
Keywords: Basalt fiber, Chemical, Durability and
aggressive environments, EDX, SEM.

Introduction & Background

The high fire resistance nature and low price of basalt fiber
(BF) compared to traditional fibers, has led to the emergence
of BF as an alternative to other fibers in fiber reinforced
polymer composites (FRP) in civil engineering over recent
years. FRP has been successfully used for strengthening,
rehabilitation and renewal of civil engineering structures Fig. 1: High Volume Fly Ash Concrete with (a) Steel Fibers (b)
for over the last two decades (Jongsung et al. 2005). Basalt Fibers at 28 days of age

More recently, continuous fibers extruded from naturally Given the use of BF as FRP in concrete and that concrete
fire-resistant basalt (at temperature 1300- 1400 degree is highly alkaline in nature, the long term durability and
centigrade) have been investigated as a replacement performance of the BF in concrete is of great concern.
for asbestos fibers, in almost all of its applications. The In addition to the alkaline nature of the concrete, a range
wide temperature range (-260/-200 to about 650/800°C) of other potentially deleterious chemicals may also be
compared to E glass: -60 to 450/460°C) enables the BF present. The most common of these are chloride ions,
to be good insulators and can replace asbestos (Van de from marine environments and de-icing salts, together
Velde, 2001).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

30 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

Table 1
Chemical Composition

Chemical components SiO2 (silica) Al2O3 (alumina) Fe2O3 (ferric oxide) CaO MgO Na2O TiO2 K 2O B2 O 3 F

Russian basalt fibers (Weight %) 57.5 16.9 9.5 7.8 3.7 2.5 1.1 0.8 - -

Chinese Basalt Fibers (Weight %) 48-60 14-19 9-14 5-9 3-6 3-6 0.5-2.5 0.8 - -

with sulphate ions from the soil and groundwater. The chemicals were also used. These were selected to simulate
study reported investigates the durability of BF in an the alkaline cementitious matrix. 1 M NaOH is to represent
alkali medium as well as alkali combined with sulphate the alkaline nature of concrete, 3% NaCl is to represent
and/or chloride ions which represents the aggressive the contamination by chloride ions, and 10% Na2SO4 is to
environments in concrete. Hence, strong alkali such as represent the contamination by sulphate ions, and their
NaOH and salts of strong alkali & strong base which are three combinations to simulate the effect of both chlorides
NaCl and Na2SO4 are considered for this study. and sulphates. The strengths of the solutions were selected
to provide accelerated testing regime.
The investigation is conducted on basalt fibers from two
sources and of having similar chemical and physical
properties, named type 1 and type 2 for distinguishing. Table 2
The effect on durability is measured by calculating the Solutions used for the experiments
weight loss after soaking fibers in different solutions over
Container No. Solution
a period of 62 days for type 1 and 165 days for type 2 fibers.
The condition of the fibres is also examined both visually 1 1 M NaOH
and using Scanning Electron Microscope (SEM) on FEI 2 3% NaCl
Quanta 200 ESEM. 1 M Na2SO4
3
4 1 M NAOH + 3% NaCl
Materials 5 1 M NaOH + 1 M Na2SO4

Basalt Fibers 6 1 M NAOH + 3% NaCl + 1 M Na2SO4 1M Ca(OH)2

The nominal diameters of these continuous filaments now 7 1M Ca(OH)2

come in the range 9 to 24 mm. Figure 2 shows the BF 8 1M CaCl 2
used in this project along with its SEM image. The basalt 9 1M CaSO4
fibers used in this experiment were provided by Kemenny
10 1M NaOH + 1M Ca(OH)2
Vek Ltd, Dubna, Moscow region, Russia (type 1) and Jiuxin
basalt Fiber Industry Co.Ltd, China (type 2). The Table 1 11 1M NaOH + 1M CaCl 2
gives the chemical composition of both the fibers. 12 1M NaOH + 1M CaSO4

Chemical Solutions
Experimental Procedure
Totally twelve different chemical solutions were used in this
The experiment is conducted in two stages. In the first
project as shown in table 2. First six chemicals are used
stage, the type 1 BF was tested under first 6 chemical
for type 1 project in stage 1, whereas for stage 2 and type 2
solutions. In second stage, the type 2 BF was tested for
BF, total twelve chemicals i.e in addition to first 6, another 6
the same 6 solutions and in addition to them the BF were
also tested for the rest of the solutions mentioned in Table
2, which resembles concrete medium.
For the entire experiment, samples were prepared by
placing 45 g of BF in 500 ml of each solution. For the
stage 1, the weight of BF was measured at fixed intervals
at 3, 7, 14, 21, 28, 42 and 62 days. Whereas in stage 2, the
measurements were taken at 1, 5, 10, 30, 35 and 165 days.
The pH of the solutions was also measured at each time
interval. The BF measurements were taken by washing
the specimens with distilled water and placing them in
an oven for 24 hrs at 105°C to obtain a dry mass at each
time interval. A photographic record of the fiber was also
Fig. 2: (a) Original BF and (b) SEM image of BF at 4K made at each point and Figure 4 shows the images after
magnification the treatment period of 62 days.

Organised by
India Chapter of American Concrete Institute 31
Session 1 A - Paper 1

conclusion of the test appeared same as the original BF,

Fig 4-a and b. At the conclusion of the test there was only
a 1.89 % variation in the initial and final weight, Table 3.
Severe damage of the fibers was observed in the BF
treated with NaOH, NaOH+NaCl and with NaOH+NaCl+
Na2SO4, Figure 4-c, d, & f respectively. The fibers had
lost all of their original characteristics such as colour and
texture and each fiber had broken and split into many thin
strands.
Fig. 3: Weight of type 1 BF immersed in different solutions Vs
Complete dissolution of the fibers was observed in the BF
time over a period of 62 days
treated with NaOH+ Na2SO4. Not only had the fibers lost all
of their original colour and texture, but most of the fibers
Prior to conducting the stage 2 experiment, the specimens
had become powder, Fig 4-e. The weight loss observed
treated in stage 1 were also analysed using SEM technique
was around 90% at the end of 62 days, Table 3.
to determine the condition of the fibers after and before
the treatment, and the presence of any precipitates on the
fibers. Results of Stage 2
The weight loss is observed as in table 4. The figure 5
shows graphical representation of weight loss of type 2
Results & Discussion basalt fibers in NaOH and combinations which are same
The mass of BF with time in each solution along with % as the solutions used for type 1 BF. Whereas the figure
weight lost and pH at the conclusion of the tests is shown 6 shows the weight loss analysis in Ca(OH)2, CaCl2 and
in Table 3 and Figure 3. Figure 3 shows the BF weight at CaSO4 and their combination solutions.
different days for stage 1 BF.
No surface damage was observed in the BF treated
with NaCl and Na2SO4. The condition of the fibers at the
Visual Analysis of type 1 BF in stage 1
conclusion of the test appeared same as the original BF,
The following observations are made based on the visual Fig 4-a and b. At the conclusion of the test there was only
analysis (Figure 4). a 1.89 % variation in the initial and final weight, Table 3.
No surface damage was observed in the BF treated Severe damage of the fibers was observed in the BF
with NaCl and Na2SO4. The condition of the fibers at the treated with NaOH, NaOH+NaCl and with NaOH+NaCl+

Table 3
Results obtained from the experiment during 62 days for type 1 BF

BF in BF in NaOH + NaCl
BF in NaCl BF in Na2SO4 BF in NaOH BF in NaOH+NaCl
NaOH+Na2SO4 + Na2SO4

Day 0 45.00 45.00 45.00 45.00 45.00 45.00

Day 3 45.96 46.05 45.65 45.50 46.40 46.80

Day 7 45.49 45.55 44.50 44.65 43.30 44.80

Day 15 44.66 44.70 41.85 41.60 39.10 38.95

Day 21 44.76 44.80 39.80 40.25 35.80 36.85

Day 28 44.39 44.45 37.50 38.80 30.65 34.45

Day 42 44.22 44.35 35.90 36.70 14.90 30.25

Day 62 44.13 44.15 31.85 32.50 3.90 27.00

Original pH of the
9.22 9.22 12.22 12.8 12.90 13.35
solution

pH after 62 days 8.33 8.19 12.55 12.68 12.93 12.83

% of Weight Loss
1.89 1.87 29.2 34.7 90 37.8
of BF after 62 days

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

32 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

Table 4
Results obtained from the experiment during 165 days for type 2 BF

Solution 1 Day 5 Days  10 Days 30 Days  35 Days 165 Days % Weight loss

NaOH 45.00 46.42 47.63 41.42 34.54 31.89 29.13

NaCl 45.00 43.55 44.14 39.27 35.96 33.70 25.11

Na2SO4 45.00 49.87 48.52 36.60 35.79 33.83 24.82

NaOH + NaCl 45.00 45.68 49.17 47.77 44.38 31.29 30.47

NaOH + Na2SO4 45.00 46.90 49.04 39.35 32.40 25.69 42.91

NaOH + NaCl + Na2SO4 45.00 46.41 49.87 40.85 35.70 30.55 32.11

Ca(OH)2 45.00 60.92 48.18 43.22 35.77 31.31 31.11

CaCl2 45.00 51.06 42.13 39.28 37.69 36.19 19.58

CaSO4 45.00 75.28 44.41 38.09 34.80 32.13 28.60

 NaOH + Ca(OH)2 45.00 55.75 51.62 55.04 44.24 40.63 9.71

 NaOH + CaCl 45.00 48.16 45.81 42.37 38.32 41.76 7.20

 NaOH + CaSO4 45.00 61.59 49.33 43.31 39.67 43.81 2.64

Fig. 6: Weight of type 2 BF immersed in Calcium solutions Vs

time over a period of 165 days

Fig. 4: Type 1 BF after 62 days of treatment with a) 3% NaCl (b) Na2SO4, figure 4-c, d, & f respectively. The fibers had lost
10 % Na2SO4 (c) 1 M NaOH d) 1M NaOH + 3% NaCl e) 1M NaOH all of their original characteristics such as colour and
+10% Na2SO4 (f) 1 M NaOH+10% Na2SO4 +3 %NaCl texture and each fiber had broken and split into many thin
strands.
Complete dissolution of the fibers was observed in the BF
treated with NaOH+ Na2SO4. Not only had the fibers lost all
of their original colour and texture, but most of the fibers
had become powder, Fig 4-e. The weight loss observed
was around 90% at the end of 62 days, table 3.

Visual Analysis of type 2 BF in stage 2

By the day 35, the type 2 BF crumbled and turned fragile
and hence easily turned into dust. This might have been
caused by the absorption of the solution within the fibers
reaching its limit.
Fig. 5: Weight of type2 BF immersed in different solutions Vs
time over a period of 165 days

Organised by
India Chapter of American Concrete Institute 33
Session 1 A - Paper 1

After 165 days, most SEM Analysis of stage 1

of the solutions The type 1 BF was examined using SEM analysis to
inside the containers determine changes in the morphology and composition
were absorbed by with time. An analysis was conducted on untreated
the fibers almost original basalt fibers and on fibers from each solution
completely. The at the conclusion of the trial. The SEM images of each
volume of the specimen are shown in Figure 8.
solution containing
NaOH mixtures were
decreasing more Discussion
than half of the initial The SEM results show that little to no degradation occurs
state indicated by the on the BF exposed to just the NaCl or the Na2SO4, small
solution line within deposits of the chloride and sulphate salts were observed
the containers; this on the fibres.
Fig. 7: BF in Ca(OH)2 after 165 days includes NaOH and All the fibers which were exposed to the alkaline (NaOH)
its combinations. solution have shown degradation. This degradation is
And in the solutions caused by the hydroxyl ions forming weak bonds with
that contained calcium component, they were almost silicon atoms on the substrate surface causing the original
absorbed completely by the fibers, leaving none of the lattice bonds to weaken.
liquid within the containers. However, the volume of the
colorless solutions, i.e just salts alone such as NaCl, The addition of NaCl to the NaOH resulted in similar
Na2SO4, CaCl2 and CaSO4 were mostly stayed similar to observations to the NaOH solution. However, when
the initial experiment day. Na2SO4 is added to the NaOH the degradation of the fibres
has increased, such that the fibres are reduced to powder
After 165 days, most of the fibers were not able to retain by the conclusion of the trial. It is hypothesised that this is
most of the initial characteristic, and in contrast, most due to the sulphate ions increasing the pH of the solution
of the samples were broken to the level of being easily due to the hydration reaction.
crushed into dust without leaving traces of once being
SO42-+ H2O <----> HSO4 + OH- (Hence release of
fibers at all. Figure 7 shows the type 2 BF in Ca(OH)2
more OH- ions) ...................................................(1)
solution after a period of 165 days. The components of
the fibers became powder and its metallic color has
This is reflected by a lower pH in the NaOH + Na2SO4
disappeared completely. solution compared to the NaOH or NaOH + NaCl
solutions.

Conclusion
Based on the stage 1 and stage 2 of the experimental
results, it can be concluded that, the BF of both types or
any type are degrading in alkaline environment. And, it is
worsening with the additional of sulphates. BF in Ca(OH)2
which resembles concrete medium also has shown
severe degradation. Hence, following conclusions have
been drawn from this study.
1. Neither chloride ions nor sulphate ions alone have any
deleterious effect on the BF however, in an alkaline
environment equivalent to a pH 12.5 severe degradation
is observed.
2. The degradation is increased in an alkaline environment
by the presence of sulphate ions.
3. The presence of chloride ions has little effect on the
degradation in alkaline solutions, but does slightly
reduce the effect when both chloride and sulphate ions
are present.
Fig. 8: SEM images of type 1 BF after 62 days treatment with a)
4. The data indicates that the use of BF in concrete as fibre
3% NaCl (b) 10 % Na2SO4 (c) 1 M NaOH d) 1M NaOH + 3% NaCl
e) 1M NaOH +10% Na2SO4 (f) 1 M NaOH+10% Na2SO4 +3 %NaCl reinforcement is susceptible to degradation due to the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

34 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

alkaline nature of the concrete. Hence modification or 3. F. N. Rabinovich., "Stability of Baslat Fibers in a medium of
Hydrating Cement, International Journal of Glass and Ceramics,
treatment of basalt fibers before using in concrete is
Vol 58, 11-12, 2001.
recommended.
4. Jongsung. Sim, Cheolwoo Park, Do Young Moon. Characteristics
of basalt fiber as a strengthening material for concrete structures,
Acknowledgement Composites: Part B Engineering vol. Vol. 36, pp. 504-512, 2005.
5. J.Militky, J.Zeisbergerova, V. Kovacic.(2003). Chemical degradation
The authors would like to acknowledge Mr Pavel Ryjkov of Basalt Fibers.Retrieved from
and other technical staff of concrete laboratory of RMIT
6. http://en.trovel.ru/u/file/chemical_degradation_of_basalt_fibers.
for their support in conducting the experiment. doc
The SEM analysis was carried out with the assistance of 7. T.H Jung and Subramanian, Aklali resistance enhancement of basalt
the RMIT Microscopy and Microanalysis Facility (RMMF) fibers by hydrated zirconia films by the sol-gel process, Journal of
at RMIT University. Also, would like to acknowledge the Materials Research, Vol 9-4, Apr 1994.
Kemenny Vek Ltd, Dubna, Moscow region, Russia and 8. Van de Velde K., Basalt Fiber as a reinforcement for a composite,
Jiuxin basalt Fiber Industry Co.Ltd, China for providing Retrieved from www.basaltex.com/files/cms1/Basalt-Fibres-as-
reinforcement-for-composites_Ugent.pdf, 2001
basalt fibers for the present research.
9. Wang Mingchao (2008), Chemical Durability and Mechanical
Properties of Alkali-proof Basalt Fiber and its Reinforced Epoxy
References Composites, Journal of Reinforced Plastics and Composites, SAGE,
1. Bin Wei, Hailin Cao and Shenhua Song. Degradation of basalt Vol 27-4, 393, Jan 2008, Beijing.
fiber and glass/epoxy resin composites in seawater, Corrosion
10. Solikin, M. (2012). High Performance Concrete with High Volume
Science,Vol 53, 426-431, 2011.
Ultra Fine Fly Ash Reinforced with Basalt Fibre. School of Civil
2. Bo Xiao, Hui Li & & Guijun Xian, Hydrothermal Ageing of Basalt Fiber Chemical and Environmental Engineering. Melbourne, Australia,
Reinforced Epoxy Composites, The 5th International Conference on RMIT Univeristy. PhD thesis.
FRP Composites in Civil Engineering, 2010, China.

Dr. Indubhushan Patnaikuni

B.E., M.Tech., Ph.D., FIEAust. CPEng.
Dr. Indubhushan Patnaikuni is an academic staff member of the School of Civil, Environmental and Chemical
Engineering at RMIT University. He is a Fellow of Engineers Australia. He has published about 160 high
quality papers in various journals and international conferences including three papers which received
best paper awards. He chaired numerous technical sessions of international conferences and delivered 26
keynote addresses in international conferences. His expertise in the area of high performance concrete,
sustainability and engineering education is well recognised internationally. He received prestigious awards
such as Vishal Bharati Gaurav Satkar (2006), Samaikya Bharat Gaurav Satkar (2005), Rashtriya Vikas
Shiromani (2003), Eminent Engineer (1997, during Golden Jubilee Celebrations of Independence of India). He
received Sir Arthur Cotton Memorial Gold Medal (1975-76).
Dr. Patnaikuni was a member of the International Committee on Concrete Model Code (ICCMC) and is one
of the Editors of Asian Concrete Model Code. He is pioneer of migrant engineer education programs in
Australia. He was honoured as a special invitee for the seminar organized by the International Organisation
on Migration (IOM), Geneva and the National Office of Overseas Skills Recognition.
He was appointed as a member of the International Advisory Committee to advise the National Working
Group of India on University Industry Science Partnership (within the framework of UNESCO-UNISPAR
Program). He is an Executive Committee Member of International Structural Engineering and Construction
Conferences. He is a Foundation Executive Committee Member and currently the President of ‘Indian
Descendent Engineers and Scientists of Australia (IDEAS)’. He is an Executive Committee Member of the
Structural Engineering Branch of Engineers Australia (Victoria division) from 2002 to 2014 and was Vice-
President of The Association for Advancement of Sustainable Materials in construction.

Organised by
India Chapter of American Concrete Institute 35
Session 1 A - Paper 2

Concrete using Siliceous Fly ash at very High Levels of Cement

Replacement: Influence of Lime Content and Temperature
G.V.P.Bhagath Singh and Kolluru V.L. Subramaniam
Department of Civil Engineering, Indian Institute of Technology Hyderabad, Hyderabad, TS, INDIA.

Abstract replacement since it allows for beneficial utilization of

Potential for producing viable binders at very high levels waste material which is generated in large quantities.
of cement replacement (60% and above) with fly ash is Since the demand and consumption of Portland cement
explored. The role of lime content and temperature on are increasing, it is becoming imperative for the cement
the efficiency of fly ash in contributing to strength gain and concrete industry to start utilizing more fly ash to
is investigated using quantitative X-ray diffraction (XRD) meet these demands rather than increase Portland
analysis. Results of fly ash characterization are presented cement production.
using quantitative X-ray diffraction (XRD) to identify Fly ash used in moderate quantities as cement
its reactive potential associated with the amorphous
replacement can significantly enhance the long-term
silica content. A new method for quantitative phase
properties of concrete. However, concrete made with
analysis of the amorphous phase contributions in the
fly ash substitution of cement often displays slow
XRD spectrum is presented. A strength-based efficiency
hydration that is accompanied by slow setting, low early
factor which provides a measure of the contriubtion of
age strength. This effect is more pronounced as the
fly ash is introduced. Temperature is shown to increase
level of fly ash replacement is increased. At high levels
the efficiency of fly ash by accelerating the dissolution of
of replacement there is a reduction even in the final
the reactive amorphous content. Efficiency of fly ash in
properties. The low reactivity and the lower quantity of
the binary high volume fly ash-cement blend is limited
by the availability of lime. Increasing the lime content in reactive components in fly ash is a major hindrance to the
the system provides significant enhancement in strength, development of concrete consisting of large volumes of fly
but it does not influence the dissolution of fly ash With ash. In concrete containing fly ash, a significant proportion
the availability of lime, the efficiency is limited by the rate of fly ash remains unreacted even after significant time. It
of contribution of reactive Silica from fly ash, which is is therefore becoming evident that optimal replacement
influenced strongly by temperature. Concrete strengths of level of fly ash in cement depends on assessing its full
30 MPa and higher were achieved with 65% replacement of reactive potential and allowing the maximum potential
cement with fly ash and total cement content of 100 kg/m3. of the cementing action provided by fly ash hydration to
The strength gain in concrete is shown to be related to the be harnessed. Effective use of high volume fly ash as
depletion of lime in the system, formation of amorphous cement replacement therefore requires development of
hydration products and the depletion of Si/Al content from an understanding of the reactivity of fly ash.
fly ash. An investigation of the underlying mechanisms
Hydration in cementitious systems with high volume fly
reveals the potential for further strength enhancement by
ash as cement replacement involves multiple complex
effectively engaging all the reactive components of fly ash.
processes. Understanding the underlying processes can
Keywords: Fly ash, Quick lime, Rietveld refinement, lead to production of alternate binders which minimize or
Efficiency factor, Si/Al dissolution, Amorphous product totally eliminate the use of cement. Proper understanding
content. of the type and quantity of products formed, the influence
of process variables and the link between the products of
Introduction reaction and the properties of concrete is required. Proper
Improvements in specific aspects of concrete performance characterization of fly ash is an essential first step. Use of
have been obtained using mineral admixtures or very high volume fly ash in concrete requires very careful
supplementary cementitious materials (SCM) at low consideration, from the characterization of the starting
levels of cement substitution (ACI report, 2002). Blended materials, through mixture proportioning and curing
cements, which are essentially binary blends of cement options to achieve desired properties, to the in-place
and SCMs, are available commercially. The use of fly ash early-age and long-term performance of the concrete in
in concrete is particularly attractive for use as a cement its fresh and hardened states.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

36 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete using Siliceous Fly ash at very High Levels of Cement Replacement: Influence of Lime Content and Temperature

In this paper results of an experimental study aimed (Maltais and Marchand,1997). Temperature increases
at producing High volume fly ash concrete suitable for the pH level of the pore solution increasing the OH- ion
warm weather concreting using 100 kg/m3 cement which concentration in the system, which causes the dissolution
achieves a target mean strength of 30 MPa at 28 days are of fly ash particles quickly (Fraay et al.,1989). A similar
presented. Starting with a mix designed for a characteristic observation was made by (Zhang et al., 2000) and they
strength of 35 MPa, the target strengths are achieved at also reported pH value decreases with increasing amount
very high levels of cement replacement with fly ash which of fly ash content.
is activated using quick lime. The study demonstrates that
Quick lime as fly ash replacement in cement-fly ash
concretes with acceptable strength can be formulated
blends has been shown have a positive influence mainly
to meet the original project specifications at 65%
on strength development and has been shown to be an
replacement of cement with fly ash. An efficiency factor
effective way for producing Ca(OH)2 in the mix (Shi, 1996).
is applied to measure the performance of reactivity of fly
The reaction rate of high Calcium fly ash with quick lime
ash in concrete compared to cement. An investigation of
the underlying reactions reveals the potential for further has been shown to produce a notable acceleration of the
strength enhancement through activation by effectively fly ash degree of reaction throughout the curing period
engaging all the reactive components of fly ash. A new (Antiohos et al., 2003&2007). Quick lime addition and its
method based on quantitative XRD is used to determine subsequent formation to Ca(OH)2 result in a higher basicity
the quantity of Si/Al cement and the amount of hydrated inside the matrix. It has been suggested that the increase
amorphous product formed in the system. in pH leads to the corrosion of the densified outer layer
of fly ash particles leaving more active cores exposed for
reacting to form additional hydration products (Ma and
Background Brown, 1997). It must be pointed out that all the studies
Fly ash as a cement replacement in concrete is very using hydrated and quick lime for increasing the Ca(OH)2
common and attractive because of its large availability, content in the mix were limited to lower levels of cement
widespread familiarity with use in construction and the substitution with fly ash. An understanding of the role
potential for high volume utilization. However, the use of temperature and lime content on the reactivity and
of fly ash is also accompanied by increased setting time efficiency of fly ash as high volume cement replacement
and decreased early strength. To improve the properties is required to produce viable high volume fly ash concrete.
of fly ash researchers have used of the methods like
thermal (Maltais and Marchand, 1997; Shi and Day, 1993),
mechanical (Bouzoubaa et al., 1997) and chemical (Shi and
Experimental program and methodology
Day,1995) activation to achieve enhanced reactivity from For evaluating very high volume cement replacement
fly ash and to compensate the loss of early strength. The with fly ash, a baseline fly ash concrete mixture with
efficacy of some of these methods is however debatable the 70% of the cementitious binder consisting of fly ash
since a number of them are too energy demanding, while was considered. The role of lime content was evaluated
others fail in simple cost-benefit analysis. by adding quick lime (QL) to the binary fly ash-cement
mixture. Reagent grade QL of 95% purity was used. When
From the different methods to enhance the reactivity of
QL was added, equal weight of fly ash was replaced with
fly ash, such as fine grinding, elevated temperature curing
QL. The total mass of the binder phase consisting of
and use of chemical activators it has been found that the
cement fly ash and QL was kept constant in all concrete
efficiency of fly ash is linked to the enhancements of the
mixtures. 5% and 10% of fly ash was replaced with QL (by
rate and extent of pozzolanic reactions between fly ash and
lime, and hence increase the strength development rate equal weight) keeping the content of cement at 30% by
and ultimate strength of hardened concrete containing fly mass of the total binder phase. To investigate the influence
ash. While chemical activation, which consider alternate of temperature, two different curing temperatures were
reaction pathways have been explored, it has been used. The role of temperature and lime content on fly ash
found that the activation of the pozzolanic reactivity of reaction were evaluated using quantitative XRD analyses
coal fly ashes is the most effective way to improve the performed on paste samples with identical water to binder
performance of the fly ash in concrete. content as the concrete mixtures.

Pozzolanic reaction starts after the breakdown of glass Commercially available ordinary Portland cement
particles in the fly ash. Fly ash does not react during confirming to Grade 53 of the Indian Standard, IS 12269
the first few days of curing, sample cured at 20 deg C and Siliceous fly ash confirming to the requirements of
and Increasing the temperature from 20 deg C to 40 IS 3812 and IS 1727 were used for all concrete mixtures.
deg C increases the reactivity of fly ash contributing to The oxide composition of the cement and fly ash used in
improvements in both early and long term the compressive this study were determined using X-ray fluorescence
strengths of fly ash cement blended concrete while it spectroscopy (XRF).The chemical and physical properties
reduces the long term compressive strength of the OPC of cement and fly ash are given in Table 1.

Organised by
India Chapter of American Concrete Institute 37
Session 1 A - Paper 2

Table 1. Chemical composition (% by mass) and binder. In concrete mixtures containing lime activators,
physical properties of cement and fly ash the equivalent weight of fly ash was reduced to keep the
Compound Name Cement Fly ash total weight of the cementitious binder equal to 340 kg/m3.
(% mass) (% mass) The batch weights of the lime activated mixtures are also
Al 2O3 3.10 28.83 shown in Table 2.
SiO2 15.76 57.35 Standard 150 mm cubes were casted from each mixture
CaO 71.33 1.92 to evaluate compressive strength gain. A slump in
Fe2O3 5.53 5.97 the range of 60-100 mm was obtained for all concrete
MgO 0.72 0.50
mixtures. No other admixtures were used or added to the
concrete mix. Immediately after casting, all specimens
K 2O 0.72 1.93
were covered with plastic covers to minimize moisture
SO3 2.06 0 loss and transferred to a temperature chamber, which
Cl 0.23 0.25 was maintained temperature 25 deg C and 40 deg C with
Na2O 0 0 the humidity range 80-85%. Specimens were demolded
TiO2 0.52 2.24
after 24 hours and kept in same temperature chambers
till the day of testing. For each mix, compressive strength
Loss of Ignition 0.8 1.89
was measured at 1, 3, 7, 14, 28, 56 and 90 days of age.
Blaine Fineness (m /kg)
2
325 320 A 5000 kN Compressive testing machine was used for
Specific gravity 3.15 2.29 testing. A loading rate of 5.25 kN/sec was prescribed
during the compressive test.
The particle size distributions of all the components
of the binder phase were determined using Microtrac Table 2. Batch weights in kg/m3 for concrete mixtures
S3500 Particle Size Analyzer. Isopropanol was used as (water to cementitious ratio 0.43)
the medium for dispersion to arrest the agglomeration Mix Designation/ Control Baseline 5% 10%
between particles. The particle size distributions of all Materials(kg/m3) Fly Ash Quicklime Quicklime
the materials of the binder phase are shown in Figure OPC 53 grade cement 340 100 100 100
1. It can be seen that while the cement and fly ash have Fly ash 0 240 221 204
comparable size distributions, QL is finer.
20 mm aggregates 573 573 573 573
10mm aggregates 573 573 573 573
Fine aggregates 767 767 767 767
Water 146 146 146 146
Quick lime -- -- 19 36
Label C F 5QL 10QL

The paste samples were prepared using a paddle mixer,

rotated at 275 rpm for 3 minutes. Immediately after
mixing, the paste was cast into 2 ml air tight vials and
placed in a temperature controlled chamber which was
maintained at a constant temperature. All vials were
kept at the constant curing temperature until tested. At
the designated age, the paste specimens were crushed
Fig. 1: Particle size distributions of the components of the binder inside the vails and ground to a finer size using a mortar
phase and a pestle. Particles passing through a 0.5 mm sieve
Concrete mix design was for a target mean strength of 43 were collected, immersed in methanol for one hour and
MPa, and the water/cementitious ratio was taken equal to oven dried at 40 deg C to evaporate the methanol, before
0.43. The cement content for the control mixture was fixed collecting XRD data.
at 340 kg/m3. In the concrete mixture fine aggregate were Laboratory X-ray diffraction measurements were in
taken as 40% of the total aggregate volume fraction. The vertical Bragg-Brentano (θ/θ) geometry between 10o and
batch weights for the control mixture, which contained no 70o at 0.02o steps using a D2 PHASER (Brucker) automated
fly ash and the baseline fly ash mixture prepared with 70% diffractometer with Cu-kα radiation (1.5418 Ao). X-ray
by mass replacement of cement with fly ash are shown tube was operated at 30 kV and 10 mA and the samples
in Table 2. Concrete mixtures were prepared with quick were rotated at 15 rpm during acquisition to improve
lime activator dosages of 5 and 10% by weight of the powder averaging and α-Al2O3 was used as an external
cementitious binder material. The weight of the quicklime standard. A first-order Chebyshev polynomial combined
activator was included in the weight of the cementitious with 1/2θ term was used to fit the background intensity

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

38 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete using Siliceous Fly ash at very High Levels of Cement Replacement: Influence of Lime Content and Temperature

before performing phase analysis. Rietveld refinement from the refinement. The mass absorption coefficients
was performed on the powder patterns, using TOPAS of samples were calculated considering the composition
4.2 software. The crystal structures used to interpret using the International tables of crystallography for CuKα
the powder patterns were taken from the inorganic radiation. The amorphous content was determined as the
crystal structure database (ICSD). The quantification of residue of the weight fractions of determined crystalline
the crystalline phases was performed using the external phases. The total amorphous content and the weight
standard method with the refined Rietveld scale factors. fractions of Mullite and Quatrtz in the fly ash determined
from X-ray analysis, are listed in Table 3. The results
Results suggest a significant percentage of the Silica and Alumina
The X-ray diffractogram of the unhydrated fly ash is are available in the non-reactive crystalline forms
shown in Figure 2. The crystalline phases associated associated with Quartz and Mullite.
with Quartz and Mullite are readily identified in the
diffractogram. The diffuse scattering produced by the Table 3. Main components of fly ash obtained from quantitive
amorphous phase present in fly ash appears as a broad phase analysis (% by mass)
hump on the diffractogram (shown in the inset for angles Reactive Amorphous Total SiO2 Quartzd Mullited
between 15 and 35 degrees). The amorphous phase of fly SiO2 a Si/Al b and Al2O3c
ash is indicative of the reactive component of fly ash. It 20.19 34.90 86.18 25.64 30.07
is mainly associated with the amorphous forms of Silica a
Value determined as specified in Indian Standard (IS 3812-part 1).
and Alumina and is referred to as amorphous Si/Al in the b
Value determined from quantitative phase analysis after Rietveld
rest of the paper. The absolute weight fraction, wα of a refinement.
crystalline phase, α was determined as c
Al 2O3 and SiO2 content obtained from XRF
Q ZMV Va S S a XS n sample X Quartz and Mullite determined by Ritveld-based external standard
d

w a = #T Y n s &w s
Q ZMV Vs S s
.............................(1) method

where (ZMV) α and (ZMV)s are the phase constants of the The reactive silica content of the fly ash determined using
phase α and the standard, respectively, S α is the scale factor the acid dissolution method given in IS 3812 is listed in
of the phase, Ss is the scale factor of the standard, µsample Table 3. The reactive Silica is indicative of its potential
and µs are the mass absorption coefficient of the sample for use as replacement of cement. For the fly ash used
and standard, respectively and ws is weight fraction of the in this study, only 20.2% of the total 57.4% Silica content
standard. The phase constant (ZMV) of a crystalline phase available in the reactive form. The results of the study
can be calculated from the crystal structure obtained indicate that the reactive components of fly ash cannot

Fig. 2: X-ray diffractogram of fly ash. (Q-quartz,M-mullite)

Organised by
India Chapter of American Concrete Institute 39
Session 1 A - Paper 2

Fig. 3: X-ray diffractograms of Cement and hydrated cement samples. ( P-portlandite (CH), C-calcite, G-gypsum)

directly be assessed from an oxide content evaluation as obained from the control mix is indiactive of pozzolanic
is traditionally done, but require a careful evaluation of the reaction involving fly ash. The QL activated fly ash mix
phases of Silica and Alumina present in the ash. showed further enhancement in strengh compared with
the base line mix (F mix), which is indicative of further
The X-ray diffraction patterns of a hydrated cement sample
enhancement of the pozzolanic reaction in the presence of
is shown in Figure 3. Hydration transforms the cement into
extra CH supplied by QL addition. The test results indicate
an X-ray amorphous calcium silicate hydrate that appears
that QL is effective in increasing the rate of strength gain
as a broad hump (between 25 and 40 degrees) on the
at a higher temperature.
diffractogram of the hydrated cement sample. The hump
associated with amorphous calcium silicate hydrates.
New crystalline phase can be identified with the formation
of Calcium Hydroxide. The hydration of cement leads to a
total consumption of Gypsum. Additionally, Ettringite was
found to form within the first day.
The compressive strength obtained from base line (F mix)
and quick lime (QL) activated fly ash mixtures cured at 25
deg C and 40 deg C are shown in Figures 4(a) and 4(b),
respectively. 30% scaled values obtained of the control
mix are also ploted in the figures for comparision. Results
indiacte that at both the curing temperatures QL activated
mixes showed improvement in compressive strength
when compared with the base line mix. At 25 deg C, the
compressive strength increase in the QL mixes is visible
only after 28 days and it is very nominal when compared to
the base line mix. At 40 deg C, the base line (F mix) and QL
activated mixes show higher strength than the 30% control
mix at all ages. QL activation indicate a clear enhancement
in the rate of ealry strength gain and improvement in the
strength at lateral ages when compared with the baseline
fly ash mixture. At both curing temperatures 5% and
10% QL activated mixed showed very little difference
in compressive strength. The increase in compressive
strength of base line fly ash mix above the 30% scaled value Fig. 4: Compressive strength gain in quick lime activated fly ash
concrete mixtures cured at (a) 25 deg C and (b) 40 deg C with age.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

40 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete using Siliceous Fly ash at very High Levels of Cement Replacement: Influence of Lime Content and Temperature

Fig. 5: X-ray diffractograms of base line (F) mix and QL activated fly ash paste sample cured at 25 deg C and 40 deg C at the age
90 days. (Q-quartz, M-mullite, P-portlandite(CH), C-calcite, G-gypsum)

The X-ray diffractograms of the baseline (F) mix and QL are also plotted in the figures for comparison. It can be
activated fly ash mixes cured at 25 deg C and 40 deg C at seen that for the control mixture, the hydraulic activity of
the age of 90 days are shown in Figure 5. The crystalline cement results in continuous increase in the CH content.
phases quartz and mullite are readily idenitifed In all the QL activated mixes showed higher CH content than the
mixes and are observed to remain unchanged with age. base line mix (F mix) and higher dosage of QL producing
A broad amorphous hump is observed in all the mixes higher CH content in the system. The results indicate that
between 15o and 38o 2 theta angles. A depression in the the QL addition immediately contributes to CH content in
amorphous hump at lower theta angles and a rightward the system and would therefore be available for reaction.
shift of the entire hump is observed with age. The diffuse At 25 deg C, CH depletion is initiated between 7 and 14 days
scattering produced by the amorphous phase of fly ash is of age and there is a steady decrease in the CH content
associated with lower 2-theta angles, in the range 15 to 30 after 14 days.The rate of CH depletion is same in baseline
degrees. The amorphous products of hydration contribute and QL activated mixes. At 40 deg C, CH depletion intiated
to diffuse scattering patter at higher 2-theta angles in at the age of 3 days and rate of depletion is very rapid up to
the range 25 to 40 degrees. In the base line (F) mix, 7 days of age. After 14 days of age there is a steady linear
the Portlandite peak is almost absent at 90 days of age decrease in the CH content up to 28 days. There is a small
indicating a complete depletion of lime in the system. In the change in the CH content after 28 days and up to 90 days
QL activated mixes, the Portlandite peaks are decreased of age. Considering the pozzloanic reaction to be the only
yet visible at the age of 90 days indicating availability of source of lime consuption, the depletion of CH can be
lime in the system. In all mixes tested the reduction of related to the rate and extent of pozzolanic reaction.
the hump associated with amorphous phases of fly ash
(at lower theta angles) and the rightward shift associated CH content is almost completely depleted in base line
with the formation of amorphous silicate hydrates is mix (F mix) while there is CH content left over in the
accompanied by a reduction in Portlandite. In all hydrated QL activated mixes at the age of 90 days. The complete
mixes the quantitied of crystalline phases like gypsum, depletion of lime in the the baseline mix suggests that
level of cement replacement is limited by lime availability
portlandite and calcite are observed to change with age.
from the cement.The 10% QL activated mixes showed
The depletion of Gypsum coincides with the formation of
more CH content than the 5QL at the age of 90 days . The
ettingite in all mixes and gypsum is totally depleted by 90
rate and pace of CH depletion in the system correlates
days in all mixes.
very well with strength development. At 40 deg C the
The Ca(OH)2(CH) content with age as a percentage of rapid initial consumption of CH in the first 7 days matches
the cementitious binder cured at 25 deg C and 40 deg C well with the observed higher rate of strength gain. At 25
obtained from Rietveld refinement is shown in Figure 6(a) deg C, the rapid uptake of CH after 14 days coincides with
and 6(b). The 30% scaled values of the control mixture the observed increase in the rate of strength gain after 14

Organised by
India Chapter of American Concrete Institute 41
Session 1 A - Paper 2

depends on the age and temperture. At 25 deg C the

efficiency factor Improved after 14 days .Intially, in the first
7 days the fly ash acts as an inert material. The resulting
decrease inefficiency in this period can be associated with
the dilution effect considering cement hydration as the
only contributor to strength in the system. Base line mix
(F mix) shows only 50% efficiency when compared with
QL activated fly ash system. This may be related to the
complete depletion of lime and leading to non-availability
of lime to support the pozzolanic reaction. However the
small increase in efficiency in the QL systems suggests
at later ages indicates the possibility of additinal factors
which influence the strength gain.
At 40 deg C, the QL activated mixes achieve higher
efficiency, close to 70%. The efficiency factor clearly
demonstrates the efficiency of QL increases with
increasing temperature. The improved efficiency in the
QL systems correlates well with the deplpetion of CH
content. The differences in the efficiency factors obtained
for baseline and QL activated blends after 3 days coincides
with the initiation of CH consumption. However, the limiting
efficiency of 70% in QL systems suggests other limiting
factors which control strength gain despite availability of
CH in the system.
A quantitative phase analysis of the amorpous phase was
Fig. 6: Calcium hydroxide content of quick lime activated fly ash performed using the available XRD data. A whole pattern
mix cured at (a) 25 deg C and (b) 40 deg C with age. based method which relies on determining the total
intensity contribution from the different components from
days. There is a larger residual CH in the 10% quicklime the total areas associated with each is used. The method
system at any age which does not appear to influence the
strength gain.

Analysis of Results
Efficiency factor of the base line fly ash mix (F mix) and
quick lime activated mixes with age was evaluated
considering the basic relationship for strength gain in a
concrete mixture (given as a proportion of strength at a
given age t, S and strength at infinity, S∞) is given as

S3 = KT " 1 + KT t %
S t .......................................................(2)

where, K T is a factor which depends on the cement type and

temperature (Carino, 2004). An expression for efficiency
factor, Ceff, obtained from strength gain was derived as

C eff = S # K t + 1&
S 1 ......................................................(3)
3 T

Ceff provides a measure of efficiency of fly ash in contributing

to compressive strength when compared with concrete
with equal weight of pure cement. The efficiency factor
for cement with no replacement is 1.0 and it provides a
basis for assessing the eficiency of other supplimentary
cementitious materials at different replacement levels.
Efficiency Factor of all the mixes as a function of age cured
at 25 deg C and 40 deg C is shown in Figure 7(a) and 7(b). Fig. 7: Efficiency Factor of base line mix and QL actiavated fly ash
The efficiency of cement fly ash blended system clearly mixes cured at (a) 25 deg C and (b) 40 deg C with age.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

42 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete using Siliceous Fly ash at very High Levels of Cement Replacement: Influence of Lime Content and Temperature

cured at 25 deg C and 40 deg C with different ages are

shown in Table 4. It can be seen that the centers of the
2 peaks and the FWHM values are relatively invariant
with time and the values are consistent at the both curing
temperatures. A1 and A 2 however continiously change with
age due to the relative decrease the amorphous Si/Al in fly
ash and the relative increase in the content of amorphous
products of hydration.

Table 4. FWHM and peak Positions of the base line fly ash mix
cured at 25 deg C and 40 deg C

Fig. 8: Pawley fit and Deconvolution of the hydrated base line mix Hydrated Amorphous
Si/Al content Area (A1)
Area (A 2)
(F mix) paste cured at 25 deg C at the age of 90 days. Mix–Temp-Age
Center
FWHM Position-2 FWHM
Position-1
for quantitative phase determination using this approach
F-25-1 22.524 7.703 30.010 6.503
was developed for determing the the degree of crystallinity
by (Riello ,2004). This method is extended for amorphous F-25-3 22.445 7.303 29.961 6.803
phase quantification in hydrating fly ash-cement system F-25-7 22.277 7.508 29.956 6.608
where the amorphous phases in both fly ash and F-25-14 22.288 7.561 29.919 6.561
products of hydration contribute to diffuse scattering.
F-25-28 22.280 7.559 29.969 6.559
The quantification of the amorphous phases requires an
accurate intensity profile of the total amorphous phase, F-25-56 22.433 7.492 29.921 6.452
which is then decomposed to determine the individual F-25-90 22.309 7.680 29.782 6.680
contribution from the two components.
F-40-1 22.402 7.317 29.895 6.517
The experimentally observed intensities for the
F-40-3 22.439 7.401 30.086 6.601
amorphous and crystalline phases were determined
using the Pawley method. The Pawley method models F-40-7 22.418 7.576 29.855 6.562
each individual intensity peak of the pattern using the F-40-14 22.353 7.510 29.764 6.510
same factors of the Rietveld method. The intensities are F-40-28 22.316 7.540 29.941 6.540
refined without the crystal structure. For the Pawley
F-40-56 22.334 7.585 29.874 6.485
intensity refinement, known crystalline components
were defined using space groups as hkl phases. Arbitrary F-40-90 22.222 7.684 29.940 6.584
space group and lattice parameters were selected for the
broad hump, which were refined in the analysis for best The amorphous Si/Al content (A1) of the total area under
fit. The decomposed intensity profiles for the amorphous the diffctogram as a function of age for samples cured
and crystalline components from the Pawley fit for the at 25 deg C and 40 deg C are shown in Figures 9(a) and
hydrating baseline fly ash-cement sample at 90 days of 9(b). Results indicate that initial amorphous Si/Al content
depends on the replacement level of fly ash with QL. In QL
age is shown in Figure 8.
systems, fly ash was substituted by an equal mass of QL.
The decomposition of the intensity pattern of the entire Therefore the reduction of fly ash content in QL systems
amorphous phase, deconvoluted using pseudo Voigt (PV) decreases the initial amorphous Si/Al content. The results
peaks. A peak fit algorithm, which uses unconstrained, clearlyindicate the differences in the rate of reduction
non-linear optimization, was used to decompose the broad of amorphous Si/Al in fly ash produced by dissolution.
amorphous intenstiy pattern obtained from the Pawley The rate of depletion is significantly higher at 40 deg C,
refinement in the 2θ region 15o-38o into its component particularly in the early ages. At 25 deg C, after some early
pseudo voigt functions. Typical deconvolution of the decrease in the amorphous Si/Al content, Si/Al dissolution
amorphous hump of the base line fly ash cement paste starts after 7 days of age and it steadly decreases up to 90
sample at the age of 90 days is shown in Figure 8. The PV days. At 40 deg C dissolution starts at 1 day and it steadly
peak (A1) at the smaller 2θ corresponds to the amorphous decrese till 28 days following which the rate of decrease is
Si/Al content in fly ash and the PV peak at the larger 2θ(A 2) very slow. The total extent of fly ash dissolution at both 25
corresponds to the products of hydration. The center of deg C and 40 deg C appears to be similar at 90 days of age.
the first PV peak was consistently found between 2θ angle The trend in the dissolution of Silica and Alumina from fly
of 22-22.5 and the other one was centered around 2-theta ash is not influenced by the QL dosage. The results of the
angles ranging between 29.5 to 30 degrees. The 2-theta quantitative XRD indicate that the dissolution of fly ash
of the centers and the full width half maximum (FWHM) is strongly influenced by temperature although the final
values of the PV peaks for A1 and A 2 for the base line mixes extent of dissolution is not dependent on the temperature.

Organised by
India Chapter of American Concrete Institute 43
Session 1 A - Paper 2

Fig. 9: Si/Al content of QL activated fly ash mix cured at (a) 25 Fig. 10: Hydrated amorphus product in base line (F mix) and QL
deg C and (b) 40 deg C with age. activated pastes cured at (a) 25 deg C and (b) 40 deg C with age.

The lime content in the system does not appear to influence Discusssion
the dissolution of fly ash. Further, at both temperatures, The information from the different experimental
there is still residual amorphous Si/Al in the fly ash, which techniques can now be combined to arrive at an
is not completely dissolved. understanding of reactions in binary mix of cement with
The fraction of the amorphous hydration product content high level of fly ash replacement. The dissolution of Si/Al
(A 2) of the total area under the diffractogram as a from fly ash in the early ages is not influenced by the lime
function of age for cured at 25 deg C and 40 deg C are content of the mix. Initially, the lime provided by cement
hydration is sufficient to support the pozzolanic reaction.
shown in Figures 10(a) and 10(b). Results clearly indicate
The excess lime available in the QL systems does not
the QL activated mixes at 40 deg C showed a significant
appear to influence the pozzolanic reaction as indicated by
high early increase in the amorphous hydration product
the similar quantity of amorphous products formed. This
content than the corresponding mixes cured at 25 deg C. is also supported by the identical efficiency factors up to 3
After 3 days of age, there are also noticeable difference days for fly ash mixtures with and without QL.
in the amorphous product formation in the QL mixes
cured at 40 deg C when compared with the baseline The strength-based efficiency factor of fly ash relates
mix. There are differences in the base line mix at early directly with the decrease in amorphous Si/Al content from
and lateral ages. At 25 deg C, differences between the fly ash resulting from dissolution. In baseline systems, the
early increase in the efficiency of fly ash correlates with
baseline and QL mixes are noticeable after 7 days of age.
a larger decrease in amorphous Si/Al content of fly ash.
At 25 deg C the amorphous hydration product content
To sustain the pozzolanic reaction, sufficient lime should
increases steadily up to 90 days and the final contents in
be available in the system. The lime made available
the baseline and QL mixes are nominally similar to the from the cement may not be sufficient for reacting with
values in the corresponding mixes at 40 deg C. At 40 deg the available silica at high level of cement replacement
C, the amorphous product content increases very rapidly with fly ash. Considering the cement and fly ash used
in the first 28 days following which it starts approaching in this study, cement contributed 15.76% Silica while fly
an symptotic valuee. The 5% and 10% QL mixes showing ash contains 20.2% reactive Silica. Therefore cement
very little difference in quantity of product at both replacement with fly ash results in larger silica content in
temperatures.The amorphous product content of all the system. However, since the fly ash contains very little lime,
mixes very well compares with strength data. the replacement of cement by fly ash results in a decrease

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

44 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete using Siliceous Fly ash at very High Levels of Cement Replacement: Influence of Lime Content and Temperature

in the lime available to support the pozzolanic reaction. At less detrimental on the long-term strength produced by
70% replacement of cement, the lime produced by cement fly ash. This has also been obseved previously (Maltais and
hydration is clearly insufficient to react with the available Marchand,1997).
reactive silica.
The strength-based efficiency factor is a good indicator
Addition of QL resulted in an increase in the CH content of the reactive potential of fly ash in the blended system
in the system which allows the pozzolanic reaction to and it compares very well with the amorphous hydration
continue, while it does not influence the dissolution of product and the reduction of the amorphous Si/Al content
fly ash. In the presence of excess CH, the increase in in fly ash. The level of cement replacement to achieve
efficiency factor is directly related to the decrease in required strength depends on the efficiency factor. There
amorphous Si/Al in fly ash. Higher level of dissolution appears to be an optimal replacement level of fly ash for
at 40 deg C results in enhanced efficiency factor. The cement considering silica replacement, lime availability
reduction in the efficiency factor at any age for 25 deg C and the silica available from fly ash dissolution. The
compared with 40 deg C is related to the lower level of fly optimal quantity of quicklime can be established based on
ash dissolution. In the lime rich system, the dissolution of the available silica to maximize the use of reactive silica.
fly ash which contributes reactive Silica to the pozzolanic
reaction is the rate controlling step. Findings and Conclusions
There is an enhancement in the quantity of amorphous The findings of this investigation pertaining to very high
hydrated reaction products in the QL systems, which is levels of fly ash replacement, 60% by mass of cement and
very evident after 7 days of age. This also corresponds with higher, can be summarized as below.
the increase in the efficiency factor in the QL systems after 1. The primary indication is that the strength gain is
7 days of age. However, increasing the quick lime dosage associated with the formation of an amorphous hydrate
to 10% did not produce further improvement compared product and is linked with the rate of depletion of CH in
to 5% quick lime dosage in both amorphous product and the system.
efficiency factor. From the results of quanitiative XRD
anslysis, it is evident that there is excess CH in the both QL 2. The rate of strength gain is directly influenced by the
systems at 90 days of age. There is however little change consumption of CH, which depends on the reactive
in the Si/Al content after 28 days of age, indicating no silica in the system. Increasing the CH content in the
additional reactive Silica is available for reaction mix does not appear to influence the release of silica
into the mix at the levels of replacement, 60% by mass
In fly ash cement blended system,the reactive silica content of cement and higher.
depend on the level of replacement of cement with fly ash.
At 5% and 10% quicklime additions, there is a reduction in The findings presented here indicate the possibility of
the silica contributed by fly ash due to a decrease in fly ash producing high volume fly ash concrete with a target
content to 65% and 60%, respectively. This suggests that strength using quick lime activation. Commercial
at 10% quicklime addition, the reduction in reactive silica products can be developed, where the quick lime required
contributed by the fly ash results in an overall excess is optimized for a given fly ash. The results presented were
unreacted Ca(OH)2 in the system. At 5% QL addition, while conducted using a specific portland cement and a specific
there is a reduction in the silica content compared to the fly ash. The degree to which the concrete mixture will
cement fly ash mixture, the available lime is sufficient to require modification, or whether the specified properties
support the pozzolanic reaction. can be achieved at all, will depend upon the specific
characteristics of the fly ash being employed. Following
The highest efficiency factor achieved in the fly ash mix points require careful consideration
was 70%. The availability of amorphous Si/Al indicated
by the presence of the amorphous hump in the XRD 1. The requirement of quick lime is currently based
spectrum suggests that a portion of amorphous Si/Al on the formation of calcium silicate hydrate as the
is still undissolved in fly ash. Further improvement in primary product of hydration. Fly ash also contains
efficiency factor is possible if all the reactive components reactive Alumina in addition to reactive Silica.
Enhancing the potential of fly ash requires utilizing the
of fly ash utilized. Further, at 90 days, the dissolution
reactive Alumina as well. Further strength gains can
percentages and amorphous hydration product contents
be achieved if the Alumina in the fly ash is activated in
are nominally similar in Ql systems at both temperatures,
generating additional reaction products.
while the the efficiency factor is higher for curing at 40 deg
C. This suggests that the differences in the strength are 2. Cement used in this study contained 15.76% Silica
related to the influence of temperature on the long-term while fly ash contained 20.2% reactive Silica. There is
strength in concrete with cement as the binder. Concrete a higher contribution of Silica when cement is replaced
made with cement has been shown to produce higher with fly ash. Therefore if sufficient lime is available,
long-term strength when cured at lower temperature. more calcium silicate hydrate should be formed.
The results from this study suggest that temperature is However, even with lime available in the system, the

Organised by
India Chapter of American Concrete Institute 45
Session 1 A - Paper 2

strengths of fly ash concrete are lower than pure 6. Bureau of Indian Standards,1999. Specification for fly ash for use
as pozzolana and admixture. IS 3812.
cement system. This suggests efficiency of fly ash is
related significantly with the dissolution of fly ash and 7. Bureau of Indian Standards,2003. Pulverized Fuel Ash-Specification.
IS 3812-part 1.
the silica made available by the fly ash.
8. Bureau of Indian Standards,2013. Ordinary Portland Cement,53
3. Considering the lower density of fly ash relative to Grade-Specifications. IS 12269
cement, for a replacement of cement by mass, fly ash 9. Carino, N,J., 2004. The Maturity Method, in Handbook on
occupies a larger volume than the cement it replaces. Nondestructive Testing of Concrete, Ed. M. Malhotra and N.J.
The effective cement content in a unit volume of binder Carino, CRC Press, ASTM International, 100 Barr Harbor Drive,
is lower than the cement content obtained from mass West Conshohocken, PA 19428.

proportions. Mix design procedures which consider 10. Caijun Shi., 1996. Early Microstructure Development of Activated
Lime fly Ash. Cement and Concrete Research, 26 (9):1351-1359.
the efficiency of fly ash are required. Volumetric-
based replacements and proportioning instead of the 11. Fraay, A.L.A., Bijen, J.M., Dehaan, Y.M., 1989. The reaction of fly ash
in concrete A critical examination. Cement and concrete Research,
conventional mass-based approaches may lead to
19 (2):235-246.
better mix designs considering the large difference in
12. Ma, W., Brown, P.W.,1997. Hydrothermal reactions of fly ash with
specific gravity between cement fly ashes. Ca(OH)2 and CaSO4.2H2O. Cement and Concrete Research,
References 27(8):1237-1248.
1. ACI 232.2R-96, 2002. Use of fly ash in concrete, ACI report. 13. Maltais, Y., Marchand, J., 1997. Influence of Curing Temperature on
cement Hydration and Mechanical Strength Development of Fly ash
2. Antiohos, S.K., Tsimas, S., 2004. Activation of fly ash cementitious
Mortar. Cement and Concrete Research, 27(7):1009-1020.
systems in the presence of quick lime (part-1). Compressive strength
and pozzolanic reaction rate. Cement and concrete Research, 14. Riello, P., 2004. Quantitative Analysis of Amorphous Fraction in
34:769-779. the Study of the Microstructure of Semi-crystalline Materials,
in Diffraction Analysis of the Microstructure of Materials, E. J.
3. Antiohos, S.K., Papageorgiou, A., Papadakis, V.G.,Tsimas, S., 2007.
Mittemeijer and P. Scardi, Editors, Springer Verlag.
Influence of quicklime addition on the mechanical properties and
hydration degree of blended cements containing different fly ashes. 15. Shi, C., Day, R.L., 1993. Acceleration of strength gain of lime-natural
Construction and Building Materials, 22:1191-1200. pozzolan cements by thermal activation. Cement and concrete
Research, 23 (4):824– 832.
4. Bouzoubaa, N., Zhang, M.H., Bilodeau, A., Malhotra, V.M., 1997.
The effect of grinding on the physical properties of fly ashes and 16. Shi, C., Day, R.L., 1995. Acceleration of the reactivity of fly ash by
a Portland cement clinker. Cement and concrete Research, 27 chemical activation. Cement and concrete Research, 25 (1):15–21.
(12):1861–1874.
17. Zhang,Ya,Mei., Sun, Wei., Yan, Han, Dong., 2000. Hydration of high-
5. Bureau of Indian Standards,1999. Methods of Test For Pozzolanic volume Fly ash cement pastes. Cement and Concrete Composites,
Materials. IS 1727. 22: 445-452.

G.V.P. Bhagath Singh

G.V.P. Bhagath Singh received his M.Tech.in Civil Engineering from IIT Hyderabad and is presently a research
scholar in the same institute. His research interests are materials characterization, micro structure of
materials, X-Ray Diffraction with Rietveld quantification, Supplementary cementitious materials and
geopolymers.

K.V.L. Subramaniam
Prof. K.V. L. Subramaniam is currently the Dean (Planning and Development) and a Professor in the
Department of Civil Engineering at Indian Institute of Technology Hyderabad (IITH). Prior to joining IITH, he
was a Professor and Catell Fellow in Department of Civil Engineering at the Grove School of Engineering,
the City College of New York (CCNY). Dr. Subramaniam obtained a B.Tech. in Civil Engineering from IIT
Delhi and Ph.D. in Structural Engineering and Materials from Northwestern University, Evanston. After
graduation, Dr. Subramaniam worked as a Research Associate at the NSF Center for Advanced Cement
Based Materials. Dr. Subramaniam’s research interests include Fracture and Fatigue of cementitious
materials, FRP-based Structural Strengthening, Fly ash based binders, Geopolymers and Non-destructive
testing and evaluation techniques for concrete structures.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

46 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Do crystalline water proofing admixtures affect restrained plastic shrinkage behavior of concrete?

Do crystalline water proofing admixtures affect restrained plastic

shrinkage behavior of concrete?

Rishi Gupta Alireza Biparva

Assistant Professor, Civil Engineering Research and Development
Program, Department of Mechanical Department, Kryton
Engineering, University of Victoria, International Inc.,
B.C., Victoria, B.C., Canada Vancouver, B.C., Canada

Abstract materials) in concrete. A complementary approach to

make concrete a sustainable material is to improve its
Durability of concrete structures is greatly dependant
service life by improving its durability. One of the key
on concrete’s permeability. Amongst various available
parameters that affects durability of concrete structures
techniques, addition of crystalline water proofing
is its permeability. Various commercial methods available
admixtures is one method used to decrease the permeability
to decrease permeability of concrete have been previously
of concrete. Most of these admixtures are added during
discussed by the authors (Biparva & Gupta, 2010). Even
the mixing stage and some are known to remain passive
though the main motivation of such crystalline admixtures
in concrete until concrete cracks and moisture comes in
is to make concrete less permeable over time, it is well
contact with the passive admixture. Even though the key
known that these admixtures also modify the early-age
function of the water proofing admixtures is to increase
properties of concrete. In this previously published work,
the water tightness of concrete during its long service life,
the authors have discussed the various advantages of
these admixtures can also have some positive effects on
using a hydrophilic crystalline water proofing admixtures
the early-age behavior of concrete. This paper describes
in concrete over other methods to make concrete
the effect of crystalline water proofing admixtures on
water proof. They mention that by using crystalline
early-age cracking in concrete. The performance of
waterproofing not only can concrete permeability be
three different types of these admixtures was compared
reduced, but also other properties such as self-sealing
to that of control. This study has been performed in two
and shrinkage will be affected. Some results are
stages. Stage one was performed under ASTM specified
summarized in this published paper, but a more focused
conditions and a modified stage where more severe
research is required to investigate the effects of different
drying conditions than that described in the ASTM test
integral waterproofing admixtures on key properties
standard were used. These modified conditions simulated
such as self-sealing and shrinkage. Most of the available
inadequate curing under extreme exposure conditions as
literature only describes the effects of these admixtures
experienced by concrete in many parts of the world. The
on the permeability of concrete (Geetha & Perumal, 2011),
test results indicate that the water proofing admixtures
but key properties such as self-sealing and shrinkage are
can effectively reduce the early-age shrinkage cracking.
still not understood.
The possible reasons for this secondary advantage of
crystalline water proofing admixture is also hypothesized The effect of various admixtures, mineral additives, and
in this paper. fibers on restrained plastic shrinkage has been studied
by many researchers. This includes studying the effect of
Keywords: Restrained plastic shrinkage cracking,
shrinkage reducing admixtures (Weiss and Shah, 2002;
crystalline water proofing admixtures, crack reduction
Lura et al, 2007; Bentur et al, 2001), silica fume (Bloom
ratio, time to first crack.
Bentur, 1995), limestone (Corinaldesi & Moriconi, 2009),
fly ash (Gupta et al 2010), and fibers (Soroushian et al,
Introduction and Research Significance 1992; Soroushian & Ravanbakhsh, 1998; Corinaldesi &
The demand for cement around the world has continued Moriconi, 2009; Gupta et al, 2010) on restrained plastic
to be strong over the last decade even though there have shrinkage of concrete. However, a literature search by the
been major concerns about the CO2 emissions associated authors to identify the effect of crystalline waterproofing
with its production. admixtures on restrained plastic shrinkage using the
ASTM C1579 test method did not result in any articles.
To make concrete more sustainable, recent measures Moreover, no previous studies reporting the effect of these
being taken include use of limestone blended cement water proofing admixtures on drying shrinkage could also
(recently launched in the Canadian market (Holcim, be identified. The research study presented in this paper
2011)), use of alternate fuels (Vaccaro, 2006) and was initiated due to the lack of understanding about the
improved energy management to fire kilns, and use of effect of these water proofing admixtures on restrained
higher amounts of SCMs (supplementary cementing plastic shrinkage potential of concrete. Research was

Organised by
India Chapter of American Concrete Institute 47
Session 1 A - Paper 3

conducted according to ASTM C1579 to study the effect of Testing was done following ASTM C1579 standard in
three different types of admixtures on shrinkage cracking. terms of materials, molds and specified environmental
The test conditions specified in the test standard were conditions. However, additional trials using higher
modified to simulate more severe curing conditions that temperatures and lower humidity were conducted to
concrete is exposed to in many parts of the world. simulate conditions more severe than those specified by
the test standard. According to the ASTM standard the
temperature must be 36 ± 3 °C, the relative humidity
Mix Design
30 ± 10% and a minimum wind speed of 4.7 m/s over the
A control mix with target strength commonly specified center of the sample. The measured evaporation rate in
in practice of 40MPa was chosen for this study. To study the environmental chamber was greater than 1.0 kg/m2/
the effect of crystalline based waterproofing admixtures hr which is specified in the test standard.
on plastic shrinkage, three commercially available
products were used to modify the control mix. Dosages Environmental Conditions
recommended by the manufacturers were used to modify
The conditions of the Environmental Chamber were
the control mix. Admixture K, P, and X were added at
regulated using a temperature and relative humidity
2.0%, 0.8%, and 1% of the cement mass respectively in
controller. The temperature and relative humidity were
the control mix. A w/c ratio of 0.55 was chosen for all the
recorded using a HOBOware data logger. A dual temperature
mixes.
and humidity sensor was placed above the center of each
slab. Readings were taken by the logger every 10 seconds
Table 1 and the results were plotted using HOBOware software.
Solutions used for the experiments
The graph above is a screen shot of the data and hence
Ingredient Quantity (kg/m3) does not have very high resolution. This graph is however
Portland cement 340 described below for clarity. Figure 2 shows a conditioning
period of one hour when the temperature and relative
Gravel 1120 humidity are brought up from ambient to standard test
Sand 820 conditions.This is indicated by the first rise in temperature
and drop in humidity. Later, thereis a sharp increase in
Water 187
humidity and decrease in temperature, which indicates
the slab being placed in the chamber. A few minutes after
Test Set-Up placing the specimens, both the temperature and humidity
Molds were used according to ASTM C 1579-06, to induce stabilize within the ASTM specified tolerances. The solid and
a crack along the center of the slab. Slabs were 355 dashed black curves represent the temperature at the top
(± 10) mm x 560 (± 10) mm x 100 (± 5) mm and contained and bottom of the chamber respectively. Similarly, the two
a metal stress riser plate that was bolted to the bottom of blue curves correspond to humidity values at the top and
the mold. The stress risers were used to induce a crack in bottom of the chamber. Since the screen shots presented
the concrete at an early age. For each mix, two specimens here have poor resolution, red lines and green lines have
were tested in the environmental chamber shown in been added to allow reading the upper and lower bound
Figure 1. temperature and humidity values respectively. Testing is
conducted for 24 hours, after which the sample is removed
and its crack size is measured. Once the specimens were
placed in the environmental chamber the temperature was
between 34°C and 36°C throughout the test (red lines),
whereas the humidity was within a tight range of 25-40%.
These values are within the limits allowed by ASTM.

Fig. 1: Instrumented environmental chamber for conducting Fig. 2: Typical screenshots of temperature and humidity vs. time
restrained plastic shrinkage tests (Standard conditions)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

48 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Do crystalline water proofing admixtures affect restrained plastic shrinkage behavior of concrete?

Modified Condition
Table 2
To study the behavior of the mixes tested previously Crack width measurements
under more severe drying conditions, the conditions in
the environmental chamber were modified. As described Average crack width (mm) Maximum crack width (mm)
earlier the solid red and blue lines have been added to Mix Standard Modified Standard Modified
read upper and lower bound temperature and humidity conditions conditions conditions conditions
values respectively during the first 8 hours of the test. The
Control 0.57 0.50 1.3 1.0
dotted lines correspond to the modified conditions in the
chamber after the first 8 hours. K 0.11 0.22 0.6 0.62

As with the standard condition there is a conditioning P 0.50 0.45 1.02 0.98
period before placing the slab and a rebound period after. X 0.47 0.62 1.0 1.2
The modified condition was intended to expose specimens
to more extreme temperature and humidity conditions.
The results in Table 2 clearly indicate that admixture
Conditions specified by the ASTM standard were used for
K significantly reduces the average crack widths and
the first 8 hours to emulate a work day were the concrete
maximum crack widths when compared to control
is monitored and maintained. After the 8 hour period
under both standard and modified conditions. Similarly
the conditions were altered so that the temperature
admixture P also shows some reduction in cracking when
was gradually increased to 46±1°C. This resulted in
compared to control. Even though admixture X showed
the humidity to drop to a 15-27% range within 4 hours
reduction in crack width under standard condition, it was
of these conditions being imposed. This approximately
higher than that measured for control under modified
represented a 30% increase in the average temperature
conditions. This indicates that some water proofing
and a decrease of about 25% in the humidity. Testing
admixtures may be less effective at reducing shrinkage
under these conditios was also conducted for 24 hours,
cracking under hotter and dryer conditions.
after which the cracks were measured.
Crack Reduction Ratio
ASTM C1579-06 specifies that a crack reduction ratio be
calculated using the following formula:

CRR = #1 - Average Crack Width of Control Concrete & # 100%

Average Crack Width of Modified Concrete

Fig. 3: Typical screenshots of temperature and humidity vs. time

(modified conditions)
(a) Standard environmental conditions (b) Modified environmental conditions
Time to cracking
Fig. 4: Crack reduction ratio over control
Even though not a requirement of the test standard, the
time to first crack was measured using two HP AutoFocus
The graphs in Figure 4 compare the crack reduction ratio
720i high definition video cameras. Video of the slab
of each admixture when compared to control concrete
during the test was recorded to determine the time of
mixture. It is evident that admixture K is the most effective
cracking. This eliminated the need to manually observe
in reducing the crack width. When admixture K was added
the specimens for cracking.
at a dosage of 2% by mass of cement, the crack reduction
ratio under standard conditions was about 80% and that
Results under modified conditions was approximately 55% (shown
Crack width: After 24 hrs, the test specimens were in Figure 5).
removed from the chamber and the cracks characterized. In comparison, the dosage for admixture X was only half
The crack size was measured using an optical handheld of K, but this resulted in a CRR of only 15% under standard
microscope at intervals of 10 mm along the length of conditions and a negative CRR greater than -20% under
the crack. The average of all the readings for the two modified conditions (meaning increasing the crack width).
specimens are presented in Table 2. Also included in Admixture P at 0.8% by mass of cement had a modest
this table is the maximum crack width recorded in the CRR of about 10% under both environmental conditions.
specimens.

Organised by
India Chapter of American Concrete Institute 49
Session 1 A - Paper 3

Crack Area of approximately 50mins considering both standard and

Based on the crack width modified conditions. The delay in time to first crack due
measurements, crack to admixture P when compared to control was only an
areas were calculated average of 5 mins.
for all the specimens.
The average crack
areas for standard and It is also observed that the time to first crack increased
modified conditions is under modified conditions not just for control but all
shown in Figures 6a specimens by an average of 16 mins. Under modified
and 6b respectively. It is conditions, when the temperature is higher, the rate of
clear from the figures strength gain would be faster. A delay in time to first
that the average crack crack under modified conditions for the control mix
area for the control was and the mix modified with admixture P translated into
Fig. 5: Crack widths of Control higher than 150mm2. narrower cracks and slightly lower total crack area
(C1) when compared with K1 All admixtures reduced (refer to Table 2 and Fig. 6). However, for admixture K
(modified conditions)- indicating the crack area under and X, larger crack areas were recorded under modified
an approximate CRR of 55% conditions specified by conditions. When compared to control, admixture K
ASTM and under modified caused the largest delay in time to first crack and
conditions. The only outlier was X that showed an increase resulted in the highest CRR. Hence, no conclusive
in cracking under modified conditions. At the specified trends can be drawn from the available data and further
dosage, admixture K seems to be the most effective in research is warranted to confirm the relation between
reducing shrinkage induced cracking. time to first crack and cracking. The influence of mixture
proportion and additives such as fly-ash on evaporation
rate has been described in a study by Banthia and Gupta
(Banthia & Gupta, 2009). Future studies should focus
on measuring the evaporation rate directly from the
specimens as opposed to measuring the evaporation
in the environmental chamber. Evaporation of moisture
from the specimens is also a function of bleeding,
hence measuring loss of moisture from the specimens
(a) Standard environmental conditions (b) Modified environmental conditions
directly can help capture the effect of bleeding as well.
Fig. 6: Average crack area (mm2) However, the large specimen size specified in ASTM
poses challenges in accurately measuring the moisture
Crack time and future scope of research loss from the specimens. The reduction in cracking
observed in this study could be due to a faster rate of
The following tables list out the time it took for each slab
strength gain in at early-ages. To confirm this hypothesis
to crack. The time of cracking was observed visually from
future studies will focus on measuring the moisture loss
a video recording.
of the specimens and also measuring the simultaneous
increase in early-age strength gain of the material using
Table 3 dog-bone shape specimens as utilized in a study by Gupta
Measured time to first crack
(Gupta, 2008).
Time to first crack (hrs:mins)
Mix
Standard conditions Modified conditions
Conclusions
The effects of crystalline waterproofing admixtures on
Control 1:21 1:41
restrained plastic shrinkage was examined under two
K 2:15 2:29 environmental conditions, one specified by ASTM C1579
P 1:24 1:49 and the other modified. The introduction of a modified
condition allowed insight on how the admixtures would
X 2:16 2:22 perform in warmer and dryer climates. Three types of
admixtures were used and added in concrete per the
The time to first crack for the control specimens when dosage prescribed by the manufacturers. Under both
compared to the mixes with admixtures is the lowest environmental conditions the samples with admixtures
under both standard and modified conditions. This tended to resist cracking better than a control of the same
indicates that all the admixtures delay in formation of mix proportions. Admixture K had a crack reduction ratio
the first crack. Admxitures K and X delayed the time to of approximately 80% and 55% under the standard and
first crack when compared to the control by an average modified conditions respectively. The marked decrease

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

50 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Do crystalline water proofing admixtures affect restrained plastic shrinkage behavior of concrete?

proves that the admixture K resists plastic shrinkage Qatar: International Conference on Future Concrete.
effectively. The Admixture P maintained a 10% crack 5. Bloom, R, Bentur, A. (1995). Free and restrained shrinkage of normal
reduction ratio in both environmental conditions. Finally and high-strength concretes. ACI Materials Journal, 92(2), 211-217.
the Admixture X showed a small decrease in crack size in 6. Corinaldesi, V., and Moriconi, G. (2009). Effect of Different Fibers
the standard condition. However, it performed poorly in the and Mineral Additions on the Performance of FRSCC. ACI Special
Publication (261).
modified condition with it cracking more than the control.
Overall, commercially available crystalline water proofing 7. Geetha, A,Perumal, P. (2011). Chemical reaction of waterproofing
admixtures on the corrosion behaviour of reinforced cement
admixtures seem to offer the secondary benefit of serving concrete. Asian Journal of Chemistry, 23(11), 5145-5148.
as a shrinkage reducing admixture especially at early
8. Gupta, R. (2008). Development, application and early-age monitoring
age. Even though further research is required to better of fiber-reinforced ‘crack-free’ cement-based overlays.Doctoral
understand this phenomenon, it is hypothesized that the thesis submitted at The University of British Columbia. Vancouver:
admixtures used in this study reduce the evaporation rate Faculty of Graduate Studies.
of concrete retaining more moisture in the mix that leads 9. Gupta, R., Banthia, N., and Dyer, P., (2010). Field Application and
to delayed initiation of cracking, which in turn allows more Monitoring of Crack Resistant Fiber- Reinforced Concrete Overlays.
ACI Special Publication (268).
time for the mix to gain early-age strength.
10. Holcim Canada ready to deliver portland limestone cement. (2011,
Feb 28). Canada NewsWire. Retrieved from http://search.proquest.
Acknowledgements com.ezproxy.library.uvic.ca/docview/854024570?accountid=14846

The authors acknowledge the involvement of various 11. Lura, P., Pease, B., Mazzotta, G., Rajabipour, F., and Weiss, W.
J., (2007) “Influence of Shrinkage- Reducing Admixtures on
students from BCIT involved on this project over the past Evaporation, Settlement, and Plastic Shrinkage Cracking,” American
many years. The assistance of Stevan Gavrilovic, Marie Concrete Institute Materials Journal, Vol. 104, No. 2, pp. 187-194.
Qian, and Amrit Basra is greatly appreciated. 12. Soroushian, P, Mirza, F, Alhozaimy, A. (1995). Plastic shrinkage
cracking of polypropylene fiber- reinforced concrete. ACI Materials
References Journal, 92(5), 553-560.
1. ASTM Standards, A. S. (2007). ASTM C1579. In ASTM, ASTM 13. Soroushian, P,Ravanbakhsh, S. (1998). Control of plastic shrinkage
STANDARDS SECTION 4 CONSTRUCTION (pp. 781-87). Maryland: cracking with specialty cellulose fibers. ACI Materials Journal ,
ASTM C1579. 95(4), 429-435.
2. Banthia, N., & Gupta, R. (2009). Plastic shrinkage cracking in 14. Vaccaro, M., "Burning alternative fuels in rotary cement kilns,
cementitious repairs and overlays. Materials and Structures, 42, "Cement Industry Technical Conference, 2006. Conference
567-579. Record. IEEE , vol., no., pp.10 pp.,, 9-14 April 2006 doi: 10.1109/
CITCON.2006.1635711
3. Bentur, A., Berke, N.S., Dallaire, M.P., Druning, T.A. (2001). Crack
Mitigation Effects of Shrinkage Reducing Admixtures. ACI Special 15. Weiss, W. J. And Shah S. P. (2002). Restrained shrinkage cracking:
Publication (204). the role of shrinkage reducing admixtures and specimen geometry,
Materials and Structures, Volume 35, Number 2, Page 85.
4. Biparva, A., & Gupta, R. (2010). Smart Waterproofing System. Doha,

Dr. Rishi Gupta

Dr. Rishi Gupta is currently an Assistant Professor in the Department of Mechanical Engineering (Civil
Engineering Program) at the University of Victoria. His previous experience includes working as faculty
and program coordinator in the Department of Civil Engineering at the British Columbia Institute of
Technology. He received both a masters and a PhD in Civil Engineering (Materials) from the University of
British Columbia. He has more than 10 years of combined academic and industry experience. His industry
experience includes working as the Director of Research of Octaform Systems Inc in Vancouver and as
an Engineer with Tata Consulting Engineers in Mumbai India. Rishi has published over 45 peer-reviewed
papers and made numerous technical presentations as an invited speaker. He was awarded the young
professional achievement award by the American Concrete Institute in 2011 for scholarly publications,
reviewing articles related to FRC and sustainable concrete, teaching concrete technology, and mentoring
students and foreign trained professionals. His areas of interest include: Development of innovative
cement-based mortars with improved bond properties for masonry structures - mortars with high volume
fly-ash, admixtures and fiber reinforcement; shrinkage of concrete and development of ‘crack-free’
cement composites for base slabs; advanced materials for structures - Hybrid Fiber Reinforced Concrete,
use of supplementing cementing materials in concrete; structural performance of insulated wall systems;
health monitoring of structures and Non Destructive Testing; durability and corrosion studies of reinforced
concrete. Rishi is a past chair of the APEGBC’s Burnaby/New West branch and also serves on the editorial
board of APEGBC. He is the Chair of the international affairs committee of the Canadian Society of Civil
Engineering (CSCE) and also the Treasurer of the Western Region of the CSCE. He is a long standing
member of the American Concrete Institute and serves on many technical committees including ACI 544
and 347. He is also a voting member of several subcommittees of ASTM C 09.

Organised by
India Chapter of American Concrete Institute 51
Session 1 A - Paper 4

Condition Evaluation & Repair of 100+ Years Old Buildings

Ashok Kakade
Concrete Science, Inc., Hayward, CA, USA

Abstract constructed in the year 1910. The building has concrete

foundation and the superstructure of steel columns and
This paper covers case history of two concrete buildings:
beams encased in reinforced concrete.
first building located near ocean and the other inland.
Both the buildings were about 100 years old and built in The objective of the condition evaluation was to assess
two different environmental conditions of California. The the magnitude and extent of corrosion of reinforcement,
paper provides insight into how initially assumed structural structural steel columns and beams, and provide repair
conditions were proven wrong after a thorough evaluation. recommendations.
The paper outlines evaluation methodology and describes 2. Inland Structure
of various destructive and nondestructive test methods.
This single story reinforced concrete building is located
The paper shows how nondestructive test methods can
in the central region of California and several hundred
minimize the destructive testing and overall testing cost.
miles away from the ocean. The building foundation was
The results of testing, conclusions drawn from the test
built on a sandy soil and consisted of perimeter reinforced
results, and repair alternatives are described.
concrete wall footings along with interior isolated column
Keywords: Condition evaluation, Nondestructive testing, footings. This building was built in 1908. The building
Corrosion, High rise building, old structures, Repair, foundation showed severe corrosion of reinforcement and
Marin environment. associated spalling.

Introduction
In order to maintain the anonymity to the projects, the
buildings are here in referred to as 1) Ocean Front building,
and 2) Inland building. The short descriptions of the two
buildings are as follows:
1. Ocean Front Building
This is a 14­-story high residential building located on top
of a hill in San Francisco, California. This architecturally
beautiful building overlooking the Pacific Ocean was

Photograph 2: Elevation of the single story inland structure

The objective of the condition evaluation was to determine

the cause and magnitude of distress due to reinforcement
corrosion and provide repair recommendations.
The details of the condition evaluation and repair
recommendations are described as follows.

Condition Evaluation & Repair

Recommendations
A. Ocean Front Building
Photograph 1: The view from the building shows the proximity The condition evaluation consisted of following scope­of­
to the Pacific Ocean work.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

52 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Condition Evaluation & Repair of 100+ Years Old Buildings

1. Visual observations & documentations the steel was in good condition; however, random minor
corrosion was observed on the beam web as well as
2. Delamination tests: hammer sounding and non­
flange. However, the minor corrosion had not caused
destructive tests (NDT) impact echo tests
loss of significant steel mass. The reinforcing steel bars
3. Carbonation Tests were square in shape and had some rust stains, but
the loss in cross section due to corrosion was minor.
4. Acid Soluble Chloride Tests
Considering the age of the building, this corrosion was
5. Half­Cell Corrosion Tests considered minor.
6. Electrical Continuity Tests
Carbonation & Chloride Content Tests
7. Evaluation of Repair products Carbonation is a chemical reaction between calcium
8. Repair Recommendations hydroxide in the hydrated cement paste of the concrete
and the atmospheric carbon dioxide to form calcium
The summary of field and laboratory testing is as follows: carbonate. Carbonation causes reduction in pH of the
concrete which results in diminished protection for the
Field & Laboratory Tests embedded reinforcing steel. The rate at which carbonation
occurs in concrete depends on the quality of concrete and
Acoustic Impact­Echo Test
the environmental conditions to which the concrete is
In acoustic impact­echo technique the pulse propagates exposed.
through the concrete and is reflected by material defects
such as voids, major crack or delamination. These The wall had 1⁄2­inch thick plaster over the reinforced
reflected waves, or echoes, are monitored by a second concrete elements. The depth of carbonation was about
transducer (receiver) coupled to the surface of the 1­inch from face of the wall. The concrete surrounding the
structure near the pulse source. The transducer output beam and column showed no sign of carbonation. In other
is displayed on a control unit in terms amplitude versus words, the concrete surrounding the beam and column
frequency. Frequencies are used to evaluate the interior had high enough pH to provide protection to the steel
condition of the structure. beam and column from corrosion.
The acoustic impact echo test was conducted at every 1 The three concrete samples obtained from an area
foot interval along approximately 10 feet length of the steel adjacent to the beam and columns were tested for acid
beam. The test did not show internal delamination in the soluble chloride content. The test results showed that the
wall. The test locations showed that the concrete had no concrete had chloride content of 0.08, 0.12, and 0.20 lb
internal delamination. per cubic yard of concrete. A chloride content level of 1 to
1.5 lb per cubic yard of concrete are typically considered
Condition of Steel Beams & Columns as threshold for corrosion initiation. The measured
The steel column was 6 inches wide. The steel beam chloride levels in concrete were well below the corrosion
was about 12 inches deep and the flange thickness was threshold level; therefore, the corrosion observed on the
3/8­inch. The beam flange and web had 2.75 inch and 5 beam was not due to the chlorides in concrete.
inches of clear cover from the face of the exterior wall.
The visual observations at the exploratory openings such Half­
Cell Electrical Corrosion Potential Test &
as the one shown in Photograph 3 showed that generally Electrical Continuity Test
Half­Cell Electrical Corrosion Potential Test
The half­cell corrosion potential test was conducted in
accordance with ASTM C 876. The test method provides
estimation of the electrical corrosion potential for the
purpose of determining the corrosion activity of the
reinforcing steel. When the readings in the test area are
greater than ­350 mV, then there is a greater than 90%
probability that reinforcing steel corrosion is occurring
in that area at the time of measurements.
If the readings are less than ­200mV, then there is a
greater than 90% probability that reinforcing steel
corrosion is not occurring in that area at the time of
measurements. A majority of the readings were under
­200mV indicating minor to no corrosion of reinforcing
steel and steel beams and columns.
Photograph 3: The view from the building shows the proximity
to the Pacific Ocean

Organised by
India Chapter of American Concrete Institute 53
Session 1 A - Paper 4

Photograph 4: The half­cell corrosion potential test showed Photograph 5: Foundation of Inland structure consisted of
low probability of active reinforcement corrosion. And the concrete beams, columns, and isolated footings
pink colored concrete from carbonation test showed that the
concrete pH was high enough to protect the steel beams from
corrosion

Electrical Continuity Test

The electrical continuity tests showed that the steel beam
and the wall reinforcement are electrically connected,
which means the galvanic anodes connected to protect
the reinforcing steel will also provide some protection to
the steel beam and columns.

Evaluation Findings & Repair Recommendations

Considering the age of the building, the corrosion was
considered minor. The nondestructive acoustic impact­
echo tests showed that the concrete wall did not have
internal delamination along the beam. The chloride levels
in concrete were low; therefore, the chances of chloride
induced corrosion were minimal. The concrete surrounding
the steel beam and column was not carbonated; therefore,
the concrete will continue to provide high pH protection to
the steel provided the concrete is not allowed to carbonate
in future.
The repair consisted of following:
1) Remove rust of the steel surfaces and apply a corrosion
inhibiting coating to the exposed steel reinforcement
and steel beams and columns prior to rebuilding the
sections with a high quality polymer modified repair Photograph 6: Typical severe corrosion observed in the column
mortar. Localized anodes were used around the reinforcement
perimeter of the repair areas.
2) The walls were then coated with a penetrating corrosion 2. NDT consisted of Schmidt hammer tests, pulse­echo,
inhibiting sealer prior to painting the building. and acoustic impact echo
3. Corrosion tests consisted of carbonation tests, half­cell
Inland Structure corrosion tests, and acid soluble chloride analysis.
The condition evaluation consisted of destructive and non­
4. Laboratory tests consisted of strength tests on
destructive tests. Following scope­of­work was performed.
concrete cores, petrographic analysis of concrete
1. Visual observations & documentations, RADAR and chemical tests on soils to determine chlorides,
scanning and coring sulfates, pH, and corrosively.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

54 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Condition Evaluation & Repair of 100+ Years Old Buildings

The main beams were 12­inch wide and 36­inch deep and where the sensors are in contact with the concrete and
secondary beams were 12­inch wide and 17­inch deep. the quality of the contact. Since the amplitude spectrum
The longitudinal twisted rebar in the beam were 1 inch of the results indicated no interior flaws, only possible
in diameter with 3⁄4 to 15⁄8 inch clear cover and vertical explanation for so much variation in frequencies is that the
stirrups were 1⁄4­inch in diameter and 1⁄2 clear cover. internal quality of concrete was non­uniform.
The vertical corroded column reinforcement had a clear
cover of 3 inches. The footing was about 46 inches wide Schmidt Hammer Testing
from the column face and 12 inches thick. The footing was The description of the test method is as follows.
constructed in two steps: first step 12 inches wide from
The Schmidt Hammer consists of a spring controlled
the column face and 5­inches thick and the lower part
hammer mass that slides on a plunger within a tubular
about 18 inches wide from the column face and more than
housing. When the plunger is pressed against the
14 inches thick.
concrete surface, it reacts against the force of the spring;
when completely retracted, the spring is automatically
Field Testing released. The hammer impacts against the concrete and
The field testing consisted of using nondestructive pulse the spring controlled mass rebounds, taking the rider with
velocity and acoustic impact echo test methods. The it along a guide scale.
general principles of the test methods and test results are
as follows. The Schmidt Hammer is principally a surface hardness
tester with little apparent theoretical relationship
Pulse Velocity Testing between the strength of concrete and the rebound
In this method, a pair of piezoelectric sensors is placed at number recorded during the testing. The reading can be
opposite ends of the test member. In one of the sensors, used to estimate the compressive strength of the concrete
electronic pulses are generated and the time it takes for using the calibrated chart provided by the instrument
the pulse to propagate through the concrete to the other manufacturer. Generally, ten tests are conducted at each
sensor is measured by a control unit. Knowing the distance test location and the readings are averaged. A better
traveled, propagation velocity is calculated and based on way to use the data is to check if the rebound numbers
the velocity, condition of the concrete is determined. Six are generally similar. Similar rebound numbers indicate
columns were tested using the pulse velocity technique. generally similar concrete quality and strength.

In general, the readings were repeatable and stable. The The estimated compressive strength of columns and
velocities varied between 9000 and 12,000 ft. per second. beams were 2,900 psi and 2,800 psi respectively. The
Most of the concrete sections away from the spalled areas estimated compressive strength of walls and 1st floor slab
showed no internal defect. was estimated to be 3,100 psi and 4,200 psi respectively.

Acoustic Impact­Echo Testing Half­cell Corrosion Potential Testing

In this technique, electrical energy in the form of a short The half­cell testing was performed at following ten locations
pulse is introduced into a member under investigation in an area along the column height, beam length, wall
through a sensor (transmitter). Reflected stress waves surface or slab soffit area of approximately 5 ft by 10 ft. The
are monitored by another sensor (receiver) and are readings showed active corrosion in beams and columns in
transformed into a frequency domain using a Fast Fourier the vicinity of concrete spalling. At all other locations, there
Transform method. The resulting amplitude spectrum was no sign of active corrosion and the half­cell corrosion
is used to determine condition of the concrete. The tests potential substantiated the visual observations.
were conducted at accessible and ‘normal’ looking (no
surface flaws) locations of the walls, beams and columns. Laboratory Tests
A total of 32 cores were drilled from columns, beams,
Testing showed some variation in the frequencies is
and footings to perform laboratory tests. The summary of
expected because of the concrete surface condition
findings from the laboratory testing is as follows.
The laboratory testing consisted of following tests.
1) Acid soluble chloride content
2) pH of concrete
3) Carbonation
4) Core compressive strength
5) Microscopic evaluation of concrete
6) Soil tests for resistivity, pH, chloride, & sulfate,

Organised by
India Chapter of American Concrete Institute 55
Session 1 A - Paper 4

1) Acid Soluble Chloride Analysis fractures in the concrete cores were attributed
to excessive gypsum crystallization dispersed
A total of twenty two powder samples from various
throughout the uppermost 2 inches of concrete.
concrete cores were tested for acid soluble chloride
content test. The test results showed that at most 6) Soil Tests
locations, the chloride content in concrete was much
A total eleven samples of soil were tested for: a) pH,
higher at many locations than the generally assumed
b) Resistivity, c) Chloride, and d) Sulfate. The soil
corrosion initiation threshold of 1 to 1.5 lb/cu yd of
samples were collected adjacent to columns and wall.
concrete. The chloride content ranged from 0.07 to 16
The soil samples were obtained from three depths:
lb/cu yd of concrete.
1) 0 to 12­inch, 2) 12 to 24­inch, and 3) near bottom of
2) Carbonation the footing. The objective was to determine if the soil
characteristics in terms of salt contents varied with
The depth of carbonation was measured on various
depth.
cores. The footings showed about 1­inch depth of
carbonation and columns, beams, and walls showed a) pH
over 4­inch depth of carbonation. The pH of the soil varied from 7.6 to 9.3. These
3) pH of Concrete pH values do not present corrosion problems for
reinforced concrete structures.
The pH of concrete was measured on several core
samples. The pH ranged from 9.5 to 11 with majority of b) Resistivity
the readings were at 9.5. The resistivity measurements on saturated soil
samples varied from 69 to 1100 with an average
4) Core Compressive Strength Tests
of 592. Based upon the resistivity measurements,
A total of 22 cores were tested for the compressive the soil samples were classified as corrosive to
strength of concrete. The average compressive severely corrosive.
strength of columns and beams were 2,750 and 2,880
c) Chlorides
psi. The average compressive strength of footings and
walls were 4,390 and 3,140 psi. The chloride ion content varied from 20 to
2,500 mg/kg with an average of 679. Chloride
ion concentrations greater than 300 mg/kg are
considered corrosive to embedded reinforcing
steel.
d) Sulfate
The sulfate ion content varied from 49 to 6,100 with
an average of 1,009. A sulfate content over 2,000
mg/kg and is determined to be sufficient to damage
concrete due to chemical sulfate attack. So the soil
at some locations was corrosive to concrete due to
high sulfate content.

Findings
Based on the test data, following were the evaluation
findings.
Photograph 7: Microscopic evaluation of concrete showed
presence of gypsum crystals in the concrete cracks 1) The soil is corrosive to severely corrosive from the
resistivity and chloride content stand point of view.
5) Microscopic Examination of Concrete The chloride contents in soil are very high at some
locations. The reinforcement in some columns at
The microscopic examination was conducted on a total and above soil level show severe corrosion. Both the
of five cores. A summary of findings is as follows. concrete as well as soil surrounding these severely
a) The beam and column concrete was porous, but corroded columns had high chloride contents. The soil
did not show sign of deterioration from chemical chlorides have migrated into concrete over the years
sulfate attack. and have caused severe reinforcement corrosion.

b) The cores from the footing showed locally significant 2) Some beams also show high chloride content. It is
weakening of the concrete surface that was possible that the original concrete mix used calcium
attributed to sulfate attack. The sub­ horizontal chloride as an admixture to accelerate cement

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

56 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Condition Evaluation & Repair of 100+ Years Old Buildings

hydration during concrete placement in the cold 4. Rebuild the columns, beams, and footings to specified
weather. The concrete may have been mixed manually dimensions using either high quality shotcrete or cast­
or in smaller machines and the dosage of calcium in­place concrete made with Type V cement or Type II
chloride may not have been consistent. This is a most cement with type F Flyash.
likely explanation for the wide variation of chlorides in
5. Apply two coats of high quality elastomeric paint on all
concrete even above the soil level.
footings, columns, beams, walls, and underside of the
3) The soil also had very high sulfate content at some 1st floor slab.
locations. The footing concrete showed presence
6. Install isolation material such as 2 layers of thick
of sulfate attack in the cement matrix. The high soil
polyethylene sheeting between the soil and the
sulfate content was most likely the cause of sulfate
foundation system.
attack in concrete.
4) Some beams and columns had high chlorides yet
Conceptual Repair Option 2
showed no reinforcement corrosion. It may be that
the corrosion is progressing at a slower rate and In addition to Option 1, apply two coats of epoxy on footings,
may cause damage in future. The 1st floor slab has columns, beams, walls, and underside of the 1st floor slab.
numerous spalls from construction activity such
as drilling holes for utility opening, but not due to Conceptual Repair Option 3
excessive reinforcement corrosion.
In addition to Option 1, install carbon fiber with epoxy resin
5) The concrete is experiencing reduction in pH due to over footings, columns, beams, and apply two coats of
carbonation. This is expected of concrete exposed to epoxy on the underside of the 1st floor slab and walls.
atmosphere for over 100 years. The buried unexposed
Due to budget constraints, the client had chosen Option 1
footing had the least depth of carbonation, ~ 1­inch;
without item 5, elastomeric paint.
whereas, the exposed walls, beams and columns
have concrete carbonated beyond the clear cover to
the reinforcement. The carbonation has reduced the Conclusion
concrete pH to 9.5. At this pH level, the reinforcing Initially, it was assumed that the ocean front building would
steel in concrete is susceptible to corrosion. have severe distress to reinforcing steel due to proximity
6) The combination of carbonated concrete and chlorides to the salt laden air from the ocean; however, a thorough
in concrete will continue to cause reinforcement evaluation proved it wrong. A good quality original
corrosion. construction withstood the harsh marine environment and
protected the building over 100 years. Minimal repair was
necessary to prolong the service life of this ocean front
Repair Recommendations building. Whereas, the inland building away from the harsh
Following three conceptual repair options were presented marine environment showed significant corrosion related
to the client. distresses. This building suffered severe distress due to
poor initial construction and due to prolong exposure to
high salt containing soil. The conclusion that can be drawn
Conceptual Repair Option 1 from these two case studies is that a thorough evaluation
1. Remove spalled concrete to a depth of at least 6­inches. of the building is very essential before finalizing the cause
2. Remove and replace corroded reinforcement. of distress and designing a repair.

3. Install galvanic anodes to protect the steel from further

corrosion.

Organised by
India Chapter of American Concrete Institute 57
Session 1 A - Paper 4

Ashok M. Kakade
Mr. Kakade, a Registered Professional Civil Engineer, working presently as Principal Engineer with Concrete
Science, Inc., USA, has over 30 years of experience in construction, investigation, and rehabilitation of
concrete and masonry structures. He has constructed reinforced concrete and masonry buildings,
retaining walls, foundations, and pavements. He has investigated foundations, pavements, warehouse
floors, swimming pools, marine structures, commercial and industrial buildings, high rise residential
buildings, parking structures, tunnels, dams and others concrete structures.
He has taught a Concrete Technology course to Engineers, Architects, & Contractors at the UC Berkeley
and UC Davis, Engineering Extensions. He has testified in numerous Arbitrations, Mediations, and Superior
Courts as an expert and has worked on developing, testing and modifying concrete mix designs using
various chemicals and admixtures.
Mr. Kakade was trained in Germany in the use of concrete admixtures, sealants, waterproofing, epoxy and
polyurethane coatings, and polymer modified cementitious repair materials. He has conducted tests on
hardened regular weight, light weight, gypsum, colored & polymer based concretes.
Mr. Kakade is trained in cement & concrete chemistry and has worked on projects having distresses due
to workmanship, corrosion of metal, sulfates, cracking, discoloration, concrete moisture related flooring
distresses, alkali-silica reactivity, and other design, material, and construction related defects in concrete,
stucco, and masonry structures.
He is associated with a number of Institutions like East Bay Structural Engineers Association, International
Concrete Repair Institute, ACI, ICRI, ASCE and ASTM.
He has graduated from the University of Bombay in Civil Engineering, and later did his M.S. from South
Dakota School of Mines & Technology, USA in Civil Engineering.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

58 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Geopolymer Cement Concrete - An Emerging Technology for the Delivery of Resilient Highway Infrastructure Solutions

Geopolymer Cement Concrete - An Emerging Technology for the

Delivery of Resilient Highway Infrastructure Solutions
B.J. Magee, A. Wilkinson, D. Woodward, S. Tretsiakova-McNally and Patrick Lemoine
Ulster University, Built Environment Research Institute, Shore Road, Newtownabbey, Co. Antrim, BT37 0QB, Northern Ireland.

Abstract claims and administration. The proportion of total budget

spent on structural maintenance in 2014 was 58% in
The 20th UK Annual Local Authority Road Maintenance
England and 52% in Wales. Similarly in North America,
survey has recently highlighted that, despite local
pavement patching is reported to represent one of the
authorities reporting an increase in overall maintenance
most extensive and expensive pavement maintenance
expenditure, one in six roads in England and Wales are
activities undertaken by highway agencies at all levels
classed as being in poor condition. The estimated cost of
rectifying this situation is £12 billion. As such, there has (McDaniel et al., 2014). Furthermore, despite high levels
never been a more important time to identify resilient and of ongoing financial investment, typical service lives of
cost effective planned/preventative highway maintenance various pavement preservation techniques is reported
solutions. as being only 2-6 years (Wei and Tighe, 2004). As such,
there has never been a more important time to identify
Geopolymer cement concrete is generally regarded as resilient and cost effective planned/preventative highway
an attractive alternative to Portland cement owing to maintenance solutions.
environmental and performance benefits. Reported in this
paper are preliminary findings of research undertaken to In terms of materials used for highway repair, a wide
further interrogate its potential as a high-performance variety of conventional options exists for flexible and rigid
repair material for specific road defects, such as potholes. pavements, including cold and hot asphaltic materials,
Undertaken collaboratively with local geopolymer cement cementitious materials and polymeric materials
producer Banah UK Ltd., metakaolin/alkali silicate-based (McDaniel et al., 2015). In recent times, interest is growing
geopolymer cement was assessed in this capacity. As part in the application of novel geopolymer cements, which are
of a mix optimisation investigation, reported are key fresh generally regarded as attractive alternatives to Portland
and mechanical material properties including setting time, cement owing to considerable environmental and
compressive/flexural strength and impact resistance. performance benefits. It is claimed (Davidovits, 2013), for
Indicative in situ performance, based on findings from instance, that geopolymer cement production can achieve
accelerated road testing, is also discussed. On-going up to 90% CO2 emission reductions relative to Portland
research to investigate composite material behaviour cement production. Improved properties reported
and optimisation of key material properties, such as for geopolymer cement concrete include dimensional
bond, modulus of elasticity and abrasion resistance, is stability (Wallah, 2010; Aurora Construction Materials,
additionally reported. 2014), compressive/flexural strength and resistance to
acids, sulphates (Shi, 2003; Ariffin et al., 2013; Glasby et
Keywords: Highways; Maintenance; Geopolymer cement al., 2014), fire and freezing-thawing cycles (Provis and van
concrete; Durability Deventer, 2009; Abdulkareem et al., 2014). Compressive
and flexural strengths in the ranges 90-125 MPa (Banah,
Introduction 2014; Ambily et al., 2014), for example, are reported.
Alarmingly, one in six roads in England and Wales are Despite these promising findings, however, research
currently classed as being in poor condition (Asphalt into the application of geopolymer cement in highway
Industry Alliance, 2015). The estimated cost of rectifying infrastructure environments is limited. Initial trials into
this situation is £12 billion. In England and Wales alone, its use in light pavement applications have be trialled
2,670,350 potholes were repaired in 2014; an average of by an Australia-based geopolymer cement concrete
around 15,706 repairs per highway authority responsible. manufacturer (Andrews-Phaedonos, 2014). In this work,
With an average reported cost of £57 per pothole, the in-service visual examinations of footpaths, precast
associated repair bill equated to around £144.3 million. walkways, and cycle lanes showed no signs of stress,
While a significant sum of money in its own right, this cracking or other failure types, resulting in the material’s
represents only a fraction of the total costs associated with inclusion within a regional road authority specification
related traffic and resource management, compensation (VicRoads, 2013). While a study undertaken in Thailand

Organised by
India Chapter of American Concrete Institute 59
Session 1 A - Paper 5

(Hawa et al., 2013) reported the potential use of geopolymer potential exploitation of this material for commercial-scale
cement concrete as a rapid road repair solution, the production of geopolymer cement (McIntosh and Soutsos,
material was based on fly ash, palm ash and parawood 2014). The manufacturing process established involves
ash which required heat curing at temperatures around initial calcination of the kaolinitic clay to dehydroxylate the
80°C. No durability testing was carried out as part of this main mineral component. The resultant powder (relative
work, with reported suitability based on compressive and density 2.89), is then activated using a silicate solution of
bond strengths only. an alkali metal (57% by mass solids) formulated to enable
dissolution of aluminosilicates and supply additional
Against this background, reported in this paper are
soluble silica to form a binder matrix with a defined Si
preliminary findings of a research programme aimed at
to Al ratio (McIntosh and Soutsos, 2014). The resultant
optimising geopolymer cement concrete’s application,
two-part (powder and activator) system (banahCEM) was
under ambient curing conditions, as a resilient highway
employed throughout this research.
infrastructure repair solution.
Mixture proportions
Experimental Investigation Research undertaken on behalf of South Carolina
DOT (Rangaraju and Pattnaik, 2008) identified a range
Geopolymer cement used of key material properties and values influencing
While numerous alternative inorganic polymer and alkali- the compatibility of parent pavement structures and
activated cement types are currently being researched subsequent repairs. Key properties reported included
and developed internationally, the focus of this study is the modulus of elasticity, flexural/tensile strength, porosity
application of geopolymer cement based on calcined clay; and dimensional stability. Against this background, and
a technology reported (British Cement Association, 2009; representing the initial stages of a more comprehensive
McLeod, 2005) to show the greatest potential for realistic body of work aimed at optimising and predicting
development and commercialisation. Having experienced performance, nine geopolymer cement concrete mixtures
successive historic volcanic episodes, multiple sources of were initially considered to assess effects of powder,
ferruginous kaolinitic clay exist in Northern Ireland. These activator and water contents on performance. As shown
usually occur in deposits ranging from 10-20 m in depth, in Table 1, ranges considered for each of these variables
many of which have been exposed at existing quarry sites. were 450-550, 300-400 and 50-60 kg/m3 respectively and
Despite its relatively high iron oxide content, this material for each mix, one variable was changed while the other
has been found to offer a good precursor for geopolymeric two remained at the middle content level. Fine aggregate
binders (Davidovits, 2011). Indeed in recent years, a local contents were adjusted in each case to maintain constant
company, Banah UK Ltd, has undertaken research into the volume. Given a lack of harmonised standards, mixing

Table 1
Geopolymer mortar mixture proportions

Mixture proportions (km/m3)

Activator/ powder Water/ powder
Mix no. banahCEM banahCEM
Sand Water ratio ratio
powder activator

Effect of activator content

1 500 300 1545 55 0.60 0.11

2 500 350 1495 55 0.70 0.11

3 500 400 1445 55 0.80 0.11

Effect of binder content

4 450 350 1545 55 0.78 0.12

5 500 350 1495 55 0.70 0.11

6 550 350 1445 55 0.64 0.10

Effect of water content

7 500 350 1500 50 0.70 0.10

8 500 350 1495 55 0.70 0.11

9 500 350 1490 60 0.70 0.12

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

60 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Geopolymer Cement Concrete - An Emerging Technology for the Delivery of Resilient Highway Infrastructure Solutions

was carried out in accordance with guidance provided Compaction was achieved initially by hand using a steel
by Banah UK (2011), which involved using a motorised tamping rod, followed by 20 seconds of compaction using a
table-top mixer to blend the powder and alkaline activator vibrating table. Excess mortar was removed using a hand
initially followed by addition of fine aggregate. trowel and no further surface texturing was applied. The
repaired slab was then covered with a polythene sheet to
Specimen preparation and testing retain moisture. After 24 hours at 20±2 °C, the polythene
sheet was removed and the specimen was stored at this
Material characterisation temperature, uncovered, for a further 6 days before the
To help identity factors potentially impacting ultimate wearing test was carried out. Air curing was selected, as
performance, materials were initially characterised using this method is likely to reflect in-situ curing applications
Fourier transform infrared spectroscopy (FT-IR) and for pothole repair material. The simulated wear test
scanning electron microscopy (SEM). FT-IR spectra were was carried out using an accelerated road test machine
recorded using a Thermo-Nicolet FT-IR, Nexus model in accordance with Appendix H of TRL Report 176 (1997).
470 operated over a 4000–500 cm-1 frequency range in This test involves a pair of loaded (5±0.2 kN), standard
attenuated total reflectance (ATR) mode. Analysis was pneumatic-tyred car wheels revolving so as to repeatedly
undertaken of both reacted geopolymer cement and pass over the surfacing of a series of 275x275x40 mm
the unreacted powder component powder used in its specimens in a turning action. As well as revolving, the
manufacture. Each sample was ground to a fine powder loaded wheels move 160±25 mm laterally across the
using a pestle and mortar to ensure homogeneity before specimens in a cycle taking 1-10 minutes. Undertaken
FT-IR recording. In terms of SEM, low vacuum Hitachi at an ambient temperature of 20±2 °C to replicate slow-
S3200N equipment operated at 25kV was used to analyse speed, high friction traffic loading, specimens in this study
geopolymer cement concrete samples. were exposed to 2,000 wheel-passes (1,000 revolutions)
at a rate of 10 revolutions per minute.
Compressive and flexural strength
For each of the nine mixes considered, 7- and 28-day Skid resistance values (SRV) of the repaired pothole
compressive and 28-day flexural strength was measured. were also assessed before and after application of the
Both 50 mm cube and 40x40x160 mm prism specimens wearing test according to RRL Road Note 27 (1969). This
were cast in steel moulds and wrapped in polythene sheet test involved pre-saturating samples and determining the
to retain moisture and stored at ambient temperature angle through which a slider attached to a pendulum rose
for 24 hours. Specimens were then de- moulded and coming into contact with tests surfaces. Losses of texture
stored at the same ambient temperature until testing was depth and skid resistance values were then calculated as:
carried out in accordance with BS EN 1015-11: 1999 (British (initial value - final value)
Loss = 100 x %
Standards Institute, 1999i). Initial value

Fresh properties
In this limited study, only the optimum mix in terms of
Results And Discussion
compressive/flexural strength was further assessed Material characterisation
for setting time and flow to ensure compliance with
typical pavement repair material requirements. Testing Previous research has identified FT-IR analysis as a
was carried out according to BS EN 196-3: 2005 (British proven technique for characterising geopolymeric
Standards Institute, 1999ii) and BS EN 1015-3: 1999 (British materials (Khale and Chaudhary, 2007; Rees et al.,
Standards Institute, 1999iii) respectively. 2007). For both the unreacted and reacted geopolymer
cement samples, and associated with Si-O-Si or Si-
Pavement wear and skidding resistance O-Al asymmetric stretching vibrations (Khale and
The optimum mix in terms of compressive/flexural Chaudhary, 2007), the most prominent feature identified
strength was additionally assessed for its resistance to was an intensive absorbance peak recorded between
simulated wear when applied as a pothole repair material. wavelengths of 1250 and 800 cm-1. In comparison to the
To achieve this, a pothole was simulated in a 275x275x40 unreacted powder sample where this peak occurred at
mm asphalt sample by removing material using a 1036 cm-1, absorbance of increased intensity at 980 cm-1
hammer and chisel to form a roughly circular defect was recorded for the reacted cement sample. This clearly
with rough, sloped sides and approximate volume of indicated the formation of geopolymeric gel, a trend
0.00104 m3. This defect was designed to satisfy reported reinforced by the presence of an additional absorbance
minimum dimensions of potholes as defined by over 60% signal at 780 cm-1 for the geopolymer cement powder
of local authorities in England and Wales (Asphalt Industry Khale and Chaudhary, 2007).
Alliance, 2015). Shown in Figure 1 are SEM micrographs at differing
The defect was then filled with geopolymer cement resolutions of samples made with plain geopolymer
concrete to the same level as the original slab surface. cement concrete (Figure 1, a-c) and basalt micro fibre-

Organised by
India Chapter of American Concrete Institute 61
Session 1 A - Paper 5

Fig. 1: SEM micrographs of basalt fibre (a-c) and plain geopolymer cement concrete (d-f)

reinforced polymer cement concrete (Figure 1, d-f).

Table 2
While not a focus of this paper, the basalt micro fibre- Mean compressive and flexural strength results
reinforced specimens (added at a rate of 2% by mass of
geopolymer cement powder) form part of an ongoing Flexural
Compressive strength (MPa)
Mix no. strength (MPa)
study to investigate geopolymer cement concrete-based
composites. Clearly, the images shown in indicate a 7-day 28-day 28-day
very dense microstructure for both samples, albeit with 1 59 62 2.7
limited micro cracking. Positively, indicate that the basalt 2 66 67 2.4
micro fibres deflect crack growth, hence toughening the 3 61 59 2.7
material. Importantly in terms of long term structural 4 54 54 2.7
and durability performance, the bond of geopolymer 5 66 67 1.7
cement paste around aggregate and fibre surfaces is
6 69 77 2.3
homogeneous, with no evidence of a defined, low quality
7 69 76 3.1
transition zone.
8 66 67 2.7
Compressive and Flexural Strength 9 58 58 2.4

Mean 7- and 28-day compressive and flexural strength

results for each mix are reported in Table 2 and presented
graphically in Figures 2-4. Clearly from Table 2, the nine
different mixes considered produced a range of 7- and 28-
day compressive (54-69 and 54-77 MPa respectively) and
28-day flexural (1.7-3.1 MPa) strength values.
Across the range of values measured, 28-day flexural
strengths were on average 3.9% of corresponding
compressive values; a relationship typical of conventional
cement-based materials containing fine aggregate only.
As expected, and encouraging in terms of potential future
performance predictions based on compressive strength,
a relatively well-defined (R2=0.68) relationship was noted
between flexure and compression as shown in Figure 2.
Positively, the upper range of flexural strengths measured Fig. 2: Relationship between 28-day compressive and flexural
strength

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

62 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Geopolymer Cement Concrete - An Emerging Technology for the Delivery of Resilient Highway Infrastructure Solutions

exceeded the minimum laboratory-based value of 2.4 to activator content, therefore, this finding clearly confirms
MPa proposed for selecting rapid-setting patch materials a dominant influence of geopolymer binder content in
(McDaniel et al., 2014). terms of ultimate compressive strength.
In terms of compressive strength development with Effect of water content
time, a more inconsistent pattern was observed. From
For mixes 7-9, which possessed constant powder and
Figure 3, for instance, it is clear that 7 to 28 day strength
activator contents of 500 and 350 kg/m3 respectively,
increases (in the range 2-12%) were noted for the majority
the influence of water content is plotted Figure 4(c).
of mixes. Mixes 4 and 9, on the other hand, exhibited no
As with conventional concrete, a significant inverse
strength increase, while mix 3 exhibited a minor strength
relationship (R2=1.0) existed between strength and water
loss (-3%). An explanation for this observation may be that
content. As water contents increased from 50-60 kg/m3
one element of the material proportions for each of these
(corresponding to a water/powder ratio range of 0.10-
mixes was at the limits of those considered. For mixes 3,
0.12), 28-strength values decreased significantly from 76
4 and 9 this was maximum activator content (400 kg/m3),
to 58 MPa.
minimum binder content (450 kg/m3) and maximum water
content (60 kg/m3) respectively, suggesting a negative Optimising performance
impact of these outer limits. This conclusion is further
The mix design influences discussed above are further
analysed in Figure 4.
analysed in Figure 5, which provides plots of 28-day
Effect of activator content compressive versus both activator/powder and water/
powder ratios for all nine mixes considered. Clearly from
Figure 4(a) demonstrates a non-linear relationship
this plot, the influence of water/powder ratio dominated
between compressive strength development and
(R2=0.87) that of activator/powder ratio (R2=0.25) across
BanahCEM activator/powder ratio for mixes 1-3. While
the mixes considered. This is perhaps surprising, but
based on a relatively limited data set, Figure 4(a) suggests
suggests that provided sufficient activator is present in
that an optimum activator/powder ratio in the region of
the system to promote dissolution of Al and Si and supply
0.70 exists. Indeed, as the activator/powder ratio increased
additional soluble silica, ultimate performance, as is
from 0.70 to 0.80, a reduction in 28-day strength from 67
the case with conventional concrete, is driven by water/
(mix 2) to 59 MPa (mix 3) was noted. As mentioned above
powder ratio. In the current study, optimum performance
for mix 3, this value of 28-day strength was additionally
in term of 28-day compressive strength was achieved by
linked to a minor reduction in strength between 7 and 28
mixes 6 and 7 (77 and 76 MPa respectively). While these
days.
mixes had differing activator/powder ratios (0.64 and
Effect of binder content 0.70 respectively), both were prepared with the minimum
water/powder ratio considered (0.10).
Compressive strength data for mixes 4-6 is plotted in
Figure 4(b), which as predicted, shows a significant
(R2=0.99) relationship between strength and geopolymer Fresh Properties
binder content. With constant activator and water contents Based on the optimisation process described above, mix
of 350 and 55 kg/m3 respectively, as powder contents 6 was selected for further testing in relation to key fresh
increased from 450-550 kg/m3 (corresponding to an properties. Mean initial and final setting times recorded
activator/powder ratio range of 0.78-0.64), 28-strength for mix 6 were 100 and 150 minutes respectively. While
values increased significantly from 54 to 77 MPa. Relative suitable for standard mortar applications, it is recognised
that accelerated setting times are typical of road
repair materials. Indeed, the minimum recommended
laboratory-based strength requirement of 20 MPa for
rapid-setting patch materials is required after only 2
hours (McDaniel et al., 2014). While research into the
performance of rapid-set geopolymer cement concrete is
on-going as part of the current study, this falls outside the
scope of this paper.
The mean mortar flow recorded for mix 6 was 143 mm,
representing a 69.9% increase from the lower diameter
of the test mould. Falling within a flow range of 140-200
mm and, therefore, classed as plastic mortar (British
Standards Institution, 2007), this indicated an adequate
level of workability for progressing to the next stage of
research.
Fig. 3: Comparison of 7- and 28-day compressive strength
values

Organised by
India Chapter of American Concrete Institute 63
Session 1 A - Paper 5

Fig. 6: Simulated pot hole before and after repair with geopoly-
mer cement concrete

Pothole Repair Performance

In terms of pavement wear, both visual inspections and
measurements of texture depth were recorded before and
after exposure to 2,000 accelerated wheel passes (see
Figure 6). While performance issues were predicted due
to strength and stiffness incompatibilities between the
parent asphalt and geopolymer repair, no visual surface
cracking, delamination, de-bonding or deterioration was
noted. This finding may, in part, have been influenced by
the relatively small-scale nature of the test specimen
and the high stiffness of the steel mould used. Equally,
no measurable decrease in surface texture was noted
(equating to a classification of E; excellent, no discernible
fault (TRL, 1997)), although some minor shining of the
pothole surface was noted. While the 2,000 wheel exposure
level reported is only 20% of the maximum recommended
for this test (100,000 passes), the early indication from
this testing regime was that geopolymer cement concrete
potentially offers a durable and compatible pavement
repair material.
Similar to the findings for pavement wear, no discernible
reduction in skid performance was measured before
and after trafficking, with an average skid-resistance
value (SRV) of 41 recorded in both instances. While this
Fig. 4: Influence of mix design parameters on compressive is a positive result, it should be noted that this value fails
strength to meet the minimum requirement for use on a public
road (Category C - minimum SRV value of 45) according
to RRL Road Note 27 (Road Research Laboratory, 1969).
Minimum SRV levels recommended for motorways/trunk
roads (category B) and difficult sites such as roundabouts/
bends (category A) are 55 and 65 respectively. As such,
further work is ongoing, via the use of surface texturing
and aggregate selection, to improve initial SRV levels of
geopolymer cement concretes. This work is in line with
the recommended minimum texture depth of 0.65mm for
category A and B roads (Road Research Laboratory, 1969).

Summary and Conclusions

The investigation reported in this paper focussed on
characterisation of geopolymer cement and its application
in nine mixes concrete mixes designed to optimise mix
design in terms of powder/cement and powder/water
ratios. Tested for each mix were 7 and 28 day compressive,
Fig. 5: Influence of mix design summary

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

64 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Geopolymer Cement Concrete - An Emerging Technology for the Delivery of Resilient Highway Infrastructure Solutions

and 28 day flexural, strengths. The optimum mix in this wear and skid resistance performance. Given that
regard was selected for further workability testing and pavement repairs are carried out on a wide range of parent
subjected to simulated wearing in a road repair application. structure types, also merited is optimisation in terms of
A skid-resistance value for the geopolymer surface was properties such as stiffness and bond. As such, composite
additionally recorded. Based on the results reported, the behaviour will be analysed using a variety of fibre and
following general conclusions may be drawn: alternative aggregate types. Secondly, the exploration of
geopolymer cement concrete mixes which enable setting
1. An ability to mix and cure geopolymer cement concrete
time reductions, without compromising other fresh and
in ambient conditions has been confirmed, enabling
mechanical properties, is required to offer a practical
future in-situ applications to be considered.
rapid road repair material alternative. As mentioned
2. For the mix designs considered, compressive and previously, attainment of desired properties, such as
flexural strength values at 28 days ranged from 54-77 compressive/flexural strength, within a 2-hour window
and 1.7-3.1 MPa respectively. This range of mechanical is the norm. Finally, while geopolymer cement concrete
properties was considered appropriate in terms of durability appears of high quality, further research is
future application in highway repair scenarios. required into asphalt-geopolymer mortar bond and
3. The mechanical properties of geopolymer cement surface texturing techniques to ensure acceptable long-
concrete are affected by both activator/powder and term skid-resistance properties. High impact for the
powder/water ratios. Optimum performance in this study will be assured by engaging local road authorities
study was achieved by mix 6, prepared with activator/ and stakeholders in on-going research to additionally
powder and water/powder ratios of 0.64 and 0.10 explore full- scale pavement trials and development of
respectively. related design and specification documentation.

4. For the optimum mix assessed, the initial and final

setting times recorded in this study (100 and 150 Acknowledgements
minutes respectively) were too slow for the material Financial and technical support provided for this research
to be considered as a ‘rapid setting’ solution. However, from the Department for Employment and Learning,
material flow indicated an appropriate level of Northern Ireland and Banah UK Ltd. is gratefully
workability for in-situ repair applications. acknowledged.
5. After exposure to simulated traffic wear in Ulster
References
University’s accelerated road test machine, the
1. Abdulkareem, O.A., Al Bakri, A.M.M., Kamarudin, H. and Khairul
potential durability performance of geopolymer Nizar, I., 2014. Fire Resistance evaluation of lightweight geopolymer
cement concretes in road pavement application concrete system exposed to elevated temperatures of 100-800°C.
appears to be excellent. Minimal surface wear was Key Engineering materials, 594-595 pp. 427-432.
measured and no surface cracking or other surface 2. Ambily, P.S., Ravisankar, K., Umarani, C., Dattatreya, J.K. and Iyer,
deformation was apparent during a visual examination N.R., 2014. Development of ultra-high- performance geopolymer
concrete. Magazine of Concrete Research, 66(2) pp. 82-89.
of the material after test completion.
3. Andrews-Phaedonos, F., 2014. Specification and use of geopolymer
6. While skid-resistance of the geopolymer cement concrete. Melbourne: VicRoads. VicRoads Standard Specification,
concrete repair considered was deemed unsatisfactory 2013. Section 703 – General Concrete Paving. Melbourne: VicRoads.
for general road applications, reductions in 4. Ariffin, M.A.M., Bhutta, M.A.R., Hussin, M.W., Mohd Tahir, M. and
performance after exposure to simulated traffic wear Aziah, N., 2013. Sulfuric acid resistance of blended ash geopolymer
concrete. Construction and Building Materials, 43 pp. 80-86.
were minimal.
5. Asphalt Industry Alliance, ‘Annual Local Authority Road Maintenance
7. The overarching conclusion from this investigation Survey’, AIA Press & Information Office, 2015.
was that geopolymer cement concretes exhibit 6. Aurora Construction Materials (ACM), 2014. E-Crete: Engineering
considerable potential for application in road pavement Properties and Case Studies. [Pdf] Available at: http://www.acm.
applications. com.au/pdf/b69822_d3b9c3173f174bf59e9d3427892ab0c6.pdf
[Accessed 23 January 2014].
Clearly in its infancy as a research programme, further 7. Banah UK, 2011. BanahCEM Cement – Instructions for Mixing.
work is merited to develop a market for geopolymer Antrim: Banah UK. Banah UK, 2014. Introduction to Geopolymer
cement concrete applications in the highway environment. Binders. Ballyclare: Banah UK.
Indeed, based on the conclusions reported in this paper, 8. British Cement Association, ‘Novel cements: Low energy, low
three main areas of future development have been carbon cements’, 2009.
identified. 9. McLeod, R.S., ‘Ordinary Portland Cement with extraordinary high
CO2 emissions,’ Newbuilder, pp.30-33, 2005.
Firstly, further optimisation of geopolymer cement
10. British Standards Institute, ‘BS EN 1015-11: 1999, Determination
concrete mix design is required. In addition to considering of flexural and compressive strength of hardened mortar’. Milton
powder/activator/water ratios, impacts of aggregate type Keynes: BSI, 1999.
and content will also be critical, particularly regarding

Organised by
India Chapter of American Concrete Institute 65
Session 1 A - Paper 5

11. British Standards Institute, ‘BS EN 196-3: 2005, Determination of Board in cooperation with the Federal Highway Administration,
setting times and soundness’. Milton Keynes: BSI, 1999. Transportation research Washington DC, 2014.
12. British Standards Institute, ‘BS EN 1015-3: 1999, Determination of 21. McIntosh JA and Soutsos MN, ‘Development of a Geoplymer Binder
consistence of fresh mortar (by flow table)’. Milton Keynes: BSI, from the Interbasaltic Laterites of Northern Ireland’, Proceedings
1999. of Civil Engineering Research in Ireland conference, Belfast, 28-29
August 2014, p487-492.
13. Davidovits, J., ‘Geopolymer Cement: A Review’. Saint Quentin:
Geopolymer Institute2013. 22. Provis, J.L. and van Deventer, J.S.J., 2009. Geopolymers: Structures,
Processing, Properties and Industrial Applications. Cambridge:
14. Wallah, S.E., 2010. Creep behaviour of fly ash based geopolymer
Woodhead Publishing.
concrete. Civil Engineering Dimension, 12(2) pp.73-78.
23. Rangaraju Rao, P. and R. Pattnaik Ranjan, Evaluation of Rapid Set
15. Davidovits, J., ‘Geopolymer chemistry and applications’, 3rd ed.
Patching Materials for PCC Applications, Report FHWA-SC-07-07,
Saint Quentin, France: Institute Geopolymere, 2011.
Clemson University, Clemson, S.C., 2008.
16. Geopolymer Institute. 2008. Technical Data Sheet, [online]. Available
24. Rees, C. A., Provis, J. L., Lukey, G. C. and van Deventer, J. S. J. In situ
at: http://w w w.geopolymer.org/science/technical-data-sheet
ATR-FTIR study of early stages of fly ash geopolymer gel formation.
[Accessed 24 October 2014].
Langmuir, 2007, 23, pp. 9076-9082.
17. Glasby, T., Day, J., Kemp, M. and Aldred, J., 2014. Geopolymer
25. British Standards Institution, ‘BS EN 1015-6:1999, Methods of test
Concrete for Durable Linings. [Online] Available at: http://www.
for mortar for masonry – Part 6: Determination of bulk density of
tunneltalk.com/TunnelTECH-Jan2014 [Accessed 22 January 2015].
fresh mortar’, Milton Keynes: BSI, 2007.
18. Hawa, A., Tonnayopas, D., Prachasaree, W. and Taneerananon, P.,
26. Road Research Laboratory, ‘Road Note 27 – Instructions for using
2013. Development and Performance Evaluation of Very High Early
the portable skid-resistance tester’. London: HMSO, 1969.
Strength Geopolymer for Rapid Road Repair. Advances in Materials
Science and Engineering. 2013 pp. 1-9. 27. Shi, C., 2003. Corrosion Resistance of alkali-activated Slag cement.
Advances in Cement Research, 15(2) pp. 77-81.
19. Khale, D. and Chaudhary, R. Mechanism of geopolymerisation and
factors influencing its development: a review. Journal of Material 28. Transportation Research Laboratory, ‘Report 176 Laboratory Tests
Science, 2007, 42, pp. 729-746. On High-Friction Surfaces for Highways’. Berkshire: TRL, 1997.
20. McDaniel RS, Olek J, Magee BJ, Behnood A, and Pollock R, ‘NCHRP 29. Wei, C. and S. Tighe, ‘Development of Preventive Maintenance
SYNTHESIS 463 - Pavement Patching Practices: A synthesis of Decision Trees Based on Cost-Effectiveness Analysis: An Ontario
Highway Practice’, sponsored by American Association of State Case Study,” Transportation Research Record: Journal of the
Highway and Transportation Officials Transportation Research Transportation Research Board, No. 1866, Transportation Research
Board of the National Academies, Washington, D.C., 2004, pp. 9–19.

Dr Bryan Magee
Dr Bryan Magee is a Lecturer in Construction Materials within Ulster University’s School of the Built
Environment. A chartered civil engineer, Bryan has worked as an educationalist, researcher and policy-
maker in the field of civil/structural engineering for over 15 years. Bryan’s previous career includes
academic posts at the University of Cape Town, Purdue University, University of New Hampshire and
Queen’s University Belfast, as well as UK-based industry appointments working for TRL Ltd. and The
Concrete Centre. His specialist area of interest is the behaviour, performance and advancement of civil/
structural materials.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

66 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 1 B
Session 1 B - Paper 1

Sustainable Concrete as a Platform for Outreach

Roger P. West, Ahmed Alawais Naveen Kwatra

Department of Civil, Structural and Department of Civil Engineering,
Environmental Engineering, Trinity College, Thapar University, Patiala,
University of Dublin, Dublin, Ireland Punjab, India

Abstract In particular, 85% of the Portland cement clinker was

This paper will describe a recent project to present the replaced by the supplementary cementitious material
best of concrete sustainability to school students as an ground granulated blastfurnace slag (GGBS), the course
inducement to improve the image of the engineering and fine aggregates by a combination of recycled tyres
profession. A concrete mix with 85% GGBS binder, (crumb rubber), recycled concrete aggregate and waste
aggregates comprising waste quarry dust, recycled from quarries (in the
concrete and recycled tyres (in the form of crumb rubber), form of grit and quarry
harvested rainwater and crushed hemp as an anti- (limestone) dust),
cracking fibre was designed and tested for strength in a with 100% harvest
laboratory. The means by which the subsequent strength rainwater replacing
loss could be restored are explored, briefly. The most tap water. As no
sustainable mix was used in a public demonstration as artificial admixtures
part of an Institution of Engineers of Ireland outreach were used, inevitably
programme into schools. A local competition was used the workability
to identify those students who would be allowed to and compressive
leave their hand prints in a thin concrete slab which was strength properties
subsequently carefully cured to ensure its strength and were detrimentally
durability to allow it to be suspended on view in public in affected, but in this
perpetuity. instance, that was
not critical for the Fig. 1: Mixing one of the most sus-
Keywords: Crumb rubber; Ground granulated
purpose in hand. tainable concretes in the world in a
blastfurnace slag; Hemp; Rainwater harvesting; Recycled
When the concrete school classroom
aggregate.
was mixed, four
children were selected from a participative audience of 60
Introduction (by setting them a structural tower competition using only
Education of the population to improve awareness of newspaper and sellotape) to impress their handprints in
sustainability matters does not begin in universities, the fresh concrete which was deemed to be one of the
but right back in primary and secondary schools. most sustainable ever made.
An appreciation of the importance of nurturing the
This paper will describe the composition and mechanical
environment has been successfully inculcated in the
properties of different concretes with various combinations
Australian school system for decades so that its citizens,
of these waste materials, to derive an awareness of
by and large, understand the fragility of the planet. In
the most significant contributors to the severe effects
Ireland, the Institution of Engineers of Ireland have a well-
of these materials on concrete compressive strength.
developed system for promoting the profession in schools
Furthermore, descriptions of potential remedies to
and, once a year, in “Engineer’s Week”, practitioner and
restore performance will be proposed, which is the
academic volunteer engineers infiltrate the schools to
subject of future work.
explain the role of the engineer in daily life.
The lead author in this paper took the theme of concrete
sustainability to address, through a highly interactive half Background to Waste Utilisation
day of making concrete (Figure 1), the matter of how it is Ground granulated blast furnace slag (GGBS) has been
technically feasible to manufacture a concrete which uses used globally as a supplementary cementitious material
97.5% recycled materials, based on previous research on (SCM) to improve the rate of the heat of hydration and
a number of specific research themes. durability potential of concrete, depending on its physical
characteristics and levels of replacement [Githachuri

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

68 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainable Concrete as a Platform for Outreach

and Alexander, 2013 and McNally and Sheils, 2012]. The RWH as a sustainable strategy [Farreny et al., 2011b].
substitution of GGBS for Portland cement (PC) clinker leads
Since the beginning of the nineties, research has
to reduced CO2 emissions, owing to GGBS’s production
developed into hemp (a naturally growing fibrous crop)
process which emits only about 60 kg of CO2/tonne,
mixed with a binder as a new building material. It led to
primarily from the grinding process, compared to at least
more sustainable buildings and this allows a significant
750kg/tonne CO2 for PC clinker [O’Rourke et al., 2009].
reduction on the carbon footprint of materials of existing
Although the usage of GGBS influences the mechanical
buildings because the concrete stores approximately 35
characteristics of concrete, such as compressive strength
kg of CO2 per square meter of wall (built with a thickness
development, modulus of elasticity and modulus of
of 25 cm) over 100 years [Arnaud and Gourlay, 2012].
rupture [Shariq et al., 2013], yet it is documented that the
Another study conducted concluded that a hemp concrete
strength of the concrete modulates at a later age (usually
wall allows carbon sequestration of 83 kg CO2/m2 of wall
after 56 days), matching the compressive strength of PC
[Pretot et al., 2014]. Hemp fibre has a strong tolerance for
concrete [Teng et al., 2013].
an alkali environment and a high tensile strength and that
At present, recycled concrete gathered from both makes it an excellent alternative reinforcement material
reinforced and pre-stressed demolished structures for non-load bearing applications [Zhijian et al., 2006].
is a reliable source of aggregates with comparatively
good quality [Ajdukiewicz and Kliszczewicz, 2002]. What
Mix Composition
differentiates recycled concrete aggregate (RCA) from
natural aggregates (NA) is the adhered mortar on its In the present study, various combinations of cement,
surfaces. Consequently, RCA has lower bulk density conventional crushed aggregates, GGBS, crumb rubber,
and lower crushing strength than NA, and this explains recycled concrete, processed grit and quarry powder
the decrease in the compressive strength, density and have been used, where the baseline mix, made from
modulus of elasticity of the RCA concrete [Bogas et al., conventional materials, is a C28/35 concrete (that is,
2014]. Researchers have been using two fractions of having a characteristic cylinder/cube crushing strength
RCA: coarse recycled concrete aggregates (CRCA) for of 28/35MPa respectively at 28 days). The particle size
the replacement of gravel and fine recycled concrete distributions of these materials are presented in Figure
aggregates (FRCA) for the replacement of natural sand 2. In this, the substitution of the 10 and 20mm aggregate
[Cartuxo, 2015]. One very clear benefit from using recycled by crumb rubber can be deduced as being feasible, but
aggregate is a direct reduction in the volume of demolition the grit and recycled aggregate are coarser than the
materials sent to landfill [Silva et al., 2014]. sand, hence quarry dust is also needed to substitute the
fine fraction and make the overall mix less harsh, more
On the other hand, studies have been undertaken with the cohesive and more finishable.
purpose of developing techniques that combine concrete
technology with recycled tyres in the form of crumb
rubber [Albano et al., 2005]. Concrete’s limited properties,
such as its stiffness and capacity to absorb energy, can be
enhanced by aggregating rubber to increase concrete’s
impact resistance. Research was carried out on the
application of crumb rubber in concrete as a substitute for
coarse aggregate or sand. To a certain extent, the studies
revealed that the addition of crumb rubber to concrete
could improve the elastic behaviour and sound insulation,
but reduced the compressive strength [Al-Tayeb et al., Fig. 2: Particle size distribution of material used
2013 and Yung et al., 2013]. Nevertheless, suggestions
were made to control the loss in strength; for example, Four different mixes have been cast (Table 1). Mix 1 is the
it might be minimized by a prior surface treatment of the control mix having water-cement ratio approximately 0.5
rubber particles used in the concrete [Yung et al., 2013]. with crushed limestone 20 and 10 mm coarse aggregates
Due to the urbanization contribution to increasing surface used in 2:1 ratio. In Mix 2, 85% of the CEM II A-L cement
runoff flooding, rainwater harvesting (RWH) is one of the has been replaced with GGBS and rest of the ingredients
best available methods for establishing sustainable water having been kept the same as Mix 1. With this high GGBS
cycles in urban developments [Lee et al., 2012]. Since content, it is expected that early age strength will drop
approximately half of the total sealed surfaces in cities significantly. In Mix 3, 85% of cement is again replaced
comprises roofs, they contribute most to the important with GGBS and all coarse aggregates has been replaced
urban storm water runoff flow. Therefore, they provide an with crumb rubber. The weight of coarse aggregates,
excellent possibility for RWH [Farreny et al., 2011a]. Urban after performing volume mapping of limestone to rubber,
water cycle management is looking towards including reduces to 365 kg/m3. In Mix 4, all by-products have been
used, including natural sand being replaced by recycled

Organised by
India Chapter of American Concrete Institute 69
Session 1 B - Paper 1

Table 1
Details of Concrete Mixes (kg/m3)

Mix Cement sSand 10 mm Aggt. 20 mm Aggt. GGBS Crumb Rubber Recycled Concrete Proc’d Grit Quarry Powder H20

1 395 590 400 800 - - - - - 200

2 55 590 400 800 340 - - - - 200

3 55 590 - - 340 365 - - - 200

4 55 - - -- 340 225 280 280 25 200

concrete aggregate and processed grit. Some quarry crumb rubber substitutes for all the coarse aggregate,
powder has also been added to improve cohesiveness and in Mix 3, not surprisingly the strength drops even more
add somewhat to early age strength. substantially, by a further 90% in fact. Here, the very
high compressibility of the crumb rubber, as much as
A small dosage of hemp was used to replace steel fibres
its low strength, suggests that the cement/GGBS paste
to prevent early age and long term drying shrinkage
has to provide most of the compression resistance and
cracking.
thus resulted in the very low compressive strength at
56 days (approximately 3.5MPa). This was unsurprising
Strength Development
as the concrete was honeycombed and hard to compact
The compressive strength development of the various and finish. Finally, the inclusion of the other substitute
mixes was monitored over a 56 day period and the results constituents (grit and quarry dust) in concrete made
of the average of three cube tests are shown in Figure 3. with Mix 4 appeared to have no beneficial effect on the
From this it may be observed that the high GGBS content strength, as anticipated, but did make the hardened
in Mix 2 caused a 30% drop in the 56 day strength (from concrete more compactable and finishable and thus with
approximately 45.5MPa to 32.0MPa), suggesting that no honeycombing.
some of the GGBS had not hydrated without artificial
stimulus and may be acting solely as a filler. When the
Conclusions
While the somewhat simplistic objective of raising
environmental awareness amongst children was readily
met, the interesting research questions as to how to
maintain performance of concrete while using alternative
and inferior recycled and waste materials which naturally
degrade the concrete’s mechanical properties raises
interesting research questions. Currently, the scope for
alkali activators, natural plasticisers to reduce the w/c
substantially, pre-coating recycled concrete to reduce
absorption and enhance the interfacial transition zone,
the practicalities of further grinding of the waste grit to
create an artificial sand and the choice of a natural fibrous
Fig. 3: Strength development with time, Mixes 1 to 4
material to replace steel/polypropylene fibres are all
suitable strands of an on- going research programme.
In the meanwhile, the positive experience of making
highly sustainable concrete in the classroom (Figure 4)
is not one which the primary school students will easily
forget and the message for an improved awareness
of the responsibility of engineers to make concrete as
sustainable as technically and economically possible will
not be lost on them and, thus, hopefully, in future on their
generation.

References
1. Ajdukiewicz, A. and A. Kliszczewicz, 2002. Influence of recycled
Fig. 4: Highly sustainable concrete slab with hand imprints, aggregates on mechanical properties of HS/HPC. Cement and
Concrete Composites, 24(2): 269-279.
retained in the school for posterity
2. Albano, C., et al., 2005. Influence of scrap rubber addition to

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

70 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainable Concrete as a Platform for Outreach

Portland I concrete composites: Destructive and non-destructive rainwater – Comparison of different roofing materials, Environmental
testing. Composite Structures, 71(3-4): 439-446. Pollution, 162, 422-429.
3. Al-Tayeb, M.M., et al., 2013. Effect of partial replacement of sand 11. Li, Z., Wang, X.and Wang, L., 2006. Properties of hemp fibre
by recycled fine crumb rubber on the performance of hybrid reinforced concrete composites. Composites Part A: Applied
rubberized-normal concrete under impact load: experiment and Science and Manufacturing, 37, 3, 497-505.
simulation. Journal of Cleaner Production, 59(0): 284-289.
12. McNally, C. and E. Sheils, 2012. Probability-based assessment of
4. Arnaud, L. and Gourlay, E., 2012. Experimental study of parameters the durability characteristics of concretes manufactured using
influencing mechanical properties of hemp concretes. Construction CEM II and GGBS binders. Construction and Building Materials,
and Building Materials, 28, 1, 50-56. 30(0): 22-29.
5. Bogas, J.A., J. de Brito, and J. Cabaço, 2014. Long-term behaviour 13. O’Rourke, B., C. McNally, and M.G. Richardson, 2009. Development
of concrete produced with recycled lightweight expanded clay of calcium sulfate–ggbs– Portland cement binders. Construction
aggregate concrete. Construction and Building Materials, 65(0): and Building Materials, 23(1): 340-346.
470-479.
14. Pretot, S., Collet, F. and Garnier, C., 2014. Life cycle assessment
6. Cartuxo, F., et al., 2015. Rheological behaviour of concrete of a hemp concrete wall: Impact of thickness and coating. Building
made with fine recycled concrete aggregates – Influence of the and Environment, 72, 223-231.
superplasticizer. Construction and Building Materials, 89(0): 36-47.
15. Shariq, M., J. Prasad, and H. Abbas, 2013. Effect of GGBFS on age
7. Farreny, R., Morales-Pinzón, T., Guisasola, A., Tayà, C., Rieradevall, dependent static modulus of elasticity of concrete. Construction
J. and Gabarrell, X., 2011a. Roof selection for rainwater harvesting: and Building Materials, 41(0): 411-418.
Quantity and quality assessments in Spain, Water Research, 45,
16. Silva, R.V., J. de Brito, and R.K. Dhir, 2014. Properties and
10, 3245-3254.
composition of recycled aggregates from construction and
8. Farreny, R., Gabarrell, X. and Rieradevall, J., 2011b. Cost-efficiency demolition waste suitable for concrete production. Construction
of rainwater harvesting strategies in dense Mediterranean and Building Materials, 65(0): 201-217.
neighbourhoods, Resources, Conservation and Recycling, 55, 7,
17. Teng, S., T.Y.D. Lim, and B. Sabet Divsholi, 2013, Durability and
686-694.
mechanical properties of high strength concrete incorporating
9. Githachuri, K.u. and M.G. Alexander, 2013. Durability performance ultra fine Ground Granulated Blast-furnace Slag. Construction and
potential and strength of blended Portland limestone cement Building Materials, 40(0): 875-881.
concrete. Cement and Concrete Composites, 39(0): 115-121.
18. Yung, W.H., L.C. Yung, and L.H. Hua, 2013. A study of the durability
10. Lee, J.Y., Bak, G. and Han, M., 2012. Quality of roof-harvested properties of waste tire rubber applied to self-compacting concrete.
Construction and Building Materials, 41(0): 665-672.

Prof. Roger West

“Prof Roger West is Director of Postgraduate Teaching and Learning in the School of Engineering at Trinity
College Dublin, Ireland. His main areas of research interests includes concrete rheology, fibre reinforcing,
durability and innovative materials. He is chairman of the Irish Concrete Durability Committee and is a
board member of the journal, Magazine of Concrete Research. He has over 100 peer-reviewed publications
and has delivered a dozen key-note papers on four continents.”

Organised by
India Chapter of American Concrete Institute 71
Session 1 B - Paper 2

Life Cycle Assessment and Durability of Concrete

Containing Limestone
D.K. Panesar, K. Seto, and M. Aqel
Department of Civil Engineering, University of Toronto, Toronto, Ontario, Canada, M5S 1A4

Abstract The negative environmental impact associated with

Over the past two decades, efforts have been made to cement and concrete production can be reduced by
‘green’ the concrete industry through material selection, improving manufacturing techniques, using alternative
manufacturing and implementation of innovative concrete fuels and/or using cement substitution (CEMBUREAU,
products. Today the challenge we are faced with is not 2009). Approximately 50% to 55% of CO2 emission from
only advancement in construction building materials, but cement production is produced from the calcination of raw
choosing among several material options while balancing materials (Huntzinger et al., 2009). Cement substitution
cost, environmental impacts, and long term durability. using supplementary cementing materials (SCM) or fillers
Life cycle assessment (LCA) is one approach to evaluate such as limestone filler (LF) are a common approaches to
the environmental impacts of ‘green’ concrete. This reduce CO2 emissions (Neuhoff et al., 2014; IEA Clean Coal
study examines the effect of limestone filler (LF) on the Centre, 2011; Imbabi et al., 2012). Fillers are typically inert
compressive strength and rapid chloride permeability
materials that are used to reduce the amount of binding
(RCPT) of self-consolidating concrete. In addition, a
material (i.e. cement).
cradle-to-grave LCA was conducted with three functional
units- namely, concrete volume, compressive strength, The use of LF as cement replacement has become a
and a combination of compressive strength and durability common practice in Canada as most GU cement has up to
(by RCPT). The LCA evaluates the relative differences 5% of interground LF. Cement containing up to 15% of LF
between concrete mixtures with and without LF on the is known as Portland limestone cement (PLC) in Canada
basis of their environmental impacts, in particular their
and the United States and has been implemented in many
potential to cause acidification, global warming, resource
areas in the construction industry. However, the use of
depletion, and water depletion. The experimental results
showed increased compressive strength and decreased PLC in precast/prestressed applications is still limited.
concrete permeability with the addition of LF. In addition, This is partly due to the limited data in the literature
the environmental performance of concrete was improved related to the influence of LF on early age properties and
with the addition of LF. long term durability performance of steam cured concrete
at elevated temperatures.
Keywords: Life Cycle Assessment, Limestone Filler,
Durability, Compressive Strength, Rapid Chloride Although alternative concrete materials such as LF
Permeability Test (RCPT) can potentially improve environmental performance,
it is important to assess these materials in a rigorous
Introduction way over their entire life cycle such that benefits are
Global cement production was estimated to be 4.2 billion not excluded nor exaggerated. Life cycle assessment
tonnes in 2014 and is responsible for approximately 9% of (LCA) enables analysis of all activities that occur during
the global CO2 emissions (U.S. Geological Survey, 2014). a product’s life cycle, including raw materials extraction,
Moreover, the global cement demand is increasing which transportation, production, use, maintenance and end
creates pressure on the concrete construction industry of life (Scientific Applications International Corporation,
to reduce the impact on the environment. Such pressure 2006). With such a broad scope of potential impacts, a
has been emphasized by governments around the world LCA perspective is useful for assessing the environmental
through imposing carbon taxes on the production of CO2. performance of LF in precast/prestressed applications.
This law has been implemented in Canada (Alberta, British The aim of this paper is to study the influence of LF on
Columbia and Quebec) and in many U.S. states (Mabee et self-consolidating concrete (SCC) strength, permeability
al., 2011). Currently, the carbon tax varies from $3.5 per
[using rapid chloride permeability test (RCPT)] and life
tonne of CO2 in Quebec to $30 per tonne of CO2 in British
cycle environmental impact.
Columbia.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

72 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Life Cycle Assessment and Durability of Concrete Containing Limestone

Experimental Program with the symbol “5SF”. Mix deigns containing 15% of LF
are identified with the symbol “LF”. All concrete mixes
Materials have a constant water-to-cement-ratio of 0.34 and a
In this study, two types of cement were used, namely, constant sand-to-total aggregate-ratio of 0.47. LF was
Calcium Sufo alyminate (CSA) type GU (general use) and used as cement replacement but was not considered
CSA type HE (high early strength) cement. The cements as a cementitious material in the water-to-cement ratio
were supplied by Holcim and Lafarge Canada. The calculation. All concrete mixtures were designed to have
reason for using two different sources of cement is to 5.5 ± 0.5% entrained air and slump flow of 650 ± 50 mm.
represent the commonly used cements in the precast/
prestressed industry in Canada. The SF used was an
Table 2
undensified powder from the production of silicon metal Concrete Mix Design
supplied by SKW Canada Inc. The chemical compositions
of the cements and SF are presented in Table 1. LF No Mix ID Cement SF LF Coarse Fine Water AEA HRWR
with a nominal particle size of 3μm (Blaine fineness of Agg. Agg.
1125 m2/kg) was used. The LF was supplied by Omya, kg/m3 ml/100kg
Canada. The fine aggregate was natural sand with a specific
1 GU-1 427.5 - - 950 850 150.3 37 900
gravity of 2.72 and fineness modulus of 2.84. The coarse
aggregate was crushed limestone with a maximum size 2 GU- 360 - 67.5 950 950 121.0 148 2450
of 13 mm. The fine and coarse aggregates were supplied 1LF
by Dufferin Aggregates. Two chemical admixtures were 3 HE1 427.5 - - 950 950 150.3 45 1000
used, namely, high range water reducer HRWR (Plastol 4 HE1- 360 - 67.5 950 950 121.0 240 2500
6400) and air entraining admixture AEA (Airex-L) which LF
were supplied by Euclid Chemical Company, Canada.
5 GU2 450 22.5 - 950 950 150.3 30 700

Table 1 6 GU2- 382.5 22.5 67.5 950 950 121.0 135 2100
Chemical and Physical Properties of Cement and SF 5SF-
LF
Cement
Chemical 7 HE2- 450 22.5 - 950 950 150.3 35 850
Composition and Holcim Lafarge 5SF
Physical Properties SF
GUI HE1 GU2 HE2 8 HE2- 382.5 22.5 67.5 950 950 121.0 165 2150
5SF-
CaCO3 (%) 2.5 3.5 2.0 0.0 - LF

SiO2 (%) 19.6 19.1 19.4 19.7 92.1

Curing Regime
Al 2O3 (%) 5.2 5.2 4.9 5.0 0.3
The curing regime consisted of steam curing for 16 hours
Fe2O3 2.3 2.4 3.3 3.3 0.6 at a maximum holding temperature of 70°C as presented
CaO (%) 61.8 61.6 61.9 61.8 0.8 in Figure 1. Following steam curing, concrete samples
were moist cured at 23°C until tested at 28 days.
MgO (%) 2.4 2.4 2.6 2.5 0.7

SO3 (%) 4.0 4.3 4.0 4.1 0.2 Testing

Na2Oeq (%) 1.0 1.0 0.6 0.7 0.9 The plastic properties of concrete were evaluated using
C3S (%) 44.9 55.2 56.0 54.0 -
slump flow (ASTM C1611-09) and air content (ASTM C231-

C3A (%) 10.3 9.8 8.0 8.0 -

C 4 AF 7.0 7.1 10.0 10.0 -

C2S (%) 22.3 13.2 13.0 14.0 -

LOI at 1150 °C (%) 2.61 2.10 1.80 0.90 2.0

Blaine (m2/kg) 417 514 371 505 -

Mix Design
The SCC mix designs evaluated in this study are
presented in Table 2. In this table, the first symbol in the
mixture ID represents cement type followed by a number
representing the source of the cement. Mix designs
containing 5% SF as cement replacement are identified Fig. 1: Steam Curing Regime

Organised by
India Chapter of American Concrete Institute 73
Session 1 B - Paper 2

10). For each concrete mixture three cylinders (100 mm impact assessment methods, including the International
diameter × 200 mm height) were tested for compressive Reference Life Cycle Data System (ILCD) method which
strength at 28 days in accordance to CSA A23.2-9C. The was used for this study (European Commission Joint
rapid chloride permeability test (RCPT) was performed at Research Centre, 2011).
28 days according to ASTM C 1202-10.
Four impact categories were selected based on their
relevance to the assessment of the environmental
LCA Methodology impact of concrete. Table 4 presents a description of each
A cradle-to-grave system boundary that includes raw impact category as well as the characterization factor
material extraction, transportation, mixing, and end-of- used in the analysis. In LCA, characterization factors are
life has been selected as presented in science-based factors that are used to convert LCI data
Figure 2. The following assumptions were made in the to actual environmental impacts (Scientific Applications
model: International Corporation, 2006). The life cycle impact
assessment results were interpreted using the three
ll As HE cement is finer than GU cement, production functional units, and were normalized by dividing the
data for HE cement was modelled by increasing the results by the result of mix GU1. These normalized results
energy input and emissions data for GU cement by 25% were then combined into a single green indicator score
while maintaining the same raw material usage. This that combines the acidification, global warming, resource
approach is consistent with previous LCAs (Glavind depletion, and water depletion potentials. A weighting of
and Haugaard, 1999; Wang et al. 2012). 1:1:1:1 was used to combine these categories.
ll The upstream production of SF is not included in the
analysis, as it is a waste product that is produced
regardless of final use. Using SF in concrete is a better
environmental choice compared to landfilling.
ll Transportation distances were assumed to be relative
to Toronto, Ontario, Canada.
ll Electricity production was based on the 2014 Ontario
grid mix (Independent Electricity System Operator,
2015).
ll Since the same steam curing regime was used for all
mix designs, and consumption data for steam curing
specifically is not typically recorded in the industry,
water and energy usage by steam curing was not
included in this system boundary. This omission is not
expected to influence the results, which are normalized
to mixture GU1.
A functional unit is the basis for analysis in an LCA, and it
allows for comparisons between diverse concrete types
based on functional equivalency (Scientific Applications
International Corporation, 2006). Durability and strength
are properties that are particularly important for concrete
construction and are often the basis for mix design, and
so these are appropriate functional units (De Schepper
et al., 2014). This approach was used in this analysis. The
three functional units are FU1 = 1 m3 of concrete, FU2 = m3
for 1 MPa of 28-day compressive strength, and FU3 = m3
for 1 MPa of 28-day compressive strength and 1 Coulomb
of 28-day RCPT. LCA requires large quantities of input
and output data [life cycle inventory (LCI)] to be analyzed
in order to determine net environmental impact. Table
3 identifies sources of LCI data used for conventional
concrete and concrete containing limestone.
In order to compile the LCI data and conduct the impact
assessment, the GaBi 6 software package was used (PE Fig. 2: System Boundaries for LCA of i) Conventional Concrete
International, 2014). GaBi 6 is packaged with several ii) Concrete with LF

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

74 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Life Cycle Assessment and Durability of Concrete Containing Limestone

Table 3
LCI Data for Conventional Concrete and Concrete Containing LF

Process Source Included Sub-Processes

Cement (Canada Centre for Mineral & Energy Raw material, energy use, atmospheric emissions, liquid effluent and water
Production Technology and Radian Canada Inc., 1993; usage data for the Canadian cement industry GU data modified for HE
Venta et al., 1999; Athena, 2005)

Limestone (Athena, 2005) Grinding energy per tonne. It is assumed that grinding is done at quarry (no
Grinding transportation).

Electricity (PE International, 2014) Energy production from biomass, hard coal, hydropower, natural gas, nuclear,
Production and wind. Datasets combined according to 2014 Ontario grid mix (Independent
Electricity System Operator, 2015).

Water Production (Racoviceanu et al., 2007) Energy use and greenhouse gas emissions for the City of Toronto municipal
water treatment system (chemical manufacturing, chemical transportation,
and water treatment facility operation).

Aggregate (Canada Centre for Mineral & Energy Energy requirements, raw material requirements, atmospheric emissions and
Production and Technology and Radian Canada Inc., 1993; liquid effluents for the extraction, processing and transportation of the fine
Transport Venta et al., 1999; Athena, 2005) and coarse aggregates.

Admixtures (European Federation of Concrete Admixture All inputs and outputs for production of 1 kg of high range water reducer and 1
Production Associations, 2005; European Federation of kg of air entraining admixture
Concrete Admixture Associations, 2006)

Concrete Mixing (Prusinski e al., 2004) Energy for batching and mixing processes to produce a unit quantity of concrete

End-of-Life (Canada Centre for Mineral & Energy Energy requirements for crushing in place (rubblizing). Emissions factors for
Technology and Radian Canada Inc., 1993; energy production from diesel.
Venta et al., 1999; Athena, 2005)

Table 4
Impact Categories

Impact Category Description Characterization Factor

Acidification Release of protons and leaching out of anions from a system Acidification Potential (mole H+ ion equivalents)

Global Warming Effect of increasing temperature in the atmosphere due to the Global Warming Potential (kg CO2 equivalents)
reflection of radiation by greenhouse gasses back to the surface

Resource Depletion of abiotic non-water resources (ex. minerals or fossil Resource Depletion Potential (kg of antimony (Sb)
Depletion fuels) equivalents)

Water Depletion Water use (reuse, degradation, or incorporation into final product) Water Depletion Potential (volume of water used (m3)

Results and Discussion SF had the greatest influence on the 28-day compressive
strength followed by LF and cement type.
Compressive Strength
The mean 28-day compressive strengths were
normalized to the value of the GU1 compressive strength.
The normalized results are presented in Figure 3.
Concrete mixes made of HE cement without SF showed
higher strength compared to counterpart mixes made
with GU cement. For concrete mixes made with 5% SF, the
compressive strength of HE and GU mixes were similar.
The addition of LF increased the compressive strength
compared to concrete mixes made without LF. This
increase was greater in concrete mixes made without SF.
The addition of 5% SF caused significant (approximately 16
to 25%) increase in the compressive strength compared to
counterpart mixes made without SF. Based on the results, Fig. 3: Normalized Compressive Strength of Concrete at 28 Days

Organised by
India Chapter of American Concrete Institute 75
Session 1 B - Paper 2

Rapid Chloride Permeability Test of 1.16. This positive effect is increased when HE cement
and/or SF are used in the mix designs. The optimal mix
The results of the RCPT were normalized by dividing over
design in terms of environmental performance was found
the result of mix GU1 as presented in Figure 4. Concrete
to be HE- 5SF-LF, which has a normalized green indicator
mixes made of HE cement showed lower concrete
of 8.98. The results highlight how the improvement in
permeability compared to counterpart mixes made
durability influences the environmental performance of
with GU cement. The addition of LF reduced concrete
the concrete.
permeability by approximately 7% to 12% compared to
counterpart mixes made without LF. The addition of 5%
SF caused significant reduction in concrete permeability Conclusions
at 28 days. This reduction ranged from 65% to 71% Based on the results of this study, the following conclusions
compared to counterpart mixes made without SF. SF was can be drawn:
the most influential variable on concrete permeability at
the 28 days followed by LF and cement type. 1. The use of LF as cement replacement increases
the compressive strength and reduces concrete
permeability at 28 days. In addition, LF improved the
life cycle environmental performance of concrete.
2. Concrete mixes made with HE cement had greater
compressive strength, lower concrete permeability
and improved environmental performance compared
to GU cement.
3. SF had the greatest influence on compressive
strength, concrete permeability and environmental
performance of concrete followed by LF and cement
type.

Fig. 4: Normalized RCPT Results of Concrete at 28 Days Acknowledgments

This research was supported by the Ministry of
LCA Results Transportation of Ontario. Opinions expressed in this report
are those of the authors and may not necessarily reflect
The life cycle impacts of concrete mixtures were
the views and policies of the Ministry of Transportation of
interpreted using three different functional units and
Ontario. The authors would like to acknowledge Holcim
the environmental performance of concrete mix designs
and Lafarge Canada for providing the cement, Omya
was calculated (quantified here as a ‘green indicator’).
Canada for providing the limestone and Euclid Admixture
Figure 5 shows the normalized LCIA results for the three
Canada Inc. for providing the chemical admixtures.
functional units. In Figure 5, the numerical value of the
labels show the green indicator corresponding to FU3. This
References
figure shows that adding LF increases the green indicator
1. American Society for Testing and Materials, 2010. ASTM C1202
of concrete regardless of cement type or the use of SF. standard test method for electrical indication of concrete’s ability
In addition, the results show that the improved durability to resist chloride ion penetration. Annual Book of ASTM Standards.
with the addition of limestone (as measured using 28-day Philadelphia, USA.
RCPT) further increases the green indicator of concrete. It 2. American Society for Testing and Materials, 2009. ASTM C1611
can be seen that GU1-LF has a normalized green indicator standard test method for slump flow of self-consolidating concrete.
Annual Book of ASTM Standards. Philadelphia, USA.
3. American Society for Testing and Materials, 2010. ASTM C231,
standard test method for air content of freshly mixed concrete by the
pressure method. Annual book of ASTM Standards. Philadelphia,
USA.
4. Athena, 2005. Cement and structural concrete products: life cycle
inventory update #2. Ottawa: Athena Sustainable Materials Institute.
5. Canada Centre for Mineral & Energy Technology and Radian
Canada Inc., 1993. Raw material balances, energy profiles, and
environmental unit factor estimates: cement and structural
concrete products. Ottawa: Athena Sustainable Materials
Institute.
6. Canadian Standards Association, 2009. CSA A23.1/A23.2. Concrete
materials and methods of concrete construction/test methods and
standard practices for concrete. Ottawa, Canada.
Fig. 5: Normalized Green Indicator of Concrete

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

76 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Life Cycle Assessment and Durability of Concrete Containing Limestone

7. CEMBUREAU, 2009. Sustainable cement production: co-processing 18. Independent Electricity System Operator, 2015. Supply Overview.
of alternative fuels and raw materials in the cement industry. Retrieved February 10, 2015 from IESO: http://www.ieso.ca/Pages/
CEMBUREAU publications. Power-Data/Supply.aspx
8. De Schepper, M., Van den Heede, P., Van Driessche, I., De Beile, 19. Mabee, W., Wood, S., Cockburn, L., Wood, T., 2011. Agricultural
N., 2014. Life cycle assessment of completely recyclable concrete. bioenergy conference report. The Canadian Farm and Food Biogas
Materials (7):6010-6027. Conference, London, Canada, 7-10.
9. Environment Canada, 2014. National pollutant release inventory. 20. Neuhoff, K., Vanderborght, B., & Ancygier, A., 2014. Carbon control
Retrieved December 4, 2014 from Environment Canada: https:// and competitiveness post 2020: the cement report. Climate
www.ec.gc.ca/inrp-npri/ Strategies.
10. European Commission Joint Research Centre, 2011. International 21. PE International. (2014). GaBi 6.0.
reference life cycle data system (ILCD) handbook. Institute for
22. Prusinski, J. R., Marceau, M. L., VanGeem, M. G., 2004. Life cycle
environment and sustainability. Luxemburg: Publications Office of
inventory of slag cement concrete. Eighth CANMET/ACI Eighth
the European Union.
CANMET/ACI International Conference on Fly Ash, Silica Fume,
11. European Federation of Concrete Admixture Associations, 2005. Slag and Natural Pozzolans in Concrete.
EFCA Environmental declaration air entraining admixtures.
23. Racoviceanu, A., Kennedy, B., Kennedy, C., Colombo, A., 2007.
Retrieved February 10, 2015 from Cement Admixtures Associations:
Life-cycle energy use and greenhouse gas emissions inventory
http://admixtures.org.uk/publications.asp
for water treatment systems. Journal of Infrastructure Systems,
12. European Federation of Concrete Admixture Associations, 2006. 13(4), 261-270.
EFCA Environmental declaration superplasticising admixtures.
24. Scientific Applications International Corporation (SAIC), 2006. Life
Retrieved February 10, 2015 from Cement Admixtures Association:
cycle assessment: principles and practice. Cincinnati, OH: U.S.
http://admixtures.org.uk/publications.asp
Environmental Protection Agency.
13. Flury, K., Jungbluth, N., Frischknecht, R., Munoz, I., 2012.
25. Stranddorf, H. K., Hoffmann, L., & Schmidt, A., 2005. Update on
Recommendation for life cycle inventory analysis for water use
impact categories, normalisation and weighting in LCA- selected
and consumption. Working Paper, ESU-services, Zurich.
EDIP97-data. Danish Ministry of the Environment, Environmental
14. Glavind, M., Haugaard, M., 1999. Survey of environmental aspects Protection Agency. Danish EPA.
of the Danish concrete industry. Taastrup, Denmark: The Nordic
26. U.S. Geological Survey, 2014. Mineral Commodity Summaries.
Concrete Federation.
27. Venta, Glaser and Associates, 1999. Cement and structural concrete
15. Huntzinger, D., Gierke, J., Kawatra, S., Eisele, T., Sutter, L., 2009.
products: life cycle inventory update. Ottawa: Athena Sustainable
Carbon dioxide sequestration in cement kiln dust through mineral
Materials Institute.
carbonation. Environmental Science and Technology, 43(6):1986-1992.
28. Wang, T., Lee, I. S., Harvey, J., Kendall, A., Lee, E., Kim, C., 2012.
16. IEA Clean Coal Centre, 2011. CO2 abatement in the cement industry. UCPRC life cycle assessment methodology and initial case studies on
United Kingdom: Gemini House. energy consumption and GHG emissions for pavement preservation
17. Imbabi, M., Carrigan, C., McKenna, S., 2012. Trends and developments treatments with different rolling resistance. University of California
in green cement and concrete technology. International Journal of Pavement Research Center. Berkeley: California Department of
Sustainable Built Environment, 1(2):194-216. Transportation.

Daman Panesar
Affiliation: Department of Civil Engineering, University of Toronto
Dr. Panesar, is an Associate Professor with the Department of Civil Engineering at the University of Toronto.
She obtained three engineering degrees in Canada: B.Eng. (McMaster University), M.A.Sc. (Western
University), Ph.D. (McMaster University), and is a licensed Professional Engineering (Ontario). Prior to
her Ph.D., she was a Design Engineer with Atomic Energy of Canada Limited. Dr. Panesar’s research
activities are focused on advancing: concrete technology, material characterization, long-term durability
performance of cement-based materials and non-destructive testing. Her work on concrete material
science is also integrated in the fields of structural engineering and constructability of buildings, nuclear
facilities and transportation infrastructure. She is a member of RILEM, ACI, and CSA.
d.panesar@utoronto.ca

Organised by
India Chapter of American Concrete Institute 77
Session 1 B - Paper 3

Ageing of old and modern concrete structures

- Observations and Research
Klaas van Breugel,
Delft University of Technology, Faculty of Civil Engineering & Geosciences 2628 CN Delft, The Netherlands

Abstract country’s mobility and economy. The same holds for

our energy infrastructure. Power plants for generating
Ageing is an inherent feature of nature. Yet it seems to be
a rather new topic in both science and engineering. The electricity and energy transport grids have to function
main reason for increasing attention for ageing as a topic is reliably for 24 hours per day, the whole year round. Failing
the growing awareness that, particularly in industrialized components may cause expensive process interruptions
countries, ageing of our assets is a financial burden for and may even constitute a risk for life and limb. Pro-
the society and affects the overall sustainability of our active replacement of vital components of systems
planet. In this contribution the urgency and challenges and structures is considered a safe strategy to prevent
of ageing of concrete structures are addressed. The catastrophic failures. But do we really know how close
complexity of ageing problems will be illustrated by we were to a catastrophic failure at the moment these
looking in more detail to the evolution in concrete mix components were replaced? Was the society really at risk
design and the consequences thereof for the long-term or did we spoil a lot of still perfectly operating components
performance of concrete structures. Emphasis will be on without improving safety substantially? In other words:
ageing of concrete infrastructure and on the justification how accurately can we predict the progress of ageing
of research on ageing phenomena. processes from which our assets suffer?

Keywords: Infrastructure, durability, sustainability, Ageing is everywhere and unavoidable. Yet it is not easy to
ageing, mix design, high performance concrete, find a clear an unambiguous definition of ageing. The term
autogenous shrinkage, codes, integral asset management ageing is used for changes in performance with elapse
of time of materials, structures, systems, organisations,
societies, governments, software, economic systems,
Introduction living organisms, etc. These changes in performance
Ageing is everywhere around us. Huge mountains seem with time can be observed at different scales. But what
to keep their shape for ever, but at a closer look we see are the real driving forces behind these changes? Before
that the surface of rocks gradually changes. Changes in starting an attempt to explain what we mean by ageing,
temperature and moisture conditions, wear, wind and light we first give an impression of the societal relevance of
are sufficiently powerful to crumble even the strongest ageing of our fixed assets, with a focus of ageing of our
rock. Mountains age! Earthquakes may split mountains, infrastructure.
causing changes in the state of stress in the newly formed
parts of the mountain. Fresh fracture surfaces become
exposed to climatic conditions and another cycle of ageing Ageing Infrastructure and Society
starts. In modern industrial countries the infrastructure makes
out over 50% of the nation’s national wealth[1]. This
Like rocks, also man-made structures are exposed infrastructure consists of roads and railway systems,
to ambient climate conditions. While exposed to water works, airports, power stations and electricity
environmental loads, structures have to carry life loads grids. Based on an inventory in twelve countries, the
and deadweight in a safe way during their entire service value of infrastructure stock averages around 70% of
life. Roads and railways need continuous maintenance. If the global gross domestic product (GDP). For a global
planned correctly, the trouble maintenance works often
GDP of ∈ 53 trillion in 2012, this makes ∈ 37 trillion. This
cause can be kept to a minimum. If maintenance comes too
infrastructure is vital for the mobility of people and for a
late, expensive repair is needed and may cause time and
country’s economy.
money consuming traffic jams, delays or even accidents.
The direct costs of failing infrastructure can be huge, but Economic growth is inconceivable without growth of a
the indirect costs are generally many times higher. country’s infrastructure. To catch up with global economic
growth the McKinsey report[2] estimates a required
Proper functioning of our infrastructure is vital for the
investment in the infrastructure of ∈ 42 trillion between

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

78 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Ageing of old and modern concrete structures - Observations and Research

Table 1
Estimated needs for global infrastructure in different categories. Period 2013-2030[2]

Required investment
Category Source
[Χ ∈ 1,000,000,000,000]

Roads OECD1) 12.2

Rail OECD 3.3
Ports OECD 0.5
Airports OECD 1.4
Power IEA 2) 8.8
Water GWI3) 8.4
Telecommunications OECD 6.8
Total 41.4
1) OrganisationforEconomicCo-operationandDevelopment 2) International Energy Agency 3) Global Water Intelligence.

2013 and 2030. This means an annual investment of

∈ 2.3 trillion, which is about 4.5% of the global GDP.
The investment of ∈ 42 trillion is needed for roads and
railways, ports, airports, power stations, water works
and telecommunication. Table 1 gives the breakdown of
investments over these categories. These figures are (in
part) based on an extrapolation from data provided by
84 countries. These countries are responsible for 90% of
the global GDP and are considered today’s best possible
basis for estimating the investments needed for our
infrastructure in the period from 2013 to 2010.

Ageing and Science Fig. 1: Evolution of the probability of failure in a complex system [3]
Change of performance with time
The early lifetime of made-made materials, structures
and systems is often characterized by a high probability of
failure. It takes some time to overcome inevitable teething
problems and to reach the required level of maturity and
stability. Once that point is reached a ‘quiet’ period follows
until we arrive again in a period of increasing probability
of failure. Exceeding a certain predefined probability of
failure then marks the end of the service life of a structure
or system. The high probability of failure in the beginning,
the subsequent period of ‘rest’ and the subsequent period
of increasing probability of failure is generally presented
with the bathtub curve (Figure 1).
Fig. 2: Evolution of the performance of ageing materials,
In essence the bathtub curve also applies to our fixed
structures and systems[3]
assets. The length of the period in which the probability
of failure is low is of crucial importance for the economic
of failure’. After a short period of teething problems the
performance of these assets. The bathtub curve suggests
material, structure or system has reached the required
that this period is a period during which ‘nothing happens’.
(high) level of performance. That is the level at which the
It is a period of ‘rest’, or ‘dormant’ period. Assuming that
material should demonstrate its capacity to meet safety
in the period of low probability of failure nothing happens,
and functional criteria, if possible without intervention for
however, is misleading. If there would really be ‘rest’, what
maintenance or repair. It is the period of ‘top-level sport’
could then be the driving force behind the increase of
for all the basic building blocks, i.e. atoms, molecules and
probability of failure with elapse of time? To illustrate the
interfaces, from which a material or structure is made.
foregoing reasoning it may help to put the bathtub curve of
When these basic building blocks give up and leave their
figure 1 upside down, as shown in figure 2. On the vertical
position, the period of decay begins. Then ageing has
axis we now put ‘Performance’ instead of the ‘Frequency

Organised by
India Chapter of American Concrete Institute 79
Session 1 B - Paper 3

started! These first tiny steps of decay will most probably The foregoing illustrates that a material ‘at rest’ is hardly
not be observed at the macroscale immediately. The conceivable. Al lower scales there is motion all the time
moment that the first basic building blocks give up to do and a variety of driving forces cause the basic building
their job can only be captured with comprehensive and blocks of a material to change their position. In essence,
appropriate material models at subsequent levels of this holds for all materials and systems. Basic building
observation. Here chemistry, physics, electrochemistry, blocks search for a position (energy level), where they feel
mechanics and mathematics meet each other and more comfortable. By designing materials in a smart way,
need each other for developing tools for describing and i.e. by minimizing internal gradients and concentrations of
predicting ageing processes at a fundamental level. stresses and strains, there will be less reason for basic
building blocks to leave their position. Hence, the ageing
Driving forces behind ageing – A closer look process will slow down and the service life of materials,
Ageing has been defined is a change of performance of structures and systems will be enhanced.
a material, structure or system with elapse of time. How In the following sections the aforementioned general
time per se can result in a change of performance is not principles of ageing constitute the framework for
easy to understand at first sight. How can a material ‘at evaluating ageing phenomena in concrete structures.
rest’ change its performance with time? At this point we
have to realize that time is the domain in which we describe
Performance of Concrete Structures
changes in performance rather than the cause of these
changes. The remaining question is then: what causes the Performance of bridge decks – An inventory
changes of a material, structure or system that is, on the
In March 2001 the results of a most interesting study were
macroscale, in a status of ‘rest’.
published by Mehta et al[4]. He analysed the performance
A closer look at any piece of matter ‘at rest’ tells us that of bridge decks of bridges built in four subsequent periods
the status of rest only applies to a certain length scale. in the twentieth century. The first period was the period
Going down to the atomic scale the world is in motion all before 1930, the second between 1930 and 1950, the third
the time! Fundamental entities, i.e. basic building blocks, from 1950 to 1980 and the fourth from 1980 to present.
are continuously moving with a certain probability to The concrete mixtures used for the bridge decks were
leave their position for one that fits them better. This characterised by the chemical composition and the
phenomenon takes place in the time domain. It is an fineness of the cement. The cements used in the first
inherent feature of matter and lies at the basis of ageing period, before 1930, had a C3S content less the 30% and
of materials. On top of this inherent feature we see, at a Blaine surface of 180 m2/kg. Consequently the rate of
different scales, a number of gradients, which may hydration was low. The performance of many of the bridge
cause the basic building blocks of matter to start moving. decks made with these cements was quite good.
Gradients are the driving forces causing changes with The cements used in the second period were ground
elapse of time. Note that at the boundary of any piece of to a Blaine fineness between 180 and 300 m2/kg. The
material with its environment gradients do exist. These construction and building technology used for the bridge
gradients concern, for example, temperature, humidity decks was similar to those used in the first period. The
and radiation and they may cause changes at the surface authors report that the bridge decks built in the second
of the material. period were less durable than those built before 1930.
Inside a material crystals are connected and build up a The structures that were built between 1950 and 1980
strong microstructure. But just at the contact points appeared to have more durability problems than those
between crystals atoms are liable to leave their position built before 1950. The cements used in this period had a
causing changes in the performance of the material. In fineness up to 400 m2/kg and a C3S content beyond 60%.
heterogeneous materials – and concrete is one of the best With the aim to get a denser and more durable concrete
examples of a heterogeneous material – numerous sites the w/c ratio was lower than in the first two periods. The
exist where stress concentrations may initiate changes higher C3S content and the higher fineness of the cement
of the microstructure of a material resulting in non- had increased the early strength of these mixtures. This
linear behaviour with elapse of time. The material ages! made it possible to build faster. This, however, had resulted
Porous materials continuously communicate with their in a higher probability of early-age thermal cracking and,
environment and never reach a condition of rest. This on top of that, higher autogenous shrinkage of the low
ongoing communication of porous materials, or porous water-cement ratio mixtures. The higher proneness to
systems, with their environment induce alternating early- age (micro)cracking was the most plausible reason
stresses and strains in the system. This gradually changes for durability problems at later ages.
the nano- and microstructure of the material and hence In the fourth period the tendency to go for higher strengths
its performance. continued. Generally this was realised by using mixtures

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

80 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Ageing of old and modern concrete structures - Observations and Research

with a low w/b ratio. The use of low w/b mixtures further shrinkage will be compared with values given by currently
increased the risk of cracking. For bridge decks moderate used design codes.
strengths between 30 and 45 MPa were found. Among 29
bridge decks the cracking in 44 MPa bridge decks was
Shrinkage – Influencing factors
twice that in 31 MPa bridge decks.
Relative humidity - - The hydration-induced drop of the
Mix design and proneness to ageing relative humidity in the pores of cement- based materials
is a major reason for the occurrence of autogenous
Mehta’s inventory of the performance of bridge decks
shrinkage. This so-called ‘internal drying’ goes along
illustrates how the pressure from the market to build
with an increase in the capillary pressure in the pore
faster has created a demand for mixtures with a high
water. As a consequence the solid microstructure is put in
early strength. This was possible by using finer cements
compression, which leads to autogenous shrinkage.
with a higher C3S content. The price of this, however, was
a higher probability of early-age cracking of the bridge Shrinkage of the drying paste is restraint by the aggregate
decks. particles in the mixture. Whether the restraint of
autogenous shrinkage strain will cause (micro)cracking
For realising more slender and more elegant structures
depends on the size and stiffness of the aggregate
a higher final strength is required. A high strength is
particles and on the time dependent properties (creep,
attainable by reducing the w/b ratio. The use of (super)
relaxation) of the hardening paste. How internal curing of
plasticisers has made it possible to reduce the w/b ratio
concrete mixtures can be used to prevent a drop of the
of concrete mixtures to values even below 0.2. With these
relative humidity, and hence of autogenous shrinkage, will
low w/b mixtures dense concretes are obtained with low
be discussed later in this contribution.
permeability. This is considered good for the concrete’s
durability. At the same time, however, we see an increase The drop of relative humidity can also be caused by drying
in the concrete’s proneness to (micro)cracking, mainly of the concrete when exposed to an environment with
because of increased autogenous shrinkage. low relative humidity. In this case moisture gradients
are created, which result in stress gradients. Also these
Another reason for a higher cracking risk of high
stresses will be subject the time dependent behaviour of
strength and ultra-high strength concretes are the high
the concrete.
temperatures that occur as a result of high cement
contents in those mixtures. By optimizing the particle Surface tension -- Besides the relative humidity also the
packing of the aggregate fractions the amount of cement, surface tension in the capillary water is an important
and hence the peak temperatures, can be reduced. A shrinkage parameter. Reducing the surface tension
low cement content is also considered positive from the will result in lower capillary stresses and hence less
sustainability point of view (lower carbon footprint of the shrinkage. The reduction of surface tension can be realised
fresh concrete mixture). A low cement content, however, by adding shrinkage reducing admixtures (SRA’s). There
also has a drawback. A low cement content reduces the is a mitigating effect indeed by these products[5], but they
inherent self-healing capacity of the concrete. From the can have some negative side effects as well, for example
self-healing point of view a not too low cement content a (small) drop in compressive strength[6].
and the use of ‘old’, coarsely ground cement is favourable. Type of cement -- Autogenous shrinkage is different
This partly explains the outcome of Mehta’s inventory that for different types of cement. Hence, a proper choice
old bridge decks performed better than newer ones. In for a certain cement could be a straightforward way to
the terminology of this paper we would say that the old mitigate shrinkage problems on the construction site.
concrete mixtures with coarse cements with a low C3S Concrete, w/c = 0.35, made with a composite cement with
content were less prone to ageing than modern mixtures a low clinker content (CEM II A-L 42.5R, 15% limestone)
with finely ground cements with a high C3S content. had a smaller autogenous shrinkage after 21 days than
concrete made with ordinary Portland cement (OPC: CEM
Autogenous Shrinkage – A Closer Look I, 42.5R[7]. Concrete made with a slag cement (CEM II B-S
For understanding ageing of traditional and modern 42.5R, 27% slag) expanded initially. However, between 1
concrete mixtures we need a clear picture of the processes and 10 days after mixing the shrinkage in that period was
that cause internal stresses in the material. As discussed larger than of the CEM I 42.5R and the CEM II A-L 42.5R.
in section 3 these internal stresses are among the driving Jensen[8] found that a high C3A or gypsum content leads
forces of ageing. One of the causes of internal stresses to a significant reduction of autogenous deformation of
is autogenous shrinkage of hardening concrete. In this cement paste. Based on that observation he suggested
section experimental results of autogenous shrinkage that potentially the crack tendency of HPC may be reduced
of traditional and high strength concrete mixtures are by increasing the C3A or gypsum content of the cement.
presented, as well as technological measures to mitigate Jensen’s observations were in contrast with those of
autogenous shrinkage. The measured autogenous Tazawa et al.[9]. As possible reasons for the discrepancy

Organised by
India Chapter of American Concrete Institute 81
 Table 2
Session 1 B - Paper 3 Mixture compositions of HPC (C55/65)[16]

Component I II III IV
Water kg/m3 133 153 156 156
CEM III/B 42.5 LH HS kg/m3 248 340 300 300
CEM I 52.5 R kg/m 3
112 110 100 100
Limestone powder1) kg/m3 60 - -- --
Water/cement ratio kg/m 3
0.37 0.34 0.39 0.39
Sand 0 – 4 mm kg/m3 942 860 830 830
Crushed aggregate 4–16
kg/m3 997 980 975 730
mm
Liapor F10, 4-8 mm kg/m3 - - -- 156
HR Superplast. CON 35 kg/m 3
5.0 - -- --
Cretoplast CON 35 kg/m3 - 1.8 -- --
Cretoplast SL01 CON 35 kg/m3 - 7.2 -- --
Addiment BV1 kg/m 3
- - 1.6 1.6
Addiment FM 951 kg/m3 - - 4.8 4.8
1) Fineness 530 m2/kg

between results of Jensen and Tazawa, Jensen mentioned effect can be achieved with mixed-in super absorbing
the way in which the C3A- content had been determined, polymers (SAP), a technology promoted by Jensen et al[14]
the way of testing and the influence of minor components and subject of the RILEM committee 225-SAP[15].
in the cement. Schachinger et al[10] found that autogenous
In the following paragraphs results of studies on the
shrinkage of cement pastes, w/b = 0.2 – 0.33, made with
magnitude of autogenous shrinkage of concrete and
blastfurnace cement (CEM III B 42.5 NW/HS), was smaller
the effect of internal curing will be presented. First test
than that of OPC pastes (CEM I 42.5 R/HS). Schachinger
results on autogenous shrinkage of concrete mixtures
also found that autogenous shrinkage was still continuing
C55/65 are discussed, followed by results obtained with
after 56 days.
mixtures C28/35 and C35/45. After that internal curing
Internal restraint – Both autogenous shrinkage and drying of ultra-high performance concrete made with rice husk
shrinkage of the paste is restrained by the aggregates in ash will be presented, followed by a discussion on the
the mixture. Even though this internal restraint causes relationship between shrinkage and ageing.
internal (tensile) stresses in the shrinking paste and
Background of the study -- In the eighties and nineties
probably microcracking, this so-called passive internal
of the past century the use of HPC with target strength
restraint of shrinkage strains has been suggested as a
C55/65 was considered for several concrete bridges in
mechanism to mitigate autogenous shrinkage (Soliman
The Netherlands. The prevailing Dutch design code did
et al.[11]). Unhydrated cores of cement particles or finely
not require designers to consider autogenous shrinkage
ground inert powders can act as restraining components
of those mixtures. However, the owner of the bridges, the
for reducing autogenous shrinkage.
Ministry of Transportation, required a check of the overall
From this brief overview it is clear that quite a number of performance of the mixtures, including a check of the
parameters affect the magnitude of autogenous shrinkage. autogenous shrinkage and the effectiveness of internal
By manipulating these parameters the consequences of curing for mitigating the risk of early-age cracking.
autogenous shrinkage can be mitigated and hence the
Mixture design and test specimen -- Four mixtures
susceptibility of concrete mixtures to age. In the next
were tested with w/b ranging from 0.34 to 0.39. The mix
sections emphasis will be on autogenous shrinkage and
compositions are given in Table 2. In mixture I, 60 kg
internal curing of HPC and UHPC and how internal curing
limestone powder was used while the amount of cement
reduces autogenous shrinkage.
was reduced by the same amount. In mixture IV, 25% of
the coarse aggregate was replaced by water-saturated
Autogenous shrinkage in C55/65 mixtures lightweight aggregate, Liapor F10. The autogenous
and internal curing shrinkage was measured on sealed specimens,
100x100x400 mm3.
As indicated in the previous section, autogenous shrinkage
of low w/c mixtures can be reduced by internal curing. Measured autogenous deformation and evaluation -- The
Internal curing can be accomplished by adding water- autogenous deformations of the mixtures are presented
saturated lightweight aggregate (LWA) particles to the in Figure 3. The measurements started after 1 day.
concrete[12,13]. When the internal RH drops, the water stored This implies that the very first part of the autogenous
in the LWA particles is released to the drying matrix, thus deformation was not recorded. This was not considered
maintaining the RH at a relatively high level. A similar a problem, since the aim of the test series was to quantify

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

82 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Ageing of old and modern concrete structures - Observations and Research

strength classes C35/45 and C28/35. In a preliminary

study on autogenous shrinkage of concrete mixtures with
w/c ≈ 0.45 Van Cappellen[17] found that particularly at early
ages the autogenous shrinkage of concrete mixtures made
with blast-furnace slag cement developed faster than that
of OPC-mixtures. At 200 days the difference was not very
large anymore. Van Cappelle’s study was continued by
Mors[18] for mixtures made with two types of aggregate,
i.e. limestone and quartz. The mixture compositions are
given in Table 3.
Fig. 4 and 5 show the autogenous shrinkage of the
traditional mixtures T (0.50) and T (0.44) made with
Fig. 3: Autogenous deformation mixture I to IV. 20oC. quartz aggregate and the mixtures N (0.50) and N (0.46)
Measurements starting after 1 day [16] made with limestone aggregate. The shrinkage curves
convincingly show that also mixtures with w/b in the range
how autogenous deformation would affect drying from 0.44 to 0.5 exhibit substantial autogenous shrinkage.
shrinkage. In the practice drying shrinkage will generally More importantly, also these mixtures exhibit ongoing
not start during the first day after casting. For the purpose autogenous shrinkage at ages beyond 28 days, the age at
of this study it was appropriate, therefore, to measure only which the concrete is generally assumed to have reached
the autogenous deformation after 1 day. a high degree of maturity already. The relevance to
these findings for the interpretation for drying shrinkage
The shrinkage curves of the mixtures I, II and III show measurements carried out in the past will be discussed in
that a greater part of autogenous shrinkage occurs in more detail later.
the first few days after mixing. But even after 28 days
autogenous shrinkage still continues. From 28 to 91 days Autogenous shrinkage in UHPC made with Rice Husk
the autogenous shrinkage of the mixtures I, II and III varies Ash
from 70 to 90 µm/m.
Ultra High Performance Concrete (UHPC) is often made
Replacing 25% of the dense aggregate by water-saturated with mixtures with a low w/b ratio and fine silica fume for
lightweight aggregate particles was sufficient to eliminate improving the particle packing. Silica fume (SF), however,
autogenous shrinkage of this paste. Obviously the internal is expensive and cheaper alternatives would be welcome.
curing by using saturated lightweight aggregate particles Tuan[20] investigated the application of finely ground rice
(Liapor F10, 4-8 mm) is very effective. husk ash (RHA) as alternative for silica fume. From the
chemical point of view the RHA and SF are quite similar, but
Autogenous shrinkage of traditional concrete the microstructure of the particles is completely different.
Silica fume particles are spherical. RHA particles have an
mixtures C28/35 and C35/45 irregular shape, are porous and are much bigger, between
The high autogenous shrinkage of mixture C55/65, much 5 μm to 10 μm. By adding combinations of silica fume and
higher than expected, was reason enough to start an RHA to the concrete very high strengths were obtained,
investigation on the autogenous shrinkage of traditional beyond 200 MPa, even higher than mixtures where either
concrete mixtures with w/b ratios between 0.44 and 0.50, only silica fume or only RHA were used[20].

Table 3
Mixture compositions of concrete mixtures C28/35 and C35/45 (Mors [18])

Concrete T (0.50) T (0.44) N (0.50) N (0.46)

Strength class C28/35 C35/45 C28/35 C35/45

CEM III/B (kg/m3) 340 340 340 360
LSP filler (kg/m3) -- -- 20 20
Design w/c 0.50 0.44 0.50 0.46
SPL (% M/Mcem) 0.2 0.2 0.2 0.2
Fine aggregate Sand 0/4 Sand 0/4 Sand 0/4 Sand 0/4
Coarse aggregate Gravel Gravel  Limestone Limestone
Fractions 4/8, 8/16 4/8, 8/16 6/20 6/20
T = Traditional mixture; N = Mixtures made with natural limestone as aggregate.

Organised by
India Chapter of American Concrete Institute 83
Session 1 B - Paper 3

Fig. 4: Autogenous shrinkage of traditional mixtures T(0.50) and Fig. 6: Autogenous shrinkage of UHPC-mixtures. Reference
T(0.44). Quartz aggregate. w/c 0.5 and 0.44 [18,19] (REF) and RHA-pastes with 10% and 20% replacement of
cement by RHA (Tuan [20]).

Fig. 5: Autogenous shrinkage of mixtures N(0.50) and N(0.46).

Limestone aggregate. w/c = 0.46 and 0.50 [18,19] Fig. 7: Autogenous shrinkage of UHPC mixtures containing RHA
(20%) with different mean particle sizes measured from the
The materials used by Tuan were Portland cement (CEM I final setting time until 28 days (Tuan [20]).
52.5N) with a Blaine specific surface of 4500 cm2/g, silica
sand with a mean particle size of 225 μm, condensed The autogenous shrinkage of UHPC mixtures was
silica fume, rice husk ash, and polycarboxylate-based measured according to the ASTM C1698 standard (ASTM
superplasticizer with 30% solid content by weight. The Standard 2009). The ambient temperature was maintained
rice husk ash contained 88.0% amorphous SiO2, 3.81% at (23±1)°C.
loss on ignition. After grinding the mean size of the RHA
particles was 5.6 μm. Mixtures with RHA with mean Figure 6 shows the development of autogenous shrinkage
particle sizes of 3.6 μm and 9.0 μm were studied as well. of UHPC with 0% (REF), 10% and 20% RHA(5.6). The figure
The mix compositions are shown in Table 4. The cement shows that most of the autogenous shrinkage occurs in
replacement percentages of the SF and RHA were 10% the first 12 hours after casting. There is a big influence,
and 20% by weight of the cement, respectively. however, of the added RHA. After 12 hours the autogenous

Table 4
UHPC mixtures for autogenous shrinkage measurements (after Tuan [20])

Amount of w/b ratio Sand/binder ratio RHA SF The mean article

Mix No.
cement, [kg/m3] (by weight) (by weight) (% by weight) (% by weight) size of RHA [μm]

REF 1140 0.18 1 0 0 -

RHA10 (5.6) 1010 0.18 1 10 0 5.6
SF10 1010 0.18 1 0 10 -
RHA20 (3.6) 885 0.18 1 20 0 3.6
RHA20 (5.6) 885 0.18 1 20 0 5.6
RHA20 (9.0) 885 0.18 1 20 0 9.0
SF20 885 0.18 1 0 20 -

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

84 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Ageing of old and modern concrete structures - Observations and Research

Autogenous Shrinkage, Drying Shrinkage and Design

Codes
In the past experimental studies on drying shrinkage
of concrete have often been carried out on 28 days
old specimens and the measured shrinkage strains
were mostly interpreted as drying shrinkage. From the
autogenous shrinkage curves presented in the previous
sections we have to conclude, however, that after 28 days
autogenous shrinkage cannot be ignored, also not for
mixtures with water-cement ratios higher than 0.4. For
those mixtures the contribution of autogenous shrinkage to
the measured shrinkage strains in drying specimens has
often been neglected. This means that in the past many
Fig. 8: Autogenous shrinkage of UHPC mixtures containing
drying shrinkage tests might have been misinterpreted.
different amounts of SF measured from the final setting time
until 28 days (Tuan [20]). A substantial part of the strains measured on drying
specimens should have been attributed to autogenous
shrinkage of the mixture with 20% RHA(5.6) is significantly shrinkage. In recent updates of design codes autogenous
lower than that of the reference mixture. The autogenous shrinkage is now explicitly mentioned, also for traditional
shrinkage of the mixture with 10% RHA (RHA10(5.6)) is mixtures with w/b > 0.4. In the new EuroCode 2 and the
found between that of the reference mixture and mixture Japanese code autogenous shrinkage is considered also for
RHA20(5.6). After 14 days mixture RHA20(5.6) even mixtures in strength classes < C55/65. For mixtures with
showed a slight expansion. This tendency to expand is not w/b 0.44 – 0.50, however, these codes still underestimated
observed for the reference mixture (REF) and the mixture the autogenous shrinkage, at least for the tested mixtures
with 10% RHA. and cement types considered in the previous sections.
Figure 9 shows autogenous shrinkage according to both the
Fig. 7 shows the effect of the mean particle size of the EuroCode 2 and the JSCE-Code, together with the measured
RHA for mixtures with 20% replacement of cement by autogenous shrinkage of normal strength concretes
RHA. The autogenous shrinkage of mixtures RHA20(9.0) C28/35 and C35/45. Both the measured autogenous
and RHA20(5.6) are almost identical. Similar to mixture shrinkage and the curve after correction for small moisture
RHA20(5.6) also mixture RHA20(9.0) exhibits a tendency loss through the sealant are presented. The autogenous
to slightly expand in the period between 14 and 22 days. shrinkage according to EuroCode 2 is presented for C28/35
The autogenous shrinkage of the mixture with the fine and C35/45 mixtures, i.e. mixtures with strengths similar
ground ash, i.e. mixture RHA20(3.6), is found between the to the measured strength of the mixtures considered in this
reference mixture and the two mixtures RHA20(5.6) and paper. In both cases the autogenous shrinkage according
RHA20(9.0). The fine RHA turned out to be less effective to EuroCode 2 is about 30% of the measured autogenous
in reducing autogenous shrinkage. The most plausible shrinkage. Underestimation of autogenous shrinkage by
reason for this is that this fine ground ash has lost too EuroCode 2 was also observed by Darquennes et al.[24].
much of its internal porous structure and could hence The predictions with the Japanese code are closer to the
store less water in its porous microstructure than the measured values, but still underestimate the measured
coarser ground RHA(5.6) and RHA(9.0)[20]. autogenous shrinkage.
In Fig. 8 the autogenous shrinkage of mixtures with 10%
and 20% SF are shown, together with the autogenous
shrinkage of the reference mixture. The curves show
that the effect of silica fume on autogenous shrinkage is
completely different from that of RHA. The difference is
largest for the mixtures with the coarser RHA particles
(compare Fig. 7 and Fig. 8). The autogenous shrinkage
of mixture SF20 is close to that of the reference mixture,
whereas the mixtures with 20% RHA, i.e. RHA20(5.6) and
RHA20(9.0), hardly show any net autogenous shrinkage
after 12 hours. It is suggested that an internal curing
mechanism as found for saturated lightweight aggregate
explains why the autogenous shrinkage of RHA-modified
mixtures is much smaller than that of the reference
mixture and the SF-containing mixtures. A mitigating effect Fig. 9: Comparison of measured autogenous shrinkage with
of RHA on autogenous shrinkage has also been found by de predictions with the EuroCode 2[22] and the Japanese Code [23]
Sensale et al.[21] for Portland cement pastes with w/b = 0.3. (after Mors [18]).

Organised by
India Chapter of American Concrete Institute 85
Session 1 B - Paper 3

Shrinkage and Ageing In the reactive approach the heterogeneity of the material,
Autogenous and drying shrinkage are both the result of and hence the internal concentrations of stresses and
the development of capillary stresses in the pore water strains and the occurrence of internal damage and ageing,
caused by a drop of the internal relative humidity. In both are considered a matter of fact. If ageing is unavoidable
cases the cement paste is the shrinking component. indeed, self-healing could be a solutions for ageing
Autogenous shrinkage strains are restraint by both problems. When dealing with concrete the presence of
aggregates and fillers in the concrete and in the cement still unhydrated cement particles provides an inherent
paste, respectively, and will result in an internal self-healing capacity. In this respect concrete made with a
state of stress. Drying shrinkage goes along with the coarse cement is considered favourable to concrete made
development of moisture and stress gradients in the with a fine cement. Mehta’s[4] observation that old bridge
cross section of a concrete element. On top of this any decks, built with coarse cement, have performed better
restraint of shrinkage of concrete elements on the than the younger ones built with finer cements, could be
macroscale will generate structural forces and, hence, explained, at least in part, by the role of self-healing in the
additional stresses in the concrete. All these shrinkage- older structures. The modern trend to, firstly, use finer
induced stresses are subject to relaxation. Relaxation cement in order to speed up the rate of strength gain and,
of stresses, however, is not ‘for free’. It requires the secondly, reduce the amount of cement in order to reduce
re-structuring of smallest building blocks of matter. In the CO2 footprint of concrete, may work out negative on
other words: the material ages! As far as autogenous the material’s resistance against ageing! In these cases a
shrinkage strains are concerned it has been proposed comprehensive life cycle analysis is needed for weighing all
that the observed long-term deformations might be the pros and cons of modern trends in concrete mix design.
creep strains following the elastic shrinkage strains
exerted by the capillary forces in the pore water. Also Ageing and design codes
these creep deformations are not ‘for free’, but require For designing and realizing concrete structures design
re-structuring of the material’s elementary building codes are indispensable. From the numerous buildings
blocks: The material ages! For quantitative analysis and fascinating construction works realised in the past a
of creep and relaxation Wittmann has applied the high degree of maturity of these codes can be inferred. In
activation energy concept[25], which is considered the foregoing sections we have seen, however, that currently
most appropriate approach for fundamental research of used prescriptive codes fall short in describing long-term
ageing phenomena of cement-based materials. performance, i.e. shrinkage, of concrete structures. In this
respect it is interesting to reflect on the recent tendency
to switch from prescriptive codes to performance based
Coping With Ageing
codes. The question is whether it is to be expected that
Materials design with this switch ageing issues will be considered more
appropriately and will become part of an integral design
Ageing is an inherent feature of materials. Solutions
approach for concrete structures. Strictly speaking the
for ageing problems require, therefore, interventions
change from prescriptive to performance-based codes
at fundamental materials level. For coping with ageing
is a return to the origin of the building profession. In the
problems, two approaches are conceivable, i.e. the
ancient past the whole building process was in the hands
preventive approach and the reactive approach.
one person: the builder. The builder had the integral
In the preventive approach the focus is on designing responsibility to meet all the safety and functional criteria
homogenous materials with as few as possible internal set by the owner. In which way the builder performed his job
gradients, stress concentrations and interfaces. For and how he managed to meet the owner’s criteria, was not
heterogeneous materials, like concrete, this is a big prescribed in detail. This was all considered the builder’s
challenge. When going through the subsequent length expertise and responsibility. In his classical book on
scales, from (sub-)nano to meso level, concrete behaves building technology Vitruvius stated that ideally the whole
as a complex system rather than a material[26]. To state building process (architectural design, structural design,
it differently, concrete is a ‘product of the mind’[27], of materials choice, execution, etc.) should be in the hands of
which the properties are determined by the properties of one person[28]. When Vitruvius wrote his book, a few years
the individual components and of the interfaces between BC, he noticed already that this ideal situation was no longer
them. Some of these components – in fact all! - change tenable. The building process became too complicated and
with time, and so do the properties of the interfaces. This one single person could not be an expert in all areas of the
makes heterogeneous materials susceptible to ageing. building process. Gradually the builder had to share his
Multiscale modelling is considered the appropriate responsibility with others. This situation started the emerge
vehicle for analysing time-dependent changes of these of certificates and, later on, prescriptive codes. The user of
complex materials and for developing concretes with a these documents could be held responsible for a correct
lowest possible proneness to ageing. interpretation of and compliance with the codes, but not for
the content of the codes.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

86 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Ageing of old and modern concrete structures - Observations and Research

Prescriptive codes can be judged as the ultimate

consequence of a process of increasing fragmentation of
the building process and, more importantly, of the vision
that everything, including quality, is engineerable[29].
The huge sustainability problems the world is facing
today illustrates, however, that this vision has lost most
of its convincing power. Prescriptive codes, even the
most detailed ones, are necessary, but insufficient for
guaranteeing quality and/or sustainability. Prescriptive
codes deal with materials properties with the primary goal
to provide the designer with data needed for designing safe
structures. Any change of performance of the materials
with time is considered a time dependent property without
addressing the cause of these changes.
With performance-based codes the building process has Fig. 10: Schematic presentation of required investment in ageing
been given back to the builder, including the challenge to research for realizing an extension of the mean service life
accomplish (long-term) quality criteria and sustainability of infrastructure of 10%. (interest / inflation not considered).
goals. The builder’s freedom to decide how to meet these Estimated average lifetime 50 years.
criteria and goals may stimulate the builder to invest in
fundamental research of traditional and new building science-oriented research on materials and structures.
materials and in innovative design concepts. Furthermore, A part of this science-oriented research has to be spent
performance-based codes, in combination with new on ageing research. A reasonable, though conservative
DBFM (Design-Built-Finance- Maintenance) contracts, assumption is that 10% of science-oriented research, i.e.
will also force the builder to focus on both the short-term ∈ 0.7 billion per year, should be spent on fundamental
and long-term performance, i.e. ageing, of materials and ageing research. This ∈ 0.7 billion is 5% of the required
structures. For that purpose the builder will need reliable research budget for realizing the savings of yearly
predictive models, including models for quantifying the replacement costs and only 1% of the targeted savings.
rate of ageing processes and the consequences thereof. Schematically this is shown in Figure 10. By varying the
assumptions in this exercise other values for required
investments are obtained, but do not change the order of
Required Investment for Generating Savings magnitude of these figures.
In the first section of this paper it has been explained that
ageing of the nation’s fixed capital goods is a huge financial Concluding Remarks
burden for the society. A way to reduce this burden is by
A nation’s infrastructure makes out about 50% if its national
reducing the maintenance costs and extending the life
wealth. This infrastructure is essential for the country’s
time of our infrastructure. This will result in savings in
economy and prosperity. However, this infrastructure is
annual replacement costs of structures that have reached
ageing. This means that 50% of the nation’s national wealth
the end of their service life. But for realizing these savings
is ageing! With the existence, growth, maintenance and
we first have to invest! Potential savings justify, and
replacement of ageing infrastructure a huge responsibility
require, investments in ageing research. Some key figures
comes to all actors involved in planning, design, building
may help us to get an indicative picture of the required
and operating our assets. It is a matter of responsible
investment for realizing a certain level of savings.
stewardship to mitigate the environmental impact that
We have seen that the global value of the infrastructure comes along with realizing and operating our infrastructure.
stock has been estimated at ∈ 37 trillion. Let us assume
Fundamental research on ageing is recommended in
an average lifetime of these infrastructure assets of 50
order to improve and extend the tools for accurate and
years. Each year ∈ 740 billion has then to be spent on
reliable predictions of the long-term performance of
replacement of obsolete assets. Let us further assume
our ageing infrastructure. The results of experimental
that through dedicated research the average lifetime
research on autogenous shrinkage of traditional
can be increased by 10%, i.e. from 50 to 55 years. The
and modern, innovative cement-based materials
yearly replacement costs would then decrease to from
illustrates the need for more research in order to better
∈ 740 to ∈ 670 billion. This is a reduction of ∈ 70 billion
understand the cause of autogenous shrinkage, as well
per year. Let us assume that for saving these ∈ 70 billion
as the (sometimes unexpected) possibilities to mitigate
we have to invest 20% of this amount in research, i.e. ∈
autogenous shrinkage, for example by using low-tech
14 billion per year. Let us further assume that 50% of the
waste products, like rice husk ash. Mitigating shrinkage
required research money, i.e. ∈ 7 billion, has to be spent
implies mitigating shrinkage-induced stresses and hence
on management-oriented research and the other 50% on
reducing the rate of ageing.

Organised by
India Chapter of American Concrete Institute 87
Session 1 B - Paper 3

Performance-based design codes, in combination with 9. Tazawa, E., Miyazawa, S. “Influence of constituents and composition
on autogenous shrinkage of cementitious materials”. Magazine of
new contracts in which the builder is made responsible
Concrete Research, 49, 178, 1997, pp. 15-22.
for the long-term performance and operation of their
10. Schachinger, I., Schmidt, K., Heinz, D., Schiessl, P. “Early-
structures, generate a strong need of knowledge of Age Cracking Risk and Relaxation by Restrained Autogenous
ageing phenomena in materials and structures. In this Deformation of Ultra High Performance Concrete”. Proc. 6th Int.
way performance-based codes may stimulate the search Symp. On Utilisation of High Strength / High Performance Concrete,
for innovative solutions. Ed. G. Konig et al. , Leipzig, 2002, Vol. 2, 2002, pp. 1341-1353.
11. Soliman, A.M., Nehdi, M.L., “Self-Restraining Ultra-High-
Like many other industries, also the building industry is Performance Concrete: Mechanism and Evidence”. ACI Mat.
under pressure. Structures have to be realised faster, but Journal, Vol. 110, No. 4, 2013, pp 355-364.
with lower environmental impact. Any product, however, 12. Van Breugel, K., de Vries, J. “Mixture optimization of low water/
realised under pressure, irrespective of what kind of cement ratio high strength concretes in view of reduction of
pressure, has an inherent tendency to age. To cope with autogenous shrinkage”. Proc. Int. Symp. on High- Performance
and reactive powder concretes. Sherbrooke, 1998, pp. 365-382
the risk of increasing ageing rates, in-depth knowledge of
the performance of materials and structures with elapse 13. Zhutovsky, S., Kovler, K., Bentur, A. “Efficiency of lightweight
aggregates for internal curing of high strength concrete to eliminate
of time is needed. An increase of the average service life autogenous shrinkage”. Proc. Int. RILEM Conf. Early Age Cracking
of our infrastructure by 10% would save tens of billions of in Cementitious Systems, AEC’01, Ed. K. Kovler et al., Haifa, 2001,
euros each year. The required investments to realise these pp. 365-374.
savings are estimated at 20% of these savings. Half of this 14. Jensen, O.M. “Use of Superabsorbent Polymers in Concrete.
amount is assumed to be needed for research on materials Concrete International”, Vol. 35, No.1, 2013, pp. 48-52.
and structures, of which 10% has been assumed to be 15. Mechtcherine, V., Reinhardt, H.-W. “Application of Super Absorbent
needed for fundamental research on ageing. Setting such Polymers (SAP) in Concrete Construction”. State-of-the-Art Report
targets for savings is not only challenging and a stimulus Prepared by Technical Committee 225- SAP, 2012, 165 p.
for research and innovation. The figures also illustrate that 16. Van Breugel, K., Ouwerkerk, H., de Vries, J. “Effect of mixture
composition and size effect on shrinkage of high strength concrete”.
caring for our infrastructure will finally pay off.
Proc. Int. RILEM Workshop Shrinkage of Concrete - Shrinkage 2000.
Ed. Baroghel Bouny et al., Paris, 2000, pp. 161-177.
Acknowledgement 17. Van Cappellen, J. “Autogenous and drying shrinkage”, MSc-thesis,
Delft University of Technology, Delft, The Netherlands. 2009. 95 p.
Part of this paper was based on the Vision Document of
18. Mors, R.M. “Autogenous shrinkage – Cementitious materials
the Ageing Centre for Materials, Structures and Systems
containing BFS”. MSc-thesis, TU Delft, 2011, 63 p.
of the Technical University of Delft. For the use of this
19. Van Breugel, K., Mors, R.E., Bouwmeester, J. “New insight in the
material the Ageing Centre is greatly acknowledged. combination of autogenous and drying shrinkage”. Fib symposium
Engineering the concrete future: Technology, Modelling and
References Construction. 2013.
1. Long, A.E., “Sustainable bridges through innovative advances”. 20. Tuan, N.V. “Rice husk ash as a mineral admixture for ultra-high
Institution of Civil Engineers, presented at Joint ICE and TRF Fellows performance concrete”. PhD Thesis, Delft, 2011, 165 p.
Lecture. 23, 2007.
21. de Sensale, G.R., Ribeiro, A.B., Conçalves, A. “Effects of RHA on
2. Dobbs R., et al. “Infrastructure productivity: How to save $ 1 trillion autogenous shrinkage of Portland cement pastes”. Cement &
a year”. McKinsey Global Institute, 2013, 88 p. Concrete Composites, 30. (10), 2008, pp. 892-897.
3. Van Breugel, K. “Caring for ageing infrastructure – Scope, strategy 22. “EuroCode 2. Design of concrete structures - Part 1-1: General
and responsible stewardship”. Proc. 3rd. Int. Conf. on Service Life rules and rules for buildings” (Ref. No. EN 1992-1-1:2004: E) (2004)
Design for Infrastructures. Zhuhai. 2014.
23. JSCE (1996, 2002), “Standard specification for design and
4. Mehta, P.K., Burrows, R.W. “Building durable structures in the 21st construction of concrete structures, part I [Construction]” (in
century”. Indian Concrete Journal, 2001, pp. 437-443. Japanese), JSCE, Tokyo, Japan (2002).
5. Lopers, A.N.M., Silva, E.F., Dal Molin, D.C.C., Pilho, R.D.T. 24. Darquennes, A., Roziere, E., Khokhar, M.I.A., Turcry, Ph., Loukili, A.,
“Shrinkage-Reducing Admixture: Effects on Durability of HSC”. Grondin, F. “Long- Term deformation and cracking risk of concrete
ACI Mat. Journal, Vol. 110, No. 4, 2013, pp. 365-374 with high content of mineral additions”. Materials and Structures,
6. Li, Z., Qi, M., Ma, B. “Influence of chemical admixtures of concrete Vo,. 45, No. 11, 2012, pp 1705-1716.
shrinkage and cracking”. Proc. Int. RILEM Conf. Early Age Cracking 25. Wittmann, F.H., “Grundlagen eines Modells zur Beschreibung
in Cementitious Systems, AEC’01, Ed. K. Kovler et al., Haifa, 2001, charakteristischer Eigenschafften des Betons”, Deutscher
pp. 291-300. Ausschuss fur Stahlbeton, Heft 290, 1977, 42-101.
7. Schiessl, P., Plannerer, M., Brandes, Chr. “Influence of binders 26. Van Breugel, K. “Concrete: A materials that barely deserves that
and admixtures on autogenous shrinkage of high performance qualification”. International Conference on Material Science and
concrete”. Proc. PRO 17, Int. RILEM workshop, Shrinkage of 64th RILEM Annual Week in Aachen - MATSCI (2010) Aachen
Concrete – Shrinkage 2000, Ed. Baroghel-Bouny et.al., Paris, 2000,
27. McCarter, R.. “Louis I. Kahn and the Nature of Concrete”. Concrete
pp. 179- 190.
International, Vol. 31, Nr. 12, 2009, pp. 27-33.
8. Jensen, O.M. “Influence of cement composition on autogenous
28. Vitruvius Pollio, M., (85-20 BC) “The Ten Books on Architecture”
deformation and change of the relative humidity”. Proc. PRO 17,
Int. RILEM workshop, Shrinkage of Concrete – Shrinkage 2000, Ed. 29. Van Breugel, K. “A critical appraisal of codes as vehicles for realizing
Baroghel-Bouny et.al., Paris, 2000, pp. 143-153. on-site quality”. Proc. fib. Symposium Engineering the concrete
future: Technology, Modelling and Construction. Mumbai, 2013.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

88 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Ageing of old and modern concrete structures - Observations and Research

Prof. Dr.ir. K. (Klaas) van Breugel

Delft University of Technology
Faculty of Civil Engineering & Technology
Delft
The Netherlands
Prof. Dr.ir. K. van Breugel is full professor at the faculty of Civil Engineering & Geosciences of Delft University
of Technology. His fields of expertise concerns design of concrete containment structures, extreme load
conditions, early age concrete and modelling and simulation of materials and structures..
Professor van Breugel is member of several international organisations ACI, RILEM, IABSE and fib. In 2014
he was appointed RILEM follow. He has chaired several national and international research committees and
was active in a number of international research projects. He is (co-)author of about 600 conference and
journal papers and editor of several books, reports and conference proceedings. In 2013 the Ageing Centre
for Materials, Structures and Systems was launched under his supervision.

Organised by
India Chapter of American Concrete Institute 89
Session 1 B - Paper 4

Influence of Supplementary Cementitious Materials on the Properties

of Ultra-high Performance Concrete
Zhengqi Li and Prasada Rao Rangaraju
Glenn Department of Civil Engineering, Clemson University, Clemson, SC 29634, United States

Abstract treatments to produce UHPC, several researchers have

conducted investigations on developing UHPC under
The effects of supplementary cementitious materials
ambient temperature and atmospheric pressure curing,
(SCM) on the workability, time of set, autogenous
in recent years. Wille et al. developed several non-fiber
shrinkage, compressive strength and drying shrinkage of
reinforced UHPC mixtures having compressive strength
paste fraction of UHPC were studied. The test results show
as high as 206 MPa without special treatment (Wille et al.
that the use of silica fume (SFU) can result in significant
2011). Rangaraju et al. developed fiber reinforced UHPC
decrease in the workability, decrease in the autogenous
mixtures having good workability, compressive strength
shrinkage and increase in the drying shrinkage of
over 150 MPa and durability without special treatment
paste. The use of fly ash (FA) can result in significant
(Rangaraju et al. 2013).
retardation in the setting behavior of paste, decrease in
the 1-day compressive strength and increase in the drying Generally, UHPC formulation is characterized by very
shrinkage of paste. The use of meta-kaolin (MK) can low water-to-cementitious materials (w/cm) ratio (< 0.2),
result in significant decrease in the workability and time high cementitious materials content (> 1000 kg/m3) and
of set, improved 1-day compressive strength, increase high high-range water-reducing admixtures (HRWRA)
in the autogenous shrinkage and decrease in the drying content (Graybeal 2006; Russell and Graybeal 2013; Wille
shrinkage of paste. The 28-day compressive strength of and Boisvert-Cotulio 2013; Wille et al. 2011). Round or
paste can be improved by using SCM at proper proportion semi-round siliceous sand with rough surface texture is
compared with that of control. The use of ternary blend preferred, and coarse aggregate is typically not used in the
of MK, FA and cement can compensate the reduction in UHPC formulation. Reinforcing fibers are used to improve
the 1-day compressive strength and increase in the drying the tensile strength of UHPC (Graybeal 2006; Russell and
shrinkage of paste due to the use of FA, and compensates Graybeal 2013; Wille and Boisvert-Cotulio 2013; Wille et al.
the reduction in the workability of paste due to the use 2011). Supplementary cementitious material (SCM) is an
of MK. UHPC mixtures can be prepared by using ternary important component in the UHPC formulation to improve
blend of MK, FA and cement, instead of using binary blend its performance. Based on the previous literature, silica
of SFU and cement. fume (SFU) is the most frequently used SCM replacing
part of the portland cement to improve the compressive
Keywords: UHPC, meta-kaolin; fly ash; silica fume;
strength and durability of UHPC (Graybeal 2006; Li et
ternary blend.
al. 2014; Randl et al. 2014; Russell and Graybeal 2013;
Tafraoui et al. 2009; Wille and Boisvert-Cotulio 2013; Wille
Introduction et al. 2011). SFU is a by-product of the silicon industry
Ultra-high performance concrete (UHPC) is a relatively (kDurekovic 1995; Mindess et al. 2003; Ping and Beaudoin
new type of concrete which is defined to have a minimum 1992). It can densify the microstructure of both the bulk
28-day compressive strength above 150 MPa, post- paste and the interfacial transition zone (ITZ), which is
cracking tensile strengths above 5 MPa, and good attributed to the micro-filler effect and the pozzolanic
durability (Graybeal 2006; Russell and Graybeal 2013; effect of SFU (kDurekovic 1995; Mindess et al. 2003;
Wille and Boisvert-Cotulio 2013; Wille et al. 2011). Ping and Beaudoin 1992). Initially as a cheap industrial
Previous research studies related to UHPC were mostly byproduct, SFU has become much more expensive than
focused on investigating UHPC mixtures cured with the portland cement nowadays. Cheap substitutes to SFU,
special treatments, such as elevated temperature curing such as fly ash (FA) and meta-kaolin (MK), have been
or pressure curing (Cheyrezy et al. 1995; Richard and increasingly attracting research interest (Randl et al.
Cheyrezy 1995; Roy et al. 1972). For instance, concrete 2014; Tafraoui et al. 2009; Yu et al. 2015).
mixture having compressive strength up to 230 MPa can FA is produced from power plant. It is well known that the
be produced under 90°C curing (Richard and Cheyrezy use of FA can reduce the water demand and improve the
1995). Considering the difficulties of applying special workability of the fresh concrete (Mindess et al. 2003).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

90 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of Supplementary Cementitious Materials on the Properties of Ultra-high Performance Concrete

However, one of the disadvantages of using FA is that it

Table 1
causes reduction in compressive strength particularly
Physical and chemical properties of cementitious materials
at early ages. Meta-kaolin (MK) is produced from a well-
controlled process of calcination of quality kaolin clay Cement SFU FA MK
within certain temperature range (650–800 oC) (Sabir et al.
2001). It contains reactive silica and considerable portion SiO2 (%) 20.5 95.5 54.1 50.4
of alumina. Similar to other types of pozzolanic materials, Fe2O3 (%) 3.5 0.3 8.01 0.45
the use of MK in concrete can densify the microstructure,
which improves the mechanical properties and durability Al 2O3 (%) 4.9 0.7 27.8 42.6
of concrete (Khatib and Wild 1996; Poon et al. 2006; Poon CaO (%) 64.1 0.4 1.34 0.02
et al. 2001; Wild et al. 1996; Zhang and Malhotra 1995).
MgO (%) 1.3 0.5 0.90 0.16
The use of MK has also been observed to accelerate the
cement hydration (da Cunha et al. 2008; r e ). However, Na2Oeq (%) 0.47 1.4 2.13 0.22
one of the disadvantages of using MK is that it reduces the
SO3 (%) 3.6 - 0.16 0.00
workability of fresh mixture (Guneyisi and Gesoglu 2008).
Studies on normal strength concrete has shown that the LOI (%) 1.34 2.0 2.39 1.63
combined use of FA and MK can compensate the early age Specific gravity 3.15 2.20 2.26 2.20
strength loss due to the use of FA and the reduction in
workability of fresh concrete due to the use of MK (Bai et
al. 2000; Guneyisi and Gesoglu 2008). meeting the requirements of ASTM C33 specification for
fine aggregates. The gradation of the fine aggregate used
The present study was focused on the properties of in this study is shown in Table 2. The specific gravity, water
UHPC influenced by SCMs including SFU, MK and FA. absorption, and fineness modulus of the sand were 2.63,
The experiment was divided into three parts. In the first 0.3%, and 2.65, respectively.
part, the effects of different types of SCMs (SFU, MK and
FA) on the properties of paste fraction of UHPC were
Table 2
comprehensively studied. The investigated properties Gradation of fine aggregate
included workability, setting time, autogenous shrinkage,
compressive strength and drying shrinkage of paste. In Sieve opening (mm) Percent Passing
the second part, workability, compressive strength and
drying shrinkage of paste using ternary blend of MK, FA 9.5 100.0
and cement were studied, in comparison with paste using 4.75 99.8
binary blend of SFU and cement. In the third part, several
2.36 97.1
paste formulations with good performance were selected
to develop UHPC by proportioning with proper amount 1.18 82.0
of fine aggregate and steel micro fibers (SMF). The
0.60 41.9
investigated properties included workability, compressive
strength and drying shrinkage of UHPC. 0.30 14.0

0.15 0.5
Experimental Program
0.075 0.1
Materials
Steel micro fiber (SMF)
Cementitious materials The steel micro fibers used in this study had dimensions
A Type III portland cement meeting ASTM C150 of 13 mm in length and 0.2 mm in diameter. The specific
specification was used for the experimental study. The gravity and ultimate tensile strength were 7.8 and 2000
specific gravity and Blaine's surface area of the cement MPa, respectively.
were 3.15 and 54 m2/kg, respectively. Three types of SCMs
were used for study. These included a processed Class F High-range water-reducing admixture (HRWRA)
fly ash (FA) with an average particle size of 3 microns, high A powder form of polycarboxylate ether-based high-
reactivity meta-kaolin (MK) with an average particle size range water-reducing admixture (Melflux® 4930F) was
of 1.4 microns, and a silica fume (SFU) with an average used to improve the workability of paste.
particle size of 0.15 microns. The chemical composition
and physical properties of the portland cement and the Mixture Proportions
SCMs are given in Table 1.
Cementitious paste
Fine aggregate In this part, 14 paste mixtures were investigated to study
The fine aggregate was semi-round natural siliceous sand the properties of pastes using binary and ternary blends

Organised by
India Chapter of American Concrete Institute 91
Session 1 B - Paper 4

of SCMs and cement. The supplementary cementitious The quantities of materials used for 1 m3 of paste are
materials content in the paste was expressed as the mass presented in Table 4.
ratio of SCM to portland cement (SCM/c). The investigated
levels of SCM/c included 0, 0.05, 0.1, 0.2 and 0.3. For the Table 4
entire investigation, the w/cm by mass was fixed at 0.2, Quantities of materials used for 1 m3 of paste
and the HRWRA dosage was fixed at 1% by mass of total
cementitious materials. The relative mixture proportions Paste Constituent (kg/m3)
of the 14 paste mixtures are shown in Table 3. ID Cement SFU MK FA Water HRWRA
C 1933 0 0 0 387 19.3
Table 3 S1 1716 172 0 0 377 18.9
Relative proportions of materials in paste (by mass)
S2 1542 308 0 0 370 18.5
Paste SFU/ MK/ FA/ SCMb/ Water/ HRWRA/
c /c
a a
ID ca ca ca ca cmc cmc (%) S3 1401 420 0 0 364 18.2

C 1.00 0 0 0 0.00 0.20 1.0 M1 1818 0 91 0 382 19.1

S1 1.00 0.10 0 0 0.10 0.20 1.0 M2 1716 0 172 0 377 18.9

S2 1.00 0.20 0 0 0.20 0.20 1.0 M3 1542 0 308 0 370 18.5

S3 1.00 0.30 0 0 0.30 0.20 1.0 F1 1731 0 0 173 381 19.0

M1 1.00 0 0.05 0 0.05 0.20 1.0 F2 1568 0 0 314 376 18.8

M2 1.00 0 0.10 0 0.10 0.20 1.0 F3 1433 0 0 430 373 18.6

M3 1.00 0 0.20 0 0.20 0.20 1.0 MF1 1562 0 78 234 375 18.7

F1 1.00 0 0 0.10 0.10 0.20 1.0 MF2 1556 0 156 156 373 18.7

F2 1.00 0 0 0.20 0.20 0.20 1.0 MF3 1428 0 71 357 371 18.6

F3 1.00 0 0 0.30 0.30 0.20 1.0 MF4 1423 0 142 285 370 18.5

MF1 1.00 0 0.05 0.15 0.20 0.20 1.0

Selected paste mixtures with good performance were
MF2 1.00 0 0.10 0.10 0.20 0.20 1.0
used to produce UHPC by adding proper amount of sand
MF3 1.00 0 0.05 0.25 0.30 0.20 1.0 and SMF. The sand-cementitious materials (s/cm) ratio
MF4 1.00 0 0.10 0.20 0.30 0.20 1.0
was 1.25. The SMF content was 2% by volume of the UHPC
mixture.
Note: acement; bsupplementary cementing materials: silica fume alone or
meta-kaolin + fly ash; ccementitious materials: cement + SCM
Specimens Preparation
As Table 3 shows, the first paste C is the control which Fresh paste and UHPC mixtures were prepared by a
only contains Type III portland cement as cementitious UNIVEX M20 planetary mixer. For cementitious pastes,
material. the mixing procedure was divided into three stages, as
some of the mixtures took much longer time to reach
The next three pastes S1, S2, and S3 contain binary
fluid state than others, in particular when MK content was
blend of SFU and cement as cementitious material, and
high. At first, the cementitious materials and the HRWRA
the corresponding levels of SCM/c are 0.1, 0.2 and 0.3,
were dry mixed for about 4 min at low speed (100 RPM).
respectively. The three pastes designated as M1, M2 and
Then the mixing water was added to the dry mixture.
M3 contain binary blend of MK and cement as cementitious
The mixing continued at low speed until the dry mixture
material, and the corresponding levels of SCM/c are 0.05,
started to behave as fluid. As soon as the paste reached
0.1 and 0.2, respectively. Similarly, the pastes designated
fluid state, the mixing speed was increased to medium
as F1, F2 and F3 contain binary blend of FA and cement
speed (300 RPM), and the fluid mixture was mixed for
as cementitious material, and the corresponding levels
another 3 min. For UHPC mixtures, the mixing procedure
of SCM/c are 0.1, 0.2 and 0.3, respectively. The effect of
was basically same as the procedure of preparing
different types of SCM and their content on the properties
cementitious paste, except that the sand together with
of paste was determined by comparing these ten mixtures.
the cementitious materials and the HRWRA was dry
The last four pastes MF1, MF2, MF3 and MF4 were mixed first. The SMF was added after the fluid non-fiber
prepared by using ternary blend of MK, FA and cement as mortar was prepared. Then, the mixing continued for
cementitious material. The SCM content studied included another 5 min. The workability of paste and UHPC was
SCM/c=0.2 and SCM/c=0.3. measured immediately after mixing.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

92 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of Supplementary Cementitious Materials on the Properties of Ultra-high Performance Concrete

Fresh paste and UHPC mixtures were cast into molds Results and Discussion
without vibration. For the study of time of set and
autogenous shrinkage of paste, the specimens were Paste Using Binary Blend of SCM and Cement
stored following the ASTM C191 and ASTM C1698,
respectively. For the study of compressive strength of Workability
paste and UHPC, the specimens were de-molded at 24 The workability of cementitious paste using binary blend
hr., and stored in the moist room (100% relative humidity of SCM and cement is shown in Figure 1.
and 23 °C) until the age of testing. For the study of free As shown in Figure 1, for pastes containing binary blend
drying shrinkage of paste and UHPC, the specimens were of cement and FA, the increase in the FA content from
de-molded at 48 hr. after casting, as some of the mixtures SCM/c=0 to SCM/c=0.1 slightly increases the flow value of
had slow development in early age, particularly when FA paste by 4%. This is similar to what observed in previous
content was high. The drying shrinkage specimens were literature that the increase in the FA content improves the
stored following the procedures in ASTM C596. workability of concrete and attributed to the ball bearing
effect and water reducing effect of FA (Mindess et al.
Test Methods 2003). However, the further increase in the FA content
Workability from SCM/c=0.1 to SCM/c= 0.3 slightly decreases the flow
value of paste by 8%. Considering the fine particles size of
A similar procedure to determine the workability of
FA used in this study, the decrease in the flow value can
mortar to the one described in ASTM C1437 was followed.
be attributed to the large surface area provided by the FA
The fresh paste or UHPC was allowed to spread freely
particles. For pastes containing binary blend of cement and
on a level plastic plate without being dropped. When
SFU, the increase in the SFU content decreases the flow
the mixture stopped spreading (about 5 min after the
value. The flow of pastes with SCM/c at 0.1, 0.2 and 0.3 was
removal of the flow mold) the diameter of the mixture was
7%, 17% and 39% lower than that of control (SCM/c=0.0),
measured for calculating the flow value as described in
respectively. Likely the large surface area provided by the
the ASTM C1437 method.
fine SFU particles contributes to the decrease in the flow
Time of set value as the SFU content increases. For pastes containing
binary blend of cement and MK, the increase in the MK
The time of set of paste was determined following the
content decreases the flow value of pastes significantly.
test method described in ASTM C191, except that the
The flow of pastes with SCM/c at 0.05, 0.1 and 0.2 is 2%,
fresh paste was kept in a metal cylindrical mold instead
14% and 37% lower than that of control, respectively.
of conical ring to prevent the highly flowable paste from
The decrease in the flow value due to the use of MK has
leaking. The diameter of the cylindrical mold is 80 mm.
also been observed in the previous literature, and it was
The height of the finished paste sample in the cylindrical
attributed to its high surface area and its ability to promote
mold is 40 mm.
cement hydration at early ages (da Cunha et al. 2008; r e;
Autogenous shrinkage Guneyisi and Gesoglu 2008). It should be noted that MK
has the most significant effect in decreasing the flow value
The autogenous shrinkage of paste was determined
than both FA and SFU at the same levels of SCM/c.
following the test method described in ASTM C1698. Three
specimens were tested for each mixture.

Compressive strength
Compressive strength of each of the pastes and UHPCs
was determined by testing three 50×50×50 mm cubes at
the ages of 1 and 28 days. The compressive strength test
was conducted in accordance with procedures in ASTM
C109.

Free drying shrinkage

Three specimens with dimensions of 25×25×285 mm Fig. 1: Workability of paste using binary blend of SCM and cement
were prepared for each of the pastes and UHPCs.
Length comparator reading of each specimen stored Time of set
in the environmental chamber was taken following the The time of set of cementitious paste using binary blend of
procedures in ASTM C596 at selected periods of exposure SCM and cement is shown in Figure 2.
up to 25 days. The environmental chamber was maintained
at 23±2 °C temperature and 50±4% relative humidity in As shown in Figure 2, for pastes containing binary blend of
accordance with the requirement of ASTM C157. cement and FA, the increase in the FA content from SCM/c=0
to SCM/c=0.2 significantly increases both the time of initial set
and the time of final set. This is in accordance with previous

Organised by
India Chapter of American Concrete Institute 93
Session 1 B - Paper 4

As shown in Figure 3, the development of autogenous

shrinkage of pastes containing different types of SCM
starts to flatten at 48 hours after the final set of paste.
The autogenous shrinkage value of pastes at 168 hours
after the final set of paste is considered the maximum
autogenous shrinkage value of pastes. For pastes
containing binary blend of cement and FA, the increase
in the FA content from SCM/c=0 to SCM/c=0.1 slightly
increases the maximum autogenous shrinkage of paste.
However, the further increase in the FA content from SCM/
c=0.1 to SCM/c= 0.2 slightly decreases the autogenous
shrinkage value of paste. For pastes containing binary
Fig. 2: Time of set using binary blend of SCM and cement blend of cement and SFU, the increase in the SFU content
decreases the maximum autogenous shrinkage of paste. It
literature on the retarding effect of FA on the hydration of can be observed from Figure 3 that autogenous shrinkage
cement, and it was attributed to the low pozzolanic reactivity of pastes with SCM/c at 0.1and 0.2 is 13% and 29% lower
of FA at early ages (Brooks et al. 2000; Fajun et al. 1985; than that of control, respectively. For pastes containing
Frias et al. 2000; Guneyisi and Gesoglu 2008). For pastes binary blend of cement and MK, the increase in the MK
containing binary blend of cement and SFU, there appears content increases the maximum autogenous shrinkage of
to be a threshold SFU content of SCM/c=0.1. When the SFU paste significantly. It can be observed from Figure 3 that
content increases from SCM/c=0 to SCM/c=0.1, the time of the maximum autogenous shrinkage of pastes with SCM/c
initial set increases by 12%, and the time of final set increases at 0.1 and 0.2 is 31% and 53% higher than that of control,
by 2%. When the SFU content increases from SCM/c=0.1 respectively.
to SCM/c=0.2, the time of initial set decreases by 21%, and Previous literature on the effect of SCM on the autogenous
the time of final set decreases by 17%. In this study the shrinkage of paste is contradictory. The increase and
influence of SFU on the setting behavior of cement can be decrease in the autogenous shrinkage of paste resulted
attributed to the increased actual water-cement (w/c) ratio from the use of SCM both have been reported (Brooks
with the increase in SFU content as w/cm is kept constant, and Megat Johari 2001; Gleize et al. 2007; Kinuthia et al.
the nucleation effect provided by the surface area of SFU 2000; Mazloom et al. 2004; Tazawa and Miyazawa 1995;
particles, and the early pozzolanic reactivity of SFU. At low Tertnkhajornkit et al. 2005; Wild et al. 1998; Yoo et al. 2012).
SFU content, the increased actual w/c is likely responsible The autogenous shrinkage behavior of paste containing
to the increased time of set of cement. At high SFU content, SCM has been attributed to possible effects of SCM which
the nucleation effect and the early pozzolanic reaction of SFU include the dilution effect which decreases the autogenous
are likely responsible to the decreased time of set of cement. shrinkage, heterogeneous nucleation which may increase
For pastes containing binary blend of cement and MK, the or decrease the autogenous shrinkage, pozzolanic effect
increase in the MK content decreases both the time of initial which increases the autogenous shrinkage, filler effect
set and the time of final set. This was been observed in the which increases the autogenous shrinkage and properties
previous literature, and attributed to the promoted cement of hydration products (Gleize et al. 2007; Mazloom et al.
hydration in the presence of MK (da Cunha et al. 2008; Frias 2004; Tazawa and Miyazawa 1995; Tertnkhajornkit et al.
et al.; Guneyisi and Gesoglu 2008). 2005). Mineral admixtures may increase or decrease
the autogenous shrinkage of paste depending on the
Autogenous shrinkage
significance of each of the possible effects of SCM. The
The autogenous shrinkage of cementitious paste using autogenous shrinkage of pastes is also related with the
binary blend of SCM and cement is shown in Figure 3. materials, proportions and test method used for study.
This study did not focus on the mechanism underlying
the autogenous shrinkage behavior of paste affected by
SCMs. More research is needed to discern these effects.

Compressive strength
The compressive strength of cementitious paste using
binary blend of SCM and cement is shown in Figure 4.
As shown in Figure 4, at same level of SCM/c, the difference
in the compressive strength of paste using different types
of SCMs is very significant at the age of 1 day. For pastes
containing binary blend of cement and FA, the increase in
Fig. 3: Autogenous shrinkage of paste using binary blend of
the FA content from SCM/c=0 to SCM/c=0.3 continuously
SCM and cement

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

94 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of Supplementary Cementitious Materials on the Properties of Ultra-high Performance Concrete

in the SFU content. The highest 28-day compressive

strength can be achieved by using SFU alone as SCM is
152 MPa. The effect of MK on the 28-day compressive
strength does not have clear trend. The highest 28-day
compressive strength can be achieved by using MK alone
as SCM is 144 MPa (SCM/c=0.2).

Free drying shrinkage

The drying shrinkage of cementitious paste using binary
blend of SCM and cement is shown in Figure 5.
As shown in Figure 5, for pastes containing binary blend
of cement and FA, the increase in the FA content results
Fig. 4: Compressive strength of paste using binary blend of in increase in the drying shrinkage of paste. At the period
SCM and cement of exposure of 25 days, the drying shrinkage of paste
with SCM/c=0.2 is 20% higher than that of control. For
decreases the 1-day compressive strength. Data from pastes containing binary blend of cement and SFU, the
Figure 4 reveals that the 1-day compressive strength of increase in the SFU content also increases the drying
pastes with SCM/c=0.3 is 90% lower than that of control shrinkage of paste. However, the difference in the drying
(SCM/c=0.0). In this study, the observed reduction in the shrinkage of paste with SCM/c=0.1 and SCM/c=0.2 is not
1-day compressive strength due to the use of FA is in significant. Data from Figure 5 reveals that, at a period of
accordance with findings from previous literature, and exposure of 25 days, the drying shrinkage of paste with
it was attributed to the low pozzolanic reactivity of FA SCM/c=0.1 and SCM/c=0.2 is 24% and 22% higher than
at early ages (Brooks et al. 2000; Fajun et al. 1985; r e that of control, respectively. The increased in the drying
2000; Guneyisi and Gesoglu 2008). For pastes containing shrinkage of paste due to the use of FA and SFU can be
binary blend of cement and SFU, there appears to be probably attributed to the increased capillary stress due
a threshold SFU content of SCM/c=0.2. When the SFU to the refined pore structure of concrete (Mokarem et
content increases from SCM/c=0 to SCM/c=0.2, the 1-day al. 2005). For pastes containing binary blend of cement
compressive strength is not affected significantly. The and MK, the increase in the MK content decreases the
1-day compressive strength of pastes with SCM/c=0.2 drying shrinkage of paste significantly. At the period of
is only 7% higher than that of control. However, when exposure of 25 days, the drying shrinkage of paste with
the SFU content is further increased from SCM/c=0.2 to SCM/c=0.2 is 43% lower than that of control. The effect
SCM/c=0.3, 1-day compressive strength is decreased by of MK in reducing the drying shrinkage of concrete has
30%. Likely the increased actual w/c, nucleation effect, been observed in previous literature (Guneyisi et al. 2012;
and the early pozzolanic reactivity of SFU contribute to Guneyisi et al. 2008; Guneyisi et al. 2010; LUO et al. 2011;
the observed phenomenon. For pastes containing binary Mermerdas et al. 2013), and it was mainly attributed to the
blend of cement and MK, the increase in the MK content reduced rate of water loss of concrete as the MK densified
continuously increases the 1-day compressive strength. the pore structure of concrete through pozzolanic reaction
Data from Figure 4 reveals that the 1-day compressive (LUO et al. 2011; Mermerdas et al. 2013).
strength of pastes with SCM/c=0.2 is 38% higher than
that of control. The increase in the early age compressive
strength of paste due to the use of MK was been observed
in the previous literature, and attributed to the promoted
cement hydration in the presence of MK (da Cunha et al.
2008; r e ; Guneyisi and Gesoglu 2008).
The use of SCM at proper proportions can improve the
28-day compressive strength of paste, compared with
the control. For pastes containing binary blend of FA
and cement, the 28-day compressive strength of paste
increases with the increase in the FA content up to SCM/
c=0.1, after which the 28-day compressive strength of Fig. 5: CDrying shrinkage of paste using binary blend of SCM
paste decreases with the increase in the FA content. The and cement
highest 28-day compressive strength can be achieved by
using FA alone as SCM is 150 MPa. For pastes containing Discussion
binary blend of SFU and cement, the 28-day compressive Based on the experimental test results presented above,
strength of paste increases with the increase in the it is recognized that the use of FA and the use of MK
SFU content up to SCM/c=0.2, after which the 28-day have distinct effects on the workability, time of set, 1-day
compressive strength of paste decreases with the increase

Organised by
India Chapter of American Concrete Institute 95
Session 1 B - Paper 4

Table 5
Properties of paste using ternary blend of MK, FA and cement

 MF1 MF2 MF3 MF4 M3  F2 S2  S3

Workability (Flow, %) 269 239 258 239 181 289 238 175

1-day comp. str. (MPa) 61 59 38 32 106 26 72 58

28-day comp. str. (MPa) 145 145 149 136 134 147 152 119

25-day dry. shrinkage (%) -0.1557 -0.1490 -0.1570 -0.1407 -0.0853 -0.1787 -0.1820 -0.1680

compressive strength and drying shrinkage of paste. It Development Of Uhpc

is anticipated that the combined use of MK and FA may
As discussed above, pastes MF2 and MF3 have good
compensate the reduction in the early age compressive
workability, high 1-day and 28-day compressive strength
strength, prolonged time of set and increased drying
and relatively low drying shrinkage. They were selected to
shrinkage of paste due to the use of FA, and may
develop UHPCs by adding proper amount of sand and SMF.
compensate the reduction in the workability of paste due
Paste S2 was also used to produce UHPC for comparison.
to the use of MK. The use of ternary blend of MK, FA and
For all of these three UHPC mixtures, the s/cm was 1.25
cement may produce pastes with properties as good as
and the SMF content was 2% by volume of total UHPC
or even better than paste using binary blend of SFU and
mixture. The mixture proportions for 1 m3 of UHPC are
cement. The properties of paste using ternary blend of MK,
shown in Table 6.
FA and cement is presented in the later part of this paper.

Paste Using Ternary Blend of MK, FA and Table 6

Quantities of materials used for 1 m3 of UHPC
Cement
The properties of pastes using ternary blend of MK, FA Paste Constituent (kg/m3)
and cement are presented in Table 5. The properties of ID Cement SFU MK FA Sand SMF Water HRWRA
pastes using binary blend of SCM and cement are also UHPC1 808 0 81 81 1212 156 194 9.7
presented in Table 5 for comparison at the same SCM
content of SCM/c=0.2 and SCM/c=0.3. UHPC2 743 0 37 186 1208 156 193 9.7

As shown in Table 5, the investigated properties of pastes UHPC3 804 161 0 0 1206 156 193 9.6
MF1 and MF2 fall into the middle of that of pastes M3 and
F2, which illustrate that the use of ternary blend of MK, FA As shown in Table 7, at the same SCM content, UHPC
and cement can compensate the shortcomings of using prepared with ternary blend of MK, FA and cement exhibits
binary blend of MK and cement or binary blend of FA and lower 1-day and 28-day compressive strength than UHPC
cement. prepared with binary blend of SFU and cement, which
is evident by comparing UHPC1 and UHPC3. However,
It is noted that, at the SCM content of SCM/c=0.2, pastes
UHPC1 exhibits significantly higher workability and lower
MF1 and MF2 present lower compressive strength than
drying shrinkage than UHPC3. It is noted that UHPC 2 has
paste S2 at the ages of 1 and 28 days. The 1-day compressive
28-day compressive strength of 150 MPa, which indicates
strength of pastes MF1 and MF2 is 27% and 29% lower than
that UHPC mixture can be produced by using ternary
that of paste S2, respectively. The 28-day compressive
blend of MK, FA and cement.
strength of pastes MF1 and MF2 is same, which is 5%
lower than that of paste S2. However, pastes MF1 and
MF2 present higher workability and lower 25-day drying
Table 7
shrinkage than paste S2. At the SCM content of SCM/c=0.3, Properties of UHPC
pastes MF3 and MF4 present lower compressive strength
than paste S3 at the age of 1 day. The 1-day compressive UHPC1 UHPC2 UHPC3
strength of pastes MF3 and MF4 is 34% and 44% lower
Workability (Flow, %) 144 150 106
than that of paste S3, respectively. However, pastes MF3
and MF4 present higher compressive strength than paste 1-day compressive strength
66 58 74
(MPa)
S3 at the age of 28 days. The 28-day compressive strength
of pastes MF3 and MF4 is 26% and 14% higher than that of 28-day compressive strength
142 150 160
(MPa)
paste S3, respectively. It is worthy to mention that pastes
MF1 and MF2 present higher workability and lower 25-day 25-day drying shrinkage (%) -0.0390 -0.0520 -0.0692
drying shrinkage than paste S3.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

96 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of Supplementary Cementitious Materials on the Properties of Ultra-high Performance Concrete

Conclusions 5. UHPC mixture can be prepared by combined use of MK

and FA as SCM, instead of using SFU alone as SCM in
In this study, the effects of supplementary cementitious
the paste formulation. This research shows that the
materials on several properties of paste fraction of
ternary blend of MK and FA with cement can be used
UHPC were studied. UHPC mixtures were developed by
to meet the formal requirements of UHPC, i.e. greater
using ternary blend of MK, FA and cement. Based on the
than 150 MPa compressive strength.
materials and proportions used in this study, the following
conclusions are drawn:
References
1. The use of FA in paste does not have noticeable effect 1. Bai, J., Sabir, B., Wild, S., and Kinuthia, J. (2000). "Strength
on the workability of paste. However, it does retard the development in concrete incorporating PFA and metakaolin."
setting behavior of paste. The increase in the FA content Magazine of concrete research, 52(3), 153-162.
results in decrease in the 1-day compressive strength 2. Brooks, J. J., Johari, M. A. M., and Mazloom, M. (2000). "Effect of
admixtures on the setting times of high-strength concrete." Cement
of paste. The highest 28-day compressive strength
Concrete Comp, 22(4), 293-301.
of paste using binary blend of FA and cement is 150
3. Brooks, J. J., and Megat Johari, M. A. (2001). "Effect of metakaolin on
MPa which is achieved at FA content of SCM/c=0.1. The creep and shrinkage of concrete." Cement and Concrete Composites,
increase in the FA content also results in increased 23(6), 495-502.
autogenous shrinkage and drying shrinkage of paste. 4. Cheyrezy, M., Maret, V., and Frouin, L. (1995). "Microstructural
Analysis of Rpc (Reactive Powder Concrete)." Cement and Concrete
2. The use of SFU in paste above a certain threshold level Research, 25(7), 1491-1500.
decreases the workability of paste. There appears to 5. da Cunha, A. L. C., Goncalves, J. P., Buchler, P. M., and Dweck, J.
be a threshold SFU content of SCM/c=0.1, below which (2008). "Effect of metakaolin pozzolanic activity in the early stages of
the use of SFU shows prolongs the setting behavior cement type II paste and mortar hydration." J Therm Anal Calorim,
of paste. When SFU content is higher than SCM/ 92(1), 115-119.
c=0.1, the time of set is decreased as the SFU content 6. Fajun, W., Grutzeck, M. W., and Roy, D. M. (1985). "The retarding
effects of fly ash upon the hydration of cement pastes: The first 24
increases. SFU content does not have significant effect hours." Cement and Concrete Research, 15(1), 174-184.
on the 1-day compressive strength up to a content of 7. Frıas, M., de Rojas, M. I. S., and Cabrera, J. (2000). "The effect that
SCM/c=0.2. When SFU content is higher than SCM/ the pozzolanic reaction of metakaolin has on the heat evolution in
c=0.2, the 1-day compressive strength decreases metakaolin-cement mortars." Cement and Concrete Research, 30(2),
as the SFU content increases. The highest 28-day 209-216.
compressive strength of paste using binary blend of 8. Gleize, P. J. P., Cyr, M., and Escadeillas, G. (2007). "Effects of
metakaolin on autogenous shrinkage of cement pastes." Cement
SFU and cement is 152 MPa which is achieved at SFU Concrete Comp, 29(2), 80-87.
content of SCM/c=0.2. Among all the twelve paste 9. Graybeal, B. A. (2006). "Material property characterization of ultra-
mixtures investigated in this study, the highest 28-day high performance concrete." Report FHWA-HRT-06-103, FHWA, U.S.
compressive strength which is 152 MPa is achieved by Department of Transportation.
using SFU at a content of SCM/c=20%. The increase 10. Guneyisi, E., and Gesoglu, M. (2008). "Properties of self-compacting
in the SFU content results in decreased autogenous mortars with binary and ternary cementitious blends of fly ash and
metakaolin." Mater Struct, 41(9), 1519-1531.
shrinkage and increased drying shrinkage of paste.
11. Guneyisi, E., Gesoglu, M., Karaoglu, S., and Mermerdas, K. (2012).
3. MK has the most significant effect in reducing the "Strength, permeability and shrinkage cracking of silica fume and
workability, decreasing the time of set, and increasing metakaolin concretes." Constr Build Mater, 34, 120-130.
the 1-day compressive strength among FA and SFU. 12. Guneyisi, E., Gesoglu, M., and Mermerdas, K. (2008). "Improving
strength, drying shrinkage, and pore structure of concrete using
The highest 28-day compressive strength of paste
metakaolin." Mater Struct, 41(5), 937-949.
using binary blend of MK and cement is 144 MPa which
13. Guneyisi, E., Gesoglu, M., and Ozbay, E. (2010). "Strength and drying
is achieved at MK content of SCM/c=0.2. The use of shrinkage properties of self-compacting concretes incorporating
MK results in increased autogenous shrinkage and multi-system blended mineral admixtures." Constr Build Mater,
decreased drying shrinkage. 24(10), 1878-1887.
14. kDurekovic, A. (1995). "Cement pastes of low water to solid ratio: An
4. The combined use of MK and FA compensates the investigation of the porosity characteristics under the influence of a
reduction in the 1-day compressive strength and superplasticizer and silica fume." Cement and Concrete Research,
increase in the drying shrinkage of paste due to the 25(2), 365-375.
use of FA, and compensates the reduction in the 15. Khatib, J. M., and Wild, S. (1996). "Pore size distribution of metakaolin
workability of paste due to the use of MK. At the same paste." Cement and Concrete Research, 26(10), 1545-1553.
SCM content, the use of ternary blend of MK, FA and 16. Kinuthia, J., Wild, S., Sabir, B., and Bai, J. (2000). "Self-compensating
autogenous shrinkage in Portland cement—metakaolin—fly ash
cement can produce paste with higher workability and pastes." Advances in cement research, 12(1), 35-43.
lower 25-day drying shrinkage than paste using binary 17. Li, Z., Kizhakommudom, H., Rangaraju, P. R., and Schiff, S. D.
blend of SFU and cement. The compressive strength "Development of ultra-high performance concrete for shear-
of paste using ternary blend of MK, FA and cement is keys in precast bridges using locally available materials in South
slightly lower than that of paste using binary blend of Carolina." Proc., 2014 60th Anniversary Convention and National
Bridge Conference, Precast/Prestressed Concrete Institute.
SFU and cement.

Organised by
India Chapter of American Concrete Institute 97
Session 1 B - Paper 4

18. LUO, J., LU, D., XU, T., and XU, Z. (2011). "Effect of Metakaolin on Concrete: A State-of-the-Art Report for the Bridge Community."
Drying Shrinkage Behaviour of Portland Cement Pastes and its FHWA-HRT-13-060.
Mechanism."Journal of The Chinese Ceramic Society, 39(10), 1687- 30. Sabir, B. B., Wild, S., and Bai, J. (2001). "Metakaolin and calcined
1693. clays as pozzolans for concrete: a review." Cement and Concrete
19. Mazloom, M., Ramezanianpour, A. A., and Brooks, J. J. (2004). "Effect Composites, 23(6), 441-454.
of silica fume on mechanical properties of high-strength concrete." 31. Tafraoui, A., Escadeillas, G., Lebaili, S., and Vidal, T. (2009).
Cement and Concrete Composites, 26(4), 347-357. "Metakaolin in the formulation of UHPC." Constr Build Mater, 23(2),
20. Mermerdas, K., Guneyisi, E., Gesoglu, M., and Ozturan, T. (2013). 669-674.
"Experimental evaluation and modeling of drying shrinkage behavior 32. Tazawa, E.-i., and Miyazawa, S. (1995). "Influence of cement and
of metakaolin and calcined kaolin blended concretes." Constr Build admixture on autogenous shrinkage of cement paste." Cement and
Mater, 43, 337-347. Mindess, S., Young, J. F., and Darwin, D. (2003). Concrete Research, 25(2), 281-287.
Concrete, Prentice Hall, Upper Saddle River, NJ.
33. Tertnkhajornkit, P., Nawa, T., Nakai, M., and Saito, T. (2005). "Effect
21. Mokarem, D. W., Weyers, R. E., and Lane, D. S. (2005). "Development of fly ash on autogenous shrinkage." Cement and Concrete Research,
of a shrinkage performance specifications and prediction model 35(3), 473-482.
analysis for supplemental cementitious material concrete mixtures."
Cement and Concrete Research, 35(5), 918-925. 34. Wild, S., Khatib, J., and Roose, L. (1998). "Chemical shrinkage and
autogenous shrinkage of Portland cement—metakaolin pastes."
22. Ping, X., and Beaudoin, J. J. (1992). "Modification of transition zone Advances in cement research, 10(3), 109-119.
microstructure—Silica fume coating of aggregate surfaces." Cement
and Concrete Research, 22(4), 597-604. 35. Wild, S., Khatib, J. M., and Jones, A. (1996). "Relative strength,
pozzolanic activity and cement hydration in superplasticised
23. Poon, C. S., Kou, S. C., and Lam, L. (2006). "Compressive strength, metakaolin concrete." Cement and Concrete Research, 26(10), 1537-
chloride diffusivity and pore structure of high performance metakaolin
1544.
and silica fume concrete." Constr Build Mater, 20(10), 858-865.
36. Wille, K., and Boisvert-Cotulio, C. (2013). "Development of Non-
24. Poon, C. S., Lam, L., Kou, S. C., Wong, Y. L., and Wong, R. (2001). "Rate
Proprietary Ultra-High Performance Concrete for Use in the Highway
of pozzolanic reaction of metakaolin in high-performance cement
Bridge Sector." FHWA-HRT-13-100.
pastes." Cement and Concrete Research, 31(9), 1301-1306.
37. Wille, K., Naaman, A. E., and Parra-Montesinos, G. J. (2011). "Ultra-
25. Randl, N., Steiner, T., Ofner, S., Baumgartner, E., and Meszoly, T.
High Performance Concrete with Compressive Strength Exceeding
(2014). "Development of UHPC mixtures from an ecological point of
150 MPa (22 ksi): A Simpler Way." Aci Mater J, 108(1), 46-54.
view." Constr Build Mater, 67, 373-378.
38. Yoo, S. W., Kwon, S. J., and Jung, S. H. (2012). "Analysis technique for
26. Rangaraju, P. R., Kizhakommudom, H., Li, Z., and Schiff, S. D. (2013).
autogenous shrinkage in high performance concrete with mineral and
"Development of High-Strength/High Performance Concrete/Grout
chemical admixtures." Constr Build Mater, 34, 1-10.
Mixtures for Application in Shear Keys in Precast Bridges." FHWA-
SC-13-04a, FHWA, U.S. Department of Transportation. 39. Yu, R., Spiesz, P., and Brouwers, H. J. H. (2015). "Development of an
eco-friendly Ultra-High Performance Concrete (UHPC) with efficient
27. Richard, P., and Cheyrezy, M. (1995). "Composition of Reactive Powder
cement and mineral admixtures uses." Cement Concrete Comp, 55,
Concretes." Cement and Concrete Research, 25(7), 1501-1511.
383-394.
28. Roy, D. M., Gouda, G. R., and Bobrowsky, A. (1972). "Very high strength
40. Zhang, M. H., and Malhotra, V. M. (1995). "Characteristics of a
cement pastes prepared by hot pressing and other high pressure
Thermally Activated Aluminosilicate Pozzolanic Material and Its
techniques." Cement and Concrete Research, 2(3), 349-366.
Use in Concrete." Cement and Concrete Research, 25(8), 1713-1725.
29. Russell, H. G., and Graybeal, B. A. (2013). "Ultra-High Performance

Dr. Prasada Rao Rangaraju

Dr. Rangaraju is a Professor of Civil Engineering and the Director of SMaRT center at the Glenn Department
of Civil Engineering of Clemson University in South Carolina, USA. He has been a faculty member at
Clemson for last 15 years. Before joining Clemson, Dr. Rangaraju was a senior research engineer at Office
of Materials and Road Research at Minnesota Department of Transportation. Dr. Rangaraju received his
PhD from Purdue University and is a registered professional engineer. Dr. Rangaraju’s research interests
are in the material science and engineering of cementitious materials and concrete, durability of concrete
and high performance concrete mixtures. He has published extensively on these subjects and he is also an
active member of several technical committees in ACI, ASTM, TRB and a member of ASCE.

Zhengqi Li
Zhengqi Li is a PhD candidate in the Glenn Department of Civil Engineering at Clemson University. He
received his B.S. in Civil Engineering (2009) and M.S. in Structural Engineering (2012) from Tongji University
in China. His current research interests are in the area of Ultra-High Performance Concrete with emphasis
on material development, modeling and structural applications of UHPC and high-performance concrete
mixtures. His previous research experience were in the area of punching shear behavior and modeling of
slab made of recycled aggregate concrete (M.S. thesis). He has authored seven journal papers and five
conference papers, and he is a member of ACI technical committees 130, 209, 239.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

98 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber Reinforced Concrete Pavement Bangalore-Mysore Corridor-NICE Project-- India’s Longest Whitetopped Road

Fiber Reinforced Concrete Pavement

Bangalore-Mysore Corridor-NICE Project-- India’s Longest
Whitetopped Road
K.R.S. Narayan
National Head-Recron3S Business-Reliance Industries Ltd.-Mumbai

Abstract than ordinary cement concrete pavement. “FRC is defined

as composite material consisting of concrete reinforced
Road transportation is undoubtedly the lifeline of the
with discrete randomly but uniformly dispersed short
nation and its development is a crucial concern. The
length fibers.” The fibers may be of STEEL, SYNTHETIC
traditional bituminous pavements and their needs for
or natural materials. FRC is considered to be a material
continuous maintenance and rehabilitation coupled with
of improved properties and not as reinforced cement
frequent repairs, points towards the scope for cement
concrete whereas reinforcement is provided for local
concrete pavements. There are several advantages of
strengthening of concrete in tension region. Fibers
cement concrete pavements over bituminous pavements.
This paper explains benefits of FIBRE REINFORCED generally used in cement concrete pavements are Steel
CONCRETE PAVEMENTS, which is a recent advancement fibers and organic Synthetic Polymer fibers such as
in the field of Reinforced Concrete Pavement design with Polyester or Polypropylene.
a Case Study of NICE ROAD-Bangalore which is India’s
Longest Whitetopped Road using Recron3S Fibers is Fiber Reinforced Concrete
presented. Concrete is well known as a brittle material when subjected
to normal stresses and impact loading, especially, with its
Keywords: Pavements; Recron3S Fibers; Concrete;
tensile strength being just one tenth of its compressive
Reinforcement.
strength. It is only common knowledge that, concrete
members are reinforced with continuous reinforcing bars
Introduction to withstand tensile stresses, to compensate for the lack
In a developing country such as India, road networks of ductility and is also adopted to overcome high potential
form the arteries of the nation. A pavement is the layered tensile stresses and shear stresses at critical location in
structure on which vehicles travel. It serves two purposes, a concrete member.
namely, to provide a comfortable and durable surface
for vehicles, and to reduce stresses on underlying soils. Even though the addition of steel reinforcement
In India, the traditional system of bituminous pavements significantly increases the strength of the concrete, the
is widely used. Locally available cement concrete is a development of micro-cracks must be controlled to
better substitute to bitumen which is the by product in produce concrete with homogenous tensile properties.
distillation of imported petroleum crude. It is a known fact The introduction of fibers was brought into consideration,
that petroleum and its by-products are getting doomed as a solution to develop concrete with enhanced flexural
day by day. Whenever we think of a road construction in and tensile strength, which is a new form of binder that
India it is taken for granted that it would be a bituminous could combine Portland cement in bonding with cement
pavement and there are very rare chances for thinking of matrices.
an alternative like concrete pavements. Within two to three Fibers are generally discontinuous, randomly distributed
decades bituminous pavement would be a history and thus throughout the cement matrices. Referring to the
the need for an alternative is very essential. The perfect American Concrete Institute (ACI) committee 544 , in fiber
solution would be SYNTHETIC FIBER REINFORCED reinforced concrete there are four categories namely
CONCRETE PAVEMENTS, as it satisfies two of the much
1. SFRC - Steel Fiber Reinforced Concrete
demanded requirements of pavement material in India,
economy and reduced pollution. It also has several other 2. GFRC - Glass Fiber Reinforced Concrete
advantages like longer life, low maintenance cost, fuel
3. SNFRC - Synthetic Fiber Reinforced Concrete
efficiency, good riding quality, increased load carrying
capacity and impermeability to water over flexible 4. NFRC - Natural Fiber Reinforced Concrete
pavements.
Fiber Reinforced concrete can be defined as a composite
Fiber Reinforced Concrete Pavements are more efficient material consisting of mixtures of cement, mortar or

Organised by
India Chapter of American Concrete Institute 99
Session 1 B - Paper 5

concrete with discontinuous, discrete, uniformly dispersed It enables easier and smoother finishing. It also helps
suitable fibers. Continuous meshes, woven fabrics and to achieve reduced bleeding of water to surface during
long wires or rods are not considered to be discrete fibers. concrete placement, which inhibits the migration of cement
and sand to the surface and the benefits of the above will
Fiber reinforced concrete (FRC) is concrete containing
be harder, more durable surface with better abrasion
fibrous material which increases its structural integrity. It
resistance. A uniform distribution of fibers throughout
contains short discrete fibers that are uniformly distributed
the concrete improves the homogeneity of the concrete
and randomly oriented. Fibers may generally be classified
matrix. It also facilitates reduced water absorption,
into two: organic and inorganic. Inorganic fibers include
greater impact resistance, enhanced flexural strength
Steel fibers and Glass fibers, whereas organic fibers
and tensile strength of concrete. The use of polymeric
include natural fibers like coconut, sisal, wood, bamboo,
fibers with concrete has been recognized by the Bureau
jute, sugarcane, etc and synthetic fibers based on Acrylic,
of Indian Standards (BIS) and Indian Road Congress and is
Carbon, Polypropylene, Polyethylene, Nylon, Aramid, and
included in the following Standard documents:
Polyester. Within these different fibers the character of
fiber reinforced concrete changes with varying concretes, IS:456:2000 – Amendment No.7, 2007
fiber materials, geometries, distribution, orientation and
IRC:44-2008 – Cement Concrete Mix Designs for
densities.
Pavements with fibers
Fibers are usually used in concrete to control cracking
IRC:SP 46-2013—Guidelines for Design & Construction
in its’ Plastic and Drying states. They also lower the
of Fiber Reinforced Concrete Pavements.
permeability of concrete and thus reduce bleeding of
water. Some types of fibers produce greater impact, ICI-Indian Concrete Institute-Technical Committee
abrasion and shatter resistance in concrete. recommendation for Fiber Reinforced Concrete-TC-01
The amount of fibers added to a concrete mix is measured IRC:SP:76:2008 – Guidelines for Ultra Thin White
as a percentage of the total volume of the composite Topping with fibers
(concrete and fibers) termed volume fraction (Vf). Vf
MOST-Section 600 -- Ministry of Surface Transport,
typically ranges from 0.1 to 3%. Aspect ratio (l/d) is
New Delhi.
calculated by dividing fiber length (l) by its diameter (d).
Fibers with a non-circular cross section use an equivalent Polymeric Fiber Reinforced concrete has been approved
diameter for the calculation of aspect ratio. If the modulus by National bodies like:
of elasticity of the fiber is higher than the matrix (concrete
or mortar binder), they help to carry the load by increasing 1. Central Public Works Department (CPWD) & Local
the tensile strength of the material. Fibers which are State PWDs.
too long tend to “ball” in the mix and create workability 2. Airport Authority of India
problems.
3. Military Engineering Services
Synthetic Fiber Reinforced Concrete (SFRC) 4. Defence Airfields
Polymeric fibers are gaining popularity because of 5. Railways.
its properties like zero risk of corrosion and cost
effectiveness. The polymeric fibers commonly used are CASE STUDY of India’s Longest Whitetopped
Polyester and Polypropylene.
Concrete Road using Fibers
These fibers act as crack arresters, restricting the
Introduction
development of cracks and thus transforming a brittle
material into a strong composite with superior crack Project—Bangalore-Mysore Infrastructure Corridor
resistance, improved ductility and distinctive post cracking Project (BMICP) – NICE Ltd.-Nandi Infrastructure
behavior prior to failure. Corridor Enterprises Limited.
Concrete pavements may be weak in tension and against
impact, but PFRC is a suitable material which may be Project Salient features
used for cement concrete pavement as it possesses extra ll First White Topped Project in India – Completed in May’
strength in flexural fatigue and impact etc. The usage 2013.
of fibers in combination with concrete also results in a
ll Longest White topped Concrete Road in India—90
mix with improved early resistance to plastic shrinkage
Lane Km.
cracking and thereby protects the concrete from drying
shrinkage cracks. It accomplishes improved durability ll First White topped project under PPP model.
and reduced surface water permeability of concrete. It
ll Designed for 489 MSA.
reduces the risk of plastic settlement cracking over rebar.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

100 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber Reinforced Concrete Pavement Bangalore-Mysore Corridor-NICE Project-- India’s Longest Whitetopped Road

ll Design Life of 60 years. Sequence of Paving Operations

ll Project using Fibers for entire White topping of approx.
72000 Cu.M of Concrete.
ll Project completed with least time & cost overrun,
completed in 120 days using RMC.
ll Concrete Pavement is light in colour adding to
luminosity during night thereby saving Energy.
ll Project executed without cutting a single tree & saving
environment.
ll Albedo effect—Surface being light in colour compared
to Bitumen roads reduces surface temperature (Heat
Island effect) having substantial effect in less heat
radiation to surroundings.
ll Savings in Natural Aggregates as Design life is 60
years. Milling of the existing asphalt pavement

Design & Construction features

ll Composite Rigid Pavement Construction with M-40
grade of RMC with designed Flexural Strength of 5.2
N/mm2.
ll Use of Fiber Reinforced M-40 grade of Concrete for
entire 72000 Cu.M with the aid of Site based RMC plant
of high capacity of 120 Cu.M/Hour.
ll Highly mechanized Paver operation used for laying
concrete.
ll Effective use of Curing compound replacing
conventional water curing system for concrete.
ll No use of Dowel bars/reinforcements for the entire 90
lane km.
ll Laying of concrete with joints/grooves of 1m X 1m
panels. Brooming milled surface by compressed air

Mix Design Details

PARAMETER M-40 ( OPC + Fibers )

Cement 430 Kgs

20mm Coarse Aggregates 681 Kgs

12mm Coarse Aggregates 435 Kgs

Manufactured Sand 739 Kgs

Water 160 Kgs

Trademark PP Fibers 0.90 Kgs

Admixture “Glenium ACE 30” 0.8 – 1.0%

Water / Binder Ratio 0.37

Unit Weight 2445Kgs Laying concrete and finishing the surface

Organised by
India Chapter of American Concrete Institute 101
Session 1 B - Paper 5

Construction Machinery

Construction Machinery
Milling Milling machine (W 100) Wirtgen make - 3 nos.;
Tippers - 6 nos.
Cleaning Mechanical Broomer - 3 nos. Air Compressor - 6
nos. & by manual means
Production of Dedicated Batching plant (100 Cum/hr capacity) - 1
Concrete no. & supporting with 3 nos. of 1 Cum. capacity
Transporting Tippers (8 Cum) - 30 nos. & Transit Mixers for
Manual Concreting
Paving Slip form Paver (SP-500) Wirtgen make - 1 no.;
Slip form Paver (SP - 850) Wirtgen make - 2 nos.;
Screed Vibrator & Needle Vibrator - 10 sets (for
manual concreting)
Spreading of Front Hoe & Loader (JCB) - 3 nos.
Concrete

Texturing the finished surface Texturing and TCM (SP - 850) - 2 nos.; Hand operated Texturing
Curing Combs & Floaters - 10 nos. (For SP - 500 & Manual
concreting); Hand operated mechanical sprayers -
10 nos. (For SP - 500 & Manual concreting)

Pavement Design
The base coarse of Dry Lean Concrete (DLC) serves as
working platform for supporting PFRC slabs which by
slab action distributes the wheel load to larger area. The
DLC base layer rests on granular sub-base which rest on
sub grade.
Over the well compacted sub grade Granular Sub base
is constructed using big stone boulders and mud. Over
that the Dry Lean Concrete of mix 1:4:8 is made, which
is compacted, leveled and floated. Surface of DLC is also
corrected for road camber. An antifriction separation
membrane of 125 micron thickness is spread over the
DLC surface so as to impart free movement of the upper
Groove cutting slab caused due to temperature warping stresses. The
separation membrane may be stuck to the lower layer
with patches of adhesives or appropriate tape or concrete
nails with washer so that polythene sheet does not move
during placement of concrete.
Many of the thickness design methods for cement concrete
pavement adopted internationally derive their origin from
the method evolved by Portland Cement Association (PCA).
In this technology thickness of the pavement is assumed
on trial basis. When dewatered concrete is provided on
lean concrete, it has no problem of water being coming
out on surface during compaction process but when
it is done over WBM, a considerable amount of water is
soaked by WBM and thus the concrete loses the water to
WMB and the water which comes out during dewatering/
compaction process is not in same quantity as in case of
lean concrete. It appears that it is better to provide base
concrete than WBM as the base.
Due to repeated application of flexural stresses by the
Covering concrete slabs with Hessian cloth traffic loads, progressive fatigue damage takes place in

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

102 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber Reinforced Concrete Pavement Bangalore-Mysore Corridor-NICE Project-- India’s Longest Whitetopped Road

the cement concrete slab in the form of cracks especially Protection and maintenance
when the applied stress in terms of flexural strength of
The joint groove is to be protected from ingress of dirt or any
concrete is high. The ratio between the flexural stress due
foreign matter by inserting performed neoprene sealant.
to the load and the flexural strength is termed as the stress
To exercise a very stringent quality control the test are to
ratio (SR). If the SR is less than 0.45, the concrete pavement
be conducted on fine and coarse stone aggregates, water
is expected to sustain infinite number repetitions. As the
cement, granular sub base, DLC etc as per standards and
SR decreases the no. of load repetitions required to cause
specification published by Indian roads congress.
cracking increases. This is also considered in the design
of pavement. Advantages and disadvantages
Polymeric fibers normally used are either Polyester Advantages
Polypropylene. It should be 100% virgin synthetic fiber
(1) Water logging is a major reason for potholes in roads.
size 12mm long and of 8 to 15 denier in diameter. It shall
WBM and Asphalt roads are permeable to water which
be mixed at the rate of 900gms per cum of concrete.
damages the road and sub grade. But PFRC roads are
These fibers reduce plastic shrinkage and substance
highly impermeable to water so they will not allow
cracking. Use of Fibers increases the toughness and post
water logging and water being coming out on surface
cracking integrity. Polypropylene is one of the cheapest
from sub grade.
and abundantly available polymers. Polypropylene fibers
are resistant to most chemical attacks. Its melting point (2) Implementation of sensors in roads will be easier
is high (about 165 degrees centigrade). So that it can while using polymer fibers for concrete.
withstand a working temp, as (100 degree centigrade) (3) Environmental load of PFRC pavement was found to
for short periods without any detrimental effect to its’ be significantly lower than the steel fiber reinforced
properties. Polypropylene fibers being hydrophobic and pavement.
can be easily mixed. Polypropylene short fibers in small (4) Maintenance activities related to steel corrosion will
volume fractions between 0.1% to 0.6% are commercially be reduced while using PFRC.
used in concrete. (5) In fresh concrete polymer fibers reduces the
settlement of aggregate particles from pavement
surface resulting in an impermeable and more
durable, skid resistant pavement.
(6) Fibers reduce plastic shrinkage and substance
cracking. Fibers also provide residual strength after
occurrence of cracking.
(7) The use of PFRC produces concrete of improved
abrasion resistance and impact resistance.
(8) PFRC also enhances ductile and flexural toughness of
concrete.
(9) All these advantages result in overall improved
Requirements for paving operations DURABILITY of PAVEMENT & ensuring design life with
drastic reduction in long term maintenance costs.
(1) Use of microfilm or antifriction layer of 125 micron in
between PFRC and DLC layers. Disadvantages
(2) The DLC layer is to be swept clean of all the extraneous (1) The use of PFRC, being a relatively new technology
materials before applying microfilm which may be poses a threat of a high initial cost of construction.
nailed to the DLC layer without wrinkles and holes.
Applications of PFRC
(3) Concreting work in hot weather should be carried out 1. Slab On Grade: All types of pavements and overlays,
in early or later hours. industrial floors, roads, taxi ways, hangars, etc.
(4) The laying temperature of concrete should always be 2. Structural Concrete: Foundations (deep and shallow),
below 35degreeCelsius. machine foundation, slabs, column beams and lintel,
bridge decks and girders etc.
Curing
3. Water retaining Structures: RCC retaining walls, water
Membrane curing is applied with the help of texture-cum- tanks, cross drains, swimming pools, hydel projects,
curing machine. The resin based curing compound is used check dams, canal lining, ETPs, jetties, ports, spillways
at the rate of 300 ml per square meter of the slab area. etc.
After about 1.5 hours moist Hessian cloth is spread over 4. Water proofing in rooftops, sunken toilets, etc.
the surface covered with curing compound spray.

Organised by
India Chapter of American Concrete Institute 103
Session 1 B - Paper 5

Conclusion References
1. Banthia, N. and Mindess S., Polyester Fiber as Concrete
PFRC can be used advantageously over normal concrete Reinforcement: A Second Look, FIBCON 2012: Conference on Fiber
pavement. Polymeric fibers such as polyester or Reinforced Concrete Global Developments, Ramdaspeth, Nagpur,
polypropylene are being used due to their cost effective India, Dec 13-14, 2012.
as well as corrosion resistance. PFRC requires specific 2. Banthia, N., A Comprehensive Study of Polyester Fiber as Concrete
design considerations and construction procedures Reinforcement, Inaugural ‘R N Raikar Memorial International
to obtain optimum performance. The higher initial Conference’ - Dr. Suru Shah Symposium on ‘Advances in Science
and Technology of Concrete’, Mumbai, India, December 20-21, 2013
cost by 15-20% is counterbalanced by the reduction in
maintenance and rehabilitation operations, making 3. Katz A., Nov/Dec 2004, “Environmental impact of steel and FRP
reinforced polymer”, Journal for composite for construction vol 8
PFRC cheaper than flexible pavement by 30-35%. In a no.6 pp 48-488
fast developing and vast country like India, road networks
4. Kenneth G. Budhinski, Michel K. Budhinski,” Engineering materials-
ensure mobility of resources, communication and in Properties & selection”, 8th edition, Prentice Hall India, pp 194-195
turn contribute to growth and development. Resistance 5. Krishna G., July 2007,”Key role of chemical admixtures for pavement
to change though however small disturbs our society; quality concrete”, NBM&BW vol 13, pp166-169.
hence we are always reluctant to accept even the best. 6. Reis J.M.L., Nov 2006,”Fracture and flexure characterization of
It’s high time that we overcome the resistance and reach natural fibers-reinforced polymer concrete” Construction and
for the peaks. PFRC opens a new hope to developing Building Materials vol 20 pp 673-678
and globalizing the quality and reshaping the face of the 7. Seehra S.S., March 2007,” An Innovative concrete technological
“True Indian Roads”. development of fully mechanized construction of cement concrete
pavement”, NBM&BW vol 12 pp76-93
8. Soni K.M, May 2007, “Fiber Reinforced Concrete in Pavements”,
Acknowledgement NBM&CW vol 12, pp 178-181.
The Author wishes to thankfully acknowledge support, 9. Agrawal B.K., Indroduction to Engineering Materials”, 4th edition,
cooperation & technical inputs from NICE Site team & Tata Mc Grawhill Publishing company ltd, pp194-195
Ultratech RMC Plant / Lab staff. 10. CECR magazine-July’14. Article on CIDC-Vishwakarma Award for
Rigid Pavement construction technique.

Dr. KRS.Narayan
Dr. KRS.Narayan, a Post Graduate from The University of Leeds, U.K., has over 24 years of experience in
Concrete Engineering with specialization in the fields of Fiber Reinforced Concrete, RMC, Acid Resistant
coatings, FRP & Water Proofing. He is currently the National Head at Reliance Industries Ltd., Mumbai,
heading Market Development & Technical Services of Recron3S range of fibers.
He has been an active member of various committees of BIS & ICI and has been a key player in formulation
of the first draft code for FRC in India through the ICI. He is a Fellow of the Indian Concrete Institute, the
ACCE, and a member of the Indian Roads Congress.
He has published 23 papers in various Construction & Concrete related magazines, represented in over
50 National conferences as a Speaker, and has so far guided 38 Masters’ theses and also 2 students who
completed Ph.D. He is currently a Member of The IC-IMPACTS-India Canada Joint Committee headed by
the world famous Dr.Prof.Nemy Banthia.
He is a Corporate Advisory Board Member of the SRM University and is in the ABET certification committee
of the Civil Engineering group. His interests are in the fields of Fiber Reinforced Concrete Pavements/Fiber
Concrete/Quality improvement in RMC and training of upcoming Civil Engineers.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

104 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 1 C
Session 1 C - Paper 1

Repair of jetty damaged by cyclone – Case Study

Manish Mokal, Dr. N. V. Nayak
Gammon India Limited
Amit Datta
Chryso

Abstract impact on the Jetty. The Jetty was not designed for such
huge magnitude of forces during the cyclone.
Hudhud cyclone made landfall on 12th Oct 2014 noon at
Visakhapatnam coast with wind speed touching to 180 The waves during cyclone hit the Pump House at Intake
to 195 Kmph. The Intake and Outfall structure which is Well No. 1. The runoff due to obstruction caused an upward
being constructed at Vishakhapatnam was badly hit by the splashing force on the Pump house slab which eventually
cyclone. The waves generated during the cyclone made an cracked due to wave forces which acted in opposite
impact on the Jetty. The Jetty was not designed for such direction of the designed loading of the slab. Also some of
huge magnitude of forces during the cyclone. The detailed the beams in the Pump house got damaged along with the
assessment and strengthening methodology is discussed slab. The cracks in the beams can be visually seen. The
in detail in this paper. general layout of the project is shown in Fig. 1.
Keywords – Cyclone, jetty, high strength cementitious
mortar, epoxy grout, carbon fibre wrapping, ultrasonic
Initial investigation
pulse velocity test, core In order to access the extent of damage the following
evaluation techniques were undertaken:
Introduction 1. Visual inspection to identify the areas of damage (Fig. 2
Hudhud cyclone made landfall on 12th Oct 2014 noon at to 5)
Visakhapatnam coast with wind speed touching to 180 2. Conducting Ultrasonic Pulse Velocity Test on concrete
to 195 Kmph. The Intake and Outfall structure which is members to check the presence and extent of cracking
being constructed at Vishakhapatnam was badly hit by the 3. Conducting Schmidt Hammer Test on concrete
cyclone. The waves generated during the cyclone made an members to check the strength of concrete

Fig. 1: Sketch showing key plan showing main components of the project

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

106 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Repair of jetty damaged by cyclone – Case Study

In order to assess the extent of cracking in the beams

PLB 37, 38, 39 & 40 Ultrasonic Pulse Velocity Test (UPV)
using indirect method was conducted on all the affected
beams. The results indicated that the crack depth ranged
between 50 to 250 mm indicating severe damage to the
beams. When testing was done by direct method away
from cracks the UPV values are above 4km/sec. This
indicated that cracking was only at locations evident and
except at crack locations concrete quality elsewhere is
GOOD as per IS:13311 (Part:II).
Rebound Hammer Test was conducted to ascertain the
strength of concrete in the structure. The results generally
Fig. 2: Slab cracked due to upward splash in Intake Well No. 1 indicated concrete grade M40 and above both at good and
near crack locations. The rebound hammer readings
were generally taken 150mm away from the crack and at
locations where UPV probes were placed. Readings taken
exactly at crack locations showed very low values and
crushing of concrete and hence were not recorded. This
implies that cracking was likely only at crack locations
evident and may not have caused general deterioration of
concrete in lateral direction in beams.

Rectification Methodology & Execution

Various options were taken into consideration before
finalizing the rectification methodology:
ll Demolishing the beams and recasting. It was
not a feasible option due to many constraints in
Fig. 3: Damage to beam junction Intake Well No. 1
reconstructing the beams
ll Strengthening the beams with steel plates. This
system was rejected due to possibility of corrosion of
steel plates which would require regular maintenance
which is practically not possible
ll Strengthening the beams with carbon plates. This
system was selected as it is light weight and easy to
execute. There is no issue of corrosion of carbon plates.
Strict supervision can be exercised during execution to
ensure that the process is correctly done.
After detailed investigation was complete, the following
Fig. 4: Cracks in beams methodology was adopted for repair and strengthening:
ll The cracked slab to be completely removed and recast
after the repair of beams is completed
ll Clean all the cracks and damaged surface with
compressed air and water to remove loose material,
dust, grease, etc.
ll The wide cracks between beam and column junction
to be repaired with epoxy mortar. Before filling the
cracks with epoxy mortar, insert perforated nozzles in
the affected area.
ll After filing the cracks with epoxy mortar, grout the
nozzles with low viscosity epoxy grout to ensure that
all voids are completely filled.
Fig. 5: Spalling of bottom concrete portion of beam due to ll All other cracks to be grouted with low viscosity epoxy
splashing of water grout.

Organised by
India Chapter of American Concrete Institute 107
Session 1 C - Paper 1

Fig. 6: UPV test before repairs Fig. 9: Sealing of cracks in progress

ll Grind the complete surface of beam to get a smooth

surface. Any depressions to be filled with epoxy mortar
and made smooth. Grind all sharp edges into round
surface. This is to ensure maximum contact area of
old concrete with carbon fibre laminates and carbon
fibre wrap.

ll Recheck the grouted area again with UPV for checking

the enhancement in velocity results.

ll All the beam edges and right angle portions shall be

treated by using high modulus carbon plates of cross
section 50mm x 1.20mm having E- modulus > 165 GPa
and tensile strength > 3.0 GPa as per the standard
manufacturer’s instructions. Epoxy mortar shall be
used to glue the carbon plates to concrete surface.

ll The entire beam surface shall then be encapsulated in

carbon fibre wrap.

Fig. 7: Schmidt Hammer Test on beams ll To protect and ensure the carbon surface from
weathering and environmental conditions surface high
strength epoxy mortar shall be applied as per standard
instructions from manufacturer.

Proper removal of loose concrete, cleaning of concrete

surface before start of repair works was done under strict
supervision. The repair work was given to authorized
applicator of M/s. Chryso.

All the repair material including carbon fibre was

procured from Chryso and the design calculation for
the quantity of carbon fibre required was done by their
German counterpart. The entire work was completed
under supervision of client’s representatives. Figures 6 to
Fig. 8: Drilling of nozzles in cracks 23 show the various activities of repair works.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

108 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Repair of jetty damaged by cyclone – Case Study

Fig. 10: Cracks sealed ready for epoxy grouting

Fig. 13: Repair of beam junction with epoxy mortar

Fig. 11: Grouting of cracks in progress Fig. 14: UPV tests after grouting

Fig. 15: Epoxy grouting complete, grinding of surface complete,

Fig. 12: Repair of beam bottom portion with cementitious grout locations marked for glueing carbon laminates

Organised by
India Chapter of American Concrete Institute 109
Session 1 C - Paper 1

Fig. 19: Wrapping the carbon laminates with carbon fibre

Fig. 16: Fixing of carbon laminates on beam sides

Fig. 20: Epoxy mortar finish for protecting carbon fibre wrapping

Fig. 17: Locations marked for glueing carbon laminates on beam

bottom

Fig. 21: Slab casting to be done

Conclusion
ll Hudhud cyclone was the worst cyclone to hit Vizag in
about 100 years
ll The structure was not designed for such forces causing
severe damage to the structure
ll The damages were very carefully investigated through
visual inspection, UPV test and hammer test
ll Based on the investigation findings, the repair
methodology was finalized and executed under strict
Fig. 18: Fixing carbon laminates on beam bottom supervision

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

110 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Repair of jetty damaged by cyclone – Case Study

Fig. 22: Fixing sacrificial shuttering and reinforcment for slab

Fig. 23: Slab casting completed

Dr. N.V. Nayak

Dr. Narayan Nayak was born in December 1936 in a small village Baggone in Kumta Taluka of Karnataka
State and in a family, highly averse to education then. But with sheer determination, he completed his
PhD in engineering from USA even after marriage and having 2 children. His academic career has been
excellent with GPA, 4.0/4.0 during Ph.D programme. Even in SSC, he was within the top 30 rank in Pune
Board. He came to India with his family to serve his country.
Dr. Nayak has got 8 years of Teaching, 8 years of Consultancy and 36 years of Construction experience.
Presently, Dr. Nayak is the Principal Advisor, Gammon India Ltd.; MD, Gammon Realty Ltd. and Chairman,
Geocon International Pvt. Ltd. Prior to this, he was Dy. Managing Director, Gammon India Limited and
before that Director AFCONS , Mumbai.
Dr. Nayak is particularly known for Speedy, Economical, Quality, Innovative and Safe Methods of Execution
of Projects by motivating hands-on field-training of all concerned. Near his home, he has executed 24
Bridges and one Tunnel for prestigious Konkan Railway project. The Mattanchery bridge in Kochi was
completed in record time.
Dr. Nayak has introduced many innovative techniques in construction field. As such, he is the First Annual
Lecturer from Construction Industry to be selected by Indian Geotechnical Society and first from the
industry to receive the prestigious Kueckelmann Award. He is also the recipient of Kanara Ratna Award.
The citation given to Dr. Nayak by India Chapter of American Concrete Institute states, “He is a fascinating
blend of knowledge and entrepreneurship for promotion and prosperity of concrete. In spite of several
achievements to his credit, he is a humble human being. He is a man of few words spreading fragrance of
his confidence and expertise around”.
Inspite of high education and entrepreneurship, Dr. Nayak has been very keen in spreading knowledge
and expertise and adopting the same in practice rather than accumulating the wealth. He has published
many technical papers (over 95 Nos) in India and abroad and authored very popular book “Foundation
Design Manual” (now in 6th edition). He is also co-editor and co-author of the “Handbook on Advanced
Concrete Technology”. He has delivered many invited keynote lectures on Geotechnical and Foundation
Engineering, Concrete Technology and Construction Management etc.

Manish Mokal
Manish Mokal graduated in Civil Engineering from University of Mumbai in the year 1997. He secured his
M.E. (Structures) from University of Mumbai in 2004.
He has 18 years of practical field experience and has worked on major infrastructure projects. He has
published many technical papers in National & International journals.
His main area of interests relate to high strength concrete and durability of concrete structures.

Organised by
India Chapter of American Concrete Institute 111
Session 1 C - Paper 2

Repair of core wall in high rise building – Case Study

Manish Mokal, Dr. N. V. Nayak, Amit Datta
Gammon India Limited, Chryso

Abstract After construction of 4th level of the core wall, severe voids
and honeycombing was observed in some portion of the
Severe honeycombing was observed after deshuttering
core wall. After lot of deliberation with the structural
of core wall lift of high rise building. The root cause for
engineer it was decided to repair the affected portion.
this honeycombing is analysed and discussed. The repair
methodology along with the testing conducted before and
after the repairs is discussed in detail. Initial investigation
Once the issue of voids and honeycombing was observed,
Keywords: Honeycombing, high strength cementitious a detailed visual inspection of the affected portion was
mortar, epoxy grout, ultrasonic pulse velocity test, core. done along with the client and structural engineer. Fig. 2
& 3 indicate the defects in the wall.
Introduction It was observed that the major defects had occurred around
This case study is of a high rise residential building 300 m the openings and corners in the wall as shown in Fig. 4.
tall in Mumbai. The building has very complex structural
form with a slenderness ratio of 14:1 making it more The detail of reinforcement is given in Fig. 1. The
slender than Burj Dubai which is currently the tallest reinforcement spacing in the core wall is divided into two
structure in the world. The major structural components zones:
are mega columns with embedded structural steel, ll Very dense reinforcement around the openings,
outrigger walls and core wall of M80 grade concrete apart corners and at specific locations along the length of the
from the slabs which are M40 grade concrete. wall (does not meet the minimum spacing requirement
Apart from the other structural components, the core as specified in IS 456, ACI 318 & BS 8110)
wall is the major structural component which takes load ll Relatively lesser dense reinforcement in the remaining
as well as imparts stiffness to the building. The plan of portion of the wall
core wall is as shown in Fig. 1. The core wall is 33 x 14.2
Primarily it is observed that the concrete has not
m with outer wall thickness of 1200 mm and inner walls
reached at some locations in the area with very dense
of 800 mm thickness. The core wall is constructed in two
reinforcement bars. The reinforcement is so dense that
parts for logistic purpose.
it makes the movement of
concrete very difficult as
well as difficult to vibrate the
concrete effectively. In the
properly compacted areas
also, the cover zone is filled
mostly with mortar and very
less aggregate. Fig. 5 below
shows the reinforcement
congestion in the core wall.
Rectification Methodology &
Execution
The following methodology
was approved for
rectification of core wall
portion.
Fig. 1: General Layout Plan of Core Wall

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

112 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Repair of core wall in high rise building – Case Study

ll The affected core wall shall be scanned with Ultrasonic

Pulse Velocity Meter (UPV) to ascertain the velocity in
concrete
ll In order to compare the densities of good concrete and
affected concrete, the next lift of core wall done above
the affected lift shall be scanned
ll All the loose concrete shall be removed with hammer
and chisel.
ll Injection nozzles shall be drilled and fixed in the
affected area.
ll Shuttering shall be fixed around the affected area and
M80 grade micro concrete (Chryso Excem GP 80) shall
Fig. 2: Voids and honeycombing defect in core wall be poured into it. The shuttering will be removed after
48 hours and the area is cured for 7 days
ll After 7 days curing is over, low viscosity epoxy grout
(Chryso Resicrete 21 LV) is pumped in the nozzles to
ensure that all micro voids are effectively filled.
ll The grouting area shall be left for 72 hours post
injection.
ll After the entire grouting operation is completed, the
Fig. 3: Voids and honeycombing in core wall nipples shall be cut off and sealed.
ll 28 days after the grouting exercise is complete, the
affected core wall shall be rescan with UPV in order
to check whether the areas with honeycombing where
the velocity was less has improved
ll Cores shall be taken from the repaired area to check
the uniformity and effectiveness of concrete
Fig. 6 to 12 shows the various activities being carried out at
site under the supervision of Client’s engineer. Every step
was checked and approved by the client before the next
activity was taken up.
Fig. 4: Mapping of honeycombing in core wall

Fig. 6: Cleaning the concrete surface with compressed air &

reinforcement with wire brush after chipping loose concrete

Fig. 5: Congestion of reinforcement in core wall Fig. 7: Cleaning surface with water

Organised by
India Chapter of American Concrete Institute 113
Session 1 C - Paper 2

Fig. 8: Cleaned concrete surface Fig. 11: Epoxy injection grouting

Fig. 9: Formwork arrangement Fig. 12: Cores extracted from repaired wall

The total volume of concrete in the affected wall is 24

cubic metres. The total consumption of cementitious
grout for filling the voids was 1.25 cubic metres and epoxy
grout consumed was 265 litres. This quantity is including
wastage which is approx. in the tune of 20%.
28 days after the entire repair works were completed,
the repaired core wall was again scanned with UPV for
velocities. The details of the UPV results before and after
the repair works are shown in Fig. 13 & 14.
Fig. 10: Repaired core wall

Fig. 13: Velocity in concrete in good lift and affected lift

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

114 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Repair of core wall in high rise building – Case Study

Fig. 14: Velocity in concrete in good lift and affected lift before and after grouting

In Fig. 13 the velocity data for affected wall lift and the good Even with best possible workmanship and supervision
concrete wall lift done above is given. The areas marked during construction, one cannot rule out possibility of such
in blue are the area which had honeycombing. UPV was honeycombing when reinforcement is so heavy. Quality
not done on this area before repair. The areas where of construction achieved for all lifts indicates excellent
the difference between the velocity of good and affected workmanship and supervision of very high standards.
wall was significantly high are shown in red. All the area Such honeycombing could have been eliminated by
demarcated in red are in the zone of honeycombing as is specifying self-compacting concrete (SCC) which is
expected to be. universally standard practice when reinforcement is very
Fig. 14 indicates the UPV results before and after repair of heavy.
the affected area. It can be seen that there is a significant In fact during the trial stages the contractor had designed
improvement in the velocity of concrete in the affected both normal concrete as well as self- compacting
area, thus indicating that the voids in the concrete have concrete and even suggested for use of SCC. But based on
been effectively filled. structural engineer’s recommendation, normal concrete
In general the velocity of good concrete is ranging is finally used.
between 4.5 to 4.7 km/s indicating excellent condition. In order to prevent such incidents from reoccurring, split
The velocity of concrete around the honeycomb area dosing of admixture was recommended in all transit
before repair indicated densities between 3.8 to 4.2 mixers, such that a calculated dosage of admixture shall
km/s which are lower than the velocities indicated in be added at batching plant and remaining dosage shall
the good concrete. After the repairs and grouting the be added at the time to pouring to adjust the workability
velocities indicated in the affected area and the adjoining as desired. This will prevent any quick loss of workability
area are in the range of 4.4 to 4.6 km/s which indicate a after pumping of concrete. Strict supervision during
substantial improvement in the velocity indicating that compaction is maintained during concreting. Almost 15
the voids are effectively filled with micro concrete and more core wall levels have been done after that without
epoxy grout.
any major issues of honeycombing.
Root Cause Analysis & Corrective Action
The core wall is having very highly congested reinforcement Conclusion
hampering free flow and compaction of concrete. In order ll One of the major issues in high rise buildings is the
to achieve good surface finish, we need a highly flowable, very dense reinforcement congestion in structural
cohesive concrete at the pouring point along with strict elements
supervision during compaction such that the concrete
ll In spite of exercising strict supervision over quality
passes through the congested reinforcement or we need
of concrete and excellent workmanship, possibility of
self-compacting concrete which can easily pass through
getting honeycombing cannot be avoided
the congested reinforcement without the need of any
compaction. ll Use of self-compacting concrete is the only solution to
avoid such incidents

Organised by
India Chapter of American Concrete Institute 115
Session 1 C - Paper 2

ll Use of Fe 600 steel to reduce the congestion can be effectively repaired with micro concrete and low
an effective option. In Japan, Fe 980 steel has been viscosity epoxy grout injection
implemented successfully.
ll UPV test can be effectively used to judge the difference
ll Use of composite section using structural steel in densities of concrete before and after repair
embedded in concrete can be option. This will
ll Extracting cores in order to visually judge the
substantially reduce the reinforcement congestion.
performance of repair can be done, but extreme care
Similar option was implemented for mega columns,
must be taken to avoid cutting of reinforcement
but though the contractor proposed this option for core
wall, structural engineer did not accept it. ll 15 lifts were done after this issue occurred without
any major issue indicating excellent workmanship and
ll In case of occurrence of honeycombing, it can be
quality control

Dr. N.V. Nayak

Dr. Narayan Nayak was born in December 1936 in a small village Baggone in Kumta Taluka of Karnataka
State and in a family, highly averse to education then. But with sheer determination, he completed his
PhD in engineering from USA even after marriage and having 2 children. His academic career has been
excellent with GPA, 4.0/4.0 during Ph.D programme. Even in SSC, he was within the top 30 rank in Pune
Board. He came to India with his family to serve his country.
Dr. Nayak has got 8 years of Teaching experience, 8 years of Consultancy experience and 36 years of
Construction experience.
Presently, Dr. Nayak is the Principal Advisor, Gammon India Limited; Managing Director, Gammon Realty
Limited and Chairman, Geocon International Pvt. Ltd. Prior to this, he was Dy.Managing Director, Gammon
India Limited and before that Director AFCONS , Mumbai.
Dr.Nayak is particularly known for Speedy, Economical, Quality, Innovative and Safe Methods of Execution
of Projects which he achieves by motivating hand on field training of all concerned. Near his home, he has
executed 24 Bridges and one Tunnel for prestigious Konkan Railway project. The Mattanchery bridge in
Kochi had completed in record short time.
Dr. Nayak has introduced many innovative techniques in construction field. As such, he is the First Annual
Lecturer from Construction Industry to be selected by Indian Geotechnical Society and 1st from the industry
to receive the prestigious Kueckelmann Award. He is also the recipient of Kanara Ratna Award.
In citation given to Dr. Nayak by India Chapter of American Concrete Institute states, “He is a fascinating
blending of knowledge and entrepreneurship for promotion and prosperity of concrete. In spite of several
achievements to the credit is a humble human being which is winning. He is a man of few words spreading
fragrance of his confidence and expertise around”.
Inspite of high education and entrepreneurship, Dr. Nayak has been very keen in spreading knowledge
and expertise and adopting the same in practice rather than accumulating the wealth. He has published
many technical papers (over 95 Nos) in India and abroad and authored very popular book “Foundation
Design Manual” (now in 6th edition). He is also co-editor and co-author of the “Handbook on Advanced
Concrete Technology”. He has delivered many invited keynote lectures on Geotechnical and Foundation
Engineering, Concrete Technology and Construction Management etc.

Manish Mokal
Manish Mokal graduated in Civil Engineering from University of Mumbai in the year 1997. He secured his
M.E. (Structures) from University of Mumbai in 2004.
He has 18 years of practical field experience and has worked on major infrastructure projects. He has
published many technical papers in National & International journals.
His main area of interests relate to high strength concrete and durability of concrete structures.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

116 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 An Experimental Investigation on the Behaviour of High Strength Restrained one Way GPC/TVC Slabs

An Experimental Investigation on the Behaviour of High Strength

Restrained one Way GPC/TVC Slabs

R.Mourougane C.Sashidhar C.G.Puttappa K.U.Muthu

Research Scholar, Civil Engg Professor, CED, JNTUA, Professor, CED, M.S.R.I.T, Dean, Brindavan College of
Department, M.S.Ramaiah Institute Anantapur, India Bangalore, India Engineering, Bangalore, India
of Technology, Bangalore, India

Abstract slabs. He noted that at failure the slabs were divided into
segments with well defined
An experimental programme consisted of casting and
testing sixteen one way restrained slabs, out of which crack patterns. Roberts E H[3], conducted tests on 36
eight were made of geopolymer concrete (GPC) and other strips representing one way slabs. The purpose here
eight were made of traditionally vibrated concrete (TVC). In was to gain an understanding of strip action as a basis
these slabs, the reinforcement percentages were varied. for explaining compressive membrane action in one way
The size of the slabs were 1080×500 mm and two different slabs and concluded that the theoretical and experimental
slab thicknesses were adopted. The results are compared deflections was not satisfactory due to the neglect of
with the various codes. An analytical method is proposed elastic curvatures in the analytical model. Park R[4],
to predict the complete load deflection behavior of laterally calculated load deflection curves corresponding to the
restrained one way slab strips subjected to uniformly strips tested by Roberts E H[3]. One such comparison
distributed loading. The development of compressive indicates the peak load predicted by the theory to be
membrane forces due to deflections were incorporated conservative. It was noted that the load on the actual slab
in theoretical formulation for prediction of load deflection decreased more rapidly than that predicted by the theory.
behavior of slabs. The results are compared with the tests Also, deflection at Johansen’s load was zero. Eyre J R and
conducted in the laboratory and the results are presented. Kemp K O[5], studied the in­plane stiffness of reinforced
concrete slabs under compressive membrane action.
Keywords: Deflection, GPC, Load, Membrane force,
Their investigation showed that a significant reduction
Restrained slab, Reinforcement, TVC.
in axial stiffness occurs in elements under combined
moment and membrane forces and hence the assumption
Introduction of an elastic full depth of element which is frequently used
Reinforced concrete slabs are one of the main structural for the slab stiffness in the theoretical predictions leads to
elements which are extensively used in civil engineering over estimation of ultimate loads. Desayi P, Muthu K U and
construction. A reinforced concrete slab behaves linearly Ashwath M U[6], presented a paper with an approach for
under smaller loads and non­linearly after cracking. the serviceability design of restrained rectangular slabs
Methods of prediction of the behaviour of reinforced with normal concrete.
concrete slabs are fairly known. Most of the designers Rankin G I B and Long A E[7], developed a method for
use simplified analytical techniques which have proven predicting the ultimate load capacity of laterally restrained
to yield safe and conservative designs. One of the needs reinforced concrete slabs by considering the arching or
of structural engineer is a reliable method for predicting compressive membrane action. The method was based on
the ultimate capacity of reinforced concrete slab. Elastic deformation theory and utilized an elastic­plastic stress­
or plastic methods are widely used for the prediction of strain criterion for concrete. The loads carried by bending
loaded slab behaviour; however an important factor that and arching action were calculated separately and then
is normally neglected in the design of slabs is the strength added to give the total ultimate load capacity. Amarnath
enhancing effect of arching/compressive membrane K [8], developed a method for predicting the load deflection
action. It is widely recognized that the load carrying capacity behavior and cracking behavior of flat plate slabs
of reinforced concrete slabs is significantly increased subjected to uniformly distributed loading, for partially
when the slab edges are restrained against lateral restrained and fully restrained boundary conditions. The
movement. This restrained induces large ‘arching forces’ results obtained by the proposed method were compared
within slab between the supports. This phenomenon is with the experimental test data of 10 slab specimens.
known as ‘compressive membrane action’. The proposed method for predicting load deflection
Timoshenko S and Woinowsky ­Krieger S[1] proposed behavior made use of deflection prior to Johansen’s load.
solutions for different shapes of slabs and boundary In this paper an attempt has been made to study the load
conditions. Danish engineer, Ingerslev A [2] conducted deflection behavior of fully restrained one way slabs of
experiments to assess the strength of reinforced concrete M60 grade concrete. An analytical method for predicting
the complete load deflection behavior for this case is

Organised by
India Chapter of American Concrete Institute 117
Session 1 C - Paper 3

proposed. The same proposed method is also applied to Stage 1: Load deflection behavior up to cracking load
study the load deflection behavior of fully restrained one The load deflection behavior up to cracking load is
way slabs made of TVC and GPC, the results are compared predicted using the gross moment of inertia function in
with the experimental test data. elastic deformation formula recommended by ACI 318. The
deflection increases linearly with respect to the load up to
Objectives Of The Present Study cracking load. The short term deflection of a restrained one
way slab under distributed loading can be estimated as,
ll To study the ultimate strength of GPC and TVC slabs
and compare the values based on the models proposed Where β1 = 1/384 constant for restrained flat plate. at
in various codes.
ll To obtain the actual load deflection plot up to failure Stage 2: Load deflection behavior beyond
load (under short­term loading) of GPC and TVC slabs cracking load and upto Johansen’s load
and compare the values obtained by the deflection
Calculation of theoretical deflection at
models proposed in various codes.
johansen’s load (δjcal)
ll An analytical model is proposed to include the load The theoretical deflection at Johansen’s load is calculated
deflection behavior beyond the Johansen’s load. using effective moment of inertia function, given as
where,
Proposed Method for Predicting Load
Deflection Behaviour of Restrained Slab Computation of Power Coefficient α
Strips The power coefficient α is determined for each slab
The proposed method has been developed in three stages. specimen from the experimental data and it involves
Figure 1 shows a typical load deflection plot in which effective moment of inertia function.
AB, BC and CD corresponds to three stages. In the first
From equation (1), we have
stage (AB in Figure 1), classical elastic theory is used for
computing the deflections up to cracking load and it follows where β1 = 1/384 constant for restrained flat plate.
the method given in ACI 318­2011. In the second stage (BC For δ = δj , Ig is replaced by Ieff
in Figure 1) an effective moment of inertia function is used
on rearranging, we get,
which reflects the reduction in flexural rigidity of the slab.
In this stage theoretical deflection at Johansen’s load (δjcal) also we can write,
is computed. Also, this stage involves computation of the on simplification and rearranging, we get
power coefficient α which is dependent on experimental
deflection at Johansen’s load i.e. (δjexp). The third stage The value of δjexp in equation (3) is obtained from the
(CD in Figure 1) corresponds to the behaviour of the experimental load deflection curve for each slab strip,
slab beyond Johansen’s load and considers the effect of corresponding to Johansen’s load calculated by using ACI
membrane action. 318. The power coefficient α is calculated for each slab
strip and is noted to be varying for every slab. Hence the
value of α is taken as 3 for better predictions in the test
data.
The load deflection curve beyond Pcr up to Pj is predicted
by using the effective moment of inertia function given by,
where β1 = 1/384
Different values of load intensity q are used in equation
(9) and the corresponding deflections are obtained from
equation (10).
Load Deflection Behaviour Beyond Johansen’s Load
In this stage the load deflection behaviour is predicted by
a procedure which incorporates the effect of membrane
action on the load carrying capacity. The procedure in
general follows Eyre & Kemp’s approach with following
modifications.
(1) The deflections prior to Johansen’s load have been
included to predict the load deflection response.
Fig. 1: Idealized load deflection curve

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

118 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 An Experimental Investigation on the Behaviour of High Strength Restrained one Way GPC/TVC Slabs

(2) Elastic shortening of slab strips under compressive Assuming both the materials to be perfectly plastic
membrane force has been included. manner, Wood gave a non­dimensional relationship as,
The proposed method is based on the following Where
assumptions.
M / Mu = ratio of moment M to pure flexural yield
(1) The membrane action begins after the mechanism has moment Mu
formed.
N / To = ratio of axial compressive force N applied at
(2) The strength of concrete in tension is neglected. the centre of the section to the yield force of tension
of steel.
Regarding the first assumption the membrane forces
develop at early stages of loading but the magnitude of the α and β = coefficients depending on the properties of the
membrane forces is small prior to yield line load. Hence concrete section.
the assumption has been made. The second assumption is
The maximum moment is,
made as the concrete is weak in tension.
The moment capacity falls from the maximum value as
Figure 2 shows the case of a uniformly loaded restrained
the compressive force is reduced and the neutral axis
flat plate slab. The supports are assumed to be infinitely
rises. The limit is reached when the neutral axis reaches
rigid and the slab is restrained against both rotation
the surface (i.e. when the section is cracked through the
and translation. The slab is provided with equal tensile
full depth) which occurs when the membrane force is
reinforcement at the support and at mid span. If there
tensile and is equal to the strength of reinforcement.
is no horizontal restraint at the support, the slab would
fail when the flexural plastic hinges are formed at the The value of the moment is then
centre of the span and at the supports and the collapse
From the yield criterion the position of neutral axis μ
load would be Pj. The horizontal restraint at the supports
(positive above the centre of the slab depth) is found as
increases the collapse load.
Where f is the yield function obtained from equation (11) as
Yield Criterion Putting equation (22) and (23) in (20), we get
The yield criterion for a singly reinforced concrete section
subjected to a bending moment M and centrally applied Geometrical Relationship
compressive force w per unit width is formulated by The collapse mechanism, the extension of the central
horizontal force and moment equilibrium (Figure 2). fibre of strip and the deflected shape are shown in Figure
2 Considering unit width of the strip, the relationship
between original length and the length of the strip beyond
yield line load for a particular deflection δp can be obtained
as follows. Initially the length of the strip is (l/2). At a
deflection of δj, the length of the strip changes to
At any load greater than the yield line load, if δp is the
deflection in excess of δj the elastic shortening of the strip
is
If eo and e1 are the extensions of the strip at the centre
and at the edge, the relationship between the lengths of
the strip before and after formation of mechanism can be
written as
Neglecting higher powers of δj greater than 2, above
equation is simplified to
The extensions eo and e1 are dependent on the plastic
deflection and hence from Figure 2
Substituting the values eo and e1 from above equation, we
get
The height of neutral axis at the edge μe is obtained from
the yield criterion and the equilibrium of membrane forces
and since it is isotropic and μo = μe , it is obtained as
Where
Fig. 2: Restrained flat plate slab with plastic deflectio

Organised by
India Chapter of American Concrete Institute 119
Session 1 C - Paper 3

Considering the equilibrium of moments the enhancement Loading Arrangement and Test Procedure
factor is obtained as The load from a single jack of 500 kN capacity was equally
The load deflection behavior beyond Johansen’s load till distributed to the loading through tiers of steel rails and
ultimate load is obtained using equation (33) Equation rods as shown in Fig 4 The loading area was divided into
(33) expresses the load as a function of plastic deflection four equal areas. The center of gravity of each area acted
δp. As plastic deflection (δp) increases the intensity of as loading point as shown in Fig 3 In order to avoid local
load (q) increases up to a maximum value and then with failure due to point loads, bearing plates of size 50 × 50
further increase of δp the load decreases. Hence the mm were provided at each load point. During testing, the
calculations done till the load deflection behaviour of the Central deflections and crack widths were measured
slabs showed a descending trend. Also the value of δj used at different stages of loading. The auxiliary specimens
in the calculations is the computed value at Johansen’s (Cubes) were tested for compressive strength on the
load (obtained in section 2.2.1). An analytical method has same day of the testing of slabs.
been proposed to predict the complete load deflection
behavior of restrained slab strips subjected to uniformly
distributed loading. To verify the results of analysis, test
data from experimental programme of fifteen slab strips
is taken. The details of the experimental work are given in
the following article.

Experimental Program
The experimental programme consisted of casting
and testing of sixteen one way restrained slabs out of
which eight were made up of GPC and other eight were
Fig. 3: Location of Loading Points
TVC. The summarized mix proportions are indicated in
Table 1. The Preparation of GPC was as follows, Fly ash,
GGBS and aggregates were dry mixed in the pan mixer
for about 3 minutes. The alkaline solution, i.e sodium
silicate solution, the sodium hydroxide solution that was
prepared one day prior to usage along with, added water
and the super plasticizer were premixed and then added
to the solids. The wet mixing was continued for another 2
minutes. The fresh concrete was used to make the slab
specimens of size 1080 × 500 mm. In these slabs, the
reinforcement percentages were varied. Two types of
thicknesses of the slab were adopted i.e., 50 mm and 65 Fig. 4: Experimental setup for Restrained Slab
mm respectively. The reinforcement adopted was 5mm in
both the directions. Clear cover to the main reinforcement
was 15 mm. Prior to casting, the inner walls of moulds Test Results and Load Deflection Behavior of
were coated with lubricating oil to prevent adhesion GPC and TVC Slabs
with the cured concrete. All geopolymer slab specimens The load ­deflection plots for all the tested slabs are
were steam cured in a curing chamber for 24 hours at a shown in Fig 5 & 6 and results of the experiment are
temperature of 60°C. The specimens were then allowed tabulated in Table 2 The deflection increased linearly
to cool in air. Similarly, TVC slabs were cured for 28 with the load up to cracking stage and later a non­linear
days using by ponding. Details of the loading set up and behavior was observed. The deflections in the descending
required dimensions for loading points on the slabs are portion of the load deflection curve could not be measured
shown in Fig 3 All the slabs were tested under four point for the specimens due to the instability of the loading
loading, simulating the uniformly distributed load. system. First crack for all the slabs were observed at

Table 1
Concrete Mix Details

kg/m3 Cement Fly ash GGBS SF CA FA NaOH Na2SiO3 Water SP

GPC ­- 366 40 - 1295 555 41 103 16.2 3

TVC 375 -­ - 42 1050 716 - - 150 2.5

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

120 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 An Experimental Investigation on the Behaviour of High Strength Restrained one Way GPC/TVC Slabs

the centre and for subsequent increment of loads the

Table 3
cracks extended parallel to the width. New cracks were Ratios of Experimental load to Calculated loads
formed with increase in the load and final crack pattern
failure mechanism was observed. At loads slightly less
Pue/Puc
than ultimate load, the bottom crack penetrated to the top Pue
and cracks were seen in compression face also, which Slab no
(kN)
ensured the development of tensile membrane action. Euro
IS 456 ACI 318 BS 8110
Code
Final failure was due to snapping of reinforcement. The
crack patterns of the typical slabs are shown in Figure GPC 1/50 54 2.17 1.85 2.16 1.84
7 & 8. Ultimate load and Deflection of all GPC and TVC
slabs are shown in the Table 3 to 6 the values obtained are GPC 2/50 46 2.14 1.83 2.13 1.82
computed using various codal equations and compared
with the experimental values. GPC 3/50 42 2.32 1.99 2.32 1.98

The proposed method has been used to predict the GPC 4/50 36 2.47 2.13 2.46 2.12
load deflection behavior of eight traditionally vibrated
concrete and seven geopolymer concrete slab strips. The GPC 1/65 56 1.70 1.46 1.70 1.46
predicted load deflection behavior is compared with the
experimental load deflection curves of the slab strips and GPC 2/65 22 0.78 0.67 0.78 0.67
also with the results obtained by using Rankin’s and Eyre
and Kemp’s analytical method. GPC 3/65 40 1.68 1.45 1.68 1.44

GPC 4/65 22 1.15 0.99 1.15 0.99

Table 2
Test Results of Restrained Slabs Mean 1.80 1.55 1.80 1.54

SD 0.59 0.51 0.59 0.50

fck Pcr δcr Pu δu δw
Slab No
(Mpa) (kN) (mm) (kN) (mm) (mm)
CV 32.92 32.83 32.90 32.79

GPC­1/50 60 8 0.861 54 11.722 6.016

GPC­2 /50 58 6 0.987 46 12.211 5.985 Table 4

Ratios of Experimental load to Calculated loads
GPC­3/50 60 12 1.102 42 8.287 4.515
Pue/Puc
GPC­4 /50 57 12 1.315 36 11.653 5.453 Pue
Slab no
(kN)
Euro
GPC­1/65 61 10 0.855 56 10.725 5.398 IS 456 ACI 318 BS 8110
Code

GPC­2 /65 59 6 2.547 22 13.933 5.550 TVC 1/50 46 1.83 1.57 1.83 1.56

GPC­3/65 60 10 0.992 40 10.084 4.960 TVC 2/50 30 1.39 1.19 1.38 1.18

GPC­4 /65 58 6 1.269 22 7.674 4.955 TVC 3/50 26 1.43 1.23 1.43 1.23

TVC­1/50 68 10 0.644 46 12.112 6.776 TVC 4/50 26 1.77 1.53 1.77 1.52

TVC­2 /50 65 10 1.826 30 13.414 5.988 TVC 1/65 50 1.52 1.30 1.52 1.30

TVC­3/50 64 12 1.460 26 11.940 4.573 TVC 2/65 40 1.41 1.22 1.41 1.21

TVC­4 /50 66 8 1.341 26 8.677 3.331 TVC 3/65 36 1.51 1.30 1.51 1.30

TVC­1/65 64 20 1.295 50 10.504 5.004 TVC 4/65 30 1.56 1.35 1.56 1.35

TVC­2 /65 58 28 1.092 40 7.165 1.035 Mean 1.55 1.34 1.55 1.33

TVC­3/65 60 12 1.209 36 11.260 2.877 SD 0.17 0.14 0.17 0.14

TVC­4 /65 62 12 1.456 30 8.145 3.630 CV 10.65 10.63 10.65 10.64

Organised by
India Chapter of American Concrete Institute 121
Session 1 C - Paper 3

Table 5
Ratios of Calculated and Experimental Deflections at
service Loads

δv/δe
δe
Slab no
(mm) Euro
IS 456 ACI 318 BS 8110
Code

GPC 1/50 6.02 0.65 0.62 0.38 0.61

GPC 2/50 5.99 0.64 0.61 0.36 0.59

GPC 3/50 4.52 0.76 0.64 0.45 0.73

GPC 4/50 5.45 0.61 0.44 0.33 0.59

GPC 1/65 5.40 0.41 0.39 0.30 0.39

GPC 2/65 5.55 0.16 0.14 0.27 0.15

GPC 3/65 4.96 0.39 0.33 0.27 0.38

GPC 4/65 4.96 0.25 0.19 0.24 0.25

Fig. 5: Load Vs Central Deflection for 50 mm GPC and TVC slabs
Mean 0.48 0.42 0.33 0.46

SD 0.21 0.19 0.07 0.20

CV 44.11 46.71 21.27 43.44

Table 6
Ratios of Calculated and Experimental Deflections at
service Loads

δv/δe
δe
Slab no
(mm) Euro
IS 456 ACI 318 BS 8110
Code

TVC 1/50 6.78 0.45 0.43 0.34 0.43

TVC 2/50 5.99 0.33 0.27 0.36 0.32

TVC 3/50 4.57 0.35 0.20 0.44 0.31

TVC 4/50 3.33 0.73 0.55 0.54 0.72

TVC 1/65 5.00 0.29 0.32 0.32 0.28

Fig. 6: Load Vs Central Deflection for 65 mm GPC and TVC slabs
TVC 2/65 1.04 0.70 1.44 1.31 0.11
The load deflection behavior beyond cracking load is
computed using the proposed method, by making use TVC 3/65 2.88 0.54 0.47 0.40 0.52
of effective moment of inertia function and determining
the value of the power coefficient α. The computation of TVC 4/65 3.63 0.38 0.32 0.24 0.36
α requires experimental deflection at Johansen’s load
(δjexp). The value of δjexp for each slab is obtained from Mean 0.47 0.50 0.49 0.38
the experimental load deflection curves of tested slab
strips, these values are corresponding to Johansen’s load SD 0.17 0.40 0.34 0.18
calculated by using ACI 318. The power coefficient α is
CV 35.57 79.19 69.50 47.56
computed for each slab strip and is noted to be varying

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

122 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 An Experimental Investigation on the Behaviour of High Strength Restrained one Way GPC/TVC Slabs

Fig. 7: Crack pattern on Compression and Tension Face of GPC/

TVC slabs 50mm

Fig. 10: Load Vs Deflection GPC and TVC 1/65

largely for all slab strips and hence an average value of

α = 3 has been used for better prediction of deflections.
The comparison is shown for typical slabs in Figures 9 &
10. It is noticed from those plots that the load deflection
curve has an ascending trend is noticed till it reaches
ultimate load and then a descending trend is observed.
Thus it is seen that the proposed method predicts the
load deflection behavior more satisfactorily than other
methods for most of the slab specimens.
Fig. 8: Crack pattern on Compression and Tension Face of GPC/
TVC slabs 65mm
Summary snd Conclusions
The following conclusions are drawn based on the
proposed analysis and the experimental investigation
conducted. A method has been proposed to predict the
load deflection behavior of restrained slab strips subjected
to uniformly distributed loading.
ll The predicted load deflection behavior is compared
with the experimental load deflection curves of the
slab strips and also with the results obtained by using
Rankin’s and Eyre and Kemp’s analytical method. It
is noted that the proposed method predicts the load
deflection behavior more satisfactorily than other
methods for most of the slab specimens.
ll The ultimate loads were computed using ACI 318, IS
456, EN 1992 and BS 8110. The results show that the
all the codes are conservative in estimating the same.
The ACI 318 and EN 1992 codes predict the ultimate
loads better.
ll The experimental ultimate load is approximately
0.67 to 2.47 times of the computed load for GPC and
approximately 1.18 to 1.83 times of the computed load
Fig. 9: Load Vs Deflection GPC and TVC 1/50

Organised by
India Chapter of American Concrete Institute 123
Session 1 C - Paper 3

for TVC respectively. This enhancement could be due 2. Ingerslev A., 1921. “The strength of rectangular slabs”, Journal of
Institution of Structural Engineers, 1(1): 3­14.
to the development of membrane action in slabs.
3. Roberts E H.,1969. “Load carrying capacity of slab strips restrained
ll Deflection of GPC and TVC slabs were calculated using against longitudinal expansion”, Concrete, 3(9): 369­378.
IS 456, ACI 318, EN 1992 and BS 8110. None of the codes 4. Park R., 1965. “The lateral stiffness and strength required to ensure
predicts the deflection satisfactorily. membrane action at the ultimate load of a reinforced concrete slab
and beam floor”, Magazine of Concrete Research, 17: 29­38.
5. Eyre J R and Kemp K O., 1994. “In plane stiffness of reinforced
Acknowledgements concrete slabs under compressive action”, Magazine of Concrete
Research, 46: 67­7 7.
The authors sincerely thank the Management and
Principal Dr. N V R Naidu of M S Ramaiah Institute of 6. Desayi P, Muthu K U and Ashwath M U., 1992. “Control of Deflections
of Restrained Rectangular Slabs”, International Journal of
Technology, Bangalore. Research and Development, Structures (Roorkee), 12(1): 21 ­28.
Jawaharlal Nehru Technological University, Anantapur. 7. Rankin G I B and Long A E., 1997. “Arching action strength
All India Council for Technical Education, New Delhi, for enhancement in laterally­restrained slab strips”, Proceedings of
funding the Research Project that was carried out in the Institution of Civil Engineers, Structures & Buildings, 122: 461­4 67.
Department of Civil Engineering, M S Ramaiah Institute of 8. Amarnath K., 1998. “Strength and deformation behavior of
Technology, Bangalore. The research work presented in restrained / partially restrained flat plate slabs”, Ph.D. thesis,
Bangalore University, Bangalore, India,
this paper is part of the ongoing Research Project.
9. Perumal K and Sundararajan R., 2003. “Experimental
Investigation on High Performance Concrete using Silica Fume and
References Superplasticizer,” Proceedings of the INCONTEST 2003: 303 ­313.
1. Timoshenko, S.; and Woinowsky ­Krieger S., 1959. “Theory of Plates 10. Rangan B.V., 2008. “Mix design and production of fly ash based
and Shells” 2nd Edition, McGraw ­Hill Co., New York. geopolymer concrete”, Indian Concrete Journal, 82(5): 7–15.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

124 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of reinforced concrete beams strengthened with Glass Fibre Reinforced Polymer laminates subjected to corrosion damage

Behaviour of reinforced concrete beams strengthened with

Glass Fibre Reinforced Polymer laminates subjected
to corrosion damage
Kolluru Hemanth Kumar, Sanjay Rajpal, Shashank Jha, Dr. P.K. Jain
Department of Civil Engineering, Maulana Azad National Institute of Technology- Bhopal

Abstract stresses, consequently leading to cracking, staining and

spalling of concrete, hence reduction of bond strength and
Referable to the adverse effects of the surroundings along
change in the bond-slip behavior between concrete and
the social system, reinforcement corrosion in concrete
steel reinforcement occurs.
may cause detrimental effects viz. The decrease of the
effective area of support which ultimately equates to Maintenance and rehabilitation of these structures
strength reduction, internal stress caused by rust products are the prime concern in the present industry.
formed during corrosion consequently lead to cracking and New techniques emerge daily for rehabilitation or
spalling of concrete, reduction of bond strength and change strengthening, one of these involves externally bonding
in the bond-slip behavior in between concrete and steel of fiber reinforced polymer (FRP) laminates to the
reinforcement. Hence the prevention of corrosion in steel structural member. Interest in this technique has been
reinforcement of RC structures has been an area of prime wide- spread and continues to flow because of their
concern for researchers in the past few years. The present high tensile strength to weight ratio, high resistance to
paper deals with the influence of GFRP on corrosion of fatigue and corrosion, low-cost, negligible clearance
steel reinforcement in RC beams. The experiment includes loss, high durability and ease of installation. Scores of
construction of ten (150*150*700 mm) reinforced concrete studies have shown that rehabilitation or strengthening
beams. Eight beams were strengthened by externally epoxy with FRP laminates is successful in restoring or
bonding of GFRP on concrete surface in different patterns. increasing the strength of concrete members (ACI
The reinforcements of four strengthened beams (one of 440-R 1996; Nanni 1993; Mufti et al. 1992; Meier et
each pattern) and one unstrengthend beam was exposed al. 1992; El- Badry 1996; Dolan et al. 1999). Another
to accelerated corrosion by the impressed current method. promising aspect of FRP is the prevention of ingress
After the corrosion phase, the beams were subjected to of aggressive chemical compounds into the concrete
three point bending. matrix which cause deterioration in the form of steel
reinforcement corrosion by confining the concrete
Keywords: Fiber reinforced polymer, GFRP-Glass Fibre
matrix. Strengthening of concrete members also
Reinforced Polymer, accelerated corrosion.
delays concrete spalling and cracking caused by the
expansive forces of the byproducts of corrosion. Thus
Introduction the structural safety of the RCC structural members
As the world's infrastructure ages day by day, reinforced will be degraded due to the adverse effects caused by
concrete structures functionally degrade due to adverse corrosion of steel reinforcement (ACI 222 1996; Lee
effects of environmental factors, increased structural et al. 1997; Uomoto et al. 1984; Soudki 1999; Soudki et
load requirements, etc. One of the major deterioration al. 2000).Thus by preventing or controlling corrosion
effect is corrosion of steel reinforcement in reinforced of steel reinforcement, a certain degree of structural
concrete. Reinforcement in concrete beams possess a strength can be enhanced in the deteriorated RCC
passive film around them due to the highly basic medium structural members.
of concrete. Corrosion occurs when this passive layer Consequently, an FRP strengthened member exhibits
encounters a breakdown due to chloride ions in deicing improved structural performance than the un-
salts or due to neutralization of concrete by atmospheric rehabilitated member after undergoing active corrosion,
carbon dioxide. Various characteristic studies of corroded due to a combination of several mechanisms which are
Reinforced concrete members have shown that corroded mentioned below (Sherwood and Soudki 1999a, 1999b):
reinforcement may cause many detrimental effects
on RC member’s viz. the decrease of the effective area (I) Confining Concrete matrix, thereby cutting down the
of reinforcement which ultimately equates to strength corrosion cracking and bond splitting cracks caused
reduction, and the corrosion products formed occupy a by real tensile stresses caused by corrosion products.
larger volume than the steel. They therefore exert internal (II) Prevention of ingress of chloride ions and carbon
stress in the concrete in the form of substantial tensile dioxide molecules, they're bringing down the rate of
erosion of the steel reinforcement.

Organised by
India Chapter of American Concrete Institute 125
Session 1 C - Paper 4

(III) An Increase in the flexural and shear strength, to Glass Fibre Reinforced Polymer specifications
overcome the deprivation caused by corrosion of steel The unidirectional Glass fibre laminates used to
reinforcement. strengthen the RC beams were the flexible tow sheets
This paper summarizes the effect of different factors with thickness of 0.324 mm (dry fibres), with ultimate
on the extent of corrosion as well as developing ways to tensile strength of 3400 MPa and ultimate elongation of
inhibit rebar corrosion using Fiber Reinforced Polymer 1.5%. The design thickness of FRP composite (fibres and
(FRP) sheets. The FRP sheets are wrapped using different epoxy) is about 2 mm.
patterns and the efficiency of each pattern is investigated.
Specimen Details
Experimental Preview This section describes the main characteristics of
the tested specimens, properties of their constituent
The experiment comprised of testing of ten reinforced material, loading equipment, instrumentation and testing
concrete beam specimens, eight beams were sequences. This study consists of casting of ten RC beams
strengthened with Glass Fibre Reinforced Polymer (GFRP) from which four were externally strengthened with
sheets while two beams were unstrengthened and are GFRP in different configurations. The cross section of
control specimens. The eight strengthened beams were all specimens was 150*150 mm with a length of 700 mm
covered with GFRP in four different patterns which are as shown in Fig1. The beams were reinforced with 2 Fe-
described below: 250 bars at bottom, 2 Fe-250 bars at top and 7 Fe-250
(i) Bottom covered beam(two beams) equally spaced stirrups throughout the length. Two tensile
bars are protruded as shown in Fig.1, which will be used
(ii) Bottom covered and sides covered till 100 mm(two as anode in impressed current technique. After 28 days
beams) of casting of beams, GFRP sheets were applied on four
(iii) Bottom covered and full sides covered(two beams) beams using epoxy as per manufacturer’s specifications.
The different patterns of FRP application include only
(iv) Only sides covered
bottom covered, only sides covered, bottom covered and
sides covered up to 100mm and completely sides and
bottom covered.

Loading Method
The loading was done by the 3 point flexure testing method
in which, the beam was supported from the bottom at
both the sides at a distance of 5 cm form the respective
edges and the load was applied at a single point from the
top using a hydraulic jack. The maximum load achieved
during loading was considered as the failure load. Figure
below illustrates the loading apparatus.

Fig. 1: Reinforcement details

The specimens were 700 mm long, 150 mm wide, and 150

mm high. The reinforcement details of the specimens are
shown in Fig. 1. It consisted of two 6mm diameter bars of
length 650 mm mild steel at the bottom and two 6 mm
bars of length 65cm mild steel at top and 6 mm diameter (a)
(b)
mild steel stirrups placed at 90 mm o/c, and the clear
cover was 25 mm. One beam from each pattern and one Fig. 2: (a) schematic of Flexure Loading Apparatus; (b) Loading
controlled beam were subjected to accelerated corrosion frame
regime. After 150 hours of accelerated corrosion the
beams were subjected to three point flexure test. The Accelerated Corrosion Technique:
beams were compared on the basis of: Five beams were subjected to accelerated corrosion
i) Flexural strength, technique where steel reinforcement acts as anode and
a stainless steel rod as cathode, as shown in Fig.3. The
ii) Strength retained after corrosion, accelerated corrosion was performed by dipping the beam
Iii) Per cent corrosion of steel calculated by Faradays law. into 5% nacl solution upto half depth so as to target only
the bottom rebars.The cathode was fixed in the tub along

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

126 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of reinforced concrete beams strengthened with Glass Fibre Reinforced Polymer laminates subjected to corrosion damage

Fig. 3: Accelerated Corrosion Apparatus

with beam as anode (at a distance of about 5 cm from the

beam). An external power supply was provided to supply Fig. 4: Corroded Beam (Plain), Corroded Beam (with FRP)
a constant voltage of 15 V. Each beam was impressed
with current for 150 hours. The decrease in area of steel Flexural Test
reinforcement was calculated using Faraday’s Law.
Table 2
Strength comparison of Plain and FRP bonded beams
Results
Load Carried before % increase in load capacity
FRP Pattern
corrosion (kN) compared to plain beam
Accelerated Corrosion
Plain Beam
The impressed current apparatus yielded the voltage (No FRP)
30.26 kN -
applied and current passing through a particular beam at Only Bottom 55.26 kN 82.61%
any particular point of time. These readings were taken
Only Sides
at different times and recorded. The current values of two Covered
36.26 kN 20.68%
consecutive times were used to calculate the mean current
Bottom + Side
during that time period as the arithmetic mean of each (10
54.26 kN 80.17%
current value which was then used to calculate the ampere- Bottom + Side
hours by multiplying with the time duration for that particular 57.26kN 90.08%
(15
average current calculated as the difference between
the times for the consecutive readings. The cumulative of
Table 3
ampere-hours when applied in the formula already stated, Exposed Strength and corrosion estimate
could be used to calculate the mass of iron that has been
% decrease
corroded. The total mass of dipped iron was calculated by FRP
Load (KN)
After
Theoretical
in load
taking the average mass of 6mm rebar as 0.22 kg/m rum Before % of
Pattern Corrosion capacity (after
corrosion Corrosion
and it came out to be 700 gram. Using this value, percent of corrosion)
corrosion was calculated at each reading and final corrosion Plain
percent after applying the beam for corrosion for 150 hours. Beam (No 30.26 22.26 27.66% 26.43%
FRP)
The extent of corrosion in all the beams at 150 hours Only
of corrosion at same voltage of 15V has been depicted 55.26 37.26 17.00% 32.57%
Bottom
in Table 1: Only Sides
36.26 32.26 13.08% 11.03%
Covered

Table 1 Bottom +
54.26 53.26 6.20% 1.84%
Extent of corrosion Side (10
Bottom +
Voltage Weight Corrosion 57.26 55.26 4.15% 3.49%
Type of beam Time Side (15
applied lost %
Uncovered 15V 150 Hours 193.64 gm 27.66%
Bottom covered 15V 150 Hours 119 gm 17.00% Conclusions
Only Sides Covered 15V 150 Hours 91.56 gm 13.08%
Study of the effect of corrosive environment on flexural
Bottom+10 cm sides 15V 150 Hours 43.4 gm 6.20%
strength of RCC beams
Bottom + sides 15V 150 Hours 29.05 gm 4.15%
Flexural capacity of plain beam decreased by 26.43%

Organised by
India Chapter of American Concrete Institute 127
Session 1 C - Paper 4

(Table-3) on applying a voltage of 15V for 150 hours and the Another bottom and sides covered beam was then
theoretical corrosion calculated on the basis of Faraday’s exposed to corrosive environment and the theoretical
Law came out to be 27.66% (Table-3). corrosion percent in this case came out to be only 4.15%
(Table-3) which was the lowest amongst all the patterns
Study of the effect of covering only bottom face of beam selected and its strength as compared to non-exposed
with FRP on its flexural strength with and without similar covered beam decreased by 3.49% (Table-3) and
exposure to corrosive environment the failure occurred by de-bonding of FRP at the bottom
The beam was covered in FRP only at the bottom and the face as well as on the sides. Thus, it could be concluded
Flexural strength without exposure increased by 82.61% that covering tensile face as well as sides was also found
(Table-2) as compared to uncovered beam thus it could pretty effective when considering the combined effect of
be concluded that using FRP on the tensile side is pretty strengthening as well as corrosion inhibition.
effective for flexural strengthening
Study of the effect of covering only side faces of RCC
Another bottom covered beam was then exposed to beam with FRP on its flexural strength with and
corrosive environment and the theoretical corrosion without exposure to corrosive environment
percent in this case came out to be 17% (Table-3) which was
The beam was covered in FRP only on the sides and the
lower than that in plain beam but its strength as compared
Flexural strength without exposure increased by 20.68%
to non-exposed bottom covered beam decreased by
(Table-2) as compared to uncovered beam thus it could be
32.57% (Table-3) because the FRP debonded due to severe
concluded that using FRP only on the side faces in not at
cracks developed just at the interface of covering. Thus, it
all feasible for strengthening as compared to the rest of
could be concluded that only tensile face being covered is
the patterns.
effective for strengthening but not for corrosion inhibition.
Another side covered beam was then exposed to corrosive
Study of the effect of covering bottom face as well as environment and the theoretical corrosion percent in this
sides (up-to 10cm and 15cm) of RCC beam with FRP case came out to be 13.08% (Table-3) which was not that
on its flexural strength with and without exposure to good as compared to the other patterns listed above and
corrosive environment the failure occurred as tensile failure of the bottom of
the beam and rupture of FRP at the sides. Thus, it could
Bottom and 10 cm side be concluded that covering only the sides is not a viable
The beam was covered in FRP at the bottom as well as option for strengthening as well as corrosion inhibition.
the sides up-to 10cm and the Flexural strength without
exposure increased by 80.17% (Table-2) as compared to Comparative study of Percent increase in flexural
uncovered beam thus it could be concluded that using FRP strength as well as corrosion inhibition using different
on the tensile side and portion of side face has almost the patterns of FRP wrapping on RCC beam
same effect on flexural strength as covering the bottom
On the basis of above studies, it could be concluded that
face only. Thus, it would not be that feasible in terms of
taking a combined effect of strengthening and strength
strengthening application. Another bottom and sides up-
retention after exposure to corrosive environment, only
to 10 cm covered beam was then exposed to corrosive
two of the selected patterns proved to be comparably
environment and the theoretical corrosion percent in this
effective i.e. bottom and 10cm sides pattern and bottom
case came out to be 6.20% (Table-3) which was pretty low
and full sides pattern. Since the aim of the project was to
than that in plain beam and its strength as compared to
determine the most efficient pattern of FRP that deals with
non-exposed similar covered beam decreased only by
strengthening as well as corrosion inhibiting, the rest of
1.84% (Table-3) and the failure occurred by debonding of
the patterns as clearly explained are not that effective and
FRP at the bottom face and rupture of FRP on the sides.
would be neglected in this study. A comparison of only the
Thus, it could be concluded that covering tensile face
two above mentioned patterns would be done henceforth.
as well as side’s up-to 10cm was pretty effective when
considering the combined effect of strengthening as well The bar chart depicting the performance of both the
as corrosion inhibition. patterns on the basis of the three deciding parameters
namely
Bottom and full sides
(i) The percent increase in flexural strength of beam on
The beam was covered in FRP at the bottom as well as
applying the particular pattern of FRP (ii) The percent
the full sides and the Flexural strength without exposure
of flexural strength retained when exposed to corrosive
increased by 90.08% (Table-2) as compared to uncovered
environment and
beam thus it could be concluded that using FRP on the
tensile side and the side faces increases the flexural (iii) The economic viability of each pattern.
strength even more but this increase is not that appreciable Based on these factors, giving weightage to each one
when considering the amount of extra wrapping required. according to the need of any project, the best fit pattern

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

128 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of reinforced concrete beams strengthened with Glass Fibre Reinforced Polymer laminates subjected to corrosion damage

could be selected. The economic viability of a particular 2. ACI Committee 440. 1996. State of the art report on fiber reinforced
plastic reinforcement for concrete structures. ACI 440R-96.
pattern has been calculated by finding the ratio of the cost
American Concrete Institute, Detroit, Mich. 68 pp.
of making one concrete beam to that of the FRP applied on
3. Dolan, C., Rizkalla, S., and Nanni, A. 1999. ACI-SP-188 on Nonmetallic
the beam and depicting it as a percentage. (FRP) reinforcement for Concrete. Baltimore, Md.
Average cost of material used for casting 1 beam = Rs. 4. El-Badry, M. 1996. Proceedings of the 2nd International Conference
200/- Cost of FRP (bottom and 10 cm sides) on Advanced Composite Materials in Bridges and Structures.
Area of FRP applied = 0.15*0.7 + 2*(0.10*0.7) m2 = 0.245 Montréal, Qué. 11–14 August.
m2 Cost of 1 m2 FRP including primer and saturant = Rs. 5. Lee, H.S., Tomosawa, F., Masuda, Y., and Kage, T. 1997. Effect of
CFRP sheets on flexural strengthening of RC beams damaged by
750/- Thus, cost of FRP = Rs. 183.75
corrosion of tension rebar. Proceedings of the Third International
Economic Viability = (Rs. 200/ Rs. 183.75)*100 = 109.14% Symposium on Non- Metallic (FRP) Reinforcement for Concrete
Cost of FRP (Bottom and full sides) Structures, Sapporo, Japan, 1: 435–442

Area of FRP applied = 3* 0.15*0.7 m2 6. Khaled A., Soudki, Ted G. Sherwood., 2000. “Behaviour of reinforced
concrete beams strengthened with carbon fibre reinforced polymer
Cost of FRP = Rs. 236.25/- laminates subjected to corrosion damage”. Canadian Journal of
Civil Engineering 2000.
Economic Viability = (Rs. 200/ Rs. 236.25)*100 = 84.65%
7. Baiyasi M. & Harichandran R., 2001; “Corrosion and wrap strains
in concrete bridge columns repaired with FRP wraps” Proceedings
of the 80th Annual Meeting, Transportation Research Board,
Washington, DC. Berver E., Fowler D., Jirsa J. & Wheat H., July
2001; “Corrosion in FRP-wrapped concrete members”, Proceedings
of the Structural Faults and Repair Conference London.
8. Bonacci J., Thomas M., Hearn N, Lee C. & Maalej M., 10–13 June
1998; “Laboratory simulation of corrosion in reinforced concrete
and repair of CFRP wraps” Proceedings of the Annual Conference
of CSCE, Halifax, Nova Scotia.
9. Debaiky A., Green M. & Hope B., 2002; “Carbon fiber-reinforced
polymer wraps for corrosion control and rehabilitation of reinforced
concrete columns”, ACI Materials Journal 2002.
10. Kash E., October-2003; “The Mechanisms of Corrosion and Utilizing
Fiber Reinforced Polymers as a Chloride Barrier”
11. Farmington Hills, Michigan, Oct.-2002; ACI 4402R-02. “Guide for
the Design and Construction of Externally Bonded FRP Systems for
Strengthening Concrete Structures” ACI.
12. Song G., Shayan A.: July-1998; “Corrosion of steel in concrete:
Causes, detection and prediction, state of the art review”
13. Kwangsuk S., Gray M., Sen R., and Winters D., 2010; “Effective Repair
Fig. 5: Comparison between the two effective patterns for Corrosion Control Using FRP Wraps, Journal of composites for
construction”
14. Rajan Sen, 2003; “Advances in the application of FRP for repairing
References corrosion damage. New materials in construction”
1. ACI Committee 222. 1996. Corrosion of metals in concrete.
ACI222R-96. American Concrete Institute, Detroit, Mich. 29 pp.

Kolluru Hemanth Kumar

Kolluru Hemanth Kumar is a final year Under Graduate student in civil engineering at NIT-Bhopal with flair
in structural engineering and concrete technology
He worked on the presentation topic ‘Behaviour of reinforced concrete beams strengthened with Glass
Fibre Reinforced Polymer laminates subjected to corrosion damage’ with his fellow groupies during his
junior year.
It deals with prevention of corrosion in steel reinforcement in RCC structures with the influence of GFRP.
Apart from this he also worked on ‘Effects of soil-structure interaction on seismic response of existing RCC
building’, and is currently working on ‘Origami Inspired Deployable Popup Structures for disaster mitigation
strategies’ as a part of his academic curriculum.

Organised by
India Chapter of American Concrete Institute 129
Session 1 C - Paper 5

Behaviour of Synthetic Fibre Reinforced Prestressed Hollowcore Slabs

under Flexure-Shear
Sameer K. Sarma Pachalla, Pradeep K. and Dr. Suriya Prakash S.
Department of Civil Engineering, Indian Institute of Technology Hyderabad, Telangana, India

Abstract Literature Review and Research Significance

Prestressed Hollowcore slabs are generally precast In the past few decades, researchers have used fibres
elements which are cast with an extrusion machine. as secondary reinforcement for the concrete elements.
Although prestressed concrete members are designed as Cuenca and Serna[1] used steel fibres to control the
uncracked members, cracking is possible in unexpected crack propagation in precast beams. They found that fibres
loading conditions. In such cases, the behaviour of these and stirrups had a synergic effect. Very recently, Cuenca
elements can be improved by the addition of synthetic et al.[2] have investigated the influence of concrete matrix
fibres during casting process with no modifications to mix and type of fibre on the shear behaviour of self-compacting
design and at very low costs. Synthetic fibre reinforced fibre reinforced concrete beams. They considered two
hollow core slabs and control slabs are tested in the different concrete compressive strength values and
laboratory to evaluate the effect of synthetic fibres on five different types of steel fibres on shear behaviour of
their behaviour. Shear span to depth (a/d) ratios of 7.5 and reinforced concrete beams. They found that the type of
3.75 are considered for evaluation. It was found that in fibres used substantially affected shear behaviour. The
specimens with high a/d ratio of 7.5, even a low volume combination of high strength concrete matrixes with low
addition of polypropylene fibres at volume fraction of 0.33 strength fibres was not found to be efficient.
(3kg/m3) has increased the peak load by 19% along with
Previous researchers[3-5] have evaluated the behaviour of
increase in ductility. The peak load got increased up to 10
hollow core slabs with the addition of steel fibres in the
% in specimens with low a/d ratio 3.75.
concrete during casting. They found similar beneficial
Keywords: Hollow core slabs, precast, prestressed, effects on that of beams on prestressed hollow core
synthetic fibres, polypropylene, flexural behaviour. slabs. Paine et al.[5] observed that the cracking and peak
loads of fibre reinforced concrete (FRC) slabs are higher
when compared to that of slabs with normal concrete. The
Introduction
authors found that at relatively low a/d ratios, the code
Precast concrete elements were developed in early 20th equation was able to predict the shear strength of plain
century to meet the increasing construction needs. The hollow core slabs appropriately. However, the modified
main advantages of using precast elements are high shear equation proposed by other researchers for Steel
quality control, construction feasibility and reduced FRC, seemed to overestimate the strength. Cuenca and
construction time. Hollow core slabs are the advanced Serna[6] carried out an experimental program consisting
products of precast industry. These are prestressed of 26 hollow core slabs for different test variables
precast slab elements which have longitudinal cores including amount of steel fibres (0, 50 and 70 kg/m3)
running along the span. This results in reduction of and shear span/ depth (a/d) ratio of 2.3–4.4 and 8.6.
dead weight and being an efficient cross section with an The authors observed that hollow core slabs with fibres
increased effect of prestressing. Hollow-core slabs under achieved greater loads than control ones with a more
ideal conditions are usually designed to resist flexural ductile behaviour. Marazzini et al.[7] in their experiments
stresses (moments) due to assumed uniformly distributed on individual cut beams of hollow core slabs with and
loads. Under service loads, prestressed elements are without steel fibres found that for high strength concretes
usually designed as uncracked elements. However, of range 80-130 MPa, the fibre reinforced concrete has
in few cases under unexpected loading conditions, tendon-to-concrete bond about twice as large as that
these elements can be cracked and may not meet the of plain concrete. They also observed that the post peak
serviceability requirements. As the slabs are cast with an response of fibre reinforced specimens is very smooth
extrusion machine, provision of extra reinforcement is not when compared to that of normal concrete. Though the
very easy. For such cases, addition of fibres in concrete amount of increase varies, all the researchers confirmed
during casting will enhance both the flexural and shear that the addition of steel fibres increased the peak load
behaviour of hollow core slabs. and ductility of hollow core slabs. The amount of increase

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

130 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of Synthetic Fibre Reinforced Prestressed Hollowcore Slabs under Flexure-Shear

depends on various factors like a/d ratio, compressive and

tensile strengths of concrete and amount of fibres added.
The main disadvantages of using steel fibres are high
specific gravity of concrete and corrosion of steel fibres.
To overcome these problems, synthetic fibres are recently
being used instead of steel fibres.
Mechanical properties including compressive and tensile
strength of fibre-reinforced (both steel and synthetic) Fig. 1: Cross Section Details of Hollowcore Slabs
concrete have been relatively studied well (Olivito and
Zuccarello [8]; Soulioti et al. [9] Barros et al.[10]). Li[11] found
Table 1
that the polypropylene fibres only marginally increased Nomenclature of Specimens
flexural tensile strength. However, after cracking, the
fibres were found to greatly increase the ultimate strain, Specimen
Label
Fibre content Volume a/d
though the load carrying capacity is decreased. However, Number (kg/m3) Fraction (%) ratio

the fracture mechanisms and fracture energy of synthetic 1 HCS-150-7.5-0 0 0 7.50
fibre- reinforced concrete is still a matter of interest. It is 2 HCS-150-7.5-3 3 0.33 7.50
in fracture processes where the fibres absorb energy and
provide ductility and toughness to the FRC. 3 HCS-150-3.75-0 0 0 3.75

4 HCS-150-3.75-3 3 0.33 3.75

Cifuentes et al.[12] in their experiments on polypropylene
fibre reinforced beams found that the effect of the fibre is
more remarkable in the case of the low strength concrete Material Properties
which is attributed to the stresses in cohesive zone which
are directly proportional to strength of concrete. The
Concrete
investigation by Lanzoni et al. [13] suggests that despite the All specimens were cast using normal weight, ready-mix
lower dosage of synthetic fibres, they effectively reduces concrete with a target compressive strength of 40 MPa at
crack formation and growth to an extent comparable to 28 days. The unit weight of concrete was 2400 kg/m3. The
that of traditional steel fibres. Four-point loading bending tested concrete cube strength is 42 MPa at 28 days.
test data showed that synthetic fibres had positive Internal reinforcement - prestressing steel tendons
contribution to the toughness index, which suggests
The type of strands used were seven-wire low-relaxation
that adoption of polypropylene-based FRC for structural
strands with an ultimate tensile strength of 1860 MPa
elements lends enhanced toughness and durability. It is
and modulus of elasticity of 196.5GPa. 9.53 mm diameter
worth mentioning that no previous work has evaluated the strands were used at the bottom with a jacking force of
behaviour of prestressed precast hollow core slabs with 70kN.
synthetic fibres as reinforcement. This paper provides the
basic understanding on the behaviour of synthetic fibre Synthetic fibres - Polypropylene
reinforced hollow core slabs at low and high shear span Structural Polypropylene (PP) fibres with a length of
to depth ratios. 60mm and 0.50mm diameter were used (Table 1). The
fibres have a modulus of elasticity of about 10 GPa and
Experimental Investigation tensile strength between 550 and 640MPa. The fibres have
continually embossed surface anchorage mechanism to
General enhance bond.
Four hollowcore slabs were tested in the laboratory to
evaluate the effect of fibres on their behaviour. Two slabs Table 2
Properties of Poly-Propylene fibres
were tested with a/d ratio of 7.5 and two were tested with
a/d ratio of 3.75. All the slabs had same geometry as Material Poly-propylene
shown in the Figure 1. The tested slabs were cut to 600mm
Form Structural fibre
width from a full slab of 1200mm wide. The slabs were
prestressed with 3 number of strands of 9.53mm diameter Specific Gravity 0.91
with a prestressing force of 70kN each. Polypropylene
Length 50mm
synthetic fibres are added to 2 slabs during casting. The
weight of fibres added is 3kg per m3 of concrete volume. Diameter 0.5 mm
All the slabs were cast on same day and were cured for Tensile Strength 618 N/mm2
the same time at same temperature. The test matrix is
Modulus of Elasticity 10 GPa
show in Table 1.

Organised by
India Chapter of American Concrete Institute 131
Session 1 C - Paper 5

Instrumentation Test Results and Discussion

All specimens had similar instrumentation details.
Deflections were recorded using linear variable Behaviour of Control Specimen at High a/d Ratio (HCS-
differential transducers (LVDT). Specific locations of 150-7.5-0)
LVDTs were chosen to capture the entire curvature Control slab (HCS-150-7.5-0) with no fiber reinforcement
profile during testing. Two 50mm LVDT were positioned was tested at high a/d ratio of 7.5. The first crack occurred
at one third clear span distance and one 100mm LVDT was at a load of 23.7kN in the constant moment zone. One
positioned at the center of the span to accurately capture further loading, there was crack distribution in the
mid-span deflection. TML strain gauges with a gauge constant moment zone and the strands yielded. The
length of 120 mm were used to measure the strains in peak load reached was 42.8kN and the ultimate failure
the concrete. Surface was thoroughly cleaned and strain occurred due to crushing of concrete under loading line.
gauges were installed at top and bottom at the center line
on the concrete slab to capture the strain profile. HBM
data acquisition system was used to capture the data.
Strain gauges of 5mm gauge length were also installed
on the prestressing strands to capture the strain in the
strands during testing.

Test Setup and Loading Details

Slabs were tested in a four-point loading configuration
(Fig. 2) by which constant moment region along the mid-
span was obtained. Figure 2 illustrates the components
in the test setup. A 250 kN MTS hydraulic actuator was
utilized to apply two concentrated loads centered over Fig. 3: Crack Distribution at Failure for HCS-150-7.5-0
the mid-span of each test specimen. The loads were
transferred to the concrete specimens via a single Behaviour of Fibre Reinforced Specimen at High a/d
longitudinal rigid steel spreader beam, stiffened with Ratio (HCS-150-7.5-3)
web stiffeners for high rigidity. The load from spreader This slab contains fibres in the concrete and is tested at a
beam was transferred to the slab through two transverse flexure dominated a/d ratio of 7.5. The first cracking load
I beams as two distributed line loads along the full width was 31.9 kN which occurred in the constant moment zone.
On further loading, crack distribution in the moment zone
occurred. The fibres came to effect after cracking, and
the peak load reached to 50.9 kN. The load vs mid span
deflection for HCS-150-7.5-0 and HCS-150-7.5-3 is shown
in Figure 5. From the load deflection curves (Figure 5),
it is observed that there is no change in initial stiffness
but the cracking load increased from 23.7 kN to 31.9 kN
(34.5%). After cracking, the response of both the slabs is
similar and the peak load increased from 42.8 kN to 50.9
kN (18.9%) due to fiber contribution in load resistance.

Fig. 2: Schematic Test Setup

of the slab (Figure 2). High strength cement mortar was

used between the two transverse spreader I-beams
and the slab to remove the surface irregularities and
to avoid stress concentrations. End supports included Fig. 4: Crack Pattern and Fibres Resisting Crack Opening in
I-beams stiffened with transverse stiffeners to avoid any HCS-150-7.5-3
sudden buckling failure and transfer the loads safely to
the loading frame. Loading was applied monotonically in Behaviour of Control Specimen with load a/d ratio
displacement control mode with a rate of 0.05 mm/sec. (HCS-150-3.75-0)
Loading was paused intermittently to observe the crack This specimen was tested low a/d ratio of 3.75. Low a/d
patterns and change in failure modes. ratio led to higher load resistance of slab in flexure with

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

132 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of Synthetic Fibre Reinforced Prestressed Hollowcore Slabs under Flexure-Shear

Fig. 5: Comparison of Load vs deflection Curves for Specimens Fig. 8: Comparison of Load vs Deflection Curves for Specimens
Tested at High a/d (7.5) Ratio Tested at Low a/d (3.75) Ratio

a larger crack distribution. The first crack occurred at a summarizes the results. Addition of fibres significantly
load of 65 kN and on further loading crack distribution increased the strain energy dissipation.
occurred in the constant moment zone. The peak load was
95 kN. Table 3
Nomenclature of Specimens

Specimen Strain energy

Label Peak load (kN)
Number (kN-m)

1 HCS-150-7.5-0 42.8 343.5

2 HCS-150-7.5-3 50.9 (19%) 433.5 (26.2%)

3 HCS-150-3.75-0 95 2420.5
Fig. 6: Crack distribution in HCS-150-3.75-0 4 HCS-150-3.75-3 103.8 (9%) 2526.5 (4.3%)

Behaviour of Fibre Reinforced Specimen with Low a/d

Ratio (HCS-150-3.75-3) Conclusions
This fibre reinforced specimen was tested at a lower a/d The behaviour of the tested fibre reinforced hollow core
ratio of 3.75. The first cracking load was 63 kN which slabs shows that the addition of fibres is very effective
occurred in the constant moment zone. On further loading, in controlling the crack opening and thereby leading to
crack distribution in the moment zone occurred (Figure 7). increased ductility. Following are the main inferences
The fibres came to effect after cracking, and the peak load from the test results:
reached to 103.8 kN. The load vs mid span deflection for i) Addition of fibres increased the peak load carrying
HCS-150-3.75-0 and HCS-150-3.75-3 is shown in Figure capacity of slabs tested at both the low and high a/d
8. From the curves we can observe that there is no change ratios. However, the increase is inversely proportional
in initial stiffness and the cracking load. After cracking, to the a/d ratio of the specimens.
the response of both the slabs is similar and the peak load
ii) Increase of cracking load only occurred for flexure
increased from 95 kN to 103.8 kN (9.2%).
dominated specimen tested at a higher a/d ratios
iii) There is substantial increase (26.2%) of strain energy
in the high a/d ratio specimens by the addition of fibres.
iv) Addition of fibres in general increased the toughness
of the specimens indicating better crack bridging by
the fibres.

Acknowledgments
Fig. 7: Crack Distribution in HCS-150-3.75-3 This experimental work is carried out as part of the project
funded by Prime Minister Fellowship grant sponsored by
Strain Energy Department of Science and Technology and Confederation of
The strain energy of the tested specimens is calculated Indian Industry (CII) India. Their financial support is gratefully
from the load deflection response curves. The area under acknowledged. We also acknowledge PRECA Ltd, Hyderabad
the load deflection curve represents the amount of strain India for their financial support and help in casting of hollow
energy stored in the elements during testing. Table 3 core slabs used in this study.

Organised by
India Chapter of American Concrete Institute 133
Session 1 C - Paper 5

References Concrete Beams: Material and Structural Behavior, ACI Special

Publication 182, 29-52
1. Cuenca, E., Serna, P., 2013. Shear behavior of prestressed precast
beams made of self-compacting fibre reinforced concrete, 8. Olivito, R.S., Zuccarello, F.A., 2010. An experimental study on the
Construction and Building Materials, 45:145–156. tensile strength of steel fibre reinforced concrete. Composites Part
B: Engineering, 41(3):246-255.
2. Cuenca, E., Echegaray-Oviedo, J., Serna, P., 2015. Influence of
concrete matrix and type of fibre on the shear behavior of self- 9. Soulioti, D.V., Barkoula, N.M., Paipetis, A., Matikas, T.E., 2011. Effects
compacting fibre reinforced concrete beams, Composites Part B of Fibre Geometry and Volume Fraction on the Flexural Behaviour
75:135-147. of Steel‐Fibre Reinforced Concrete, Strain 47 (S1): 535-541.
3. Paine, K.A., 1998. Steel fibre reinforced concrete for prestressed 10. Barros, J. A., Figueiras, J.A., 1999. Flexural behavior of SFRC:
hollow core slabs, PhD thesis, University of Nottingham:325 testing and modeling. Journal of materials in civil engineering,
11(4):331-339.
4. Paine, K.A., Elliott, K.S., Peaston, C.H., 1997. "Increasing the Shear
Strength and Ductility of Prestressed Hollow Core Slabs using 11. Li, V., 2002. Large Volume, High-performance applications of fibres
Metal Fibres" Fourth International Symposium on Noteworthy in civil engineering. Journal of Applied Polymer Science, 83:660-
Developments in Prestressing and Precasting, Singapore:116-124 686.
5. Peaston, C., Elliott, K., Paine K., 1999. Steel fibre reinforcement for 12. Cifuentes, H., García, F., Maeso, O., Medina, F., 2013. Influence of
extruded prestressed hollow core slabs. ACI Special Publication, the properties of polypropylene fibres on the fracture behaviour of
182:87–107. low-, normal- and high-strength FRC, Construction and Building
Materials, 45:130-137
6. Cuenca, E., Serna, P., 2013. Failure modes and shear design of
prestressed hollow core slabs made of fibre-reinforced concrete, 13. Lanzoni, A., Nobili, A.M., Tarantino, 2012. Performance evaluation
Composites: Part B 45 952–964. of a polypropylene-based draw- wired fibre for concrete structures,
Construction and Building Materials, 28:798-806
7. Marazzini, M., Rosati, G., 1999. Fiber Reinforced High-Performance

Sameer Kumar Sarma Pachalla

Sameer Kumar Sarma Pachalla is a research scholar at Department of Civil Engineering, Indian Institute of
Technology Hyderabad, India. His research interests include behavior of reinforced/prestressed concrete
members and strengthening of concrete members using FRP composites. Before joining at IITH he worked
as a design engineer for Mumbai Monorail and Hyderabad Metro projects. He received his M-tech from IIT
BHU, India and B-tech from Osmania University.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

134 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 2 A
Session 2 A - Paper 1

Shrinkage Cracking of Self Compacting Concrete (SCC) with

Supplementary Cementitious Materials
Salah Al Toubat , Moussa Leblouba, Deena Badran
Department of Civil & Environmental Engineering, University of Sharjah, United Arab Emirates

Abstract Introduction
The use of supplementary cementitious materials Self Compacting Concrete which was described as the
(SCMs) such as ground granulated blast furnace slag quiet revolution in the concrete construction[9] can be
(GGBS), micro silica (MS), and fly ash (FA) in self- defined as the concrete that does not require vibration
compacting concrete (SCC) is widely recognized as a for placing and compaction, it is able to flow under
good way to enhance its fresh properties and to reduce its own weight, completely fill formwork and achieve
the amount of cement. Such use of SCMs contributes to full compaction, even in the presence of congested
a more sustainable and green construction. However; reinforcement whilst maintaining its homogeneity,
without the need for any additional compaction[21]. Self
SCC is more prone to shrinkage cracking due to the
Compacting Concrete comparing with normal concrete
low water-to-cementitious ratio and high content of fine
is subjected to low content of coarse aggregate, w/c
materials typically specified in SCC. Shrinkage cracking
ratio and high content of binder, superplasticiser and
of SCC with SCMs still needs further investigation, and sometimes the use of viscosity enhancer admixture. In self
an experimental program was conducted to assess the compacting concrete, cement is usually partially replaced
restrained shrinkage behavior and cracking potential by supplementary cementitious materials (SCMs) such as;
of SCC mixes with various SCMs. The effect of type and Fly Ash, GGBS and Micro Silica to increase its fluidity or
proportion of SCMs, degree of restraint and curing regime, cohesiveness or to minimize the heat generated from its
were specifically addressed. In this paper, the effect hydration process. Supplementary cementitious materials
of GGBS, MS, and a combined GGBS-FA on shrinkage are defined as pozzolanic materials that influence the
cracking is presented. A test program was carried out hydration reactions and make a significant contribution to
on a control SCC with pure cement, four SCC mixes with hydration products [26]. These materials in the presence of
different amounts of GGBS replacing cement by 35%, Calcium Oxide (CaO) act as hydraulic cement, they react
50% and 70% and one SCC mix with combined GGBS- with water to form Calcium Silicate Hydrate (C-S-H) and
FA (35%-35%) equally replacing cement by 35% each, this reaction; which called a pozzolanic reaction, can
and two SCC mixes with 5% and 10% of MS replacing the occur at ordinary temperatures. The calcium oxide (CaO)
cement. The results revealed that the cracking potential needed to start the pozzolanic reaction, is provided by the
of the SCC mixes was affected significantly by the type hydration products of cement; calcium hydroxide (CH) and
and proportion of the supplementary cementitious calcium silicate hydrate (C-S-H)[26],[23]. Studies showed
that the resistance level of SCC to shrinkage cracks was
materials, curing regime and degree of restraint.
quite different depending on type of powdered materials
In addition, it was found that a 50% of GGBS can be
used[30]. In addition, studies show that the degree of
effectively used for concrete applications subjected to
autogenous shrinkage of SCC depends on powdered
high degree of restraint and around 70% for applications material types used and it becomes especially larger
with low degree of restraint without compromising its with increasing of GGBS content[10],[8],[19]. Akkaya, et al.
shrinkage cracking resistance provided that adequate on the other hand concluded that using supplementary
moist curing is adopted. Furthermore, it is demonstrated cementitious materials increases the drying shrinkage
that the proportion of MS in SCC should be less than 10% and decreases the autogenous shrinkage; but no
for all applications regardless of the degree of restraint. significant difference in the total shrinkage will be noticed
The results with GGBS and MS presented in this paper and under restraint condition using such materials
are also discussed together with those obtained with FA will delay the age of cracking[1]. Curing is an important
in this research program to better assess the shrinkage factor and should be taken in to consideration to achieve
cracking resistance of SCC with various SCMs. a sufficient strength required to reduce the risk for
cracking of concrete[20] especially when one or more
Keywords: Self compacting material; GGBS; Fly ash; of supplementary cementitious materials are used
Micro silica; Cracking; Shrinkage.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

136 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Shrinkage Cracking of Self Compacting Concrete (SCC) with Supplementary Cementitious Materials

because the pozzolanic reaction can only proceed in the Seven self compacting concrete mixes were designed
presence of water [23]. Researches illustrated that using and tested. All the mixes were designed for a maximum
moist curing for seven days with mixes contain fly ash aggregate size of 10 mm which is the maximum size
gives the maximum compressive strength[6], in contrast specified by ASTM 1581 Standard Ring Test. The total
the continued presence of water is required for GGBS cementitious materials content and the w/c used are
reaction to continue and attain the maximum strength[24]. typically recommended for the Arabian Gulf Peninsula [27].
This is due to the slower, but prolonged reaction (hydraulic In addition, the content of coarse and fine aggregate is in
and pozzolanic) between cement hydration products line with the recommendation of the European Guidelines
and GGBS, which contributes significantly to strength. for Self Compacting Concrete[21]. The proportions of fly ash
Curing is probably the most important aspect of micro adopted generally match with the values in British Code
silica concrete as the material undergoes virtually zero BS 8500-2: 2006 and those recommended for Arabian
bleeding. If the rate of evaporation from the surface is Gulf Peninsula [7], [27].
faster than the rate of migration of water from interior to
the surface, plastic shrinkage takes place. In the absence The dosages of admixtures were adjusted such that
of bleeding and slow movement of water from interior to the mixes exhibited similar fresh properties. The fresh
the surface, early curing of membrane is essential[23]. properties of SCC were assessed based on the slump flow,
T500 time and L-Box ratio. The slump flow is the mean
diameter of the spread of fresh concrete using a conventional
Materials and Methods slump cone. The time to reach a spread of 500mm is the
T500 Time. L- Box ratio measures the passing ability of SCC
Materials and Mix Properties mix. The recommendation of a slump flow in the range of
The concrete mix proportions for the seven mixes are 600 to 750mm, a T500 time between 3 to 10 seconds and
presented in Table 1. Ordinary Portland cement (Type I) was L-Box ratio ≥ 0.7 were targeted in this study[11],[16]. The fresh
used in this research. The Fly Ash used in the mixtures is properties measured for all mixes were presented Table 2.
originally from India and meets the requirements of Class
F Fly Ash according to ASTM specifications while the Micro
silica is originally from Norway. The coarse aggregates Experimental Work
are 10 mm crushed Gabro while the fine aggregates are In the laboratory tests, three different curing regimes
mainly crushed dry sand, crushed washed sand, and were observed as shown in Table 3. In the first curing
dune sand. CHRYSO® Fluid Optima 230, which is a Poly regime, the specimens were sealed for 24 hours using
Carboxylated based high range water reducing admixture plastic sheet and then exposed to air drying. In the second
was used in addition to Feyplast SUB-AQUA that comes in curing regime the specimens were sealed for 24 hours
a powder form to increase the viscosity of the water. using plastic sheet then moist-cured for three days using

Table 1
SCC Mixes Proportions

Mix Total
No. Supplementary Cementitious Coarse Fine
w/c Cement (Cement + HRWRA VEA
Materials (SCMs) Aggregate Aggregate
SCMs)

% used to
Kg/m3 Type replace the Kg/m3 Kg/m3 Kg/m3 Kg/m3 L/m3 Kg/m3
Cement

1 0.36 450 - - - 450 736 1101 6.5 0.6

2 0.36 292 GGBS 35 158 450 747 1096 5.5 -

3 0.36 225 GGBS 50 225 450 748 1094 6 0.5

4 0.36 135 GGBS 70 315 450 738 1083 5.5 0.6

GGBS 35 157
5 0.36 136 450 775 1017 6.5 0.5
Fly Ash 35 157

6 0.36 405 Micro Silica 5 45 450 742 1093 5.5 0.5

7 0.36 427.5 Micro Silica 10 22.5 450 748 1097 5 0.5

Legend: HRWRA: High Range Water Reducing Admixture. VEA: Viscosity Enhancer Admixture. SCMs: Supplementary Cementitious Materials.

Organised by
India Chapter of American Concrete Institute 137
Session 2 A - Paper 1

Table 2
Fresh Properties of SCC Mixes

Fresh Properties
Mix
Slump Flow (mm) Flow Rate (S) L-Box Ratio

CONTROL 620 10 0.70

35%GGBS 650 6 0.85

50%GGBS 680 3 0.90

70%GGBS 680 7 0.90

35%GGBS+35%FLY ASH 620 9 Not Measured

5% MICRO SILICA 620 10 0.70

10% MICRO SILICA 600 9 Not Measured

Table 3
Experimental Program & Test Matrix

Restrained Shrinkage Test (Ring Test)

Mix % of Cement Replaced by

Mix no. ASTM C1581 Standard Ring Test AASHTO PP34 Standard Ring Test
Category GGBS/ Micro Silica

Curing Curing Curing Curing Curing Curing

Regime No.1 Regime No.2 Regime No.3 Regime No.1 Regime No.2 Regime No.3

Control 1 - - X X X X X X

2 GGBS 35 X X X X

3 GGBS 50 X X X X
GGBS- SCC
4 GGBS 70 X X
Mixes
GGBS 35 X X
5
Fly Ash 35

6 Micro Silica 5 X X
Micro Silica-
SCC Mixes
7 Micro Silica 10 X X

Note: Curing Regime No.1: Sealing for 24 hrs before exposure to air drying. Curing Regime No.2: Sealing for 24 hrs then moist curing for 3 days before exposure to air
drying. Curing Regime No.3: Sealing for 24 hrs then moist curing for 7 days before exposure to air drying.

Fig. 1: AASHTO-PP34 and ASTM 1581 standard rings Fig. 2: Experimental Setup

wet burlap and in day four they were exposed to air drying. instead of three days. All curing regimes were adopted for
The third curing regime was similar to the second one but the control mix to be as a reference for other mixes. For
the specimens were moist cured for approximately 7 days other mixes, the method of curing was selected depending

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

138 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Shrinkage Cracking of Self Compacting Concrete (SCC) with Supplementary Cementitious Materials

on the earlier expectation about the pozzolanic reaction.

The restrained shrinkage test was performed using two
standard ring tests; ASTM C1581 and AASHTO PP34 Rings
(Figure 1). The two standard rings provide tawo different
degree of restraint. In general, the degree of restraint
provided by ASTM Ring varies between 70 to 80% while in
the AASHTO Ring it varies between 50 to 60%. Two ASTM
Rings and Two AASHTO Rings were cast for each mix
under the curing regime specified in Table 1 (see Figure 2).
Furthermore, two prismatic specimens with dimensions
equal to 280x 80x 80 mm were cast for each SCC mix to
measure their free drying shrinkage with time according to
ASTM C157 Standard Test. The free shrinkage specimens Fig. 4: Cont. Steel Ring Strain vs. Time of GGBS- SCC Mixes in
ASTM Rings
had the same curing regime as their corresponding ring
specimens. In addition, standard cubes (150x150x150 mm)
shrinkage potential and the degree of relaxation due to
and cylinders (150mm x 300mm) were used to measure
tensile creep[2],[22]. The net effects of these factors govern
the compressive and tensile strength at different ages; 3,
the cracking potential of SCC which in this research was
7, 28, 56 and 100 days.
found sensitive to the proportion of SCMs, curing regime
and degree of restraint. The effect of curing regime, type
Results and Discussion and proportion of SCMs and degree of restraint on the
The steel strains of the ASTM and AASHTO Ring factors mentioned above was analyzed carefully to explain
Specimens that used GGBS and Micro Silica to replace the cracking behavior of the SCC mixes.
the cement by different proportions and under different
curing regimes are shown in Figure 3, Figure 4, Figure 6 Restrained Shrinkage Behavior of SCC with GGBS
and Figure 7, respectively. The sudden drop in the strain High Degree of Restraint (ASTM Rings)
value refers to the age at first cracking which was used The results presented in Table 4 show that the use of
in addition to the net time-to-cracking as assessing moist curing for three days with the mix containing 35%
parameters to compare the cracking potential of different GGBS shifted its age at cracking from 3.3 (immediate
mixes. The net time-to-cracking is defined as the time exposure to air drying) to 7.4 days. It also increased the
from the initiation of air drying to the age at first cracking. corresponding net time-to-cracking from 2.6 to 3.7 days.
The age at first cracking and net time-to-cracking for the Furthermore, the results show that the increase of moist
SCC mixes in ASTM and AASHTO Rings are summarized curing period from three to seven days shifted the age at
in Table 4. It should be emphasized herein that the ASTM cracking for the mix containing 50% GGBS from 8.3 to
Ring provides relatively high degree of restraint and the 11.6 days. This trend was manifested in the net time-to-
AASHTO Ring provides relatively low degree of restraint. cracking which increased from 4.6 to 4.8 days.
Therefore the results and findings were discussed under The results presented in Table 4 reveal that the age at
two separate categories; namely high degree of restraint cracking delayed from 7.4 to 8.3 days for specimens
and low degree of restraint. The key parameters that subjected to three days of moist curing when the proportion
influence cracking potential of concrete includes the rate of GGBS increased from 35% to 50%. The net time-
of induced tensile stress and strength developments, to-cracking followed the same manner and increased
from 3.7 to 4.6 days. On the other hand, the increase in
GGBS proportion from 50% to 70% in the specimens
subjected to seven days of moist curing decreased the age
at cracking from 11.6 to 10.7 days. It also decreased the
net time to cracking from 4.8 to 3.9 days. Therefore, the
GGBS can be used to replace up to 50% of the cement with
significant improvement in the crack resistance. However;
the resistance to cracking decreased when the GGBS was
added at 70% to replace the cement. This suggests an
adequate proportion of GGBS in the neighborhood of 50%
for structures subjected to high degree of restraint as that
induced in the ASTM Rings, provided that the mix should
be accompanied with moist curing for at least 3 days and
preferably 7 days.
Fig. 3: Steel Ring Strain vs. Time of GGBS- SCC Mixes in the
ASTM Rings
Significant increase was noticed in the net time-to-
cracking when GGBS was combined in equal proportion

Organised by
India Chapter of American Concrete Institute 139
Session 2 A - Paper 1

when the GGBS was used in combination with fly ash at

equal proportion to replace 70% of cement, the potential
for shrinkage reduced significantly relative to all GGBS
mixes provided that a moist curing for at least 3 days is
adopted. This mix also exhibited the highest cracking
resistance among all mixes that contained GGBS alone as
shown in Table 4.

Low Degree of Restraint (AASHTO Rings)

The results obtained from AASHTO Rings and summarized
in Table 4 suggest that the curing of the GGBS significantly
affected its cracking resistance. The age at cracking for the
Fig. 5: Free Shrinkage Strain of GGBS- SCC Mixes up to 14 Days mix containing 35% GGBS delayed from 14.5 to 21.8 days
when it was subjected to three days of moist curing instead
with fly ash to replace 70% of cement. The net time-to- of immediate air drying after de-molding. The difference
cracking as shown in Table 4 increased from 3.9 days for between the ages at cracking was 7.3 days compared to
the mix containing 70% of GGBS only to 11.4 days for the 4.1 days in the corresponding ASTM Specimens. For the
mix containing both Fly ash and GGBS. specimens containing 50% GGBS, the increase in the
moist curing period from three to seven days shifts the
The shrinkage potential as revealed from the free age at cracking from18.0 to 29.2 days. The difference
shrinkage strains shown in Figure 5 affected significantly was 11 days compared to 3.3 days for the corresponding
by the proportion of GGBS and the curing regimes. The ASTM Specimens. The low rate of stress development in
results show that moist curing is necessary for mixes the AASHTO Rings helped the effect of curing regimes to
containing GGBS. The period of the moist curing depends manifest itself on the age at cracking. The results also
on the proportion of GGBS used in the mix to replace the indicate that the moist curing for three days increased the
cement. Apparently, as the proportion increased beyond net time-to-cracking for the mix containing 35% GGBS
35%, the curing period should be extended beyond 3 from 13.8 (air drying after de-molding) to 18.1 days. The
days and preferably for 7 days. The results show that the net time to cracking also increased from 14.1 to 22.3 days
mix containing 50% GGBS exhibited similar shrinkage when the mix containing 50% of GGBS was subjected to
potential as that for the 35% GGBS mix when it was seven days of moist curing instead of three days.
subjected to 3 days moist curing. However; when the
moist curing extended to 7 days, significant reduction in In addition to curing regime, the effect of GGBS
the shrinkage was observed. The shrinkage potential for proportions can also be seen in the results presented
GGBS mixes is in line with their resistance to cracking in in Table 4. The age at cracking reduced from 21.8 to18.0
ASTM Ring Test which is characterized as high degree of days in specimens subjected to three days of moist curing
restraint. Therefore, the results support the hypothesis when GGBS proportion was increased from 35% to 50%.
that for high degree of restraint condition, the optimum Similarly, the net time-to-cracking decreased from 18.1 to
proportion of GGBS is a round 50% provided that moist 14.1 days. This may be attributed to the need for subjecting
curing for 7 days is adopted. The results also show that the high proportion GGBS- SCC mixes (GGBS≥50%) to at

Table 4
LCI Data for Conventional Concrete and Concrete Containing LF

Mix ASTM Rings AASHTO Rings

Age at Cracking (Days) Net Time to Cracking (Days) Age at Cracking (Days) Net Time to Cracking (Days)

35%GGBS-Air Drying 3.3 2.6 14.5 13.8

35%GGBS-3Days Moist 7.4 3.7 21.8 18.1

50%GGBS-3Days Moist 8.3 4.6 18.0 14.1

50%GGBS-7Days Moist 11.6 4.8 29.2 22.3

70%GGBS-7Days Moist 10.7 3.9 34.7 27.9

35%GGBS+35%Fly Ash- 15.3 11.4 - -

3Days Moist

5%Micro Silica- Air Drying 5.9 5.0 40.1 39.0

10%Micro Silica-Air Drying 4.3 3.4 24.7 23.8

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

140 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Shrinkage Cracking of Self Compacting Concrete (SCC) with Supplementary Cementitious Materials

least seven days of moist curing. Table 4 also indicated to 10%. The strain curves with time for the two mixes
that the increase in GGBS proportion from 50% to 70% presented in Figure 6 were approximately the same in the
in 7days- moist-cured specimens delayed the age at first two days from casting the concrete. After which the
cracking from 29.2 to 34.7 days and increased the net strain for the mix containing 10% of micro silica becomes
time-to-cracking from 27.9 to 39.0 days. This indicates higher. This means that the rate of stress development in
that curing is essential for mixes containing GGBS. The the mix contains 10% of micro silica increased after 2 days.
GGBS can be added up to 70% to replace the cement; Therefore, the results suggest that micro silica should be
but curing period should be extended beyond three days, used at a proportion below 10%.
implying that for the concrete structure subjected to low
degree of restraint as that provided by the AASHTO Rings,
the high proportion of GGBS (up to 70%) can be used
without compromising the shrinkage crack resistance for
the mix. However; this proportion was not recommended
for a high degree of restraint condition as that provided by
the ASTM rings where the proportion of GGBS was limited
to 50%. This attributed to the low rate of tensile stress
induced by the AASHTO rings relative to that induced by
the ASTM rings. The low stress rate allowed the mixes
to exhibit a good resistance to shrinkage cracks at early
age. Therefore, the cracking resistance for the mixes
was affected by their behavior at later ages. This support
the previous argument that moist curing is essential for Fig. 7: Steel Ring Strain vs. Time of Micro Silica- SCC Mixes in
the GGBS mixes and that GGBS can be used as high as AASHTO Rings
70% for application with relative low to medium degree
of restraint similar to that induced by AASHTO rings Low Degree of Restraint (AASHTO Rings)
(50 to 60%) provided that moist curing for 7 days is adopted. The results for the AASHTO Rings indicate that the age at
cracking for SCC mixes contain 5% and 10% of micro silica
Restrained Shrinkage Behavior of SCC with Micro are significantly greater than that for the corresponding
Silica ASTM Rings. The results also show that the age at cracking
significantly decreased when the proportion of micro silica
High Degree of Restraint (ASTM Rings) increased from 5% to 10%. This trend matches with that
Table 4 presents the age at cracking and net time-to- in the corresponding ASTM rings. Therefore, micro silica
cracking for the ASTM Ring Specimens that used micro should be used at a proportion below 10%.
silica to replace 5% and 10 % of cement. One curing
regime was only used with SCC mixes containing micro Comparison between the Cracking Potential of SCC
silica which was exposure to immediate air drying after Mixes with GGBS, MS, and FA
de-molding. The net time-to-cracking for SCC mixes in ASTM Rings is
The results indicate that the age at cracking decreased summarized in Figure 8. Careful look at the results for the
when the proportion of micro silica increased from 5% to mixes in ASTM Rings reveal that the percentage of fly ash
10%. The net time-to-cracking also decreased from 5.0 used to replace the cement affect the crack resistance
to 3.4 days by increasing the proportion of micro silica of the mix and govern the level of required moist curing.
The results suggest that fly ash can be added up to 35%
without compromising the cracking resistance of the SCC
mix. In addition, curing for at least 3 days is recommended
for this optimum proportion. The degree of restraint in
the ASTM Test is in the range of 70 – 80% which means
that for high restraint structure the use of fly ash is not
recommended to be beyond 35%. However; if fly ash is
used beyond this limit, the results suggest that 7% micro
silica must be added in this case as can be seen from the
results obtained from the mixes containing 35 and 50% fly
ash in which the crack resistance of the mix significantly
improved when 7% micro silica was added. The addition of
micro silica in combination with fly ash generally improves
the crack resistance of the mix. The results of the GGBS
Fig. 6: Steel Ring Strain vs. Time of Micro Silica- SCC Mixes in
mixes suggest that the crack resistance is influenced
the ASTM Rings

Organised by
India Chapter of American Concrete Institute 141
Session 2 A - Paper 1

Fig. 8: Net Time-to-Cracking of the SCC Mixes in ASTM Rings

Fig. 9: Net Time to Cracking for SCC Mixes in AASHTO Rings

by the proportion of the GGBS and the curing regime. However; the addition of micro silica by 5% improved the
Furthermore, adding the GGBS by up to 50% improves the resistance to cracking. As a consequence, the micro silica
crack resistance of the mix provided that moist curing is proportion is recommended to be lower than 10%.
adopted for at least 3 days. This limit should be linked to
the high degree of restraint in the ASTM Test. However; The net time to cracking results obtained from AASHTO
if lower degree of restraint is expected, the limit may go test is shown in Figure 9. Careful examinations of the
beyond 50% as will be discussed later in the AASHTO results indicate that the fly ash proportion and curing
Test. For a high degree of restraint the combination of fly regime influenced the crack resistance. The results
ash and GGBS leads to effective enhancement in the crack suggest that fly ash can be added by up to 50% with
resistance of the mix as can be seen from the results of significant improved crack resistance provided that the
the mix containing 35% GGBS and 35% fly ash. The results mix is moist curing for at least 3 days. However; extending
also suggest that the addition of micro silica at 10% did not the moist curing to 7 days is further improve the mix
improve the crack resistance relative to the control mix. resistance to cracking. The 50% proposed limit of fly ash

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

142 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Shrinkage Cracking of Self Compacting Concrete (SCC) with Supplementary Cementitious Materials

Fig. 10: Normalization of the Results obtained from ASTM Rings

Fig. 11: Normalization of the Results obtained from AASHTO Rings

is more effective than that recommended based on the significantly improving its shrinkage crack resistance. The
ASTM Test results which was 35%. This may be attributed results of the mixes with GGBS showed that the shrinkage
to the degree of restraint provided by the AASHTO ring cracks resistance of the mix improved as the moist curing
which is in the range between 50 and 60%. This implies period increased. The results propose that GGBS can be
that for concrete structure with low degree of restraint, the added by up to 70% to replace the cement with improved
fly ash can be used up to 50% to replace the cement while crack resistance provided that moist curing for at least 7

Organised by
India Chapter of American Concrete Institute 143
Session 2 A - Paper 1

days is adopted. The addition of GGBS at 70% to replace ASTM results revealed that the combination of fly ash
the cement is a viable option in the AASHTO ring while it and GGBS to replace 70% of cement leads to effective
was limited to 50% in the ASTM Ring. This attributed to enhancement in the cracking resistance of the mix. In
the different degree of restraint provided by the rings. contrast to the results exhibited by the ASTM rings,
The results implies that for concrete structure subjected the AASHTO ring results suggest that for concrete
to low degree of restraint, the GGBS can be added by up structure subjected to low degree of restraint, the
to 70% to replace the cement without compromising the GGBS can be added by up to 70% to replace the cement
shrinkage crack resistance for the mix. The results for without compromising the shrinkage crack resistance
the mix with micro silica follow the same trend observed for the mix provided that moist curing for at least
in the ASTM Rings which suggest that the micro silica is 7days is adopted. The results obtained from ASTM and
recommended to be lower than 10%. The addition of 5% AASTO rings indicate that moist curing is essential for
micro silica significantly improved the crack resistance GGBS mixes. In addition, the moist curing should be
in the AASHTO Rings which propose that the optimum extended beyond 3 days for high proportion of GGBS
proportion of micro silica around 5% exist. (Proportion of GGBS ≥50%).
The net time to cracking for the SCC mixes in ASTM and 3. The results obtained from ASTM and AASHTO rings
AASHTO Rings were normalized to that for the control suggest that the micro silica should be added to SCC
mix as shown in Figure 10 and Figure 11. The figures show in a proportion less than 10% for all application under
different degree of improvement to the cracking resistance different degree of restraint.
of the control mix by using different supplementary
cementitious materials together with different curing 4. The results obtained for SCC mixes showed that
regimes. The figures also highlight those mixes that the proportion of GGBS and moist curing affected
tremendously improved the cracking resistance such as significantly the shrinkage potential. The mix contained
the mix contained 35% fly ash in combination with 7% 50% GGBS and subjected to seven days of moist curing
micro silica and tested by ASTM rings. The presented exhibited the least potential for shrinkage. Based
normalization may prove to be an important tool to device on the results; the moist curing for at least 7 days is
mixes best suited for on-site conditions. essential for the GGBS mixes and the addition of 50%
GGBS is adequate for high degree of restraint whereas
GGBS can be added by up to 70% in relatively low
Summary & Conclusions
degree of restraint provided that both are subjected to
This research investigated the cracking potential of 7 days of moist curing. Furthermore, the use of GGBS
various SCC mixes. The effect of type and proportion of in combination with fly ash provided excellent cracking
supplementary cementitious materials (SCMs) which
resistance provided that it is subjected to at least 3
are used to partially replace the cement, curing regimes
days of moist curing.
and degree of restraint on cracking potential of the SCC
mixes was particularly examined. The GGBS was added 5. The ASTM rings results for FA suggests an optimum
to replace the cement by 35%, 50%, 70%, a combined proportion of FA in the neighborhood of 35% for
GGBS-FA (35% - 35%) while the micro silica was replaced a structural element subjected to high degree of
the cement by 5% and 10%. Three curing regimes were restraint, provided that moist curing for at least three
adopted to assess the effect of curing regimes on the days is adopted. If higher fly ash proportions are used,
cracking potential of the SCC mixes. The restrained the results suggest that 7% micro silica must be added
shrinkage cracking was investigated using two types of to ensure the crack resistance for the mix. Unlike
standard rings; ASTM and AASHTO Rings. The ASTM ASTM rings, the results obtained from the AASHTO
Rings provide a degree of restraint in the range of 70 rings imply that for concrete structure with low degree
to 80% while the AASTO Rings provide a low degree of of restraint, fly ash can replace cement by up to 50%
restraint which is in the range of 50 to 60%. The results with significant improved crack resistance provided
reveal the following: that the mix is moist cured for at least 3 days. However;
1. The results obtained from the ASTM and AASHTO extending moist curing to 7 days further improves the
ring tests revealed that the cracking potential for the resistance to cracking.
SCC mixes was affected significantly by the type and
proportion of supplementary cementitious materials, Acknowledgments
curing regime and degree of restraint.
Support for this research was provided in part by the
2. The ASTM results also suggest an adequate proportion Sustainable Construction Materials and Structural
of GGBS in the neighborhood of 50% for the structures Systems (SCMASS) research group at the University
subjected to high degree of restraint as that induced of Sharjah. The support provided by TRIMIX Company,
in the ASTM rings provided that the mix accompanied Dubai, United Arab Emirates is also gratefully
with moist curing for at least 7 days. Furthermore, the acknowledged.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

144 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Shrinkage Cracking of Self Compacting Concrete (SCC) with Supplementary Cementitious Materials

References Concrete Used in Repair.” ACI Materials Journal 499 - 509.

1. Akkaya, Yilmaz, Chengsheng Ouyang, and Surendra P. Shah. 2007. 16. Hwang, Soo-Duck, Kamal H. Khayat, and Olivier Bonneau. 2006.
“Effect of supplementary cementitious materials on shrinkage and “Performance-Based Specifications of Self-Consolidating Concrete
crack development in concrete.” Cement & Concrete Composites Used in Structural Applications.” ACI Materials Journal 121-129.
117-123. 17. Moon, Heum Jae, and Jason Weiss. 2006. “Estimating residual
2. Altoubat, Salah A., and David A. Lange. 2001. “Creep, Shrinkage, stress in the restrained ring test under circumferential drying.”
and Cracking of Restrained Concrete at Early Age.” ACI Materials Cement & Concrete Composites 486-496.
Journal. 18. Moon, Jae Heum. 2006. SHRINKAGE, RESIDUAL STRES, AND
3. ASTM C 157. 1999. “Standard Test Method for Length Change of CRACKING IN HETEROGENEOUS MATERIALS. PhD Thesis, West
Hardened-Cement Mortar and Concrete.” Lafayette, Indiana: Purdue University.
4. ASTM C1581. 2004. “Standard Test Method for Determining Age at 19. Nakae, T., H. Matsushita, T. Makizumi, and H. Tsuruta. 1997. “The
Cracking and Induced Tensile Stress Characteristics of Mortar and Properties of Shrinkage in Concrete Containing High Volume of
Concrete under Restrained Shrinkage.” Granulated Blust-Furnace Slag.” Proceedings of the 52nd Annual
Conference of the Japan Society of Civil Engineers V. p.p. 840-841.
5. Attiogbe, Emmanuel K., Jason Weiss, and Heather T. See. n.d.
“A LOOK AT THE STRESS RATE VERSUS TIME OF CRACKING 20. Rols, Sébastien, Jean Ambroise, and Jean Péra. 1999. “Effects
RELATIONSHIP OBSERVED IN THE RESTRAINED RING TEST.” of different viscosity agents on the properties of self-leveling
concrete.” Cement and Concrete Research 261-266.
6. Bilodeau, A., N. Bouzoubaâ, C. Nkinamubanzi, and R. L. Chevrier.
2008. Properties of Fly Ash Concrete Mixtures Made with Materials 21. SCC European Project Group. 2005. “The European Guidelines for
from the United Arab Emirates. EcoSmart Foundation Inc. Self-Compacting Concrete.”
7. BS 8500-2. 2006. Specification for constituent materials and 22. See, Heather T., Emmanuel K. Attiogbe, and Matthew A. Miltenberger.
concrete. BRITISH STANDARD. 2003. “Shrinkage Cracking Characteristics of Concrete Using Ring
Specimens.” ACI Materials Journal 239-245.
8. Chikamatsu, R., N. Takeda, N. Minura, and S. Sogo. n.d. “Study
on lower Shrinkage of High Strength, High Fluidity Concrete.” 23. Shetty, M.S. 2004. Concrete Technology (Theory and Practice). New
Proceedings of the Japan Concrete Institute. Vol. 19, No. 1, p.p. Delhi: Chand & Company Ltd.
169-174. 24. SONEBI, M., P. J.M. BARTOS, W. ZHU, J. GIBBS, and A. TAMIMI.
9. Cocrete Society & BRE. 2005. Self - Compacting Concrete - A 2000. PROPERTIES OF HARDENED CONCRETE (Self Compacting
Review. Technical Report No.62, Cocrete Society. Concrete). University of Paisley, Scotland, United Kingdom:
Advanced Concrete Masonry Centre.
10. Edamatsu, Y., S. Nagaoka, and A. Yasumoto. 1997. “Method for
Estimating Autogenous Shrinkage of Self-Compacting Concrete.” 25. p. 179–82. Standard practice for estimating the crack tendency of
Proceedings of the 52nd Annual Conference of the Japan Society concrete. AASHTO Designation PP-34-89.
of Civil Engineers V. p.p. 838-839. 26. Taylor, H F W. 1997. Cement Chemistry. London, UK: Thomas Telford
11. EFNARC, European Federation of National Trade Associations. Services Ltd.
2002. Specifications and Guidelines for Self-Compacting Concrete. 27. Walker, Mike. 2002. Guide to the construction of reinforced concrete
UK. in the Arabian Peninsula. CIRIA & Concrete Society.
12. Grzybowski, Miroslaw, and Surendra P. Shah. 1990. “Shrinkage 28. Weiss, Jason, Brad Pease, Alexandra Radlinski, Gaurav Sant, Pietro
Cracking of Fiber Reinforced Concrete.” ACI Materials Journal Lura, Jae-Heum Moon, and Mette Geiker. 2005. “The Influence on
138-148. the Cracking of Concrete: Causes and Effects.” ACBM Cracking
13. Hossain, Akhter. 2003. A SSESSING RESIDUAL STRESS Workshop.
DEVELOPMENT AND STRESS RELAXATION IN RESTRAINED 29. Wiess, Jason, and Hossain Akhtar. 2006. “The role of specimen
CONCRETE RING SPECIMENS. PhD Thesis, Purdue University. geometry and boundry conditions on stress developement and
14. Hossain, Akhter B., and Jason Weiss. 2004. “Assessing residual cracking in the restrained ring test.” Cement and Concrete Research
stress development and stress relaxation in restrained concrete 189-199.
ring specimens.” Cement & Concrete Composites 531–540. 30. Yasumoto, A., Y. Edamatsu, M. Mizukoshi, and S. Nagaoka. 1998.
15. Hwang, Soo Duck, and Kamal H. Khayat. 2008. “Effect of Mixture “A Study on the Shrinkage Crack Resistance of Self-Compacting
Composition on Restrained Shrinkage Cracking of Self-Consolidating Concrete.” ACI Materials Journal 651-670.

Dr. Salah Altoubat

Ph.D., MASCE, MACI
Current Position: Department of Civil and Environmental Engineering University of Sharjah, UAE
www.sharjah.ac.ae
Dr. Salah Altoubat is an associate professor in the Department of Civil and Environmental Engineering,
and a coordinator of the Sustainable Materials and Structural Systems Research Group at the University
of Sharjah. He received his Ph.D in Civil Engineering from the University of Illinois at Urbana-Champaign,
USA in 2000. He has industrial and academic experience in North America and in the Middle East. His
expertise spans materials engineering, structural engineering and civil engineering infrastructure. Dr.
Altoubat is a recipient of the American Concrete Institute (ACI) Wason Award for the most meritorious
paper in the year 2003. Previously, he was a senior research associate at W. R. Grace, Cambridge, USA.
He has been involved in consulting work at the national and international levels. His current research
interest includes durability aspects and sustainability of concrete structures; structural applications of
fiber reinforced concrete; self compacting concrete; high performance concrete; early age behavior of
concrete; creep, shrinkage and cracking of concrete.

Organised by
India Chapter of American Concrete Institute 145
Session 2 A - Paper 2

Concrete cracking in two marine micro-climates

A. A. Torres-Acosta
P. Castro-Borges, J.A. Cabrera-Madrid and Universidad Marista de Querétaro, Marte No. 2, Col.
M.G. Balancán-Zapata Centro, 76000, Querétaro, Querétaro, México
Centro de Investigación y de Estudios Avanzados Permanentaffiliation: Instituto Mexicano del
del IPN Unidad Mérida, Antigua Carretera a Transporte, Carretera Querétaro-Galindo Km 12 + 000,
Progreso, Km 6, 97310, Mérida, Yucatán, México 76700 Sanfandila, Querétaro, México

Abstract Introduction
The mechanism of corrosion-induced concrete cover cracking Corrosion of reinforcing steel is one of the most important
is very important for durability forecasting of reinforced causes of reinforced concrete structures degradation
concrete structures. Accelerated corrosion tests have (Yuan et al, 2007; Angst et al, 2009; Fang et al, 2004). The
provided very useful information on corrosion propagation environment and climate change effects contribute also
and residual life stages, but tests and data from natural to the acceleration of this deterioration. The humidity,
environments are required to validate and corroborate rainfall, and different concentrations of sea chloride ions
previous findings. Concrete cylinders with different water/ in the environment enhance its deposition on the concrete
cement (w/c) ratios, exposed at two natural sites, 50 m and surface and the degradation rate significantly depends
100 m from the seashore (two micro-climates), in the Port of on the seashore distance (Castro-Borges and Mendoza-
Progreso, Yucatán, México, were used to measure corrosion Rangel, 2010)
parameters such as time-to corrosion initiation, apparent
corrosion rate, surface crack propagation, rebar radius loss Among the different damages caused by the reinforcing
to generate concrete cover cracking, and rebar pit depth due steel corrosion process, concrete surface cracking is the
to natural corrosion. A slight but clear correlation between most common, and it is produced by the steel corrosion
data of concrete cracking from both micro-climates was products formation at the steel/concrete interface and
found. The w/c ratio of concrete was the most important expansion (Meyer et al, 2006; Jaffer and Hansson, 2009;
parameter to in the durability performance of concrete in Torres-Acosta and Sagüés, 2000). Therefore, it has been
such tropical environment. Empirical correlations between presented elsewhere empirical correlations between
natural and accelerated corrosion tests were also obtained corrosion- induced cross-section loss of the reinforcing
to corroborate data available in the literature regarding steel and concrete cracking, to forecast service life
residual life of corroding reinforced concrete structures. reductions of reinforced concrete structures (Torres-
Acosta and Martinez, 2003; Ahmad, 2003; Weyers, 1998;
Keywords: concrete cracking, corrosion initiation, Kwon et al, 2009). The study of any possible relationship
corrosion propagation, micro-climates, marine between concrete cover cracking and reinforcing steel
environment, prevention. corrosion parameters will help to structural inspectors to
define possible repair strategies.
Nomenclature Description
w/c ratio Water/cement ratio Such relationships between structural distress and
CT Curing Time corrosion from accelerated corrosion tests performed in
∆W GRAV Gravimetric steel mass loss at end of experimental period laboratory have been widely presented (Torres-Acosta,
Leffec Rebar effective length showing corrosion during autopsy 2010; Vidal et al 2007; Poupard et a, 2004; El-Maaddawy
PITmax Maximum pit depth measured during autopsy
and Soudki, 2003; Chi et al, 2002; Li and Sagüés, 2001;
Ecorr Corrosion potential
Torres-Acosta and Sagüés, 2000; Meade, 1999; Simpson
icorr Apparent corrosion rate
et al 1991), while natural corrosion investigations are
i Acorr Accumulated corrosion rate
SCE Saturated Calomel Electrode
scant (Cabrera-Madrid et al, 2014; Torres-Acosta and
CSE Copper Sulfate Electrode
Castro-Borges, 2013; Torres-Acosta, 2010; Hernández-
XCRIT or X aver Critical or Average radius loss López et al, 2009) due to its complexity on data analysis,
r0 Nominal diameter of reinforcing bar which may include various parameters involved in this
Rp Polarization resistance corrosion phenomenon (Vidal, Castel and R. François,
Rs Ohmic drop 2007). In the case of structures exposed in a marine
ZRA Potentiostat/galvanostat environment, chloride ions are deposited on the surface
B Tafel constant of the concrete and penetrate until they reach the surface
CWmax Maximum Crack Width of the reinforcing steel, causing the breakdown of the
m1 or m2 Slope 1 or Slope 2 passive layer (Moreno et al, 2004; Caré et al, 2008). Once

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

146 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete cracking in two marine micro-climates

reinforcing bar corrosion starts, corrosion products entire length (Figure 1). A 50-mm activated titanium rod
starts to build up around the reinforcing bar, causing (ATR), characterized elsewhere was used as a reference
volumetric expansions that will result in concrete cover electrode for electrochemical measurements (Castro et
cracking (Zhao et al., 2012, Torres-Acosta and Sagüés, al, 1996, Pech-Canul et al, 1998). For this study, it was
2010; Montemor et al, 2003). calibrated vs a saturated calomel electrode (SCE). An
epoxy coating layer was applied to the cylinder’s ends
The literature has reported important investigations
to delimit concrete and steel exposure area (Figure 1).
related to the concrete cover cracking in reinforced
Concrete mixture proportions and physical/mechanical
concrete structural elements, but under controlled
properties of the hardened concrete have been published
conditions in the laboratory. The methods that have been
elsewhere (Castro, 2001; Castro et al 1999, Castro et al,
used in these investigations to accelerate the corrosion of
1996, Pech-Canul et al, 1998). A total of 120 cylinders were
the steel reinforcement include chloride contamination of
tested at two micro-climates (50 and 100 m) having two
the concrete during mixing water, or by spraying saline
cylinders from each combination of plain and reinforced
water on the surfaces of the specimens (Hernández-
concrete for a total of 60 plain concrete and60 reinforced
López et al., 2009; Torres-Acosta and Martínez-Madrid,
concrete cylinders,
2003; Li, 2000). Another accelerated method includes
not only chloride contamination but also accelerated One of the most important parameters for defining
electrochemical methods by applying anodic currents concrete quality is the w/c ratio (Haacha, Vasconcelos
or potentials, among others (Torres-Acosta et al., 2007; and Lourenço, 2011): A low w/c ratio reduces concrete
Torres-Acosta and Sagüés, 2000; Torres-Acosta, 1999; porosity, and concrete permeability to external agents
Mangat and Elgarf, 1999; Alonso et al., 1998; Andrade et (i.e., chlorides) decreases when porosity is low, improving
al, 1996; Rodríguez et al, 1997; Cabrera, 1996; Andrade et its service life (Li, 2001). To study the effect of w/c ratio on
al., 1996; Andrade et al., 1993). A serious concern is to find service life, concrete was made using five w/c ratios: 0.76,
possible relationships between accelerated and natural 0.70, 0.53, 0.50, and 0.46.
tests in terms of several variables like crack width, radius
Ratios as high as 0.7 and 0.76 are commonly used in the
loss due to corrosion of the reinforcing steel. This could
Yucatan Peninsula region, although public buildings in
help to predict cross section loss if crack width is known,
Mexico (i.e., schools, hospitals, etc.) are constructed
among other prediction possibilities.
using lower w/c ratios (0.53 or 0.5). The 0.46 ratio is
The objective of this work is to provide information and not commonly used in this region, but was included in
discuss empirical correlations between natural corrosion this study for comparative purposes since this value is
(two exposure sites: marine environments at 50 m and frequently used in the relevant literature. The effect of
100 m away from the seashore) and
accelerated corrosion tests.

Experimental Procedure
Materials and Sample Dimensions
ASTM Type I ordinary Portland
cement (OPC) and crushed limestone
aggregates were used for fabrication
of concrete cylinders (diameter ~
75 mm, height~150 mm) (Figure 1).
Plain concrete cylinders were used
to determine 28-day compressive
strength and chloride penetration
during the experimental period. Typical
no. 3 deformed steel rebar (~0.4 wt%
carbon content) was used to simulate
reinforcement. Rebar sections were
weighed before cylinder casting, and
section ends were protected with
epoxy and adhesive tape. This limits
the exposed area inside the concrete
specimen and isolates any rebar areas
directly exposed to the environment.
Rebar sections were positioned axially
within the cylinders and extended the Fig. 1: Specimen configuration

Organised by
India Chapter of American Concrete Institute 147
Session 2 A - Paper 2

CT on crack evolution was also investigated. Curing time At the end of the experimental period, all specimens
periods for concrete structures in Mexico range from were examined carefully for signs of deterioration i.e.,
7 days to 28 days, although 7 days is by far the most crack formation, corrosion stains, and delaminations).
common CT period. This variable was tested at 1, 3, and Crack morphology (i.e., position, length and width) was
7 days, with two cylinders tested per each combination of recorded. The steel rebar was then removed from the
w/c ratio and CT. concrete cylinder to determine final mass and calculate
gravimetric loss. Average radius loss from corrosion
Environmental Exposure (x AVER) was estimated with the equation:
Specimens had been exposed to the marine environment 10DW C ................................................(2)
x AVER =
of Progreso’s Port, located at the north coast of the t $ r $ ] $ L EFFEC
Yucatan Peninsula (21°18’ N, 89°39’ W), in Mexico, and
Where ∆WG is gravimetric mass loss (in grams); ρ =7.86
is characterized as having a tropical humid climate.
g/cm3; and LEFFEC is the actual rebar length affected by
Specimens were placed about 50 m and 100 m away from
corrosion (in cm). Depending on the recorded crack
the seashore, on top of a roof (approximately 4m above
width, estimated radius loss was defined as x AVER. Finally,
ground level). Both specimen types (plain and reinforced
pit depth was estimated using a magnifying lens and a
concrete cylinders) were oriented vertically. The carbon
caliper. Detailed descriptions of these parameters are
steel corrosivity category in this particular marine
available elsewhere (Torres-Acosta et al., 2007; Torres-
environment is higher than C5 (environment with t4 time
Acosta and Martínez-Madrid, 2003; Torres-Acosta and
of wetness, or time that relative humidity ≥ 85% and
Sagües, 2000).
temperature > 30°C and ≤ 60°C, between 2,500 h/y and
5,500 h/y; atmospheric chloride concentration 300 mg/ Because of the presence and non-uniformity of rebar
m2d to 1,500 mg/m2d) with an average annual temperature corrosion and corrosion of the rebar ends under the
of 26°C and average annual relative humidity of 79% protective tape, x AVER and L could only be defined
(Maldonado and Veleva, 1999; ISO 9223, 1992). Chloride approximately. Corrosion of the taped rebar ends was
was not added to the concrete mixtures prior to specimen adjusted for by assuming a modified total length (effective
casting to ensure that any chloride contamination was due length) of the anodic zone, LEFFEC. Therefore, LEFFEC was
to natural exposure during the experimental period. obtained by measuring the corroded rebar area with a
caliper and magnifying lens (x20) before cleaning off the
Corrosion Measurements corrosion products.
Rebar electrochemical properties were measured using All pit depths were measured and recorded using a caliper
an external conductive rubber counter electrode tightly mounted on a custom-made base to provide a baseline
attached to the concrete cylinders with aluminum clamps position during measurements. To locate pits, rebar
for each test (Figure 1). Electrochemical measurements sections were placed on a flat, leveled table, allowing
were taken with a commercially available potentiostat/ the caliper with its base to move along the rebar. Once
galvanostat/ZRA, and included corrosion potentials (ECORR) a pit was located, the caliper was used to measure its
and apparent corrosion rates (iCORR) determined using the deepest point. Caliper position was then changed to one
polarization resistance (Rp) technique (Andrade et al, 1990; end of the rebar (that closest to the measured pit) and the
Andrade et al, 1984). The Rp was measured by applying10 rebar diameter was recorded to create a pre-corrosion
mV in the cathodic direction from the ECORR at a 0.06 mV/s baseline. Special care was taken if a pit was near or
scan rate. Ohmic drop (Rs) as a result of the high concrete on top of a rebar deformation. In these cases, pit depth
resistivity was measured with acommercially available was estimated using a non-corroded deformation as the
resistance meter. The adjusted Rp estimates were then baseline and following the procedure explained above.
converted to apparent corrosion rate (iCORR) values using Finally, pit depth was estimated as the difference between
the equation 1 (Andrade and González, 1978): the baseline measurement and pit depth measurement.
B
i corr = R - R ..............................................................(1)
Results and Discussion
p s

Where B = 0.026 when the rebar exhibited corrosion

activity (i.e., when rebar potentials were more negative Reinforcing steel corrosion potential for cylinders
than –200 mV vs CSE); or B = 0.052 (Andrade et al, 1990; exposed at 50 m from seashore
Andrade et al, 1984) when the rebar exhibited passive Figure 2 shows reinforcing steel ECORR (vs CSE) results for
behavior (i.e., when rebar potentials were more positive cylinders exposed at 50 m from seashore, five different
than –200 mV vs CSE); and Rp or RS are in Ω•cm2 (calculated w/c ratios, and 7-day CT. Every data point is an average
via multiplication by the nominal rebar surface area ~33 value from two cylinders with same CT and w/c ratio.
cm2 in contact with concrete). These measurements
ECORR values a quite clear performance: all w/c ratio
were taken every two to three months during the entire
concretes showed values more positive than - 200
experimental period.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

148 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete cracking in two marine micro-climates

Fig. 2: Average Ecorr vs CSE as a function of time of specimens Fig. 4: Average apparent iCORR values as a function of exposure
at 50 m from the seashore and 7-day CT. time, for specimens at 50 m from the seashore, and 7-day CT.

mV (vs. CSE), up to 6 months of exposure, where ECORR values presented fluctuations, influenced mainly by to
values shifted to more negative ones, which means that the environmental conditions (i.e. pluvial precipitations,
reinforcing steel corrosion activation was reached (ASTM relative humidity, and temperature.
C876). With exception of 0.46 w/c ratio concrete, the
other concrete types maintained active after 6 months of Reinforcing steel apparent corrosion rate for cylinders
exposure and during the entire experimental period. exposed at 50 m from seashore
The positive values reached during the last four months of Figure 4 shows average (of two specimens) apparent iCORR
exposure could be due to the presence of cracks appeared values as a function of exposure time for the specimens
at the cylinders’ surface. From the 30 specimens exposed at the 50 m from seashore exposure site, for all w/c
in this micro-climate, 70% (21 cylinders) presented surface ratio concretes and 7-day CT. As observed, all w/c ratio
cracks after approximately 48 months of exposure. concretes showed apparent iCORR higher values than
0.5 μA/cm2, indicative of steel active corrosion (DURAR
Reinforcing steel corrosion potential for cylinders Network, 1997). This performance corroborated ECORR
exposed at 100 m from seashore tendency in this microclimate.
Figure 3 shows ECORR results obtained with 7-day CT
specimens and exposed at 100 m away from seashore. It Reinforcing steel apparent corrosion rate for cylinders
is clear to observe that ECORR values were passive a longer exposed at 100 m from seashore
period of time (20 months) than specimens exposed to the 50 Figure 5 shows average (of two specimens) apparent iCORR
m site (ASTM 876). However, 0.76 and 0.70 w/c ratio concrete values as a function of exposure time for the specimens
cylinders shifted to an active potential zone values after 24 at the 100 m from seashore exposure site, for all w/c ratio
months of exposure, which indicated probably corrosion concretes and 7-day CT. Similar trend is observed between
activation of the reinforcing steel. These high w/c ratio ECORR (Figure 3) and iCORR (Figure 5): the first 20 months of
cylinders were retrieved from the 100 m exposure site at 76 exposure, the five w/c ratio concretes presented below
and 85 months, respectively, due to the presence of surface threshold values, 0.1 a 0.5 µA/cm2, (DURAR Network,
cracks in those cylinders. Next cylindrical specimens 1997). It was until 24 months of exposure, 0.70 and 0.76
showing high risk corrosion potential values, ECORR< -200 w/c ratio concretes showed higher apparent iCORR values
mV (vs CSE) where the 0.53 w/c ratio specimens, retrieved (explanations of specimen retrieving after surface
after 166 months of exposure, as observed in Figure 3. The cracking appearance is explained above). For 0.53 w/c
other w/c ratio specimens, 0.50 and 0.46, did not showed ratio concrete, depassivation occurred after 50 months of
active ECORR values up to 192 months of exposure. exposure, and the exposure time lasted up to 152 months.
At the moment, apparent iCORR values for 0.50 and 0.46
For all w/c ratios concretes and micro-climates (50 and w/c ratio concretes, are still not showing active corrosion
100 m from seashore) used in this investigation, ECORR (DURAR Network, 1997).

Fig. 3: Average ECORR vs CSE as a function of exposure time,for Fig. 5: Average apparent iCORR values as a function of exposure
specimens at 100 m from the seashore, and 7-day CT. time, for specimens at 100 m from the seashore, and 7-day CT.

Organised by
India Chapter of American Concrete Institute 149
Session 2 A - Paper 2

Information described at this point was similar and Both figures include arrows representing the time of
reproducible for the other two CT’s (1 and 3 days). On corrosion activation for each w/c ratio concrete. It is
the other hand, electrochemical results presented well clear the performance of iACORR estimates as a function
defined trends: higher w/c ratio concretes presented higher of concrete types and presented similar trend than the
apparent iCORR values. But additional analysis is required proposed by Tuutti’s two stage durability model (1982).
with the data obtained from field measurements to obtain At the beginning, iACORR vs t data presented a small slope
other important parameters as, for example, the corrosion (m1) with time and when the reinforcing steel begin to
initiation times as a function of w/c ratio and distance from corrode, the slope of iACORR vs t switched to a higher value
the seashore, and the corrosion degradation parameters (m2) indicative of the end of the corrosion initiation stage
like for example, the steel reinforcement corrosion (T1) in Tuutti’s durability model and the starting point of the
penetration and pitting distribution, and concrete surface corrosion propagation stage (T2).
cracking distress, as discussed as follows.
Another important trend observed with the experimental
results is the differences between T1 stage for 50 m and
100 m exposure sites. For 50 m exposure site (the closest
from the seashore), T1 estimates were always smaller than
the ones obtained at the 100 m exposure site for all w/c
ratio concretes. For example, concretes with higher w/c
ratio (0.76 and 0.70) reached typical T1 estimates between
2.5 and 4 months at the 50 m exposure site, compared
with ~ 25 months for the 100 m exposure site. On the other
hand, lower w/c ratio concretes (0.53, 0.50, and 0.43)
reached T1 estimates between 15 and 24 months at the
50 m exposure site; in comparison with 50, 90, and 130
months, for w/c ratios of 0.53, 0.50, and 0.43, respectively,
for the 100 m exposure site.
Fig. 6: Typical i ACORR estimates as a function of exposure time,
At the moment, the data is still in the process of scrutinized
for specimens at 50 m from the seashore, and 7-day CT.
analysis to determine the slope values for each one of the
Accumulated apparent corrosion rate, iACORR two exposure sites, three CT’s, and five w/c ratios, and will
be presented in the near future.
From apparent icorr as a function of time graphs,it is
possible to estimate Tuutti’s service life stages, initiation Experimental determination of x AVER and PITMAX
(T1) and propagation (T2) (Tuutti, 1982),for each w/c ratio
Average corrosion penetration depth estimates, x AVER,
concrete. However, this is difficult to visualize with the
from Equation (2) using the gravimetric mass loss, are
way the results are showed. Thus, additional iCORR data
listed in Tables 1 and 2 for 50 m and 100 m exposure sites,
processing wasAconsidered to compare each w/c ratio
respectively. As observed from these tables, data from
concrete types for the three CT’s, which includes the
0.46 and 0.5 w/c ratios presented the lowest x AVER values
accumulated iCORR estimates as a function of time. These
(median x AVER ~ 0.2 mm for 50 m exposure site, and x AVER
i ACORR estimates were calculated as the integral of the
~ 0.32 mm for 100 m exposure site), following data from
i ACORR vs time results (∫iCORR·dt). Results of these i ACORR
0.53 and 0.7 w/c ratios (median x AVER ~ 0.25 mm for 50 m
estimates are shown in Figures 6 and 7 for each w/c ratio
exposure site, and x AVER ~ 0.36 mm for 100 m exposure
and 7-day TC.
site). For 0.76 w/c ratio, x AVER was ~ 0.35 mm, which was
the highest of all concrete mixtures tested at the 50 m
exposure site, but for the 100 m exposure site, no cracks
were observed up to date. Again, no correlation was
observed between CT and x AVER for both exposure sites.
Maximum pit depths (PITMAX) obtained in this investigation
are listed in Tables 1 and 2 for both exposure sites 50
m and 100 m, respectively. As observed from these
results no differences were observed between PITMAX
results from the two exposure sites: Data from the 50
m exposure site ranged between 0.81 mm and 4.45 mm
(average 2.29 mm), and data from the 100 m exposure
site ranged between 0.99 and 3.31 mm (average 1.68 mm),
but there are still some cylinders that are been exposed
Fig. 7: Typical iACORR estimates as a function of exposure time, at the present date with no corrosion distress observed
for specimens at 100 m from the seashore, and 7-day CT. (concrete cracking or corrosion stains).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

150 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete cracking in two marine micro-climates

Table 1
Experimental result set of specimens exposed at 50 m from the seashore (Torres-Acosta and Castro-Borges, 2013).

Maximum
Curing Time, w/c ∆ WGrav LEFFEC Average radius PITMAX
Specimen crack width,
CT (days) Ratio (g) (mm) loss x AVER(mm) (mm)
CW MAX (mm)

505 1 0.46 7.67 156.6 0.21 0.80 2.16

506 1 0.46 9.06 148.7 0.26 1.50 1.42

43 3 0.46 5.76 160.6 0.15 0.30 2.69

434 3 0.46 6.68 129.7 0.22 2.00 2.08

469 7 0.46 2.87 35.9 0.34 - 1.80

470 7 0.46 2.67 53.5 0.21 - 0.81

361 1 0.50 8.89 134.1 0.28 1.00 2.92

362 1 0.50 4.37 129.3 0.14 1.00 1.47

397 3 0.50 15.2 139.0 0.47 1.50 2.29

398 3 0.50 8.38 128.9 0.28 4.00 1.60

325 7 0.50 6.14 140.3 0.19 1.00 1.45

326 7 0.50 11.88 168.9 0.30 1.50 2.54

289 1 0.53 9.02 144.0 0.27 - 2.16

290 1 0.53 6.37 127.8 0.21 0.60 1.98

253 3 0.53 5.96 132.4 0.19 0.08 2.08

254 3 0.53 11.79 196.0 0.26 0.60 1.78

217 7 0.53 6.41 164.4 0.17 - 2.54

218 7 0.53 - - - - -

181 1 0.70 - - - 1.50 -

182 1 0.70 10.86 180.0 0.26 - 1.65

145 3 0.70 11.36 176.3 0.27 - 2.46

146 3 0.70 10.54 172.7 0.26 - 3.68

109 7 0.70 13.76 208.0 0.28 0.08 2.90

110 7 0.70 21.3 197.0 0.46 1.00 2.64

73 1 0.76 8.86 165.7 0.23 - 2.49

74 1 0.76 16.41 201.0 0.35 0.80 2.79

1 3 0.76 17.21 201.7 0.36 1.00 4.45

2 3 0.76 11.43 149.7 0.33 0.08 2.92

37 7 0.76 18.76 185.0 0.43 0.80 1.96

38 7 0.76 19.76 204.0 0.41 0.08 2.41

Organised by
India Chapter of American Concrete Institute 151
Session 2 A - Paper 2

Table 2
Experimental result set of specimens exposed at 100 m from the seashore

Maximum
Curing Time, w/c ∆ WGrav LEFFEC Average radius PITMAX
Specimen crack width,
CT (days) Ratio (g) (mm) loss x AVER(mm) (mm)
CW MAX (mm)

510 1 0.46 7.17 98.00 0.31 0.80 1.41

509 1 0.46 14.12 183.00 0.33 0.80 2.22

293 1 0.53 9.75 135.00 0.31 1.00 1.67

294 1 0.53 9.03 145.00 0.27 1.00 1.56

234 7 0.53 14.87 141.00 0.45 3.00 1.59

233 7 0.53 5.98 133.00 0.19 0.40 0.99

185 1 0.70 7.26 123.00 0.25 1.00 1.07

186 1 0.70 12.24 169.00 0.31 0.80 1.89

149 3 0.70 15.14 137.00 0.47 2.00 2.51

150 3 0.70 17.10 141.00 0.52 0.80 1.88

114 7 0.70 13.87 125.00 0.47 2.00 3.31

Concrete surface crack survey

The crack monitoring procedure has been presented
elsewhere (Torres-Acosta et al., 2007; Torres- Acosta and
Martínez-Madrid, 2003;Torres-Acosta 1999). From all the
widths measured along the cracks, only the maximum value,
CWMAX, measured in all cracked specimens (50 m and 100 m
exposure sites) were recorded in Tables 1 and 2. As observed
form results presented in these tables, most of the concrete
cylinders at the 50 m exposure site presented surface
cracks (21 out of 30 cylinders). On the other hand, most of
the concrete cylinders (19 out of 30 cylinders) exposed at the
100 m site are still showing no corrosion distress (concrete
cracking of corrosion stains), and electrochemical and crack
survey monitoring are still in process.

Corrosion-induced crack propagation in concrete

As well known, the cross section of the reinforcement
gradually decreases in function of corrosion advance
whose corrosion products produce a higher volume than
the original steel (Jaffer and Hansson, 2009; Andrade
et al., 1993). This increment in volume is responsible of
concrete cracking because tensile strength is lower than
stresses produced by the volumetric expansion of oxides.
At the beginning, cracks generate in the interface concrete/
steel and then they propagate radially until reaching
concrete surface. Several studies to correlate cross
section loss with crack width, can be found in old and recent
literature evidence, being most of them from accelerated Fig. 8: CWMAX vs. x AVER /r0 composite plot for accelerated corrosion
corrosion tests, which include anodic corrosion current tests (above) and natural corrosion tests (below).
applicationand/or the addition of certain quantities of
chloride to concrete during mixing (Hernández-López, and cross section loss in naturally exposed environments.
et al, 2009; Torres-Acosta and Martínez-Madrid, 2003; This works intend to correlate some data obtained from
Andrade et al, 1996; Cabrera, 1996; Andrade et al, 1993). natural exposure during several years with the available
There have been some concerns about crack generation data using accelerated corrosion tests.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

152 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete cracking in two marine micro-climates

Figure 8 was plotted to observe a general trend with

the data from accelerated and natural corrosion tests
separately. It shows a slight but clear correlation
between crack width and loss of cross section for data
from accelerated tests, several authors, and natural
tests, this research. The empiric relation shown in Figure
8 is similar to that from other works, being y=21.8x with
R2=0.4596 (Torres-Acosta and Martínez-Madrid, 2003;
Andrade et al, 1993).
Although most of the data fell inside the cloud of data from
other authors, it is also clear that natural environment
measurements show similar tendency and larger scatter
between the data than the accelerated corrosion data. A Fig. 10: PITMAX vs. x AVER empirical correlation for accelerated and
slight difference between the two tendencies was found: natural corrosion tests
the slope for natural corrosion data (18.41) is higher than
for accelerated corrosion (15.34). It is also observed that 10%, and the reinforcement cross section loss becomes
generation of a crack with accelerated tests requires a more complex process under such circumstances. As
less corrosion products than for the natural case. This a consequence, when x AVE /r0 reaches values of 0.1, the
could be explained by the fast accumulation of corrosion crack width might be more influenced by the distribution
products on the pores during the accelerated tests that of tensile and compressive stresses than the corrosion
generate tensile stresses between reinforcement and product stress itself. Previous results have considered
concrete inducing cracks more easily. It is also reported that a 10% loss of section produce about a 50% of carrying
that acidification of the reinforcement surface because of capacity of a concrete element (Torres Acosta and
corrosion can produce a loss of bond and then generate Martínez Madrid, 2003).
early cracks (Hernández-López et al., 2009).
Corrosion-induced pit formation at the corroding steel
It is also observed from natural test data in Figure 8, more
reinforcement
corrosion products are needed for crack appearance. This
is due to the slower corrosion process and longer periods After determining mass loss, each rebar was further
of time that allow a migration of corrosion products from inspected for evidence of pitting. All pit depths were
the rebar to the concrete surface through the pores (Figure measured and recorded using a caliper, as explained in
9). Therefore, more corrosion products/cross section loss the experimental procedure section. Tables 1 and 2 list
is needed to produce surface cracks. the values for the two exposure sites, 50 m and 100 m.
Results from these tables and previous investigations
Figure 8 also shows an apparent limit value for x AVER / were plotted as a correlation between x AVER and maximum
r0 which is 0.1. This may mean that loss of section does pit depth, PITMAX (see Figure 10).
not increase as crack width does. This may be due to a
higher quantity of cracks close to the surface that might With the information presented in this section, it is
be generated when cross section loss is in the order of possible to obtain an empirical relation between x AVER and
PITMAX . This relation is presented as the discontinuous
line in Figure 10, which represents the equation PITMAX
= 9.154•x AVER with a correlation coefficient R2 = 0.763.
This empirical relation shows that PITMAX values is
approximately ten times the value of x AVER, similar to
findings reported in previous references (Alonso et al,
1998) with PITMAX ~ α • x AVER where α ~10, where also the
accelerated corrosion tests were included in Figure 10
and the obtained empirical correlation.

Conclusions
Based on the information obtained from this investigation,
the following conclusions were obtained:
Electrochemical monitoring of naturally exposed
reinforced concrete cylinders at two different exposure
Fig. 9: Cylinder #293 photograph after crack survey and before sites determine important differences between corrosivity
reinforcement was retrieved (1-day CT, 053 w/c ratio, and 28- parameters, even though the two exposure sites were
day f’C=25MPa) located at the same port city, Progreso, Yucátan, México:

Organised by
India Chapter of American Concrete Institute 153
Session 2 A - Paper 2

the closest is the exposure site to the shore, the faster is 5. Andrade C., Alonso M.C. and González J.A., 1990. An initial effort
to use the corrosion rate measurements for estimating rebar
the corrosion activation of the reinforcing steel and the
durability. In Corrosion rates of steel in concrete, Eds. N.S. Berke,
concrete cracking due to the corrosion products formed V.Chaker, D. Whiting, ASTM STP 1065 (West Conshohocken,
at the steel concrete interface. PA:ASTM International):29-37

Between the parameters chosen in this investigation 6. Andrade C., Castelo V., Alonso C. and González J. A., 1984. The
determination of the corrosion rate of steel embedded in concrete
regarding concrete quality, the water to cement ratio (w/c) by the polarization resistance and AC impedance methods. In
is the most important of all, giving better performance Corrosion effect of stray currents and techniques for evaluating
in tropical marine environment of Progreso, Yucátan, corrosion of rebars in concrete, Ed. V. Chaker, ASTM STP 906 (West
México, concretes with w/c ratios less than 0.50: corrosion Conshohocken, PA: ASTM International):43-63
initiation stage, at both exposure sites, increased as w/c 7. Andrade C. and González J. A., 1978. Quantitative measurement of
ratios decrease. Curing times did not affect the durability corrosion rate of reinforcing steels embedded in concrete using
polarization resistance measurements. Werkstoffe und Korrosion,
performance of such concretes. 29(8):515- 519
The crack generation obtained in this investigation under 8. Angst U., Elsener B., Larsen C. K. and Vennesland O., 2009. Critical
natural environment exposure was produced at smaller chloride content in reinforced concrete. Cement and Concrete
corrosion rates (0.9 - 12.0 μA/cm2) than those reported in Research, 39(12):1122 -1138
the literature with an accelerated corrosion technique (10 9. ASTM-C-876-91, 1991. Standard test method for half-cell potential
of uncoated reinforcing steel in concrete.ASTM, Philadelphia, USA.
– 3000 μA/cm2).
10. Broek D., 1986. Elementary engineering fracture mechanics. 4th
This investigation has experimentally established the ed. Martinus Nijhoff Publishers.
correlation between visible corrosion degradation 11. Cabrera J. G., 1996. Deterioration of concrete due to reinforcement
(appearance of crack at the concrete surface, CWMAX) steel corrosion. Cement and Concrete Composites, 18(1):47–56.
and metal loss (x AVER) using naturally exposed reinforced 12. Cabrera-Madrid J. A., Balancán-Zapata M, Torres-Acosta A. A.,
concrete cylinders: CWMAX~ 18.4 x AVER /r0 for natural Castro-Borges P., 2014. Effect of tropical marine microclimates
corrosion. on depassivation and corrosion-induced cracking of reinforced
concrete. International Journal of Electrochemical Science,
Also an empirical correlation was obtained from metal 9(12):8211-8225
loss and maximum pit depths: PITMAX~ 9.154 x AVER, 13. Caré S., Nguyen Q. T., L’Hostis V. and Berthaud Y., 2008. Mechanical
corroborating empirical calculations form previous properties of the rust layer induced by impressed current method in
investigations using accelerated corrosion tests. reinforced mortar. Cement and Concrete Research, 38(8):1079-1091
14. Castro P., 2001. The chloride threshold for corrosion onset of
Small differences between crack opening and extension reinforced concrete in two tropical marine micro-climates of
results in accelerated and natural corrosion tests were Yucatán, México. In Proc. 3rd International Conference: Concrete
observed, but with other corrosion parameters such as Under Severe Conditions, Eds. N. Banthia, K. Sakai, O. Gjørv
mass loss and pit depth, trends were similar, supporting (London, U.K.:CRC Press, Taylor & Francis Group:151-158
the use of such constant current accelerated method for 15. Castro-Borges P., Balancán-Zapata M and López-González A. 2013.
remaining life forecasting. Analysis of tools to evaluate chloride threshold for corrosion onset
of reinforced concrete in tropical marine environment of Yucatán,
México. Journal of Chemistry, http//dx.doi.org/10.1155/2013/208619
Acknowledgements 16. Castro P., Sagüés A.A., Moreno E.I., Maldonado L. A, Genescá
J., 1996. Characterization of activated titanium solid reference
The authors acknowledge the partial support of their electrodes for corrosion testing of steel in concrete. Corrosion,
Institutions and CONACyT. One of the authors, J. A. 52(8):609-617
Cabrera-Madrid performed his PhD studies at CINVESTAV 17. Castro P., Troconis O. and Pazini E., 1999.Chloride penetration
and acknowledges its support as well as that from profiles in marine environment. In 2nd CANMET/ACI International
CONACYT through his PhD grant. Conference. Eds. V.M. Malhotra, P. Helene, L.R. Prudencio, Jr.,
D.C.C. Dal Molin, ACI Special Publication SP 186, Farmington, MI:
American Concrete Institute:371.
References
18. Castro-Borges P. and Mendoza-Rangel J. M., 2010.The influence
1. Ahmad. S., 2003. Reinforcement corrosion in concrete structures,
of climate change on concrete durability in the Yucatan Peninsula.
its monitoring and service life prediction–– A review. Cement and
Corrosion Engineering Science and Technology, 45(1):61-69
Concrete Composites, 25(4-5):459-471.
19. Chi J.M., Huang R. and Yang C., 2002 Effects of Carbonation on
2. Alonso C. et al, (1998). Factors controlling cracking of concrete
mechanical properties and durability of concrete using accelerated
affected by reinforcement corrosion. Materials and Structures,
testing method. Journal of Marine Science and Technology, 10(2):14-
31(Aug.-Sept.):435 – 441.
20.
3. Andrade, C., Alonso Rodríguez J., and García M.,1996. Cover cracking
20. DURAR Thematic Network XV.B. 2000. Manual for inspecting,
and amount of rebar corrosion: Importance of the current applied
evaluating, and diagnosing corrosion in reinforced concrete
accelerated test. In Concrete repair rehabilitation and protection,
structures, 1997. CYTED DURAR Network (Durability of rebars).
R.K Dhir and M. R. Jones, E & FN Spon, London, U.K: 263-273.
Maracaibo, Venezuela.
4. Andrade C., Alonso C and Molina F. J. 1993.Cover cracking as
21. El-Maaddawy T. A. and Soudki K.A., 2003. Effectiveness of impressed
function of bar corrosion. Part I: experimental tests. Materials
current technique to simulate corrosion of steel reinforcement in
and Structures, 26(8):453-464
concrete. Journal of Material in Civil Engineering, 15(1):41-47.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

154 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Concrete cracking in two marine micro-climates

22. Fang C., Lundgren K., Chen L. and Zhu C., 2004. Corrosion influence 38. Rodríguez J., Ortega L. M. and Casal J., 1997.Load carrying capacity
on bond in reinforced concrete. Cement and Concrete Research, of concrete structures with corroded reinforcement. Construction
34(11):2159–2167 and Building Materials, 11(4):239–248.
23. Haacha V.G., Vasconcelos G., Lourenço P. B., 2011. Influence of 39. Simpson C. H., Ray C. J. and Skerry B. S., 1991.Accelerated
aggregates grading and water/cement ratio in workability and corrosion testing of industrial maintenance paints using a cyclic
hardened properties of mortars. Construction and Building corrosion weathering method. Journal of Protective Coatings &
Materials, 25(6):2980-2987 Linings.8(5):28-36
24. Hernández-López Y., et al. 2009. Relationship between loss of 40. Tachibana Y., Maeda K., Kajikawa Y and Kawuamura M.,1990.
reinforcement section and width of cracks in concrete beams (in Mechanical behavior of RC beams damaged by corrosion of
Spanish). In CONPAT, Mexico 2009. reinforcement. In Corrosion of Reinforcement in Concrete. Eds.
C.L. Page, K.W.J. Treadaway, and P.B. Bamforth: 178-187.
25. ISO 9223, 1992. Corrosion of Metals and Alloys. Corrosivity of
Atmospheres. Classification, Geneva, Switzerland, International 41. Torres-Acosta, A.A., 2010. Accelerated vs. natural corrosion
Standardization Organization. experimental results for remaining life stage forecasting. In
Proceedings of the 6th International Conference on Concrete Under
26. Jaffer S.J. and Hansson C. M., 2009. Chloride-induced corrosion
Severe Conditions (CONSEC’10), Merida, Yucatan, Mexico, Eds. P
products of steel in cracked-concrete subjected to different loading
Castro-Borges, E.I. Moreno, K. Sakai, O.E. Gjorv, and N. Banthia,
conditions. Cement and Concrete Research, 39(3):116-125.
Taylor & Francis Group, London, UK, 1:35-42.
27. Kwon S. J., Jin Na U., Park S. S. and Jung S. H., 2009. Service life
42. Torres-Acosta, A.A., 1999. Cracking Induced by Localized Corrosion
prediction of concrete wharves with early-aged crack: Probabilistic
of Reinforcement in Chloride Contaminated Concrete, Ph.D.
approach for chloride diffusion. Structural Safety, 31(1):75-83
Dissertation, Department of Civil and Environmental Engineering,
28. Li. C. Q., 2000. Corrosion initiation of reinforcing steel in concrete University of South Florida, Tampa FL, 33620-5350, USA.
under natural salt spray and service loading - Results and analysis.
43. Torres-Acosta A. A. and Castro-Borges P., 2013. Corrosion -
ACI Materials Journal, 97(6):690-697
induced cracking of concrete elements exposed to a natural marine
29. Li L. and Sagüés A.A., 2001. Chloride corrosion threshold of environment for five years. Corrosion, 9(11):1123-1131
reinforcing steel in alkaline solutions-Open- circuit immersion
44. Torres-Acosta A.A., Navarro-Gutiérrez S. and Terán-Guillen J.,
tests. Corrosion Science, 57(1):19–28.
2007. Residual flexure capacity of corroded reinforced concrete
30. Maldonado L. and Veleva L., 1999. Corrosivity category maps of beams. Engineering Structures, 29:1145-1152.
a humid tropical atmosphere: The Yucatán Peninsula, México.
45. Torres-Acosta A. A. and Martínez-Madrid M., 2003. Residual life of
Materials and Corrosion, 50(5):261-266
corroding reinforced concrete structures in marine environment.
31. Mangat P. S. and Elgarf S., 1999. Bond characteristics of corroding Journal of Materials in Civil Engineering, 15 (4):344–353.
reinforcement in concrete beams. Materials and Structures,
46. Torres-Acosta A.A. and Sagüés A. A., 2000. Concrete cover cracking
32(216):89-97
with localized corrosion of the reinforcing steel. In Fifth CANMET/
32. Meade C. L., 1999. Accelerated corrosion testing. Metal finishing, ACI Int. Conference on durability of concrete, SP-192-36, Ed. V.M.
Supplement 1, 97(5):526-531 Malhotra (Farmington Hills, MI: American Concrete Institute):591-611
33. Meyer M. D., Watson L. S., Walton C. M. and Skinner R.E., 2006. 47. Torres-Acosta A. A., Castro-Borges P. and Sagües A. A., 1999. Effect
Control of cracking in concrete. State of the art. Transportation of corrosion rate in the cracking process of concrete. In Proceeding
Research Board of the National Academies. No. E-C107. Washington XIV National Congress of the Mexican Electrochemical Society,
DC. Mérida, México (in Spanish): 24–28
34. Montemor M.F., Simoes A. M. P. and Ferreira M. G. S., 2003. Chloride- 48. Tuutti, K., 1982. Corrosion of steel in concrete. Swedish Cement and
induced corrosion on reinforcing steel: from the fundamentals Concrete Research Institute, Estocolmo, Dissertation (Monograph).
to the monitoring techniques. Cement and Concrete Composites,
49. Vidal T., Castel A. and François R., 2007. Corrosion process and
25(4-5):491- 502.
structural performance of a 17 year old reinforced concrete beam
35. Moreno M., MorrisW., Alvarez M. G. and DuffóG. S., 2004.Corrosion stored in chloride environment. Cement and Concrete Research,
of reinforcing steel in simulated concrete pore solutions. Effect of 37(11):1551- 1561.
carbonation and chloride content. Corrosion Science, 46(11):2681-
50. Weyers. R. E. 1998. Service life model for concrete structure in
2699
chloride laden environments. ACI Materials Journal, 95(4):445 - 453
36. Pech-Canul M.A., Sagüés A. A. and Castro P., 1998.Influence
51. Yuan Y., Ji Y.and Shah S. P., 2007. Comparison of two accelerated
of counter electrode positioning on the solution resistance in
corrosion techniques for concrete structures. ACI Structural
impedance measurements of reinforced concrete. Corrosion
Journal, 104(3):344 - 347.
54(8):663-667
52. Zhao Y., Yu J., Hu B. and Jin W., 2012. Crack shape and rust
37. Poupard O., Aït-MokhtarA. and Dumargue P., 2004. Corrosion
distribution in corrosion-induced cracking concrete. Corrosion
by chlorides in reinforced concrete: Determination of chloride
Science, 55(1):385 - 393
concentration threshold by impedance spectroscopy. Cement and
Concrete Research. 34(6):991–1000

Organised by
India Chapter of American Concrete Institute 155
Session 2 A - Paper 2

Dr. Pedro Castro-Borges

Dr. Pedro Castro-Borges works at CINVESTAV-IPN Unidad Merida, Yucatán, México since 1986 where he
is researcher. He is Civil Engineer and M. Eng. from the Yucatán Autonomous University, he has a Ph.D.
in Engineering from the National Autonomous University of México and a post-doctorate from the Torroja
Institute of Construction Sciences from Madrid, Spain. His main interest is the durability of reinforced
concrete. He is Editor in Chief of ALCONPAT Journal.

Dr. A. A. Torres-Acosta
Dr. A. A. Torres-Acosta is a Research professor at Mexican Institute of Transportation (IMT), Querétaro,
México. He is a Civil Engineer from the Yucatán Autonomous University and he has MSc and Ph.D. degrees
from the University of South Florida at Tampa. His main interest is the rehabilitation of corrosion-damaged
infrastructure and the effects of aggressive marine environments on durability of concrete infrastructure.

Dr. J. Alejandro Cabrera Madrid

Dr. J. Alejandro Cabrera Madrid works at Civil Engineering Campus of the Chiapas Autonomous University
since January 2012 where he is a professor and researcher. He is a Civil Engineer from UNACH, Tuxtla
Gutierrez, Chiapas (2007), he has a Ph.D. in Physicochemical Sciences from the CINVESTAV-IPN Unidad
Merida (2014). His main interest is the new cement materials, particularly focusing on waste material
valorization and durability.

M. Balancán-Zapata
M. Balancán-Zapata is a Civil Engineer from the Autonomous University of Yucatán and she is a research
assistant at Center of Research and Advanced Studies (CINVESTAV-Mérida), Yucatán, México. She is
involved in laboratory and field tests on concrete and concrete structures.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

156 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

Improving segmented Functionally Graded Concrete concept by using

SCC technology

Olga Río, Khanh Nguyen Xabier Turrillas, Judith Oró-Solé, Ana E. Carrillo
E. Torroja Institute for Construction Sciences, Institute of Materials Science of Barcelona, CSIC,
CSIC, Madrid, Spain Barcelona, Spain

Abstract material, maintains a position- dependent microstructure

or material composition, which may result in gradually or
Segmented Functionally Graded Self-consolidating
stepwise variations of material properties with position
Cementitious Composite (FGSCC) is a new trend of
(Miyamoto et al., 1999). Already, in 1972, the usefulness
Functionally Graded Material (FGM) enabling the
of the gradient idea was documented (Bever and Duwez,
controlled spatial variation of material properties through
1972) and then a large variety of Functionally Graded
stepwise gradation in functional concrete components.
Materials (FGM) production methods developed, but it
FGSCC can be used to design breakthrough FGC suitable
took up to the beginning the XXI century, to investigate
to withstand uneven environments or mechanical forces
some fabrication concepts applied to cement-based-
while avoid sharp interfaces. Rheological control is vital in
materials. Nevertheless, new challenges awaited most
successful production of not only a robust interlayer, but
of these fabrication processes for rapidly transforming
also of layers and interlayers depth as designed. A simply
them in promising precast useful applications.
method using SCC rheological capabilities, in terms of
design and characterization, was found to be effective to This investigation bestows an advanced production
prepare layered- or segmented-FGC sections capable to procedure that utilizes the controlled rheological
meet different structural requirements by only varying properties of the SCC for transforming actual stacking
SCC composition. Different FGSCC materials were and gradation processes -mainly based on vibration, co-
successfully cast. Its feasibility and potential sectional extrusion or extrusion and pressing- used for building
robustness also were experimentally demonstrated layered-FGC in a self-controlled production. Thus, a
using SEM (Scanning Electron Microscopy), EDX (Energy process based in the termed in FGM as segregating
Dispersed X-Ray Spectroscopy), XRD (X-Ray Diffraction) process (Kieback et al., 2003), which does not depend on
and ST (Splitting test) analyses. Results show that auxiliary methods after stacking for producing gradient
properly designed Self Consolidating Concrete (SCC) had is presented. Segregating processes for forming an
the ability of forming a graded interlayer of about 2-4 mm interlayer start with two macroscopically homogeneous
or less, moreover, this size is enough for avoiding a sharp materials (i.e. two cement-based mixes) which are
interface. converted into a graded substance by material transport
caused by using an external field (i.e. a gravitational or an
Keywords: FGSCC, segmented-FGC, SCC, rheology,
electrical field).
mechanical bond, microstructure.
Currently, layered-FGC presents two basic structures
of gradation: either gradually or stepwise Miyamoto
Introduction
et al., 1999). The stepwise gradation and the stacking
The structural concrete industry has advanced greatly followed by a segregating procedure are the most
in the past decades (Aitcin, 1998, De Schutter, 2011). widespread methods to build cement composites. Even
Considerable work has been done on enhancing properties if, a continuous grading is considered preferable to help
of the cement-based systems with fibers (Li, 2002; Brandt, relaxing interlayer stresses (Nogata and Takahashi, 1995;
2008), nano-particles (Kawashima et al., 2012), etc., but Ruys et al., 2001; Oxman et al., 2011), casting in layers or
the structural concrete practice is still firmly based on stacking is a common practice in the construction sector,
using concrete technologies in a bulk scale, in spite of being i.e. for strengthening or repairing, or for reducing thermal
most costly and rather inefficient, i.e. significant fractions gradient when casting mass concrete. On the other
of material are not used to their full potential (Shen et al., hand, as it was denoted by (Dias et al., 2009) the stacking
2008). To overcome these drawbacks, different layered- procedure is a straightforward way to generate functional
Functionally Graded Concrete (layered-FGC) elements gradation. However, it has been considered sometimes
have been designed (Maalej et al., 2005; Chen et al., 2006; insufficient for producing FGM because it leads to the
Baoguo et al., 2009; Dias et al., 2009; Li and Xu, 2009; Quek formation of sharp interfaces (Ruys et al., 2001). The
et al., 2010). This new trend of highly developed composite production of layered-FGC has revolved mainly around this

Organised by
India Chapter of American Concrete Institute 157
Session 2 A - Paper 3

approach, where an extraordinary effort has been made Halloran, 1998) it is reported that all the materials in the
for avoiding sharp interfaces inherent to layer methods adopted solution should have similar rheological behavior
that use vibration (Ho et al., 2001; Li and Xu, 2009; Baoguo for ensuring design criteria.
et al., 2009), extrusion and subsequent process (Shen et
Then, the widespread adoption of SCC concepts for
al., 2008; Dias et al., 2009), and co-extrusion (Chen et al.,
designing the whole solution will be an important step
2006).
forward in the FGC technological field, as some authors
Although there is relatively scarce literature focused on have shown (Ho et al., 2001; Bosch, 2010). However,
the interlayer formation using cement-based materials, neither the use of self-consolidating concrete/conventional
it is possible to derive that the best way up to now of vibrated concrete (SCC/CVC) nor the high performance
producing accurate and slim graded interfaces is based self-consolidating concrete/pervious concrete (HPSCC/
in well-controlled rheology when applying the stacking PC) may be considered more than just a first approximation
and subsequent segregating processes (Chen et al., 2006; for improving the process. They stated that only when
Shen et al., 2008; Dias et al., 2009) because those based SCC is placed over the other concrete solution is possible
on vibration only allow forming interfaces of about 2-5 cm to obtain a self-penetration or material transport without
(Ho et al., 2001; Baoguo et al., 2009). It was also reported by the need of using complex and/or limited systems for this
Baoguo and colleagues the need of using complementary aim. Preliminary works of the authors (Alonso, et al., 2011;
forms for producing regular and even layers. Thus, the only Nguyen et al., 2014; Río et al., 2015) based on SCC/SCC
possibility for meeting dimensional requirements similar solutions built accordingly (Río, 2010) show that it is possible
to those proposed for other layered-FGM (in the order of to obtain regular layers and thin graded interlayers (2-4
millimeters or less) is using extrusion based systems. On mm) if the SCC mixes have similar rheological properties.
the works of Chen and Shen (Chen et al., 2006; Shen et al., Nevertheless, they also stated the need to explore in depth
2008) it is pointed out that results are quite reliant on a the bond design of the interfacial zone. Currently, in spite
good control of mix rheological properties. They also have of the importance of quality control to design a feasible
showed that successful extrusion of mix can be reached if and robust grading processes, for the case of cement
the design and/or assessment considerations provided in based materials only limited interlayer monitoring and
some state of art papers (Benbow et al., 1987; Peled and testing have been performed (Chen et al., 2006; Shen et
Shah, 2003; Shen, 2003; Zhou and Li, 2005) are followed. al., 2008; Río et al, 2013). Moreover no simple methods for
Moreover, when doing their extruded samples, Shen and characterize mixes are used (Chen et al., 2006).
colleagues reported the appearance of wavy layers and
This paper highlights significant findings on a research
uneven thickness resulting from inadequate control of
effort focused on developing stepwise-FGC solutions with
plasticity after pressing although layers meet extrusion
gradually interfaces of about 2-4 mm by applying casting
requirements. On the other hand, Chen and colleagues
(wet-wet stacking) and subsequent self-homogenizing
reported that sometimes a tangling between layers could
and standard for concrete consolidating methods.
appear due to unmatched viscosity.
In the present paper, the focus has been put on the
Even though the viscosity seems relevant for stacking- requirements of the processing procedure as well as in the
grading and viscosity can be well controlled using SCC no microstructural composition and preliminary mechanical
literature was found on grading processes and SCC up to performance capabilities of the graded interlayer rather
recently (Bosch, 2010; Río, 2010). SCC is a fairly new type than in the FGSCC overall sample performances. The
of high performance material introduced as first time in study here, using different materials science analytical
Japan (Okamura and Ouchi, 2003), which as was pointed tools, and the splitting test for investigating the critical
out is a different approach to mix design and rheological issue of the fabrication process, has demonstrated its high
characteristics and also a new approach to casting concrete potential for extending the FGSCC procedure to different
enabled by adjusted its fresh properties (Khayak and De structural elements and material functionalities.
Schutter, 2014). Thanks to these SCC characteristics and
the systematic research on SCC performed (Skarendahl,
Experiments
2000 and 2006; Wallevik, 2003; Štirmer and Banjad Pečur,
2009; De Schutter 2011) different design and easy to apply This section addresses the experimental methods used to
assessment issues are available now-a-days (Concrete, fabricate and characterize the selected bi- layer FGSCC
2005; ACI 237R, 2007; ACI 238.1R, 2008). Therefore, this samples of φ8x10 cm, 10x10x10 cm or 10x10x40 cm. It
nowadays young but mature concrete can be considered includes a discussion of materials, mix proportions,
a promising technology for producing layered-FGC. When mixing procedure, the actual layered casting process and
establishing FGM design concepts, Kieback and colleagues the methodologies proposed for interlayer investigation.
(Kieback et al., 2003), reported that rheological conditions It involves 3 of the 16 mixes used for preparing the
should be chosen in such a way that gradient is not whole experimental program, which includes different
destroyed or altered in uncontrolled fashion. On the other performance tests (Alonso et al., 2011; Nguyen et al., 2014;
hand, in other two papers (Hoy et al., 1998) and (Crumm and Río et al., 2015).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

158 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

Materials are shown in Table 1, where quantities are expressed in

As the powder-type SCC design was chosen to meet the kg/m3 for the aggregates in saturated and surface dry
viscosity (T50= in the range of 5-6s) and flowing ability (SSD) conditions. The fluidity level at which the mix
(spread in the range of 550-650mm) requirements, so becomes unstable was determined using the variable
the powder components were selected accordingly. The dosage of the HRWRA. How to select mix proportions for
main powder components used for preparing the three ensuring a plastic viscosity allowing reaching thicknesses
FGSCC sample layers were ordinary Portland cement of interlayer of 2- 4 mm is discussed in a latter section.
Type I (OPC-I), calcium aluminate cement type E (CAC-E), A rotary planetary-type mortar mixer was used for mixing.
limestone filler (LF) with a fineness of 78.8% < 63μm and The dry ingredients were first mixed for 1- 2 min. Water
silica fume (SF) in powder form with the content of SiO2 was then added slowly as well as the HRWRA after which
larger than 92%. As the size of the small samples prevents mixing was continued for about 8 min to reach a paste-
the use of coarse aggregates, siliceous aggregates (SA) of like state with no segregation and/or bleeding. The mixing
grading (0-4 mm) with specific gravity, fineness modulus time was slightly larger for the LWA-SCC mixes.
and absorption of 2.61, 2.59 and 1.22%, respectively,
lightweight aggregate perlite (LWA-P) of grading Mix characterization
and density (0-1.5 mm and 80 Kg/m3) and vermiculite
The three SCC were tested for their plastic requirements
(LWA- V) --with grading and density of 0.5-4 mm and
at the fresh state. The test chosen was focused on the
100 Kg/m3 repectively-- were used for the granular
assessment of the flowing ability and resistance to
skeleton. Also a polycarboxylate-based superplasticizer
segregation using standard equipment, thus, inverted
(HRWR), Viscocrete TSG 30, was incorporated in all
cone (ASTM C1611M-14, 2014) were selected. As it was
mixes. Except for the cements, the powder materials
pointed out (Koehler and Fowler, 2003), in this way it is
were selected to enable the fabrication technique, a
not only easy to complete the test by one person but to
combination of casting and self-segregating process.
increase the precision of viscosity determination (T50) too.
Nevertheless, in spite of the recognized importance Viscosity control is of utmost importance to managing the
of using SF for improving the durability (Cohen and penetration area for the proposed methodology. It is a fact
Bentur, 1988) or LWA and CAC components to reduce that differences exist but it was also reported that these
Powder-type SCC proneness to fire spalling LF (Parr and differences were not influencing the interpretation of the
Wohrmeyer, 2006; Nguyen et al., 2013; Bakhtiyari et al., results. Moreover, the target values proposed for reaching
2014), the selection of such materials was done here only uniformly layer thickness between 1-3 mm were also
for being used as trace indicators due to their different determined with the inverted position (Alonso et al., 2011;
mineralogical composition in addition to meet their mix Río et al., 2015). Furthermore, this is more representative
rheological capabilities. Moreover, CAC was selected of what happens when filling the small elements in the way
also for checking if the use of cements with differentiated describes later on. The spread and T50 were determined
setting (Rabiet and López 1999) could lead to a failure for all the mixes and the values obtained are shown in
during fabrication. Table 1. The resistance to segregation was estimated
through visual observation. The results show that mixes
Mixture proportions and preparation were in general highly stable (neither segregation nor
The primary goal was to obtain mixes with similar water bled were observed during the tests).
rheological properties, although having differences in In addition of the plastic requirements at fresh state,
composition. Taking into account this overall aim, three mechanical performance was assessed by means of
mortar mix initial proportions were selected following compressive and splitting tensile strengths, obtained
the general criteria for powder-type SCC, according to from at least two cubic samples of 10x10x10 cm, tested at
(Concrete, 2005) and (Yoğurtcu and Ramyar, 2009) for the the age of 28 days after standard curing (Table 1).
specific LWA-SCC mixes. The definite mix proportions

Table 1
Mix proportions and characterization at fresh and hardened states

SA LWA-P LWA-V HRWR Flow

Mix OPC-I CAC-E SF LF w/c w/p t50 fc fct
(0-4) (0-1.5) (0-4) TSG30 spread

Kg/m3 mm s Mpa Mpa

M1 500 - 45 2150 - - - 5.4 0.40 0.26 610 6.0 86.6 4.7

M2 400 - - 47 669 13.2 42.4 6.5 0.40 0.36 570 6.5 22.5 1.9

M3 - 400 - 47 669 13.2 42.4 6.5 0.43 0.38 580 6.4 22.3 2.1

Organised by
India Chapter of American Concrete Institute 159
Session 2 A - Paper 3

Stacking and segregating

A cost-effective method based on SCC technology was
used to fabricate the bi-layered FGC samples. It involves
just two stages stacking and segregating. The stacking
method is quite similar to the one described by (Shen et
al., 2008), excepting that, in this proposed FGSCC process,
neither the use of an extrusion system for casting layers,
nor special reduction in dimensional considerations
are needed; pressing cause an extension of length and
correspondingly reduction in thickness, thus sample length
must be lesser than mold length. Moreover, no external Fig. 1: Casting of big (left) and small (right) elements
fields are necessary for the segregating process. The low
yield stress and plastic viscosity of SCC impart the required
characteristics that enable the mix to flow laterally or
vertically into the forms until the layer thickness is reached
and then self-consolidate without the need of any vibration
or external field –see Figure 1 left reproduced from (Río,
O. et al., 2015)–. Through this simply procedure proposed
in 2010 (Río, 2010) it is possible that the mix of upper layer
penetrates into the lower layer mix 1-3 mm due only to the
hydrostatic head caused by the second layer if viscosity
Fig. 2: SCC1/SCC2 cubic (left) and cylindrical (right) samples
determined by T50 is in agreement. The different tests
performed up to now considering more than 16 different
Scanning Electron Microscopy (SEM) and Energy-
mixes and more than 100 elements –different in size or
dispersive X-ray Spectroscopy (EDX) observation
shape– showed that mixes with T50 in the range 6-8s result
appropriate. Through this “self” stacking and segregating The study was carried out with a SEM FEI Quanta 200 and
process an opening bond can ensure the formation of a an X-ray energy dispersive analyzer attached to it. The
preliminary continuous element. microscope operated at 20 KV and under high vacuum
conditions. The small samples were used for such
Different layers of either 5/5 cm thick were stacked one experiments. The samples were polished by hand with an
after the other into glass cylindrical molds of φ8x10 cm abrasive sheet of silicon carbide to remove any possible
and cubic steel molds of 10x10x10 cm or 10x10x40 cm. carbonate or inlay material contamination. This last, due
These last ones were used for cutting slides at different to drops on upper part of mold surface, appeared during
lengths for visually checking the interface of the different the casting of the first mix. Images were obtained with
sections along element. The gap time of placing between backscattered electrons. In order to cover an observation
layers was ≅15 minutes, to ensure a wet-wet casting. It is field containing both layers, several consecutive images
advisable that, if more time for preparing the subsequent were taken in the gradation direction. Moreover, a most
mix is needed, the upper part of mold must be covered in detailed analysis was made in the close to the interface
order to avoid drying, which would prevent penetration and area and the composition was checked by EDX along the
cause defects in the surface, leading to a non-compliance hand cut sample.
grading depth. A mark was put in mold for checking that
the first layer reached freely the level in the whole mold. X-Ray Diffraction Analysis
Bearing in mind the size of the glass molds, a special
injection system to pump the mixtures was employed for A D5000 powder diffractometer from Bruker in 2Θ/ Θ
preparing the cylindrical elements (Figure 1 right) while mode was used. The Cu anticathode operated at 45 KV
for the remain elements the one shown in prior papers and 35 mA. A beta filter was employed to select Kα1 and
(Río, et al., 2015) was used. Kα2 wavelengths. Samples were spun and diffraction
patterns were acquired in a 2Θ angular domain ranging
from 5 to 90 degrees by steps of 0.02 counting 5 seconds
Curing
per step. Divergence, sample and detector slits were set
After casting, samples were standard curing (T=20±2°C to 2, 0.2 and 0.6 mm respectively and used in conjunction
and RH=98±2%) up to the testing date: 28 days. As it with conventional Soller slits. The quantitative Rietveld
is customary after 24 h samples were demolded (by Analysis was made with Materials Studio 5.539. Its Reflex
carefully breaking the molds in the case of the glass ones) QPA module was used to carry out the computations.
and then kept in the curing chamber. A total of 14 FGSCC
samples were prepared: 2 beams, three cubic and two Splitting test
cylinders of each FGSCC composition: the SCC1-SCC2 and
the SCC1-SCC3. Figure 2 shows the aspect of two of the The test was performed as it is described in the EN
FGSCC samples. standard (UNE-EN 12390-6, 2010) with the only difference

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

160 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

that the load was applied in correspondence with the

transition zone. The tests were conducted, by applying
the load continuously at the stress rate of 0.05 N/mm2.s,
on the hydraulic testing machine from IBERTEXT (with
maximum load capacity of 3000 kN), up to failure of
sample. The measured splitting tensile strength fct, of
the specimen was calculated to the nearest 0.05 N/mm2
using the formula provided in the standard, as for all the
studied samples the transition or graded area was in the
middle of the specimen.

Interlayer Mechanical Properties Fig. 4: Average results of splitting tension strength on three
As it was pointed out in the introduction the layering identical samples
method is considered sometimes insufficient for producing
FGM because it leads to the formation of sharp interfaces reducing the tension in the middle of the sample leading to
(Ruys et al., 2001). As it was also pointed out by Sánchez- a situation similar to that shown in Figure 3, although with
Herencia (Sánchez-Herencia, A.J., 1996) this is due to the only a single point it is not possible to show the gradient
shear residual stresses that appear between dissimilar trend. Thus, the used splitting test allows determining
materials. In his paper he also showed that these shear that the formation of a gradient structure actually
stresses, although could not fully avoid, at least could be happens, and also give useful insights for better designing
reduced if between the two layers a gradient was created. the gradient structure. However, more studies need to be
Figure 3 adapted from (Sánchez-Herencia, 1996), performed for using these values as design parameters.
reproduced what should the situation be for the analyzed Results also indicate that the values were higher for the
cases if a finite element analysis is done. Moreover, materials using same cement type than for those using
although concrete is not normally designed to resist direct different cements. This effect, which in turn shows that
tension, the knowledge of tensile strength is of value in the M1/2 material has a potential outstanding structural
estimating the load under which cracking will develop but behavior compared to the M1/3, is probably due to the
also to explain if the formation of an intermediate material differentiated hydration trends of both cements (Milestone,
is formed or not. Cubic specimens of both series were 2006). Although this topic would need further studies, it
tested under splitting test up to failure as it was described seems that an increasing on the gradient thickness or the
in the prior section. Results are those presented on the definition of a step by step graded structures (Sánchez-
bar chart of Figure 4, where also the results of bulk Herencia, 1996) --using in this case different proportions
specimens (M1-M3) are introduced. of cement to make the transition-- should contribute to
reduce the tensional strengths.
It is noteworthy to say that the values obtained for the
mismatched graded materials were in all cases higher
than those of the individual bulk materials presenting Microstructure Analysis
lower splitting tension strength. This means: 1) that a
Preliminary quality control and naked-eye observation
different material was formed, which in turn demonstrated
that a penetration of one material in the other occurs A ruler and a digital caliber (Mitutoyo) were used to
and 2) results show that this gradient structure allows measure the size of the different FGSCC samples. It was
perceived that their size and layer thickness were as
designed within an admissible error of 0.5 mm. Surface
appearance and cracking of samples were primarily
observed visually (Figures 2 and 5).
No cracking were observed on the interfacial areas but a
small rather edgeless transition zone. It is noted that on
the surface of small FGSCC samples there were some
tiny cracks due to the outer cement paste formation
and superficial carbonate concentration (see Figure
2 right) and also some small pores (see Figure 5) but
they disappeared after polishing, necessary to analyze
them by Scanning Electron Microscopy (see Figure 6).
The use of proper size molds and the use of a suitable
demolding agent may be also important to achieve the
Fig. 3: Schematic of laminated section shear: (a) without and
required surface, especially when non-permeable molds
(b) with FG formation

Organised by
India Chapter of American Concrete Institute 161
Session 2 A - Paper 3

Table 1
Mix proportions and characterization at fresh and hardened
states

Quartz CaCO3 Ca(OH)2 Rwp

Sample Mix
%

M1 78.3 17.3 5.3 20

Fig. 5: FGSCC Surface appearance M1/2 (a) and M1/3 (b) FGSCC1/2
M2 87.3 5.9 6.7 26
macrophotography images
M1 80.5 15.2 4.3 20
FGSCC1/3
are used. No other similar defects where observed in the M3 94.1 4.6 - 21
medium size elements as can be observed in Figure 2 left.
Regarding the transition zone, they were more visible in
the M1/3 FG elements due to the different coloration of weaker diffraction peaks were present, but their intensities
OPC and did not allow using them for quantitative analysis.
Quartz is by far the main phase; its intensity dwarfs the
CAC used for the specimens (see Fig. 5b) than on M1/2
diffraction signal of the other crystalline phases. The
(see Fig. 5a) samples where the same cement was used
quantification of the crystalline phases does not take into
for both layers.
account the amorphous or poorly crystalline phases, i.e.
As it was mentioned, none of the FGSCC samples, either gels of calcium silicate hydrates (C-S-H) and aluminum
the one made with OPC/CAC or those made with OPC/ hydroxide present in Portland and calcium aluminate
OPC have shown any defect after being polished. Figure 6 mixtures respectively, therefore they are not taken into
shows the set of images taken at SEM, for the FGSCC1/2 consideration. That is why the percent composition seems
sample show in Figure 2 right, prior to take a most detailed to be distorted. In any case, the diffraction analysis proves
scanning of the central area of sample or area BC (where that the hydration of the cementitious mixtures occurred,
the interface was located). The slightly darken area AB as expected at the temperature set.
corresponds to the mix M1 while the light one, from C to E,
The calcium hydroxide levels for Portland mixtures and
corresponds to the mix M2 containing the LWA. Contrarily
the CaAl2O4.10H2O values for calcium aluminate mixtures
the dark area in the FGSCC1/3 was located in the zone CE
are within the normal levels. On the other hand, other
due to the presence of the CAC in this area.
phases such as vermiculite or perlite were not detected.
Vermiculite, which is crystalline and easily prone to
preferential orientation, was not seen, and perlite could
not be detected either since a separate diffraction analysis
for this phase proved it is a non-crystalline phase. The
specimen M3 contained also 1.3% of CaAl2O4.10H20. The
figures are affected with an error uncertainty of ±1. Last
column shows the goodness of fitting for every Rietveld
refinement.

SEM/EDX analysis
Elemental analysis by EDX on different parts of the
Fig. 6: FGSCC longitudinal scan SEM images of polished sample specimens reveals that the most conspicuous grains
of Figure 2 right are constituted by quartz. The elemental composition
of some spots seems to indicate that it is made of belite
X-Ray Diffractometry Ca2SiO5 and other parts could be identified as CaCO3. On
the extreme part of FGSCC1/3, samples corresponding to
To test that the cementitious mixtures were properly
the M3 mix, some spots rich in Ba and Al were detected.
hydrated, samples taken at 1cm of each edge (M1 and
On the contrary, on the extreme part, corresponding
M2 or M3 zones) of every FGSCC sample series were
to M1, regions reach in Fe was observed. Obviously, the
analyzed by X-ray diffraction. Portions of approximately
belite presence is due to a partial hydration of the OPC,
one gram were ground in an agate mortar and used to
whereas the calcium carbonate is the main component of
get statistically representative diffraction patterns. The
the filler used. The intrinsic inhomogeneous nature of the
results are presented in Table 2.
mixtures present in the specimens – with many phases
The main crystalline phases found were: quartz, calcium of variable particle size distribution – does not allow to
hydroxide, calcium carbonate and CaAl2O4.10H2O –this procuring a composition profile that would provide clues
phase present only in the calcium aluminate pastes. Other on the exact extension of a transitional zone that was in

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

162 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

Fig. 8: Wavy layers and uneven thickness resulting from

inadequate control of rheological parameters

regarding quality control

issues. Compared
to other accurate
production methods the
methodology has two
main advantages: i) it
is not necessary to use
complex methods that
are only available on lab
Fig. 7: FGSCC1/3 SEM/EDT images of the interface zone Fig. 9: Sharp interface due to for the mix quality control
an inappropriate layer placing and ii) it does not present
any case not greater of 3 mm considering the EDT analysis order limitations in terms of
of the paste (Figure 7). Figure 7 shows an enlarged image mix composition since,
of the transition region of one of the FGSCC1/3 samples also when different cements are used, not wavy surfaces
and the corresponding analysis of the paste of areas name are formed. As it was reported by Shen et al., the wavy
as 1 and 2. As it is possible to observe the percentage of interfaces between layers were due to the use of different
aluminum increased from 2.6% by weight in zone 1 up to cements, which produced different plasticity between
10.1% by weight in zone 2 while the percentage of calcium layers (Shen et al., 2008).
decrease from 39.8% in zone 1 to 25.4% in zone 2. The
distance between both zones is approximately 2mm. On The presence of coarse aggregates and/or fibers do not
the other hand, the SEM photographs do not show any produce changes since in all the specimens fabricated
discontinuity, which indicates that the bonding between up to now (Río, 2010 and 2015; Alonso et al., 2011,
the layers is neat. Nguyen, 2015), the graded interface was formed and the
depth was in the range of 1-3 mm; such values for these
mixes are in compliance (do not exhibited segregation).
Processing Interpretations It has been observed that when a section is made of
Matching the paste true viscosity and flow ability of mixes high performance and normal strength mixes, placing
is a key issue for FGSCC fabrication. If the viscosities of the higher packing layer below the lower packing
mixes are quite different from each other, and especially layer results in a deficient penetration that leads to
if the one placed down has significant lower viscosity, in the formation of a sharp interface. One example of this
comparison to the upper one, the penetration results in phenomenon is shown in Figure 9.
wavy interfaces as the hydrostatic head cannot be well
controlled. Two examples are shown in Figure 8, in which In this particular case, the order of fabrication was
mixes with two different viscosities and with and without deliberately inverted to test other possibilities (M3 layer
different cement compositions were used. This defect is was placed on top of M1). Taking into account that no major
much higher if the mix presents segregation, although the differences exist regarding composition and rheological
segregation observed was not as higher as to reject the properties, the sharp interface formation could only be due
mix. to a hydrostatic head --insufficient to allow penetration
from a mix with lower content of finer particles into a more
This phenomenon that can occur independently of mix compact mix, i.e. with a higher content of finer. It should be
composition can easily be controlled. We found that a pointed out that although this problem has not appeared
T50 in the order of 5.5-6.5s and spreads between 550- in mixes with similar packing, it could be a limiting factor
650 mm is sufficient to produce interlayer bonding strong for cases where mixes with different packing degree are
enough to avoid delamination while producing layers of put together.
uniform thickness. Therefore, the present methodology
as it is based in SCC technology has several advantages

Organised by
India Chapter of American Concrete Institute 163
Session 2 A - Paper 3

Conclusions observed only when a higher load was applied. This

demonstrates not only the formation of a combined
Processing of Layered-FG cement based materials has
material --as a skeleton change could anticipate--
been explored at laboratory scale using different stacking,
but also provides insights on how the graded can be
segregating and consolidation –curing- methods.
designed for properly relaxing stresses.
Therefore, a variety of processing methods, based on
different stacking sequences and segregating features, 6. Despite of all these achievements, the authors believe
is available today for almost any cement based material that there will be some open questions for the future,
combination, since consolidating or curing do not depend when industrial applications of FGSCC develop.
on the method but on the age of testing. State-of-the-art These include: i) adaption of manufacturing to mass
seems to indicate that at present this craft depends not only production and upscaling, ii) determination of most
on the concrete technology, but on the type and extension cost-effective gradient solutions for improving tension
of the gradient and on the geometry of the required relaxation, iii) on-site quality control. Modeling based
component too. Our main priority has been to contribute on accurate testing may contribute to solve some of
to state-of-the-art of layered-FGC by introducing a these problems and first steps. In this sense it has
novel freeform and component independent processing been undertaken by means of present results and
technique --based on the SCC technology mix design methodologies.
and quality control methods-- and that has been termed
by the authors as FGSCC. The main concluding remarks
of this work, which is focused on evaluating the gradient
Acknowledgement
extension and its strenght capabilities experimentally, The authors want to express their thanks to other CSIC
using SEM (Scanning Electron Microscopy), EDX (Energy collaborators involved in other different aspects of
Dispersive X-Ray Spectroscopy), XRD (X-Ray Diffraction) the experimental program. Thanks are also extended
and ST (Splitting test) are: to SIKA España for their support in the experiments.
The work presented herein was partially funded by
1. Rheology among different layers should be matched the Spanish Ministry of Economy and Competitiveness
to obtain good FGSCC products. Too low viscosity or (Projects BIA2008-06673-C02-00 BOHEC, BIA2013-
mixes that exhibit segregation would result in wavy 48480-C2-1&2-R SCC-Pump and Proposal for
layers and uneven graded thickness. On the other Sincrotron ALBA Mineral quantitative determination on
hand mixes having too high viscosity could lead to a the interface of cementitious bi-layers by synchrotron
sharp interface. micro diffraction).
2. Slump test spread and T50 determination were found
adequate to assess SCC rheology properties. Mixes References
having spread values in the range of 550-650 mm 1. ACI 237R, 2007. Self Consolidating Concrete Practice. In: ACI Manual
and T50 in the range of 5-6s allow obtaining the target of Practice. s.l:American Concrete Institute.
graded thickness: 3-4 mm. 2. ACI 238.1R, 2008. Report on Measurements of Workability and
Rheology of fresh concrete.. s.l.:American Concrete Institute.
3. Naked-eye observation by close-up views of FGSCC 3. Aitcin, P., 1998. High Performance Concrete. London: E & Fn Spon.
samples showed no distinctive boundaries between
4. Alonso, M.C., Rio, O., Rodríguez, C., Nguyen, V. D., 2011. Avoiding
layers. The transition between layers demonstrated spalling of precast HPC& UHPC elements by using a novel layered
the small and rather gradual material transition and concept. In: E. D. F. Koenders, ed. Pro080: 2nd International RILEM
that its straightness, parallel to the casting direction, Workshop on Concrete Spalling due to Fire Exposure. s.l.:RILEM
was well preserved. This was most easy to observe Publications: 393-400.
when trace material was present because of the 5. ASTM C1611M-14, 2014. Test method for slump flow of self-
different color. consolidating concrete, s.l.: ASTM International.
6. Bakhtiyari, S., Allahverdib, A., & Rais-Ghasemic, M., 2014. A case
4. SEM images inspection indicated that the bonding study on modifying the fire resistance of self-compactiong concrete
between layers is continuous without presence of with expanded perlite aggregate and zeolite powder additives. Asian
any abrupt borderline though materials of each layer L. of Civil Eng. , 15(3): 339-349.
have different mineralogical composition, as seen by 7. Baoguo, M.A., Dinghua, Z., Li X.U., 2009. Manufacturing technique
EDX and XRD. Also the distribution of aggregates is and performance of functionally graded concrete segment in shield
tunnel. Front. Archit. Civ. Eng. China, 3(1): 101-104.
uniform along the specimens without any signal of
segregation or discontinuity in a similar way as in a 8. Benbow, J. J., Oxley, E. W., Bridgwater J., 1987. The Extrusion
Mechanics of Pastes-the Influence of Paste Formulation on
bulk CC material. Extrusion Parameters. Chem. Eng. Sci., 42(9):. 2151-2162.
5. Although compression load in splitting tension test 9. Bever, M.B., Duwez, P.E., 1972. Gradients in composite materials.
was applied right onto transition zone, no adhesive Materials Sci Eng, 10(1): 1-4.
failure was found for strength value similar to the one 10. Bosch, C. R. O. F.-L. L., 2010. Fire resistance segment for tunnels
of material with load tension capacity. Failure was and manufacturing process. Spain, Patent No. ES.2 326 936.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

164 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Improving segmented Functionally Graded Concrete concept by using SCC technology

11. Brandt, A., 2008. Fibre reinforced cement-based (FRC) composites 32. Nogata, F., Takahashi, H., 1995. Intelligent functionally graded
after over 40 years of development in building and civil engineering. material: bamboo. Comp. Eng., 5(7): 743-751. Okamura, H., Ouchi,
Compos Struct, 86(1-3): 3-9. M., 2003. Self-compacting concrete. J. of advanced concrete
technology, 1(1):. 5-15.
12. Chen, Y., Struble, L.J., Paulino, G. H., 2006. Fabrication of
Functionally Graded-cellular Structures of Cement-based Materials 33. Oxman, N., Keating, S., & Tsai, E., 2011. Functionally graded rapid
by Co-extrusion. Applied Ceramic Technology, 5(5):532-537. prototyping.. Proceedings of VRAP in Innovative Developments in
Virtual and Physical Prototyping: 483-490.
13. Cohen, D., Bentur, A., 1988. Durability of Portland Cement-Silica
Fume Pastes in Magnesium and Sodium Sulfate Solutions. Materials 34. Parr, C., & Wohrmeyer, C., 2006. The advantages of calcium
Journal, 85(3):148-157. aluminate cement as a castable bonding system., Paris: Lafarge
Aluminates.
14. Concrete, S., 2005. The European Guidelines for Self-Compacting
Concrete. s.l.:s.n. 35. Peled, A., Shah, S.P., 2003. Processing Effects in Cementitious
Composites: Extrusion and Casting. J. Mater. Civil Eng., 15(1): 189-
15. Crumm, A.T., Halloran, J.W., 1998. Fabrication of Microconfigured
192.
Multicomponent Ceramics. J. Am. Ceram. Soc., 81(4):1053-1057.
36. Quek, S.T., Lin, V.W.J., Maalej, M., 2010. Development of functionally-
16. De Schutter, G., 2011. Self-compacting concrete after two decades
graded cementitious panel against high-velocity small projectile
of research and practice. Rotorua, New Zealand, s.n.:151.
impact. International Journal of Impact Engineering, 37(8): 928-941.
17. Dias, C.M.R., Savastano, Jr. H. and John, V.M., 2009. Mixture
37. Rabiet, C., López, D., 1999. La reactividad de las mezclas de cemento
Screening Design to Choose Formulations for Functionally Graded
portland y aluminosos. Construcción y Tecnología en Concreto, p.
Fiber Cements. Materials Science Forum, 10(65):631-632.
http://www.imcyc.com/revistacyt/oct11/anteriores.html.
18. Ho, D.W.S., Sheinn, A.M.M., & Tam, C.T., 2001. The sandwich concept
38. Río, O, Nguyen, V.D., Turrillas, X., 2013. Functionally-Graded
of construction with SCC. Cement and Concrete Research, 31(9):
Self Compacting Cement Composites. In: Fifth North American
1377-1381.
Conference on the Design and Use of Self-Consolidating Concrete.
19. Hoy, C.V., Barda, A., Griffith, M., Halloran, J.W., 1998. Microfabrication Chicago: K Wan and SP Sha Eds.
of Ceramics by Co-Extrusion. J. Am. Ceram. Soc., 81(2): 152-158.
39. Río, O., Nguyen, V. D., Nguyen, K,, 2015. Exploring the Potential of
20. Kawashima, S., Hou, P., Corr, D., Shah, S.P. , 2012. Modification of the Functionally Graded SCCC for Developing Sustainable Concrete
cement-based materials with nanoparticles. Cement and Concrete Solutions. Journal of Advanced Concrete Technology, 13(3): 193-204.
Composites, 36(0): 8-15.
40. Río, O., 2010. Pre-fabricated cement-based hybrid section and
21. Khayak, K., De Schutter, G. Ed. , 2014. Mechanical properties of method for the production thereof. s.l. Patent No. WO2010122201 A2.
self-compacting concrete. State-of-the-Art Report of the RILEM
41. Ruys, A.J., Popov, E.B., Sun, D., Russell, J.J., Murray, C.C.J., 2001.
Technical Committee 228. s.l.:Springer.
Functionally Graded electrical/thermal ceramic systems. J. Eur.
22. Kieback, B., Neubrand, a., Riedel, H., 2003. Processing techniques Ceram. Soc., 21(10-11): 2025-2029.
for functionally graded materials. Materials Science and
42. Sánchez-Herencia, A.J., 1996. Functional gradient ceramic
Engineering, 362(1-2): 81-106.
materials. Bol. Soc. Esp. Ceramica Vidrio, 35(4): 247-256.
23. Koehler E.P., Fowler, D.V., 2003. Summary of Concrete Workability
43. Shen, B., Hubler, M., Paulino, G.H., Struble, L.J., 2008. Functionally-
Test Methods. ICAR 105-1, Austin, Texas: International Center for
graded fiber-reinforced cement composite: processing,
Aggregates Research .
microstructure, and properties. Cement and Concrete Composites,
24. Li, Q. H., Xu, S. L., 2009. Experimental investigation and analysis 30(8):663-673.
on flexural performance of functionally graded composite beam
44. Shen, B., 2003. Experimental Approaches for Determining
crack-controlled by ultrahigh toughness cementitious composites.
Rheological Properties of Cement-Based Extrudates, Hongkong:
Science in China series e-technological sciences, 52(6): 1648-1664.
s.n.
25. Li, V., 2002. Large volume, high-performance applications of fibers
45. Skarendahl, Å. &. Petersson O. eds., 2000. : Self Compacting
in civil engineering. J Appl Polym Sci, 83(3): 660-686.
concrete, State of the Art Report of RILEM TC 174-SCC. s.l.:RILEM
26. Maalej, M., Leong, K. S., 2005. Engineered cementitious composites publications.
for effective FRP-strengthening of RC beams. Compos Sci Technol,
46. Skarendahl, Å., Billberg P. ed., 2006. Report 35: Casting of
65(7-8): 1120-1128.
Self Compacting Concrete-Final Report of RILEM TC 188-CSC .
27. Milestone, N., 2006. Reactions in cement encapsulated nuclear s.l.:RILEM publications.
wastes: need for toolbox of different cement types.. Advances in
47. Štirmer, N., Banjad Pečur, I. , 2009. Mix design for self-compacting
Applied Ceramics, 105(1): 13-20.
concrete. Građevinar, 61(4), pp. 321-329. UNE-EN 12390-6, 2010.
28. Miyamoto, Y., Kaysser, W., Rabin, B., Kawasaki, A., Ford, R., 1999. Tensile splitting strength of test specimens.. s.l.:AENOR.
Functionally graded materials: design, processing and applications.
48. Wallevik, O., 2003. Rheology- A scientific Approach to develop Self-
s.l.:Kluwer Academic Publishers.
Compacting Concrete. In: Proc. of the third Int. Symposium in Self
29. Nguyen V.D., 2015. Mechanical behaviour of laminated functionally Compacting Concrete. Reykjavik: s.n.:23-31.
graded fibre-reinforced self-compacting cementitious composites.
49. Yoğurtcu, E., Ramyar, K., 2009. Self-compacting lightweight
Madrid: Doctoral Thesis presented at the Pol. Univl of Madrid.
aggregate concrete: design and experimental study. Magazine of
oa.upm.es/33772/.
Concrete Research, 61(7): 519-527.
30. Nguyen, V.D., Alonso, M.C., Aráoz, G., Rio, O., 2013. Hybrid cement-
50. Zhou X., Li Z., 2005. Characterization of Rheology of Fresh Fiber
based materials exposed to fire. Proceedings of CILASCI congress:
Reinforced Cementitious Composites through Ram Extrusion. Mater.
429-438.
Struct., 38(275): 17-24.
31. Nguyen, V.D., Río, O., Sánchez-Gálvez, V., 2014. Performance of
hybrid cement composite elements under drop-weight impact load.
Materiales de Construcción, 64(314): p. e017.

Organised by
India Chapter of American Concrete Institute 165
Session 2 A - Paper 3

Olga Río
Olga Río holds degree in Civil Engineer (University of Tucumán, Argentina), a PhD in Civil Engineer (Technical
University of Madrid). Scientific Researcher at the Spanish High Council for Scientific Research (CSIC) since
1988, at Eduardo Torroja Institute for Construction Sciences. Her main research topics are Structural
Concrete Technology, Structural Analysis and Evaluation of structural concrete performance.

Khanh Nguyen
Khanh Nguyen, born 1982, received his CEng in 2007 and PhD in 2013 from Technical University of Madrid.
He is since 2013 in the Eduardo Torroja Institute for Construction Sciences, CSIC. Recently his main areas
of research are Structural Concrete technology, Dynamic of Structures and Structural Health Monitoring.

Xabier Turrillas
Xabier Turrillas holds degrees in Chemistry and Pharmacy, a PhD in Materials science (University of
Grenoble) and a PhD in Chemistry (University of Navarre). Permanent Scientist at the Spanish High Council
for Scientific Research since 1996, first at Eduardo Torroja Institute for Construction Sciences and from 2014
at the Institute of Materials Science of Barcelona. His main topic of research is Cement Chemistry studied
with synchrotron radiation.

Judith Oró
Judith Oró is Dr in Chemistry by the materials science program. She is technical supervisor of the Electron
Microscopy Service at ICMAB-CSIC. She is an expert in Transmission Electron Microscopy (TEM), Electron
Difraction, Energy Dispersive X-ray
(EDX), Scanning Electron Microscope (SEM) and electron beam lithography, techniques applied for the
microstructural characterization of complex materials.

Anna Esther Carrillo

Anna Esther Carrillo has a degree in Geological Sciences and Master degree in Materials Science (UAB).
At present she is Technical Supervisor at Electron Microscopy Service at the Institute of Materials Science
of Barcelona. She is an expert in SEMEDX and electron beam lithography, techniques applied for the
microstructural characterization of complex materials.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

166 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Controlling plastic shrinkage cracking in concrete using polypropylene microfiber

Controlling plastic shrinkage cracking in concrete using

polypropylene microfiber
M. Sappakittipakorn and P. Sukontasukkul
Department of Civil Engineering, King Mongkut’s University of Technology North Bangkok
N. Banthia
Department of Civil Engineering, University of British Columbia

Abstract water to fill the pores, the drying of pores induces surface
tension and pore contraction resulting in the shrinkage of
This research is aimed to study the use of polypropylene
concrete (Wittmann, 1976). If there is no any restraint, the
microfiber to reduce the plastic shrinkage cracking
of concrete. The test method to determine the plastic concrete shrinks freely that only shortens its dimension.
shrinkage crack was carried out as per ASTM 1579. After On the other hand, when a restraint exists, a tensile stress
mixing, fresh concrete was cast in a modified mold having develops internally. If the tensile stress exceeds the
20 cm wide, 10 cm deep, and 35.5 cm long and immediately ultimate tensile strength of concrete, it will begin to crack
placed in a hot and dry environment chamber controlled at (Grzybowski and Shah, 1990).
40-45 °C, 30-35 % relative humidity and evaporation rate Since the free water in capillary pores plays important role
was 0.79 kg/m2/hr. Subsequently the length, width, and to balance the water loss by evaporation, the change of
amount of cracks were measured by using a microscope w/c ratio in concrete would considerably affect the plastic
at every one hour for six hours. Two sizes of polypropylene shrinkage. It was thus interested to determine such effect
fiber were tested; one is 12 mm long (aspect ratio = 375) in the present study by varying the w/c ratio at 0.4, 0.5,
and another is 19 mm (aspect ratio = 1056). The volume and 0.6. Following to that, in the concrete mix proportion,
fraction of the fiber was varied at 0.1%, 0.2%, and 0.3%. the amount of water was 200 kg/m3 constant while the
The water to cement ratio was also varied at 0.4, 0.5, and cement content was varied respectively at 500, 400, and
0.6. From the test results between plain concrete and 333 kg/m3. The addition of polypropylene microfiber has
polypropylene fiber reinforced concrete, it was found that been known as one of the effective means to minimize
both 12 mm and 19 mm polypropylene fibers can reduce the plastic shrinkage. Therefore, its effectiveness on
plastic shrinkage crack in concrete. However, the 19 mm impeding plastic shrinkage crack was also examined.
fiber (having higher aspect ratio) performed better as it Two types of polypropylene fiber – one is 12 mm long
has higher aspect ratio. Moreover, the crack reduction was with the aspect ratio of 375 and another is 19 mm long
also proportional to the number of fiber volume fraction. with the aspect ratio of 1056 – were evaluated at the fiber
Keywords: Plastic Shrinkage Cracking, Polypropylene dosages of 0.1%, 0.2%, and 0.3% by volume. To evaluate
Microfiber, Concrete the plastic shrinkage crack in these concrete specimens,
the ASTM C1579 was followed with some modification to
its specimen base.
Introduction
Plastic shrinkage cracks where aggressive chemicals can
Experimental Plan
easily seep in to promptly degrade concrete and reinforcing
steel inside impair concrete durability. In general, it can Materials and Mix Proportion
be prevented by applying some construction methods The test specimens in this study were divided into two
to reduce the rate of water evaporation from concrete,
sets; one was ordinary plain concrete (OPC) and another
adjusting the concrete mix proportion (Almusallam,
was fiber reinforced concrete (FRC). In the OPC set, the
et.al., 1998), using shrinkage reducing admixtures (Shah,
influence of cement amount was solely examined by
Karaguler, and Sarigaphuti, 1992), or implementing
varying the water to cement ratio (w/c) at 0.4, 0.5, and 0.6
fiber reinforcement (Banthia, Azzabi, and Pigeon, 1993;
while keeping the amount of water constant at 200 kg/m3.
Soroushian, Mirza, and Alhozaimy, 1993; Balaguru, 1994;
The OPC mix proportions were listed as shown in Table 1.
Soroushian and Ravanbakhsh, 1998; Banthia and Gupta,
For the FRC set, all of the OPC mixes were repeated with
2006; Boghossian and Wegner, 2008).
the addition of polypropylene and polyester fiber (except
Plastic shrinkage commonly happens in fresh concrete for the 0.6 w/c OPC mix in which no cracking was found).
when exposed to high temperature, low relative humidity, Two sizes of the polypropylene fiber were evaluated – one
and windy environment. As the free water in capillary is 12 mm long with the aspect ratio of 375 called as PPM
pores evaporates so rapidly that there is not enough bleed (as shown in Figure 1) and another one is 19 mm long with

Organised by
India Chapter of American Concrete Institute 167
Session 2 A - Paper 4

the aspect ratio of 1056 called as PPG (as shown in Figure

2). The fiber amount was varied at 0.1%, 0.2%, and 0.3% by
total volume as summarized in Table 2. The designation of
all mixes was also listed in Table 2.

Table 1. Mix proportion of the ordinary plain concrete set

Composition (kg/m3) Water to Cement Ratio (w/c)

0.4 0.5 0.6

Portland Cement 500 400 333
Water 200 200 200
River Sand 660 744 798
Crushed Limestone 992 992 992

Table 2. Mix proportion and designation of the ordinary plain

concrete and the fiber reinforced concrete set Fig. 2: Polypropylene fiber having the length of 19 mm and the
aspect ratio of 1056 – called PPG
Mix Designation w/c Polypropylene Fiber (% by volume)
PPM PPG
Preparation of Specimens and Test Apparatus
OPC-0.4 0.4 - -
The OPC and FRC mixes were prepared by using a pan
OPC-0.5 0.5 - -
mixer, consolidated into a mould with a vibrating table,
OPC-0.6 0.6 - -
and finished to achieve a flat and smooth surface. For the
PPM-0.1% 0.1 - evaluation of plastic shrinkage crack, all OPC and FRC
PPM-0.2% 0.2 - specimens were tested immediately after casting. The
PPM-0.3% 0.4 0.3 - specimens were casted in a mould (as shown in Figure
0.5
PPG-0.1% 0.6 - 0.1 3), which was partially modified from the mould stated in
PPG-0.2% - 0.2
ASTM C1579. The dimension of the mould is 120 mm wide,
100 deep, and 560 mm long. To create a restraint to the
PPG-0.3% - 0.3
specimens, three triangular prisms were installed at the
Note : OPC represents “ordinary plain concrete”
base similar to the standard mould plus four extra screws
PPM-a represents “concrete reinforced with
polypropylene (12 mm long and 375 aspect ratio) at the ends were implemented.
PPG-a represents “concrete reinforced with polypropylene
(19 mm long and 1056 aspect ratio) To simulate a hot and dry condition to induce plastic
a represents the volume fraction of the polypropylene shrinkage in concrete, an environment controlled chamber
fiber
was built. It utilized a heater, a dehumidifier, and three
8 inch electric fans to generate a flow of hot and dry air.
Inside the chamber, the temperature was 40-45 degree
Celsius and the relative humidity was 30-35%. At this
condition, the rate of water evaporation in the chamber
was 0.79 kg/m2/hr. Three replicates of the specimens with
the moulds were placed in the chamber and were tested
at a time.

Fig. 1: Polypropylene fiber having the length of 12 mm and the Fig. 3: Specimen base for the plastic shrinkage test (modified
aspect ratio of 375 – called PPM from ASTM C1579)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

168 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Controlling plastic shrinkage cracking in concrete using polypropylene microfiber

Results and Discussion

Plastic Shrinkage Cracks in OPC
In OPC specimens, the influence of w/c ratio (keeping the
amount of water constant) on the plastic shrinkage crack
was examined. From the test results, the crack areas
occurred within the first six hours after casting of each
concrete mix were demonstrated as shown in Figure 7.
During the first one hour of testing, there was no crack
Fig. 4: Temperature and relative humidity control chamber for found in all OPC specimens. In the 0.4 and 0.5 w/c ratio
the plastic shrinkage test OPC mixes, the crack was observed at the second hour
and expanded until the fifth hour. Between 0.4 and 0.5, the
Measurement of Cracks lower the w/c ratio, the higher the crack. But, in the 0.6
Once the test specimens were put in the chamber, the w/c ratio OPC mix, no crack was found throughout the six
crack monitoring was begun. As a consequence of the hours of testing.
restraint system in the mould, cracks in all specimens
were controlled to take place in transverse direction above
the middle prism. By using a microscope camera, the
crack quantification was performed as shown in Figure
5 by measuring the width and length of cracks at every
one hour time for six hours after the casting. As a crack in
concrete is commonly not a perfect straight line and does
not have equal width, the measurement of crack width
and length was approximated as shown in Figure 6. On a
crack line, its length was determined from a displacement
between its ends. If the crack length is less than 10 mm,
the crack width was averaged from the measurement at
three points, which were located equally apart from each
other. If the length is more, the crack width was averaged
from five points. Then, a crack area was computed by Fig. 7: Relation between the measured crack area and the testing
multiplying the crack length with the averaged crack time of the OPC set
width to represent the degree of crack.
Plastic Shrinkage Cracks in FRC
Since the 0.6 w/c ratio OPC mix had no crack, there was no
reason to add any fiber in it. Hence, the 0.6 w/c ratio FRC
mix was discarded. Only were the 0.4 and 0.5 w/c ratio
FRC mixes tested and their results were shown in Figure
8. At the first hour of testing, the plastic shrinkage crack
was not yet occurred in the FRC specimens similar to the
OPC. However, between the first and the second hour, the
crack was largely developed as found on the crack area
measured at the second hour. Subsequent to that the
crack was grown very little with time and was quite steady
at the end of the testing. The figure clearly indicated that
the addition of polypropylene fiber could reduce the plastic
Fig. 5: Process of crack measurement by using the microscope shrinkage crack.
As shown in Figure 9 in which the final crack area at sixth
hour was plotted with respect to the fiber volume fraction,
the crack resistance of the FRC mixes were depended on
the fiber type and the fiber volume fraction. At the same
fiber volume fraction, the PPG fiber was superior to the
PPM. For the same fiber type, the higher the volume
fraction, the better the resistance. It is noted that, among
the FRC mixes, only the FRC mix with the PPG fiber at
Fig. 6: Example of the measurement of the averaged crack width 0.2% volume fraction can completely prevent the plastic
and the crack length shrinkage crack at both 0.4 and 0.5 w/c ratio.

Organised by
India Chapter of American Concrete Institute 169
Session 2 A - Paper 4

Fig. 8: Relation between the crack area and the testing time of the test specimens; w/c = 0.4 (on the left) and w/c = 0.5 (on the right)

Fig. 9: Variation of the crack area after 6 hours of testing with the fiber volume fraction; w/c = 0.4 (on the left) and w/c = 0.5 (on
the right)

Conclusions Acknowledgement
Following the ASTM C1579, the plastic shrinkage crack in The authors would like to thank the Faculty of Engineering,
the ordinary plain concrete (OPC) and the polypropylene King Mongkut’s University of Technology North Bangkok
fiber reinforced concrete (FRC) was evaluated. In the for financial support on this research.
OPC set, the influence of cement content on the crack References
was determined. Then, in the FRC set, the performance 1. Almusallam, A.A., Maslehuddin, M., Abdul-Waris, M., and Khan,
of polypropylene fiber to reduce the crack was examined. M.M., 1998. Effect of mix proportions on plastic shrinkage cracking
of concrete in hot environments. Construction and Building
From the test results discussed above, the conclusion can Materials, 12(6–7):353-358.
be made as follows. 2. Balaguru, P., 1994. Contribution of Fibers to Crack Reduction of
Cement Composites During the Initial and Final Setting Period. ACI
1. In a concrete mix proportion, when the amount of Materials Journal, 91(3):280-288.
water was 200 kg/m3 constant, the increase of cement 3. Banthia, N., Azzabi, M., and Pigeon, M., 1993. Restrained shrinkage
content (in the range of 0.4 to 0.5 w/c ratio) resulted cracking in fiber reinforced cementitious composites. Materials and
Structures, 26(161):405–413.
in the increase of plastic shrinkage crack. In higher
4. Banthia, N. and Gupta, R., 2006. Influence of polypropylene fibre
cement content concrete, the amount of free water was geometry on plastic shrinkage cracking in concrete. Cement and
lower and inadequate to compensate the water loss Concrete Research, 36(7):1263–1267.
from evaporation on the surface. As a consequence, 5. Boghossian, E. and Wegner L.D., 2008. Use of flax fibres to reduce
the concrete was contracted and cracked. plastic shrinkage cracking in concrete. Cement and Concrete
Composites. 30(10):929–937.
2. The polypropylene fibers – either the PPM (12 mm 6. Grzybowski, M. and Shah, S.P., 1990. Shrinkage cracking of fiber
reinforced concrete. ACI Materials Journal, 82(2):138-148.
long with the aspect ratio of 375) or the PPG (19 mm
7. Shah, S.P., Karaguler, M.E., and Sarigaphuti, M., 1992. Effects of
long with the aspect ratio of 1056) – could reduce the shrinkage-reducing admixtures on restrained shrinkage cracking
plastic shrinkage crack. Based on the maximum fiber of concrete. ACI Materials Journal, 89(3):291–295.
dosage at 0.3% by volume used in this study, it was 8. Soroushian, P., Mirza, F., and Alhozaimy, A., 1993. Plastic shrinkage
found that the PPM could not completely inhibit the cracking of polypropylene fiber reinforced concrete. 92(5):553-560.
9. Soroushian, P. and Ravanbakhsh, S., 1998. Control of Plastic
crack while the PPG simply achieved that at the fiber Shrinkage Cracking with Specialty Cellulose Fibers. ACI Materials
dosage of 0.2%. At the shrinkage level of concrete in Journal, 95(4): 429-435.
plastic state, the finer PPG fiber was more effective as 10. Wittmann, F.H., 1976. On the action of capillary pressure in fresh
it yielded higher number of fibers. concrete. Cement and Concrete Research, 6(1):49-56.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

170 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Controlling plastic shrinkage cracking in concrete using polypropylene microfiber

Dr. Manote Sappakittipakorn

Dr. Manote Sappakittipakorn is presently working as an assistant professor in the department of civil
engineering, faculty of engineering, King Mongkut’s University of Technology North Bangkok, Thailand.
He received his PhD degree at the University of British Columbia, Canada under the supervision of Prof.
Nemkumar Banthia in 2010. His research interest includes fiber reinforced cementitious materials,
durability of concrete material, and the recycling of wastes in concrete.

Organised by
India Chapter of American Concrete Institute 171
SESSION 2 B
Session 2 B - Paper 1

Fiber/Textile Reinforced Permanent Formwork

that can Provide Shear Capacity to Concrete Beams
Changli Yu and Christopher KY Leung
Department of Civil and Environmental Engineering, Hong Kong University of Science and Technology

Abstract this observation, Leung and Cao (2010) proposed the

use of permanent formwork made with pseudo-ductile
To enhance the durability of reinforced concrete
cementitious composites (PDCC) for the construction of
structures, permanent formwork has been developed to
concrete members under severe environment. PDCC, also
provide a barrier to water and/or chemical penetration
known as Engineered Cementitious Composites (ECC), is
so steel reinforcement corrosion can be delayed. If the
a class of material designed through micromechanics (Li
permanent formwork can also provide part of the load
and Leung, 1992, Li, 1993, Leung, 1996) to achieve high
carrying capacity, the amount of steel reinforcements can
tensile ductility (up to 5% strain) through the formation of
be reduced or even eliminated in some cases to simplify
closely spaced fine cracks with opening below 50 mm. By
the construction process. In the literature, permanent
controlling crack opening, the permanent formwork can
formwork with embedded fiber reinforced polymer (FRP)
serve as an effective barrier to the penetration of water
to provide flexural resistance to concrete members has
and chemicals to enhance durability.
been developed. This study will focus on permanent
formwork providing shear capacity to concrete beams. The permanent formwork need to exhibit sufficient
The formwork is made with pseudo-ductile cementitious load carrying capacity. During construction, it has to be
composites (PDCC) containing textile reinforcements. able to carry the loading from the fresh concrete with
The design of formwork is first described, followed by an minimal supporting falsework. This aspect was the
experimental investigation on the shear capacity of beam major consideration in earlier work on the development
members. Finite element analysis of the final member is of permanent formwork for concrete slabs (Brameshuber
then conducted. Numerical and experimental results on et al, 2004), where flexural resistance was provided by
shear capacity are found to be in good agreement. geometric design of the formwork and the incorporation of
textile reinforcement. In more recent work (Kim et al, 2008,
Keywords: PDCC, Permanent formwork, Textile, CFRP
Leung and Cao, 2010, Papanicolaou and Papantoniou,
2010), the formwork is also relied upon to carry structural
Introduction loading in the final member. Reinforcements in the
Steel corrosion is a major problem with reinforced form of Glass Fiber Reinforced Polymer (GFRP) bar or
concrete structures around the world. The conventional textile reinforcements have been employed. It is indeed
wisdom to control corrosion is to employ concrete with desirable to have the formwork providing structural load
low water/binder ratio and the addition of pozzolans. carrying capacity, as site construction can become more
Such kind of concrete will exhibit high resistance to efficient by reducing the amount of steel reinforcements,
the penetration of water and chemicals (chloride and or even eliminating them under some situations with light
carbon dioxide in particular) so corrosion initiation can or moderate loading.
be delayed. A limitation of this approach is the fact that
All existing work in the literature on concrete members
concrete structures are allowed to crack under service
made with permanent formwork has focused on the
loading condition. Experimental results (Wang and Shah,
flexural capacity. The main objective of our work is to
1997, Djerbi et al, 2008) indicate that when the crack
develop permanent formwork that can provide shear
opening reaches 0.15mm to 0.20mm, which is allowed
capacity to beam members. The design of permanent
in design codes, the water permeability and chloride
formwork with PDCC and textile reinforcement will
diffusivity would be significantly increased. In other words,
be covered first, followed by detailed description of an
the penetration of water and chloride will be greatly
experimental program to determine the shear capacity
accelerated and long-term durability can no longer be
for beams members made with the formwork. After
guaranteed.
discussing the results and major observations, nonlinear
According to Wang and Shah (1997) and Djerbi et al (2008), finite element analysis will be conducted to calculate the
for cracking to have little effect on transport properties, the shear behavior of the member. Comparison between
crack opening should be kept below 50-60 mm. Following numerical and experimental results will be carried out.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

174 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber/Textile Reinforced Permanent Formwork that can Provide Shear Capacity to Concrete Beams

Materials Before putting the textile into the mold for casting the
PDCC formwork, it was first tightened by stretching and
The formwork in this study is prepared with PDCC and
fixing to a wood frame (Figure 1 (a)). This process ensures
embedded textile reinforcement to provide additional
the proper alignment and separation of the fibers. Epoxy
shear capacity. The matrix of PDCC is made with
was brushed onto the fibers until the fibers were fully
cement, fly ash, water and silica fume in the proportion
wetted (Figure 1(b)). The epoxy coating can protect the
of 0.18:0.8:0.22:0.02. To provide pseudo-ductility to the
fibers from surface damage during handling and facilitates
material, 2% (in volume) of PVA is added. The PVA fiber is
stress transfer to produce a more even stress distribution
12mm in length and 38 mm in diameter. The strength and
among the fibers. The epoxy used for the textile requires
Young’s modulus are 1530MPa and 33GPA, respectively. a relatively long hardening time. After 48 hours of curing
From the previous tests of ours, this PDCC composition in open space, the coated textile was removed from the
shows hardening behaviour in direct tension, with cracking frame and ready to be embedded inside PDCC to make the
strength, ultimate strength and tensile ductility (which is formwork.
the strain at ultimate strength) of 3.5MPa, 5MPa and 3.5%
respectively. To provide additional reinforcement, a hybrid
Beams made with C-Shape PDCC Permanent
textile with CFRP fibers in one direction and GFRP fibers
in an orthogonal direction is employed. The properties Formwork with CFRP/GFRP Textile as Shear
of the CFRP yarn in the textile are shown in Table 1. The Reinforcement
spacing between fiber yarns is around 20mm. As the Preparation and assembling of the formwork
length of PVA fiber in the PDCC is only 12mm, it is easy for
fresh PDCC to flow into the space between fiber yarns to C-shape formwork elements, reinforced with CFRP/GFRP
form a well-compacted formwork. In the formwork, CFRP textile, will be placed on the two sides of a concrete beam
fibers of the textile will be kept in the vertical direction, to provide shear capacity. The idea of the C-shape comes
from the shear strengthening of concrete beams with FRP
which is the direction of steel stirrups. As the inclination of
sheets. To prevent the peeling of FRP from the side of the
shear cracks is usually below 45 degrees to the horizontal
beam, the sheet needs to be bent around the corners of
(except for deep beams), vertical fibers should be more
the beam so it is anchored at the top and bottom. Here,
effective in controlling crack propagation. The GFRP yarns
the upper and lower sides of the C-shape were expected
are then lying in the horizontal direction. In view of the low
to serve as anchors for the middle part that contains the
stiffness of GFRP and the fiber orientation to the shear
textile. To prepare the formwork element, three wood
crack, the contributions of GFRP to the shear resistance
planks, with two hinges in between and lined with a layer
can be neglected. This aspect is indeed verified by finite
of rubber sheet, was employed. The joined planks were
element analysis to be discussed in a later section.
first placed horizontally on the ground for PDCC to be cast
over the middle plank and part of the two side planks (for
forming the top and bottom parts of the C-shape). The
textile was embedded in the PDCC layer on the middle
plank. After first setting, the planks on both sides were
bent up by 90 degrees to form the C-shape. After bending
up the sides, transverse grooves were introduced on the
inner surface of the formwork (Figure 2 (a)). In order to
prevent the upper part of the C-shape formwork from
interfering with the casting of concrete, the bent part
was designed to be intermittent, as shown in Figure 2
(b). (Note: for wider members used in practice, this may
(b) not be necessary.) The intermittent C-shape can be
easily achieved by placing wooden blocks at appropriate
locations of the side planks during casting, to break the
continuity of the PDCC layer that will be bent up.
Before assembling the formwork for concrete casting,
(a)
sufficient curing is crucial because the C-shape element
Fig. 1: Preparation of the textile is relatively thin and can hence be fragile during

Table 1. Properties of the Hybrid Textile along the Direction of CFRP

Type of reinforcement Tensile strength Modulus of Ultimate Theoretical area of Titer of single Density
(MPa) elasticity Strain yarn (mm2) roving (Tex) (g/cm3)
T700S(carbon) 3518 193.3 1.82 0.44 800 1.80

Organised by
India Chapter of American Concrete Institute 175
Session 2 B - Paper 1

handling if not properly cured. After 20 days of curing in bonding, the PDCC should reach the upper surface of the
the laboratory, the full formwork was assembled as in lower legs of the C-shape element, as shown in Figure 3.
Figure 2(b).

Fig. 3. Control beam and beams made with permanent formwork

After initial setting of the fresh PDCC, surface

treatment can be applied on the inner surface of the
bottom part. After 7 days of curing, the PDCC at the
bottom was found to connect the C-shape elements on
(a) Prepared C-shape formwork elements the sides effectively.

Preparation and testing of beam elements

To prepare beam members for testing, fresh concrete
was cast directly into the prepared permanent
formwork shown in Figure 2 (b). As our focus is on the
shear capacity, two high yield steel rebar of 20mm
diameter were added to prevent flexural failure of
the beams. The cross-section details of the beams
are shown in Figure 3. In this set of test, a total of 5
beams were first cast, while an extra beam with purely
PDCC formwork was prepared later to study the shear
contribution of PDCC itself.

The details of all beams are listed in Table 2. The control

beam was made with concrete only, with longitudinal
steel reinforcements but without using the permanent
formwork. BC_1 members are beams with C-shape
formwork containing one layer of textile reinforcement,
while BC_05 members are beams with half layer of
textile reinforcement. Half layer refers to the situation
where a layer of textile is cut into strips and only half
of the strips are placed inside the formwork (as shown
in Figure 3). For both BC_1 and BC_05, two identical
members were tested.

As shown in Figure 4, the height of the beam was

270mm and the shear span was 600mm, so the shear
(b) C-shape formworks before concrete casting span/depth ratio was about 2.2. The load was applied
centrally by a 1000kN hydraulic jack, and a spreader
Fig. 2: Preparation of C-shape permanent formwork
beam was used to distribute the load to two loading
To form a complete formwork, a pair of C-shape elements points placed symmetrically about the middle of the
were placed inside a wooden box with width equal to that beam at 30cm apart (Figure 4). The test was carried
of the beam to be cast. Instead of adding a bottom plate, out under displacement control with a loading rate
which can be bonded to the C-shape elements with epoxy, of 0.4mm/min. All data were collected by a data
fresh PDCC was cast directly into the box to form the acquisition system at one second intervals.
bottom part of the formwork. In order to achieve good

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

176 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber/Textile Reinforced Permanent Formwork that can Provide Shear Capacity to Concrete Beams

Table 2. Details and Test Results for Beams with C-shaped Formwork containing Textile

Type Details Layer of Textile Load Capacity (kN) Total Contribution (kN) Textile Contribution (kN)

Control Only Concrete 0 135 - -

BC0 C-shape Forms 0 160 12.5 -

BC1_1 C-shape Forms 1 208 39 26.5

BC1_2 C-shape Forms 1 205 37 24.5

BC05_1 C-shape Forms 0.5 192.5 29 16.5

BC05_2 C-shape Forms 0.5 188 27 14.5

figure showing the fine multiple cracks is shown in Figure

6. The pictures in the figure are taken at a high testing load
of about 60% of the ultimate load. Under serviceability
condition, the damage condition should be much less
severe. With the control of cracks by the formwork,
durability during service can be enhanced.

Fig. 4: Testing configuration for members with C-shape form- When loading approaches the ultimate state, the single
work crack of the control beam kept opening up and propagating
until final failure. For the other beams, final failure is also
Test results and observations accompanied by the formation of localized shear cracks
All the specimens in this series showed typical shear (as shown in Figure 7).
failure with inclined cracks running along the shear spans.
There was no sign for the formwork to separate from the
concrete, despite the fact that concrete was cast after the
formwork had been cured for a number of days (which
would be the situation in a real application where the
formwork is pre-fabricated). The top and bottom parts of
the C-shape are hence shown to be effective in anchoring
the middle part so the formwork and cast concrete can
work together to carry the applied loading.
The failure of the control beam made with plain concrete
(without PDCC formwork) is shown in Figure 5. Only a few
shear cracks are formed and one of them opened up and
propagated quickly to result in final failure. For members
with the C-shape formwork, multiple cracks can be seen
on the surface of the members (Figure 6). A magnified
Fig. 6: Fine multiple cracks on the surface of the beam
The load vs. displacement curves for all tested members
are shown in Figure 8 and the ultimate load values are
also given in Table 2. According to the results, the shear
capacity can be significantly enhanced by the C-shape
formwork elements. Comparing the member with pure
PDCC formwork (no textile) and the control plain concrete
member, the increase in shear capacity is about 18.5%.
For the formwork with half a layer and one layer of textile,
the increase, relative to the control, is 40% and 53%
respectively.
To see if the permanent formwork can replace steel
stirrups, it is informative to calculate the equivalent
volume of steel reinforcements that can provide the same
Fig. 5: Crack for the control beam shear resistance. To provide a given shear capacity (Vs),

Organised by
India Chapter of American Concrete Institute 177
Session 2 B - Paper 1

Fig. 7: Failure modes of members with C-shape formwork: (a) control with PDCC alone, (b) with half layer of textile, (c) with one
layer of textile.

the area of steel ( A sv) is given by: Figure 9. Note that all the elements are made transparent
A to clearly reveal the interface between formwork and
t sv = bssv ........................................................................(1) concrete (in brown color). In the model, the x-direction
is taken to be along the length of the beam, while the
where s is the spacing between the stirrups and ho the y-direction is along the width. The z-direction is then the
effective depth of the beam. After obtaining A sv/s, the direction of loading. In view of symmetry, it is sufficient to
percentage of steel reinforcement rsv is given by: model a quarter of the whole beam, with the cutting surface
A
Vs = 1.25fyv ssv h o ..........................................................(2) X (perpendicular to x-direction) fixed in x-direction and
free in y- and z-directions, and surface Y (perpendicular
where b is the width of the member. When there is one to y- direction) fixed in the y-direction and free in x- and
layer of textile in the PDCC formwork, rsv is calculated to z-directions. These boundary conditions are illustrated in
be 0.57%. According to this result, for applications where Figure 9. In the model, only one horizontal strip is placed
failure is dominated by flexure so only nominal shear around the upper corner of the beam (see Figure 9). In the
reinforcements (with r of 0.5% or below) is required, test specimen (Figure 2), the upper part of the C-shape
the permanent formwork can provide sufficient shear is made intermittent to facilitate the casting of concrete.
capacity and no steel stirrups will be required. An exact model should therefore also have a number
of horizontal strips spaced along the upper edge of the
beam. Initial numerical simulations have indeed been
carried out to compare the cases with multiple horizontal
strips and a single strip, but no difference was observed.
Subsequent simulations were therefore carried out with
the single strip configuration.
Also shown in Figure 9 are two blocks for load application
and support of the beam, respectively. The support is kept
fixed in the z-direction, but free in the x- and y- directions.
Loading is applied through prescribed displacement, with
a rate of 0.1mm for each step.

Fig. 8: Load vs displacement curves for the various members

Modeling of Shear Capacity provided by

Permanent Formwork
Details of the model
Boundary conditions
The ATENA program is employed to perform nonlinear
Finite Element Analysis. The FEM model is shown in Fig. 9: Boundary conditions of the model

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

178 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber/Textile Reinforced Permanent Formwork that can Provide Shear Capacity to Concrete Beams

Assignment of material properties up the C-shape has a length of 200mm along the
longitudinal direction. The width of the blocks for support
The assignment of material properties to different parts
and load application (along the longitudinal direction) is
of the model is shown in Figure 10. Elements inside the
100mm. A typical mesh is shown in Figure 11. To model
load application and support blocks were assigned elastic
the members made with C-shape formwork, 4 node
material properties with modulus similar to the materials
tetrahedral elements are employed for both the PDCC and
under contact, so the stress concentration at the support/
the concrete. The size of the mesh inside concrete is taken
loading positions can be minimized. Note that by using the
to be 20mm where that for the PDCC formwork is taken
modulus of concrete (the material under contact, rather
to be 10mm.
than that of steel (the material making up the block), the
stress distributions under the support and loading blocks
may not be accurately captured. However, as failure is
not governed by stresses under the blocks, there should
be little effect on the results. Steel reinforcements were
represented by truss elements connected to the concrete
elements with bond-slip behavior following the CEB-FIP
model. For the elements inside the formwork, a strain
hardening material model developed by Kabele (2002)
was used. At the surface between formwork and concrete,
perfect bonding was assumed. This is because no
debonding between the PDCC formwork and cast concrete
was observed in the experiments. Instead of modelling Fig.11: Meshing of the model
the CFRP/GFRP fiber reinforcements with discrete truss
elements, they are smeared into a continuum with the Results and analysis
same stiffness, which is bonded perfectly to the PDCC. To ensure that the material parameters adopted for
To model the smeared reinforcement, besides knowing concrete, steel and the bond-slip relation between steel
the material properties and fiber angles, the reinforcing and concrete are all properly chosen, the behavior of
ratio is also required. According to measurements on the the control beam (without PDCC formwork) was first
CFRP/GFRP textile, the thickness and width of the fiber simulated by the FEM and compared to experimental
bundle was around 2mm when the spacing between results obtained in the laboratory. The result is shown
the bundles is 20mm. For the PDCC formwork with a in Figure 12. As shown in the figure, the load capacity
thickness of 20mm, the reinforcing ratio for one layer of from both experimental and numerical results is similar,
textile can be calculated to be around 1%, while that for but the mid-span displacement shows some difference.
half layer of textile is around 0.5%. Part of the difference is related to the way displacement
was measured in the experiment. As the holder of the
Simulations have been carried out with only the CFRP
LVDT was fixed on the loading system rather than the
fibers and with both kinds of fibers smeared into the
beam member, the measured displacement included
PDCC elements, and including the GFRP fibers would only
deformation of the supporting fixtures and is hence higher
change the results by less than 2%. In the final numerical
than the actual value.
results, the GFRP fibers are therefore not considered.

Fig. 10: Material assignment in the model

The dimensions of the member follow the ones in our Fig. 12: Comparison of results for the control beam prepared
experimental investigation. The horizontal strip making with concrete only

Organised by
India Chapter of American Concrete Institute 179
Session 2 B - Paper 1

beams made with the C-shape textile/PDCC permanent

formwork.

Fig. 13: Comparison of results for beams with plain PDCC

formwork

For the beam member with formwork made with PDCC

alone, the results are shown in Figure 13 and good
agreement between the experimental and simulated load
capacities can be obtained. As in the case for the control
beam, there is a difference in the mid-span displacement.
The difference is around 1.5mm, which is within the same
range for the case of the control beam.
Simulation was then performed for beams with formwork
made using both PDCC and textile reinforcements. The
results are shown in Figure 14 (a) and (b) for the cases with
half layer of textile and one layer of textile respectively. For
both cases, the results show a larger deflection than the
numerical results, and the difference is within the same
range as those for the control member and the member
with PDCC formwork. On the other hand, the simulated
shear capacity is in good agreement with test result.
To have an overall comparison of the results, the Fig. 14: Comparison of results for beams with textile reinforced
experimental curves for all cases shown in Figure 8 can PDCC formwork: (a) with half layer of textile, (b) with one layer
be compared to the numerical curves shown in Figure 15. of textile.
From the test results, the contribution from half a layer of
textile reinforcement is over half of that from a full layer.
Comparing with numerical results, the beam with half
a layer of textile reinforcement also exhibited a higher
shear capacity in the tests. A plausible explanation is as
follows. As described in the experimental part, the textile
was cut into strips and put inside the formwork. As shown
in Figure 16, it is then possible for a certain main shear
crack to interact with more than half of the reinforcement.
In other words, the actual effective reinforcement ratio in
the test can be above 0.5%, which is the average value for
the PDCC formwork with half a layer of textile.
For quantitative comparison of the experimental and
simulated shear capacity, the values for all members
are given in Table 3. Good agreement between the values
indicates that the finite element analysis can provide a
reasonably accurate prediction of the load capacity for Fig. 15: Contributions provided by the embedded textile

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

180 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fiber/Textile Reinforced Permanent Formwork that can Provide Shear Capacity to Concrete Beams

Table 3. Summary of the results the computed and measured shear capacity. The finite
Experiment Model Difference
element method is hence applicable to the prediction
(kN) (kN) (%) of shear capacity and can be employed for the practical
Concrete 135 138 2.2 design of beams made with textile reinforced PDCC
formwork on the sides.
Only PDCC in formwork 160 164 2.5
Half layer of textile 190 183 3.6
Acknowledgement
Full layer of textile 208 212 1.9
Financial support of this research by the Hong Kong
Research Grant Council through and GRF 615508 is
gratefully acknowledged.
References
1. Brameshuber, W., Koster, M., Hegger, J., Voss, S., Gries, T., Barle, M.
Reinhardt, H.W. and Krugger, M. (2004) Textile Reinforced Concrete
(TRC) for Integrated Formworks, In Thin Reinforced Cement-Based
Products. ACI-SP 224, pp.45-54.
2. Djerbi, A., Bonnet, S., Khelidj, A. and Baroghel-bouny, V. (2008)
Influence of Travsering Crack on Chloride Diffusion into Concrete,
Cement and Concrete Research, Vol.38, No.6, pp.877-883.
3. Kabele, P. (2002) Equivalent Continuum Model of Multiple Cracking,
Engineering Mechanics, Vol.9, No.1/2, pp.75-90.
4. Kim, G.B., Pilakoutas, K., Waldron, P. (2008) Development of thin
FRP Reinforced GFRC Permanent Formwork Systems, Construction
and Building Materials, Vol.22, 2250-2259
5. Li, V.C. and Leung, C.K.Y. (1992) Steady-State and Multiple Cracking
of Short Random Fiber Composites, ASCE Journal of Engineering
Fig. 16: Possible textile distribution along the main shear crack Mechanics, Vol.118, No.11, pp.2246-2264.
6. Li, V.C. (2003) On Engineered Cementitious Composites (ECC): A
Conclusions Review of the Material and its Applications, Journal of Advanced
Concrete Technology, Vol.1, No.3, pp. 215-230.
The use of C-shape formwork element on the side of
beams is an effective and convenient approach to enhance 7. Leung, C.K.Y. (1996) Design Criteria for Pseudo-ductile Fiber
Composites, ASCE Journal of Engineering Mechanics, Vol.122, No.1,
shear resistance while preventing debonding failure. pp.10-18.
Under situations where only nominal shear reinforcement
8. Leung, C.K.Y. and Cao, Q. (2010) Development of Pseudo-ductile
is required, the use of textile/PDCC formwork on the Permanent Formwork for Durable Concrete Structures, RILEM
sides of the beam can potentially eliminate steel stirrups. Materials and Structures, Vol.43, No.7, pp. 993-1007
The PDCC/textile formwork is proved to be effective in 9. Papanicolaou, C. and Papantoniou, I. (2010) “Mechanical Behavior of
controlling surface cracks, thus beneficial to long-term Textile Reinforced Concrete (TRC) / Concrete Composite Elements”,
durability. Experimental results on the shear behavior Journal of Advanced Concrete Technology (ACT), 8(1), pp. 35-47.
of beams with textile/PDCC formwork on the sides 10. Wang, K.J., Jansen, D.C., Shah, S.P. and Karr, A. (1997) Permeability
were compared to numerical results from FEM models. Study of Cracked Concrete, Cement and Concrete Research, Vol.27,
No.3, pp.381-393.
Good agreement (within 4%) can be achieved between

Dr. Christopher K. Y. Leung

Dr. Christopher K. Y. Leung (PhD, MIT, 1990) is a Professor at the Hong Kong University of Science and
Technology, where he headed the Civil and Environmental Engineering Department from July 2009 to June
2015. His research interests are in the general area of construction materials, with particular focus on
composite mechanics, fracture mechanics, optical fiber sensing and the application of composites in civil
engineering. He has received research-related awards from ASCE, the China Ministry of Education and
China State Department. He served as the Honorary President of RILEM in 2011 and is elected Fellow of 6
international institutions/associations.

Organised by
India Chapter of American Concrete Institute 181
Session 2 B - Paper 2

Fire Resistance of Fibre Reinforced Concrete Beam

Piti Sukontasukkul Sunisa Sukchoo Manote Suppakittipakorn

Professor, Department of Civil Graduate Student, Department of Civil Assistant Professor, Department of Civil
Engineering, King Mongkut’s University Engineering, King Mongkut’s University Engineering, King Mongkut’s University
of Technology North Bangkok of Technology North Bangkok of Technology North Bangkok

ABSTRACT both compressive and flexural strength of both plain and

1% SFRC subjected to high temperature from 105°C to
In this study, the flexural behavior of fibre reinforced
1200°C. However, the SFRC appeared to perform better
concrete beams subjected to gradient heat was
than plain concrete as seen by higher residual strength.
investigated. Three types of concrete beams were tested:
Poon[6] carried out the experiment on SFRC and PPFRC
plain concrete (OPC), polypropylene fibre reinforced
samples subjected to temperature ranging from 200°C to
concrete (PPFRC) and steel fibre reinforced concrete
800°C. At 200°C, the compressive strength of both plain
(SFRC). The specimens in form of beam with dimension
and FRC remained unchanged. The strength was found
of 100x100x350 mm. were cast and cured for 28 days
to decrease linearly with the increase temperature above
in water and 7 days in air. After curing, the specimens
200°C. In the case of compression toughness, SFRC was
were subjected to the fire on one side (at bottom surface)
found to maintain its energy absorption better than plain
to create gradient heat across the section for 15, 30, 45
concrete even at the highest temperature. While, in the
and 60 minutes. After cooling down, the specimens were
case of PFRC, the compression toughness decreased
then tested under 4 point loading. Results are discussed
more rapidly and became lower than that of plain concrete
in terms of cross-section investigation and flexural
at the highest temperature.
response.
Recently, in our previous study[7], the effect of high
Key Words: Fire, Gradient Heat, Flexural Response, Fibre
temperature on the post-peak response of FRC was
Reinforced Concrete.
investigated. Results indicated that under flexural load,
fiber type and content, and temperature level played an
Introduction important role. At the temperature of 400oC, all FRCs
Concrete structures occasionally may be subjected to exhibited higher flexural strength, better post-peak
heat from fire accident. However, the pattern of exposure response and toughness. Significantly drop in strength,
on each structural member may be different depending toughness and load-deflection response can be observed
on its location. In the case where members are located in in the case of synthetic or plastic FRC (PFRC) at the
the middle of the fire, a fully exposed pattern (all side) can temperature above 600oC. But for the steel FRC (SFRC),
be expected. But in the case where the members are not drop of strength, toughness and load response was
located in the middle of the fire incident, partial exposure relatively small. This showed that the SFRC had better
(only on one or two sides) is a dominant pattern. This thermal resistance than the PFRC.
partial exposure to heat causes a so called ‘non-uniform This study is a continuing part of the previous study [],
or gradient heat’ through the thickness of a concrete however, it focuses mainly on the effect of gradient heat
member which may somehow affect the ability of concrete on the flexural performance of plain and FRC. In this
to resist certain types of load. Therefore, it becomes the experiment, three types of concrete (plain concrete and
objective of this study which is to investigate the effect of FRC (steel and polypropylene)) were cast in form of beams
gradient heat on flexural properties of concrete and fibre and directly exposed to fire at one side (bottom) at four
reinforced concrete. different durations: 15, 30, 45 and 60 minutes. The burned
Several studies on the effect of high temperature on specimens were then tested for flexural performance to
plain concrete indicated several physical changes include study the effect of temperature and time of exposure.
cracking and spalling, the debonding of cement paste and
aggregates, and the weakening of the hardened cement Experimental Procedure
paste[1]. The compressive strength was also found to
Materials used in this study consist of Portland cement
change differently depending the level of temperature[2-4].
Type I, river sand, crush rock and fibres. Two types of
For fibre reinforced concrete (FRC), several studies have fibres are used: Hooked end steel fibre and Crimped
been carried out. Lau et al[5] reported the decrease in polypropylene fibre (Table 1 and Fig.1).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

182 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fire Resistance of Fibre Reinforced Concrete Beam

Table 1
Mix proportions and characterization at fresh and hardened states

SA LWA-P LWA-V HRWR Flow

Mix OPC-I CAC-E SF LF w/c w/p t50 fc fct
(0-4) (0-1.5) (0-4) TSG30 spread

Kg/m3 mm s Mpa Mpa

M1 500 - 45 2150 - - - 5.4 0.40 0.26 610 6.0 86.6 4.7

M2 400 - - 47 669 13.2 42.4 6.5 0.40 0.36 570 6.5 22.5 1.9

M3 - 400 - 47 669 13.2 42.4 6.5 0.43 0.38 580 6.4 22.3 2.1

Fig. 1: Polypropylene and Steel Fibre

The mix proportion for plain concrete was set at 1 : 2.4 :

2.9 : 0.37 (Cement : Fine : Coarse : Water). For FRC, the
volume fraction was set at 1%. Concretes were mixed
using pan mixer and cast in form of beam with dimension
Fig. 3: Temperature Development of the Furnace
of 100x100x350 mm. The specimens were cured in water
and air for 28 and 7 days, respectively. Prior to the test, the
specimens were burned at the bottom side with burning
period of 15, 30, 45 and 60 minutes using a high pressure
open furnace constructed at the Department of Civil
Engineering, KMUTNB (Fig. 2).
The furnace is constructed of four steel partitions with an
opening at the top. The base consists of four high pressure
gas-heads with separate controlling valve. The furnace is
tested and calibrated in the lab to obtain a temperature-
time reference curve as shown in Fig. 3.
The flexural performance test was carried out in according
to ASTM C1609 (Fig.4). Two LVDTs were set at both sides
of the beam. Data in terms of load and displacement were
measured and recorded using a data acquisition system
Fig. 4: Flexural Performance Test according to ASTM C1609
and computer notebook.
Experimental Results
Cross-section Investigation
After the tests, all the specimens were cut in half along
the crack line to investigate any physical changes that may
happen to the specimen. Results are shown in Fig. 5 to 7.
For plain concrete, a few changes in the heated zone can be
observed. For example, the change in color, as the burning
time increased the grayish cement like color began to fade
Fig. 2: Four-Gas Head Open Furnace and Test Set Up away and replaced with a burning (brownish/blackish)

Organised by
India Chapter of American Concrete Institute 183
Session 2 B - Paper 2

(a)

(b)

Fig. 7: Temperature vs. Time along the Beam Thickness

Fig. 5: Cross-section of Plain Concrete Specimen after

Subjected to Fire

shade. The burning shade can be seen clearly at the

bottom portion closed to the fire. Another thing that can
be observed was the increasing porosity at the surface
with time (Fig.5b).
In the case of polypropylene FRC, beside from the change
in color which was similar to that of plain concrete, the
disappearance (evaporation) of fibre at the heated zone
was also observed (Fig. 6). Theoretically, an evaporation
point of polypropylene fibre is about 190oC. Since our
experiment was set up to create gradient heat along the Fig. 8: Cross-section of SFRC Specimen after Subjected to Fire
thickness of beam, so the temperature was highest at
the bottom (closest to the fire source), decreased along or exceed 190°C. This “fibre evaporation zone” appeared to
the thickness (as it moves away from the fire source) and moving up along the thickness of the beam with increasing
lowest at the top of beam. The heat distribution along the burning period.
beam thickness at different locations is shown in Fig.7. The SFRC also exhibited similar affect as that of plain
Based on this phenomenon, the PP fibres were found to concrete. Both color change and increasing porosity were
evaporate at the location where the temperature reached found in SFRC as shown in Fig. 8. However, in the case of
SFRC, there was no fibre evaporation occurred because
the temperature was even not high enough to effect the
steel fibres. The only change in the steel fibre that can be
observe is change in color from metallic to dark charcoal
color.

Flexural Behavior
The flexural responses of plain concrete beams are as
illustrated in Fig. 9. The flexural strength of the beam was
found to decrease with the increasing burning time. The
decrease in strength is perhaps due to the degradation
of bond between the paste and aggregates. As the
temperature increases the mismatch between the thermal
expansion coefficients of both paste and aggregate begins
to play an important role on the internal expansion. This
non uniform expansion creates internal stresses and also
Fig. 6: Cross-section of PFRC Specimen after Subjected to Fire leads to internal cracking and debonding.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

184 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Fire Resistance of Fibre Reinforced Concrete Beam

Fig. 9: Flexural Responses of Plain Concrete Beams after Fig. 10: Flexural Responses of PFRC Beams after subjected
subjected to Fire to Fire

toughness to increase. Our test results also confirmed

the above explanations, as seen by the decrease peak
strength and the increase post-peak response with the
burning time (Fig.11)

Conclusion
Gradient heat or high temperature affects the flexural
performance of plain and FRC in similar manner. The
reduction of strength is commonly found in all type of
concrete. However, in the case of FRC, the post-peak
response becomes an important factor that separates
both types of fibre. For steel fibre, the reduction in
post-peak response is found due to the degradation of
bond between fibre and paste. However, in the case of
Fig. 10: Flexural Responses of SFRC Beams after subjected polypropylene, the mechanism of melting and evaporation
to Fire causes a combination of increasing post-peak response
and decreasing peak strength.
In the case of SFRC, the responses of the SFRC beams
(both pre- and post-peak) worsened with the increase of References
burning time as seen in Fig. 10. The reduction of the flexural 1. Georgali, B., Tsakiridis, P.E., “Microstructure of Fire-Damaged
responses is perhaps the combined effect between the Concrete. A Case Study”, Cement & Concrete Composites No. 27,
degradation of concrete strength (as explained previously) 2005, pp. 255–259
and degradation of bond strength between fibre and paste. 2. Handoo, S.K., Agarwal. S., Agarwal, S.K., “Physicochemical,
Mineralogical, and Morphological Characteristics of Concrete
The mismatch between the thermal expansion of steel
Exposed to Elevated Temperatures”, Cement and Concrete
fibre and paste also plays an important role. Research, No. 32, 2002, pp. 1009–1018
In the case of PFRC, the mechanism becomes more 3. Powers-Couche, L., “Fire Damaged Concrete-Up Close”, Concrete
complex because polypropylene fibres melt when Repair Digest, 1992, pp. 241–8.
exposed to temperature lower than 190oC and evaporate 4. Gustafero, A.H., “Experiences from Evaluating Fire-Damaged
at the temperature higher than 190oC. For any PFRC Concrete Structures––Fire Safety of Concrete Structures,”
American Concrete Institute SP-80, 1983.
beam when subjected to gradient heat, the fibres far away
5. Lau, A., and Anson, M, “Effect of High Temperatures on High
from the heat source remain unaffected, while the fibres
Performance Steel Fibre Reinforced Concrete,” Cement and
closer to the heat source either melt or evaporate. The Concrete Research, No.36, 2006, pp 1698–1707
zone where fibres are melted, the bond between fibres 6. Poon, C.S, Shui, Z., and Lam, L., “Compressive Behavior of Fiber
and paste improved (increase toughness). However, in Reinforced High- Performance Concrete subjected to Elevated
the zone where fibres are evaporated and left behind the Temperatures,” Cement and Concrete Research, No. 34, 2004,
voids, this causes the strength to decrease. Both melting pp. 2215–2222.
and evaporation zones affect the flexural response as 7. Sukontasukkul, P., and Pomchiengpin, W., Post-Crack (or Post-
follow: the fibre evaporation causes the pre-peak strength Peak) Flexural Response and Toughness of Fiber Reinforced
Concrete after Exposure to High Temperature, Construction and
to decrease while the fibre melting causes the post-peak Building Materials (JCBM), Vol 24, 2010, pp. 1967-1974.

Organised by
India Chapter of American Concrete Institute 185
Session 2 B - Paper 2

Dr. Piti Sukontasukkul, PhD

Dr. Piti Sukontasukkul is professor in the Department of Civil Engineering at the King Mongkut University of
Technology, Bangkok. Prior to his doctorate in 2011 from University of British Columbia, Vancouver, Canada,
he did his Graduation and Masters in Civil Engineering from institutes based in Thailand. He is in teaching
profession for more than 20 years and is presently Professor in the Department of Civil Engineering, King
Mongkut’s University of Technology in Bangkok, Thailand. He is serving on many national and international
committees.
He has many awards and scholarships to his credit and has made numerous presentations at the
international conferences in a number of countries world over. He has published more than 50 papers in
journals and conferences.
His research interests cover topics like Behavior of Concrete under Elevated Temperature, Fiber reinforced
concrete (FRC), Impact behavior of concrete, Polymer concrete, Lightweight concrete, Rubberized Concrete,
Sustainability of Concrete Materials.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

186 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Enhancement of substrate-repair bond strength and durability in concrete structures

Enhancement of substrate-repair bond strength

and durability in concrete structures
C. Zanotti
Department of Civil Engineering, University of British Columbia, Vancouver (BC) Canada

Abstract The root of the problem lies in the complexity of repaired

systems, composed of different parts -the old concrete,
While there is much information available on substrate-
the new repair, and the interface in between- with
repair bond strength, most of the data obtained through
different physical, chemical, mechanical properties and
standard bond tests are of limited applicability and
different responses to mechanical forces and various
practicality as inherently bound to the stress condition
applied and affected by disturbed stress paths. chemo-physical attacks. The nature and effects of those
Furthermore, little is known about substrate-repair interactions, although acknowledged, are still poorly
interface fracture, crack propagation, and damage understood. For instance, substrate-repair interface
resistance. However, such information can be very useful behavior is commonly characterized by standardized
to prevent interfacial damage, cracking, and debonding bond tests. However, most of the data obtained through
induced by loading, shrinkage, thermal gradients, standard bond tests routinely adopted such as pull-off,
and severe environmental conditions. By increasing twist-off, and slant shear test are of limited applicability
robustness and durability of the substrate-repair and practicality as intrinsically bound to the stress
interface, one can enhance the overall durability of the condition applied. Those tests have also been proven to
repair. have a limited reliability, as the stresses applied at the
substrate-repair interface can be highly non uniform
The application of Poly-Vinyl-Alcohol (PVA) fiber and are often affected by disturbed stress paths (ASTM
reinforcement to a cement-based repair is investigated 2005, ASTM 2013, Austin et al. 2009, Momayez et al.
in this study as a possible effective strategy to enhance 2005, Zanotti et al. 2014c). Although those results offer an
substrate-repair bond strength and fracture resistance. assessment of bond quality, they are far from exhaustively
Three fiber volume fractions are considered, namely: 0%, characterizing the interface.
0.5%, and 1%. Substrate-repair bond tests for various
combinations of normal and shear stresses along the bond Furthermore, little is known about substrate-repair
plane are carried out, supported by splitting (indirect) interface fracture, crack propagation, and damage
tension bond test. The substrate-repair bond strength resistance. Yet, such information can be very useful since a
domain of each repair material is determined and the thin region of weak material forms between substrate and
correlated interfacial cohesion and friction coefficients repair, the Interfacial Transition Zone (ITZ), similar to the
are extrapolated. The Contoured Double Cantilever Beam one forming between aggregate and cement paste, whose
(CDCB) test is performed to assess Mode-I fracture porosity, micro-cracking, and microstructural changes
properties of the substrate-repair interface, based on under stresses affect the vulnerability of the repaired
a modified Linear Elastic Fracture Mechanics Approach structure to corrosion, deterioration, and mechanical
(LEFM). Bond strength and fracture tests exhibit coherent damage and, thus, the overall durability of the repair.
results since PVA fibers have a beneficial effect on both In particular, interfacial toughness and crack growth
substrate-repair interface cohesive strength and interface resistance affect energy and displacement capacity,
crack growth resistance. The observed enhancement of failure mode, response under loading during service life,
the interface quality is analyzed and discussed. and resistance to shrinkage and thermal cracking.

Keywords: Repair; Bond; Cohesion; Fracture; Durability; Substrate-repair bond optimization in terms of robustness
Fiber. and durability is therefore instrumental to the attainment
of durable effective repairs. Although substrate-repair
interfaces coated with bonding agents have exhibited
Introduction satisfactory bond strength, their performance in the
Our society is in need for expansive structural and long term and under some extreme conditions remains
nonstructural repair of infrastructure, buildings, and of concern. Concrete-concrete bond, conversely, can be
historical heritage. Unforeseen failures and rapid considered more reliable in the long run. As we are moving
deterioration demonstrated that such interventions are towards the application of specialized concretes with
often nor as effective as required neither as durable. high fracture toughness and strain-hardening abilities,

Organised by
India Chapter of American Concrete Institute 187
Session 2 B - Paper 3

we have a unique opportunity to critically optimize the Table 2. Properties of PVA fibers
substrate-repair interface that is resistant to cracking PVA fibers Property Value
and damage. Specific gravity 1.3
The effect of the addition of Poly-Vinyl-Alcohol (PVA) fibers Tensile strength [N/mm ]2
1600
to a cement-based repair mortar on its adhesion to the Average length [mm] 8
concrete substrate is assessed in this study. Various fiber Diameter [µm] 40
volume fractions are considered, namely: 0%, 0.5%, and Young’s Modulus [GPa] 40
1%. Slant shear tests with various bond plane inclinations
Elongation [%] 7
are performed to characterize bond under various stress
conditions and determine inherent interfacial parameters
(i.e., cohesion and friction) that can be adopted to predict - The plain concrete halves of the specimen are cast and
bond under various combinations of normal and shear covered with polyethylene sheets soon afterwards.
stresses. The specimen geometry is optimized to minimize Proper forms are adopted to ensure that only half
the effect of disturbed stress regions on the bond plane. of the cylinder is cast. This technique is preferred
Mode-I (opening mode) substrate-repair interface over casting full cylinders and obtaining two halves
fracture properties are assessed through Contoured from each by saw cutting, since cutting would create
Double Cantilever Beam (CDCB) test. Slant shear bond damage in the concrete substrate and jeopardize bond
test and Mode-I fracture tests are supported by splitting at a the microscopic level.
(indirect) tensile test. The effect of the application of
PVA fiber reinforcement to the repair on the mechanical - Samples are demolded 24 hours later and cured in
properties of the substrate-repair interface is analyzed lime-saturated water for 27 days.
and discussed. - Samples are removed from the water and allowed to
dry for 24 hours.
Materials - Contact surfaces are sandblasted to achieve a
Mixture proportions of both substrate and repair are listed satisfactory level of roughness.
in Table 1. The substrate is a plain concrete and the repair
- The concrete halves are cleaned and inserted in
material is a mortar reinforced with various amounts of
the molds where the new mortar is cast. Molds are
Poly-Vinyl-Alcohol (PVA) fibers. The fiber volume fraction
oiled before the concrete substrate is inserted back
(Vf) in the repair is equal to 0% (plain mortar), 0.5%, or
in. Extreme care is taken to avoid contamination of
1%. PVA fibers are selected in this study for their ability
the bond surface. The cylinders are covered with
to develop a chemical bond with the cement paste leading
polyethylene sheets soon after casting.
to a significant increase of fracture toughness and strain
capacity (Kanda and Li 1998, Li 2003). PVA has a relatively - Specimens were demolded 24 hours later and cured in
simple chemical structure with a pendant hydroxyl group lime-saturated water for an additional 27 days.
and is produced by the polymerization of vinyl acetate first
to Poly-Vinyl-Acetate (PVAc), followed by hydrolysis of Substrate-repair shear bond test
PVAc to PVA. The properties of the PVA fibers employed Shear bond is assessed through a modified configuration
are listed in Table 2. In order to promote compatibility, of the standard slant shear test (ASTM 2005). In addition
compositions of substrate and repair are identical, except to the standard bond plane inclination angle (α) of 30
for the addition of coarse aggregate in the substrate and
degrees, two additional angles of descent, α = 25° and α =
fibers in some of the repairs.
20°, are adopted (Figure 1). During the test, a compressive
load is applied to the specimen edges, resulting in a
Table 1. Mixture proportions combination of shear and normal stresses along the
Material Cement Fly Sand Coarse Water Vf inclined bond plane (Figure 2). Each bond angle, α,
ash aggregate
generates a different shear-normal stress ratio along the
Substrate 1 0.25 2 0.48 0.5 - bond plane and, thus, a different shear bond strength is
Repair 1 0.25 2 - 0.5 0%, 0.5%, obtained for each value of α. Since bond strength may be
1%
affected by the area of bond contact over which the load
gets transmitted, the diameter of the cylinders is varied
Methods in order to keep a similar area of bond contact. Cylindrical
rather than prismatic geometry is selected to avoid stress
Specimen preparation concentration in the corner regions. Friction effects
A minimum of four replicates is prepared for each between loading plate and specimen edge can restrain
material and test. The specimen preparation consists of the lateral specimen expansion and induce disturbed
the following steps: stress paths in the specimen. In order to avoid that the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

188 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Enhancement of substrate-repair bond strength and durability in concrete structures

bond plane is affected by those disturbed stress regions,

the distance between the specimen loaded edges and
the bond plane is designed equal to the diameter of the
sample.

Fig. 3: Cylinders undergoing tension splitting: (a) test set-up, (b)

initial crack formation, and (c) failure

(a) (b) (c) of change of the compliance function (introduced in the

following Chapter) is theoretically constant during the
Fig. 1: Specimen geometries for shear bond test with different crack growth, leading to a strain release rate independent
bond plane inclinations from the crack length and, thus, to a more stable crack
growth (Mostovy et al. 1967). The CDCB specimen is
also called linear compliance change specimen. The
CDCB geometry adopted in this work is adapted from
the one developed by Genois (1995) to test concretes
and mortars. In order to test the interface between two
different materials, the specimen is subdivided into two
different parts, with the concrete substrate on one side,
the repair material on the other side, and the bond plane
in between, running exactly along the CDCB middle
section (Figure 4b). A side groove along the desired crack
plane promotes a proper crack growth to avoid any crack
deviation (Figure 4c).

Fig. 2: Normal (σn) and shear (τn) stress configuration along the
bond plane in a slant shear test

Substrate-repair tension bond test

Splitting (indirect) tension test is performed for an
estimation of substrate-repair tensile bond strength
(ASTM 2011). The feasibility of applying the splitting tension
test, originally developed for monolithic concrete, to testing
of substrate-repair bond has been proven (Geissert et al.
1999, Momayez et al. 2005). Cylinders with a length of 250 Fig. 4: (a) CDCB test apparatus, (b) specimen with crack propa-
mm and a transverse diameter of 100 mm are adopted. gating along the substrate-repair interface, (c) CDCB geometry,
Two opposite line loads are applied perpendicular to the (d) crack propagation and crack parameters, and (e) forces
cylinder axis (Figure 3a), resulting in a state of tensile transferred to the sample
stresses over the interface, which is located along the
diametrical plane covering the load lines. The test is performed with an Instron universal testing
machine. The specimen is placed on a hinge fixed to
Mode-I substrate-repair crack growth test the lower plate of the testing machine, while the upper
Mode-I crack growth resistance is assessed through part is supported by a steel profile with two identical
the Contoured Double Cantilever Beam (CDCB) test, wedges (Figure 4b). Each wedge is positioned between
which consists of propagating a crack in a pre-notched bearings mounted on both sides of the CDCB sample
specimen while recording the load applied and the opening with steel rods. The steel profile is connected to a load
displacement of the crack faces (Figure 4a). The tapered cell connected to the cross head of the machine. The
shape of the specimen was developed as an improvement vertical displacement of the cross head results in the
of the original Double Cantilever Beam (DCB) geometry, steel wedges pushing against the four bearings with the
and was designed by imposing the condition that the rate system of forces shown in Figure 4e. Since the angle of the

Organised by
India Chapter of American Concrete Institute 189
Session 2 B - Paper 3

wedge, α = 15°, is small, the vertical component of the load α, lower is the normal stress component, lower is the
can be ignored. Low friction needle bearings are adopted friction contribution against debonding, and thus lower
so that the coefficient of friction, µ, between wedge and is the shear bond strength. Experimental data fit a linear
bearings is negligible and the effect of residual vertical plot that represents the shear bond failure domain and
forces is avoided. The resulting horizontal Splitting Load can be adopted to predict bond strength under various
(SL) applied to the specimen is shown in Figure 4e. The combinations of normal and shear stresses. A linear
Crack Mouth Opening Displacement (CMOD) is recorded failure criterion is significant in the range of compressive-
by using a clip-gauge displacement transducer (fixed at shear stress ratios applied in these experiments and for
the level where the opening load is applied, Figure 4d) and lower values of normal stresses, including the case of
a digital data acquisition system running at 5 Hz. Initially, pure shear (i.e., intersection with the vertical axis) and
a standard vertical cross-arm rate of 0.1 mm-1 is adopted. the case of combined tensile and shear stresses. The
Eventually, the rate is reduced up to 0.025 mm-1 to ensure failure domain presents a tension cut-off -such as the
quasi-static crack propagation also within the most brittle one shown in Figure 5d for repair mortar with 1% PVA
specimens (plain concrete substrate, and substrate- fiber volume fraction- that coincides with the bond tensile
repair interface). strength and needs to be assessed through tensile bond
test. For compressive-shear stress ratios higher than the
ones applied in this study (i.e., for α>35°-40°), the linear
Results and Discussion
approach does not apply and the failure envelope curve is
Shear bond strength nonlinear, presenting a decreasing gradient for increasing
normal-shear stress ratio (Robins and Austin 1995).
Debonding failure is observed for all the slant shear
specimens (Figure 5a). For each repair material, three The linear failure plot is described by the Mohr-Coulomb
different substrate-repair shear bond strength values criterion as follows:
are obtained, in relation to the three different bond plane x n = c + v n tan z ............................................................(1)
inclinations, α, considered, which generate three different
shear-normal stress combinations along the bond plane Where c is the adhesion strength (pure shear bond
(Figures 1,2). Shear bond strength, τn, and associated strength), or cohesion coefficient, and φ is the angle of
applied normal stress, σn, are calculated as per Figure friction. The two parameters, c and φ, are computed
2, and their average values are plotted in the shear- from polynomial fitting of three average data points at
normal stress interaction diagrams shown in Figure different slants for each repair material, that is, for each
5b-d. The statistical analysis of the results is available fiber volume fraction. Cohesion and friction are plotted
elsewhere (Zanotti et al. 2014a). As expected, lower is in Figure 6 as function of fiber content; the statistical

Fig. 5: (a) Typical repair bond failure and shear-normal stress interaction diagram for (b) 0%, (c) 0.5% and (d) 1% PVA fiber rein-
forcement in the repair

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

190 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Enhancement of substrate-repair bond strength and durability in concrete structures

confidence of the trends obtained is demonstrated

(Zanotti et al. 2014a). Cohesion and friction are inherent
interfacial factors, characterizing the interface regardless
of the stress condition applied. Notice that cohesion
substantially increases as the fiber content in the repair
material is increased (Figure 6a). This encouraging
result indicates that the quality of the bond is enhanced
by the addition of fibers. The ability of PVA fibers to
reduce shrinkage strains and consequent damage is
well established (Kanda and Li 1998, Li 2003) and, very
likely, this increase in cohesion occurred due to a reduced (a) (b)
shrinkage cracking in the Interfacial Transition Zone (ITZ),
Fig. 6: (a) Cohesion and (b) angle of internal friction for various
which is expected to reduce the size of the unsupported fiber volume fractions
cracks and in turn increase the area of contact along the
interface. Another potential reason is the reduction in and cutting operations can be avoided as they may lead
the bleeding at the interface due to fiber reinforcement. to substrate-repair interface damage. Moreover, different
Bleeding would increase the water-cement ratio at the casting directions can be adopted. On the other hand,
interface and produce a more porous interface in the the test may lead to strength overestimation due to the
case of plain mortar, thereby reducing cohesion; previous “indirect” nature of tensile stresses. The indirect tensile
studies have cited these reasons for bond enhancements strength, fct, is calculated as follows (ASTM 2011, Geissert
in repair (Banthia and Dubeau 1994, Banthia and Yan et al. 1999, Momayez et al. 2005):
2000). Furthermore, the noted increase in cohesion may
also be attributed to the ability of the fibers to reduce 2P
fct = rA ........................................................................(2)
operational damage at the interface. Beyond the effects of
the enhanced ITZ robustness, increased cohesion is also where P is the peak load and A is the area of the bond
the result of micro-crack kinking within the two materials plane.
and especially in the repair material, leading to increased
Results are summarized in Table 3 for substrate, repair,
toughness and failure resistance. Crack kinking is a well-
and substrate-repair interface for the case of 1% PVA fiber
known mechanism (He and Hutchinson 1989, Lim and Li
reinforcement in the repair. As expected, average tensile
1997, van der Giessen and Needleman 2003), enhanced
strengths of plain substrate and FRC repair are similar.
under a mixed-mode fracture such as the one of slant
Plain concrete exhibits a brittle failure mode occurring
shear test. This resisting mechanism is further analysed
along the middle section of the cylinder. The FRC failure
and discussed in the following section titled “Mode-I
mode is softer than the one of plain concrete; once the
substrate-repair interface crack growth resistance”.
maximum load is reached, an unstable loss of bearing
A slight decrease in the angle of internal friction is capacity is recorded but the sample is not yet split up. A
observed (less than 6%) due to fiber reinforcement (Figure re-loading stage follows until the cylinder is fully broken;
6b). These variations appear to be a second order effect the peak stress of the re-loading stage (failure stress) is
and can be considered a consequence of minor differences around 2/3 the tensile strength.
in the condition of the substrate surface at the time of
Failure occurs along the bond plane for all the concrete-
casting, cleanliness, soundness, and moisture content.
FRC specimens. Macro-crack formation is accompanied
However, this trend may also be a consequence of the
by a loss of bearing capacity. A visible crack around the
increase in cohesion. As is well known, bond failure can
interface is observed at that stage but the two parts of
occur as ideal pure frictional failure, ideal pure cohesive
the sample are still adhering to each other (Figure 3b).
failure, or mixed frictional-cohesive failure. If cohesion is
After re-loading, a second peak load higher than the first
enhanced, bond failure is delayed and fracture involves
one is reached when the specimen splits up (Figure 3c).
more the adhesive failure of the interfacial transition zone
The average first cracking stress is 1.64 MPa and failure
or the repair material and less the pure frictional interlock
strength is 3.44 MPa (twice greater than the first cracking
between repair and substrate, since the two materials
stress). The concrete-FRC bond strength is lower than
adhere better to each other.
the tensile strength of the two materials. However, the
interface is able to resist further loading in open loop
Splitting (indirect) tensile strength
after first cracking. Such behaviour can be due to a stress
Splitting tests are performed for estimating the indirect redistribution along the edge of the bond plane, promoted
tensile strength of plain concrete, FRC, and plain concrete- by the presence of PVA fibers in the repair mortar and by
FRC interface. In comparison to the pull-off test (ASTM the development of a tortuous crack that deviates from the
2013), which is more commonly employed for tension roughened interface into substrate and FRC repair. These
bond test, the splitting test has the advantage that drilling aspects are further analysed in the following section

Organised by
India Chapter of American Concrete Institute 191
Session 2 B - Paper 3

regarding crack propagation in tension. equations that apply to LEFM of perfectly brittle materials,
and the toughening effect of the Fracture Process Zone
Shear-normal stress interaction diagrams obtained from
(FPZ), i.e., the large zone of damaged material in the highly
slant shear test can be updated with splitting tensile bond
stressed region surrounding the crack tip, is represented
strength values, which are located on the tension cut-off
by fictitiously increasing the length of the crack. The
of the bond strength domain (Figure 5d).
resulting crack length is called effective crack length, aeff.
Table 3. Splitting (indirect) tensile strength [MPa] According to the LEFM approach, the fracture instability
Plain FRC FRC Interface Interface of a brittle material stressed in Mode-I (the opening
concrete peak second first failure mode), can be completely defined by means of one only
stress peak cracking strength parameter, either the critical stress intensity factor (KIC) or
4.29 4.24 2.71 1.64 3.44 the specific fracture energy (Gf). According to the energy
concept proposed by Griffith (1920), unstable propagation
Mode-I substrate-repair interface crack growth of a crack is deemed to occur when the rate of change in
resistance the elastic energy released by the system for a unit crack
Contoured Double Cantilever Beam (CDCB) tests are extension (U), is equal or greater than the energy required
performed on the plain concrete substrate, the repair for the crack growth of the material (W), for a unit crack
reinforced with 1% volume fraction of PVA fiber (FRC), extension:
and the concrete-FRC interface. The desired propagation 2U 2W ...................................................(3)
of a vertical crack along the middle section is observed G IC = 2a = 2a = R
(Figure 4b).
where a = crack length, GIC = Mode-I critical strain
Experimental Splitting Load (SL) vs Crack Mouth Opening energy release (or else, specific fracture energy, Gf),
Displacement (CMOD) curves of concrete-FRC interface and R = crack growth (or fracture) resistance. The stress
are shown in Figure 7a, while SL-CMOD curves of concrete, intensity factor, KI, characterizes the intensity of the
FRC, and concrete-FRC interface are plotted in Figure 7b. elastic stress field ahead of the crack and is sufficient
The three curves in Figure 7b are selected as the most by itself to completely define the entire stress field in the
representative of the average behavior of each species. vicinity of the crack (Figure 4d). When sufficient energy
Interestingly, concrete-FRC interface exhibits a higher is supplied to the system, unstable crack propagation
splitting load bearing capacity than the one of monolithic occurs; at this stage, KI approaches a critical value, KIC.
plain concrete. Typical crack-growth-resistance effective crack length
SL-CMOD curves are analysed to calculate fracture curves (R-curves) of brittle and quasi-brittle materials
properties of materials and interfaces. In this study, are displayed in Figure 8a. Crack growth resistance is
a modified Linear Elastic Fracture Mechanics (LEFM) not a function of crack extension in brittle materials and
approach based on the effective crack model is adopted. R = GIC (LEFM). In quasi-brittle mediums, toughening
The quasi-brittle behaviour of the materials and the mechanisms and stable FPZ development occur during
interfaces investigated can be represented by the the initial crack growth (pre-critical or sub-critical crack

Fig. 7: (a) Splitting Load (SL) vs. Crack Mouth Opening Displacement (CMOD) curves of concrete-FRC interfaces and (b) representa-
tive SL-CMOD curves of concrete, FRC, and concrete-FRC interface

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

192 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Enhancement of substrate-repair bond strength and durability in concrete structures

growth), represented by a rising branch prior to unstable simple cantilever beam made of perfectly elastic material.
propagation. Thorough information on fracture of quasi With reference to the strength-of-materials approach,
brittle materials, FPZ modelling, nonlinear fracture theoretical compliance can be calculated as follows:
6Q 1 + v V
mechanics, and modified LEFM approaches is available a a

in the literature (Bažant and Cedolin 1979, Hilleborg 1980, 2 3 CMOD 24 # x2 # H1 dx ....(5)
C th = SL = SL = EB dx + EB
H3
de Borst 1984, Rots 1988, Rots and Blaauwendraad 1989, 0 0

Bažant 1992, Reinhardt and Cornelissen 1994, Zanotti et where ∆ = arm displacement, E = Young’s Modulus, ν =
al. 2014b).
Poisson’s coefficient, B = beam width (that is, specimen
Crack growth resistance analysis can be addressed by depth), H = beam height (specimen width), and a = crack
different methods. The procedure followed in this work is length (Figure 4c,d). The beam height is variable due to
based on the determination of the stress intensity at the the specimen tapered shape; however, a constant height
crack tip through the use of experimental and theoretical equivalent to the mean height of the corresponding
compliances (direct method, Visalvanich and Naaman, contoured cantilever beam (Hc) is assumed:
1981) and is implemented with the software MatLab. The H0 + Ha ma
Hc = = H0 + 2 ............................................(6
compliance (C) is given by the ratio between the elastic 2
component of the Crack Mouth Opening Displacement
(CMODe) and the Splitting Load (SL): where Ho is the width of the top part of the tapered beam,
Ha is the beam height at distance a from the loading
CMOD e .................................................................(4)
C= SL point, and m is the contour profile (Figure 4d). Equation
5 is written by assuming that the end of the cantilever
Cement-based quasi-brittle materials typically show beam is immobile; the actual rotation of the bottom edge
a SL-CMOD response curve similar to the one sketched can be roughly accounted for by fictitiously increasing
in Figure 8b, which, indeed, is consistent with the the crack length in the calculation of the compliance
experimental results shown in Figure 7. An initial linear term due to bending deformation. Mostovoy et al. (1962)
elastic branch (A-BOP), characterized by a constant initial experimentally determined that the rotational effect was
value of the compliance (Cexp,i), is followed by a nonlinear equivalent to considering a crack length of about (a +
part of stable sub-critical crack propagation (BOP-B) 0.6Hc). By assuming ν = 0.2, the following expression is
prior to instability (point B). The first point of nonlinearity obtained:
24 !a + 0.6H c (a)$
is named Bend Over Point (BOP). The calculation of the
C th (a) = EB G
0 . 3a J
3
experimental compliance can be simplified by ignoring + ....................(7)
permanent deformations during the nonlinear crack 3H 3c (a) H c ( a)
growth, i.e., assuming a return to the origin upon unloading
The effective modulus of elasticity (Eeff) can be derived
(CMODe = CMOD, Figure 8b). Therefore, the experimental
by equating the initial experimental compliance to the
values of the compliance, Cexp, for each loading stage can
initial theoretical compliance. The initial experimental
be calculated from the experimentally recorded values of
compliance (Cexp,i) is obtained as the experimental
CMOD and SL.
compliance at the first point of nonlinearity (BOP). The
By assuming a LEFM behavior, one can consider that each crack length before BOP, ao, is equal to the length of the
specimen arm at one of the two crack sides behaves as a notch and, thus, the initial theoretical compliance (Cth,i) is

Fig. 8: (a) Typical crack-growth-resistance curves of brittle and quasi-brittle materials, (b) typical response of cement-based mate-
rial undergoing CDCB test

Organised by
India Chapter of American Concrete Institute 193
Session 2 B - Paper 3

given by: splitting load, SLBOP = 0.4 kN, for both plain concrete
C th,i = C th (a 0) .................................................................(8) and concrete-FRC interface, while SLBOP is higher (= 1
kN), for the FRC (Figure 7b). Accordingly, higher stress
Once the effective modulus of elasticity is known, the intensity at the crack tip is obtained in the FRC (Figure
effective crack length (aeff) can be calculated by equating 9b). The substrate-FRC interface in FRC repairs depicts
the experimental compliance to the theoretical compliance a subcritical crack growth pattern similar to the one in
at each step of the nonlinear crack growth (after BOP, monolithic plain concrete samples, but instability occurs
Figure 8b). The corresponding Mode-I Stress Intensity earlier in the plain concrete (Figure 9b), which withstands
Factor (KI) for a CDCB specimen is given by: a lower peak Splitting Load (Figure 7b). Compared to the
K 2I = h 2 SL2 B -2 H -a 3 Q a 2 + 1.4aH a + 0.5H 2a V ......................(9) monolithic plain concrete samples, the concrete-FRC
interface endures significant subcritical crack growth at
where η is a function of the slope (m) and is equal to 3.1 in greater stress intensity and fails at higher critical stress
this case (m = 0.222). intensity. In other words, an overall amelioration of the
It must be noted that one B parameter in B-2 term holds interfacial cohesive fracture behaviour is observed.
for the beam width while the other B represents the crack This can be explained with an increase of the damage
width (b). Since the specimen adopted contains a side- tolerance of the substrate-repair Fracture Process
groove along the middle plane, one of the B has to be Zone. As a matter of fact, the macro-crack propagation
changed for the cracked width b (Figure 4c): in the stable stage is accompanied by the development

K 2I = h 2 SL2 Q Bb V-1 H -a 3 Q a 2 + 1.4aH a + 0.5H 2a V ...............(10)

of very thin micro-cracks, as experimentally observed
in the FRC, near to the concrete-FRC interfacial macro-
Crack growth resistance curves, in terms of stress crack. Thus, one can conclude that the enhanced fracture
intensity factor (KI) plotted as function of the effective resistance of the repair material as in FRC can positively
crack length (aeff), are shown in Figure 9a for concrete- affect the fracture properties of the interface. In this
FRC interface. The curves are computed based on particular case, the enhanced toughness of the interface is
the experimental results shown in Figure 7a, and the promoted under the following conditions: (i) a mechanical
analytical procedure described above. The responses compatibility between the two materials in term of similar
obtained from the different concrete-FRC samples are elastic stiffness and first cracking strengths; (ii) enhanced
in good agreement and crack growth resistance curves interface roughness, which promotes the tortuosity of the
have similar trends. Also, coherent values of KI at the crack path and causes directional deviations; and (iii) the
onset of the stable crack growth are obtained, as well ability of the PVA fibers added to cause stress redistribution
as coherent KI values at the initiation of the unstable around the main crack, create crack blunting as well as
stage. Representative crack-growth resistance curves develop a chemical bond with the cementitious matrix
for monolithic concrete, monolithic FRC, and concrete- (Zanotti et al. 2014b). Interestingly, the enhanced interface
FRC interface, computed from the SL-CMOD curves bond toughness observed in PVA FRC repairs under CDCB
shown in Figure 7b, are plotted in Figure 9b. The initial test is coherent with the enhanced cohesiveness observed
elastic stiffness of the three series is similar and satisfies for the same repair material undergoing modified slant
mechanical compatibility requirements as desired. The shear test (see Section above on “Bond shear strength”).
Bend Over Point (BOP) is reached upon the same average It is well know that the observed micro-crack deviation, or

Fig. 9: (a) Crack growth resistance curves (in term of stress intensity factors, KI, vs. effective crack length, aeff) of concrete-FRC
interface and (b) representative K I - aeff curves of concrete, FRC, and concrete-FRC interface

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

194 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Enhancement of substrate-repair bond strength and durability in concrete structures

kinking, has a beneficial effect on the fracture properties surfaces and the bond strength or tensile strength of concrete
repair and overlay materials by direct tension (Pull-off Method).
of interfaces in bi-material systems. However, most of
ASTM C1583/C1583M
the theories available in the field of concrete to describe
interfacial failures in layered or composite materials, 4. Austin, S., Robins, P., Pan, Y., 1999. Shear Bond Testing of Concrete
Repairs. Cement and Concrete Research, 29(7):1067–1076.
and possible crack or micro-crack deviation, are mostly
restricted to elastic fracture theories, while more insights 5. Banthia, N., Dubeau, S., 1994. Carbon and Steel Micro-Fiber
Reinforced Cement Based Composites for Thin Repairs. ASCE J.
on the role of plastic dissipation in setting the apparent of Materials in Civil Eng., 6(1): 88-99.
toughness of interface cracks are available in the field
6. Banthia, N., Yan, C., 2000. High Performance Micro-Fiber Reinforced
of metal-ceramic interfaces (He et al. 1991, Wei and
Concrete for Thin Repairs. ACI Special Publication on High
Hutchinson 1999, Van der Giessen and Needleman 2003). Performance Materials for Repairs (Eds. Krstulovic-Opara et al),
In the light of these experimental results, it is suggested ACI SP-185:69-80.
that future research work on substrate-repair interface 7. Geissert, D.G., Li, S., Frantz, G.C., Stephens, J.E., 1999. Splitting
optimization should be based on non-elastic fracture prism test method to evaluate concrete-to-concrete bond strength.
approaches. ACI Materials Journal, 96(3):359-366.
8. Bažant, Z.P., 1992. Fracture mechanics of concrete structures.
Proceedings of the First International Conference (FraMCoS-1),
Conclusions Breckenridge, Colorado, Elsevier, London, 1040 pp.
An experimental study is carried out to assess the effect 9. Bažant, N., Cedolin, L., 1979. Blunt crack band propagation in finite
of the addition of Poly-Vinyl-Alcohol (PVA) fibers to a element analysis. Journal of Engineering Mechanics Division, ASCE,
cementitious repair material on its bond to the existing 105 (EM2) Proc., Paper 14529, 297–315.
substrate. The substrate is sandblasted first and a repair 10. de Borst, R., 1984. Application of advanced solution techniques to
mortar with similar tensile elastic stiffness and first concrete cracking and non-associated plasticity. Numerical Methods
cracking strength is adopted for repair; no bonding agent for Non-linear Problems, Taylor, C. et al. eds, v. 2, Pineridge Press,
Swansea, UK, 314–325.
is employed. The main results can be summarized as
follows: 11. Geissert, D.G., Li, S., Frantz, G.C., Stephens, J.E. (1999). Splitting
Prism Test Method to Evaluate Concrete-To-Concrete Bond
1. The addition of PVA fibers to the repair improves the Strength. ACI Materials Journal, 96(3):359-366.
strength of its bond to the substrate. In particular, 12. Griffith, A.A., 1920. The phenomena of rupture and flow in solids.
the cohesive component of bond is substantially Philosophical Transactions of the Royal Society of London, Series
increased. As expected, only second order variations A, 221:163–198.
of the friction component are observed, since friction 13. He, M.Y., Hutchinson, J.W.,1989. Kinking of a crack out of an interface.
is mainly affected by mechanical interlock. Journal of Applied Mechanics, 56:270-278.

2. PVA fibers also improve the substrate-repair 14. Hilleborg, A., 1980. Analysis of fracture by means of the fictitious
crack model, particularly for Fiber Reinforced Concrete.
interface Mode-I crack growth resistance. Compared International Journal of Cement Composites, 2(4):177–184.
to monolithic concrete, the concrete-FRC interface
15. Kanda, T., Li, V.C., 1998. Interface property and apparent strength of
endures significant subcritical crack growth at greater
a high strength hydrophilic fiber in cement matrix. ASCE J. Materials
stress intensity and fails at higher critical stress in Civil Engineering, 10(1):5-13.
intensity.
16. Li, V.C., 2003. On engineered cementitious composites – A review
3. The enhanced substrate-repair bond strength and of the material and its applications. Journal of Advanced Concrete
Technology, 1(3):215-230.
toughness observed in repairs reinforced with PVA
fibers is a result of enhanced robustness of the ITZ 17. Lim, Y.M., Li, V.C., 1997. Durable repair of aged infrastructures
and enhanced damage tolerance of the FPZ. These using trapping mechanism of Engineered Cementitious Composites.
Cement and Concrete Composites 19:373-385
results are encouraging as they set the ground for
bond optimization under any stress condition and for 18. Momayez, A., Ehsani, M.R., Ramezanianpour A.A., Rajaie H., 2005.
Comparison of methods for evaluating bond strength between
the design of durable substrate-repair interfaces. concrete substrate and repair materials. Cement and Concrete
Further developments of the research are including Research, 35 (4):748–757.
the experimental investigation of the effect of different
19. Momayez, A., Ehsani, M.R., Ramezanianpour A.A., Rajaie H., 2005.
types of fibers and admixtures, microscopic analysis of Comparison of methods for evaluating bond strength between
the interface, and interface fracture test under mixed concrete substrate and repair materials. Cement and Concrete
modes. Research 35:748-757.
References 20. Mostovy, S., Crosley, P.B. Ripling E.J., 1967. Use of crack-line-loaded
1. ASTM, 2005. Standard test method for bond strength of epoxy-resin specimens for measuring plane strain fracture toughness, Journal
systems used with concrete by slant shear. ASTM C 882/C 882M - of Materials, 2 (3):661–681.
05e1. 21. Reinhardt, H., Cornelissen, H.A.W., 1994. Post-peak cyclic behavior of
2. ASTM, 2011. Standard test method for splitting tensile strength of concrete in uniaxial tensile and alternating tensile and compressive
cylindrical concrete specimens. ASTM C496/C496M. loading. Cement and Concrete Research, 14(2):263–270.

3. ASTM, 2013. Standard test method for tensile strength of concrete 22. Rots, J.G., 1988. Computational modeling of concrete fracture.

Organised by
India Chapter of American Concrete Institute 195
Session 2 B - Paper 3

Ph.D. thesis, Civil Engineering and Geosciences, Delft University accompanied by plastic dissipation at multiple scales, International
of Technology, Delft, The Netherlands. Journal of Fracture, 95(1–4):1–17.

23. Rots, J.C., Blaauwendraad, J., 1989. Crack models for concrete: 28. Zanotti, C., Banthia, N., Plizzari, G., 2014a. A study of some factors
Discrete or smeared? Fixed, multi-directional or rotating. HERON, affecting bond in cementitious fiber reinforced repairs. Cement and
34(1):1–59. Concrete Research, 63:117-126.
24. Robins, P.J., Austin, S.A., 1995. A Unified Failure Envelope from the 29. Zanotti, C., Banthia, N., Plizzari, G., 2014b. Towards sustainable
Evaluation of Concrete Repair Bond Tests. Magazine of Concrete repairs: Substrate-repair interface Mode-I fracture analysis.
Research 47(170):57-68. International Journal of Sustainable Materials and Structural
25. Van der Giessen, E., Needleman, A., 2003. Dislocation plasticity Systems, 1(3):265-281.
effect on interfacial fracture. Interface Science 11:291-301.
30. Zanotti, C., Talukdar, S., Banthia, N., 2014c. A State-of-the-Art on
26. Visalvanich, K., Naaman, A.E., 1981. Fracture methods in cement Concrete Repairs and Some Thoughts on Ways to Achieve Durability
composites. Journal of the Engineering Mechanics Division, ASCE in Repairs. Infrastructure Corrosion and Durability - A Sustainability
Proc., 107(EM6):1155–1171. Study, ed. Yang Lu, OMICS Group EBook, www.esciencecentral.org/
27. Wei, Y., Hutchinson, J.W., 1999. Models of interface separation ebooks.

Dr. Cristina Zanotti

Dr. Zanotti works at the Department of Civil Engineering of the University of British Columbia (Canada),
where she researches “Structural & Non-structural Repair and Rehabilitation”, with special focus
on “Durability” and “Carbon footprint”. Cristina holds a PhD degree in “Structural Rehabilitation and
Maintenance of Ancient and Modern Buildings” from the University of Brescia (Italy). Her research objectives
include the development of effective, durable, and sustainable repair and rehabilitation techniques for both
modern infrastructure and Historical Heritage (with focus on bond, compatibility, and carbon footprint
optimization). Her research interests also include Fracture Mechanics of cementitious materials and
interfaces, mitigation of shrinkage and thermal cracking, and phased nonlinear structural and heat
transfer analysis techniques. Through the IC-IMPACT center of excellence, she is currently involved in
a joint project between Canada and India for the development of “Sustainable Ultra-thin Rural Roads in
Canada and India”. Dr. Zanotti is a member of the International Concrete Repair Institute, the American
Concrete Institute, and is associate member of the ACI Committee 546 - Repair of Concrete.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

196 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical and Durability properties of recycled aggregate concrete made with different replacement levels of recycled coarse aggregate

Mechanical and Durability properties of recycled aggregate concrete

(RAC) made with different replacement levels of recycled coarse
aggregate (RCA)
Sumaiya Binte Huda and M. Shahria Alam
School of Engineering, The University of British Columbia, Kelowna, BC, Canada

Abstract help reducing the environmental impacts of concrete

production, researchers and industries are continuously
The purpose of this study is to investigate the mechanical
investigating to find new ways of producing concrete.
and durability properties of recycled concrete made
Recycled aggregate concrete (RAC) as a coarse aggregate
with recycled coarse aggregate (RCA) obtained from
has potentials to become a trade mark for construction
demolished concrete. While having a control mix as a
projects to achieve LEED (Leadership in Energy and
reference, six different RCA replacement levels (e.g.
Environmental Design) points (USGBC, 2014).
0, 30, 40, 50, 75, and 100%) were used to compare the
performance of recycled aggregate concrete (RAC). The Moreover, before 1970s durability of concrete was
physical properties of RCA and the compressive strength occasionally considered as a design criteria. At that time
of different mixes were analyzed to evaluate the different government officials, design professionals, developers,
variations due to the application of RCA. The durability and builders rarely considered durability property in a
properties of RAC were examined under chloride and construction project. Later in 1980, “concrete cancer”
sulphate exposure conditions. The test results show that became a popular phrase in the society and media. This
recycled concrete made with 30% RCA’s mechanical and influenced peoples to grab their attention towards the
durability performance is quite comparable to control durability performance of concrete. Durability of concrete
mix. This finding will boost up the confidence level of structures exposed to aggressive or harsh environments,
construction industry regarding the use of RCAs, and can be significantly affected. This can reduce the life span
will pave the way towards eco-friendly and sustainable of civil infrastructures, and consequently increase the
construction material. maintenance cost. The use of RCA in concrete may make it
more susceptible to degradation as these aggregates are
Keywords: Recycled aggregate concrete · Durability ·
more porous in nature than the natural coarse aggregate
Chloride exposure · Sulphate exposure · Wet-dry cycle.
concrete.
Introduction The reuse of demolished concrete in the commercial
Engineers and researchers are always striving and application is limited due to the absence of proper
exploring different ideas to utilize various industrial guideline like standards along with lack of knowledge and
wastes to produce concrete for construction. Determining expertise. Concerning this issue, a number of important
studies were carried out by various researchers (Gomes
the characteristics and behavior of these different types
and di Brito, 2009, Huda and Alam 2014). Smith (2009)
of concrete has become an important research stream
revealed that due to the variation in sources, recycled
in order to utilize them in mix design (Alam et al. 2013).
concrete aggregate may possess impurities along with
Compressive strength is one of the most important
the adhered mortar content. This significantly influences
characteristics of concrete that dictates its durability.
the properties of RAC, and makes it difficult to predict the
Due to its worldwide availability, comparatively low cost,
properties of new concrete. RCA has a higher absorption
and ability to take any form and shape, concrete has
capacity than natural coarse aggregate due to the attached
emerged as the most widely used construction material
mortar where 3.2% to 12% range of water absorption is
all over the world. According to Cement Association of
seen in the case of fine and coarse recycled aggregates
Canada (2012), every year 15 billion tonnes of concrete
(Catz 2003). Due to the higher absorption capacity of
is produced throughout the world. This widely accepted
recycled aggregate, the concrete mixes become stiffer and
construction material, however, has some disadvantages
less workable compared to natural aggregate concrete
such as greenhouse gas emissions during the production,
(NAC) (Salem et al. 2003).
consumption of limited natural resources. The overall
production process of concrete contributes approximately Gomez-Soberon (2002) explained that replacing natural
5% of the greenhouse gas (GHG) emissions produced aggregates (NAs) with RCAs yielded an increase in
each year (Concrete Association of Canada (2012). To porosity. The tensile and compressive strengths of RAC

Organised by
India Chapter of American Concrete Institute 197
Session 2B - Paper 4

Table 1.Mix proportions

Mix component Mix-1 Mix-2 Mix-3 Mix-4 Mix-5 Mix-6
(Control) (30% RCA) (40% RCA) (50% RCA) (75% RCA) (100% RCA)
Cementing GU cement 208 208 208 208 208 208
materials (kg)
Fly ash 52 52 52 52 52 52
Fine aggregate Natural 807.6 807.6 807.6 807.6 807.6 807.6
(kg) aggregate
Coarse NA 1040.7 728.5 624.4 520.4 260.2 0
aggregate (kg)
RA 0 312.2 416.3 520.4 780.5 1040.7
Water (kg) 150 150 150 150 150 150
Water reducer Glenium 3030 468 ml 468 ml 468 ml 468 ml 468 ml 468 ml
Air entraining Micro air 120 ml 120 ml 120 ml 120 ml 120 ml 120 ml
admixture

decrease with increased porosity. Several researchers adjustment of total amount of water was done depending
(Yang et al. 2008, Huda and Alam 2015) found that similar/ on the absorption capacity and moisture content of RCA.
comparable strength can be achieved by concrete made Coarse aggregate, natural sand, cementitious materials,
with RCA. However, Ulloa et al. (2013) suggested that the water, Glenium 3030NS, and Micro Air were used to
compressive strength of RAC is greatly influenced by the produce different concrete mixes. The concrete mixes also
recycled aggregate replacement ratio and the effective utilized a 20% cement replacement with Fly Ash (Class F)
water cement (w/c) ratio. that helps lower the CO2 embodied in the concrete. The use
of fly ash also aids in the sulphate and chloride resistance
RAC with variable percentages of RCA (0%, 50% and
by forming a tighter concrete matrix, thereby reducing the
100%) were studied by Olorunsogo and Padayachee
permeability and rate of chemical infiltration.
(2002). They found that if the RCA replacement level is
increased, the durability property of RAC decreases. RCA source plays a vital role for achieving desired
The quality of RAC can be enhanced with the curing age. properties of concrete. The source of parent concrete was
Shayan and Xu (2003) found that RCA concrete’s sulphate unknown, which is similar to the case in many recycling
resistance was more than NAC, and after one year plants. This unknown source of RCA is beneficial for this
exposure, the related expansion was less than 0.025%. study to represent practical situations. As presented in
Different types of opinions can be found in the literature Table 1, six different concrete mixes were designed with
regarding the chloride resistance of RAC. Shayan and Xu varying levels of RCA replacement. The RCA content
(2003) obtained almost 2.2 to 2.3 mm higher penetration used to replace a portion of the natural coarse aggregate
depth for RAC compared to NAC after exposed to chloride varies from 30%-100 % with a 0% RCA replacement as the
solution. Gomes and de Brito (2009) also came up with control mix (Mix-1). In many codes 20% is the maximum
similar type of findings. But Hansen and Hedegkd (1984) allowable limit for the use of recycled aggregate. Thus,
obtained 0.69% soluble chloride ion by weight of cement the choice of replacement levels in this study will boost
in concrete made with recycled concrete aggregate which up the confidence level of ready mix industry. All other
was higher than ACI acceptable limit. mix components remained constant to ensure that the
test results only reflect the effect of changing the RCA
The main objective of this study is to investigate the proportions.
mechanical and durability characteristics of 25MPa
concrete made with different replacement levels of RCA, Assessment of Fresh and Hardened Properties of RAC
with an aim to escalate the commercial production of this
To investigate the mechanical property of concrete,
new concrete and its application. Durability properties of
compressive strength test was done according to CSA
RAC are assessed in terms of sulphate attack and cyclic
A23.2-9C (CSA 2009). At least three specimens were cast
wetting and drying along with chloride exposure.
to do the compressive strength testing on the specified
date. The test specimens (100 x 200 mm cylinders) were
Experimental Program cured inside a moisture monitored curing chamber, and
The mix design used in this study for the production of 25 were taken out from the curing chamber, and dried just
MPa concrete was provided by local ready-mix company, before the testing on the specified dates. Moreover, fresh
where effective w/c ratio was 0.56. Table 1 shows the mix concrete properties were investigated by performing
proportions for the various mixes that were tested for slump and air content test during casting. Fresh concrete
this study. Same effective w/c ratio was used throughout samples for slump and air content tests were taken from
the production of different concrete mixes where only the same batch to maintain the consistency of results of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

198 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical and Durability properties of recycled aggregate concrete made with different replacement levels of recycled coarse aggregate

fresh and hardened concrete properties. chamber before being placed in the sulphate bath. Moist
curing was done for 7 days to replicate the field condition.
Test Method to Assess The Sulphate Resistance of RAC Before being placed in the sulphate bath, the diameter and
RCA itself can work as a potential source of sulphate height of each cylinder were measured. Cylinders from
attack since these aggregates are usually collected each mix were broken under compression to measure the
from landfill where they can easily get contaminated. 7th day strength of that mix. Sulphate bath was prepared
Moreover, groundwater, industrial effluents, sea water, one day before the use with 5% sodium sulphate (Na2SO4)
soil, and decaying organic matters are potential sources solution, and stored at 23 ± 2oC. In the storage container,
of sulphate and affect the performance of concrete. The the ratio of the volume of sulphate solution to the volume
cement paste’s composition and permeability are key of concrete cylinder was 4± 0.5. The pH of sulphate
factors in terms of resisting sulphate ions ingression solution was always maintained around 7 which is close
into concrete. Once the dissolved sulphate ions enter into to pH (7.2) of typical field condition. Sulfuric acid (H2SO4)
concrete, sulphate ions attack in the form of chemical was added into the sulphate bath to maintain the pH of
reaction which causes strength degradation along with sodium sulphate solution. During the testing period pH
expansion, cracking, and spalling (Monteiro et al. 2000). was monitored twice everyday with pH strips. As a result,
the concentration of sulphate ion remained constant
Currently, there is no guideline or standard to assess the over the testing period. Visual inspection was also done
sulphate resistance of concrete. This is due to the fact to determine any sign of deterioration on the specimens.
that the delayed visual damage along with the expansion Compressive strength test was done at the age of 56 days
usually shows up after several years even when those and 148 days to evaluate the loss of strength during the
are exposed to high concentrated sulphate solutions. sulphate exposure. At least three cylinders were used per
Here, the sulphate resistance of 25 MPa RAC made with mix to perform any test on the specified date. The used
different RCA replacement levels was assessed following sodium sulphate solution was discarded after taking the
an accelerated test method used by California Department measurement of cylinders at certain intervals such as at
of Transportation. Monteiro et al. (2000) used this method 56 days, 90 days, and 148 days.
to investigate the sulphate resistance of five different
types of cement. This accelerated test method can Every week all the specimens were taken out from the
represent the field conditions and results obtained from sulphate bath to measure the dimension and height of
this method could yield similar behavior as found in field the cylinders. The height and volume changes were
conditions. Moreover, ASTM C1012 (2012) and ASTM C452 determined using the following equations
(2012) cannot represent the real deterioration pattern of Ht - Hi ...................(1)
% Height change, 3H (%) = H i # 100
concrete (Mehta and Gjorv 1974) since these guidelines
are for mortars. It is very important to investigate the
where, Hi = average initial height of the cylinder (mm);
sulphate resistance of RAC concrete due to the presence
and Ht= average height of the cylinder after a prescribed
of interfacial transition zone which differentiate it from
exposure period (mm).
mortar. Almost similar type of experimental approach
was followed for evaluating the sulphate resistance of Vt - Vi ..................(2)
% Volume change, 3V (%) = Vi # 100
RAC by Shayan and Xu (2003). In accelerated test method,
sulphate resistance of RAC was measured in terms of where V i= average initial volume of cylinder (mm3); and
the strength loss (compressive strength loss) along with Vt = average volume of cylinders after a prescribed
expansion during the sulphate exposure. Expansion only exposure period (mm3).
reveals the ettringite formation due to sulphate attack.
However, loss in strength indicates that the cracking Test Method to Assess The Chloride Ion Ingression into
occurred due to the gypsum and ettringite formation RAC
during the sulphate exposure (Cohen and Mather 1991, In North America sodium chloride (NaCl) is widely used as
Mehta and Gjorv 1974). a de-icing salt on a regular basis during winter. Therefore,
In this study, the sulphate resistance performance was a high concentration of NaCl solution forms on concrete
evaluated using 75x150 mm cylinders. This size was chosen surface, and subsequently it tries to penetrate inside the
to maximize the surface area to volume ratio. The increased concrete. To represent this phenomena, cyclic wetting
surface area to volume ratio helped accelerate the effects and drying along with NaCl solution was considered
of sulphate by increasing the exposure area. Prism and to evaluate the performance of 25 MPa RAC. It is well
bar specimens can represent the length expansion more documented that cyclic wetting and drying increases the
accurately due to their geometry compared to cylindrical chloride ion ingression into concrete (Moukawa 1990), and
specimens since it has more surface area compared to its thus accelerates the effects to yield faster test results
volume. Due to unavailability of such mold, we had to work (Yeomans 1994, Hong and Hooton 1999, Hong 1998).
with cylindrical specimens. These cylinders were moist For chloride exposures the specimen size selected was
cured for seven days in the moisture monitored curing

Organised by
India Chapter of American Concrete Institute 199
Session 2B - Paper 4

100x200 mm cylinders, and these were cured inside the The results of different types of aggregate properties
moist curing chamber (relative humidity 100%) for 28 tests are shown in Table 2. The specific gravity of RCA
days before being placed in 5% sodium chloride solution was 2.48 which was smaller than that of natural coarse
for six hours. After 6 hours of wetting, those cylinders aggregate. It is due to the adhered mortar of RCA. The
were taken out from chloride bath, and placed on a shelf adhered mortar also increased the absorption capacity
at normal room temperature (21oC) and relative humidity of RCA which was 3.75 times higher than that of natural
of 65% for drying. Those cylinders were left there for coarse aggregate. The bulk density of natural and
18 hours for drying. That means one day wet-dry cycle recycled coarse aggregates were 1576.8kg/m3 and 1374.8
where 6 hours of wetting followed by 18 hours of drying. kg/m3, respectively. Procedure outlined in CSA A23.2-
McCarter and Watson (1997) found that the wetting rate is 6A (2009) was followed to determine the relative density
faster than drying rate, and in some situations it is 3 to 7 and absorption capacity of fine aggregate. The bulk SSD
times faster. 5% sodium chloride solution was prepared specific gravity and moisture content of fine aggregate
using locally available table salt. Salt was oven dried at was 2.64 and 4.14%, respectively.
a temperature of 110oC before being mixed with water.
To investigate the chloride ion ingression concentration Results of Fresh Concrete Properties
1, 4, 9, 16, 28, 90, and 120 cycles were considered. After The 25 MPa concrete mix was designed for a 90 mm
being subjected to these numbers of cycles, chloride target slump. The results of the fresh concrete properties
concentration was measured using Ion Chromatography are provided in Table 3. This table shows that the slump
test. Ion Chromatography test is usually used for water value of different concrete mixes remained unaffected
chemistry analysis. In this method, ion concentration is due to the utilization of different replacement levels of
measured by separating them based on their interaction RCA except Mix-3 and Mix-4. This can be attributed to the
with resin. In this method, small discs were cut from the unexpected weather condition since the aggregates were
surface of concrete cylinder using chisel and hammer. stored outside which eventually got wet due to overnight
Then these small discs were pulverized using pulverizer. snowfall. The CSA requirement of air content for the 14-
The powdered concrete was then analyzed using “Ion 20 mm nominal maximum sizes of aggregates is 5-8% for
Chromatography Test”. The chloride concentration was category-1. From Table 3 it can be seen that the air content
obtained in units of parts per million (ppm). of Mix- 1 (control) and Mix-6 (100% RCA) were 5.7% and
5.5%, respectively.
Results and Discussions
Results of Compressive Strength
Properties of Aggregates
The cylinder compression tests were conducted after
Different types of testing were performed for the natural 7, 28, 56, and 148 days of curing, and the results are
coarse aggregate, natural fine aggregate (sand), and presented in Figure 1 and Table 4. Compressive strength
RCA. The results of various aggregate property tests are versus age graph is depicted in Figure 1 illustrating that
discussed in the following sections. as the percentage of RCA replacement increases the
compressive strength decreases. Inadequate hydration
Bulk density, specific gravity, and moisture content of
and weak interfacial transition zone (ITZ) between the
aggregates
components of concrete cluster caused by the high amount
The specific gravity (relative density) and absorption of attached mortar on the surface of recycled aggregate
capacity of natural and recycled coarse aggregates are the major reasons behind the strength degradation
were determined according to CSA A23.2-12A (2009). of RAC with the increased replacement level of RCA (Tu

Table 2. Properties of Aggregates

Bulk dry Bulk SSD Apparent Bulk density Absorption Moisture
specific gravity specific gravity specific gravity (kg/m3) capacity (%) content (%)
Natural coarse aggregate 2.62 2.65 2.71 1576.8 1.2 0.22
RCA 2.37 2.48 2.66 1374.8 4.5 1.9
Fine aggregate 2.54 2.64 2.77 - 1.99 4.14

Table 3. Properties of Fresh Concrete

Mix-1 Mix-2 Mix-3 Mix-4 Mix-5 Mix-6

Slump (mm) 110 110 140 140 100 95

Air content (%) 5.7 5.8 5.8 6 5.6 5.5

[Mix-1 = Control, Mix-2 = 30% RCA, Mix-3 = 40% RCA, Mix-4 =50% RCA, Mix-5 =75% RCA, and Mix-6 =100% RCA]

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

200 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical and Durability properties of recycled aggregate concrete made with different replacement levels of recycled coarse aggregate

et al. 2006, Yang et al. 2008). It is also influenced by the Several researchers worked with recycled concrete and
low bulk density and adhered mortar of RCA. On the found that long term strength development of recycled
other hand, 148 days’ compressive strength of Mix-2 (30% concrete is not as good as NA concrete (Yang et al. 2008).
RCA replacement) was higher than Mix-1 (control). Mix-1 On the other hand, many researchers found that long
compressive strength was 24.1 MPa at 148 days which was term strength gaining pattern of recycled concrete is
5.5% less than the compressive strength of Mix-2 at the better than NA concrete (Khatib 2005, Etxeberria et a.
same age. This was due to the rough texture and higher 2007). The results of this study revealed that the percent
absorption capacity of RCA. Presence of adhered mortar difference of different concrete mixes decreased at
increases the absorption capacity of RCA, and crushing 148thday which indicates that the RCA aggregate concrete
of demolished concrete makes aggregate surface rough. is more favourable than NAC while considering its long
Both of these properties of RCA lead to better interlocking term strength development. This long term strength
and bonding between the RCA and cement paste as is contributed by the unhydrated cement paste on the
compared to natural aggregate concrete (Etxeberria et exterior surface of RCA (Khatib 2005). Besides, it may be
a. 2007, Salem and Burdette 1998). Within the considered attributed by the absorbed water of RAC that may work as
time period the highest compressive strength was gained a source of water to complete the hydration process (Yang
by Mix-2 (25.5MPa) and the lowest was found for Mix- et al. 2008).
4 (19.8 MPa) due to higher slump value. The strength Statistical analyses were carried out to evaluate the
gaining pattern of Mix-5 and Mix-6 were almost similar variation of the compressive strengths of six different
except at the age of 28 days the compressive strength of mixes. Figure 2 presents the box plot of data found from the
Mix-6 was 9.3% smaller than that of Mix-5. Table 4 shows 7, 28, 56, and 148 day compressive strength of RAC made
the results of compressive strength at different test days with different RCA replacement levels and its comparison
and their percent difference in strength gain with respect with control mix (Mix-1). Figure 2 shows the variation of
to NAC (Mix-1) at the same respective age. The percent compressive strengths in individual mix proportion where
difference in compressive strength between the Mix-1 the numerical range (maximum and minimum values) of
(control mix) and Mix-6 (100% RCA replacement) at 148 data is represented by the height. The boxes represent
days was 14.5 %. This illustrated the true loss in strength the 1st quartile through 3rd quartile. The horizontal line
as a result of replacing RCA with NA. As the replacement inside the box represents the 50th percentile (median
level of natural aggregate by RCA increases, the percent value). It can be observed that all the mixes at early ages
difference also increases. (7days) were having uniform distribution of strength
where the control specimen had the highest dispersion
after Mix-2 (30% replacement) whereas mixes with
higher replacement levels had lower range of dispersion.
Mixes-1, 2, 5 and 6 showed negative skewness for later
strength whereas Mix-3 showed positive skewness in
its strength distribution. Mix-4 was showing closer to
uniform distribution.

Results of Sulphate Resistance Test

The sulphate durability test was based on the strength
loss as well as the changes in volume and height of the
specimen which were measured over time. This indicated
how reactive the specimens were to sulphate and whether
one mix was more reactive than the other, thereby making
Fig. 1: Compressive strength of concrete made with different it less durable. The compressive strengths of concrete
replacement levels with different RCA replacement levels at different

Table 4. Compressive Strength Results of Different Concrete Mixes

Mix-1 Mix-2 Mix-3 Mix-4 Mix-5 Mix-6
(MPa) (MPa) (MPa) (MPa) (MPa) (MPa)
7 days 12 13.4(11.7%) 9.6(-19.6%) 10.1(-15.5%) 8.7(-27.8%) 8.9 (-25.4%)
28 days 17.1 16.5(-3.6%) 16.3(-4.8%) 14.7(-14.2%) 15.1(-11.6%) 13.7(-19.9%)
56 days 23 22(-4.2%) 18(-21.6%) 18.9(-17.9%) 17.7(-23%) 16.6(-27.8%)
148days 24.1 25.5(5.8%) 19.9(-17.4%) 19.8(-17.8%) 21(-12.9%) 20.6(-14.5%)
[Mix-1 = Control, Mix-2 = 30% RCA, Mix-3 = 40% RCA, Mix-4 =50% RCA, Mix-5 =75% RCA, and Mix-6 =100% RCA]
Note: the value in braces represents the percent difference in strength gain with respect to Mix-1

Organised by
India Chapter of American Concrete Institute 201
Session 2B - Paper 4

interval of time under sulphate exposure condition are

shown in Figure 3. The bar diagram reveals that even
under sulphate exposure up to 56 days, the compressive
strengths of all mixes were increasing and later a
decreasing phenomenon was observed by all mixes.
Strength increase does not indicate anything about
sulphate attack. It only reveals that cement continues
to hydrate in sodium sulphate solution during that time
and the pores get filled up with hydrated products along
with gypsum and ettringite. Further formation of these
products are responsible for the micro crack development
and degradation of concrete strength in the later period as
these products have a considerably greater volume than Fig. 3: Compressive strength of concrete made with different
the compound they replace during the reaction in sulphate RCA replacement level under sulphate exposure
exposure (Neville 2011). The results also indicate that as
percent strength reduction as found for Mix-6 though its
the RCA replacement increases the compressive strength
RCA replacement level was less than the other different
decreases. The compressive strength of Mix-1 at the age
of 148 days under sulphate exposure was 14.5 MPa which replacement levels of RCA.
was 9.8% higher than that of Mix-2 (13.2 MPa).

Fig. 2: Variation in compressive strength of different concrete Fig. 4: The percent (%) change in compressive strength of
mixes at the age of 7, 28, 56, and 148 days concrete made with different RCA replacement levels at the
age of 148 days under sulphate exposure with respect to the
The relative change in compressive strength (∆Cs) found
compressive strength of moist cured specimens at the same
at the age of 148 days after 141 days of sulphate exposure
respective age
is shown in Figure 4. The change in compressive strength
due to sulphate exposure was measured as a percentage As mentioned earlier, the specimens used for sulphate
of the strength of each cylinders found after 148 days of testing were measured on a weekly basis to monitor
moist curing (C148m). The equation is given below:
the physical changes occurring over time. The height
C 148s - C 148m .........................................(3) change was increasing with the increased level of RCA
3C s (%) = C 148m # 100
replacement. The change of height is important since it is
where, C148s= average initial compressive strength of the larger dimension of the cylinder, and will experience
cylinders under sulphate exposure at the age of 148 days more change than the diameter. Figure 5a shows the
(141 days sulphate exposure) (MPa) and C148m= average variation in terms of height change experienced by the
compressive strength of moist cured cylinders at the age specimens over time under sulphate exposure. The
of 148 days (MPa). percent (%) height change of Mix-3 and Mix-4 were similar
Figure 4 shows that after 141 days of sulphate exposure, and lower than Mix-5 and Mix-6 at 56th day. The crumbling
Mix-6 showed the highest strength reduction (48%) of concrete cylinders of Mix-4 was responsible for this
among the six considered mixes. This is due to the similar value. At the age of 148 days, the highest change
formation of micro crack for the production of gypsum and was observed for Mix-6 (0.2%) and it was 48%, 43%, 38%,
ettringite. Presence of old interfacial transition zones also 33%, and 18% higher than the percent height change of
significantly influences this phenomenon. Mix-1 and Mix- Mix-1, Mix-2, Mix-3, Mix-4, and Mix-5, respectively. This is
3 performed almost in a similar fashion. Only exception due to the increased porosity and old interfacial transition
was Mix-2 (48%) which showed the highest amount of zone of RCA.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

202 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical and Durability properties of recycled aggregate concrete made with different replacement levels of recycled coarse aggregate

Fig. 5: (a) Height change (%) (b) Volume change (%) of concrete cylinders under sulphate exposure condition

Similar trend is observed for the results in terms of 1 after being exposed to 4 wet dry cycles was 1.56 times
change in volume as seen in the height change where the higher than that of first cycle and the value of chloride
volume change increases as the RCA content increases. ion concentration gradually increased with the number
Figure 5b shows the average volume change for each of cycles which was 120.96 ppm/m2 after 120 cycles. Mix-
mix over the considered time under 5% sodium sulphate 2’s chloride ion concentration was 2.1% higher than Mix-
solution. RAC made with different RCA replacement levels 1 after being subjected to 120 cycles. After 120 wet dry
experienced higher volume expansion than the NAC (Mix- cycles, the chloride ion concentration of Mix-6 was 0.8%
1). At the age of 28 days, Mix-6 showed the highest volume and 6.8% higher than that of Mix-5 and Mix-4, respectively.
change which was 20% higher than that of Mix-5 (0.125%). This reveals that permeability of RAC increases with the
From Figure 5b, it can be observed that there was a increased amount of RCA replacement, which is due to
significant expansion in the volume of concrete cylinders the presence of old interfacial transition zone (ITZ) and
with increased period of time under sulphate exposure. attached mortar on the surface of recycled aggregate.
Mix-2 experienced 3.8% higher volume change than Mix-1
and 1.2%, 2.4%, and 6.9% lower than that of Mix-3, Mix-
4, and Mix-5, respectively at 148thday. Highest volume
change of 0.495% was observed by Mix-6 at 148th day.

Results of Chloride Ion Ingression into RAC

Figure 6 shows the chloride ion concentration of different
concrete mixes after being exposed to considered number
of wet dry cycles along with sodium chloride solution.
The results obtained from ion chromatography test were
divided by the surface area of the cylinder to get the
chloride ion concentration per unit area of concrete and
are shown in Figure 6. This approach is different from
pervious researchers’ approach. The results illustrate Fig. 6: Concentration of chloride ions per unit surface area of
that the concentration of chloride ion increased with concrete cylinder
the increased number of RCA replacement levels. It
was found that chloride ion concentration significantly
Conclusions
increased with increased number of wetting and drying
cycles. No chloride concentration was found for moist Previous researchers showed that it is possible to produce
curing samples. Initially, chloride ion ingression rate was low strength recycled concrete using construction and
higher, and for Mix-6 it was the highest. From Figure 6, it demolition waste. Still, limited research works have been
can be observed that after being exposed to 90 wet dry done for the industrial production of recycled concrete
cycles Mix-2’s chloride concentration is higher than Mix- with recycled aggregate collected from landfill, ensuring
1 (91.59 ppm/m2). This can be attributed to the presence a valid comparison with the control mix, as well a similar
of attached mortar and pores of RAC. The difference in effective water-cement ratio. The results from this
chloride concentration between Mix-1 and Mix-2 after study will help overcome industry’s fear regarding the
exposed to 90 wet dry cycles was 17.96 ppm/m2. It can application of RCA in concrete production. The following
be seen that the chloride ion concentration value of Mix- conclusions are drawn from this study.

Organised by
India Chapter of American Concrete Institute 203
Session 2B - Paper 4

ll The bulk density and specific gravity of recycled Mortars Exposed to Sulphate.
coarse aggregate were 12.8% and 6.4% less than that 4. Cement Association of Canada, 2012. Available online: http://www.
of natural aggregate, respectively. cement.ca/en/Concrete-and-the-Environment.html. Accessed: 13th
Jan 2013.
ll The absorption capacity of recycled coarse aggregate 5. Cohen, M.D., and Mather, B. 1991. Sulfate Attack on Concrete -
was 3.75 times higher than that of natural coarse Research Needs, ACI Mater J, 88(1):62-69.
aggregate. The findings of this study show that the CSA 6. Concrete Association of Canada, 2012. Available online: http://www.
A23.2-12A can be used for investigating the absorption cement.ca/en/Concrete-and-the-Environment.html. Accessed: 28th
of recycled coarse aggregates Oct 2012.
7. CSA A23.2-12A, Canadian Standards Association (CSA), “Relative
ll As the RCA replacement level increases the Density and Absorption of Coarse Aggregate, Toronto, Ontario,
compressive strength decreases. Only exception was 2009.
Mix-2 (30% RCA) which achieved 5.8% higher strength 8. CSA A23.2-6A, Canadian Standards Association (CSA), “Relative
than that of Mix-1 (control) at 148th day. This can be Density and Absorption of Fine Aggregate, Toronto, Ontario, 2009.
attributed to the rough texture and better interlocking 9. CSA A23.2-9C, Canadian Standards Association (CSA), “Compressive
properties of RCA. This study shows that up to 30% Strength of Cylindrical Concrete Specimens, Toronto, Ontario, 2009.
RCA replacement level, it is possible to achieve similar 10. Etxeberria, M., Vázquez, E., Mari, A., and Barra, M. 2007. Influence
or higher compressive strength than the natural of Amount of Recycled Coarse Aggregates and Production Process
coarse aggregate concrete. on Properties of Recycled Aggregate Concrete, Cement and Concr
Res, 37:735-742.
ll The long term strength development of RCA concrete 11. Gomes, M., and de Brito, J. 2009. Structural Concrete With
is more favorable than NAC. Incorporation of Coarse Recycled Concrete and Ceramic Aggregates:
Durability Performance. Mater and Struc, 42:663–675.
ll The durability performance of recycled concrete is
12. Gomez-Soberon, J.M.V. 2002. Porosity of Recycled Concrete With
affected by the higher absorption and porosity of RCA. Substitution of Recycled Concrete Aggregate – An Experimental
Study, Cement and Concr Res, 32,(8):1301–1311.
ll The result of sulphate resistance of recycled concrete
was quite comparable to NAC. 13. Hansen, T.C., and Hedegkd, S.E. 1984. Properties of Recycled
Aggregate Concretes as Affected by Admixtures in Original
ll Chloride ion concentration increased with the Concretes, J of the Amer Concr Ins, 81(1):21–26.
increased number of wetting and drying cycles. 14. Hong, K., and Hooton, R.D. 1999. Effects of Cyclic Chloride Exposure
on The Penetration of Concrete Cover, Cement and Concr Res,
ll Significant influence of the RCA replacement level (50% 29:1379-1386.
and above) was found on the chloride concentration of 15. Hong, K. 1998. Cyclic Wetting and Drying and Its Effects on Chloride
RAC. It showed that the chloride ion ingression of RAC Ingress in Concrete, Master’s Dissertation, Department of Civil
increased with the increased RCA replacement level. Engineering, University of Toronto, Toronto, ON, Canada.
After 120 wet-dry cycles, the chloride ion concentration 16. Huda, S.B., and Alam, M.S. 2014. Mechanical Behavior of Three
of Mix-6 (100% RCA) was 198.19 ppm/m2 , and the Generations of 100% Repeated Recycled Coarse Aggregate
Concrete. Constru and Build Mater, 65:574-582.
highest among all the considered mixes.
17. Huda,S.B., and Alam,M.S. 2015. Mechanical and Freeze-Thaw
The application of RAC made with different replacement Durability Properties of Recycled Aggregate Concrete Made with
levels exhibited reduction in terms of their strength Recycled Coarse Aggregate, J Mater Civ Eng, 10.1061/(ASCE)MT.
and durability properties with increased level of RCA 18. Katz, A. 2003. Properties of Concrete Made with Recycled Aggregate
replacements. From mechanical and durability point from Partially Hydrated Old Concrete, Cement and Concr Res,
of view it is possible to produce good quality industrial 33(5):703-711.

concrete with low RCA replacement levels. The 19. Khatib, J.M. 2005. Properties of Concrete Incorporating Fine
Recycled Aggregate, Cement and Concr Res 35:763-769
compressive strength and durability properties of Mix-
2 (30% RCA) were quite comparable to the control 20. Olorunsogo, F.T., and Padayachee, N. 2002. Performance of
Recycled Aggregate Concrete Monitored by Durability Indexes,
mix. Moreover, environmental benefits may be able to Cement Concr Res, 32:(2):179–185.
compensate to some extents the negative effect due to the
21. McCarter, W.J., and Watson, D. 1997. Wetting and Drying of Cover-
use of RCA, and can lead us significantly closer towards Zone Concrete. In: Proceedings of the institute of civil engineers
sustainable construction for infrastructure. - structures and buildings 122(2):227-236.

References 22. Mehta, P.K., and Gjorv, O.E. 1974. A New Test for Sulfate Resistance
of Cements, J of Testing and Evaluation, 2(6):510-15.
1. Alam, M.S., Slater, E., and Billah, A. 2013. Green Concrete Made with
23. Monteiro, P.J.M., Roesler, J., Kurtis, K.E, and Harvey, J. 2000.
RCA and FRP Scrap Aggregate: Fresh and Hardened Properties, “J
Accelerated Test for Measuring Sulphate Resistance of Hydraulic
Mater Civ Eng, 25(12):1783-1794.
Cements for Caltrans LLPRS Program, California Department
2. ASTM C1012, American Society for Testing and Materials. 2012. of Transportation, Report No:FHWA/CA/OR-2000/03, Pavement
Standard Test Method for Length Change of Hydraulic-Cement Research Centre, Institute of Transportation Studies, University of
Mortars Exposed to A Sulfate Solution. California, Berkeley, USA, 2000.
3. ASTM C452, American Society for Testing and Materials. 2012. 24. Moukawa, M. 1990. Deterioration of Concrete in Cold Sea Waters,
Standard Test Method for Potential Expansion of Portland-Cement Cement and Concr Res, 20(3):439-446.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

204 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical and Durability properties of recycled aggregate concrete made with different replacement levels of recycled coarse aggregate

25. Neville, A.M. (2011). Properties of concrete, 5th edition, Pearson 30. Tu, T.Y., Chen, Y.Y., and Hwang, C.L. 2006. Properties of HPC with
Education Limited, Essex, England, 2011. Recycled Aggregates, Cement and Concr Res, 36:943-950.
26. Salem, R.M., and Burdette, E.G. 1998. Role of Chemical and Mineral 31 Ulloa, V.A., Garcia-Taengua, E., Pelufo, M., Domingo, A., and Serna,
Admixture on Physical Properties and Frost-Resistance of Recycled P. 2013. New Views on Effect of Recycled Aggregates on Concrete
Aggregate Concrete, ACI Mater J, 95(5):558–563. Compressive Strength, ACI Mater J, 110(6):687-696.
27. Salem, R.M., Burdette, E.G., and Jackson, N.M. 2003. Resistance to
32. U.S. Green Building Council (USGBC), 2014. Available online: http://
Freezing and Thawing of Recycled Aggregate Concrete, ACI Mater
www.usgbc.org/leed/rating-systems. Accessed : 22nd Jan 2014.
J, 100(3):216–221.
33. Yang, K.H., Chung, H., and Ashraf, A.F. (2008). Influence of Type and
28. Shayan, A., and Xu, A. 2003. Performance and Properties of
Structural Concrete Made with Recycled Concrete Aggregate, ACI Replacement Level of Recycled Aggregates on Concrete Properties,
Mater J,100(5):371-380. ACI Mater J, 105(3):289-296.

29. Smith, J.T. 2009. Recycled Concrete Aggregate – A Viable Aggregate 34. Yeomans, S.R. 1994. Performance of Black, Galvanked, and
Source for Concrete Pavements, PhD Dissertation, Department of Epoxy-Coated Reinforcing Steels in Chloride Contaminated
Civil Engineering, University of Waterloo, Waterloo, ON, Canada. Concrete,“Corrosion, 50(1):72-81.

Dr. Shahria Alam

Dr. Shahria Alam is an Associate Professor in the School of Engineering at The University of British
Columbia’s Okanagan campus. He received his PhD in Civil Engineering from Western University in 2008.
His research interests include smart materials and their structural applications in bridges and buildings;
seismic isolation devices, seismic rehabilitation of structures; performance-based design; recycle/reuse
of industrial wastes. Dr. Alam is the Chair of Concrete Structures Sub-Committee and Vice-Chair of the
Mechanics and Materials Division of Canadian Society for Civil Engineering. He is an active member of
Joint ACI-ASCE Committee 441, Reinforced Concrete Columns and ACI Committee 341- Earthquake-
Resistant Concrete Bridges. His research interests include structural application of smart and advanced
materials, seismic rehabilitation of buildings and bridges, and their performance-based design, and
sustainable construction. He has published more than 100 peer reviewed articles in these areas. He is
also the recipient of many national and international awards including CSCE Pratley Award 2015 and UBC
Moldovan Memorial Award 2014.

Organised by
India Chapter of American Concrete Institute 205
SESSION 2 C
Session 2 C - Paper 1

Compression Behavior of Synthetic Fiber Reinforced Cellular Concrete

Masonry Prisms
Abdur Rasheed, M, Dr. Suriya Prakash. S
Dept. of Civil Engineering, IIT-Hyderabad, Telangana, India

Abstract structures. Both the masonry load bearing walls and

infill walls were heavily damaged during the earthquake
This paper presents the effect of addition of synthetic
in Bhuj, India in the year 2001. The Masonry Society and
fiber reinforcement on compression behavior of Cellular
Federal Emergency Management Agency (FEMA) of USA
Lightweight Concrete (CLC) masonry prisms. CLC prisms
identified that the falling of unreinforced masonry walls is
of dimension 470 mm x 200 mm x 150 mm were cast with
the main source for the loss of human life and damage to
and without synthetic fiber reinforcement. Poly-olefin
the property during earthquakes.
structural fiber reinforcement is used at volume fractions
of 0.22%, 0.33%, 0.44%, and 0.55% with and without micro The strength and the seismic performance of masonry
fiber dosage of 0.02% volume fraction. Use of micro- structures can be improved by engineering fiber
fibers (Fibrillated) enhances pre-cracking behavior of reinforcement into masonry system. Fibers in masonry
masonry by arresting cracks at micro-scale, while Macro increase structural integrity by reducing permeability
(structural) fibers induce ductile behavior in post-peak which prolongs the life of the masonry structure. Residual
region by arresting the crack propagation soon after the strength of masonry can be provided by the addition of
crack initiation. Test results show that fiber reinforcement fibers. Using fiber reinforcement in masonry means a
enhanced the strength, stiffness and ductility of CLC better, long-lasting performance. However, a thorough
masonry prisms under compression. Test results indicated knowledge about the behavior and the failure modes of
that the hybrid fiber reinforcement provided better crack engineered fiber reinforced brick masonry is necessary
bridging both at micro and macro levels. to arrive at the design guidelines. It is essential to develop
low-cost brick masonry systems with improved tensile
Keywords: CLC Prisms, Compression, Hybrid-synthetic
and shear strength. It is worth mentioning that bricks of
Fibers, Stress-Strain Curves, Fiber Dosage.
low strength (varying from 4 to 10 MPa) are commonly
used for masonry load bearing and infill wall construction
Introduction in the developing countries. The total rural housing
A large percentage of the building stocks in India and around shortage in 2012 is estimated to be 47.3 million as per the
the world comprise of non-engineered unreinforced report of the working group setup by the Government of
masonry (URM). The performance of these buildings in the India on rural housing for 11th five year plan. Out of this, 90
past has shown that these masonry buildings are highly percent is for the below poverty line families. Therefore,
vulnerable to failure under seismic loads. In particular,
URM exhibits brittle failure modes under seismic loading[1]
and are prone to complete collapse leading to loss of life
and property (Fig. 1). The most widespread collapsing
mechanisms commonly encountered in URM buildings
under seismic loading involve both the out-of-plane and
in-plane failure modes[2]. As the unreinforced masonry
walls contribute to the lateral seismic resistance of the
building, the first possible failure mode is in-plane shear
failure. The other type of failure is represented by the out-
of-plane flexural failure due to the orthogonal inertial
forces induced by the earthquake. Excessive out-of-plane
bending also reduces the vertical load carrying capacity
of URM walls and thereby leading to failure under in-
plane conditions. Earthen brick masonry load bearing
constructions are common in India but fail in earthquakes
mainly due to their low tensile and shear resistance. Fig. 1: Common Failures of Unreinforced Masonry Systems (a)
Brick masonry infill walls are used as exterior and & (b) Diagonal cracking; (c) Vertical cracking at the corner joint;
interior partitions in reinforced concrete and steel framed (d) Out of plane failure

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

208 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Compression Behavior of Synthetic Fiber Reinforced Cellular Concrete Masonry Prisms

the purpose of this study is to explore the development of Critical review of literature indicates that only a handful of
sustainable low cost fiber reinforced blocks for structural studies have focused on fiber reinforced CLC for structural
applications of masonry that can provide better housing applications of masonry. Improved compression, shear
solutions. In particular, the focus is on developing a high- and tensile resistance can be expected with hybrid
performance fiber reinforced cellular concrete blocks addition of structural/macro fibers along with micro-
without the high- pressure steam curing process as an fibers for superior crack resistance at both micro and
alternative to clay brick and aerated autoclaved concrete macro levels. Hence, the purpose of this study is to (i)
(AAC) blocks that can be used in rural constructions. investigate the stress-strain behavior of masonry prisms
cast with different fiber dosages and (ii) understand the
effectiveness of fibers on energy dissipation capacity
Literature Review
(toughness index) and failure modes on the compression
The concept of light weight concrete is not new. The behavior of CLC prisms.
building structure of ‘The Pantheon’ of lightweight
concrete material is still standing in Rome until now for
about 18 centuries. It shows that the lighter materials can Experimental Program
be used in concrete construction and has an economical
advantage. The light-weight concrete can be broadly Materials
categorized into three groups: (i) No-fines concrete, (ii) The materials used for the nonfibrous control CLC
Lightweight aggregate concrete (iii) Aerated concrete. mixture consisted of 53 grade Ordinary Portland Cement,
The aerated concrete is the basis of CLC technology can Flyash from NTPC (National Thermal Power Corporation),
be further classified Based on method of pore formation potable water and sunlite foam. The mix proportion of
such as (i) Air-entraining method (gas concrete) (ii) flyash: cement: water: foam was 833: 277:277:1.4 kg/m3.
Foaming method (foamed concrete) (iii) Combined pore The additives are coarse bi-component macrofiber and
forming method. It can be further classified based on type fibrillated fibre as shown in the Fig. 2 and Fig. 3 respectively.
of binder used as (i) Cement based (ii) Lime based and (iii) The physical properties of fibers are mentioned in Table 1.
Pulverized fuel ash or slate waste as partial replacement
to binder. Moreover, it can be classified based on method Mixing and Curing
of curing as (i) Non autoclaved aerated concrete and (ii) The dry ingredients such as cement and flyash were fed
Autoclaved aerated concrete. into the mixer first and thoroughly mixed to ensure even
Rudolph and Valor [3] carried out tests on cellular concrete distribution of cement. Thereafter, water was added and
and suggested that flexure strength of CLC was 1/3 to 1/5 the mixing process continued. The preformed foam[9]
of compressive strength. Sengupta [4] used flyash as partial was added at 35 gm/ sec for 40 seconds to the slurry
replacement of binder and concluded that, utilizing flyash of cement, flyash and water in the batch mixer. After an
to produce aerated concrete is an economically attractive additional mixing of three minutes along with fibers to
proposition, which will help in mitigating the environmental get uniform consistency, the slurry form CLC was poured
damage caused by flyash. The usage of Polypropylene into dimension 600 mm length, 150 width and 200 mm
fibers has gained more prominence in the recent years height rectangular moulds. These rectangular moulds
for reinforcing cementitious materials[5-6]. C-cement, were demoulded after 24 hours and curing was done
L-lime, S-sand, F-flyash, Q-quartz, W-slate waste, mc- as per IS-456 2000[10]. Masonry prisms were cast using
moist curing, ac-autoclave curing. Usage of CLC blocks in
masonry construction has gained tremendous popularity in
recent decades owing to its sustainability, low density, low
thermal conductivity and less mortar joints in construction.
However, they have few disadvantages such as low tensile
and shear strength. This leads to very brittle failure under
lateral loads. Previous investigation have revealed that
addition of fiber has improved post-cracking features of
masonry, showing ductile behavior by arresting the crack
propagation soon after the crack initiation. However, Fig. 2: Poly-Olefin Macrofiber
such studies in CLC masonry is very scarce and needs
attention to better understand the fracture behavior under
flexure and shear. Tests carried out by Ronald and Carol[7]
indicates the ability of fiber reinforcement to transform the
basic material character of cellular concrete from brittle
to ductile elasto-plastic behavior The authors found that
the performance of the fiber reinforced CLC was better
compared to the control ones. Fig. 3: Poly-Olefin fibrillated fiber

Organised by
India Chapter of American Concrete Institute 209
Session 2 C - Paper 1

Table 1
Physical Properties of Poly-Olefin Fiber [8]

Macro Fiber Fibrillated Fiber

Specification Bi-component fiber Interlinked fiber

Material Poly-olefin Poly-olefin

Form Structural fiber Fibrillated fiber

Specific Gravity 0.91 0.91

Length 50mm 19mm

(a) Specimen ready (b) Specimen after
Tensile strength 618 N/mm2 400 N/mm2 for testing Compression Failure

Modulus of Elasticity 10 GPa 4.9 GPa Fig. 4: Masonry Prism Details for Compression Testing

Diameter 0.5mm 0.08mm

Results and Discussions
blocks of 200 x 150 x 110 mm. Four blocks were used for Slump
constructing the prisms. A cement mortar with cement: The CLC mix which flows into the moulds like self-
sand ratio of 1:6 was used for joints. Dimensions of the compacting concrete, remained unaffected by addition of
cast prism is shown in Fig. 4. After curing the blocks for fibers and showed equally good mobility into the moulds
28 days, the testing was done under displacement control. on reinforcing with fibers. This can be attributed to free
movement of air voids around the fibers which is restricted
had there been the coarse aggregate of normal concrete.

Behavior under Compression

Toughness Index is the measure of energy absorbed by
the material in undergoing a specified amount of strain,
being the area under the Stress-strain graph. A limiting
strain of 0.01 was used for calculation of strain energy.
Three series of specimen were tested. Series 1 had control
specimen with no fiber. Series 2 had specimen with only
macro fibers. Series 3 had macro fibers with a constant
micro fiber dosage of 0.02%. Unreinforced CLC exhibited
brittleness with the post-peak strength decreasing rapidly
with increase in strains after the peak stress. However,
Fig. 4: Masonry Prism Details for Compression Testing for the fiber reinforced specimens, the post-peak strength
degradation was more gradual indicating the addition of
fibers have enhanced the toughness as shown by increase
Test Method
in the strain energy in Table 2.
The testing code for fiber reinforced CLC prisms under
compression are not available yet, though ASTM C1609/C[11] Stress-stain curve under compression for the unreinforced
and IS 1905[12] was used as a guideline to establish specimen showed a linear behavior upto 30% the peak
stress strain curves of prisms under compression. The load (Fig. 6). Thereafter, non-linear behavior was observed
specimen was tested using servo controlled hydraulic upto the peak stress. After the peak load, the failure was
testing machine and loading was increased at a rate of quite sudden as the specimen collapsed showing little
0.1 kN/sec for 90% of the peak load and then 0.001mm/ resistance to the applied strain. For masonry prisms
sec displacement control loading was used for the rest with the structural fibers, the behavior until the peak load
of the test. All the specimen were tested under pure was similar to that of unreinforced specimen but with a
compression as shown in Figure 5. marginal increase in the initial modulus of elasticity (Fig.
6). The increase in modulus of elasticity can be attributed
Loads were measured using the load cell and strains to higher modulus of elasticity of fibers (about 10,000 MPa)
were measured using LVDTs mounted on the specimen as compared to that of CLC (about 3000 MPa). As the CLC
shown in Figure 5. All the measurements were collected material forms the major volume fraction of the prisms,
through Controls data acquisition software. the behavior is close to that of CLC blocks than that of
mortar. The peak strength increased with the increase in
fiber dosage.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

210 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Compression Behavior of Synthetic Fiber Reinforced Cellular Concrete Masonry Prisms

Table 2
Test Results of CLC Masonry Prisms in Compression with and without Fibers

Peak Compressive strength(MPa) Mean Strength

Series Specimen Std. Dev (kN) CTI (10-3)
1 2 3 ‘f’m’ (kN)

I Control 4.11 3.78 - 3.87 0.34 14.69

ma-0.22-mi-0.0 4.40 3.73 4.03 4.05 0.69 29.00

II ma-0.33-mi-0.0 4.73 4.02 5.21 4.73 0.84 36.68

(only Macro) ma-0.44-mi-0.0 4.99 5.37 4.69 5.04 0.23 42.43
ma-0.55-mi-0.0 5.66 4.26 4.89 4.96 0.33 42.72
ma-0.11-mi-0.02 3.77 3.18 4.87 3.96 0.61 28.57

III ma-0.22-mi-0.02 4.18 4.29 - 4.23 0.85 32.93

(hybrid) ma-0.33-mi-0.02 5.18 5.43 3.86 4.85 0.08 39.28
ma-0.44-mi-0.02 6.34 4.52 5.99 5.62 0.96 44.24

The post-peak region of fiber reinforced specimen showed strain curves for specimens with only macro fibers and
a very ductile behavior. The area under the stress- strain hybrid fibers are compared in Fig. 7. Peak compressive
curve increased with increase in fiber dosage. The strength in hybrid specimen increased compared to that
stress in the post-peak remained almost close to that of cylinders with only macro-fibers. This can be explained
of peak compressive load. Hybrid-fiber reinforcement by the better arresting of cracks at micro- scale by micro-
on the other hand also showed appreciable increase fiber and synergetic role of both fibers which led to the
in modulus of elasticity upto the peak load, while the increase in peak compressive strength and better post-
strength degraded in the post-peak region without much peak behavior. It is worth mentioning that clay bricks of
degradation in modulus of elasticity (Fig. 6b). The stress- low strength (varying from 4 to 10 MPa) are commonly
used for masonry load bearing and infill wall construction
in the developing countries. Compressive strength of 4
to 6 MPa was achieved in CLC through addition of fibers
in compression. Therefore, the fiber reinforced CLC can
potentially replace the existing clay bricks with superior
mechanical properties. Cost optimization of the developed
fiber reinforced CLC can be a scope for future work.

Fig. 7: Behavior of Masonry Prisms under Compression: Macro

versus Hybrid Fiber Dosage

Conclusions
CLC is sustainable, light in weight and has good acoustic
Fig. 6: Behavior of CLC masonry prisms under axial compression and thermal insulation. The performance of CLC can be
(a) with Macro-fibers and (b) Hybrid Fibers enhanced with regard to ductility and inelastic behavior

Organised by
India Chapter of American Concrete Institute 211
Session 2 C - Paper 1

by addition of fibre reinforcement. Fiber reinforced CLC plant, Hyderabad India for helping with mixing and casting
can be used for developing low cost masonry blocks of CLC blocks used in this study.
for structural applications. The effect of synthetic fiber
reinforcement addition on the compression behavior was References
studied by testing prisms in compression. Based on this 1. Albert ML , Elwi AE , Cheng JR. Strengthening of unreinforced
study, the following conclusions can be drawn: masonry walls using FRPs. J Comp Construct 2001; 2:76-84
2. Evaluation of earthquake damaged concrete and masonry wall
ll Masonry compressive strength results were consistent buildings. basic procedures manual. California: Applied Technology
with cylinder compressive results. It was found to Council, Federal Emergency Management Agency (FEMA); 1999.
increase progressively with fiber dosage. It increased Report No.: ATC-43,FEMA 306.
up to 28.3% for 0.55% when compared to that of 3. Valore, R. C.,(1954) “Cellular Concretes-Physical Properties,”
control specimen. Increase in strength decreased with Journal of the American Concrete Institute, No. 25, pp.817-836.
increase in fiber dosage. Mobasher, B., Li, C.Y., (1996) “Mechanical Properties of Hybrid
Cement Based Composites," ACI Materials Journal, 93 (3) 284–292.
ll There was a minimal change in strength and post- 4. Sengupta J. (1992) Development and application of light weight
peak behavior between 0.44% and 0.55% volume aerated concrete blocks from fly ash. Indian Concr J; 66:383±7.
fraction. With further addition of micro-fibers of 5. Perez-Pena, M., Mobasher, B. (1994) “Mechanical properties of fiber
0.02%, the compressive strength increased up to reinforced lightweight concrete composites”, Cem. Concr. Res. 24
45.22% for 0.44%. This indicates that the hybrid (6), 1121–1132.
reinforced specimens performed better compared to 6. Qian, C.X., Stroeven, P. (2000) “Development of hybrid polypropylene-
the specimen with only macro structural fibers. steel fibre-reinforced concrete”, Cem. Concr. Res. 31 (1) 63–69.
7. Ronald, F., Carol, D. H., (1998) “Engineering Material Properties
ll The CLC masonry prisms showed a good composite of a Fiber Reinforced Cellular Concrete,” Journal of the American
behavior due to the proximity of strength between Concrete Institute, No. 95-M61, pp.631-635.
mortar and block. Therefore, the strength of the 8. Concrix-Technical Datasheet, Brugg Contec AG, CH-8590
composite prisms was closer to the low strength block Romanshorn . http://w w w.bruggcontec.com/English/Home/
due its higher volume fraction in the prism. Concrix/tabid/474/language/en- US/Default.aspx
9. ASTM C 869-91 (1992) Standard specification for foaming agents
used in making preformed foam for cellular concrete, American
Acknowledgments Standards for Testing Materials.
This experimental work is carried out as part of the 10. IS: 456 (2000), Plain and Reinforced Concrete-Code of Practice
project funded by Ramanujan fellowship grant sponsored (Fourth Revision), Bureau of Indian Standards, New Delhi, India.
by Department of Science and Technology, India. Their 11. ASTM C 1609/C 1609M – 07, (December, 2007) “Standard Test
financial support is gratefully acknowledged. Fiber Method for Flexural Performance of Fiber-Reinforced Concrete
(Using Beam With Third-Point Loading)”, Annual Book ASTM
reinforcement used in this study was donated by Brugg Standards, American Standards for Testing Materials.
Contec AG. We also acknowledge Srinivasa CLC block
12. IS 1905 –1987; Structural use of unreinforced masonry (third
revision), Bureau of Indian Standards, New Delhi, India.

Dr. Suriya Prakash

Dr. Suriya Prakash’s research expertise on structural concrete behaviour and design. He is currently
working as assistant professor at IIT Hyderabad. Before joining IITH, he worked with STRUCTURAL Inc,
USA a renowned firm in strengthening design and construction using advanced construction materials.
He has designed strengthening solutions for several buildings in the US and middle east. Before that,
he had completed his PhD from Missouri University of Science and Technology, USA. He has authored
more than twenty journal papers on behaviour of reinforced concrete columns and strengthening with FRP
composites. He is a member of ASCE and ACI, USA.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

212 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Flexural Behavior of Synthetic Fiber Reinforced Cellular Light Weight Concrete

Flexural Behavior of Synthetic Fiber Reinforced Cellular

Light Weight Concrete
Abdur Rasheed, M, Dr. Suriya Prakash. S
Dept. of Civil Engineering, IIT-Hyderabad, Telangana, India

Abstract positive effect on compressive strength when added in

optimum amount[6]. Moreover, no emission of pollutants
Cellular light weight Concrete (CLC) masonry has gained
during manufacturing makes CLC a viable alternative
tremendous popularity in recent decades owing to its
to red clay burnt bricks. Burnt clay bricks uses top soil
sustainability, low density, low thermal conductivity and
as raw material[7] and require approximately 50 tons of
less mortar joints in construction. Investigations in the
firewood for 1,00,000 bricks (direct thermal requirement).
past have revealed that addition of fiber has improved
In addition, CLC offer strength, dead load reduction and
post- cracking features of masonry, showing ductile
thermal insulation[8], but their limited ability to absorb
behavior by arresting the crack propagation soon after
earthquake energy raises concerns. On the other hand,
the crack initiation. This paper presents the effect of
fiber reinforced concrete has greater energy absorbing
addition of synthetic structural fiber reinforcement on
ability called ductility or inelastic deformation capacity[9-10],
flexural behavior of CLC blocks. The objective of this study
than normal concrete, but its weight possess problems.
was to understand the flexural behavior of CLC blocks
Fiber reinforced CLC has a promising future for precast
with and without fiber reinforcement. CLC blocks of
concrete panels that can be used in both low and high
dimension 600mm x 200mm x 150mm were cast with and
rise construction as it possess the comfort of light weight
without fiber reinforcement. Poly-olefin structural fiber
concrete and reliability of FRC. The purpose of this study is
reinforcement is used as reinforcement with dosages of
to investigate the behavior of fiber reinforced CLC blocks
2kg/m3, 3kg/m3, 4kg/m3 and 5kg/m3 of structural fibers
under flexure with different fiber dosages and understand
and another group with same dosages of structural fibers
the effectiveness of fibers on load-deflection behavior and
and 0.2kg/m3 dosage of fibrillated fiber. Five samples
ductility of the composite.
were tested under four-point bending for each dosage
under displacement control mode to understand its effect
on strength and ductility. Test results indicate that the Literature Review
addition of fibers improved the strength marginally and The concept of light weight concrete is not new. The building
good improvement in post-peak ductility behavior under structure of ‘The Pantheon’ of lightweight concrete material
flexure. The fibers enhanced the post peak behavior by is still standing in Rome until now for about 18 centuries. It
overtaking the load soon after the formation of crack and shows that the lighter materials can be used in concrete
resist the crack growth by pull out mechanism. construction and has an economical advantage. The light-
Keywords: CLC Block, Flexure, Hybrid-synthetic Fibers, weight concrete can be broadly categorized into three
Load-Displacement Curves, Fiber Dosage. groups: (i) No-fines concrete, (ii) Lightweight aggregate
concrete (iii) Aerated concrete. The aerated concrete is the
basis of CLC technology can be further classified Based on
Introduction method of pore formation such as (i) Air-entraining method
Cellular light weight concrete (CLC) is produced by mixing (gas concrete) (ii) Foaming method (foamed concrete) (iii)
cement, fly-ash, foam and water in required proportions Combined pore forming method. It can be further classified
using ready mix plant or ordinary concrete mixer. The based on type of binder used as (i) Cement based (ii) Lime
foam is pumped through specialized equipment that adds based and (iii) Pulverized fuel ash or slate waste as partial
fixed volume of air voids at constant pressure[1]. Millions replacement to binder. Moreover, it can be classified
of isolated tiny air bubbles with protein-hydrolyzed based on method of curing as (i) Non autoclaved aerated
covering is formed. The foam formation does not involve concrete and (ii) Autoclaved aerated concrete. Rudolph
any gas releasing chemical reaction and hence it does not and Valor[11] carried out tests on cellular concrete and
expand thereby maintaining constant density[2]. Previous suggested that flexure strength of CLC was 1/3 to 1/5 of
environment impact assessment (EIA) studies shows compressive strength. Sengupta J[12] used flyash as partial
that CLC technology is sustainable and produces a green replacement of binder and concluded that, utilizing flyash
building material[3], owing to its low direct CO2 emission to produce aerated concrete is an economically attractive
and due to usage of by-products from industries as raw proposition, which will help in mitigating the environmental
materials[4-5]. Flyash a by-product of industries, shows a damage caused by flyash. The usage of Polypropylene

Organised by
India Chapter of American Concrete Institute 213
 Table 1
Overview of salient literature pertaining to the structure and properties of aerated concrete

214
Reference Parameter studied

Method of Curing Micro- Chemical Salient features of the st udy

Ingredients Properties
Aeration method structure compostion

Functional
Binder Filler Gas Foam mc ac strength density shrinkage Porosity
proportion
Session 2 C - Paper 2

Valore RC 1954 C,L S         Review on propeties

Hoff GC 1972 C S     strength porosity relation

Mitsuda T 1977 C S    Anomalous tobermorite

Ziembika H 1977 C S    Micro pore-shrinkage

   Structure-Mechanical
Alexanderson 1979 C S
properties

Watson KL 1980 C,W S       Use of slate waste

Leitch FN 1980 C S   Fire resistance and acoustics

Schubert P 1983 C S     Shrinkage behavior

Tada S, Nakuno S 1983 C S    Micro and macrocapillaries

Prim P 1983 C S      Structure and water absorption

VermaCJ1983 L F      Lime-flyash cellular concrete

Tam CT 1987 C S    Strength-composition

Georgiades 1991 C S    Micropore-shrinkage

PospisilF1992 C F     Density change-pore structure

Tada S1992 C S     Pore structure, properties

Sengupta J 1992 C,L F      Flyash cellular concrete

Lauret JP 1995 C S    Thermal conductivity

Odler I, Robler M 1995 C Q    Particle size on properties

Haneck et,al 1997 C S       Carbonation

Durack JM 1998 C F     Strength-gel space ratio

Kearsley and Wainwrigth      Porosity compressive strength

C F
2002 relation

   Potential of CLC as structural

Jones and Macathy C S
material

2005 nambiar and      

C F Air-void characterisation
ramamurthy 2007

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Esmaily and Nuranian 2012 C    Alkali slag cellular concrete

Ameer et.al 2015 C S       Pore size distribution

 Flexural Behavior of Synthetic Fiber Reinforced Cellular Light Weight Concrete

fibers has gained more prominence in the recent years

for reinforcing cementitious materials[13-14]. Table 1 reports
a summary of previous research that has been done in
the past with respect to aerated concrete. C-cement,
L-lime, S-sand, F-flyash, Q-quartz, W-slate waste, mc-
moist curing, ac-autoclave curing. Usage of CLC blocks in
masonry construction has gained tremendous popularity in
Fig. 2: Poly-Olefin fibrillated fiber
recent decades owing to its sustainability, low density, low
thermal conductivity and less mortar joints in construction.
However, they have few disadvantages such as low tensile Table 2
and shear strength. This leads to very brittle failure under Physical Properties of Poly-Olefin Fiber [16]
lateral loads. Previous investigation have revealed that
Macro Fiber Fibrillated Fiber
addition of fiber has improved post-cracking features of
masonry, showing ductile behavior by arresting the crack Specification Bi-component fiber Interlinked fiber
propagation soon after the crack initiation. However, such
Material Poly-olefin Poly-olefin
studies in CLC masonry is very scarce and needs attention
to better understand the fracture behaviour under flexure Form Structural fiber Fibrillated fiber
and shear. Tests carried out by Ronald and Carol[15]
Specific Gravity 0.91 0.91
indicates the ability of fiber reinforcement to transform the
basic material character of cellular concrete from brittle Length 50mm 19mm
to ductile elasto- plastic behavior The authors found that
Tensile strength 618 N/mm2 400 N/mm2
the performance of the fiber reinforced CLC was better
compared to the control ones. Critical review of literature Modulus of Elasticity 10 GPa 4.9 GPa
indicates that only a handful of studies have focused on
Diameter 0.5mm 0.08mm
fiber reinforced CLC for structural applications of masonry.
Improved compression, shear and tensile resistance can
be expected with hybrid addition of structural/macro fibers Mixing and Curing
along with micro-fibers for superior crack resistance at The dry ingredients i.e., cement, flyash were fed into
both micro and macro levels. Hence, the purpose of this the mixer first and thoroughly mixed to ensure even
study is three-fold (i) to develop low cost fiber reinforced distribution of cement. Thereafter, water was added and
CLC blocks for masonry applications and (ii) to investigate the mixing process continued. The preformed foam[17]
their mechanical properties under flexure with different was added at 35 gm/ sec for 40 seconds to the slurry of
fiber dosages and (iii) to understand the effectiveness of cement, flyash [18] and water in the batch mixer. After an
fibers on energy dissipation capacity (toughness index) of additional mixing of 3 minutes along with fibers to get
the developed CLC blocks. uniform consistency, the slurry form CLC was poured into
dimension 600 mm length, 150 width and 200 mm height
rectangular moulds. Specimens were demoulded after 24
Experimental Program hours and curing was done as per IS-456 2000[19]. After
curing the blocks for 28 days, the testing was done under
Materials
displacement control.
The materials used for the non-fibrous control CLC
mixture consisted of 53 grade Ordinary Portland Cement,
Flyash from NTPC (National Thermal Power Corporation),
potable water and sunlite foam. The mix proportion of
flyash: cement: water: foam was 833: 277:277:1.4 kg/m3.
The additives are coarse bi- component macrofiber and
fibrillated fibre as shown in the Figure 1. The physical
properties of fibers are mentioned in Table 1.

Fig. 1: Poly-Olefin Macrofiber Fig. 3: Manufacturing of CLC blocks

Organised by
India Chapter of American Concrete Institute 215
Session 2 C - Paper 2

Distribution of fibers
The complexity involved in mixing fibres with slurry like
paste is lesser due to the absence of coarse aggregate
in the CLC. Moreover, mixing process is carried out in
a mechanical drum mixer which ensures the uniform
distribution of fibers. Segregation of fibers is avoided as
the mix is having same density as that of fibers. Table
3 gives the fiber count of macro-fibers on the cores of
dimension 100mm diameter and 200mm height extracted
from the blocks. Due to continuous interlinked form,
the micro-fiber count could not be estimated. However,
uniform distribution was observed during mixing and (a) Specimen ready for testing
manufacturing of the CLC blocks. As expected, the count
of fibers increased with increase in dosage of fibers.
(Table 3).

Table 3
Distribution of Poly-Olefin Macro-Fiber from Cored Samples

Fiber Dosage Sample Sample Sample Average

Specimen
(kg/m3) -1 -2 -3 count

Control
0 0 0 0 0
Specimen
(b) Failed specimen under flexure
ma-0.22-
2 127 112 118 119
mi-0.0
Fig. 5: Testing of CLC blocks under Flexure
ma-0.33-
3 236 233 257 242
mi-0.0
LVDTs mounted on the specimen as shown in Figure 5. All
ma-0.44-
mi-0.0
4 330 342 313 328 the measurements were collected through Controls data
acquisition software.
ma-0.55-
5 426 450 436 437
mi-0.0
Results and Discussions
Test Method Slump
The testing code for fiber reinforced CLC blocks under The CLC mix which flows into the moulds like self-
flexure are not available yet, though ASTM C1609/C[20] and compacting concrete, remained unaffected by addition of
JSC SF-4[21] was used as a guideline to establish load- fibers and showed equally good mobility into the moulds
deflection curve of flexure specimen The specimen was on reinforcing with fibers. This can be attributed to free
tested using servo controlled hydraulic testing machine movement of air voids around the fibers which is restricted
and loading was increased at a rate of 0.1 kN/sec for 90% of had there been the coarse aggregate of normal concrete.
the peak load and then 0.001mm/sec displacement control
loading was used for the rest of the test. All the specimen Stiffness and Peak strength
were tested in third-point loading as shown in Figure 4 ,
Figure 6, 7 shows the load-displacement response of CLC
using rigid steel mechanism on a servo controlled flexure
blocks under with different fiber dosages. The progressive
testing machine. Loads were measured using the load cell
increase in peak flexural load of CLC blocks as the dosage
of the frame and displacements were measured using
of fiber increases is shown in observed from the Table
4. Number of specimen tested under flexure and the
corresponding scatter of results have been included in
Table 5
Figs. 8, 9 shows the close up view of load-displacement
curve. The increase in ductility can be observed from the
figures. To account for the quantitative measurement of
ductility, Re,3.6 factor from JSC SF-4[21] has been used.

Fig. 4: Third point loading scenario for a simply supported beam

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

216 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Flexural Behavior of Synthetic Fiber Reinforced Cellular Light Weight Concrete

Fig. 6: Full behavior of CLC blocks without microfiber dosage Fig. 8: Close-up look at behavior of CLC blocks without
in flexure microfiber dosage in flexure

Fig. 7: Full behavior of CLC blocks with microfiber dosage in Fig. 9: Close-up look at behavior of CLC blocks with microfiber
flexure dosage in flexure

Table 4
Average peak flexural load and Re,3.6 values for corresponding fiber dosage

Volume Fraction of Macro/ Increase in fct due to

Specimen fct (kN) Re,3.6 value % increase in Re, 3.6
Micro Fibers addition of fibers (%)
Control Specimen 0/0 6.2967 - 0.0445 -
ma-2-mi-0.0 0.22/0 7.0340 11.7 0.5492 11.34
ma-3-mi-0.0 0.33/0 8.1910 30.1 0.6514 13.64
ma-4-mi-0.0 0.44/0 7.9879 46.7 0.6729 14.12
ma-5-mi-0.0 0.55/0 8.5944 59.3 0.8014 17.00
ma-1-mi-0.2 0.11/0.02 9.4361 14.6 0.3847 7.65
ma-2-mi-0.2 0.22/0.02 10.6784 36.6 0.5563 11.51
ma-3-mi-0.2 0.33/0.02 9.2357 49.8 0.6637 13.91
ma-4-mi-0.2 0.44/0.02 10.0313 69.6 0.7259 15.31

Table 5
Average peak flexural load and Re,3.6 values for corresponding fiber dosage

Peak Flexural strength in kN Std. Dev (kN)

Series Specimen Mean Fct (kN)
1 2 3 4 5
I Control 6.234 7.016 5.735 6.963 5.5405 6.297 0.680
ma-2-mi-0.0 8.402 7.092 6.780 6.951 5.9450 7.034 0.884

II ma-3-mi-0.0 7.846 8.361 7.569 8.793 8.386 8.191 0.483

(only Macro) ma-4-mi-0.0 8.036 9.748 10.391 8.745 9.2585 9.235 0.905
ma-5-mi-0.0 10.674 11.277 8.975 9.172 10.0585 10.031 0.977
ma-1-mi-0.2 9.372 7.245 8.294 7.650 7.3785 7.988 0.873

III ma-2-mi-0.2 8.428 7.882 8.902 9.287 8.473 8.594 0.530

(hybrid) ma-3-mi-0.2 9.815 8.777 7.844 10.871 9.8735 9.436 1.158
ma-4-mi-0.2 10.586 8.517 9.186 12.588 12.515 10.678 1.865

Organised by
India Chapter of American Concrete Institute 217
Session 2 C - Paper 2

ll Increase in stiffness and the peak flexural load

resulted in the increment under the area of load-
displacement curve which led to increase in toughness
index. The Re,3.6 values increased up to 11.34% for 2 kg/
m3 and upto 17% for 5 kg/m3. It increased up to 7.65%
for 1kg/m3 and upto 15.31% for 4 kg/m3 with constant
microfiber dosage of 0.2 kg/m3. This can be attributed
to the synergetic role played by fibers in bridging the
(a)Control specimen (b)Macro-fiber (c) Macro and cracks.
reinforced specimen fibrillated reinforced

Fig. 7: Failure of blocks under flexure with and without fibers Acknowledgments
This experimental work is carried out as part of the
Failure Modes project funded by Ramanujan fellowship grant sponsored
The failure pattern followed by unreinforced specimen is by Department of Science and Technology, India. Their
predominantly a single explicit crack as shown in Figure financial support is gratefully acknowledged. Fiber
10a. Thereafter the beam fails. On the other hand, the reinforcement used in this study was donated by Brugg
FRCLC blocks show post-peak resistance to the opening Contec AG. We also acknowledge Srinivasa CLC block
of the large crack at the failure as shown in Figure 10b, plant, Hyderabad India for helping with mixing and casting
10c. The fibers in the matrix form a closed network of CLC blocks used in this study.
which hinders the formation of crack. Even when the
crack forms, fibers in the matrix bridge the crack and References
slow down further crack propagation. Thus, the fibers in 1. Narayanan, N., Ramamurthy, K., (2000) “Structure and properties
of aerated concrete: a review”, Cement & Concrete Composites
the post peak region continue to take load due to strain 22, 321-329
localization. However the serviceability criteria may
2. Vine-Lott, K., (1985) “Production of foam concrete by microcomputer”,
restrict the amount of deflection undergone in the post The Concrete Society, volume 19, page no. 12-14, ISSN: 0010-5317.
peak region.
3. Satheeshbabu, S., (2010) “Life cycle assessment of cellular
lightweight concrete block – a green building material,” Journal of
Environmental Technology and Management (1), ISSN: 2010-1554
Conclusions
4. Hassan, K.E., Cabrera, J.G., Bajracharya, Y.M., (1997) “The Influence
CLC is sustainable, light in weight and has good acoustic of Fly Ash Content and Curing Temperature on the Properties of
and thermal insulation. The performance of CLC can be High Performance Concrete”, in Proceedings of the 5th International
enhanced with regard to ductility and inelastic behavior Conference on Deterioration and Repair of Reinforced Concrete in
the Arabian Gulf, Bahrain pp. 311–319.
by addition of fibre reinforcement. Fiber reinforced CLC
can be used for developing low cost masonry blocks 5. Stuart, K.D., Anderson, D.A., Cady, P.D., (1988) “Compressive
strength studies on portland cement mortars containing fly ash
for structural applications. Developing fiber reinforced and superplasticizers,” Cem. Concr. Res. 10, 823-832.
CLC for masonry applications was explored through
6. Kearsley, E.P., Wainwright, P.J., (2002) “Ash content for optimum
addition of macro-fiber reinforcement and hybrid-fiber strength of foamed concrete,” Cement and Concrete Research 32,
reinforcement. The effect of synthetic fiber reinforcement 241-246
addition on the flexural behavior was studied by testing 7. Krishna Bhavani Siram, K.,(December 2012) “Cellular Light-
beams in flexure. Based on this study, the following Weight Concrete Blocks as a Replacement of Burnt Clay Bricks,”
conclusions can be drawn: International Journal of Engineering and advanced Technology
(IJEAT) ISSN: 2249 – 8958, Volume-2, Issue-2.
ll Flexural strength increased with increase in fiber 8. IS: 6598 (1972), Cellular concrete for thermal insulation, IS: 6598,
dosage when compared to unreinforced CLC blocks. New Delhi, India.
Their performance was further enhanced with addition 9. Bentur A, Mindess S. (2007) Fiber reinforced cementitious
of microfibers. A positive fiber synergy was observed composites. 2nd ed. Taylor and Francis; [601 pp.].
between macro and micro-fibers under flexural 10. Hsie, M., Tu, C., Song, P.S., (2008) “Mechanical properties of
behavior. polypropylene hybrid fiber-reinforced concrete,” Materials Science
and Engineering A 494, 153–157
ll Due to addition of macro-fibers, the flexural strength
11. Valore, R. C., (1954) “Cellular Concretes-Physical Properties,”
increased upto 11.7% for 2 kg/m3 and upto 59.3% for Journal of the American Concrete Institute, No. 25, pp.817-836.
5 kg/m3. With further addition of micro-fibers of 0.2
12. Sengupta J. (1992) Development and application of light weight
kg/m3, the flexural strength increased upto 14.6% for aerated concrete blocks from fly ash. Indian Concr J; 66:383±7.
1 kg/m3 and upto 69.6% for 4 kg/m3. This indicates that
13. Perez-Pena, M., Mobasher, B. (1994) “Mechanical properties of fiber
the hybrid reinforced specimens performed better reinforced lightweight concrete composites”, Cem. Concr. Res. 24
compared to the specimen with only macro structural (6), 1121–1132.
fibers. 14. Qian, C.X., Stroeven, P. (2000) “Development of hybrid polypropylene-

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

218 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Flexural Behavior of Synthetic Fiber Reinforced Cellular Light Weight Concrete

steel fibre-reinforced concrete”, Cem. Concr. Res. 31 (1) 63–69. Lightweight Concrete Containing Pozzolan Materials,” The Twelfth
East Asia-Pacific Conference on Structural Engineering and
15. Ronald, F., Carol, D. H., (1998)“Engineering Material Properties of
Construction, Procedia Engineering 14, 1157–1164
a Fiber Reinforced Cellular Concrete,” Journal of the American
Concrete Institute, No. 95-M61, pp.631-635. 19. IS: 456:(2000), Plain and Reinforced Concrete-Code of Practice
(Fourth Revision), India.
16. Concrix-Technical Datasheet, Brugg Contec AG, CH-8590
Romanshorn . http://w w w.bruggcontec.com/English/Home/ 20. ASTM C 1609/C 1609M – 07, (December, 2007) “Standard Test
Concrix/tabid/474/language/en- US/Default.aspx Method for Flexural Performance of Fiber-Reinforced Concrete
(Using Beam With Third-Point Loading)”, Annual Book ASTM
17. ASTM C 869-91 (1992) Standard specification for foaming agents
Standards, American Standards for Testing Materials.
used in making preformed foam for cellular concrete, American
Standards for Testing Materials. 21. JSCE-SF4, (1984) “Method of Test for Flexural Strength and Flexural
Toughness of Steel Fiber Reinforced Concrete”, Concrete Library,
18. Jitchaiyaphum, K., Sinsiri T, Chindaprasirt, P. (2011) “Cellular
JSCE No 3.

Dr. Suriya Prakash

Dr. Suriya Prakash’s research expertise on structural concrete behaviour and design. He is currently
working as assistant professor at IIT Hyderabad. Before joining IITH, he worked with STRUCTURAL Inc,
USA a renowned firm in strengthening design and construction using advanced construction materials.
He has designed strengthening solutions for several buildings in the US and middle east. Before that,
he had completed his PhD from Missouri University of Science and Technology, USA. He has authored
more than twenty journal papers on behaviour of reinforced concrete columns and strengthening with FRP
composites. He is a member of ASCE and ACI, USA.

Organised by
India Chapter of American Concrete Institute 219
Session 2 C - Paper 3

Development of Self-Compacting Concrete using Potable Water

Treatment Sludge as a Costless Self-Curing Agent
Rampradheep G. S.
Assistant Professor, Department of Civil Engineering, Kongu Engineering College, Erode, Tamilnadu, India - 638 052.
gsramcivil34@gmail.com, 9750033500.
Dr.Sivaraja M.
Principal, N.S.N. College of Engineering and Technology, karur, Tamilnadu,India- 639 603.
Arunkumar N., Gopinath K., Dhanasekaran M., Saranya R.
UG Student, Department of Civil Engineering, Kongu Engineering College, Erode, Tamilnadu, India - 638 052.

Abstract This high cement paste volume works as lubricant for

coarse aggregates but this high cement paste volume
Experimental study made on Potable Water Treatment
end up with high shrinkage cracks and prone to high
Sludge (PWTS) as a Self-Curing agent in Self-Compacting
heat generation during hydration of cement paste[2]. The
Concrete is presented in this paper. The major concept
high water content in concrete results in segregation
behind the usage of PWTS as a Self-Curing agent is
of materials and therefore Super plasticizers (SP) and
due to the presence of bond water in it and its calcium
Viscosity Modifying Agents (VMA) are employed to reduce
carbonate content. Results from using other commonly
liquid limit. It also enhances resistance to segregation of
used curing agents such as Poly-Ethylene Glycol (PEG),
concrete matrix [2].
Super Absorbent Polymers (SAP), Light Weight Aggregate
(LWA) and External Coating Paint (ECP) are compared Crack formed in the concrete due to shrinkage reduces
with results of PWTS. Different tests were conducted the durability of concrete[3]. Providing internal curing
to determine the workability properties, mechanical agent releases moisture slowly day by day and may
properties, microstructural properties and durability extend for weeks[4]. Cement paste getting hydrated
properties of the concrete. Test such as compression shrinks due to two major mechanisms namely drying
test, split tensile test and flexural test were conducted shrinkage and autogenous shrinkage. Drying shrinkage
to determine the mechanical properties of concrete. Test can be reduced by proper curing but were we are facing
such as acid attack test, rapid chloride penetration test, many difficulties practically. Self-Curing agents reduces
shrinkage test were conducted to determine the durability the early age shrinkage[5]. Curing of concrete plays a
of concrete. Test on bond of PWTS with concrete matrix major role in the development of microstructures and
and dispersion of curing agents in concrete were carried pore structures in the concrete. The concept behind the
out using Scanning Electron Microscope (SEM) and Self-Curing concrete is to increase the water retention
X-Ray Diffraction (XRD). Results shows that PWTS is an capacity of concrete by controlling the evaporation of
effective curing agent and improves the cement hydration, water from the concrete[6,7]. Usage of Self-curing agent
compressive strength, reduces shrinkage and enhance in the concrete results in better hydration with time,
the durability of concrete. Their results are found to be under drying conditions compared to the conventional
satisfactory as per IS standards, EFNARC specifications concrete [8]. According to ACI 308 committee, internal
and ASTM codal provisions. curing is the process which occurs due to availability of
additional water that is not the part of water added for
Keywords: Self-Compacting concrete; Self-Curing agent;
mixing[9]. The concrete mix with high water-cement ratio,
Potable Water Treatment Sludge; Compressive Strength;
the additional water for curing is supplied by external
Relative Humidity; Shrinkage.
curing source but in concrete with low water-cement
ratio the permeability of concrete soon becomes too
Introduction low and efficiency of curing through external water (i.e.
Concrete, as a composite construction material, is surface curing) gets reduced [10].
generally composed of cement, sand, aggregate, water, New Self-curing agents in form of chemical admixtures
mineral admixtures and chemical admixtures. Concrete and aggregates are introduced into concrete by many
is the most durable material when it is made free from researchers to enhance the self-curing process and to
cracks and material of low permeability. The Self- solve the problems faced by improper curing. Some self-
Compacting Concrete (SCC) are used to achieve high curing agents such Poly-Ethylene Glycol (PEG), Super
strength with less man power and it must have high Absorbent polymers (SAP), Light Weight Aggregate (LWA)
cement paste volume and so powder content should be are found to be very effective in Self-Curing and increases
higher in SCC than conventional concrete, in this case physical and microstructure properties resulting in high
high water-powder ratio is adopted to reduce viscosity [1]. durability of concrete [4,11].

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

220 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Development of Self-Compacting Concrete using Potable Water Treatment Sludge as a Costless Self-Curing Agent

Background on Potable Water Treatment Table 1

Sludge Properties of Materials Used
Various processes are involved in treatment of drinking
Material Type Materials Used Specification
water/potable water used for drinking and for industries.
Hardness present in the drinking water produces Bonding Material Cement – OPC IS: 12269–1987 [17]
cardiovascular disease and some other diseases are
correlated with hardness of water such as anencephaly and Coarse Aggregate
Crushed Aggregate IS: 383–1970 [18]
(C.A.)
various type of cancer [12-14]. Hardness is a major problem
in water used in boilers, which causes scum and scale Fine Aggregate River Sand –Grading zone
IS: 383–1970 [18]
formation in walls of boiler tanks. Hardness is caused by (F.A.) III
dissolved carbonates and bicarbonates of calcium and
Water Water IS: 456-2000
magnesium. Most Commonly used process to remove
hardness form water is Lime softening process. In this Super Plasticizer EFNARC
CONPLAST SP 430
process first step is addition of quick lime or CaO, which (SP) specifications [19]
reacts with H2O to form calcium hydroxide Ca(OH)2. The
Viscosity Modifying EFNARC
calcium present in water reacts with dissolved CO2 to form Glenium Stream B233
Agent (VMA) specifications [20]
a calcium carbonate (CaCO3) precipitate, when the pH is
maintained around 10. Similarly magnesium precipitates Super Absorbent Polymer
Self-Curing Agent -
in the form of magnesium hydroxide Mg(OH)2 at pH level (SAP)
around 11.3. This pH can be attained by adding additional lime
Light Weight Aggregate
to the water. So, the PWTS contains high calcium content Self-Curing Agent -
(LWA) - Vermiculate
which would increase the bond strength of the concrete.
The fine particles of calcium carbonate and magnesium Self-Curing Agent Poly-Ethylene Glycol (PEG) -
hydroxide precipitates are settled in clariflocculator by
External Coating Paint
adding coagulants which induces the flocculation process. Self-Curing Agent -
(ECP)
The PWTS obtained from potable/ drinking water Self-Curing Agent PWTS -
treatment process is in semi-solid state containing large
amount of water content in it. The water present in sludge
is broadly classified as bulk water and bond water. Bond
Table 2
water is further classified as interstitial water and vicinal Chemical Composition of Potable Water Treatment Sludge
water [15]. The bulk water can be easily removed from PWTS
by adopting proper drying process while the bond water Treated Waste Type – Sludge based (semi-solid)
can’t be removed by simple drying process. The interstitial
Ca 46.03 %
water is the water molecules found between the flocs,
which can be removed breaking the floc structure. The O 38.10 %
vicinal water is the water molecules bond to the surface C 6.66 %
of sludge particles by hydrogen bonding. These bond
Si 3.97 %
waters in sludge can be used efficiently used for internal
curing of the concrete and it reduces drying shrinkage and Fe 3.16 %
autogenous shrinkage in the concrete[16]. S 0.89 %

Al 0.59 %
Experimental Investigation
Mg 0.35 %
Materials Used Na 0.26 %

 Specific Gravity 2.55

Mix Proportion
The mix composition was chosen to satisfy all performance
criteria of the concrete in both fresh and hardened
states. Here, the four trail mixes were adopted as per
EFNARC specification [19] to attain medium strength
Self-Compacting concrete of grade M40. In this study
different percentage of fly ash content were adopted for
each trail mix. The different percentage of fine aggregate
Fig. 1: X-ray Diffraction Graph – Potable Water Treatment Sludge with respect to total weight of aggregate was adopted for

Organised by
India Chapter of American Concrete Institute 221
Session 2 C - Paper 3

each trail mix. The percentage of SP and VMA added to all

Table 4
mix combinations were kept constant. The proportion of Test Combinations
materials used for each trail mix is shown in the Table-3.
The main constituent parameters in the study were the
Chemical dosage
different percentages of five Self-curing agents. PEG, Test
S.No
SAP, PWTS, LWA, ECP were added in three different Combination
Appropriate Self-Curing agent SP VMA
percentages. The each different percentage of five Self-
curing agents was added to all four trail mix to determine
1. SP + VMA - 2% 2%
the best trail mix combination. So, totally sixty trail mix
combinations were tested in this study. Table-4 shows the
different test combinations adopted. For each four trail 2. PEG + SP + VMA 0.02 % 0.05 % 0.07 % 2% 2%
mix conventional concrete was cast without adding Self-
Curing agent and cured by conventional method. SP, VMA, 3. PWTS + SP + VMA 0.50 % 1.00 % 1.50 % 2% 2%
PEG, SAP were added to the concrete mix as a percentage
of weight of cement. PWTS and LWA were added as partial 4. ECP + SP + VMA 0.2 mm 0.50 mm 1.00 mm 2% 2%
replacement of fine aggregate in concrete.
5. LWA + SP + VMA 0.30 % 0.35 % 0.40 % 2% 2%
Table 3
General Mix Composition 6. SAP + SP + VMA 5% 10% 15% 2% 2%

Trial Cement Fly Ash F.A. C.A. Water

SP VMA
Mix (Kg/m3) (Kg/m3) (Kg/m3) (Kg/m3) (l/m3) Specimens and Testing Methods
1 435 145 720 760 180 2% 2% The cubes of size 150mm were cast to conduct
Compressive strength test, Relative humidity test and
2 445 155 740 810 180 2% 2% Acid attack test. The cylinders of diameter 150 mm and
3 470 130 730 760 180 2% 2%
height 300 mm were cast to conduct Split Tensile test. The
beams of size 500 mm X 100 mm X 100 mm were cast to
4 460 140 790 720 180 2% 2% test the flexural strength at 7th, 14th, 28th day.

Table 2
Testing Methods

Test Property Test Name Apparatus / Instruments Used Specification

Slump flow test Slump cone

U-Box test U-Box

Workability Tests EFNARC specifications [19]
L-Box test L-Box

V-Funnel test V-Funnel

Compressive Strength HEICO compression testing machine

IS: 516–1959 [21]& IS: 5816 – 1970 [22]
Mechanical Split tensile strength Two point loading system
Property T ests
Vaisala Structural Humidity Measurement Kit
Flexural strength ASTM F2170 [23]
SHM40

Relative Humidity Test Scanning Electron Microscope -

Microstructure
Property Tests AASHTO T277-83 & ASTM C1202
Scanning Electron Microscope (SEM) Techno care Rapid Chloride Penetration Setup
specifications [24, 25]

Specimens dipped in the container with diluted Visual observation & Strength
Rapid chloride Penetration Test
acid Deterioration Factor. (SDF)
Durability Tests
Acid Attack Test – 5% H2SO4 Autoclave Apparatus -

Drying shrinkage Test

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

222 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Development of Self-Compacting Concrete using Potable Water Treatment Sludge as a Costless Self-Curing Agent

Results and Discussion Trail mixes. The minimum time taken for flow in Trail-4 is
6.5 sec, whereas minimum flow time in Trail-1, 2, 3 was
Workability Test Results 8 sec, 8.5 sec and 7 sec respectively. In Trail-4 out of 16
Figure 2. shows the graph obtained from plotting the combinations all percentages of PEG, PWTS, LWA, SAP
values determined from slump flow test for different Trail shows good flowing property but the conventional mix and
mixes. The Y-axis represents the slump flow length in ECP mixes shows low flowing property compared to other
mm and the X-axis represents the Trail mixes. This test trail combinations.
determines the unconfined flow potential of the Self- Figure 4. shows the graph of L-Box test. Y-axis in the
Compacting concrete[26]. The test is carried out as per the graph represents the H2 / H1 ratio of the L-Box while X-axis
ASTM C1611 / C1611M standards[27]. If the slump value of represents the four Trail mixes. The different colour bars
concrete increases then the flow potential of the concrete in the graph represent the different percentage of five
is high and vice versa Figure 2. shows that the Trial-4 has Self-curing agent mix combinations. Trail combinations
high slump value when compared to the other trail mixes. in Trail-4 has high H2 / H1 ratio which proves that Trail-4
The peak value attained in the Trail-4 test combination has good flowing nature compared to other three trail
of PEG 0.07% and PWTS 1.5% whose slump value was mixes. The maximum H2 / H1 ratio obtained in the Trail-4 is
745mm. This slump value is higher when compared with 0.93 were as maximum ratio obtained in Trail-1, 2, 3 was
the conventional mix. Trail-2 mix has the lowest slump 0.88, 0.88, 0.92 respectively. This again shows that Trail-4
value of abut 695mm which is 6% lesser than the Trail-4 mixes has high workability compared to other mixes due
mix. to the high paste volume and high percentage of F.A. in it.
In Trail-4 the different percentage of chemical mixtures
slightly various the H2 / H1 ratio in which PWTS 1.5% and
LWA 15% each showed 0.93 which is higher compared to
conventional mix and mixes with ECP because they both
have no extra chemical other than SP and VMA to affect
the workability.

Fig. 2: Slump Flow Test Graph

Fig. 4: L Box Test Graph

Figure 5. shows the graph for U-Box test conducted on

sixty four different trail mix combination with different
five Self-Curing agents added. Y-axis in graph represents
Fig. 3: V Funnel Test Graph the value of difference in level of concrete in two different
limbs of U-Box (H1-H2) while X-axis represents different
Figure 3. Shows the graph of V-Funnel test carried trail mixes. Each different colour bar represents the
out to find the workability of concrete. In graph Y-axis various test combinations of five Self-curing agents. In
represents the time in seconds that concrete taken to flow four different trail mixes, Trail-4 shows lesser values
down the through the V-Funnel and X-axis represents the compared to other three trail mixes because of the higher
Trail mixes. The bars in the graph represent the different paste volume which lubricates the coarse aggregate flow
combinations of Self-curing agent in each Trail mix. In in the Trail-4 mix compared to other trail mix[2]. The Trail-4
Trail-1-3 concrete took little higher time to flow when mix has 52% of F.A. of total volume of aggregates, but
compared to the Trail-4. The flow time of the V-funnel Trail-1, 2, 3 has 48%, 47% and 49% of F.A. of total volume
test is to some degree related to the plastic viscosity[26]. of aggregate respectively. Fig.9. shows clearly that flow
The time taken to flow through the V-Funnel is directly increases with increase in F.A. percentage. In Trail-4,
proportional to the plastic viscosity of concrete. If time test combinations with higher percentage of chemical
taken for flow is high then the viscosity of the concrete content have lesser U-Box value due to increase in flow
is high and vice versa. Hence, the Trail-4 mix of all which occurred due the reduction in viscosity by these
combinations has low viscosity compared to other three chemicals. In Trail-4, PEG 0.07% and PWTS 1.5% have
higher flow than the conventional mix of Trail-4.

Organised by
India Chapter of American Concrete Institute 223
Session 2 C - Paper 3

Fig. 8: Flexural Strength of Trail-4 mix with Optimum Dosage of

Self-Curing Agents at 28th Day

Fig. 5: U-Box Test Graph

complete hydration of cement paste have occurred
when the specimen is placed inside the water and good
Mechanical Property Test Results microstructure properties were developed in those
The optimized values of compressive strength, split specimens. The compressive strength of specimens
tensile strength and flexural strength obtained from cured by Self-Curing agent such as PEG 0.05% was
the graph plotted for three different dosages of five 1.4% lesser than the compressive strength attained
Self-Curing agents added to each of the four Trial mixes by conventional curing method. Similarly compressive
are tabulated. The graphs are plotted for compressive strength of specimen with Self-Curing agents such as
strength, split tensile strength and flexural strength PWTS 1%, ECP 1.0mm, LWA 5% and SAP 0.35% were
values of the conventionally cured concrete, PEG, PWTS, lesser than compressive strength of conventionally cured
ECP, LWA, SAP respectively for 7th day, 14th day and 28th concrete by 2.92%, 4.48%, 4.28% and 3.39% respectively.
day results are separately tabulated. From each separate These compressive strength values various only by lesser
graph plotted, Trail mix with optimum dosage of each percentage. The strength achieved by these Self-Curing
Self-Curing agents which gives maximum strength was agents was up to the considerable range, since they have
analysed and determined. It was found that the Trail-4 achieved 92%-95% of target mean strength of the mix
mix gives maximum mechanical strength for all Self- design. Good microstructure and pore structure formed
curing agents added. So, the compressive strength, split in concrete will results in high compressive strength. This
tensile and flexural strength at 28th day of Trail-4 mix shows that internal curing and hydration of cement paste
with optimized dosage of each Self-Curing agent were were enhanced by the Self-Curing agents added without
obtained. any supply of external water for curing. The Trail-4 mix
with PWTS 1% has achieved 94% of target mean strength
of the design mix which is higher than other Self-Curing
agents added. This shows that PWTS added to concrete
has effectively involved in increasing the bond strength
due to presence of high calcium content in it.
Figure 7. & Figure 8. shows the Split Tensile strength and
Flexural strength of the Trail-4 mix with different Self-
Fig. 6: Compressive Strength of Trail-4 mix with Optimum Curing agents. The Split Tensile strength and Flexural
Dosage of Self-Curing Agents at 28th Day strength of the Trail-4 mix with PWTS 1% was 3.38%
and 2% lesser than the conventionally cured concrete of
Trail-4 mix respectively. However strength was achieved
to considerable range.

Microstructure Properties Test Results

Relative Humidity Test Results
Figure 9. shows the results of Relative Humidity(RH) test
Fig. 7: Split Tensile Strength of Trail-4 mix with Optimum Dosage carried on concrete specimen of Trail-4 mix with optimum
of Self-Curing Agents at 28th Day

Figure 6.shows the graph in which Y-axis represents

the compressive strength at 28th day and the X-axis
represents the Trail-4 mix with optimum dosage of five
different Self-Curing agents. Concrete specimen cured
conventionally by immersing into the water shows the
maximum compressive strength when compared to
Fig. 9: Relative Humidity of Trail-4 mix Specimens with Optimum
any other Self-cured concrete. This is because that Dosage of Self-Curing Agents

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

224 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Development of Self-Compacting Concrete using Potable Water Treatment Sludge as a Costless Self-Curing Agent

dosage of five different Self-Curing agents added. RH SEM photo of conventionally cured concrete of Trail-4
inside the concrete is directly related to the availability of mix. SEM photo of concrete with PWTS and conventionally
water in the concrete for curing. The RH of all concrete cured concrete were compared. It was found that the
mix is more over same on the first day after 28 days of microstructure formed in the concrete matrix with PWTS
curing. As the day passes, RH of the conventional concrete was more over similar to that of the conventionally cured
with no Self-Curing agent falls at higher rate than RH of concrete matrix. Hence, the PWTS has the tendency
all other specimen with Self-Curing agent. The concrete of getting distributed within the concrete matrix and to
with Self-Curing agent releases moisture at slow rate [4]. effectively act as Self-Curing agent.
This increases the durability of the concrete and hence
high performance is achieved from the concrete. Here the
Durability Test Results
specimen with PWTS 1.0% added as Self-Curing agent has
RH of 88.4% at 28th day which was little lesser than the Rapid Chloride Penetration Test (RCPT) Results
specimens with PEG 0.05%, SAP 0.35%, LWA 5%. But the
difference in RH between PWTS and other specimen was
around 0.5%-2% which was negligible. This good relative
RH level maintained results in high degree of hydration
in concrete and develops the microstructural properties.
So, the PWTS also effectively acts as Self-Curing agent.
Relative Humidity is usually affected by water-cement
ratio and cementitious materials content [28, 29]. But
in this study there is no difference in RH between any of Fig. 12: Penetration Value Graph for Trail-4 mix with Optimum
the Trail mixes due to cement content and water-cement Dosage of different Self-Curing Agents
ratio because the water-cement ratio and cement content
where kept constant in all Trail mixes. Figure 12. shows results obtained from the Rapid Chloride
Penetration Test conducted on Trial-4 mix with optimized
percentage of different Self-Curing agents. Chloride
penetration into the concrete structures in under water
construction or structures near the industries producing
chloride as a waste product was high. Chloride is the major
reason for corrosion of reinforcement in the concrete
which reduces the durability of concrete. Reduction in
water-cement ratio reduces the penetration of chloride
Fig. 10: Relative Humidity Apparatus into concrete and also introducing cementitious materials
to reduce the penetration of chloride into the concrete
Scanning Electron Microscope Analysis [30]. In all the Trail mixes water-cement ratio was kept
Figure 11(b). shows the SEM photo of a concrete with minimum i.e. water- cement ratio was 0.3. As a result
PWTS. Concrete matrix after 28 days curing was of adopting 0.3 water-cement ratio and using fly ash as
examined by means of SEM. C-S-H gel formation is the mineral admixture, penetration value comes under the
important phenomenon to achieve high strength. Figure category of low and moderate penetration as per ASTM
11(b). shows that C-S-H gel formation in concrete with C1202 specifications for Trail-4 mix with optimum dosage
PWTS is high because of calcium content present in the of different Self-curing agents added. In Rapid Chloride
concrete. The distribution PWTS in the concrete matrix Penetration test the amount of current passed through
was homogeneous and hence the curing occurs uniformly the specimen is directly proportional to the penetration of
throughout the concrete matrix. Figure 11(a). shows the chloride into the concrete. Trail-4 mix with PWTS 1.0% as
Self- Curing agent shows penetration value of 1834.26C
which is higher than the penetration in the conventionally
cured concrete, PEG, LWA and SAP having difference
around 291.02C - 637.94C. However, according to ASTM
C1202 specifications the penetration value of PWTS was
in category of low penetration [25]. So, the PWTS doesn’t
affect the durability of concrete.

Acid Attack Test Results

Figure 13. shows the results of Acid Attack Test conducted
on Trial-4 mix with optimized percentage of different Self-
(a) Conventional Specimen (b) PWTS Specimen Curing agents. From the results of visual scale it was
Fig. 11: Scanning Electron Microscope Images clear that all the specimens of different test combinations

Organised by
India Chapter of American Concrete Institute 225
Session 2 C - Paper 3

ll It had been optimized from Mechanical strength

results that 1.0% of PWTS as an effective percentage
for self-curing.
ll Water holding capacity of concrete have got increased
by addition of PWTS, as found by Relative Humidity
analysis. The Relative Humidity of concrete with other
Fig. 13: Acid Attack Test Graph for Trail-4 mix with Optimum
Self-Curing agents was little higher than concrete with
Dosage of different Self-Curing Agents
PWTS. Although, the difference in Relative Humidity
was negligible.
kept in H2SO4 acid for 30 days shows visual scale reading
of 3 which is categorized under moderate attack. The ll Introduction of PWTS into concrete results in good
compressive strength obtained after 30 days of immersion Microstructural properties. It enhances the formation
of Trail-4 mix with PWTS 1.0% shows 10.69% deterioration. of calcite due to high calcium content in it, as found by
The deterioration of the specimen with PWTS 1.0% was SEM analysis.
slightly higher than the conventionally cured concrete,
ll Curing enhanced by PWTS results in reduction of
PEG, SAP by 1.75%, 3.22%, 3% respectively. However the
chloride penetration which was evident from the RCPT
deterioration of PWTS compared with other Self-Curing
as the penetration value was low as per ASTM C1202
agents has negligible difference.
specifications. The penetration value was little higher
for concrete with PWTS when compared to concrete
Shrinkage Test Results with other Self-Curing Agents. Even though, the
penetration value was within the range.
ll It was evident from the Acid Attack Test that concrete
with PWTS was less prone to attack due to acids. The
deterioration value for concrete with PWTS was in the
range of deterioration value of concrete with other
Self-Curing Agents.
ll Drying shrinkage in concrete was controlled by PWTS
Fig. 14: Shrinkage Test Graph for Trail-4 mix with Optimum to a considerable extent as the shrinkage test shows
Dosage of different Self-Curing Agents that shrinkage value of specimen with PWTS was
negligibly higher than specimen with other Self-Curing
Figure 14. shows the results of Shrinkage test conducted Agents.
with Autoclave apparatus on Trial-4 mix with optimized ll PWTS is a waste product which has to be processed
percentage of different Self-Curing agents. The Trail-4 before it was disposed into the environment which
mix with PWTS as Self- Curing agent, at end of 1 month increases the operating cost of water treatment plant.
specimen length gets reduced for about 1.90mm. This Hence, recycling of PWTS as a Self-Curing agent
reduction in length was higher than the specimen cured would be costless.
conventionally, with PEG and SAP whose reduction
in length has difference of 0.4mm, 0.8mm, 0.1mm
respectively with PWTS specimen. But this difference in Acknowledgements
reduction of length between PWTS specimen and other We, the authors heartfully thank the Management,
specimens were negligible. Principal, Deans and HOD's of various departments,
Teaching, Non - teaching staffs, our friend Aarthi.N -
Conclusions Research Scholar, Division of Geotechnical Engineering,
IITM, Tamilnadu and our student friends Priyanka, Shruthi
ll Calcium content was found to be 46.03% in PWTS from S Kumar, Nathiya and Rudhra of Kongu Engineering
X-Ray diffraction analysis. College, Tamilnadu for their support in successful
completion of the project.
ll The Trail mix with higher paste volume was good in
workability when compared to other Trail mixes, as
References
found from the workability test results.
1. Zoran Grdić., Iva Despotović., Gordana Topličić Ćurčić., “ Properties
of Self-Compacting Concrete with different Types of Additives”,
ll Mechanical Strength of concrete was found Architecture and Civil Engineering, Vol.6, No.2, 2008, pp.173 – 177.
to be higher for Trail-4 mix when compared to other
2. Hardik Upadhyay., Pankaj Shah., Elizabeth George., “Testing and Mix
Trail mixes. Introduction of PWTS into the concrete Design Method Of Self-Compacting Concrete”, National Conference
have not disturbed the Mechanical Strength of the on Recent Trends in Engineering & Technology.
concrete. 3. Kazem Reza Kashyzadeh., Neda Aghili Kesheh., “Study Type of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

226 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Development of Self-Compacting Concrete using Potable Water Treatment Sludge as a Costless Self-Curing Agent

Cracks in Construction and its Controlling”, International Journal Environment Research, Vol.66, No.1 (Jan. - Feb., 1994), pp. 4-11.
of Emerging Technology and Advanced Engineering, Vol.2, Issue
16. Qiwei Cao Nowasell., John T. Kevern., “Using Drinking Water
8, August 2012.
Treatment Waste as Low-Cost Internal Curing Agent for Concrete”,
4. Moayyad Al-Nasra., Mohammad Daoud., “Investigating the Use of ACI Materials Journal, Vol.112, No.1, January-February 2015.
Super Absorbent Polymer in Plain Concrete”, International Journal
17. IS: 12269–1987, “Indian Standard Specification for 53 grade ordinary
of Emerging Technology and Advanced Engineering, Vol.3, Issue 8,
portland cement”.
August 2013.
18. IS: 383–1970, “Indian Standard Specification for coarse and fine
5. Dhir R. K., Hewlett P. C., Lota J.S., and Dyer T.D., “An Investigation
aggregates from natural sources for concrete (second revision)”.
into the Feasibility of Formulating Self-Cure concrete”, Materials
and Structures/Materiaux et Constructions, Vol. 27, No. 174, 1994, 19. “EFNARC Specification and Guidelines for Self-Compacting
pp. 606-615. Concrete”, May 2005.
6. Wang J., Dhir R K., Levitt M., “Membrane Curing of Concrete”, 20. “EFNARC Guidelines for Viscosity Modifying Admixtures For
Cement Concr Res 1994:24(8) : 1463-74. Concrete”, September 2006.
7. Dhir PK., Hewlett P.C., Dyre T.D., “Mechanism of Water Retention 21. IS: 516–1959, “Indian Standard for Methods of tests for strength
in Cement Pastes Containing a Self-Curing Agent”, Mag Concr Res of concrete”.
1998;50(1):85-90. 22. IS: 5816-1999, “Indian Standard for splitting tensile strength of
8. El-Dieb A.S., “Self-curing concrete: Water retention, hydration and concrete - method of test (first revision)”.
moisture transport”, Construction and Building Materials, vol.21, 23. ASTM F2170, “Standard Test Method for Determining Relative
2007, pp.1282–1287. Humidity in Concrete Floor Slabs Using In Situ Probes”.
9. ACI Committee 308., “Guide to curing concrete (ACI 308R–01,Re- 24. AASHTO T277, “Standard Method of Test for Rapid Determination
approved 2008)”, American Concrete Institute, 2001, p. 31. of the Chloride Permeability of Concrete”.
10. Bentz D.P., Snyder K.A., “Protected Paste Volume in Concrete – 25. ASTM C1202, “Standard Test Method for Electrical Indication of
Extension 10 Internal Curing using Saturated Light Weight Fine Concrete’s Ability to Resist Chloride Ion Penetration”.
Aggregate”, Cem. Concr. Res. 29, 1999, pp.1863–1867.
26. De Schutter G., “Measurement of Properties of Fresh Self-
11. Magda I. Mousa., Mohamed G. Mahdy., Ahmed H. Abdel-Reheem., Compacting Concrete”.
Akram Z. Yehia., “Physical Properties of Self-Curing Concrete
(SCUC)”. 27. ASTM C1611/C1611M, “Standard Test Method for Slump Flow of
Self-Consolidating Concrete”.
12. Masironi R., Pisa Z., Clayton D., “Myocardial infarction and water
hardness in the WHO myocardial infarction registry network”, 28. Mjornell k., “Self-Desiccation in Concrete”, Chalmers University of
Bulletin of the World Health Organization, 1979, 57:291-299. Technology, 1994;2(556):94.

13. Crawford MD., Gardner MJ., Sedgwick PA., “Infant Mortality and 29. MeGrath PF., “Internal Self-Desiccation of Silica Fume Concrete”,
Hardness of Local Water Supplies”, Lancet, 1972, 1(758):988-992. MASc thesis, Civil Engineering Department, University of Toronto,
1989.
14. Zemla B., “Geographical Incidence of Gastric Carcinoma in
Relation to Hardness of Water for Drinking and Household Needs”, 30. Detwiler, R.J., Fapohunda C., and Natale, J., “Use of Supplementary
Wiadomosci lekarskie, 1980, 33(13):1027-1031 (in Polish). Cementing Materials to Increase the Resistance to Chloride Ion
Penetration of Concretes Cured at Elevated Temperatures”, ACI
15. Aarne Vesilind P., “The Role of Water in Sludge Dewatering”,Water Materials Journal, Vol.91, No.1, January-February 1994, pp. 63-66.

Rampradheep G.S.
Rampradheep G.S. has completed his graduation in Civil Engineering from Kongu Engineering College,
Perundurai, Erode, Tamil Nadu, India in the year April 2008 and completed his Masters in Structural
Engineering from Bannari Amman Institute of Technology on May 2010. He was awarded two Gold Medals
for his outstanding performance in Structural Engineering. He pursuing his Ph.D in the field of Concrete
under the Anna University, Chennai. He has published more than 14 papers in reputed conferences and a
paper in reputed journal. His project Electricity Generation from cement matrix incorporated with self-
curing agent has been awarded the second best project by the IEEE, Consumer Society, Bangalore. He
was the recipient of various awards and his team has obtained two national awards in the year 2014 and
2015. He has successfully awarded and completed a research and seminar grant over a lakh from various
agencies like AICTE, CSIR, TBI-KEC etc., At present, he is working as an Assistant Professor in Kongu
Engineering College, Perundurai, Erode Tamil Nadu, India.

Dr.Sivaraja M.
Dr.Sivaraja M. has completed his graduation in Civil Engineering from Madurai Kamaraj University and
Masters in Structural Engineering from Periyar University. He completed his Ph.D from Anna University,
Chennai during 2008. He has done his Post Doctoral Research at “University at Buffalo”, The State
University of New York, USA on “Multifunctional Cementitious Composites” during 2009 under a prestigious
scholarship called “BOYSCAST Fellowship” given by DST, New Delhi. He has published more than 25
papers in reputed journals and 60 papers in conferences. He has successfully awarded and completed a
research grants over 50 lakhs from various agencies like DST, AICTE, IEI, Coir Board of India etc., He is the
reviewer of reputed journals. At present he is working as a principal at N.S.N College of Engineering and
Technology, Karur, Tamil Nadu.

Organised by
India Chapter of American Concrete Institute 227
Session 2 C - Paper 4

Partial replacement of natural sand with recycled waste materials in

concrete for sustainable construction practices

Ram Prasad V S S Ashwin Bharathwaj A. V. Marckson

Department of Civil Engineering, Department of Civil Department of Civil
Malaviya National Institute of Engineering, National Institute Engineering, Indian Institute
Technology Jaipur, India. of Technology Trichy, India. of Technology Madras, India.

Abstract costs of land filling and the environmental problems

concerned with the waste disposal. Many researchers and
Use of concrete for construction purposes has been
scholars have explored
increasing over the years and resources for natural sand,
which is used as fine aggregate in concrete, are depleting the idea of utilization of these wastes in concrete as a
very quickly of late. Thus it is imperative for the researchers partial replacement for fine aggregates: fly ash (Gesoglu
in the concrete industry to find an alternative for natural et. al., 2006; Seo et. al., 2010 and Christy and Tensing,
sand. Waste materials such as fly ash, recycled concrete, 2010); blast furnace slag in many forms (Escalante-Garia,
recycled crumb rubber, steel slag etc. are hard to dispose. 2009; Topcu et. al., 2010 and Binici et. al., 2012); bottom ash
Therefore to save the natural sand resources as well as (Trakool, 2006; Sani et. al., 2010 and Topcu and Turhan,
using these waste materials beneficially, researchers 2010); metakaolin (Badogiannis et. al., 2004; Badogiannis
have suggested using them as a partial replacement for and Tsivilis, 2009 and Batis et. al., 2005); waste glass (Park
fine aggregates. This review focusses on the properties of et. al., 2004; Kou and Poon, 2009 and Lee et. al., 2013).
fresh and hardened concrete (viz. compressive strength, Apart from the above mentioned wastes, many other
flexural strength, splitting tensile strength, modulus waste products such as olive oil wastes, marble cutting
of elasticity, density, workability, water absorption and sludge, fresh and recycled concrete waste, incinerated
abrasion resistance), after incorporating the concrete with sludge sewage ash, sheet glass powder, recycled crumb
varied percentages of waste materials as a replacement rubber etc. have been subjected to a lot of research for
for natural sand by the various researchers and scholars. their incorporation in concrete as a replacement for
The waste materials considered are waste glass, fine aggregates. Using these waste products reduces
fresh concrete waste, recycled crumb rubber, recycled negative impacts caused by the natural sand usage in
concrete aggregates, foundry sand, copper slag, bottom concrete on the economy and the environment (Yuksel
ash, crushed bricks and steel slag. After compilation of et. al, 2007). However, to entirely embrace these wastes
the various research works all over the world, it has been as a fine aggregate replacement in the concrete industry,
found that the waste materials can be successfully used it is necessary for understand the performance of such
as a partial replacement for natural sand in concrete. At concrete. It is necessary that concrete made with these
ideal percentages for each waste material, the properties aggregates is mechanically strong and durable. This
of the modified concrete have been observed to be better review aims to bring together the various research works
than those of the conventional natural sand concrete. on fine aggregate replacement and discuss the properties
of fresh and hardened concrete viz. compressive strength,
Keywords: Eco-friendly, Natural sand, Replacement,
flexural strength, splitting tensile strength, modulus
Sustainability, Waste products.
of elasticity, density, workability, water absorption and
abrasion resistance, in order to justify the viability of
Introduction these wastes pertaining to their usage in concrete as a
With an increase in the construction activities across replacement for fine aggregates.
the world and a decrease in the amount of natural sand
available, there has been an urge to find materials Waste Glass
as replacement for natural sand which have similar
There have been many researchers who have studied on
properties as natural sand and also behave similar to
the incorporation of waste glass as a partial replacement
the natural sand when used in concrete. Other important
for fine aggregates. The main reason for consideration
reason for our quest to find a replacement for natural
of waste glass as a potential replacement for natural
sand is the environmental concerns pertaining to its
aggregates is that it creates serious environmental issues
extraction. Utilization of waste products in concrete as a
and the concrete industry has provided an ideal base for
replacement for fine aggregates has become one of the
its use in an advantageous way. Theoretically, glass has
most productive topics of research due to the increasing
been observed to be a completely recyclable material

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

228 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial replacement of natural sand with recycled waste materials in concrete for sustainable construction practices

and that too without any loss in the quality of the material in the near future (Tam and Tam, 2007). Recycling this
(Sobolev et. al, 2006). The construction industry has material is advantageous because it will not only help in
found waste glass to be an ideal replacement for natural waste storage but also aide in the preservation of natural
aggregates (though only partially) which not only counters aggregates. Fresh concrete waste does not contain a lot
the environmental concerns pertaining to the landfills and of impurities in comparison to other recycled aggregates
other disposal methods of this waste but also reduce the and thus use of this waste as aggregates in concrete may
continuous extraction of the natural aggregates (Rakshvir prove to be more economical when compared with other
and Barai, 2006). But the effective replacement of these recycled aggregates.
wastes depends on the effect of their replacement on a
Though limited research has been conducted on the
number of factors such as compressive strength, flexural
potential of fresh concrete waste, there have been a few
strength, abrasion resistance, modulus of elasticity,
successful studies on the use of this waste as aggregates
splitting tensile strength, density, water absorption etc.
in concrete (Correia et. al., 2009 and Kou et. al., 2012).
There have been numerous studies which have extensively
However these studies mainly concentrated on their use
carried out research on these parameters.
as coarse aggregates. Thus there is a lack of knowledge
Park et. al. (2004) observed that there was a decrease in about the use of fine fraction of fresh concrete waste in
the compressive strength, tensile strength and flexural concrete.
strength when the percentage of waste glass was
increased in the concrete. Shayan and Xu (2006) found
Recycled Crumb Rubber
that incorporation of 30% waste glass did not have any
significant influence on the long term performance of the Approximately, 275 million and 180 million used rubber
concrete and recommended the use of glass powder and tires get accumulated every year in the United States
glass aggregate together in 40 MPa concrete mixes without (Papakonstantinou and Tobolski, 2006) and the European
any considerable effect on properties of concrete. Turgut Union (Silvestravieiete and Sleinotaite- Budriene, 2002)
and Yahlizade (2009) reported that using 20% waste glass respectively. The estimated value for total number of
as a replacement for natural fine aggregates achieved waste tires in India is 112 million per year (Thomas et. al.,
maximum compressive strength and it decreased as the 2014). The easiest way of disposing off rubber tires is by
percentage was increased to 30%. Corinaldesi et. al . burning them but it generates a lot of smoke and degrades
(2005) stressed on the feasibility of use of waste glass as the environment which makes the process undesirable.
fine glass as they reported no alkali-silica reaction when The other prominent method of disposal is dumping
waste glass particles having sizes up to 0.1 mm is used in into the landfills. But due to the low density and poor
the concrete or mortar. Higher mechanical strength has degradation of rubber, this method is also not so favored
been observed to be achieved when glass particles with (Serge and Joekes, 2000). Thus many studies have tried to
size less than 0.15 mm have been used in the concrete due explore the possibilities of incorporation of crumb rubber
to their ability to form calcium silicate hydrates (C-S-H) as fine aggregates in the concrete in order to device an
during the cement hydration process (Shao et. al., 2000 environmental friendly method of disposal of rubber tires.
and Shayan and Xu, 2004). Chen et. al. (2006) and Metwally Crumb rubber is generated by shredding the rubber tires
(2007) stressed on the adverse effect of incorporating (Chesner et. al., 1998). They are fine particles whose size
waste glass on the workability of concrete, though both ranges from 0.075 mm to 4.75mm (Issa and Salem, 2013).
of them agreed on the fact that it significantly enhanced Aggregating rubber in concrete enhances concrete’s
the mechanical properties of concrete at later ages. When impact resistance (Topcu and Avcular, 1997) and helps
waste glass is used as fine aggregate in the concrete, combat its limiting properties such as low flexibility,
reduced water absorption and higher abrasion resistance elasticity and capacity to absorb energy (Wang et. al.,
is observed due to higher hardness of glass (Meyer et. al., 2000). Mohammed et. al. (2012) and Pelisser et. al. (2011)
2001). studied the effect of sand replacement with crumb rubber
in concrete. Eldin and Senouci (1993) were among the first
Fresh Concrete Waste to study the use of crumb rubber in concrete and they found
that the concrete had lower workability, lower compressive
Due to the uncertainty in the exact quantity of the concrete and lower tensile strength. Improving on the same, Serge
required in a project, every day a concrete batching plant and Joekes (2000) reported that treating the rubber with
receives a huge amount of over-ordered fresh concrete NaOH enhances the adhesion of rubber to cement paste
from various construction sites. In France, 2.6 million thereby improving the mechanical properties, abrasion
tons of fresh concrete waste is generated annually resistance and decreasing the water absorption of the
(Serifou et. al., 2013). Currently, as is common with concrete. Bravo and de Brito (2012) also replaced sand
most waste materials around the world, the practice for with crumb rubber at replacement percentages of 5,
disposal of these wastes is dumping into landfills which 10 and 15% and reported that while the compressive
is not so beneficial. Additionally, due to the saturation strength, carbonation resistance and chloride resistance
of the landfill areas, this procedure will be of high cost

Organised by
India Chapter of American Concrete Institute 229
Session 2 C - Paper 4

decreased with increase in percentage replacement, FOUNDRY SAND

shrinkage and water absorption of the concrete increased.
The metal alloy casting process in the foundry industries
Khaloo et. al. (2008) studied the toughness of concrete
yields Foundry sand (FS) as one of the by- products.
after replacing the fine aggregates with crumb rubber
The molding sand is generally recycled for about 8-10
at percentages of 12.5, 25, 37.5 and 50% and found that
times before disposing it. Repeated exposure to high
the toughness was enhanced at all replacement levels
temperatures and abrasive materials leads to the sand
and the maximum toughness was experienced at the
losing its molding characteristics (Dayton et al. 2010
replacement percentage of 25%. Workability was found
and Guney et al 2010). Foundry sands are generally
to decrease with increasing replacement levels by Nayef
sub-angular to round in shape. Waste foundry sand has
et. al. (2010) and Ozbay et. al. (2011). However, Aiello and
a uniform grain size distribution and a specific gravity
Leuzzi (2010) and Wang et. al. (2013) found the workability
ranging from 2.39 to 2.79. They are characteristic of low
to slightly improve with increasing replacement levels.
absorption capacity and plasticity.
Etxeberria et al. (2010) evaluated the properties of
Recycled Concrete Aggregates concretes by using chemically bonded foundry sand,
Concrete is used in a humungous quantity daily throughout green foundry sand as the partial replacement for fine
the world. Majority of the buildings that are being aggregate and blast furnace slag as the course aggregate
constructed or that have been constructed in the past replacement. 25%, 50% and 100% replacements were
are made from concrete and for the sake of advancement done and it was concluded that (i) at higher water cement
in the way the buildings are built, the existing buildings ratio, the concrete mixes with replacements acquired
are being demolished. This generates a large amount of higher strength properties than the conventional
concrete waste which can be recycled and used as a partial concrete. Bakis et al (2006) reported the usage of foundry
replacement for natural aggregates in concrete. The sand as fine aggregate replacement in asphalt concrete
recycled concrete waste is different from the construction and reported that 10% replacement level had the best
and demolition waste in a way that it is not processed and suitability. Khatib et al. (2013) in his research replaced
thus the properties of the recycled concrete aggregates natural sand with 0%,30%,60% and 100% of waste foundry
differ significantly from the natural aggregates, thereby sand and observed a gradual increase in the absorption of
hindering its use regularly (Pedro et. al., 2014). Many water by the specimens and decreasing trend of strength
researchers have been drawn towards investigating the properties and ultra-pulse velocity with increasing level of
properties of the recycled concrete aggregates and the replacements. Khatib and Ellis (2001) observed decrease
effect of their incorporation in concrete as aggregates in strength properties and increase in shrinkage.
(Matias et. al., 2013, Amorim et.al., 2012, Kwan et. al.,
2012 and Guedes et. al., 2013). But concrete made with
recycled concrete aggregates is not only a research field Copper Slag
now, it has already become a practical reality (Grubl and The manufacture of copper in industries yields copper
Nealen, 1998). This calls for the emergence of continuous slag as a by-product. About 2.2 tons of copper slag is
suggestions from the research field in order to improve obtained for every ton of copper produced (Gorai et al,
the concrete made with the incorporation of recycled 2003). The copper slag generated globally is over 25
concrete aggregates. It is of paramount importance that million tons which finds its use in concrete production due
the concrete not only fulfills the criteria for mechanical to its pozzolonic properties and particle size gradation.
properties but also satisfies the durability criteria Copper slag is generally black and glassy in appearance
before it can be efficiently put to use. According to Tighe with a bulk density between 1.9- 2.1 g/cc and a fineness
et. al. (2013), for a particular batch of recycled concrete modulus around 1.9. Al-Jabri (2011) characterized copper
aggregate to be put in use for structural applications, its slag with low water absorption, higher specific gravity,
aggregate relative density should be a minimum of 2.3, low CaO content and higher silica content compared to
maximum mortar content should be 50% and maximum the conventional fine aggregate. Hwang and Laiw (1989)
absorption should be 3%. have observed the compressive strength and workability
to enhance on inclusion of copper slag. Li (1999) and Zong
Studies on recycled aggregates have mostly been
(2003) observed similarities in mechanical properties
restricted to the coarse fraction and very less attention
between conventional concrete and copper slag replaced
has been given to the fine fraction of these aggregates.
concretes. Ayano et al. (2000) evaluated the strength,
The use of fine recycled aggregates in concrete is limited
durability and setting time of concretes which used copper
because of its undesirable properties (Wainwright et. al.,
slag as fine aggregate. Ishimaru et al. (2005) studied the
1994 and Zaharieva et.a l., 2003) which are caused due
fundamental properties of concrete incorporating copper
to the adhered mortar/ cement/ concrete in the recycled
slag as partial replacement for natural sand and reported
aggregates and extreme porosity of fine particles which
that values up to 20% replacement were acceptable. Wu et
leads to reduction in the performance of any composite
al. (2010) investigated the dynamic compressive strength
containing them (Evangelista and de Brito, 2010).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

230 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial replacement of natural sand with recycled waste materials in concrete for sustainable construction practices

of concrete reinforced with copper slag and found the substitution and then decreasing on further replacement
values to increase compared to control mix till 20% of (b) flexural strength of bottom ash replace samples were
replacement. Naganur and Chethan (2014) reported generally comparable or higher to control mix. Bai et al.
enhanced workability, density, corrosion and acid attack (2005) & Shi-Cong and Chi-Sun (2009) investigated the
vulnerability as copper slag replaces sand. They also prospect of substituting sand with bottom ash and found
reported that 40% as the optimum content of copper slag that at (a) at constant water content, workability increased
in terms of strength properties, which were similar to the and strength decreased as replacement levels increased
observations by Brindha et al. (2010). while (b) at constant slump, compressive strength at all
levels were comparable to control mix (c) workability
increases. Aramraks (2006) substituted natural fine
Bottom Ash
aggregates with 50% and 100% coal bottom ash and found
Combustion of coal at thermal power plants produces that the strength, workability and abrasive resistance
bottom ash which is ultimately disposed at enormous decreased with increase in bottom ash content.
quantities. Coal bottom ash resembles natural sand in
physical appearance and has sizes varying from fine sand
to fine gravel. Bottom ash are angular, rough textured and Crushed Bricks
is comparatively lighter and porous with a specific gravity Millions of tons of brick waste are obtained from ceramic
varying from 1.39 to 2.33. Bakoshi et al. (1998) concluded industries, building and demolishing operations. These
that the optimal replacement to for compressive strength, bricks, when crushed into powders can be used to replace
tensile strength and abrasive resistance of concrete is natural sand as a fine aggregate in concrete. Fine recycled
10%-40% of bottom ash as fine aggregates. Andrade brick aggregate properties are found to have relatively
et al. (2007) studied the influence of bottom ash as fine higher water absorption ratio and lower bulk modulus.
aggregates replacement on mechanical properties Khatib (2005) and Poon and Chan (2009) replaced
such as compressive strength and modulus of elasticity sand with recycled crushed brick waste and observed
and other properties related to moisture transport and lower strength values as compared to conventional
reported decreasing trends of elastic modulus and concrete. Debieb and Kenai (2008) studies the influence
increasing sorptivity. Karuma and Kaya (2008) used of fine and coarse recycled brick aggregates on the
bottom ash as partial fine aggregates replacement and compressive strength, tensile strength, elastic modulus,
found that (a) the compressive strength increased till 10% water absorption, permeability, shrinkage etc. by using

Table 1
Properties of concrete incorporated with different waste materials

Splitting
S. Authors (% Compressive Flexural Modulus of Workability/ Water Abrasion
Aggregate tensile Density
No. replacement) strength strength elasticity flow Absorption resistance
strength

Park et. al.,

2004 (0, 30, 50 Decrease Decrease Decrease - Decrease Decrease Decrease -
& 70%)

Topcu and
Canbaz, 2004 Decrease Decrease Decrease Decrease Decrease Decrease Increase -
(0-60%)

Ismail and Al-

Hashmi, 2008 Increase Increase - - Decrease Decrease - -
Waste glass

(10, 15 & 20%)

1.
Kou and Poon,
2009 (10, 20 Decrease - Decrease Decrease Decrease Constant Decrease Increase
&30%)

Ali and Al-

Tersawy, 2012 Decrease Decrease Decrease Decrease Decrease Increase Increase Increase
(0-50% in SCC)

Lee et. al.,

2013 (25, 50, Increase - - - Decrease - Increase -
75 &100%)

Continued on next page

Organised by
India Chapter of American Concrete Institute 231
Session 2 C - Paper 4

Continues from previous page

Splitting
S. Authors (% Compressive Flexural Modulus of Workability/ Water Abrasion
Aggregate tensile Density
No. replacement) strength strength elasticity flow Absorption resistance
strength

Correia et. al.,

2009 (10, 20 & Decrease - - - - Constant Increase -
Fresh concrete waste

30%)

Kou et. al.,

2. 2012 (0, 15, 30 Decrease - Decrease Decrease Decrease Increase Increase Increase
& 50%)

Serifou et. al.,

2013 (0, 50 & Decrease Decrease Decrease - - - Increase -
100%)

Issa and
Salem, 2013 Decrease - - - Decrease - - -
(0-100%)

Ganjian et. al.,

Recycled Crumb Rubber

2009 (5, 7.5, Decrease Decrease Decrease Decrease - - Increase -

10%)

3. Thomas et. al.,

Decrease Decrease - - Decrease - Increase Increase
2014 (0- 20%)

Al-Tayeb et.
Slight
al., 2013 (10 & Decrease Decrease Decrease - - - -
Increase
20%)

Gupta et. al.,

Decrease Decrease - Decrease Decrease Decrease Increase Decrease
2014 (0-20%)

Ravindrarajah
and Tam, 1987 Slight decrease Increase Increase Decrease - Constant - -
(10%)

Evangelista
and de Brito,
Slight decrease Decrease - Decrease - - - Increase
2007 (0, 10, 20,
30, 50 & 100%)

Evangelista
Recycled concrete aggregate s

and de Brito,
- - - - - - Increase -
2010 (0, 30 &
100%)

Increase
4. Increase up up to
to 20%, then 20%, then
Neno et. al.,
decrease up to decrease
2013 (20, 50 & - Increase Decrease Decrease Increase -
50% and again up to 50%
100%)
increase up to and again
100% increase up
to 100%

Levy and
Helene, 2007
Decrease - - - - Decrease Increase -
(20, 50 &
100%)

Lotfy and Al-

Slight Slight
Fayez, 2015 (10 Slight decrease - Increase Decrease Increase -
decrease decrease
& 20%)

Continued on next page

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

232 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial replacement of natural sand with recycled waste materials in concrete for sustainable construction practices

Continues from previous page

Splitting
S. Authors (% Compressive Flexural Modulus of Workability/ Water Abrasion
Aggregate tensile Density
No. replacement) strength strength elasticity flow Absorption resistance
strength

Prabhu et al.,
2014 (10, 20, Decreases Decreases Decreases - - Decreases - -
30, 40 & 50%)

Siddique and
Singh, 2011 (5, Increases - Increases Increases - - - -
10, 15 & 20%)

Siddique et al.,
2009 (10, 20, & Increases Increases Increases Increases - - - -
Foundry Sand

30%)
5.
Khatib et
al.,2013 (30,60 Increases Decreases Decreases Decreases - - - -
& 100%)

Guney et al.,
10% > 0% > 5% 10% > 0% > 10% > 0% >
2010 (5,10 & Decreases Decreases - Decreases -
> 15% 5% > 15% 5% > 15%
15%)

Basar et al.,
2012 (10,20,30 Decreases - Decreases Decreases Decreases - Decreases -
& 40%)

Decreases
Al-Jabri et al., & then
2009a (30, 50, Increases Increases Increases - Increases Increases increases -
70, 80 & 100%) (optimum-
50%)

Decreases
Al-Jabri et al.,
& then
2011 (10, 20,
Increases Increases Increases - Increases Increases increases -
40, 50, 60, 80
(optimum-
& 100%)
40%)

Al-Jabri et al.,
2009b (10, 20,
Increases Decreases Increases - Increases Increases - -
40, 50, 60, 80
Copper Slag

& 100%)

6.
Madheswaran
et al., 2014 (25, Increases - - - Increases Increases - -
50, 75 & 100%)

Naganur and
Chethan, 2014
Increases - Increases - Increases Increases - -
(10, 20, 30, 40,
50 & 60%)

Tamil Selvi et
al., 2014 (20, Increases upto Increases Increases
- Increases Increases - -
40, 60, 80, 40% upto 40% upto 40%
100%)

Wu et al., 2010
(20, 40, 60, 80 Decreases Decreases Decreases - Increases- Increases - -
& 100%)

Continued on next page

Organised by
India Chapter of American Concrete Institute 233
Session 2 C - Paper 4

Continues from previous page

Splitting Modulus
S. Authors (% Compressive Flexural Workability/ Water Abrasion
Aggregate tensile of Density
No. replacement) strength strength flow Absorption resistance
strength elasticity

Kim and Lee, 2011

Decreases Decreases - Decreases Decreases Constant - -
(25, 50, 75 & 100%)

Singh and Siddique,

2015 (30, 50, 75 & Decreases - - - - - Decreases Decreases
100%)
Singh and Siddique,
2014 Constant - Increases Decreases Decreases Decreases - -
Bottom Ash

(30, 50, 75 & 100%)

7.
Ghafoori and
Decreases Decreases Constant Decreases Decreases Decreases - Decreases
Bucholc, 1996

Aggarwal et al.,
2007 (20 ,30, 40 & Decreases Decreases Decreases - - Decreases - -
50%)
Decreases
Yuksel and Genc,
till 10%
2007 (10, 20, 30, 40 Decreases Decreases - - Increases - -
& then
& 50%)
constant

Khatib, 2005 (25,

Decreases - - Decreases - Decreases - -
50, 75 & 100%)

Debieb and Senai,

2008 (25, 50, 75 & Decreases Decreases - - Decreases Decreases Decreases -
100%)
Crushed Bricks

Alves et al., 2014

8. Decreases - Decreases Decreases Decreases Decreases - Decreases
(20, 50 & 100%)

Poon and Chan,

Decreases - - Decreases Decreases Increase - -
2007 (20%)

Aliabdo et al., 2014

Constant - Decreases Decreases Increase - Increase -
(25, 50, 75 & 100%)

H. Qasrawi et al.,
2009 (15, 30, 50 & Increases - Increases Increases Decreases
100%)

Q. Wang et al., 2013

Decreases - - - - - - -
(15, 30 & 45%)

V. Subathra Devi
and B.K. Gnanavel, Increases till
Increases Increases - - Decreases - -
2014 (10, 20, 30, 40 40%
Steel Slag

& 50%)
9.
P.S Kothai and B.K.
Gnanavel, 2014 (10, Increases Increases Increases Increases Increases Decreases - -
20, 30, 40 & 50%)
M.Soundar Rajan,
2014 (10, 20, 30, 40 Increases Increases Increases Increases - - - -
& 50%)
Prasanna &
Kiranmayi, 2014
Increases Increases Increases - - Decreases - -
(5% to 25% at 5%
increment)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

234 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial replacement of natural sand with recycled waste materials in concrete for sustainable construction practices

25%,50%,75% and 100% substitution levels. Alves at at fresh and hardened state. Waste Management; 30:1699–704.
al. (2014) observed decrease in strength, workability, 3. Al-Jabri KS, Hisada M, Al-Oraimi SK, Al-Saidy AH, 2009b. Copper
abrasive resistance and density of concrete with partial slag as sand replacement for high performance concrete. Cement
& Concrete Composites 31, 483–488.
crushed bricks replacement for fine sand. On the contrary,
Poon and Chan (2007) observed higher workability at 20- 4. Al-Tayeb Mustafa Meher, Abu Bakar B H, Ismail Hanafi and Akil
Hazizan Md, 2013. Effect of partial replacement of sand by recycled
80 ratio of brick waste to sand content. fine crumb rubber on the performance of hybrid rubberized-normal
concrete under impact load: experiment and simulation, Journal of
Cleaner Production 59, 284-289.
Steel Slag
5. Ali Ersaa Emam and Al-Tersawy Sherif H., 2012. Recycled glass as
Steel slag is a by-product obtained in the steel making a partial replacement of fine aggregate in self-compacting concrete,
process. It is produced when the impurities are separated Construction and Building Materials (35), 785-791.
from the molten steel produced. While its influence as 6. Ali A. Aliabdo, Abd-Elmoaty M. Abd-Elmoaty, Hani H. Hassan, 2014.
substitute for course aggregate is investigated widely, Utilization of crushed clay brick in concrete industry: Alexandria
not much research is done on the effect of steel slag as Engineering Journal 53, 151–168.
partial replacement of fine aggregate in concrete. Steel 7. Alves A V, T.F. Vieira, J. de Brito J.R. Correia, 2014. Mechanical
slag aggregates are highly angular in structure and has properties of structural concrete with fine recycled ceramic
aggregates. Construction and Building Materials 64, 103–113.
a rough texture. They have higher specific gravity and
8. Amorim P, de Brito J, Evangelista L., 2012. Concrete made with coarse
moderate water absorption value (<3%). They process
concrete aggregate: influence of curing on durability. ACI Materials
very good abrasive resistance, bearing strength and Journal; 109(2):195–204.
are highly sound. Wang et al. (2013) replaced sand with 9. Andrade LB, Rocha JC, Cheriaf M, 2007. Evaluation of concrete
15%, 30% and 45% steel slag and reported decreased incorporating bottom ash as a natural aggregates replacement.
compressive strength, comparable shrinkage, increased Waste Management; 27(9):1190–9.
chloride permeability values at constant water cement 10. Andrade LB, Rocha JC, Cheriaf M., 2009. Influence of coal bottom
ratio. Prasanna & Kiranmayi (2014) reported enhanced ash as fine aggregate on fresh properties of concrete. Construction
strength and lower workability at increasing content of and Building Materials 23, 609–14.
steel slag in place of sand. 11. Aramraks T, 2006. Experimental study of concrete mix with bottom
ash as fine aggregate in Thailand. In: Symposium on infrastructure
development and the environment, p. 1–5.
Conclusion 12. Ayano T, Kuramoto O, Sakata K., 2000. Concrete with copper slag
The research works on replacement of fine aggregates as fine aggregate. Journal of the Society of Material Science Japan
; 49(10):1097–102.
with waste materials in concrete is not new to the
13. Bai Y, Darcy F, Basheer PAM., 2005. Strength and drying shrinkage
concrete industry. For long, researchers have studied
properties of concrete containing furnace bottom ash as fine
the properties of the waste materials and the effect of aggregate. Construction and Building Materials; 19:691–7.
incorporating these waste materials on the properties of 14. Bakis R, Koyuncu H, Demirbas A., 2006. An investigation of waste
fresh and hardened concrete. Using the waste materials foundry sand in asphalt concrete mixtures. Waste Management
such as fly ash, foundry sand, recycled crumb rubber Resources; 24:269–74.
etc. as fine aggregate replacement in concrete not only 15. Bakoshi T, Kohno K, Kawasaki S, Yamaji N., 1998. Strength and
serves the purpose of waste disposal for these materials durability of concrete using bottom ash as replacement for fine
in an environmental friendly way, but also helps preserve aggregate. ACI Special Publications, 179:159–72.
the natural sand, which has been found to be depleting 16. Basar Merve H., Nuran Deveci Aksoy, 2012. The effect of waste
drastically. It is observed that though these materials foundry sand (WFS) as partial replacement of sand on the mechanical,
leaching and micro-structural characteristics of ready-mixed
cannot be used as a complete replacement for fine concrete. Construction and Building Materials 35, 508–515.
aggregates, but they can be successfully used as a partial
17. Binici H, Durgun MY, Rizaoglu T, Kolucolak M., 2012. Investigation of
replacement for the natural sand. At the ideal percentage durability properties of concrete pipes incorporating blast furnace
of replacement, the concrete incorporated with the waste slag and ground basaltic pumice as fine aggregates. Scienctia Iranica
materials might even behave better than the conventional A; 19(3):366–72.
natural sand concrete. Further research wok is required 18. Bravo Miguel, de Brito Jorge, 2012. Concrete made with used tire
on the durability properties such as chloride migration, aggregate: durability-related performance. Journal of Cleaner
Production; 25:42–50.
oxygen permeability, abrasion resistance etc. Also new
materials such as plastic, metakaolin and olive oil wastes 19. Brindha D, Baskaran T, Nagan S, 2010. Assessment of Corrosion
and Durability Characteristics of Copper Slag Admixed Concrete,
can be explored as a possible fine aggregate replacement International Journal of Civil and Structural Engineering, Vol. 1, No.
in concrete. 2, pp. 192 – 211.
20. Chen CH, Wu JK, Yang CC., 2006. Waste E-glass particles used in
References cementitious mixtures. Cement and Concrete Resources; 36(3):449–
1. Aggarwal P, Aggarwal Y, Gupta SM, 2007. Effect of bottom ash as 56.
replacement of fine aggregates in concrete. Asian Journal of Civil 21. Chesner WH, Collins RJ, MacKay MH., 1998. Users guidelines for
Engineering (Building and Housing); 8:49–62. waste and byproduct materials in pavement construction. Report
2. Aiello MA, Leuzzi F, 2010. Waste tire rubberized concrete: properties No. FHWA-RD-97-148, Commack: Chesner Engineering, P.C., April.

Organised by
India Chapter of American Concrete Institute 235
Session 2 C - Paper 4

22. Christy CF, Tensing D., 2010. Effect of class-F fly ash as partial unprocessed steel slag in concrete as fine aggregate: Construction
replacement with cement and fine aggregate in mortar. Indian Journal and Building Materials 23, 1118–1125.
of Engineering Materials and Sciences; 17(April):140–4.
43. Hwang CL, Laiw JC, 1989. Properties of concrete using copper slag as
23. Corinaldesi V, Gnappi G, Moriconi G, Montenero A., 2005. Reuse of a substitute for fine aggregate. In: Proceedings of the 3rd international
ground waste glass as aggregate for mortars. Waste Management; conference on fly ash, silica fume, slag, and natural pozzolans in
25(2):197–201. concrete,SP-114-82;p.1677–95.
24. Correia S. L., Souza F. L., Dienstmann G., and Segadaes A. M., 2009. 44. Ishimaru K, Mizuguchi H, Hashimoto C, Ueda T, Fujita K, Ohmi M,
Assessment of the recycling potential of fresh concrete waste using 2005. Properties of copper slag and second class fly ash as a part
a factorial design of experiments, Waste Management, vol. 29, no. of fine aggregate. Journal of the Society of Material Science Japan;
11, pp. 2886–2891. 54(8):828–33.
25. Dayton EA, Whitacre SD, Dundan RS, Basta NT, 2010. Characterization 45. Ismail Zainab Z. and Al-Hashmi Enas A., 2008. Recycling of waste
of physical and chemical properties of spent foundry sands pertinent glass as partial replacement of fine aggregates in concrete, Waste
to beneficial use in manufactured soils. Plant Soil; 329:27–33. management (29), 655-659.
26. Debieb F, Kenai S., 2008. The use of coarse and fine crushed bricks 46. Issa Camille A. and Salem George, 2013. Utilization of recycled crumb
as aggregate in concrete. Construction and Building Materials; rubber as fine aggregates in concrete mix design, Construction and
22(5):886–93. Building Materials 42, 48-52.
27. Eldin N, Senouci A, 1993. Observations on rubberized concrete 47. Jayapal Naganur, Chethan B, 2014. Effect of Copper Slag as a
behavior. Cement and Concrete Aggregates; 15(1):74–84. Partial Replacement of Fine Aggregate on the Properties of Cement
Concrete, International Journal of Research (IJR) Vol-1, Issue-8,
28. Escalante-Garia JI, Magallanes-Rivera RX, Gorokhovsky A, 2009.
September, ISSN 2348-6848.
Waste gypsum-blast furnace slag cement in mortars with granulated
slag and silica sand as aggregates. Construction and Building 48. Khalifa S. Al-Jabri,Abdullah H. Al-Saidy, Ramzi Taha, 2011. Effect of
Materials; 23:2851–5. copper slag as a fine aggregate on the properties of cement mortars
and concrete: Construction and Building Materials 25, 933–938.
29. M, Pacheco C, Meneses JM, Beerridi I, 2010. Properties of concrete
using metallurgical industrial by-product as aggregate. Construction 49. Khalifa S. Al-Jabri, Makota Hisada, Abdullah S. Al-Saidy, S. K. Al-
and Building Materials; 24:1594–600. Oraimi, 2009a. Performance of high strength concrete made with
copper slag as a fine aggregate. Construction and Building Materials
30. Evangelista L. and de Brito J., 2007. Mechanical behavior of concrete
23, 2132- 2140.
made with fine recycled concrete aggregates, Cement and Concrete
Composites, 397-401. 50. Khaloo, A.R., Dehestani, M., Rahmatabadi, P., 2008. Mechanical
properties of concrete containing a high volume of tire-rubber
31. Evangelista L. and de Brito J., 2010. Durability performance of
particles. Waste. Management 28 (12), 2472-2482.
concrete made with fine recycled concrete aggregates, Cement and
Concrete Composites 32, 9-14. 51. Khatib JM, Herki BA, Kenai S, 2013. Capillarity of concrete
incorporating waste foundry sand. Construction and Building
32. G. Ganesh Prabhu, Jung Hwan Hyun, Yun Yong Kim, 2014. Effects of
Materials; 47:867–71.
foundry sand as a fine aggregate in concrete production. Construction
and Building Materials 70, 514–521. 52. Khatib JM., 2005. Properties of concrete incorporating fine recycled
aggregate. Cement and Concrete Resources; 35(4):763–9.
33. Ganjian Eshmaiel, Khorami Morteza, Maghsoudi Ali Akbar, 2009.
Scrap-tire-rubber replacement for aggregate and filler in concrete. 53. Kim H.K., H.K. Lee, 2011. Use of power plant bottom ash as fine
Construction and Building Materials; 23:1828–36. and coarse aggregates in high-strength concrete. Construction and
Building Materials 25, 1115–1122.
34. Ghafoori N, Bucholc J, 1996. Investigation of lignite based bottom
ash for structural concrete. Journal of Materials in Civil Engineering; 54. Kothai P. S., Dr. R. Malathy, 2014. Utilization of Steel Slag In Concrete
8(3):128–37. As A Partial Replacement Material for Fine Aggregates: ISSN: 2319-
8753.
35. Gesoglu M, Ozturan T, Guneyisi E, 2006. Effects of cold-bounded
fly ash aggregate properties on shrinkage cracking lightweight 55. Kou SC and Poon CS., 2009. Properties of self-compacting concrete
concretes. Cement and Concrete Composites; 28(7):598–605. prepared with recycled glass aggregate. Cement and Concrete
Composites; 31:107–13.
36. Gorai B, R.K. Jana, Premchand, 2003. Characteristics and utilization
of copper slag a review. Resources, Conservation and Recycling. 56. Kou S., Zhan B., and Poon C., 2012. Feasibility study of using recycled
39(4), 299–313. fresh concrete waste as coarse aggregates in concrete, Construction
and Building Materials, vol. 28, no. 1, pp. 549–556.
37. Grübl P, Nealen A, 1998. Construction of an office building using
concrete made from recycled demolition material. Darmstadt 57. Krishna Prasanna P and Venkata Kiranmayi K, 2014. Steel Slag as a
Concrete; 13:163–77. Substitute for Fine Aggregate in High Strength Concrete: International
Journal of Engineering Research & Technology (IJERT) ISSN: 2278-
38. Guedes M, Evangelista L, de Brito J, Ferro A., 2013. Microstructural
018 October.
characterization of concrete prepared with recycled aggregates.
Microscopy and Microanalysis; 19(5): 1222–30. 58. Kurama H, Kaya M., 2008. Usage of coal combustion bottom ash in
concrete mixture. Construction and Building Materials; 22(9):1922–8.
39. Guney Y, Sari YD, Yalcin M, Tuncan A, Donmez S, 2010. Re-usage of
waste foundry sand in high-strength concrete. Waste Management 59. Kwan WH, Ramli M, Kam KJ, Sulieman MZ., 2012. Influence of the
(Oxford); 30:1705–13. amount of recycled coarse aggregate in concrete design and durability
properties. Construction and Building Materials; 26(1):565–73.
40. Gupta Tirlok, Sandeep Choudhary and Ravi K. Sharma, 2014.
Assessment of mechanical and durability properties of concrete 60. Lee Gerry. Poon Chi Sun, Wong Yuk Lung and Ling Tung Chai, 2013.
containing waste rubber tire as fine aggregate, Construction and Effects of recycled fine glass aggregates on the properties of dry-
Building Materials 73, 562-574. mixed concrete blocks, Construction and Building Materials (38),
638- 643.
41. Hamemrnik, J.D., Frantz, G.C., 1991. Physical and chemical properties
of municipal solid waste fly ash. ACI Materials Journal 88 (3), 294–301. 61. Levy Salomon and Helene Paulo, 2007. Durability of concrete mixed
with fine recycled aggregates, Exacta, Sao Paulo, Vol. 5, No. 1, 25-34.
42. Hisham Qasrawi, Faisal Shalabi, Ibrahim Asi, 2009. Use of low CaO

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

236 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial replacement of natural sand with recycled waste materials in concrete for sustainable construction practices

62. Li F., 1999. Test research on copper slag concrete. J Fuzhou University 82. Shao Y, Lefort T, Moras S, Damian R., 2000. Studies on concrete
(Natural Science Edition); 127(5):59–62. containing ground waste glass. Cement and Concrete Resources;
30(1):91–100.
63. Lotfy Abdurrahmaan and Al-Fayez Mahmoud, 2015. Performance
Evaluation of Structural Concrete Using Controlled Quality Coarse 83. Shayan A, Xu A., 2004. Value-added utilization of waste glass in
and Fine Recycled Concrete Aggregate, Cement and Concrete concrete. Cement and Concrete Resources; 34(1):81–9.
Composites.
84. Shayan A, Xu A, 2006. Performance of glass powder as a pozzolanic
64. Madheswaran C. K., P. S. Ambily, K. Dattatreya, N. P. Rajamane, 2014. material in concrete, a field trial on concrete slabs. Cement and
Studies on use of Copper Slag as Replacement Material for River Concrete Resources; 36(3):457–68.
Sand in Building Constructions: Journal of Institutions of Engineers
85. Shi-Cong K, Chi-Sun P., 2009. Properties of concrete prepared with
India Ser. A (July–September) 95(3):169–177.
crushed fine stone, furnace bottom ash and fine recycled aggregate
65. Matias D, de Brito J, Rosa A, Pedro D, 2013. Mechanical properties as fine aggregate. Construction and Building Materials; 23:2877–86.
of concrete produced with recycled coarse aggregates – influence
86. Siddique R, Schutter G, Noumowe A, 2009. Effect of used-foundry
of the use of superplasticizers. Construction and Building Materials;
sand on the mechanical properties of concrete. Construction and
44:101– 9.
Building Materials 23, 976–980.
66. Metwally IM, 2007. Investigations on the performance of concrete
87. Siddique R, Gupta R, Kaur I., 2007. Effect of spent foundry sand as
made with blended finely milled waste glass. Advanced Structural
partial replacement of fine aggregate on the properties of concrete.
Engineering; 10(1):47–53.
In: 22nd international conference on solid waste. Technology and
67. Meyer C, Egosi N, Andela C., 2001. Concrete with waste glass as Management Widener University, Philadelphia, USA.
aggregate. In: Recycling and re-use of glass cullet, proceeding of the
88. Siddique Rafat and Gurpreet Singh, 2012. Effect of waste foundry
international symposium concrete technology unit of ASCE, University
sand (WFS) as partial replacement of sand on the strength, ultrasonic
of Dundee.
pulse velocity and permeability of concrete: Construction and Building
68. Mohammed, B.S., Anwar Hossain, K.M., Eng Swee, J.T., Wong, G., Materials26;416–422.
Abdullahi, M., 2012. Properties of crumb rubber hollow concrete
89. Silvestravieiete I, Sleinotaite-Budriene L, 2002. Possibility to use
block. Journal of Cleaner Production. 23, 57-67.
scrap tires as an alternative fuel in cement industry environmental
69. Nayef AM, Fahad AR, Ahmed B, 2010. Effect of micro-silica addition research. Engineering Management; 3(21):38–48.
on compressive strength of rubberized concrete at elevated
90. Singh Gurpreet and Rafat Siddique, 2012. Abrasion resistance and
temperatures. Journal of Material Cycles and Waste Management;
strength properties of concrete containing waste foundry sand (WFS).
12:41–9.
Construction and Building Materials 28, 421–426.
70. Ozbay E, Lachemi M, Sevim UK., 2011. Compressive strength, abrasion
91. Singh, M., Siddique, R., 2014. Strength properties and micro-
resistance and energy absorption capacity of rubberized concretes
structural properties of concrete containing coal bottom ash as
with and without slag. Materials and Structures; 44:1297–307.
partial replacement of fine aggregate. Construction and Building
71. Papakonstantinou CG, Tobolski MJ., 2006. Use of waste tire steel Materials, 50, 246-256.
beads in Portland cement concrete. Cement and Concrete Resources;
92. Sobolev K, Turker P, Soboleva S, Iscioglu G, 2006. Utilization of
36(9):1686–91.
waste glass in ECO cement, strength properties and microstructural
72. Park SB, Lee BC, Kim JH., 2004. Studies on mechanical properties observations. Waste Management; 27(7):971–6.
of concrete containing waste glass aggregate. Cement and Concrete
93. Soundar Rajan M, 2014. Study on strength properties of concrete by
Resources; 34(12):2181–9.
partially replacement of sand by steel slag: International Journal on
73. Pedro D., de Brito J. and Evangelista L., 2014. Influence of the use of Engineering Technology and Sciences. ISSN (P): 2349-3968, ISSN
recycled concrete aggregates from different sources on structural (O): 2349-3976.
concrete, Construction and Building materials 71, 141-151.
94. Subathra Devi V and B.K. Gnanavel, 2014. Properties of concrete
74. Pelisser, F., Zavarise, N., Longo, T.A., Bernardin, A.M., 2011. Concrete manufactured using steel slag. Procedia Engineering 97, 95–104.
made with recycled tire rubber: effect of alkaline activation and silica
95. Tam V. W. Y. and Tam C. M., 2007. Economic comparison of recycling
fume addition. Journal of Cleaner Production 19 (6), 757-763.
over-ordered fresh concrete: a case study approach, Resources,
75. Poon C-S, Chan D., 2007. The use of recycled aggregate in concrete in Conservation and Recycling, vol. 52, no. 2, pp. 208– 218.
Hong Kong. Resources Conservation and Recycling; 50(3):293–305.
96. Tamil Selvi P, Lakshmi Narayani P and Ramya G, 2014. Experimental
76. Rakshvir M, Barai SV., 2006. Studies on recycled aggregates-based Study on Concrete Using Copper Slag as Replacement Material of
concrete. Waste Management Resources; 24(3):225–33. Fine Aggregate: Journal of Civil and Environmental Engineering, 4:5.
77. Ravindrarajah Sri R. and Tam C. T., 1987. Recycling concrete as 97. Thomas Blessen Skariah, Ramesh Chandra Gupta, Pawan Kalla
fine aggregate in concrete, The International Journal of Cement and Cseteneyi Laszlo, 2014. Strength, abrasion and permeation
Composites and Lightweight concrete, Volume 9, Number 4, 235-241. characteristics of cement concrete containing discarded rubber
fine aggregates, Construction and Building Materials 59, 204-212.
78. Sani Mohd Syahrul Hisyam, Muftah Fadhluhartini bt, Muda Zulkifli,
2010. The properties of special concrete using washed bottom 98. Tighe S, West S, Butler L, 2013. Guidelines for the Selection and Use
ash (WBA) as partial sand replacement. International Journal of of Coarse Recycled-Concrete Aggregates in Structure Concrete.
Sustainable Construction, Engineering and Technology; 1(2):56–76. Transportation Research Record: Journal of the Transportation
Research Board; 2335:3-12.
79. Segre N, Joekes I., 2000 Use of tire rubber particles as addition to
cement paste. Cement Concrete and Resources; 30(9):1421–5. 99. Topcu, I.B., Canbaz, M., 2004. Properties of concrete containing waste
glass. Cement and Concrete Research 34 (2), 267–274.\
80. Seo T, Lee M, Choi C, Ohno Y., 2010. Properties of drying shrinkage
cracking of concrete containing fly ash as partial replacement of fine 100. Topcu, I.B., Avcular, N., 1997. Collision behaviors of rubberized
aggregate. Magazine of Concrete Research; 62(6):427–33. concrete. Cement and Concrete Resources. 27 (12), 1893-1898.
81. Serifou Mamery, Sbartai Z. M., Yotte S., Boffoue M. O., Emeruwa 101. Topcu Llker Bekir, Bilir Turhan, 2010. Effect of non-ground-
E. and Bos F., 2013. A study of concrete made with fine and granulated blast-furnace slag as fine aggregate on shrinkage
coarse aggregates recycled from fresh concrete waste, Journal of cracking of mortars. ACI Materials Journal; 107– M61 (November–
Construction Engineering, Volume 2013, Article ID 317182. December):545–53.

Organised by
India Chapter of American Concrete Institute 237
Session 2 C - Paper 4

102. Topcu Llker Bekir, Bilir Turhan, 2010. Effect of bottom ash as fine of steel slag on mechanical properties and durability of concrete:
aggregate on shrinkage cracking of mortars. ACI Materials Journal; Construction and Building Materials 47, 1414–1420.
107–M08 (January–February):48–56.
109. Wu W, Zhang W, Ma G, 2010. Mechanical properties of copper slag
103. Trakool Aramraks, 2006. Experimental study of concrete mix reinforced concrete under dynamic compression. Construction and
with bottom ash as fine aggregate in Thailand. Symposium on Building Materials ; 24(6):910–7.
infrastructure development and the environment. University of
110. Wu Wu, Zhang Weide, Ma Guowei, 2010. Optimum content of copper
Philippines, Diliman, Quezon City; 7–8 December. p. 1–5.
slag as a fine aggregate in high strength concrete. Materials and
104. Turgut P, Yahlizade ES, 2009. Research into concrete blocks with Design 31, 2878–2883.
waste glass. International Journal of Environmental Science and
111. Yuksel Isa, Bilir Turhan and Ozkan Omer, 2007. Durability of concrete
Engineering; 1(4):202–8.
incorporating non-ground blast furnace slag and bottom ash as fine
105. Wainwright P, Trevorrow A, Yu Y, Wang Y, 1994. Modifying the aggregate, Building and Environment 42, 2651-2659.
performance of concrete made with coarse and fine recycled
112. Yuksel I, Genc A., 2007. Properties of concrete containing non ground
aggregates. In: Third international RILEM symposium on demolition
ash and slag as fine aggregate. ACI Materials Journal; 104(4):397–
and reuse of concrete and masonry; October. p. 319–30.
403.
106. Wang, Y., Wu, H.C., Li, V.C., 2000. Concrete reinforcement with
113. Zaharieva R, Buyle-Bodin F, Skoczylas F, Wirquin E., 2003.
recycled fibers. Journal of Materials and Civil Engineering 12 (4),
Assessment of the surface permeation properties of recycled
314-319.
aggregate concrete. Cement and Concrete Composites; 25(2):223–
107. Wang HY, Chen BT, Wu YW, 2013. A study of the fresh properties 32.
of controlled low strength rubber lightweight aggregate concrete
114. Zong L. The replacement of granulated copper slag for sand concrete,
(CLSRLC). Construction and Building Materials; 41:526–31.
2003. Journal of Qingdao Institute and Architectural Engineering ;
108. Wang Qiang, Yan Peiyu, Yang Jianwei, Zhang Bo, 2013. Influence 24(2):20–2.

V.S. Ram Prasad

Under-Graduate student,
Department of Civil Engineering
Malaviya National Institute of Technology, Jaipur

S. Ashwin Bharathwaj
Under-Graduate student,
Department of Civil Engineering,
National Institute of Technology, Tiruchirapalli

Anto Vibin Marckson A.

Research Associate,
Department of Civil Engineering,
Indian Institute of Technology Madras (IITM),
Research interests include Durability of concrete, Cement based composites, Porosity characterisation,
Cement chemistry, Repair and Rehabilitation, Recycled Aggregates and Corrosion of concrete structures.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

238 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 3 A
Session 3 A - Paper 1

Strength and Permeability of Porous Concrete

with Polypropylene Fibres
Ominda Nanayakkara* and Puyue Gong**
Lecturer*, Graduate Student**
Department of Civil Engineering, Xi’an Jiaotong Liverpool University, Suzhou, Jiangsu, China

Abstract Strength of porous concrete can be improved by

introducing other additives, in addition to the chemical
The strength of porous concrete is a critical factor as
porous concrete should be able to carry a designated admixtures, into the concrete mix. Those are fibres,
load during its applications in addition to the rate of water polymers, rubbers and pozzolanic materials. It has
permeability. This project aimed to obtain enhanced been investigated (Huang, 2010) that the use of the
compressive strength by introducing polypropylene fibres combination of latex polymer, natural fine aggregate, and
in to porous concrete. The coarse aggregate is of 10 mm polypropylene fibre could produce acceptable pervious
maximum single graded and fibres are of 19 mm length concrete with both enough permeability and strength
throughout the project. A series of mix proportions were properties. The same investigation has concluded,
used to investigate the effect of fine aggregate content, however, that the addition of fibre does not significantly
fibre content, GGBS content and the effect of compaction. enhance the strength properties of pervious concrete.
It is experimentally observed that the addition of fine This observation has been explained considering the fact
aggregate beyond 8% with 10 mm coarse aggregate that fibre was not fully dispersed and evenly distributed
does not improve the compressive strength of concrete. in the mix. The amount of fibre used is 0.9 kg/m3 which
Results also show that the addition of polypropylene fibres approximately is 0.1% in volume basis. In another study
into porous concrete could improve the compressive (Gesoglu, 2014) it has been investigated that rubber
strength if the amount of fibres is in the range of 0.1% incorporated pervious concretes had lower compressive
0.3%, however showing minimum level of permeability. It strength, splitting tensile strength, and modulus of
can be concluded that addition of fibres could significantly elasticity and a lower rate of permeability. However, the
improve the compressive strength, however further compressive strength achieved in rubberized pervious
studies are needs as the type of coarse aggregate could concrete is 6.5 MPa with a permeability of 0.025 to 0.61
have a greater influence on the amount of fibres needed to cm/s, which lies in the recommended range. The addition
obtain the desired strength. of cement replacement materials in porous concrete has
not been widely studied; this could be due the fact that
Keywords: Porous concrete, Permeability, Polypropylene the replacement materials can significantly reduce the
fibres, Compressive strength level of permeability of concrete. The contribution of silica
fume to the mechanical properties of porous concrete has
Introduction been investigate and concluded that effect is minor (Agar-
Porous concrete is widely used mainly as pervious Ozbek, 2013). The mechanical properties tests of porous
concrete when the water permeability through concrete is concrete have been conducted at 28 days.
essential. These types of concrete’s main applications are
This study has been set to investigate the effectiveness
pavements, residential roads, parking lots and sidewalks.
of polypropylene fibres in enhancing the compressive
Pervious concrete can be considered as environmental
strength of porous concrete while expecting to achieve
friendly mainly because of its high level of permeability
desired level of permeability. The size of coarse aggregate
and noise absorption compared to conventional concrete.
is 10 mm and the length of fibre is 19 mm. An attempt
The porous condition is achieved by introducing no or
has also been made to investigate the effect of Ground
little amount of fine aggregate into the concrete mix, as a
Granulated Blast Furnace Slag (GGBS) on the compressive
result, a large amount of voids between coarse aggregate
strength development.
particles remain at hardened concrete. However, as a
result of large void content, the strength capacity reduces
significantly and it can be considered as poor compared to Experimental Setup
conventional concrete. Strength reduction can obviously Although the purpose of permeable concrete is to achieve a
limit its applications especially when the strength significant level of permeability, the materials used in this
demands, for example, applications in a high traffic project are same as conventional concrete. To achieve a
pavement (Tennis et al., 2004). permeable condition, the amount of fine aggregate used is

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

242 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Strength and Permeability of Porous Concrete with Polypropylene Fibres

restricted. To investigate the effect of fibres, polypropylene

fibres are added to concrete. The effect of GGBS for the
strength development of the permeable concrete is also
investigated in the project. The compaction method was
uniform throughout the experimental setup, except,
one concrete specimen which was heavily compacted to
investigate the effect of the level of compaction on the
strength and the permeability.

Binding Material
Normal Ordinary Portland Cement of 42.5 strength class
and having the specific gravity of 3.15 was used in this
study. Ground Granulated Blast Furnace Slag (GGBS) Fig. 1: Sieve analysis curves of fine and coarse aggregate
having the specific gravity of 2.91 was used to investigate
permeable concrete. Polypropylene fibres with 19 mm
the effect of fine particles of the binding materials and its
length were chosen in this study. The diameter of fibre is
effectiveness in permeable concrete. The total amount of
in the range of 18 to 45 μm and the density is 0.91 g/cm3.
binding materials was kept at 350 kg/m3 while introducing
Moisture absorption rate of fibre is less than 0.1%.
30% and 50% GGBS replacement for two specimens. The
water to binder ratio was kept at 0.35 throughout the
experimental programme. Concrete Mix Design
The concrete mix design was carried out based on the
Coarse and Fine Aggregate concept that the volume of paste depends on the volume
of voids in a dry rodded coarse aggregate (NRMCA) and
The type of coarse aggregate can be categorized as
it can be shown as in the Equation 1. This concept is
maximum size of 10 mm as the percentage passing on
fundamentally used to design pervious concrete where the
the 10 mm sieve is 95.7% and the percentage retaining on
void content of concrete is considered to be significantly
the 5 mm sieve is 7.5%. All particles larger than 1.18 mm
larger with a minimum or no fine aggregate.
was removed in fine aggregate to minimize the particle
size interaction between coarse and fine aggregate. The Vp (%) = Aggregate void content (%) + CI (%) - V void (%)....1
percentage passing is 1.8% and 0.7% through the sieve
Where, Vp is the volume of paste required (%), aggregate
size of 0.15 mm and 0.075 mm, respectively. The Figure
void content is the void content of dry rodded coarse
1 shows the particle size distributions of both coarse and
aggregate (%), CI is the compaction index (%) with 1-2%
fine aggregate. Loose and compacted bulk density, void
for greater compaction and 7-8% for lighter compaction
ratio, Saturated Surface Dry (SSD) density and the water
and V void is the design void content of the pervious concrete
absorption ratio of fine and coarse aggregate are shown
mix (%).
in the Table 1.
Compaction index is taken as 5% which is reasonable for a
Polypropylene Fibres laboratory experiment. The design void content, V void, is set
at 20%. However, the design void content can be altered if
Permeable concrete strength is significantly low
fine aggregate is used in the concrete mix as that amount
compared to conventional concrete as the amount of fine
of fine aggregate enters into the void of coarse aggregate.
aggregate is low and hence the amount of mortar paste.
This is a typical phenomenon in pervious concrete and the
Fibres are introduced into permeable concrete to recover
Equation 1 cannot provide a solution to this condition.
the strength loss due to lack of mortar paste. Though
there are different types of fibres available, only plastic The mix design of all concrete specimens is shown in the
fibres can be effectively used in permeable concrete as Table 2. Specimens 1 to 5 are set to investigate the effect
water can drastically degrade natural or steel fibres. of fine aggregate content which varies from 0% to 16% as a
Polypropylene as an economical material have high volume replacement with coarse aggregate. Specimens 2
strength, elasticity, good wear-resisting and corrosion and 6 to 11 are set to investigate the effect of fibre content
resistance which make it becomes an ideal material in which varies from 0% to 0.8% as a volume percentage of

Table 1. Physical properties of aggregate

Aggregate type Loose bulk Compacted bulk Loose void ratio Compacted void SSD density (kg/ Water absorption
density (kg/m3) density (kg/m3) (%) ratio % m3) ratio (%)
Fine 1459 1666 40.4 32.0 2510 2.62
Coarse 1360 1575 48.8 40.7 2680 0.92

Organised by
India Chapter of American Concrete Institute 243
Session 3 A - Paper 1

Table 2. Concrete mix design values

Specimen Water Cement Coarse Fine aggregate Fine Fibre content GGBS content
(kg) (kg) aggregate (% coarse aggregate volume) aggregate (% volume) (% cement)
(kg) (kg)
1 123 350 1340 0 0 0 0
2 123 350 1313 2 25 0 0
3 123 350 1286 4 50 0 0
4 123 350 1233 8 100 0 0
5 123 350 1126 16 201 0 0
6 123 350 1313 2 25 0.1 0
7 123 350 1313 2 25 0.2 0
8 123 350 1313 2 25 0.3 0
9 123 350 1313 2 25 0.5 0
10 123 350 1313 2 25 0.7 0
11 123 350 1313 2 25 0.8 0
12 123 245 1313 2 25 0.5 30
13 123 175 1313 2 25 0.5 50
14 123 350 1313 2 25 0 0

concrete. Specimens 2, 12 and 13 are set to investigate Results and Discussions

the effect of GGBS content which varies from 0% to 70%.
The results are based on the concrete specimens shown
The amount of fine aggregate content was set at 2%
in the Table 2. Three cube specimens of 100x100x100
to investigate the effect of fibre content and the GGBS
mm size and one prism beam of 100x100x500 mm
content. Specimen 14 has same mix design as specimen
size were used for compressive and flexural strength
2; however, the compaction method was changed to a
measurements, respectively. One cylinder of 100 mm
continuous compaction by vibration table while all other
diameter and 200 mm height was used to determine the
specimens were compacted by both hand and vibration
permeability coefficient.
table in three layers.

Compressive Strength, Flexural Strength and

Permeability Measurement
All specimens were water submerged and cured for 28
days. Compressive strength and the flexural strength
measurements were taken according to the British
Standard (BS EN 12390-3 and 5) respectively. The centre-
point loading test method was used for the flexural
strength measurement. The effective span of the prism
beam is 300 mm and the cross sectional dimension is
100x100 mm.
Falling head permeability test was used to determine the
coefficient of permeability (k) using cylindrical concrete
Fig. 2: Sample of slump tested concrete
specimens. The k is determined the by the Equation 2
(Gesoglu, 2014). Fresh Properties
aL h ........................................................... 2 All concrete specimens were tested for the slump and a
k = At ln ( h 1 ) t c
2
zero slump was observed as shown in the Figure 2. This
Where, k is the coefficient of permeability (cm/s), a is inside is mainly due to the selected low water to cement ratio of
cross-sectional area of the buret (cm2), L is the average 0.35 and the lack of fine aggregates in the concrete mix.
thickness of the test specimen (cm), A is the average It is worth to note here that no chemical admixtures were
cross-sectional area of the test specimen (cm2), t is the used in this set of experiments. Low water to cement ratio
elapsed time between h1 and h2 (seconds), h1 is the initial is ideal for porous concrete as the there is a possibility to
head across the test specimen (cm), h2 is the final head flow away all cement particles if the water to cement ratio
across the test specimen (cm), and tc is the temperature is high. However, care should be taken to adjust the water
correction for viscosity of water (taken as 1.0 in this study). to cement ratio when GGBS as a cement replacement

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

244 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Strength and Permeability of Porous Concrete with Polypropylene Fibres

material is used since small particle size absorb more 4(a) and the 4(b), respectively. For this investigation,
water during the mixing. One of the main difficulties the fine aggregate content was kept at 2% as it is also
which was observed in this study is the difficulty in mixing aimed to obtain an appropriate permeability of porous
polypropylene fibre into concrete. The lack of mortar paste concrete. According the experimental results, it can be
in porous concrete media results significant amount of seen that the largest compressive strength is achieved
unmixed and bundled fibres at the 0.8% volume addition. when the added fibre content is 0.1%. At this level, the
compressive strength increment is about 30% compared
Compressive and Flexural Strength to concrete with no fibres. It can also be observed that
The relationship of the compressive and the flexural the compressive strength do not significantly vary when
strength against the fine aggregate content is shown the fibre content is between 0.1% and 0.3%. Compressive
in the Figure 3. This investigation was carried out as a strength cannot be improved by adding more fibres than
preliminary work to understand the effect of addition of 0.3% into porous concrete. This phenomenon can be
fine aggregate into the coarse aggregate used in this study. explained by considering the fact that the porous concrete
As expected the compressive strength increases with the is having a minimum amount of mortar paste compared to
amount of fine aggregate content and this is obviously conventional concrete. Mortar paste can positively support
because of the reduction of porous media in concrete. The the proper mixing of fibres in the concrete and provide
compressive strength values are approximately equal at appropriate bond between the paste and the fibres. It is
8% and 16% fine aggregate replacement. No experimental noted here that the compressive strength results at the
results are available for fine aggregate replacement fibre content of 0.5% seems erroneous and this can be due
beyond 16%. The above observation could be explained to improper mixing and compaction at the casting stage.
based on the fact that the size of coarse aggregate is
It can be argued that the compressive strength at 28
comparatively small and hence the addition of more sand
days could not be improved when the GGBS is added as
cannot significantly increase the density of concrete.
a cement replacement material and this is because of the
Therefore, the compressive strength remains the same.
slow strength of GGBS material. Although the Figure 4(b)
The variation of the flexural strength of concrete is not
shows that there is a compressive strength increment
comparatively significant although a slight increment of
from 0% to 30% GGBS, the strength decrease from
the strength is observed.
30% to 50%. It is noted that he strength at 0% GGBS is
same as the strength at 0.5% fibre content in Figure
4(a) and it assumed erroneous. Another disadvantage
of adding GGBS into porous concrete is because of the
Calcium Hydroxide (Ca(OH)2) dissolution into water at
the submerged curing condition for 28 days. Ca(OH)2 is
essential for the strength development in GGBS added
concrete.

Permeability
Permeability measurements against the fine aggregate
content and the fibre content taken by falling head
permeability test are shown in the Figure 5 (a) and (b).
Fig. 3: Compressive and flexural strength against the fine The results show the obvious expectation of the behaviour
aggregate content
of porous concrete, i.e. the permeability decreases with
The compressive strength variation against the fibre increasing fine aggregate content. It is desirable to have
content and the GGBS content is shown in the Figure a permeability rate of 0.2 cm/s or above if the porous

Fig. 4: Compressive strength against the Fibre and GGBS content

Organised by
India Chapter of American Concrete Institute 245
Session 3 A - Paper 1

Fig. 5: Permeability against the fine aggregate content

concrete is to be considered as pervious concrete. permeability of concrete. The Figure 6 shows the strength
However, in this study, the permeability values for all and permeability results of specimens 2 and 14, where,
concrete specimens show a value less than 0.2 cm/s. This the specimen 14 was subject to continuous vibration table
is mainly due to the fact that the presence of small size compaction. The compressive strength is approximately
coarse aggregate with a large amount of cement, 350 kg/ doubled and the permeability is approximately reduced by
m3, in the mix. Considering the amount of mortar paste 30% due to the variation of compaction.
needed for mixing with fibres and the permeability rate,
This result can highlight the importance of compaction
a 2% of fine aggregate content was therefore used for the
investigation of the effect of fibre and GGBS content. to achieve a higher strength; however, a care should be
taken as the permeability significantly decreases. Hence,
Although the strength increases with the amount of fibre in porous concrete, the compaction method must be
content in the range of 0.1% to 0.3%, the permeability consistent to achieve the required design specifications.
decreases. It is, generally, considered that the porosity
increase with the amount of fibres in concrete as the Conclusions
fibres in concrete make the concrete mixing difficult and
not smooth. This contradicts the results in this study. This Some experimental results are not conclusive enough;
behaviour is due to the fact that the significant amount of therefore, authors suggest that further studies about
voids in porous concrete could be filled with fibres and polypropylene added porous concrete are needed. The
hence reducing the permeability. A drastic increment permeability results can be sensitive to the method of
in the permeability at the fibre content of 0.5% could be compaction, the experimental setup and the number of
linked with the erroneous strength data shown in Figure specimens used in the test which could also be considered
4(a). However, the permeability results are not conclusive in further improvements of this work. The conclusions
for other values of fibre contents. based on the experimental results obtained in this study
can be summarized as follows.
Effect of Compaction 1. If the maximum size of coarse aggregate is 10 mm,
One of main uses of porous concrete is as pervious replacement of fine aggregate beyond 8% with
concrete so that the water could easily pass through the coarse aggregate does not improve the compressive
concrete media. However, the compaction condition itself strength. Also this would drastically reduce the
can significantly affect the compressive strength and the permeability level of porous concrete.
2. It can be concluded that the addition of polypropylene
fibres into porous concrete could improve the
compressive strength if the amount of fibres is in the
range of 0.1% 0.3%. Further increment of the fibre
content do not support for the strength improvement.
3. As the strength increases, the permeability at the
fibre content of 0.1% to 0.3% shows its minimum
values. Further increment of fibre content could
increase the permeability level as the void content
increase, however this is satisfactorily conclusive.
4. GGBS addition in porous concrete will not improve
Fig. 6: Relationship between the strength and the permeability the strength in short term. A further study is needed
against the compaction to investigate the effect on long term strength

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

246 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Strength and Permeability of Porous Concrete with Polypropylene Fibres

development and to investigate whether Ca(OH)2 Skokie, Illinois, and National Ready Mixed Concrete Association,
Silver Spring, Maryland, USA, 2004, 36 pages.
dissolution could adversely affect the strength
development of porous concrete. 2. Huang, B., Wu, H., Shu, X., and Burdette, E. D., 2010. Laboratory
evaluation of permeability and strength of polymer-modified
5. Compressive strength of porous concrete significantly pervious concrete. Construction and Building Materials, 24: 818-
depends on the method of compaction; therefore, the 823.
compaction method should be consistent to achieve 3. Gesoglu, M., Guneyisi, E., Khoshnaw, G., and Ipek, S., 2014.
the designated level of permeability. Investigating properties of pervious concretes containing waste
tire rubbers. Construction and Building Materials, 63: 206-213.

Acknowledgement 4. Agar-Osbek, A. S., Weerheijm, Jaap., Schlangen, E., and Breugel, K.

van., 2013. Investigating porous concrete with improved strength:
Authors would like to acknowledge the experimental Testing at different scales. Construction and Building Materials, 41:
facilities and the financial support received from the Xi’an 480- 490.
Jiaotong-Liverpool University and the Teaching assistants 5. Pervious concrete: Mixture proportioning, National Ready Mixed
and Technicians for their contribution to the work reported Concrete Association (NRMCA), http://www.perviouspavement.org/
materials.html, Accessed: July 2015.
in the paper.
6. British Standard, 2009. Compressive strength of test specimens.
References BS EN 12390-3. British Standard, 2009. Flexural strength of test
1. Tennis, P. D., Leming, M. L., and Akers, D. J., 2004. Pervious specimens. BS EN 12390-5.
Concrete Pavements. EB302.02, Portland Cement Association,

Dr. Ominda Nanayakkara

Dr. Ominda Nanayakkara is a Lecturer in the Department of Civil Engineering at Xi’an Jiaotong-Liverpool
University, China. His research interests include durability of concrete and non-destructive testing of
concrete. His current research projects focus on patch repairing of concrete, effect of permeability on
corrosion initiation, chloride ion diffusion in concrete-repair interface and fibre reinforced permeable
concrete.

Organised by
India Chapter of American Concrete Institute 247
Session 3 A - Paper 2

Dynamic Shear Resistance of RC Beams

based on Modified Field Compression Theory
K. Fujikake and A. Somraj
Department of Civil and Environmental Engineering, National Defense Academy

Abstract extending the modified compression field theory to

dynamic loading.
The aim of this study was to develop an analytical model
to estimate the dynamic shear capacity of RC beams
which may exhibit diagonal tension failure under impact Dynamic shear resistance based on modified
and blast loadings. Thus, the modified compression field compression field theory
theory has been extended to dynamic loading in this study. The modified compression field theory proposed by
The developed analytical model has been applied to the Vecchio and Collins (1986) is a generalized analytical
experimental results obtained from rapid loading tests of method to evaluate the shear resistance of a RC member.
RC beams. As a result, the developed analytical model has In this theory, it is assumed that a concrete element, in
been in good agreement with the experimental results. which diagonal cracks are uniformly formed, should
Keywords: dynamic shear resistance, RC beam, modified be resisted by two principle stress components after
field compression theory cracking which are parallel to and perpendicular to the
diagonal cracks. In addition to this basic assumption,
Introduction the equilibrium condition, compatibility condition and the
stress-strain relationships for concrete and reinforcing
It is very important to develop an analytical method
steel make it possible to evaluate the shear resistance of
to evaluate the shear capacity of RC members under
the RC member.
impact and/or blast loading to prevent unfavorable brittle
failure. Thus, we already developed a strut-and-tie model Nakamura and Higai (1994) proposed an analytical method
considered loading rate effects to predict the shear with the modified compression field theory to evaluate
capacity of RC beams with a shear-span-to depth ratio the shear capacity of RC beams subjected to axial force
of less than 2.5, which are normally called deep beams and bending moment. In their analytical method, the RC
(Smraj et al., 2014). However, very few analytical models beams are divided into a number of thin-layered concrete
are available for predicting the dynamic shear resistance elements as shown in Figure 1. By assuming the shear
of RC beams with a shear span-to-depth ratio larger than stress distribution and axial strain distribution due to
2.5, which may exhibit diagonal tension failure. Therefore, bending moment, the modified compression field theory
this study intends to develop an analytical model to is applied to each thin-layered concrete element as shown
estimate the dynamic shear capacity of RC beams by in Figure 2.

Fig. 1: Application of modified compression field theory to RC beam

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

248 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Dynamic Shear Resistance of RC Beams based on Modified Field Compression Theory

in which fcl = compressive strength of concrete under

static loading, fo sc = 1.2 # 10 (1/s) [4]. The dynamic
-5

tensile strength of concrete ( ftd ) can also be given by

ftd = ft #0.00126 G Log 10 S fo X J &

fo 3.373 ..................................(4)
ss

where ft = tensile strength of concrete under static

loading= = 0.23fcl , fo st = 1.0 # 10 (1/s) . The dynamic
2/3 -7

yield strength of reinforcing steel ( fsyd ) can be determined

by
fsyd = fsy (1.202 + 0.040 # Log 10 fo ) $ fsy ..........................(5)

in which fsy = yield strength of reinforcing steel under

static loading. The stress-strain relationships considered
the rate effects for concrete and reinforcing steel are
shown in Figures 4 and 5.

Fig. 2: Stresses and strains at i-th thin-layered concrete element

(a) in compression

Fig. 3: Relation between loading rate and strain rate in pure

shear deformation
In this study, the analytical method with the modified
compression field theory proposed by Nakamura and
Higai is extended to dynamic loading, in which the loading
rate effects on the compressive and tensile strengths of
concrete and the yield strength of reinforcing steel are
fully considered. As shown in Figure 3, we assume that a
simply supported RC beam subjected to midspan loading (b) in tension
may be deformed in pure shear. Thus, the relationship
Fig. 4: Stress-strain relationship for concrete
between the midspan deformation rate ( do ) and shear
strain rate ( co ) can be given as
co = 2do /L .......................................................................(1)

where L = span.
By considering the relationship between the principle
strain rate ( fo ) and shear strain rate ( co ), the following
equation can be derived.
fo = do /L ........................................................................(2)

In this study, by using the principle strain rate given by

Eq.(2), the dynamic compressive strength of concrete ( fcdl
) can be calculated by

fcdl = fcl @ fo C
0.006# Log 10 S fo X&
fo 1.05
fo cc
................................................(3)
cc Fig. 5: Stress-strain relationship for reinforcing steel

Organised by
India Chapter of American Concrete Institute 249
Session 3 A - Paper 2

Fig. 6: Calculation flow for developed analytical model

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

250 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Dynamic Shear Resistance of RC Beams based on Modified Field Compression Theory

Figure 6 shows the calculation flow for evaluating the loading rates controlled by a midspan deflection rate were
dynamic shear resistance of RC beam. In this calculation, 4.0x10-5, 4.0x10-2, 4.0x10-1 and 2.0x100 m/s.
the dynamic ultimate resistance of a RC beam is calculated
Figure 9 shows the typical failure modes of the RC
by increasing the curvature of the RC beam’s section beams obtained in the rapid loading test. As the shear
under constant shear and axial force condition. Based on reinforcement ratio increases, the failure behavior
the result by Nakamura and Higai, the following failure changes from brittle to ductile. Diagonal tension failure
criteria are employed to determine the failure mode of the was observed for S0 specimens. The failure mode of S35
RC beam. While the equilibrium condition of axial force in was characterized as flexure-shear failure which started
a section is satisfied, if any stress and strain component with flexure failure and was followed by shear failure due
cannot be resolved to meet the other conditions, the RC to crushing of concrete in the compression zone. S70
beam should fail in shear. On the other hand, when the specimens failed in flexure.
maximum stain at the extreme compression fiber exceeds
the strain of -0.002 corresponding to the maximum Figures 10 to 12 shows the relationship between the
compressive strength, the RC beam should fail in flexure. maximum capacity and shear span-to-depth ratio
calculated by the developed analytical model for each
shear reinfrcement ratio, in which the dynamic flexure
Veryfication of Developed Analytical Model capacity and static shear capacity were also calculated by
To assess the validity of the modified compression field Fujikake et al. (2009) and by Niwa et al. (1986), respectively.
theory extended to dynamic loading, rapid loading tests The calculated results by the developed analytical model
for RC beams were performed as shown in Figure 7. generally agree well with the experimental results.
Figure 8 shows the detail of RC beams used in this study. According to the calculated results for the RC beams
All the RC beams had the same cross-sectional dimension without shear reinforcement as shown in Figure 10, the
of 120mm in width and 220mm in height and 1,500mm in RC beams with shear span-to-depth ratios from 4 to 6
length. fail in shear at the loading rate of 4.0x10-5 (m/s), but fail in
flexure at the loading rate of 2.0x100 m/s. Therefore, the
developed analytical results clearly show that, depending
on the shear span-to-depth ratio, the failure mode may
change from shear at the static loading rate to flexure at
the high loading rate.

Fig. 7: Rapid loading test for RC beams Fig. 8: RC beam specimens

The shear reinforcement ratios were 0, 0.35 and 0.70%

Conclusions
(S0, S35, S70). Deformed reinforcing bars of D19 and
D6 were used for longitudinal reinforcement and shear Based on the results presented in this paper, the following
reinforcement, respectively. The yield strengths of D19 and conclusions were drawn.
D6 were 366 and 316 MPa, respectively. The compressive 1. The analytical model based on the modified
strength of concrete was 30.0 MPa. To avoid the bond compression field theory was developed to estimate
failure between the longitudinal bars and concrete, all the the dynamic shear capacity of RC beams. Finally,
longitudinal bars were welded to 3 mm thick steel plates the developed analytical model has been in good
at the both ends of the RC beam as shown in Figure 8. The agreement with the experimental results.

Organised by
India Chapter of American Concrete Institute 251
Session 3 A - Paper 2

Fig. 9: Typical failure mode

(a) Loading rate = 4.0x10-5 (m/s) (b) Loading rate = 2.0x100 (m/s)
Fig. 10: Calculation results for S0

(a) Loading rate = 4.0x10-5 (m/s) (b) Loading rate = 2.0x100 (m/s)
Fig. 11: Calculation results for S35

(a) Loading rate = 4.0x10-5 (m/s) (b) Loading rate = 2.0x100 (m/s)
Fig. 12: Calculation results for S70

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

252 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Dynamic Shear Resistance of RC Beams based on Modified Field Compression Theory

2. The developed analytical model confirmed that 3. Nakamura, H. and Higai, T. 1994. “Evaluation of shear strength of
depending on the shear span-to-depth ratio, the failure RC beam section based on extended modified compression fields
mode may change from shear to flexure as the loading theory”, Proceedings of JSCE, No.490/V-23, pp.157-166.
rates increase. 4. Fujikake, K., Li, B. and Sam, S. 2009. “Impact Response of Reinforced
References Concrete Beam and Its Analytical Evaluation”, Journal of Structure
1. Somraj, A., Fujikake, K. and Li, B. 2014. “Influence of loading rate on Engineering, ASCE, pp.938-950.
shear capacity of reinforced concrete beams”, Journal of Protective
Structures, Vol.4, No.4, pp.521-544. 5. Niwa, J., Yamada, K., Yokozawa, K. and Okamura, H. 1986.
“Revaluation of the equation for shear strength of reinforced
2. Vecchio, F. J. and Collins, M. P. 1986. “The Modified Compression
Field Theory for Reinforced Concrete Elements Subjected to Shear”, concrete beams without web reinforcement”, Proceedings of JSCE,
ACI Journal, Vol.83, No.2, pp.219-231. No.372/V-5, pp.167-176.

Kazunori Fujikake
Affilation: National Defense Academy, Japan
Dr. Kazunori Fujikake received his B. Eng, M. Eng and Ph. D from the University of Tsukuba, Japan. He
is a professor and head of the department in the Department of Civil & Environmental Engineering at
National Defense Academy, Japan. Before joining National Defense Academy in 1994, he worked for 8 years
as a structural engineer in a construction company, Japan. His research interests include the mechanical
properties of cementious materials under dynamic triaxial stress states and the structural performance
of reinforced structures under impact and blast loadings.

Organised by
India Chapter of American Concrete Institute 253
Session 3 a - Paper 3

Role of Advanced Non Destructive Testing in Health Assessment of

Cooling Tower Structures

Vinayak Samal and Chetan R. Raikar

M/s Structwel Designers& Consultants Pvt. Ltd., Navi-Mumbai, Maharashtra, India

Abstract Methodology to carry out visual inspection

Presented in this paper is a case study of the use of and non destructive testing
advanced non destructive testing in a hyperboloid ll Detailed visual inspection and distress mapping was
structure. The study was initially carried out by visually carried out with the available monkey ladder and
inspecting the structure with the available basic binoculars up to a height of 91m (accessible height with
resources like monkey ladder, binoculars etc. Further, the available monkey ladder).
a detailed thermography survey was carried out to
locate the hot and cold spots which would justify the ll The Thermography survey was carried out on the
hyperboloid structure at different angles to locate the
necessity to carry out NDT at those locations. Advance
hot and cold spots in order to finalise the locations for
NDT such as Ground Penetration RADAR, Auto clam
the required Non Destructive Testing.
and Permit, Microscopic Petrography and basic NDT
as Ultrasonic Pulse Velocity, Rebound Hammer, Half ll The challenging task was to erect scaffolding along the
cell Potentiometer, chemical analysis, cover meter etc hyperbolic profile of the shell. This was necessary as a
were used at site to arrive at the respective properties of conventional vertically standing scaffolding would not
concrete and its quality. allow access to the testing surface due to the peculiar
shape of the profile.
Keywords: Hyperboloid, RADAR, Autoclam, Permit,
Petrography
Thermography Survey
Principle: Infrared Thermography can be defined as the
Introduction
science of acquisition and analysis of data from non contact
The client intended to carry out a detailed visual inspection thermal imaging devices. The invisible infrared radiation
followed by an advanced non destructive testing for the emitted by bodies is converted into temperature and
qualitative and quantitative assessment of the condition of displayed as thermal images.Thermography is not only
the structure. This assessment would also assist in the a non contact technique but also totally non destructive
estimation of the required repair cost. (figure 1).
The biggest challenge faced was to evaluate the strength
and durability of the reinforcement and concrete Advance Non Destructive Testing
quantitatively while the structure was in an operational
Ground Penetration RADAR Survey
condition. Being a high rise structure (130m) the difficulty
of execution was further amplified. Ground penetration RADAR Survey (GPR) is an
electromagnetic (EM) geophysical method for high
Due to the above said challenges, it was mandatory
resolution detection, imaging and continuous mapping of
to plan a very systematical execution procedure with
sub surface. It transmits short frequency electromagnetic
certain adaptations to carry out NDT on the hyperboloid
pulses of radio frequency into the medium through
structure.
transmitted antenna.

Brief of Structure When the pulse reaches an electric interface in the

medium, some of the energy will be reflected back while
The cooling tower unit is a typical very large hyperboloid
the rest will be refracted due to a change in the di electric
structure and has a smooth curvilinear hyper parabolic
permittivity and electric conductivity of the medium.
surface shape as observed on inside and outside. The
throat area in the shell portion has a diameter of 70m A Typical GPR system has three main components:
and the bottom diameter of shell is around 118m. The Transmitter and receiver that are directly connected to an
diameter at the top level stands at about 90m. antenna and a control unit (timing).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

254 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Role of Advanced Non Destructive Testing in Health Assessment of Cooling Tower Structures

Fig. 1: Infrared Image of the cooling Tower locating hot and cold spots

In this case study, we

have carried out the
survey on the erected
scaffolding up to the
height of 25.0m at
four locations along
the external profile
of the cooling tower.
The testing for the
rest of the height was
done using a rope
and harness access
system as shown in the
photograph (figure 2).
It is recommended that
rope access technique Fig. 2: Ground penetration Radar Survey
is practiced only by
trained technicians under strict supervision to avoid fatal
accidents.
It was noticed that some of the reinforcement had de-
bonded with the cover concrete and some other areas
showed voids in the concrete (honeycombing).

Auto clam
Purpose: To determine the rate of decay of air pressure
and the volume of water penetrating into the concrete
and other porous material at a constant pressure of 0.02
bar and 0.5 bar (figure 3).
Significance: This test is used to assess the durability of
the concrete in-situ for its air and water permeability.
Based on the test results, it has been concluded that the Fig. 3: Auto clam test
water permeability is low, as is desired.
Significance: This test is used to assess the chloride ion
Permit diffusion resistance of the near surface zone of concrete
Purpose: To determination of chloride diffusion coefficient The average value recorded for the permeability index is
and to assess the resistance to chloride ingress (figure 4). 1.04x10-12 m2/s.

Organised by
India Chapter of American Concrete Institute 255
Session 3 a - Paper 3

ll Rebound hammer (IS 13311 Part 2): This test is

based on the hardness of the concrete surface and it
gives the approximate strength of the concrete surface
.Total sixteen impact readings were taken at each
location and average of middle ten was calculated after
discarding the top three and bottom three readings.
The probable accuracy of prediction of concrete
strength by the rebound hammer is + 25%.
ll Half Cell Potentiometer (ASTM C-876-80): The
test is based on the principle of measuring milli-
voltage in the circuit of reinforcement and covers
concrete using copper sulphate half cell. This method
essentially consists of measurement of the absolute
potential at concrete surface with reference to an
Fig. 4: Permit test electrode. It should have direct electrical connections
to the embedded steel. The test is fairly indicative of
Petrography corrosion response of present and future.

Purpose: The Petrography is the study of rocks/concrete The following are the limits:-
using thin section, under a petrographic microscope using ll Negative than -350mV indicate high probability of
transmitted light (figure 5). active corrosion
ll Positive than -200mV indicate high probability of No
corrosion
ll Between -200mV and -300mV indicate uncertainty
of corrosion
ll Positive readings indicate probability of insufficient
moisture in concrete
The existence of corrosion in steel is further confirmed
by high percentage of chloride and less pH value by other
methods of testing.
ll Concrete Core Extraction: This test is performed
to evaluate the actual strength of the concrete in
Fig. 5: Plane Polarized Light, Magnification PPL (Magnification the structure. It is done by extracting 75mm and
5X PPL) 100mm dia cores from the structural members of the
structures. The extracted core is taken to laboratory
Significance: This test is used to identify the minerology for the dressing and capping on both surfaces for
and to evaluate the quality of concrete through microscopic testing in the compression testing machine.
parameters such as micro cracks, texture and extent of The average test results show the strength is 64 N/
weathering etc. mm2 in pedestal area and ranges from 39 N/mm2 to
63.01 N/mm2 in shell area.
Basic Non Destructive Testing
ll Chemical Analysis (pH, Cl, So3)(Ref:IS 456:2000)
ll Ultrasonic Pulse velocity Test (IS 13311 Part 1):
This test is based on the principle of passing high It is important to check the pH, sulphate and chloride
frequency sound waves through the body of concrete of concrete and cross verification as per IS code since
and measuring the time taken. Velocity of the waves it directly correlates to durability.
through the concrete members is calculated by dividing This test indicates chloride content in concrete
the path length to the time taken. Direct method and indicates extent of corrosion of steel and pH value. If it is
indirect method are used at column and shell area low, it indicates acidic in nature. The higher percentage
depending on the access. of chlorides and sulphates indicates deterioration of
The test results obtained show good quality of concrete concrete and possibility of disintegration.
in columns and general homogeneity in concrete in Quality of the concrete in terms of pH, chloride and
shell area. sulphate content are within the specified limits.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

256 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Role of Advanced Non Destructive Testing in Health Assessment of Cooling Tower Structures

ll Carbonation Test (BS 1881 Part 201:1986) Conclusions

This test is carried out to measure the depth of 1 Due to the peculiar shape of the cooling tower
carbonated concrete from the external face. 2% structures, it is extremely expensive to use scaffolding
phenolphthalein solution is sprayed on the extracted for inspection & NDT
core sample (i.e. 25mm or 50mm diameter). If sprayed 2 It is cost effective to use techniques of access like rope
concrete turns pink it is considered as non carbonated. access coupled with scaffolding to bring in economy in
The depth of carbonation is measured in millimeters investigation.
as the depth from the external face of concrete to
the point beyond which the phenolphthalein sprayed 3 Use of Infrared Camera is an efficient way of identifying
concrete turns pink in color. probable porous/honeycombed locations in concrete.
This can be further verified by conducting NDT at these
The readings show the carbonation depth is in the locations.
range of 15-22mm for pedestal area and 11-22mm for
shell area. 4 RADAR undoubtedly is the ‘next generation’ scanning
technique for detection of imperfections in concrete
Carbonation does not indicate depth beyond 22mm. and far supersedes age old techniques like ultrasonic,
The average depth of carbonation is around 19 mm for impact echo etc.These testing are for spot testing
the pedestal and 18mm for the shell concrete. whereas RADAR can scan the whole concrete surface
It indicates, from these results that major depth of in substantially lesser time with equal or higher
cover concrete is carbonated. accuracy.
ll Cover Meter 5 In situ permeability test like Autoclam & PERMIT
This test indicates the cover of the concrete to the improve the confidence level of the engineers
reinforcement. In this case the cover is without plaster. conducting the health assessment and are also useful
The cover of concrete above the reinforcement is to carrying out quantitative /qualitative assessment of
adequate at almost all locations. concrete for durability.

Mr. Chetan Raikar

Civil Engineer by basic qualification and post graduate from UK in the field of Conservation of Heritage
Structures.
Currently CMD of M/s. Structwel Designers & Consultants Pvt. Ltd., a 48 year old Consultancy Organisation
based in Pune, Mumbai, Bangalore, Dubai
Structural Engineer, Conservationist of Heritage Structure, and material Scientist by expertise.
Responsible for carrying out conservation of several prestigious projects like CST Railway Station, Taj after
26/11 terrorist attack, Haji Ali Dargah, etc.
Responsible for Health Assessment of several prestigious projects like Kolkata Metro, entire plant of
Reliance Jamnagar, etc.
Recipient of following awards :-
s American Concrete Institute (ACI) Young Member Award for Professional Achievement for the year 1999.
s Appointed as Governor’s Nominee on the Buildings & Works Committee of the University of Mumbai, 1998. The youngest person
to be appointed for this post in the record of University of Mumbai.
s Has two patents in innovation in machine development in his name.
s Appointed Expert Member on Board of Governors of the Mumbai Metropolitan Region - Heritage Conservation Society - 2008.
s Excellence Vocational Award by Rotary International District: 3140 – Year 2009.
s Recipient of “Maharashtracha Kohinoor” Award on 16th March, 2013.
s Certified Assessor by National Accreditation Board for Testing & Calibraton Laboratories,(N.A.B.L.) Ministry of Science &
Technology, Government of India, New Delhi – 2001 onwards
s Lead Auditor of ISO 9001-2008 and ISO/IEC 17025:2005.

Mr. Vinayak D. Samal

Bachelor of Engineering {Civil} in 2006 from Sardar Patel College of Engineering, Mumbai University,
Mumbai.
Working as a Engineer R&D (since 9 year) for Structwel Designers and Consultants Pvt. Ltd in Turbhe lab.
Management of Material Testing Laboratories of all kinds like, Geotechnical, Soil, Concrete, Cement and
Analytical testing services.
Management for execution and interpretation of Geotechnical works such as SPT, PLT and Boring.
Supervision, Execution and interpretation of specialized NDT tests like Ground penetration Radar Survey
and Thermography Survey.

Organised by
India Chapter of American Concrete Institute 257
Session 3 A - Paper 4

Study of the Post-Cracking Behaviour of Steel and Polymer Fibre

Reinforced Concretes
Sujatha Jose, Stefie J. Stephen, Ravindra Gettu
IIT Madras, Chennai 600036, India

Abstract the work of various researchers (Østergaard, 2003; Sousa

In the present work, an evaluation of the performance & Gettu, 2006; Löfgren et al., 2005); the cracked hinge
of concrete with steel and polymer fibres is done using model is employed to perform the inverse analysis.
notched beam tests as per EN 14651:2005 and RILEM
TC162-TDF. The steel fibre reinforced concrete is found Experimental Programme
to have higher flexural toughness than the polymer fibre
reinforced concrete, for the same volume fractions, The test procedure followed in this work for the
although the scatter in the former is higher. A comparison flexural toughness characterization conforms to the
of the equivalent flexural strength with the residual EN 14651:2005 (E) standard and the RILEM TC 162-TDF
strengths is also presented for both types of fibre Recommendations. The configuration of the notched beam
reinforced concretes. The bilinear stress-crack opening test with center-point loading (CPL) is given in Figure 1.
diagram (σ-w) is obtained from inverse analysis using
the experimentally-obtained load-crack mouth opening,
and the differences among the different concretes
are discussed.
Keywords: Flexural toughness, Notched beam test,
Crack mouth opening displacement, Inverse analysis

Introduction
The benefit of fibre reinforced concrete (FRC) over
conventional concrete is its improved energy absorption
capability, fatigue resistance, fracture energy, ductility in
tension and compression, flexural strength and toughness,
resistance against impact loading, abrasion and durability
(Rossi et al., 1986; Gopalaratnam & Gettu, 1995; Mindess
et al., 1986; Granju et al., 2005). Polymer microfibres
have been typically used to mitigate plastic shrinkage Fig. 1: Experimental Setup
cracking though more recently polymer macrofibres are
The specimens were cured in the moulds for 24 h after
being employed in structural applications (Buratti et al.,
casting, demoulded and maintained in a mist room for
2011). However, knowledge regarding the mechanical
the next 27 days. A notch of 25 mm depth and about 3
characteristics of synthetic fibre reinforced concrete is
limited. The present work characterizes and compares the mm width was cut at mid-span on a face perpendicular
flexural toughness of polymer and steel fibre reinforced to the cast surface at least 3 days before testing.
concretes. There are various guidelines and standards The properties of the fibers used are listed in Table 1. In
for such characterization (Gopalaratnam & Gettu, 1995;
this study, two series of specimens were cast with M35
Banthia & Mindess, 2004; ACI 544.2R-89, 2009), and
grade concrete, one with the steel fibres incorporated at
Gopalaratnam & Gettu (1995) advocated the use of the
notched beam test in a servo-controlled testing machine the dosages of 10, 15, 20 and 30 kg/m3 and the second
using Crack Mouth Opening Displacement (CMOD) with the polymer fibres at the dosages of 2.5, 3.75 and
control, which is followed here. Further, the load-CMOD 5 kg/m3, and in both cases reference concrete specimens
curve obtained experimentally is used for determining the were made without any fibres. Compression test results
tensile behaviour of fibre reinforced concrete, extending did not indicate any significant change in strength.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

258 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Study of the Post-Cracking Behaviour of Steel and Polymer Fibre Reinforced Concretes

Table 1 Details of the fibres used (as given by the manufacturer)

Material Fibre type Specific Length Diameter Tensile

gravity (mm) (mm) strength
(g/cc) (MPa)
Steel Cold drawn, 7.8 60 0.75 1225
hooked-ended,
collated
Polymer Polypropylene 0.92 40 0.44 620

Experimental Results and Discussions

Flexural response
Typical Load-CMOD curves for the steel fibre reinforced
concrete (SFRC) and the polymer fibre reinforced concrete Fig. 4: Typical Load-CMOD curves of PFRC
(PFRC) are presented in Figure 2.
Flexural Strength and Toughness Parameters
The flexural strength (fctf) or limit of proportionality (LOP)
and residual flexural tensile strength are obtained from
the load-CMOD curve following EN 14651:2005 (E), and
the equivalent flexural strengths are calculated from
the load-deflection curve as given in RILEM TC 162-TDF.
The value of fctf is obtained using the first peak load in the
elastic beam formula for maximum stress. The equivalent
flexural strength is calculated from the load deflection
curve based on the average load in the post crack region;
feq.2 and feq.3 are obtained by dividing DBZ, 2 and DBZ, 3 by 0.5
mm and 2.5 mm, respectively, where DBZ, 2 and DBZ, 3 are the
areas under the load-deflection curve up to the deflection
Fig. 2: Typical Load-CMOD curves of SFRC of δ2 = δL+0.65 and δ3 = δL+2.65, and δL is the deflection
at the LOP. The residual flexural tensile strength (fR,j) is
Figure 3 and 4, respectively. Note that in the mix notation, an estimate of the flexural strength retained by FRC after
M35 denotes the concrete grade, SF and PF denotes the cracking at particular crack widths of j = 0.5, 1.5, 2.5 and
use of steel and polymer fibres, and the number at the end
3.5 mm.
denotes the dosage in kg/m3. It is clear from the curves
that in the pre-peak region there is not much influence The mean values of flexural strength (fctf), and the
of the incorporation of fibres. However, in the post-crack residual flexural strengths at crack mouth openings of
phase, there is a softening behaviour for plain concrete, 0.5, 1.5, 2.5 and 3.5 mm for SFRC and PFRC are given in
whereas the curves for the SFRC are flat, with residual Table 3 and 4, respectively. The fctf values indicate that
hardening behaviour for higher dosages. In the case of there is a no significant change in flexural strength and
PFRC, flat curves were observed after a sharp drop in the the variability due to the addition of fibres. As expected,
post-peak regime, for all dosages. the residual flexural tensile strength (fR,j) values show
that the performance is better as the dosage increases,
irrespective of the type of fibre. It was also observed that
for SFRC, the residual strengths increase after a crack
mouth opening of 0.5 mm, for all dosages, reflecting
the ‘hardening-type’ behaviour. For PFRC, the residual
flexural strengths are almost the same, indicating little
change in the load-carrying capacity at higher crack
openings.
The values of the equivalent flexural tensile strength feq.2
and feq.3 for SFRC and PFRC given in Table reflect the
increase in toughness as the dosage of fibres increases.
The feq.3 values are higher than feq.2, especially at higher
dosages, for both SFRC and PFRC, as expected after the
Fig. 3: Typical Load-CMOD curves of SFRC observation of the shape of the load-CMOD curves.

Organised by
India Chapter of American Concrete Institute 259
Session 3 A - Paper 4

Table 3. Flexural strength and residual flexural tensile strength values of SFRC (mean ± std deviation)

Concrete Fibre dosage fctf fR,0.5 fR,1.5 fR,2.5 fR,3.5

(kg/m3); volume (MPa) (MPa) (MPa) (MPa) (MPa)
fraction Vf

M35SF0 0 4.83±0.48 - - - -

M35SF10 10; 0.13% 5.07±0.28 1.64±0.41 1.56±0.25 1.63±0.37 1.60±0.34

M35SF15 15; 0.19% 4.73±0.34 2.05±0.29 2.46±0.52 2.14±0.41 2.19±0.46

M35SF20 20; 0.26% 5.35±0.35 3.06±0.55 3.39±0.83 3.59±0.64 3.49±0.58

M35SF30 30; 0.38% 5.26±0.50 4.35±1.05 5.16±1.32 5.25±1.09 5.20±1.14

Table 4. Flexural strength and residual flexural tensile strength values of PFRC (mean ± std deviation)

Concrete Fibre dosage fctf fR,0.5 fR,1.5 fR,2.5 fR,3.5

(kg/m3); (MPa) (MPa) (MPa) (MPa) (MPa)
volume
fraction Vf
M35PF0 0 5.00±0.19 - - - -

M35PF2.5 2.50; 0.27% 5.13±0.39 1.52±0.13 1.30±0.08 1.27±0.06 1.16±0.14

M35PF3.75 3.75; 0.41% 5.06±0.44 2.00±0.17 1.92±0.11 2.01±0.11 1.90±0.12

M35PF5 5.00; 0.54% 5.51±0.27 2.33±0.27 2.55±0.27 2.58±0.35 2.29±0.30

Table 5. Equivalent flexural strength values of SFRC and PFRC (mean ± std deviation)

Concrete Fibre feq,2 feq,3 Concrete Fibre feq,2 feq,3

dosage (MPa) (MPa) dosage (MPa) (MPa)
(kg/m3) (kg/m3)
M35SF10 10 1.70±0.60 1.83±0.48 M35PF2.5 2.5 1.26±0.14 1.33±0.05

M35SF15 15 1.92±0.46 2.11±0.42 M35PF3.75 3.75 1.66±0.27 2.05±0.23

M35SF20 20 2.87±0.84 3.33±0.72 M35PF5 5 2.13±0.21 2.49±0.24
M35SF30 30 4.28±1.15 4.89±1.33

In order to compare the flexural toughness of steel and same volume fractions, at all crack openings. However,
polymer fibre reinforced concretes, the data for the the scatter in fR,j-values for PFRC is lower, probably due
volume fractions of about 0.26% and 0.4% are presented to larger number of fibres in PFRC than SFRC, for a
in Figure 5 and 6. It is clear that SFRC has better residual particular volume fraction. Comparing feq.2 and feq.3, it can
and equivalent flexural strengths than PFRC, at the be observed that feq.3-values are higher than feq.2-values in

Fig. 5: Comparison of flexural toughness parameters at 0.26% Fig. 6: Comparison of flexural toughness parameters at 0.40%
volume fraction volume fraction

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

260 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Study of the Post-Cracking Behaviour of Steel and Polymer Fibre Reinforced Concretes

the case of SFRC, whereas for PFRC, they are almost the Inverse Analysis
same.
Inverse analysis has been used to obtain the tension
According on the fib Model Code 2010, fR0.5 and fR2.5, are softening diagram (σ-w) from the notched beam data
related to the serviceability and ultimate limit states, (P-CMOD) using a module based on the analytical
respectively. These two parameters, along with the formulation of Stang and Olesen (Stang and Olesen, 1998,
equivalent flexural strength, feq,3, are compared for SFRC 2000; Sousa and Gettu, 2006). The analysis is repeated
and PFRC, in Figures 7-9, respectively. It can be observed for different σ-w curves until the relation that gives the
that all three parameters are similar for SFRC and PFRC P-CMOD response that best fits the experimental data.
when the volume fraction in the former is about half that The optimization algorithm used here is the probabilistic
of the latter; e.g., the parameters obtained for 0.2%Vf of global search Lausanne (PGSL), which is founded on
SFRC are almost equal to those of 0.4%Vf of PFRC. the assumption that by focusing the search around a set
of good solutions, an optimal solution can be identified
(Raphael & Smith, 2003). The parameters required for
σ-w curve are found for each experimental result, and the
mean curve is obtained consequently for each concrete.
The shape of the σ-w curve has to be decided prior to
inverse analysis. The bilinear model has been shown to
provide a reasonable representation of fibre reinforced
concrete (Sousa and Gettu, 2006). The bilinear model is
given by:

v=G
f, E precrack state
v w Q w V = g Q w V, ft
..........................(1)
cracked state
Fig. 7: Comparison of fR,0.5-values of SFRC and PFRC

where E – elastic modulus, ε – elastic strain, σw(w) - stress

crack opening relationship with w - crack opening and f t-
uniaxial tensile strength and function g (w) is given by,

gQ w V = bi - ai w = G
b1 - a1 w 0 # w # w1
.....................(2)
b2 - a2 w w1 # w # w2

1-b b
w 1 = a 1 - a22 ; w 2 = a22 ...................................................(3)

where b1 = 1, and w1 and w2 are the limits given by the

intersection of the two line segments, and the intersection
of the second line segment and the abscissa respectively(
Figure 10).
Fig. 8: Comparison of fR,2.5-values of SFRC and PFRC

Fig. 10: Model for precracked and cracked state of concrete

The P-CMOD curves of concretes with different dosages of
steel and polymer fibres are used to perform the inverse
analysis, and the resulting σ-w curves are compared in
Figures 11 and 12. It can be seen that the incorporation of
Fig. 9: Comparison of feq.3-values of SFRC and PFRC fibres does not increase the tensile strength of concrete,

Organised by
India Chapter of American Concrete Institute 261
Session 3 A - Paper 4

confirming the conclusions made earlier. As the dosage The tension softening diagram derived by inverse analysis of
of steel fibre increases, there is an upward shift in the the notched beam test data helps in characterising the fibre
second part of the curve, implying that the ductility and reinforced concrete. The results reflect the improvement in
fracture energy increase. In polymer fibre, the addition of the toughness due to the incorporation of fibres.
fibres improves the ductility though to a lesser extent.
Acknowledgements
The authors thank Bekaert and W.R. Grace for providing
the fibres used in this work. They are grateful to Prof.
Benny Raphael for programming the optimization
algorithm for the inverse analysis.
References
1. ACI 544.1R-96 (1997) ‘State-of-the-Art Report on Fibre Reinforced
Concrete’, American Concrete Institute.
2. Buratti, N., Mazzotti, C., and Savoia, M. (2011), ‘Post-cracking
behaviour of steel and macro-synthetic fibre-reinforced concretes’,
Construction and Building Materials (25), pp 2713-2722.
3. EN 14651 (2005), Test method for metallic fibred concrete -
Measuring the flexural tensile strength (limit of proportionality
(LOP), residual) European specification for sprayed concrete-
Fig. 11: Comparison of σ-w curves of SFRC EFNARC Guidelines.
4. fib bulletin 55: Model Code 2010,First complete draft –Volume 1.
5. Gopalaratnam, V.S., and Gettu, R. (1995), ‘On the characterization
of flexural toughness in fibre reinforced concretes’, Cement and
Concrete Composites,17, pp. 239-254.
6. Granju, J.L., and Balouch, S.U. (2005), ‘Corrosion of steel fibre
reinforced concrete from the cracks’, Cement and Concrete
Research, 35, pp. 572-577.
7. Löfgren, I., Stang, H. and Olesen, J.F. (2005) ‘Fracture properties
of FRC determined through inverse analysis of wedge splitting
and three-point bending tests’, Journal of Advanced Concrete
Technology, 3(3),pp. 423-434.
8. Mindess, S. and Banthia, N. (2004), ‘Toughness characterization
of fibre-reinforced concrete: Which standard to use?’, Journal of
Testing and Evaluation, 32(2), pp.1-5.

Fig. 12: Comparison of σ-w curves of PFRC 9. Olesen J. F. (2001), ‘Fictitious crack propagation in fibre-reinforced
concrete beams’, Journal of Engineering Mechanics, 127(3), pp.
273-80.
Conclusions 10. Østergaard, L. (2003), “Early-age fracture mechanics and cracking
The post-cracking behaviour of steel and polymer fibre of concrete, comparative study of fracture mechanical test methods
reinforced concrete was assessed using three-point for concrete.” PhD Thesis, Technical University of Denmark.
bending tests on notched specimens. From the results 11. Raphael, B., and Smith, I.F.C., (2003), ‘A direct stochastic algorithm
for global search’, Journal of Applied Mathematics and Computation,
obtained, the following conclusions can be drawn:
146 (2-3), pp. 729-758.
For both the steel and polymer fibre reinforced concretes, 12. RILEM TC 162-TDF (2002), ‘Test and design methods for steel fibre
the equivalent flexural tensile strengths feq,2 and feq,3 reinforced concrete. Bending test. Final recommendation’, Materials
increase as the dosage of fibres increases, and the feq,3 and Structures, 35(253), pp. 579-581.

values are generally more than the feq,2 values, at higher 13. Rossi, P., Coussy. O, Boulay, C., Acker, P. and Malier, Y. (1986),
‘Comparison between plain concrete toughness and steel
dosages. fibre reinforced concrete toughness’, Cement and Concrete
For a particular volume fraction, SFRC had better Research,16(3), pp. 303-313.
toughness than PFRC. Considering the toughness 14. Sousa, J.L.A.O. and Gettu, R. (2006), ‘Determining the tensile stress-
crack opening curve of concrete by inverse analysis’, Journal of
parameters related to the serviceability and ultimate limit
Engineering Mechanics, ASCE, 132(2), pp 141-148.
states, namely fR,0.5 and fR,2.5. However, the performance
15. Stang, H., and Olesen, J. F. (1998). ‘On the interpretation of
of concrete with 15 kg/m3 steel fibres was found to be bending tests on FRC-materials.’ Fracture Mechanics of Concrete
comparable to that with 3.75 kg/m3 of polymer fibres. Structures, Proceedings of FRAMCOS-3, AEDIFICATIO Publishers,
D-79104, Freiburg, Germany, pp. 511 – 520.
The scatter in the values obtained for polymer fibre
16. Stang, H. and Olesen, J.F.(2000), ‘A fracture mechanics based
reinforced concrete is less compared to that of steel fibre
design approach to FRC’, Proc. Fifth RILEM Symposium on Fibre-
reinforced concrete, as the number of fibres present in the Reinforced Concretes, Lyon (Eds. P.Rossi and G. Chanvillard, RILEM,
former is much higher than in the latter. Cachan, France), pp. 315-324.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

262 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Study of the Post-Cracking Behaviour of Steel and Polymer Fibre Reinforced Concretes

Sujatha Jose
Sujatha Jose holds a Masters degree in Construction Engineering and Management from College of
Engineering, Guindy, Anna University and is pursuing her PhD at Building Technology and Construction
Management Division, IIT Madras.

Stefie J.Stephen
Stefie J.Stephen holds a Masters degree in Structural Engineering from Government College of Technology,
Coimbatore and is pursuing her PhD at Building Technology and Construction Management Division, IIT
Madras.

Ravindra Gettu
Ravindra Gettu is a Professor of Civil Engineering at the Indian Institute of Technology Madras (IITM),
Chennai. His current research interests are in fiber reinforced concrete, the effective use of chemical
admixtures, self-compacting concrete, housing and the mechanical characterization of construction
materials. He is currently the Vice-President of RILEM, the International Union of Laboratories and Experts
in Construction Materials, Systems and Structures

Organised by
India Chapter of American Concrete Institute 263
Session 3 A - Paper 5

Validation Needs for Concrete Modeling

J.E. Bolander
Department of Civil and Environmental Engineering, University of California, Davis, USA

Abstract capabilities of numerical models are discussed,

acknowledging the need to account for various
The computational modeling of concrete materials
uncertainties in the input parameters for practical
and structures serves several interrelated purposes,
applications. The comparisons are limited to early-
including elucidation of structure-property relationships,
age temperature development, but are being extended
design support for new concrete technologies, and
to consider additional factors that influence crack
predictions of concrete performance in field applications.
formation.
The success of these efforts depends on our abilities to
assess the accuracy of the models in the presence of
various sources and forms of uncertainty. This paper Early-Age Cracking In Structural Concrete
discusses some of the gaps between validated capabilities
of current models and practical needs. Attention is given Relevant Phenomena
to the modeling of early-age cracking in concrete bridge The use of higher cementitious materials content and
decks and slabs. Simulations of temperature evolution lower water to cementitious materials ratio (w/cm) in
in freshly cast bridge decks demonstrate some of the structural concrete has been motivated by the desire
principle ideas and serve as a basis for further study. for higher strength and better durability. As witnessed
Keywords: Concrete modeling, cracking, heat of in many bridge decks, however, such concretes are
hydration, uncertainty quantification. susceptible to early-age cracking (Krauss and Rogalla,
1996). Such cracking tends to increase permeability
(Wang et al., 1997) and negate the anticipated
Introduction improvements in durability associated with densification
Early-age cracking of concrete remains an important of the cement- based matrix. Susceptibility to early-age
concern for high-performance concrete applications, cracking depends on several phenomena (CEB, 1992):
such as bridge decks. Whereas early-age cracking has ll Plastic shrinkage - Cracking is associated with an
been the focus of much study, the subject area remains inability to supply bleed water to a drying surface.
complicated by the large number of influential parameters Such cracking is often minor and not problematic by
and their interactions. Many of the parameters are to itself, but large plastic shrinkage cracks can affect
some degree uncertain, which further complicates the durability. Plastic shrinkage cracks might also be
assessment of cracking and its consequences. Whereas precursors to other early-age cracking or cracking
physical testing and field observations of concrete under service loads.
provide invaluable understandings of parameter
significance, such methods of investigation are limited ll Plastic settlement - The solid fraction of the concrete
by both technical and practical constraints. mixture tends to settle, since is heavier than the water
that bleeds upward. Reinforcing bars locally restrain
Numerical modeling is a means for conducting virtual such settlement, which can lead to cracking over top
experiments within such large parameter spaces. Model bars, especially in deep sections.
validation is prerequisite to its use for either fundamental
study or practical applications. Validated models can ll Drying shrinkage - Drying shrinkage is caused by
provide insight into the relative importance the influential departure of water from the pore structure, either
parameters and the robustness of concrete mixture by evaporation or diffusion (Altoubat and Lange,
designs for the anticipated structural and environmental 2001). The remaining water in partially filled pores
boundary conditions. develops menisci that exert an underpressure on the
pore structure (i.e., negative pressure within the pore
This paper highlights several outstanding issues of fluid puts the solid skeleton into compression), which
concrete modeling through the simulation of early- causes shrinkage of the concrete member.
age behavior of concrete bridge decks. The predictive

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

264 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Validation Needs for Concrete Modeling

ll Autogenous shrinkage - Originates from the in which k is the number of parameters and ni represents
stresses developed within the menisci appearing in the number of values assignable to parameter i. If each of
the capillaries of hydrated cement paste due to the the 29 parameters appearing in Fig. 1 could take on one
development of chemical shrinkage (Durán-Herrera of three values, for example, there would be N = 329 = 6.9
et al., 2008). The absolute amount of autogenous x 1013 possible test cases. This large number of test cases
shrinkage increases and the shrinkage tends to start presents several problems.
at earlier ages when w/c decreases (Tazawa and
ll Existing data sets, which are needed for model
Miyazawa, 1995). Autogenous shrinkage can start
validation, cover only a small portion of the parameter
immediately after setting, which is problematic since
space.
the matrix has little strength at that time.
ll Data needed for model validation with respect to
ll Temperature variation - Heat of cement hydration
actual full-scale systems (e.g. spatial descriptions of
can produce high temperatures within the young
the relevant field quantities in concrete bridge decks
concrete. Volumetric changes associated with heating
constructed and cured under actual conditions) are
and subsequent cooling cause thermal stresses within
scarce.
the hardening concrete, which can contribute to early-
age cracking (Springenschmid, 1995). ll The production of practical, validated models will
involve compromises between model simplicity and
The relative importance of each phenomenon depends
the accuracy with which physical processes (and their
upon the prevailing mechanisms of restraint and a variety
mutual couplings) are represented. Mutual couplings
of other factors, many of which can be anticipated or
between variables are difficult to validate and interpret
controlled through design (Mihashi and Wittmann, 2002;
when investigating cause-and-effect relationships at
ACI-231, 2010). Early age conditions influence cracking
the system level.
under service loads, which can further modify the
transport properties of the concrete (Hoseini et al., 2009). ll Presuming the existance of an adequately validated
model, there is a crucial need for methods to
Design Parameters strategically explore the parameter space for viable
The factors that affect cracking potential can be categorized designs. An exhaustive search, or random sampling by
in different ways, such as the grouping presented in Fig. 1. classical Monte Carlo methods, is prohibitive in terms
These factors can be viewed as design parameters, since of computational expense.
they can be controlled (at least to some degree) prior to ll There are uncertainties with respect to both model
and during the time of construction/curing. For the sake of accuracy and the assignable values to each parameter i.
simplicity, the many environmental boundary conditions
The parameter count could be reduced by excluding
are represented in terms of heat and moisture exchange.
parameters likely to have only secondary influence
A full factorial examination of the discretized parameter on cracking potential. For a given region, practical
space involves N test cases considerations (e.g., restriction to using locally available
N = n 1 n 2 n 3 ....n k ....................................................(1) aggregates) may also allow for reductions in the parameter
count. Existing designs can be used as bases from which
to conduct sensitivity analyses, although such explorations
might be confined to the neighborhoods of local optima.

Assessment of Cracking Potential

Although the prediction of cracking is complicated by
many factors, it is reasonable to view cracking potential
as a competition between measures of restraint stress
σt and tensile strength ft that develop over time, as
depicted in Fig. 2a. The comparison of restraint stress
and strength holds at each point throughout the domain.
In deterministic terms, cracking is assumed to occur
when σt ≥ ft (CEB, 1992) or, equivalently, when cracking
index Icr is less than unity:
f t (t ) ...............................................................(2)
I cr (t) =
v t ( t)

Considering the stochastic nature of the factors affecting

Fig. 1: Parameters affecting early-age cracking potential of both restraint stress and strength, however, cracking
concrete bridge decks potential ought to be estimated in probabilistic terms. At

Organised by
India Chapter of American Concrete Institute 265
Session 3 A - Paper 5

any time t, there is a scatter in both restraint stress and

strength produced within a set of nominally identical
systems, differing only by random variations of the
parameters and processes defining the systems (Fig 2b).
The estimation of cracking potential is also affected by
spatial variation of these values.
A practical means for assessing the risk of cracking
involves comparison of the cracking index of Eq. (2) with
an empirically derived safety index (JSCE, 2010). The
comparison provides a probability of cracking (Fig. 3).
Level of cracking risk can then be estimated according to:
$ 1.75 : cracking prevented
c cr$ 1.45 : number of cracks controlled
$ 1.00 : cracks allowed, but crack width controlled ...............................(3) Fig. 3: Probability of thermal cracking (adapted from JSCE,
2010)
These relations have been developed with respect to risk
of thermal cracking (JSCE, 2010), but the concept could Property Development
be applied more generally to early-age cracking. The
requirement for Icr ≥ 1.75 to prevent cracking implicitly Simulation of property development is a key component
accounts for uncertainties in the models and processes being of early-age cracking models, as property development
represented. The preceding discussions are incomplete, but affects restraint stress and strength evolution. A common
they serve to illustrate the complexity of assessing early-age approach is to relate property development to equivalent
cracking potential. The comparison of restraint stress and age of the concrete or degree of cementitious materials
strength development relies on the robustness and accuracy reaction (de Schutter and Taerwe, 1996). For the latter
of the computational model. As a goal, the model should case, degree of reaction can be expressed by
resolve the spatial distributions of temperature, relative Q (t ) ...............................................................(4)
a ( t) = Q
humidity, and displacement within the aging concrete, as 3

well as account for the development of material properties

where Q(t) is heat of hydration produced up to time t and
at early ages (di Luzio and Cusatis, 2013).
Q∞ is the total heat potential of the cementitious materials.
This, in effect, accounts for the influences of both time
and temperature on property development. A generic
expression has been used to relate degree of reaction and
property development
g (a) = g 1 S 1 - a 0 X for a $ a 0 .................................(5)
a-a a
0

where g1 is a measure of property development for a

completely reacted cementitious material; α0 is the degree
of hydration associated with initial setting; and coefficient
a depends on the property being modeled (de Schutter and
Taerwe, 1996; Krauß and Rostásy, 2002). By curve fitting
to experimental data, a = 0.5 for elastic modulus; a = 1 for
tensile strength; and a = 1.5 for compressive strength. The
value of α0 can be determined by Vicat needle penetration
or some other assessment of setting. Alternatively,
concrete property development can be modeled
according to the physical processes of microstructure
formation. Solidification theory is one prominent example
(Bazant and Prasanan, 1989; di Luzio and Cusatis, 2013).
Measurements of property development within the first
24h are of particular interest (Roziere et al., 2015).

Modeling Example: Temperature Evolution in

Structural Concrete
Fig. 2: Comparison of restraint stress and strength development:
(a) deterministic view; and (b) results for a set of nominally This example covers one of the primary tasks associated
identical systems with assessing the cracking potential of early-age concrete:

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

266 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Validation Needs for Concrete Modeling

simulation of cementitious materials hydration and use the multi-species hydration model of Riding et al.
temperature development within the structural concrete (2012). The calculations of heat production rate are made
members. This is of importance since degree of hydration on a nodal basis; the volume of material associated with
governs properly evolution, which plays a central role in a given node is defined by its corresponding Voronoi cell.
the competition between restraint stress and strength The input parameters, based on laboratory assessment
development, as illustrated in Fig. 2. A lattice model is of the cement used for the Markham Ravine Bridge, are
used for this purpose, although alternative approaches indicated in Table 1.
(e.g., those based on finite element technology) would
serve the analysis needs equally well (Faria et al., 2006; Table 1
Zanotti et al., 2010). Input values for hydration model
Simulated temperatures are compared with those
measured within the Markham Ravine Bridge deck in portland cement composition* mixture composition
northern California (WJE, 2011). Measurements were C 3S 62% OPC 75%
taken from the time of casting and included ambient
C2S 13% fly ash 25%
temperature and wind speed. Using a partially validated
model, Bolander et al. (2014) considered the influences of C 3A 5% slag 0%
cementitious materials content, amount of solar radiation
C 4 AF 12% silica fume 0%
on the day of casting, and time of casting (e.g., morning
versus afternoon) on early-age temperature development. MgO 1.6% cm content 401 kg/m3
Of interest herein is the sensitivity of the model results to
SO3 2.7% retarder 0%
parameters that are poorly defined, namely the thermal
properties of curing blankets placed after a short period (Na2O)eq 0.31% accelerator 0%
of curing.
Blaine 401 m2/kg w/cm 0.42

* Based on laboratory testing.

The boundary conditions for Eq. (6) involve either prescribed

temperatures or heat flux across the boundary. For bridge
deck systems, heat exchange with the environment is of
primary importance. Herein, the following types of heat
exchange are modeled:

Fig. 4: (a,b) Delaunay/ Voronoi dual tessellations of an ll Convection- Convective heat exchange across exposed
unstructured set of points; and c) lattice element ij surfaces depends on the difference between the solid
surface temperature Ts and that of the surrounding
Simulation Approach ambient medium Ta
The lattice topology is defined by the Delaunay tessellation q conv = h c (Ts - Ta) .................................................(7)
of a set of randomly placed nodes (Fig. 4); the dual Voronoi
tessellation defines element properties (Bolander and where hc is the coefficient of convective heat transfer,
Berton, 2004). For transport analyses, element ij can which depends on wind speed (Bentz, 2000).
be regarded as a conduit for transport between nodes ll Solar radiation- The amount of solar radiation
i and j. Routines used for modeling moisture diffusion reaching the concrete surface depends on several
and drying from an exposed surface (Bazant and Najjar, factors including the structure’s location, surface
1971; Bolander and Berton, 2004) are adapted herein for orientation, altitude, atmospheric conditions, time of
thermal analyses. The governing differential equation for day, and day of the year. Incoming heat due to solar
heat conduction is radiation is
Q 2T
d $ (kdT) + pc c = pc 2t ...........................................(6) q sun = c abs q inc .........................................................(8)

in which T is temperature (K), Q. is the rate of heat in which γabs is the solar absorptivity of the concrete,
production (J/g/s), k is thermal conductivity (W/m/K), ρ is which is influenced by the color and texture of the concrete
density (kg/m3), and c is specific heat capacity (J/g/K). surface, and qinc is the incident solar radiation acting on
a horizontal surface (W/m2). The latter of the two values
General requirements of the hydration model include the can be obtained from recorded weather data for the time
ability to simulate the hydration of multi-species blends period and region of interest.
and, to the extent possible, account for the chemical
composition of the cementitious materials. Herein, we ll Thermal radiation - Heat exchange due to thermal
radiation is dependent on the temperature of the concrete

Organised by
India Chapter of American Concrete Institute 267
Session 3 A - Paper 5

surface, emissivity of the concrete, and the ambient

(sky) temperature. Heat loss to the surroundings due to
grey-body irradiation is calculated using
4
q sky = ve (T sK - T 4sky) .....................................................(9)

where σ is the Stefan-Boltzmann constant (5.669 x 10-2 W/

(m2 K4)), ε is the emissivity of the concrete (= 0.9, in this
study), and TsK is the temperature of the concrete surface
(K). Tsky is the sky temperature, which depends on the Fig. 5: Lattice representation of symmetric portion of Markham
sky emissivity, the dew point temperature and the cloud Ravine bridge deck
conditions (Bentz, 2000).
Simulation Results
Heat exchange also occurs due to evaporation and
The simulated temperature histories are compared
condensation, but those mechanisms are considered to be
with the field measurements in Figure 6, for the case of
of secondary importance for ordinary concrete structures.
thermocouples (TC1, TC2, TC3) located above the soffit
Some other relevant aspects of the thermal analyses are:
region. The recorded ambient temperature history is also
ll The heat capacity of the cement paste is estimated plotted in the figure.
using an approach given by Bentz (2007), in which
One point of interest is the use of thermal/curing blankets
heat capacity is a function of degree of reaction of
on the cast surface. A combination of burlap and plastic
the cement. The heat capacity of the concrete is then
membrane (Transgard 4000) was used to cover the
determined from the heat capacities of the cement
bridge deck at about t = 6 h after concrete placement.
paste and aggregates, according to the mass fractions
The product sheet for Transgard 4000 indicates a light
of each using an ordinary rule of mixtures.
reflectance of 0.85. This is roughly twice the reflectance
ll Thermal conductivity of the concrete is estimated by value of ordinary portland cement concrete, which ranges
taking the average of the Hashin-Shtrikman bounds from about 0.34 to 0.48 (Marceau and Vangeem, 2008).
for a two-phase composite formed of paste and For this reason, qsun was reduced by a factor of η = 0.5
aggregates (Bentz, 2007). for t > 6 h. For lack of information, the same reduction
factor was assumed for both the convection and radiation
Model Definition boundary conditions.
As part of a recent Caltrans sponsored project (WJE, hlc = hh c VWW
2011), the early-age behavior of concrete bridge decks was W
studied through their instrumentation and data acquisition
qlsun = hq sun WW for t $ 6h .......................................(10)
W
for a period of time after concrete casting. Sensors, qlsky = hq sky W
positioned within the freshly cast concrete, monitored
X
Comparing the measured and simulated temperatures
temperature, internal relative humidity, and strain. Along
in Fig. 6(a), however, the rate of cooling in response to
with these sensor readings, field observations of cracking
ambient temperature drops is too large. Changing to η =
behavior were made.
0.3 provides better agreement between the experimental
Thermal analyses of the freshly cast deck is a reasonable and simulated results, as evident in Fig. 6(b). The point here
starting point for the ultimate objective, which is assessing is not that better results can be obtained through this type
early-age potential for cracking. It remains to be seen of parameter adjustment. Rather, the example highlights
whether stresses induced by heat of hydration, and other the dependence of the results on poorly understood and
sources of heating/cooling, contribute significantly to measured quantities, i.e., the thermal blanket properties.
cracking potential of bridge decks. However, the degree of As seen in other studies (Macobatti et al., 2014), improved
cementitious materials reaction is closely associated with thermal insulation reduces the temperature gradient and
the development of relevant properties, such as stiffness, variations in property development through the member
strength, creep behavior, and transport. In this sense, the thickness.
modelling of cementitious materials reactions is at the
Figure 7 depicts the temperature distribution within the
heart of the analysis program.
slab/girder system at 10h after casting, which is close
Discretization of the bridge deck/girder system is shown to the time of peak temperatures in the deck. The lower
in Fig. 5. Concrete forming the existing girders is assumed temperatures over the supporting girders are due to
to be mature. Simulation of cementitious materials conduction of heat toward the cooler substrate concrete.
hydration is limited to the freshly cast deck. Whereas Conversely, the insulative properties of the plywood
the computational framework is three-dimensional, formwork give rise to higher temperatures between the
the simulations presented herein are for planar supporting girders.
representations of the structure.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

268 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Validation Needs for Concrete Modeling

Fig. 9: Simulated property variations: a) tensile strength; and

b) elastic modulus

Figure 9 presents histograms of tensile strength and

elastic modulus based on Eq. (5), where the degree of
hydration at initial setting α0 = 0.17 (Krauß and Rostásy,
2002). The normalizing factors f t1 and Ec1 are the tensile
strength and elastic modulus, respectively, associated
Fig. 6: Temperature variation at thermocouple locations within with complete hydration of the cementitious materials (i.e.
soffit region of Markham Ravine Bridge. Effect of thermal/curing α = 1). As anticipated from the dependence on exponent a,
blanket representation: a) η= 0.5; and b) η= 0.3 the rate of increase in elastic modulus is greater than that
in tensile strength. Stiffer, stronger regions result from
higher temperatures over the duration of curing. Figure
9 also indicates the simulated dependence of strength
and elastic modulus on the insulating properties of the
thermal blanket placed at t = 6h. The lower value of η,
which corresponds to better insulating properties, shifts
the f t and Ec histograms toward higher values.

Discussion
Fig. 7: Simulated temperature distribution at t = 10h (based The preceding example illustrates one barrier to the
on η= 0.3) validation and practical application of models for assessing
early-age cracking potential: the existence of various forms
of uncertainty. In the example, temperature evolution in
the concrete was significantly affected by the insulative
properties of thermal/curing blankets placed on the deck
surface. Such properties are typically poorly understood and
quantified. It can be argued that such forms of uncertainty
pervade these types of analysis. The quantification of
uncertainty in support of structural concrete design is a
vital area of research (Wendner et al., 2014).
Fig. 8: Simulated degree of cementitious materials reaction It remains to be seen whether thermal stress is a major
factor with respect to early-age cracking of bridge
Figure 8 shows contours of degree of reaction α at decks. Gradients in temperature appear at early ages,
selected times after casting. Since hydration of the mature which likely contribute to restraint stress. However, the
substrate concrete was not modeled, only the deck region simulations need to be extended to include the coupled
is shown. As expected, α develops nonuniformly within effects of the displacement, hygral, and thermal fields.
the concrete deck. Comparing the results of Figs. 7 and The role of creep in lessening, or possibly increasing
8, one sees the expected positive correlation between (Faria et al., 2006), restraint stress must also be included
temperature and degree of reaction. in such study.

Organised by
India Chapter of American Concrete Institute 269
Session 3 A - Paper 5

Conclusions diffusion problem. Cem. Concr. Res. 1(5):461-473.

4. Bentz, D.P., 2000. A computer model to predict the surface
Numerical modeling is means for parametric study of the temperature and time-of- wetness of concrete pavements and
various parameters and processes that affect concrete bridge decks. Technical Report NISTIR 6551, National Institute of
performance, as part of efforts to improve the design of Standards and Technology.
structural concrete, particularly for high-performance 5. Bentz, D.P., 2007. Transient plane source measurements of the
applications such as bridge decks. The utility of such thermal properties of hydrating cement pastes. Mater. Struct.
parametric study depends on the validity of the model. 40:1073-1080.
Rigorously validated models can be used as surrogates 6. Bažant, Z.P. and Prasannan, S., 1989. Solidification theory for
for physical experimentation. concrete creep. I: Formulation. Journal of Engineering Mechanics,
ASCE, 115(8):1691-1703.
Presently, however, model validation is arguably the most 7. Bolander, J.E. and Berton, S., 2004. Simulation of shrinkage induced
difficult and pressing need. Some basic concepts have been cracking in cement composite overlays. Cement and Concrete
discussed in this paper, reinforced through simulation Composites, 26(7):861-871.
examples of early-age temperature development in 8. Bolander, J.E., Kim, K. and Sasaki, K., 2014. Thermal effects
freshly cast bridge decks. Several remarks can be made: on early-age cracking potential of concrete bridge decks. In N.
Bicanic, H. Mang, G. Meschke, and R. de Borst (Eds.), Computational
1. To support model validation, there is a continuing need Modeling of Concrete Structures, EURO-C 2014, Volume 2, pp.
723–729.
for data sets that reflect the complexity and variability
of actual systems. The instrumented bridge decks, 9. Comite Euro-International du Beton (CEB), 1992. Durable concrete
structures – Design guide, Thomas Telford Ltd., London.
examined herein, serve that purpose albeit for a
limited range of possibilities. 10. de Schutter, G. and Taerwe, L., 1996. Degree of hydration-based
description of mechanical properties of early age concrete.
2. The degree of cementitious materials reaction serves Materials and Structures, 29:335-344.
as a basis for modeling property development. Spatial 11. di Luzio, G. and Cusatis, G., 2013. Solidification – microprestress
differences in time-temperature histories lead to – microplane (SMM) theory for concrete at early age: Theory,
validation and application. International Journal of Solids and
differential property development, which may be
Structures, 50(6):957-975.
significant in modelling cracking potential. It remains
12. Durán-Herrera, A., Petrov, N., Bonneau, O., Khayat, K. and Aitcin, P.-
to be seen whether thermal stress is a significant C., 2008. Autogenous control of autogenous shrinkage. In: Internal
factor with respect to early-age cracking of bridge Curing of High- Performance Concretes: Laboratory and Field
decks. Experiences, eds. D. Bentz and B. Mohr, ACI SP- 256, 2008, pp. 1-12.
13. Faria, R., Anzenha, M. and Figueiras, J.A., 2006. Modelling of
3. Boundary conditions, which are often poorly understood
concrete at early ages: application to an externally restrained slab.
or inadequately represented by the modelling efforts, Cement and Concrete Composites, 28:572-585.
can have a primary effect on early-age temperature 14. Hoseini, M., Bindiganavile, V. and Banthia, N., 2009. The effect of
development within structural concrete. Dependence mechanical stress on permeability of concrete: A review. Cement
of the model results on such poorly understood and Concrete Composites, 31:213-220.
quantities complicates the prediction of early- age 15. JSCE, 2010. Standard Specifications for Concrete Structures – 2007
behavior in practical situations. Validation efforts (English Version), Japan Society of Civil Engineers.
should include the quantification of such uncertainties 16. Krauss, P.D. and Rogalla, E.A., 1996. Transverse cracking in newly
and their potential influence on system behavior. constructed bridge decks. NCHRP Report 380, National Academy
Press.
4. The parameter set affecting early-age behavior is 17. Krauß, M. and Rostásy, F.S., 2002. Determination of initial degree
large, such that a full factorial examination of the of hydration by means of ultrasonic measurements. In Control of
parameter space is immensely cost prohibitive. Cracking in Early Age Concrete, Proceedings of the International
Methods are needed for effectively examining the Workshop on Control of Cracking in Early Age Concrete, H. Mihashi
and F.H. Wittmann (eds.), Sendai, Japan, A.A. Balkema Publishers,
parameter space. 19-28.
18. Macobatti, F., Zanotti, C., Meda, A. and Plizzari, G., 2015. Jacketing of
Acknowledgement existing piers: Evaluation of the risk of cracking due to hydration heat
when different types of application techniques are used, In: High
The author gratefully acknowledges support provided by Performance Fiber Reinforces Cement Composites (HPFRCC7),
Caltrans through technical agreement #65A0532 with the RILEM, PRO 94:465-472.
University of California, Davis. 19. Marceau, M.L. and Vangeem, M.G., 2008 (August). Solar reflectance
values for concrete. Concrete International: 52-58.
References 20. Mihashi, H. and Wittmann, F.H. (editors), 2002. Control of Cracking
1. ACI-231, 2010. Report on Early-Age Cracking: Causes, Measurement, in Early Age Concrete, Proceedings of the International Workshop
and Mitigation. ACI Committee 231, American Concrete Institute. on Control of Cracking in Early
2. Altoubat, S.A. and Lange, D.A., 2001. Creep, shrinkage, and 21. Age Concrete, Sendai, Japan, A.A. Balkema Publishers, 399 pp.
cracking of restrained concrete at early age, ACI Materials Journal, 22. Riding, K.A., Poole, J.L., Folliard, K.J., Juenger, M.C.G. and
98(4):323-331. Schindler, A.K., 2012. Modeling hydration of cementitious systems.
3. Bažant, Z.P. and Najjar, L.J., 1971. Drying of concrete as a non-linear ACI Materials Journal, 109(2):225-234.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

270 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Validation Needs for Concrete Modeling

23. Roziere, E., Cortas, R. and Loukili, A., 2015. Tensile behaviour of 27. Wendner, R., Hubler, M.H. and Bažant, Z.P., 2015. Optimization
early age concrete: New methods of investigation. Cement and method, choice of form and uncertainty quantification of Model B4
Concrete Composites, 55:153-161. using laboratory and multi-decade bridge databases, Materials and
Structures, 48:771-796.
24. Springenschmid, R. (editor), 1995. Thermal cracking in concrete at
early ages, Proceedings of the International RILEM Symposium, E 28. WJE, 2011. On-call structural concrete bridge deck cracking
& FN Spon, 470 pp. investigation services. Technical Report WJE No. 2009.2643,
Prepared for Caltrans, Wiss, Janney, Elstner Associates, Inc.
25. Tazawa, E. and Miyazawa, S., 1995. Influence of cement and
admixture on autogenous shrinkage of cement paste, Cement and 29. Zanotti, C., Meda, A., Plizzari, G. and Cangiano, S., 2010. Evaluation
Concrete Research, 25(2):281- 287. of the risk of cracking in thin concrete walls due to hydration heat.
In: Proceedings of the 6th International Conference on Concrete
26. Wang, K., Jansen, D.C., Shah, S.P. and Karr, A.F., 1997. Permeability
Under Severe Conditions (CONSEC’10), Merida, Yucatan, Mexico.
study of cracked concrete. Cement and Concrete Research,
27(3):381-393.

Prof. John E. Bolander

John E. Bolander is a professor within the Department of Civil and Environmental Engineering at the
University of California, Davis. He received his Ph.D. degree in civil engineering from the University of
Michigan, Ann Arbor, in 1989. He belonged to the faculty of engineering at Kyushu University, Japan, for five
years prior to arriving at UC Davis in 1994. He received the Outstanding Faculty Teaching Award from the
College of Engineering in 2006. Bolander has served as the chief editor of the international journal Cement
and Concrete Composites for the period of 2006-2012. He received fellowship awards for computational
mechanics research from the Japan Society for the Promotion of Science in 1997 and 2008. Bolander’s
research and teaching interests involve the effective use of high-performance materials within the civil
infrastructure, with emphasis on the life-cycle performance of concrete materials and structures. Over
the past two decades, he has been a pioneering developer of discrete methods for the modeling of concrete
materials and structures.

Organised by
India Chapter of American Concrete Institute 271
Session 3 A - Paper 6

Alkali Activated Slag: Reaction Kinetics and Hydration Products

Akash Dakhane Zihui Peng, Robert Marzke, Narayanan Neithalath,

Graduate student, School of Graduate student, Department Associate Professor, Associate Professor, School of
Sustainable Engineering and of Physics, Department of Physics, Sustainable Engineering and
the Built Environment, Arizona State University, Arizona State University, the Built Environment,
Arizona State University, Tempe AZ 85287 Tempe AZ 85287 Arizona State University,
Tempe AZ 85287 Tempe AZ 85287

Abstract the type of curing[10,11]. The use of slag as a precursor for

This paper investigates the influence of the alkali cation alkali activated concretes has been reported in several
(Na or K) on the reaction kinetics, product formation, gel studies [12–14]. Slag activation using alkali silicates typically
structure, and mechanical properties of alkali activated produces C-S-H gel as the reaction product, similar to
slag binders. For the same activator Ms, i.e., molar SiO2- that in OPC systems, but with a lower Ca/Si molar ratio
M2O ratio (M = Na or K), a shorter induction period, a (0.6-1.0, as opposed to 1.5-1.8 in OPC systems [15,16]). Over
larger acceleration peak, and consequently, a higher time, Al is also incorporated into the gel structure to form
amount of total heat release under isothermal conditions a C-(A)-S-H gel.
is observed for the K-silicate activated slag pastes. The Sodium or potassium silicates and/or hydroxides are
early age compressive strengths in these systems roughly commonly used as the activating agents for alkali activated
relate to the heat release characteristics. The later-age binder systems [13,17–19]. In alkali activated slag systems, the
compressive strengths are observed to be higher for the alkali ions are likely incorporated into the C-(A)-S-H gel
Na-silicate activated systems, which is corroborated by: (i) structure. It is thus reasonable to expect differences in
higher amounts of C-(A)-S-H gel in this system indicated the kinetics of chemical reaction, the composition of the
by a thermal analysis-based approximate quantification reaction products formed, and the final binder properties
method, and (ii) higher combined intensities of Q1 and Q2 when the alkali cations are different. This could be a result
structures that point to increased degrees of reaction, and of several factors including the cationic size differences
lower amounts of unreacted slag obtained from29 Si MAS and/or the charge densities. In this paper, an in-depth
NMR spectroscopy. comparison of the effects of Na and K silicate solutions of
Keywords: Slag, Alkali activation, Cationic type, varying SiO2-to-M2O ratio (where M is the alkali cation; Na
Isothermal calorimetry, Thermal Analysis, NMR or K) on the early age reaction kinetics and compressive
spectroscopy strength development of activated slag systems is
reported. Thermal analysis is used to discern the reaction
products formed, and facilitate an approximation of the
Introduction amounts of reaction products whereas 29Si MAS NMR
Several strategies are being recommended and adopted spectroscopy is used to further probe the gel structure.
towards reducing the production of ordinary Portland It is anticipated that such a comprehensive understanding
cement (OPC) which is proven to be a large contributor of the cationic influence on alkali activation of slag will
of greenhouse gas emission[1–3]. Among these methods, help choose the ideal activator type and characteristics
one of the most studied one is the development of binder (including the SiO2-to-M2O and M2O-to total powder ratios)
systems using alkaline activation of aluminosilicate for desired performance of the final product.
materials such as fly ash (coal combustion by-product), or
ground granulated blast furnace slag (GGBFS, hereinafter
termed slag)[4–6]. This methodology is extremely beneficial Experimental Investigation
in concrete because the process of forming a value- added
material by utilizing large volumes of an industrial waste/ Materials and Mixture Proportioning
byproduct alleviates concerns related to its disposal and The source material used in this study is ground granulated
results in a lower ecological foot-print for the concrete blast furnace slag (GGBFS) Type 100 conforming to ASTM
thus produced. Previous studies have shown that alkali C 989 [20] , with the chemical composition shown in Table
activated concretes can be proportioned so as to obtain 1. The median particle size (d50) of the slag is 8.9 μm as
mechanical and durability properties that are similar to or determined using dynamic light scattering.
superior than those of OPC-based binders[7–9]. The material Sodium silicate (waterglass) and potassium silicate
properties are inherently linked to the physico-chemical solutions were used to activate slag in this study. The
characteristics of the source materials, activators, and
activator solutions were proportioned based on the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

272 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Alkali Activated Slag: Reaction Kinetics and Hydration Products

Table 1
Chemical composition and physical properties of slag

SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K 2O LOI LOI SSA

36.0% 36.0% 0.67% 39.8% 7.93% 2.10% 0.27% 0.80% 3.01% 2.90 487 m2/kg

following two parameters: M2O- to-slag (binder) ratio (n), mm cube specimens were determined at ages of 1, 3, 14,
and the silica modulus (molar SiO2-to-M2O ratio) (Ms[21–23]. 28 and 56 days of moist curing.
The n value dictates the total amount of Na2O or K2O in the
Thermo-gravimetric analysis (TGA) was performed using
mixture. Based on previous studies, the optimum n value
a Perkin Elmer simultaneous thermal analyzer (STA
was found to 0.05 and the increasing amounts of alkalis
6000). TGA was performed on samples at ages of 1, 7, and
will only contribute to carbonation and leaching [24–26].
28 days. The thermal analysis procedure used was as
Two different sodium silicate solutions were selected
follows: the sample was heated to 50°C and then held for
for use in this study, one having a solids content of 36%
1 minute, and then heated to 995°C at a rate of 15oC/min.
and molar Ms of 3.3 and the other with a solids content of
An approximate quantification of the amount of C-(A)-S-H
44% and molar Ms of 2.06. The potassium silicate solution
gel is attempted using the mass loss between 150oC and
has a solids content of 44% and a molar Ms of 3.29. For
300°C[28–30]. The stoichiometry of the gel to be used in the
both the Na and K silicates, NaOH or KOH was added if
approximation was obtained from published literature
necessary to bring the activator Ms down to 1.0 to 2.5 since
and the NMR spectroscopy studies carried out as part of
beneficial strength development required Ms values in this
this research.
range[23,25]. The water-to-powder ratio (w/p) of 0.40 is used
in this study[22]. The water consists of the liquid portion of 29
Si Magic Angle Spinning (MAS) NMR spectroscopy
the activator and the additional water required to obtain (Varian Solids NMR spectrometer) was performed on
the desired w/p (mass-based) ratio. Table 2 summarizes the unreacted slag and the Na and K- silicate activated
the amount of activating agents required for 1000 g of slag, paste samples after 28±2 days of reaction in a sealed
for an n value of 0.05 and the different Ms values used in environment. For the 29Si NMR, the resonance frequency
this study. For the compressive strength tests, mortars used was 59.7 MHz with a spinning rate of 10 kHz, and
were prepared with a 44±2% sand volume fraction. The the spectra were obtained after irradiating the samples
mortars were stored in a moist chamber (23±1°C, >98% with a π/2 pulse. The time between accumulations was
RH) until the respective testing durations. The paste 5s, and 29Si chemical shift was referenced externally
specimens for thermal analysis and NMR spectroscopy relative to tetramethylsilane (TMS) at 0 ppm. The 29Si
were stored in sealed conditions until the age of testing. resonances were analyzed using the standard Qn (mAl)
classification, where n is the number of bridging oxygens
Test Methods per SiO4 tetrahedra, and m is the number of neighboring
Isothermal calorimetry experiments were carried out AlO4 tetrahedra.
for 72 hours at a temperature of 25°C in accordance with
ASTM C 1679[27]. The activator solutions were prepared Results, Analysis and Discussions
separately and allowed to cool down to the ambient
temperature. The activator was then added to the binder Reaction Kinetics of Na- and K-Silicate Activated Slag
(slag) and the mixture placed in the calorimeter chamber. Pastes
The time elapsed between mixing and placing it in the Figures 1(a) and (b) show the heat evolution of sodium
calorimeter was not more than 2 min[22]. The compressive and potassium silicate activated slag pastes respectively,
strengths of the mortars were determined in accordance proportioned using an n value of 0.05 and three different Ms
with ASTM C 109. The strengths of the three replicate 50 values (1, 1.5 and 2.5). The as-obtained Na- and K-silicate

Table 2
Amounts of the alkali silicate activators for 1000g of slag

Activator solutions Na-Si (Ms = 3.3, 36%) Na-Si (Ms = 2.0, 44%) K-Si (Ms = 3.3, 36%)

Ms 1 1.5 2.5 1 1.5 2 1 1.5 2.5

Na/K silicate solution (g) 174.7 262.0 436.7 164.6 246.9 340.1 130.9 196.3 327.2

NaOH/KOH (g) 45.1 35.4 16.0 33.3 17.7 0.0 41.5 32.4 14.3

Water (g) 317.9 274.1 186.6 339.9 307.1 269.9 342.3 308.3 240.2

Organised by
India Chapter of American Concrete Institute 273
Session 3 A - Paper 6

Fig. 1: Heat evolution of slag pastes activated using (a) Na-silicate and (b) K-silicate at 25°C after 72 h of reaction

activators with 36% solids content were used here, larger peak at around 12h could be attributed to the reaction
and NaOH or KOH solution was added to reduce the Ms between the dissolved Ca2+ in solution and the silicate
values. In general, two distinct peaks appear as the result anions from the activator[12]. This effect is not noticed
of reaction of slag with the alkaline solution – an initial for the pastes activated using Na-silicates or when the
dissolution peak and a later acceleration peak, similar to activator Ms is decreased. When the Ms is lowered through
the heat release signature of OPC systems. The initial peak the addition of alkali hydroxides, the initial dissolution rate
noticed during the first few hours is caused by particle is enhanced for both Na- and K-silicate activated mixtures,
wetting and the dissolution of the slag particles in the highly with a consequent precipitation of an insoluble, low Ca/Si
alkaline activator[14,22]. The Ca-O bonds and some amount molar ratio product (Al-rich gel). There is a higher amount
of Al-O and Si-O bonds in slag are broken by the highly of OH- ions in solution to break further Si-O and Al-O bonds
alkaline solution. Since external mixing was adopted, the to react with Ca ions to form the C-(A)-S-H gel in these
dissolution peak magnitudes would not be accurate, even cases. For the same Ms value (of 2.5, here), the Na-silicate
though they can be used for comparative purposes[22]. The activated paste does not demonstrate the two initial peak
setting times of all the activated slag pastes studied here behavior. This could be attributed to the K+ ion being able to
are relatively short, indicating the formation of insoluble retain the soluble silicate anions stable for longer durations
reaction products around the slag particles. as compared to the Na+ ion[26]. These anions react with the
dissolved Ca2+ to form the second initial peak. Notice that a
Higher Ms values of the activator induces more
weak acceleration peak appears only at about 48h for this
precipitation of Ca-containing reaction products early on
system that is much less alkaline because of the higher Ms.
but in the absence of large amounts of alkali cations, the
polysilicate ion linkages are soluble in water [31]. This For both the Na- and K-silicate activated pastes, decreasing
explains the general increase in the heat release rate the Ms (increasing alkalinity) increases the magnitude of
associated with dissolution as the activator Ms increases, the acceleration peak. The pastes proportioned with an
and the subsequent delay in the acceleration peak. The Ms of 2.5, for both the silicate solutions, show extremely
dormant period after the dissolution peak is followed by delayed and wide acceleration peaks of smaller magnitude.
the acceleration peak that corresponds to the strength- This is a result of the lower levels of alkalinity in these
imparting reaction product formation in these systems. pastes, which consequently reduces the ability of the ions
In systems with high alkalinity (lower values of Ms), the to penetrate the initially formed reaction product layer as
dormant period is relatively short as observed from these described earlier [32]. The dormant period is found to be
figures, attributable to the fact that the dissolution rates shorter for the K-silicate activated pastes as compared
are enhanced and the ionic diffusion through the reaction to the Na-silicate activated pastes at the same Ms. The
product layer over the slag particles is enhanced. For the K-silicate activated pastes also demonstrate increased
paste activated using K-silicate of Ms = 2.5 alone, there are intensities of the acceleration peak, and a higher rate of
two heat release signatures corresponding to the initial the acceleration phase of the reaction as evidenced by
reaction stage. While the first, smaller peak corresponds the slope of the acceleration curve. The alkalinity of the
to the particle wetting and dissolution (higher Ms, and activator solutions in these systems is significant, as
therefore reduced degree of dissociation), the second high alkalinity favors better dissolution of the silica and

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

274 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Alkali Activated Slag: Reaction Kinetics and Hydration Products

Compressive Strength Development and Reaction

Product Formation

Compressive strength
The compressive strengths of Na- and K-silicate activated
mortars were determined after 1, 3, 14, 28 and 56 days
of moist curing (Figure 3(a)) to understand the cationic
influence on the mechanical properties of the binder. The
rate of strength development provides an indication of the
formation of the C-(A)-S-H gel over time in these systems.
In general, it can be noticed that, at later ages (28 days and

Fig. 2: Comparison of heat release response of slag pastes

activated using Na-silicates of different solids content and Ms.

alumina species in the binder and the formation of reaction

products [33,34]. The smaller ionic radius of the solvated K+
ion [35–37] likely results in a faster ionic movement through
the system and thus a reduced dormant period and a
more intense acceleration peak. It has also been found in
a companion study that the potassium silicate activator is
about 3 to 5 times less viscous than the sodium silicate
activator of the same Ms[38]. This also contributes to faster
ionic movement and consequently increased early age
reactivity in the K-silicate activated slag pastes.
Figure 2 shows the influences of the solids content in the
Na-silicate solution and that of reducing the activator Ms
through the addition of external NaOH on the heat release
response of the activated pastes. The solution with a
solids content of 36% had an Ms of 3.3 and the one with
a solids content of 44% had an Ms of 2.0. The amount of
NaOH to be externally added is lower for the solution with
a solids content of 44% (Ms = 2.0) than for the one with
solids content of 36% (Ms = 3.3) to bring both the solutions
to an Ms of 1.5. The magnitude of the heat release peak
(and hence the extent of reaction) more than doubles when
the Ms is reduced from 2.0 to 1.5, indicating the influence
of relative amounts of alkali and silica in solution that
results in desirable rates and amounts of reaction product
formation and thus property development. The net amount
of Na2O in the system is the same irrespective of the solids
content of the starting solution which makes the initial
solids content in the solution irrelevant with respect to
the heat release response as observed from Figure 2. For
the pastes activated with the solution having an Ms of 2.0,
the acceleration peak of a much lower magnitude appears
after a longer dormant period as compared to the systems
made using an activator of Ms = 1.5. There is virtually no
heat released when slag is activated using a Na-silicate
solution of Ms = 3.3. This is attributed to an excess of Fig. 3: (a) Compressive strength development of Na- and K- sili
cat e (36 % soli ds content activated slag mortars, (b) com-
silicates and scarcity of alkalis in the system at an Ms of 3.3
pressive strength at early ages (1 and 3 days) for the Na- and
which prevents the polycondensation reaction as alkalis K- silicate activated mortars, and (c) compressive strength
are required to sever the silica ions in silicate chains and development of Na silicate activated slag mortars as a function
incorporate the aluminum ion in the gel structure to form of the solids content in the Na silicate solution.
the C-(A)-S-H gel in the presence of M+ ion[10,39].
Organised by
India Chapter of American Concrete Institute 275
Session 3 A - Paper 6

beyond), the Na-silicate activated mortar proportioned TG/DTG analysis and reaction product quantification
using a lower Ms (1.5) is stronger than the corresponding The thermogravimetric (TG) and differential
K-silicate activated mortar. This could be attributed to the thermogravimetric (DTG) curves of alkali activated slag
increased amounts of reaction product formation in Na- pastes are shown in Figures 4 and 5. Figures 4(a) and
silicate activated systems (quantified in the later section) (b) show the time-dependent evolution of the reaction
and the likely enhancement in the degrees of silicate products for Na- and K-silicate activated slag pastes
polymerization attributable to the Na+ cation being able respectively when activated using alkali silicate solutions
to better coagulate with monomeric silicates species[21,36]. of Ms = 1.5. The main mass loss peak in the DTG curves
There is however no distinguishable strength difference in the 50°C-300°C region can be attributed to the major
between the Na- and K-silicate activated mortars when reaction product, C-S-H gel. It needs to be noted that the
an activator Ms of 2.5 is used. Thus the cationic effect C-S-H gel in alkali activated slags incorporates significant
on mechanical properties is more evident when the amounts of Al, depending on the reactive Al content of the
cationic concentrations are higher, i.e., at lower molar slag. When activators of high alkalinity are used, a more
Ms values of the activator. At higher activator Ms values, crystalline C-(N)-A-S-H gel has also been reported[41].
the silica content dominates the strength response, It can be readily observed from these figures that the
and the variations in the reaction product as a result of total mass loss up to about 600°C, which reasonably
incorporation of either Na or K is expected to be minimal. approximates the degree of reaction of these systems,
Figure 3(b) depicts the 1- and 3-day strengths of the is approximately the same at 28 days irrespective of
Na- and K-silicate activated slag mortars. The higher the activator type. However, at an early age (1 day), the
early age strengths observed for K-silicate activated K-silicate activated paste demonstrates a much higher
systems, especially at lower Ms can also be ascribed to degree of reaction as shown by the bound water content,
the explanations provided for the heat evolution rates which is in line with the observations from isothermal
described earlier. With an increase in the alkali ion calorimetry and compressive strengths. The later-age
availability (such as for a lower Ms), the effect of increased DTG curves also show the presence of a hydrotalcite-
amounts of silica in solution when K-silicates are used, is like phase with a mass loss in the range of 300-400°C.
magnified. Hence the strength of the K-silicate activated Hydrotalcite phases have been commonly observed in
mortar is higher at 1 and 3 days than the Na-silicate alkali activated slag systems[24,42–44] and they are found to
activated mortar when the activator Ms is lower. increase with increasing reaction time.

Figure 3(c) depicts the compressive strength development Figures 5(a) and (b) show the TG and DTG curves for 28-
as a function of the solids content of the as-obtained Na- day hydrated slag activated using Na- and K-silicates
silicate solution. As mentioned earlier, Na-silicate with a
solids content of 36% had an as-received Ms of 3.3 and
the one with a solids content of 44% had an as-received
Ms of 2.0. In this case also, NaOH was added to reduce
the Ms values to 1.5 so as to ensure adequate property
development[11,40]. It can be observed that the strength is
lower at all ages when the solids content of the activator
is higher. The amount of soluble silica and Na2O is same
in both the activators, with the only difference being in
the amount of NaOH and water added externally. For the
sodium silicate activator with a lower solids content, a
higher level of NaOH addition was required as explained
in the previous section.
This results in a compressive strength difference of about
20 MPa at later ages when more NaOH is added to lower
the Ms of the solution containing a lower solids content.
In addition to improving the dissolution of ionic species
from slag and facilitating increased reaction, higher NaOH
addition within limits also helps incorporate more Al in
the tetrahedral structures and higher degrees of silicate
polymerization[10,22], which are responsible for strength
increase. Thus the external addition of the alkali hydroxide
plays a significant role in the degree of polymerization
of the aluminosilicate gel structure and the resultant
mechanical properties. Fig. 4: TG and DTG curves of: (a) Na-silicate and (b) K-silicate
activated slag pastes (Ms = 1.5) after 1, 7 and 28 days of reaction

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

276 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Alkali Activated Slag: Reaction Kinetics and Hydration Products

Quantification of the amount of C-(A)-S-H gel from TG

data is difficult in general because the mass loss of other
reaction products like hydrotalcite overlaps with that of
C-(A)-S-H in the temperature range considered. However,
in this study, only a relative quantification of the reaction
products is attempted. It is assumed, and rightly so, that
the amount of C-(A)-S-H gel is significantly higher than
the other reaction products. Also, it is assumed that the
reaction product stoichiometry does not significantly
change based on the type of alkali cation, Ms, or with age
of reaction after 7 days. The only compositional difference
that is considered here is on the Al/Si ratios of the reaction
product, which were determined from 29Si MAS NMR
spectroscopy using the integrated intensities of resonance
lines [41], as detailed in the forthcoming section. For an
Ms of 1.5, the Al/Si ratio was found to be 0.12, whereas
it decreases to 0.10 when the Ms was increased to 2.5,
irrespective of the alkali cation type. The Ca/Si ratio was
fixed at 0.80 for all the pastes based on past studies
[26,43] and the H/S ratio was taken as 1.2 [12].
The amount of C-(A)-S-H gel formed over time as a
function of the activator Ms is shown in Figure 6(a) and (b).
As expected, the reaction product quantity increases with
Fig. 5: TG and DTG curves of: (a) Na-silicate and (b) K-silicate time. All the pastes show a reduction in reaction product
activated slag pastes at 28 days as a function of activator Ms content with increasing Ms, which is reflected in the
compressive strength results also (Figure 3(a)). While the
respectively, of various Ms values. At a low Ms (or a higher amounts of reaction products are about the same for Na-
alkalinity), the hydrotalcite peaks in the DTG curves are and K-silicate activated slag pastes when the Ms value is
stronger for both Na- and K-silicate activation, indicating low, they are seen to diverge at higher activator Ms values,
the influence of Ms on the hydration product constitution. with the Na-silicate activated paste showing a higher gel
A closer examination of the DTG plots also reveals that content, and consequently a higher strength. This could be
the hydrotalcite formation is higher for the K-silicate the result of Na+ being strongly hydrated than the K+ ions
activated pastes, attributed to the increased basicity of that influence the rate of silica dissolution, and thus the
the K-based activators. product formation at higher Ms values [45]. Notwithstanding
the potential sources of errors in the quantification of the
When the bound water contents (mass loss up to 600oC)
amount of reaction products, the trends mirror that of the
were considered, no perceptible differences were found
measured mechanical properties of these systems. The
between the Na- and K-silicate activated pastes (of the
NMR spectroscopic information presented in the following
same activator Ms) at reaction ages of 7 and 28 days.
section also provides a confirmation of the accuracy of these

Fig. 6: Amount of C-(A)-S-H (g/100g of sample) gel as a function of activator Ms for Na and K silicate activated slag pastes after: (a)
7 days, (b) 28 days of reaction; and (c) relationship between the amount of C-(A)-S-H gel and compressive strength of the mortars.

Organised by
India Chapter of American Concrete Institute 277
Session 3 A - Paper 6

trends, where it is shown that the amount of unreacted slag, differences between the combined intensities of the Q1,
after a reaction age of 28 days, is higher for the K-silicate Q2(1Al), and Q2(0Al) structures (degrees of reaction) for Ms
activated paste for Ms values of 1.5 and 2.5. Figure 6(c) shows values of 1.5 and 2.5 are higher for the K-silicate activated
the relationship between the amounts of C-(A)-S-H gel and pastes, a trend that is consistent with Figure 6(b).
the compressive strengths. A generally increasing strength
The mean silicate chain lengths of the C-(A)-S-H gels
with increasing gel content is observed; the deviations from
calculated using the expression provided in[10] range
linearity can be attributed to the differences in degrees of
polymerization of the gel likely induced by the alkali cation between 10.6 and 12.4 units, with a slight increase with an
type and the silica modulus[46]. increase in Ms and almost invariant with the alkali cation
type. Since the mean chain length is related to the Ca/Si
NMR Spectroscopy ratio of the reaction product, the use of the same Ca/Si
ratio in the previous section for quantification of C-(A)-
Figure 7 depicts the 29Si MAS NMR spectra for slag S-H gel in both Na- and K-silicate activated pastes can
activated using Na- and K-silicate solutions of Ms 1.5 and be justified. The tetrahedral condensation ratios, given
2.5 after 28±2 days of reaction. A Gaussian deconvolution
byΣQ2 /QTOTAL, are found to be slightly higher for the Na-
procedure was adopted for the overlapping peaks in the
silicate activated slag pastes than those for the K-silicate
MAS NMR spectra of the reacted pastes. The component
activated pastes at the respective Ms values. This indicates
peaks and simulated spectra are shown in Figure 7. It
that there is more of tetrahedrally coordinated Al in Na-
is assumed in the deconvolution process that the shape
silicate activated slag systems. Similarly the ratio Q2
of the remnant anhydrous slag resonance line does not
(0 Al)/ Q2 (1Al ) is lower in the reaction product formed
change during the reaction. The peaks are assigned to Qn
through Na-silicate activation of slag, demonstrating
structures based on available literature for cement and
that the amount of Al incorporated in bridging positions
activated slags[47,48]. The broad deconvoluted peak around
-75.8 ppm is attributed to the anhydrous slag remaining in is higher in the pastes activated using Na-silicates. This
the system. The relative amounts of Si in the different Qn quantity also is lower when the Ms values are lower for
structures are proportional to the integrated intensities of both the alkali cation types showing that higher alkalinity
each of the resonance lines in the deconvoluted spectra. facilitates better Al incorporation in the reaction products.
Figure 8 represents the fraction of Qn sites as a function Increased SiO2 content with higher Ms causes this ratio to
of the cation type and Ms, obtained from the relative increase.
integrated intensities of the deconvoluted spectral lines. The spectral deconvolution also points to peaks at -89 and
It can be noticed that, for the same Ms value, the K-based -93 ppm as shown in Figure 8, which are generally assigned
activating solutions result in a higher amount of unreacted to Al-substituted Q3 and Q4 species. Recent studies have
slag in the system, which is reflected in the compressive interpreted this to mean the existence of C-N(K)-A-S-H
strength (Figure 3(a)) of these systems and the amount of gel[50] and lends credence to the fact that the C-S-H gels
C-A-S-H gel (Figure 6(b)).
The assignment of resonances between -75 and -80 ppm
for alkali activated slags is not easy, which is generally
attributed to either Q0 or Q1 sites. It has been reported
that the chemical shifts of Q1 sites in C-S-H gels can be
widely different depending on the number of cations in
charge balancing sites or the type of the charge balancing
species[49]. The predominant Q1-related resonance occurs
around -78.5 ppm for the pastes studied here, with a more
negative chemical shift when K is used as the cation. The
major contribution to the NMR spectra from the reacted
phases, as shown in Figure 8, is provided by the Q2(1Al)
and Q2 resonances around -82 and -85 ppm respectively.
This is consistent with the occurrence of an Al-substituted
C-S-H (C-A-S-H) gel. The combined intensities of the Q1,
Q2(1Al), and Q2(0Al) structures were found to be around
60% of the total integrated intensities of the deconvoluted
spectra for the Na-silicate activated pastes and 53% for the
K-silicate activated pastes. These values can essentially
be considered to be the degree of reaction of slag in these Fig. 7: Amount of C-(A)-S-H (g/100g of sample) gel as a function
systems. The integrated NMR spectra results are in of activator Ms for Na and K silicate activated slag pastes after:
agreement with the quantification of the reaction products (a) 7 days, (b) 28 days of reaction; and (c) relationship between
shown in Figure 6(b) obtained from thermal analysis. The the amount of C-(A)-S-H gel and compressive strength of the
mortars.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

278 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Alkali Activated Slag: Reaction Kinetics and Hydration Products

analysis results along with the gel stoichiometry from

published literature and the NMR studies conducted as
part of this study showed reducing amounts of reaction
product with increasing Ms for both activator types beyond
7 days of reaction. At later ages, the Na-silicate activated
pastes showed higher amounts of reaction product
formation from TG analysis, and K-silicate activated
systems showed higher amounts of unreacted slag from
the NMR spectra, both of which generally corresponds
to the trends in compressive strength. While the mean
chain length of the silicate chains in the reaction product
(related to the Ca/Si ratio of the gel) remained relatively
Fig. 8: Fraction of Qn sites for 28-day Na and K silicate activated
slag pastes from integrated intensities of 29Si NMR spectra
unchanged with the activator type, the ratios ΣQ 2 /QTOTAL
and Q 2 (0Al) / Q2 (1Al) were higher for the Na-silicate
in alkali activated slags are more polymerized than those activated pastes, pointing to more Al incorporation in the
in portland cement-based systems. The contribution of gel when activated using Na-silicates.
these resonance lines to the total integrated intensity
of the spectra is less than 5%, and hence have not been Acknowledgements
considered in the analysis. It is however interesting to note The authors gratefully acknowledge Holcim US and PQ
from Figure 8 that the intensities of Q3 and Q4 are higher Corporation for providing the materials used in this study.
when a higher Ms is used irrespective of the activator The work was carried out in the Laboratory for the Science
cation type. This potentially indicates that, at longer and Sustainable Infrastructural Materials (LS-SIM) and
durations of curing, more amounts of alkali-incorporated, the Physics NMR facility at Arizona State University,
highly cross-linked gel is formed when the activator and the support that has made the establishment and
alkali content is lower. Increased amounts of alkalis are operation of these laboratories are also acknowledged.
thus only essential in enhancing the dissolution process
and forming reaction products that provide high early References
strengths as shown in Figure 3(b).
1. Worrell E, Price L, Martin N, Hendriks C, Meida LO. Carbon Dioxide
Emissions from the Global Cement Industry. Annu Rev Energy
Environ 2001;26:303–29.
CONCLUSIONS
2. Taylor M, Tam C, Gielen D. Energy efficiency and CO2 emissions
This paper has presented the results of experimental from the global cement industry. Int Energy Agency 2006.
studies carried out with an aim of exploring the influence of 3. Marland G, Boden TA, Griffin RC, Huang SF, Kanciruk P, Nelson
the alkali cation type on the reaction kinetics, compressive TR. Estimates of CO/sub 2/emissions from fossil fuel burning
strength, and reaction product structure in alkali and cement manufacturing, based on the United Nations energy
activated slag systems. It was observed that the K-silicate statistics and the US Bureau of Mines cement manufacturing data.
Oak Ridge National Lab., TN (USA); 1989.
activated slag pastes demonstrate higher heat of reaction
after 72h. The induction period was shorter and the 4. Roy DM. Alkali-activated cements Opportunities and challenges.
Cem Concr Res 1999;29:249–54.
intensity of the acceleration peak higher for the K-silicate
activated systems at lower Ms values than those activated 5. Palomo A, Grutzeck MW, Blanco MT. Alkali-activated fly ashes: A
cement for the future. Cem Concr Res 1999;29:1323–9.
with Na-silicate showing the effectiveness of K-based
6. Xu H, Van Deventer JSJ. The geopolymerisation of alumino-silicate
activators, including those contributed by the smaller
minerals. Int J Miner Process 2000;59:247–66.
solvated radius of K+ and lower viscosity of the activator,
7. Bernal SA, Mejía de Gutiérrez R, Provis JL. Engineering and
in early-age reactions. The early-age compressive
durability properties of concretes based on alkali-activated
strength results also validate this observation. At later granulated blast furnace slag/metakaolin blends. Constr Build
ages, the compressive strengths are higher for the Na- Mater 2012;33:99–108.
silicate activated systems at lower Ms attributable to the 8. Bijen J. Benefits of slag and fly ash. Constr Build Mater 1996;10:309–
effects of increased amounts of Na that likely result in a 14.
more crystalline reaction product. Increasing the silica 9. Fernandez-Jimenez A, García-Lodeiro I, Palomo A. Durability
content of the activator resulted in comparable later-age of alkali-activated fly ash cementitious materials. J Mater Sci
strengths irrespective of the alkali cation type. 2007;42:3055–65.
10. Fernández-Jiménez A, Puertas F, Sobrados I, Sanz J. Structure
29
Si MAS NMR spectra of the activated pastes and of Calcium Silicate Hydrates Formed in Alkaline-Activated Slag:
spectral deconvolution indicated major occurrences of Influence of the Type of Alkaline Activator. J Am Ceram Soc
Q2 (1Al) and Q2 resonances, providing evidence of the 2003;86:1389–94.
formation of an Al- substituted C-S-H gel. An approximate 11. Bakharev T, Sanjayan JG, Cheng Y-B. Effect of elevated temperature
quantification of C-(A)-S-H gel utilizing the thermal curing on properties of alkali-activated slag concrete. Cem Concr
Res 1999;29:1619–25.

Organised by
India Chapter of American Concrete Institute 279
Session 3 A - Paper 6

12. Chen W, Brouwers HJH. The hydration of slag, part 1: reaction performance of alkali activated fly ash and slag systems. Clarkson
models for alkali-activated slag. J Mater Sci 2007;42:428–43. University, 2012.
13. Glukhovsky VD, Rostovskaja GS, Rumyna GV. High strength slag- 33. Depasse J. Coagulation of Colloidal Silica by Alkaline Cations:
alkaline cements. 7th Int. Congr. Chem. Cem., vol. 3, 1980, p. 164–8. Surface Dehydration or Interparticle Bridging? J Colloid Interface
14. Shi C, Day RL. A calorimetric study of early hydration of alkali-slag Sci 1997;194:260–2.
cements. Cem Concr Res 1995;25:1333–46. 34. Depasse J. Simple Experiments to Emphasize the Main
15. Holzer L, Figi R, Gruskovnjak A, Lothenbach B, Winnefeld F. Characteristics of the Coagulation of Silica Hydrosols by Alkaline
Hydration of alkali- activated slag: comparison with ordinary Cations: Application to the Analysis of the Model of Colic et al. J
Portland cement. Adv Cem Res 2006;18:119–28. Colloid Interface Sci 1999;220:174–6.

16. Yip CK, Lukey GC, van Deventer JSJ. The coexistence of geopolymeric 35. Bach TTH, Chabas E, Pochard I, Cau Dit Coumes C, Haas J, Frizon
gel and calcium silicate hydrate at the early stage of alkaline F, et al. Retention of alkali ions by hydrated low-pH cements:
activation. Cem Concr Res 2005;35:1688–97. Mechanism and Na+/K+ selectivity. Cem Concr Res 2013;51:14–21.

17. Davidovits J. Geopolymers and geopolymeric materials. J Therm 36. Provis JL, van Deventer JSJ. Geopolymerisation kinetics. 1.
Anal Calorim 1989;35:429–41. In situ energy-dispersive X-ray diffractometry. Chem Eng Sci
2007;62:2309–17.
18. Duxson P, Fernández-Jiménez A, Provis JL, Lukey GC, Palomo A,
Deventer JSJ van. Geopolymer technology: the current state of the 37. Khale D, Chaudhary R. Mechanism of geopolymerization and factors
art. J Mater Sci 2007;42:2917–33. influencing its development: a review. J Mater Sci 2007;42:729–46.
19. Duxson P, Provis JL, Lukey GC, van Deventer JSJ. The role of 38. Vance K, Dakhane A, Neithalath N. Rheological behaviour of alkali
inorganic polymer technology in the development of “green activated fly ash suspensions: Influence of the activator type and
concrete.” Cem Concr Res 2007;37:1590–7. chemistry n.d.
20. Hunt JM, Wisherd MP, Bonham LC. Infrared Absorption Spectra of 39. Hong S-Y, Glasser F. Alkali sorption by C-S-H and C-A-S-H gels:
Minerals and Other Inorganic Compounds. Anal Chem 1950;22:1478– Part II. Role of alumina. Cem Concr Res 2002;32:1101–11.
97. 40. Živica V. Effects of type and dosage of alkaline activator and
21. Ravikumar D, Neithalath N. Effects of activator characteristics on temperature on the properties of alkali-activated slag mixtures.
the reaction product formation in slag binders activated using alkali Constr Build Mater 2007;21:1463–9.
silicate powder and NaOH. Cem Concr Compos 2012;34:809–18.
41. Ben Haha M, Le Saout G, Winnefeld F, Lothenbach B. Influence
22. Ravikumar D, Neithalath N. Reaction kinetics in sodium silicate of activator type on hydration kinetics, hydrate assemblage and
powder and liquid activated slag binders evaluated using isothermal microstructural development of alkali activated blast-furnace slags.
calorimetry. Thermochim Acta 2012;546:32–43. Cem Concr Res 2011;41:301–10.
23. Chithiraputhiran S, Neithalath N. Isothermal reaction kinetics and 42. Haha MB, Lothenbach B, Le Saout G, Winnefeld F. Influence of slag
temperature dependence of alkali activation of slag, fly ash and chemistry on the hydration of alkali-activated blast-furnace slag —
their blends. Constr Build Mater 2013;45:233–42. Part II: Effect of Al2O3. Cem Concr Res 2012;42:74–83.
24. Wang S-D, Scrivener KL. Hydration products of alkali activated slag 43. Lothenbach B, Gruskovnjak A. Hydration of alkali-activated slag:
cement. Cem Concr Res 1995;25:561–71. thermodynamic modelling. Adv Cem Res 2007;19:81–92.
25. Chithiraputhiran SR. Kinetics of Alkaline Activation of Slag and Fly 44. Puertas F, Fernández-Jiménez A. Mineralogical and microstructural
ash-Slag Systems. Arizona State University, 2012. characterisation of alkali-activated fly ash/slag pastes. Cem Concr
26. Fernández-Jiménez A, Zibouche F, Boudissa N, García-Lodeiro I, Compos 2003;25:287–92.
Abadlia MT, Palomo A. “Metakaolin-Slag-Clinker Blends.” The Role 45. Phair JW, Van Deventer JSJ. Effect of the silicate activator pH on
of Na+ or K+ as Alkaline Activators of Theses Ternary Blends. J Am the microstructural characteristics of waste-based geopolymers.
Ceram Soc 2013;96:1991–8. Int J Miner Process 2002;66:121–43.
27. C09 Committee. Practice for Measuring Hydration Kinetics of 46. Dakhane A. Reaction kinetics and quantification anaysis of sodium
Hydraulic Cementitious Mixtures Using Isothermal Calorimetry. and potassium silicate liquid and powdered activators in alkali
ASTM International; 2009. activation of slag based binders. Arizona State University, 2013.
28. Barbosa VFF, MacKenzie KJD. Thermal behaviour of inorganic 47. Lecomte I, Henrist C, Liégeois M, Maseri F, Rulmont A, Cloots
geopolymers and composites derived from sodium polysialate. R. (Micro)-structural comparison between geopolymers, alkali-
Mater Res Bull 2003;38:319–31. activated slag cement and Portland cement. J Eur Ceram Soc
29. Duxson P, Lukey GC, Deventer JSJ van. Physical evolution of Na- 2006;26:3789–97.
geopolymer derived from metakaolin up to 1000 °C. J Mater Sci
48. Wang S-D, Scrivener KL. 29Si and 27Al NMR study of alkali-
2007;42:3044–54.
activated slag. Cem Concr Res 2003;33:769–74.
30. Duxson P, Lukey GC, van Deventer JSJ. The thermal evolution of
49. Barnes P, Bensted J. Structure and Performance of Cements,
metakaolin geopolymers: Part 2 – Phase stability and structural
Second Edition. Taylor & Francis; 2002.
development. J Non-Cryst Solids 2007;353:2186–200.
50. Bernal SA, Provis JL, Walkley B, San Nicolas R, Gehman JD, Brice
31. Vail JG, Wills JU. Soluble Silicates: Their Properties And Uses, Vol.
DG, et al. Gel nanostructure in alkali-activated binders based on
2: Technology 1952.
slag and fly ash, and effects of accelerated carbonation. Cem Concr
32. Ravikumar D. Property development, microstructure and Res 2013;53:127–44.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

280 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Alkali Activated Slag: Reaction Kinetics and Hydration Products

Narayanan Neithalath
Narayanan Neithalath is an Associate Professor in the School of Sustainable Engineering and the Built
Environment at Arizona State University, Tempe AZ. He received his PhD in Civil Engineering Materials
from Purdue University in 2004. His research interests are in the areas of sustainable construction
materials including high volume cement replacement materials for concrete, development of novel
materials including unconventional cement-less binders through carbonation and alkaline activation,
characterization and modeling of microstructure and properties of such systems. He has received several
awards for his work on novel cementitious systems. He is the Editor of the Cementitious Materials section
of the ASCE J of Materials in Civil Engineering and an Editorial Board Member of Cement and Concrete
Composites.

Organised by
India Chapter of American Concrete Institute 281
SESSION 3 B
Session 3 B - Paper 1

Studies of Strength, Durability and Microstructural Properties of Cow

Dung Ash Modified Concrete

D.Ramachandran, Vinita Vishwakarma, R.P. George Kalpana Kumari,

N.Anbarasan, K. Viswanathan Corrosion Science and V. Venkatachalapathy
Centre for Nanoscience and Nanotechnology, Technology Group, IGCAR, Civil Engineering Group,
Sathyabama University, Chennai - 600119 Kalpakkam-603102 IGCAR, Kalpakkam - 603102

Abstract may also improve workability, act as additional binder,

has more hardening enzymes and adhesive qualities (J.
This study is done by partially replacing cow dung ash
Ashurst and N.Ashurst 1988a). Pearson, T. Gordon (1992)
(CDA) instead of ordinary portland cement (OPC) in the
and Holmes and Wingate (1997) both suggest dung modifies
production of concrete. Five sets of concrete mixes of
lateritic soil plasticity, acts as a binder and so improves
M40 grade had CDA replaced with OPC in the ratio of 5.0,
durability. The application of cow dung with respect to
7.5, 10.0, 12.5 and 15.0% CDA by weight of cement. It was
concrete tanks was studied by Prithwiraj Jha et al (2004).
compared with normal concrete (NC). Energy dispersive
Sirri Sahin et al (2006) investigated partial replacement
X-Ray diffraction (XRD) study confirmed presences of
of concrete with Cattle Waste Ash (CWA). Pam Billy Fom
nano-silica in CDA. After 7d of curing, there was not much
et al (2011) found the higher effect of cement-cow dung
appreciable increase in compressive strength of CDA
on the compressive strength of lateritic-soil blocks and
aided mixes. But after 28d both compressive and split
low- cost improved lateritic-soil in rural housing projects.
tensile strength was increased in the 15% modified CDA.
It could also reduce the emission of carbon-dioxide and
Superplasticizer requirement was increased with increase
global warming due to cement production. CDA contains
in CDA percentage for achieving similar workability. Rapid
approximately 60% of silica and other elements (G.
chloride Penetration Test (RCPT) results confirmed, CDA
Sivakumar and K. Amutha, 2012). The main component
modified concretes are low in permeability as well as
of bio waste ashes such as sugar cane bagasse, sugar
surface and internal pH study confirmed that alkalinity
cane leaf, rice straw, and corn leaf is silicon dioxide
is also well maintained. The chemical composition
(SiO2) (Kuen-Song Lin et al. 2003). Pavan et al (2012)
and microstructure properties were analyzed by X-ray
found the significance and necessity of using CDA for the
diffraction (XRD), Field Emission Scanning Electron
manufacture of sustainable concrete for construction
Microscopy (FESEM) and Thermogravimetry/ /Differential
of green buildings in future. Chemical properties of cow
thermal analysis (TG/DTA).
dung showed it is a nitrogen rich material with, potassium,
Keywords: Cow dung ash, Nano-silica, Compressive phosphorous and calcium (Smith and Wheeler 1979). Cow
strength, RCPT, TG/DTA. dung has a relatively high carbon to nitrogen ratio. The
chemical composition of the two cow dung was compared
Introduction and it was found that, there was no difference in the
organic matter (OM), nitrogen (N) and manganese (Mn)
The concrete industry is the largest consumer of natural but the contents of calcium (Ca), phosphorus (P), zinc (Zn)
resources such as water, sand, gravel and crushed rock and copper (Cu) were higher by 10.8, 8.0, 84.1 and 21.7
(Karim MR et al. 2013). One of the major issues in the percent in the dung of desi cows as compared to that of
construction sector is to develop sustainable concrete to crossbred cows (Garg Mudgal 2007). Microstructure
reduce the emission of carbon dioxide (CO2) and pollution properties of the concrete always influence their strength,
during the manufacture of concrete. Production of dimensional stability and durability. However, knowledge
cement leads to higher energy consumption, emission of the microstructure and properties of the individual
of greenhouse gases, as well as air pollutants (Karbassi components of concrete and their relationship to each
et al. 2010 and Ghrici et al. 2007). Partial replacement other is useful for exercising control on properties (Mehta
of cement with non-polluting materials is to be done to 2014). The chemical compounds formed in various
balance the performance and cost. Arshad, A et al. 2014 chemical reactions are quantified using thermal analysis
studied on reusing or recycling waste material in concrete. (Vedalakshmi 2008).
Dumping the waste materials will affect the environment
directly (Aliabdo et al. 2014). In the recent years, few Thermogravimetry is used to study the interaction
researchers have approached to modify the mortar phase between polymers and cement, as well as the extent
of the concrete using CDA, as it improves the durability of of pozzolanic reaction of the mortars with silica fume
concrete structures. It has been suggested that cow dung (Alessandra and Eduvaldo 2006).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

284 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Studies of Strength, Durability and Microstructural Properties of Cow Dung Ash Modified Concrete

There has been very less research on the CDA modified pH study
concrete and its properties. This study is attempted to All sets of specimens were tested for surface (WTW
identify and improve concrete properties using nano size SenTix- 3110) as well as internal pH (Hanna, HI-2211).
CDA mineral admixture as a partial replacement of OPC. The internal pH of concrete was tested by crushing the
specimens of 28d of cured NC and concrete modified with
Experimental CDA.

Materials and methods Rapid chloride Penetration Test (RCPT) Study

Ordinary Portland cement (OPC) (43 Grade) of specific Chloride corrosion is one of the most common
gravity 3.1 and confirming the requirements of IS 8112 environmental attacks that lead to the deterioration of
(2013) was used. Other ingredients such as river sand, fine concrete due to internal pores. Durability represented by
aggregates conforming zone II as per IS 383 classification, RCPT, was performed as per ASTM C 1202, to determine
coarse aggregates (blue Granite rock aggregate - the electrical conductance of concrete and concrete
machine crushed) with a maximum nominal size 20mm modified with CDA after 28d of curing. The test method
and 12.5mm, superplasticizer (SNF based- type G) as per consists of monitoring the amount of electrical current
ASTM C494 and potable water were used. The cow dung passed for 6h through 50 mm thick samples which were
was collected, dried in sunlight, burnt to obtain the ash sliced from 100mm diameter concrete cylinders.
and sieved through 425μ mesh. Specific gravity of the ash
was 2.39. XRD studies of NC and concrete modified with CDA
The X-ray diffraction studies were performed by powder
Field Emission Scanning Electron Microscopy (FESEM) X-ray diffractometer, Rigaku (9kW) smartLab and Copper
with EDAX analysis of OPC and CDA (K α) was used as a target material. The NC and concrete
The FESEM with EDAX is a multifaceted instrument used modified with CDA of the entire mix ratio of 5.0, 7.5, 10.0,
to find out the morphology, size and chemical composition 12.5 & 15.0 % were crushed to make the powder. Brag
of powder sample. The OPC and CDA were studied by Carl Brentano method was used to analyze and identify the
Zeiss, SUPRA® 55 with GEMINI® Technology. Both the unknown crystalline compounds.
samples were chosen for FESEM and EDAX were sputter
The scan step size was 0.02°, the collection time 1s, and in
coated with gold for electrical conduction.
the range 2θ Cu Kα from 10° to 80°. The X-ray tube voltage
and current were fixed at 30 kV and 100 mA respectively.
Specimen Preparation The standard database (JCPDS) was used for phase
Five sets of concrete mixes of M40 grade were prepared identification for a large variety of crystalline phases in
by replacing OPC with 5.0, 7.5, 10.0, 12.5 and 15% CDA by NC and concrete modified with CDA.
weight of cement and were compared with NC (0 % CDA).
The water content and total powder content (cement + FESEM with EDAX of NC and concrete modified with
CDA) were kept constant. In order to obtain a reasonably CDA
similar workable concrete, the chemical admixture
FESEM has the ability to characterize cement and concrete
dosage was altered. The total powder content was 365
microstructure and will aid in evaluating the influence of
kg/m3 and water to powder ratio was 0.44. Specimens
supplementary cementing materials (Scrivener 1988).
were cast and then stripped after 24h, initial curing at
Through FESEM, we can observe the pores, cracks,
standard control environment and then immersed in
micro-cavities, morphology of phases and deposition area
water for 28d curing. Thereafter they were tested for
of minerals and chemical composition of mineral phases.
compressive strength, split tensile strength, pH and
Energy dispersive X-ray analysis (EDAX) together with
chloride permeability.
X-ray analysis software is standard accessories mounted
with FESEM. EDAX can simultaneously analyze the entire
Studies on Mechanical Properties surface under the electron beam and provide qualitative
The mechanical properties of the concrete tested, include analysis of the major elements on the surface. The
compressive and split tensile strength. Specimens of analysis is referred to as energy dispersive X-ray analysis
150 x 150mm cube and 150 x 300mm cylinder were cast (EDAX). The cylindrical specimens of NC and concrete
and stripped after 24h and then cured for 28d in normal modified with CDA of the entire mix ratio (5.0, 7.5, 10.0,
water. The compressive strength test was performed 12.5 & 15.0 %) were cured for 28d and used for FESEM
as per provisions of IS 516 and split tensile strength test study. The surface morphological characteristics of the
was carried out as per IS 5816. Compressive strength specimens were observed under Carl Zeiss, SUPRA® 55
of concrete is evaluated using automatic compression with GEMINI® Technology at 20 kV. All specimens selected
testing machine of 3000kN capacity for the analysis were coated with gold for electrical
conduction.

Organised by
India Chapter of American Concrete Institute 285
Session 3 B - Paper 1

TG/DTA analysis of NC and concrete modified with CDA Compressive and Tensile Strength
The quantity of mass loss in the concrete with increase Compressive strength of the NC cured for 7 and 28d
in temperature was studied by using TG/ DTA, which is showed an average of 42.95 and 57.65 N/mm2 respectively.
important to know the stability of the concrete at high Concrete modified with CDA by 5.0 to 15.0 % showed the
temperature. The preparation of samples of NC and average of 30.37-34.21 N/mm2 for 7d and 53.75 – 54.50 N/
concrete modified with 15% CDA for TGA/DTA (NETZSCH mm2 for 28d. Strength variation for NC and CDA replaced
STA 449 F3 Jupiter ®) were ground by using agate mortar concretes have been depicted in Figure. 2.
and pestle. To prevent the carbonation and humidity, the
powdered samples were kept in the vacuum till the TG/DTA
started. The powder sample of NC (30.723 mg) and 15%
CDA modified concrete (35.978 mg) were weighed exactly
using a quartz cell. The samples were taken in ceramic
crucible and Al2O3 powder was used as the reference
material. The TG/DTA were performed by heating the
concrete samples starting from room temperature to
1300 oC at the heating rate 20k/min under nitrogen gas
dynamic atmosphere.

Results
Results of FESEM with EDAX of OPC and CDA
The FESEM analysis of OPC and CDA powder explains the
change in the morphology, size and chemical composition
in both OPC and CDA particles. OPC analysis image shows
presence of bigger size particles and some voids between
Fig. 2: Compressive strength at 7 and 28d
the grains, whereas CDA is closely filled and packed by
small grains. The size of the particles in OPC was observed
as 160 -250nm (figure.1a) while CDA is having 70-80nm The split tensile strength was found to be 4.04 N/mm2 for
(figure.1b). This naturally found nano-silica particles can NC and 3.05, 3.57, 3.60, 3.00 and 4.13 N/mm2 for 5.0, 7.5,
fill the voids further and hence it has the potential for 10.0, 12.5 and 15.0 % respectively for the concrete modified
densification of the concrete in mortar phase. The weight with CDA. It was observed from Figure. 3 that equivalent
percentage of OPC and CDA was analysed by EDAX which tensile strength is obtained for concrete modified with
showed different elements. It was found that basic nature
CDA only when replacement level is reaching to 15% than
of CDA had more silica content than OPC. (Table 1).
that of NC.

Variation of pH
The surface pH of NC was 10.6 and that of CDA replaced
concrete was shown as 10.5, 10.33, 10.25, 10.22 and 10.24
for 5.0 to 15.0 % modifications respectively. The pH of the
crushed samples of NC was 11.31 and that of CDA replaced
concrete samples showed 11.41, 11.78, 11.96, 12.26 and
12.32 for 5.0 to 15.0 % modifications respectively.
Fig. 1: FESEM images of (a) OPC and (b) CDA.

Table 1
EDAX of OPC and CDA is showing weight percentage of different elements

Ingredients C O Na Mg Al Si P S K Cl Ca Ti Fe

OPC 6.56 43.01 - 0.27 1.02 6.21 - 1.08 0.63 - 39.28 0.26 1.69

CDA 6.47 49.81 1.70 3.05 1.45 25.85 1.49 - 1.62 0.63 7.30 - 0.63

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

286 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Studies of Strength, Durability and Microstructural Properties of Cow Dung Ash Modified Concrete

hydrate (JCPDS: 42-1452), as the CDA percentage increases

the intensity of this peak decreased and the calcium silicate
hydrate peak at 27.45 increases. Peak observed at 29.49,
32.39 and 39.60 shows the presence of calcium aluminum
silicate hydrate (JCPDS: 39-1372). Calcium hydroxide
(JCPDS: 04-0733) peak were formed at 33.99 and 54.99
in control NC where as there is an absence of calcium
hydroxide peak at 33.99 shows this calcium hydroxide
reacts with silicon dioxide of CDA and forms calcium
silicate hydrate. This type of C-S-H formation is due to the
secondary hydration reaction in the CDA modified concrete.

FESEM with EDAX of NC and concrete modified with

CDA
The FESEM analysis of all the mix ratio of NC and concrete
modified with CDA were observed after 28d of curing in
normal water. Figure. 5 shows the FESEM micrographs
Fig. 3: Split tensile strength at 28d
which reveal the better structures in all the mix ratio of
concrete modified with CDA (5.0, 7.5, 10.0, 12.5 and 15.0
Chloride Permeability Test %) compared to NC. The presence of nanosilica seems
RCPT was conducted at 28d and the test results showed to be a homogenous dispersion in the concrete structure
low permeability (1000-2000 coulombs) in NC as well and changes the microstructure of concrete. The particles
as concrete modified with CDA up to 15 % replacement, in the concrete structures are self- assembled together
as per ASTM C1202-09. The NC has 1507.5 and 1759.5, and arranged in the spherical form. They have filled the
1663.6, 1698.3, 1564.2, and 1306.8 with respect to 5.0, 7.5, voids between the cement and aggregates. The nanosilica
10.0, 12.5 and 15.0 % CDA modified concrete. not only acted as filler but it is also acts as a promoter
for the pozzolanic reaction (Mohamed and Ragab 2014)
XRD studies of NC and concrete modified with CDA Therefore it will enhance the performance of concrete.
X ray diffraction of the NC generally shows more intense EDAX analysis also confirmed the presence of more silica
peaks as the percentage of CDA increases the intensity in CDA modified concrete than NC (Table 2).
of the peaks decreases. Peak observed at 21.06, 26.77,
42.46, 50.26, 68.22 and 81.55 shows the presence of silicon
dioxide (JCPDS: 46-1045) (Fig. 4). Calcium silicate hydrate
(C-S-H) peaks were observed in 27.45 and 60.09 (JCPD: 33-
0305). The intensity of C-S-H peaks goes on increasing as
the percentage of CDA increases. And also the peak at 28.16
in NC shows the presence of calcium silicate hydroxide

Fig. 5: FESEM micrograph of NC and concrete modified with

CDA (5.0, 7.5, 10.0, 12.5 and 15.0%)

Results of TG/DTA analysis of NC and concrete modified

with CDA
The four endothermic effects are observed in NC and
concrete modified with 15% CDA samples. The first
endothermic effect at 27-100°C indicates the evaporation
of surface absorbed water. The second endothermic
reaction starts at 100 to 350°C shows the dehydration of
calcium silicate hydrates and calcium aluminum silicate
hydrates. The third endothermic effect attributed to the
decomposition of calcium hydroxide into calcium oxide and
Fig. 4: XRD of studies NC and concrete modified with CDA (5.0, water that occurs in the range 370 to 545°C (Fig.6 and 7.).
7.5, 10.0, 12.5 and 15.0%).

Organised by
India Chapter of American Concrete Institute 287
Session 3 B - Paper 1

Table 2
Quantitative analysis of NC and concrete modified with CDA (Wt. %) with EDAX

Ingredients C O Na  Mg Al Si K   Cl Ca  Fe

NC 13.72 47.33 2.03 - 3.86 14.86 0.92 - 15.25 2.03

5.0% of CDA 22.66 34.99 - 2.05 5.71 19.22 4.81 1.83 2.39 6.34

7.5% of CDA 17.07 42.61 0.65 0.58 2.55 16.49 0.59 0.94 16.77 1.74

10.0% of CDA 16.0 43.83 0.33 0.47 1.29 30.83 0.71 - 6.04 0.69

12.5% of CDA 23.06 37.67 - - 2.41 29.41 0.72 - 12.30 1.48

15.0% of CDA 13.70 45.52 1.10 0.37 2.99 21.11 1.42 - 24.14 1.92

Ca(OH)2 → CaO + H2O↑ (Loss of Water)

And the final endothermic reaction at 546 to 752°C shows
the decomposition of calcium carbonate to calcium
oxide and carbon-dioxide gas emission (Alessandra and
Eduvaldo 2006 and Neven et al. 2006).
CaCO3→ CaO + CO2↑ (Loss of Carbon dioxide)
Table 3 confirmed the weight loss of surface water
dehydration H2O, dehydration of C-S-H, decomposition
of calcium hydroxide and decarbonation of calcium
carbonate in both NC and concrete modified with 15%
CDA.

Fig. 7: DTA of NC and concrete modified with CDA in nitrogen

atmosphere

Table 3
TG/DTA of NC and concrete modified with CDA

Concrete Temperature Compound Weight loss

specimens range (°C) evaporation (mg)

29.75 - 98.21 H2 O 0.485

NC 458.75 - 416.75 CO 0.160

546.75 - 752.75 CO2 1.027

27.60 – 99.72 H2 O 0.364

Fig. 6: TGA of NC and concrete modified with CDA in nitrogen CDA 321.60 – 440.61 CO 1.052
atmosphere
589.61-724.61 CO2 0.404

Discussion
The production of cement for concrete is contributing 15%. Several studies have shown that natural pozzolans
nearly 8% of global carbon-dioxide emission (Olivier et have been widely used as a substitute for Portland
al. 2012), which is a matter of concern with regards to the cement in many applications because of its advantageous
environment. Present study is focused to examine basic properties which include cost-reduction, reduction in
mechanical properties such as compressive and tensile heat evolution, decreased permeability and increased
strength, alkalinity through pH, chloride permeability by chemical resistance (Belaidi et al. 2012). Massazza (1993)
means of RCPT when OPC gets replaced by CDA up to declared that if cements contain small amounts of very
active pozzolans such as silica fume then both early

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

288 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Studies of Strength, Durability and Microstructural Properties of Cow Dung Ash Modified Concrete

and ultimate strengths may be higher than those of the research areas which are being focused by the authors.
substituted cement. Nano-silica is found naturally in Results obtained from this research could be the basic
CDA, acts as a pozzolanic material which is substantial information about CDA replaced concretes.
in development of about 40% compressive strength from
7 to 28d period. When nano-silica reacts with Ca(OH)2
Conclusions
produced from the primary hydration of cement, it is
converted into additional C-S-H which is the supporting Following conclusions can be drawn from this
constituent for strength and density in the hardened experimental study.
cementatious system (Singh et al. 2013). Compressive (1) The presence of nano-silica in CDA was confirmed
strength of 28d cured NC and all modified concretes by FESEM and EDAX. It can be added as a mineral
replaced with CDA of 5.0, 7.5, 10.0, 12.5 and 15.0 % have admixture in conventional concrete manufacturing
shown equivalent performance to that of an ordinary M40 process.
grade concrete. Since rate of strength gain is relatively
less during early stage of hydration, CDA concrete mixes (2) CDA replaced concrete show lower and equivalent
can be used for mass concrete applications where heat of compressive strength at early and 28d curing periods
hydration is a concern. However split tensile strength was respectively compared to that of a NC produced using
less around by 15- 20% for lesser replacement of CDA and OPC.
it significantly gained after 15% replacement of CDA. (3) It was observed that the superplasticizer requirement
The surface pH of the NC as well as concrete modified was increased when CDA percentage was increased
with CDA at 28d shows high alkalinity which is higher than for obtaining similar workability.
the NC. Langan et al. (2002) suggested that dissolution (4) Replacement of CDA has not increased the split tensile
of silica increases with pH and chemisorptions of Ca2+ strength. However, there was a marked increase in
can only occur after silica has gone into a solution when tensile strength for a 15% CDA replacement.
attack by OH- ions. Hence CDA replaced concrete can be
used for constructing the reinforced concrete structures. (5) Surface and Internal concrete pH test confirmed that
The study of durability in concrete structures is important CDA replacements do maintain required alkalinity for
because chloride ion ingress is depending on the internal a concrete similar to that of controlled mix.
pore structures of the concrete. The test results of RCPT (6) RCPT confirms low chloride permeability in all CDA
for NC and concrete modified with CDA showed “low modified concretes.
permeability” for chloride penetration, which is also
(7) XRD analysis showed concrete modified with CDA
comparable than that of NC. Zhang (2011) explained that
contains more SiO2 compound and CaO (lime) than NC.
the resistance to chloride ion penetration is positively
increasing with compressive strength. Since natural (8) The FESEM analysis confirmed smaller particle size in
nano-silica is observed in cow dung ash which can also 15% replaced concrete which may fill the air voids.
ensure concrete durability, there is scope for development
(9) TG/DTA examined that after 650oC heat; concrete
of sustainable high performance concrete structures
modified with 15% CDA was thermally stable compared
using CDA replacement. Presence of calcium silicate
to NC.
hydrate and calcium silicate iron oxide in CDA replaced
modified concrete is the main source of concrete strength
and hardening. XRD analysis shows that there is not much Acknowledgement
difference observed in 28d of cured concrete compared to Financial support from Department of Biotechnology,
NC. This concrete modified with CDA contains more SiO2 Government of India (BT/PR7436/BCE/8/946/2012) is
compound and CaO (lime) so there will be a secondary greatly acknowledged. Authors thank to Dr Jeppiaar,
hydration reaction in CDA modified concrete. The CDA Chancellor Sathyabama University, Chennai for his
react with excess lime from the cement hydration forming guidance, encouragement and motivation.
addition as C-S-H compound, which will give more strength
to the concrete. This secondary hydration reaction will References
be slow and it will take place after 28d only. The FESEM 1. Alessandra Etuko Feuzicana de Souza Almeida, Eduvaldo Paulo
analysis showed that in the 15% replaced CDA concrete, Sichieri. (2006). Thermogravemetric Analyses and Mineralogical
the size of the particles were less when compared to NC. Study of Polymer Modified Mortar with Silica Fume. Materials
The results of TG/DTA confirmed that more degradation Research. 9: 321 - 326.
occurs in NC than concrete modified with 15% CDA. The 2. Aliabdo, Ali A., Abd Elmoaty, M., Abd Elmoaty., Auda Esraa, M.
CDA modified concrete shows more stability than NC (2014). Re-use of waste marble dust in the production of cement
and concrete. Construction and Building Materials. 50: 28–41.
after 720oC. The effect of different curing environments,
3. Arshad, A., Shahid, I., Anwar, U. H. C., Baig, M. N.1, Khan, S.2 and
change in alkalinity, improvements in micro-structure due
Shakir, K. (2014). The Wastes Utility in Concrete. International
to pozzolonic activity of CDA, etc are some of the further Journal of Environmental Research. 8: 1323-1328.

Organised by
India Chapter of American Concrete Institute 289
Session 3 B - Paper 1

4. Ashurst J and Ashurst N. Practical Building Conservation. Volume 17. Sahin, Sirri., Kocaman, Bahar., Orung, Ibrahim., Memis, Selcuk.
2: Brick, Terracotta and Earth. Aldershot: Gower, 1988a. (2006). Replacing Cattle Manure Ash as Cement into Concrete.
5. Garg Anil Kumar and Mudgal Visha. (2007). Organic and mineral Journal of Applied Science. 6: 2840 - 2842.
composition of Gomeya (cow dung) from desi and crossbred cow. 18. Scrivener, K.L, Bentur, A, and Pratt, P.L. (1988). Quantitative
International Journal of Cow Science. 3:17-19. Characterization of the Transition one in High Strength Concrete.
6. Ghrici, M., Kenai, S., Said-Mansour, M. (2007). Mechanical properties Advances in Cement Research. 1: 230 - 237.
and durability of mortar and concrete containing natural pozzolana
19. Massazza Franco. Pozzolanic cements (1993). Cement and Concrete
and limestone blended cements. Cement & Concrete Composites.
Composites. 15:185–214.
29: 542–549.
7. Holmes Stafford and Michael Wingate. Building with Lime. 20. Neven Ukrainczyk, Marko Ukrainczyk, Juraj Šipušić1, Tomislav
Intermediate Technologies Publications. 1997; 241. Matusinović. XRD and TGA investigation of hardened cement paste
degradation. Conference on Materials, Processes, Friction and
8. Karim, MR., Zain, MFM., Jamil, M., Lai, FC., Islam, MN. (2011).
Wear MATRIB’06 2006.
Necessity and Opportunity of Sustainable Concrete from Malaysia's
Waste Materials. Australian Journal of Basic and Applied Sciences. 21. Olivier JGJ, Greet J-M, Peters JAHW. Trends in Global CO2 Emissions
5: 998 - 1006. 2012 report. PBL Netherlands Environmental Assessment Agency,
9. Karim, MR., Zain, MFM., Jamil, M., Lai, FC., Islam, MN. (2011). 2012; 17.
Necessity and Opportunity of Sustainable Concrete from Malaysia's 22. Pavan Kumar.Rayaprolu, V.S.R P. Polu Raju. Incorporation of
Waste Materials. Australian Journal of Basic and Applied Sciences. Cow dung Ash to Mortar and Concrete. International Journal of
5: 998 - 1006. Engineering Research and Applications (IJERA). 2012. 2: 580 - 585.
10. Karbassi AR., Jafari, HR., Yavari, AR., Hoveidi, H., Kalal, H. (2010). 23. Saleh Abd El-Aleem Mohamed and Abd El-Rahman Ragab.(2014).
Reduction of environmental pollution through optimization of energy
Physico-Mechanical Properties and Microstructure of Blended
use in cement industries. International Journal of Environment
Cement Incorporating Nano-Silica International Journal of
Science and Technology. 7: 127-134.
Engineering Research & Technology (IJERT). 3 (7) 339- 358.
11. Kumar Mehta, P., Ph.D.; Paulo J. M. Monteiro. (2014). Concrete:
Microstructure, Properties and Materials, Fourth Edition. 24. Scrivener KL, Bentur A and Pratt PL. (1988). Quantitative
Microstructure of Concrete, Chapter (McGraw-Hill Professional). Characterization of the Transition one in High Strength Concrete.
Access Engineering. Advances in Cement Research. 1: 230 - 237.
12. Kuen-Song Lin, H. Paul Wang, Ni-Bin Chang, C. J. G. Jou and M. 25. Singh, L.P., Karade, S.R., Bhattacharyya, S.K., Yousuf, M.M.,
C. Hsiao. (2003). Synthesis of ZSM- type Zeolites from Biowaste Ahalawat, S. (2013). Beneficial role of nanosilica in cement based
Gasification Ashes. Energy Sources. 25: 565 - 576. materials – A review. Construction and Building Materials. 47:
13. Langana, B.W., Wengb, K., Ward, M.A. (2002). Effect of silica fume 1069 - 1077.
and fly ash on heat of hydration of Portland cement. Cement and 26. Sivakumar, G., and Amutha, K. (2012). Studies on Silica Obtained
Concrete Research. 32: 1045 - 1051. from Cow Dung Ash. Advanced Materials Research. 584: 470 - 473.
14. Pam Billy Fom,Uche, O.A.U and Joseph Elma Lagasi. (2011). Effect of 27. Smith, L W and Wheeler, W E (1979) Nutritional and economic value
cement, cow-dung on the compressive strength of lateritic bricks.
of animal excreta. Journal of Animal Science. 48: 144 - 156.
Journal of Science and management. 1:31 - 34.
28. Vedalakshmi, R. Raj., Sundara, A., Palaniswamy, N. (2008).
15. Pearson, T. Gordon. (1992). Conservation of Clay and Chalk Buildings
London: Donhead. Identification of various chemical phenomena in concrete using
thermal analysis. Indian Journal of Chemical Technology. 15: 388
16. Prithwiraj, Jha., Kripan, Sarkar., Sudip, Barat. (2004). Effect of - 396.
Different Application Rates of Cow dung and Poultry Excreta on
Water Quality and Growth of Ornamental Carp, Cyprinus carpio 29. Zhang, MH Li H. (2011). Pore structure and chloride permeability of
vr. koi,in Concrete Tanks. Turkish Journal of Fisheries and Aquatic concrete containing nanoparticles for pavement. Construction and
Sciences. 4: 17 - 22. Building Materials. 25: 608 - 616.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

290 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Studies of Strength, Durability and Microstructural Properties of Cow Dung Ash Modified Concrete

Dr. Vinita Vishwakarma

Dr. Vinita Vishwakarma is a Prof. & Scientist - F at the Centre for Nanoscience and Nanotechnology,
Sathyabama University, Chennai. She received her Ph.D in Genetics from Ranchi University (Ranchi) in 2002
and has since published more than 50 research articles, book chapters and book in the field of genetics,
surface modifications, thin films, bio-corrosion studies, concrete corrosion, environmental technology,
nanotechnology and microbiology. Her research is focused on the developments of metals and materials,
corrosion studies and microbiology. She is handling sponsored research projects from various agencies
of Government of India. She is the members of many professional bodies such as Indian Science Congress
Association, Indian Institute of Metals, Indian Women Scientist Association and Organization for Women in
Science for the Developing World. She has also edited several proceedings books and reviewed the journal
papers.

D. Ramachandran
D. Ramachandran is a Research Scholar at Centre for Nanoscience and Nanotechnology, Sathyabama
University, Chennai. He received his masters from Annamalai University, Chidambaram in 2009 and
has published 16 research articles in the field of microbial induced concrete corrosion, In-situ High
temperature X- ray diffraction studies, Synthesis and characterization of various Nanoparticles and basic
microbiological studies. He is currently doing Ph.D in the field of “Studies on strength and durability of
biodeterioration resistance of flyash modified concrete exposed to marine environment”. He is the member
of various professional bodies such as Indian Science Congress Association (ISCA) and Indian Institute of
Metals (IIM).

Organised by
India Chapter of American Concrete Institute 291
Session 3 B - Paper 2

The Shear Strengthening of (Rc) Beams with

Textile-Reinforced Mortar
Mouza Abdullah Al-Salmi
Senior Structural Engineer, Sultanate of Oman, Royal Court Affairs, The Royal Estates, Central Design Office
Muscat City, P.O.Box:949 P.C: 100, Oman

Abstract advanced composite materials, such as textile reinforced

mortar (TRM) or fibre-reinforced polymer (FRP), will solve
The overall objective of this study was to experimentally
many of the problems caused by steel plates and offers
investigate the effectiveness of TRMs as a means of
significant advantages, such as flexibility in design, ease
increasing the shear resistance of a reinforced concrete
of installation, reduced construction time and improved
beam. The specific aims were to address the factors
durability.
affecting the shear strength and to study the shear
performance and failure modes of RC beams strengthened TRM materials are considered as an interesting alternative
with externally-bonded carbon (CTRM) sheets. To achieve to FRP, providing solutions to many problems associated
these objectives, an experimental programme consisting with the application of FRP. The uses of TRM are expected
of testing seven simply supported beams deficient in to extend across a broad spectrum of applications. (du
shear was implemented. The variables investigated in this Béton, 2001; Triantafillou and Papanicolaou, 2006).
experimental study included three different configurations
There are different techniques for RC member
of CTRM and a number of TRM layers.
strengthening, which is usually accomplished by
The experimental results conclude that the contribution of constructing external reinforced jackets. One of the
externally bonded CTRM to the shear capacity is significant most prevalent upgrading techniques is the use of FRP
and provides substantial gain in shear resistance; where jackets, which are intended to increase and enhance
this gain becomes higher as the number of textile layers the shear resistance in areas of inadequate transverse
increases. TRM can be considered as an interesting reinforcement. The use of FRP has gained substantial
alternative to FRP, providing solutions to many problems popularity in the civil engineering community. This is
associated with the application of FRP. because of the suitable properties that these materials
possess, such as extremely high strength-to-weight
ratio (making them much easier for personnel to handle
Introduction
on site), immunity to corrosion, ease and speed of
The issue of upgrading existing reinforced concrete (RC) application on site, and excellent mechanical strength
structures has become very popular over the past few and rigidity. In spite of all these advantages, however, FRP
years, both in non-seismic and seismic areas. In non- retrofitting techniques do have a few drawbacks, which
seismic areas, this is because of the deterioration that pertain to the epoxy resins used to bind or impregnate
may result from existing civil engineering infrastructure, the fibres.
such as bridge beams, columns and buildings that have
not been maintained, or which were poorly designed The main disadvantages of FRP can be summarised
and constructed in the first place. However, in seismic as: poor epoxy resin behaviour at high temperatures;
areas, seismic retrofitting becomes an essential the dependence of FRP on concrete substrata; the high
issue, particularly in regions of high earthquake risk. cost of FRP; the unsuitability of FRP for application on
(Triantafillou and Papanicolaou, 2006). wet surfaces or at low temperatures; the lack of vapour
permeability, which may lead to damage, and the danger
On several occasions, existing RC beams have been found and hazards for manual labourers. FRP materials also
to be deficient in shear and in need of strengthening. The generally behave in a linear elastic manner, with no
use of conventional strengthening techniques, such as yielding or plastic deformation, thus potentially leading
epoxy-bonded steel plates, steel jacketing and reinforced to premature disintegration and collapse (Al- Salloum
concrete (RC) jacketing in the tension zone of the external and Elsanadedy et al., 2012).
surface of RC members may be problematic. This is due
to the difficulty in manipulating heavy steel plates on However, TRM materials are considered as an interesting
construction sites, steel plate corrosion, and the limitation alternative to FRP, providing solutions to many problems
of plate length. However, replacing the steel plates with associated with the application of FRP. The uses of
TRM are expected to extend across a broad spectrum

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

292 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

of applications. For example the advantages of textile (cement-based mortar). An investigation of seven simply
materials are generally that they have very good load- supported beams which are deficient in shear (with no
bearing capacity. They also resist corrosion, have stirrups in the shear span) was carried out.
excellent ductility, can be attenuated to reduce bulk and
have lightweight components with no magnetism being
Experimental Procedure
created through bonds (Xu and Krüger et al., 2004).
Nevertheless, studies on textile reinforcement for the Experimental method
upgrading of concrete structures are limited. Research The experimental programme was carried out by testing
on the use of TRM materials began in the early 1980s, seven rectangular beams deficient in shear (with no
but development in this area proceeded slowly until stirrups in the shear span) in three-point bending. The
the late 1990s. During the past few years, the research beams were 1.677 m long and had a cross-section of
community has focused on the use of textiles to reinforce 102 x 203 mm. All the beams were tested monotonically
cement-based products, particularly in new construction and reinforced with two 10 mm diameter type S550s
(Triantafillou and Papanicolaou, 2006). longitudinal rebars at the top of the beam and 16 mm at
the bottom of the beam.
In this regard, Triantafillou and Papanicolaou (2006) found
that the replacement of organic binders with inorganic Stirrups were hand made with a diameter of 8 mm; their
ones, such as cement-based mortars, leading to the spacing was about 75 mm in the beam, except in the shear
replacement of FRP with fibre-reinforced mortar (FRMs), span, which had no stirrups. Details of the reinforcement
can be considered as one possible solution to the above are shown in Figure1
problems arising as a result of using FRPs. However,
although these materials have a comparatively long track
record in structural engineering, they are not without
their shortcomings. For example, as a consequence
of the granularity of the mortar, the penetration and
impregnation of the fibre sheet is not easy to achieve.
Moreover, mortars, unlike rain, cannot moisten individual
fibres. The behaviour of carbon fibre sheets with an
Fig. 1: Reinforcement details of simply supported beams
inorganic matrix, consisting of aluminosilicate powder
plus a water-based activator and used for the externally-
bonded flexural strengthening of reinforced concrete These beams were strengthened with different CTRP
beams or prisms, has been evaluated by Kurtz and schemes, but one of the seven beams was tested without
Balaguru (2001) and Garon and Balaguru et al. (2001), strengthening materials, as a control specimen, while the
respectively. In terms of shear resistance and rigidity, remaining six beams were strengthened using different
these materials have been found to display similar schemes applying a TRM composite. Three of these were
behaviour to epoxy- impregnated sheets, along with some strengthened with one layer and the other three, with two
reduction in ductility. Furthermore, when continuous layers.
fibre sheets are replaced by textiles, the bond condition
in the cementation composites can be amended and the Test Specimens
interaction in the fibre-matrix can be rendered tighter and The beams were grouped into three main series designated
stronger (Triantafillou and Papanicolaou, 2006). as A, B and C. The first series of tests, Series A, focused
on simply supported beams without strengthening, as a
control specimen known as AC. The second series, Series
Research Significance B, addressed the shear strengthened beam with one layer
From the limited results of TRM application obtained of three different strengthening configurations, namely:
from previous studies, it is believed that TRM jacketing fully-wrapped, U-jacketed and side-bonded specimens
is an extremely promising solution for enhancing the (BW1, BU1 and BS1, respectively). The third series of tests,
shear resistance of reinforced concrete members. Series C, was identical to Series A, but this time, with two
Thus, the application of TRM as a means of increasing layers instead of one (CW2, CU2 and CS2, respectively).
the shear resistance of reinforced concrete beams will
be investigated in this paper, while considering various The variables investigated in this research study are
parameters. the CTRM quantity (one layer versus two) and different
strengthening schemes. To summarise, the notation of
The main aim of this research and the experimental specimens is L-X, where L defines the series type and
programme was to provide a better understanding of the whether one or two layers are used and X denotes the type
effectiveness of the shear reinforcement of RC beams of jacket; F refers to ‘fully-wrapped’, U to a three-sided
offered by jackets made from continuous textile fibre jacket (U-shaped), and S indicates a two-sided bond.
(carbon) in combination with inorganic matrix materials

Organised by
India Chapter of American Concrete Institute 293
Session 3 B - Paper 2

Material Properties tensile strength was 670MPa, and the average modulus
The main four materials in this research are concrete, steel of elasticity was 200 GPa. In the case of the 10 mm bars,
reinforcement, carbon textile reinforcement and mortar. the values were 542MPa for the average yield stress, 630
Each of these has specific properties, as discussed below. MPa for the average ultimate tensile strength, and 200
GPa for the average modulus of elasticity.
Concrete
Carbon Textile Reinforcement
The seven beams were cast on four separate days. The
concrete for each casting was provided by the laboratory The textile contained equal quantities of high strength
technician in two batches. All seven beams were made carbon fibre made of long woven, knitted or unwoven
in the same concrete batches. The water: cement: sand: fibre rovings in two orthogonal directions, as shown in
gravel ratio was 0.08:0.52:0.3:0.1 by weight. A three-mix Figure 2. The density of the TRM materials in terms of
design was used prior to the experiment in order to select quantity and the spacing of the rovings in each direction
the most suitable. The concrete mix selected for the test can be controlled independently, therefore influencing
was specified as having a 28-day strength of 20.48 MPa, the mechanical properties of the textile and the degree of
maximum aggregate size of 20 mm, specified slump of penetration of the mortar which can be achieved through
100 mm, and no air entrainment. the mesh. The textile rovings in each direction were simply
placed one on top of the other and bonded on a secondary
On the day of each casting, three cube and three cylinder polypropylene grid. (Triantafillou,T.C.etal 2006).
specimens were cast. The cubes had a dimension of 150
x 150 x 150 mm3 and the cylinders had a diameter of 150 Each fibre roving was 3 mm wide and the clear spacing
mm and a height of 300 mm. The cube compression tests between each roving was 7 mm. The weight of the carbon
and the cylinder tensile strength were allowed to cure fibre in the textile was about 1.8g/cm3 and the areal weight
in the laboratory adjacent to the beam specimens. The was about 348g/m2. The mean tensile strength of the
compressive and tensile strength of the concrete was carbon was taken from the data sheet and was thus equal
determined from seven cubes and cylinders obtained to 3,800 MPa. The elastic modulus of the carbon fibres
on the day of the beam testing. The resulting average was 225Gpa and the ultimate strain was about 2% (εu =
values for the 28-day compressive strength and tensile 0.02).The thickness of the strengthening materials was
strength were 22.3 and 2.6 MPa, respectively. The obtained from:
average compressive strength was close to that which
was measured prior to the test, i.e. about 20.48 MPa.
The average compressive and tensile strengths of the
concrete for the seven different specimens applying where n is the number of layers of strengthening material,
different schemes are shown in Table 1. d is the density of the materials and w is the areal weight.
The nominal thickness of each layer was 0.09 mm (based
Table 1 on the equivalent smeared distribution of fibres).
Average compressive and tensile concrete strength

Specimens Compressive strength Tensile Strength Mortar

notation MPa MPa The textile fabric sheets
AC 21.6 2.4 were bonded to the
BS1 21.6 2.6
concrete surface using
cement paste or mortar.
BU1 23.8 2.7
A large circular steel
BF1 21.6 2.6 container was used for
CS2 22.2 2.8 mortar mixing. Moreover,
CU2 23.8 2.7
for the specimens
receiving mortar as
CF2 21.6 2.4
a binding material, a Fig. 2: The typical carbon textile
commercial inorganic used in this study
Steel Reinforcement dry binder was used. This
The longitudinal steel reinforcement bars were deformed, binder consisted of fine cement and polymers at a ratio of
hot-rolled and had high-yield strength, with a 10 mm and 8:1 by weight .The binder to water ratio was 0.23:1 by weight,
16 mm diameter. The stirrups were made from deformed resulting in a super-plasticiser that brought the mix to a
steel bars with an 8 mm diameter. Three coupons of steel plastic consistency, low shrinkage, good workability, high
bars were tested under uniaxial tension, in accordance adhesion and compatibility with the textile fabric.
with ASTM specifications. For the 16 mm bars, the
The strength of the mortar applied to the strengthened
average yield stress was 555 MPa, the average ultimate
beams was obtained through flexural and compression

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

294 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

testing using a servo-hydraulic MTS testing machine.

The mortar was prepared over two separate days and
the prisms were prepared and cured in the laboratory
until the day of testing, in conditions which were identical
to those for cubic and cylindrical tests, except for the two
days when the prisms were inside the moulds.
Flexural testing was carried out on two different days Fig. 3: A deeply inclined line applied in the shear span
on three hardened mortar prisms 40×40×160mm3 at 28
days. The average flexural and compressive strength
values were 4.16 and 27.43 MPa, respectively.

Concrete Casting and Strengthening Application

All the beams were cast simultaneously in rigid
steel moulds. Vibrators were used to ensure proper
consolidation between each lift and to prevent any Fig. 4: Applying the first TRM layer
segregation. Next, the concrete was levelled off to the
appropriate height and the surface was finished with Instrumentation
trowels. A few hours after casting, the concrete surface Different types of instrumentation were used for different
was covered with wet burlap to promote moist curing purposes, such as strain gauges, a potentiometer
and avoid shrinkage cracking. The concrete was kept and digital image correlation. The application of these
moist over a five-day period, after which time the forms instruments is discussed below.
were opened and the beams removed and left in room
temperature conditions until the date of the test. Steel Strain Gauges
Textile sheets were attached to the concrete surface For the placement of strain gauges, reinforcement bars
using manual lay-up. The installation procedure were filed and the steel surface prepared with chemicals.
employed to apply the TRM sheets can be summarised
Strain gauges were attached to the prepared steel surface.
as follows:
Furthermore, strain gauges were named according to
ll Surface preparation for the beam, such as grinding, the type of reinforcement bar. Two strain gauges were
cleaning and leavening should be carried out until any attached to the beam. The 6 mm strain gauge was attached
loose material is removed and vacuumed up. at the bottom of the specimens in the tension zone and the
three 3 mm strain gauge was attached to the top of the
ll The concrete was ground to a radius of about 10 mm
specimen in the compression zone, as clarified in Figure
at the four edges in the shear span region, which was
5. The purpose of these strain gauges was to record strain
wrapped by the textile reinforcement.
at the level of steel reinforcement during loading and to
ll Deeply inclined lines were provided in the beams’ ascertain whether or not the steel reinforcement had
shear span in order to ensure high bond quality yielded. The strain gauge is depicted in Figure 6.
between the concrete and the TRM jackets, as shown
in Figure 3.
ll The textile was cut to the required length and the
outline of the jackets marked on the specimens.
ll As a preparation, the concrete surface was water-
basted to open the pore structure. The beams were
allowed to dry prior to the TRM application.
ll A mortar layer of about 1.5-2 mm thick was applied in
the beams with a smooth metal trowel. Fig. 5: Location of the strain gauges
ll After application of the first mortar layer on the
dampened concrete surface, the textile was applied
and pressed slightly into the mortar, which protruded
through all the perforations between rovings, as
shown in Figure 4.
ll The second layer of mortar was applied and the
textile covered completely.
Fig. 6: Picture of a strain gauge

Organised by
India Chapter of American Concrete Institute 295
Session 3 B - Paper 2

Potentiometers All specimens were tested as simply supported beams

Seven potentiometers were utilised to measure vertical subjected to a three point load using a servo-hydraulic
displacements at various locations.The instrumentation MTS testing machine. A 100 kN capacity actuator testing
layout and labelling scheme are given Figure 7. Six machine was used to apply a concentrated load on a simply
potentiometers were located at the shear span of the supported control beam (AC), while a 500 kN actuator
specimen and one potentiometer was located at the testing machine was used for the other strengthening
support to record support displacement. beams. All specimens were monotonically loaded at
a displacement of 0.02 mm/sec till failure. A load cell
was mounted between the machine and the rigid beam
in order to record load during the experiment. The data
generated from the load cell, the potentiometer, and the
strain gauges were collected by a data acquisition system.
Figure 9 consists of a photograph of the test setup of the
specimen and Figure 10 (a) and (b) shows a 100 kN and
500 kN actuator, respectively.

Fig. 7: Location and layout of potentiometer Strengthening Scheme

Parameters
The applied load was measured by a load cell. The
In shear strengthening
potentiometers were connected via national instruments
situations involving a RC
linked to a computer and a schema is given for the data
beam, externally-bonded
acquisition system. In order to process the test data more
TRM reinforcement is used
effectively, readings were taken only at specified load
to wrap each scheme tested
steps. Four load readings were taken in a second. The
with one or two layers in one
readings obtained between zero and peak displacement
of several configurations.
and the actuator were stopped and held in a position to
The common method
obtain readings. This was accomplished by developing a
of textile reinforcement
data-logging programme to record the logged data as a
strengthening include side-
spreadsheet. Furthermore, the programme also allowed
Fig. 8: The Specimen frame bonding (S) in which TRMs
real-time data visualisation for several potentiometers/
gauges and provided a load displacement plot during the
test.

Digital Camera Correlation

The digital 3D image correlation (DIC) system shown in
Figure 8 is instrumented with acoustic emission sensors,
in order to assess the evaluation of damage during
test loading, as well as providing a three-dimensional
measurement. Furthermore, this system was used to
measure the shape and displacements of shear cracking.
In addition, it provides a clear picture of crack patterns
in each segment and the width of shear cracks can be
measured using such instruments. Moreover, load and
displacement values may be taken using this instrument.

Fig. 10: (a) A 100 and (b) A 500 actuator

are merely bonded to the sides of a beam, as shown in

Figure 11; (U) U-jacketing in which TRMs are bonded to
both sides and the tension face, as shown in Figure 12, and
fully-wrapped (F), in which TRMs are wrapped around the
entire cross-section of the beam, as shown in Figure 13.

Experimental Result
The results obtained from testing seven RC beams were
reported. Furthermore, for each test measurement of
Fig. 8: Digital Image correlation

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

296 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

Figure 15 shows the load versus the mid-span displacement

of the AC specimens. It is clear from the graph that the
maximum load was 51.83 kN. An interesting observation
during the test was that there was no sudden drop in the
load recorded, as is clear from the below Figure. This
means that shear failure is a ductile failure, because
some resistance was provided in the tension zone due to
the contribution of the dowel action activated by the two 16
Fig. 11: Side jackets
mm diameter longitudinal rebars. The crack width of the
control beam calculated using data from DIC was about
0.9803mm at the failure load and increased to 8.467mm
at the final load.

Fig. 12: U- shaped jacket

Fig. 15: Load versus mid-span displacement

Moreover, Figure 16 shows load versus strain for both

the 10 mm diameter bar of the top longitudinal rebar in
the compression zone and the 16 mm diameter bar of the
Fig. 13: Fully-wrapped jacket bottom longitudinal rebar in the tension zone. As is clear
from the figures, the strain at the support is near zero
deflection, reinforcement strain gauges and the use of and then starts to increase steadily further away from
digital image correlation were implemented. the support until it reaches maximum load. In addition,
neither reinforcement yielded since they did not exceed
Test Result of Control Beam (AC) the yield point, which was is 2750 με.
In the first test of the control beam (AC), diagonal shear
cracks started to form close to the middle of the shear
span with a load of 45 kN. The first shear crack was the
critical crack in the specimen and as the load increased,
the diagonal cracks increased. The ultimate load was
51.83 kN. The crack started to widen and was propagated
near the support as the load increased. This was expected
since there were no stirrups in the shear span. Figure 14
shows the crack pattern of the control beam at ultimate
load.

Fig. 14: Failure of the (AC) Fig. 16: Load versus strain (a) 16 mm and (b) 10 mm diameters

Organised by
India Chapter of American Concrete Institute 297
Session 3 B - Paper 2

Test Result of BS1 maximum load was 57 kN. Then, the load started to drop
The side-bonded beam with one layer (BS1) failed in shear, gradually (ductile behaviour). This was due to the shear
as expected, through the formation of diagonal shear resistance provided by both the TRM and the dowel action
cracks close to the middle of the shear span, as shown in (activated by two 16 mm diameter longitudinal rebar).
Figure 17. The initial shear crack started to appear when the
load was 54.42 kN. The ultimate load was approximately
57 kN. An interesting observation during this test was
that the beam cracking was clearly visible on the mortar-
based jacket and there was no debonding failure between
the TRM and the concrete. Neither was there a sudden
drop recorded in the load following the formation of the
diagonal crack. This means it was a ductile failure which
may be attributed to the considerable contribution to
shear resistance provided by both the TRM and the good
bonding between the concrete and the TRM. Furthermore, Fig. 19: Load versus mid-span displacement
a local failure occurred due to textile slippage from the
mortar because the bonding and interlocking mechanism Furthermore, Figure 20 shows load versus strain for both
between the mortar and the textile was not strong enough the 10 mm diameter bar and the 16 mm diameter bar, so
to hold the two materials together as well as the mortar it is clear from the Figures that neither reinforcement
could not impregnate well inside the textile , as can yielded.
be seen from Figure 18. An interesting feature in this
specimen was that the rupturing of the roving fibres in
the mortar-based jacket was gradual, starting with a few
fibre bundles being pulled and propagated towards the
neighbouring fibres, demonstrated in Figure 18.
The load versus the mid-span displacement curve is
presented in Figure 19. From the graph it is clear that the

Fig. 17: The formation of diagonal shear cracks in BS1

Fig. 20: Load versus strain for (a) 16 mm and (b) 10 mm diameters

Test Result of BU1

In the case of the U-jacketing of one layer, a shear failure
occurred in the middle of the shear span and there
was a small debonding in the top left of the beam, as
shown in Figure 21 (a). This was due to a very good bond
between the TRM and the concrete. The ultimate load
was approximately 79.8 kN. Moreover, during this test, a
diagonal shear crack appeared in both sides of the beam,
as shown in Figures 21 (b) and (c).
An interesting feature of this beam is that the rupture
Fig. 18: Textile slippage and rupture
of the fibres in the mortar-based jacket was gradual,

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

298 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

Fig. 23: Load versus mid-span displacement

Fig. 21: U-jacketing failure: (a) top view, (b) front view and (c)
back view of beam

starting with a few fibre bundles and propagating slowly

Fig. 24: Load versus strain for (a) 16 mm and (b) 10 mm diameters
into the neighbouring fibres. The beam cracking was
clearly visible on the mortar-based jacket, as shown in
Figures 22. Test Results of BW1
The fully-wrapped beam with one layer failed in shear
The load versus the mid-span displacement curve it
through the formation of diagonal shear cracks in both
presented is shown in Figure 23. From the graph, it is
sides of the beam, as shown in Figure 25. In addition, the
clear that the maximum load was 79.05 kN. Then, the
rupturing of the roving fibres in the mortar-based jacket
load started to drop gradually, which indicates ductile
occurred as the BS1 and BU1 specimens.
behaviour.
Furthermore, Figure 24 shows load versus strain for both
the 10 mm diameter and 16 mm diameter bars and as is
clear from the Figures, neither reinforcement yielded.

Fig. 25: (a) Shear crack at the front of the beam, and (b) shear
Fig. 22: Fibre ruptures crack at the back of the beam

Organised by
India Chapter of American Concrete Institute 299
Session 3 B - Paper 2

The load versus the mid-span displacement curve it Test Result of CS2
presented is shown in Figure 26. From the graph, it is In the case of a side-strengthened beam with two layers of
clear that the maximum load was 112.82 kN. After the TRM, the specimen failed in shear. A debonding from the
maximum load had been reached, the load started to drop concrete cover occurred on both sides, as is clear in Figure
gradually, as with the previous specimens. 29 (a) and (b). Furthermore, a shear crack appeared in the
Load and y displacement under the point load were other side of the beam through the formation of diagonal
taken from the DIC system. Figure 27 shows load versus cracks in the shear span. However, there was no slippage
displacement. It can be seen that the two Figures have the in the fibre roving, as there was with the BS1. This is due to
same pattern, except for some fluctuation in Figure 27. the high bonding of the two layers, which can lead to a new
This was due to some noise during the experiment.in the pattern as a result of the overlapping of the layers, which
digital image correlation. improves and increases the bonding and interlocking
of textiles and mortar. The ultimate load was 90.05 kN.
Furthermore, Figure 28 shows load versus strain for Figure 29 (c) shows a view of the beam from the bottom. It
both the 10 mm diameter and 16 mm diameter bars; is clear that many cracks appear under the tension.
it is therefore clear from the Figures that neither
reinforcement yielded.

Fig. 26: Loads versus mid-span displacement curve

Fig. 27: Load versus mid-span displacement curve by DIC

Fig. 29: Failure of CS2: (a) side view, (b) top view and (c) bottom
view

The load versus the mid-span displacement curve is

presented as shown in Figure 30, both from the data
produced by Potentiometer Seven under the point load of
the beam and the data from the digital image correlation.
From the graph, it is clear that both curves had the same
trend and the maximum load was 90.05 kN. After reaching
maximum load, the load started to drop gradually, as with
the previous specimens.
Fig. 28: Load versus strain for (a) 16 mm and (b) 10 mm diameters

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

300 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

Fig. 33: Shear crack on the other side of the beam

Fig. 30: Load versus mid-span displacement curve
The load versus the mid-span displacement curve
is presented in Figure 34, with data produced by
Potentiometer Seven under the point load of the beam
and data from digital image correlation. From the graph,
it is clear that both curves had the same trend and the
maximum load was 120.99 kN. After the maximum load
had been reached, the load started to drop. Furthermore,
Figure 35 shows load versus strain for both the 10 mm and
16 mm diameter bars. As is clear from the Figures, the
values of both the top and bottom reinforcement strains
are less than the ultimate strain value, which indicates
that both reinforcements are not yielded.

Fig. 31: Load versus strain for (a) 16 mm and (b) 10 mm diameters
Fig. 34: Load versus mid-span displacement curve
Furthermore, Figure 31 shows load versus strain for the
10 mm and 16 mm diameter bars. As is clear from the
Figures, neither reinforcement yielded.

Test Result of CU2

A debonding failure occurred in the case of the two-layer
U-jacket beam. In addition, part of the TRM removed
concrete, as shown in Figure 32. The ultimate load was
120.99 kN. Furthermore, a shear appeared in the other
side of the beam through the formation of diagonal cracks
in the shear spans, as shown in Figure 33.

Fig. 32: U-jacket failure mode Fig. 35: Load versus strain for (a) 16 mm and (b) 10 mm diameters

Organised by
India Chapter of American Concrete Institute 301
Session 3 B - Paper 2

Test Result of CW2

The behaviour of the fully-wrapped beam indicated that
the shear failure was suppressed and that failure was
controlled by flexure. In addition, minor cracks appeared
in the TRM in both sides of the beam, shown in Figures
36. These cracks started to appear when the load was
approximately 101.26 kN. Figure 37shows the buckling
of the top reinforcement. The ultimate load was 152.11
kN. Furthermore, both the compression and tension
reinforcement yielded. As is clear from Figure 38, the
value of the strain at 16 mm was about 2900 με and for 10
Fig. 39: Load versus mid-span displacement curve
mm, was approximately 3530 με.
The load versus the mid-span displacement curve
it presented is shown in Figure 39, with data from
Potentiometer Seven under the point load of the beam
and data from digital image correlation. From the graph,
it is clear that both curves had the same trend and the
maximum load was 152.11 kN. After the maximum load, the
load started to drop gradually, indicating ductile behaviour.
This is due to the contribution to shear resistance provided
by both fully- wrapped jackets of the TRM.
Fig. 36: Failure of CW2
Test Results Evaluation and Discussion
The load versus the mid-span displacement curve it
presented for all specimens is shown in Figure 40. From
the graph it is clear there was no sudden drop in the load
after it reached ultimate load. Moreover, the load started
to drop gradually, which indicates that all specimens
displayed ductile behaviour. This is due to the shear
resistance provided by both the TRM and the dowel action
(activated by two 16 mm diameter longitudinal rebar).The
Fig. 37: Top reinforcement buckling maximum loads in specimens AC, BS1, BU1, BW1, CS2, CU2
and CW2 were 51.83 kN, 57 kN, 79.8 kN, 112.82 kN, 90.05
kN, 120.99 kN and 152.1 kN, respectively. Furthermore, it
can be clearly seen from the Figures that all the beams
which underwent shear failure had the same curve trend,
except for the last specimen, which was fully-wrapped
with two layers that failed in flexure. This brings us to the
conclusion that as the number of TRM layers increased,
the effect of the number of layers on the shear strength
provided by the textile fibre was enhanced, as well as the
mortar thickness increasing while the number of TRM
layers increased, having an effect on the shear, but this
failed in flexure, whereas the other specimens failed in
shear.
The shear load carrying capacity of the beam can be
noticeably increased by applying TRM strengthening
layers. This means that the contribution of TRM to shear
capacity increases as the number of layers increase. This
can be seen very clearly, especially in the case of side-
bonding and U-jackets, where there is a huge difference
in the value of the TRM contribution, with one layer being
compared to two layers, as can be seen in Table 2. Thus,
a parameter of increasing layers works effectively and
gives a good result.
Fig. 38: Loads versus strain for (a) 16 mm and (b) 10 mm diameters

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

302 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

the test results which include the number of layers (one

versus two), the applied load, and the different TRM
configurations and failure modes.

Table 2
Summary of TRM contribution to shear capacity (one layer
compered to two layers)

V fo V for V for
Number
fully-wrapped U-jacketing side-bonding
of layers
Fig. 40: Load versus mid-span displacement curve for all (kN) (kN) (kN)
specimens
The control beam AC failed in shear, as expected. All the One layer 60.82 27.8 5
strengthening beams failed in shear and this was evident
from the diagonal cracking in the shear, except for the CW2
beam, which underwent a flexural yield failure. The CS2 Two layers 100.11 68.99 38.05
and SU2 beams were controlled by debonding beside the
shear crack. From the results of the experiment, it shows
that debonding failure is not likely to dominate in the case Summary, Conclusions and Future
of fully-wrapped beams. This is because the shear failure Recommendations
can be suppressed in the case of full wrapping. The performance of strengthening materials, such
An interesting feature for BS1 and BU1 specimens is that as TRM and FRP retrofits for shear strengthening
the rupturing of the fibre roving in the mortar-based RC members was examined through a combination
jacket was gradual, starting with a few fibre bundles and of experimental and analytical work. From previous
propagating slowly into the neighbouring fibres, as well as studies on the topic of TRM shear-strengthened beams
local failure occurring due to textile slippage in the mortar. revealed that the majority of experimental tests were
This happened because the bonding and interlocking carried out using small-scale specimens, whether
mechanism between the mortar and the textile were not under monotonic or cyclic loading conditions. Therefore,
strong enough to hold the two materials together. further research was warranted regarding the TRM
retrofitting of critical larger shear RC beams. Moreover,
On the basis of the test result, it may be proved that the studies on textile reinforcement in the upgrading of
TRM contribution to shear capacity increased as the concrete structures are limited. However, from these
number of layers increased .In addition, the contribution of relatively scant studies, it is evident that TRM jacketing
the fully-wrapped configuration to shear capacity is higher is considered as an extremely promising solution for
than that of U-shapes or side-bonding and U-jackets are increasing the shear capacity of reinforced concrete
more effective than side-bonding. Moreover, regardless members. In fact, the use of TRMs in the retrofitting of
of the strengthening scheme involved, two TRM layers RC structures reveals their effectiveness as a means
perform better than a single layer. Table 3 summarises of increasing the shear resistance of RC members. The

Table 3
Summary of experiment test results

Specimen Total applied shear force at Contribution of TRM to the

Test No. TRM layers &schemes Failure mode
designation ultimate load (kN) shear capacity (kN)

1 AC - Shear 52 -

2 BS1 One layer, side jacket Shear 57 5

3 BU1 One layer, U-wrap Shear (debonding) 79.8 27.8

4 BW1 One layer, fully- wrapped Shear 112.82 60.82

5 CS2 Two layers, side jacket Shear (debonding) 90.05 38.05

6 CU2 Two layers, U-wrap Shear (debonding) 120.99 68.99

7 CW2 Two layers, fully- wrapped Flexural 152.11 100.11

Organised by
India Chapter of American Concrete Institute 303
Session 3 B - Paper 2

effectiveness of these composite materials has been ll The test results show that regardless of the TRM
investigated both experimentally and numerically by a strengthening configuration, two layers of textile
number of researchers. reinforcement with mortar binder perform better than
one layer.
The main objectives of this research study were to
investigate the effectiveness of external TRM shear ll The contribution of externally-bonded CTRM
reinforcement and the contribution of these materials to reinforcement to shear capacity is influenced by layers
the shear capacity of RC beams, as well as studying the and it appears to increase as the layers are increased.
failure modes of RC beams strengthened with externally-
ll The test results show that applying TRM to the beam
bonded CTRM sheets. Moreover, factors influencing shear
sides only is less effective than a U-wrap configuration.
strength were addressed, such as the number of layers
and different strengthening schemes, as well as the use ll The fully-wrapped schemes make a higher contribution
of various design models for computing the contribution to the shear capacity, compared to the other schemes.
to the shear capacity of the strengthened beams. A
ll The experimental test results for all the specimens
comparison was also drawn between the results of the
show that after the beam reaches maximum load, no
models and the current experiment.
sudden drop is recorded in the load. This is attributed
In order to fulfil these objectives, an experimental to the considerable contribution to shear resistance by
programme consisting of tests on seven simply supported both TRM strengthening materials and the dowel action
rectangular RC beams. The beam specimens were activated by 16 mm rebar diameter in the tension zone.
grouped into three main series. The first series tested
ll The experimental test results show that debonding
the beam without applying any strengthening materials.
failure is not likely to dominate in the case of full
This represented the control beam. The second series
wrapping.
investigated the capability of one CTRM layer to enhance
the shear capacity of the RC beams. The third series was ll The test results show that the strain gauge only yields
identical to the second, but applied two layers (instead in the case of TRM strengthening with two layers of full
of one). wrapping.
The variables investigated in this experimental study ll Existing evidence from the experiment results clearly
included three different CTRM configurations, different indicate that TRM jackets have a high level of bonding
numbers of TRM layers (one versus two) to investigate between the textiles and the concrete.
the shear performance and failure modes of RC beams Based on the findings and conclusion of the current study,
strengthened with externally-bonded carbon TRM (CTRM) it is to be believed that further work and more investigation
sheets. are needed to obtain a better understanding of textile
In order to achieve the study goals, three different types of reinforced mortar behaviour. Therefore, new parameters
instrumentation were used for different purposes, such as should be considered in future, like the interaction between
strain gauges, potentiometers and digital image correlation the contribution of external TRM and internal steel shear
to investigate the strengthening effectiveness of TRM in
Based on the experimental and analytical results, the
long shear spans and for this, additional specimens need
following conclusions have been drawn:
to be tested and explored using different numbers of TRM
ll TRM materials can be considered as an interesting layers and configurations to create a large database and
alternative to FRP, providing solutions to many testing the beams with a range of TRM orientation.
problems associated with the application of FRP. The
Because of the limited results obtained in this and previous
uses of TRM are therefore expected to be extensive.
studies, and due to differences between the experimental
ll Externally bonded TRM reinforcement can be used to loading conditions and those modelled analytically, more
enhance the shear capacity of RC beams. research is needed to develop an analytical model to predict
the shear behaviour and failure mode of RC members
ll The experimental test results confirm that the TRM
strengthened with externally-bonded TRM composite
system strengthening technique is applicable and can
materials. The effect of different parameters on the overall
increase the shear resistance of reinforced concrete
behaviour of the RC members must also be evaluated.
members.
It is also highly recommended for future experimental
ll Existing evidence clearly indicates that TRM jackets
work that all shear reinforcement should be
provide substantial gain in shear resistance, and this
systematically instrumented to capture the full variation
gain is higher as the number of layers increase.
of steel reinforcement and TRM strain throughout the
ll • The test results show that one layer of textile RC members. This is because a complete set of strain
reinforcement is less effective than two layers but still information can be allowed for the accurate estimation of
sufficient for providing substantial shear resistance. steel and TRM shear strength capacity.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

304 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 The Shear Strengthening of (Rc) Beams with Textile-Reinforced Mortar

Acknowledgements References and Bibliography

I would like to express my deep gratitude to Dr Dionysios 1. Al-Salloum, Y.A. et al (2012). "Experimental and numerical study
for the shear strengthening of reinforced concrete beams using
Bournas for the great help and close supervision he has textile-reinforced mortar." Journal of Composites for Construction
afforded me in carrying out this work. 16(1): 74-90.

Sincere thanks are also extended to the technical staff of 2. Çavdar, A. (2012). "A study on the effects of high temperature on
mechanical properties of fiber reinforced cementitious composites."
Nottingham University for their genuine assistance in the Composites Part B: Engineering 43(5): 2452-2463.
laboratory work.
3. Chen, J. and Teng, J. (2003). "Shear capacity of fiber-reinforced
Finally, I would like to thank my husband for his continuous polymer-strengthened reinforced concrete beams: Fiber reinforced
support, encouragement and patience. polymer rupture." Journal of structural engineering 129(5): 615-625.
4. Curbach, M. and Zastrau, B. (1999). "Textilbewehrter Beton—
Aspekte aus Theorie und Praxis." AA Balkema, Potterdam.
5. du Béton, F.I. (2001). "Technical Report Bulletin 14." Externally
Notations Bonded FRP Reinforcement for RC Structures.
A sw: area of steel shear reinforcement within a distance 6. Garon, R. et al (2001). "Performance of inorganic polymer-fiber
b: width of the beam in cross-section composites for strengthening and rehabilitation of concrete beam."
Thomas Telford, London, 1: 53-62.
bw: minimum width of the cross-section over the effective depth
d: depth from the top of the section to the tension steel reinforcement 7. Khalifa, A. et al (1998). "Contribution of externally bonded FRP to
shear capacity of RC flexural members." Journal of Composites for
df: effective depth of the CFRP shear reinforcement
Construction 2(4): 195-202.
Ef: elastic modulus of the FRP in the principal fibre direction (GPa)
8. Khalifa, A.M. (1999). "Shear perfrmance of reinforced concrete
(Efρf): limiting value of product Efρf
beams strenghened with advanced composite." PhD Thesis,
fc: concrete compressive strength in MP University of Missouri-Rolla ,Misdouri united states.
fcd: design value of compressive strength of the concrete cylinder
9. Kim, G. et al. (2008). "Shear strength of strengthened RC beams
t: thickness of the strengthening material with FRPs in shear." Construction and Building Materials 22(6):
t f: thickness of the FRP on each side of the RC element 1261-1270.
Le: effective bond length (mm) 10. Kurtz, S. and Balaguru, P. (2001). "Comparison of inorganic and
P: applied load organic matrices for strengthening of RC beams with carbon
R: reduction coefficient (ratio of the average effective stress or strain in sheets." Journal of structural engineering 127(1): 35-42.
the FRP sheet to its ultimate strength or elongation) 11. Larrinaga, P. (2011). "Flexural strengthening of low grade concrete
s: spacing of steel stirrups through the use of new cement- based composite materials." PhD
sf: spacing of FRP strips Thesis, University of the Basque Country, Spain.
t f: thickness of FRP sheet on one side of the beam (mm) 12. Larrinaga, P. et al (2013). "Non-linear analytical model of composites
Vexp: experimentally derived shear capacity provided by the FRP based on basalt textile reinforced mortar under uniaxial tension."
Composites Part B: Engineering 55: 518-527.
Vc: nominal shear strength provided by concrete
V f: nominal shear strength provided by the FRP shear reinforcement 13. Melek, M. and Wallace, J.W. (2004). "Cyclic behavior of columns
with short lap splices." ACI Structural Journal 101(6).
Vn: nominal shear strength (ACI format)
Vs: nominal shear strength provided by the steel shear 14. Monti, G. and Liotta, M.A. (2007). "Tests and design equations for
reinforcement FRP-strengthening in shear." Construction and Building Materials
21(4): 799-809.
w f: width of the FRP strip
w fe: effective width of the FRP sheet (mm) 15. Nammur, G., Jr and Naaman, A.E. (1989). "Bond stress model for
fiber reinforced concrete based on bond stress-slip relationship."
α: angle between the inclined stirrups and longitudinal axis of the
ACI Materials Journal 86(1).
member
β: angle between the principal fibre orientation and the longitudinal 16. Ohno, S. and Hannant, D. (1994). "Modeling the Stress-Strain
axis of the beam Response of Continuous Fber Reinforced Cement Composites."
ACI Materials Journal 91(3).
εfe: effective strain of FRP
εfu:ultimate tensile elongation of the fibre material in the FRP 17. Ožbolt, J. et al (2001). "Microplane model for concrete with relaxed
composite kinematic constraint." International Journal of Solids and Structures
εmax: limiting value of the characteristic effective FRP strain
38(16): 2683-2711.
γc: partial safety factor of concrete 18. Peled, A. and Bentur, A. (2000). "Geometrical characteristics and
efficiency of textile fabrics for reinforcing cement composites."
γf: partial safety factor of the FRP
Cement and concrete research 30(5): 781-790.
ρf: FRP fraction area = (2tf / bw) (wf / sf)
19. Peled, A. et al (1994). "Woven fabric reinforcement of cement
Acronyms and Abbreviations matrix." Advanced Cement Based Materials 1(5): 216-223.

CTRM: Carbon textile-reinforced mortar 20. Reinhardt, H. and Krüger, M. (2001). Vorgespannte dünne
Platten aus Textilbeton. Proc., Textilbeton-l. Fac-kolloquium der
DIC: Digital image correlation
Sonderforschungsbereiche.
FRP: Fibre-reinforced polymer
21. Si Larbi, A. et al (2010). "Shear strengthening of RC beams with
TRM: Textile-reinforced mortar
textile reinforced concrete (TRC) plate." Construction and Building
RC: Reinforced concrete Materials 24(10): 1928-1936.

Organised by
India Chapter of American Concrete Institute 305
Session 3 B - Paper 2

22. Täljsten, B. (1997). "Strengthening of beams by plate bonding." 26. Triantafillou, T.C. and Papanicolaou, C.G. (2006). "Shear
Journal of Materials in Civil Engineering 9(4): 206-212. strengthening of reinforced concrete members with textile
reinforced mortar (TRM) jackets." Materials and structures 39(1):
23. Teng, J.G. et al. (2002). "FRP: strengthened RC structures."
93-103.
Frontiers in Physics 1.
27. Triantafillou, T.C. et al. (2006). "Concrete confinement with textile-
24. Triantafillou, T.C. (1998). "Shear strengthening of reinforced concrete
reinforced mortar jackets." ACI Structural Journal 103(1).
beams using epoxy-bonded FRP composites." ACI Structural
Journal 95(2). 28. Xu, S. et al. (2004). "Bond characteristics of carbon, alkali resistant
glass, and aramid textiles in mortar." Journal of Materials in Civil
25. Triantafillou, T.C. and Antonopoulos, C.P. (2000). "Design of concrete
Engineering 16(4): 356-364.
flexural members strengthened in shear with FRP." Journal of
Composites for Construction 4(4): 198-205.

Mouza Abdullah Al-Salmi

This Moza from Royal Court Affairs-Muscat-Oman.
Moza Abdullah Al-Salmi, Senior Structural Engineer, Royal Court Affairs, The Royal Estates,
Central Design Office, Muscat City, P.O.Box:249 P.C: 100, Sultanate of Oman

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

306 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Properties of Lightweight Dry-Mix Shotcrete for Mine Insulation

Thermal Properties of Lightweight Dry-Mix Shotcrete for

Mine Insulation
Vivek Bindiganavile, Victor Liu and Derek Apel
Department of Civil & Environmental Engineering, University of Alberta, Edmonton, Canada, T6G 2W2

Abstract Shotcrete may be produced using the so-called dry-mix

or the wet-mix techniques and, both of them are routinely
This paper describes the thermal properties of lightweight
used in tunnels and mines. Note that in the dry-mix
dry-mix shotcrete using expanded perlite aggregate
process, the bone- dry ingredients – a mixture of cement,
(EPA). Mixes made with different EPA/sand ratios were
aggregates and admixture – are pneumatically sprayed
sprayed through the dry-mix shotcreting technique
towards the target substrate, at which point the mix water
onto wooden molds to produce panels for mechanical
is introduced under pressure from another hose. Given
and thermal testing. The density, uniaxial compressive
the brief mixing time and the high momentum when
strength (UCS) and splitting tensile strength (STS) were
applied, dry-mix shotcrete differs from conventionally
measured at various ages. Further, the ISO approved
cast concrete both in its rheology and in its hardened
Transient Plane Source (TPS) technique was employed to
properties[9]. Here, using lightweight aggregates has
measure the thermal properties at 28 days. The results
been shown to favour both shootability and pumpability[10].
illustrate that shotcrete mixes with EPA have similar UCS
Whereas shotcrete has been extensively characterized
and superior STS compared to cast concrete. Adding
for its mechanical performance as well as early age
EPA led to a drop in thermal conductivity and diffusivity.
characteristics[7, 11], there is very limited understanding
When compared with cast concrete of equal EPA content,
of its thermal properties, especially as resulting from
shotcrete had lower specific heat capacity. This study
lightweight inclusions [12].
found dry-mix shotcrete incorporating EPA at up to 75%
sand substitution by volume, as a mechanically viable and In this study, expanded perlite aggregate (EPA) was
thermally resistant alternative to cast concrete containing incorporated as a lightweight substitute for sand. Raw
regular aggregates. perlite occurs in nature as a siliceous volcanic rock, which
contains 2-5% water [13]. Heating at over 870 °C causes a
Keywords: expanded perlite; mechanical properties;
4 to 20 fold expansion[14], which results in the lightweight,
shotcrete; thermal characterization.
porous, expanded perlite. When expanded perlite is
used in lieu of regular aggregates in conventionally
Introduction cast systems, many benefits accrue including lower
The demand for minerals is continuously increasing the density, lower thermal conductivity, acoustic insulation
world over. Needless to say, it takes mining ever deeper. One and shrinkage resistance[15]. Such mixtures have been
therefore must contend with increasing temperatures in applied in tiles, stucco, brick/block masonry, precast
the working area due to the geothermal heat trapped in the products, roof fill, pipe coating, oil-gas and geothermal
surrounding rocks. A direct consequence is an escalating wells, etc. The mechanical and thermal advantage to cast
energy cost arising from cooling the working environment concrete through incorporating expanded perlite are well
for the miner’s comfort[1]. A structural lightweight thermal known[16-19]. As stated earlier, the U.S. Bureau of Mines
insulation that satisfies both mechanical and thermal (USBM) sprayed an insulation of shotcrete containing
requirement is therefore, attractive [2-4]. expanded perlite and achieved a thermal conductivity of
0.36 W/(m·K) along with a 90-day uniaxial compressive
Shotcrete refers to a cement-based mixture that is strength (UCS) of 20 MPa[12]. The present authors here
projected pneumatically at high velocity towards the examine this promising research track, which was
target surface[5]. Further, it is stipulated that it must be apparently discontinued following the disintegration of
compacted by its own momentum[6]. This technique has USBM in 1995.
been used in a wide range of applications in construction
and mining industries. The latter has now become a major
consumer of shotcrete especially for use in underground Research Significance
rock support[7], and has been used in several projects Lightweight cement-based liners promise low thermal
across India[8]. conductivity, so essential to ensure an energy-efficient and

Organised by
India Chapter of American Concrete Institute 307
Session 3 B - Paper 3

comfortable work environment. Whereas in underground and contrasts dry-mix shotcrete with conventional cast
applications, the spraying technique is attractive, concrete for thermo-mechanical performance.
the process dependence of the strength and thermal
constants of the resulting lightweight sprayed concrete
Experimental Details
has not been well understood. Using expanded perlite as
a lightweight inclusion in lieu of sand, this study compares Materials and Mixtures
The expanded perlite aggregate (EPA), Type GU Portland
cement[20] and sand were locally sourced in Edmonton.
The EPA was mainly composed of SiO2 (70-75%) and Al2O3
(12– 18%). As shown in Figure 1, it had a porous structure
and with a bulk density of 71kg/m3 in oven-dry conditions,
it was rated to absorb water at 100% of its dry mass. The
sand was at SSD condition, with a moisture content of 2.04
% on oven-dry mass. It had a bulk density of 1675 kg/m3 in
oven-dry conditions. As shown in Table 1, the mixes were
designed in accordance with ACI 506.5R-09[21]. The sieve
analyses of the EPA and the sand used along with their
blends were conducted as per ASTM C 126[22] by means
of a mechanical shaker. As plotted in Figure 2, the grain
size distribution for the aggregate blends in all the mixes
Fig. 1: Scanning Electron Micrograph of Expanded Perlite

Fig. 2: Grain Size Distribution of Fine Aggregate Blends in a) SP0 and SP 100; (b) SP25; (c) SP50; (d) SP75

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

308 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Properties of Lightweight Dry-Mix Shotcrete for Mine Insulation

and all five mixes were shot on the same day, inside a
Table 1
Mix Proportions of Shotcrete and Corresponding Rebound specially fabricated hut. A plastic sheet was placed on
the floor to collect the rebound after each shooting. As
Mix
Replacement
Cement
Sand EPA
(%)
shown in Figure 3, the molds were propped on the hut’s
percentage# (kg/m3) (kg/m3) facing wall at an angle of 45°. After shotcreting, all panels
number (kg/m3) Rebound
(%) (OD) (OD)
were immediately finished and covered with plastic wraps
SP0* 0 520 1624 0.00 21 and then transported to the curing room that was set to a
temperature of 25 ± 2°C and relative humidity between 95
SP25 25 520 1218 17 11 - 100 %. The specimens were obtained by coring cylinders
SP50 50 520 812 35 17
from each panel at 6 stages of curing namely, after 1 day,
3 days, 5 days, 7 days, 14 days and 28 days. The cores (50
SP75 75 520 405.0 52 15 mm in diameter) were designated as follows: three cored
cylinders were examined for density as well as the volume
SP100 100 520 0.00 69 29
of permeable voids as per ASTM C642[25]; three cylinders
#
EPA substitution of sand was by volume. *SP0 was pre-mixed by drum mixer were tested per ASTM C42[23] to establish compressive
one day before the shooting process to tighten schedule, but the mix started to strength and a further four cylinders were tested to
hydrate, resulting in hardened blocks. However it was still shot to obtain the
rebound. evaluate the splitting tensile strength in accordance with
ASTM D3967[26]. Thus, at each stage 10 cylinders were
cored out for each shotcrete mix. In addition, three cored
was in each case, within the shotcrete grading zone No.1 cylinders were slated for thermal testing at the end of
given in ACI 506R-05[23]. Besides the reference mix with the 28th day. The drilled cores were end ground and this
no EPA, four other mixes were produced where sand was resulted in a variation in the Length/Diameter ratio (L/D)
replaced with EPA at four levels of volumetric substitution for the shotcrete cylinders between 1 and 1.78. Only those
namely, 25, 50, 75, and 100%. First, a series was cast in the cores that conformed to Grade 1 and Grade 2 as specified
conventional manner and these five mixes were designated in ACI 506.2-95[27] were accepted for further testing in this
CP0, CP25, CP50, CP75 and CP100, respectively. They study.
have been described separately in an earlier report[24].
A second series was sprayed using the dry-mix process Testing methods
and the resulting five mixes were designated SP0, SP25,
SP50, SP75 and SP100, respectively. Mechanical Evaluation
The density and water absorption in the shotcrete
Shotcreting process and sample preparation specimens were determined according to ASTM C 642[25].
The dry-mix shotcrete was prepared with the assistance The mechanical properties namely, the unconfined
of a local industrial facility, and the shotcreting process compressive strength (UCS) was evaluated per ASTM
was supervised by a certified nozzleman. In all, 15 panels C42[28], as illustrated in Figure 4a. As stated earlier, the
were prepared upon spraying on to wooden moulds splitting tensile strength (STS) was evaluated per ASTM
610 mm (width) × 610 mm (length) × 90 mm (depth) in D3967[26].
dimension. The SP0 batch was pre-mixed in the late
afternoon on the day before the shooting, whereas the Thermal Evaluation
other 4 mixes were pre-mixed early in the morning of A thermal constants analyser that is based on the
the shoot. Three specimens were prepared for each mix transient plane source method (TPS) and conforms to
ISO 22007-2.2[30] was employed, as illustrated in Figure
4c. This technique was developed by Gustaffson[31] during
the 1990s and it is being used widely to characterize the
thermal properties of construction materials[32]. It is
found to yield rapidly and simultaneously, information

(a) (b) (c)

Fig. 4: Test in Progress: (a) Compression of Cored Shotcrete

Cylinder; b) Splitting Tensile Test; c) Thermal Evaluation via
Fig. 3: Shotcreting into Panels Propped Against the Wall at 45° TPS Method

Organised by
India Chapter of American Concrete Institute 309
Session 3 B - Paper 3

on the thermal conductivity and thermal diffusivity [32-36].

The key component of the TPS device is the combined
heat source-and-temperature sensor (shown in Figure
4d), which contains the electrically conductive Nickel
bifilar spirals, sandwiched by two thin Kapton sheets.
This Kapton probe of radius, r, equal to 6.4 mm is shown
sandwiched between two cylindrical disks each of
whose thickness is 15 mm, in Figure 4c. The disks were
sawn out from the shotcrete cylinders, such that each
cylinder yielded one pair of disks. This test requires that
the penetration depth, i.e. the minimum distance into
the specimen at which there is no temperature change,
does not exceed the sample’s thickness. For the samples
examined here, the calibrated constants were found
as follows: for an output power, Po, equal to 0.1 W, the
probe detected 200 points for the average temperature
increase (∆T), within an interval, t, of 20 s. Under this
condition, the penetration depth was around 9 mm,
which was less than the sample’s thickness of 15 mm.
An illustrative example is shown in Figure 5, using a test
on samples from shotcrete mix SP25. It depicts the data
output from the TPS device and its treatment. Figure
5a shows the time-history for the average transient
temperature increase (∆T). Figure 5b shows the linear
plot for the function ̅ versus D(τ), where the latter is
a known dimensionless specific time function[31]. It is
worth pointing out at this stage that the plot in Figure
5b was rendered linear after a series of computational
iterations to obtain the optimised thermal diffusivity Fig. 5: Illustration to Explain Data Analysis from the TPS Device
using Specimens from Mix SP25: (a) Time-History of Average
(α). And at the same time, the thermal conductivity (k)
Temperature Increase; (b) Linear Fitting between Average Tem-
was calculated from the slope of this linear fitting. The perature Increase and the Dimensionless Specific Time Function
authors also point out that the first 9 points were skipped
over as shown in Figure 5a. This helps to eliminate the diffusivity are obtained by this single TPS test by means of
influence of the Kapton insulation layer at initial time[30]. the data shown in Figure 5.
The average transient temperature increase (∆T) on the
bifilar spiral of Kapton probe has been given the exact
analytical solution[31] as follows: Results and discussion
TT = P0 Q r 3/2 rk V D Q x V ......................................(1)
-1 The rebound associated with spraying each mix is listed in
Table 1. The pre-mixed powder for the SP0 mix underwent
where, ∆T is the average transient temperature increase setting overnight due to moist sand. Thus, this mix was ignored
(K); Po is the output power (W); r is the radius of the in further discussion. Nonetheless, it appears from Table 1
Kapton probe (m) and, k is the thermal conductivity of the that there was an increase in the rebound with an increase in
testing sample (W/(m·K)). D(τ) is a known dimensionless the amount of EPA in the fine aggregate. An earlier study on
specific time function[31], which depends on τ, which is the the effect of particle density on rebound in dry-mix shotcrete
dimensions time, as evaluated by Equation (2): revealed that a lower density was beneficial to the reduction
of rebound[10]. The results of the present investigation clearly
x = at /r ..............................................................(2) contradict this premise. In that earlier study, the aggregate
density ranged from 600-15,000 kg/m3, whereas in this
where, α is the thermal diffusivity (m2/s); and t is the time of investigation, the bulk density of EPA was less than 72 kg/m3.
measurement (s). In Equations (1) and (2), the parameters As demonstrated by Armelin and Banthia[37, 38], the tendency
∆T, Po, r and t are all known parameters, whereas only k for a particle to rebound is a trade off between its ability to
and α are unknown. A series of computational iterations indent the substrate on the one hand, and its susceptibility to
was conducted to fit the function ∆T versus D (τ), until ejection on the other. Bindiganavile and Banthia[10] illustrated
an optimised thermal diffusivity (α) was found to obtain that once a lightweight aggregate reached the substrate, it
the linear straight line seen in Figure 5b. The thermal was more likely to resist ejection than would be a denser
conductivity (k) was calculated from the slope of this aggregate. However, a particle as light as EPA, may not even
straight line. Thus, both thermal conductivity and thermal

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

310 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Properties of Lightweight Dry-Mix Shotcrete for Mine Insulation

be able to reach the substrate in the first place. The authors

Table 2
confirm that when spraying the SP100 mix, a large portion Mechanical Data
of dry material, predominantly EPA particles, were seen
fanning out from the stream and clearly not reaching the Batch
SP25 SP50 SP75 SP100
substrate. One may question whether the resulting wastage name
is to be classified as rebound or perhaps a special case of 1st day UCS
overspray. Either way, it appears that in the case of dry-mix 13.06 ± 0.69 12.01 ± 0.65 9.26 ± 0.27 1.12 ± 0.07
(MPa)
shotcreting, there exists a lower limit to the particle density,
below which the benefit on rebound is not witnessed. 3rd day UCS
28.58 ± 0.57 19.41 ± 0.16 22.70 ± 1.2 1.90 ± 0.03
(MPa)

Density and water absorption 5th day UCS

31.11 ± 0.49 20.70 ± 3.08 24.01 ± 3.28 1.97 ± 0.05
As seen in Figure 6, the oven dry density for the shotcrete (MPa)
samples lay between 2100 kg/m3 and 800 kg/m3. A 7th day UCS
comparison was made with conventionally cast mixes 34.39 ± 0.85 23.96 ± 3.67 25.37 ± 0.61 2.20 ± 0.07
(MPa)
batched to identical proportions, wherein the oven-dry
density ranged from 2150 kg/m3 to 1200 kg/m3[24]. The 14th day
41.66 ± 2.74 29.78 ± 1.28 25.55 ± 3.73 3.80 ± 0.66
UCS (MPa)
difference between cast and sprayed mixes was not
significant except for the extreme case where all sand 28th day
45.92 ± 0.45 31.38 ± 0.93 29.72 ± 1.61 4.26 ± 0.12
was replaced with EPA. Note that the water absorption in UCS (MPa)
the shotcrete mixes as recorded after firstly, immersion
1st day STS
in water and subsequently, boiling, were nearly identical (MPa)
1.80 ± 0.18 1.55 ± 0.15 1.54 ± 0.08 0.36 ± 0.06

3rd day STS

3.41 ± 0.14 3.10 ± 0.08 3.31 ± 0.12 0.62 ± 0.06
(MPa)

5th day STS

5.50 ± 0.10 3.64 ± 0.34 3.94 ± 0.17 0.72 ± 0.13
(MPa)

7th day STS

5.59 ± 0.32 4.12 ± 0.56 4.02 ± 0.31 0.75 ± 0.05
(MPa)

14th day STS

6.54 ± 0.09 4.37 ± 0.45 4.15 ± 0.18 0.80 ± 0.06
(MPa)

28th day
6.90 ± 1.34 4.44 ± 0.52 4.49 ± 1.29 1.16 ± 0.17
STS (MPa)

except for that in SP100. Generally speaking, as seen in

Figure 7, the water absorption went up with an increase
in the EPA content, from a value of 5% in the SP25 mix to
55% in the SP100 mix wherein EPA replaced all sand.
Fig. 6: Effect of EPA on Density of Shotcrete; S: Shotcrete; C:
Concrete
Mechanical performance
The compressive (UCS) and tensile (STS) strength of the
shotcrete samples are listed in Table 2. Due to the inherent
lack of control over the water content (and consequently,
the water- binder ratio) in the dry-mix shotcrete mixes,
there is more merit to comparing the UCS values against
the oven-dry density, as illustrated in Figure 8a. Note
that the uniaxial compressive strength (UCS) for the
corresponding cast mixes, as taken from Liu et al.,[24]
is shown alongside. Although the two processes result
in products of varying density, it is clear that the UCS in
both cases obeys a similar trend with oven-dry density.
The trend so obtained can be attributed to the porous
microstructure of EPA itself, and possibly, a weaker
interfacial transition zone influenced by the introduction
of EPA. As expected, the only significant difference in the
Fig. 7: Effect of EPA on Water Absorptions in Shotcrete UCS as a result of the shotcreting process was seen with

Organised by
India Chapter of American Concrete Institute 311
Session 3 B - Paper 3

Fig. 8: Effect of EPA Mechanical Properties of Shotcrete: a)

Variation in UCS with Oven-dry Bulk Density; b) Variation in
STS with UCS

mix SP100, which is in keeping with the difference in their Fig. 9: Thermal Constants Evaluated for Different Moisture
oven-dry densities discussed earlier. Similar UCS-density States: a) Thermal Conductivity and b) Thermal Diffusivity
relationships were obtained by previous researchers
also[17-19]. suggest a non-linear relationship with an exponent of
0.90. The difference seen between the sprayed and cast
The relationship between UCS and STS is plotted in Figure
mixes may be attributed to superior consolidation in the
8b, along with the ACI 363R prediction[39]. Whereas other
researchers have published plots with exponents equal former, an explanation that is consistent with research
to 0.80[40, 41], present authors suggest an exponent of 0.72 on self compacted concrete wherein a power-law with an
from cast samples. The data from the shotcrete samples exponent of 1.04 was obtained[42].

Table 3
Thermal Data

Batch name SP25 SP50 SP75 SP100

Air-dry shotcrete thermal conductivity (W/(m· K)) 1.95 1.66 1.24 0.41

Air-dry shotcrete thermal diffusivity (mm2/s) 1.56 1.05 0.83 0.57

Air-dry shotcrete volumetric heat capacity (MJ/(m3·K)) 1.26 1.58 1.50 0.72

Air-dry shotcrete specific heat capacity ((J/(kg·K)) 568.36 814.89 807.3 0.18

Oven-dry shotcrete thermal conductivity (W/(m· K)) 1.44 1.11 0.83 0.37

Oven-dry shotcrete thermal diffusivity (mm2/s) 0.88 0.81 0.68 0.48

Oven-dry shotcrete volumetric heat capacity (MJ/(m ·K)) 3

1.65 1.37 1.22 641.77

Oven-dry shotcrete specific heat capacity ((J/(kg·K)) 793.56 783.12 735.76 688.67

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

312 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Properties of Lightweight Dry-Mix Shotcrete for Mine Insulation

Thermal properties that shotcreting resulted in higher thermal diffusivity, and

The thermal constants as derived in this study are listed this may again be attributed to the higher water demand
in Table 3. Recall that this TPS device is rated to yield associated with EPA.
data within 5% error deviating from the mean value In common numerical tools based on the finite element
shown. Figure 9a illustrates the thermal conductivity for method, the specific heat capacity is used to evaluate the
the shotcrete samples. As expected, there was a drop in thermal resistance of the insulation. It is obtained from
this parameter with an increase in the EPA replacement. the following expression:
C p = k/ Q ta V ...........................................................(3)
Given that EPA has a thermal conductivity of 0.04 W/
(m·K)[18,43] compared with a value of 0.78 to 2.2 W/(m·K)
for sand[44], it is clear that an increase in EPA content
where, α is the thermal diffusivity (m2 /s); k is the
should result in lower thermal conductivity. Now, as is
thermal conductivity (W/(m•K)); ρ is the corresponding
common in the production of dry-mix shotcrete, the water
density (kg/m3); and Cp is the specific heat capacity (J/
content is adjusted by the nozzleman to achieve adequate
(kg•K)). The product, ρCp, is known as the volumetric
shootability. This ensures a minimum rebound and
heat capacity (J/(m3•K)). Note that this parameter
adequate penetration resistance. It is well known that the
is dependent upon the actual density and hence, the
penetration resistance is largely dependent on the water
corresponding density as evaluated for the oven-dry
demand of the dry-mix powder [37, 45]. Note that the air and
and air-dry conditions must be used. The specific heat
water have a thermal conductivity of 0.026 W/(m.K) and
capacity and the volumetric heat capacity as found here
0.60 W/(m.K), respectively[44]. Therefore, it is likely that
are listed in Table 3. As expected, the air-dry samples
an increase in the highly water absorptive EPA leads
have larger heat capacities than the oven-dry samples.
to an increase n the water content in situ. It should be
In cement-based systems, it is possible to predict the
pointed out that the dry-mix shotcrete process produces
specific heat capacity of the hardened composite based
a more compacted system as compared to conventional
on the specific heat capacity of the ingredients using
casting[46]. However, Nguyen et al.[47] found that superior
the rule-of- mixtures[34, 48, 49]. The authors have shown
compaction does not lead to higher thermal conductivity.
this to be true for conventionally cast samples [24] made
In fact, when the results of the present study were
with the same mixes as sprayed here. The specific
compared with those for cast concrete prepared with
heat capacity of shotcrete was found to be less than
the same mix[24], the thermal conductivity of shotcrete
that of corresponding cast concrete, and this is likely
samples was consistently lower.
due to the lower density obtained with the former,
As seen in Figure 10, more moisture is trapped when when incorporating EPA. However, due to the inherent
shotcreting as against during casting in the conventional difficulty in estimating the actual water content in situ,
manner. This is especially true at higher EPA contents. it was not possible to test the rule-of-mixtures on the
Notice in Figure 9b, that there was a decrease in thermal dry-mix shotcrete samples in the present study.
diffusivity with an increase in the EPA content. Once again,
when compared with results obtained with conventionally
cast mixes made with identical proportions[24], it appears
Concluding Remarks
Thermal constants were investigated for cast concrete
and dry-mix shotcrete, both containing expanded perlite
aggregate (EPA) as a lightweight inclusion. Five mixes,
with increasing EPA/sand ratio were prepared and the
results show that dry-mix shotcrete with up to 75% of sand
substituted with EPA offers superior thermal properties
without compromising its mechanical performance. The
principal findings are enumerated as follows:
ll Although replacing sand with EPA results in lower
densities, the spraying process led to high rebound
due to the very low density of the EPA particles.
Furthermore, as evidenced from comparing air-dry
and over-dry samples, the water demand in shotcrete
was seen to increase consistently with the EPA dosage.
ll Whereas the split tensile strength is seen to be
proportional to the square root of the uniaxial
compressive strength in conventional cast concrete,
there was a linear relation obtained for the shotcrete
Fig. 10: Linear fitting between moisture contents and the ratio samples examined here.
of thermal conductivity between air-dry and oven-dry conditions

Organised by
India Chapter of American Concrete Institute 313
Session 3 B - Paper 3

ll As expected, the thermal conductivity and the thermal 18. Topçu İB, Işıkdağ B. Effect of expanded perlite aggregate on the
properties of lightweight concrete. Journal of Materials Processing
diffusivity dropped with an increase in the EPA content.
Technology. 2008;204(1-3):34-8.
Superior compaction and the higher water demand
19. Sengul O, Azizi S, Karaosmanoglu F, Tasdemir MA. Effect of expanded
associated with the dry- mix shotcrete process did not perlite on the mechanical properties and thermal conductivity of
noticeably affect these thermal constants. lightweight concrete. Energy and Buildings. 2010;43(2-3):671-6.
20. CSA. CAN/CSA A23.1/A23.2:Concrete materials and methods of
Acknowledgements concrete construction/methods of tests and standard practices
for concrete. Mississauga, Ontario, Canada: Canadian Standard
The authors thank Mr. Chaoshi Hu and Mr. Lang Liu for Association; 2009.
their assistance with the shotcrete production and sample 21. ACI. Guide for specifying underground shotcrete (506.5R-09).
preparation. Continued financial support from the Natural Farmington Hills, MI: American Concrete Institute; 2009.
Sciences and Engineering Research Council (NSERC), 22. ASTM. Standard test method for sieve analysis of fine and coarse
Canada, is gratefully acknowledged. aggregate (ASTM C126- 06). West Conshohocken, Pennsylvania:
ASTM International; 2006.
References 23. ACI. Guide to shotcrete (506R-05). Farmington Hills, MI: American
1. Hartman HL, Mutmansky JM, Ramani RV, Wang YJ. Heat sources Concrete Institute; 2005.
and effects in mines. Mine ventilation and air conditioning Third 24. Liu WV, Apel DB, Bindiganavile V. Thermal characterisation of a
Edition1998. p. 592. lightweight mortar containing expanded perlite for underground
2. Bottomley P. The reduction in heat flow due to the insulation of rock insulation. International Journal of Mining and Mineral Engineering.
surfaces in mine airways. 2nd US Mine Ventilation Symposium. 2011;3(1):55-71.
Reno, NV1985. p. 457-61. 25. ASTM. Standard test method for density, absorption, and voids
3. Bottomley P, van Rensburg B. The insulation of rock surfaces-an in hardened concrete (ASTM C642 -06). West Conshohocken,
empirical study. Journal of the Mine Ventilation Society of South Pennsylvania: ASTM International; 2006.
Africa. 1987;40(4):41-5. 26. ASTM. Splitting tensile strength of intact rock core specimens
4. Chellam S. Thermal insulation of underground mine workings at (ASTM D3967-08). West Conshohocken, Pennsylvania: ASTM
deep levels Master of Science. Rapid City, South Dakota, South International; 2008.
Dakota School of Mines and Technology; 1992. 27. ACI. Specification for shotcrete (ACI 506.2-95). Farmington Hills,
5. Warner J. Understanding shotcrete–the fundamentals. Concrete MI: American Concrete Institute; 1995.
International. 1995:59-64. 28. ASTM. Standard test method for obtaining and testing drilled
6. Yoggy GD. The history of shotcrete. Shotcrete (American Shotcrete cores and sawed beams of concrete (ASTM C42 M-12). West
Association). 2000;2(4):28-9. Conshohocken, Pennsylvania: ASTM International; 2012. p. 8.

7. Hoek E, Brown ET. Chapter 15: Shotcrete support. Underground 29. ASTM. Standard test method for pulse velocity through concrete
excavations in rock: Spon Press; 1990. p. 205. (ASTM C 597-09). West Conshohocken, Pennsylvania: ASTM
International; 2009.
8. Manuel, A., Herbert, A., Ali, E. Denis, A.F. To study the properties
& materials of shotcrete used in tunnel construction. Int. J. of Res. 30. ISO. Plastics — Determination of thermal conductivity and thermal
in Eng. Tech. and Mngmt. 2014; 2(6):1-3. diffusivity —Part 2:Transient plane heat source (hot disc) method.
Geneva, Switzerland: International Organization for Standardization;
9. Leung CKY, Lai R, Lee AYF. Properties of wet-mixed fiber reinforced 2008.
shotcrete and fiber reinforced concrete with similar composition.
Cement and Concrete Research. 2005;35(4):788- 95. 31. Gustafsson SE. Transient plane source techniques for thermal
conductivity and thermal diffusivity measurements of solid
10. Bindiganavile V, Banthia N. Effect of particle density on Its rebound materials. Review of Scientific Instruments. 1990;62(3):797- 804.
in dry-mix shotcrete. Journal of Materials in Civil Engineering.
2009;21(2):58-64. 32. Log T, Gustafsson SE. Transient plane source (TPS) technique for
measuring thermal transport properties of building materials. Fire
11. Kaiser PK, Tannant DD. The role of shotcrete in Hard-Rock Mines. and Materials. 2004;19(1):43-9.
Underground Mining Methods: Engineering Fundamentals and
International Case Studies. 2001:579-92. 33. Al-Ajlan SA. Measurements of thermal properties of insulation
materials by using transient plane source technique. Applied
12. USBM. An 'insulating' shotcrete for heat abatement in deep mines. Thermal Engineering. 2006;26(17-18):2184-91.
Technology News 4341994.
34. Bentz DP. Transient plane source measurements of the thermal
13. Kramar D, Bindiganavile V. Mechanical properties and size effects properties of hydrating cement pastes. Materials and Structures.
in lightweight mortars containing expanded perlite aggregate. 2007;40(10):1073-80.
Materials and Structures. 2010;44(4):735-48.
35. Huang L, Liu L-S. Simultaneous determination of thermal
14. Ciullo PA. The industrial minerals. Industrial minerals and their conductivity and thermal diffusivity of food and agricultural
uses: a handbook and formulary. Westwood, New Jersey: Noyes materials using a transient plane-source method. Journal of Food
Publication; 1996. p. 52. Engineering. 2009;95(1):179-85.
15. Brouk JJ. Perlite aggregate: its properties and uses. Journal of the 36. Bouguerra A, Aït-Mokhtar A, Amiri O, Diop MB. Measurement
American Concrete Institute. 1949;46(11):185-90. of thermal conductivity, thermal diffusivity and heat capacity of
16. Demirboğa R, Gül R. The effects of expanded perlite aggregate, silica highly porous building materials using transient plane source
fume and fly ash on the thermal conductivity of lightweight concrete. technique. International Communications in Heat and Mass Transfer.
Cement and Concrete Research. 2003;33(5):723-7. 2001;28(8):1065-78.

17. Karakoc MB, Demirboga R. HSC with expanded perlite aggregate 37. Armelin H, Banthia N. Mechanics of aggregate rebound in
at wet and dry curing conditions. Journal of Materials in Civil shotcrete—(Part I). Materials and Structures. 1998;31(2):91-8.
Engineering. 2010;22(12):1252-9.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

314 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Properties of Lightweight Dry-Mix Shotcrete for Mine Insulation

38. Armelin H, Banthia N. Development of a general model of aggregate 44. Sukhatme SP. Some thermophysical properties of selected
rebound for dry-mix shotcrete—(Part II). Materials and Structures. materials. A Text Book on Heat Transfer. Cambridge, Massachusetts:
1998;31(3):195-202. Phlogiston Press; 2011. p. 713,8.

39. ACI. State-of-the-art report on high-strength concrete (ACI-363R). 45. Jolin M, Beaupré D, Mindess S. Tests to characterise properties
Farmington Hills, MI: American Concrete Institute; 1992. of fresh dry-mix shotcrete. Cement and Concrete Research.
1999;29(5):753-60.
40. Oluokun FA, Burdette EG, Deatherage JH. Splitting tensile strength
46. Bindiganavile V, Banthia N. Process dependence of shotcrete for
and compressive strength relationships at early ages. ACI Materials
repairs. International Journal of Materials and Product Technology.
Journal. 1991;88(2):115-21. 2005;23(3):240-56.
41. Gardner NJ, Poon SM. Time and temperature effects on tensile, 47. Nguyen TT, Picandet V, Carre P, Lecompte T, Amziane S, Baley C.
bond, and compressive strengths. Journal of the American Concrete Effect of compaction on mechanical and thermal properties of hemp
Institute. 1976;73(7):405-9. concrete. European Journal of Environmental and Civil Engineering.
2010;14(5):545-60.
42. Sekhar TS, Rao PS. Relationship between compressive, split tensile,
flexural strength of self compacted concrete. International Journal 48. Bentz DP, Peltz MA, Duran-Herrera A, Valdez P, Juarez CA. Thermal
of Mechanics and Solids. 2008;3(2):157-68. properties of high- volume fly ash mortars and concretes. Journal
of Building Physics. 2011;34(3):263-75.
43. Khan MI. Factors affecting the thermal properties of concrete and
49. Choktaweekarna P, Saengsoyb W, Tangtermsirikulb S. A model for
applicability of its prediction models. Building and Environment.
predicting the specific heat capacity of fly-ash concrete. ScienceAsia.
2002;37(6):607-14. 2009;35(2):178-82.

Vivek Bindiganavile
Vivek S. Bindiganavile, PhD; PEng
Associate Professor, Department of Civil and Environmental Engineering
7-281, Donadeo ICE Building
9211, 116 St. NW, University of Alberta
Edmonton, AB, CANADA T6G 1H9
(T) 1-780-492-9661 (E) vivek@ualberta.ca
Vivek Bindiganavile is an Associate Professor at the University of Alberta, Edmonton Canada. He obtained
his B.Tech in Civil Engineering from the Indian Institute of Technology, Banaras Hindu University and
completed his MASc and PhD in Civil Engineering at the University of British Columbia, Canada. He is a
registered Professional Engineer in Alberta, Canada. His research interests include the development
and characterization of lime, gypsum and cement based systems for their rheology, fracture mechanics,
thermal properties and durability.

Organised by
India Chapter of American Concrete Institute 315
Session 3 B - Paper 4

Mechanical Properties of Cementitious Composites Containing

Phase Change Materials: Experimental Data and Effective Medium
Approximations

G. Falzone, G. Puerta Falla, Z. Wei, M. N. Neithalath L. Pilon

Zhao, A. Kumar, School of Sustainable Engineering and Department of Mechanical and
M. Bauchy, and G. Sant the Built Environment, Arizona State Aerospace Engineering, University of
Department of Civil and Environmental University California, Los Angeles
Engineering, University of California, Los Angeles

Abstract Proper selection of phase change materials may realize

their use in a multitude of smart building materials
Implementation of phase change materials (PCMs) in
applications that may enhance the service lifetime of the
concrete is an attractive means to engender energy
built environment, by mitigating thermal cracking, or
savings due to their capability for latent thermal energy
preventing damage from freeze­thaw cycles (Bentz and
storage. In order for PCM­composite cement elements to
Turpin, 2007; Fernandes et al., 2014). Generally, increasing
be properly designed, the influence of these soft particles
the volumetric energy storage of the PCM­composite
on the mechanical properties must be quantified. The
building material (by increasing PCM volume fraction)
present study provides experimental data representing
improves performance in the lens of thermal properties,
the effects of PCM inclusion on key mechanical properties
but may substantially degrade mechanical properties.
(elastic modulus and compressive strength) as a function
PCMs suitable for ambient temperature applications
of time and inclusion volume fraction. A competition
(based on their phase transition temperature window)
between the influences of stiff quartz and soft PCM
are generally composed of a paraffinous core encased in
inclusions on the elastic modulus is noted in mixed (PCM
a thin polymeric shell. These microencapsulated PCMs
+ quartz + cement paste) inclusion mortars. The ability of
have effective mechanical properties greatly inferior to
effective medium approximations (EMAs) to predict the
those of the inorganic cement hydrates which make up
effective elastic modulus of PCM­mortar composites is
the inorganic matrix of cementitious composites (Taylor,
examined. Both the EMAs of Hobbs and of Garboczi and
1997). Therefore, the mechanical properties of PCM­
Berryman (G­B) reliably described the elastic modulus of
composite materials requires extensive study before they
composites containing both PCM and quartz inclusions.
can be made practical in structural elements.
Based on its strong fit to experimental data and ease of
implementation, the Hobbs EMA was further applied to As would be expected, multiple researchers have observed
develop a design rule of performance equivalence, such compressive strength degradation in PCM mortars which
that elastic modulus of the composite can be maintained scales with PCM content (Fernandes et al., 2014; Hunger
equivalent to that of the cementitious matrix, in spite of the et al., 2009; Fenollera et al., 2013). However, more in­depth
addition of soft PCM inclusions. discussions of mechanical properties are rare, especially
in regards to stiffness (elastic modulus). As PCMs are
Keywords: Elasticity; Mechanical properties; Micro­
generally soft polymeric materials, we can look to trends
mechanics; Phase change materials.
in analogous composites, i.e., those containing inclusions
of scrap tire rubber and expanded polystyrene beads
Introduction (Meshgin et al., 2012; Eldin and Senouci, 1993; Miled et
There is significant interest in embedding al., 2007; Shin et al., 2011). While these studies indicate
microencapsulated phase change materials (PCM) in similar trends as in the case of PCM additions, i.e., the
concrete to enhance its thermal performance. As a effective mechanical properties of concrete reduce
result of their capability to store thermal energy during with increasing inclusion content, they seldom present
reversible changes between (typically solid and liquid) experimental data of elastic modulus, and even more
states, embedment of these materials increases the rarely provide means for predicting composite properties
thermal inertia of building materials, and is thus a based on inclusion characteristics. Therefore, the present
promising method for reducing energy expended in study seeks first to experimentally quantify the effects
heating or cooling building interiors (Cabeza et al., 2007; of microencapsulated PCM addition on the compressive
Arce et al., 2012; Tyagi and Buddhi, 2007; Khudhair strength and elastic modulus of cement­based composites
and Farid, 2004; Kuznik et al., 2011; Tyagi et al., 2011). over time. Second, the application of effective medium
approximations (EMAs) in predicting the elastic modulus

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

316 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical Properties of Cementitious Composites Containing Phase Change Materials: Experimental Data and Effective Medium Approximations

of these composites is investigated, with the goal of In this equation Kl and Gl are the least bulk and shear
accurately modelling the combined effects of mixed stiff moduli of the components, while Kg and Gg are the
and soft inclusions (PCM and quartz sand). Finally, an greatest. The subscript “r” refers to the rth component
appropriate EMA is used to develop a design rule for the in the composite. Comparison of experimental elastic
design of inclusion dosage as to ensure that the reduction moduli to these bounds can provide useful information,
in elastic modulus is negated. The findings are significant i.e., data consistently outside of the bounds may suggest
in guiding design to limit the detrimental impacts of PCM that a critical microstructural component is not being
on mechanical behaviour of concretes, while providing considered. Such a methodology has been used to argue
maximal thermal performance. for the consideration of the interfacial transition zone as a
discrete component when modelling the elastic modulus of
cementitious composites (Nilsen and Monteiro, 1993). While
Effective Medium Approximations
this argument should apply to the PCM­mortar composites
The microstructures of cementitious composites consist discussed herein, the extreme width of H­S bounds under a
of at least three components, i.e., the cement paste large mismatch in component stiffnesses makes violation
matrix, inclusions, and the interfacial transition zone (ITZ) of the bounds very unlikely in this case (Hashin, 1983).
formed at the surface of inclusions. The ITZ occurs due
to “wall effects” which hamper particle packing near flat Hobbs developed an EMA for the effective bulk modulus
surfaces, creating a surrounding layer of porous, and Keff of composites composed of aggregate particles
therefore weaker hydration products. Generally, the effect perfectly bonded to a continuous matrix (Hobbs, 1971).
of porosity in the cement paste matrix is often lumped into Equations for elastic modulus were subsequently derived
its effective properties, rather than by considering pores a for two specific cases: (a) identical Poisson’s ratio between
discrete component. In order to describe the mechanical matrix and inclusion, and (b) inclusions have negligible
properties of these complex systems (which are made stiffness (Ei ≈ 0, as air voids). The respective EMAs
even more complicated with the addition of PCM), a brief resulting from these two assumptions are

E eff = E m #1 + & .....................(2a)

review of effective medium approximations for the elastic 2z (E i - E m)
modulus of composites is provided herein. In general, (E i + E m) - z i (E i - E m)
the EMAs mentioned below assume that the individual
and E eff = E m # &
material components are homogeneous, isotropic, and (1 - 2y eff ) (1 - z i) ...................................
(2b)
linear elastic. (1 - 2y m) (1 + z i)

The widest bounds for the effective elastic modulus where vi and vm are the Poisson’s ratios of the inclusion
of composites are the parallel (Voigt, 1889) and series and matrix components, respectively. Additionally, veff is
(Reuss, 1929) models. The parallel EMA assumes that the effective Poisson’s ratio of the composite, which can
each component is equally strained, as they would be if be approximated by the Voigt­Reuss­Hill (VRH) average as

y eff = #Q o i z i + o m z m V + S X& /2 ..................(2c)

welded side­by­side. The series EMA assumes equivalent oi om
stress in all components, such that the components are oi zm + om zi
stacked in the direction of loading. It has been noted that
the equivalent stress condition is a better assumption The differential effective medium theory (D­EMT) is another
when the matrix is more stiff than the inclusion, and the method for predicting the elastic moduli of composite
equivalent strain assumption is better when the matrix is materials. This scheme is distinctive in its treatment of
stiffest (Hansen, 1965). one component as particles within a continuous matrix.
This prevents phase inversion, which is necessary when
The tightest bounds for the elastic modulus of a multi­ the ratio of moduli between the components is large, e.g.,
component composite, without assumptions of the in the case of air voids or soft inclusions. The bulk and
component geometry, were developed by Hashin and shear moduli of the composite (isotropic matrix containing
Shtrikman (H­S) based on variational principles for elastic spherical inclusions) are the solutions to a set of coupled
strain energy (Hashin and Shtrikman, 1963; Walpole, 1966). non­linear ordinary differential equations, which can be
Walpole’s identical formulation, more readily applicable to solved by a 4th order Runge­Kutta scheme (Hildebrand,
N­component composites, is expressed as 1987). These equations are given in a scalar form as
#| z r (K ll + K r) -1& - K ll # K eff # #| z r (K gl + K r) -1& - K gl .....(1a)
N -1 N -1
(McLaughlin, 1977)

=# &# &
r=1 r=1
dK eff K i - K eff K eff + K l ....................................(3a)
dz i 1 - zi Ki + K l
#| z r (G ll + G r) -1& - G ll # G eff # #| z r (G gl + G r) -1& - G gl ..(1b)
N -1 N -1

=# &# &
r=1 r=1
dG eff G i - G eff G eff + G l
where: K ll = 4 G l K gl = 4 G g G ll = 3 S 1 + X
10 -1 , dz i 1 - zi Gi + Gl ....................................(3b)

G l = S 6 X K + 2G
3 3 2 G l 9K l + 8G l G eff 9K eff + 8G eff
4
K l = 3 G eff
and G ll = 3 S 1 + 10 X , and K eff = K m and
-1 eff eff

2 G l 9K l + 8G l G eff = G m at z i = 0

Organised by
India Chapter of American Concrete Institute 317
Session 3 B - Paper 4

The weakening effect of the interfacial transition zone in In accordance with ASTM C109 (ASTM International,
cementitious composites can be considered by combining 2014a, p. 109), the compressive strength of composites
the ITZ shell with the inclusions as an “effective particle.” was measured on cubic specimens (50 mm on edge).
Garboczi and Berryman (G­B) combined the generalized Three replicate samples were tested for each mixture, and
self­consistent method (GSCM) (Christensen and Lo, 1979) the average coefficient of variation was ≈ 5%. Following
and differential effective medium theory to first homogenize ASTM C469 (ASTM International, 2014c), the elastic
an effective particle of aggregate and ITZ, and subsequently modulus was measured on cementitious composites of
homogenize this particle into the cement paste matrix a cylindrical geometry (diameter x height, 101.6 mm x
(Garboczi and Berryman, 2001). D­EMT is well­suited to the 203.2 mm) using an MTS 311.31 closed­loop servo­hydraulic
consideration of discrete particles in a continuous matrix, instrument. Prior to measurement, the cylindrical
but exhibits a slight dependency on the order in which specimens were capped with a quick­ setting gypsum
inclusions of differing properties are embedded into the plaster to ensure proper alignment and contact with the
matrix. GSCM is ideal for mapping the core­shell particles compression platens. The experimental chord elastic
of inclusions + ITZ, thus simplifying the microstructure into modulus Eeff (equivalent to Em in the case of cement paste)
two components. This method has been shown to agree very of the composite specimens was calculated according to
well with both experimental datasets and results obtained ASTM C469 (ASTM International, 2014c, p. 469)
via finite element methods in cementitious composites v -v ..............................................................(4)
E eff = f 22 - f 11
(Garboczi and Berryman, 2001).
where σ2 is the stress developed at 40% of the peak
Materials and Methods load, σ1 is the stress developed at strain ε1=50 με, and
The materials used in this study include: ASTM C150 (ASTM ε2 is the strain produced by stress σ2. The data reported
International, 2012, p. 150) compliant Type I/II ordinary correspond to the average of three replicate samples,
portland cement (OPC), microencapsulated phase change with a coefficient of variation of ≈ 7%.
material (MPCM24D, Microtek Laboratories Inc.), and an
ASTM C778 (ASTM International, 2013) compliant graded Experimental Results
quartz sand. The particle size distributions (PSDs) of each
Figure 1 presents the particle size distributions of
material were determined by static light scattering via an
the respective materials utilized in preparation of the
SLS Particle Analyzer (Beckman Coulter, LS13­320). Each
cementitious composites (OPC and inclusions). The
material was subjected to ultrasonication in isopropanol
median diameters of the OPC, PCM, and quartz are
(also used as the carrier fluid) to disperse it to primary
revealed to be d50,c≈9 μm, d50,p≈20 μm, and d50,q≈365
particles. The complex refractive indices of OPC, PCM,
μm, respectively. These particle size distributions are
and quartz at were taken as 1.70 + 0.10i (Ferraris et al.,
critical to providing insight into the microstructure of
2004), 1.53 + 0.00i (Gao et al., 2001), and 1.54 + 0.00i
the cementitious composites studied herein, as they give
(Ghosh, 1999), respectively. The maximum uncertainty in
some basis for assumptions regarding the interfacial
the median diameters (d50) of the materials was 3.5%
transition zone around each inclusion.
based on 5 replicate measurements.
Cementitious composites were prepared in accordance
with ASTM C305 (ASTM International, 2014b, p. 305)
under four regimes: cement paste (OPC + water mixture),
PCM mortar (PCM + cement paste), quartz mortar (quartz
+ cement paste) and mixed mortar (PCM + quartz + cement
paste). A fixed water­to­cement ratio w/c = 0.45 (mass
basis) was used for each mixture. In the case of PCM
mortars, PCM was added by volume of the total composite
in fractions φp =0.05, 0.10, 0.20, and 0.30. Quartz was
added at φq=0.10, 0.20, 0.30, and 0.55 for quartz mortars.
Two series of mixed mortars were also evaluated for total
inclusion volume fractions of 0.30 (0.10, 0.15, 0.20 PCM; Fig. 1: The particle size distributions of OPC, PCM, and
remainder quartz) and 0.55 (0.10, 0.20 PCM; remainder quar tz sand used in the cementitious composites. The
OPC, microencapsulated PCM, and quartz inclusions have
quartz), respectively. A high range water­ reducing
median diameters d50,c≈9 μm, d50,p≈20 μm, and d50,q≈365 μm,
admixture (Glenium 7500®, BASF Corporation), added at a respectively.
maximum dosage 2% of the (dry) cement mass, enhanced
the workability of the fresh cementitious mixtures. All Figure 2 shows the measured elastic modulus of
specimens were cured for either 1, 3, 7, or 28 days, under cementitious composites as a function of time. Expectedly,
immersion in saturated limewater at 25.0 ± 0.2 °C before increasing PCM volume fraction reduced the stiffness
testing. of the PCM mortars (Figure 2a) (Shin et al., 2011), while

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

318 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical Properties of Cementitious Composites Containing Phase Change Materials: Experimental Data and Effective Medium Approximations

an age of 28 days. This age was chosen for specific focus

as it is often used in structural design codes (e.g., ACI 318
(ACI 318, 2002)).

Discussion
Inputs to Effective Medium Approximations
Fig. 2: Experimentally measured elastic modulus of (a) PCM In order to properly predict the elastic modulus of
mortars, (b) quartz mortars, and (c) mixed mortars with total cementitious composites, appropriate inputs of the
inclusion volume fraction φtot= φq + φp= 0.30 properties for each component must be provided. The
elastic modulus of the plain cement paste at 28 days
increasing quartz volume fraction increased stiffness measured as Em(t=28 days) = 16.75 GPa, is used to
of the quartz mortars (Figure 2b) (Hirsch, 1962; Hansen, represent the paste matrix. The Poisson’s ratio of the
1965; Zimmerman et al., 1986; Nilsen and Monteiro, 1993). cement paste matrix is selected as vm=0.22 (Mindess et
When these two types of inclusions were combined to al., 2003). The elastic modulus and Poisson’s ratios of
form mixed mortars, the stiff quartz particles produced the microencapsulated PCM (subscript “p”) and quartz
some offsetting effect to the stiffness reduction caused (subscript “q”) inclusions are taken from the literature as:
by the PCM. Based on this observation, proper dosage of Ep = 0.0557 GPa, vp =0.499, and Eq=72 GPa, vq=0.22 (Hossain
stiff quartz inclusions may provide a means to mitigate et al., 2009; Chen et al., 2004). The wall effects which
the decrease in stiffness engendered by addition of PCM hamper packing around aggregate surfaces decrease
inclusions. It is also notable that the effects of inclusion as the distance from the surface increases. However,
dosage in the composites tested herein are approximately the simplified assumption of a homogeneous, constant
constant throughout all testing ages, that is, the effects of thickness ITZ around aggregate particles is often made
inclusions on the elastic modulus are apparent after one (Garboczi and Berryman, 2001; Sun et al., 2007; Yang,
day of hydration. 2003). In the literature, the effect of increased porosity in
the ITZ has been shown to reduce its elastic modulus by 20­
The compressive strength of the same cementitious 50% from that of the bulk cement paste (Lutz et al., 1997).
composites was also examined, as shown in Figure Therefore, in the present study, the ITZ is assumed to have
3. The general trend in compressive strength of PCM an elastic modulus 50% that of the matrix, i.e., EITZ=0.5·Em
mortars (Figure 3b) is similar to that of their elastic (Hashin and Monteiro, 2002). The Poisson’s ratio of the
modulus, i.e., a monotonic decrease with PCM dosage. ITZ is assumed equivalent to the paste matrix (Lutz et
However, in the case of quartz mortars (Figure 3b), an al., 1997; Wang et al., 1988; Königsberger et al., 2014b).
independence of compressive strength on quartz dosage The ITZ thickness around quartz particles is assumed
is observed, as strength data from all mixtures may be to be uniform, independent of the particle diameter, and
fitted well to a single logarithmic trend. This indicates that taken as tITZ,q=10 μm, which represents the lower end of
compressive strength in this case is dictated ultimately by experimental observations of ITZ thickness (≈10­50 μm)
the strength of the weakest component, the cement paste. (Li et al., 2005; Königsberger et al., 2014a; Zheng et al.,
Therefore, in the case of mixed mortars (Figure 3c), the 2005; Sun et al., 2007). To simplify calculations, the quartz
quartz inclusions exert no strengthening influence on the particles are assumed to be spherical with a size equal to
system, as behaviour is controlled by the presence of PCM their median diameter (d50,q, Figure 1). These assumptions
microcapsules which behave approximately as flaws. produce a relation for quartz­related ITZ volume fraction
φITZ,q =((tITZ,q+d50,q /2)3/(d50,q /2)3­1)·φITZ,q for a given quartz
volume fraction φq. This equation yields an ITZ volume
fraction φITZ,q=0.025 for a quartz volume fraction φq=0.30
and median diameter d50,q=365 μm.
The existence and properties of ITZs around particles sized
near the cement grains (as are PCM microcapsules) have
rarely been studied (Narayanan and Ramamurthy, 2000;
Ma and Li, 2014). Based on the relatively similar particle
Fig. 3: Experimentally compressive strength of (a) PCM mortars, sizes of PCM and OPC, the effect of PCM in producing an
(b) quartz mortars, and (c) mixed mortars with total inclusion ITZ is expected to be reduced in comparison to quartz
volume fraction φtot= φq + φp= 0.30 inclusions. In this particle diameter regime, it would also
logical that the ITZ thickness decreases linearly with PCM
Although the time­dependent properties are presented, size as the particles will exert decreasing influence on
the following sections are focused on selecting predictive packing, so the thickness is modelled as tITZ,p =C· rp, where
models for the elastic modulus of composite specimens at C=0.25 and rp is the PCM microcapsule radius (μm). The

Organised by
India Chapter of American Concrete Institute 319
Session 3 B - Paper 4

value of C was selected such that the ITZ thickness at the

maximum PCM microcapsule size matched that of the
quartz particles. As a result of this assumption about ITZ
thickness, the volume fraction of PCM­related ITZ in the
composite follows a radius­independent relation given by
φITZ,p =(C3 + 3C2 + 3C)·φITZ,q. As such, for a PCM volume
fraction φp =0.30, the associated ITZ volume fraction
within the composite is predicted as φITZ,p =0.29 when
C = 0.25. These assumptions are used throughout the
study unless otherwise noted.

Comparing Measurements of Elastic Modulus to EMA

Fig. 4: Predictions of EMAs for (a) PCM mortars and (b) quartz
Predictions mortars compared with experimental data as a function
First, the effective medium approximations described of inclusion volume fraction, φq and φq for PCM and quartz,
previously are applied to predict the elastic modulus of respectively. Solid lines represent two component (inclusion
cementitious composites containing PCM, as shown in + cement paste) predictions, while dashed lines denote three
Figure 4(a). The EMAs presented in solid lines result from component (inclusion + cement paste + ITZ) predictions.
consideration of only two components (PCM and cement
paste), while those in dashed lines represent three­ acceptable prediction of the effective composite modulus.
component results (PCM, bulk cement paste, and ITZ). In The consideration of an ITZ reduces the predicted
these composites, the Hashin­Shtrikman (H­S) and Voigt­ modulus in all cases, although only slightly. Notably, the
Reuss (VR) bounds are extremely wide, as expected due fit of experimental data within the H­S bounds is improved
to the large ratio of cement paste elastic modulus to that with the consideration of the ITZ, although the error of the
of the PCM (Em/Ep ≈ 300 when t=28 days). Introduction measurements is substantial to make these effects minor.
of the ITZ component into model consideration narrows Based on the observations of the EMAs in the single
the H­S upper bound, while leaving the H­S lower bound inclusion mortars, the Hobbs and D­EMT/G­B EMAs are
unchanged. Hobbs EMA, implemented for the case of applied to mixed inclusion mortars with varying total
substantial mismatch in the Poisson’s ratio between the inclusion volume fractions (φtot = φp + φq). In cases without
matrix and inclusions (Equation 2b), is able to favourably consideration of the ITZ (three­component composite), a
describe the elastic moduli of PCM mortars across the two­ step homogenization process is followed, wherein
entire range of volume fractions considered. Predictions the quartz particles are embedded first into the cement
arising from the D­EMT lie near the H­S upper bound, paste matrix (using Equation 2a), yielding an effective
and therefore overestimate the elastic modulus. Such matrix into which the PCM microcapsules are embedded
overestimation must be avoided in order to provide more using Equation 2(b). The EMAs calculated in this method
conservative predictions for structural design with PCMs. demonstrate a dependence on the order of inclusion
While the H­S bounds encompass all other predictions, homogenization, with quartz first homogenization (models
the overestimation by the D­ EMT may stem from the denoted “ii”) providing better fits in general. When the ITZ
lack of consideration of the ITZ component. Therefore, is included to yield a four­component composite, a similar
the formulation of D­ EMT with application to the case process is followed, in which the ITZ is first homogenized
including an ITZ component (the model of Garboczi and with its associated inclusion to produce an effective
Berryman, G­B) markedly improves predictions (Garboczi particle. The two types of effective particle (ITZ + quartz
and Berryman, 2001). The success of the G­B EMA was and ITZ + PCM) are then embedded in the cement paste
perhaps unsurprising as it was previously demonstrated matrix in a manner identical to the case without the ITZ.
to accurately predict the elastic response of cementitious The results of these calculations, along with experimental
materials containing an ITZ (Garboczi and Berryman, data of elastic modulus for mixed mortar systems with
2001). φtot = 0.30 and 0.55 are displayed in Figure 5(a) and (b),
Figure 4(b) displays predictions of the same EMAs when respectively. Regardless of total inclusion volume fraction,
applied to quartz mortars. In this case, essentially all the Hobbs EMA produces better predictions of elastic
EMAs predict effective elastic modulus well, and lie within modulus, with the D­EMT showing highly non­conservative
the narrow H­S bounds. Unlike in PCM mortar, the ratio results. When the ITZ is included in EMA predictions, the
of matrix to inclusion moduli in this case is small (Em/ effective modulus of the composite reduces substantially,
Eq ≈ 0.23 when t=28 days). Once again, the Hobbs EMA and more so at large PCM volume fractions, improving the
(this time in the formulation of Equation 2a) favourably fit of both EMAs. These observations are consistent with
predicts the effective modulus, in this case coinciding those of the single inclusion mortars depicted in Figure 4.
with the lower H­S bound. In both two/three­component Next, the influence of characteristics selected for the ITZ
systems containing quartz, the H­S lower bound provids an and inclusion components is investigated in regards to

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

320 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical Properties of Cementitious Composites Containing Phase Change Materials: Experimental Data and Effective Medium Approximations

Fig. 7: (a) The maximum PCM volume fraction that can

Fig. 5: Composite moduli of mixed mortar systems containing be accommodated mixed mortars (at t=28 days) to yield a
total inclusion volume fractions: (a) φtot= φq + φp= 0.30 and composite modulus greater than or equal to the cement
(b) φtot= φq + φp= 0.55 as predicted by D­EMT, and the EMAs of paste. (b) The critical ratio of quartz to PCM inclusions which
Garboczi and Berryman (G­B) and Hobbs. The solid horizontal yields an effective modulus Eeff equal to the elastic modulus
line shows the elastic modulus of the neat cement paste matrix. of the cement paste matrix. Each of these calculations are
implemented using the Hobbs EMA (form “ii”) including the
ITZ component.

of quartz­heavy composites when Eq=80 GPa (Lee and

Park, 2008). Next, the elastic modulus of the ITZ is varied
within the range typically reported in the literature (0.8 ≥
EITZ /Em ≥0.5), keeping all other parameters constant. The
ITZ stiffness more substantially influences the predicted
composite modulus at the PCM­rich end, due to the larger
ITZ volume fraction produced around these inclusions.
Finally, the assumptions of ITZ thickness around quartz
and PCM particles re investigated to study their influence
on modulus predictions. In both cases, increasing the
ITZ reduces the volume fraction of matrix such that the
composite reaches an unphysical circumstance where
there is no paste matrix present, which occurrs at smaller
PCM volume fraction as ITZ thickness increased. The
inability of the EMAs, as formulated herein, to handle
large inclusion volume fractions results from the neglect
of ITZ overlap between neighbouring particles, as the ITZ
volume becomes unreasonably large at modest inclusion
volume fractions. To avoid this issue, ITZ thicknesses are
chosen within a range which accommodates modelling of
physically realizable, and realistic mixtures.
Fig. 6: The influences of parametric variations of: (a) quartz
modulus, (b) ITZ modulus, and (c­d) ITZ thickness on the
properties of composites containing 0.55 inclusion volume A Design Rule for PCM Dosage
fraction, when using the EMA of Garboczi and Berryman As the mixed inclusion mortars investigated have elastic
(Garboczi and Berr yman, 2001). Unless stated, the ITZ moduli greater than or less than the modulus of the plain
thickness is fixed at tITZ, q=10μm around quartz inclusions cement paste dependent on relative proportions of PCM
and tITZ,p=0.25·r p around PCM particles, withEq =72GPa and and quartz inclusion, there must exist a critical ratio of
EITZ=0.5Em. The solid horizontal line shows the elastic modulus inclusion volume fractions (φq /φp) such that the elastic
of the neat cement paste.
modulus of the composite remains: (a) similar, or greater
the G­B model, as shown in Figure 6(a­d). The system with than the paste matrix, i.e., Eeff/Em ≥ 1. Due to its relative
total inclusion volume fraction 0.55 is considered in this simplicity and strong fit to experimental moduli, Hobbs
case, as the predictions in this case are most significantly EMA including the ITZ is used to calculate this ratio. First,
impacted by inclusion properties. Given the already large the desired modulus ratio between composite and paste
mismatch between paste and PCM moduli, altering the matrix are input into the EMA, which is solved numerically
elastic modulus of the PCM by ≈10x only negligibly affects for the inclusion volume fractions. Figure 7 shows the
EMA predictions, so these results are not shown. The maximum allowable PCM volume fraction to maintain
elastic modulus of the quartz inclusions is varied between composite elastic moduli ranging from 1.0 to 2.0 times the
70 and 100 GPa, better representing the elastic modulus cement paste at t=28 days. Note, that to achieve stiffness

Organised by
India Chapter of American Concrete Institute 321
Session 3 B - Paper 4

greater than that of the paste matrix, PCM may not be (Contract: PIR: 12­032), the University of California, Los
introduced at any volume fraction if the total inclusion Angeles (UCLA) and National Science Foundation (CMMI:
volume fraction is not substantial. 1130028) The contents of this paper reflect the views
and opinions of the authors, who are responsible for the
The critical ratio of quartz­to­PCM inclusion volume fractions
accuracy of the datasets presented herein, and do not
is determined for specimens at t=1, 3, 7, and 28 days for the
reflect the views of the funding agency, nor do the contents
case where Eeff/Em = 1, as shown in Figure 7(b). This ratio
constitute a specification, a standard or a regulation.
calculated shows that the quartz volume fraction should
This research was conducted in the Laboratory for the
be approximately 2.5­3.5 times that of PCM to ensure no
Chemistry of Construction Materials (LC2) and the core­
loss in stiffness relative to the plain cement paste. This
facility Molecular Instrumentation Center (MIC) at the
value increases with age due to the development of matrix
University of California, Los Angeles. As such, the authors
stiffness over time, while the inclusion stiffnesses remain
acknowledge the support of these laboratories in making
constant with time. As the total inclusion volume fraction
this research possible. The last author would also like
increases, more quartz must be proportioned relative to
to acknowledge discretionary support provided by the
PCM. This trend can be tied to the increasing influence
Edward K. and Linda L. Rice Endowed Chair in Materials
of the PCM­associated ITZ in diminishing the composite
Science.
modulus. Choice of stiffer aggregates would reduce
the critical ratio, allowing for improved PCM addition at
equivalent stiffness. Similarly, reducing ITZ porosity by References
1. ACI 318, 2002. Building Code Requirements for Reinforced Concrete.
the addition of supplementary cementitious materials
American Concrete Institute, Detroit, Michigan.
would reduce the critical ratio. As only fine aggregate is
2. Arce, P., Castellón, C., Castell, A., and Cabeza, L.F., 2012. Use
utilized in the present study, effects of aggregate interlock of microencapsulated PCM in buildings and the effect of adding
and stress transfer are not considered, which may render awnings. Energy and Buildings, 44:88–93
the results presented herein somewhat conservative 3. ASTM International, 2014a. ASTM Standard C109/C109M.
when extended to concretes (Fernandes et al., 2014). Compressive strength of hydraulic cement mortar (Using 2­in. or
[50­mm] cube specimens). ASTM International, West Conshohocken,
PA.
Conclusions 4. ASTM International, 2012. ASTM Standard C150/C150M. Standard
Although attractive for their improvements in thermal specification of portland cement. ASTM International, West
performance, the very low stiffness and strength of Conshohocken, PA.
typical microencapsulated phase change materials 5. ASTM International, 2014b. ASTM Standard C305/C305M. Standard
(PCMs) are known to diminish the mechanical properties practice for mechanical mixing of hydraulic cement pastes
and mortars of plastic consistency. ASTM International, West
of cementitious composites. Quantitative evaluation and Conshohocken, PA.
prediction of the influence of PCM microcapsules on the
6. ASTM International, 2014c. ASTM Standard C469/C469M. Standard
elastic modulus of PCM­containing composites is necessary test method for static modulus of elasticity and Poisson’s ratio of
to make practical their use in building materials. To this concrete in compression. ASTM International, West Conshohocken,
end, the title study presents the following outcomes: PA.
7. ASTM International, 2013. ASTM Standard C778. Standard
1. Experimental elastic moduli and compressive strength specification for standard sand. ASTM International, West
are presented for PCM mortar, quartz mortar, and Conshohocken, PA.
mixed mortars (PCM + quartz + cement paste). 8. Bentz, D.P., and Turpin, R., 2007. Potential applications of phase
change materials in concrete technology. Cement and Concrete
2. Effective medium approximations (Hobbs; D­ EMT;
Composites, 29(7):527–532
Garboczi and Berryman) are applied to predict effective
9. Cabeza, L.F., Castellón, C., Nogués, M., Medrano, M., Leppers, R.,
composite elastic moduli as a function of component and Zubillaga, O., 2007. Use of microencapsulated PCM in concrete
volume fractions. walls for energy savings. Energy and Buildings, 39(2):113–119
3. The influence of the interfacial transition zone, as 10. Chen, X., Zhang, S., Wagner, G.J., Ding, W., and Ruoff, R.S., 2004.
Mechanical resonance of quartz microfibers and boundary condition
well as that of the individual component properties
effects. Journal of Applied Physics, 95(9):4823–4828
is assessed on the predictions of EMAs for mixed
11. Christensen, R., and Lo, K., 1979. Solutions for effective shear
mortars.
properties in three phase sphere and cylinder models. Journal of
4. A design rule is developed using the Hobbs EMA to the Mechanics and Physics of Solids, 27:315–330
negate the detrimental effects of PCM inclusions on 12. Eldin, N., and Senouci, A., 1993. Rubber­tire particles as concrete
composite stiffness by dosage of rigid inclusions. aggregate. Journal of materials in civil engineering, 5(4):478–496
13. Fenollera, M., Míguez, J., Goicoechea, I., Lorenzo, J., and Ángel
Álvarez, M., 2013. The influence of phase change materials on the
Acknowledgement properties of self­compacting concrete. Materials, 6(8):3530–3546
The authors acknowledge full financial support for this 14. Fernandes, F., Manari, S., Aguayo, M., Santos, K., Oey, T., Wei, Z.,
research provisioned by the California Energy Commission Falzone, G., Neithalath, N., and Sant, G., 2014. On the feasibility of
using phase change materials (PCMs) to mitigate thermal cracking

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

322 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Mechanical Properties of Cementitious Composites Containing Phase Change Materials: Experimental Data and Effective Medium Approximations

in cementitious materials. Cement and Concrete Composites, modulus of concrete considering interfacial transition zone. Cement
51:14–26 and Concrete Research, 38(3):396–402
15. Ferraris, C.F., Hackley, V.A., and Avilés, A.I., 2004. Measurement 33. Li, C.Q., Zhou, X.Z., and Zheng, J.J., 2005. Thickness of interfacial
of particle size distribution in portland cement powder: Analysis transition zone and cement content profiles around aggregates.
of ASTM round robin studies. Cement, concrete and aggregates, Magazine of Concrete Research, 57(7):397–406
26(2):1–11
34. Lutz, M., Monteiro, P., and Zimmerman, R., 1997. Inhomogeneous
16. Gao, G., Moya, S., Lichtenfeld, H., Casoli, A., Fiedler, H., Donath, E., interfacial transition zone model for the bulk modulus of mortar.
and Mohwald, H., 2001. The decomposition process of melamine Cement and Concrete Research, 27(7):1113–1122
formaldehyde cores: the key step in the fabrication of ultrathin
35. Ma, H., and Li, Z., 2014. Multi­aggregate approach for modeling
polyelectrolyte multilayer capsules. Macromolecular Materials and
interfacial transition zone in concrete. ACI Materials Journal, 111(2)
Engineering, 286(6):355–361
36. McLaughlin, R., 1977. A study of the differential scheme for composite
17. Garboczi, E., and Berryman, J., 2001. Elastic moduli of a material
materials. International Journal of Engineering Science, 15:237–244
containing composite inclusions: effective medium theory and finite
element computations. Mechanics of Materials, 33 37. Meshgin, P., Xi, Y., and Li, Y., 2012. Utilization of phase change
materials and rubber particles to improve thermal and mechanical
18. Ghosh, G., 1999. Dispersion­equation coefficients for the refractive
properties of mortar. Construction and Building Materials,
index and birefringence of calcite and quartz crystals. Optics
28(1):713–721
Communications, 163(1–3):95–102
38. Miled, K., Sab, K., and Le Roy, R., 2007. Particle size effect on
19. Hansen, T., 1965. Influence of aggregate and voids on modulus of
EPS lightweight concrete compressive strength: Experimental
elasticity of concrete, cement mortar, and cement paste. ACI Journal
investigation and modelling. Mechanics of Materials, 39(3):222–240
Proceedings, (62):193–216
39. Mindess, S., Young, J.F., and Darwin, D., 2003. Concrete. Prentice
20. Hashin, Z., 1983. Analysis of composite materials—a survey. Journal
Hall, Upper Saddle River, NJ.
of Applied Mechanics, 50(September 1983):481–505
40. Narayanan, N., and Ramamurthy, K., 2000. Microstructural
21. Hashin, Z., and Monteiro, P.J.M., 2002. An inverse method to
investigations on aerated concrete. Cement and Concrete Research,
determine the elastic properties of the interphase between the
30(3):457–464
aggregate and the cement paste. Cement and Concrete Research,
32(8):1291–1300 41. Nilsen, A.U., and Monteiro, P., 1993. Concrete: a three phase
material. Cement and Concrete Research, 23(1):147–151
22. Hashin, Z., and Shtrikman, S., 1963. A variational approach to the
theory of the elastic behaviour of multiphase materials. Journal of 42. Reuss, A., 1929. Berechnung der fließ grenze von mischkristallen auf
the Mechanics and Physics of Solids, 11(2):127–140 grund der plastizitätsbedingung für einkristalle. ZAMM ­Zeitschrift
für Angewandte Mathematik und Mechanik, 9(1):49–58
23. Hildebrand, F.B., 1987. Introduction to Numerical Analysis. Courier
Corporation, 706 p. 43. Shin, K.J., Bucher, B., and Weiss, J., 2011. Role of Lightweight
Synthetic Particles on the Restrained Shrinkage Cracking Behavior
24. Hirsch, T., 1962. Modulus of elasticity of concrete affected by
of Mortar. Journal of Materials in Civil Engineering, 23(5):597–605
elastic moduli of cement paste matrix and aggregate. ACI Journal
Proceedings, 59(12):427–452 44. Sun, Z., Garboczi, E.J., and Shah, S.P., 2007. Modeling the elastic
properties of concrete composites: Experiment, differential effective
25. Hobbs, D.W., 1971. The dependence of the bulk modulus, Young’s
medium theory, and numerical simulation. Cement and Concrete
modulus, creep, shrinkage and thermal expansion of concrete
Composites, 29(1):22–38
upon aggregate volume concentration. Matériaux et Constructions,
4(2):107–114 45. Taylor, H.F.W., 1997. Cement Chemistry. Thomas Telford, 488 p.
26. Hossain, M., Ketata, C., and Islam, M., 2009. Experimental study of 46. Tyagi, V.V., and Buddhi, D., 2007. PCM thermal storage in buildings: A
physical and mechanical properties of natural and synthetic waxes state of art. Renewable and Sustainable Energy Reviews, 11(6):1146–
using uniaxial compressive strength test, in Proceedings of third 1166
international conference on modeling, simulations, and applied 47. Tyagi, V.V., Kaushik, S.C., Tyagi, S.K., and Akiyama, T., 2011.
optimization, p. 1–5. Development of phase change materials based microencapsulated
27. Hunger, M., Entrop, A.G., Mandilaras, I., Brouwers, H.J.H., and technology for buildings: A review. Renewable and Sustainable
Founti, M., 2009. The behavior of self­compacting concrete Energy Reviews, 15(2):1373–1391
containing micro­encapsulated phase change materials. Cement 48. Voigt, W., 1889. Ueber die beziehung zwischen den beiden
and Concrete Composites, 31(10):731–743 elasticitätsconstanten isotroper körper. Annalen der Physik,:573–587
28. Khudhair, A.M., and Farid, M.M., 2004. A review on energy 49. Walpole, L., 1966. On bounds for the overall elastic moduli of
conservation in building applications with thermal storage by inhomogeneous systems—I. Journal of the Mechanics and Physics
latent heat using phase change materials. Energy Conversion and of Solids, 14(1964)
Management, 45(2):263–275
50. Wang, J.A., Lubliner, J., and Monteiro, P.J.M., 1988. Effect of ice
29. Königsberger, M., Pichler, B., and Hellmich, C., 2014a. formation on the elastic moduli of cement paste and mortar. Cement
Micromechanics of ITZ­Aggregate Interaction in Concrete Part II: and Concrete Research, 18(6):874–885
Strength Upscaling. (J. Olek, Ed.) Journal of the American Ceramic
Society, 97(2):543–551 51. Yang, C., 2003. Effect of the interfacial transition zone on the
transport and the elastic properties of mortar. Magazine of Concrete
30. Königsberger, M., Pichler, B., and Hellmich, C., 2014b. Research, 55(4):305–312
Micromechanics of ITZ­aggregate interaction in concrete part I:
Stress concentration. (J. Olek, Ed.) Journal of the American Ceramic 52. Zheng, J.J., Li, C.Q., and Zhou, X.Z., 2005. Characterization of
Society, 97(2):535–542 microstructure of interfacial transition zone in concrete. ACI
Materials Journal, 102(4):265–271
31. Kuznik, F., David, D., Johannes, K., and Roux, J.­J., 2011. A review
on phase change materials integrated in building walls. Renewable 53. Zimmerman, R., King, M., and Monteiro, P., 1986. The elastic moduli
and Sustainable Energy Reviews, 15(1):379–391 of mortar as a porous­granular material. Cement and Concrete
Research, 16(2):239–245
32. Lee, K.M., and Park, J.H., 2008. A numerical model for elastic

Organised by
India Chapter of American Concrete Institute 323
Sessioin 3 B - Paper 5

 Advanced Experiments and Models to Predict the Effective Elastic

Properties of Cementitious Systems: Application to a Geopolymer

Sumanta Das, Pu Yang, Sudhanshu S. Singh, James C.E. Mertens, Xianghui Xiao
Narayanan Neithalath Nikhilesh Chawla Advanced Photon Source,
School of Sustainable Engineering and the Materials Science and Engineering, Arizona State Argonne National Laboratory,
Built Environment, Arizona State University, University, Tempe, AZ, USA Argonne, IL, USA
Tempe, AZ, USA.

Abstract conventional cementitious systems. Scanning electron

microscopy (SEM) on two-dimensional images also
A microstructural and micromechanical study of a fly
are used for pore structure characterization[7,8]. The
ash-based geopolymer paste including synchrotron x-ray
drawback of these methods in probing the pore structure
tomography (XRT) to determine the volume fraction (size
can be alleviated to a large extent through the use of X-ray
> 0.74 m) and tortuosity of pores, scanning electron
microtomographic (XRT or μCT) techniques such as those
microscopy (SEM) and multi-label thresholding method
to identify and characterize the solid phases in the implemented in cementitious systems [9,10].
microstructure, and nanoindentation to determine the The mechanical properties of a heterogeneous material
phase elastic properties, is reported in this paper. The are also influenced significantly by the individual
pore volume from XRT, solid phase volumes from SEM, and phase amounts (both solids and pore) and their elastic
the phase elastic properties are used in multi-step mean properties. The individual phase elastic properties can be
field homogenization (Mori-Tanaka and double inclusion) obtained through the use of nanoindentation experiments
models to determine the homogenized macroscale elastic [11–14]. Micromechanical homogenization models help
modulus. The homogenized elastic moduli are in good estimate the composite properties at the macro- scale.
agreement with the flexural elastic modulus determined The successful implementation of this approach demands
on macroscale paste beams. It is shown that the accurate quantification of the phase volume fractions. The
combined use of microstructural and micromechanical use of empirical models to determine the reaction product
characterization tools at multiple scales provides valuable and pore volume fractions in highly heterogeneous
information towards the material design of such systems. systems containing multiple phases such as fly ash-
Keywords: Geopolymer, X-ray tomography, Pore based geopolymers is unlikely to yield satisfactory
structure, Nanoindentation, Micromechanical models, results. Moreover, it has been reported elsewhere that the
Elastic modulus classification and identification of different solid phases in
the fly ash-based geopolymer using synchrotron XRT is
challenging due to the low absorption contrast of different
Introduction solid phases in the available x-ray energy range[10].
It is well understood that the use of industrial waste/by- In this study, we have carried out a comprehensive
product materials including fly ash, blast furnace slag, investigation of a fly ash-based geopolymer based
and clays can be activated using alkaline media to produce on 3D x-ray synchrotron imaging, nanoindentation
binders that have comparable or superior properties to of the individual constituents in the geopolymer, and
those of OPC-based binders[1–3]. Geopolymers belong to micromechanics-based modeling of the mechanical
the class of alkali-activated binders and are based on fly properties. Synchrotron XRT is used to interrogate the
ash or metakaolin, resulting in an alkali aluminosilicate pore structure using a pixel size of 0.74 μm. We have
reaction product that is morphologically and behaviorally also conducted quantification of the microstructure
different from the calcium (alumino) silicate gels resulting using image segmentation and thresholding techniques.
from OPC hydration [4]. Similar to conventional OPC- Multiple-phase quantification was accomplished using
based binders, geopolymeric systems are also inherently a multiple-thresholding image analysis procedure
porous and heterogeneous, and display widely differing implemented on several high resolution backscattered
mechanical and durability properties. Thus, proper SEM images. The quantified solid phase fractions are
characterization of the pore- and-micro-structure is also validated through the frequency of occurrences of
critical for a fundamental understanding of material different solid phases in the statistical nanoindentation
behavior, as well as developing methods for adequate technique. The microscale properties are up-scaled using
material design. Recent studies have evaluated the pore mean field homogenization schemes including Mori-
structure of fly ash geopolymers using mercury intrusion Tanaka and double inclusion methods[15,16] to extract the
porosimetry (MIP)[5,6], which is a common method for
homogenized Young’s modulus, which is validated through

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

324 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
  Advanced Experiments and Models to Predict the Effective Elastic Properties of Cementitious Systems: Application to a Geopolymer

Table 1
Chemical composition and physical properties of fly ash

Median
SiO2 Al2O3 Fe2O3 CaO MgO SO3 Na2O K 2O LOI
particle size

58.4% 23.8% 4.19% 7.32% 1.11% 0.44% 1.43% 1.02% 0.5% 19 µm

macro-scale experiments. Linking the microstructure and at ~10 μm spacing in a grid on an area approximately 250
micromechanical properties of a heterogeneous material, μm x 250 μm in size, which is considered representative for
and using this information to predict the macroscopic cementitious materials[10,25]. All the indentation locations
mechanical performance, provides efficient means of were carefully selected prior to testing to ensure that the
optimizing the material design and mechanical behavior pores or cavities are not encountered in the process. The
of fly ash-based geopolymers. depth of penetration was chosen to be 500 nm which is
smaller than the characteristic size of unreacted fly ash
inclusions in order to avoid phase-interactions during
Materials and Experimental Procedure
penetration. Continuous stiffness measurement (CSM)
Materials and Mixture Proportions technique [26–29] was employed here to measure the contact
stiffness. CSM technique allows measurement of contact
A Class F fly ash conforming to ASTM C 618 was used as
stiffness at any point of the loading curve corresponding
the starting material. The chemical composition of fly ash,
to any depth of penetration. CSM is accomplished by
determined using x-ray fluorescence is shown in Table 1.
imposing a harmonic excitation of constant amplitude and
8M NaOH solution was used to activate the fly ash. The
frequency to the normally increasing load on the indenter.
alkaline solution was added to fly ash, and mixed for 4
For any excitation frequency, the displacement response
minutes in a laboratory mixer. The mixtures were then
of the indenter and the phase angle are measured
filled in molds and subjected to heat curing in a laboratory
continuously as a function of the penetration depth. The
oven at 60°C for 48 hours, in sealed conditions. This curing
contact stiffness, as a function of penetration depth, is
process was previously shown to provide a compressive
obtained by solving for the in-phase and out-of phase
strength of 25-30 MPa [2], which is commonly adopted for
portions of the response which are described in detail in
many types of structural concretes.
[27]. Young’s modulus for an individual indentation was
measured as the average value over a depth range where
Synchrotron X-ray Tomography both modulus and hardness were independent of depth.
Synchrotron X-ray tomography (XRT) was performed
at the 2-BM beamline of the Advanced Photon Source A constant Poisson’s ratio of 0.20 was used in elastic
(APS) at Argonne National Laboratory. Details of APS modulus calculations since the effect of variation of
beamline 2-BM have been described elsewhere [17,18]. Poisson’s ratio in the range 0.18-0.22 has been reported
A monochromatic beam with energy of approximately 27 to be insignificant for similar systems [14,30].
keV was focused on samples of approximately 1.5 mm
in size. The transmitted x-rays were converted to visible Scanning Electron Microscopy (SEM)
light using a LuAG:Ce scintillator screen coupled with an The plane polished sample was prepared as explained in
objective lens and CoolSnap K4 CCD camera to achieve the section on nanoindentation. The sample was imaged
a pixel size of 0.74 μm. 2D projections were acquired at under a Philips XL30 field emission environmental scanning
angular increments of 0.12° over a range of 180°. The 2D electron microscope (FESEM) in the backscattered mode
projections were then reconstructed to 3D using Fast to obtain high resolution micrographs that distinguishes
Fourier Transform (FFT)-based Gridrec algorithm [19– different solid phases. Several high resolution Images
21]. Synchrotron microtomography is considered highly were obtained at different magnifications so that volume
suitable for the evaluation of pore structure [10,22–24]. fractions of different solid phases could be determined
using a multi-label segmentation algorithm.
Nanoindentation
For nanoindentation, a cylindrical sample of 25 mm in Flexural Test on Geopolymer Beams
diameter and 75 mm in height was prepared and heat Three-point bend tests were performed on prismatic
cured at 60°C for 48 hours. A cubic piece with 4 mm sides beams of size 330 mm x 76 mm x 25 mm (span of 305
was cut and polished to a 0.04 μm colloidal silica finish. mm). Six replicate beams were tested. The beams were
The nanoindentation measurements were carried out on tested in a mid-span displacement-controlled mode
the polished sample in a commercial Nanoindenter (MTS at a constant displacement rate of 0.004 mm/sec. The
Nanoindenter XP) using a Berkovich tip. Samples were mid-span displacement was measured continuously
mounted on aluminum stubs for nanoindentation testing using a LVDT. The test was terminated at a center- point
using superglue. Indentations were carried out at initially displacement value of 0.018 mm when the load reaches

Organised by
India Chapter of American Concrete Institute 325
Sessioin 3 B - Paper 5

about 80% of the peak load which was determined unreacted/partially reacted fly ash, reaction product (N-A-
previously from simple flexural strength tests. S-H) and the pore phase in the fly ash-based geopolymer.
N-A-S-H gel (brighter grey phase) is found around the
fly ash particles as well as uniformly distributed in the
Results and Discussion
microstructure. The micrographs also show unreacted
Multi-Scale Evaluation of the Pore- and Microstructure rounded fly ash particles and partially reacted fly ash
of Fly ash-based Geopolymer particles. In addition, the pore phase (darker phase) can
also be distinguished, both in the N-A-S-H gel as well as
This section provides a detailed look into the pore- and
within some fly ash particles (cenospheres) [6,10]. While
micro-structure of the fly ash-based geopolymer The pore
a small fraction of the partially reacted fly ash particles
network features (total and connected volume, critical pore
retain their rounded shape, there are also particles
size, and tortuosity) and the distribution of the pore and solid
that have undergone non-uniform dissolution on their
phases are characterized using X-ray microtomography
surfaces.
(XRT), and scanning electron microscopy (SEM). The
phase volumes and micromechanical phase properties 3D Characterization and quantification of component
are employed in the prediction of homogenized composite phases using Synchrotron X-Ray Tomography (XRT)
properties as elucidated in a later section.
Synchrotron XRT provides valuable 3D information on
Microstructure of alkali-activated fly ash from the pore structure of random heterogeneous materials,
backscattered SEM which is yet unavailable through other commonly adopted
methods, enabling the prediction of transport properties
Figure 1 shows representative backscattered SEM
and thus the material durability. In order to facilitate
micrographs of the geopolymer paste after 2 days of
efficient computation, cubic volumes of interest of sizes
sealed heat curing at 60°C. The reaction product in
ranging from 503 voxels to 4003 voxels were extracted
NaOH-activated aluminosilicate materials such as fly ash
from different areas of the original synchrotron XRT
and metakaolin is a sodium aluminosilicate (N-A-S-H) gel
reconstruction corresponding to the sample size used,
[5,31,32]. These micrographs demonstrate the presence of
in order to determine the influence of the size of the
representative volume element (RVE) on the accuracy
(a) of the extracted pore structure features. Representative
microstructures for the selected RVE sizes (503, 1003,
2003 and 3003 voxels) are shown in Figures 2(a1 to d1).
The process of quantification of phase volumes from
the representative microstructures requires accurate
segmentation. Phase segmentation in heterogeneous
materials such as cementitious systems with minimal
greyscale contrast between the component phases is
a difficult problem[10]. For fly ash-based geopolymers,
the difficulty is exacerbated by the presence of partially
reacted/unreacted fly ash particles, compositional
aspects of fly ash particles that make some appear
brighter and some less so, and the relative densities of
(b) the N-A-S-H gel based on its location of formation in the
microstructure.
In this study, the phase segmentation process was
implemented in Avizo FireTM software package that
employs advanced filters and operators for image
processing and analysis. First, a transition point (in the
grey scale histogram) where a small increment in the
threshold value causes a sharp change in the detected
phase quantities, is employed. However, this global
thresholding method alone was found not completely
reliable for the fly ash-based geopolymer systems by virtue
of a wide range of greyscale intensities and the presence
of pores within some of the fly ash particles. Therefore,
Fig. 1: Backscattered scanning electron micrographs of 8M the transition point-based thresholding is augmented with
NaOH activated fly ash after 2 days of heat curing at 60 °C: (a) the application of appropriate discrete thresholding in the
lower (800X) and (b) higher (1200X) magnification. The scale bars microstructure to obtain the most realistic representation
correspond to 20 µm.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

326 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
  Advanced Experiments and Models to Predict the Effective Elastic Properties of Cementitious Systems: Application to a Geopolymer

Fig. 2: Representative microstructures of different sizes. Figures

a1-d1 shows cubical volumes of interest extracted from the XRT
dataset corresponding to 503, 1003, 2003, and 3003voxels respec-
tively. Figures a2-d2 shows the segmented 3D pore structure
corresponding to the above mentioned RVE sizes. Please refer
to the online version of the paper for color images.(800X) and
(b) higher (1200X) magnification. The scale bars correspond to
20 µm.
Fig. 3: 3D segmented (volumetric) and connected (percolated)
of known microstructural features such as rounded porosities
unreacted fly ash particles, cenospheres, and N-A-S-H
gel around partially reacted fly ash particles. The discrete The average connected porosity decreases slightly as the
operations were carried out for 2D images in the 3D stack RVE size is increased from 503 to 2003 voxels; however
using routines implemented in the analysis software. no further change is observed with increasing size. For
Figures 2(a2-d2) show the thresholded microstructures RVE sizes of 3003 and 4003 voxels, the variability in the
corresponding to the different RVE sizes, where the segmented and connected porosities is very low, signifying
segmentation algorithm described above has been the applicability of such RVE sizes for phase quantification
implemented to separate the pore and solid phases. and property estimation. It is also noted from Figure 3
that the connected porosity is significantly lower than the
Figure 3 depicts the influence of the RVE size on porosity segmented (overall) porosity, attributed to the hollow fly
of the fly ash-based geopolymer. A number of RVEs of ash particles.
different sizes were extracted from different regions of the
original 3D XRT image. For the RVE sizes of 503, 1003, 2003, Figure 4 demonstrates the application of the segmentation
3003 and 4003 voxels, the number of 3D images analyzed procedure that employs discrete thresholding based on
were 40, 20, 10, 5 and 4 respectively. Here, segmented known microstructural features for multiple phases.
porosity refers to the extracted porosity obtained after Figure 4(a) is the original 3003 voxel image from XRT,
segmentation and it includes all the connected and isolated while Figure 4(b) shows the pore (red) and solid (grey)
pores (e.g., fly ash cenospheres) in the microstructure. phases segmented using the procedure similar to the
On the other hand, connected porosity is the pore cluster one described in the discussion relating to Figure 2.
that percolates in all three orthogonal directions. While Figure 4(c) exhibits the 3D XRT image after multiple
the total volumetric porosity influences the mechanical thresholding. Here three phases (pores in red, fly ash in
properties of the material, the connected porosity yellow, and the reaction product in grey) are shown. The
significantly influences the durability characteristics unreacted and partially reacted fly ash phases could
of the material such as its resistance to moisture and not be separated due to indistinguishable greyscale
ionic transport. The percolated porosity is calculated intensities. While the sequential thresholding method
using a pore-centroidal algorithm implemented in Avizo could provide an indication of the relative quantities of
FireTM. The algorithm involves determination of the pore- these phases, a closer observation to compare several
network centroids within each slice, and then tracking its random thresholded 2D slices with the actual XRT images
movement between each slice in all three directions. The showed that some fly ash particles were labeled as N-A-
summation of total volume of such connected pathways is S-H gel while the actual gel around the fly ash particles
divided by the total sample volume to obtain the connected in some images were labeled as part of fly ash. Thus
porosity. the classification and identification of such phases using
method of edge detection in synchrotron XRT images is
Figure 3 shows the influence of the RVE size on the challenging due to low absorption contrast in available
segmented and connected porosities of the geopolymer x-ray energy range. Hence the use of synchrotron XRT
paste. The average segmented porosity remains relatively data is limited to quantification of 3D pore structure in
invariant of the RVE size even though the variability is this study. A quantification of volume fractions of distinct
much larger, as expected, when smaller RVEs are used.

Organised by
India Chapter of American Concrete Institute 327
Sessioin 3 B - Paper 5

Fig. 4: (a) Sequential, discrete thresholding implemented on a

3003 voxel RVE obtained from XRT: (a) original XRT image, (b)
3D image after pore segmentation (two-phase; pore and solid),
and (c) 3D image after further segmentation for two different
solid phases (fly ash and N-A-S-H gel).

solid phases, towards mechanical property prediction, is

carried out using several high resolution backscattered
SEM images on plane polished samples which are
presented in detail in a forthcoming section.

Micromechanical Behavior of the Fly Ash-Based

Geopolymer and Modeling
In this section, the mechanical performance of the
individual phases in the geopolymer system is evaluated
using nanoindentation. The individual phase responses are
homogenized using established micromechanical theories
to obtain predictions of elastic properties of the geopolymer,
which is then validated through macroscale tests.

Nanoindentation for component phase elastic properties

For the classification of different phases present in alkali-
activated fly ash paste and a quantification of Young’s
modulus of different phases, a statistically significant
number of indentations were performed on carefully
chosen solid regions in the paste through microscopic
observation of the indentation locations. This ensures
that the material microstructure within each interaction
volume can be estimated with a high degree of confidence, Fig. 5: (a) Load-penetration depth curves for different phases in
and the validity of isotropicity and homogeneity within the fly ash-based geopolymer, and (b) Elastic modulus as a func-
tion of the penetration depth for the four distinct solid phases.
the interaction volume can be ensured. Penetration
depths of up to 500 nm were employed to ensure that the
measured elastic properties are not influenced by the obtained from experimental measurements by employing
sample preparation process that could introduce surface a bin size (b) of 1 GPa which is presented in Figure 6(a)
effects. Representative load- penetration depth plots to (symbols). Four characteristic peaks are observed in the
identify the phases of differing stiffness (the methodology histogram. Statistical deconvolution [30,33,34] is applied
to assign phases is explained below) are presented in to the elastic modulus histogram to obtain individual
Figure 5(a). The elastic modulus values as a function phase elastic properties. Assuming Gaussian distribution
of penetration depth were calculated using the CSM for all the four phases (which is a reasonable assumption
technique described in the experimental program section based on the data points in Figure 6(a)), the theoretical
and representative curves are shown in Figure 5(b). For probability distribution function (pdf) is given as:
all the measurements, average elastic modulus in a -Q x - n r V
4 2
1
C (x ) = | f r exp ............................(1)
penetration depth range of 100-to-200 nm is computed r=1 2rs 2r 2s 2r
(since the E is invariant of depth in this range) and used for
further analysis. The mean Young’s modulus, along with In Equation 1, x represents the elastic modulus, fr is the
a phase quantification procedure described later, is used volume fraction of the r th phase, and sr and sr are the mean
in a homogenization scheme for quantification of bulk and standard deviation of the r th phase. The deconvolution
Young’s modulus of the fly ash-based geopolymer. algorithm involves random seeding using a Monte Carlo
simulation and minimization of quadratic deviations
The experimental probability density functions (PDFs) are

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

328 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
  Advanced Experiments and Models to Predict the Effective Elastic Properties of Cementitious Systems: Application to a Geopolymer

between the experimental and theoretical PDFs to find but since an estimation of the gel porosity is tedious, the
the relevant parameters for the Gaussian distribution measured elastic modulus is considered to be the value
functions. The sum of the deconvoluted peaks is identified for the N-A-S-H gel considered virtually to be a “solid”.
using the solid line in Figure 6(a). Similar values of elastic moduli for these phases have
been reported in previous studies [34,35].
Figure 6(b) shows the deconvoluted PDFs for the four
different phases. SEM observation of the microstructure A comprehensive evaluation of the microstructure
at the indented locations, along with an understanding was carried out towards attributing the component
of the phases present in fly ash-based geopolymers microstructural phases to the two remaining deconvoluted
based on past studies, was used for the assignment of peaks of the elastic modulus distribution. The peak
the deconvoluted peaks to the individual peaks. The solid corresponding to the lower among the two peaks
phase with the highest elastic modulus can be attributed corresponds to the fly ash particles that have reacted
to the unreacted fly ash present in the matrix, while the partially, and the reaction products are retained around
phase with the lowest elastic modulus can be attributed or close to the particles, thus providing them with a higher
to the N-A-S-H gel that is formed through diffusion of the elastic modulus than the N-A-S-H gel that has diffused
ionic species and gel precipitation in the empty spaces in away from the particles. These phases are referred to as
the microstructure. This gel is intrinsically nanoporous, partially reacted fly ash. Similar values for this phase have
been reported in[34,35]. The identification of the other peak
was somewhat difficult. The indentation locations were
mostly on fly ash particles and it is hypothesized here
that the peak could be attributed to particles with some
reaction products on the surface, or to unreacted fly ash
particles with cavities, many of which are observed in the
micrographs. Image analysis on backscattered SEM to
quantify these phases have been reported in a forthcoming
section. The deconvolution results are presented in Table 2.

Table 2
Elastic properties of individual solid phases and their
frequency of occurrence in the microstructure as
determined by nanoindentation

Partially
Partially
N-A-S-H activated fly Unreacted
activated fly
gel ash or fly ash fly ash
ash
(Phase 1) with cavities (Phase 4)
(Phase 2)
(Phase 3)

Elastic
modulus 16.3 ± 4.1 32.4 ± 5.1 43.7 ± 2.8 72.0 ± 6.8
[GPa]

Frequency
of 60.4% 22% 8.2% 9.4%
occurrence
(in the solid
phases)

Mean-field homogenization methods for elastic property

prediction
Homogenization procedures are applied to
heterogeneous materials such as cementitious systems
to determine the homogenized elastic properties using
Fig 6: (a) Distribution of nanoindentation-based Young’s modulus individual phase properties (e.g., determined from
(raw data) in the fly ash- based geopolymer sample (symbols) nanoindentation) and phase amounts (e.g., determined
and sum of deconvoluted component peaks (solid line), and (b) from XRT or SEM) to be used in material design. Mean
deconvoluted component peaks for the four distinctly identifiable field homogenization techniques are based on Eshelby’s
microstructural phases in the solid component of the paste. tensors [36] to determine the average stresses and
The area under each deconvoluted peak is the fraction of that strains in inclusions embedded in an elastic matrix. Two
respective solid phase in all solids in the paste.

Organised by
India Chapter of American Concrete Institute 329
Sessioin 3 B - Paper 5

homogenization methods are employed in this study: (1) A three-step homogenization is performed as shown in
Mori-Tanaka scheme[15],(2) double inclusion model [16]. Figure 7(a). In Step I, unreacted fly ash is the inclusion in
The analytical schemes of these models have been well a homogeneous N-A-S-H matrix, and these two phases
documented [15,16,37–40,30]. are homogenized. In Step II, partly reacted fly ash is
incorporated into the homogenized medium. In Step
Mori-Tanaka method has been used for determination of
III, pores are added as inclusions to the homogenized
effective properties in cement-based materials [14,40,30].
medium from Step II.
It considers a discrete spherical inclusion embedded in
an infinitely extended homogeneous reference medium While the Mori-Tanaka model consists of an ellipsoidal
(matrix). In this method, the homogenized bulk and shear inclusion in an infinitely extended homogeneous
moduli for two-phase materials can be quantified from reference medium, a modification implemented in the
the individual phase properties as given in Equation 2. The double inclusion model consists of an ellipsoidal inclusion
individual phase shear and bulk moduli are obtained from v
S = "v
of stiffness
-1
S I-1 + f1 a + f2 b% in another ellipsoidal matrix of
embedded
-1

their elastic modulus and Poisson’s ratio. v

S = "v
stiffness
-1
S I-2,+which 2 b%
f1 a + fis
-1
further embedded in an infinitely
extended homogeneous mediumv "v
of   + f1 a + f2 b% in
S   interaction
fi k i T 1 + a ref $ S k i - 1 X Y
-1 -1
k -1 S=
ref
addition to matrix-inclusion interactions considered in the
k hom =
fi T 1 + a ref $ S k i - 1 X Y
k -1
.............................(2a) Mori-Tanaka approach[16]. The average elastic moduli can
ref be calculated as:
v
S = "vS + f1 a + f2 b %
-1 -1 ................................................(4)

fi n i T 1 + b ref $ S n ref - 1 X Y
-1
ni In the above equation, f1 and f2 are volume fractions of two
inclusion phases, and α and β are functions of f1 and f2 and
v -1"v v v
n hom =
fi T 1 + b ref $ S n ref - 1 X Y v v "
S = " S + fI-11 a + f2 b%I-2 % %
-1
ni ...........................(2b) -1 -1-1 -1 -1
the Eshelby’s tensorsS = , S ,
+Sand
f
= 1 a S
+ f 2+.
b Detailed
f 1 a + f 2 bderivation
and analysis procedure are described in [16,39,40]. In line with
Here i is the number of inclusions; fi is the volume fraction microstructural observations, here unreacted fly ash is
of ith inclusion; ki and kref are the bulk moduli for the ith considered to be embedded in partially reacted fly ash, and
inclusion and reference medium respectively; μi and the double inclusion of unreacted and partially reacted fly
μref are shear moduli for the ith inclusion and reference ash particle is embedded in an infinitely extended N-A-
medium respectively and αref and αref are given in S-H gel. The obtained homogeneous property of the solid
Equation 3[30,33]. A detailed methodology to extract the phase is homogenized with the pore phase using the Mori-
component volume fractions is presented in the following Tanaka scheme to obtain the overall homogenized Young’s
section. modulus, as shown in Figure 7(b).
3k ref .....................................................(3a)
a ref = 3k + 4n Determination of volume fractions to be used in the
ref ref
homogenization schemes
The mean-field homogenization schemes that are used
6k ref + 12n ref for the estimation of macro-level properties require
b ref = 15k + 20n .................................................(3b)
ref ref precise input in terms of phase volume fractions. It was
mentioned earlier that synchrotron XRT delivers good
From homogenized bulk and shear modulus, the contrast between the solid and the pore phases (at the
homogenized Young’s modulus can be calculated as resolution considered), which makes it an attractive tool
[30,33]: for 3D characterization of the pore structure. Thus, the
9k hom $ n hom ......................................................(4) overall pore volume determined from XRT is directly
E hom = 3k + n
hom hom used in the homogenization schemes for elastic property

Fig. 7: Schematic illustration of the homogenization process according to: (a) Mori-Tanaka method, and (b) double inclusion method

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

330 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
  Advanced Experiments and Models to Predict the Effective Elastic Properties of Cementitious Systems: Application to a Geopolymer

prediction. However, identification and quantification of corresponding volume fraction obtained for this phase
different solid phases in the fly ash-based geopolymer was attributed completely either to the partly reacted fly
systems (especially distinguishing partially reacted and ash, or to the hollow fly ash particles during the analysis.
unreacted fly ash phases) is challenging for reasons Hence, this volume was added to that of partially reacted
described earlier in this paper. Hence, the XRT information fly ash or to that of unreacted fly ash for convenience,
is supplemented in this study through the determination thereby effectively reducing a four-phase solid to a three-
of volume fractions of different solid phases using two phase solid as shown in Figures 7(a) and (b). The smaller
approaches: (i) frequency of occurrences of different gel pores, which will be a part of N-A-S-H gel are also not
phases using statistical nanoindentation technique considered.
(Figure 6b, where the integrated intensities correspond
to the volume fraction of the phase assigned to that Homogenized elastic modulus and comparison with
peak), and (ii) multi-label thresholding (combination of macro scale experiments
global thresholding using a transition point approach and Table 3 shows the homogenized Young’s modulus for
discrete thresholding considering known microstructural fly ash-based geopolymer pastes computed using the
features of fly ash-based geopolymers) of several high Mori-Tanaka approach and the double inclusion method.
resolution backscattered SEM images. The multi-label A comparison of homogenized elastic moduli based
thresholding was selected based on global thresholding on phase volume fractions predicted by the SEM and
and/or based on known microstructural features such nanoindentation-based approaches are also provided.
as: (a) unreacted fly ash particles which are rounded Table 3 also shows the macroscale flexural elastic
and without any reaction products surrounding them, modulus of the pastes determined from three-point
(b) partially (in some cases, almost fully) reacted fly bend tests on geopolymer pastes. The tensile elastic
ash where the particles have reacted to form the gel modulus was extracted from the flexural load-deflection
and the reaction products are retained around or close data using a moment- curvature-based inverse analysis
to the particles, (c) pores both in the matrix as well methodology described in detail in[41,42], which is also
as part of the fly ash particles (cenospheres) which shown in Table 3.
are generally distinguishable easily through global
The macro-scale estimates of elastic modulus are
thresholding method using a transition point approach,
in good agreement with those predicted using mean
and (d) N-A-S-H gel which is the remaining solid phase
field microstructural homogenization. Mori-Tanaka
after considering (a) and (b). Figures 8 (a) and (b) show
approach provides a higher value of Young’s modulus as
a representative micrograph and its color thresholded
compared to double inclusion method, because of the
version respectively, showing all the distinct phases.
inter-inclusion interactions not being considered. The
The porosity obtained from 3D synchrotron XRT using the double inclusion model captures the heterogeneity of
analysis scheme described earlier was used to infer the fly ash-based geopolymer more accurately, especially
volume fraction of pores in the bulk material. It needs to be with its capability to consider reaction products around
noted that the smaller peak in the deconvoluted spectrum fly ash particles. Results of this comprehensive study
of the nanoindentation results that was attributed to partly establish the material property-microstructure link
reacted fly ash/hollow fly ash particles (Figure 6b) could for fly ash-based geopolymers using a combination of
not be isolated with a desirable degree of confidence synchrotron XRT, statistical nanoindentation and mean-
based on careful observation of micrographs. Hence the field homogenization models.

Fig. 8: (a) Representative backscattered SEM of a fly ash-based geopolymer paste, (b) multi- phase threshold applied to the SEM
image for phase identification and quantification

Organised by
India Chapter of American Concrete Institute 331
Sessioin 3 B - Paper 5

Table 3
Homogenized elastic modulus and the experimental values from flexural test on bulk specimens

Homogenized elastic modulus (GPa)

Macro-scale measurement of Young’s
modulus (GPa)
Volume fractions from SEM Volume fractions nanoindentation

Mori-Tanaka Double inclusion  Mori- Tanaka Double inclusion Flexural Tensile

16.7 ± 2.9 15.1 ± 2.1 17.5 15.6 14.7 ± 0.7  14.9 ± 0.7

Conclusions contents of this paper reflect the views of the authors who
are responsible for the facts and accuracy of the data
A comprehensive characterization of the pore- and
presented herein, and do not necessarily reflect the views
micro-structure of a fly ash-based geopolymer has been
and policies of the funding agency, nor do the contents
reported here. Pore structure-based determination of the
constitute a standard, specification, or a regulation.
elastic properties of solid microstructural phases using
nanoindentation, and the use of homogenization models
to predict the bulk elastic properties were detailed. References
Synchrotron XRT was used to extract the 3D information 1. Rashad AM. A comprehensive overview about the influence of
different admixtures and additives on the properties of alkali-
of the pore structure. The influence of the size of the RVE activated fly ash. Mater Des 2014;53:1005–25.
on the total pore volume and tortuosity has been brought
2. Ravikumar D, Peethamparan S, Neithalath N. Structure and
out. The XRT data showed that pores in the size range 10 strength of NaOH activated concretes containing fly ash or GGBFS
μm-to- 20 μm contributes the most to the total porosity. as the sole binder. Cem Concr Compos 2010;32:399–410.

A multi-label thresholding method was implemented on 3. Chithiraputhiran S, Neithalath N. Isothermal reaction kinetics and
temperature dependence of alkali activation of slag, fly ash and
several high resolution SEM images for identification and their blends. Constr Build Mater 2013;45:233–42.
quantification of solid phase volume fractions. The phase
4. Lecomte I, Henrist C, Liégeois M, Maseri F, Rulmont A, Cloots
volumes thus calculated were also validated through R. (Micro)-structural comparison between geopolymers, alkali-
frequency of occurrences of different solid phases from activated slag cement and Portland cement. J Eur Ceram Soc
statistical nanoindentation. The individual phase elastic 2006;26:3789–97.
properties determined using nanoindentation along with 5. Peng Z, Vance K, Dakhane A, Marzke R, Neithalath N. Microstructural
the pore and solid phase volume fractions were used in and 29Si MAS NMR spectroscopic evaluations of alkali cationic
effects on fly ash activation. Cem Concr Compos 2015;57:34–43.
multi-step micromechanical mean-field homogenization
models to determine the homogenized Young’s modulus 6. Ma Y, Hu J, Ye G. The pore structure and permeability of alkali
activated fly ash. Fuel 2013;104:771–80.
of the composite. The homogenized Young’s modulus,
especially the one determined using the double inclusion 7. Ma H, Li Z. Realistic pore structure of Portland cement paste:
experimental study and numerical simulation. Comput Concr
method, was found to be in good agreement with the 2013;11:317–36.
experimental macro-scale Young’s modulus. This study,
8. Igarashi S, Chen W, Brouwers HJH. Comparison of observed
linking the microstructure and properties of a highly and simulated cement microstructure using spatial correlation
heterogeneous material (at the microscale), and the use functions. Cem Concr Compos 2009;31:637–46.
of up-scaling models to predict the bulk elastic response, 9. Gallucci E, Scrivener K, Groso A, Stampanoni M, Margaritondo G. 3D
is expected to lead to better material design of fly ash- experimental investigation of the microstructure of cement pastes
based geopolymers, which currently relies greatly on using synchrotron X-ray microtomography (μCT). Cem Concr Res
2007;37:360–8.
experimentally obtained bulk mechanical response only.
10. Provis JL, Myers RJ, White CE, Rose V, van Deventer JSJ. X-ray
microtomography shows pore structure and tortuosity in alkali-
Acknowledgements activated binders. Cem Concr Res 2012;42:855–64.

The authors gratefully acknowledge the National Science 11. Da Silva WRL, Němeček J, Štemberk P. Methodology for
nanoindentation-assisted prediction of macroscale elastic
Foundation (NSF) for the partial support for this research properties of high performance cementitious composites. Cem
(CMMI: 1068985 and CMMI: 1353170). This research Concr Compos 2014;45:57–68.
was conducted in the Laboratory for the Science of 12. Hu C, Li Z. Micromechanical investigation of Portland cement paste.
Sustainable Infrastructural Materials, the Structural Constr Build Mater 2014;71:44–52.
Engineering Laboratory, the Mechanical Behavior of 13. Constantinides G, Ulm F-J. The effect of two types of C-S-H on the
Materials Facility, and 4D Materials Science Laboratory elasticity of cement-based materials: Results from nanoindentation
at Arizona State University and the supports that have and micromechanical modeling. Cem Concr Res 2004;34:67–80.
made these laboratories possible are acknowledged. The 14. SorelliL ,ConstantinidesG,UlmF-J,ToutlemondeF.Thenano-
2-BM beamline of the Advanced Photon Source (APS) at mechanical signatureofUltr aHigh Per formance Concrete
Argonne National Laboratory is also acknowledged. The by statistical nanoindentation techniques. Cem Concr Res
2008;38:1447–56.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

332 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
  Advanced Experiments and Models to Predict the Effective Elastic Properties of Cementitious Systems: Application to a Geopolymer

15. Mori T, Tanaka K. Average stress in matrix and average elastic energy 29. Singh SS, Schwartzstein C, Williams JJ, Xiao X, De Carlo F, Chawla
of materials with misfitting inclusions. Acta Metall 1973;21:571–4. N. 3D microstructural characterization and mechanical properties
of constituent particles in Al 7075 alloys using X-ray synchrotron
16. Hori M, Nemat-Nasser S. Double-inclusion model and overall
tomography and nanoindentation. J Alloys Compd 2014;602:163–74.
moduli of multi-phase composites. Mech Mater 1993;14:189–206.
30. D a Silva WRL, Nemecek J, Štemberk P
17. Williams JJ, Flom Z, Amell AA, Chawla N, Xiao X, De Carlo F. Damage
Applicationofmultiscaleelastichomogenizationbasedon
evolution in SiC particle reinforced Al alloy matrix composites by
nanoindentation for high performance concrete. Adv Eng Softw
X-ray synchrotron tomography. Acta Mater 2010;58:6194–205.
2013;62–63:109–18.
18. De Carlo F, Tieman B. High-throughput x-ray microtomography
31. Provis JL, Van Deventer JSJ. Geopolymers: structures, processing,
system at the Advanced Photon Source beamline 2-BM. vol. 5535,
properties and industrial applications. Elsevier; 2009.
2004, p. 644–51.
32. Garcia-Lodeiro I, Palomo A, Fernández-Jiménez A, Macphee DE.
19. Dowd BA, Campbell GH, Marr RB, Nagarkar VV, Tipnis SV, Axe L, et
Compatibility studies between NASH and CASH gels. Study in the
al. Developments in synchrotron x-ray computed microtomography
ternary diagram Na 2 O–CaO–Al 2 O 3–SiO 2–H 2 O. Cem Concr
at the National Synchrotron Light Source. vol. 3772, 1999, p. 224–36.
Res 2011;41:923–31.
20. Rivers ML. tomoRecon: High-speed tomography reconstruction on
33. Constantinides G, Ravi Chandran KS, Ulm F-J, Van Vliet KJ. Grid
workstations using multi- threading. vol. 8506, 2012, p. 85060U –
indentation analysis of composite microstructure and mechanics:
85060U – 13.
Principles and validation. Mater Sci Eng A 2006;430:189–202.
21. Marone F, Stampanoni M. Regridding reconstruction algorithm for
34. Němeček J, Šmilauer V, Kopecký L. Nanoindentation characteristics
real-time tomographic imaging. J Synchrotron Radiat 2012;19:1029–
of alkali-activated aluminosilicate materials. Cem Concr Compos
37.
2011;33:163–70.
22. Bossa N, Chaurand P, Vicente J, Borschneck D, Levard C, Aguerre-
35. Němeček J, Šmilauer V, Kopeckỳ L. Characterization of alkali-
Chariol O, et al. Micro- and nano-X-ray computed-tomography: A
activated fly-ash by nanoindentation. Nanotechnol. Constr. 3,
step forward in the characterization of the pore network of a leached
Springer; 2009, p. 337–43.
cement paste. Cem Concr Res 2015;67:138–47.
36. Eshelby JD. The Determination of the Elastic Field of an Ellipsoidal
23. Zhang M, He Y, Ye G, Lange DA, Breugel K van. Computational
Inclusion, and Related Problems. Proc R Soc Lond Math Phys Eng
investigation on mass diffusivity in Portland cement paste based
Sci 1957;241:376–96.
on X-ray computed microtomography (μCT) image. Constr Build
Mater 2012;27:472–81. 37. Miled K, Sab K, Le Roy R. Effective elastic properties of porous
materials: Homogenization schemes vs experimental data. Mech
24. Monteiro PJM, Kirchheim AP, Chae S, Fischer P, MacDowell AA,
Res Commun 2011;38:131–5.
Schaible E, et al. Characterizing the nano and micro structure of
concrete to improve its durability. Cem Concr Compos 2009;31:577–84. 38. Hu GK, Weng GJ. The connections between the double-inclusion
model and the Ponte Castaneda– Willis, Mori–Tanaka, and Kuster–
25. Garboczi EJ, Bentz DP. The effect of statistical fluctuation, finite size
Toksoz models. Mech Mater 2000;32:495–503.
error, and digital resolution on the phase percolation and transport
properties of the NIST cement hydration model. Cem Concr Res 39. Yang CC, Huang R. Double inclusion model for approximate elastic
2001;31:1501–14. moduli of concrete material. Cem Concr Res 1996;26:83–91.
26. Oliver W c., Pharr G m. Measurement of hardness and elastic 40. Yang CC. Approximate Elastic Moduli Of Lightweight Aggregate.
modulus by instrumented indentation: Advances in understanding Cem Concr Res 1997;27:1021– 30.
and refinements to methodology. J Mater Res 2004;19:3–20.
41. Das S, Aguayo M, Sant G, Mobasher B, Neithalath N. Tensile Behavior
27. Li X, Bhushan B. A review of nanoindentation continuous stiffness in the Fracture Process Zone of Blended Binders Containing
measurement technique and its applications. Mater Charact Limestone Powder. Cem. Concr. Res 2015; 73:51–62.
2002;48:11–36.
42. Mobasher B, Dey V, Cohen Z, Peled A. Correlation of constitutive
28. Moody NR, Gerberich WW, Burnham N, Baker SP. Fundamentals of response of hybrid textile reinforced concrete from tensile and
nanoindentation and nanotribology. Warrendale, PA (United States); flexural tests. Cem Concr Compos 2014;53:148–61.
Materials Research Society; 1998.

Narayanan Neithalath
Narayanan Neithalath is an Associate Professor in the School of Sustainable Engineering and the Built
Environment at Arizona State University, Tempe AZ. He received his PhD in Civil Engineering Materials
from Purdue University in 2004. His research interests are in the areas of sustainable construction
materials including high volume cement replacement materials for concrete, development of novel
materials including unconventional cement-less binders through carbonation and alkaline activation,
characterization and modeling of microstructure and properties of such systems. He has received several
awards for his work on novel cementitious systems. He is the Editor of the Cementitious Materials section
of the ASCE J of Materials in Civil Engineering and an Editorial Board Member of Cement and Concrete
Composites.

Organised by
India Chapter of American Concrete Institute 333
Session 3 B - Paper 6

Evolving Acceptance Criteria for Concrete Durability Tests In

Construction Projects
Manu Santhanam, Sarath Kumar, Ravindra Gettu and Radhakrishna Pillai
Department of Civil Engineering, IIT Madras

Abstract from these samples. In-place testing, while being most

desirable as it a true reflection of the concrete inside the
Durability of concrete is affected by factors like binder
structure, is very difficult because of the poor reliability
content, water to binder ratio, aggregate grading,
of tests available for the same. Further, the in-place
admixture compatibility etc. Durability of concrete can
measurement can be affected by a number of factors,
be measured by different tests based on the mechanism
including temperature, moisture, and other site-related
of deterioration. Tests like Rapid chloride permeability
issues that are difficult to control.
and water permeability are mainly used in India for
qualification of concrete in construction projects. The Bickley et al. (2006) provide a good overview of the various
criteria for acceptance are largely based on a target durability tests, their strengths and weaknesses. It is
requirement for these parameters. The lack of acceptance generally accepted that no single test can assess the
criteria for durability tests is a major impediment in the long-term durability of concrete. Further, there is also
implementation of durability specifications. This study is an agreement on the fact that the test should adequately
an attempt to present a way forward for the development reflect the transport mechanisms that pertain to the
of acceptance criteria for rapid chloride permeability test specific service environment. The use of durability tests
and water permeability test. also calls for the provision of criteria for acceptance of the
concrete. This can be a tricky process. Typical durability
Keywords: Concrete durability; RCPT; Water penetration;
criteria used in some major Indian codes and construction
Performance specifications; Acceptance criteria
projects, along with examples of field experience in North
America (from Bickley et al. 2006) are provided in Tables
Introduction 1 to 3.
Durability is the measure of concrete’s resistance for The data in these tables shows that the primary tests
deterioration. While durability pertains to the long term covered in the Indian codes and specifications are DIN 1048
performance of concrete in a service environment, water permeability test and rapid chloride permeability
the testing for durability is performed at 28 days (and test. On the other hand, the North American experience
sometimes 56 / 90 days), in a manner similar to the primarily deals with the rapid chloride permeability test.
compressive strength. Several durability index tests In most of the cases, only limiting values are prescribed,
are available for concrete, which typically address the and there is no clear mention of what would be the
penetration of water, gas, or ionic species into concrete. acceptance criteria in the case of day to day sampling of
These tests are often used in construction projects that use the production concrete (except in the last case in Table
specifications for the durability performance of concrete. 3). In other words, the application of durability testing
The application of durability tests in a construction project in construction projects, especially in India, is oriented
could be in the following ways: towards single valued acceptance i.e. when a concrete
1. Prequalification tests on concrete mixtures, specified with a particular test value is not satisfying
the test result, the concrete is deemed to be rejected.
2. Routine testing on samples at the delivery point, or Achieving specified value is practically difficult due to
3. In-place testing. the variability associated with the testing procedure and
construction practices. The acceptable level of variability
The prequalification tests are routinely used in the of the test results on the field is to be established. This
selection of materials and mixture proportioning to achieve necessitates a careful analysis of data from several
the specified durability. However, the performance in such research reports, which is presented in the next section.
tests has no bearing on the actual quality of concrete that
is delivered at site, and this leads to the need for sampling
concrete from the construction site on a routine basis, Acceptance Criteria for Durability Tests
and performing durability tests on specimens prepared This section explores the international developments with

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

334 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolving Acceptance Criteria for Concrete Durability Tests In Construction Projects

Table 1
Durability clauses in various Indian concrete codes

Code Organisation Highlights

Ministry of Water Permeability Test specified in the same code is mandatory for all RCC/PSC bridges under severe, very
IRS 1997
Railways severe and extreme environment s; limit for the depth of penetration of moisture is 25 mm for all cases
Regarding the tests, the code says “there is no specified test method for durability which can be completed
within a reasonably short time”
Indian Roads For high performance concrete (HPC), Rapid Chloride Permeability Test (ASTM C 1202) and Water
IRC 112 :2011
Congress Permeability Test (DIN 1048 part 5) or Initial Surface Absorption Test (BS 1881 part 1) can be specified
Upper limits for total charge passed in RCPT for the exposure conditions, such as severe (1500 Coulombs),
very severe (1200 Coulombs) and extreme (800 Coulombs) conditions are provided.
Min. of Surface Tests and standards of acceptance: cube compressive strength, chloride and sulphate content, density of
MOST / Transport/ Min. fresh concrete and hardened concrete, permeability tests
MoRTH of Road Transport Water permeability test on cylindrical specimen is specified (Section 1716.5) – the maximum permissible
and Highways depth of penetration is 25 mm
The clause on durability says that one of the main characteristics influencing the durability of concrete is
its impermeability to the ingress of water, oxygen, carbon dioxide, chloride, sulphate and other potentially
deleterious substances. Impermeability is governed by the constituents and workmanship employed in
Guidelines making the concrete
for the use Ministry of The acceptance tests specified – Compressive Strength, Rapid Chloride Ion Permeability test (ASTM C-1202 or
of HPC in Railways AASHTO T-277), Water Permeability test as per DIN: 1048 Part 5-1991 or Initial Surface Absorption test as per
bridges BS:1881 Part 5
The permissible value of chloride-ion permeability is 800 Coulombs. The permissible values in water
permeability and surface absorption test shall be decided taking into account the severity of the exposure
conditions
From the consideration of strength, the acceptance criteria of concrete used for construction of structures
should be as per IS: 456 and other related codes. The permeability requirements of concrete are deemed to
AERB Safety Atomic Energy satisfy when the rapid chloride penetration test (RCPT) value is less than 3000 coulombs. The requirement is
Guide Regulatory Board to be satisfied for 56 days test results on 28 days water cured sample. For containment and water retaining
structures, RCPT value of concrete should be lower than 2000 coulombs. Effort should be made to achieve a
concrete mix with RCPT value lower than the above values.

Table 2
Durability clauses in Metro project specifications

Project Clauses related to durability

Mandatory Test - Cube Compressive Strength Test

Chennai Metro Rail
Additional Test – Permeability test for Concrete as per IS: 3085-1965, Section 1716.5 of MOST Specification and DIN 1048
For all the main structures, water permeability test on concrete sample is specified
Hyderabad Metro Rail
No other details or acceptance criteria regarding the permeability test are given
Mandatory Test - Cube compressive strength test
Kolkata Metro Rail Additional Test – Permeability test for Concrete as per IS: 3085-1965, Section 1716.5 of MOST Specification and DIN 1048
Limiting value of water penetration depth when tested as per DIN is less than 25 mm

Table 3
North American experience in implementing durability testing (Bickley et al. 2006)

Project / specification Test specified Limits prescribed

New Brunswick draft specification for bridges RCPT < 1000 C without corrosion inhibitor
< 1500 C with corrosion inhibitor
Shrinkage < 0.04% at 7 days (superstructure)
< 0.05% at 7 days (substructure)
Calgary city (for high performance concrete) RCPT < 600 C (values of 601-1200 C acceptable with $40/m3
penalty)
Port Authority of New York and New Jersey RCPT For pre-qualification < 1000 C Production concrete < 1500
C in 80% of the tests

Organised by
India Chapter of American Concrete Institute 335
Session 3 B - Paper 6

Table 4
Two tables (a) and (b) on North American data reproduced from Bickley et al. (2006)
(a) Summary of ASTM C 1202 Data from Canadian Projects

CSA A23.1 Mix Class No. of Test Results No. out of spec % Failure Max Coulombs Comments
HPC (C-XL) 785 60 7.6 1,000
HPC+CI 12 No spec - - 2 > 1,500
CI 23 12 52.2 1,500 **
CI/C2 6 No spec - - 5> 1,500**
C2 24 No spec - - 24> 1,500**

CI + CI 2 No spec - - 2> 1,500

*CI = corrosion inhibitor, ** = mixtures without SCMs gave much higher values

(b) Field Sorptivity Data, Ontario Ministry of Transport

Sorptivity: mm/ √min

Finish
Mean Value Standard Deviation
Machine 0.054 0.012
Hand 0.090 0.002

respect to acceptance limits for durability parameters, South African experience

specifically for the RCPT and DIN 1048 water permeability. In South Africa, durability design has been practiced for
Experiences with other test methods are also discussed more than a decade using three principal test methods –
in order to throw more light on the degree of variability in the Chloride Conductivity method, Oxygen Permeability
tests from field-sampled specimens. method, and the Water Sorptivity method. While the
mechanisms and transport properties represented by
North American experience these methods are considerably different compared to the
Referring to the North American experience provided in RCPT and water permeability, the guidance provided on
the report by Bickley et al. (2006), it can be understood limiting values would be of major help in the current project.
(Table 3.7.2.1 and Table 3.7.2.3 reproduced from the report
In a comprehensive evaluation of the variability in on-site
as Table 4 (a) & (b)) that the failures in RCPT in the projects
tests in projects executed by the South African National
considered were less than 7.6%, for a maximum charge
Roads Agency Ltd (SANRAL), Nganga (2011) reported
passed specification of 1000 C. However, the acceptance
the summary of the statistical values obtained from the
criteria specified have not been given in this study, and
projects considered, and this is reproduced in Table 5.
only the performance of samples collected on site was
The acceptance criteria given (to decide on the maximum
analysed. Further, the data also shows the statistics of the
permissible % defectives) were based on the data by
results from field sorptivity measurements from projects
Alexander et al. (2009), which are presented in Table 6.
in Ontario, which indicate a variability of 3 – 18%.

Table 5
Variability in SANRAL projects

Parameter Average CoV (%) Proportion of defectives (%)

OPI 9.8-10.3 1.8-4.6 0 – 40.1

K Values (E -10 m/s) 0.7-1.7 46-85 3-32.7

Sorptivity (mm/√hr) 7.1-9.7 15.4-22.5 3.7-50.5

Porosity (%) 11.5-13 12.7-13.7 -

Cover depth (mm) 46.6-58 14.7-29.4 8.7-19.6

OPI is the oxygen permeability index, K = coefficient of oxygen permeability.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

336 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolving Acceptance Criteria for Concrete Durability Tests In Construction Projects

Table 6
Acceptance criteria followed in SANRAL projects, as suggested by Alexander et al. (2009) (from Nganga, 2011)

Acceptance OPI (log scale) Sorptivity (mm/√ hr) Conductivity (mS/cm)

Laboratory > 10 <6 < 0.75

Full acceptance > 9.4 <9 < 1.00

9 to 12 1.00 to 1.50
As-built Conditional acceptance 9.0 to 9.4
Structures 12 to 15 1.5 to 2.50
Remedial measures 8.75 to 9.0

Rejection < 8.75 > 15 > 2.50

In other independent analyses provided by South African Asian experience

researchers, shown in Tables 7 to 9, further data on the Although evidence from actual projects in Asia is not
three durability index tests are provided. While the data yet extensively available to come up with a realistic idea
in Table 7 pertain to site samples, the results in Table of acceptance criteria, one specific experimental study
8 are from an inter-laboratory exercise, and Table 9 is quite useful in understanding how to go about setting
summarizes a larger set of results from the three tests. limits for the acceptance criteria for chloride permeability
From the South African experience, it is evident that the and water penetration. Both these tests were performed
degree of variability in the results of water and chloride on a number of concretes with different binder types,
based tests is much greater than that for gas based test binder contents and water to binder ratios by Ahmad et
(oxygen permeability). al. (2008). A summary of their proposals for compliance
criteria for RCPT, as well as DIN 1048 water penetration,
is provided in Table 10.
Table 7
Single operator coefficient of variation for site-based samples
(Bouwer, 1998 from Nganga, 2011)
Rapid Chloride Permeability Test
Concrete source OPI
Water Chloride The ASTM C1202, which is the standard test method
sorptivity conductivity prescribed for the RCPT, suggests that “When using
Actual structure (%) 3 13 14 this test for determining acceptability of concrete
mixtures, statistically-based criteria and test age for pre-
Wet cured site mixed qualification, or for acceptance based on jobsite samples,
2 12 7
concrete (%)
should be stated in project specifications. Acceptance
Wet cured ready mixed criteria for this test should consider the sources of
1 7 5
concrete (%)
variability affecting the results and ensure balanced
risk between supplier and purchaser. The anticipated
Table 8 exposure conditions and time before a structure will be
Summary of the range of results from an inter lab exercise put into service should be considered.” In other words, it is
(Grieve et al., 2003, from Nganga, 2011)
imperative for acceptance criteria to be specified explicitly
Water Chloride in the case of a regular testing requirement on site based
OPI
sorptivity conductivity specimens. In terms of precision of the test, the following
is specified:
Max. Repeatability (%) 2.8 17.8 57.4
Single-Operator Precision – The single operator coefficient
Max. Reproducibility (%) 2.8 22.6 39.7 of variation of a single test result has been found to be

Table 9
Precision for various durability index tests (Stanish et al., 2004, from Nganga, 2011)

Water sorptivity test Chloride conductivity test

OPI K value
Sorptivity Porosity CCI Porosity

Repeatability (%) 1.4 32.2 9.9 5.5 9.1 5.5

Reproducibility (%) 1.8 36.6 12.8 6.4 21.1 8.9

Organised by
India Chapter of American Concrete Institute 337
Session 3 B - Paper 6

Table 10
Compliance criteria for RCPT and water penetration, from Ahmad et al. (2008)

Compliance criteria for chloride permeabiity for plain and blended cement concretes

Charge Passed (Coulombs)

Cementitious materials
w/cm
content (kg/m3)
Plain cement concrete Silica fume cement concrete Fly ash cement concrete

350 2500-3000 700-1000 1300-2000

0.35
400 2000-2500 600-1000 1300-1600

350 3000-4000 800-1200 1600-2000

0.40
400 2500-3500 800-1000 1600-1800

300 3500-4500 1000-1500 2000-3000

0.45
350 3500-4000 1000-1500 2000-3000

400 3500-4000 900-1500 2000-2500

0.50 350 5000-6000 2500-3000 2500-4500

400 4000-4500 1500-2500 2200-3500

Compliance criteria for depth of water penetration for plain and blended cement concretes

Water penetration depth (mm)

Cementitious materials
w/cm
content (kg/m3)
Plain cement concrete Silica fume cement concrete Fly ash cement concrete

350 35-45 20-30 20-35

0.35
400 35-40 20-25 20-35

350 40-50 30-40 40-45

0.40
400 40-45 30-35 35-40

300 50-60 40-50 45-50

0.45 350 45-50 35-45 45-50

400 40-45 30-35 35-45

350 75-90 40-50 50-60

0.50
400 70-85 35-45 40-55

12.3%. Therefore, the results of two properly conducted acceptance range for RCPT and strength test is shown in
tests by the same operator on concrete samples from the Table 11.
same batch and of the same diameter should not differ by
Applying acceptance criteria based on strength to
more than 42%.
the durability tests would, therefore, be unjustified.
Multi-laboratory Precision – The multi-laboratory In this context, Obla and Lobo (2007) present a sound
coefficient of variation of a single test result has been methodology to determine the acceptance limit for the
found to be 18.0%. Therefore, results of two properly RCPT. This employs the suggestions of the ASTM C1202.
conducted tests in different laboratories on the same Using a statistical approach with a 99% confidence level,
material should not differ by more than 51%. The average they propose that an average of 5 consecutive tests be
of three test results in two different laboratories should within a limiting value, while any individual test should
not differ by more than 42%. not be greater than 30% of the proposed limiting value.
This is stated as a reasonable modification of the current
Obla and Lobo (2007) report that the variability of results
acceptance limits based on the criteria for compressive
in RCPT is about 4 times that of standard cylinder
strength used in ACI 318. The authors had proposed
compression test. The comparison of variability and

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

338 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolving Acceptance Criteria for Concrete Durability Tests In Construction Projects

Table 11
Precision for various durability index tests (Stanish et al., 2004, from Nganga, 2011)

Test Variability Acceptance Range

Strength test (ASTM-39) 2.9% 8%

RCPT (ASTM C-1202) 12.3% 42%

Table 12
Data from Andrews-Phaedonos (2012) (Precision of various durability test methods)

Water
De.28 day (10-12 Sorptivity 28 day RCPT 28 day Chloride diffusion
permeability VPV28 day (%)
m2/s) (mm/s1/2) (Coulomb) (10-12 m2/s)
(mm)
ASTM ASTM ASTM DIN 1048 NORDTEST AS
C1566 C1585 C1202 (under pressure NT Build 443 1012.21

Repeatability 14% 6% 12.3% N/A 15% (?) 1.8%

Reproductibility 20% NA 18% 16.8% 15% 6.5%

the modification to account for the greater variability RCPT and WPT. The mixes were prepared with different
seen in the RCPT. Using this approach, if the proposed binder contents, w/b, and with various dosages of mineral
limiting value is 1000 C, individual values should be lower admixtures. For each test, three specimens prepared from
than 1300 C. The data provided in Table 3 from the Port the same concrete cylinder were tested. The coefficient of
Authority of New York and New Jersey is also similar in its variation was found to be in the range of 12 – 24%. This
assessment, as it says that the individual result (in at least seems to indicate that the suggestion by ASTM C1202 of
80% of the test) should be less than 1500 C, for a proposed 42% difference (i.e. COV of 21%) between specimens holds
limiting value of 1000 C. good in the case of laboratory tests.
In the case of the DIN water permeability test, the cubes
Water Penetration Test were conditioned by drying at 50 oC in an oven for seven
The standard for the water permeability test (DIN 1048 - days. A total of 24 tests were conducted. The average
Part 5, or EN 12390-8), in contrast to the RCPT, does not coefficient of variation in the depth of penetration readings
provide any criteria to determine the level of precision along a single water profile in the same specimen is found
expected in the test. Research experience with the test to be 22.5%.
has not been conclusive to establish any acceptance
criteria, as reported by Day et al. (2014). The variations Field Experience with Tests
in the test are reported to be high, one example being a
The variability in the test results in the field was assessed
case cited by Day et al. of 65% COV for a study in the UAE.
using the data provided by M/s Chennai Metro Rail Ltd
In fact, Day et al. suggest that the water permeability test
(CMRL) for concrete used in the construction of the Chennai
is not suitable for performance specifications and on-site
Metro. The target RCPT value for the concretes was 600
compliance testing.
Coulombs at 90 days, but there was no clear definition of
In his positive assessment of the VPV (Volume of the acceptance criteria. Applying the equation (1) suggested
permissible voids) test, Andrews-Phaedonos provides by Obla and Lobo (2007), and using a coefficient of variation
data from other research studies, and a table indicating of 21%, the acceptability limit for an individual test result
the precision of various durability methods is reproduced works out to 1175 Coulombs, while the acceptable average
as Table 12. The data for ASTM C1202 are similar to the of a set of 3 specimens is 835 Coulombs.
cases explored from North America, while the data from P1 ....................................................(1)
DIN 1048 tests indicate an inter-laboratory variability of P2 =
1 - 2.33/ n * V
16.8% (no data is provided for repeatability).
where P2 = Acceptable (critical) value
Data from Research Projects at IIT Madras P1 = Target value (in this case, 600 Coulombs)
In a recent study at IIT Madras (Dhanya, 2015), a total n = number of specimens
of 41 mixes were tested at 28, 56 and 90 days using the
V = Coefficient of variation (assumed as 21%)

Organised by
India Chapter of American Concrete Institute 339
Session 3 B - Paper 6

Using the three test average value of 835 Coulombs and 2. Ahmad, S., Al-Kutti W. A., Al-Amoudi, O. S. B., and Maslehuddin
M., “Compliance Criteria for Quality Concrete”, Construction and
the individual limit of 1175 Coulombs, a review of the data
Building Materials 22, 2008, pp. 1029 – 1036.
supplied by CMRL indicates that all the tests are passing.
3. ALEXANDER, M.G. AND BEUSHAUSEN, H. (2008),"Performance-
Whereas if a single value acceptance of 600 Coulombs is based durability testing, design and specification in South Africa:
considered, several concretes would fail. Alternatively, if latest developments", International Conference on Excellence
coefficients of variation that are typical of compressive in Concrete Construction through Innovation, London, UK, 9-10
strength tests (< 10%) are chosen, several sets of September 2008, CRC Press, pp. 429 - 43
concrete indicate unsatisfactory performance. 4. ALEXANDER, M.G., BALLIM, Y., and STANISH, K. 'A framework
for use of durability indexes in performance-based design and
On the whole, the assessment from lab and field studies specifications for reinforced concrete structures'. Materials &
seem to suggest that 21% COV is valid for results of RCPT, Structures, Vol. 41, No. 5, June 2008, pp. 921-936.
while the method suggested by Obla and Lobo (2007) could 5. Andrews-Phaedonos, F. “Assessment of Concrete Durability
be considered for evolving acceptance criteria for RCPT. Using a Single Parameter with a High Level of Precision – The VPV
Test,” 25th ARRB Conference – Shaping the Future: Linking policy,
In the case of WPT, the target value was 25 mm water Research and Outcomes, Australia, 2012.
penetration. As in the case with RCPT, a single value 6. ASTM C 1202, Standard test method for electrical indication of
acceptance would have meant that several mixes end up concrete’s ability to resist chloride penetration, Annual book of
failing. The data collected from the laboratory study by ASTM standards, West Conshohocken, United States, 2010.
Dhanya (2015) and the data from the CMRL project seem 7. Bickley, J., Hooten, R. D. and Hover, K.C., “NRMCA Guide to
specifying concrete performance Phase II Report of Preparation
to indicate that a coefficient of variation of 35 – 40% can
of a performance based specification for cast-in-place concrete”,
be safely considered for this test. Using this COV and the RMC research foundation and NRMCA P2P steering committee,
method suggested by Obla and Lobo (2007), the individual 2006.
and three test average values are satisfied by each set of 8. Day, K. W., Aldred, J., and Hudson, B., Concrete Mix Design, Quality
data from the CMRL project. Control and Specification, Fourth Edition, CRC Press, 2014, 329 pp
9. DIN 1048 part 5, ‘Testing of concrete, Testing of hardened concrete
(Specimens prepared in mould)’ (German standards, 1991).
Recommendations for Durability Testing and
10. Durability Index Testing Procedure Manual (2009), Concrete
Acceptance Criteria durability index testing, South Africa.
Based on the analysis of the lab and field data presented in 11. Elevated Mass Rapid Transit System through Public Private
the paper, as well as data presented from the international Partnership, Manual of Specifications and standards, Government
experience, the following suggestions should be of Andhra Pradesh, Hyderabad Metro Rail Ltd., Hyderabad, 2008.
considered for active implementation in projects using 12. Guidelines for the use of High Performance Concrete in bridges,
durability tests: Ministry of Railways, Government of India, Research Designs and
Standards Organisation, Lucknow, 2008
1. The limiting values provided by the client may be 13. IRS 1997 Code of practice for plain, reinforced and prestressed
treated as the ‘performance criteria’. In addition to concrete for general bridge construction (Concrete Bridge Code),
these performance criteria, the acceptance limits, Indian Railway Standard IRS 1997, Research Designs and Standards
Organisation, Lucknow, 2003.
to allow for variability in site sampling, must be
specified. While performance criteria are applicable 14. IRC 112:2011 Code of practice for concrete road bridges, IRC 112,
Indian Roads Congress, New Delhi, 2011.
to the pre-qualification laboratory mixes, acceptance
criteria address the results from day-to-day sampling 15. IS 456 (2000), Plain and Reinforced Concrete – Code of Practice
(Fourth Revision), Bureau of Indian Standards, New Delhi.
of concrete as produced for the actual application.
16. Kolkata metro Rail Corporation Limited, East West Metro Project,
Suggested acceptance limits for RCPT and DIN water
Tender documents, Volume 3, Technical Specifications, Kolkata,
permeability can be determined using the methodology 2009.
suggested by Obla and Lobo (2007). 17. Metro Rail project phase-1, volume -2, Structural specifications
2. The specification used for the project should also clearly (Part – 1) and Design basis report (Part – 2), Chennai Metro Rail
Limited, Chennai.
describe the sampling criteria for durability testing.
Further, when mineral admixtures and supplementary 18. MOST /MoRTH Specification for road and bridge works, Ministry
of Surface Transport or Ministry of Road Transport and Highways,
cementing materials are used, the specified testing Government of India, 2000.
ages should also be altered to consider the long-term
19. Nganga, G.W., “Practical implementation of the durability index-
contribution expected from such materials. based performance approach”, M.S. Thesis, University of Cape
Town, South Africa, 2011.
References 20. Obla, K.H. and Lobo, C.L., “Acceptance Criteria for Durability Tests,”
1. AERB Safety Guide No. AERB/NF/SG/CSE-4, Materials for ACI Concrete International, Vol. 29, No. 5, May 2007, pp. 43-48.
Construction of Civil Engineering Structures Important to Safety
of Nuclear Facilities, AERB, 2011.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

340 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Evolving Acceptance Criteria for Concrete Durability Tests In Construction Projects

Manu Santhanam B. Tech, M. S., Ph.D. (Purdue University)

Professor, Department of Civil Engineering, IIT Madras, Chennai – 600036, India
E-mail: manus@iitm.ac.in, manusanthanam@gmail.com
Work Experience: Senior R&D Chemist, Sika Corporation, USA, May 1996 – Nov 1998
Associate Professor, IIT Madras, Oct 2001 – Mar 2009
Associate Professor, IIT Madras, Mar 2009 – Jul 2013
Professor, IIT Madras, Jul 2013 onwards
Professional Affiliations: Life member of Indian Concrete Institute, Member of RILEM
Awards Received
1. Indian Concrete Institute – Prof. V. Ramakrishnan Award for Outstanding Young Researcher in Concrete
Technology, 2006.
2. Indian National Academy of Engineering (INAE) Young Engineer Award 2008
3. Indian Concrete Institute – Tamil Nadu Centre Ultratech Award for Outstanding Young Concrete Engineer
2011.
4. Corps of Engineers Prize from Institution of Engineers (India), 2011.
Research Highlights
• Guided 8 PhD and 5 MS Theses; currently guiding 9 PhD and 3 MS.
• Principal Investigator on research and industry-sponsored projects worth Rs 9 Crores
• Publications – More than 60 refereed journal papers, and more than 50 papers in national and
international conferences; 3 book chapters

Organised by
India Chapter of American Concrete Institute 341
SESSION 3 C
Session 3 C - Paper 1

Some Aspects of Concrete Science and Concrete Technology

Dr S.C.Maiti Raj K. Agarwal

Ex-Joint Director, National Council for Cement Managing Director, Marketing and
and Building Materials, New Delhi Transit (India) Pvt. Ltd., New Delhi

Abstract in order to obtain the desired characteristics of fresh and

hardened concrete. The cementitious material content e.g.
Concrete science and concrete technology have many
(OPC + flyash) or (OPC +g.g.b.s.) needs to be increased,
aspects.This article covers the role of different types of
in order to obtain the desired 28-day compressive
cement and the role of different chemical and mineral
strength of concrete. Depending on the shape, size and
admixtures in producing different characteristics of
grading of aggregates, the quantity of coarse aggregate,
concrete. Different concrete mix proportioning methods
fine aggregate and water are fixed in order to obtain the
using different chemical and mineral admixtures have been
desired workability and compressive strength of concrete.
examined. Deleterious alkali-aggregate reaction in concrete,
and accelerated test methods developed to correlate field
performance have been discussed and recommendations Role of Different Types of Cement in Concrete
for combating the reactions have been presented. Ordinary Portland cement is the basic cement produced
Keywords: Cements,concrete admixtures,concrete mix from cement clinker. The lower grade of OPC, say
proportioning,alkali-aggregate reaction. 33-grade1, has low heat of hydration, and so it can safely be
used in mass concrete construction. High-strength OPC
(53-grade1) is needed to develop high-strength concrete,
Introduction as high as 100 MPa. For precast and pre-stressed
Concrete science and concrete technology play their roles concrete construction, OPC is suitable, so that the pre-
in the construction of concrete structures. Cement is a stressing forces can be released early, or the formwork
binding material and different types of cement produce can be reused early, in case of precast concrete. For such
different characteristics in concrete. Setting times, purposes, blended cements i.e. PPC and PSC are not
heat of hydration and compressive strength of cements suitable, as their rate of hardening is slower than that of
influence the characteristics of fresh and hardened OPC. They are suitable in mass concrete construction e.g.
concrete. The Portland pozzolana cement (PPC) with in the foundation of multi-storeyed buildings, as their heat
flyash as a constituent material and Portland slag cement of hydration is lower than that of OPC 43- grade1 or OPC 53-
(PSC) with ground granulated blast furnace slag (g.g.b.s.) grade. For combating the deleterious alkali-silica reaction
as a constituent material, change the characteristics in concrete, low alkali (less than 0.6% as Na2O equivalent)
of concrete to a great extent. They retard setting times, OPC is recommended. For resisting chloride and sulphate
reduce the rate of hardening and thereby reduce the attack in soil or ground water, PSC having more than 50%
early strength, but they reduce the temperature inside g.g.b.s. is recommended2. The rapid hardening Portland
the concrete. Also PPC, PSC or ordinary Portland cement can be used in the construction of concrete roads,
Cement(OPC)+flyash or (OPC+g.g.b.s.) are recommended as roads can be opened to traffic early. Railway sleeper
in concrete to resist the deleterious alkali-silica reaction manufacturers also use such high-early strength (with
in concrete structures, if aggregates are reactive. minimum 7-day compressive strength of 37.5MPa) and
The mineral admixture silica fume in concrete makes finer OPC (with Blaine’s fineness of 370m2/kg or more).
the concrete stronger and abrasion-resistant. All these
materials including cement should have good compatibility Role of Chemical Admixtures in Concrete
with the chemical admixtures used in concrete. When Chemical admixtures change the characteristics of
compatible, these materials in concrete along with water concrete in the fresh state. There will be increase in
and aggregates make the fresh concrete mix cohesive, the workability of concrete by using plasticizers or
workable and the concrete is placed comfortably in the superplasticizers. Alternately, the superplasticizers
formwork. can cause reduction of the mixing water (for a given
With the specific concrete making materials in hand, the workability), and thereby the water-cement ratio of
concrete mix proportioning procedure needs modifications concrete is reduced, and hence there will be increase

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

344 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Some Aspects of Concrete Science and Concrete Technology

in the 28-day compressive strength of concrete. Also Because of abrasion-resistance characteristics, silica
by reducing both the water content and cement content fume concrete is being used in the spillways of concrete
(keeping w/c ratio constant), for a fixed workability and a dams. In the spillway of the Tehri dam of Uttarakhand, 8%
fixed 28-day compressive strength of concrete, the cement silica fume concrete is performing satisfactorily(Fig.2).
can be saved in concrete. The other admixtures e.g. air-
entraining, set-retarding or set-accelerating admixtures
do their specific jobs, depending on the requirements.
Polycarboxylic (PC) ether-based superplasticizers,
however, are most efficient. They can reduce the mixing
water of concrete by more than 30%, thereby making the
concrete of very low w/c ratio, sometimes as low as 0.22,
0.23 or 0.24, and developing very high-strength concrete,
as high as 100 MPa at 28 days. Alternatively, this type of
superplasticizer can produce very high workability (as
high as 700 or 800mm slump flow), so that concrete can
be of self-compacting type.
But the chemical admixtures have to be compatible with
the cement used. A particular admixture may not be Fig. 2: The spillway of 1000 MW Tehri Hydro-Electric Project, 60
compatible with a particular brand or type of cement. MPa concrete with 8% silica fume
When they are not compatible, there will be segregation
and bleeding in fresh concrete. Mineral admixtures flyash and g.g.b.s. can also control
the temperature of mass concrete structures. The
temperature inside the fresh concrete increases upto
Role of Mineral Admixtures in Concrete 40-50 hours, and then reduces. The peak temperature
Mineral admixtures used in concrete are many e.g. is due to the heat of hydration of cement. Lower grade
pulverised fuel ash (PFA) i.e. flyash, g.g.b.s., rice husk OPC (say 33-grade) exhibits lower heat of hydration.
ash, metakaoline, silica fume etc. They (except g.g.b.s.) But the 43-or 53-grade OPC has high heat of hydration,
have pozzolanic properties. G.G.B.S. is a latent hydraulic and develops high temperature inside the concrete. The
material. Mineral admixtures, being fine, contribute to the flyash and g.g.b.s. used in the production of PPC and
cohesiveness of concrete, help in producing pumpable PSC respectively or when mixed with OPC at the sites
concrete, and also can increase the strength of concrete of construction, develop less temperature inside the
in the long run. Silica fume needs a special mention, as concrete, and therefore, there is no distress in concrete
it is the finest of all (about 50 times finer than cement), due to higher temperature differential inside the concrete,
and hence is highly reactive. For high strength concrete whose limit is about 19°C. Using 20% flyash in a mass
(60 MPa and above), it is absolutely necessary, about 5% concrete foundation (of M40 grade concrete) of a multi-
for concrete strength of 60-70 MPa, and about 10% for storeyed building, the temperature differential observed
concrete strength of 80-100 MPa. between the centre and the outer portion of the foundation
was in the range of 15.3°C to 16.4°C (Fig.3).
Concrete with silica fume can also resist abrasion.
The Southern Illinois University test result3 indicates
more than 100% improvement in abrasion-resistance of
concrete (Fig.1).

Fig. 3: Temperature profile in a mass-concrete foundation

(1500mm inside & 300mm inside concrete; bottom curve is
Fig. 1: Abrasion index of silica fume concrete ambient temperature)

Organised by
India Chapter of American Concrete Institute 345
Session 3 C - Paper 1

Silica fume, no doubt is a good pozzolana and highly content of OPC concrete, to achieve comparable 28-day
reactive, but it develops high heat inside the concrete. compressive strength of concrete6.
Therefore, one has to use it in less quantity (say about
In the concrete mix proportioning procedure, the Indian
8%) in the mass concrete of Stilling basins or aprons of
method selects absolute volume of coarse aggregate
concrete dams. There have been instances of cracks
per unit volume of total aggregate for different maximum
developing in the stilling basin of a concrete dam, where
sizes of aggregate and for different grading zone of fine
11% silica fume was used in concrete.
aggregate, whereas the American method7 selects
the bulk volume of dry-rodded coarse aggregate per
The Science of Concrete Mix Proportioning unit volume of concrete for different maximum sizes
Different countries have different concrete mix of aggregate and for different fineness modulus of fine
proportioning methods. But the common basis is the aggregate. The two procedures have different approaches,
Abram’s water-cement ratio law. So, the water-cement but the fundamental basis is that the volume of coarse
ratio (w/c) for the target 28-day compressive strength of aggregate, and the volume of fine aggregate in an unit
concrete has to be fixed from an established relationship volume of concrete are dependent on their particle shape,
between the two. But this relationship is also dependent size and grading, for a given workability of concrete. The
on the 28-day compressive strength of cement. Typical Indian mix proportioning method is similar to that of ACI
relationships are shown in Fig. 4. mass concrete mix proportioning method7 in selecting
the absolute volume of coarse aggregate per unit volume
of total aggregate.
In the British and American methods, the air content of
concrete is considered within the absolute volume of 1
m3 of concrete for different maximum sizes of aggregate,
whereas the Indian method ignores it. Air content in
concrete needs to be considered, as air will be there
in compacted concrete, and it cannot be removed by
vibration.

Alkali-Aggregate Reaction in Concrete

The reaction between the alkali of cement and the reactive
siliceous constituents of some of the aggregates is
deleterious. The reaction produces alkali-silica gel inside
the concrete and the gel expands when imbibes water.
Fig. 4: Typical relationships between water-cement ratio or Besides the alkali of cement, the source of alkali inside
water-cementitious materials ratio Vs 28-day compressive the concrete may be from the chloride being present. The
strength of concrete alkali as Na2O equivalent from the chloride will be about
0.76 times the chloride ion8.
For producing high strength concrete (say more than 60
Petrographic examination under the microscope detects
MPa at 28 days), the efficient PC - based superplasticizer
the reactive constituent minerals e.g. opal, chalcedony,
and silica fume will be necessary. Superplasticizer
cristobalite, tridymite, metamorphic greywacke etc. But
reduces the water content, and hence reduces the water-
this examination may not clearly identify some micro
cementitious materials ratio. Typically, for high-strength
crystalline, strained or micro fractured quartz9. The
concrete of the order of 100 MPa, the water-cementitious
strained quartz type reactive mineral was observed in
materials ratio is around 0.22. However, with so low
the Himalayan aggregates. Two Indian dams (i.e. Hirakud
water-cementitious materials ratio in concrete, some of
Dam and the Rihand Dam) suffered distress due to this
the cement particles remain unhydrated.
reactive mineral. The alkali-silica reaction is a very slow
With the use of mineral admixtures in concrete, the reaction, and it took a long time (about 30 years) to cause
cementitious material needs to be modified. With the use of the distress inside the power houses of the concrete
flyash (say about 25-30%), the British4 and the Indian5 mix dams. The Indian test method as per IS: 2386 (Part VII)10
design method stipulate use of 10% higher (OPC+ flyash) is an accelerated one to be conducted at 60°C, similar to
content than that of OPC content only in the concrete mix, the test method of ASTMC126011 which is to be conducted
in order to achieve equal 28-day compressive strength of at 80°C. But this ASTM method is not foolproof, as the test
concrete. For g.g.b.s. concrete mix proportioning, for use results sometimes do not corroborate field performance.
of different % of g.g.b.s. content (say 25-70%), the British The ASTM Standard itself indicates that “some granite
mix design method recommends an additional (cement + gneisses and metabasalts have been found to be
g.g.b.s.) content of 10kg/m3 to 50kg/m3 above the cement deleteriously expansive in field performance, even though

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

346 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Some Aspects of Concrete Science and Concrete Technology

their expansion in this test was less than 0.10% at 16 days”. MPa at 28 days. Alternatively, they can produce very
The ASTMC 1260 is somewhat conservative in that it uses high workability of concrete, as high as 700-800mm
aggressive 1N NaOH solution in which the specimens are slump flow, and hence they are used to produce self-
immersed, and the specified temperature is quite high i.e. compacting concrete. The mineral admixture silica
80°C. In reality, no external NaOH source may exist and fume, being a highly reactive pozzolana, can be used to
the reaction may terminate earlier 9. develop high-strength concrete, along with PC- based
superplasticizer. Also, silica fume concrete can resist
The University of Taxas’ test results on concrete
abrasion in the spillways of concrete dams, and in
specimens exposed to outside environment for more
concrete roads.
than 6 years, compared the performance of a number of
coarse aggregate samples with that of the accelerated 3. Concrete mix proportioning procedures using mineral
mortar bar test of ASTMC1260 and that of the concrete admixtures need modification in the cementitious
prism test (ASTMC1293) at 38°C. They concluded that materials content, in order to develop the desired 28-
several coarse aggregate samples that were predicted day compressive strength of concrete. Different mix
to show innocuous behaviour in ASTMC1260 test, showed proportioning procedures have different approaches
deleterious expansion in the more representative to select the volume of coarse and fine aggregates in
ASTMC1293 test and the outdoor simulated field concrete, which are dependent on the particle size,
exposure test12. shape and grading of the aggregates for the desired
workability and 28-day compressive strength of
For combating the probable alkali-silica reaction in
concrete.
concrete, especially for hydro-electric projects, IS: 456
recommends use of at least 25% flyash or at least 50% 4. The deleterious alkali-silica reaction in concrete is a
g.g.b.s. as part replacement of low alkali (not more than very slow reaction, but the accelerated test methods
0.6% alkali as Na2O equivalent) OPC. The advantage is developed to predict the reactivity do not always
that, the alkali of flyash and the alkali of g.g.b.s. are not corroborate the field performance. For combating
fully reactive8. the deleterious alkali-carbonate reaction, the
recommendations include use of very low alkali (as
Similar to alkali-silica reaction, the deleterious alkali-
low as 0.4% alkali as Na2O equivalent) OPC. But such
carbonate reaction has also been observed in some of
low alkali OPC is not easily available.
the Indian aggregates. For reactive carbonate rocks,
ASTMC 58613 specifies the rock cylinder test method.
But experimental studies have found that pozzolanas REFERENCES
1. IS 269 Ordinary Portland cement – Specification.Bureau of Indian
are not effective for controlling the alkali-carbonate
Standards,New Delhi.
reaction. The pozzolanic reaction is found to be too slow
2. IS 456 Code of practice for plain and reinforced concrete. Bureau
to prevent dedolomitization14. For preventive measures, of Indian Standards, New Delhi.
ACI Committee15 recommends the use of low alkali (0.4%
3. Ghafoori, N. and Diawara, H. Abrasion resistance of fine aggregate
alkali as Na2O equivalent or lower) OPC. But such low replaced silica fume concrete. ACI Materials Journal, September-
alkali OPC is not easily available. October 1999, pp. 559-567.
4. BS 5328 : Part 1 : 1991 Guide to specifying concrete. British
Standards Institution, London.
CONCLUSIONS
5. IS 10262 Concrete mix proportioning – guideline. Bureau of Indian
From various aspects of concrete science and concrete Standards, New Delhi.
technology discussed above, the following conclusions
6. Teychenne, D.C., Nicholls, J.C., Franklin, R.E. and Hobbs, D.W.
are drawn: Design of normal concrete mixes. Department of Environment,
London, 1993, 42p.
1. Different cements develop different characteristics
in concrete. The blended cements i.e. PPC and PSC 7. ACI 211.1 Standard practice for selecting proportions for normal,
heavy weight and mass concrete. American Concrete Institute.
are not suitable for prestressed or precast concrete
construction, as their rate of hardening is slower than 8. BS : 5328: Part 4 Specification for the procedures to be used in
sampling, testing and assessing compliance of concrete. British
that of OPC. But these blended cements or the mixture Standards Institution, London.
of (OPC+flyash) or (OPC + g.g.b.s.) when used at the
9. Malvar, L.J. Cline, G.D., Burke, D.F, Rollings, R, Sherman, T.W.
sites of construction, develop lower temperature in and Greene, J.L. Alkali-silica reaction mitigation : State of the art
fresh concrete,and hence they are suitable in mass and recommendations. ACI Materials Journal, September-October
concrete construction. 2002, pp.480-489.
10. IS 2386 (Part VII) Methods of test for aggregates for concrete. Alkali
2. Different chemical admixtures influence the aggregate reactivity (with Amendment No.1, June 1999). Bureau of
characteristics of fresh concrete in different ways. The Indian Standards, New Delhi.
PC- based superplasticizers can reduce the mixing 11. ASTMC 1260 Standard test method for potential alkali reactivity of
water of concrete by more than 30%, and thereby aggregates (Mortar bar method). American Society for Testing and
can produce high-strength concrete, as high as 100 Materials.

Organised by
India Chapter of American Concrete Institute 347
Session 3 C - Paper 1

12. Idekar, J.H., Bentivegna, A.F, Folliard, K.J. and Juenger, M.C.G. Do American Society for Testing and Materials.
current laboratory test methods accurately predict alkali-silica
14. Grattan Bellew, P.E. Preventive measures to counteract expansion
reactivity? ACI Materials Journal, July-August 2012, pp. 395-402.
of concrete containing alkali-reactive aggregates. Durability of
13. ASTMC 586 Standard test method for potential alkali reactivity of Building Materials 1, 1983, pp. 363-376.
carbonate rocks as concrete aggregates (Rock-cylinder method).
15. ACI Committee 201. Guide to Durable Concrete, 1977.

Dr. S.C. Maiti

Dr. S.C. Maiti holds a PhD (Structural Engineering) from IIT Kharagpur and is former Joint Director of National
Council for Cement and Building Materials (NCCBM), New Delhi. He is currently working with UltraTech
Cement Ltd. as Technical Consultant. He has been a member of the panel for revision of IS 456: 2000. His
areas of interest are Cement and Concrete, Advances in Concrete Technology, Concrete Mix Design and
Quality Assessment of Concrete Structures.

Raj K. Agarwal
Raj K. Agarwal is Managing Director of marketing and transit (India) Pvt. Ltd. New Delhi. He has been
interacting with project authorities of Hydro Electric projects in the Country for the last 15 years, and providing
guidance for the use of proper construction materials in order to achieve durable concrete structures. He
has published several papers and participated in ICOLD and international seminars.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

348 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial Replacement of Cement in Mortar with Red Mud and Ultrafines

Partial Replacement of Cement in Mortar with Red Mud and Ultrafines

Deshmukh M.P Sarode D.D Pendhari S.S I. Alam

Research Scholar, Dept Associate Professor, Dept. Associate Professor, Dept. M. Tech. student, Dept. of
of General Engg, Institute of General Engg, Institute of Structural Engg, Veermata Structural Engg, Veermata
of Chemical Technology, of Chemical Technology, Jijabai Tech. Institute, Jijabai Tech. Institute,
Mumbai. Mumbai. Mumbai Mumbai

Abstract used in various cement and cement based composites

for production of building blocks, tiles, paver blocks,
Cement and cement products are largest consumable
bricks, precast components,etc. It is found that red
materials in the world, next to water. Production of cement
mud concrete components offer better resistance to
is energy intensive process and a large amount of carbon
chloride penetration, thus helps to improve durability of
di­oxide is liberated in this process. Thus, there is need to
end product. The presence of calcium, silica and soda
explore the material that would replace cement in mortar
in red mud are beneficial in the formation of vitreous
and concrete.
glazes. Red mud has been successfully used as additives
An inventory of about 3 billion tons of red mud is awaiting in cement production, geo­ polymers, development of
in stock­pilling yards for its utilization and 120 million tons special types of cement, mortar, concrete, etc. Red
of red mud is added every year in it. This high alkaline red mud contains high percentage of iron oxide and hence
mud generated during the production of alumina is posing imparts red colours to the products made with (more
a serious threat to the environment. Ultrafine is a new than 15%) specific content. Thus, red mud mortar helps
generation supplementary cementitious material and in improving aesthetic appearance when applied as
helps in reducing the water demand at a given workability. coloured mortar for the interiors.
An attempt is made here to partially replace cement in
mortar with raw red mud and ultrafine.
Research Significance
Keywords: Ultrafine, bauxite­ residue, compressive, By replacing cement in mortar, an attempt has been made
flexural, mortar, red mud, tensile. to reduce the consumption of cement content in mortar.
This may be useful in reducing the carbon foot­prints.
Introduction Exploration of alternatives to cement consumption is an
Ever increasing needs of housing and infrastructural essential component for green technology and sustainable
development are leading to very high consumption of development.
cement and aggregates. This is posing a great threat Utilization of an industrial waste from alumina industry
to environmental protection and sustainability issues. (red mud) will reduce the burden of residues lying in
Cement production process releases 0.8­1 tonn of carbon stock­piling yards and abandoned sites. Thus conservation
dioxide per tonn production of cement. Hence there is an of large land areas and the associated air, water and land
urgent need to explore alternatiive substitutes for cement. pollution could be minimised.
Bayer’s process for the production of alumina results Ultrafine would help in correction of water demand,
in the generation of significant amount of dust­ like, workability and associated strength benefits.
high alkaline bauxite residue known as red mud. It is
one of the largest industrial waste p ­ roduct in modern
society. Source of bauxite and the minerological process Experimental Investigation
parameters determine the chemical and minerological Materials
composition of bauxite residue. Stock ­piling needs large The materials used in the presented research are
land areas and are liable to contaminate by air,water and described as the following:
land pollution. About 1­-1.6 tons of red mud is generated
i Cement: 53 Grade grade OPC, (specific gravity­3.15)
per tonne production of allumina. The disposal cost of red
and fineness 2%
mud is also very high (1-­2% of alumina price).
ii Red mud : NALCO, Damanjodi,Orissa,India,(specific
There is an urgent need to explore methods of gravity­3.10, PH­12.5)
utilisation of this high alkaline industrial waste for
iii Fine aggregate: Grade 1, grade 2 and grade 3 sand
some constructive purpose. It is well established that
confirming to IS:650
red mud can be used as pozzolanic material and can be

Organised by
India Chapter of American Concrete Institute 349
Session 3 C - Paper 2

iv. Water : Phase 1­Potable water with W/C ratio in the replaced by red mud in the varying proportions such as 0%,
range of 0.38-0.40. Phase 2­Potable water with W/c 5%, 10% and 15% of cement content. In phase 2, red mud and
of 0.35 . ultrafines are added in the proportion 1:1 as a cementitios
v. Ultrafine : PSD 12000 cm2/gm and specific gravity­ mass and this composite is being used to replace cement in
2.86. 10%, 20% and 30% of the cement content.

Chemical properties of red mud checked by XRD indicates Design mix of 1:3 is prepared and mortar cubes of size
low percentage of CaO as compared to percentage of CaO 50 mm x 50 x 50 mm are cast, cured for determination
in cement and gypsum. Low cementatious properties of compressive strength of mortar after 3, 7 and 28 days.
put forth limitations on the use of red mud as binder in This mix is taken as a Control mix and then trials are
concrete. However, it is found that red mud reacts with conducted for 5%, 10% and 15% replacement of cement
water & cement and forms calcium silicate hydrate (CSH) with raw red mud. Cubes are then tested for compressive
gel that improves strength characterstics of composites. strength after 3, 7 and 28 days using compression testing
Pozzolana is aluminous and siliceous material which machine. Flexural strength is determined in the similar
reacts with calcium hydroxide in presence of water. manner for mould of size 70.6mm x 70.6mm x 70.6mm.
Tensile strength is determined for control and other
The chemical composition of XRD of Red mud from Nalco, trials using briquette mould.
Damanjodi ( ACC lab /R & d/2014/56 dtd 7/7/14) is given in
Figure 1. Following table shows results of compressive, tensile and
flexural strength for various trials.
Trials are conducted in two phases. In phase 1 cement is

Chemical composition of Red mud from Nalco, Damanjodi, Orissa, India in percentage content.

SiO2 Al2O3 Fe2O3 CaO Mn2O3 MgO LOI Na2O K 2O TiO2 P 2O 3 SO3 Chloride

6.6 17.1 49.2 2.0 0.17 0.1 15.6 3.12 0.12 5.52 0.22 0.2 0.018

Table 1
Compressive, tensile and flexural strengths after 3,7 and 28 days

Compressive strength in MPA Tensile strength in MPA Flexural strength in MPA

Trial
Compositi on of trial
no
3 7 28 3 7 28 3 7 28
day day day day day day day day day

1 Reference 24.08 36.69 46.11 1.62 2.72 2.93 4.05 5.96 6.95

2 5% Red mud 25.25 32.11 45.18 1.78 2.56 3.25 4.54 5.94 7.16

3 10% Red mud 22.7 30.02 42.06 1.94 2.78 3.38 4.59 6.33 7.68

4 15% Red mud 22.57 26.1 37.15 2.17 2.83 3.56 4.78 5.25 8

5 5% Red mud + 5% ultrafine 22.3 29.5 38.85 1.68 2.67 2.9 4.46 5.83 6.99

6 10% Red mud+ 10% ultrafine 23.48 31.59 39.57 1.86 2.82 2.95 4.73 5.93 7.24

7 15% Red mud +15% ultrafine 23.48 32.02 43.88 2.04s 2.92 3.22 5.35 5.93 7.36

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

350 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial Replacement of Cement in Mortar with Red Mud and Ultrafines

and 30% of combined mix of

red mud and ultrafine in the
ration 1:1. Compressive , tensile
and flexural strengths are
determined for all the trials after
3, 7 and 28 days.
From phase 1 results, it can be
concluded from trial no 1,2,3
and 4 that compressive strength
of mortar reduces with the
increment of red mud content as
a replacement of cement. This is
due to additional water demand
required for mixing of red mud.
Flexural and Tensile strength
increases with the increment of
red mud percentage.
Fig. 1: XRD red mud at Nalco ( ACC lab /R & d/2014/56 dtd 7/7/14) From phase 2 results, it can be
concluded from trial no 5.6 and
7 that compressive strength of
mortar increases consistantly with the increment of red
mud and ultrafine content as a replacement of cement
as compared to trial no 4. This is due to reduction of
water demand because of substitution of ultrafine along
with red mud. Flexural and tensile strength found to be
increasing progressively with respect to reference control
mix results.

Fig. 2: Compressive strength of mortar after 3,7 and 28 days

for various trials

Fig. 4 : Flexural strength of mortar after 3, 7 and 28 days for

various trials

Summary and Conclusions

1. Compressive strength result sheet indicates that
replacement of cement with red mud in mortar mix
Fig. 3: Tensile strength of mortar after 3, 7 and 28 days for does not have significant effect on its compressive
various trials strength. This shows that 30% cement replacement
by composite mix of 15% red mud and 15% ultrafine
Discussion of Results reduces the 28th day compressive strength merely by
4.81%. Maximum compressive strength of 46.11 MPA is
Trial nos 1,2,3 and 4 represents control mix 0%,5%, observed in trial no 1.
10% and 15% of cement replacement with raw red
mud conducted in phase 1. Trial 5,6 and 7 represents 2. Tensile strength result sheet indicates that
replacement of cement from control mix with 10%, 20% replacement of cement with red mud and ultrafine

Organised by
India Chapter of American Concrete Institute 351
Session 3 C - Paper 2

gives better tensile strengths to the mortar. Tensile 3. Cooling, D.J., “Improving the Sustainability of Residue Management
Practice”, from Paste 2007, Proceedings of the Tenth International
strength of mortar increases from 2.93 MPA to 3.56
Seminar on Paste and Thickened Tailings, edited by Fourie, A. B.,
MPA at 15 % replacement of cement with red mud . and Jewell,
Replacement of cement with composite mix of red mud 4. Eliz Paula Manfroi, MalikCheriaf, Junaide Cavalcante Rocha, 2013,
and ultrafine makes no sizeable change in the tensile Microstructure, Minerology and and Environmental Evaluation
strength of morter. Maximum tensile strength of 3.56 of cementitious Composites Produced with Red Mud waste ,
MPA is observed in trial no 4.. Construction and Building Materials, Vol.no.xxx , 1­08.
5. Gray R, J., “Engineering Properties and Dewatering Characteristics
3. Flexural strength result sheet indicates that of Red Mud Tailings”, (1974) University of Michigan, DRDA project
replacement of cement with 15% of red mud gives 340364
maximum flexural strength of 8MPA amongst the trials 6. Jamaican Bauxite Institute and the University of the West Indies,
on mortar. Flexural strength of mortar increases from “Bauxite Tailings “Red Mud”, Proceedings of International Workshop
6.95 MPA to 8 MPA at 15 % replacement of cement with Kingston, Jamaica, October 1986.
red mud . Replacement of cement with composite mix 7. Jones, B.E. H., and Haynes, R.J., “Bauxite Processing Residue:
of red mud and ultrafine makes no sizeable change in A Critical Review of Its Formation, Properties, Storage, and
Revegetation”, (2011)
the flexural strength of morter.
8. Critical Review Environ. Sci. and Tech., (41) 271­315. Klauber, C.,
It can be concluded from the above results that, Grafe, M., and Power, G, “Review of Bauxite Residue “Re­u se”
replacement of 30% of the cement with a composite mix Options”, CSIRO Document DMR­3609 (2009)
of red mud and ultrafine gives fairly good compressive 9. Pawar M.S., Saoji A.C.,2013 ,Effect of Alccofine on Self Compacting
strength and better tensile as well as flexural strengths,as Concrete The Internatiional Journal Of Engineering And Science,
Vol 2, 5­9.
compared to controll mix.
10. Pinnock, W.R.: “Measurement of Radioactivity in Jamaican Building
Reduction of cement content in mortar will reduce carbon Materials and Gamma Dose Equivalents in a Prototype Red Mud
emmisions and energy consumption in cement production. House”, J. Health Physics, (1991), 61 (5), 647­651.
Utilization of industrial waste like red mud and ultrafine 11. Patel P.J, Patel H.S., 2013, Effect of Compressive and Flexural
material against cement will help in environmental Strength of High Performance Concrete Incorporating Alccofine and
Flyash International Journal of Civil , Structural, Environmental and
protection ,sustainable development and in the promotion
Infrastructure enginnering Research and Devlopment Vol 3, 109­114.
of clean technologies.
12. Power, G., Grafe, M., and Klauber, C., “Review of Current Bauxite
Acknowledgements : We are grateful to the officials of Residue Management, Disposal and Storage: Practices, Engineering
Nalco, Damanjodi, Orissa for the help extented in supplying and Science.” CSIRO Document DMR­3609 (2009).
red mud to Mumbai . We are eqqually grateful to R and D 13. Purnell, B.G., “Mud disposal at the Burntisland alumina plant”, Light
division of Associated cement companies, Thane, India. Metals (1986), 157­159.
14. Zainab Z Ismail and Enas A. Al­Hashmi,2009, Recycling of waste
glass as a partial Replacement for Fine aggregate in Concrete,
References Waste management, Vol.no.29.655­659.
1. Banvolgyi, G. Huan, T. M., “De­w atering, disposal and utilization of
red mud: state of the art and emerging technologies”. 15. Zhihua Pan, Lin Cheng, Yinong Lu, Nanru yang,,2002, Hydration
Products of Alkali­Activated Slag­Red Mud cementitious material,
2. Cajjun Shi and KerenZeng , A Review on the use of Waste Glasses Cement and Concrete Research, Vol 32,357­362.
in the production of Cement and Concrete Resourses, Conservation
and Recycling 52 (2007) 234­247.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

352 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Partial Replacement of Cement in Mortar with Red Mud and Ultrafines

Mr. Manoj Prakash Deshmukh

Mr. Manoj Prakash Deshmukh is presently working as Research scholar in general engineering department
at Institute of Chemical Technology, Matunga, Mumbai. He has done his graduation- Bachelor of Civil
engineering with First division from Government College of Engineering, Amaravati, Maharashtra, India
in 1991. He obtained his Master of Civil engineering degree with First division from Sardar Patel College of
Engineering, Andheri, Mumbai in the year 1999.
Currently, his research work areas are focused on utilization of bauxite industry residue (Red mud) for
various construction material substitutes in housing, construction and infrastructure industry. This aims
to reduce environmental pollution, reduction in natural construction material consumption, and address
sustainability issues up to certain extent.

Dilip D. Sarode
Dr. Dilip D. Sarode, a doctorate from IIT, Mumbai. Prior to his doctorate, he completed graduation in civil
engineering and master in structers from VJTI, Mumbai. He is presently working as Associate Professor,
General Engineering department in Institute of Chemical Technology, Mumbai. He has guided 8 ME and 1
PhD students and is presently guidng 2 ME and 4 PhD students. He has 10 international and 25 national
publications and 1 patent to his credit. He is fellow member of Indian Geotechnical Society and member of
Institution of Engineers and Indian Society for Technical Education. His research interest includes concrete
technology, construction chemicals and composite materials.

Dr. S.S. Pendhari

Dr. S.S. Pendhari is presently Associate Professor of Structural Engineering Department, Veermata Jijabai
Technological Institute (VJTI), Matunga, Mumbai. He obtained his B.E. Civil from Mumbai University (VJTI)
and PhD from IIT Bombay. He has published 16 international peer reviewed research paper and presented
22 research papers in international and national conferences. He has guided 14 MTech dissertations and
examined more than 20 MTech dissertations from Mumbai and Pune Universities so far. Presently 4 MTech
students and 4 PhD students are working under his supervision. His research area includes computational
mechanics, composite mechanics, modeling of smart materials, finite element method, repair and
rehabilitation of structures.

Organised by
India Chapter of American Concrete Institute 353
Session 3 C - Paper 3

Production of M60 Grade of Concrete in Difficult &

Underdeveloped Conditions
Durga P Shrestha

Abstract ground construction are going higher and higher. To meet

the challenge, leading architects, structural designers
Under the conditions where material supply quality
and concrete technologists started to investigate the
and quantity including choices of brands of additives &
possibilities of making High­Strength / High­Performance
admixtures are very limited it could be very difficult task to
concrete at site with manual batching in drum mixer
make decision what grade & quality of major construction
machines.
materials like concrete to be used for the first time.
Economic challenges sometimes could be so high that one Due to Earthquake risk, high rise buildings are very slowly
has no choice but to take up the challenge. picking up. Therefore, High –Strength/ High­Performance
concrete is also slowly coming in use. Concrete industry
Approach to the mix design for M60 grade of concrete was
as such is very much in infant stage with only two Ready
to look at what has been done in the past to achieve this
mix Plants operating in the country. Almost all concrete is
grade of concrete with high workability as in our case it
mixed at sites in the batching plants for large projects and
was going to be used in the circular columns of 600mm
by drum mixer machines of normally 50­100 kilograms
diameter. The Building this M60 grade of concrete is going
cement per batch of concrete capacity.
to be used has been registered with LEED USA for their
certification, most probably platinum/gold, so concrete to For new corporate building of M/S HAMA IRON & STEEL
be used must use 20%of recycledcementationsmaterials Pvt. Ltd. at Kathmandu, it was decided to use highest
replacing OPC. Therefore, this component is already fixed. achievable grade of concrete and with the consent of
M60 grade of concrete is going to be used in Kathmandu, concrete technologist the grade of concrete was finalized
where day temperature in summer is around 33 degree C as 60MPa @ 28 day cube strength. The above building
and occasionally goes as high as 36 degree C. Therefore was registered at LEED, USA for their certification
OPC 53 grade can be safely used. possibilityplatinum. As per LEED’S requirement for
concrete to be used in the construction of building,we must
All the concrete is going to be batched manually and also
use minimum of 20% recycled cementations material as
all the additive & admixture to be measured manually.
replacement to OPC, to be qualified as green concrete.
Therefore, the mix has to be versatilei.e. its workability
must not change much with minor changes of moisture Looking at very slow development of the High ­Strength /
content and if possible with minor changes in quantity of High – Performance concrete in Nepal, there could three
admixture. Consistency of mix in workability and in strength factors:
is extremely important. Please refer to the analysis of
1. Poor demand
the 28days crushing strengths of 130number of sample
taken over year (Appendix­1). The Standard Deviation is 2. Site mixed concrete as common practice
2.37only and coefficient of deviation is 3.85. Therefore it
3. Supply of aggregates, chemical and even cement has
can be concluded that design mix is very good in the view
not been dependable in quality and even in quantity.
that batching was done manually. Also concrete was made
throughout the year i.e. seasonal affect in moisture content Poor demand due to only few high rise buildings are coming
did not affect the strength much. All manual batching can up. This may be due high risk as Nepal being Earthquake
produce consistent quality of high grade/high performance prone area. Recently, Earthquake of magnitude 7.8 Richter
concrete in the difficult and technically under developed scale struck the country and did lot of damage to high rise
Conditions. building in Kathmandu.
Mostly concrete is mixed at site till this date. Due to this
Introduction practice structural designers do not have confidence
High – Strength /High –Performance Concrete is the in production of high grade concrete consistently by
need of the day as in highly densely build up area like the contractors. At site contractor do not have skilled
Kathmandu where the construction area is very small. manpower to batch multi blend additive and admixtures
Consequently, Basements are going deeper and above of various types and in doses for large concreting works.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

354 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Production of M60 Grade of Concrete in Difficult & Underdeveloped Conditions

Supply of material for high greatconcrete is very much sand was stacked under cover andon sloping ground
uncertain. In and around of Kathmandu good quality with brick soling. Moisture content of aggregate was
course aggregate is getting scare. Recently there was measured at least once before mixing first batch
total stoppage in production & supply of aggregates of concrete and added water quantity was decided
for several months. Cement supply also is uncertain in on this finding of moisture content. Therefore total
quality and in quantity. High grade concrete suffers most water content was always consistent in the concrete.
from these uncertainties, Imports are also often not As is could be seen in the mix design the water in the
regular. Cementitious supplementary additive which are chemical admixture was also taken into account.
very much required to produce high grade concrete, their
2. Slump were measured regularly and when in doubt of
sources of supply,their quality and prices are not known to
slump of concrete,
many buyers & users. Therefore, it is very difficult to get
good material at reasonable price. Therefore, production 3. Test cubes making curing & testing of all the test cubes
of High ­Strength /High­Performance concrete suffers a lot were made as per British Standard practice. These
due to this shortcoming. Cubes were cured in the temperature control. Water
tank and tested at right time as per British Standard
Similar is the situation with regard to chemical admixtures.
procedures.
This shortcoming is the biggest drawback in producing
High ­ Strength./ High ­ Performance concrete. Now we
can see that producing High ­Strength.­High ­Performance Conclusion
concrete here in Nepal is very much risky decision Following are advantages of mix design of M60 grade of
but progress must go on. So, production of M60 grade concrete used
concrete has been started recently. There are certain
conditions we have to meet in the Mix Design we are going 1. At site by manual batching High ­ Strength/High
to do. They are: ­Performance concrete of consistent quality, strength
and workability can be produced over a long period of
1. OPC quantity to be used, 20% of this has to be replaced time.
by recycled cementitious material.
2. To use high strength concrete in built up congested
2. Slump 110mm minimum areawhere land price is very high, it is highly economical
3. Course Aggregate maximum size 20mm for contractor and the owner. To builder/contractor it is
not only money saving in material but also cost of labor
4. Concrete should flow when vibrated for handling reduced quantity including time saving.
5. Concrete should be versatile i.e. slump should not 3. Continuous increase of concrete strength during long
change much with changes in water content& within period of time gives huge safety factor againstnatural
short duration of time of mixing concrete. disaster. And accidental over loading including
Keeping above conditions in mind, several trial mix were miscalculation by structural design designer.
tried and concrete cubes were tested for 4. Due to pozzolanicreaction it is more dense concrete than
minimum period of 56days. Finally one was selected using only OPC therefore more water proofconcrete so
which met all above conditions. This has been shown as less prone to chemical and more safety to reinforcing
works cube ID1 for various period of testing. steel bars.
5. Mix is very versatile and slump does not change due
Materials were selected on the following basis: to small changes in water content & therefore it is
1. Meets all the technical quality requirement user finely which has huge advantage for use in under
developed area& not so much skilled workforce is
2. quality is consistent
required.
3. Constant supply of ingredient construction material is
6. Present days practice of using 28 days cube crushing
ensured
strength, as design strength which can be replaced by
4. Price is nominal and variations is small over the year 56 days cube crushing strength. Therefore saving can
be achieved if we follow similar line of mix Design and
Methodology followed for quality control use 56 days cube strength as design strength.
1. Aggregate Quality –Every time new supply of aggregate
arrive at site, aggregate were tested for grading and Ackndwledgement
also for crusting strength etc. specially when it appears
Author would like to thank Mr. Malay Sahof M/S SWC
to be different than approved aggregate. Especially
INDIA for giving solid & liquid content of SUPAPLAST­PC.

Organised by
India Chapter of American Concrete Institute 355
Session3 C - Paper 4

Role of Supplementary Cementitious Materials on

Chloride Induced Corrosion - An overview
Ms. Anita N. Borade and Dr. B. Kondraivendhan
Applied Mechanics Department, SV National Institute of Technology,
Ichchhanath, Surat- 395007.

Abstract to premature failure is a serious problem worldwide

(Elsener B. 1997). This involves the enormous cost for
Reinforced concrete is a strong, durable and long lasting
inspection, maintenance, restorations and replacement of
material proven by many prestigious structures throughout
infrastructures (Neville, 1995).
the world. Over the last few decades durability of concrete
structures and their long-term performance have The source of harmful chloride ions into the concrete is
emerged as a primary concern for structural engineers, either a internal or external. Internal chlorides may be
infrastructure owners and researchers. Corrosion due to added through the ingredients of the concrete mix through
chloride attack is one of the major worldwide durability chloride contaminated aggregate, water or through
problems for reinforced concrete structures. Naturally admixtures. But the major part of chloride entered in
the reinforcement is protected by the thin passive layer of hardened concrete through deicing salts in bridge deck
iron oxide till the pore solution is highly alkaline ( pH above and parking structures, penetration of chloride ions
12.5 and 13.5). When chloride concentration exceeds from sea water in marine structures or from the soil
threshold limit, breaks passivity of reinforcement leads to and ground water contains chloride salts (C Andrade, 93,
corrosion. The chlorides are normally contributed by mix Pradhan and Bhattacharjee, 2011, Montemor M., 2003).
constituents of concrete or they penetrate into hardened Corrosion is takes place in presence of moisture and
concrete from surrounding chloride bearing environment. oxygen is basically an electrochemical process. After the
Corrosion of the reinforced concrete is an electrochemical initial hydration process of cement, due to high alkalinity
process which takes place in presence of moisture and (pH - 12.5 to 13.5) a protective impermeable passive layer
oxygen ingresses due to porous concrete which plays an of ferric hydroxide is generated which adheres the steel
important role in chloride transportation. Chloride ions surface tightly. Steel remains protected as long as the
exist either chemically bound to the hydration products, passive layer of oxide around the steel is stable (Angust
physically held to the surface of the hydration products or and Vennesland, 2007). When chloride ion concentration
free which are soluble in water. Chloride binding of cement- reached to the reinforcement level, get reacted with other
based material is very complicated and influenced by corrosion products (water and oxygen). Due to which pH of
many factors. To restrict chloride ingress in the concrete the pore solution falls (below 12.5), oxide film is destroyed
various mineral admixtures are used to improve the and corrosion starts due to loss of passivity. Once the
micro structure of concrete hence the control the rate of corrosion is started, the corrosion product (iron oxide and
corrosion Numerous researches aimed at assessing the hydroxide) increases about four to six times of the volume
suitability of using the electrical conductivity of concrete of steel reinforcement.
to indirectly evaluate corrosion activity and hence predict
the service life of RC structures. An overview of mineral An intensive research is going on improving the durability
admixtures like Ground Granulated Blast Furnace Slag of reinforced concrete (RC) structures. The use of mineral
(GGBS), Fly Ash (FA) and Metakaolin (MK) on chloride admixtures with Ordinary Portland Cement (OPC) and
induced corrosion and their contribution in durability addition of alternative cementitious materials such as
enhancement is discussed. ground granulated blast furnace ash (GGBS), fly ash (FA)
and Metakaolin (MK) are known to enhance the durability.
Keywords: Durability, electrochemistry, chloride binding, It has been reported widely that concrete containing
mineral admixtures. alternative cementitious materials excellently performs
in marine environment and in highway structures. In
Introduction addition to improving the durability, the use of these
materials in construction reduces the waste deposits
The long term performance of the structures exposed and the demand of cement. Hence helps in reduction in
to the marine environment is emerged on the top of environmental pollution. From various literatures there
durability. The deterioration of reinforced concrete is different opinion about the use of supplementary
structures due to chloride induced corrosion which leads cementitious materials on chloride induced corrosion.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

356 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Role of Supplementary Cementitious Materials on Chloride Induced Corrosion - An overview

Here the review on effect of GGBS, FA and MK on chloride

induced corrosion is discussed.

Electrochemical Mechanism of Corrosion

The conversion of iron or steel into rust is natural
electrochemical process. Due to passage of electrical
charge, electrochemical cell develops along the steel in
concrete. The potential difference between the electrode
and the adjacent electrolyte represents the electrode
potential and it is the driving force for an electrochemical
reaction to occur where the anode and the cathode are on
the same steel bar.
Corrosion takes place in presence of oxygen and water but Fig. 2: Volumetric Change in Corrosion Process
without chlorides in several steps. Electrons are dissolved
in oxidation process at anode and consumed in reduction in figure 2. The final corrosion product is about 4 -6 times
process at cathode. (Broomfield, 1997) of original in volume.

At the anode of the steel reinforcing bar, iron is oxidized The tensile stresses are developed due to increase
to iron ions dissolved in pore water to become ionic (Fe2+). in volume which results in development of cracks,
separation of steel and concrete bond. Further the cracks
Anodic reaction: Fe → Fe+2 + 2e- ...................................(1) reach to surface of concrete through and widened will
The free electrons travel through the steel to the local cause spalling and entry of corrosion product to the rebar
cathode. At the cathode on the same steel bar, the level further accelerate the corrosion (see figure 3) which
reduction of oxygen in the presence of moisture accepts causes to reduction in cross section, hence ductility and
electrons and reduced to hydroxyl ions(OH−) as shown in strength of reinforcement (Benture, 2013, Liu, 1996)
eq.2 (Broomfield, 1997 and 2002) as shown in figure 1
Cathodic reaction: O2 + 2H2O + 4e- → 4(OH)- ................(2)
It is observed that the rate of flow of ions between anode
and cathode is mainly affected by the porosity of concrete
(Sadowski, 2008). The electrochemical process of
corrosion is explained in figure 1 below. Cathodic reaction
always produces hydroxide ion which increases the pH of
the pore solution near cathode. The anodic and cathodic
reactions are necessary for the corrosion to occur, also
they need to take place simultaneously and there must be
equilibrium between these processes. Fig. 3: Initiation of Corrosion and Spalling of Concrete
After oxidation and reduction process, further corrosion
continue in few steps. The hydroxyl ions combine with the Further corrosion of steel in concrete is in the presence
ferrous ions to form ferrous hydroxide, in the presence of chlorides but without oxygen (at anode) takes place in
of water and oxygen, the ferrous hydroxide is further several steps. At the anode, iron reacts with chloride ions
oxidized to form Fe2O3.H2O which is known as rust shown to form an intermediate soluble iron-chloride complex,
see equation 3.
Anodic reaction: Fe + 2Cl− → (Fe+2 + 2Cl−)+2e- ...............(3)
When the iron-chloride complex diffuses away from
the bar to an area with higher pH and concentration of
oxygen, it reacts with hydroxyl ions to form Fe (OH)2. This
complex reacts with water to form ferrous hydroxide as
per equation 4.
Cathodic reaction:
(Fe+2 + 2Cl−) + 2e− + 2H2O → Fe (OH)2 + 2H + 2Cl− .......(4)
From the above reaction above it seen that chloride is
regenerated, hence the rust do not contain chloride. There
Fig. 1: Electrochemical Mechanism of Corrosion in Reinforced will be limited corrosion in dry concrete (i.e. when the
Concrete

Organised by
India Chapter of American Concrete Institute 357
Session3 C - Paper 4

relative humidity inside concrete is less than 60 percent) Physical Binding

and also in concrete fully immersed in water (due to the lack Calcium silicate hydrates (C-S-H) is the primary binding
of oxygen). The relative humidity that is most susceptible for phase and main hydration product about 75% of hardened
corrosion is between 70 and 80 percent (Andrade C, 2001). Portland cement. Being a cementitious material C-S-H
has variable composition and wide range of Ca/ Si ratio.
Chloride Ingress in Concrete It is found that with decrease in Ca/ Si ratio, the alkali
binding in C-S-H gel will be increased (Tang Y M, 2012).
Chlorides enter into concrete through contaminated
It is also reported that alkali binding influences the OH-
aggregate, from sea water or through admixtures at the
ion concentration in pore solution (Rusheeduzzffer, 1992).
time of mix preparation are called as internal chlorides.
As it is porous in nature it physically absorbs chloride
Chlorides ingresses in concrete from chloride bearing
ions on the surface. It is found that physical chloride
environment, de-icing salts in parking structures and
absorption capacity of OPC is depends on C-H-S content
bridge deck, chloride bearing soils and ground water are
in the hardened concrete. It is irrespective of w/c ratio and
called as external chlorides. Chlorides from the external
the aggregate to be added (Tang and Nelson, 1992). The
chloride bearing environment penetrates to the rebar
absorption of chloride by C-H-S gel greatly reduces the
level through concrete reduce the alkalinity of pore water
free chloride causing corrosion.
and breaks the passivity of steel further lead to corrosion
(Montemor M F, 2003). It is observed that the chlorides
mixed in the mix (internal) are more aggressive than the Chemical binding
chloride ingresses from the environment even quantity of Chloride ions are chemically bounded with the initial
chlorides is same (Deb S. 2011, Tareq S. 2011). Ingress of hydration product of the cement paste and the remaining
chlorides is depends on the resistance of the hardened termed as free chlorides or water soluble are dissolved in
concrete which depends on various factors like porosity, w/ pore solution which actually involved in corrosion initiation
c ratio, cement content, curing age, temperature, oxygen (Page 1986, Song and Sarswati, 2007). In chemical
and moisture content, source of chloride concentration, binding chloride ions in gel pores reacts with tricalcium
permeability and OH- ion concentration. Practically aluminates (C3A), forms calcium chloroaluminates,
corrosion is controlled by increasing the thickness of 3CaO.Al2O3.CaCl2.10H2O known as Fridel’s salt (Neville,
concrete cover (Neville A,. 1995). Chloride may ingress 1995, Song and Saraswathy, 2007). It reduces the porosity
externally by diffusion or through penetration (Andrade and chloride in the pore water solution. It is seen that
C, 93, Deb S, 2011). The minimum quantity of chlorides cement with higher C3A content have more chloride
available to initiate rebar corrosion under optimal binding capacity, hence good in corrosion resistance as
moisture, temperature and oxygen is already mentioned reducing the free chlorides in the hardened concrete. In
in many literature (Song and Saraswath,y 2007). Various chloride binding, chemical binding by forming Fridel's
codes have restricted the limits of minimum chloride salts contribute more than physically absorbed with C-S-H
concentration in the concrete. According to the European gel (Arya and Newman, 1990, Neville, 1995, Pradhan and
standard - EN 206, the maximum allowed chloride Bhattacharjee, 2007).
contents are 0.2– 0.4% chloride ions by mass of binder
for reinforced concrete and 0.1–0.2% for prestressed Free Chlorides
concrete, while USA has limit of 0.15% for chloride The free chlorides which are soluble in gel pores are
exposures and 0.30% for chloride free exposure In India responsible for corrosion initiation. Once chloride gets
the limit is 0.15%. As per BS 8110-Part I-1995, the total chemically bound do not participate in corrosion (G. Glass
chloride count is –0.4% by the mass of cement. and Buenfeld, 1997, Reddy, 2002). However, a recent study
of chemical aspects of binding suggests that the bound
Chloride Binding chloride dissolves and may subsequently be involved
in initiation of corrosion by releasing the part of bound
The interaction between chloride ions and the porous
chloride when the pH drops to values below 12, which
concrete matrix is defined as chloride binding. It results
may happen locally in voids at the steel/concrete interface
in effective removal of free chloride ions from the pore
( Bertolini, 2004). The free chloride has strongest effect
solution hence reducing the risk of corrosion (Glass and
in defining the different zones of corrosion of concrete
Buenfeld, 1997). Total chlorides available from various
irrespective to the type of steel and w/c ratio but influenced
sources in the concrete are in different forms like
by type of cement (Broomfeld, 2002).
physically absorbed, chemically bound and free chlorides
or water soluble chlorides. Chloride binding is a very Generally about 40% to 50% of total chloride get bound
complicated process affected by various factors but the (Gaynor, 1985). It is also mentioned that the ratio of bound
main factor influences is C3A content of the cement which chloride to free chloride is 1:3 (Dhir et al., 1990). When the
having higher chloride binding capacity. A higher chloride chloride content reached to the level of steel reinforcement
binding capacity is associated with less corrosion risk exceeds the threshold level (critical chloride content- is
(B.Pradhan, 2007, Song and Saraswathy, 2007). the ratio of chloride ion to hydroxyl ions (Cl-/OH-)), it will

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

358 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Role of Supplementary Cementitious Materials on Chloride Induced Corrosion - An overview

break the passivity and contribute in accelerating the rate Influence of Supplementary Cementitious
of corrosion (S Fleu et al, 1990, Xu Shi, 2011). It is also Materials on Corrosion in Concrete
reported that there is linear relationship between pH
and logarithm of chloride concentration which show that It is now accepted that there is need to use supplementary
with increase in pH of pore solution, OH- effect become cementitious material to replace Portland cement (OPC)
stronger (Li & Sagues, 2001). Equilibrium conditions is for enhanced environmental and durability performance.
established between free chloride ions, chemically and From last few decades’ intensive research is going
physically bound chlorides, depending on the composition worldwide on the addition of widely and easily available
of the cement and its binding capacity. Therefore, it is pozzolanic materials. The urge to reduce emissions of
possible that the concentration of free chloride ions in carbon dioxide (1 tonne of OPC generates almost 1 tonne of
the pore solution of different concrete varies, even if the CO2) (A. Robert, 1998, Absora, 2011). Pozzolanic material
total chloride content is the same. The corrosion rate may be artificial or natural rich in reactive silica and are
increases with the increase of free chloride content in glassy in nature. It is not having cementitious property but
concrete (A Sharopoulou, 2009). Critical chloride level when mixed with water the finely grinded pozzolanas will
leading to corrosion initiation is not a unique value and form cementing compounds due to chemical reaction with
varies with steel type, cement type, and w/c ratio. With the alkalis in cement (Mike, 2014, Xamning, 2012, Ahmed and
increase in C3A content it found that there is considerable Bhattacharjee, 2003). When these materials are mixed
reduction in water soluble chlorides in internal chlorides with cement in proper proportion, improves the pore
(Rusheeduzzffer, 1992). It is also noted that pure C3A and structure hence increase the resistivity of the concrete
gypsum has better chloride binding capacity compared to and hence controls the rate of corrosion by controlling
OPC and C3A paste when it exposed to different chloride the ingress of harmful chloride ions. Along with durability
concentration (Q Yuan et al, 2009). Other research also improvement by using supplementary cementitious
shows that when OPC and sulphate resisting cement material it also helps in reduction in cost of material,
subjected to 20g/l NaCl solution, OPC give considerably environmental protection and improving the properties
more chloride bound than sulphate resisting Portland of concrete. It is also noted that fly ash and slag can be
cement (SRPC) as it contents less C3A content. From many added to OPC at the time of preparation of concrete or
other research it is concluded that the cement with higher can be blended with OPC clinkers and inter grinded at
C3A content have higher chloride binding capacity in the time of manufacturing. The effect of such blending on
case of internal chloride (Ethesham and Rusheeduzzffer, durability properties of the concrete may or may not be
1994 , Deflagrate Anil, 1997). The reaction between C3A similar to that of addition at the time of casting (Pradhan
or C4AF and chloride ions in the pore solution results and Bhattacharjee, 2007).
in formation of Fridel’s salt in chemical binding. Though Many literatures reported that corrosion initiates when
higher alkalis in pore solution reduces chloride binding the chloride concentration in gel pores surround the
capacity of cement hydrates, they helps in increasing the reinforcement reached to critical level. Pozzolana either
OH- concentration which is a dominant factor in reducing artificial or natural acts as binding materials and reduce
the Cl-/OH- ratio, hence reducing the corrosion risk chloride content hence less free chlorides causing
(Rasheeduzzaffer 1992). It is to be noted that out of total corrosion. Chlorides adversely affect sulfate resistance.
C3A in the cement, 5% is consumed by 3% of gypsum Sulfate attack results in a decomposition of calcium
(CaSO4.2H2O) to avoid flash set (P. K Meheta, 2006). It chloroaluminate, thus making some chloride ions available
has been seen that more chloride binding will reduce the for corrosion and calcium sulfoaluminate is formed. The
corrosion risk (Ann and Song, 2003). It has been reported risk of corrosion in sulphate resistant Portland cements,
that use of super plasticizer in concrete lowers the which are characterized by a low content of tricalcium
chloride binding capacity (Song G and Ahmed S, 1998). The aluminate (C3A), is higher than in normal Portland
sulphate ion concentration influences the chloride binding cement, considering an equal content of total chloride (C.
capacity for a given binder (Sansbergs, 1993, Byfors ,1990). Page 1986, Apostolopoulos and Papdkis, 2008). The use of
It is also reported that there is different effect of sodium mineral admixtures lowers the permeability and refines
sulphate and calcium sulphate on chloride binding and the micro structure of cementitious materials reduces the
chemistry of pore solution. Calcium sulphate has higher conductivity of electrolytic concrete pore solution (Ahmed
binding capacity than sodium sulphate for the same and Bhattacharjee, 2003, Pradhan and Bhattacharjee,
sulphate content of sulphate in cement because of their 2009, Polder, 1996). These materials have proved not
different effect on OH- ion concentration in pore solution. only to be good in improving the durability of concrete but
Sodium Sulphate increases the alkalinity where calcium also to protect the environment and to conserve energy
sulphate decreases it (Xu, 1997). It is also found that high resistance to chemical attack at later ages due to
chloride binding is improved significantly by the increase lower permeability and less calcium hydroxide available
in concentration of hydroxyl ion (OH-) in external chloride for reaction, and low diffusion rate of chloride ions that
bearing environment (Sandbergs, 1993, Page CL, 1991). results in a higher resistance to corrosion of steel in
concrete (ACI-222R-01, Sharma, Mukharjee, 2011). A

Organised by
India Chapter of American Concrete Institute 359
Session3 C - Paper 4

partial re-placement of Portland cement by fly ash, GGBS Fly Ash(FA)

results in a refinement of the pore structure of the cement It is a by-product of thermal power electricity generation
paste, have lower free chloride concentrations than the plant by combustion of coal. It is the most common
original Portland cement (Pradhan and Bhattacharjee, pozzolan used worldwide. Using fly ash with OPC
2007). Although their pH is slightly lower which leads to increases porosity at early stage but average pore size
a higher resistance of the concrete against the ingress of is decreased which makes concrete less permeable and
chloride ions (ASTM, 2005, Neville, 1995). hence more resistance to harmful chloride penetration
(Mohammed 2002). Other researcher, in his research
Supplementary Cementitious Materials replaced OPC by 30% fly ash with three different fineness
and two w/c ratio, used rapid chloride penetration test
Type of cement greatly influence the relation between
and found that chloride penetration is decreased with
bound and free chlorides pore structure of cement paste,
increased fineness of fly ash and increase with increase
degree of hydration which is the main cause to transport
in w/c ratio (P Chindaprasirt, 2007). OPC with different
mechanism of chloride and required corrosion products
percentage of fly ash and found that chloride binding
i.e. moisture and oxygen ( Arya, 1990). However for the
capacity is increased with increase in replacement upto
corrosion to occur the requirement of chloride and oxygen
50% and deceased after 67% (Dhir, et al). The high alumina
are at different locations i.e. chlorides required at anode
content in fly ash tends to form more fridel’s salts result
and oxygen at cathode. The pore solution is the electrolyte
in more chloride binding (Arya, 1990). Fly ash from a coal-
which is alkaline in nature due to alkalis in cement i.e.
fired power station is a pozzolana that results in low-
KOH and NaOH which keep the ph of cement concrete 12
permeability concrete, which is more durable and able to
and above. Pozzolanic materials like fly ash (FA), ground
resist the ingress of deleterious chemicals (Newmann,
granulated blast furnace slag (GGBS) and metakaolin are
2003). In other study it is reported that total amount of
widely used to improve the durability of concrete in marine
alumina and iron oxide in OPC replaced with fly ash shows
environment (U Angust 2009).
higher chemical chloride binding capacity (Arya, 1992).
It is found that fly ash containing high alumina increases
Ground Granulated Blast Furnace Slag (GGBS
the chloride binding capacity of the binder. Therefore,
It is the major by-product of steel manufacturing process. there is an increasing tendency worldwide to use blended
The fineness of slag is more (>400 m2/kg) than OPC (310 cements containing pozzolanic materials (Absora, et al
m2/kg). The basic constituents of the slag are Alumina, 2011). Portland Pozzolanic cement (PPC) improves in
Silica and quicklime. When mixed with water, it forms resistance to chloride penetration than OPC (B Pradhan
calcium silicate hydrates as like OPC but the properties of and B Bhattacharjee, 2007).
cement paste like strength, porosity and heat of hydration
are influenced by addition of slag (Beushausen, A 2012). Metakaolin (MK)
Slag cement affects the permeability and improved
Mineral Kaolinite is fired at 500 to 800°c in kiln; the product
microstructures (Arya, 1992).In initial reactions of physical
formed is calcined clay. There is removal of chemically
and chemical binding of chloride ions refines the pore
bonded hydroxyl ions by evaporating water during heating.
structures of hydrated products reduces the diffusion rate
It is rich in reactive alumina silicates pozzaolana. ASTM C
of surface chloride by blocking the path ( Arya and Xu, 1995,
168 also recommended using MK as a part of cementitious
Glass and Buenfeld, 1997). GGBS is having 50% smaller
material. It give good early strength with low w/c ratio. It
diffusion coefficient than that of normal OPC (T Darren
is also found that MK consumes portlandite in pozzolanic
et al. 2011). It is suggested that the total replacement is
activity which refines the pore structures (Justine and
about 50 % of the cement but generally between 40 to
Kuris, 2007). In one of the research OPC was replaced by
65 % (ACI 233R 95). Chloride binding capacity increases
10 and 20 % by MK by the weight of cement and concluded
with replacement level of slag which dilutes the effect
that it reduced the chloride penetration by decreasing the
of sulphates by slag (Xu Y, 1997). It also increases with
mean pore size also improving the uniformity of pore size
increase in w/c ratio (Tang and Nelson, 1996). In another
distribution (Xi she et al, 2012).Other study found that MK
research different type of slag by replacing the OPC
is better than GGBS in early age compressive strength
by 0%, 20%, 35%, 50% by mass of total binder with
gaining (Khatib J, 2005). In another research MK was
two different w/c ratio(0.40 and 0.60) concluded that
replaced by 20% for different w/c ratio (.35, 0.55). Result
chloride ion penetration and porosity decreased with
shows that MK reduces chloride penetration to great
increased with increase in replacement level (Mike, 2014).
extent. Permeability is mainly depending on replacement
Partial replacement of ordinary Portland cement (OPC)
level, w/c ratio, curing condition and chloride exposure
with GGBS has been reported to increase the chloride
period (Erhan Guneyisi, 2007). But there are very few
binding capacity of the binder (Page, 1986, Neville, 1995,
literatures on chloride binding capacity and corrosion
Apostolous, 2008). Portland slag cement (PSC improves
electrochemistry of MK. There is further need to study in
in resistance to chloride penetration than PPC and OPC (B
this aspect.
Pradhan and B Bhattacharjee, 2007).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

360 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Role of Supplementary Cementitious Materials on Chloride Induced Corrosion - An overview

Discussion Conclusion
The purpose of present review summarises the state of The most common cause of reinforcement corrosion
art regarding the effect of pozzolana or supplementary is chloride ingression. Apart from the total chloride in
cementitious materials on chloride induced corrosion. concrete gel pores, only free chloride initiates corrosion,
Numerous studies in past few decades were undertaken other get chemically or physically bound. The mineral
to overcome the durability issue by adding pozzolanic admixtures increases the chloride binding hence reduces
material with cement increase the chloride binding which the free chloride. Mineral and chemical admixtures can
need detailed study. By adding this material the rate of be used to reduce permeability by using permeability
corrosion can reduced. reducing admixtures such as fly ash, blast furnace
slag and metakaolin. Corrosion of reinforced concrete
1. The process related to corrosion mechanism of
is influenced by may parameters like concrete mix
reinforced concrete have been investigated intensively
design, water-cement ratio, type of cement and type of
and depth also. However there is still conflict in the
reinforcement, chloride exposure condition. The effect of
relation between free chloride and bound chloride.
type of binder to resist the chloride ingress, type of steel,
Different values are found in different literature.
its type, its composition, content of alloy materials and its
2. The corrosion rate in blended cement concretes is manufacturing process need to study in depth. Corrosion
governed mainly by the high concrete resistivity, in concrete is unavoidable, so the best combination of
whereas in OPC concretes, it is governed by the steel, blended cement which can control the corrosion and
availability of oxygen (i.e. cathodic controlled) and hence the rate of corrosion is required to study in depth.
consequently, by cover depth. Analysing data on bound
and free chloride from literature found that among the References
factors influencing bound versus free chloride were: 1. Arya, C., et al., 1990. Factors influencing chloride-binding in
C3A content, FA addition, addition of GGBS. concrete. Cement and Concrete Research, 20(2):291-300.
2. Andrade, C., et al., 1993. Calculation of Chloride diffusion in concrete
3. Slag blended cement is likely to enhance the corrosion
from ionic migration measurements. Cement Concrete Research,
initiation period due to lack of available free chloride in 23: 724.
the pore solution of the concrete compared with PPC 3. Ahmad, S., Bhattacharjee B., 1995. A simple arrangement and
and OPC. procedure for in-situ measurement of corrosion rate of rebar
embedded in concrete. Corrosion Science Journal, 37(5):781–91.
Also there is no significant variation in the values of
critical free chloride contents with steel type and w/c 4. Arya, C., Xu Y., 1995. Effect of cement type on chloride binding
and corrosion of steel in concrete. Cement Concrete Research,
ratio. But it significantly affects by the type of cement 25(4):893–902.
and concluded that PSC exhibited highest chloride
5. Andrade, C., Alonso C., 1996. Corrosion rate monitoring in the
tolerance against risk of pitting or localized corrosion laboratory and on site. Construction and Building Materials,
compared to PPC and OPC. 10(5):315–328.

4. The effect of corrosion product on chemical and 6. Andrade, C., et al., 2001. Electrochemical behavior of steel rebar in
concrete: influence of environmental factors and cement chemistry.
electrochemical process is also need to refine. There Electrochemica Acta, 46 : 3905–3912.
is also need to study the effect of chloride exposure
7. Andrade, C., et al., 2002. Corrosion rate evolution in concrete
condition on corrosion, because in practice and lab structures exposed to the atmosphere. Cement Concrete
condition exposure varies considerably. Composites, 24: 55–64.

5. Chloride has to be introduced by combination of 8. Ahmad, S., Bhattacharjee B., 2003. Reinforcement corrosion in
concrete structures, its monitoring and service life prediction– A
capillary suction and diffusion and not internally review. Cement and Concrete Composites, 25: 459–471.
through ingredients.
9. ASTM. 2005. Standard test method for determining the effects
6. The various factors affecting chloride ingress of chemical admixtures on the corrosion of embedded steel
reinforcement in concrete exposed to chloride environments. ASTM
porosity (permeability), w/c ratio, chloride content
G109.
and transport mechanism need to study. Reducing
10. Ann, K.Y., Song H.W., 2007. Chloride threshold level for corrosion
w/c ratio, increasing cover, increasing compressive of steel in concrete. Corrosion Science, 49:4113–4133.
strength is commonly suggested to improve durability
11. Apostolopoulos, C.A., Papadakis V.G., 2008. Consequences of
but they may affect one or other way to corrosion. steel corrosion on the ductility properties of reinforcement bar.
Construction and Building Materials, 22:2316–2324.
7. New materials like Metakaolin and micro silica which
are more effective and corrosion resistance might be 12. Angus,t U., et al., 2009. Critical chloride content in reinforced
concrete — A review. Cement and Concrete Research, 39:1122–1138.
introduced into reinforced concrete. Also reinforced
concrete combination of metal and cement, both are 13. Abd, E. E., et al, 2009. Environmental factors affecting corrosion
behaviour of reinforcing Steel Measurement of pitting corrosion
having different chemistry, while comparing results of current of steel in CaOH2 solution under natural corrosion condition.
corrosion as a whole. Corrosion Science, 51:1611-1618.

Organised by
India Chapter of American Concrete Institute 361
Session3 C - Paper 4

14. Abosrra, L., et al. 2011. Corrosion of steel reinforcement in concrete 39. Glass, G. K., Buenfeld N. R., 2000. The influence of chloride binding
of different compressive strengths. Construction and Building on the chloride induced corrosion risk in reinforced concrete.
Materials, 25:3915–3925. Corrosion Science, 42:329-344.
15. Aleksandra, M., et al., 2013. Properties of composite cement with 40. Fajardo, G. P., et al., 2009. Corrosion of Rebar embedded in natural
commercial and manufactured Metakaolin. Technical Gazette MK, pozzolan based mortar exposed to chlorides. Construction and
20(4): 683-687. Building Materials, 23:768-774.
16. Byfors, K., 1990. Chloride initiated reinforcement corrosion, chloride 41. Glass, G.K. & Buenfeld, N.R. 1997b. The presentation of the chloride
binding. CBI report 1: 90. threshold level for corrosion of steel in concrete. Corrosion Science
17. Bautista, A., Gonzalez, J. A., 1996. Analysis of the Protective Efficiency 39: 1001–1013.
of Galvanizing against Corrosion of Reinforcements Embedded in
42. Glass, G.K. et al., 1996. An investigation of experimental methods
Chloride Contaminated Concrete. Cement and Concrete Research,
used to determine free and total chloride contents. Cement and
26(2):215-224.
Concrete Research 26: 1443–1449.
18. Broomfield, J. P. 1996. Field Measurement of the Corrosion Rate
of Steel in Concrete Using a Microprocessor Controlled Unit with 43. Glass, G.K., Buenfeld, N.R. 1997a. Chloride threshold levels for
a Monitored Guard Ring for Signal Confinement. ASTM STP 1276. corrosion induced deterioration of steel in concrete. In Proc. RILEM
Int. Workshop Chloride penetration into concrete, Paris.
19. Bamforth, P., 1999. The derivation of input data for modeling chloride
ingress from eight-years UK coastal exposure trials. Magazine of 44. Gu, X. L., et al., 2010. Flexural Behavior of Corroded Reinforced
Concrete Research, 51: 87–96. Concrete Beams. Earth and Space, ASCE: 3545-3552.
20. Broomfield, J. P. 1997. Corrosion of steel in concrete, Understanding, 45. Hossain, A. B., Shrestha S, Summers J. Properties of concrete
Investigation and Repairs. Chapman and Hall. incorporating ultrafine fly ash and silica fume. Transp Res Rec: J
21. Bentur, A., 1997. Steel corrosion in concrete. E. & F.N. Spon, London. Transp Res Brd 2009, 2113:41–52

22. Broomfield, J.P., Millard, S.G., 2002. Measurement of concrete 46. Jiirgen, M., Bernd I., 1996. Monitoring of concrete structures with
resistivity to assess corrosion rate. Institute of Corrosion / Concrete respect to rebar corrosion. Construction and Building Materials,
Society: Good Practice Guide, 128:37-39. 10:361-313.
23. Bertolini, L et al., 2004. Corrosion of Steel in Concrete: Prevention, 47. Kouloumbi, N., et al., 1995. Efficiency of Natural Greek Pozzolan
Diagnosis, Repair. Wiley and Sons, Cambridge, UK, 409. in Chloride-induced Corrosion of Steel Reinforcement. Cement
24. Beushausen, H., et al., 2012. Early age properties, strength Concrete and Aggregates, CCAGDP, 171:18-25
development and heat of hydration of concrete containing slag. 48. Kyle, D. S., 2002. The migration of chloride ions in concrete. PhD
Construction and Building Materials, 29:533-540. thesis, Department of Civil Engineering, University of Toronto,
25. Luo, R, et al., 2003. Study of chloride binding and diffusion in GGBS Canada.
concrete. Cement Concrete Research, 33:1–7. 49. Kapat, C., Pradhan B., et al. 2006. Potentiostatic study of reinforcing
26. Chindaprasirt, P, et al., 2007. Influence of fly ash fineness on steel in chloride contaminated concrete powder solution extracts.
chloride penetration of concrete. Construction and building Corrosion Science, 48:1757–1769.
Materials, 21:356-361. 50. Liu, Y., Weyers, R. E., 1998. Modelling the time-to-corrosion cracking
27. Dhir, R.K. et al. 1990. Determination of total and soluble chlorides in chloride contaminated reinforced concrete structures. ACI
in concrete. Cement and concrete Research, 20(4):579-585. Materials Journal, 956:675-681.
28. Dhir, R.K. et al. 1996. Chloride binding in GGBS concrete. Cement 51. Li, L., Sagues, A.A., 2001. Chloride corrosion threshold of reinforcing
and concrete Research, 26:1767-1773. steel in alkaline solutions—Open-circuit immersion tests. Corrosion
29. Dhir, R. K., Jone M.R., 1999. Development of chloride-resisting 57(1): 19–28.
concrete using fly ash Fuel, 78:137-142. 52. Luo, R, et al., 2003. Study of chloride binding and diffusion in GGBS
30. Deb, S., Pradhan B., 2011. Rebar corrosion in chloride environment. concrete. Cement Concrete Research, 33:1–7.
Construction and building materials, 25:2565-2575.
53. Mangat, P.S., Khatib J. M., 1994. Molloy BT. Microstructure, chloride
31. Elsener, B, Bohm H., 1990. Potential Mapping and Corrosion of Steel diffusion and reinforcement corrosion in blended cement paste and
in Concrete, Corrosion Rates of Steel the Concrete, ASTM STP 1065. concrete. Cement Concrete Composites, 16(2):73–81.
32. Elsener, B., Bohni, H., 1994.Galvanostatic pulse measurements. 54. Mehta, PK., 1997. Durability – Critical issues for the future. Concrete
Rapid on-site corrosion monitoring. International Conference on International, 19-27.
Corrosion and Corrosion Protection of Steel in Concrete. R.N.
Swamy, Sheffield, 236-246. 55. Montemor, M.F., et al. 2003. Chloride-induced corrosion on
reinforcing steel: from the fundamentals to the monitoring
33. Ehtensham, H.S., et al., 1994. Influence of sulphates on chloride
techniques. Cement & Concrete Composites, 25:491–502
binding in cements. Cement and concrete Res., 24:8-24.
34. Elsener, B., et al., 1997. Assessment of reinforced corrosion by 56. Millard, S., 2003. Measuring the corrosion rate of reinforced
means of galvanostatic pulse technique. International conference- concrete using linear polarisation resistance. Cement & Concrete
Repairs of concrete structures: 391-402. Composites, 6:36–38.

35. Elsener, B., et al., 2003. Non destructive determination of the 57. Magdy, A.M., 2009. Corrosion behaviour of some austenitic stainless
free chloride content in cement based materials. Materials and steels in chloride environments. Materials Chemistry and Physics,
Corrosion, 54: 440–446. 115:80–85.
36. Erhan, G., et al., 2013. Corrosion behaviour of reinforcing steel 58. Mike, B., et al., 2010. Suitability of Various Measurement Techniques
embedded in chloride contaminated concretes with and without for Assessing Corrosion in Cracked Concrete. ACI Material Journal,
Metakaolin. Composites: Part B, 45:1288-1295. 107:481-489.
37. Feliu, S et al., 1990. Influence of counter electrode size on the on- 59. Mike, B., et al., 2014. Effect of chemical composition of slag on
site measurement of polarization resistance in concrete structures. chloride penetration resistance of concrete. Cement & Concrete
NACE, Corrosion 90:142. Composites, 46:56-64.
38. Glass, G. K., Buenfeld N. R. 1997. Chloride threshold level for 60. Neville, A.M., 1995. Chloride attack of reinforced concrete: an
corrosion of steel in concrete. Corrosion Science, 39:1001- 1013. overview. Materials and Structures, 28:63-70.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

362 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Role of Supplementary Cementitious Materials on Chloride Induced Corrosion - An overview

61. Oh, B. and Jang S., 2007. Effects of material and environmental 76. Stanish, K., et al., 2004. A novel method for describing chloride
parameters on chloride penetration profiles in concrete structures. ion transport due to an electrical gradient in concrete. Cement &
Cement and Concrete Research, 37:47–53. Concrete research, 34:43-49.
62. Page C. L., et al., 1986. The influence of different cements on 77. Song, H., Saraswathy V., 2007. Corrosion monitoring of
chloride induced corrosion of reinforcing steel. Cement and Reinforced concrete structures -A review. International Journal of
Concrete Research, 16:79–86. Electrochemistry Science, 2:1-28.
63. Page, C. L., et al., 1990. The influence of different cements on 78. Sadowski, L., Millard S., 2008. Non-destructive assessment of
chloride induced corrosion of reinforcing steel. Cement and the corrosion risk to reinforcing steel using a measurement of the
Concrete Research, 16:79–86. resistance of concrete. 5:35–39.
64. Pavla, H., Rache D., 1995. Water permeability and chloride ion 79. Sharopoulou, A., et al., 2009. Thaumasite form of sulphate attack
diffusion In portland cement mortars. Cement and Concrete in lime stone cement motars: a study on long term efficiency of
Research, 25:790-802. mineral admixtures. construction and building Materials, 23:2338-
2345 .
65. Pech, M.A.,Canul P., 2002. Corrosion measurements of steel
reinforcement in concrete exposed to a tropical marine atmosphere. 80. Sharma, S., Mukherjee A., 2011. Monitoring Corrosion in Oxide and
Cement and Concrete Research, 32:491–498. Chloride Environments Using Ultrasonic Guided Waves. Journal of
Materials and Civil Engineering, 23:207-211.
66. Poon,, C.S., et al., 2006. Compressive strength, chloride diffusivity
and pore structure of high performance metakaolin and silica fume 81. Shi, X., et al., 2011. Strength and corrosion properties of Portland
concrete. Construction and Building Materials, 20:858–865. cement mortar and concrete with mineral admixtures. Construction
Building Materials, 25(8):3245–3256.
67. Pradhan, B., Bhattacharjee B., 2007. Role of Steel and Cement Type
on Chloride-Induced Corrosion in Concrete ACI Material Journal, 82. Tang, Y, Nelson , 1992. Effect of condition of Chloride diffusivity in
104:612-619. silica fumes high strength concrete. 9th International congress on
chemistry of cement, 5:100.
68. Rasheeduzzaffar, S. E., Hussain, S., 1991. Effect of cement
composition on chloride binding and corrosion of reinforcing steel 83. Thomas, M. D., et al., 1999. Use of ternary cementitious systems
in concrete, Cement and Concrete Research, 21(5):777–794. containing silica fume and fly ash in concrete. Cement Concrete
Res, 29:1207–14.
69. Rodriguez, P., et al., 1994. Methods for Studying Corrosion in
Reinforced Concrete. Magazine of Concrete Research, 167(46):81-90. 84. Tareq, S., Muftar S., 2011. Effect of chloride ions source on corrosion
of reinforced concrete, Buletinul AGIR, 2:107-112.
70. Raupach, M., 1996. Chloride-induce macro-cell corrosion of steel
in concrete-theoretical background and practical consequences. 85. Tang, Y M., et al., 2012. Corrosion behaviour of steel in simulated
Construction and Building Materials, 105:329–338. concrete pore solutions treated with C-S-H. construction and
building Materials, 30:252-256.
71. Reddy, B., et al., 2002. On the corrosion risk presented by chloride
bound in concrete. Cement Concrete Composites, 24(1):1–5. 86. Xianming, Shi et al., 2012. Durability of steel reinforced concrete
in chloride environment – An Overview. Construction and building
72. Rangaraju, PR, Desai J. 2009. Effectiveness of fly ash and slag in
Materials, 30:125-138.
mitigating alkali– silica reaction induced by deicing chemicals. J
Materials Civil Eng, 21:19–31. 87. Yeomans, S.R., 1994, Performance of black, galvanized and epoxy-
coated reinforcing steels in chloride-contaminated concrete.
73. Sandburg, P., Larsson J., 1993. Chloride binding in cement pastes in
Corrosion, 50:72–81.
equilibrium with synthetic pore solution. International proceedings
Nordic Seminar: 13-14. 88. Yuan, Q., et al., 2009. Chloride binding of cement based materials
subjected to external chloride environment- A review. Construction
74. Song, G., Ahmad S., 1998. Corrosion of steel in concrete: causes,
and building Materials, 23:1-13.
detection and prediction. ARRB Transport Research:1-56.
89. Zhang, D., et al., 2011. Effects of mineral admixtures on the chloride
75. Suryavanshi, A K., et al., 1998. Corrosion of reinforcement steel
permeability of hydraulic concrete. Advanced Materials Res, 8:168–
embedded in high water-cement ratio concrete contaminated with
70.
Chloride. Cement and Concrete Composites, 20:263-381.

Organised by
India Chapter of American Concrete Institute 363
Session3 C - Paper 4

Ms. Anita N Borade

Ms. Anita N Borade is a full time Ph.D. Research Scholar at S.V. National Institute of Technology (SVNIT),
Surat, India. She received her bachelor’s degree from North Maharashtra University (NMU), Jalgaon,
Maharashtra, India, her master’s degree from Pune University (PU), Maharashtra. She is having 9 years
experience as Assistant Professor in Pune University, Maharashtra, 03 years experience as Sub Engineer
(SE), Brihan Mumbai Corporation (BMC), Mumbai and 01 year as Design Engineer in MNC in Mumbai
Maharashtra, India. Her research interests include durability of concrete, fibre reinforced concrete,
pozzolanic materials and high performance concrete.

Dr. B. Kondraivendhan
Dr. B. Kondraivendhan is an Assistant Professor at S.V. National Institute of Technology, Surat. He received
his bachelor’s degree from Alagappa Chettiar College of Engineering and Technology (ACCET), Karaikudi,
Tamil Nadu, India, his master’s degree from Annamalai University, Chidambaram, Tamil Nadu and his
doctoral degree from Indian Institute of Technology (IIT) Delhi, India. He has Post-Doctoral Fellowship
(P.D.F) experience as well at Nanyang Technological University (NTU) Singapore. He has published 9
highly reputed journal papers, 6 conference proceedings and one book on service life prediction model on
reinforced concrete structures. His research interests include strength, durability, repair and rehabilitation
and microstructure of cement-based composites.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

364 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Pervious Concrete - A Value added material

Pervious Concrete - A Value added material

S B Kulkarni and Clinton Pereira
UltraTech Cement Limited, Mumbai

Introduction
According to world statistics, the potable water on the
earth’s surface is only 3% as compared to the total water
cover which is approximately 71% of the earth’s surface.
Because of merciless cutting down of trees to develop
habitable areas, the intensity of rainfall has considerably
decreased in many parts of the world. Several locations
which once received heavy rainfall are presently suffering
from droughts and as a result they are experiencing
an acute shortage of fresh water, which in turn has an
adverse effect on human life. The major issue faced today
by several nations across the world is effective control
& efficient use of the available fresh water for human
habitation. Even though some areas receive considerable
amount of rainfall the water does not percolate into the
subsoil because of man-made impervious surfaces,
like concrete & bituminous pavements, which results Pervious concrete:
in lowering of the ground water table. Hence, it is very
American Concrete Institute’s committee 522 defines
important to recharge the ground water table to ensure
Pervious concrete as hydraulic cement concrete
percolation of rain water by using efficient rain water
harvesting systems. During monsoons, water logging is proportioned with sufficient interconnected voids that
seen in almost all metro cities because we have converted result in a highly permeable material allowing water
our land into a concrete jungle. The concrete & asphalt to readily pass through. Pervious concrete is a special
pavements constructed around buildings, city roads and concrete designed with a high porosity generally used for
footpaths prevent the rain water from seeping into the concrete pavements that allow rain water or water from
ground. The storm water drains are also not capable of any other sources to pass through it and percolate into the
discharging rain water appropriately, resulting in water ground below, thereby reducing runoffs and recharging
logging. On the other hand, this fresh water gets wasted ground water levels. Pervious concrete is a mixture of
and does not help in enhancing the ground water table, Portland cement, Pozzolanic materials like fly ash, GGBS,
indirectly affecting the balance of the ecosystem. etc. coarse aggregate, water and with very little or no
sand. A typical pervious concrete pavement has a 15-30%
Recent, Research & development in concrete technology void structure, with compressive strengths of 5-25 MPa
has resulted into development of a special type of and allows approximately 10-30 liters of water per minute
concrete known as Pervious concrete or No fines
to pass through each square foot.
concrete, especially used for rain water conservation
& management. Pervious concrete is found to be an
excellent solution to the above referred issue, which helps
Manufacturing Pervious Concrete
in protecting our Mother Earth. As this concrete contains Carefully controlled amounts of water and cementitious
around 15-30% voids, it allows water to percolate and gets materials are used to create a paste that forms a thick
absorbed in the sub soil, thereby recharging the ground coating around aggregate particles without allowing the
water table and reducing storm water runoff. This article slurry to run off during mixing and placing. Using just
deals with the various aspects of Pervious Concrete in enough paste to coat the particles, maintains a system of
terms of mix design parameters, manufacturing process, interconnecting voids. The result is a very high permeable
construction, maintenance, advantages, limitations & also concrete that drains the water quickly, termed as Pervious
emphasizes on the Green aspect of Pervious concrete. Concrete.

Organised by
India Chapter of American Concrete Institute 365
Session 3 C - Paper 5

After placement, pervious concrete resembles a sponge. concrete is less. However, joints in pavements must
Its low paste content and low fine aggregate content make be provided as per the design parameters or as per the
the mixture harsh, with a very low slump. The compressive relevant code for practice (reference IS 456:2000 &
strength of pervious concrete is limited since the void IS 11817). The recommended depth is 1/3 depth of the
content is so high. However, compressive strengths of 5 pavement and mandatorily to be cut within 24 hours of
to 25 MPa are typical and sufficient for many applications. concrete placement. Joints may be saw cut mechanically
or manually by using a rolling jointing tool.
Concrete Materials
ll The cementitious content varies in the range of 250- Curing
350 Kgs/CuM. Ordinary Portland cement or Portland This in one of the most important stage of construction of
Pozzolana cement can be used with the additions of a pervious pavement, as the open pore structure makes it
mineral admixtures Fly Ash, GGBS, etc. prone to plastic shrinkage. Curing must start immediately
after finishing the pavement and continue for a minimum
ll The aggregate content varies in the range of 1200- period of 7 days (as per IS-456:2000, clause 13.5) As
1700 Kgs/CuM. Coarse aggregates of 10 mm size is to regular water spraying and ponding method of curing
be used, but larger coarse aggregate size upto 20 mm cannot be used in case of pervious pavements, covering
may also be used. with polythene sheets is usually recommended.
ll Water cementitious ratios can be used in the range of
0.30-0.40, with the use of water reducing admixtures. Testing of Pervious concrete
ll Colouring pigments may also be used for decorative As compared to any other regular concrete, pervious
concrete applications. concrete is tested for compressive strength, density
& permeability. The compressive strength of pervious
Construction of pervious concrete pavements concrete usually ranges from 5-25 MPa. The samples of
fresh concrete must be taken as per specifications of IS
The success of any concrete pavement depends on the 1199-1999/ASTM C172 and cubes must be made, cured
subgrade preparation. The subgrade must be properly and tested at 28 days in accordance with IS 516-1999/
compacted to provide a uniform & stable surface, to ASTM C39. The density and air content or percentage of
avoid any sub soil settlements leading to pavement voids may be measured and calculated in accordance with
failures. The type of soils (clayey, silty, sandy, gravely, etc.) IS1199-1999. The percentage of void must be in the range
encountered decides the level of subgrade compaction for of 15-30% and dry rodded density in the range of 1600-
a particular pavement design. Geo textile fabrics may be 1900 Kgs/CuM. Alternatively, the void content of pervious
used to separate fine grained soils from the stone layers. concrete is measured as a percentage of air by ASTM
Depending on the hydrological design of pervious concrete C138/ C138M and the density is measured using ASTM
pavements, an aggregate reservoir bed of pre designed C1688/ C1688M.
thickness should be placed between the subgrade &
concrete layer. The subgrade must be properly moistened Case study- UltraTech Pervious.
before concrete placement, to prevent loss of water from
the Pervious concrete. Perforated pipes may be placed Case study of Pervious concrete pavement construction
in the sub grade or aggregate reservoir bed to distribute by UltraTech Concrete, Chennai.
runoff evenly and provide additional percolation volume. UltraTech Concrete team has recently placed 40 CuM.
The concrete may be placed in wooden or steel forms of pervious concrete for a private bungalow located
placed on the subgrade, having a depth of the pavement at Nandavakkam in Chennai. This concrete was used
design. Manual or vibratory screeds may be used to level for the walkway & driveway admeasuring 3000 sq.ft
the concrete surface and partially compact it in order to (approximately) & the customer was very concerned
ensure that aggregate interlock takes place. A roller float regarding conservation of water and had approached
may be used to level the top surface of the pavement and UltraTech RMC specifically for this purpose. The
consolidate the fresh concrete and provide a smooth riding construction of the concrete pavement was completed
surface. Excessive vibration and rolling the top surface in three major stages: excavation and removal of surface
must be avoided as this lead to decrease in the void content soil, sub-base fill with aggregate layer and pervious
of pervious concrete & accumulation of laitance on the top concrete placement and curing. The details of each are
surface of the pavement, leading to blocking of voids and described below.
decreasing the drainage efficiency of pervious concrete. Excavation & sub base preparation-The top surface clayey
soil and other construction debris was removed up to 150
Contraction Joints mm depth. The surface was well compacted & levelled
With significantly less water content in pervious concrete using a plate compactor. After levelling, 150 mm thick
as compared to regular concrete, shrinkage of this layer of aggregate was placed above the subgrade. The fill

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

366 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Pervious Concrete - A Value added material

was then compacted using a manual roller and levelled to

gain a uniform depth (Photograph-1 shows the sub base
prepared for the concrete pavement)
Concrete placement & curing-The concrete placement
was completed in a single pour to avoid any construction
joints. The concrete mix was supplied by UltraTech RMC
from Poonamallee plant located at Chennai. (Photograph-2
illustrates discharge of concrete form a transit mixer and
manual placement & levelling of concrete). The typical
target parameters of the pervious concrete mix are given
in Table 1. The concrete finishing was done manually using
steel floats. Curing was done for a period of 10 days by
covering the concrete surface with polythene sheets. The
compressive strength of the concrete at 28 days was in
the range of 16-18 MPa and the slump of concrete at site
Fig. 3: Illustrates discharge of concrete from a transit mixer
was between 75-80 mm.
and manual placement & levelling of concrete
Table-1 Mix proportion & characteristics of Pervious concrete
concrete, thereby blocking the pores and decreasing the
Cement 250-350 Kgs/CuM drainage efficiency of a pervious concrete pavement. This
Fly Ash 100-125 Kgs/CuM calls for maintenance of pervious concrete which can be
Aggregate 1200-1700 Kgs/CuM
achieved by pressure washing & power vacuuming. Power
vacuuming helps in extracting foreign matter near the
W/C ratio 0.3-0.4
surface, whereas a pressurized water jet cleans the pores
Admixture 0.5-1.0% beneath. Structural failures include cracking- transverse
Slump 60-90 mm & longitudinal, sub-grade settlement, pot-holes, surface
erosion, etc. which can be taken care of by patch repairs or
Compressive strength (28 days) 10-25 MPa
replacement of concrete in extremely damaged sections.
Bulk density 1400-1900 Kgs/CuM
Void content 15-30% Advantages & Benefits of use Pervious
Concrete.
Pervious concrete provides various environmental
benefits:
ll Storm water Management: By allowing water to
infiltrate, pervious pavements arrests storm water
flow to a great extent and enhances the efficiency of
storm water management systems.
ll Green & Sustainable development: It provides an
opportunity to utilize waste materials like fly ash,
GGBS, etc., thus making it a Green Concrete.
ll Temperature control: Pervious concrete absorbs less
heat than conventional pavements. The open void
structure helps in keeping the pavement cool during
higher environmental temperatures.

Fig. 2: Shows the sub base prepared for the concrete pavement ll Increase in water table: Pervious concrete increases
the ground water table in the surrounding areas, thus
making it possible to get perennial availability of water.
Maintenance
The maintenance of concrete pavement is mainly required ll Soil conservation: The area covered with pervious
due to clogging of pores & structural failures. Clogging concrete helps to arrest erosion of the soil and
occurs when foreign materials enter the pores of pervious subsequent air pollution.
concrete and prevent the designed quantity of water to ll Enhanced green cover: The enhanced water table &
flow through its body. These foreign materials may be fine porous nature of pervious concrete helps growth of
mud, sand, vegetative matter and other debris that gets trees in the surrounding area and also controls the
collected on the surface and penetrates into the body of ambient temperatures.

Organised by
India Chapter of American Concrete Institute 367
Session 3 C - Paper 5

Applications concrete can be used for light traffic, pedestrian walk

Pervious concrete has numerous applications; some of ways, alley ways, and parking lots for light commercial
them are listed below: vehicles. Foreign materials enter the pores of pervious
concrete, resulting in frequent clogging and decreasing
ll Green buildings
the percolating efficiency of pervious concrete pavements
ll Residential roads, alleys, and driveways as per the hydrological design. At times the foreign
ll Parking lots matter may become an integral part of the pavement and
ll Sidewalks and pathways permanently disrupt the functioning & design of a pervious
ll Pavement edge drains
concrete pavement.
ll Foundations/floors for greenhouses, fish hatcheries,
aquatic, amusement centres, and zoos, etc.
Concluding remarks
ll Sub base for conventional concrete pavement
Pervious Concrete has been successfully used across
the world in various applications such as parking lots,
ll Well linings
walkways, light duty pavements, etc. However in order to
ll Tree grates in sidewalks make it more popular government bodies & local municipal
ll Low-traffic pavements authorities have to specify the use of pervious concrete
and make it mandatory for parking lots, driveways,
Using Pervious Concrete to earn LEED points. walkways & other open spaces in all government and
When pervious concrete is used in a project, it contributes private projects. Also this initiative of going Green has to
to qualifying for LEED (Leadership in Energy and be seriously adopted by the private sector, to popularize
Environmental Design) points in the areas of storm water this sustainable concrete which helps in rain water
design, water efficient landscaping, optimized energy conservation and management. To be able to adequately
performance, design to reduce Heat Island Effect, site feed and support the world’s growing population; our
development, protection & restoration of habitat and global economy needs to grow continuously. Water is
innovation in design & use of recycled and locally sourced critical to future growth. By adopting the use of pervious
materials. concrete lets join hands to protect the future of the world
we live in today. Pervious concrete is the answer to take
Specific credits where pervious concrete can aid the
the mission of sustainability forward for the benefit of
designer include:
mankind.
ll LEED Credit SS-C6.1 Storm Water Management – Rate
and Quantity
Acknowledgements
ll LEED Credit SS-C6.2 Storm Water Management –
We would like to thank Mr. Anil Kulkarni & Mr. Anil Kumar
Quantity Control
C. from UltraTech Concrete for sharing a case study on
ll LEED Credit SS-C7.1 Landscape and Exterior Design to pervious concrete walkway & driveway which has been
Reduce Heat Island Effect recently executed at Chennai.
ll LEED Credit WE C1.1 Water-Efficient Landscaping
References
ll LEED Credits MR-C4.1 AND MR-C4.2 Recycled Content
1. American Concrete Institute, ACI 522-Specifications for pervious
(Fly Ash, GGBS, etc.) concrete pavements
ll LEED Credit MR-C5.1 AND MRC5.2 Regional Materials 2. Ashok Kakade, P.E.- “Water conservation & pervious concrete
(within a distance of 800 Kms from manufacturing pavement”.
location to the site) 3. Vinod Vanvari & Dr. Sumedh Mhaske- “Use of pervious concrete in
storm water drain construction in redevelopment building projects”.
Limitations 4. NRMCA- Code in practice- CIP-38 Pervious concrete.
As all other materials have benefits and limitations, 5. Erin Ashley, 2008, NRMCA. “Using pervious concrete to achieve
pervious concrete also has some drawbacks. Due to the LEED points”.
large amount of voids the strength of pervious concrete 6. ASTM codes pertaining to concrete-C-29, C-172, C-138, C-1688.
is lower as compared to normal concrete; hence this 7. BIS codes pertaining to concrete- IS-456: 2000, IS-1199, IS-516,
concrete cannot be used for structural applications. This IS-11817.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

368 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Pervious Concrete - A Value added material

S.B.Kulkarni
Head Technical Services, Key accounts Cell - UltraTech Cement Ltd. Mumbai
Mr. S.B.Kulkarni Civil Engineering Graduate of 1980 from VJTI Mumbai, he has secured Masters in
Management Science in 1992. After his graduation, he worked as a Construction Engineer for 11 years
at Industrial and Residential projects. He has been working in Technical Services department UltraTech
Cement Limited since 1991. He also has experience in the field of Premix Plasters and Construction
Chemicals. Currently, he is working as Vice President and Head Technical Services department for
Key Customers Cell of UltraTech Cement Limited. He has written many technical articles in National &
International conferences and has delivered more than 600 technical lectures.
Mr. S.B. Kulkarni is Life Member of various institutions like ICI, Institution of Engineers, ACI- India Chapter,
Bombay Management Association (BMA) and the Member of American Society of Civil Engineers (ASCE).
He is also member of BIS committee on waterproofing.

Clinton Pereira
Zonal Technical Services Manager, Key accounts Cell - UltraTech Cement Ltd. Mumbai
Mr. Clinton Pereira is a Civil Engineer with Management background having 15 years of professional
experience, with over 9 years at UltraTech Cement. He secured his degree in Civil Engineering from
Shah & Anchor Kutchhi College of Engineering, Mumbai, in the year 2000 & part time Master’s Degree in
Management Science from Mumbai University. Currently, he is working in the Technical Services department
of UltraTech Cement for Key Account customers based at Mumbai. His current job responsibilities includes
supporting the marketing team for all technical matters pertaining to cement & concrete for Key Account
customers in Maharashtra. He has delivered more than 100 Technical lectures on Cement and Concrete
Technology & good construction practices for Engineers & consultants. He has also co-authored articles
on topics related to concrete. He has been certificated as “Concrete technologist of India” by the NRMCA.

Organised by
India Chapter of American Concrete Institute 369
Session 3 C - Paper 6

Preparation of an ultra-high performance concrete

with superior mechanical properties using corundum sand
as fine aggregate under normal heat treatment
Fangyu Han, Jianzhong Liu, Qianqian Zhang, Jianfang Sha Jiaping Liu
State Key Laboratory of High Performance Civil Engineering Materials, State Key Laboratory of High Performance Civil
Jiangsu Research Institute of Building Science, Engineering Materials, Jiangsu Research Institute
Nanjing 211103, China of Building Science, Nanjing 211103, China
Jiangsu Key Laboratory of Construction Materials, Jiangsu Sobute New Materials Co., Ltd., Nanjing 211103, China
School of Materials Science & Engineering, Jiangsu Key Laboratory of Construction Materials,
Southeast University, Nanjing 211189, China School of Materials Science & Engineering,
Southeast University, Nanjing 211189, China

Abstract broad the innovative application of UHPC liking slender

A carefully designed formulation of ultra-high performance structures, thinner or perforated facades, decorated
concrete, which was expect to have superior mechanical panels and especially recursive spatial architectures[6]
properties, was obtained by introducing an hard fine etc., it is becoming imperative to design UHPC with much
aggregate of corundum under normal heat treatment. more superior mechanical properties under normal heat
Besides, to further improve the mechanical strengths of treatment.
UHPC samples, a research strategy, including aggregate Therefore, this study aimed to design and produce an ultra-
size, water-binder ratio and fiber dosage, was proposed. high performance concrete with extremely strengthened
The underlying mechanism was investigated by scanning mechanical properties by replacing natural sand with
electron microscope and nitrogen sorption isothermal corundum sand. UHPC mixture was tailoring by achieving
measurement. Results revealed that both the hydration
a densely compacted and strong cementitious matrix. The
reaction and packing degree of binder system are critical
influenced factors, including aggregate size, water-binder
for designing the mix proportion of UHPC. A remarkable
ratio and fiber dosage, on the mechanical properties
enhancement to the mechanical strengths of UHPC
of UHPC were also directly focused. Additionally, the
sample can be observed by replacing natural sand with
underlying mechanism was further evaluated by scanning
corundum sand and this effect can be strengthened by
electron microscope (SEM) and nitrogen sorption
further reducing the corundum sand size. Furthermore,
coupled with the inclusion of steel fibers, an UHPC sample isothermal measurement.
with the compressive strength of higher than 300 MPa can
be easily prepared under normal heat curing. Experimental methodology
Keywords: Ultra-high performance concrete, Corundum Raw materials
sand, Mechanical properties, Mechanism, Microstructures The specific gravity of ordinary Portland cement, ultra-
fine slag and silica fume were 3.15, 1.87 and 2.84 and their
Introduction chemical compositions were given in Table 1. Moreover,
Investigations of UHPC with superior mechanical strength their particle size distributions (PSD) are given in Fig. 1. A
subjected to normal heat treatment only have been a straight brass-coated steel fiber with a tensile strength of
high-priority topic, and the application of steaming plant 2800 MPa was selected as reinforcement, where its length
has been provided the feasibility of curing UHPC in site was 13 mm and aspect ratio was 65. A polycarboxylate-
construction instead of factory fabrication. However, the based superplasticizer was adopted as water-reducer.
reported compressive strength value of UHPC specimens Two types of fine aggregate, natural sand and corundum
incorporating 2% fiber by volume is only located in the sand, were investigated. The natural sand had an apparent
range of 160 to 200 MPa under the curing temperature density of 2.63 g/cm3, whereas the corundum sand had an
in the range of 80oC to 150oC and water to binder ratio apparent density 3.94 g/cm3. The PSD of natural sand was
in the range of 0.13 to 0.20 [1-2]. Although more fibers also given in Fig. 1. For the sake of comparison, the PSD of
(3%-4% by volume) can be incorporated into UHPC corundum sand was fixed to consistent with that of natural
mixture[3-5], the reported maximum compressive strength sand. Morphology of both fine aggregates was presented
value is just improved to 261 MPa under the curing in Fig. 2, where both of them had angular particle shapes.
temperature of 100oC[3]. To improve the resistance of The corundum sand had much more edges and a highly
buildings and infrastructure mentioned above under rough and irregular surface while the natural sand had a
extreme mechanical and environmental loads and to smoother surface.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

370 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preparation of an ultra-high performance concrete with superior mechanical properties

Table 1 Chemical composition of cementitious materials

Al2O3 CaO SiO2 Fe2O3 K 2O MgO Na2O SO3 TiO2 MnO LOI
(%) (%) (%) (%) (%) (%) (%) (%) (%) (%) (%)
Cement 4.94 63.05 19.95 2.92 0.66 1.33 0.15 3.83 0.27 - 2.9
Silica fume 0.16 0.18 97.31 0.15 0.39 0.82 0.2 0.54 - 0.01 0.24
Ultra fine slag 16.8 37.1 31.3 0.43 0.35 9.08 0.38 2.8 0.9 0.34 0.52

maximum particle size and minimum particle size (µm),

respectively, q is the distribution modulus (-), it is fixed at
0.22 in this study.
Subsequently, applying the Least Squares Method to
determine the proportions of each individual
material. The developed UHPC mixture is presented in
Fig. 3 (R2=0.912) and the rounding results (CTRL) was
listed in Table 2, in which the maximum content of cement
was fixed around 750 kg/m3 and the reminder was
accumulated to ultra-fine slag due to the similar PSD.
To balance the effect of filling and pozzolanic activity of
silica fume, its optimal dosage was investigated by further
replacing cement with silica fume (labeled as M-I and
M-II). Based on the optimized formulation, the corundum
sand was introduced into the mix by replacing river sand
with equal volume (labeled as C-I mix) to investigate its
effect on the mechanical strengths of UHPC. Moreover,
Fig.1: Particle size distribution of constituents of UHPC the effect of corundum sand size on the performance of
UHPC was also addressed by minimizing its size to the
range of 0.15~1 mm (labeled as C-II mix).

Fig. 2: Morphology of two aggregates: (a) and (b) natural sand,

Fig. 3: Comparisons of the target curve and the resulting integral
(c) and (d) corundum sand.
grading curve of the mixtures.

Mix design of UHPC Specimens preparation and characterization

The modified Andreasen and Andersen model was The fresh mixes were cast into the molds of 40 • 40 •
applied as optimization algorithms for designing the mix of 160 mm, and the specimens were cured in the molds for
UHPC [7], ran as follows: 1 day at the room condition. Then they were de-molded
and cured in 90 oC steam tank for another 2 days. The
Dq - Dq .........................................................1 air content of fresh mixes was determined by a air
P (D) = D q - Dminq
max min
entrainment meter. Mechanical properties of UHPC were
where P(D) is a fraction of total solids being smaller than determined according to Chinese Standard GB/T 17671-
size D (-), D is particle size (µm), Dmax and are Dmin the 1999 [8]. The micro structure of hardened mortar was

Organised by
India Chapter of American Concrete Institute 371
Session 3 C - Paper 6

Table 2 Mix proportions of UHPC (kg/m3)

Cement Silica fume Ultra-fine slag River sand Water Superplacitizer
CTRL 750 100 150 1200 106 55
M-I 650 200 150 1200 106 55
M-II 550 300 150 1200 106 55

examined by SEM and the pore structures was measured As known, the mechanical strengths of UHPC is resulted
by nitrogen sorption isothermal measurement. from the joint effect of packing density and hydration
degree [7]. Although the packing density of CTRL mix is
Results and discussion highest, its compressive strength is lowest as shown
in Fig. 4, due to the smallest hydration degree. This can
Effect of corundum sand
be verified the results of the porosity of UHPC pastes in
The determined air content of UHPC in the fresh state is Fig. 5. As silica fume dosage increases, the total porosity
presented in Table 3. As shown in Table 3, the air content is decreased from the 1.20 mm3/g to 0.91 mm3/g and the
of UHPC increases with increasing silica fume dosage pores are refined towards smaller size, revealing that the
(CTRL, M-I, M-II), indicating the particle packing becomes mix is gradually densified with the increased hydration
loosen [9]. This phenomenon verifies the validity of mix products. Moreover, the reduced gap of compressive
algorithm. strength of UHPC samples, induced by the decrement
in packing density, would be bridged by the effect of
Table 3 Variation of air content in the fresh UHPC mixtures
increased hydration degree if the replacement of silica
Mixes CTRL M-I M-II C-I C-II
fume is higher 20 wt. %. The trend in compressive strength
Air content 4.7 5.6 5.9 6.7 6.3 of UHPC samples is reversed for flexural strength. This

Fig. 4: Comparisons of mechanical strengths of UHPC samples with different formulation under normal heat curing: (a) Compressive
strength, (b) Flexural strength.

Fig. 5: Relationship between pore size distribution of UHPC samples and silica fume dosage:(a) cumulative pore volume, (b) Pore
size distribution.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

372 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preparation of an ultra-high performance concrete with superior mechanical properties

Fig. 6: Schematic of failure pattern of fine aggregates in UHPC sample: (a) River sand, (b) Corundum sand.

Fig. 7: Morphology of ITZ of UHPC sample with the introduction of corundum sand: (a) low-magnification (b) high-magnification.

could be attributed to the fact that flexural strength is electron microscope images as shown in Fig. 6. The river
more sensitive to features/ defects in the microstructure sand is obviously crushed with the cracks going thought
[10]. The addition of silica fume may result in more micro- the aggregate, and thus resulting in the disappear of
shrinkage cracking. aggregate skeleton. Whereas, the aggregate skeleton still
exists with the introduction of corundum sand and a very
Because of the superior compressive strength, binder
tortuous fracture surface is observed where the cracks
system of M-II mix was selected as optimized
mainly propagate along the aggregate boundaries and
formulation of UHPC sample for the following studies. matrix. Second, the rougher texture and more angular
With the introduction of corundum sand, the air content shape of corundum sand could improve the bonding and
of UHPC is almost 43% higher than that of CTRL mix, interlocking between the aggregates and the cement
as shown in Table 3. This is contributed by the rougher paste [12-13]. This can be verified by the SEM images
texture and more angular shape of corundum sand which observed in Fig. 7, no visible cracks appears between the
subsequently hindered the full compaction of concrete corundum sand and pastes. Especially, this factor affects
[11]. Furthermore, a smaller corundum sand size leads to the flexural strength to a greater degree than compressive
a denser compaction of concrete (C-II mix). However, it still strength [14], where a significant mprovement of 71.6%
higher than that of M-II mix due to the texture features. in flexural strength obtained in Fig. 3. As expected, due
As shown in Fig. 3, the mechanical strength of UHPC to the increased packing density, the C-II mix shows a
samples remarkably improved with the introduction of compressive and flexural strength of 2.9% and 25.7%
corundum. The C-I mix shows a compressive strength higher than those of C-I mix.
and a flexural strength of almost 30.7% and 71.6% higher
Improvement in mechanical strengths
than those of M-II mix. Two factors could contribute to
the strength increase. First, the stiffness and hardness Based on the mix proportion of UHPC samples
(with a Mohs hardness of 9.0) of the corundum sand is incorporating corundum sand, the steel fibers were
dramatically higher than those of the natural sand (with added into the UHPC mixes in the amount of 2%, 3% and
a Mohs hardness of about 7.0). This is agreed with the 4% by the volume of concrete (Labeled as F2, F3, F4),

Organised by
India Chapter of American Concrete Institute 373
Session 3 C - Paper 6

respectively, and the water to binder ratio was attempt fiber, irrespective the fiber dosage. This may be resulted
to reduce to 0.13 (Labeled as W). Fig. 8 shows the effect from the stronger bonding between the matrix and steel
of fiber dosage and water to ratio on the mechanical fiber. Furthermore, it can be seen from Fig. 8 that the
strengths development of UHPC samples. With the flexural strength of C-IIF3 mix and C-IIF4 mix reaches
addition of varying amount of steel fibers, the UHPC to 44.9 MPa and 62.1 MPa and the improvement ratios
sample shows a considerable improvement of 20.0% are 31.5% and 21.7%, respectively. Compared with the
(C-IF2), 44.7% (C-IF3) and 61.3% (C-IF4) in compressive improvement ratio of 25.7% that C-II mix obtained, it is
strength, respectively, compared to that of C-I mix. This found that the increased trend of improvement ratio is not
is attributed to that fact that the steel fibers can bridge true for flexural strength with the addition of steel fibers,
cracks and retard their propagation. Generally, with the where it shows higher scatter. This may be attributed to
increase of steel fiber dosage, the compressive strength that the uniform dispersion of fibers becomes hard to
improvement ratios tendency basically follow a straight reach and a severe overlay of fibers is obtained with the
line as shown in Fig. 9. This is agreed with the investigation addition of a large amount of fiber, resulting in a complex
of Yu et al [7]. In addition, the introduction of fiber is microstructure of UHPC sample.
more effective at enhancing the flexural strength than
Hence, coupled with the introduction of corundum sand
compressive strength, where a significant improvement
and steel fiber, the UHPC sample with compressive
of 42.0%, 82.9% and 172.9% is observed as the steel fiber
strength of higher than 300 MPa is easily accessed under
dosage increases and a parabolic increase tendency
normal heat curing.
of improvement ratios can be observed in Fig. 9. This
phenomenon is also consistent with previous studies
[4,7]. Moreover, by reducing the reducing the water to
binder ratio to 0.13, the C-IW mix, with the introduction
of 3% steel fibers, shows a remarkably enhancement of
68.2% and 176.3% in compressive strength and flexural
strength, and even more strengthened than those of C-IF4
mix. However, coupled with the overlap of steel fibers, the
sample becomes hard to introduce more steel fibers and
shape at this water to binder ratio, although a lower water
to binder ratio can brings in a higher mechanical strengths
of UHPC sample.
As shown in Fig. 8, the compressive strength of C-IIF3
mix and C-IIF4 mix attains to 301.1 MPa and 335.3 MPa,
respectively. It is already known that the mechanical
strengths of UHPC samples will increased with the
reduction of corundum sand size. However, compared
with that of the C-II mix without fiber introduction, the Fig. 9: Correlation between steel fiber dosage and mechanical
strengths improvement ratios of UHPC sample
compressive strength improvement ratio significantly
increased from 2.9% to 12.8% with the addition of steel

Fig 8: Mechanical strengths of UHPC samples incorporating steel fibers with varying amount: (a) Compressive strength, (b) Flexural
strength

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

374 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preparation of an ultra-high performance concrete with superior mechanical properties

Conclusions 2. Ambily PS, Umarani C, Ravisankar K, Prem PR, Bharatkumar BH,

Iyer NR. Studies on ultra high performance concrete incorporating
In this study, the feasibility of preparing an UHPC with copper slag as fine aggregate. Constr Build Mater. 2015; 77: 233-
superior mechanical properties under normal heat 240.
treatment is investigated. Both hydration reaction 3. Yazıcı H, Yardımcı MY, Yig˘iter Y, Aydın S, Türkel S. Mechanical
and packing degree of binder system are critical for properties of reactive powder concrete containing high volumes of
ground granulated blast furnace slag. Cem Concr Compos. 2010;
designing the mix proportion of UHPC. Increasing 32: 639-648.
the silica fume dosage will lows the packing density,
4. Xiao R, Deng ZC, Shen CL. Properties of ultra high performance
however, the reduced gap of compressive strength, concrete containing superfine cement and without silica fume. J
induced by the decrement in the packing density, can be Adv Concr Technol. 2014; 12: 73-81.
bridged by the effect of increased hydration degree if the 5. Yoo DY, Shin HO, Yang JM, Yoon YS. Material and bond properties of
replacement is higher 20 wt.%. Moreover, by replacing ultra high performance fiber reinforced concrete with micro steel
the natural sand with corundum sand, the mechanical fibers. Composites: Part B. 2014; 58: 122-133.
strengths of UHPC sample can be dramatically improved, 6. Lafarge Ductal: Structures, architecture, design. http:// http://www.
especially for flexural strength, due to the couple effect ductal.com. Acquired by 2014.07.10.
of hard stiffness and rougher texture. Also, reducing 7. Yu R, Spiesz P, Brouwers HJH. Mix design and properties assessment
of Ultra-High Performance Fibre Reinforced Concrete (UHPFRC).
the corundum sand size is another beneficial method to
Cem Concr Res. 2014; 56: 29-39.
improve the mechanical strengths of UHPC. Additionally,
8. GB/T17671-1999. Method of testing cements – determination of
with the introduction of steel fiber, an UHPC sample with strength. 1999. (in Chinese)
a compressive strength of higher than 300 MPa can be
9. Wille K, Naaman AE, Parra-Montesinos GJ. Ultra high performance
easily prepared under normal heat treatment, and the concrete with compressive strength exceeding 150 MPa (22 ksi): a
improvement ratio of mechanical strengths induced by simpler way. ACI Mater J. 2011;108(1):46–54.
reducing corundum sand size can be increased, especially 10. Toutanji HA, Bayasi Z. Effect of curing procedures on properties of
for compressive strength. silica fume concrete. Cem Concr Res. 1999;29:497–501.
11. Alengaram UJ, Al Muhit BA, bin Jumaat MZ. Utilization of oil palm
Acknowledgements kernel shell as lightweight aggregate in concrete–a review. Constr
Build Mater. 2013; 38:161-172.
The authors gratefully acknowledge financial support from
12. Donza H, Carera O, Irassar EF. High-strength concrete with different
the Key Project of National Nature Science Foundation of fine aggregate. Cem Concr Res. 2002;32:1755–61.
China (Grant No. 51438003).
13. Huda SB, Alam MS. Mechanical behavior of three generations of
References 100% repeated recycled coarse aggregate concrete. Constr Build
1. Yang SL, Millard SG, Soutsos MN, Barnett SJ, Le TT. Influence of Mater. 2014; 65: 574-582.
aggregate and curing regime on the mechanical properties of ultra- 14. Richardson DN, Whitwell BA. Concrete production plant variables
high performance fibre reinforced concrete (UHPFRC). Constr Build affecting flexural strength relative to compressive strength. ASCE
Mater. 2009; 23: 2291–2298. J Mater Civil Eeg. 2014; 26(8).

Fangyu Han
Affiliations:
1. State Key Laboratory of High Performance Civil Engineering Materials, Jiangsu Research Institute of
Building Science, Nanjing 211103, China
2. Jiangsu Sobute New Materials Co., Ltd., Nanjing 211103, China
Tel.: +86-25-52837006; Cell: +86-15251856320
Address: Liquan Road 118, Jiangning district, Nanjing, Jiangsu Province, China
Fangyu Han was born in 1988. He received his M.Sc. in Civil Engineering Materials from the Southeast
University in Nanjing, China, and continued his studies in State Key Laboratory of High Performance Civil
Engineering Materials of Jiangsu Research Institute of Building Science, Nanjing. Now, he has been involved
in a member of Key Project of National Nature Science Foundation of China and national 973 projects. He
is currently focused on researches and application of high and ultra-high performance concrete, including
their preparation, flowability, pumpability and durability.

Organised by
India Chapter of American Concrete Institute 375
SESSION 4 A
Session 4 A - Paper 1

Sustainability Evaluation of Two Iconic Bridge Corridors under

Construction using Fuzzy Vikor Technique: A Case Study
Shishir Bansal
Research Scholar, Environmental Engg. Deptt., Delhi Technological University (DTU), Delhi-110092
Email: bansal.shishir@gmail.com Mailing Address : A-304, Lake View Apartments; G 17, Sunder Vihar, New Delhi -110087.
Amandeep Singh
M. Tech Student, Environmental Engineering Department., Delhi Technological University (DTU), Delhi-110092
S.K. Singh
Professor and Head, Environmental Engineering Department, Delhi Technological University, (DTU), Delhi-110042

ABSTRACT key transportation system sustainability issues through

construction in the Urban Environment in a Metropolitan
The idea of sustainability has been distinguished as a
city like Delhi. In this research, Sustainability indicators
worldwide need and is most ordinarily characterized as
of the transportation corridor through development in
“Improvement that addresses the issues of the present
an urban domain have been perceived and itemized out.
without trading off the capacity of future eras to address
The research has been made on Signature Bridge (an
their own particular issues”. This idea has infested
asymmetric cable stayed bridge) being constructed
whole ranges of Engineering involving transportation
on River Yamuna by Delhi Tourism and Transportation
frameworks building. Despite the fact that there is no
Development Corporation Limited (DTTDC) and
standard meaning of the sustainable transportation,
Barapulla Elevated Corridor Project (an extradose
sustainability is mostly defined in terms of transportation
bridge over rover Yamuna) being constructed by the
system efficiency and its impact on environmental integrity,
Public Works Department (PWD), Govt. of Delhi.
economic productivity, and social quality of life (Mihyeon
Jeon & Amekudzi 2005). The goals of providing sustainable Amid the research study was made at both the sites in their
features in the design and construction of transportation construction period, and it was found that Sustainability
corridor in an Urban Environment are to minimize impacts of these transportation corridors while the development
on the environmental resources, consumption of material stage is just not restricted to just three Pillars, but rather
resources, energy consumption, encourage the use of in actuality much beyond that. Finally, the real center of
new and innovative approaches, enhance the historic, study lies on showing a correlation between the afore
scenic and aesthetic context and integrate into the mentioned two construction sites by two government
community in a way that helps to preserve and enhance organizations, that is PWD and DTTDC, under the identical
community life, encourage community involvement in the urban environment, by utilizing the Fuzzy rationale
transportation planning process, encourage integration strategy to assess sustainability taking into account the
of non-motorized means of transportation and finally find perceived sustainability pointers utilizing information
a balance between what is important to the community, collected by directing different reviews (survey proforma)
to the natural environment and is economically sound. from the field specialists and the general population
The three parameters i.e. social, economic and (occupants/suburbanites). This research work obtains its
environmental are most commonly accepted as three motivation and guidance from similar project undertaken
pillars of sustainability. In this paper we have compared at other construction sites in Delhi by S K Singh et al.
two iconic road bridge projects of Delhi i.e. Signature “Sustainability Indicators of a Transportation Corridor
bridge and Barapullah elevated road projects, both during Construction in an Urban Environment”.
of which are being constructed over river Yamuna for This study is based on application of fuzzy technique. Fuzzy
sustainability analysis using the Fuzzy Vikor method. It logic is referred to as a way of “reasoning with uncertainty.”
has finally been concluded that construction of Barapulla It gives an all-around characterized system to manage
Elevated road project is better sustainable in the terms of dubious and not completely characterized information, so
various indicators classified during the study. one can make exact findings from uncertain information
The fuzzy theory provides a mechanism for representing
Introduction linguistic constructs such as “many,” “low,” “medium,”
This Research task begins with depicting the eminent “often,” “few.” Notions like rather tall or quick can be
thinking on what constitutes sustainability of the figured numerically and prepared by PCs, with a specific
transportation framework amid development and how end goal to apply a more human-like mindset in the
to perform it. Further the study identifies some of the programming of PCs. As a rule, the fuzzy rationale gives

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

378 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainability Evaluation of Two Iconic Bridge Corridors under Construction using Fuzzy Vikor Technique: A Case Study

a surmising structure that empowers suitable human 7 Both the projects have their major portions constructed
thinking capacities. away from the urban parts of city and there has been
least disturbance to the public. The normal life has not
Selection of Site been hindered in any manner.
Two iconic bridges of Delhi that are Signature Bridge
and Barapulla elevated Corridor have been taken into Methodology Adopted for the Research
consideration for sustainability review. Following procedure has been followed in this research to
identify the sustainability indicators.
SIGNATURE BRIDGE AT WAZIRABAD: Signature bridge
project or Wazirabad bridge project is an upcoming project 1 Selection of a corridor under construction and defining
that is of national as well as international significance. the infrastructure criteria for the corridor.
Wazirabad Bridge Project over River Yamuna consists
of an asymmetric cable stayed bridge having 154 m high 2 Developing sustainability indicator categories
pylon and 251 m clear water way along with Eastern and 3 Identifying sustainability indicators
Western approaches and creation of tourist destination
along the east and west banks of River Yamuna. Later 4 Compiling a proforma that includes sustainability
on it was decided to implement these in two phases viz. indicators and columns for rating
construction of Bridge in Phase-I and creation of a Tourist
Destination in Phase-II. 5 Assigning quantitative as well as qualitative ratings to
the recognized indicators by furnishing ratings from
BARAPULLA ELEVATED ROAD CORRIDOR: Elevated Road the expert’s opinions.
Project over Barapulla Nallah is a corridor connecting
East and South Delhi. The Project has been conceived in First of all preliminary survey of the selected sites was
three phases with nodal locations as Mayur Vihar in East carried out at different times during both day and night.
Delhi and Aurobindo Marg in South Delhi with intermediate Its main purpose was to identify certain issues that hinder
locations as Sarai Kale Khan and Jawahar Lal Nehru the smooth movement of traffic and also those which
Stadium. The main attraction will be Extra-dose Bridge are problematic in execution and protection of ongoing
over river Yamuna having multiple spans with clear water project. The list of 43 such issues was developed and then
way of 128 m in each span. they were classified into six categories and each category
is defined as Sustainability Indicators (Table 1). For an
It was found out that both the projects have striking Urban Environment and developing city like New Delhi,
similarities, which led to formation of common the triple bottom line concept of sustainability does not
ground for unbiased comparison of sustainability. The get fit. It requires extension to accommodate the local
aforementioned similarities are as follows:- conditions. Accordingly the triple bottom line concept is
extended to six broad sustainability indicators. Based
1 Both of the projects are on water bodies
on the classification of these indicators, a questionnaire
2 Both of the projects are iconic bridges and one of their was framed and opinion of experts in this field from CRRI,
kind: Signature Bridge is an asymmetric cable stayed PWD, BRO, Consultants, RITES etc. was obtained and
bridge with main span of 251 m, while the Bridge over with the opinion of experts, rating to these indicators
River Yamuna in Barapulla Phase III is Extra Dose was assigned based on Fuzzy methodology.
bridge with multi spans of 128 m. In both the cases the
deck is supported on Cables. Identification of Sustainability Indicators
Based on Fuzzy theory, the ratings were assigned to
3 Both projects are conceived on new alignments that these 43 indicators, as reflected in Table 1. In later stages
are they are neither the upgradation nor the expansion a survey was conducted in commuters and residents
of existing corridors. nearby to evaluate the measures adopted by client and
the construction agency in the form of questionnaire with
4 Both the projects are carried out in phases where
rating scale of 0 to 9. Where 9 meant best arrangements
partially completed sites have been opened for public
and 0 signifies least arrangements causing maximum
use
inconvenience.
5 Both the projects were constructed in same period i.e.
their construction works begin prior to commonwealth Fuzzy Logic
games of 2010 Linguistic variables and fuzzy set theory
6 Both the projects boast about usage of new and highly In fuzzy set theory, conversion scales are used to transform the
improvised technologies. Segmental constructions qualitative terms into fuzzy numbers. A scale of 0–9 is used to
have been adopted in both the projects. rate the criteria and the alternatives. Following Tables represent

Organised by
India Chapter of American Concrete Institute 379
Session 4 A - Paper 1

Table 1. Identified Sustainability Indicators the conversion schemes for the qualitative, alternative and
criteria ratings.
S.No. SUSTAINABILITY INDICATORS
ENVIRONMENTAL Quantitive Rating Membership Function

1. Air Pollution Very poor (VP) (1,1,3)

2. Existing Drainage system Poor (P) (1,3,5)

3. Noise pollution during day Fair (F) (3,5,7)
4. Noise pollution during night Good (G) (5,7,9)
5. Depletion of Green Belt Very Good (VG) (7,9,9)
6. Plantation scheme Very Low (VL) (1,1,3)
7. Alternate schemes for make the project more sustainable Low (L) (1,3,5)
SOCIAL Medium (M) (3,5,7)
8. Health of workers High (H) (5,7,9)
9. Welfare activities for family of workers Very High (VH) (7,9,9)
10. Sanitation conditions
Fuzzy transformation for qualitative alternative site ratings
11. First Aid facilities
Fuzzy transformation for qualitative criteria ratings
12. Safety measures
13. Increase in stress level of residents/commuters Fuzzy Number
14. Impact on Health of residents/commuters
A fuzzy number is a quantity whose value is ambiguous,
15. Impact on safety of residents/ commuters rather than exact as is the case with “conventional” (single-
16. Preserving the social spaces like cremation ground, Sur Ghat valued) numbers. Any fuzzy number can be assumed as
17. Public attraction with the aesthetics of the Project a function whose domain is a specified set (usually set of
18. Utility of the Project to Public the real numbers) and whose range lies within the span
19. Preserving the heritage structures of non-negative real numbers 0 and 1000 (both included).
ECONOMICS Each numerical value in the domain is allotted a specific
20. Increase in Travel time “grade of membership” where 0 represents the minimum
21. Increase in travel cost possible grade, and 1000 is the maximum possible grade.
22. Disturbance to the business/Employment of nearby residents
Vikor Method
23. Increase in cost of Construction due to lack of funds
24. Increase in cost of Construction due to time overrun In 1998 VIKOR (Vlsekriterijumska Optimizacija I
TECHNICAL Kompromisno Resenje) method was developed by the
25. Display of Project Details Opricovic for the multi-criteria optimization of the complex
26. Traffic Diversions systems. VIKOR method focuses on ranking and sorting
27. Visibility and sight distance to moving traffic
a set of alternatives against various decision criteria
assuming that compromising is only adequate to resolve
28. Lighting of Construction site
conflicts. Alike some other MCDM methods like TOPSIS,
29. Barricading the site
VIKOR depends on an aggregating function that signifies
30. Effectiveness of Technology used
closeness to the ideal, but unlike the TOPSIS, introduces
31. Handling of C & D Waste
the ranking index based on the particular measures of
32. Quality Assurance on the Project
closeness to the ideal solutions and hence this method
GOVERNANCE uses linear normalization for eliminating units of the
33. Ensuring the mobility of Traffic within the project area by criterion functions (Opricovic & Tzeng, 2004).
traffic Marshalls
34. Maintenance of existing drainage system The VIKOR strategy was introduced as one appropriate
35. Maintenance of Barricades method for actualizing within MCDM issue and was
36. Maintenance of existing utilities produced as a multi criteria choice for making a
37. Maintenance of existing greenery procedure to tackle a discrete decision making problem
38. Time over run due to delay in Govt. decisions with non-commensurable and clashing criteria. This
39. Time over run due to mismanagement at site
method focuses on the ranking and selection from a set
of alternatives, and evaluates the compromise solution
INNER ENGINEERING
for a problem within conflicting criteria, which can aid
40. Facilities of Yoga/meditation
the decision makers to reach a final solution. The multi-
41. Performance of Rituals at site like Vishvakarma Puja, May Day
criteria measure for bargain positioning is produced from
42. Celebration during Festivals at site
the LP–metric utilized as a totaling capacity as a part of a
43. Motivation to workers by reward policy or otherwise
trade off programming method.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

380 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainability Evaluation of Two Iconic Bridge Corridors under Construction using Fuzzy Vikor Technique: A Case Study

Assuming that each alternative is evaluated according Step 4: To defuzzify the elements of the fuzzy decision
to each criterion function, comparing the measure of matrix corresponding to the alternatives and the
closeness to the ideal alternative could perform the criteria weights into crisp values.
compromise ranking. The various m alternatives are
For example a fuzzy number a~= (a1, a2, a3) can be
denoted as A1,A 2,……Am. For alternative Ai, the rating of the
converted into a crisp number a by employing the below
jth aspect is denoted by fij (i= 1,2,…. m; j=1,2,… n), i.e., fij is
equation:
the value of jth criterion function for the alternative Ai, n is
the number of criteria. Development of the VIKOR method a=(a1+4a2+a3)/6 ...........................................................(4)
starts with the following form of LP-metric:
Step 5: To Determine the best and worst values of
Lp, i = G | !wj Q f *j - fij V / Q f *j - f -j V$ J , 1 # p # 3
n 1/p
p
critera rating where fj* is best and values fj- is worst
j=1
value
In the VIKOR method L1,i (as Si) and L∞;i (as Ri) are used f *j = max i E x ij H
.........................................................................(5)
f -j = min i E x ij H
to formulate ranking measure. The solution obtained by
min Si is with a maximum group utility (‘‘majority’’ rule),
and the solution obtained by min Ri is with a minimum z
individual regret of the opponent. Step 6: To compute the values of Si and Ri using the
equations given below
The compromise-ranking algorithm of the VIKOR method
f *j - x ij
has the following steps: Si = | j = 1 w j .....................................................(6)
n

f *j - f -j
Step 1: To Assign ratings to various alternatives and
f *j - x ij .....................................................(7)
criteria Ri = max j w j
f *j - f -j
Let us take a set of m alternatives called A = {A1, A2,., A m }
which we need to evaluate against a set of n criteria, that Step 7: To compute the values of Qi using
is C = {C1 ,C2 ,., Cn }. The criteria weights are represented Si - S* Ri - R- ........................................(8)
by wj where (j=1,2,..,n). The performance ratings of the Qi = v - * + (1 - v)
S -S R- - R*
decision maker Dk (k = 1,2,…, K) for each alternative Ai
(i=1,2,..,m) according to criteria Cj (j= 1,2,..,n) are denoted Where:
by : S* = minimum Si
Rk = xijk =(aijk, bijk, cijk), where j = 1, 2,…, n; i= 1,…., m; S- = maximum Si
k =1, 2 ,.., K with membership function μRk (x).
R* = minimum Ri
Step 2: To compute the aggregate fuzzy ratings R- = maximum Ri
corresponding to alternatives and criteria.
And v is the weight for the strategy of maximum group
When fuzzy ratings for all the decision makers are utility and here it is taken to be 0.5
described as the triangular fuzzy number Rk=(ak, bk, ck),
where k=1,2,...,K, then the aggregated fuzzy rating is Step 8: To rank the alternatives by sorting the values
defined by R=(a, b, c), k=1,2,...,K where; Q, R and S in ascending order.
a = min E a k H, b = K bk, c = max E c k H
1 | k
...................(1)
k=1 Step 9: To propose a compromise solution for the
alternative (A (1)) which is the best ranked by the measure
The aggregated fuzzy weights (wij) corresponding to each Q(minimum) if the following two conditions are satisfied
criterion are calculated as wj = (wj1; wj2; wj3) where
C1: Acceptable advantage
w j1 = min E w jk1 H, wj2 = K wjk2, w j3 = max E c jk3 H ....(2)
1 | k
k=1
If Q (A (2)) – Q (A (1)) ≥ DQ .................................................(9)
Step 3: To compute the fuzzy decision matrix Where A (2) is the alternative that holds second position in
the ranking list according to Q and
The fuzzy decision matrix for the criteria (W) and the alternatives
(D) is constructed as follows: DQ = 1/J-1 where j is number of criteria
C1 C2 ..... Cn C2: Acceptable stability in decision making .................(10)
A1 RSS X11 X12 ..... X1n VWW
A2 SS X21 X22 ..... X2n WWW , i= 1, 2,….m ; j= 1,2,.n .... The alternative A(1) should also be the best ranked by R
D = SS W (3) or/and RS. The settlement solution is stable only within
A3 SS ... ... ..... ... WW
SS WW a specific decision making process, and that could be the
A4 Xm1 Xm2 ..... Xmn
T X strategy of maximum group utility (when v>0.5 is needed),
W = (w1, w2… wn) or --by consensus when v = 0.5, or -with veto i.e. (v<0.5).

Organised by
India Chapter of American Concrete Institute 381
Session 4 A - Paper 1

Table 2 : Qualitative Assessments and Aggregate fuzzy criteria ratings

Criteria Qualitative rating Aggregate Fuzzy Rating Crisp Value
E1 E2 E3 E4 E5 E6 E7 E8 E9 E10
OPS LP SKR PK HKS VKS VSK PKS S sri SS
C1 VH VH VH H H VH VH VH VH (3,8.2,9) 7.466667
C2 H VH M VH M M H VH VH H (3,7.2,9) 6.8
C3 M H H H L H M M M M (1,5.6,9) 5.4
C4 H VH VH VH VH H H H H VH (5,8,9) 7.666667
C5 VH VH M H H VH VH H H H (3,7.6,9) 7.066667
C6 VH VH VH H H M H M H H (3,7.2,9) 6.8
C7 H M VH H VH H H (3,6.6,9) 6.4
C8 VH VH H VH H VH H H H VH (5,8,9) 7.666667
C9 VH H L VH H H H M H H (1,6.8,9) 6.2
C10 VH VH H VH H VH H H H H (5,7.8,9) 7.533333
C11 VH VH VH VH H VH VH VH H VH (5,8.6,9) 8.066667
C12 VH VH VH VH VH VH VH VH VH VH (7,9,9) 8.666667
C13 H VH VL VH M VH VH H M VH (1,7,9) 6.333333
C14 VH VH VH VH M VH VH H H H (3,8,9) 7.333333
C15 VH VH VH VH H VH VH VH VH H (5,8.6,9) 8.066667
C16 H VH M VH H VH H (3,6.8,9) 6.533333
C17 M H L H H VH M M H M (1,6,9) 5.666667
C18 VH VH M H M H VH VH H VH (3,7.6,9) 7.066667
C19 VH M M M VH VH H (3,6.4,9) 6.266667
C20 VH VH VH VH M VH H H VH H (3,8,9) 7.333333
C21 VH VH VH VH M M H H VH H (3,7.6,9) 7.066667
C22 H H VH VH L M H H H M (1,6.6,9) 6.066667
C23 H H H VH VH VH H H VH VH (5,8,9) 7.666667
C24 H H H VH VH VH H H VH VH (5,8,9) 7.666667
C25 H H M H L L H VH L (1,5.6,9) 5.4
C26 VH VH VH VH VH VH VH VH VH H (5,8.8,9) 8.2
C27 VH VH H VH M VH VH VH H H (3,8,9) 7.333333
C28 VH VH H VH VH VH VH VH VH H (5,8.6,9) 8.066667
C29 VH VH H VH H VH VH H VH VH (5,8.4,9) 7.933333
C30 VH H H H M M M H VH H (3,6.8,9) 6.533333
C31 H H M VH H VH VH H H H (3,7.4,9) 6.933333
C32 VH H VH VH VH VH VH VH H H (5,8.4,9) 7.933333
C33 VH VH VH VH VL VH H VH VH VH (1,8,9) 7
C34 VH VH H VH H VH H VH VH VH (5,8.4,9) 7.933333
C35 H M H VH H VH H VH H M (3,7.2,9) 6.8
C36 VH H VH VH VH VH H VH H H (5,8.2,9) 7.8
C37 VH VH M VH H H H H H VH (3,7.6,9) 7.066667
C38 H H VH VH VH H H M VH VH (3,7.8,9) 7.2
C39 H H VH VH M M VH VH VH H (3,7.6,9) 7.066667
C40 M M M H VL M VL L VL L (1,3.6,9) 4.066667
C41 VL L H VH VH M VL M H L (1,5,9) 5
C42 M VL VH H VL M VL M M VL (1,4,9) 4.333333
C43 VH VH H H VH VH H H VH VH (5,8.2,9) 7.8

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

382 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainability Evaluation of Two Iconic Bridge Corridors under Construction using Fuzzy Vikor Technique: A Case Study

If one of the above conditions is not satisfied, then a set of Table 3:

settlement solutions is proposed, which consists of: The best values fj* and the worst values fj- of the 43 criteria
CRITERIA CRISP RATING WORST BEST
ll Alternatives A(1) and A(2) if only the condition C2 is not VALUE VALUE
satisfied Or A1 (PWD) A2 (DTTDC)
Fj - Fj *

ll Alternatives A(1), A(2),…. A(M) if the condition C1 is not C1 6.146667 6.013333 6.013333 6.146667
satisfied; A(M) is determined by the relation Q(A(M))
C2 6.173333 6.146667 6.146667 6.173333
- Q(A(1)) < DQ for maximum M (the position of these
alternatives are in closeness). C3 6.28 6.173333 6.173333 6.28
C4 6.32 6.186667 6.186667 6.32
Numerical Application of Fuzzy Logic C5 6.826667 6.533333 6.533333 6.826667
In this section sustainability evaluation of the two C6 4.44 4.573333 4.44 4.573333
transportation corridors namely A1 and A2, in Delhi, under C7 6.746667 6.586667 6.586667 6.746667
construction have been carried out using the Fuzzy VIKOR C8 6.146667 6.013333 6.013333 6.146667
technique and is presented in this. These project sites are C9 4.333333 4.52 4.333333 4.52
Barapulla Elevated Corridor (A1) constructed by PWD and
C10 4.36 4.573333 4.36 4.573333
Signature Bridge (A2) constructed by DTTDC.
C11 6.853333 6.533333 6.533333 6.853333
A committee of 10 experts (E1, E2… E10) was formed to C12 6.146667 6.013333 6.013333 6.146667
obtain the qualitative ratings for the criteria and the
C13 6.093333 6.12 6.093333 6.12
alternatives (refer Table 2).
C14 6.64 6.586667 6.586667 6.64
We convert the qualitative ratings into fuzzy triangular C15 6.8 6.506667 6.506667 6.8
numbers and then we generate aggregate ratings using
C16 6.64 6.533333 6.533333 6.64
the equation (2).The Table 3 presents the aggregate fuzzy
decision matrix for the both the alternative sites. C17 6.146667 6.013333 6.013333 6.146667
C18 6.853333 6.506667 6.506667 6.853333
Generate aggregate crisp ratings for both the alternative
C19 6.853333 6.533333 6.533333 6.853333
sites using equation (4). Based on these values, we will
calculate the best fj* and the worst fj- values of all 43 C20 6.933333 6.613333 6.613333 6.933333
criteria using equation (5) C21 6.853333 6.533333 6.533333 6.853333
C22 6.826667 6.533333 6.533333 6.826667
Following table 4 presents the values of Si, Ri and Qi for
the two alternatives calculated using equation (6,7,8). C23 6.853333 6.56 6.56 6.853333
The values of S*= 0.736, S- = 5.76, R*= 0.163, R- =0.188 are C24 6.746667 6.506667 6.506667 6.746667
computed using equation (9). C25 6.826667 6.533333 6.533333 6.826667
C26 6.826667 6.586667 6.586667 6.826667
Table 4 values of Si, Ri, Qi
C27 6.826667 6.506667 6.506667 6.826667
A1 (PWD) A2 (DTTDC) C28 6.853333 6.586667 6.586667 6.853333
Si 0.7359 5.7531 C29 6.826667 6.533333 6.533333 6.826667
Ri 0.1634 0.188 C30 6.88 6.56 6.56 6.88
Qi 0 1 C31 6.853333 6.533333 6.533333 6.853333
C32 6.64 6.533333 6.533333 6.64
Table 5 ranks the two alternatives, by sorting the values of
C33 6.64 6.56 6.56 6.64
Si, Ri and Qi obtained from Table 4 in the ascending order
C34 6.586667 6.533333 6.533333 6.586667
Table 5 Ranking the alternatives C35 6.826667 6.56 6.56 6.826667
Si A1 A2 C36 6.613333 6.56 6.56 6.613333

Ri A1 A2 C37 5 5.72 5 5.72

C38 6.853333 6.533333 6.533333 6.853333
Qi A1 A2
C39 6.586667 6.533333 6.533333 6.586667
It can be seen from the above results as presented in C40 6.64 6.56 6.56 6.64
Table 5 that site 1 that is Barapulla Elevated Corridor by C41 6.64 6.56 6.56 6.64
the PWD is the best ranked by the measure of least value
C42 6.44 6.146667 6.146667 6.44
of Qi. Therefore we now cross-examine it for the given two
conditions those have been earlier discussed. C43 6.64 6.506667 6.506667 6.64

Organised by
India Chapter of American Concrete Institute 383
Session 4 A - Paper 1

1). C1: acceptable advantage i.e. equation 9 Sustainability Evaluation

Using equation 9 DQ = 1/43-1 = 1/42 = 0.0238. The Fuzzy VIKOR technique was applied for sustainability
evaluation of two major transportation corridors under
Now to satisfy the condition Q (A(2)) – Q(A(1)) ≥ DQ ,where A(1)) is
construction i.e. (A1, A2) in New Delhi city. These projects
the best ranked by the measure Q(minimum) and in our case it is A1
were Barapullah Elevated Corridor being constructed
We have by PWD (A1) and Signature Bridge being constructed by
DTTDC (A2).
Q(A2) - Q(A1) = 1 - 0= 1 > 0.0238, hence the condition
QA(1) – QA(2)≥DQ is satisfied. The Final outcomes after the numerical application of
Fuzzy VIKOR method exhibit that the site A1, i.e. Barapullah
2). C2: Acceptable stability in decision-making using Elevated Corridor being constructed by PWD is found to
equation 10 be more sustainable under the given conditions and the
Since site A1 is best ranked by Si and Ri (considering the identified sustainability indicators..
-”by consensus rule v =0.5”), therefore it is declared to be
as a more sustainable corridor. Conclusions
Following conclusions are drawn from the above study:
Results & Discussions 1 Through this research study it has been furnished that
Results of this study has been illustrated in Table 5, sustainability is not only based on three parameters
which depicts that alternative A1 i.e. Barapullah Elevated but also depend on various other indicators that has
Corridor by PWD is a more sustainable corridor, in light been identified as per study.
of the recognized sustainability indicators, among the two 2 Various Sustainability Indicators through the
corridors chosen for the case study. construction stage has been identified for an elevated
transportation corridor and hence are classified under
Discussion various categories as covered in this research.
Identifying Sustainability Indicators 3 The three pillars of sustainability namely social,
economic and environmental are viable only for
The five-step methodology defined in this research can be
developed countries whereas in developing economies
used for any transport corridor to develop sustainability
like India where various other factors such as
indicators. The five steps are
exponential increase in population etc. come into play,
1 Selection of a corridor under construction and defining the need to introduce additional parameters arises.
the infrastructure criteria for the corridor 4 The comparative study of 2 iconic transportation
2 Developing the sustainability indicator categories corridors through construction, Barapullah Elevated
Corridor being constructed by PWD (A1) and
3 Identifying the sustainability indicators Signature Bridge being constructed by DTTDC (A2)
4 Compilation of a proforma that include sustainability has defined a methodology for future sustainability
indicators and corresponding columns for rating studies
5 Assigning the quantitative as well as qualitative ratings 5 The results of this study yield that the construction of
to the recognized indicators by furnishing the ratings Barapullah Elevated corridor is more sustainable as
from the field expert’s opinions compared to the Signature Bridge.

Each of these steps can be applied to evaluate a sustainable Scope of Future Work
transportation corridor through construction in an urban In this research, the study has been limited to only
environment. This process began with the requisite developing the indicators and demonstrating application
for categorization of the sustainability from its existing of FUZZY technique for sustainability evaluation of
three pillars i.e. Economic, Social and Environmental transportation corridors. In later stages we wish to develop
aspects and excelled with the development of three more a green rating system for transportation corridors, similar
vital categories namely Inner Engineering, Technical an to those for the green buildings. Moreover, This research
Governance. In later stages the individual parameters/ will serve as a platform or guide for the implementation
indicators under these 6 sustainability categories were of most suitable sustainability indicators through
recognized by visiting the corridors through construction construction of a transportation infrastructure.
and consultation with the field experts. Finally, the References
process completed with the compilation of a proforma
1. Bansal Shishir, Verma Sameer, Singh Santosh K (2014),
that furnishes Qualitative as well as Quantitative ratings “Sustainability Indicators of a Transportation Corridor during
to each identified sustainability indicator from the experts. Construction in an Urban Environment”, ICSCI 2014, Hyderabad

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

384 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainability Evaluation of Two Iconic Bridge Corridors under Construction using Fuzzy Vikor Technique: A Case Study

2. Anjali AWASTHI, Hichem OMRANI, Philippe GERBER2 CIISE, Conference of Indian Institute of Bridge Engineers Held In Mumbai
(January 2013), “Multi-criteria decision making for sustainability On 24th November 2012
evaluation of urban mobility projects”, CEPS Instead, Working Paper
19. Shishir Bansal, Jose Kurian and Santosh K. Singh, Assessment of
No 2013-01
Degradation Caused to the Environment During Constructions And
3. Awasthi A, Chauhan SS, Omrani H, Application of fuzzy TOPSIS in The Extent Of Its Replenishment, First International Conference
evaluating sustainable transportation systems”, Expert Systems On Concrete Sustainability, Tokyo, Japan, Being Organised By Japan
with Applications 38 (2011) 12270–12280 Concrete Institute , May 27-29, 2013
4. CIRIA. (2008). Sustainability. Retrieved March 8, 2008, from http:// 20. Jeon, C. M. and Amekudzi, A. (2005) “Addressing Sustainability
www.ciria.org/ complianceplus/images/sustainability2.gif. in Transportation Systems: Definitions, Indicators, and Metrics”
Journal of Infrastructure Systems, American Society of Civil
5. Sunita Bansal, Srijit Biswas, SK Singh, (2015), “Selection of most
Engineers (ASCE). Volume 11, Number 1, pp 31-50.
economical green building out of n-alternatives: approach of vague
fuzzy logic”, International Journal of Research in Engineering and 21. Jeon Christy Mihyeon,(2007), “Incorporating Sustainability
Technology, 4(2): 164-168. Into Transportation Planning And Decision Making: Definitions,
Performance Measures, And Evaluation” Ph.D. Thesis in the
6. Shishir Bansal, S.K. Singh(2015),“Sustainable handling of
School of Civil and Environmental Engineering, Georgia Institute
construction and demolition (c & d) waste”, International journal
of Technology (2007).
of sustainable Energy and Environmental Research, 4(2): 22-48
22. Litman T. (2009), A Good Example of Bad Transportation
7. Sunita Bansal, FIE Srijit Biswas, SK Singh, (2015), “Fuzzy
Performance evaluation, Working paper, Victoria Transport Policy
Modelling For Selection Of Most Economical Green Building Out
Institute.
Of N-Alternatives”, International Journal of Advanced Information
Science and Technology (IJAIST), Volume 36, Issue 3: 7-11 23. Litman, T.,(2009), “Sustainable transportation indicators- A
recommended research program for developing sustainable
8. Shishir Bansal, S.K. Singh, (2014), “A Sustainable Approach Towards
transportation indicators and data”. In Proceedings of the
The Construction and Demolition Waste”, International Journal of
2009 transportation research board annual conference, CD-
Innovative Research in Science, Engineering and Technology, Vol.
ROM,Washington, DC, January 11-15
3, Issue 2, February, 2014.
24. Litman,Todd. (2003), “Sustainable Transportation Indicators.”
9. Shishir Bansal, S.K. Singh, (2014), “An Elevated Road Over Barapulla
Victoria Transport Policy Institute (VTPI), Victoria, Canada. (http://
Nallah In New Delhi”, Proc., FIB 2014, Mumbai, India.
www.vtpi.org/sus-indx.pdf)
10. Shishir Bansal , Vinay Gupta, And S K Singh (2012) , A 3-Level
25. Litman, T., and Burwell, D. (2006), “Issues in Sustainable
Grade- Separator At Ghazipur In East Delhi And Sustainability
Transportation”, Journal of Global Environmental Issues, 6(4),
Considerations During The Construction, , Proceeding of
pp331-346.
Conference of Indian Institute of Bridge Engineers Held In Mumbai
On 24th November 2012 26. Litman, T.,(2008), “Well Measured: Developing Indicators for
Comprehensive and Sustainable Transport Planning”. British
11. Shishir Bansal, Jose Kurian and Santosh K. Singh, Assessment of
Columbia: Victoria Transport Policy Institute.
Degradation Caused to the Environment During Constructions And
The Extent Of Its Replenishment, First International Conference 27. Opricovic, G.H. Tzeng, Compromise solution by MCDM methods: A
On Concrete Sustainability, Tokyo, Japan, Being Organised By Japan comparative analysis of VIKOR and TOPSIS, European Journal of
Concrete Institute , May 27-29, 2013 Operational Research, 156 (2004), pp. 445–455
12. Shishir Bansal ,Jose Kurian ,and Santosh K. Singh, “Application 28. Opricovic, G.-H. Tzeng, Multicriteria planning of post-earthquake
Of Environmental Friendly Systems To Protect The Environment sustainable reconstruction, Computer-Aided Civil and Infrastructure
During Construction Of Grade Separators In New Delhi” ¸ first Engineering,, 17 (3) (2002), pp. 211–220
international conference on concrete sustainability, Tokyo, Japan,
29. World Commission on Environment and Development (WCED),
Being Organised By Japan Concrete Institute , May 27-29, 2013
(1987) “Our Common Future”. Oxford University Press, Oxford,
13. Sunita Bansal, FIE Srijit Biswas, SK Singh, (2015), “Approach of fuzzy England
logic for evaluation of green building rating system”, International
30. Ott, W,(1978), “Environmental Indices: Theory and Practice”. Ann
Journal of Innovative Research in Advanced Engineering (IJIRAE),
Arbor: Ann Arbor Science.
Volume 2, Issue 3: 35-39.
31. Samberg Stuart, RK&K Engineers, BassokAlon, (2011) Puget Sound
14. Shishir Bansal, S.K. Singh(2015),“Sustainable handling of
Regional Council and Holman Shawna, Parametrix, “Method for
construction and demolition (c & d) waste”, International journal
Evaluation of Sustainable Transportation toward a Comprehensive
of sustainable Energy and Environmental Research, 4(2): 22-48
Approach”
15. Sunita Bansal, FIE Srijit Biswas, SK Singh, (2015), “Fuzzy
32. Bansal Shishir, Singh S K, Kurian Jose, (2014), “Environmental
Modelling For Selection Of Most Economical Green Building Out
Impacts And Mitigation Measures In Infrastructure Projects In New
Of N-Alternatives”, International Journal of Advanced Information
Delhi”, 1st First International Conference on Concrete Sustainability,
Science and Technology (IJAIST), Volume 36, Issue 3: 7-11
27-29 May 2013, Tokyo Japan, Program & Paper abstracts, pp 177
16. Shishir Bansal, S.K. Singh, (2014), “A Sustainable Approach Towards
33. Bansal Shishir, Singh S K (2014), “Sustainable Construction Of
The Construction and Demolition Waste”, International Journal of
Grade Separators At MukarbaChowk And Elevated Road Corridor
Innovative Research in Science, Engineering and Technology, Vol.
At Barapulla, Delhi”, International conference of Advance Research
3, Issue 2, February, 2014.
and Innovations 2014, New Delhi, India, Feb 1, 2014
17. Shishir Bansal, S.K. Singh, (2014), “An Elevated Road Over Barapulla
34. Steg Linda, (2005) Department of Psychology, University of
Nallah In New Delhi”, Proc., FIB 2014, Mumbai, India.
Groningen, Grote Kruisstraat 2/I, 9712 TS Groningen, The
18. Shishir Bansal , Vinay Gupta, And S K Singh (2012) , A 3-Level Netherlands, Gifford Robert, Department of Psychology, University of
Grade- Separator At Ghazipur In East Delhi And Sustainability Victoria, Victoria BC V8W 3P5, Canada, “Sustainable transportation
Considerations During The Construction, , Proceeding of and quality of life”.

Organised by
India Chapter of American Concrete Institute 385
Session 4 A - Paper 1

Shishir Bansal
Shishir Bansal obtained the degree of Bachelor in Civil Engineering from Punjab Engineering College in
1985 and Master in Highways Engineering from same institution in 1987. Thereafter during the service, he
did LL.B from Delhi University in 1999.
He started his service carrier in teaching line as faculty of Punjab Engineering College from 1987 to 1990.
Thereafter qualifying 1988 Examination of Engineering Services, he joined CPWD in 1990 as Assistant
Executive Engineer in Kolkata where he was instrumental in designing several buildings.
In 1994, on promotion to Executive Engineer, he came to Delhi and was responsible for construction
of Police stations, Police lines, Hospital complexes, School Building and Industrial Training Institute.
In addition to buildings he supervised construction of many infrastructure projects like Clover Leaves
with voided slabs, Flyovers with Precast segmental construction, and Underpass with soil Anchors
Technology.
In 2006, he was promoted to Superintending Engineer and was responsible for construction of various
official and residential buildings of Paramilitary forces land Central Govt. offices and residential
complexes. Thereafter, he was posted as Project Manager in PWD Delhi for the famous Barapullah
Elevated Road Project
Presently he is the Chief Project Manager, Signature Bridge Project being executed by DTTDC.
He is Member Secretary, IRC G-2 Committee (Human Resource Development), Member, Technical
assessment committee of BMTPC. Member B-4 and B-7 Committee, IRC,
He has written about 20 papers which have been presented in various prestigious conferences in India
and abroad like FIB conference 2006 at Italy, FIB Conference 2010 at USA, FIB Conference 2014 at
Mumbai, First International Conference on sustainability of Concrete at Japan in 2013, International
Conference on Sustainable Civil Infrastructures at Hyderabad in 2014.
He is Fellow member, Institution of Engineers Fellow member, Indian Concrete Institute, Member, Indian
Roads Congress Member, Indian Buildings Congress Member, Indian Council of Arbitration and member
American Concrete Institute.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

386 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Design and Construction of Wirewound Circular Precast, Prestressed Concrete Tanks with Sliding Base

Design and Construction of Wirewound Circular Precast, Prestressed

Concrete Tanks with Sliding Base
Sanjay Mehta
Chief Technical Officer and Vice President, Preload Inc., 49 Wireless Blvd, Suite 200, Hauppauge, NY 11788, USA

Abstract: results in vertical bending of the tank wall. Flexural

Sliding base circular precast prestressed concrete tensile stress well in excess of modulus of rupture
tanks, although common in the North American of concrete can develop due to this base restraint,
Continent, have not found widespread applications in especially when shrinkage and thermal stresses are
the Indian subcontinent. Such tanks, continuously and added to flexural stresses.
spirally wrapped with high tension wires, offer superior 3. In case of a seismic event, the load path from tank
liquid tightness and service life as compared with the wall, subjected to liquid convection and impulse, goes
reinforced concrete tanks. Once hydro tested after initial through the wall-footing joint. Although this joint can
construction, such tanks can be virtually maintenance be designed and reinforced to provide the required
free. In context of Indian design and construction practice, ductility, it is not possible to control cracking in case
replacing a system of Underground Sump Reservoir of an extreme seismic event. The liquid tightness of
(USR) and Elevated Storage Reservoir (ESR) with a single, the tank will be compromised during such a seismic
tall prestressed concrete tank can result in significant event, precisely the time when water requirement is
savings in pumping electricity cost in addition to savings very critical to control post-earthquake fires.
in land and excavation cost, so long as distribution system
can be adequately designed. Ground supported tanks Figure 1 illustrates various stages of loading and stresses
as large as 133m in diameter and 30m in height have in a fixed base reinforced concrete tank. Figure 2 shows
been successfully designed and constructed across the similar stages for a circular prestressed concrete tank
United States. This article is written to explain design and with a sliding base. The problems associated with fixed
construction technology of such tanks. base reinforced concrete tanks are alleviated in a sliding
base, prestressed concrete tank explained as follows:
Keywords: precast, prestressed, wirewinding, circular,
tanks, domes, shells, seismic, water 1. The circular tank wall is prestressed in excess of the
hoop tension caused by liquid load. Thus, even under
full hydrostatic and hydrodynamic pressures, concrete
Introduction remains in compression.
Circular reinforced concrete tank walls are designed and
constructed to keep concrete hoop stress within modulus 2. Elastomeric bearing pads separate base of the tank
of rupture when subjected to liquid load. Amount and wall from the ring footing. When subjected to prestress
spacing of reinforcement is designed to control cracking loading, the tank wall moves in with minimum restraint
within the tank wall as per the applicable code provisions. allowing compression to develop without significant
Typically, base of the tank wall is either fixed or hinged vertical bending. Similarly, the tank wall moves out
to the footing. This traditional approach to reinforced with minimum restraint when subjected to liquid
concrete tank design has three major problems: pressure. However, total outward radial movement
is always less than total inward radial movement,
1. Concrete is in tension in circumferential direction ensuring residual compression in the tank wall under
when subjected to liquid load. No amount of mild steel full hydrostatic loads. The liquid tightness of the wall-
reinforcement can prevent concrete from cracking or footing joint is achieved by use of a specially designed
micro- cracking. The problem is exacerbated when PVC waterstop. This waterstop, about 275mm long,
shrinkage and thermal stresses are added to tensile has ability to move radially by about 75mm without
hoop stresses. At best, it is possible to design for crack compromising liquid tightness of the tank.
control as exemplified in IS 33701, Euro Code2, or ACI
3503. 3. Typically, seismic restraint cables are placed on the
outside of the precast core walls. The cables are
2. The fixed or hinged base prevents lateral movement
activated only when tank wall begins to slide either
of the wall base when subjected to the liquid load. This
during earthquake or wind or any such lateral loads.

Organised by
India Chapter of American Concrete Institute 387
Session 4 A - Paper 2

Fig. 1: Loads and Moments in Conventional reinforced Concrete Tanks

Fig. 2: Loads and Moments in Circular Wirewound‐Prestressed Concrete Tanks

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

388 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Design and Construction of Wirewound Circular Precast, Prestressed Concrete Tanks with Sliding Base

Under normal operating conditions, the restraint The ring footing is connected to the membrane floor
cables remain inactive and almost unstressed. Thus, using haunch and reinforcement as shown in Figure
the lateral load resistance is separated from the 4. The membrane floor acts as a “tie” which prevents
vertical and hydrostatic load resistance, providing footing rotation due to eccentricity of the vertical load.
superior seismic performance. The design approach requires checking tie stress at floor-
footing haunch to ensure that the tensile tie stress is well
Benefits of Precast Technology: within the modulus of rupture of concrete to ensure liquid
tightness. The width of the footing is selected such that
It is well known that better concrete quality control can
stresses from vertical loads are within the allowable soil
be exercised when concrete is poured under controlled
bearing capacity.
condition. This is particularly true in case of tank walls
because poor quality of concrete is the primary reason As far as possible, it is preferable to cast floor and footing
for leakage from liquid containing concrete structures. in one pour to avoid construction joint. Construction
Furthermore, if it were possible to cast circular walls joints in liquid containment structure are a maintenance
in sections, shrinkage stresses could be drastically problem. Special care is required to vibrate concrete
reduced. Small sections of walls are easier to pour and around horizontal water-stop at the construction joint
finish, especially when concrete pour is in horizontal to prevent “honeycombing”. Membrane floors as large
direction. Since wall panels are cast in horizontal position, as 65m have been cast in one pour. Figures 5 through 8
they have to be tilted up and placed in position on the show various stages of membrane floor and ring footing
circular ring footing. The tilt-up operation generates high construction.
flexural stresses in the wall panel. Thus, the strength and
quality of concrete is tested even before the tank is placed Wall Base Joint
in service, due this tilt-up operation. In contrast to cast-
As mentioned earlier, unlike reinforced concrete tanks,
in- place walls, the precast wall casting operation can
circular prestressed concrete tanks are designed to move
start simultaneously with the floor pour. This significantly
in and out in the radial direction. This is achieved by setting
reduces the construction time of the entire project since
elastomeric bearing pads between top of the footing and
wall concreting is no longer along the critical path.
wall base. Bearing pads allow radial tank movement
with minimum resistance. The design approach requires
Discussion of Critical Design and calculation of bearing pad resistance and vertical moment
Construction Concepts: in the wall due to this resistance. The hoop forces in the
Figure 3 shows a typical section of wirewound precast, tank wall are computed assuming no reduction in stresses
prestressed concrete tank. Accordingly, there are five critical due to pad resistance. As can be seen in Figures 3 and 4,
elements in the design and construction (1) Floor and Footing liquid tightness at the wall base joint is achieved by casting
(2) Wall Base Joint (3) Precast Wall Panels and Joints (4) Wall
Prestressing and Shotcreting and (5) Dome Construction
and Prestressing. Each element is discussed as follows:

Floor and Footing

Typically, 300mm deep footing is cast monolithically with
100m membrane slab. Figure 4 shows cross sectional
details of footing-floor connection. As per ACI 350-063,
Appendix G, floors upto 150mm in thickness are classified
as “membrane” floors. Membrane floors are designed
with a single layer of reinforcement in each direction for
creep and shrinkage of concrete. They have very little
flexural capacity and liquid load is transferred “directly”
to the supporting soil. That is, the slab is flexible enough
to conform to the deformed shape of supporting soil. The
key to the successful performance of a membrane floor
lies in the uniformity and strength of subgrade. So long as
supporting soil is compacted properly to avoid localized
settlement, the membrane slab will deform in “dish-type”
settlement shape. ACI 372R -035, Appendix A, Section
A.3.2 describes deflection limits on various deformation
modes of membrane slab-on-grade, along with design
Fig. 3: Typical Cross Section of Wirewound‐Prestressed Con-
considerations. crete Tank

Organised by
India Chapter of American Concrete Institute 389
Session 4 A - Paper 2

Fig. 4: Wall‐Floor‐Footing Details

Fig. 5: Floor‐Footing Reinforcement Fig. 6: Floor Concrete Pour

a special waterstop into the footing on the inside face of The design approach requires computation of wall base
the wall. After wall panels are erected, the waterstop is joint displacement when subjected to combined effect
encased in a concrete curb, as shown in Figures 9 and 10. of circumferential prestressing and liquid load. The wall
base movement is restricted to about 40mm to ensure

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

390 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Design and Construction of Wirewound Circular Precast, Prestressed Concrete Tanks with Sliding Base

Fig. 7: Screeding and Finishing Operation

Fig. 10: Concrete Encasement for Waterstop

sufficient factor of safety on the displacement capacity

of waterstop. In some situations, staged prestressing
may be required. That is, curb may have to be cast after
prestressing the tank to some extent. The remaining
prestressing will be applied after casting curb. Design of
elastomeric bearing pads is based on established rules
found in literature6,7,8,9.

Precast Wall Panels and Joints

Typically, the wall panels are cast in 3m sections. The
panel curvature is developed by computing ordinates
for 3m section based on the tank radius. A sand bed is
prepared on the ground with proper curvature. A thin
steel sheet metal diaphragm (similar to corrugated steel
Fig. 8: Flooding and Curing of Concrete Floor (Waterstop and sheet) is placed on the sand bed. It is this steel diaphragm
Seismic Cables Embeded in Footing) that separates the concrete wall panels from shotcrete
and prestressing wires. The barrier provided by the
steel diaphragm prevents the water from reaching the
prestressing wires. Liquid tightness provided by the steel
diaphragm, when combined with alkaline rich shotcrete
encapsulating the prestressing wires, prevents corrosion
of wires. Hence, it is possible to use plain prestressing
wires without any need for galvanizing. Thousands of
prestress concrete tanks using this approach have been
designed and constructed in the United States using non-
galvanized prestressing wires.
Figure 11 shows field preparation of steel diaphragm
for casting 3m wide panels. Figure 12 shows concrete
operation for panel casting. A wood screed cut to the
curvature of the tank radius is used to “strike off” concrete.
AWWA Standard D110-0410 restricts allowable stress in
the steel diaphragm to 125MPa (18ksi) when counted as
reinforcement in the vertical direction. The bond stress
between plain steel metal diaphragm and concrete in
the vertical direction is one of the reasons for restricting
Fig. 9: Waterstop Embeded in Footing the allowable stress to 125MPa. This bond is fully tested

Organised by
India Chapter of American Concrete Institute 391
Session 4 A - Paper 2

Fig. 11: Diaphragm for Wall Panels

Fig. 13: Tilt Up Operation for Precast wall Panel

Fig. 12: Panel Casting and Finishing

during the panel pick up operation because of high flexural Fig. 14a: Wall Panel Set on Bearing Pads
stresses developed in the precast wall panel. Frequently,
the panel tilt up operation governs the reinforcing steel
requirements in the panel. Figure 13 shows panel tilt up
operation.
Figure 14 shows adjacent wall panels set on the top of the
bearing pads over footing. As can be seen, there is vertical
joint between wall panels, abut 178mm to 203mm wide.
Horizontal reinforcement placed at 1200mm on center
projects out from the panel ends. This reinforcement
is welded to 18mm X 18mm X 6mm thick steel angles,
150mm long. Welding of reinforcement to steel angles
provides stability to wall panels during construction
and prestressing operation. This joint is then filled with
shotcrete applied against the projecting ends of the
diaphragm sheets. Figure 15 shows joint shotcreteing
operation.
Fig. 14b: Vertical Joints between Wall Panels

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

392 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Design and Construction of Wirewound Circular Precast, Prestressed Concrete Tanks with Sliding Base

Prestres, at a given height from the tank floor, is not uniform

along the circumference due to curvature and friction
losses. Figures 18 and 19 show wirewinidng operation.

Fig. 15: Shotcreting of Wall Panel Joints

The design approach requires that the wall panel joints

remain in compression under full hydrostatic and Fig. 17: Tank Exterior After Shotcrete Coat Over Diaphragm
hydrodynamic fluid pressure. Minimum prestressing
is provided even above the waterline to ensure residual
compression in the wall panel joints. Seismic design
approach to compute hydrodynamic pressure and base
shear requirement is based on ACI 350.3-06 and related
literature.11, 12, 13

Wall Presrtessing and Shotcreting

After shooting all vertical wall panel joints, 12mm thick coat
of shotcrete is applied over the steel diaphragm. Figure
16 shows steel diaphragm on the outside of the tank wall.
Figure 17 shows similar view after shooting shotcrete
over diaphragm. The prestressing operation starts after
shotcrete coat has gained the required strength. Typically,
5mm diameter high strength steel wire (1517Mpa) is pulled
through the die using a prestressing machine. Pulling high
Fig. 18: Wirewinding Machine for Circular Prestressing
strength steel wire through die reduces the wire diameter
before it is placed on the wall. The reduction in the wire
diameter corresponds to prestressing force applied on
the wall. This method of “wirewinding” or prestressing
ensures continuous and uniform prestress along the tank
circumference at a particular elevation from the tank floor.
This is the most distinguishing feature of wirewound tanks
as compared to the tendon prestressed tanks, in which case
(1) The circumferential prestressing is discontinuous and (2)

Fig. 19a: Prestressed Wires on Wall

Design and detailing requirements dictate that wires be

spaced at least 1 wire diameter apart along the height of
the wall so as to completely encapsulate them in shotcrete
wire coat. This requirement restricts number of wires to
approximately 22 per 300mm height of wall. Thus, many
“layers” of wires would be required at the base of the
Fig. 16: View Showing Steel Diaphragm on Outside of Wall Panels

Organised by
India Chapter of American Concrete Institute 393
Session 4 A - Paper 2

Fig. 19: Measurement of Wire Prestress Force

wall to counteract higher liquid pressure as compared to

only one “layer” of wires at the top of the wall. In terms
of construction, the first layer of wires is applied along
the full length of the wall over diaphragm shotcrete coat.
This layer of wire is covered with shotcrete to provide
minimum 6mm cover coat over wires. The next layer of
wire is placed on the wall after shooting shotcrete over
the first layer of wires. As many as 15 layers of wires may Fig. 20: Falsework for Dome Construction
be required for bottom 2m of wall in case of tall, large
diameter tanks. At least 25mm of shotcrete cover coat is
applied over the final layer of prestressing wire to provide
adequate cover protection. The normal practice is to coat
the wall shotcrete with “breathable” paint after applying
final shotcrete cover coat.

Dome Construction and Prestressing

The most popular method of roof over potable water tanks
is spherical dome. Clear spanning domes are aesthetically
pleasing and provide tank interior that is free of any
obstruction for cleaning and maintenance purposes. This
is particularly important in case of digesters because
rotating equipment inside the tank will function only when
there are no obstructions. Furthermore, as compared
with flat slabs, domes require less concrete and steel and
hence, are economically attractive.
Fig. 21: Formwork before Dome Pour
Typically, concrete domes are cast in-situ due to double
curvature. However, many precast domes have also The design approach requires computing flexural
been built, mainly for small diameter tanks. Seismic stresses in the end regions of the spherical dome due to
performance of cast-in- place domes is far superior the “edge effect”. The dome tension ring is prestressed
to precast domes and hence precast domes are not to counteract the dead load and live load thrust. Typical
preferred in high seismic zones. Construction of cast-in- dome design requires “hinged” connection between dome
place domes requires false work erection and formwork edge and top of the wall. Detailed discussion of all design
as shown in Figures 20 and 21. The dome concrete is aspects is beyond the scope of this introductory paper.
poured in “pie” sections. Typically, 6 to 8 “pie” sections are However Billington19, Timoshenko20 and T.Y. Lin21 have
required to complete the dome pour. covered this topic in great detail.
The thickness of domes is in the range of 75mm to 125mm Dome prestressing operation begins after dome concrete
and hence domes are classified as “thin shell structures”. has gained the required strength. Typically, dome is
Design of concrete domes is governed by resistance to prestressed for both dead and live loads. Most of the times,
buckling, creep and imperfections. Thin spherical shells dome “lifts” off the formwork support after application
are very sensitive to buckling if imperfections get very of full prestress because full live load is never applied
large and hence it is necessary to set limits on the extent on the dome surface. This “lifting” up of dome facilitates
and curvature of “flat spots” and “thin spots”. 14,15,16,17,18 falsework and formwork removal operation.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

394 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Design and Construction of Wirewound Circular Precast, Prestressed Concrete Tanks with Sliding Base

Hydrotesting of Tank Summary

Most of the design and construction specifications Circular wirewound, prestressed concrete tanks offer
require that prestressed concrete tank be hydro tested superior performance and durability as compared to
to demonstrate the liquid tightness. Depending on the reinforced concrete tanks because it eliminates all the
intended use of the tank, the extent of “liquid tightness” problem areas associated with reinforced concrete tanks. In
varies from bottle tight to measurable loss within the addition, it is possible to construct ground supported tanks
specified limit. As per AWWA D110-04, an ANSI standard as tall as 25m which can replace a system of USR and ESR.
for potable water tank, the tank is filled with potable water Replacing two tanks (USR and ESR) with one prestressed
to the maximum level for 24 hours before starting the concrete tank provides significant savings in pumping
hydro test. The liquid drop is measured over the next 72 electricity cost, increased storage capacity and savings
hours to determine the liquid volume loss for comparison in land as well as construction costs. The superior water
with the allowable leakage. The tank is accepted if the net tightness of prestressed concrete tanks, demonstrated by
liquid loss for a period of 24 hours does not exceed 1⁄20 hydro testing after completion of construction, is particularly
of 1% of the tank capacity. Furthermore, wet spots on the important in the Indian context because of significant water
exterior of tank walls and flowing water at the base of the loss in storage and distribution system.
wall footing joint are not permitted. Necessary repairs
are carried out if tank does not pass the hydro test at no References
expense to the owner. 1. IS 3370, Vol. 1 through Vol. 4- “Code of Practice- Concrete Structures
for Storage of Liquids,” Bureau of Indian Standards, 2000
2. ACI 350-06, “Code requirements for Environmental Engineering
Painting And Architectural Treatment Concrete Structures and Commentary”, American Concrete Institute,
2006
The exposed surface of dome and wall is painted to improve 3. EN 1992-1-1 (2004), Eurocode 2, “Design of Concrete Structures- Part
aesthetics and durability of the tank. In addition, various 1-1: General Rules and Rules for Buildings.”
architectural treatment, as shown in Figures 22 through 23 4. Sanjay Mehta and Donald Cameron, “Bond Strength of Diaphragm-
can be applied on the tank exterior to improve aesthetics. Shotcrete Interface in the Vertical Direction, Technical Note, “
Earthquake Spectra, August 2013.
5. ACI 372R-03, Design and Construction of Circular Wire- and Strand-
Wrapped Prestressed Concrete Structures, American Concrete
Institute, 2003
6. AASHTO, “Standard Specifications for Highway Bridges,” Figteenth
Edition, 1992
7. Charles W. Roeder and John F. Stanton, “State-of-the-Art Elastomeric
Bridge Bearing Design,” ACI Strucutral Journal, February 1991.
8. “Handbook of Molded And Extruded Rubber,” The Goodyear Tire &
Rubber Company, Third Edition, Akron, Ohio 44316
9. James K. Iverson and Donald Pieifer, “Criteria for Design of Bearing
Pads,” Technical Report No. 4, PCI, June 1985
10. AWWA D110-04, “Wire-and Strand-Wound Circular, Prestressed
Concrete Water Tanks,” American Water Works Association, 2004
11. ACI 350.3-06, “Seismic Design of Liquid-Containing Concrete
Structures and Commentary,” 2006
12. Housner G.W., “Dynamic Pressure on Fluid Containers,” Technical
Information Document (TID) 7024, Chapter 6 and Appendix F, U.S.
Atomic Energy Commission, 1963
13. John A. Blume & Associates, “Report of Testing Program on Earthquake
Cable Detail for the Preload Company, Inc.,” July 1958
14. Bushnell David, “Nonlinear Axisymmetric Behavior of Shells of
Revolution,” AIAA Journal 5.3, 432-29, 1966
Fig. 22: Tank Coated with Breathable Paint 15. Krenzke, Martin A., and Thomas J. Kiernan, “The Effect of Initial
Imperfections on the Collapse Strength of Deep Spherical Shells”,
Report 1757, Model Basin In-house Independent Research Program,
1965
16. Huang, Nai-Chen, “Unsymmetrical Buckling of Thin Shallow Spherical
Shells,” Division of Engineering and Applied Physics, Harvard
University, Cambridge, Massachusetts, 1963
17. Zarghamee, Mehdi S., and Frank J. Heger, “Buckling of Thin Concrete
Domes,” ACI Journal, 487-500, 1983
18. Report of ACI Committee 344, “Design and Construction of Circular
Prestressed Concrete Structures,” J. ACI, No. 9, Proc. V. 67, p. 664,
September 1970
19. David P. Billington, “Thin Shell Concrete Structures,” 2nd Edition,
McGraw-Hill Book Company
20. S. P. Timoshenko and S.W. Krieger, “Theory of Plates and Shells,” 2nd
Edition, McGraw- Hill Book Company
21. T.Y. Lin and Ned H. Burns, “Design of Prestressed Concrete Structures,”
Fig. 23: Architectural Treatment‐ Brick Pilasters Third Edition, John Wiley & Sons

Organised by
India Chapter of American Concrete Institute 395
Session 4 A - Paper 3

Damage Assessment in Concrete Structures using PZT patches

Arun Narayanan and Kolluru V.L. Subramaniam
Department of Civil Engineering, Indian Institute of Technology Hyderabad, INDIA, kvls@iith.ac.in

Abstract application or from internal sources such as shrinkage.

Damage initiation takes place in the form of distributed
Piezoelectric based PZT smart sensors offer significant
microcracks, which eventually localize to form cracks.
potential for continuously monitoring the development and
Often the damage, particularly in the incipient stages
progression of internal damage in concrete structures.
is not directly visible and by the time signs of distress
PZT-based damage sensors consisting of piezo-electric
appear on the surface of the structure significant damage
patches, which are bonded to the surface of a concrete
would have accrued in the structure and there may be
structure can be developed for assessing the damage
significant degradation of the capacity of the structure.
progression of concrete members. The primary challenge
Early detection of damage, before visible signs appear on
in developing a PZT-based sensor lies in developing a
the surface of the structure is essential to initiate early
methodology to infer about the level of damage in the
intervention, which can effectively increase the service
material from measurement. Changes in the resonant
life of structures. Methods to detect incipient damage in
behavior in the measured electrical conductance obtained
the form of microcracks are required to provide effective
from electro-mechanical (EM) response of a PZT bonded
methods of monitoring structural health and service life
to a concrete substrate is investigated for increasing
performance of structures.
levels of damage. The sensitivity of EM impedance-
based measurements to level of damage in concrete is Use of PZT patches and wafers has become popular in
reported. Incipient damage in the form of microcracks in structural health monitoring. Due to the coupled electro-
the concrete substrate produces a change in the electrical mechanical constitutive response of a PZT material, the
conductance signature associated with the resonant mechanical response of a bonded PZT patch subjected to
peaks. Changes in the conductance resonant signature an applied electrical potential is influenced by the elastic
from EM conductance measurement are detected before restraint provided by the substrate material. Coupling the
visible signs of cracking. The root mean square deviation structure to the PZT changes the mechanical impedance
of the conductance signature at resonant peaks is shown of the PZT, which produces a change in its vibration
to accurately reflect the level of damage in the substrate. characteristics. Monitoring changes in the electrical
The findings presented here provide a basis for developing impedance signature due to changes in the effective
a sensing methodology using PZT patches for continuous mechanical impedance of the substrate is the basis for
monitoring of concrete structures. electromechanical impedance-based measurements.
Information about the surrounding material is contained
Keywords: PZT, impedance, Conductance, Microcracks.
in the electromechanical impedance (EMI) signature of a
PZT. By comparing the impedance signature taken in the
Introduction pristine state and at any other time, structural damage can
Structural Health Monitoring (SHM) is a process of be determined. Generally, both frequency and amplitude
assessing the structural integrity of the constituent shifts are produced relative to the pristine state (without
parts and the level of damage level in the structure damage). [Chaudhry et al. (1994), Sun et al. (1995), Ayres
during its life period. SHM relies on non-destructive et al (1998), Giurgiutiu et al (1997, 1999), Park et al (2000),
evaluation (NDE) procedures and continuous monitoring Zagrai (2001), Giurgiutiu (2002, 2004), Peairs (2004)].
of structural parameters to determine the intensity Application of EMI technique for damage detection in
and location of the damage. This involves sensors, data concrete structures requires a careful study of the
acquisition and signal processing tools. Signs of distress changing compliance of the substrate for different forms
in concrete are often associated with visible cracking. of damage in the substrate material from the incipient to
Since concrete is a brittle material, which is weak in the visible stages. The use of PZTs for health monitoring of
tension, cracking is the manifestation of damage in the concrete structure was demonstrated by the ability of EMI
material which results from tensile stress in the material. technique to register changes due to formation of cracks
Stress induced damage in concrete could result from load well in advance of failure [Park et al. (2000)]. Several

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

396 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Damage Assessment in Concrete Structures using PZT patches

other studies of damage in concrete using impedance- Average of five measurements was collected. Impedance
based measurements of PZTs have been conducted using data was collected from the PZT patch in the free-state
embedded defects and artificial damage in the form of before attaching the PZT to the concrete cube. The
machine cuts [Tseng (2004), Lim (2006), Dongyu (2010), baseline EM conductance signature and image were
Wang (2013)]. The EM impedance method has also been taken prior to the start of loading. Cubes were subjected
used to determine the location of a crack by inducing crack to cyclic compressive loading of increasing magnitude
at different positions and depths and cross correlation where the load amplitude was increased in increments of
as damage index [Wang et al. (2013)]. While the use of 10% of the average compressive strength in every cycle.
artificial damage provides meaningful insight, it is not The loading procedure consisted of alternate loading and
representative of substrate compliance with stress/load unloading cycles as shown in Figure 1b. During the loading,
induced damage in the material. the conductance signatures and the image for DIC were
recorded on top of the load cycle and after unloading.
Potential of using electro mechanical impedance based
measurements of surface mounted PZT to identify the
formation of incipient damage in concrete structures EM Impedance of PZT
is presented in the paper. The relationships between
forms of material damage, visual indication of damage,
mechanical compliance of the material and resonant
modes in the conductance signature of PZT bonded to
a concrete substrate are investigated. The variation (a)
in surface strains for incremental levels of loading is
monitored using Digital Image Correlation (DIC) and
compared with the conductance plot of the PZT. Root
mean square deviation (RMSD) of the EM conductance
close to the resonant peak is used as a damage index and
variation in RMSD at different damage states is presented.

Experimental Program
Experiments were performed using 150 mm concrete
cubes. Six cubes were cast and cured for 90 days before (b)
testing. The properties of the cube is given in table1.
The three cubes were tested to failure to determine the Fig. 1: (a) Experimental set up (b) Applied loading history
compressive strength of the concrete.
In a PZT material, the application of an electrical field
Table 1
results in mechanical strain in the material due to the
Properties of materials coupled electro-mechanical constitutive relations. For a
PZT patch attached to a substrate subjected to an applied
Average Young’s Density electrical input, the motion of the interface subjected
Poisons
Type Failure stress Modulus (ρ)
(Mpa) (GPa) (kg/m3)
ratio (υ) to continuity conditions is governed by the combined
mechanical impedance of the structure and the PZT.
Concrete cube 52 36 2300 0.2
The constrained motion in turn produces a change in
Epoxy - 2 1400 0.36 the measured electrical impedance. The first systematic
attempt to derive the electrical impedance of the PZT
which is mechanically connected to a structure using a
The cubes were bonded with PZT patches exactly at the
1D idealization of the system was developed by Liang et
center of the side face of the cube using two component
al. (1994). Subsequent improvements in modelling the
epoxy. A 20mm x 20mm PZT patch, which was 1mm in
PZT response have included the effective 1-D model of
thickness, was used for the experimental study. The front
the PZT and varying levels of idealization of the structural
faces of the cubes were smoothened and a sprayed-on
impedance [Bhalla et al. (2004), Xu (2002), Yang et al
speckle pattern was created for measurement of surface
(2005)]. Most of the available analytical solutions are
displacements using the full-field optical technique
applicable for 1 or 2-D idealizations of the PZT, substrate
known as digital image correlation (shown in Figure 1a).
or both. Typically the electrical impedance of the PZT
The baseline signatures of the PZT when attached to the
patch for a given electrical input at a frequency can be
substrate are taken.
represented as
In a typical impedance measurement, the frequency was Yr = Yr (Z A, Z S, ~, l i, E) .....................................................(1)
varied between 1 kHz and 0.5 MHz at an applied voltage
of 1V and data was collected at 800 discrete frequencies. where Z A And ZC are the mechanical impedance of the PZT

Organised by
India Chapter of American Concrete Institute 397
Session 4 A - Paper 3

and substrate respectively. li, represent the dimensions of Analysis of results

patch and E is the electric field applied for actuation. The From the results of the numerical analysis, the second
functional form of Equation (1) is not readily available for peak in the EM conductance response of the bonded PZT
the test configuration used in this study. was selected for evaluating the influence of load-induced
The conductance, which is the real part of admittance of damage. The conductance signatures at the second peak
the free PZT and the PZT bonded to the 150 mm concrete of the bonded PZT response after unloading from different
cube are shown in Figure 2. It can be seen that resonance load levels are shown in figures 3. The second peak is
peaks associated with the free vibration of the PZT can centered on 255 kHz. The response between 245 and 265
also be identified in the response of the PZT attached to the kHz is plotted in the figures. Contours of horizontal strain at
concrete cube. Only three prominent peaks are identified distinct loading obtained from the DIC technique are shown
in the conductance spectrum of the bonded PZT. Peaks in figure 4. It can be clearly identified from the plot that the
1 and 2 in the conductance spectrum of the bonded PZT unloading signature at 40%u shows a left ward shift. This
correspond with modes 1 and 3 respectively of the PZT. The is due to the incipient damage produced in the concrete.
third peak in the conductance response of the bonded PZT Horizontal strain contour shows an increase in strain levels
has contributions from closely spaced modes 5 and 6 of (figure 4). As the load level increases, the resonance peak
the PZT. There are several prominent changes associated in the conductance signature shows a consistent leftward
shift. Comparing with the measured DIC response, there
with the frequency of the resonant modes and the relative
is no visible sign of distress or cracking up to 70% of
magnitude of the resonant peaks. There is a noticeable
strength, while some signs of localization are evident at
decrease in values of conductance, an increasing baseline
60% of peak. Localization of damage into a crack occurs
trend which increases the magnitude of conductance
at 70% of strength. Significant changes in the resonant
with increasing frequency and a change in the relative
peak associated with the localization are observed. After
magnitudes of the resonant peaks in the bonded state.
localization, significant changes are observed in the shape
There is also a significant broadening of the resonance of the resonant peak. At 90% of the compressive strength,
peaks compared with the free-state. The resonance peaks the peak showed a significant decrease in amplitude and a
shift to higher frequencies, with a larger frequency shift in flattening of the peak. The flattening of the peak is associated
lower modes. with the formation of a major crack on the surface. The
The resistance to the motion of the PZT by the substrate is conductance signatures associated with the resonant peak
reflected in the overall decrease in the value of conductance. has a very good agreement with the indication of damage
While the conductance of the free PZT is essentially zero obtained from surface strain measurements. Further,
between resonant peaks, the conductance is non-zero changes in EM conductance are observed before any visible
between the resonant peaks for the bonded PZT. The sign of distress.
resistance to the motion of a point located on the surface of
the cube, given by the driving point impedance, influences
the motion of the bonded PZT. The frequency dependency
of the substrate driving point impedance is reflected in the
relative shifts in the amplitudes and the general increasing
trend in the background of the measured conductance.
The influence of the substrate can also be identified with
the overall increase in the frequency and broadening of the
resonant peaks.

6
4 5
2

Fig. 2: Conductance spectrum of PZT in the free condition and Fig. 3: Electrical conductance signatures: a. 30%-50% of
coupled with a 150 mm concrete cube. strength b. 60%-90% of strength

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

398 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Damage Assessment in Concrete Structures using PZT patches

(a) (b)

(c) (d)

(a)

(e) (f

(a)

(b)

Fig. 5: (a) RMSD of the second resonance peak (b) Average

Fig. 4: Contours of horizontal strain (exx) obtained using digital vertical strain (εyy) obtained From DIC
image correlation (a) at 40%; (b) at 50%; (c) at 60%; (d) at 70%;
(e) at 80%; and (f) at 90% of Strength. Conclusions
Potential of using EM impedance measurements of
The root-mean-square deviation (RMSD) is used to
surface mounted PZT patches for structural health
measure the differences between values of baseline
monitoring of concrete structures is presented in the
measurement of conductance signature at the second
paper. It is shown that there are changes in resonant
resonant peak and the corresponding signatures at
behavior of the EM conductance response of the PZT
different load levels. The RMSD for the frequency range 245
bonded to a concrete substrate with increasing damage.
kHz-260 kHz with respect to the baseline measurement
The PZT sensor detects incipient damage significantly
were calculated using equation 2, where xi and yi are the
earlier than the appearance of visible signs of damage.
signatures obtained from the PZT transducer bonded to
There is an amplitude reduction and frequency shift of
the structure before and after damage (or loading). The
the PZT resonance peak with an increase in damage in
scatter in the results obtained from all the specimens is
the concrete substrate. At higher damage levels, there is
also plotted in the Figure 5a. It can be seen that despite
flattening of the resonant peak associated with localization
the scatter, there is an increasing trend of RMSD with each
and formation of a major crack.
level of loading as shown in the figure. The variation in the
average vertical strains recorded at the top and bottom
of the load cycles obtained from DIC measurements are References
1. Ayres, J.W., Lalande, F., Chaudhr y, Z., and Rogers, C.A.,
also plotted in Figure 5b. It can be seen that the level of
1998. Qualitative impedance-based health monitoring of civil
damage assessed using the RMSD variation of the second infrastructures. Smart Materials and Structures, 7(5):599-605.
resonant peak compares well with the evolution of plastic 2. Bhalla, S., and Soh, C.K., 2004. Structural Health Monitoring by
strain and increase in mechanical compliance. There is an Piezo-Impedance Transducers. I: Modeling. Journal of Aerospace
exponential increase in the evolution of plastic strain with Engineering, 17(4):154–165.
loading. Plastic strain is an indicator of level of damage in 3. Chaudhry, Z., Joseph, T., Sun, F., and Rogers, C.A., 1995. Local-Area
the material. This corresponds with the observed trend in Health Monitoring of Aircraft via Piezoelectric Actuator/Sensor
the RMSD measure with loading. Patches. Proceedings, SPIE North American Conference on Smart
Structures and Materials, 2443:268-276.
| Qy - x V
N
i=1 i i
2

...............................................(2) 4. Giurgiutiu, V., and Rogers, C. A., 1997. Electro-mechanical (E/M)

RMSD =
| x N
i=1 i
2 impedance method for structural health monitoring and non-
destructive evaluation. International Workshop on Structural Health
Monitoring, Stanford University, CA, September 18–20:434–444.

Organised by
India Chapter of American Concrete Institute 399
Session 4 A - Paper 3

5. Giurgiutiu, V., and Zagrai, A.N., 2000. Characterization of of the Impedance-based Structural Health Monitoring Method.
Piezoelectric Wafer Active Sensors. Journal of Intelligent Material Journal of Intelligent Material Systems and Structures, 15(2):129-
Systems and Structures, 11(12):959-976. 139.
6. Giurgiutiu, V., Reynolds, A., and Rogers, C.A.,1999. Experimental 13. Sun, F.P., Chaudhry, Z., Liang, C., and Rogers. C.A., 1995. Truss
Investigation of E/M Impedance Health Monitoring for Spot-Welded Structure Integrity Identification Using PZT Sensor-Actuator.
Structural Joints.Journal of Intelligent Material Systems and Journal of Intelligent Material Systems and Structures, 6(1):134-139.
Structures, 10(10):802- 812.
14. Tseng, K.K., and Wang, L., 2004. Smart piezoelectric transducers
7. Giurgiutiu, V., Zagrai, A.N., and Bao, J.J., 2002. Piezoelectric Wafer for in situ health monitoring of concrete. Smart Materials and
Embedded Active Sensors for Aging Aircraft Structural Health Structures, 17(5):1017-1024.
Monitoring. Structural Health Monitoring, 1(1): 41-61.
15. Wang, D., Song, H., and Zhu. H., 2013. Numerical and experimental
8. Giurgiutiu, V., Zagrai, A.N., and Bao,J.J., 2004. Damage Identification studies on damage detection of a concrete beam based on PZT
in Aging Aircraft Structures with Piezoelectric Wafer Active Sensors. admittances and correlation coefficient. Construction and Building
Journal of Intelligent Material Systems and Structures, 15(9):673- Materials, 49:564–574.
687.
16. Xu, Y.G., and Liu, G.R., 2002. A Modified Electro-Mechanical
9. Liang, C., Sun, F.P., and Rogers, C.A., 1994. An Impedance Method for Impedance Model of Piezoelectric Actuator- Sensors for Debonding
Dynamic Analysis of Active Material Systems. Journal of Vibration Detection of Composite Patches. Journal of Intelligent Material
and Acoustics, 116(1):120-128. Systems and Structures, 13(6):389-396.
10. Lim, Y.Y., Bhalla, S., and Soh, C.K., 2006.Structural identification and 17. Yang, Y., Xu, J., and Soh, C.K., 2005. Generic Impedance-Based
damage diagnosis using self-sensing piezo-impedance transducers. Model for Structure-Piezoceramic Interacting System. Journal of
Smart Materials and Structures, 15(4):987-995. Aerospace Engineering, 18(2):93-101.
11. Park, G., Cudney, H., and Inman, D., 2000. Impedance-Based Health 18. Zagrai, A.N., and Giurgiutiu, V., 2001). Electro-Mechanical Impedance
Monitoring of Civil Structural Components. Journal of Infrastructure Method for Crack Detection in Thin Plates. Journal of Intelligent
Systems, 6(4):153–160. Material Systems and Structures, 12(10):709-718.
12. Peairs, D.M., Park, G., and Inman, D.J., 2004. Improving Accessibility

Arun Narayanan
Arun Narayanan is a research scholar in Department of civil engineering at Indian institute of technology
Hyderabad,India. His areas of interest are Sensor Development for Structural Health Monitoring;Structural
dynamics, Non- destructive Evaluation.

K. V. L. Subramaniam
Prof. K. V. L. Subramaniam is currently the Dean (Planning and Development) and a Professor in the
Department of Civil Engineering at Indian Institute of Technology Hyderabad (IITH). Prior to joining IITH, he
was a Professor and Catell Fellow in Department of Civil Engineering at the Grove School of Engineering,
the City College of New York (CCNY). Dr. Subramaniam obtained a B.Tech. in Civil Engineering from IIT
Delhi and Ph.D. in Structural Engineering and Materials from Northwestern University, Evanston. After
graduation, Dr. Subramaniam worked as a Research Associate at the NSF Center for Advanced Cement
Based Materials. Dr. Subramaniam’s research interests include Fracture and Fatigue of cementitious
materials, FRP-based Structural Strengthening, Fly ash based binders, Geopolymers and Non-destructive
testing and evaluation techniques for concrete structures.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

400 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Prestress of Reinforced ASR-Expansive SHCC Beams and Possibility of Other Expansive Materials

Chemical Prestress of Reinforced ASR-Expansive SHCC Beams

and Possibility of Other Expansive Materials
Hiroo TAKADA, Keitetsu ROKUGO, Koichi KOBAYASHI, Masashi KAWAMURA and Yukio ASANO
Department of Civil Engineering, Gifu University, JAPAN
Hyundo YUN
Department of Architectural Engineering, Chungnam National University, KOREA

Abstract to shrinkage as shown Figure 1. It is therefore difficult to

maintain expansiveness by using an expansive additive.
The purpose of this contribution is to develop new
cementitious materials that are capable of keeping If a concrete that continues moderate-expansion over a
expanding without shrinkage over a long time in a stable long time without cracking is developed, it is hopefully
manner. ASR (Alkali-Silica Reaction)-expansive SHCC applicable to anchor bolt fixing, filling of drilled holes and
(Strain-Hardening Cement Composite) was developed. form tie holes, backfilling of tube hole grooves in concrete,
Chemical prestress was introduced into this material by machine installation and so on.
restraining the ASR expansion using reinforcing bars.
Bending tests on the reinforced ASR-expansive SHCC
beams showed that the cracking strength increased due
to chemical prestress induced by ASR. The expansive
behavior of the matrix mortar (SHCC without fibers)
containing Ex (expansive additive), NEDA (non-explosive
demolition agent), or WG (water glass cullet) was
examined. NEDA suggested a possibility of being used
as a material to impart expansiveness to SHCC after
hardening without damage.
Keywords: SHCC, alkali-silica reaction, expansion, Fig. 1: Conceptual drawing of the length change behavior of
chemical prestress, expansive materials. cementitious materials
The purpose of this contribution is to develop new
Introduction cementitious materials that are capable of keeping
Concrete is a material prone to shrinkage, and restrained expanding without shrinkage over a long time in a stable
shrinkage is prone to cause cracking. For this reason, manner. This contribution consists of two parts. In the
expansive additives that expand at early ages and first part, an SHCC which slowly expands over a long time
shrinkage-reducing admixtures have been used for by incorporating alkali-silica reaction (ASR) expansion is
concrete to reduce its shrinkage. Excessive expansion proposed, and chemical prestressing of this material by
causes cracking in concrete. Figure 1 shows a conceptual placing reinforcing steel bars is investigated (Rokugo et
drawing of the length change of cementitious materials. al., 2014). The authors report on the effect of prestressing
Strain-Hardening Cement Composites (SHCC) based on changes in the neutral axis depth of small R/SHCC
demonstrate excellent mechanical behavior showing beams and the results of shrinkage compensation by ASR
tensile strain-hardening and multiple fine cracks based on their length changes due to drying shrinkage. In
the second part, the authors investigated a non-explosive
under tensile forces. Because of such excellent tensile
demolition agent (NEDA) and water glass cullet (WG) as
performance, it is expected that expansion of SHCC after
materials that show expanding after hardening with the
hardening does not cause a defect.
effect of offsetting the later-age shrinkage of SHCC that
In a past study (Takada et al., 2010), the authors reported has been expanded at an early age using an expansive
that chemical prestressing by adding a large amount of an additive (Ex), as shown in Figure 1. NEDA for demolishing
expansive additive is effective in increasing the cracking concrete pile heads in construction sites is designed to
load and decreasng the number of cracks of SHCC utilize the expansive pressure generated by reaction
beams reinforced with steel bars (hereafter referred between calcium oxide and water, with the hydration rate
to as R/SHCC). The expansion induced by an expansive being extremely suppressed by grading and a retarder
additive generally subsides at early ages and then turns (Ishii, 2006). WG shows so significant water-absorbing

Organised by
India Chapter of American Concrete Institute 401
Session 4 A - Paper 4

expansion with alkali-silica gel that it is used as a model appearance of NEDA. WG with a silica-alkali ratio (SiO2/
material for ASR in substitution for natural reactive Na2O) of 3.6, which was found to show significant ASR
aggregate (Iwatsuki et al., 2009). expansion (Iwatsuki et al., 2009), was prepared, with the
particle size adjusted to 0.15 to 0.3 mm.
Experimental Program
Table 2. Mixture proportions of SHCC for the first part
Materials and Mix Proportions
Unit mass (kg/m3)
Table 1 gives the properties of materials used in the test. W/C W/P
SHCC Powder
In the first part of this contribution, river sand as an alkali- PE:1.25% (%) (%) W* SA
SP MC PE
(S6+S7)
silica reactive aggregate was used as the fine aggregate, (Control) C LP
and silica sand was used as the control aggregate. Figure 50 38 390 780 234 550 (558) 9.1 1.0 12.1
2 shows the alkali-silica reactivity of both aggregates
*W includes superplasticizer (SP)
determined by the chemical method of JIS A 1145. This
river sand was taken from an upper part of a river rising
Table 3. Mixture proportions of SHCC for the second part
from a volcanic area.
Expansive W/C Unit mass (kg/m3)
Table 1. Properties of materials used in the tests material (%)
W C LP Ex NEDA WG S6+7 PE
Materials Properties and specifications
Control 50 390 780 234 - - - 558 (12.1)
High-density Diameter: 0.012 mm, Length: 12 mm,
polyethylene fibers Density: 0.98 g/cm3, Strength: 2. 6 GPa, Ex 50 390 733 234 47 - - 558 (12.1)
(PE) Elastic modulus: 88 GPa
NEDA 50 390 780 234 - 5 - 552 (12.1)
Cement (C) JIS R5210 High-early-strength Portland
cement, Density: 3.13 g/cm3 WG 50 390 780 234 - - 39 515 (12.1)
Expansive additive JIS A6202 Ettringite and calcium hydrate
(EX) combined formation type, Density: 3.05g/ SP:9.13 kg/m3, MC: 1.01 kg/m3
cm3
Non- explosive Pile head remover after bore piling work,
demolition agent Main component : calcium oxide
(NEDA)
Water glass cullet SiO2/Na2=3.6, Patricle size 0.3-0.15 mm
(WG)
Limestone powder Density: 2.71 g/cm3, Specific surface aera:
(LP) 3050 cm2/g
Alkali silica reactive Maximum diameter: 600 µm, Density: 2.55
sand (SA) g/cm3
Silica sand No.6 (S6) Maximum diameter: 600 µm, Density: 2.60
g/cm3
Silica sand No.7 (S7) Maximum diameter: 200 µm, Density: 2.60
g/cm3
Superplasticizer (SP) JIS A 6204 High-range water reducing
admixture, Polycarboxylate ether based Fig. 2: Alkali-silica reactivity of fine aggregate
Methylcellulose (MC) Nonionic water-soluble type

Table 2 gives the mixture proportions of the SHCC

containing 1.25% of polyethylene (PE) fibers (hereafter
referred to as ASR-SHCC) for the first part. Part of
cement of this mixture is replaced with limestone powder
to adjust the W/C to 50%, and the fine aggregate content
was increased to 550 kg/m3 to make ASR expansion
evident. Also, particle sizes less than 0.6 mm were used in
consideration of the pessimum aggregate size of this sand
for ASR obtained from another experiment (Takada et al.
2014). For control specimens, silica sands No. 6 and No. 7
were blended at a mass ratio of 4:6 to equalize the grading
with the reactive aggregate. Table 3 shows the mixture
proportions of SHCC containing expansive materials and
polyethylene fibers for the second part. NEDA 0.6 mm or
less in particle diameter was used. Figure 3 shows the Fig. 3: Appearance of NEDA before crushing

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

402 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Prestress of Reinforced ASR-Expansive SHCC Beams and Possibility of Other Expansive Materials

Specimen Preparation and Testing Method

In the first part, specimens were subjected to accelerated
curing to confirm the effect of ASR rapidly. Based on our
earlier study, the equivalent alkali content was increased
to 12 kg/m3 (N2O equivalent) at the time of mixing using
NaCl, and the specimens were immersed in a 1 mol/liter
NaOH solution at 60OC.
The chemical prestress of R/SHCC beams was evaluated
directly from the length change along the central axis of
each beam embedding gauge plugs. The length change
was physically measured in accordance with JIS A 1129-3
(Method with dial gauge). The ASR-accelerating container
Fig. 5: Flexural beam test setup
containing the specimens was transferred from the
thermostatic chamber at 60OC to thermo-hygrostatic The flexural loading tests were conducted on beams by
chamber at 20OC, 24 h before measurement to attain the symmetrical 2-point loading as shown in Figure 5. The
temperature equilibrium. load and the displacements at the support and loading
points were measured using a load cell and sensitive
D6 bars were placed in two levels, doubly or singly
displacement transducers, respectively. A strain
reinforced. When varying the restraining steel ratio,
the bars on the compression side were removed. The gauge with a gauge length of 30 mm was glued on the
restraining steel ratios in doubly and singly reinforced top (compression side) surface, and those with gauge
were 2.11% and 1.06%, respectively. Figure 4 shows a lengths of 30 mm and 60 mm were glued parallel to the
cross-section of a beam specimen. The yield strength and bottom (tension side) surface of each specimen. After
tensile strength of reinforcing bars were 332.1 N/mm2 and moist-air-curing up to an age of 14 days and datum length
457.8 N/mm2, respectively. The R/SHCC beams of 11 types measurement at 20oC, the specimens were subjected
were prepared as shown in Table 4. to accelerated curing. With the target expansion levels
being 0.05%, 0.1%, and 0.15%, the rate of length change
was continually measured. The specimens were then
transferred to a moist air environment at 20oC until
simultaneous loading testing. Prior to loading, the
length change due to drying shrinkage during storage
was measured to examine the rate of length change of
specimen.
In the second part of this contribution, the expansive
behavior of each expansive material was examined by
adding each to the matrix mortar (SHCC without fibers)
using cylindrical tin molds, with strain gauges attached
Fig. 4: Cross-section of beam specimen to the side wall, referring to the Japan Concrete Institute

Table 4. Description of R/SHCC beam specimens

Curing age (days) Rate of length change (%) Chemical
Material Steel
Beam ID prestress
type reinforcement ASR accelerated Humid air After ASR acceleratd Before flexural test (N/mm2)
ASR-0-1 Without 35 9 0.3103 0.2664 -
ASR-S-1 Singly 21 23 0.1550 0.1036 -
ASR-D-1 14 30 0.0844 0.0353 1.59
ASR-SHCC
ASR-D-2 21 23 0.1100 0.0608 3.52
ASR-D-3 Doubly 35 9 0.1442 0.1044 4.53
ASR-D-4 21 - 0.1078 - -
Control-D-1 Control 35 9 0.0503 0.0131 -
ASR-0-0 Without 0 44 - -0.0458 -
ASR-S-0 ASR-SHCC Singly 0 44 - -0.0392 -
ASR-D-0 0 44 - -0.0381 -1.61
Doubly
Control-D-0 Control 0 44 - -0.0244 -

Organised by
India Chapter of American Concrete Institute 403
Session 4 A - Paper 4

(JCI) Standard (Test method for restrained expansion of Results and Discussion
expansive concrete using cylindrical molds).
Chemical Prestress of Reinforced ASR-Expansive
The matrix mortar was proportioned as given in Table SHCC Beams
3, without fibers, with the air-entraining and high-range
water-reducing admixture (SP) reduced to 1/6 to equalize Table 4 gives the list of R/SHCC beam specimens tested.
the flow with the SHCC. A ratio of 6% in place of part of Figure 8 shows the changes in the rate of length change
cement was selected as the Ex content based on the of ASR-expanded beams. This figure reveals that the
authors’ previous study (Takada et al. 2010). NEDA 0.6 mm expansion of R/SHCC beams due to ASR varies depending
or less in particle diameter was added at a ratio of 5 kg/m3 on the restraining steel ratio. Each beam specimen was
based on our preliminary study (Kawamura et al., 2015), removed from the accelerated curing condition at the
For WG, a ratio of 5% by mass of cement was adopted point of expansion given in Table 4. Figure 9 shows the
based on a study by Iwatsuki et al. (2009). No alkali was state of chemical prestressing of R/SHCC beams by
added during mixing, as 3.9 kg/m3 of alkali was calculated reinforcing bars that restrain the expansion of ASR-0 with
to be present in the matrix mortar based on the amount of no reinforcing bars. In this study, chemical prestress was
cement with an assumed alkali content of 0.5%. evaluated by Eq. (1) referring to the equation for expansive
concrete.
Cylindrical tin molds 50 mm in inside diameter and 100 v cp = p $ E s $ f sp = p $ E s $ f cp ............................................(1)
mm in inside height were used, with strain gauge with
where σcp = chemical prestress to be introduced (N/mm2)
a length of 30 mm being attached to the side as shown
in Figure 6. These molds were embedded in foamed εsp = tensile strain generated in steel
polystyrene containers as shown in Figure 7 and stored εcp = expansive strain of R/SHCC
in a thermostatic chamber at 20oC, in consideration of the Es = Young’s modulus of steel (N/mm2)
fact that the molds smaller than those specified in the JCI
p = restraining steel ratio (= A s/Ac)
standard cause the expansive material to be less prone
to the effect of cement hydration heat. A thermocouple (Ac = cross-sectional area of SHCC,
was connected to one of the three specimens from each A s = cross-sectional area of steel)
mixture. A 3-lead system was adopted to minimize the
effect of temperature changes on the leads.

Fig. 6: Cylindrical tin mold

Fig. 8: Curing age and rate of length change of beams

Fig. 7: Insulated molds Fig. 9: Curing age and chemical prestress

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

404 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Prestress of Reinforced ASR-Expansive SHCC Beams and Possibility of Other Expansive Materials

Similarly to expansive concrete, chemical prestress

induced by ASR is affected by the restraining steel ratio.
A greater restraining steel ratio tends to lead to a greater
chemical prestress.
This is presumably because as our earlier study showed
SHCC involving ASR retains its uniaxial tensile properties,
that is, ASR expansion of dispersed aggregate particles is
led to stable crack propagation owing to bridging fibers, to
be restrained by the reinforcing bars thoroughly.

Flexural Loading Tests on Reinforced ASR-Expansive Fig. 10: Load-deflection relationship of ASR accelerated
SHCC Beams specimens
Flexural loading tests were conducted on beam specimens
given in Table 4 at an age of 58 days. The rate of length
change of the specimens due to ASR-accelerating curing
was partly offset by drying shrinkage and creep, reaching
the values given in Table 4 at the time of loading testing.
Figure 10 shows the load-deflection relationship under
flexural loading of beam specimens with ASR-accelerating
curing. Figure 11 shows the load-deflection relationship
of moist-air-cured beams. No marked difference is
observed between the curve shapes of both ASR and non-
ASR steel bar-reinforced beams, but the curve shapes of
ASR-0-1 and ASR-0-0, both having no reinforcing bars, Fig. 11: Load-deflection relationship of moist-air cured specimens
show different tendencies. This can be explained as
follows: As described in a past study (Takada et al., 2014),
SHCC involving ASR lead to a small ultimate strain under
uniaxial tension testing. Therefore, in flexural loading
testing on SHCC beams involving ASR, ASR-expanded
ASR-0-1 with no reinforcing bars leads to a small
deflection-hardening zone, with softening beginning at an
early stage. In contrast, the deflection-hardening zone of
ASR-0-0, which is non-ASR with no reinforcing bars, is
much longer.
To confirm the effect of chemical prestress induced by
ASR, Figure 12 shows the load-deflection relationships Fig. 12: Load-deflection relationship of R/SHCC at an early stage
of loading
of four R/SHCC beams, ASR-D-0, ASR-D-1, ASR-D-2, and
ASR-D-3, which were designed to lead to different rates of
length chage. This figure reveals that the load-deflection
relationships of ASR-D-1 to -3 rise with steeper slopes at
an early stage of loading due to the chemical prestrain
of reinforcing bars. The chemical prestressing’s effect
of reducing the deflection is therefore confirmed when
compared with ASR-D-0 at the same loading level. The
relationship between this chemical prestress and the
flexural loading test results is summarized as shown in
Figure 13. The asterisk in the figure represents the value
for ASR-0-1 with no reinforcing bars as the cracking load
of the material. Figure 13 demonstrates the ASR-induced
chemical prestress’s effect of reducing the deflection at
yield and increasing the cracking load of R/SHCC beams Fig. 13: Relationship between chemical prestress and flexural
in proportion to the chemical prestress. loading test results

Organised by
India Chapter of American Concrete Institute 405
Session 4 A - Paper 4

Expansive Behavior of Matrix Mortar (SHCC without

Fibers) Containing Expansive Materials
Figure 14 shows the expansive behavior of matrix mortar
with each material in cylindrical tin molds conforming
to the JCI standard. The expansion rates are averages
of three specimens simultaneously fabricated. In these
tests, specimens were stored, with the plastic film
to seal the placing surface being removed 24 h after
placing to allow moisture escape only from this surface.
This deviates from the JCI standard for the purpose of
material development. The storage environment was 20
± 3oC and 60 ± 5% R.H. Figure 15 shows the temperature
of the thermostatic chamber and the molds embedded
in foamed polystyrene containers in consideration of the
effect of cement hydration heat. Fig. 14: Expansive behavior of matrix mortar
As shown in Figure 14, the expansion rate of specimens
containing Ex in the present test environment peaks at 3
days and begins a downward trend, decreasing by around
60% by 30 days. Specimens containing NEDA hit the
peak later than those containing Ex at 7 days and slowly
begin to shrink, decreasing by around 20% by 30 days.
It has been pointed out that the shrinkage behavior of
cementitious materials is affected by their pore structure
(Imamoto, 2007). The pore structure of mortar that is
uniformly expanded at an early age by a powdery Ex may
differ from that of mortar in which granular NEDA locally
expands during the process of cement hardening. Such
a difference is presumably the reason for the different
shrinkage tendencies of these materials under test.
Fig. 15: Temperature at early age
As found from Figure 14, matrix mortar containing WG
demonstrates moderate expansion. In order to confirm
possible expansion due to ASR, specimens with WG in
Figure 14 were transferred to an environment at 40oC and
≥95% R.H. Strain gauges protected from water by butyl
rubber tape were attached to the molds, and the molds
were coated with plastic film to prevent moisture intrusion
to be subjected to heat curing.

Figure 16 shows the expansive behavior of specimens

subjected to ASR-accelerating curing beginning at an
age of 33 days. Figure 17 shows the expansive behavior
of specimens subjected to accelerated curing in the same
environment beginning at an age of 2 days. Since the JCI
standard specifies the evaluation range of restrained
expansion by this test method to be 600 • 10-6 based on the Fig. 16: Expansive behavior of matrix mortar with WG under
accelerated curing after age of 33 days
linear elastic limit of tin plates, expansion rates beyond
this value cannot be absolutely evaluated. However, Figure Figure 18 shows the comparison of compressive strength
16 demonstrates that this material abruptly expands when of restrained expansion specimens containing Ex, NEDA
the ambient temprature rises. Figure 17 shows a risk of at the age of 33 days and WG at the age of 40 days afer
excessive expansion of this material. In order to use WG as accelerated curing from the age of 33 days shown in Figure
an expansive marerial, control of its expansion remains a 16 by the JCI method with respect to control specimens.
problem for future studies. The compressive strength under restrained expansion of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

406 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Prestress of Reinforced ASR-Expansive SHCC Beams and Possibility of Other Expansive Materials

Fig. 17: Expansive behavior of matrix mortar with WG under Fig. 18: Comparison of compressive strength of matrix mortar
accelerated curing after age of 2 days with different expansive material

the material containing NEDA, which continued to expand Acknowledgement

during the strength development phase of high-early-
This study was supported by Grants-in-Aid for Scientific
strength cement as shown in Figure 14, is comparable
Research of the Japan Society for the Promotion of Science
to the compressive strength of control specimens
(15K14014). The authors express their gratitude to Taiheiyo
and specimens containing Ex. On the other hand, the
Material Corporation for the supply of the granular non-
compressive strength obtained from the specimens with
explosive demolishing agent.
WG is lower than those with other materials.
References
This suggests, within the range of the present tests, 1. Imamoto, K., 2007. Specific surface areas and pore volumes
expansion of materials containing NEDA do not lead to of shrinking cementitious materials. Proceedings of the Japan
fracture similarly to Ex. It also suggests a possibility of Concrete Institute, 29(1):603-608. (in Japanese)
using SHCC containing NEDA as a cementitious material 2. Ishii, S., 2006. Technology of chemical admixtures for concrete.
that continues expansion even after hardening, without CMC Publishing. (in Japanese)
reaching failure and with no strength loss. 3. Iwatsuki, E., and Morino, K., 2009. Study of mechanism of ASR
using water glass cullet. Proceedings of the Japan Society of Civil
Engineers, 64(5):199-200. (in Japanese)
Conclusions
4. Kawamura, M., Takada, H., Asano, Y., YUN, H., and Rokugo, K., 2015.
The purpose of this contribution is to develop new Expansive behavior of concrete and SHCC containing water glass
cementitious materials that are capable of keeping cullet or non-explosive demolition-agent particles. Proceedings of
the Concrete Structure Scenarios, the Society of Materials Science,
expanding without shrinkage over a long time in a stable Japan, 15. (in Japanese) (submitted)
manner. SHCC that slowly expands over a long time
5. Takada, H., Takahashi, Y., Sakaguchi, Y., Kobayashi, K., and Rokugo,
incorporating ASR expansion was proposed. Chemical K., 2010. Control of cracking properties of steel bar-reinforced
prestress was introduced into this material by restraining SHCC beams by addition of large amount of expansive admixture.
the ASR expansion using reinforcing bars. Bending Proceedings of Symposium on Fracture and Damage of Advanced
Fibre-reinforced Cement-based Materials (ECF 18, Dresden), 43-50.
tests on the reinforced ASR-expansive SHCC beams
were carried out. ASR-expanded R/SHCC beams were 6. Takada, H., Rokugo, K., Tanabe, K., and Asano, Y., 2014. Cracking
Properties and Length Change Behaviour of SHCCs Utilizing ASR
chemically prestressed, and thus their cracking loads in Expansion. Proceedings of the 3rd International RILEM Conference
the flexural loading tests increased in proportion to the on Strain Hardening Cementitious Composites (SHCC3, Dordrecht),
degree of chemical prestress. 17-24.
7. Rokugo, K., Takada. H., Onda, Y., Fujishiro. M., Kobayashi, K. and
The expansive behavior of the matrix mortar (SHCC without Asano, Y., 2014. Chemical restressing of steel bar-reinforced
fibers) containing Ex (expansive additive), NEDA (non- concrete beams using SHCC that retains ASR expansion for a long
destructive demolition agent), or WG (water glass cullet) time. Proceedings of the 3rd International RILEM Conference on
was examined. There is a possibility that NEDA can be Strain Hardening Cementitious Composites (SHCC3, Dordrecht),
43-50.
used as an expansive material that impart expansiveness
to SHCC after cement hardening without damaging
it, provided the grading and content are appropriately
established. In order to use WG as an expansive material,
control of its expansion remains a problem for future
studies.

Organised by
India Chapter of American Concrete Institute 407
Session 4 A - Paper 4

Hiroo TAKADA
Department of Civil Engineering, Gifu University, JAPAN

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

408 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Factors Influencing the Bonding between Steel Fiber and Magnesium Phosphate Cement Mortars

Factors Influencing the Bonding between Steel Fiber and Magnesium

Phosphate Cement Mortars

Caijun Shi, NanYang, Zemei Wu, Jianming Yang,

YuanChang, Linlin Chong Department of Civil Engineering, Yancheng Institute of
College of Civil Engineering, Hunan University,
1 Technology, Yancheng 224051, China
Changsha 410082, China;

Abstract fiber with Portland cement matrix and proposed some

methods to improve the bonding, which includes decrease
This paper investigated the bonding between straight
of porosity in ITZ, plasma treatment and deformation
smooth steel fiber and magnesium phosphate cement
of fiber. Sun (Sun and James, 1986; Sun, 1987) studied
(MPC) mortars through double-sided fiber pullout testing.
the influence of water-to-cement (w/c) ratio, mixture
The effects of mineral admixtures such as limestone and
proportions, and the shape, embedded length and aspect
blast furnace slag powders on the bonding were also
studied. The results showed that the bonding between ratio of fiber on the bonding between steel fiber and
steel fiber and MPC mortar was much better than that Portland cement paste (PC). It was found that the bonding
with ordinary Portland cement (OPC) mortar due to its strength of interface and the enhancement effect of steel
higher strength and lower porosity of the interfacial fiber were strengthened by changing the thickness of
transition zone (ITZ) between fiber and the matrix. The water film and strength of the ITZ. Monterio (1989) and
bonding strength of straight fiber with MPC was 93% Gjorv (1990) reported that the addition of condensed
higher than that with OPC. The addition of slag into the silica fume could decrease the porosity in the ITZ due to
binder was beneficial to the bonding strength. From physical filling effects and pozzolanic reactions. Thus, the
scanning electron microscope observation, the addition bonding between fiber and matrix can be improve by: (1)
of limestone and slag could generate more cementitious changing the surface and shape of the fiber to strengthen
hydration products, densify the interfacial transition the mechanical interlocking between fiber and matrix; (2)
zone and improve the bonding strength due to filling and improving the strength of the matrix; and (3) densifing ITZ
chemical activity effects. by the addition of mineral admixtures.

Keywords: Phosphate cement-based materials; steel The good bonding is critical for the cement-based material
fiber; interface; bonding; mineral admixtures to achieve high flexural strength and ductility. Jean (1998)
found that the addition of fiber significantly increased the
toughness of cement-based materials. The use of E-glass,
Introduction polyester and metallic fibers could alter the failure
Magnesium phosphate cement (MPC) based materials mode of MPC. Ding (2005) investigated the influence of
have high early strength and good bonding with old glass and PVA fibers on compressive strength, load-
Portland concrete. Extensive researchers (Adbelrazig deflection response, modulus of rupture and toughness
et al., 1988; Ding et al., 2012; Hu et al 2014; Yang et al., index of MPC. Wang (2006) analyzed the effects of steel
2000; Yang et al. 2013; Shi et al. 2014) have studied fiber on mechanical properties, shrinkage and abrasion
their preparation, hydration and mechanical properties resistance of MPC. The results showed that the flexural
and durability. Recent publications reviewed the strength was increased by 43.5% when 1% steel fiber by
preparation and durability of MPC (Yang et al 2014; volume was added. The addition of fiber also reduced the
Chang et al 2014). However, like Portland cement-based shrinkage and improved abrasion resistance of MPC.
materials, MPC based materials are also brittle and Although extensive researches has been done on the
their ductility decreases with the increase of strength. interfacial bonding between fiber and PC, very limited
Consequently, it is necessarily to improve their tensile literature can be located on the interfacial bonding
strength and toughness for some applications. It has between fiber and MPC. As there are fundamental
been acknowledged that there is a water film between differences between MPC and PC in hydration
the steel fiber and Portland cement pastes, which allows mechanisms and microstructure, it is important to
ettringite and calcium hydroxide to precipitate and grow understand the development and improvement measures
orientally without constrain. This would not only increase of bonding properties. This paper investigated the bonding
the porosity in the interfacial transition zone (ITZ) but also properties of steel fiber with MPC pastes through fiber
decrease the bonding between fiber and C-S-H gel. Chan pull-out testing. The difference in bonding properties
(1994) systematically investigated the bonding of steel between fiber and MPC or OPC was compared. The effects

Organised by
India Chapter of American Concrete Institute 409
Session 4 A - Paper 5

of admixtures such as limestone and slag on the bonding for 90s. The mortar mixture was then poured into 40×40×40
characteristics were studied as well. Finally, the bonding mm3 cubic molds for compressive strength testing.
and reinforcement mechanisms of fiber were discussed.
The specimens for fiber pullout testing were prepared
according to the Chinese standard CECS 13:89. A dog bone
Raw Materials and Test Methods shape mold, as shown in Figure 1, was used to prepare
the specimens. The mold was divided into two halves by
Raw material 1mm thick plates. Four fibers were arranged evenly on the
MPCs derived from the mixture of dead burnt magnesia plates bridging the two halves of the specimen. In order to
(MgO), monopotassium phosphate (KH2PO4), borax ensure the fiber will be pullout in one end, the embedded
(Na2B4O7·10H2O), disodium hydrogen phosphate (DHP, length of fixed end and pullout end was carefully arranged
Na2HPO4·12H2O) and chloride based retarding agent (M). in 8mm and 5mm respectively. The mixed MPC mortar
The specific surface area of the MgO was 238m2/kg based was then poured into the mold for each batch.
on a previous study (Chang et al, 2013) and the particle The specimens were cured in a room at temperature of
size of the KH2PO4 ranged between 245~350μm. 20 °C and relative humidity of 50%, and demolded after 24
Limestone and slag powders were employed as mineral hours of casting. Then they were cured until testing ages
admixtures. Their replacement level was 15% of the mass of 3 and 2d.
of MgO. The MPC mortars with limestone and slag were
designated as L-M and S-M respectively. The chemical Testing method
composition of limestone and slag are shown in Table
Interfacial bonding strength
1. The composition of MPC mortars is listed in Table 2.
Grade 42.5 ordinary Portland cement was adopted as a The interfacial bond strength was determined according
reference. The ratio of cement: sand: water was 1:1:0.3. to the Chinese standard CECS 13: 89. The pullout testing
Straight smooth steel fiber with length of 13mm and setup is shown in Figure 2. The average bonding strength
diameter of 0.2mm were used. π was calculated as follows:
P
x = nrmax ....................................................................(1)
Preparation and curing Dl
Borax, DHP, M and water were poured into the mixer in Where Pmax is the maximum pullout load; n is the number
sequence and mixed for 60s at low speed. Sand was then of embedded fibers, which is 4; D is the fiber diameter (0.2
added and mixed for another 60s at low speed and 30s at high mm); and l is the embedment length of steel fiber (5 mm).
speed. At last, magnesia was added slowly and mixed for 90s The pullout energy is the work during the process of fiber
at low speed and paused for 30s before mixing at high speed pullout from the matrix, which is defined as the covered

Table 1
Chemical composition of magnesia, slag, limestone and Portland cement (wt%)

Material CaO SiO2 Al2O3 Fe2O3 SO3 MgO K 2O Na2O Na2Oeq

Magnesia 1.33 0.92 0.16 0.34 - 96.80 1.49 - -

Slag 39.11 33.00 13.91 0.82 - 10.04 1.91 - -

Limestone 53.76 0.92 0.55 0.06 - 1.08 - - -

Portland
63.12 22.51 5.32 3.78 2.20 2.56 0.71 0.19 0.66
cement

Table 2
Composition of MPC mortars (wt%)

Types of Sand / Water /

Batch MgO/% M/% Broax/% DHP/% Mixture/%
mixtures cement binder

MPC - 35.11 12.76 0.88 1.25 0 1:1 0.18

L-M Limestone 29.84 12.76 0.88 1.25 5.27 1:1 0.18

S-M Slag 29.84 12.76 0.88 1.25 5.27 1:1 0.18

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

410 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Factors Influencing the Bonding between Steel Fiber and Magnesium Phosphate Cement Mortars

1 -MPC paste; 2 -Plate; 3 - Steel fiber

Fig. 1: Specimen for fiber pullout testing Fig. 3: Fiber pullout curves of straight steel fiber from MPC
and OPC mortars
area of the pullout
curves with the axis.
Table 3
Bonding of straight fiber with different mortars
Scanning Electron
Microscope (SEM)
observation Pullout peak load Bonding strength Pullout energy
(N) (MPa) (N·mm)
The specimens for
microstructural
observation were cast OPC MPC OPC MPC OPC MPC
and cured in the same
condition as for strength 30.56 59.17 2.43 4.71 76.85 127.84
testingand dried for
24 hours at 60°C.
SEM observation of ITZ between fiber and matrix is
The microstructure
illustrated in Figure 4. The microstructure of the ITZ
of the hardened MPC
between steel fiber and cement matrix is shown in Figure
mortars was examined
5. Figure 5(a) is the microstructure for the pullout zone of
with QUANTA200
Fig. 2: Illustration of fiber pullout fiber and the dash line is the border of fiber and the matrix.
Environmental Scanning
testing setup Large quantity of hydration product MgKPO4·6H2O (MKP)
Electron Microscope
could be seen around the steel fiber at the distance of
(SEM) equipped with an
20μm. It was obvious that the MKP was well crystallized
energy dispersive spectrometer (EDS).
and existed in cluster distributions.
Figure 5(c) shows the microstructure of the area at the
Results and Discussion distance of 20μm to the fiber after magnified to 800 times.
Bonding of steel fiber with MPC and OPC mortars The structure of the surface of the fiber and the interface
of the matrix was dense and continuous, which was
The pullout curves of fiber from different matrixes favorable to the bonding strength between fiber and MPC
are shown in Figure 3. The pullout peak load, bonding matrix.
strength and pullout energy are summarized in Table
3. It can be seen that the bonding strength pulled out Figure 6 is the microstructure of fibers pullout from MPC
from MPC mortar was 93% higher than that from OPC and OPC mortars. As can be seen from Figure 6, some
mortar. hydration products were adhered to the surface of fiber
and were destroyed during the pullout process. Compared
According to the literature (Wang et al. 2006), the bonding Figures 6(a) with (b), it can be found that more hydration
strength of straight steel fiber with M70 Portland cement products adhered to the fiber pullout from MPC than those
mortar was 2.18 MPa, which is lower than that with MPC from OPC, which indicated that the hydration products of
mortar from this study. Besides that, the pullout energy MPC could grow and adhere to the surface of the fiber and
of fiber pullout from MPC mortar was larger than that results in stronger interlocking and adhesive bonding than
from OPC mortar. those of OPC.

Organised by
India Chapter of American Concrete Institute 411
Session 4 A - Paper 5

Bonding between fiber with mortars with different

mineral admixtures
The influences of mineral admixtures on bonding at 3
and 28 days are shown in Figure 7. From 3 to 28 days,
the development of bonding strength of the reference
MPC matrix (M-F) was not significant while the bonding
strength of specimens with limestone (L-F) and slag (S-F)
demonstrated rapid strength development. This indicated
that the addition of limestone and slag was favorable to the
bonding strength. The bonding strength of the specimens
with limestone and slag powders was 23% and 40%
greater than that of the reference at 3 days. The bonding
strength of the specimen with slag was 2.5 times greater
Fig. 4: Illustration of SEM observation of ITZ than that of the reference at 28 days.

(a) Fiber pullout zone (b) Interfacial transitional zone (c) Interfacial transitional zone

Fig. 5: Microstructure of the ITZ between fiber and matrixes after fiber pullout

(a) 3d

(a) Fiber pullout zone

(b) 28d

(b) Fiber pullout from OPC

Fig. 6: Microstructure of fibers pullout from Fig. 7: Pullout curves of fiber from MPC mortars at different ages
different MPC and OPC mortars

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

412 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Factors Influencing the Bonding between Steel Fiber and Magnesium Phosphate Cement Mortars

The ITZ between the fiber and matrix is the most vulnerable between the fiber and the matrix and the unreacted MgO
area in fiber reinforced cement based materials due to grain, and thus improving the packing density of the ITZ.
its higher water to binder ratio (w/b) and more porous
Figure 8 shows the microstructure of the holes after fiber
structure. The improvement of ITZ with addition of mineral
was pulled out from different matrixes. As shown in Figure
admixtures can be attributed to both physical and chemical 8(a), the ITZ of the matrix without mineral admixtures
aspects. The physical aspect includes particle size and was more porous and the area surrounding the hole
micro aggregate filling effects. After the addition of the was destroyed and fell down. Microcracks could also
mineral admixtures, the fluidity of the mortar was improved be observed along the radial direction as fiber centered
due to the small particle and filling effects of the mineral in other part of the matrix. As Figure 8(b) shows, the
admixtures. Besides, because of the water reduction hydration products growedand interlocked to the surface
effect, the w/b could be reduced to reach the same fluidity of the fiber. Besides, there were much more hydration
compared to the reference, which lead to the lower w/b products attached on the fiber in the specimen with slag
and less porous structure in the ITZ. On the other hand, than those in the reference. This suggests that the addition
the relative larger specific surface area of the mineral of the mineral admixtures could reduce the porosity of the
admixture could prevent the water gathering around the ITZ and improved the bonding strength between the fiber
fiber which in turn reduced the w/b and porosity in the ITZ. and the matrix. The surface of the fiber and the interface
Moreover, the mineral admixtures with small particle size of the matrix after the fiber was pulled out are shown in
could be well dispersed in the matrix and filled the gap Figures. 9(a) and (b) respectively.

(a) MPC matrix

(a) Surface of pullout fiber

(b) MPC Matrix with slag

(b) Surface of pullout matrix
Fig. 8: Microstructure of holes of fiber pullout from different
matrixes Fig. 9: Surface of fiber and matrix after fiber pullout

Organised by
India Chapter of American Concrete Institute 413
Session 4 A - Paper 5

Longitudinal scratches could be seen on the fiber surface Figure 11 is the XRD pattern of the specimens with and
due to the abrasion by the matrix during pullout process. without limestone. A new hydration product, CaHPO4, was
Similarly, the scratches could also be obviously seen in detected. These products, both amorphous and crystal,
the interface of the matrix. EDS analysis detected iron filled the gaps and the pores, and thus decreasing porosity
in the scratches, which indicated the good mechanical of the ITZ and strengthening the bonding.
interlocking and friction. The main composition of
limestone and blast furnace slag is CaO. Phosphate ion
could combine with Ca2+ to form calcium phosphate.
Besides, it could also react with other oxides like Al2O3,
Fe2O3 and MgO to produce Al2(H2PO4)3, Al3H14(PO4)8·14H2O
and Fe3H14(PO4)8·14H2O etc. (Boybay et al. 1983; Tomic
1983).
Figure 10 shows the morphology and EDS analyses of
hydration products of MPC with limestone. The crystal in
cluster is the main hydration product of MPC - MgKPO4·6H2O
(MKP). Elements, such as Ca, Si and Al etc. were detected
by EDS in the amorphous hydration product.

Fig. 11: X-ray diffraction patterns of specimens with and without

limestone

Mechanism of bonding and reinforcement of fiber

The good bonding of fiber with MPC can be attributed to
several reasons as follows. In fiber reinforced Portland
(a) Microstructure of hydration products
cement-based materials, the w/b in ITZ is relatively
higher than that in bulk matrix due to the excessive water
in the ITZ. Ettringite and calcium hydroxide crystallize and
grow without much restraint. The high porosity of ITZ that
riched in crystallized products can lead to a weak zone
around the fiber at certain distance.
Figure 12(a) illustrates the typical trend of micro-
hardness with distance to the surface of the fiber in
Portland cement based materials. As can be seen,
the micro-hardness adjacent to the fiber is relative
high. It then gradually decreases to the lowest point at
certain distance before it revives and finally reaches the
hardness of the matrix beyond some distance. A previous
(b) EDS analysis of area A
study (Sun and James, 1986; Sun, 1987) indicated that
the thickness of the interfacial transition zone between
the fiber and the cement matrix was 50~100μm, and
the distance from the fiber to the weakest ring in the
ITZ was 20~35μm.The weakest spot in the ITZ is not the
area closely against to the fiber, but the one with certain
distance due to the distribution of the ion concentration
in this area.
When a fiber is mixed with cement, water would surround
the fiber and form a film. In Portland cement, the quantity
of ion in this film is limited and the ions diffuse to the water
film in the sequence of chemical activity, i.e. Na+, K+, SO42-,
(c) EDS analysis of area B Al3+, Ca2+ and Si4+ etc. Whereas in MPC, abundant active
icon such as Na+, K+ and Mg2+ diffused to the film rapidly
Fig. 10: Morphology and EDS analyses of hydration products of when mixing and thus increase the ion concentration
MPC with limestone

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

414 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Factors Influencing the Bonding between Steel Fiber and Magnesium Phosphate Cement Mortars

Acknowledgement
The research was conducted at Hunan University, and
financially supported by the National Science Foundation
of China under project Nos. U1305243 and 51378196.

References
1. Abdelrazig, B., Sharp, J., El-Jazairi B. 1988. The chemical
composition of mortars made from magnesia- phosphate cement.
Cement and Concrete Research, 18(3): 415-425.
2. Boybay, M, Demirel T, Lee D-Y. 1982. Process of producing hydraulic
Fig. 12: Interfacial transition zone and the distribution of ion cement from fly ash. Washington, DC: US Patents.No. 4,328,037.
concentration 3. Chan Y.W., 1994. Fiber/cement bond property modification in relation
to interfacial microstructure. Ph.D dissertation, Civil Engineering,
Ann Arbor: The University of Michigan.
in the water film. Besides, the pH value of the MPC is
4. Chang Y., Shi C., Yang N., Yang J.M., 2013. Effect of fineness of MgO
relative lower. In the early stage of the hydration, the on properties of magnesium phosphatecement, Journal of The
acid environment can lead to iron ions in the steel fiber Chinese Ceramic Society, 41(4): 492-499.
dissolve and diffuse into the water film as shown in Figure 5. Ding Z., Dong B., Xing F., et al. 2012. Cementing mechanism of
12(b). As a consequence, the thickness of the water potassium phosphate based magnesium phosphate cement.
film is reduced compared to that in Portland cement. Ceramics International, 38(8): 6281-6288.
Microscopic examination has proved that the distance 6. Ding Z., 2005. Research of magnesium phosphosilicate cement.
from the fiber to the weakest spot is about 10μmin the Ph.D dissertation, Hong Kong: Hong Kong University of Science
and Technology.
MPC matrix.
7. Gjorv O.E., Monteiro P.J., Metha P.K., 1990. Effect of condensed
Actually, soluble phosphates are often applied for the silica fume on the steel-concrete bond. ACI Materials Journal, 87(6).
surface treatment of steel to prevent the steel from 8. Hu Z.L., Shi C, Yang N. et al, 2014. Factors influencing setting times
corrosion (Huang, 1989) because MPC could result in an of magnesium phosphate cements by bayesian network, Journal of
impermeable iron phosphate film at the surface of the the Chinese Ceramic Society, 42(8): 38-44.
steel fiber (Yang et al., 2013). The pH value increased 9. Huang Y.C. 1989. Metal corrosion and protection principles.
at later stages and the dense passive protection film Shanghai: Shanghai Jiaotong University Press.
was generated on the surface of the steel fiber due to 10. Monteiro P J, Gjorv O, Mehta P K., 1989. Effect of condensed silica
the phosphorization and inactivation process, which in fume on the steel-cement paste transition zone. Cement and
Concrete Research, 19(1): 114-123.
turn improves the density and continuity of the ITZ, and
reduces the micro-defects and cracks. Through the SEM 11. Pera J, Ambroise J., 1998. Fiber-reinforced magnesia-phosphate
cement composites for rapid repair. Cement and Concrete
observation, the microstructure of the surface of the Composites, 20(1): 31-39.
fiber and the interface of the MPC matrix is dense and
12. Sun W., Mandel J.A.,and Said S., 1986. Study of the interface strength
continuous, with large quantity of well crystallized MKP in steel fiber-reinforced cement- based composites. ACI journal,
detected in cluster growth. 83(4): 597-605.
13. Sun Wei. 1987. The effect of silica fume and polymer on the interface
layer of between fiber and cement- based material. Journal of The
Conclusions Chinese Ceramic Society, 15(6): 6.
Through the fiber pullout testing and the SEM observations 14. Tomic, E.A., 1983. Phosphate cement and mortar. Washington, DC:
of the interfacial transition zone, the bonding properties U.S. Patent No. 4,394,174. 19.
between fiber with OPC and MPC matrix with different 15. Wang H.T., Qian J.S., Cao J.H., 2006. Properties and application
admixtures were studied. The bonding mechanisms of of steel fiber reinforced magnesia phosphate cement mortar.
the fiber with MPC were proposed. Specific findings of this Archetecture technology, 37(6): 462-464.
research include the following: 16. Yang Q.B., Zhu B.R., Zhang S.Q., et al. 2000. Properties and
applications of magnesia–phosphate cement mortar for rapid repair
(1) Compared to ordinary Portland cement, the bonding of concrete. Cement and Concrete Research, 30(11): 1807-1813.
properties of fiber with MPC was much more favorable 17. Yang J.M., Shi C, Chang Y. and Yang N., 2013. Hydration and
than that of OPC. The bonding strength of straight fiber Hardening Characteristics of Magnesium Phosphate Cement Pastes
with MPC was 93% higher than that with OPC. Containing Composite Retarders, Journal of Building Materials
(Chinese), 16(01): 43-49.
(2) The addition of admixtures such as limestone and the 18. Shi C., Yang J.M., Yang N, et al., 2014. Effect of waterglass on water
slag could generated more cementitious hydration stability of potassium magnesium phosphate cement paste, Cement
products, densify the interfacial transition zone and Concrete Composites, 53, 83-87.
and improve the bonding strength due to filling and 19. Yang N., Shi C., Yang J.M., et al. 2014. A Review on Research
chemical activity effects. Progresses in Magnesium Phosphate Cement-based Materials.
Journal of Materials in Civil Engineering, 26(10):04014071-1 -
04014071-8.

Organised by
India Chapter of American Concrete Institute 415
Session 4 A - Paper 5

Caijun Shi, Ph.D., P. Eng., FACI

Affiliation: College of Civil Engineering, Hunan University, Changsha, China
cshi@hnu.edu.cn
Dr. Caijun Shi received his B. Eng and M. Eng from Southeast University, Nanjing, China and his Ph.D from
University of Calgary, Canada. He is currently a Chair Professor of College of Civil Engineering, Hunan
University and China Building Materials Academy, Editor-in-Chief of Journal of Sustainable Cement-based
Materials, editorial board member of Cement and Concrete Research, Cement and Concrete Composites,
Materiales de Construccion, Journal of Chinese Ceramic Society, Journal of Building Materials, Cement,
and former associate editor of Journal of Materials in Civil Engineering. Dr. Shi serves on many ACI and
RILEM technical committees.
His research interests include characterization and utilization of industrial by-products and waste
materials, design and testing of cement and concrete materials, development and evaluation of cement
additives and concrete admixtures, and solid and hazardous waste management. He has developed several
novel technologies and products, and has been granted four US patents and more 15 Chinese patents.
One of his inventions - self-sealing/self-healing barrier has been used as a municipal landfill liner in the
world's largest landfill site in South Korea. He has authored/coauthored more than 220 technical papers,
five English books, two Chinese books and edited/co-edited five international conference proceedings.
Dr. Shi has been invited to give presentations on a variety of topics all over the world. In recognizing his
contributions to researches in waste management and concrete technology, he was elected as a fellow of
International Energy Foundation in 2001, and a fellow of American Concrete Institute in 2007.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

416 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
Session 4 B
Session 4 B - Paper 1

Preserving the life of infrastructure through effective monitoring

and intervention
Professor Peter Robery FREng
Visiting Professor, the University of Leeds, University of Birmingham, UK
Director, Robery Forensic Engineering Ltd, UK

Abstract combination to minimise cost and reduce impact for road

Asset owners require effective management of their users. The latest developments in this field include LiDAR
infrastructure over the service life to deliver best value point cloud surveys to capture automatically all relevant
return on investment, through an asset management data along a route, so that features such as white lining,
plan. For many new assets, this is realized through the stud reflectors, road signs and street furniture can be
design and construction conforming to the applicable added to the inventory.
national codes and standards. The asset management This paper sets out some of the current and developing
plan therefore principally deals with facilities or approaches that can be used to link monitoring techniques
operations management and ensuring planned preventive into an asset management strategy that helps ensure our
maintenance is carried out. infrastructure remains safe and reliable.
Increasingly, new infrastructure is being built in Keywords: Asset Management, Building Information
environments that are outside of the scope of codes and Modelling, Deterioration Modelling, Maintenance,
standards. This is for example, either because the design Monitoring, Whole-life Costing.
life is beyond the normal 100 years found in codes, or the
durability risks from the environment are far more severe
than are covered by the codes (or both). Examples include Introduction
the Singapore Mass Rapid Transit System, the Bahrain- Historically, concrete infrastructure has performed
Qatar Causeway and the latest: the Hong Kong – Zhuhai variably in service: some assets have given long years
– Macau crossing. For new assets, Building Information of trouble-free service; others have required significant
Modelling is giving us the opportunity to deliver asset unplanned investment in repair and maintenance to
data models effectively and efficiently that can be used enable them to achieve their design life; worse still some
to develop the asset management plan over the intended have had to be replaced prematurely, well before their
service life. intended design life, because it was uneconomic to repair
them (Robery, 2008).
Older infrastructure presents different challenges,
because most of it will have been built to a durability level Kept away from contamination by seawater or de-icing
that would be considered inadequate by today’s standards. salt, much of our well-built infrastructure has produced
For our older infrastructure the challenges that need to be elegant and long-lasting buildings and civil engineering
addressed include: gaining a fundamental understanding structures. The 40th annual Concrete Society Awards
of the numbers of assets in the infrastructure portfolio, Dinner in the UK in 2006 celebrated by holding a special
including location, dimensions and type; identifying the “Outstanding Structures” award for concrete assets
condition of each asset through a programme of inspection built since 1968 that had stood the test of time; Figure 1
and testing; monitoring changes in the condition of the illustrates the award-winning cathedral at Clifton, Bristol,
asset over time; a valuation for each asset for cost-based built 40 years ago and still performing well (Robery, 2008).
maintenance decision-support and an understanding of
the level of service expected. In Europe and North America, much of the transport
infrastructure was built in the 1960s and 1970s, using this
Some countries and some industry sectors are more “easy to use” material called concrete that is mouldable to
familiar with these concepts than others. In the UK, the complex shapes and patterns and was thought to last for
newly formed Highways England has had a structures ever without any maintenance. Unfortunately the lack of
management information system in place for nearly 20 general understanding about the importance of design for
years, which has data on all motorway and trunk road durability, low water/cement ratio, the three “Cs” of cover,
structures and the condition assessment process is fed by compaction and curing and the effect chloride salts would
uploads of regular inspection reports that inform decisions have in causing reinforcement corrosion, led to many
about the most cost effective maintenance and scheme failures early in the life of bridges. With this background

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

418 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preserving the life of infrastructure through effective monitoring and intervention

not until BS 8110 was released in the UK that the first

code-based references for general building construction
referred to “Design for durability” (BSI, 1985).
European design approaches and construction techniques
from the 1970s were also being exported around the world
to satisfy the demand for infrastructure, resulting in
concrete structures being built in the Middle and Far East
that had the same lack of design for durability, but also
combined with less suitable cements and aggregates and
even brackish water to make the concrete. The structures
were then subjected to harsh environments, far worse than
northern European conditions, creating problems such as
Fig. 1: Cathedral Church of Saints Peter & Paul, Bristol, UK, the Dubai Dry Dock deterioration and driving the design of
built 1974 (Robery 2008) corrosion mitigation measures such as cathodic protection
applied to reinforced concrete (Pynn, 2004). Considerable
it is understandable perhaps that the structures that difficulties were experienced with a particularly difficult
suffered the most were those that interfaced with the aspect of concrete deterioration, termed the “hollow
car, such as bridges and car parks, where de-icing salts leg” effect in the UK, where rapid evaporation from the
were used to deal with freezing road conditions without inside surface of hollow concrete tunnel structures was
realising that the chloride ions would penetrate through concentrating the chloride salts in brackish groundwater
the cover zone to the reinforcement and cause severe to produce highly elevated levels in the inside cover zone
galvanic corrosion. of the reinforced concrete tunnel segments. This afflicted
the new Hong Kong Mass Transit Railway that opened in
With good attention to architectural details and proper 1979, causing corrosion of reinforcement and spalling
consideration of exposure conditions, we can achieve into the tunnel. These lessons were fed into the design of
excellent weathering characteristics and create durable the first phase of the new Singapore Mass Rapid Transit
infrastructure that is a credit to the construction industry, system (Doran, 1987).
but we have had to learn a very expensive lesson.
Gradually our approach to durability design has been
improved, with measures that include setting minimum
History of Durability Design limits on cement content, cement type, cover and water/
The problems caused by one of the most common cement ratio and these durability provisions were
ionic compounds on the planet, common salt, have progressively tightened through British and European
cost the global economy countless billions of dollars standards such as BS EN 206 (BSI, 2000) and BS 8500
for repair, unplanned maintenance and re-building of (BSI, 2006) and are now incorporated into BS EN 1991, a
infrastructure. fully integrated suite of design codes for buildings and
for bridges (BSI, 2005). For general applications with
The first attempts at improving the UK’s approach to
a normal design life of either 50 or 100 years, the latest
durability came through CP110 (BSI, 1972) but even
standards are expected to provide durable structures in
this standard still permitted to add calcium chloride to
normal European exposure conditions.
reinforced concrete as a set accelerator, so distributing
chloride throughout the concrete matrix and it was not But much of the infrastructure required today is not
until 1977 that a revised edition of CP110 banned its use in in “normal” European climatic exposure and so these
structural concrete (CIRIA, 1987). The UK’s Department of structures are outside of the latest code provisions.
Transport then produced its own specifications for bridges Does that mean we stop building assets in extreme
such as BS 5400 (BSI, 1978) with a target life of 120 years. environments and keep our infrastructure out of the sea?
No, of course not: the structures currently being built
The full realisation of the poor condition of the UK
across the Pearl River Estuary to create the 29.6km long
bridge network began to become apparent, with reports
bridge and tunnel that comprises the HKZM Crossing
such as the 200 bridge study being highly critical of the
(Figure 2), are the latest testament to how much more
standard of construction and maintenance (DOT, 1989),
confident the industry has become about asset durability
the Department of Transport (now “Highways England”)
(Robery, 2014). The 6-lane road crossing comprises
began to introduce new design requirements in its Design
22.9km of steel/concrete composite bridge and 6.7km
Manual for Roads and Bridges (DOT, 1992), including the
of immersed tube tunnel, with advanced monitoring and
waterproofing concrete bridge decks and the treating
corrosion control measures built into it.
concrete surfaces in with hydrophobic impregnation BD
43/90 (DOT, 1990) as methods to slow down the rate of We learned over the past 20 years that we cannot build
penetration of de-icing salts into the cover zone. It was and forget our infrastructure, particularly where it is in

Organised by
India Chapter of American Concrete Institute 419
Session 4 B - Paper 1

organisation. Sudden, unexpected reduction in the value

of the infrastructure, due to impairment or previously
undetected deterioration or overload, can challenge the
growth of a company owner or even a country when we
consider Government- owned assets. Equally, investing
in purchasing assets that have latent defects, which
will develop into problems later, will also be costly and
unexpected. The value of the asset should be depreciated
over time and with good management will achieve its
Fig. 2: Artist’s impression of the Hong Kong-Zhuhai-Macau expected life and will maintain a predictable value. If
bridge across the Pearl River Estuary (HKIE, 2015) there is impairment, such as the discovery of corrosion
inside post-tensioning ducts that had otherwise no
an extreme exposure environment outside of codes and outward signs of problem, the use of the asset may
standards. We now know and accept we must actively be affected (e.g. load restrictions required) or the life
monitor and protect our infrastructure over the service shortened.
life to preserve the value of the asset and prevent the
Figure 3 illustrates the financial implications for an asset
sudden discovery of deterioration problems that were not
that is suddenly found to be impaired many years after it
foreseen. We also know that most of our infrastructure
was acquired (without a structural survey). The book value
does not really have a design life and must be designed
and built to be maintained and upgraded for an indefinite at a point in time would be its assumed value, but now, it
life. has an actual depreciation that accounts for a shorter life
than that expected, which will require a substantial re-
To manage our assets effectively and preserve their adjustment on the company’s balance sheet.
value, we need an effective plan to monitor changes and
take proactive steps to address deficiencies, using tools
and techniques that give early warning of problems and Linear Asset Depreciation based on Book Value, with Impairment
100%
thereby help owners make informed finance-led decisions Acquired
Problems
as to the best time to intervene. without a
full survey Discovered
Asset Value

Who Needs Asset Management Systems? Impairment:

Shorter life
has a
Assumed value
The answer is that we all need asset management substantial
financial
systems and we cannot afford not to invest! The technical Actual value impact on
corporate
press and Government reports on the state of our global True
accounts

infrastructure regularly refer to cash shortfalls for both Depreciation

new-build and maintenance programmes that have

created a massive funding backlog. In the USA for example, 0%
Revised Expected

a recently published government report (ASCE, 2013) Life Life

identified US$3.6 trillion was needed by 2020 to extend,

Fig. 3: Impact of buying trouble, as the asset is found to have a
replace and preserve the existing infrastructure stock. 59 Presentation title - Presenter's name

shorter life than expected (Robery 2014)

A CH2M HILL COMPANY

Without a sound fiscal approach to the management of

infrastructure, it is difficult to justify to Governments and
private sector that investment is needed for repair.
Asset Management Solutions
Solutions have to be sought that deploy technologies and
Investment is committed over many decades to create materials to enhance lifespan and improve predictability.
our infrastructure and the value of each asset makes up The driving factors for more effective asset management
a large part of the balance-sheet for the owner, be this as set out in the UK’s PAS 55-1 include (BSI, 2008):
Government or private sector owners. Deterioration
of assets and the maintenance regime necessary to • Demand for greater capacity in mature and immature
preserve and maintain them, will inform decisions about markets
prioritisation and method – in short is it more economic to • Increased interest in resilience, carbon and
demolish or to repair. sustainability
Assets are normally assigned a value based on their age • Developing markets do not want to repeat the historic
and so accurate valuation and allowance for predictable mistakes:
future maintenance are important for fiscal compliance
of any business (Robery, 2006). The asset value is • Seeking comprehensive strategies from authoritative
normally reported as part of any financial report on an organizations

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

420 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preserving the life of infrastructure through effective monitoring and intervention

Asset-rich clients need assistance to find the right

balance between performance, risk and cost over the ASSET MANAGEMENT

life of the asset, usually split into capital expenditure • Identify Need
(CapEx) for the planned replacement of serviceable items •
•
•
Outline design
Business case
Funding

that have a known limited life (e.g. bridge bearings) and

• Economic modelling
• Life-cycle cost modelling
• Performance standards

operational expenditure (OpEx) that would include incident Operate

•Inspect & assess
•Maintain

management, asset inspection and routine maintenance. Transfer/Re-

finance
•Preserve
•Rehabilitate
•Regulation/performanc
•Business case
e standards
Operate •Technical audit/due-

Asset management has now become a professional

diligence •Automate
•Inspect & assess (BIM/GIS/Geospatial
•Financing
•Maintain
Construct •Equity/loan

discipline that links together the original CapEx of

•Preserve
•Build •Rehabilitate • Decommission
•Supervise •Regulation/performanc • Business case
• Manage e standards

conception, design, construction, along with OpEx

• Replacement need
•Advise •Automate • Demolition works
Procure •Commission
(BIM/GIS/Geospatial)
•Contract strategy

and planned maintenance CapEx over the life of the •Detailed design
•Technical input
•Program management
OPERATIONS MANAGEMENT

asset. Asset Management is defined in ISO 55,000 as

the coordinated activity of an organisation to realise
19 Presentation title - Presenter's name A CH2M HILL COMPANY

Fig. 4: Asset Management and Operation over the Service Life

value from its assets (ISO, 2014); in other words
(Robery, 2014)
asset management is primarily concerned with the
preservation of asset performance at an acceptable
the asset (e.g. as shown in Figure 4). It is very important
level of current and long-term risk and cost for the
that there are reliable records to show the current
benefit of users. The term can be applied equally to civil
condition of the asset, to confirm it has been regularly
infrastructure or buildings, but currently it is most widely
inspected and that all backlog maintenance has been
applied to infrastructure, arguably because these assets
are most at risk of failure in service and need effective carried out, so demonstrating its true value. A second
management to preserve them. It is a whole life strategy and more difficult question with PPP & DBFO is the asset
that starts with the concept, follows through the design handback condition at the end of concession period, when
and construction and then sets in place a strategy for the the asset returns to a government ownership allowing for
safe operation of the asset over its service life. de wear and tear.

Asset Operation concerns the day-to-day operational Both externalised asset operation and DBFO-type
activities necessary to support asset users, including contracts will commonly use a comprehensive asset
maintenance, and the delivery of the activities identified management system that provides a list of the components
through the asset management strategy. The relationship of each asset and contains the maintenance records and
between asset management and asset operation is forward workload. A proper life-cycle Asset Management
illustrated in Figure 4. Transportation infrastructure system includes:
comprises a network of generally horizontal (linear)
• Asset inventory, input into an asset management
assets that may include road pavements, bridges and
system
tunnels, each component having its own requirements for
ongoing asset condition assessment, risk assessment, • Asset management plan
routine maintenance, preservation works, incident
• Process assessment and improvement to the
management and planned component replacement. In the
framework of ISO 55,000
building sector the operation of “vertical” assets is more
commonly called Facilities Management, but typically • Whole life cost valuing, modelling and budgeting,
need less proactive asset management tools as neither including future depreciation rates with ageing and
the exposure environment nor live loading levels are performance prediction
sufficient to justify this cost, coupled with less stringent
Regulation and lower risk of failure. • Asset condition survey, monitoring and assessment,
including evaluating present condition and residual
New Assets value, particularly with a change in owner
Using private finance has been a very successful way • Developing life extension options and estimating the
of creating new infrastructure, but it does have its costs to preserve/extend the life of the asset (future-
own particular problems. Public Private Partnerships proofing the assets)
(PPP) and Design Build Finance and Operate (DBFO)
• Risk/criticality assessment and prioritization,
schemes are used increasingly by those requiring new
including impairment and/or loss of service function
infrastructure, often with a long concession period of
25 years or more, during which time the cost of the • Performance-based maintenance procurement/
construction and operation is paid back to a consortium. monitoring
Once the asset has been built, sometimes the composition
of the consortium changes or re- financing is needed by • Asset operation and maintenance delivery
banks who will need to establish the current condition of • Decommissioning and replacement

Organised by
India Chapter of American Concrete Institute 421
Session 4 B - Paper 1

Some of the components above are explored in the

following sections.

Existing Assets
For large and complex infrastructure, asset operation is
increasingly operated by external companies on behalf of
the owner to meet challenges such as strict government
Regulation, high volume use, deterioration risks, high
structural loads and high profile consequences from
failure. In the UK, the Department of Transport initially
appointed Managing Agents and Term Maintenance
Contractors on a 5 year contract term to look after the
trunk roads in England, Wales and Scotland in the 1990s,
splitting England into 24 contract areas. By 2000, joint Fig. 5: LiDAR data demonstrating the various features that can
be detected and interrogated (Robery, 2014)
venture companies were invited by England’s Highways
Agency to bid for new “MAC” contracts (Managing Agent
Contractor) that provided a single organisation to operate This technology has considerable promise for asset
the network and the number of areas has slowly reduced inventory collection in the future, allowing structures to
to the current list of 14; a similar scheme operates in be scanned, identified and dimensioned to provide the
Scotland (by Transport Scotland). These led to a modified backbone of the asset management system. In the UK, the
form of the MAC contract in England called the Asset Highways Agency has awarded contracts to scan its road
Support Contract (ASC) being awarded in the late 2000s network at normal traffic speed, to collect data on routine
(NCE, 2015), with the overall standards of the network maintenance items, such as deteriorating road lining,
now overseen by Highways England. reflective studs, the positioning and condition of signage,
as illustrated in Figure 5 (Highways Agency, 2009).
For older built assets, the challenge is to understand
what comprises the “network” of assets. Collection of LiDAR scanning technology also has the potential to
construction design and build data has not happened provide comparisons between scans, to monitor changes
consistently and reliably over the years, with paper records such as damage, theft or movement of components of the
and photographs providing the only evidence of what has infrastructure network.
been built. Worse still is the backlog of maintenance that
may exist, where only limited information is available on Asset Management Ready
current condition. To be asset management ready, new infrastructure must
Experience shows that while there is a wealth of digital be built to the quality and standards required by the design
asset management systems available commercially to (and any agreed changes), through proper management
carry out all manner of routine tasks, from an inspection of the construction process. Implicit is that the design
report repository to a works ordering system, the costs makes the infrastructure fit for its intended purpose.
are not in the software itself. Rather, it is in the time spent Construction handover should also be completed with
collecting basic data about the geometry and components proper management documentation, including O&M
of the assets and their condition and then inputting this manuals, details of consumables and spares, safety
into the computer system. These costs are hindering the certification on materials, and a briefing on the use of
implementation of asset management at local government operational systems.
level in the UK and elsewhere. All construction and component details should be
Considerable advances are now being made in uploaded into an asset inventory as part of the asset
visualisation technologies, allowing built assets to be management system, ready to drive forward with asset
scanned and the resulting point cloud of data interpreted operation. Effective asset management needs appropriate
for input into asset management systems. The data- and reliable information about assets, as it supports
rich video imagery and LiDAR (Light Detection And decision making, planning and execution of activities on
Ranging) point clouds affords opportunities for a wide infrastructure (ICE, 2014).
range of mobile asset data capture applications. LiDAR For new build schemes, digital data design and CAD
data is very accurate, high resolution 3D data, captured modelling is commonplace, but even today this data is
using special sensors, from the air or the ground, to traditionally not recorded with sufficient detail on the
produce a set of “dots” suspended in a 3- dimensional component parts and their life expectancy to be of direct
space. These dots can be displayed in special software use in the asset inventory and no simple upload conversion
or converted into a 3D mesh for use in many modern 3D exists. As a result, many large and complex structures
software packages. are built with no asset inventory at all; asset owners have

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

422 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preserving the life of infrastructure through effective monitoring and intervention

Asset Monitoring
Changes in the condition of infrastructure can be
caused by a wide variety of circumstances. Many tools
are available today that network and communicate
instantly, across continents if necessary, to relay real
time information about asset condition. Computer-aided
structural health monitoring has been used increasingly
over the last 20 years to describe a range of systems
implemented on full-scale civil infrastructure to assist
and inform operators about continued ‘fitness for purpose’
of the assets (Brownjohn, 2007). Sensors can detect: load
behaviour, such as defections, dynamic response, strain
and inclination; environmental exposure, including wind
Fig. 6: Bond Street station complex BIM model (Vernikos, 2012) speed, seismic effects, precipitation and temperature;
and deterioration, such as the effect of sea salt, causing
been known to commission asset surveys immediately resistivity, half-cell and linear polarisation changes.
after construction to measure and record what has been
built at the correct level of detail for entering into an asset Assurance
management system.
Monitoring of heavily loaded infrastructure such as
Recently, the design and construction process has been bridges has been used for decades to examine changes
better aligned through the increasing use of project in the response to known loads. Small, subtle changes
Building Information Modelling (BIM). This allows not only in the response of the element, caused by deterioration,
for fast and efficient design changes and the recording of fatigue or overload, may affect the structural reliability of
these changes, but also trouble-free construction with the asset and for example prompt load or lane restrictions
best fit and clash avoidance and concept 3-D visualisation on a bridge.
“fly-through” as shown in Figure 6 (Vernikos, 2012). The UK
Environmental conditions can also lead to overload and
Government estimated recently that if BIM was extended to
failure of components, particularly in areas subject to
all major projects, then between £1 billion and £2.5 billion
typhoons, and the technologies have advanced faster in
per annum savings can be realised in the construction
the Far East in general and Hong Kong in particular, as
phase alone, with potentially even greater savings in the
a result of the challenging and impressive infrastructure
post-construction in-service phase (Pocock, 2014).
that has been built over the past two decades.
Once complete and built, the model and all of its embedded
Data capture can be by static probes or by mobile
component details can be uploaded into an asset
inspectors, tying the data instantly into the asset inventory
management system to form the inventory and allow the
and updating the condition on road surfacing, drainage
whole of life decision making process to begin. Therefore,
and other static features.
the move to BIM models has the potential to provide all
the necessary details when information is sought during Infrastructure owners can now call up information on
operation of the infrastructure (Pocock, 2014). However, their portfolio “dashboard” to confirm that all quality
while BIM provides an obvious route for the lifecycle assurance checks are complete, inspections carried out
information requirement of assets, a common standard and value preserved.
needs to be in place for the model and issues of ownership
of the BIM data model need to be resolved at the outset to Deterioration
allow it to be used during the operation stage. Gradual (undetected) impairment will commonly occur
The BIM technology also lends itself to more effective as a result of environmental exposure conditions, such
offsite construction through better design coordination and as penetration of chloride salts into the fabric of the
avoidance of late changes. Factory-built quality-assured infrastructure. Sources of chloride can be natural, such
components manufactured to the highest standards can as seawater splash, spray or intertidal immersion,
then be brought to site for assembly (Vernikos, 2014). While or induced, such as by de-icing salt exposure that is
offsite construction is not news for the building sector, the commonplace closer to the geographic poles; contractors
increasing adoption in civil engineering infrastructure is even used to add calcium chloride to the concrete as a set
gaining more attention. This is particularly the case for accelerator. Salt also accelerates deterioration of paint
the refurbishment of existing assets, such as that shown systems on steelwork, and in the UK led to the expression
in Figure 6, where for efficiency, the largest possible “painting the Forth Bridge” – as soon as the contractors
prefabricated components need to be brought through finished painting it from one end to the other, they had to
narrow entrances and bends. start over again.

Organised by
India Chapter of American Concrete Institute 423
Session 4 B - Paper 1

Even the air we breathe out is harmful to reinforced

concrete, as the carbon dioxide in the air will slowly
Optimised

Total cost (£/year)

intervention timing Cost/year of maintenance

neutralise the alkaline cement paste of concrete. Once

(reduces as it becomes less frequent)

the protection to the reinforcement is lost, corrosion of

the reinforcement can begin, leading to either expansion Total risk cost /year increases

and spalling, or more worrying, undetected section loss,

(as hazard becomes more likely)

Lost revenue from

with a possible risk of collapse. Programmes of careful lost availability
Direct and management cost
inspection and assessment are required to detect these of interventions
Reputation damage
problems and allow appropriate intervention to be made. Maintenance cycle (years)
Accidents
Environmental damage

Impairment
Fig. 8: Optimisation of maintenance works to yield minimum
Sudden impairment due to overload most commonly
total cost over the life of the asset (Robery, 2014)
occurs as a result of unplanned impact, from a ship or 31 1 October 2012 Presentation title - Presenter's name A CH2M HILL COMPANY

vehicle. One of the most common causes of over-bridge

This evidence-based approach should be compared with
impairment with transport infrastructure is impact from
the more traditional approach to managing infrastructure:
an over-height vehicle passing beneath, which damages
the “unplanned maintenance” approach. This is where no
the soffit. As illustrated in Figure 7, some bridges are more
detailed strategy exists and it is assumed the asset will
unlucky than others (Skagit Co, 2004). All such sudden and
last indefinitely with no maintenance. When problems
unexpected damage must be detected early, using sensors
are suddenly detected and reported, they may have to
such as accelerometers, and the repaired as an exceptional
be remedied by a “fire-fighting” approach, dipping into
one-off cost, impacting on the asset’s value.
financial reserves and affecting the asset value and the
corporate balance sheet.
Ultimately we are comparing investment in an asset
management system from construction, to operate and
maintain the asset and to assure the owner risks are
managed, with a high risk strategy that undetected and
unmonitored deterioration is taking place and waiting
for the emergency call to undertake the “unplanned”
reactive maintenance.

Fig. 7: 2nd St Bridge, Mount Vernon, USA struck 3 times in 2004 Conclusion
(Skagit Co, 2004) This paper summarises how effective management of
infrastructure over the whole operational life can deliver
Impairment can also be the identification of deterioration best value return on investment for owners of existing
that was not previously known or recorded, and the assets in both Government and private sector, as well
finance required to carry out unplanned repairs will also as for PPP & DBFO projects. It requires a strategy that
impact on the asset value (Figure 3). turns the as-built design and construction details into a
detailed asset management plan, operated over the life of
Asset Maintenance the asset.
The development of asset lifecycle plans and value 1. Engineers tasked with managing strategic
engineering of intervention options will deliver an asset infrastructure need a thorough understanding of
preservation strategy with whole life cost comparisons. asset management plans and objectives to engage
Financial decisions can then be made based on fact- effectively in this field.
based evidence of the cost-effectiveness and return on 2. Asset management practitioners need assistance
investment from the works programme, using a life cycle from engineers and specialists to calibrate asset
costing analysis. A build-up of estimated costs can then management plans and so be able to predict accurately
be prepared for the different items and adjustments the future deterioration rates in the exposure
made to variable items. Analysis of these will show the environments and identify when significant risks exist
effect not only on the expenditure in any given year, but that threaten the structural capacity and asset value if
also the costing of different strategies, such as carrying intervention maintenance is not carried out.
out the bare minimum of works, sufficient to ensure
the asset reaches its intended Target Life, and delaying 3. The asset management plan will help define the point
unnecessary works or not undertaking any works at all. in time when it will be cheaper to demolish and replace
From this analysis, the optimised timing can be identified the infrastructure, rather than expensively retain and
for carrying out the works (Figure 8). repair, in whole life value terms.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

424 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Preserving the life of infrastructure through effective monitoring and intervention

4. A good asset management plan comprises a 12. DOT, 1990. Criteria and materials for the impregnation of concrete
highway structures. BD 43/90, Department of Transport, London
database inventory, inspection reports, operation
UK, 1990 (current version is BD 43/03, 2003 at http://www.
and maintenance plans, a valuation for each asset standardsforhighways.co.uk/dmrb/vol2/section4/bd4303.pdf,
for cost-based maintenance decision- support and an viewed 4 June 2015).
understanding of the level of service expected. 13. DOT, 1992. Design manual for roads and bridges (DMRB). Department
of Transport, London, UK. DMRB, 1992 (current version at http://
5. Asset management plans and accurate asset valuation www.standardsforhighways.co.uk/dmrb/index.htm , viewed 4 June
have become common requirements for banks and 2015).
other financiers of DBFO schemes in particular, 14. Doran, S. R., et al, 1987. Corrosion protection to buried structures.
to demonstrate their financial investment is being Proc. Singapore Mass Rapid Transit Conference ‘87, Construction
protected and Building Materials, 1(2): 18-26
15. Highways Agency, 2009. Project 512935: Development of traffic speed
6. Externalisation of the management of transportation LiDAR Phase 1 & 2, UK: http://www.highways.gov.uk/knowledge/
infrastructure is set to continue and with new projects/development-of-traffic-speed-lidar-phase-1-2/, viewed 1
developments such as BIM for new construction Sept 2014, London UK.
and LiDAR to capture existing asset information and 16. HKIE, 2015. Bridging the Pearl River Delta. J Hong Kong Institution
condition, the process of populating asset management of Engineers, Hong Kong. Available at http://www.hkengineer.org.
systems is becoming increasingly affordable and hk/program/home/articlelist.php?cat=cover&volid=116, viewed 4
June 2015.
quick.
17. ICE, 2014. Realising a World Class Infrastructure – ICE’s Guiding
Principles of Asset Management, Institution of Civil Engineers,
References London, UK, 26 June 2014.
1. ASCE, 2013. 2013 Report Card for America’s Infrastructure – 18. ISO, 2014. Asset management - Over view, principles and
Executive summary, American Society for Civil Engineering Report, terminology, International Organisation for Standardisation, Geneva.
USA: http://www.infrastructurereportcard.org/a/browser-options/ BS EN ISO 55,000:2014.
downloads/2013- Report-Card-Executive-Summary.pdf , viewed
4 July 2014. 19. NCE, 2015. Highways Agency reveals new-style maintenance deal,
London, UK, 11 Feb 2015. Available at http://www.nce.co.uk/news/
2. Brownjohn, J.M.W., 2007. Structural health monitoring of civil transport/highways-agency-reveals-new-style-maintenance-
infrastructure, Phil. Trans. R. Soc. A, 365(1851); 589-622 deal/8678398.article viewed 4 March 2015.
3. BSI, 1972. Code of practice for the structural use of concrete. 20. Pynn, C.R., 2004. Durability – a vigilant approach: Corrosion
Design, materials and workmanship. British Standards Institution, monitoring of reinforced concrete structures, Corrosion, New
London UK. CP 110-1:1972. Orleans USA, NACE-04328, 2004.
4. BSI, 1978. Steel, concrete and composite bridges. General statement. 21. Pocock, D.C., et al, 2014. Leveraging the Relationship between
British Standards Institution, London UK. BS 5400-1:1979. BIM and Asset Management, Infrastructure Asset Management,
5. BSI, 1985. Structural use of concrete – Part 1: Code of practice for 1(1) 12-16.
design and construction. British Standards Institution, London UK. 22. Robery P.C., et al, 2014. Keynote Paper: Role of Asset Management
BS 8110-1:1985. in Monitoring Durability Changes to Minimise Safety Risks
6. BSI, 2000. Concrete – Part 1: Specification, performance, production to Infrastructure, Proc. 3rd Int. Conf. Service Life Design for
and conformity. British Standards Institution, London UK. BS EN Infrastructure, Zhuhai, PRC.
206-1:2000. 23. Robery, P.C., 2008. Keynote Paper: Effective Management of
7. BSI, 2005. Eurocode 2: Design of concrete structures – Part 2 Concrete Assets, Intl. Congr. Concrete: Construction’s Sustainable
Concrete bridges – Design and detailing rules. British Standards Option, Dundee, UK. 8-10 July 2008
Institution, London UK. BS EN 1992-2:2005. 24. Robery P.C., et al, 2006. Valuation and whole life costing of concrete
8. BSI, 2006. Concrete – Complementary British Standard to BS EN bridges and other infrastructure assets in the UK, Proc. Structural
206-1 – Part 1: Method of specifying and guidance or the supplier. Faults and Repair ‘06, Edinburgh, UK. 13-15 June 2006.
British Standards Institution, London UK. BS 8500-1:2006. 25. Skagit Co, 2004. Second Street Bridge Construction Begins Monday,
9. BSI, 2008. Asset management Part 1. Specification for the optimised Four Days After Bridge’s Latest Hit, Skagit County, Wa, USA, http://
management of physical assets. British Standards Institution, www.skagitcounty.net/Departments/Home/press/080804.htm ,
London UK. PAS 55-1: 2008. viewed 1 Sept 2014.
10. CIRIA, 1987. Corrosion damaged concrete – assessment and repair. 26. Vernikos, V. K., et al, 2012. Realising offsite construction and
The Construction Industry Research and Information Association, standardization within a leading UK Infrastructure consultancy.
London, UK. Butterworths publ, ISBN 0-408-02556-5, 1987, 99p. Proc 7th AEC Conference, Sao Paulo, Brazil, 68–77
11. DOT, 1989. Performance of concrete in bridges - a survey of 200 27. Vernikos, V. K., et al, 2014. Building information modelling and its
highway bridges. Department of Transport, London UK. ISBN 0-11- effect on off-site construction in UK civil engineering. Proc. ICE-
550877-5, 1989. Management, Procurement and Law, 167(3): 152-159.

Organised by
India Chapter of American Concrete Institute 425
Session 4 B - Paper 1

Peter Robery
BSc, PhD, CEng, FREng, FICE, FCS, MICT, MIAM
Affiliation: Visiting Professor, the University of Leeds, UK
Director, Robery Forensic Engineering Ltd, UK
RAEng Visiting Professor in Forensic Engineering, University of Birmingham, UK
I am Visiting Professor at the University of Leeds, School of Civil Engineering, where I teach on the MEng
course and serve on the Industrial Advisory Committee. I am also founder/director of my own company,
Robery Forensic Engineering Ltd undertaking investigations of failures in construction, and Royal Academy
of Engineering Chair in Forensic Engineering at the University of Birmingham, UK.
Previously I was employed by Halcrow/CH2M HILL (2004 to 2015) in the UK and Chair of the International
Technology Forum, coordinating research and innovation outside of the North America operation. I was
their Technology Fellow in Concrete Materials, advising on concrete durability for new-build structures and
the forensic engineering examination of older structures to develop maintenance and repair strategies.
Prior to this I have been a director with AECOM in the UK, testing company Harry Stanger and contractor
Taylor Woodrow.
I am a registered (chartered) civil engineer, Fellow of the Royal Academy of Engineering, Fellow of the
Institution of Civil Engineers and Fellow of the Concrete Society. I am a member of RILEM, the American
Concrete Institute, the Institute of Asset Management and the Institute of Concrete Technology. I and am
a past President of the Concrete Society (2006/07 & 2007/08). I have authored or co-authored over 60
publications, including papers and book chapters, on the management of concrete infrastructure assets.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

426 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Rebuilding Nepal for Next earthquake

Rebuilding Nepal for Next earthquake

Er. Badan L Nyachhyon
Past President, Society of Consulting Architectural and Engineering Firms (SCAEF), Nepal
Managing Director, Multi Disciplinary Consultants (P) Ltd.

Abstract assets will never be the same again and will repeat the
The April 25, 2015 Nepal Earthquake created huge same fate with ad hoc reconstruction in the absence of
damages to life and properties in Kathmandu Valley and guided recovery plan.
28 other Earthquake affected districts along with massive However, the cities of Kathmandu Valley do not give
landslides and avalanches in Mount Everest and Lamtang impression that they were stricken with a deadly
Himalayas. Strong aftershocks counted over 400 had Earthquake with Peak Ground Acceleration of 1.2g much
drastically weakened the remaining building stock, which exceeding previous records of 0.52g. The whole world
has created unaccounted strain on the recovery plan. By expected a flattened Nepal. But to the contrary, amid the
that time, many of the buildings either will be dismantled vast destruction of poor quality and weak construction,
without any rationale and post mortem study or rebuilt the whole city including most of the cultural heritage
at the owners’ risk without any value addition of retrofit, sites, high rise buildings and residences are marvelously
recovery and protection of assets. Surely, the need for standing safe and intact. The credit for this state-of-the-
deriving lessons and confidently facing challenges of art performance goes to the stakeholders who had worked
recovery is paramount for enhancing the safety from hard during last 30 years to create Earthquake resilient
future earthquakes. The biggest challenge is to convince community following the provisions of Building codes and
the government and donors that any commitment made requirements of quality construction. But, however, the
after the earthquake has less value addition and need to challenge remains aggravated with the lack of coherent
focus on proactive initiatives. vision, ownership of the Earthquake affairs in general and
The objectives of this paper are to draw attention of rebuilding for Next earthquake.
the local and international community including the Keywords: Retrofitting and Rebuilding, Hunting
government and donors to gear up for investing in Post Dangerous Buildings, Dismantling or Protection,
earthquake recovery and Earthquake Safety Initiatives Earthquake Safety Commission, Proactive Investment for
before next Earthquake strikes. This is not an act for Rebuilding.
waiting and seeing but very urgent one that calls for
development of recovery plan and retrofit of remaining
building stock accounting over 5.5 million. At the same
Introduction
time, it is to draw attention of local community that the The Nepal Earthquake of April 25, 2015 followed by two
recovery and restoration need to be linked to economic major aftershocks of Magnitude 6.1 of April 26 and of
regeneration, and never be limited and restrained since it Magnitude 6.8 of May 12, 2015 and 400 aftershocks of
is the issue of preserving national identity. Magnitude over 4 has left Nepal in a state of devastation
making it difficult to come back to normal life style.
The methodology and procedures adopted for preparation Perhaps, the meaning of devastation is fully revealed with
of this work are mostly the field observations and review the experience of this Earthquake, which has destabilized
of various literature, buildings codes and declarations of the urban and rural physical setting and the mindsets. The
International Conventions. conglomeration of mindsets over flown from all over the
The Earthquake did not forgive the negligence shown in world is vividly colored.
effective implementation of Building Code and the poor Many aid workers were frustrated because of the lack of
quality construction and became the source of huge ability to visualize how the aid could be delivered to the
casualties. Particularly, the massive destruction of 744 needful communities in the hinterlands and over supply in
cultural heritage monuments comprising of traditional the accessible places around urban areas and the Airport.
vernacular esthetics in the form of temples of centuries Many supplies that did not met the national or international
old and 850,000 building units collapsed or damaged in standards were dumped in open ground in the Airport and
numerous traditional urban settlements is a big challenge could not get into the country, a real pathetic scenario.
for post earthquake recovery. These world heritage

Organised by
India Chapter of American Concrete Institute 427
Session 4 B - Paper 2

The first few days were the time when many people forced 30 years in advance preparation against the Hazards of
themselves out of the country in a panicky situation caring potential Earthquakes. It was their unprecedented hard
about oneself and forgetting the local partners with whom works that had been effective in minimizing the damages
they have shared so many of their valuable time. Each and casualties. The airport was running 24/7; all bridges
country was worried about the safety of its own citizens were intact to deliver the emergency supplies; the high
and many were quick enough to rescue their own people rise buildings though developed non-structural damages
leaving others in despair. Perhaps nobody has wisdom are standing well; thousands of houses, commercial and
enough to think about the rationale of actions – some were institutional buildings are standing intact except those
running away while others rushing in. that compromised on quality; devastated by earthquake
but people are still smiling.
The recovery of devastated people under the rubble of the
Earthquake aftermath was a spontaneous efforts of local
people and authorities who worked without any proper Problems and Issues
instructions – the Red Cross and local volunteers works The devastation of April 25, 2015 and over 400 aftershocks
were much appreciated that had helped to recovery of accounted for huge toll of life and property in the country
several lives from the rubbles. It was not amazing that that was otherwise was possible to reduce provided
those staying at the top floors escaped the death traps that proper attention was given in due course of time for
experienced by the people staying in the Ground Floor. capacity building of the local community, government and
The rapid relief works delivered by the international and non-government agencies and a dedicated agency was
the local communities were very much instrumental given charge. It was known to all that a large earthquake
in bringing the Earthquake ravaged society to a safe is inevitable and the only way to face such earthquakes
mode taking refuge in the temporary shelters as tents, is to make adequate preparations. Various tasks that
tarpaulins and semi-tunnels of corrugated sheets. That were very glaringly visible such as need for updating
was pretty instrumental in bringing the society to a building codes and urban development bylaws, removing
condition of resilience to earthquake and paying adequate the weaknesses and mischief in them, putting sincere
attention that the normal-after- earthquake epidemics efforts to implementation of the bylaws and codes,
as cholera, typhoid, swine flu, dysentery and diarrheria
do not occur. The volunteerism and SMS message across
the country warning about the precautions that need to be Table 1
taken was the State- of–the-performance. Damages and Toll

The overall scenario of the cities in Nepal after the April Description Expected Toll Actual Toll
earthquake did not resemble the scenario of Earthquake
Stricken cities but of a normal one with visibly intact Human Toll 100,000 8,969
cityscape. The vital infrastructure as water supply and
Injuries 300,000 22,321
sanitation facilities, electricity, telephone, internet, roads
and bridges, and airports remained unaffected and the Collapsed Buildings in Nepal 546,000 893,539
services were not interrupted. That was very instrumental Fully /Partially Damaged Private 887,074
in effective delivery of the international and domestic Houses
relief works across the affected 14 districts. Fully /Partially Damaged Health 963
facility
The damages though accounted as significant did not
extend to the magnitude forecasted by previous studies1. Government Offices 6,465
The estimated and actual casualties and damages are Schools 6,308
presented in Table 1.
Industries 133
Apart from the damages of the buildings that had made
Collapsed/Damaged Cultural 745
over 4.5 million people homeless, numerous landslides Heritage
and rock falls were triggered in the mountain areas,
Endangered Cultural Heritage 1500 ?
temporarily blocking roads.
Hydropower damaged 18
The 1934 Bihar-Nepal Earthquake produced strong
shaking in the Kathmandu Valley, destroying 20 percent Bridges 1
and damaging 40 percent of the Valley’s building stock. Few places
Roads
In Kathmandu itself, one quarter of all homes were
destroyed along with several historic sites (USGS). Water Supply Few Days

Observation of the cityscape across Kathmandu hardly Telephone None

indicates that the city was stricken by an Earthquake. Source: Kathmandu Valley Earthquake Risk Mapping Project, UNDP 1992;
This is the result of hard work of many people since last http://drrportal.gov.np

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

428 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Rebuilding Nepal for Next earthquake

checking the strengths of buildings and strengthening for ll To help update building bylaws and building codes based
design earthquake, quality improvement of design and on the lessons learnt from the recent Earthquakes and
construction, verification and certification were knowingly international experience, and
or unknowingly neglected and not implemented. ll To draw attention on the need to establish an apex
Despite for several voices called for paying attention on agency for developing ownership of Earthquake Affairs
need for declaring policy on building Earthquake Safer in the country and be responsible for.
cities and protecting important premises as historic
cultural monuments, schools, hospitals, industries, The Grand Reharsal for Future Erthquakes
communication and tourism infrastructure, the country
had no pronounced program to the effect needed. The The risk of potential earthquakes in Nepal was already
priority of actions for conservation of heritage and made immediately after the 1988 Earthquake of Dharan
cultural values verses modern engineering technology and Rajbiraj when 722 people were killed in Nepal and
needs to be established. The technology for safeguarding India, injured 12,000 and 450,000 were left homeless. The
millions of existing structures needs to be identified. The best part of Dharan Earthquake was the triggering of the
encouragement and motivation factors for investment in awareness within the Government in Nepal and the donor
Earthquake Safer Cities are missing. communities that had established the Kathmandu Valley
Earthquake Risk Management Project 1997.
The need for training of municipal and practicing engineers
in the design and construction of small buildings was initially The USGS quick report on the April 25, 2015 Gorkha
addressed through the support of UNDP2 but at later time it Earthquake made reference to very large Nepal
could not continue for lack of support and initiatives. earthquakes, with a moment magnitude of 7.5 or more,
observed in the historic periods in 1100, 1255, 1505, 1555,
This state of adamant situation cannot be anymore 1724, 1803, 1833, 1897, 1947, 1950, 1964, 1988. Three
continued. There is a strong need felt to find ways for earthquakes of similar size to the Gorkha Earthquake
creating earthquake resilient community through creating occurred in the Kathmandu Valley in the 19th Century: in
credible institutions, developing coordinated programs,
1810, 1833, and 1866. The seismic record of the region,
creating environment for effective delivery mechanism,
extending back to 1100, suggests that earthquakes of this
checking and verification of the actual deeds, and assuring
size occurred approximately every 75 years, indicating
the plans and programs were effectively implemented.
that a devastating earthquake is inevitable in the long
term.
The Objectives The strong motion network of Nepal is quite limited.
The objectives of the work are: Nevertheless, Kanti Path (Kathmandu) recorded the
ll To draw attention of the local and international maximum ground acceleration of 0.164 g. The USGS
communities to make significant investment for preliminary estimation of the maximum ground
capacity building of the country as a whole for the acceleration (PGA) in the epicenter area was about 0.35g
facing challenges of potential next large earthquakes, and 0.1 - 0.15 g for Kathmandu. In Western Nepal, PGA
ll To strengthen the already weakened, by the current range was between 0.5 g and 0.6 g, whereas in Eastern
earthquakes and aftershocks, premises in the country Nepal that ranged between 0.3 g and 0.6 g. The PGA
counted for over 5.5 million and comprising mostly of estimate was based on the empirical relations developed
adobe construction in Brick/Stone in mud mortar, by Aydan (Aydan and Ohta, 2011; Aydan 2007, 2012).
ll To draw attention on the need to target for minimizing Mr. Jean Ampuero, California Institute of Technology, in
the human toll below 1,000 in Next Earthquake, his paper “Salient Features of the 2015 Gorkha, Nepal
ll To draw attention of the community and the government Earthquake in Relation to Earthquake Cycle and Dynamic
on the need for recovery and conservation of lost Rupture Models” indicates that the high-frequency (HF)
cultural heritage and ancient heritage settlements as ground motions produced in Kathmandu by the Gorkha
priority, recovery of vast urban and rural settlements, Earthquake were weaker than expected for such a
and help to regenerate local economy to sustain the magnitude. The static slip reached close to Kathmandu
post earthquake recovery needs, but had a long rise time. An important observation
ll To provide training to the structural engineers, (Katsuichiro Goda, Department of Civil Engineering,
architects and urban planners for post earthquake University of Bristol, Bristol, UK and et el) is that the
recovery, seismic resistant planning and construction, ground motion shaking in Kathmandu during the 2015
and artisans training for quality construction, main shock was less than the PGA estimates (with 10%
probability of exceedence in 50 years i.e., a return period
ll To encourage preparing documentation of all premises
of 475 years). This may indicate that ground motion
for assuring earthquake safety,
intensity experienced in Kathmandu was not so intense,
ll To help develop recovery guidelines, in comparison with those predicted from probabilistic

Organised by
India Chapter of American Concrete Institute 429
Session 4 B - Paper 2

seismic hazard studies for Nepal. Therefore, a caution is Vulnerability Assessment and Certification
necessary related to future earthquakes in Nepal because Needs
the 2015 earthquake is not necessarily the worst-case
Most of the existing buildings stock in rural and urban areas
scenario and more intense Earthquakes may be in making.
comprises of Non-engineered traditional construction of
The surface deformation measurements including Brick/stone in mud mortar, and recent buildings in cement
Interferometric Synthetic Aperture Radar (InSAR) data and RCC structure. In the aftermath of the April Earthquake,
acquired by the ALOS-2 mission of the Japanese Aerospace it is assumed that over 80% of the damaged buildings fall
Exploration Agency (JAXA) and Global Positioning in the first category of brick and mud construction, and
System (GPS) data were inverted for the fault geometry remaining buildings to second category. Though there
and seismic slip distribution of the 2015 Mw 7.8 Gorkha is no post earthquake detailed vulnerability assessment
report of damaged and existing building stock available at
Earthquake in Nepal. The rupture of the 2015 Gorkha
this time. However, it is absolutely necessary to determine
earthquake was dominated by thrust motion that was
whether the existing building stock can withstand the next
primarily concentrated in a 150-km long zone 50 to 100
Most Considered Earthquake or Design Earthquake. This
km northward from the surface trace of the Main Frontal
question demands for carrying out a detailed vulnerability
Thrust (MFT), with maximum slip of ~ 5.8 m at a depth of assessment of the building stock covering four issues: 1)
~8 km, and 1.5 m at surface in Kathmandu Valley. In 1988, lack of documentation of the building stock, 2) Updating
Roger Bilham estimated this slip would be of magnitude of building code with consideration of recommended
of at least 10 m (Figure 1). Thus, based on the observed Design Earthquake Model. Many of these buildings are
values of Maximum Land Slip and the Maximum PGA, the not designed to sustain that kind of load; 3) construction
April Earthquake could be termed as a Grand Rehearsal quality and change in occupancy, and 4) maintenance
for the potential future earthquakes in Nepal. (Samir Chidiac, McMaster University, May 21, 2008).

Fig. 1: Earthquake Gap in Himalayan Arc

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

430 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Rebuilding Nepal for Next earthquake

It doesn't matter how well the design was, if it's not built or The distribution of the category of these buildings is not
used as specified, then the building is not the one it should known. There are two major challenges: 1) Demolition
be. This is what we have seen happening quite often. The of collapsed buildings and disposal or reuse of debris,
buildings designed and constructed have hardly had the and 2) Rehabilitation of Partially damaged buildings and
quality monitoring certificates, and further the operational buildings with minor damages. The general psyche is that
monitoring as maintaining design loading, occupancy and buildings with cracks (whatever may be the extent and
maintenance certification. It doesn't matter how well built cause) are no more useful for habitation and many started
a building is, it will age, which means its properties are demolition without giving any thoughts on potential for
changing, and if we don't address those matters, then restoration or rehabilitation. That has created strain
potentially we'll have a problem. on building stock deficiency creating huge price rise on
rental. But the rational for recovery is on rise.
One of the most important actions carried out in Nepal
immediately after the earthquake was the rapid visual Quick recovery of damaged buildings immediately after
the Earthquake was a very important aspect that would
vulnerability assessment of buildings. But the action faced
reduce the strain on the building stock. But in the absence
controversy because of lack of adequate preparation and
of recovery guidelines, access to resources as technology
legal provisions. The tools used were informally borrowed
of recovery and financing, the people gradually lost the
from ATC 40 without proper legal backup and training.
nostalgia of the Earthquake and started recovery on their
Most controversial action was the issue of Stickers (Green,
one way, mostly guided by the approach to quick repair
Yellow and Red) categorizing the buildings into Safe, and to demonstrate that the buildings were not affected by
Caution, Unsafe (Figure 2). The actions created confusion the Earthquake. They could no more wait for proper things
in the community about its rationale and appropriateness. to happen but to make efforts to quickly make financial
Surely, that was the result of lack of preparedness for recovery through use of the premises at the earliest,
such rapid action. neglecting the safety issues. The buildings demolished
The stickers were the good example of lack of adequate during the relief works period was never recorded and
preparation. They were issued in very unprofessional analysed to discover the root cause of the damages and
manner and illegally since there were no such laws or actual effect of the Earthquake.
guidelines that provide authority to do so. The Rapid
Vulnerability Assessment forms were borrowed from Conserve and Earn
elsewhere without authorization, proper guidelines and Most challenge is faced by the traditional residential
did not match with the typology of the buildings in the buildings and heritage monuments with vernacular
country. aesthetics that represented the identity of the country
and carried the value of history and culture of over 2,500
The Challenges years. Recovery of these buildings in the original form and
shape would be a strain on investment unless specific
Recovery of Damaged Buildings measures are taken to recover the lost heritage and
The Table 1 given above indicates the extent of damages generate economic return. The Traditional residences
without modern infrastructure and vehicular access could
to the building stock that include various category
be very redundant. There are several approaches being
of buildings such as 1) Low rise Concrete Buildings,
forwarded under the principles of “Integrated settlement
2) Residence in Brick Masonry in cement mortar, 3)
development” which will be developed following massive
Residence in Brick in mud mortar, 4) Residences in
dismantling of damaged buildings to give an outlook of the
traditional heritage Buildings in Brick and mud Mortar, traditional aesthetics. This will be totally new construction
and 5) Rural construction in stone in mud mortar, and 6) and nothing will resemble the value of history and
Rural Construction in Bamboo and thatch roof. culture carried by these settlements. Contrary to these,
the modern trends for quick recovery will change the
landscape dominated by modern technology that will lead
to the extinction of the ancient values. This will be a total
loss of the whole heritage assets. Some of the live cases
where the regeneration based on the recovery of cultural
heritage settlements promoted under the principle
of “Conserve and Earn” have carried the message
for paying attention on heritage conservation. These
schemes are getting more popular as “Home stay” tourist
accommodation. Some of the examples are: Shrestha
House and Swotha Café (Figure 3). The innovative concept
Fig. 2: Rapid Vulnerability Assessment of “Conserve and Earn” was recognized by UNESCO and

Organised by
India Chapter of American Concrete Institute 431
Session 4 B - Paper 2

Fig. 5: Krishna Mandir at Patan: Damaged by unplanned inter-

Fig. 3: Shrestha House and Swotha Café converted to “Conserve vention by erection of timber strut during Gorkha Earthquake.
and Earn” Projects Note damage to ancient inscription on Stone

inadequate and incomplete (Box 1). There is a dire need for

updating the Nepal Building Code3 making it independent
from other codes or reduce it to a guideline that will help
to make choice of better codes.

Fig. 4: Heritage monuments restored with international as- Box 1: Nepal Building Code Deficiency
sistance damaged during Gorkha Earthquake
Nepal Building Code is divided into four sections: Part 1) State-of-
the-Art Buildings, Part 2) Professionally Engineered Buildings,
given “World Heritage” recognition. These structures did 3) Non-Engineered Buildings (Mandatory Rule of Thumb), and
not suffered during Gorkha Earthquake. 4) Rural Construction. The code is divided into 22 parts and the
seismic design method is specified in NBC 105.
Some of the cultural heritage monuments restored with
In the preface, NBC 105 has included IS 4326 - 1993 Code of
International assistance suffered severe damages and Practice for Earthquake Resistant Design and Construction of
totally collapsed (See Figure 4). Probably, Earthquake Buildings as related code. There is a marked difference between
resistant construction of these monuments was not in these two codes with various values of the seismic parameters
their task. and giving different results. This anomaly has confused most
of the practicing engineers and NBC is practically not used.
Similarly, there are few instances where the structures Other factor affecting the use of NBC is the non-accessibility of
of cultural heritage were intervened and damaged them International software as SAP, ETAB and STAAD Pro where NBC
is not included as one of the recognized Codes.
by local authorities post Gorkha Earthquake (Figure 5).
The temporary timber struts were erected without any The basic parameters for Seismic Resistant design of NBC are
highlighted herewith. NBC 105 has divided the country into three
purpose and without any knowledge of the technical zones, illustrated in Figure 6, with Seismic zone factors of 0.9, 1.0
unit of the municipality and without consultation with and 1.1 and the Department of Mines and Geology has produced
local community. After some time the struts were the Kathmandu Valley Liquefaction Map (Figure 7), which gives
removed again without any information and taking any certain very vague information but again does not represent the
actual condition of liquefaction potential of the area. The major
measurement of strengthening or precaution. deficiency is the lack of program for updating the Liquefaction
These factors are considered the lack of ownership at map with the information on bore logs carried out by various
agencies.
the Government level and lack of consultation with the
professional and local community. During Gorkha Earthquake, a lot of buildings designed under
NBC 105 Part MRT (Non-engineered Buildings) were damaged.
The part of the code is considered in adequate in terms of
Updating Buildings Codes and Peer structural safety by several professionals and recommended
for elimination and replacement with standard designs for ready
Reviewing use. This part of the code is most misused by the municipality
registered designers and most used by coping and pasting of the
The lessons from the Earthquake clearly indicated that
details without giving design considerations and not verified for
the building damages observed during the Earthquake its acceptability.
were largely dependent on the appropriate use of the
Building codes, quality of construction, proper operation
and maintenance, and monitoring occupancy change.
The use of Building code itself is a complex process that
requires considerable time for carrying out design of
building based on the code requirements and inelastic
design based on computer modeling. The owners hardly
understand the complexities and time consumption of
seismic resistant design. More complex is the situation
in Nepal where the need for following other international
codes is paramount since Nepal Building Code in itself is Fig. 6: Seismic Zoning Map of Nepal (NBC 105)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

432 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Rebuilding Nepal for Next earthquake

The maximum Basic seismic coefficient specified by NBC

105 gives the max value of 0.08, which is much less a value
of 0.336 specified by IS 1893.
Given consideration of above earthquake design
parameters, the level of risks of structures will depend
on the choice of Building code selected. Hence, the
level of risk in every project becomes different and
level of earthquake hazard risk in Nepal also becomes
heterogeneous based on the source of funding sources.
In this context, NBC 105 may need updating to reflect
the demand of recent Earthquake and future probable
earthquakes and may need to develop consensus among
the leading professionals and academia about the choice
of appropriate Earthquake design model.
Fig. 7: Kathmandu Valley Liquefaction Map (DOGM)
Having said all above issues, it is imperative that the
It is obvious that these maps (Figure 6 & 7) do not consistency of design principles is lost and the compliance
represent the local conditions and give vague indication. to the Building Code requirements or application of
These maps are considered useful mostly for academic correct design criteria and analysis is assured. Need for a
practices. During the April 25 Earthquake, the high unified code acceptable at international level has become
potentiality of the liquefaction hazard indicted in Figure 7 imperative.
by the red color zones are not observed and not supported Apart from this, the assurance of use of appropriate
by the geotechnical information available from various code provisions and correctness of its interpretation and
boreholes investigations around the city of Kathmandu compliance is very important to insure consistency and
and Lalitpur as recorded in Annex 1. These maps are not eliminating any deficiency through making provisions for
based on adequate information, need to be used with peer review of seismic resistant design through a third
caution and shall not be made public. As a result, the need party verification of the quality of design and construction.
of micro-zoning maps has become very imperative.
In the Gorkha Earthquake Damage Survey report4
Rebuilding Approach
recommended that a basis for seismic design may be
considered the PGA estimates with 10% Probability of After the donors meet called by the Government in May
Exceedence in 50 years as the design earthquake model 2015, the International Community and the country as a
for Nepal. whole expected that rebuild initiatives will be launched
very quickly and the recovery initiatives will be started.
IS 1893 has included two categories of Design The Government’s effort to establish an independent
Earthquake: 1) 2 percent Probability of Exceedence in 50 authority met the political and legal jargon and practically
years (Maximum Considered Earthquake - MCE) and 10 went to coma. The Government’s post earthquake
percent in 50 years (Design Basis Earthquake - DBE). The instructions, related to 1) building bylaws restricted new
Structures in Category 1 shall be designed for MCE, which construction until further notice, 2) reduction of interest
is twice of DBE, whereas Structures in Category 2, 3 and 4 for bank loans, 3) short training of fresh engineers and
are designed for DBE for the project site. 4) finally the nomination of National Rebuilding Authority,
ATC 40 has specified 3 levels of earthquake ground became redundant because of inadequate home work and
motions: 1) Serviceability earthquake (SE) with 50 percent preparation and could not be formally established even
Probability of Exceedence in 50 year period, 2) Design after 6 months. The lack of consultation with experts and
Earthquake (DE) with 10 % Probability of Exceedence in 50 unilateral decisions were quite visible. The Government’s
years period and 3) Maximum Earthquake (ME) with 5% attitude of “Making the Decision in Haste and Repent in
Probability of Exceedence in 50 years period. Leisure” was very prominent.

ATC 40 has related the level of Earthquake with the The well wishers from all over the world are quite in
performance level of Buildings which is not the case with panic that Nepal could not practically gear up for post
NBC 105. earthquake recovery and loosing time. There were
practically no guidelines for post earthquake recovery and
The return period, as specified by NBC 105, for the onset of rebuilding. People started repair and recovery without
damage for a typical building of ordinary importance has any engineering or government support and many of the
been chosen as 50 years. The return period for the strength buildings started to return to the same status as before.
of buildings has been chosen as 300 years. NBC 105 specified
return period may be an under scored value compared to There is a strong voice that Nepal should learn from the
Katsuichiro Goda recommendation (See above). experience of Earthquake Recovery from other countries

Organised by
India Chapter of American Concrete Institute 433
Session 4 B - Paper 2

as Japan and New Zealand and should send Fact Finding care about the four corners of foundation) and don’t
missions for learning the lessons and developing right consider about the effect on foundation as foundation soil
approach and policy. The New Zealand’s approach of consolidation and risk associated with it, minor tilts and
post earthquake recovery through nomination of the settlements in foundation, and need for taking protection
Rebuild Team comprising of industry representatives needs. The gaps in Insurance payable damages shall be
ie the government, consultants, contractors, bankers, carefully included in the Rebuild Policy.
suppliers and manufacturers, insurance and community
was a unique model that helped New Zealand to recover
from 2011 Earthquake in a fast track manner with most Proposed Earthquake Safety Commission
effective way in terms of cost, time saving and employment Earthquake issues and remedies discussed above in
creation. relation to Nepal has lead to identification of a specific
need of permanent institution that will be in-charge of
Recently, the September16, 2015 Earthquake with
earthquake affairs, be responsive and act as an apex
magnitude 8.3 Mw in Chile caused only 13 fatalities. Why
national body that will provide leadership and guide.
only 13 fatalities in this earthquake, which is considered
the world’s strongest earthquake to date this year. While Possibly there is no common approach how earthquake
far weaker earthquakes in Haiti and, more recently, in issues are dealt at national or regional or state level.
Nepal, killed tens of thousands? The Chileans very proudly In the context of Nepal, there is apparently no agency
report that the resilience of Chile has three dimensions: a) at the apex level, which is responsible to deal with
Strong evacuation plans in coordination with international earthquake issues. It is widely felt that an Earthquake
community as the UN humanitarian affairs office and Safety Commission may be required for dealing with
the International Search and Rescue Advisory Group the vast scope of rebuilding, preparing for facing next
[Insarag], b) Strict building code that demand all new earthquakes and mobilizing national and international
buildings must be able to survive a 9.0-magnitude ressources. The Commission may be an independent
earthquake. The buildings can crack, tilt and even be and autonomous body charged with the mandate to deal
declared unfit for future use but it must not collapse and c) with all aspects of earthquake including research and
Strong sensitiveness to the Earthquake Disaster carried studies, development of the technology and policies,
by Ricardo Toro5, a former army general, in-charge of performance evaluation, development of strategy
Chile’s disaster relief agency, ONEMI. for the future, review and updating of building codes,
The lack of institutional model for rebuilding and generally bylaws, guidelines and manuals, conducting training
dealing with Earthquake Affairs in general is instrumental and capacity building, and assuring the overall safety
in the context of current chaos in Rebuilding Arrangement. including supporting for total insurance of residences.
Dissemination of these information and knowledge to
professionals and community leaders is a major tool that
Hunting Dangerous Buildings6
helps to upgrade the capacity of the local community for
Prior to April Earthquake, there were several seminars creating Earthquake Resilient Society.
and interactions conducted where the need for identifying
dangerous buildings which may be susceptible to future Given the situation in Nepal where there is a big gap
earthquakes and for development of programs to identifying in terms of ownership of earthquake matters, a lot of
such buildings, review their structural safety and encouraging unscrupulous actions emerged that have created panic
for investment strengthening. The recommendations if were within the community and created negative image. The
considered in a timely manner and appropriate programs issue of Red and Green Stickers described above is
were developed and implemented, a lot of achievements in an example. The post earthquake scenario and delay
protecting life and property would be seen. This would create in organizational set up for post earthquake recovery
a situation for developing Earthquake Resilient Communities. actions has made it paramount for establishing an
This could be part of the rebuild policy. independent and autonomous apex bode - Earthquake
Safety Commission. The leadership created through the
Commission will be beneficial for permanent surveillance,
Damage Recovery Insurance
developing policies and strategy, performance monitoring
There are very few buildings and structures in Nepal and evaluation, timely review and updating, and building
which were insured against earthquake damages. But the
consensus, which will be instrumental in creating
insurance as such has hardly helped to recover the losses
Earthquake Resilient Society.
due to earthquake damages. The insurance policies of
most of the Insurance companies of Nepal has hardly The Recommendation Study for update needs of NNBC
defined the effect of Earthquake and paying mechanism also suggested for establishment of National Code Council
for the losses. Insurance companies define the earthquake in order to keep the building code updated to comply with
damages within four corners of the buildings (even don’t international trends and needs.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

434 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Rebuilding Nepal for Next earthquake

Conclusion Creation of Earthquake Resilient Societies and traditional

settlements require advanced preparation in the form of: a)
Nepal is a highly earthquake prone area with noted
Overall Plan for rebuilding and recovery of the lost assets
earthquakes of magnitudes 4-5Mw two times a year, one
after the Gorkha earthquake, b) strengthening of existing
in summer and one in winter. The Gorkha Earthquake of
buildings and structure including vulnerability assessment
April 25, 2015 is considered as a grand rehearsal for future
and preparing data base of buildings and infrastructure,
potential earthquakes. The large energy accumulated in
c) Implementation of strong building codes, and d) be
the Himalayan Range, particularly around Kathmandu,
sensitive towards Earthquake Disaster. These tasks need
that could rock the area with a land slip of 10 m is not fully
meticulous planning, developing priorities, formulating
released since the land slip was just 1.5 m.
programs, developing tools, available expertise, building
The huge toll of life over 8,900 and loss of property about capacity, mobilizing resources, effective implementation,
600,000 collapsed buildings and 500,000 damaged and verifying compliance with standards. Along with the
buildings, though a very sad result, is considered preparation for plans for new construction, strengthening
significantly much less damage compared to the loss and retrofitting plans of existing buildings, which are given
estimates of previous studies. This is a positive achievement low priority in general, needs to be given equal priority.
of the efforts made during last decade towards creating
The planning, design and effective implementation of
Earthquake Safer Cities. At the same time, it is also
Earthquake Resilience Plans require an effective and
considered that pre-earthquake preparation was grossly
responsive agency that can take leadership and guide
inadequate.
the stakeholders for delivery of the services required for
Nepal’s Earthquake resilience march carries a huge creating Earthquake Resilient Societies. Two dedicated
score of challenges. In the wake of recent earthquake institutions are in high demand: 1) Earthquake Safety
and next potential disastrous earthquake, Nepal needs Commission, and 2) National Building Council, if the
to rebuild over 800,000 buildings and strengthen other country has to gear up to reduce the hazards of next
existing 5.5 million buildings of adobe construction. earthquake.
Apparently, there is no effective technology to restore,
rebuild and strengthen the existing adobe construction. At
the same time, updating of the building code and its strict Recommendation
implementation that would help to assure Earthquake The Post Earthquake Rebuilding Course of Nepal seems at
Resilient Society is of essence in terms of assessment, this time in great strain since the initiatives required for its
planning, implementation in a timely manner. commencement are still not visible. The foremost important
task probably is the establishment of an institutional
The rebuilding initiatives already have been delayed by model that will be able to take full responsibility and be
6 months. It had disappointed the whole world and the responsive. Following movement may be worth to take into
devastated people. But the Government is still not in consideration following priority activities:
move. This is a very pathetic situation, aggravated by the
economic embargo since September 2015 that will further ll Formalization of establishment of the National
delay the progress of rebuilding and overall progress. Rebuilding Authority promulgated by the Government
The country is slowly going back to the same status of in May 2015 and agreeing on its Terms of Reference
vulnerability it was before the earthquake. and mandates,
There are several models of recovery and rebuilding ll Formulation of Lessons learnt from the Gorkha
from earthquake disaster. Gujarat, Haiti, Chile and Earthquake,
Christchurch are recent models. The Chile model has ll Mobilisation of fact finding missions to various
very strong search and rescue plan, strict building codes countries to learn lessons from previous devastating
that demand for no collapse design, and sensitivity earthquakes and indentifying effective rebuilding
towards Earthquake Disasters. Christchurch model technology,
mobilized resources within the country with formulation ll Taking initiatives for updating of Building Code and
of a strong and dedicated rebuild team based on non- defining strict implementation procedures including
profit job distribution. Probably, Nepal need to combine third party peer review, and in long term authorizing
and blend a suitable rebuild course based on world a permanent agency as National Code Council to take
experience. charge of Building Code update,
The NNBC 105 that deals with earthquake resistant ll Nominating a Rebuild Team or other appropriate
design contains several anomalies including inadequate mechanism to undertake the rebuilding tasks with
design earthquake model, unacceptable design of non- spelled out time frame, targets, and deciding on the
engineered buildings, and requires immediate updating method of procurement of services and works,
including establishment of National Code Council that will ll Mobilising international and local community
be responsible for timely updating of the codes. partnership in rebuild initiatives,

Organised by
India Chapter of American Concrete Institute 435
Session 4 B - Paper 2

Annex 1: Bearing capacity in KN/m2 and Liquefaction Potential in Acknowledgement

certain areas in KV ( Multi Lab (P) Ltd.)
ll Department of Urban Development and Building
Site Adopted Bearing Capacity Based on: Lique- Construction, Government of Nepal Society of
Bearing faction Consulting Architectural and Engineering Firms
Capacity Shear SPT Direct Un- Potential
Value confined ll Structural engineers Association of Nepal
Compression ll Society of Nepalese Architects
Site-1 150 150 150 150 None ll Nepal Engineers’ Association
Site-2 90 90 325 200 None ll Indian Chapter of American Concrete Institute,
Site-3 85 85 85 85 None Mumbai, India Shivam Cement (P) Ltd
Site-4 150 150 150 150 None ll Venus Group of Companies
Site-5 350 350 None
ll Sika India
Site-6 225 225 None
Site-7 100 150 100 150 None References
Site-8 150 650 150 None 1. Kathmandu Valley Earthquake Risk Mapping Project, UNDP, 1997
Site-9 90 90 150 100 None 2. Young Engineers Training for Earthquake Resistant Design, UNDP/
Site-10 150 230 150 None Earthquake Safety Initiatives, 2008

Site-11 140 140 None 3. Recommendation for updating of Nepal National Building Code 1994,
Multi Disciplinary Consultants (P) Ltd, KD Associates (P) Ltd and
Site-12 1175 1175 None
Khwopa Engineering College/ Department of Urban Development
Site-13 200 200 250 None and Building Construction, Ministry of Physical Planning and Works,
Site-14 190 200 190 None Government of Nepal/ Earthquake Risk Reduction and Recovery
Project(UNDP/ERRRP: NEP/07/010(2009))
Site-15 140 325 140 None
4. Katsuichiro Goda and et el, Department of Civil Engineering,
University of Bristol, Bristol, UK
ll Developing consultation mechanism for addressing
professional and community concerns and establishing 5. Former Army general stationed in Port au Prince, Haiti in 2010
when the 7.7-magnitude earthquake destroyed the city. He lost his
ownership at local level,
wife in the earthquake disaster.
ll Develop Support mechanism for developing appropriate
6. Badan Lal Nyachhyon, 2008
technology to address local demand for recovery and
rebuild of lost buildings and strengthening of existing 7. ALOS-2 mission of the Japanese Aerospace Exploration Agency
buildings in adobe construction, (JAXA)

ll Providing priority to Conservation of Cultural Heriatge 8. Applied Technical Council ATC 40, California, USA
Monuments and Settlements, 9. American Concrete Institute
ll Develop mechanism to give economic value return for 10. American Society of Civil Engineers
the recovery and rebuild products for sustainability, 11. Aydan and Ohta, 2011; Aydan 2007, 2012
ll Establishing incentives and motivation packages 12. Badan Lal Nyachhyon, Hunting Dangerous Buildings, 2008
including reduced interest rates for bank loans,
13. Federal Emergency Management Agency
elimination of Government and municipal taxes
on rebuild activities and seismic strengthening of 14. International Building Code, 2012
properties, 15. Jean Ampuero, California Institute of Technology, USA
ll Making policy reforms in updating Building Bylaws, 16. Kathmandu Valley Earthquake Risk Mapping Project, UNDP, 1997
Building Act, and Earthquake hazard insurance, 17. Katsuichiro Goda, Department of Civil Engineering, University of
ll Encourage Third party review for assuring quality in Bristol, Bristol, UK and et el
design, construction and implementation. 18. National Forum for Earthquake Safety
ll Developing evacuation plan before earthquake strikes, 19. Nepal National Building Code 1994 and 2009
ll Developing search and rescue plan during an 20. Recommendation for Update of Nepal National Building Code
earthquake, 1994, Earthquake Risk Recovery and Rehabilitation Project, UNDP/
ll Establishing Earthquake Safety Commission to take ERRRP-Project: NEP/07/010, Multi Disciplinary Consultants (P)
charge and be responsible for all affairs related to Ltd., 2009.
Earthquake hazard in long run. 21. United States Geological services

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

436 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Rebuilding Nepal for Next earthquake

Er. Badan Lal Nyachhyon

Er. BL Nyachhyon is a senior civil engineer with over 44 years of professional experience. He has had MS in
Structural Engineering and Construction management from USSR in 1971, obtained training on Earthquake
Engineering from Uzbek Academy of Sciences in 1980 and Project Management in Germany in 1985. He
has founded several professional societies as Society of Consulting Architectural and Engineering Firms
(SCAEF Nepal), National Forum for Earthquake Safety (NFES), Zero Waste Nepal, and Three Star Club - a
national sports club.
To his credit, Er. Badan has had published several articles dedicated to creation of Earthquake Resilient
Cities. He has significantly contributed in formulation of Nepal National Building Code 1994, carried out
recommendation study for updating of Nepal National Building Code 2010 and initiated Master Degree
Course in Earthquake Engineering in Nepal. He also had contributed in formulating suggested National
Policy for development of consulting services in Nepal. Currently, he is the Managing Director of Multi
Disciplinary Consultants (P) Ltd.

Organised by
India Chapter of American Concrete Institute 437
Session 4 B - Paper 3

Building Durable Concrete Infrastructure Using Fibre- Reinforced

Polymer (FRP) Bars
Hamdy M. Mohamed and Brahim Benmokrane
Dept. of Civil Eng., University of Sherbrooke, Sherbrooke, Qc, Canada

Abstract great alternative to steel reinforcement. FRP materials

in general offer many advantages over the conventional
In the last decade, there has been a rapid increase in
using noncorrosive fibre-reinforced polymers (FRP) steel, including one quarter to one fifth the density of
reinforcing composite bars for concrete structures due steel, no corrosion even in harsh chemical environments,
to enhanced properties and cost- effectiveness. The FRP neutrality to electrical and magnetic disturbances, and
bars have been used extensively in different applications greater tensile strength than steel (Benmokrane et al.
such as bridges, parking garages, water tanks, tunnels 2006; 2007).
and marine structures in which the corrosion of steel The objective of this paper is to show that FRP bar is
reinforcement has typically led to significant deterioration on its way toward gaining widespread acceptance in
and rehabilitation needs. Many significant developments worldwide. Clearly, the most tangible successes are in
from the manufacturer, various researchers and Design the area of highway reinforced concrete bridges, parking
Codes along with numerous successful installations have garages, tunnelling and marine structures in which the
led to a much higher comfort level and exponential use corrosion resistance of FRP reinforcements as well as
with designers and owners. After years of investigation their installation flexibility are taken advantage of. In the
and implementations, public agencies and regulatory following sections, development of codes and guidelines,
authorities in North America has now included FRP as recent field applications of FRP bars in bridges, tunnels,
a premium corrosion resistant reinforcing material in parking garages and water storage tank are presented.
its corrosion protection policy. This paper presents a
summary and overview of different recent field applications
of FRP bars in different types of civil engineering concrete Design Codes and Guidelines
infrastructures. A number of committees from professional organizations
around the world have addressed the use of FRP bars in civil
structures. These have published several guidelines and/
Introduction or standards relevant to FRP as primary reinforcement
Electrochemical corrosion of steel is a major cause of for structural concrete. The recommendations ruling
the deterioration of the civil engineering infrastructure. the design of FRP RC structures currently available are
It is becoming a principal challenge for the construction mainly given in the form of modifications to existing steel
industry world-wide. An effective solution to this problem RC codes of practice, which predominantly use the limit
is the use of corrosion resistant materials, such as high- state design approach. Such modifications consist of
performance fibre- reinforced polymer (FRP) composites, basic principles, strongly influenced by the mechanical
(Benmokrane et al. 2002; Mohamed and Benmokrane properties of FRP reinforcement, and empirical equations
2014). The applications of FRP reinforcements in the last 10 based on experimental investigations on FRP RC elements.
years have been approved that the cutting-edge technology
has emerged as one of the most cost-effective alternative In North American, several codes and design guidelines
solutions compared to the traditional solutions. The use for concrete structures reinforced with FRP bars have
of concrete structures reinforced with FRP composite been published from 2000 to 2014. In 2000, the Canadian
materials has been growing to overcome the common Highway Bridge Design Code (CHBDC) [CAN/CSAS6-00,
problems caused by corrosion of steel reinforcement. (CSA 2000)] has been introduced including Section 16 on
The climatic conditions where large amounts of salts are using FRP composite bars as reinforcement for concrete
used for ice removal during winter months may contribute bridges (slabs, girders, and barrier walls). Design manual
to accelerating the corrosion process. These conditions (ISIS-M03-2001) for reinforcing concrete structures
normally accelerate the need for costly repairs and may with FRP was presented by the Canadian Network of
lead to catastrophic failure. Centres of Excellence on Intelligent Sensing for Innovative
Structures (ISIS). In 2002, CAN/CSA-S806-02 has been
Known to be corrosion resistant, FRP bars provide a
published by the Canadian Standards Association (CSA

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

438 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Building Durable Concrete Infrastructure Using Fibre- Reinforced Polymer (FRP) Bars

2002) for design and construction of building components components of structures (e.g., bridges, buildings, and
with FRP bars. marine structures). The FRP bars are classified on the
basis of their fibres, strength, stiffness, and durability.
The American Concrete Institute (ACI) introduced the first,
Only FRP bars made with aramid, carbon, or glass fibres
second and third guideline (ACI 440.1R) for the design
are considered in this Standard.
and construction of concrete reinforced with FRP bars in
2001, 2003 and 2006, respectively. The ACI 440.1R design
guidelines are primarily based on modifications of the ACI- FRP Composite Reinforcing Bars
318 steel code of practice (ACI 318-02, 2002).As a result
of the valuable, enormous and great research efforts on Advantages
different types of FRP-reinforced concrete structures in The technology of reinforced concrete is facing a serious
worldwide during the last decade, the aforesaid North degradation problem in structures due to the corrosion
American codes and design guidelines have been updated of steel bars. In North America, the repair costs are
and modified to encourage the construction industry to estimated to be close to 300 billion dollars. Several
use FRP materials [CAN/CSAS6-14; CAN/CSA-S806-12; options have been explored, most notably the use of
ACI 440.1R-15]. Nowadays, the CAN/CSA-S806-12 (2012) galvanized steel rebar, epoxy coated or stainless steel.
is the most recently issued Canadian guidelines on the The results, however, have been disappointing as these
design and construction of building components with solutions have turned out to be less than effective or cost
FRP. The CSA S806 has been completely revised. Many prohibitive. The FRP bars have proven to be the solution.
of its provisions have been improved based on the latest Lightweight, corrosion resistant, and offering excellent
research results and experience in the field. The CSA tensile strength and high mechanical performance, FRP
S806-12 contains new provisions on: punching shear bar is installed much like steel rebar, but with fewer
at slab-column connections with or without moment handling and storage problems. The material cost might
transfer, confinement of columns by FRP internal ties or still be higher compared to the costs of conventional steel
hoops, design of FRP reinforced member for combined products, but this fact is more than compensated with the
effects of shear, torsion and bending, reinforcement lesser maintenance work involved during the lifetime of
development length and detailing, strut and tie model for the structure. Also, the weight of a FRP bar is only a fourth
deep beams, corbels and brackets, shear strengthening of its steel counterpart, having the same dimensions.
of reinforced concrete members by externally bonded Combined with the flexibility of the bars this allows an
reinforcement, and FRP retrofit of reinforced concrete easy installation even in confined working space or where
members for enhanced seismic resistance. The new the support of lifting equipment is not available. The most
standard covers all the basic design requirements for FRP commonly manufactured fibers employ glass and carbon.
reinforced and retrofitted structures. E-glass is the most common fiber because of its strength
In addition to the design of concrete elements reinforced and resistance to water degradation. It is also used as an
or prestressed with FRP, the guidelines also include electrical insulator.
information about characterization tests for FRP internal On the technical level, FRP products have important
reinforcement. As for the predominant mode of failure, the advantages. FRP reinforcement bars can be used in tunnel
CSA S806-12 remarks that “all FRP reinforced concrete application as soft-eyes have a very high tensile strength
sections shall be designed in such a way that failure of which can reach far over 1200 N/mm2. Besides flexibility,
the section is initiated by crushing of the concrete in the elasticity and the minimal environmental impact the
compression zone”. In this code, new design equations are GFRP bars can be cut with working tools like saws, pilling/
included for design punching shear capacity of FRP-RC drilling equipment and TBM tools. This avoids damages
flat slab. Also, it is of interest to mention that this code to cutter heads and does not delay the work progress as
permits of using FRP bars in columns and compression piling or cutting through GFRP bars is unproblematic.
members. The fiber bars are split in small pieces which do not harm
In order to establish stringent guidelines and values for slurry pipes.
FRP manufacturers and quality control mechanisms for
owners to ensure a high comfort level of product supplied, Mechanical Properties
ISIS Canada together with the manufacturer had initiated The mechanical properties of FRP bars are typically quite
the “Specifications for product certification of FRP’s as different from those of steel bars and depend mainly on
internal reinforcement in concrete structures”. (ISIS both matrix and fibers type, as well as on their volume
Canada Corporation 2006) This document was the basis fraction, but generally FRP bars have lower weight, lower
for the new Standard CSA S-807-10 on Specification for Young’s modulus but higher strength than steel. The most
Fibre Reinforced Polymer (FRP). This Standard covers the commonly available fiber types are the carbon (CFRP), the
manufacturing process requirements of fibre-reinforced glass (GFRP) and the aramid (AFRP) fibers. Table 1 gives
polymer (FRP) bars or bars that are part of a grid for use the most common tensile properties of reinforcing bars,
in non-prestressed internal reinforcement of concrete in compliance with the values reported by CSA S- 807-10.

Organised by
India Chapter of American Concrete Institute 439
Session 4 B - Paper 3

States Federal Highway Administration (FHWA) estimates

Table 1
Typical Mechanical Properties of GFRP Bars that eliminating the nation’s bridge deficient backlog by
2028 would require an investment of $20.5 billion annually
Grade
Tensile Strength Modulus of Ultimate Tensile because of corroded steel and steel reinforcement. The
(MPa) Elasticity (GPa) Strain report also states that “the nation’s 66,749 structurally
I 588 - 804 40 - 47 0.0134 – 0.0189 deficient bridges make up one-third of the total bridge
decking area in the United States, showing that those
II 703 - 938 50 - 59 0.0133 – 0.0179 bridges that remain classified as structurally deficient are
III 1000 - 1372 60 - 69 0.0151 – 0.0211 significant in size and length, while the bridges that are
being repaired are smaller in scale.” Problems related to
expansive corrosion could be resolved by protecting the
Recent Frp Field Applications steel reinforcing bars from corrosion-causing agents or
by using noncorrosive materials such as fiber-reinforced-
Highway Bridge Structures polymer (FRP) bars. Therefore, since the late 1990s, the
Corrosion of steel reinforcing bars stands out as a significant Structures Division of the MT at different provinces has
factor limiting the life expectancy of reinforced concrete been interested in building more durable bridges with an
infrastructure worldwide. In North America in particular, extended service life of 75–150 years. For example, the MT
the corrosion of steel reinforcement in concrete bridges at Québec (MTQ), Canada has carried out, in collaboration
subjected to deicing salts and/or aggressive environments with the University of Sherbrooke, (Sherbrooke, Québec),
constitutes the major cause of structure deterioration, several research projects utilizing the straight and bent
leading to costly repairs and rehabilitation as well as a non-corrodible FRP rebar in concrete deck slabs and
significant reduction in service life. According to the 2013 bridge barriers (Mohamed et al. 2014; Ahmed et al. 2014;
Report Card for America’s Infrastructure findings, ASCE, Mohamed and Benmokrane 2014). The use of FRP bars as
nearly one-tenth of the 607,380 bridges in the National reinforcement for concrete bridge provides a potential for
Bridge Inventory were classified as structurally deficient. Of increased service life and economic and environmental
this total, over 235,000 are conventional reinforced concrete benefits.
and 108,000 were built with prestressed concrete (NACE
In the last ten years, the FRP bars have been used
International). The report further states that $76 billion are
successfully in hundreds bridge structures across
needed for deficient bridges across the United States for
Canada and USA, see Figure 1 (a, b and c). These bridges
maintenance and capital costs for concrete bridge decks
were designed using the Canadian Highway Bridge
and for their concrete substructures. In addition, the United

(a) FRP decks and barriers – Gateway Blvd/23rd Ave –Alberta (2009) (b) FRP decks/app slabs/ barriers, Skagit River – BC MOT (2009)

(c) bridge deck slab, 410 overpass bridge Qc (2012) (d) cable stayed bridge, Nipigon River Bridge, ON (2015)

Fig. 1: Recent FRP-reinforced concrete bridges

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

440 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Building Durable Concrete Infrastructure Using Fibre- Reinforced Polymer (FRP) Bars

Design Code or the AASHTO LRFD Bridge Design Guide that supports the vertical walls and top slab. The design
Specifications for GFRP- Reinforced Concrete Bridge of the tank was made according to CAN/CSA-S806-02,
Decks and Traffic Railings. Straight and bent FRP Design and Construction of Building Components with
bars (carbon or glass) were used mainly as internal Fibre-Reinforced-Polymers. This included the use of High
reinforcement for the deck slab and/or for the concrete Modulus GFRP reinforcing bars (Grade III, CSA S807)
barriers and girders of these bridges. In general, all the as main reinforcement for the foundation, walls and top
bridges that included with FRP reinforcements though slab. The tank is well instrumented at critical locations for
the ten years ago are girder-type with main girders made strain data collection with fiber-optic sensors. Figure 2.a
of either steel or prestressed concrete. The main girders and b shows the FRP bar reinforcements in the vertical
are simply supported over spans ranging from 20.0 to walls and overview of the complemented FRP tank. The
90.0 m. The deck is a 200 to 260 mm thickness concrete field test results under actual service conditions for the
slab continuous over spans of 2.30 to 4.0 m. Most of these strain behavior in the FRP bars at different location in the
bridges have been reinforced with the glass FRP bars tank are indicated a significant value less the 1.0 % of the
as a result of their relatively low cost compared to other ultimate strain. In conclusion, the construction procedure,
types of FRPs (carbon and aramid). The FRP bars were serviceability performance under real service conditions
used mainly as reinforcement to the deck slabs, barriers (water and earth pressure), and monitoring results of the
and girders. Recently, the GFRP bars have been used as FRP-reinforced walls and slabs of the tank, in terms of
the main reinforcement in the deck slab of cable stayed strain, cracking and deflection were very conservative
bridges, Nipigon River Bridge, ON, Canada. The Nipigon and satisfactory when compared with the serviceability
River Bridge spans the Nipigon River in Nipigon, Ontario requirements and strength needed.
on Highway 11/17. A new four lane cable stayed bridge
is replacing the old two lanes, four span plate girder
structures. The new bridge includes cable-supported
spans of 112.8 m and 139 m. The 36.2 m wide deck is
comprised of concrete deck panels totally reinforced with
GFRP bars and supported on transverse steel beams, see
Figure 1(c). The objectives were to implement FRP bars in
RC cable stayed bridge to overcome the steel expansive-
corrosion issues and related deterioration problems; to
assess the in-service performance of the FRP-RC bridge
deck slab after several years of operation; and to design
durable and maintenance-free concrete for cable stayed
bridge. The deck slab was design to sustain significant
axial compression force resulted from the cables and
bending moment as resulted from the live and dead loads (a) Formworks and FRP reinforcement bars of the wall
(Mohamed and Benmokrane 2012).

Water Tank
Reinforced concrete (RC) tanks have been used for water
and wastewater storage and treatment for decades. Design
of these tanks requires attention not only to strength
requirements, but also to crack control and durability. RC
water treatment plant structures are subject to severely
corrosive environments as a result of using the chlorine
to treat the wastewater before it is released. So, the
challenge for the structural engineer and municipalities
is to design these structures using noncorrosive fibre-
reinforced polymers (FRP) reinforcing bars. The first (b)Over view of the tank
worldwide concrete chlorination water treatment tank
totally reinforced with FRP bars was designed in 2010 and
Fig. 2: FRP-reinforced concrete tank, Qc, Canada
the construction started and finished in 2012. The project
is located in Thetford Mines city, Quebec, Canada and it is
considered as one component of water treatment plant for GFRP Soft Eyes in Tunnels
municipality. The volume capacity of the tank is 4500 m3, Building tunnels with Tunnel Boring Machines (TBM)
and it has the dimensions 30.0 m wide, 30.0 m length and is today state of the art in different ground conditions.
5.0 m wall height. The structural system of the tank is Launching and receiving the TBM in shafts and station
rectangular under- ground tank resisted on raft foundation boxes has in earlier years required a considerable

Organised by
India Chapter of American Concrete Institute 441
Session 4 B - Paper 3

construction effort. Breaking through the steel reinforced to 1100 mm). Highest grade 60 GPa 32.0 m vertical bars
walls of the excavation shaft with a TBM required extensive were used with #5 (16.0 m) 50 GPa continuous spirals
measurements and preparation works, (Mohamed and with 150 mm pitch, see Figure 3 (b and C) (Mohamed and
Benmokrane 2015; Schürch and Jost 2006). FRP is an Benmokrane 2015).
anisotropic composite material with a high tensile strength
in axial direction and a high resistance against corrosion. Parking Garages
The anisotropy of the material is quite advantageous at The need for sustainable structures has motivated the
excavation pits for the starting and finishing processes at Public Works and Government Services Canada (PWGSC)
automated excavation like tunnel boring machine (TBM) in the use of FRP rebar as internal reinforcement in
and Pipe jacking, see Figure 3(a). Therefore, using FRP concrete infrastructure applications. One of the most
bars in reinforced walls and piles of the excavation shaft important successful applications is using FRP rebar in
allows the designer and contractor today to find innovative reinforcing the parking garage. An agreement between
solutions for the well-known situation and save time and PWGSC and the University of Sherbrooke was reached to
costs on site. Soft-Eyes consist usually of bore piles or reconstruct the interior structural slabs of the Laurier–
diaphragm walls which are locally reinforced with GFRP Taché parking garage (Hull, Quebec) using carbon and
bars and stirrups, see Figure 3(b). The sections below glass FRP rebar, see Figure 4. The design was made
and above the tunnel opining are reinforced steel bars. according to CAN/CSA-S806-02. This project allows
Depending on the designer and contractors preferences direct field assessment and long-term monitoring of
full rectangular sections are built out of GFRP bars or FRP composite bars in a structure subjected to harsh
the fibre reinforcement follows more closely the tunnel environmental and loading conditions. In 2010, the
section resulting in a circular arrangement of the GFRP new large parking garage (La Chancelière parking
links and similar adjustments for the vertical bars. garage, area 3000 m2) in Quebec City was designed and
Building the corresponding reinforcement cages out of constructed using the FRP rebar. This design was made
GFRP bars on site requires the same working procedures according to the CAN/CSA-S413-07 for parking structures
as for an equal steel cage, see Figure 3 ( c and d). Recently, and CAN/CSA-S806-02 [13] for design and construction of
GFRP bars have been used in different tunnel projects in building components with fibre reinforced polymers. The
Canada (South Tunnels, Keele Station, Hwy 407 Station- two-way flat slabs of La Chancelière had maximum span
TTC Subway North Tunnels and Eglinton Crosstown LRT: of about 9.0 m. The thickness of the slabs was 250 mm
Toronto, ON). Whereas, GFRP bars were used to reinforce which increased to 355 mm over the columns through the
GFRP cages up to 19.0 m long (diameters ranged from 600 drop panels, see Figure 6. The increased thickness over

(a) TBM cutting through FRP-reinforced concrete drilled shaft wall (b) GFRP Soft-Eyesa

(c) Handling and lifting the GFRP Soft-Eyes (d) Soft-eye reinforcement for a diaphragm wall

Fig. 3: GFRP reinforcement for soft-eyes

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

442 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Building Durable Concrete Infrastructure Using Fibre- Reinforced Polymer (FRP) Bars

the columns was devoted to satisfy the punching stresses construction of Canada’s first roadway with continuously
around the columns’ area. The punching strength of the reinforced concrete pavement (CRCP). Five years later,
two way slabs were verified using the new punching however, concerns were raised about the long-term
equations that are being incorporated in the new version performance of CRCP, as portions of this initial installation
of the S806 Standards, Benmokrane et al. 2012). were found to have insufficient cover over the bars and
core samples showed that the longitudinal reinforcement
was corroding at transverse cracks (Thébeau 2006). These
observations, coupled with the knowledge that up to 60
tonnes (65 tons) of salt per year can be spread on a 1 km
(0.6 mile) long stretch of a two-lane pavement in Montréal
(nearly three times the amount of salt used on roads in the
State of Illinois), led the MTQ to select galvanized steel as
the standard reinforcement for subsequent CRCP projects
and to continue investigating other systems with enhanced
corrosion resistance. As part of these investigations, the
MTQ and the University of Sherbrooke has been studying
the use of glass-fiber-reinforced-polymer (GFRP) bars
for CRCP since 2006. In September 2006, a 150 m long
section of eastbound Highway 40 (Montréal) was selected
(a) Laurier-Taché Parking Garage as a demonstration project (Benmokrane et al. 2008), see
Figure 5. Through the initial 18 months of pavement life, the
maximum measured strain value in the reinforcement was
0.0041. This is within the design limit recommended in ACI
440.1R-06. In February 2008, the measured results showed
that the average crack spacing varied between 1.5 and 4 m
in most CRCP-GFRP slabs. In addition, the average crack
width varied between 0.7 and 0.9 mm, which is less than the
AASHTO design limit of 1.0 mm (Benmokrane et al. 2008).

(b) La Chancelière parking garage

Fig. 4: GFRP bars applications in parking garage

Continuously reinforced concrete pavement with GFRP

bars
Continuously reinforced-concrete-pavement (CRCP)
designs are premium pavement designs often used for
heavily trafficked roadways and urban corridors. Although
CRCP typically is an effective, long- lasting pavement
design, it can develop performance problems when the
aggregate–interlock load transfer at the transverse
cracks has degraded. The prevalence of wide cracks in
CRCP has frequently been associated with ruptured steel
reinforcement and significant levels of corrosion. This has
generated recent interest in identifying new reinforcing
materials that can prevent or minimize corrosion-related
issues in CRCP. Glass-fiber-reinforced-polymer (GFRP)
bars are one product being investigated for use in CRCP
instead of conventional steel bars.
Since the early 1990s, the Ministry of Transportation of
Quebec (MTQ) has renewed emphasis on building long-
lasting concrete pavements suited to local traffic and
climatic conditions. In 2000, these efforts led to the Fig. 5: GFRP reinforcement bars in Highway 40 (Montréal)-2006

Organised by
India Chapter of American Concrete Institute 443
Session 4 B - Paper 3

Fig. 6: GFRP bar placement in center lane in Highway 40 (Montréal)-2013

In September 2013, it was decided to use GFRP bars in Application of FRP reinforcement in different structures
one of Quebec’s CRCP highways (300 m long). A stretch of has been proved to be very successful to date. From the
test pavement has since been constructed on westbound construction point of view it was felt by the construction
Highway 40 in Montreal. The project is located on eastbound personnel that the lightweight of the FRP reinforcements
Highway 40 in Montréal, QC, and presents a collaboration were easy to handle and place during construction.
between the Ministry of Transportation of Quebec (MTQ) Concrete bridges, water tank, soft eye-tunnel application,
and the University of Sherbrooke. A variety of sensors were parking garage structures and continuously reinforced-
installed in this project to monitor the early-age behavior concrete-pavement provide an excellent application for
and the effects of repeated traffic loads and environmental the use of FRP in new construction.
conditions on the performance of CRCP slab. The test slab
was 315 mm (12.4 in.) thick with a GFRP reinforcement
Acknowledgement
ratio of 1.2%. The reinforcement ratio for steel bars in the
CRCP steel- reinforced slab is 0.1% for transverse rebar The authors would like to thank and express their sincere
and 0.74% for longitudinal bar, see Figure 6. According to appreciation to the NSERC–Canada Research Chair and
observations at 16 months, the crack spacing and crack Industrial Research Chair group (Canada Research Chair
width in the steel-reinforced CRCP test section were larger in Advanced Composite Materials for Civil Structures &
than those of the GFRP-reinforced CRCP section. The field NSERC Research Chair in Innovative FRP Reinforcement
performance of the GFRP CRCP appeared satisfactory, for Concrete Infrastructure) at Department of Civil
particularly because the crack widths satisfied the AASHTO Engineering, Faculty of Engineering, University of
limiting criterion for crack width as ≤ 1 mm (0.04 in.), which Sherbrooke, QC, Canada, for providing technical data along
is essential in maintaining pavement integrity by securing with numerous testing and reports on FRP reinforcement
adequate aggregate interlock at the crack. Data from this for concrete infrastructure. The author acknowledges
experimental phase will allow for finite-element modeling technical data from Pultrall Inc..
of the CRCP-GRFP slab.
References
1. Ahmed, E., Settecasi, F., and Benmokrane, B. (2014). “Construction
Conclusions and Testing of GFRP Steel Hybrid- Reinforced Concrete Bridge-
The observations and the outcomes from the different field Deck Slabs of Sainte-Catherine Overpass Bridges.” J. Bridge Eng.,
19(6), 04014011.
applications reported in this paper can be summarised
2. American Concrete Institute (ACI). (2001; 2003; 2006). Guide for the
into the following: corrosion resistance is without a doubt
design and construction of concrete reinforced with FRP bars, ACI
the main motive and attraction to use FRP over steel. 440.1R-01;03; 06; 15, Farmington Hills, Mich.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

444 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Building Durable Concrete Infrastructure Using Fibre- Reinforced Polymer (FRP) Bars

3. American Concrete Institute (ACI). (2004). Guide test methods for Association, Rexdale, Ontario, Canada, 44 p.
fiber reinforced polymers (FRPs) for reinforcing or strengthening
10. Canadian Standards Association (CSA). (2000). “Canadian highway
concrete structures, ACI 440.3R-04, Farmington Hills, Mich.
bridge design code.” CAN/CSA-S6- 00, Rexdale, Toronto.
4. Benmokrane, B., Ahmed, E., Dulude, C., and Boucher, E. (2012)
11. Canadian Standards Association (CSA). (2002-12). “Design and
“Design, Construction, and Monitoring OF the First Worldwide Two-
construction of building components with fiber reinforced polymers.”
Way Flat Slab Parking Garage Reinforced with GFRP Bars.” 6th
CAN/CSAS806-02, Rexdale, Toronto.
International Conference on FRP Composites in Civil Engineering.
CD proceeding, CICE 2012, Rome, Italy. 12. Canadian Standards Association (CSA). (2006-Edition 2010).
“Canadian highway bridge design code— Section 16, updated version
5. Benmokrane, B., Eisa, M., El-Gamal, S., Thébeau, D., and El-
for public review.” CAN/CSA-S6-06, Rexdale, Toronto.
Salakawy, E., (2008) “Pavement system suiting local conditions:
Quebec studies continuously reinforced concrete pavement with 13. ISIS Canada (2001; 2006; 2007). “Reinforcing concrete structures
glass fiber- reinforced polymer bars” ACI Concrete international / with fiber reinforced polymers.” ISIS- M03-2001; 2006; 2007, The
November 2008 pp. 34-39. Canadian Network of Centers of Excellence on Intelligent Sensing
for Innovative Structures, Univ. of Winnipeg, Manitoba, Canada.
6. Benmokrane, B., El-Salakawy, E., El-Ragaby, A., and El-Gamal, S.
(2007). Performance Evaluation of Innovative Concrete Bridge Deck 14. Marc Schürch and Peter Jost (2006) “GFRP Soft-Eye for TBM
Slabs Reinforced with Fibre- Reinforced Polymer Bars. Can Jour of Breakthrough: Possibilities with a Modern Construction Material.”
Civil Eng, 34(3): 298–310. International Symposium on Underground Excavation and Tunnelling
2-4 February 2006, Bangkok, Thailand.
7. Benmokrane, B., El-Salakawy, E., El-Ragaby, A., and Lackey, T.
(2006). “Designing and Testing of Concrete Bridge Decks Reinforced 15. Mohamed, H.M., Afifi, M., and Benmokrane, B. (2014). “Performance
with Glass FRP Bars.” Journal of Bridge Engineering, 11(2): 217-229. Evaluation of Concrete Columns Reinforced Longitudinally with FRP
Benmokrane, B., El-Salakawy, E.F., Desgagné, G., and Lackey, T. Bars and Confined with FRP Hoops and Spirals under Axial Load.”
(2004). “Building a New Generation of Concrete Bridge Decks using J. Bridge Eng., 19(7), 04014020.
FRP Bars.” Concrete International, the ACI Magazine, Vol. 26, No. 16. Mohamed, H.M., and Benmokrane, B. (2014). “Design and
8, August, pp. 84-90. Performance of Reinforced Concrete Water Chlorination Tank
8. Benmokrane, B., Wang, P., Ton-That, T. M., Rahman, H., and Robert, Totally Reinforced with GFRP Bars.” ASCE J. Compos. Constr.,,
J. F. (2002). “Durability of glass fiber-reinforced polymer reinforcing 18(1), 05013001-1 – 05013001-11.
bars in concrete environment.” J. Compos. Constr., 6(3), 143–153. 17. NACE International. “Corrosion Costs and Preventive Strategies
Canadian Standard Association (CSA). (2006). Canadian Highway in the United States” Publication No. FHWA-RD-01-156, National
Bridge Design Code. CAN/CSA S6-06, Rexdale, Ontario, Canada, 788. Association for Corrosion Engineers, (www.nace.org).
9. Canadian Standards Association (2010). CAN/CSA S807-10 18. Pultrall Inc, 2012, “Composite Reinforcing Rods Technical Data
“Specification for fibre-reinforced polymers, Canadian Standards Sheet.” Thetford Mines, Canada, www.pultrall.com

Dr. Brahim Benmokrane

Affiliation: P. Eng., FRSC, FACI, FCSCE, FIIFC, FCAE, FEIC
Current Position: Professor of Civil Engineering. Fellow of the Royal Society of Canada
Dr. Benmokrane (FRSC, FACI, FCSCE, FIIFC, FCAE, FEIC) is Professor of Civil Engineering at the Department
of Civil Engineering at the University of Sherbrooke. He has obtained his engineering degree from Swiss
Federal Institute of Technology of Lausanne (Switzerland), and his Ph.D. in civil engineering from University
of Sherbrooke. He was a Project Leader in the Canadian Network of Centers of Excellence on Intelligent
Sensing for Innovative Structures – ISIS Canada – during its two 7-year phases (1995- 2009). Since April
2000, he is a Chair-holder of a Natural Science & Engineering Research Council of Canada (NSERC)
Research Chair which involves research and development of Innovative Composite FRP reinforcement
for concrete structures. In 2009, Professor Benmokrane was awarded a Tier 1 Canada Research Chair in
Advanced Composite Materials for Civil structures. He is an active member in the CSA (Canadian Standard
Association) committees on Design and Construction of Building Structures with FRP, Specification for
Fibre-Reinforced Polymers as internal and external reinforcements (CSA S807, CSA S808) and Canadian
Highway Bridge Design Code (CSA S6; subcommittee S16), where he contributed for developing new Design
Codes and Standards. Professor Benmokrane is an internationally renowned leader on the innovative use
of FRP composite materials in construction. In addition to membership in many professional organizations
such as ACI, CSCE, ASCE, and ASTM, he has published more than 400 technical papers. Over the last twenty
years, he has trained more than 100 graduate students and postdoctoral fellows currently holding positions
in academic and research institutions and in public and private organizations in Canada as well as around the
world. More than 15 of his doctoral students and postdoctoral trainees hold faculty positions in universities
worldwide. He has received numerous awards and distinctions such as the NSERC Synergy Award for
Innovation, the CSA Medal of Merit, the CSCE P.L. Pratley Award, the ConMat Life-Time Achievements
Award, and elected Fellow by the Canadian Academy of Engineering, the ACI, the CSCE, the International
Institute in FRP for Construction, the Engineers Institute of Canada, and the Royal Society of Canada.

Organised by
India Chapter of American Concrete Institute 445
Session 4 B - Paper 4

Experimental Investigations on
Use of Rubber Concrete in Railway Sleepers
A.P. Shashikala*, Anilkumar P. M., George Joseph, Jestin John and Lijith K. P.
Department of Civil Engineering, National Institute of Technology Calicut, Kerala-673601
Corresponding author* (email id): apka@nitc.ac.in

Abstract Introduction
Main objective of the paper is to report the results of Sleepers are members generally laid transverse to
experimental investigations carried out on high strength the direction of rails, on which the rails are fixed and
rubber concrete in railway sleepers. In the history of supported through fasteners. The best chosen material
railways, sleepers were made with wood and wooden for sleeper has been timber since the beginning of railway
composites. Problems related with shortage of wooden constructions but due to improper deforestation methods
and rapid deterioration of wood all over the world, a need
sleepers and incomparable increase in price of wooden
for developing better alternative composites to wooden
composites led to the development of cement concrete
sleeper was realized by Engineers, thus began the era of
and prestressed concrete sleepers. These sleepers concrete sleepers. Indian Railways have mostly swapped
have huge weight and low impact strength. Rubber from wooden sleepers to prestressed concrete sleepers.
concrete composite is developed to overcome the defects Several investigations have been conducted in an attempt
like low ductility, low toughness and brittle failure of to examine the most durable, the strongest and the most
conventional concrete. Rubber particles can be used as a effective composite material for replacing deteriorated
partial replacement for fine aggregate and act as a filler and damaged sleepers. Composite sleepers made
material. These particles can be added to the conventional from waste products are environmental friendly, cost
concrete mixes either as rubber chips or as rubber fibres effective and can be categorized as the green product. It
after proper treatment. Crumb rubber is fine rubber is eco-friendly in the sense that it ensures reuse of the
particles ranging in size from 0.075- 4.75mm made by waste material in an efficient manner. Some of these
re-processing (shredding) disposed automobile tyres. approaches to composite sleepers directed to use fibre
composites as reinforcement for existing railway sleepers
Shredding waste tyres and removing steel debris found
[Humpreys M.F., 2004; Qiao P, 1998; Shokrieh M.M., 2006;
in steel-belted tyres generates crumb rubber particles.
Ticoalu A.N.E., 2008]. Other approaches focused on the
Aim of the present study is to examine the feasibility of replacement of deteriorated sleepers using alternative
using high strength concrete mixes for railway sleepers materials such as polymer concrete, reinforced plastics,
with partial replacement of fine aggregate by crumb rubber and fibre composite material [Lampo R.; Hoger D.I,
rubber particles. Three mix ratios namely M50, M55 and 2000; Jordan R., 1987; Miura S., 1998].
M60 were chosen for the current investigation. Basic
Crumb rubber can be considered as an alternative material
mix proportioning was carried out with the help of IS:
for the development of sustainable concrete composites.
10262-2009 and the suggestions provided in ACI codes Although the workability decreases with increase in
to attain high strength concrete mixes. Experimental rubber aggregate content, the density of concrete get
investigations on the use of high strength concrete with reduced and air content get increased. Strength is
different volume fractions of fine aggregate (0% to 20%) reduced with increase in rubber aggregate volume
with partial replacement of crumb rubber were carried content. The material exhibits enhanced toughness and
out in the laboratory. Investigations on the use of high a slight transition from brittle to ductile failure mode.
strength rubber concrete for railway sleepers include Impact resistance and thermal insulation properties also
static load test and impact load test on sleeper specimens are improved. Plastic cracking and rigidity of the concrete
as per Indian Railway specifications. From the studies, it structures can be reduced by adding crumb rubber. One of
has been observed that rubber concrete composite can be the defects of railway concrete sleeper is the poor impact
used in the fabrication of railway sleepers as it possesses resistance. This is more obvious in high speed train travel,
greatly reducing the life of sleeper. Hence it is necessary
high impact strength and energy absorption.
to modify the conventional high strength concrete used in
Keywords: Compressive strength, concrete, crumb concrete sleepers. Crumb rubber added to concrete can
rubber, impact, railway sleeper, tensile strength. significantly improve the impact resistance of concrete.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

446 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Experimental Investigations on Use of Rubber Concrete in Railway Sleepers

The crumb rubber, fine aggregates, coarse aggregates Rubber

and cement are mixed and pressed in to moulds of railway
Crumb rubber shown in Fig. 1 was used in the concrete
sleepers. This sleeper not only does have the advantages
mix to partially substitute fine aggregates (sand) in various
of light weight, impact resistance and abrasion, but also
percentages by volume. Crumb rubber used in the study
reduces train travel noise and vibration, smooth running
was obtained from a local industrial unit by grinding of
and has a significant damping effect. Aim of the paper is
near to end life tyres that were accumulated in the rubber
to examine the suitability of high strength rubber concrete
waste industry. The nominal size of the rubber aggregate
as a composite material for railway sleeper. Preliminary
ranges between 4.75 mm to less than 0.075 mm with
material study was carried out to finalise the mix for high
fineness modulus 2.72 and bulk density of 670 kg/m3.
strength concrete. Mechanical characteristics of the high
Fig.2 shows the grading curve of rubber aggregates.
strength rubber concrete are assessed based on the
Indian Railway standards.

Materials and Mix Design

Materials
Proper selection of materials for high strength concrete
should be carried out to improve the mechanical and
durability characteristics. Common materials used for
the study are cement, fine aggregate, coarse aggregate,
crumb rubber, super plasticizer, silica fume and water.
Standard test procedures were followed as per Indian
standards to find the properties of constituent materials.
Fig. 1: Crumb rubber particles
Cement
Ordinary Portland cement of 53 Grade, conforming to IS:
12269-1987 (reaffirmed 2004) was used in the present
investigation. Test results on cement used in the present
study were comparable with the requirements as per IS
12269-1987(reaffirmed on 2004).

Fine and Coarse aggregates

River sand passing through 4.75 mm IS sieve conforming
to Grading Zone II of IS: 383-1970 (reaffirmed 2002) was
used as fine aggregates. Crushed stone with a particle
Fig. 2: Grading Curve of Crumb rubber
size less than 20 mm were used as coarse aggregates.
Samples were tested as per IS: 2386-1997 and IS: 383- Super plasticizer
1970 (reaffirmed 2002). Properties of fine and coarse
aggregates used in the study are depicted in Table 1. Naphthalene based Super plasticizer (Conplast SP 430)
was used to obtain the required workability. Table 2
Table1. Properties of fine and coarse aggregates
gives the properties given by the manufacturer Fosroc
Chemicals.
Sl.No. Properties Fine aggregates Coarse
aggregates Table 2. Properties of Super plasticizer
1 Fineness Modulus 2.48 6.49
Sl.No Property Details
2 Specific Gravity 2.42 2.72
1 Specific gravity 1.265-1.280 at 27o C
3 Bulk density 1637 kg/m 3
1585 kg/m 3
2 Chloride content Less than 0.045%
4 Loose density 1496 kg/m 3
1437 kg/m3
3 Air entrainment Less than 2% over control
5 Water absorption 1.21% 0.67%

6 Crushing strength - 24% Silica fume

7 Impact value - 27%

Silica fume is a by-product resulting from the reduction
of high quantity quartz with coal in electric arc in the
8 Abrasion value - 23.2% manufacture of silicon or ferrosilicon alloy. Physical and
chemical properties of silica fume used in the study were

Organised by
India Chapter of American Concrete Institute 447
Session 4 B - Paper 4

comparable with IS 15388:2003 (Indian Standard for Silica

Fume — Specification).

Water
Clean potable water which satisfies drinking standards
was used for the preparation of specimens and its
subsequent curing.

Mix Design
Table 3 gives the details of mix proportions used in the
experimental investigations.

Table 3. Mix Proportion after Number of Trials

Particulars Grade of concrete

M50 M55 M60

Fig. 3. Compression test on concrete cubes
Target Strength (N/mm2) 58.25 63.25 68.25

Cement (kg) 495 501 505 Table 4. Compressive strength of rubber concrete cubes

Fine aggregate for 0% rubber (kg) 684 694 510 Mix Designation 7day 15 day 28 day
Compressive Compressive Compressive
% of Rubber X X X Strength Strength Strength
(MPa) (MPa) (MPa)
Coarse aggregate (kg) 1097 1126 1280
M50R0 40.54 50.57 59.29
Water (l) 158 147 145
M50R5 38.80 45.34 53.20
Super plasticiser (kg) 0.0098 0.0104 0.0136 M50 M50R10 35.31 44.03 51.44
Silica Fume (kg) 34 40.7 38 M50R15 33.13 43.60 49.70
X= Volume fraction of crumb rubber (0%, 5%, 10%, 15% and 20%)
M50R20 29.64 37.49 43.16
As per the Ministry of Railways specifications for pre- M55R0 43.01 54.06 64.96
stressed concrete sleepers, high strength mixes like
M55R5 39.67 49.26 58.42
M55 and M60 should be used for manufacturing concrete
sleepers. Three mix ratios namely M50, M55 and M60 M55 M55R10 37.88 47.97 56.70
were designed for the present investigation. The basic mix
M55R15 35.72 47.08 53.62
proportioning was carried out based on IS: 10262-2009
and the suggestions provided in ACI codes to attain high M55R20 32.26 40.32 47.96
strength concrete mixes. Because of the uncertainties M60R0 46.21 57.99 68.88
in the material properties of ingredients, a trial and
M60R5 42.72 53.63 62.34
error procedure was adopted for the development of
high strength concrete to obtain desired mechanical M60 M60R10 41.42 51.50 60.61
characteristics. M60R15 38.08 47.52 57.55

M60R20 34.88 40.54 50.57

Test on Concrete specimens
Compressive strength of rubber concrete
It is observed that as the percentage of partial replacement
Compressive strength is one of the main parameter based of fine aggregate by crumb rubber increases, the
on which most of other properties of concrete is evaluated. compressive strength reduces. However, the target
Preparation of specimens and testing were done as per IS strength of concrete at different grades were obtained for
516-1959. The specimens of size 150 ×150×150 mm were 10% rubber content as a replacement for fine aggregates.
cast in the required mix proportion and were cured for 7,
15 and 28 days. Compressive strength of the specimens Flexural strength of rubber concrete
were measured using 300T universal testing machine Flexural strength test was conducted on prisms of size
(Fig.3). Table 4 shows the compressive strength of 100 ×100×500 mm at the age of 28 days and conforming
different concrete mixes tested. In the mix notation MNRX, to IS 516-1959. A 40T Universal Testing Machine (Fig.4)
MN denote the specific chosen mix, R shows the presence was used for the test. On the bed of testing machine,
of rubber and X shows the percentage replacement of fine two steel rollers were placed, 38 mm in diameter, on
aggregate with crumb rubber. which the specimen was simply supported over a span

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

448 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Experimental Investigations on Use of Rubber Concrete in Railway Sleepers

of 400mm. The load was applied through two similar Casting of Railway Sleeper Models
rollers mounted at the one third points of the supporting Fig.7 shows the scaled down model of railway sleeper as
span that was spaced at 133 mm center to center. The load per T 2496 drawing of the prototype. The specifications for
was applied without shock and increasing continuously at sleeper models are listed below:
a rate of180 kg/min. The maximum load corresponding
1. Length 917 mm.
to failure load was noted and the flexural strength was
calculated. 2. Gauge length 558 mm
3. Drop weight load of 40 kg.
4. Mould was made of Timber.
5. Drop height of 750mm.
Reinforcement Details: 6mm diameter bars connected in
the form of an equilateral triangle

Fig. 4: Flexural strength test on concrete beams

Fig. 5: Variation in flexural strength

Fig. 7: Scaled sleeper s[ecifications

Casting and curing of concrete sleeper models were

carried out in the similar manner of standard concrete
specimens. Casting followed by a curing period of 28 days
was preferred in the present investigation.

Test on Sleeper Specimens

Static load test
Static load test was carried out to assess the adequacy
of design. Sleepers were loaded gradually (30-40 kN/min)
at a rate of 5t/min upto the specified load of 50t, which
will be retained at this level for one minute for observing
cracks, if any. For the purpose, a crack is defined as one
which is barely visible to the naked eye and is at least 15
Fig. 6: Failure specimens after flexural test mm long from the tension edge of the sleeper. However,
if crack appears at a load smaller than the specified load,
Flexural strength of rubber concrete for different Grades of that value shall be recorded. Deflection measurement and
concrete is shown in Fig.5 and the failure pattern is shown observation of the cracks under rail seat shall be done at
in Fig. 6. Reduction in modulus of rupture of high strength the interval of 5t load e.g. at 5t, 10t, 15t etc. Acceptance
concrete with 20% partial replacement of fine aggregate criterion is the absence of visible cracks on the outer
with crumb rubber was found to be 21%, 22% and 24% for surface of the sleepers on holding the 50t load for 5
M50, M55 and M60 mix proportions respectively. minutes.

Organised by
India Chapter of American Concrete Institute 449
Session 4 B - Paper 4

Fig.8 shows the experimental setup for the static load

test on sleeper models. Test results are depicted in Table
5 which shows that as the rubber content increases,
the maximum static load carrying capacity of sleeper
decreases, but the maximum deflection increases. This
shows the ductile behaviour of high strength rubber
concrete imparted by the crumb rubber particles.
Fig.9 shows the failed specimens after the static load
test. The crack pattern was observed to originate from
the tension bottom and propagated upwards in a vertical
manner.
Fig. 9. Specimens after static load failure

be subjected to during the passing of train as well as during

derailment. The design load and dropping distances for
the impact test on railway sleepers given by the Ministry
of railways specifications gave due considerations for
derailment conditions. The test scheme facilitates the
dropping of a wheel on sleeper placed at 30° slope to
horizontal plane at following two locations;
294 mm away from centre line of rail toward centre and (ii)
200 mm away from sleeper end.
Wheel drop of weight 500 kg from a height of 750 mm is
dropped on the sleeper specimen. As per the standard
Fig. 8.Static loading setup specifications, the criterion for success is the failure to
develop visible cracks even after two consecutive drops.
Table 5. Static load test results on sleeper models
As per the scaled models, the distance can be modified
Mix Designation Maximum Maximum Deflection into (i) 100 mm away from centre line of rail toward centre
load before crack formation and (ii) 66.67 mm away from sleeper end. The wheel drop
(t) (mm)
was reduced to 40 kg. Test set up for impact test is shown
M50RX M50R0 6.12 6.91 in Fig. 10.
M50R5 5.89 7.01

M50R10 5.25 7.45

M50R15 4.32 7.92

M50R20 4.01 8.00

M55RX M55R0 7.21 6.55

M55R5 6.91 6.76

M55R10 6.4 7.00

M55R15 6.21 7.32

M55R20 5.98 7.54

M60RX M60R0 7.91 6.01

Fig.10. Test Setup for Impact Testing
M60R5 7.63 6.32
Effect of crumb rubber on impact test on sleeper
M60R10 7.21 6.45
models
M60R15 6.98 6.87
A wheel load of 40 kg was used to determine the impact
M60R20 6.31 7.01
resistance of sleeper. Table 6 shows the Impact strength
of the sleeper models and the nature of failure of the
Impact Test
specimens are shown in Fig. 11. All specimens made
Impact load test is performed to assess the shock with crumb rubber did not show any crack after two
absorption capacity of the sleeper, which the sleeper will blows, which shows that it is suitable for railway sleeper.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

450 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Experimental Investigations on Use of Rubber Concrete in Railway Sleepers

Percentage increase in impact strength of high strength

concrete with 20% partial replacement of crumb rubber
was found to be 133%, 150% and 160% for M50, M55 and
M60 mix ratios respectively.

Table 6. Impact strength of sleeper models

Mix Designation Number of blows Impact strength

(Nm)
Fig.12: Crack pattern under impact loading
Initial Ultimate Initial Ultimate
Crack Crack Crack Crack
rubber concrete with different volume fractions (5%, 10%,
M50RX M50R0 3 4 882.9 1177.2
15% and 20%) of crumb rubber as partial replacement of
M50R5 5 7 1471.5 2060.1 fine aggregates. Three mix ratios namely M50, M55 and
M60 were considered in the present investigation. Based
M50R10 6 9 1765.8 2648.7
on the analysis of experimental results, the following
M50R15 6 12 1765.8 3531.6 conclusions were arrived at:
M50R20 7 13 2060.1 3825.9 1. The 28 day compressive strength of concrete
was found to decrease by 10 to 27% when the fine
M55RX M55R0 4 5 1177.2 1471.5
aggregate was replaced by 5 to 20% of crumb rubber
M55R5 5 7 1471.5 2060.1 by volume irrespective of the mix ratios. However, for
10% replacement of fine aggregate by crumb rubber,
M55R10 7 10 2060.1 2943.0
the compressive strength was found to be within the
M55R15 8 15 2354.4 4414.5 limits specified by Ministry of Railways.
M55R20 10 18 2943.0 5297.4 2. It was observed that as the rubber content increases,
the static load carrying capacity of the sleeper reduces
M60RX M60R0 5 6 1471.5 1765.8
by 8-14%. But, the maximum deflection was found to
M60R5 6 9 1765.8 2648.7 increase 5 to 15% with increase in rubber content by 5
to 20% making the rubber concrete less brittle.
M60R10 8 14 2354.4 4120.2
3. Impact strength of concrete railway sleepers was
M60R15 10 13 2943.0 3825.9
found to increase with increase in rubber content. The
M60R20 13 20 3825.9 5886.0 impact strength was enhanced by 60 to 160% for 5 to
20% replacement of fine aggregate by crumb rubber.
The amount of energy required for the ultimate crack
was found to increase with the percentage increase in
rubber content.
References
1. Alex M Remennikov, Sakdirat Kaewunruen, 2007, Resistance of
Railway Concrete Sleepers to Impact Loading, 7th International
Conference on Shock & Impact Loads on Structures, 17(19), 489-
496.
2. G. Kumaran, Devdas Menon and K. Krishnan Nair, 2003, Dynamic
Studies Of Rail Track Sleepers In A Track Structure System,
Journal of Sound and Vibration, 268, 485–501.
Fig. 11: Specimens after impact failure 3. Hoger DI., 2000, Fibre composite railway sleepers. University of
Southern Queensland, Toowoomba, Queensland, Australia.
Amount of energy required for the ultimate crack from
4. Humpreys MF, Francey KL., 2004, An investigation into the
initial crack increases with the percentage increase in rehabilitation of timber structures with fibre composite materials.
rubber content. Fig. 12 shows the typical failure pattern. Australia: Queensland University of Technology.
5. Indian Railway standard specification for pre-tensioned
Conclusions prestressed concrete sleepers for broad gauge and meter
gauge, serial no. T - 39 (fourth revision Aug 2011).
Experimental investigations were carried out to study
6. IS 10262:2009, Concrete Mix Proportioning – Guidelines, Bureau
the mechanical characteristics of high strength rubber
of Indian Standards, New Delhi.
concrete sleepers in order to assess the suitability of rubber
7. IS 2386 (Part III):1963(Reaffirmed 1997), Methods of Test for
concrete in Railway sleepers as per Indian standards, ACI Aggregates for Concrete, Bureau of Indian Standards, New Delhi.
standards and Indian railway standards. Properties of
8. IS 516:1959, Methods of Tests for Strength of Concrete, Bureau of
high strength concrete was compared with high strength Indian Standards, New Delhi.

Organised by
India Chapter of American Concrete Institute 451
Session 4 B - Paper 4

9. IS 383:1970, Specification for Coarse and Fine Aggregate from 14. Miura S, Takai H, Uchida M, Fukuda Y., 1998, The mechanism of
Natural Sources for Concrete, Bureau of Indian Standards, New railway tracks. Jpn Rail Transport Rev, :38–45.
Delhi.
15. Shokrieh MM, Rahmat M., 2006, On the reinforcement of concrete
10. IS 456:2000, Ordinary and Reinforced Concrete- code of Practice, sleepers by composite materials. Compos Struct, 76:326–37.
Bureau of Indian Standards, New Delhi.
16. Ticoalu ANE., 2008, Investigation on fibre composite turnout
11. IS 12269:1987(Reaffirmed 1999), Specification for 53 Grade Ordinary sleepers. Master of engineering dissertation, University of Southern
Portland cement, Bureau of Indian Standards, New Delhi. Queensland, Toowoomba,Queensland, Australia.
12. Jordan R., 1987, Testing plastic railway sleepers, Testing 17. Olli Kerokoski, Antti Nurmikolu, Tommi Rantala, 2012, Loading Tests
Research Laboratory, UK, <http://www.trl.co.uk/content/ main. of Concrete Mono-block Railway Sleepers, 1-14.
asp?pid=164> [viewed 03.08.08].
18. Qiao P, Davalos JF, Zipfel MG., 1998, Modelling and optimal design
13. Lampo R. Recycled plastic composite railroad crossties.<http:// of composite reinforced wood railroad crosstie. Compos Struct,
www.cif.org/Nom2002/Nom13_02.pdf> [viewed 18.08.2014]. 41:87–96.

Dr. SHASHIKALA A P
Dr. Shashikala, presently a Professor in Department of Civil Engineering at NIT, Calicut, India has completed
her Ph.D from IIT, Chennai in 1996. She has teaching and research experience of 32 years and her research
interests include; Development of alternate materials, Seismic Performance of buildings, Wave modeling
and prediction, Floating body analysis, Multi-body Offshore Structures and Response analysis of ships in
viscous flow.
She has received ‘National Award for Best M. Tech. Thesis in Civil Engineering 2010’, instituted by Indian
Society for Technical Education, India and ‘’Maritime Award 2005 Instituted by Government of India,
Ministry of Shipping, Road Transport and Highways.
She is Member of American Concrete Institute, Member of Indian Concrete Institute, Fellow Member of
Institution of Engineers (India), Life Member of Indian Society for Technical Education and Member of
Computer Society of India. She has 36 publications to her credit.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

452 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chloride induced corrosion of steel bars in alkali activated slag concretes

Chloride induced corrosion of steel bars

in alkali activated slag concretes
Qianmin Ma
Faculty of Civil Engineering and Mechanics, Kunming University of Science and Technology, China
Sreejith V. Nanukuttan
School of Planning, Architecture and Civil Engineering, Queen’s University Belfast, UK
P. A. M. Basheer
School of Civil Engineering, University of Leeds, UK
Yun Bai
Department of Civil, Environmental & Geomatic Engineering, University College London, UK
Changhui Yang
College of Material Science and Engineering, Chongqing University, China

Abstract composition has a significant influence on chloride

induced corrosion (Broomfield, 2007), it is essential
The studies on chloride induced corrosion of steel bars
to understand the influence of the pore solution of AAS
in alkali activated slag (AAS) concretes are scarcely
concretes on the corrosion of the steel bars embedded
reported in the past. In order to make this issue clearer
in an attempt to address any potential shortcoming in
and compare the corrosion performance of AAS with
knowledge for AAS concretes.
Portland cement (PC) counterpart, an investigation was
carried out and the results are reported in this paper. During the exposure of concrete to a chloride
Corrosion properties were assessed with the help of environment, the accompanying cations (Buenfeld et
corrosion rate, electrical resistivity and pore solution al., 1998; Zhang and Gjorv, 1996; Tang, 1999), the ionic
chemistry. It was found that: (i) steel corrosion resistance exchanges between chlorides and the ions existing in
of the AAS concretes was comparable or in some cases concrete (Buenfeld et al., 1998) and the binding of both
even worse than that of PC concrete under intermittent chlorides and accompanying cations (strongly influenced
chloride ponding regime; (ii) the corrosion behaviour by pore solution composition and alkalinity) (Broomfield,
of the AAS concretes was significantly influenced by 2007; Zhang and Gjorv, 1996; Nachbaur et al., 1998;
ionic exchange, carbonation and sulphide concentration; Tritthart, 1989), may significantly influence the diffusion of
(iii) the increase of alkali concentration of the activator chloride ions into the concrete achieving to the surface of
generally reduced corrosion rate, and a value of 1.5 was steel bars. As stated previously, AAS has a different pore
found to be an optimum modulus for the activator for solution chemistry compared to that of PC. Therefore, the
improving the corrosion resistance. influence of pore solution chemistry on chloride diffusion
could be different in the case of AAS compared to PC.
Keywords: alkali activated slag, electrical resistivity, pH
profile, sulphide concentration, rate of corrosion Presence of sulphides in pore solution of concrete can
significantly reduce the redox potential of the pore
Introduction
solution. Redox potential measurement is a reflection
It is widely accepted that alkali activated slag (AAS), of oxidation and reduction activities (Schuring, 1999).
which has a potential to 100% replace Portland cement The reduction of the redox potential is a result of the
(PC) as the cementitious material, is not only highly increase of reduction atmosphere, which would protect
desirable to environmentally friendly reduce the high the embedded steel from its oxidation to a certain extent
carbon footprint of concrete, but also could satisfy the to reduce the corrosion rate of the steel. Besides, the
strength and durability requirements (Shi et al., 2006). oxidation of chemically reduced sulphides would form
However, the physical and durability characteristics of elemental sulphur to deposit in the pores of the damaged
AAS concretes depend on both the physical and chemical passive film allowing the corroded steel to regain
characteristics of the slag and the type of activator used passivation (Shoesmith et al., 1978). However, Shoesmith
(Shi et al., 2006). It is known that when sodium silicate et al. (1978) suggested that sulphides also play a role to
solution is used as the activator, the alkali concentration break down the passive film on the surface of steel
(Na2O%) and the modulus of the activator (silica modulus, and then the corrosion of the steel would be initiated.
Ms) influence some of the properties of concrete (Shi Sulphides are rich in slag, which is why the redox potential
et al., 2006). It has been reported that AAS concretes of slag cement can be lower than -400mV compared to
have a different pore solution composition compared PC, which is normally between 200mV to 300mV (Glasser,
to that of PC (Puertas et al., 2004). As pore solution 1997). Therefore, it could be anticipated that the corrosion

Organised by
India Chapter of American Concrete Institute 453
Session 4 B - Paper 5

behaviour of the steel in AAS would be different from that Sodium silicate solution (or commonly known as water
in PC. glass, WG) with Na2O% of 12.45 and SiO2% of 43.60, which
is commercially available as ‘Crystal 0503’ from Charles
Experimental Programme Tennant and Co. (NI) Ltd., UK was used as the activator
for GGBS. Industrial grade sodium hydroxide powder with
A programme of investigation was developed with the
a purity of 99% supplied by Charles Tennant and Co. (NI)
objective of studying the initiation and the propagation of
Ltd., UK was used to adjust the Ms to the required values.
corrosion of embedded steel when these concretes were
exposed to an intermittent chloride ponding environment. A barium based retarder ‘YP-1’® (Yang and Pu, 1993)
Full details of the experimental programme shall be given was used in the AAS concretes to delay the setting
in sub- sections below, but a summary of the experimental time. The retarder was dry-blended with GGBS before
variables and the tests shall be given next. Twelve AAS mixing. A polycarboxylic polymer based superplasticiser
concretes with alkali concentrations (Na2O% of mass (commercially known as CHEMCRETE HP3 and
of slag) of 4, 6, 8 and modulus (Ms) of sodium silicate manufactured by Larsen, Northern Ireland, UK) with
solution activator of 0.75, 1.00, 1.50, 2.00 were studied. a water content of 40% was used in the PC concrete
Corrosion of the embedded steel bars was quantified by mix. The water content of the superplasticiser was taken
measuring their gravimetric mass loss after the cyclic into account whilst determining the mixing water content.
chloride ponding test regime. Pore solution composition Crushed basalt from local sources in Northern Ireland
of the concretes was studied in an attempt to explain the with size fractions of 20mm and 10mm combined in
a ratio of 1:1 was used as the coarse aggregate. Natural
physical and chemical changes that occur during the
sand with fineness modulus of 2.53 was used as the
chloride transport and the corrosion. Effects of Na2O%
fine aggregate. Properties of both the fine and coarse
and Ms on the corrosion behaviour of the concretes
aggregates are reported in Table 2. Water from the mains
were also investigated.
water supply was used to mix and cure concretes.

Materials Table 2. Specific gravity under saturated surface dry (S. S. D.)
Ground Granulated Blast-furnace Slag (GGBS) confirming condition and 1-hour water absorption of the aggregates

to BS EN 15167-1 (2006) was used to manufacture all Specific gravity 1-hour water
the AAS concretes. The GGBS was supplied by Civil and (S. S. D.) absorption (%)
Marine Ltd., UK. Class 42.5N PC (CEM-I) conforming to 20mm Coarse aggregate 2.69 3.44
BS EN 197-1 (2000), from Quinn Group Ltd., Northern 10mm Coarse aggregate 2.80 2.12
Ireland, UK was used to manufacture the PC control
Fine aggregate 2.73 0.75
concrete. The chemical compositions and physical
properties of both GGBS and PC are reported in Table 1.
Mix proportions
Table 1. Chemical compositions and Twelve AAS concretes mixes with Na2O% of 4, 6, and 8 and
physical properties of the GGBS and the PC Ms of WG of 0.75, 1.00, 1.50 and 2.00 were investigated
Chemical composition (%) Physical properties in this research. The total binder content, which is the
sum of GGBS and solid component of the WG, was kept
GGBS PC GGBS PC
constant at 400kg/m3 for all mixes. The water/binder
CaO 39.4 61.3 Specific surface area 527 386 (W/B) was constant at 0.47 for all the AAS concretes. The
(m2/kg)
SiO2 34.3 23.0 water content in the WG was taken into account while
determining the mixing water content. The retarder was
Al 2O3 15.0 6.15 Specific Gravity 2.90 3.16
added at a dosage of 0.3% of the mass of GGBS. This
MgO 8.00 1.80 dosage of the retarder was found to result in acceptable
Sulfide 0.80 - levels of the setting times of the AAS mixes.
SO3 - 2.50 For the purpose of comparison, one PC concrete mix
was manufactured with the total binder content as that
TiO2 0.70 -
of the AAS concretes. A W/B of 0.42 was determined
MnO 0.50 - for the PC concrete to guarantee its performance in
Na2O 0.45 0.22 exposure classes XS3 and XD3 as defined in BS EN 206-
1 (2000). The use of the superplasticiser of 0.4% of mass
Fe2O3 0.40 2.95
of the cement allowed the PC concrete to just achieve
K 2O 0.38 0.68 the minimum value of the slump class of S2 (50mm)
Cl 0.02 0.01 specified in BS EN 206-1 (2000). It may be noted that the
W/B of the AAS concretes was 0.47, not the same as that
LOI 0.05 1.40
used for the PC mix. This was essential to ensure that the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

454 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chloride induced corrosion of steel bars in alkali activated slag concretes

AAS concretes met the slump requirement. A reduction fresh concrete was transferred into the moulds in two
of W/B to 0.42 for AAS concretes to match with that of layers. Each layer was compacted by placing the moulds
PC concrete was found to result in harsh mixes which on a vibrating table until no air bubbles appeared on the
are not workable. The impact of this on the strength and surface of the concrete. The surface of the concrete
other properties studied in this research is highlighted in was finished by a metal float. After casting, moulds
the discussion of results later on in this paper. were covered with thin polythene sheets to minimise
The last aspect of the mix design was the ratio of fine evaporation of water from the surface of concrete.
aggregate content to total aggregate content. This was Approximately one hour after the concrete surface
kept at 36% for both the AAS and the PC concretes. became stiff, the moulds were covered with a layer of
Several trials were carried out before arriving at these previously wetted hessian and then covered with a layer
mix proportions to ensure that the resulting concretes of polythene sheet. The samples were stored in this
satisfied the specifications for both the exposure classes condition for three days. The hessian was maintained
and the slump class considered. in moisture condition at every six hours. At the end of
this period, the concrete specimens were demoulded,
Preparation of samples wrapped in wet hessian and plastic bags and stored in a
constant temperature room at 20±1°C for 91 days. The
Six 250×250×110mm blocks (three with embedded steel hessian was checked for the moisture condition at
bars as shown in Figure 1, while the other three had no every two weeks and rewetted if needed.
steel) and nine 100×100×100mm cubes for each mix
were cast for the tests to be detailed in the next section.
Figure 1 also shows the stainless steel bars (hereafter
Test procedures
referred to as electrodes) used for measuring the Slump and compressive strength
electrical resistivity of the concretes during the cycling
The slump test specified in BS EN 12350-2 (2009) was
ponding/drying regime. The top bar served as the anode
carried out on the fresh concretes and the compressive
and the three bottom bars together acted as the cathode
strength test according to BS EN 12350-3 (2009) was
for electrochemical corrosion measurements. Before
carried out on the 100mm cubes at 3, 28 and 91 days of
embedding the steel bars, they were cleaned first with a
age.
wire brush and then with a dry cleaning cloth to remove
any rust and debris, at which stage they were weighed. Corrosion of the embedded steel bars
As shown in Figure 1, the stainless steel electrodes One month before the test age of 91 days, the concrete
were embedded at four different depths, viz. 15mm, blocks with the embedded steel bars (see Figure 1)
25mm, 35mm and 45mm, from the ponding surface in were conditioned at a constant temperature of 23(±3)°C
a staggered arrangement to facilitate the distribution of and relative humidity of around 55% for a duration of
both coarse and fine aggregates around the electrodes. two weeks. The epoxy resin was applied onto all of the
The electrodes were sleeved for the whole length with surfaces of the blocks in three layers except the ponding
a heat-shrink tube except for a middle region of 50mm surface and the surface opposite to the ponding surface.
before embedding in the blocks so that any change The blocks were then stored at the above condition for
in electrical resistance measured from the pair of another two weeks. During the whole conditioning and
electrodes wound be due to the changes in the middle
testing periods, the blocks were supported by two timber
region.
strips with thickness of above 13mm to allow air flow
Both the blocks and the cubes were cast by following the under the blocks. Approximately 200ml of NaCl solution
procedure given in BS 1881-125 (1986). After mixing, the with concentration of 0.55M (≈31.85g/l) was used to pond

Fig. 1: Diagrammatic representation of the concrete block for the corrosion tests

Organised by
India Chapter of American Concrete Institute 455
Session 4 B - Paper 5

the blocks for 1 day followed by 6 days for drying. This an Inductively Coupled Plasma-Optical Emission Mass
cycle of intermittent chloride intermittent ponding was Spectrometer (ICP-MS) technique. The concentration of
continued until the end of the test (250+ days). Before S2- is discussed in this paper due to its specific relevance
each ponding cycle, the electrical resistance of the to the topic of this paper. However, the full data on the
concrete cover was measured at the depths of 15mm, concentration of the other ions are available somewhere
25mm, 35mm and 45mm from the surface of the else (Ma, 2013).
concrete block with the help of embedded stainless steel
electrodes shown in Figure 1. Results and Discussion
At the end of the test, the blocks were profile ground Slump and compressive strength
to obtain concrete dust from different depths of 3mm,
The slump and the compressive strength of the concretes
6mm, 9mm, 12mm, 15mm, 20mm, 25mm, 30mm,
are reported in Table 3. The compressive strength results
35mm, 40mm and 45mm from the exposed surface.
reported are the average values from three cubes.
The concrete dust was dissolved in deionised water in
accordance with RILEM TC 178-TMC recommendations Table 3.
(2002) to measure pH value of the suspension. After the Slump and compressive strength (± standard deviation) of the
profile grinding, the blocks were split opened, the anodic concretes (retarder of 0.3% by mass of slag was used for AAS
mixes while superplasticiser of 0.5% by mass of cement was
steel bars were taken out, scrubbed with wire brush, used for PC mix)
wiped with a dry cloth and weighed to determine the mass
Mix No. Slump Compressive strength (MPa)
loss caused by the corrosion. The corrosion rate (mm/ (Na2O%-Ms) (mm)
3d 28d 91d
year) of the steel bars was calculated using the following
equation: 4%-0.75 55 22.3±0.1 44.7±0.2 46.4±1.0
4%-1.00 55 21.8±0.1 46.7±1.0 55.6±0.6
mm 1000 $ m ....................................1
corrosion rate ( year ) = A $ t $ t 4%-1.50 55 1.7±0.0 49.5±0.2 52.6±2.3
4%-2.00 55 1.4±0.0 33.3±0.4 44.1±0.1
where m is the mass loss in gram, ρ is the density of steel
6%-0.75 65 31.7±0.7 47.3±0.0 51.8±2.4
of 7.87g/cm3, A is the complete surface area of the steel
6%-1.00 65 37.3±0.2 53.6±0.0 59.1±0.8
attacked by corrosion in mm2 and t is the test duration in
6%-1.50 65 20.3±0.7 60.8±0.1 67.4±2.6
year.
6%-2.00 75 8.0±0.0 59.6±0.2 68.7±2.1
Pore solution expression 8%-0.75 70 32.3±0.0 51.9±0.1 56.2±0.2
One day before the test age of 91 days, cores with diameter 8%-1.00 105 32.7±2.4 53.6±0.1 67.1±0.6
of 60mm were cut from the concrete blocks containing 8%-1.50 145 34.1±0.7 59.3±3.2 70.5±2.5
no steel bars. A slice with a thickness of 10mm from the 8%-2.00 180 11.7±0.2 55.4±0.2 65.0±0.8
casting surface (trowel finished face) was cut off, the rest PC 50 35.4±1.2 58.9±1.8 66.3±2.3
was kept for carrying out the test. The vacuum saturation
regime similar to that of NT BUILD 492 (1999) was used The slump values of the concretes show that all the mixes
to precondition the cores. In NT BUILD 492 after the had a slump value greater than 50mm. The principle of
application of the vacuum, saturated Ca(OH)2 solution is mix design for the AAS concretes consisted of using a fixed
introduced into the container. However, it was considered W/B whilst meeting the minimum slump requirement for
that this is likely to lead to leaching of ions from the their use in chloride environments, such as S2 specified
samples. Therefore, in this research, after the application in BS 8500-1 (2006) for marine environments. The mix
of the vacuum, the vacuum was released, samples were design for the PC reference concrete also followed the
wrapped in deionised water saturated hessian and placed same principle. Therefore, it can be seen that the slump
back in the container. The vacuum was again applied. results of all the 13 mixes met the minimum requirement
The purpose of this change in saturation regime was to for S2 slump class, which is 50mm. It is also found that
prevent calcium hydroxide solution used for saturation the slump values of the AAS concretes increased with
affecting the pore solution chemistry of the concrete the increase of Na2O%. This is in agreement with the
samples as well as to minimise the loss of ions from the results reported by Allahverdi et al. (2010) and Karahan
concrete samples. The pore solution within the cores and Yakupoglu (2011). The plasticising effect provided by
was extracted by using a specialist pore fluid expression Na2O component is considered to be responsible for this
(Allahverdi et al., 2010).
device when the cores were subjected to pressures up to
300 tonnes. Once the solution was collected, pH and The compressive strength of the concretes at the ages of
conductivity of the solution were measured immediately 3, 28 and 91 days of age reported in Table 3 shows that
by using a pH meter and a conductivity meter, the presence of the retarder had some detrimental effect
respectively. The concentration of Na+, Ca2+, Mg2+, Al3+, Si4+ on compressive strength in the case of AAS concretes,
and S2- in the solution was analysed subsequently by using particularly at the age of 3d. The mix design for the PC

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

456 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chloride induced corrosion of steel bars in alkali activated slag concretes

reference concrete was to meet the requirement for (Al-Otaibi, 2008). From the electrical resistance results
XS3 and XD3 exposure environments specified in BS EN of the cover concrete (shown in Figure 4) it can be seen
206-1 (2000). From the compressive strength results at that, unlike the PC concrete, the electrical resistance of
the age of 28 days it can be seen that not all the AAS the AAS concretes at the first layer generally increased
concretes achieved the required strength of 58MPa for at the slowest rate compared to the electrical resistance
the two exposure classes. However, most of the AAS at the other depths. This confirms the poor condition of the
concretes met the strength requirement of 50MPa for surface of the AAS concretes.
the exposure classes XS1, XD1 and XD2. It should be
noted that the W/B used for the AAS concretes (0.47) was
higher than that for the PC concrete (0.42). Therefore, an
improved compressive strength could be expected if a
lower W/B was used for the AAS concretes. However,
this would require the development of a superplasticiser
that works with AAS to improve the workability.

Corrosion of the embedded steel

In the authors’ work published earlier (Ma et al., to
be published), it was found that the AAS concretes
had better pore structure than the PC concrete.
Consequently, the former had lower chloride diffusivity.
Similarly, their resistance to the ingress of the other
aggressive substances such as water and O2 should also Fig. 3: pH profiles determined at the end of the chloride ponding
be favourable and a low corrosion rate of the embedded exposure regime.
steel bars was anticipated. However, the corrosion rate
of the steel bars in the AAS concretes was comparable From Figure 5 it can be seen that except AAS concretes
to that of the PC concrete or even worse, as shown in with Ms of 2.00, the corrosion rate of the steel bars
Figure 2. in the AAS concretes decreased with the increase of
Na2O% from 4 to 6, and then was nearly constant when
the Na2O% increased further to 8. Pore structure of AAS
becomes better with the increase of Na2O% (Karahan and
Yakupoglu, 2011; Al-Otaibi, 2008) to resist the ingress of
the aggressive substances. The oxidation of sulphides
is known to reduce the corrosion extent of the steel
bars (Shoesmith et al., 1978). The lower sulphides
concentration in the AAS concretes with Na2O% of 8
(see Table 4) may have resulted in the high corrosion
rate of the embedded steel bars. The steel bars in the
AAS concretes with Ms of 1.50 generally gave the lowest
corrosion rate as shown in Figure 6. Both the higher
sulphides concentration (see Table 4) and the lower
chloride diffusivity (Ma et al., to be published) could have
contributed to the lower corrosion rate.
Fig. 2: Corrosion rates calculated from the gravimetric mass
loss of steel at the end of the exposure regime
The reason for the comparable or even worse performance
of the steel bars in the AAS concretes could be attributed
to the outward diffusion of ions from the concretes into
the exposure solution during the intermittent chloride
ponding. It is considered that the outward diffusion of
alkali materials may have resulted in the reduction of the
pH values (see Figure 3). Without the buffering of Ca(OH)2
in the AAS concretes, the continuous diffusion could
have resulted in the dissolution of the binder. Besides,
carbonation may have occurred during the drying period,
which could also have resulted in the disintegration of
the binder. This is particularly true for AAS concrete Fig. 5: Effect of Na2O% on the corrosion rate

Organised by
India Chapter of American Concrete Institute 457
Session 4 B - Paper 5

Fig. 4: Change in the ratio of electrical resistance during chloride ponding test (R(t,x) is the resistance at any time ‘t‘for each depth
(x), R0 is the resistance at time ‘0’ for the depth of 45mm)

Relationship between the steel corrosion rate and the the embedded steel in concrete is also dependent on
chloride diffusivity the availability of oxygen and water, and concentration
of sulphides in the pore solution of the concrete; and (2)
It was expected that corrosion rate of the embedded absorption occurs in the chloride intermittent ponding
steel bars would decrease with the reduction of chloride regime which is not the case for the immersion test and
diffusivity. However, such correlation does not exist for consequently the chloride transport in these two cases
the AAS concretes, as shown in Figure 7. The possible will be different. The disintegration of the binder at the
reasons for the lack of any correlation are as follows: surface of the AAS concretes may also have contributed
(1) in addition to chloride diffusivity, corrosion rate of to such a difference.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

458 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chloride induced corrosion of steel bars in alkali activated slag concretes

Table 4. Free sulphide concentration in the pore solution of the the increase of the concentration of sulphides due to the
concretes after 3 months of curing reducing environment provided by sulphides.
Mix No. S (ppm) Mix No. S (ppm)
(Na2O%-Ms) (Na2O%-Ms)
4%-0.75 2458 6%-0.75 5661
4%-1.00 1953 6%-1.00 6210
4%-1.50 3786 6%-1.50 6245
4%-2.00 4348 6%-2.00 6292
8%-0.75 664.0 PC 329.6
8%-1.00 590.0
8%-1.50 618.3
8%-2.00 608.0

Fig. 8: Relationship between the corrosion rate and the sulphide

concentration

Conclusions
On the basis of the mixes studied and test methods
applied for the characterisation of chloride induced
corrosion of the steel bars in the AAS concretes, the
following conclusions have been drawn:

1. The corrosion rate of the steel bars in the AAS

concretes was similar to that observed in the PC
concrete or in some cases even worse under the
intermittent chloride ponding regime. It was noted that
the binder in cover of concrete disintegrated during
Fig. 6: Effect of Ms on the corrosion rate
the ponding test possibly due to the continuous loss
of alkali materials and the carbonation.

2. The corrosion rate of the steel bars in the AAS

concretes was significantly influenced by the free
sulphide concentration.

3. An increase in Na2O% from 4 to 6 results in a

decrease in corrosion rate of the steel bars in the AAS
concretes, but the corrosion rate reduced with the
further increase of Na2O% to 8. Ms of 1.50 is optimum
to give the lowest corrosion rate. In addition to the
hydration of AAS, the additional effect of sulphides
concentration was thought to be the reason for such
a trend for the corrosion rate of the embedded steel
bars.
Fig. 7: Relationship between the corrosion rate and the non-
steady state chloride diffusion coefficient Acknowledgement

Relationship between the steel corrosion rate and the The authors express their thanks to the sponsorship
sulphide concentration by UK-China Science Bridge project (EPSRC/G042594/1),
China Scholarship Council studentship and internal
From Figure 8 it can be seen that there is a good funds provided by Queen’s University Belfast. The slag
correlation between the corrosion rate of embedded steel used in this research was supplied by Civil and Marine
and the concentration of sulphides in the pore solution Ltd. Comments and suggestions by Prof. Luping Tang
for the concretes studied. As expected (Shoesmith et of Chalmers University of Technology are also greatly
al., 1978), the corrosion rate generally decreased with acknowledged.

Organised by
India Chapter of American Concrete Institute 459
Session 4 B - Paper 5

References 15. Ma, Q., “Chloride transport and chloride induced corrosion of steel
reinforcement in sodium silicate solution activated slag concrete”,
1. Al-Otaibi, S., “Durability of concrete incorporating GGBS activated
PhD thesis, Queen‟s University Belfast, UK, 2013
by water-glass”, Construction and Building Materials, V.22, No.10,
2008, pp.2059-2067 16. Ma, Q., Nanukuttan, S. V., Basheer, P. A. M., Bai, Y. and Yang, C.,
“Chloride transport and the resulting corrosion of steel bars in
2. Allahverdi, A., Shaverdi, B. and Najafi K. E., “Influence of sodium
alkali activated slag concretes”, Materials and Structures, to be
oxide on properties of fresh and hardened paste of alkali-activated
published
blast-furnace slag”, International Journal of Civil Engineering, V.8,
No.4, 2010, pp.304-314 17. Nachbaur, L., Nkinamubanzi, P.-C., Nonat, A. and Mutin, J.-C.,
“Electrokinetic properties which control the coagulation of silicate
3. Broomfield, J. P., “Corrosion of steel in concrete: understanding,
cement suspensions during early age hydration”, Journal of
investigation and repair”, second edn., London: Taylor & Francis,
Colloid Interface Science, V.202, 1998, pp.261-268
2007
18. NT BUILD 492, “Concrete, mortar and cement-based repair
4. BSI, “Testing concrete. Methods for mixing and sampling fresh
materials: chloride migration coefficient from non-steady-state
concrete in the laboratory”, BS 1881 Part 125, 1986
migration experiments”, 1999
5. BSI, “Cement. Composition, specifications and conformity criteria
19. Puertas, F., Fernandez-Jimenez, A. and Blanco-Varela, M. T., “Pore
for common cements”, BS EN 197 Part 1, 2000
solution in alkali-activated slag cement pastes. Relation to the
6. BSI, “Concrete. Specification, performance, production and composition and structure of calcium silicate hydrate”, Cement
conformity”, BS EN 206 Part 1, 2000 BSI, “Concrete. Complementary and Concrete Research, V.34, 2004, pp.139-148
British Standard to BS EN 206-1 – Part 1. Method of specifying
20. RILEM TC 178-TMC, “Testing and modelling chloride penetration
and guidance for the specifier”, BS 8500 Part 1, 2006
in concrete. Analysis of water soluble chloride content in concrete
7. BSI, “Ground granulated blast furnace slag for use in concrete, Recommendation”, Materials and Structures, V.35, 2002, pp.586-
mortar and grout. Definitions, specifications and conformity 588
criteria”, BS EN 15167 Part 1, 2006
21. Schuring, J., Schulz, H. D., Fischer, W. R., Bottcher, J. and
8. BSI, “Testing fresh concrete. Slump-test”, BS EN 12350 Part 2, 2009 Duijnisveld, W. H. M., “Redox, fundamentals, processes and
9. BSI, “Testing hardened concrete. Compressive strength of test applications”, Springer, 1999
specimens”, BS EN 12350 Part 3, 2009 22. Shi, C., Krivenko, P. V. and Roy, D., “Alkali-Activated Cements and
10. Buenfeld, N. R., Glass, G. K., Hassanein, A. M. and Zhang, J.- Concretes”, London: Taylor & Francis, 2006
Z., “Chloride transport in concrete subjected to electrical field”, 23. Shoesmith, D. W., Taylor, P., Bailey, M. G. and Ikeda, B., 1978.
Journal of Materials in Civil Engineering, V.10, 1998, pp.220-228 “Electrochemical behaviour of iron in alkaline sulphide solutions”,
11. Glasser, F. P., “Chemical, mineralogical, and microstructural Electrochimica Acta, V.23, 1978, pp.903-916
changes occurring in hydrated slag- cement blends”, Materials 24. Tang, L., 1999. “Concentration dependence of diffusion and
Science of Concrete, V.II, 1991, pp.41-81 migration of chloride ions Part 1. Theoretical considerations”,
12. Gomez, R. T., Aperador, W., Vera, E., Mejía de Gutiérrez, R. and Ortiz, Cement and Concrete Research, V.29, 1999, pp.1463-1468
C., “Study of steel corrosion embedded in AAS concrete under 25. Tritthart, J., 1989. “Chloride binding in cement II. The influence
chlorides”, Dyna, V.77, No.164, 2010, pp.52-59 of the hydroxide concentration in the pore solution of hardened
13. Holloway, M. and Sykes, J. M., “Studies of the corrosion of cement paste on chloride binding”, Cement and Concrete Research,
mild steel in alkali-activated slag cement mortars with sodium V.19, 1989, pp.683-691
chloride admixtures by a galvanostatic pulse method”, Corrosion 26. Yang, C. and Pu, X., “Retarder of alkali activated slag”, Chinese
Science, V.47, No.12, 2005, pp.3097-3110 patent, 91108316.2, 1993
14. Karahan, O. and Yakupoglu, A., “Resistance of alkali-activated slag 27. Zhang, T. and Gjorv, O. E., 1996. “Diffusion behavior of chloride ions
mortar to abrasion and fire”, Advances in Cement Research, V.23, in concrete”, Cement and Concrete Research, V.26, 1996, pp.907-917
No.6, 2011, pp.289-297

Dr. Qianmin Ma
Dr. Qianmin Ma received her PhD at Queen’s University Belfast in 2013 with the thesis titled of “Chloride
transport and chloride induced corrosion of steel reinforcement in sodium silicate solution activated slag
concretes”. Since then she works as a lecturer in Faculty of Civil Engineering and Mechanics, Kunming
University of Science and Technology, China. Her reasearch focuses on durability of concrete, utilization
of industrial slags in building materials and performance of concrete at high temperature. She has led 3
research projects and published more than 10 technical papers.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

460 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 4 C
Session 4 C - Paper 1

Electro-Mechanical Impedance Technique based Monitoring of

Concrete Health using Piezo Transducers
Sarvesh Goel and Sumedha Moharana
Civil Engineering Department, Shiv Nadar University

ABSTRACT employs piezoelectric transducers has successfully

During the construction of a concrete structure, emerged as a promising technique for monitoring the
strength monitoring is important to ensure the safety health of concrete in the works of Shin et al, (2008) and
of both personnel and the structure itself. Hydration Shin and Oh, (2009). But these studies have a drawback
(strength gain) of cement is a complicated physical and - they can’t monitor the early strength of concrete. An
chemical process involving various phases (initial and extension of this work was the introduction of embedded
final setting time, hydrated compound formation etc.) PZT transducers inside the mass of concrete as smart
which determine the strength development in concrete. aggregates by Yang et al, (2010). In this study the objective
Use of supplementary cementitious material in cement was to compare how the process of strength gain/
concrete to accelerate the hydration and strength gain hydration and bond strength between rebar and concrete
through hydraulic or pozzolanic activity is also seen changes on adding mineral admixtures like fly ash in
in common practice. This research work deals with normal concrete. Fly ash, a by-product from coal fired
complete monitoring of hydration of cement composite power plants will be used in concrete in lieu of Portland
using piezo impedance transducers. PZT (Lead zirconate cement in this study. There are many benefits of using
titanate) patch, which shows piezoelectric effect, is used fly ash in concrete as it increases plasticity, reduces
for detecting the health of concrete using EMI (Electro- hydration process time and also changes the overall
mechanical impedance) technique. The PZT has been strength. Fly ash can replace 8-40% of Portland cement
installed in embedded and steel bonded configuration for as per IS 10262 (2009) which makes great significance to
continuous hydration monitoring. This paper extends the the environment by lowering the levels of extraction for
study of admixture effect on concrete hydration through virgin silica and limestone as well as the reduction of
piezo sensors. GHG (greenhouse gas) emission tied to concrete itself.
In this work, the monitoring of initial time detection of
Keywords: SHM (Structural Health Monitoring), EMI cement hydration process has been done using surface
(Electro-mechanical Impedance), concrete structure, bonded (PZT transducers with RC bar) and embedded
hydration, PZT patch transducers. Various piezo impedance signatures are
acquired at different time intervals to identify the strength
Introduction gain and other parameters of concrete.
In principle, the mechanical behavior of concrete can be
predicted by modelling in detail the complex phenomena Electro-Mechanical Impedance Technique
arising from the hydration reaction which develops during Electro-mechanical Impedance (EMI) is a non-destructive
the setting and hardening phases. This results in heat flow evaluation technique used in structures to measure any
which is often responsible for gain of ultimate strength. incipient damage. This technique is based on monitoring
Hydration of cement is a complicated physical and EM (electro-mechanical) admittance variations in the
chemical process which determines the microstructure
PZT caused due to structural damage by observing
of concrete. Furthermore, monitoring of concrete at an
conductance and susceptance signatures. The EM
early stage of casting increases the efficiency of in situ
signature consists of the conductance (real part-G) and
casting of large reinforced concrete specimen or precast
susceptance (imaginary part-B). Conductance has been
concrete blocks and also determines the optimal time of
conventionally used for structural health monitoring due
demoulding. NDT (Non-Destructive Technique) has been
to its better reflection of structural changes.
employed for determining strength gain/ hydration but up
to a certain extent (like the ultrasonic pulse velocity meter A PZT patch is either surface bonded or embedded in the
needs access to two sides of the structure for strength structure using high strength adhesive and excited at
determination). The advent of smart materials such as high frequency (50-300 KHz) by an impedance analyzer.
piezo electrics has attracted many researchers to develop Piezoelectric material shows direct and converse effect
new techniques for structural health monitoring. The which means they produce electrical charge on applying
electromechanical impedance sensing technique which mechanical stress and show mechanical strain on

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

462 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Electro-Mechanical Impedance Technique based Monitoring of Concrete Health using Piezo Transducers

applying electrical field. Measuring electrical impedance Bond Strength between RC Bar and Concrete
is very difficult hence electrical admittance is measured
Bond strength is a measure of the grip between RC bar
which is the inverse of electrical impedance. When
and concrete. There are many destructive and non-
there is any damage or some parameter changes in a
destructive methods like pull-out or break-off tests and
structure, a change in stiffness and damping of structure
AE (acoustics emission) technique which could have been
subsequently leads to change in admittance. Measurement
used to monitor the progression of bonding between RC
of conductance and susceptance at different frequencies
bar and concrete. But these methods are time consuming
forms a ‘signature’. Liang et al (1994) first proposed the
and ambiguous. Tawie and Lee (2010) studied the bond
one-dimensional (1-D) theoretical model of this technique
strength on the basis of how the concrete is gaining its
as shown in Figure 1.
strength. In this study we will compare how the bond
strength between concrete and rebar changes on adding
fly ash in lieu of cement in normal concrete.

Fly Ash as a Cementitious Material

Fly ash, a by-product from coal fired power plant, has
been used as a pozzolan (a material that has cementitious
properties) in lieu of Portland cement in concrete. Recently,
fly ash has also been used to improve the environmental
Fig. 1: PZT-Structure Interaction (Liang et al. 1994) footprint of concrete by lowering the amount of portable
water in the mix. The ball bearing shape of fly ash
The piezo-mechanical coupled signature for 1-D structure
particles gives the same workability and slump in less
can be expressed as (Bhalla and Soh, 2003[1])
amount of water. Depending on the source and properties
Yr = 2~ h #Q fr33T - d 231 V + S Z +a Z X d 231 Yr E S kl X&
wl Z tan kl .......(1) of coal being burned, the components of fly ash vary
p s a
considerably but mostly it includes SiO2 (silicon dioxide)
Where; and CaO (calcium oxide). The product of cement hydration
Yr is complex admittance [ohm-1], components (C) is CSH (microcrystalline hydrate) with
Z a is the effective mechanical impedance of the PZT patch some lime separating out as Ca(OH)2 (calcium hydroxide)
but on adding fly ash the lime component reduces as given
[Ns/m],
below in equation 3. The short term strength of concrete
Z s is the effective mechanical impedance of the host reduces after adding pozzolanic material reduces but the
structure [Ns/m], long term strength is higher in pozzolanic based concrete.
f T33 is the complex electrical permittivity [N/m],
CS + H = CSH + CH .....................................................[3]
Yr E is complex Young’s modulus [N/m2] CS + H + Si = CSH
d 31 is piezoelectric displacement [m],
Where;
k is wave number [m-1], CS is calcium silicate,
w, l and hp are dimensions of PZT patch [m] and CSH is calcium silicate hydrate,
~ is angular frequency [rad/s]
CH is calcium hydroxide,
The effect of structural damage or change in material Si is silica and
characteristic on the PZT admittance signature is the H is hydrogen
horizontal and vertical shifting of conductance value
and frequency shift with respect to the baseline. These
are the main indicators for damage or material change.
Experimental Study
Statistical techniques such as RMSD (root mean square Piezoelectric Transducer Setup
deviation) have been used to associate the damage or
Four PZT transducers (PIC 151) measuring 10 mm X 10
material changes with the changes in the PZT admittance
mm X 0.3 mm were used. Two of them were fabricated
signatures.
on small metal piece for embedded configuration as
| QG - G 0i V shown in Figure 2 and the other two were bonded to
n
1 2
i

Q G 0i V2
RMSD = i=1 ...........................................(2) 12 mm diameter RC bar for steel bonded configuration as
shown in Figure 3. All the PZT transducers were bonded
using a thin layer of Aradlite epoxy and to protect them
Where;
from water a thick layer of Aradlite epoxy was pasted
G 0i is the baseline signature of PZT conductance and
on top of the sensors. The PZT transducers were tested
G 1i is the corresponding conductance for each monitoring using oscilloscope before casting them into the concrete
time at the ith measurement point. specimen.

Organised by
India Chapter of American Concrete Institute 463
Session 4 C - Paper 1

Table 1: Mix Design for 1 m3 concrete with and without fly-ash

All values Cement Water Coarse Fine Fly-ash

in kg Aggregate Aggregate

Specimen-1 430 191 1180 580 0

(Figure 4)

Specimen-2 301 191 1180 580 129

with fly ash
(Figure 5)
Fig. 2: Embedded PZT configuration

EMI Measurement
Figure 6 shows the laboratory set up for the overall
monitoring system. It consists of a concrete specimen
with embedded RC bar bonded and metal bonded PZT
transducers, an impedance analyzer (Agilent 4980A 30-
300 kHz) and a laptop equipped with data acquisition
software (VEEPro 9.2). The EMI measurement of both
specimens was evaluated by taking conductance and
Fig. 3: PZT patch bonded to RC bar susceptance signature for PZT sensors for the frequency
range of 50-300 kHz. In this study, the measurement was
Initial Hydration Test taken after every hour for initial setting time i.e. 10 hours
The concrete sample was prepared for compressive and then after every 24 hours for next 7 days.
strength of 30N/m2 as per IS 10262(2009). PZT is embedded
after concrete compaction in both embedded and RC bar
configuration. The amount of used construction material
for casting has been tabulated in Table 1.

Fig. 6: Experimental setup in laboratory

Experiment Results
Fig. 4: Normal concrete specimen

Fig. 7: PZT signature variation of normal concrete in embedded

Fig. 5: Concrete specimen with fly-ash
configuration

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

464 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Electro-Mechanical Impedance Technique based Monitoring of Concrete Health using Piezo Transducers

Fig. 8: PZT signature variation of normal concrete in RC bar Fig. 10: PZT signature variation of fly ash based concrete in RC
bonded configuration bar bonded configuration

Fig. 11: RMSD values for normal and fly ash based concrete
Fig. 9: PZT signature variation of fly ash based concrete in em- sample in embedded configuration
bedded configuration

Discussion
When structural properties change, the impedance of
the PZT shifts. Any change in the conductance signature
such as magnitude or frequency is attributed to structural
changes. Shift in the magnitude and frequency of
conductance signature was observed at the peak and
trough of resonance. This shift reflects the changes in
concrete during hydration.
The conductance signature for normal concrete sample
showed a progressive change in frequency and shift in
conductance value as the day progressed - see Figure 7 Fig. 12: RMSD values for normal and fly ash based concrete
and Figure 8. Similar conductance spectra was observed sample in RC bar bonded configuration
for fly ash based concrete sample - see Figure 9 and
Figure 10.
The bonding mechanism between rebar and concrete
During initial stage, the signature acquired immediately depends on the concrete properties. In other words, the
after the completion of concrete casting work has higher bond development can be attributed to the hydration of
value and sharp peak around the resonant frequency as concrete. By monitoring the hydration of concrete with
compared to the signature acquired after initial setting,
respect to time, the status of bonding between rebar and
where the peaks flatten and shift rightwards. As time
concrete can be known. As seen in Figure 8 and 10; the
passes, concrete becomes stronger and the signature
shifts to the right and the conductance value reduces. conductance signature of PZT bonded to RC bar starts
The variation of conductance value and resonance peak shifting towards right which suggests that hardening of
is more evident for the embedded sensor specimen - see concrete started at a higher rate, as reported by Tawie and
Figure 7. Lee (2010).

Organised by
India Chapter of American Concrete Institute 465
Session 4 C - Paper 1

Figure 11 and 12 show that RMSD values change rapidly 4. Liang C, Sun F.P, Rogers C.A (1994), “Coupled electro-mechanical
initially and remain almost constant after 40th hour. It is analysis of adaptive material systems determination of the actuator
power consumption and system energy transfer”. Journal of
observed in fly ash based concrete that the early strength
Intelligent Material Systems and Structures. Volume 5, no. 1
gain of concrete reduces but the long term strength is
higher. The bond strength between rebar and concrete is 5. Naidu, A. and Bhalla, S. 2002. Damage Detection in Concrete
Structures with Smart Piezo ceramic Transducers. Inter. Conf. on
also higher for fly ash based concrete.
Smart Materials Structures and Systems, July 17- 19, Bangalore,
India, ISSS2002/SA-538.
Conclusion 6. Park, S., Yun, C.B. and Inman D. J. (2008), “Structural health
This paper presents the EMI based piezoelectric method monitoring using electro-mechanical impedance sensors”, Fatigue
to investigate the hydration of cement composite. & Fracture of Engineering Materials & Structures, volume 31, no.
Results obtained show that the physical changes in 8, pp 714724.
cement paste during hydration can be monitored by 7. Shin, S.W (2008),” Piezoelectric sensor based nondestructive active
piezo coupled admittance signature and RMSD based monitoring of strength gain in concrete,” Smart Mater. Struct. 17
statistical interpretation. The embedded PZT transducer 055002.
shows higher sensitivity for monitoring the hydration 8. Shin, S.W. and Oh, T.K. (2009), “Application of electro-mechanical
of concrete compared to the surface bonded PZT patch. impedance sensing technique for online monitoring of strength
The PZT transducers can effectively detect various rates development in concrete using smart PZT patches,” Construction
of hydration as evident from fly ash based concrete and Building Materials, volume 23, no. 2.
specimen. 9. Soh C. K and Bhalla S. (2005), “Calibration of Piezo-Impedance
Transducers for Strength Prediction and Damage Assessment of
References
Concrete”, Smart Materials and Structures, Vol. 14, No. 4 (August),
1. Divsholi, B.S. and Yang, Y. (2014), “Combined Embedded and pp. 671-684.
Surface-Bonded Piezoelectric Transducers for Monitoring of
Concrete Structures” in NDT & E International 65. 10. Tawie, R. and Lee,H. K., “Piezoelectric-based non-destructive
monitoring of hydration of reinforced concrete as an indicator of
2. Fu, X. and Chung, D.D.L. (1999) “Interface between steel rebar and
concrete, studied by electromechanical pull-out testing” Composite bond development at the steel-concrete interface,” Cement and
Interfaces, volume: 6, no.2. Concrete Research, volume 40, no. 12, pp. 1697–1703

3. Lee, H.K.; Lee, K.M.; Kim, Y.H.; Yim, H.; Bae, D.B.(2004) “Ultrasonic 11. Yang, Y., Divsholi, B.S., and Soh, C. K.(2010), “A reusable PZT
in situ monitoring of setting process of high-performance concrete” transducer for monitoring initial hydration and structural health
Cement and Concrete Research, volume 34, no. 4, 631–640. of concrete,” Sensors, volume 10, no. 5, pp. 5193–5208

Sarvesh Goel
Mr. Sarvesh Goel is a 3rd year undergraduate civil engineering student at Shiv Nadar University, India. His
research interests include concrete health monitoring using electro-mechanical impedance techniques,
fatigue life assessment of RCC structures and study of new structural systems.

Dr. Sumedha Moharana

Dr. Sumedha Maharana is currently working as assistant professor in Shiv Nadar University. Her major
research work is, smart structure, piezo-impedance based structural health monitoring, shear lag
effect and piezo-structure interaction. In addition, she is also interested in composite structure, dynamic
behaviour of thin wall structure. She has expertise in the field of modelling piezo-elastic systems and
structural diagnosis using piezo sensor.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

466 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Attack on Foundation Concrete and Modern Preventive Measures

Chemical Attack on Foundation Concrete and Modern Preventive

Measures
Dr. R. A. Hegde and Dr. Shirish Vichare
Civil Engineering Department
Mukesh Patel School of Technology Management and Engineering Ville Parle (W), Mumbai – 400 056

Abstract Deteriorating Process

Foundations of almost all the structures of this era are made
of Reinforced Concrete. The foundations are underground Sulphate Attack
and are susceptible to a number of chemical attacks that Sulphate attack is a chemical breakdown mechanism
can cause severe deterioration, resulting in loss of strength where sulphate ions attack components of the cement
and unsafe conditions. Study of various chemical attacks on paste. The compounds responsible for sulphate attack
concrete, will help in proposing preventive measures and are water-soluble sulphate-containing salts, such as
solutions to make the foundations safe. This paper reports alkali-earth (calcium, magnesium) and alkali (sodium,
experimental results of such study. Study was conducted potassium) sulphates that are capable of chemically
by casting M-50 grade concrete cubes and subjecting them reacting with components of concrete. The essential
to chemical attacks from Sulphates, Chlorides, Coatings agents for sulphate attack are sulphate anions
and Inhibitors. The observations were limited to evaluation (SO4 2-). These are transported to the concrete in various
of effect on compressive strength and permeability. It was concentrations in water, together with cations, the more
observed from the results obtained that sulphate attack is a common of which are calcium, magnesium and sodium.
cause of concern, while chlorides actually help in improving Where porous concrete is in contact with saturated
the strength. Coatings and Inhibitors reduce permeability. ground the water phase is continuous across the ground/
It was however, observed that coatings reduced the concrete interface and sulphate ions will be readily carried
compressive strength by 10%. into the body of the concrete. Sulphate attack might show
itself in different forms depending on the chemical form
Keywords: Concrete, Chemical attack, Corrosion, of the sulphate or the atmospheric environment which
Permeability, Compression strength, Durability. the concrete is exposed to (Kejin Wanga, et. al.). Sulphate
attack can be 'external' or 'internal'. External attack is due
Introduction to penetration of sulphates in solution (in groundwater for
Foundations of almost all the structures are made of example) into the concrete from outside whereas Internal
Reinforced Concrete. Foundations are always underground attack is due to a soluble source being incorporated
and are susceptible to the attack of chemicals from into the concrete at the time of mixing, gypsum in the
contaminated ground water and other sources. Ground aggregate.
water contamination and ground pollution is a serious Sulphate in concrete combines with the C-S-H, or
and undesirable bi-product of rapid urbanisation and concrete paste, and begins destroying the paste that holds
unplanned development. Repairs of foundation are difficult the concrete together. As sulphate dries, new compounds
and risky. Even though the foundations are designed are formed, which is often called Ettringite. These new
properly, there is a possibility that due to adverse effect crystals occupy empty space, and as they continue to
of chemical attack on foundation concrete the structures form, they cause the paste to crack, further damaging the
may be at risk. In fact, this is an incentive for undertaking concrete. The chemical equations which describe these
our study. processes are:
Our study intends to understand the effect of chemical C3A.Cs.H18 + 2CH +2s+12H = C3A.3Cs.H32 and
attacks on reinforced concrete. We have studied effect of Na2SO4+Ca(OH)2 +2H2O = CaSO4.2H2O +2NaOH.
sulphate, chloride, inhibitors and coatings on strength and
The complex physico-chemical processes of "sulphate
permeability of M50 concrete cubes. In our presentation
attack" are interdependent as is the resulting damage.
we have first described the deteriorating process,
Physical sulphate attack, often evidenced by bloom
explained methodology of our testing, presented the
(the presence of sodium sulphates Na2SO4 and/or
experimental results and finally put forth the conclusions
Na2SO4.10H2O) at exposed concrete surfaces. It is not
of our study.
only a cosmetic problem, but it is the visible displaying of
possible chemical and micro structural problems within
the concrete matrix.

Organised by
India Chapter of American Concrete Institute 467
Session 4 C - Paper 2

The most important mineralogical phases of cement that Methods of controlling sulphate attack
affect the intensity of sulphate attack are: Use of sulphate resisting cement: The most efficient
C3A, C3S / C2S ratio and C4AF. method of resisting the sulphate attack is to use cement
with low C3A content. It has been found that C3A content
Sulphate attack process decreases the durability of of 7% gives rough division between cements of good &
concrete by changing the chemical nature of the cement poor performance in sulphate waters.
paste, and of the mechanical properties of the concrete.
The sulphate attack tends to increase with an increase in Quality concrete: A well designed, placed and compacted
the concentration of the sulphate solution up to a certain concrete which is dense and impermeable exhibits a
level. higher resistance to sulphate attack. A concrete with
low W/C ratio also demonstrates a higher resistance to
The addition of a pozzolanic admixture such as fly ash sulphate attack.
reduces the C3A content of cement.
Use of air-entrainment: Use of air-entrainment to 6% has
beneficial effect on sulphate resisting qualities of concrete.
The beneficial effects are possibly due to reduction of
segregation, improvement in workability, reduction in
bleeding and in general better impermeability of concrete.
Use of pozzolana: Admixing of pozzolana convert
the leachable calcium hydroxide into insoluble non
leachable cementitious product. This pozzolanic action
is responsible for impermeability of concrete. Removal
of calcium hydroxide reduces susceptibility of concrete to
attack by magnesium sulphate.
High pressure steam curing: It improves the resistance
of concrete to sulphate attack. This improvement is due
to change of C3AH6 into less reactive phase & also the
removal or reduction of calcium hydroxide by reaction of
silica which is invariably mixed when high pressure steam
curing method is adopted.
Use of high alumina cement: High alumina cement
contains approximately 40% alumina, a compound very
susceptible to sulphate attack. The primary cause of
resistance is attributed to formation of protective film
Fig. 1: Relationship between ratio of deterioration to C3A which inhibit the penetration or diffusion of sulphate ions
into interior. High alumina cement may not show higher
resistance to sulphate attack at higher temperature.

Chloride Attack
Chlorides, particularly calcium chloride, have been used
to shorten the setting time of concrete. However, calcium
chloride and (to a lesser extent) sodium chloride have been
shown to leach calcium hydroxide and cause chemical
changes in Portland cement, leading to loss of strength,
as well as attacking the steel reinforcement present in
most concrete (Verbeck. G. J, 1975).
Chloride attack is one of the most important aspects for
consideration when we deal with the durability of concrete.
Chloride attack is particularly important because it
primarily causes corrosion of reinforcement. Statistics
have indicated that over 40 percent of failure of structures
is due to corrosion of reinforcement.
Due to high alkalinity of concrete a protective oxide film
is present on the surface of steel reinforcement. The
protective passivity layer can be lost due to carbonation.
Fig. 2: Relation between expansions of concrete to the time

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

468 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Attack on Foundation Concrete and Modern Preventive Measures

Table 1
Limits of Chloride Content of Concrete (IS 456 of 2000)

Maximum total acid soluble chloride content.

Sr. No. Type or Use of Concrete
Expressed as kg/m3 of Concrete

Concrete containing metal and steam cured at elevated temperature and pre
1. 0.4
stressed concrete

2. Reinforced concrete or plain concrete containing embedded metal 0.6

Concrete not containing embedded metal or any material requiring protection from
3. 3.0
chloride

This protective layer can be lost due to the presence of Methodology

chloride in the presence of water and oxygen (Hoff, G. C,
This section explains the various methods which were used
1991). In reality the action of chloride in inducing corrosion
for testing of the chemical attacks. This includes the step
of reinforcement is more serious than any other reasons.
by step method, starting from type of specimen prepared,
Chloride enters the concrete from cement, water, casting of specimens and the detailed description of the
aggregate and sometimes from admixture. The present various tests performed on concrete.
day admixtures generally contain negligible quantity of
chloride. Chloride can enter the concrete by diffusion Specimen Description and Preparation
from environment. The Bureau of Indian Standard earlier
specified the maximum chloride content in cement 0.005 Description
percent. But it is now increased the allowable chloride Firstly we have defined a specific set of samples as 1 set.
content in cement to 0.1 percent. IS 456 of 2000 limits the This set includes various numbers of samples of different
chloride content as in the concrete at the time of placing is dimensions and components. This set is commonly casted
shown in the Table 1. for all conditions of attacks. The Table 3 gives detail of the
specifications and number of samples in 1 set.
The amount of chloride required for initiating corrosion is
partly dependent on the pH value of the pore water in the This each set was casted for each of the attacking
concrete. At a pH value less than 11.5, corrosion may occur conditions as shown Table 4.
without the presence of chloride. At pH value greater than
11.5 a good amount of chloride is required.
Table 3
Limiting value of chloride contents, above which corrosion Table showing specimen casted for each case (1Set)
may be imminent, for various values of pH are indicated in
the following Table 2. Number of samples
Tests Samples size (mm)
(each test)
The total chloride in concrete is present partly as insoluble Compression test Cube (150X150X150) 3
chlori-aluminates and partly in soluble form. It is the
soluble chloride, which is responsible for corrosion of Permeability Cube (150X150X150) 3
reinforcement. Modulus of elasticity Cylinder(150X300) 3

Table 2
Limiting Chloride Content Corresponding to pH of Concrete Table 4
Table showing sets used for various attacks
Chloride content in Chloride content in
pH
g/litre ppm Exposure
condition Ambient
13.5 6.7400 6740 Chloride Acid Sulphate
condition
Protection
13.0 2.1300 2130

12.5 0.6720 672 No protection (exposed to

1 set 1 set 1 set 1 set
water solution)
12.0 0.2130 213
No protection
11.5 0.0670 67 1 set 1 set 1 set 1 set
( exposed to air)
11.0 0.0213 21
Coatings 1 set 1 set 1 set 1 set
10.5 0.0021 2

10.0 0.00020 0.2 Inhibitors 1 set 1 set 1 set 1 set

Organised by
India Chapter of American Concrete Institute 469
Session 4 C - Paper 2

Table 5
Table showing casting of different specimens

Exposure
condition Ambient condition Chloride Acid Sulphate
Protection

No protection (exposed to water

solution)
Added 3% Sodium Chloride Added 1% Nitric Acid in Added 2% Magnesium Sulphate
No chemicals added
No protection (exposed to air) in proportion with cement proportion with cement in proportion with cement.

Coatings

0.75% of inhibitors
0.75% of inhibitors 0.75% of inhibitors added in 0.75% of inhibitors added in
Inhibitors added in addition to
added addition to above chemicals addition to above chemicals
above chemicals

Specimen preparation Testing of specimen

All the specimens are casted according to above After the period of 30 days, the specimens were allowed to
mix design. To accelerate the attacks on concrete dry for 1 day. Then the tests are carried out in a sequential
during the preparation of concrete different attacking pattern.
chemicals were added in definite proportion to the
cement content. These chemicals were mixed with Respective test, their result and their interpretations has
water while casting. been explained in further sections.

Mixing of different chemicals had the required attacking Tests Performed

effect throughout the concrete, giving better results in
Compression and Permeability Tests were performed
lesser time duration. This condition may not project the
on the specimen exposed to (chloride, acid and sulphate
on situ-condition, but can be used to get results and their
attacks)
effect in worst conditions. Application of protections
to prevent such worse attack is on conservative side.
The following table shows the details of different Compression test
chemicals added and their proportions during casting Description of test
for each set. Compression test is the most commonly conducted test
The set for coatings and without protection exposed to air on hardened concrete, partly because it is an easy test
and water have been casted together for each attack. The to perform, and partly because most of the desirable
specimen set for inhibitors has been added by respective characteristic properties of concrete are qualitatively
chemicals as well as inhibitor (Epco -KP-200). related to its compressive strength.
Specimen for bond was specially prepared by placing This method describes the procedure for making and
reinforcement rods (Fe-415) of 16mm dia. at the centre of curing compression test specimens from fresh concrete
the cube while placing it on vibratory machine. and for determining the compressive strength of the
specimens.
All these specimens were allowed to cure for 28 days in
normal water. Apparatus and materials
The cube specimen is of the size 15 x 15 x 15cm. If the
Preparation of specimens before attack largest nominal size of the aggregate does not exceed
After period of 28 days of curing, all the specimens were 20mm. Cylindrical test specimens have a length equal
removed from water and allowed to air dry for 3 days. to twice the diameter. They are 15cm in diameter and
Then 1 set from each attack was coated with epoxy coating 30cm long. The mould and tamping rods used are as the
(KARAIKOTE-6545). This specimen was coated in two recommendations os relevant codes.
layers and allowed to air cure for 3 days.
Sampling Equipment
Exposing specimens to attack Scoop or shovel, trowel, containers, saran wrap, tape.
After curing period of 28 days and 6 days of drying,
the specimens were exposed to 30 days of attacking Capping Compound
atmosphere. All set of specimen were kept in water bath A mixture of sulphur and granular materials having
of same above chemical concentration respectively. a compressive strength equal to or greater than the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

470 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Attack on Foundation Concrete and Modern Preventive Measures

anticipated strength of the specimen is used. be reduced and hydration ceases. The smaller sized
aggregate may have strength advantages in that the
Capping Device internal weak planes may be less likely to exist. The bond
A device for applying a capping compound to the cylinder between the mortar and coarse aggregate particles will be
and surfaces in the form of plane surfaces at right angles stronger for smaller sized aggregates which have a higher
to the axis of the cylinder is used. curvature. When concrete bleeds, the bleed water is often
trapped beneath the coarse aggregate thus weakening the
Curing Equipment bond within the interfacial zone and allowing for weaker
stress paths for cracks to initiate. Excessive bleeding
A moist storage cabinet or room capable of maintaining will produce a high water cement ratio at the top portion
specimens at a temperature within ± 1 degrees of 23oC leading to weakened wearing surfaces and dusting.
and capable of maintaining a moist condition in which free
water is maintained on the surfaces of the specimens is
Permeability test
recommended.
This test is performed to determine the permeability of
Testing Machine concrete. Below is the whole process explained in detail.
A machine of sufficient capacity which will apply a load
Specimen
continuously without shock within a range of 0.140 to 0.350
MPa per second is recommended. The testing machine Concrete cubes of size 150 mm*150 mm*150 mm
shall be equipped with two steel bearing blocks with
hardened faces. One bearing block shall be spherically Procedure
seated and the other rigidly mounted. The testing machine Firstly clean all the four faces of concrete with wire bush
shall be accurate within a tolerance of ± 1.0 percent of the for cement paste and then wash it and let it dry. At the same
compressive strength of the specimen. time prepare the permeability apparatus by maintaining
the pressure to 5 bars.
Procedure
Now, after drying the specimen keep it on the permeability
Sample Preparation apparatus. While keeping it see to it that it placed correctly
Samples of concrete for tests will be obtained in above the water jet and there is no way for leakage.
accordance with STP 106. Now fix the specimen in position tightly so that it cannot
be disturbed. After fixing start the water supply under
Results and calculations pressure. The pressure should be 5 bars and this has to
Calculate the compressive strength in megapascals by be maintained for 3 days.
dividing the maximum load in Newtons by the average After 3 days, remove the specimen and immediately keep
cross sectional area of the specimen in square millimetres. it in UTM for further procedure.
The sulphur capping compound is allowed to harden at Now make arrangement in such a way that specimen
least two hours before applying the load. Specimens will should break in two equal halves right from middle. After
be kept moist until time of test. The machine is operated the specimen gets broken measure the distance which is
at a constant rate within the range of 0.140 to 0.350 MPa wet from the edge with the help of scale. The distance got
per second. is the permeability of the concrete.

The Concrete Compression

Test Notable Affecting
Factors
Retempering water of the
mixture in the concrete
may decrease the mortar
strength due to the
uneven dispersion of the
retempering water which
leads to pockets of mortar
having a high water cement
ratio. Even so the concrete
is allowed to dry rapidly,
the available moisture for
hydration reaction will
Fig. 3: Crushed Cylinder Fig. 4: Permeability apparatus

Organised by
India Chapter of American Concrete Institute 471
Session 4 C - Paper 2

Material Properties Observation from compression test

To have proper proportions of various materials used 1. The results are observed to be inconsistent. The results
while casting, material properties have been evaluated. do not show any relationship between the attacks and
These properties have helped us to determine the amount the protection with their compressive values.
of inhibitors to be added and decide the mix design. 2. Though a peculiar property can be seen in the results.
Therefore the properties of various materials by obtained The values of epoxy coating are very less compared
from different tests have been detailed below. to other values. This can be explained by phenomena
known as “suffocation of concrete”. It is phenomena
Properties in which concrete is unable to achieve its desirable
Fineness Test:-The weight of remaining cement in sieve is properties due to non-curing of specimens. As the
8% of initial weight. Setting Time Test: coating acts barrier to both air and water, it does not
Initial Setting Time = 45 min. get cured to its counterparts and hence gives lower
Final Setting Time = 300 min. results.
Compressive Strength of Cement = 40MPa Soundness Test
of Cement = 8mm Aggregate Impact Value of Aggregate = Permeability Test
26 Crushing Value of Aggregate = 16 For this experiment cubes of M 50 were casted and studied
PH of Water = 7.4. under different chemical attacks. These specimens after
exposing to sever conditions were air dried for three
days. After this process specimens were tested for
Testing Results And Interpretation permeability. The Table 7 gives the penetration depth for
various specimens. These results provided some concrete
Compression test
observations which have been graphically represented in
M - 50 grade concrete cubes has prepared for the chemical Fig.6.
attack on concrete were tested for compressive strength
under various aggressive conditions. The various results Observations from permeability test
for compressive strength were noted and compared with
each other Table 6 gives the comparison of compressive In this test depth of penetration is the criteria.
strengths of cubes under various conditions. The variation ll The depth of penetration of water for sulphatic
in compressive strength for different condition can easily be specimen is high. This can be explained by its intensive
visualise by the graphical representation show in Figure 5. cracking due to sulphate attack.

Table 6
Compressive Strength of different specimen

Compressive Strength

Ambient condition Chloride Sulphate Acid

Mpa Mpa Mpa Mpa

N MPa kN MPa kN MPa kN MPa
(average (average (average (average

087 418.31 1062 47.20 1093 48.58 1202 53.4 2

No
Protectio 415 612.89 54.74 1040 46.22 48.40 1145 50.89 53.97 1234 54.8 4 52.70
n (water)
193 513.02 1165 51.78 1405 62.44 1121 49.8 2

216 514.04 1092 48.53 976 43.38 1183 52.5 8

No
Protectio 119 419.73 52.44 1181 52.49 50.30 1200 53.33 50.81 1212 53.8 7 53.78
n (air)
205 513.56 1122 49.87 1254 55.73 1235 54.8 9

28 491.24 1019 45.29 859 38.18 1311 58.2 7

Coatings 132 510.31 45.61 1106 49.16 47.38 1106 49.16 45.79 927 41.2 0 45.14

019 415.29 1073 47.69 1126 50.04 809 35.9 6

097 418.76 1181 52.49 1047 46.53 1383 61.4 7

Inhibitors 038 416.13 46.81 1185 52.67 51.57 970 43.11 49.64 1212 53.8 7 56.74

025 415.56 1115 49.56 1334 59.29 1235 54.8 9

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

472 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Chemical Attack on Foundation Concrete and Modern Preventive Measures

ll The chlorides have least penetration as in early stages

it act as concrete strengtheners and binds concrete
together.
ll The specimen exposed to air has greater permeability
as they tend to crack due to exposing to air.
ll Epoxy coated specimens have least or negligible
penetration as the coating act as physical barrier
between concrete and chemicals.
ll Inhibitors reduce penetration to certain level.

Conclusion
Fig. 5: Graph showing compressive strength for different In the present study an attempt was made to understand
conditions
the effect of chemical attack on concrete to be used in
foundations. Conclusions from permeability test and
compressive test are given below.
Table 7
Permeability values for various specimens
Compressive Strength
PERMEABILITY (Depth of penetration mm) The above results show that the effects of attacks and their
protections on compressive strength are inconsistent.
Ambient
Chloride Sulphate Acid
condition These results cannot be related to any level.
24.00 20.00 31.00 32.00 If strength is the property of structure needed then epoxy
No
Protection coating is not recommendable.
28.00 25.00 22.00 21.67 38.00 35.00 27.00 29.33
(exposed to
water) Permeability
23.00 23.00 36.00 29.00
The above results suggest that, sulphate protection must
30.00 30.00 41.00 38.00
No be the priority in conditions of water retaining structures.
Protection Acids are found increase the permeability by 20 – 30 %.
33.00 30.67 26.00 27.67 44.00 41.33 33.00 35.67
(exposed to
air) The epoxy coating can be successfully being applied to
29.00 27.00 39.00 36.00
reduce the penetration to minimum.
1.00 3.00 4.00 5.00
Epoxy
Coatings
2.00 1.33 5.00 3.33 6.00 4.67 4.00 3.67 Acknowledgement
1.00 2.00 4.00 2.00 The authors wish to thank under-graduate students
Mr. Aditya Potdar, Mr. Amol Saraf, Mr. Varad Kelkar, Mr.
19.00 15.00 33.00 32.00 Akshay Pise and Mr. Abhijeet Deshmukh for their sincere
efforts in completing the project.
Inhibitors 21.00 20.67 19.00 17.67 34.00 32.00 33.00 31.67

22.00 19.00 29.00 30.00 References

1. Kejin Wanga, Daniel E. Nelsena and Wilfrid A. Nixon, Damaging
effects of deicing chemicals on concrete materials", Cement and
Concrete Composites Vol. 28(2), pp 173-188.
2. Nilson, Darwin , Dolan (2003). Design of Concrete Structures. The
MacGraw-Hill Education.
3. Verbeck. G. J. (1975) ‘Mechanism of corrosion in concrete’, in
‘corrosion of Metals in concrete’, SP- 49, American Concrete
Institute, pp.21-38.
4. Hoff, G. C, (1991) ‘Durability of Concrete’’,SP-126, American Concrete
Institute, Detroit, MI.
5. S.H. Kosmatka, B. Kerkhoff, and W.C. Panarese (2002), 'Design
and Control of Concrete Mixtures’, 14th Edition, Portland Cement
Fig. 6: Comparison of permeability of different specimens Association.

Organised by
India Chapter of American Concrete Institute 473
Session 4 C - Paper 2

Dr. R. A. Hegde
Dr Hegde, a graduate in civil engineering with doctorate from IIT, Bombay, is presently Professor & Head,
Dept. of Civil Engineering at Mukesh Patel School of Technology Management and Engineering, Mumbai,
He is actively involved in Soil / rock testing and consultancy works and is a Mumbai University approved
guide for M.E., M. Tech. and Ph. D. candidates.
He was Coordinator for 1-week Refresher Course on ‘Bridge Engg.’ for CPWD Engineers at Daman,
Coordinator for 1-Day Training Course on Bridge Foundations’ for MCGM Civil Engineers and has conducted
1-Day Training program for TCE Field Engineers at Jamshedpur and at Delhi.
He is Life Member of Indian Society for Technical Education, New Delhi, Fellow of Indian geotechnical
Society, New Delhi and Life Member of Indian Society for Non-destructive Testing, Chennai.
He has undertaken Consultancy Work and has submitted geotechnical solution reports to PWD (MS),
CPWD, BMC, TMC, CIDCO, Mumbai Metro One, MHADA, Developers and Builders.
He has Research Publications to his credit including 5 in International Journals , 4 in National Journals, 14
in International Conferences and Seminars and 25 in National Conferences and Seminars.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

474 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Strengthening using Active Prestressed CFRP in RCC Structures

Strengthening using Active Prestressed CFRP in RCC Structures

Dr. Gopal L Rai
Dhirendra Group of Companies, A-3027, Oberoi Garden Estate, Chandivali, Andheri (East), Mumbai, India – 400 072

Abstract out extensive studies on the use of FRP for strengthening

of flexural (Irwin and Rahman, 2002] and compression
Need for rehabilitation of reinforced concrete structures
(Sheikh and Li, 2007) members. Application of FRP in
is rapidly increasing. Fibre reinforced polymer (FRP)
masonry structures have shown to exhibit increase in
composite materials used for strengthening of reinforced
load carrying capacity and ductility (Grande, Imbimbo and
concrete structures have higher advantages in comparison
Sacco, 2008).
with conventional methods of strengthening. These include
its high tensile strength, light weight, non corrosive, non
magnetic and its high fatigue resistance. Its property of high Basics of Passive and Active Strengthening
strength-to-weight ratio combined with fatigue resistance System
can provide high prestressing forces while adding minimal
Passive strengthening system strengthens the
additional weight to a structure. They exhibit low relaxation
structural member without introducing external forces
losses and can thus increase the service lives and the
in the member. It increases the load carrying capacity of
load carrying capacities of reinforced concrete structures.
member by change in cross section of the members or by
Prestressing the carbon fiber reinforced polymer (CFRP)
the introduction of high strength materials. It is relatively
laminates results in the material being used more
easy in application. It doesn’t utilize the full properties of
efficiently as a part of its tensile capacity is utilised and it
the materials. Examples of passive system are concrete
contributes to the load bearing capacity under both service
jacketing, steel jacketing and addition of high strength
and ultimate load condition. Proper mechanical anchorage
materials.
has been developed so as to help in prestressing the CFRP
laminates. This paper details on the use of prestressed In case of active strengthening system external forces
CFRP laminates in strengthening of RCC structures are introduced in the member which counteracts the
including its practical applications in buildings and bridges. effects of internal forces. It is not easy in application
Also it briefly mentions about the post strengthening in comparison with passive system but utilizes the full
testing carried out for the substantiation of this method of material properties of the materials. Such a system brings
strengthening. about an improvement in the serviceability by reducing
cracks and deflections. It brings an improvement in the
Keywords: FRP, Prestressed, strengthening, active,
performance of the structure. External post tensioning
passive, bridges
system, bonding prestressed FRP laminates on the
structural member, etc are some of the examples of this
Introduction method. Whether a particular structure requires an active
CFRP composites are being increasingly used for or passive strengthening system depends on the type of
enhancing the load carrying capacity of R.C.C structural deficiency in the structure. Active and passive system
members. They consist of high strength fibres which act can be used in a combination depending on the deficiency
as the main load carrying elements which are bonded faced (Mandara, Piazza, Perdikaris and Schaur, 2002).
in a resin matrix acting as load transfer medium.
These materials are better in comparison with other Non Prestressed Cfrp System and
conventional methods as they exhibit superior properties.
Prestressed Cfrp System
They are easy to install and can be made into any size or
geometry. Application of FRP takes relatively less amount Strengthening using Non prestressed CFRP sheets and
of time and does not cause any hindrance in the normal laminates is easiest in application. It doesn’t require any
functionality of the structure. highly specialized equipment and can be done in a short
period of time. They are bonded on the structural members
Various codes and guidelines have been published for the with the help of epoxy resin to enhance their load carrying
ease of application of FRP on structures (ACI, 2008; fib capacity. The main advantages it has is that it is light in
CEB-FIP, 2001). FRP can be used for the strengthening of weight, having high strength to weight ratio, non corrosive
all types of structures. Various researchers have carried and non magnetic in nature. The properties are not better

Organised by
India Chapter of American Concrete Institute 475
Session 4 C - Paper 3

utilised in comparison with the prestressed ones. Several end anchors are placed at the ends of the laminates to
researchers have carried out studies on various methods avoid failure due to debonding of the laminates and to
of prestressing FRP and have found that bonding of FRP prevent or delay crack opening at the onset of failure of
laminates and prestressing increases the load carrying the concrete substrate. Proper end anchorages after the
capacity of the member in service and dead load condition prestressing operation of laminates on beams helps in
(El-Hacha, Wright and Green, 2001). It has been found that eliminating debonding failure mode and can increase the
using prestressed FRP laminates on the tensile face of service life of the structure (El-Hacha and Aly, 2013)
the flexural member improves the serviceability to much
Prestress can be induced by bonding the FRP laminates
greater extent. A typical load deflection curve for a beam
on the tension face of the member followed by application
strengthened with non prestressed and prestressed FRP
of prestress in the members with the help of prestressing
composite system is shown in Fig.1 below.
jacks at the ends of laminates. The laminates are kept
in this stretched condition until the adhesive is cured
following which the hydraulic jacks are removed.
This helps in effective transfer of the stresses to the
concrete. This is one of the most widely used methods
of strengthening as it is much easier in application in
comparison with other prestressing methods. Other
methods of prestressing include indirect methods such
as by cambering beam system or by inducing prestress
against external steel frame followed by bonding on the
structural member to be strengthened (El-Hacha, Wright
and Green, 2001). Proper anchorages need to be placed
at the ends of laminates to avoid debonding failures. The
direct method of strengthening finds its application in
both field as well as laboratory as it is highly effective. The
indirect method finds less use in field applications as it is
Fig. 1: Typical load-deflection curves (El-Hacha, Wright and cumbersome and the prestress induced is comparatively
Green, 2001) less (El-Hacha, Wright and Green, 2001).

Fig. 3: Strengthening with prestressed FRP laminates (Rai, 2013)

Fig. 2: Application of Non prestressed FRP laminates in slabs

(Courtesy:- M/s R & M International Pvt Ltd)

There are various methods to induce prestress in the FRP

composites for increase in the load carrying capacity. This
can be induced within the structure or outside of it. The
strengthening is also affected by the amount of prestress
to be provided as very high prestress can lead to failure
of the structural member itself. The minimum amount
of prestress required for improvement in the strength
properties is 25% of the ultimate strength of the laminates
and it should not exceed 50% of the same (El-Hacha,
Fig. 4: Prestressing operation in slab (Rai, 2013) (Courtesy:- M/s
Wright and Green, 2001; fib CEB-FIP, 2001). Mechanical R & M International Pvt Ltd)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

476 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Strengthening using Active Prestressed CFRP in RCC Structures

Case Study 1 - Flat Slab System with

Excessive Deflection and Cracks
Strengthened uvsing Prestressed Frp
System
An RCC framed commercial building structure required
strengthening of its flat slab panels due to excessive
deflections and cracking. Construction of the two top
storeys of the existing multi-storeyed structure was
carried out after a gap of two years post construction of the
entire structure. It was noticed that excessive deflection
and cracks were observed in the flat slab panels in the
range of 20 to 100mm. It was observed that few of the
slabs showed severe distress and had acquired saucer
shape. Crack pattern was observed running along column Fig. 6: Alternate Prestressed CFRP laminates in slab (Courtesy:-
drop region and progressing towards the column strip. M/s R & M International Pvt Ltd)
Non Destructive testing was carried out on the slab panels
which showed that concrete grade was sufficient as per Post strengthening of one slab panel it was subjected to
design. On site inspection it was found that the negative load testing to check the efficiency of the strengthening
moment reinforcement provided at column drop region carried out. Water loading was done on the slab for
was terminated abruptly and cracks were observed in this testing. Loading was done on slab as per the provisions
region. After carrying out analysis of the flat slab system it given in IS 456:2000. Loading and unloading was carried
was decided to provide prestressed and non prestressed out stage wise. Monitoring of strain and deflection was
CFRP laminates in alternate manner at the tension face carried out for various stages of loading and loading and
of column strip regions. The prestressing force provided after placement of loading on slab for 48 hours the reading
in the laminates was 5 Ton. Properties of the materials was taken. Dewatering of the slab was done in stages and
considered are as per Table 1 below. readings were recorded to check the rebound in the slab.

Table 1

Alternate Prestressed
Type of laminates
CFRP laminates

Thickness of CFRP laminate (mm) 1.4

Width of laminate (mm) 100

Ultimate tensile strength (MPa) 3200

Ultimate strain (%) 1.5

Fig. 7: Load testing on strengthened slab (Courtesy:- M/s R &

M International Pvt Ltd)

It was observed that sufficient recovery was observed

post strengthening. After 48 hours loading on slab the
recovery observed was 80.97% which proved that the
strengthening had worked.

Case Study 2 - Rob Strengthened using

Prestressed Frp System
The Rail Over Bridge at Karal junction in Navi Mumbai
required strengthening as it showed severe signs of
distress and was in need of immediate strengthening. It
Fig. 5: Excessive deflection in flat slab (Courtesy:- M/s R & M was structurally designed for Indian Road Congress (IRC)
International Pvt Ltd)

Organised by
India Chapter of American Concrete Institute 477
Session 4 C - Paper 3

45R loading but as per as per revised Codal provisions it

required to sustain much higher load i.e. of IRC Class 70
R in the current scenario. The structure seemed to exhibit
severe deflections and vibrations. To increase the flexural
strength of the girder CFRP prestressed laminates were
placed at the bottom of the girder and to enhance the shear
capacity FRP fiber wrapping was done on the girder. The
prestressing force given to the CFRP laminates was 8 ton.

Fig. 10: Prestressed laminates with anchor plates (Rai &

Bambole, 2010) (Courtesy:- M/s R & M International Pvt Ltd)

Conclusions
Strengthening using active prestressed CFRP composite
Fig. 8: ROB at Karal junction of JNPT (Rai & Bambole, 2010) system yields in maximum utilization of FRP material
(Courtesy:- M/s R & M International Pvt Ltd) properties. This results in better performance of the
structural member post strengthening. Also, looking
Static and dynamic load tests on 14 spans out of total at the economic factor it can be said that use of active
30 spans were carried out on the bridge before and prestressed FRP system results in lesser use of materials
after strengthening. Monitoring of flexural strain, shear which results in more economical solution compared to
strain, deflection and vibration was carried out during the passive system. The two case studies elaborated in this
same. Omega type displacement transducers were used paper details on the use of active prestressed FRP system
for measurement of flexural and shear strain at critical for strengthening. It was subjected to load testing post
locations. Vibration measurement was carried out using strengthening and were found to show an improvement in
piezoelectric accelerometers. There was significant their performance as mentioned below:-
reduction in strain, deflection and vibration observed post 1. It was observed from load testing on the strengthened
strengthening (Rai & Bambole, 2010). bridge girder an average reduction in parameters
namely deflection of 26 %, flexural strain of 53% and
shear strain reduction of 56.8% (Rai & Bambole, 2010)
2. Reduction in acceleration in the bridge girder as
measured was 50% (Rai & Bambole, 2010).
3. The flat slab which was strengthened showed a
recovery of 80.95% post 48 hours loading on the
strengthened slab.

References
1. American Concrete Institute, 2008. ACI 440.2R-08, Guide for the
design and construction of externally bonded FRP systems for
strengthening concrete structures.
2. Task group 9.3 FRP reinforcement for concrete structures,2001,
Externally bonded FRP reinforcement for RC structures, fib CEB-
FIP, Technical report bulletin 14.
3. Irwin, R., Rahman. A., 2002. FRP strengthening of concrete
structures – Design constraints and practical effects on construction
detailing, New Zealand Concrete Society Conference 2002., Section.
4 Paper 2.
4. Sheikh, S.A., Li, Y., 2007. Design of FRP confinement for square
Fig. 9: Prestressing of FRP laminates at bottom of girder (Rai concrete column, Engineering Structures 29 (2007) 1074–1083
& Bambole, 2010) (Courtesy:- M/s R & M International Pvt Ltd)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

478 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Strengthening using Active Prestressed CFRP in RCC Structures

5. Grande, E., Imbimbo, M., Sacco, E., 2008. A simple model for 8. El-Hacha, R., Aly, M.Y.E., 2013. Anchorage system to Prestress FRP
evaluating the structural response of masonry structures Laminates for flexural strengthening of Steel-concrete composite
strengthened with FRP, The 14th World Conference on Earthquake girders, Journal of Composites for Construction © ASCE ( 2013).
Engineering, October 12 – 17, Beijing, China Vol 17, pp. 324- 335.
6. Mandara, M., Piazza, M., Perdikaris, P., Schaur, C., 2002. Repairing 9. Rai, G.L., 2013. Pre-stressing Techniques for structural
and strengthening for new requirements: Use of Mixed Technologies, strengthening using FRP composites, The Masterbuilder, June 21
Improving Buildings Structural Quality by New Technologies 2013.
Seminar, Lisbon.
10. Rai, G.L., Bambole, A.N., 2010. Strengthening of bridge by Pre-
7. El-Hacha, R., Wright, R.G., Green, M.F., 2001. Prestressed fiber- Stressing at JNPT and testing its efficacy, ACI Fall 2010 Convention
reinforced polymer laminates for strengthening structures, Prog. ACI SP 285.
Struct. Engng Mater .2001., vol. 3, pp. 111-121.

Dr.Gopal Rai
Dr. Gopal Rai, Chairman of Dhirendra Group of Company is a doctorate in Civil Engineering (Structural
Engineering) from Indian Institute of Technology, Bombay.
He has been involved in planning, designing and execution of major challenging projects like Mumbai
International Airport Runway Bridge strengthening, Strengthening of bridge over the Mithi River in
Mumbai, JNPT ROB bridge strengthening, T2 terminal Sahar Bridge, HPCL PDA structure, Mantralaya
(Maharashtra) Fire Strengthening, Western Railway distressed Bridge and Rajiv Gandhi Cancer Hospital.
He has also conducted 50 national and international seminars and conferences and participated in more
then 200 conferences. He has also published 4 international publications in prestigious Journals, 32 in
national publications, 5 Interviews with prestigious publications.
He is Committee Member for ISSE (Mumbai Centre), Bureau of Indian Standards and American Concrete
Institute. He is also serving as General Secretary for Association of Structural Retrofitting, and Co-
Chairman for Indian Institute of Bridge Engineers, IIBE, MH Center.

Organised by
India Chapter of American Concrete Institute 479
Session 4 C - Paper 4

Effect of Fiber and Silica Fume on High Performance Concrete

Mr.Sushil Kumar Swar, Dr. Sanjay Kumar Sharma, Mr.Paaras Gupta
Department of Civil Engineering, National Institute of Technical Teachers Training and Research. (NITTTR), Chandigarh.. India.
Dr. Hari Krishan Sharma
Department of Civil Engineering, National Institute of Technology (NIT), Kurukshetra, India.

Abstract have played an important role in prolonging structural

life, increasing the capacity of characteristic load and
High Performance Concrete (HPC) is concrete having
delaying the damage process of concrete. The presence
improved properties such as high strength, high workability
of fibers reduces the crack width increases number
and toughness, energy absorption capacity and corrosion
of cracks and ductility and delaying final crushing of
resistant and is designed to satisfy a list of requirements
concrete. The effectiveness of steel fibers in cracks in the
based on the hardened properties. The main difference
concrete is related to the average spacing of fibers inside
in the tensile load deformation behaviour between high
the matrix[8]. The greatly improved ductility and of fiber-
performance and conventional Fiber Reinforced Concrete
reinforced concrete make it quite effective under cyclic
(FRC) is in the multiple (MC) micro-cracking stage, which
and dynamic loads and have led to its applications such
may not exist in conventional FRC. Several trial mixes were
as blast-resistant construction, highway pavements, and
designed and concrete cubes casted and tested to achieve
repair of infrastructure facilities [9].
the desired compressive strength. Steel fibers used in
this study are thin strands of length 10mm and diameter Extensive research has been conducted on the
0.3mm. The fiber content varied from 0 to 15percent at development of HPC using steel fibers. However, limited
intervals of 3% by weight of cementitious material in the literature is available on the development of high strength
concrete. By the experimental investigations the value of high performance concrete .Several researchers have
compressive strength, split tensile strength and flexure studied the individual effect of addition of silica or steel
strength with various percentages of fibers is calculated fibers on the strength of high performance concrete of
and compared. In this paper, an attempt has been made to higher grades but no study has yet been conducted to
present the results of an experimental investigation carried investigate combined effect of addition of silica fume and
out on steel fiber reinforced high strength high performance steel fibers on the HSHPC of higher grade. Therefore, an
concrete of grade near to M95 and higher which is designed attempt has been made in the present study to investigate
using silica fume as the filler material thus forming as a combined effect of addition of silica fume and steel fibers
High Performance Fiber Reinforced Cement Composites on the structural characteristics of HSHPC.
(HPFRCC).
Keywords: HPC High Performance Concrete, FRC Fiber High Performance Concrete
Reinforced Concrete, MC Micro-Cracking, HPFRCC High Generally speaking, the definition of ‘High Performance’
Performance Fiber Reinforced Cement Composites. is meant to distinguish structural materials from the
conventional ones, as well as to optimize a combination
Introduction of properties in terms of final applications related to the
civil engineering. The most interesting properties are e.g.
Traditionally, high performance concrete (HPC) may be
strength, ductility, toughness, durability, stiffness and
regarded as synonymous with high strength concrete
thermal resistance. The specific term “high-performance
(HSC). In the recent years, there is great development
fiber-reinforced cement-based composites” refers to
in the mineral and chemical admixtures, Pozzolanic
high performance cement-based materials, particularly
admixtures like fly ash, silica fume, and micro silica are
developed for specific applications, for which toughness,
extensively used these days to enhance performance
ductility, and energy absorption are fundamental
characteristics of concrete. Silica fume is highly reactive
properties. Developments of these materials were
filler material due to high proportion of non-crystalline
possible due to:
SiO2 and the large surface area. At silica fume dosage of
10 percent by mass of cement, there exist 50,000–100,000 (a) The introduction of new reinforcement systems.
silica fume particles per cement grain [1].
(b) The study of fiber-matrix interface, and consequent
From the previous studies, it is known that the addition of optimisation of adhesion properties.
steel fibers in concrete resulted in crack resisting effect
(c) The development of high-performance cementitious
on the initiation and propagation of crack. Steel fibers
matrixes, which greatly improved micro-structural

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

480 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fiber and Silica Fume on High Performance Concrete

properties in terms of strength and durability. Under conventional concrete mixing, substantial contents
of agglomerates almost always remain in the concrete.
(d) The innovation in processing techniques (including
Thus the assumption that the densification process is
controlling chemical reactions) which allows obtaining
somehow ‘reversible’ is not generally warranted. The
materials with surprising high toughness and low
sizes of undispersed agglomerates remaining in concrete
porosity properties.
after mixing often exceed the sizes of Portland cement
The composition of HPC usually consists of cement, water, particles, thus limiting any potential benefits attributed
fine sand, super plasticizer, fly ash and silica fumes. to the fine particle filler effect. Large undispersed grains
Sometimes, quartz flour and fiber are the components appear to always undergo chemical reaction in concrete,
as well for HPC having ultra-strength and ultra-ductility, but such reactions may induce ASR damage only under
respectively. The key elements of high performance especially unfavorable circumstances.
concrete can be summarized as follows:
Fischer G. et al. (2007)[4] described the results of
ll Low water-to-cement ratio, investigations on FRC particularly Engineered
Cementitious Composites (ECC) that the transition from
ll Large quantity of silica fume (and/or other fine mineral
ECC material properties to the behaviour of structural
powders),
members steel reinforcement in tension. Their similar
ll Small aggregates and fine sand, elastic/plastic material properties lead to compostable
ll High dosage of super plasticizers, deformation of both components in the elastic and
inelastic deformation regime. Consequently, damage
ll Heat treatment and application of pressure which are induced by local slip and excessive interfacial bond stress
necessary for ultra high strength concrete after mixing between steel reinforcement and cementitious matrix
(at curing stage). was prevented, resulting in improved performance of the
reinforced ECC element in terms of axial loading capacity,
Literature Review ductility, and composite integrity.
Shannag M. J. et al. (2005)[10] investigated that using high Ganesan N. et al. (2007)[5] described the experimental
performance steel fiber reinforced concrete (HPSFRC) in results of ten steel fiber reinforced high performance
place of ordinary concrete in the joint region significantly concrete (SFRHPC) exterior beam-column joints under
improved the seismic behaviour of non-seismically cyclic loading. The M60 grade concrete used was
designed beam column joints such that higher load designed by using a modified ACI method. Volume fraction
level moments and curvatures were attained and larger of the fibers used in this study varied from 0 to 1% with
displacements and curvatures ductility were recorded. an increment of 0.25%. The key findings were (1) Addition
Slower stiffness degradation and higher energy dissipation of fibers to the beam-column joints decreased the rate of
was achieved, and significant increase in the joint strength stiffness degradation appreciably when compared to the
was recorded. joints without fibers. Hence the technique of inclusion of
steel fibers in beam column joints appeared to be a useful
Naaman A.E. et al. (2006)[7] presented that behaviour
solution in the case of joints subjected to repeated or
of strain hardening FRC composites can be essentially
cyclic loading. (2) During the testing it has been noted that
characterized by providing a minimum of three key
addition of fibers could improve the dimensional stability
properties, which were their tensile strain capacity, and
and integrity of the joints. Also it was possible to reduce
elastic modulus. The tensile response of fiber reinforced
the congestion of steel reinforcement in the beam-column
cement (FRC) composites can be generally classified in
joints by replacing part of the ties in t3he columns by
two distinct categories depending on their behaviour after
steel fibers. (3) Load carrying capacity of the joints also
first cracking namely either strain hardening or strain-
increased with the increasing fiber content.
softening.
TsonosA.G.et al. (2008)[14] carried out an experimental
Sidney Diamond and Sadananda Sahu (2006)[11]
investigation to evaluate the retrofitting methods to
investigated use of dry densified silica fumes which was
address the particular weaknesses that were often found in
the most common form of silica fume used in current
reinforced concrete structures, especially older structures
concrete practice as the alternative to slurred silica
which lack sufficient flexural and shear reinforcement
fume has become unavailable in many places. Densified
within the columns and lack adequate shear reinforcement
silica fume was commonly supplied and consisted of
within the joints. Thus, the use of a reinforced concrete
particles of sizes up to several millimetres which were
jacket and a high-strength fiber jacket for cases of post-
generally not dispersible into individual silica fume
earthquake and pre-earthquake retrofitting of columns
spheres. Densified silica fumes from some sources can
and beam–column joints was investigated experimentally
be dispersed by moderate ultrasonic treatment into small
and analytically. The effectiveness of jacket styles was
clusters or chains of spheres and others resist such
also compared. The results indicated that the beam-
treatment and mostly remain as large agglomerates.

Organised by
India Chapter of American Concrete Institute 481
Session 4 C - Paper 4

column joint specimens strengthened with carbon-epoxy floor subassembly typical of a (half-scale) mid storey of a
jacketing were effective in transforming the brittle joint high-rise RC moment frame building in New Zealand which
failure mode of 14 specimens into a ductile failure mode was tested under quasi-static reversed cyclic loading to
with the development of flexural hinges into the beams. investigate the effect of precast-prestressed floor units
and transverse beams on the seismic performance of
Liberato Ferrara et al.(2010)[6] presented the study
RC moment-resisting frames. The design, fabrication,
mentioning that the orientation of steel fibers within a
instrumentation, and structural behaviour of the three-
SFRC structural element can be effectively governed
dimensional specimen tested. The results show that the
through a well balanced fresh state performance, as
damage to external plastic hinges (closer to the corner
obtainable by virtue of an appropriate mix composition,
columns) was significantly more severe than the internal
and a suitably designed casting process. The orientation of
hinges (adjacent to the intermediate columns), and that
the fibers significantly affects as expected the mechanical
the elongation measured in the external hinges was more
performance of the fiber reinforced cementitious
than double that in the internal plastic hinges.
composites. This may be a discriminating factor to obtain
a high mechanical performance of the material, such as
for example deflection hardens or even reliable strain Extract of Literature Review
hardening behaviour, besides suitable selection of fiber 1. The introduction of fibers as secondary reinforcement
type, fiber volume fraction, fiber-matrix bond and matrix in a cementitious matrix bridges across the cracks that
characteristics. develop in concrete and provide a significant increase
in toughness. Fibers uniformly improve the structural
Xilin Lul et al. (2012)[15] analyzed by detailing beam–column quality and also the inherent flexural strength of
joints of RC frame structures in regions of high earthquake concrete. Internal stresses due to shrinkage are
risk which were normally governed by code provisions restrained by uniformly mixed fibers in the concrete.
(GB 50010-2002, ACI 318-05, NZS 3101 1995, EC8
2003) that require a considerable amount of transverse 2. There is an improvement in the properties like strength,
reinforcement to resist the horizontal joint shear forces. plastic shrinkage, impact resistance and crack width
In recent years, a lot of non-conventional methods have reduction with the addition of fibers in both normal and
been developed to improve the performance of RC beam- high strength concrete with higher cost effectiveness.
column joints under seismic loading, such as the joints 3. The improvement in the properties of concrete due to
with fiber reinforced polymer (FRP). This paper focuses on addition of fibers depends upon the fiber type, strength,
the design of interior beam–column joints with additional elastic modulus, geometry, surface characteristics,
bars of different sizes and configurations. Ten full-scale aspect ratio and percentage content of fiber.
specimens were tested and the detailing had been
shown to be effective in improving the seismic resistance 4. The workability is influenced by the incorporation of
of joints. Initial findings on the use of cross diagonal fibers, the main factor being the fiber length.
reinforcement in interior beam–column joints have been 5. The addition of steel fibers enhances the fatigue
rather promising. strength, ductility and strain at peak load and energy
Durai S. et al. (2013)[3] reported reinforced concrete absorption capacity significantly. Hence SFRHPC
buildings with portion of columns that are common to appears to be useful material in case of structures
beams at their intersections are called Beam-Column subjected to cyclic/impact/repeated/seismic loading.
joint. The joints constituent materials have limited
strength and limited force carrying capacity. The authors Some Gap Areas
reported that the practical joint detailing using hairpin- The critical review of published literature illustrated
type reinforcement is a competitive alternative to closed that most of the studies are limited to normal strength
ties in the joint region. The reinforcement detailed as concrete and research in the area of high performance
per IS 456-2000 and IS 13920-1993 had been tested with densified small particles concrete beam-column joint is
and without diagonal cross bars in the joint region and limited. Most of the investigators have used SFRC in the
achieved better results. For reducing the plastic shrinkage beam-column junctions and fiber volume content was
cracks in concrete, HPC mixes with 10% silica fume and restricted to 2% by volume. Some of the researchers
0.3% glass fiber showed better results. In the future utilized SIFCON in the beam-column junctions with normal
line of work, exterior beam-column joint with diagonal strength concrete. It was reported that the enhancement
cross bars in the joint and HPC mix with 10% silica fume, in strength and ductility associated parameters and
0.3% glass fiber was tested and behaviour of specimen energy absorption capacity was not significant, primarily
with reinforcement detailing as per construction code of because of low fiber volume contents. The effect of other
practice and ductile detailing has been studied. variables like aspect ratio, fiber volume content and
Dhakal R. P. et al. (2014)[2] studied and analyzed behaviour fibers types was also not investigated. The mechanism
of one-storey, two-bay reinforced concrete (RC) frame- for improved tensile strain capacity of discontinuous

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

482 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fiber and Silica Fume on High Performance Concrete

fiber composites is also yet not well understood. Further,

the research work was confined to study behaviour of
structural members independently using normal weight
concrete only. Since, beam-column, beam-column-slab
& slab-column junctions are vulnerable locations which
are subjected to high horizontal and vertical forces, use
of fiber reinforced concrete with low fiber volume fraction
like SFRC was found to be inadequate.[12]

Materials
The material required is collected and the data required
for mix design are obtained by sieve analysis and specific
Fig. 1: Compressive Strength of Cement
gravity test. Sieve analysis is carried out from various
fine aggregate and coarse aggregate samples and the
samples which suit the requirement are selected. Specific Table 2
gravity tests are carried out for fine aggregate and course Sieve Analysis Results of Fine Aggregate as per IS-383
aggregate. The various materials are tested as per Indian
standard specifications. Raw material required for the %
Weight % Cumulative Passing
concreting operations of the present work are cement, Sieve Grading
Retained Weight % Wt. Zone
Size Zone
fine aggregate, coarse aggregate, silica fumes, steel (KG) Retained Retained
fibers, water and super plasticizer.
4.75mm 0.002 0.2 0.2 99.8
Cement
Portland Pozzolana Cement (43 MPa+) confirming to IS: 2.36mm 0.048 4.8 5 95
1489-1991 (Part-I) is used in all fiber tests & different trial
mixes also done with Ordinary Portland Cement (43 MPa+). 1.18mm 0.236 23.6 28.6 71.4
The test results of physical properties of cement are given Conformin
below in Table 1 and Figure 1 shows strength of cement. g
600 µ 0.202 20.2 48.8 51.2 to
Zone
II
Table 1 300 µ 0.434 43.4 92.2 7.8 as
Physical Properties of Cement per
150 µ 0.064 6.4 98.6 1.4 IS
Recommendation ­383
Sr. Observed
Characteristics Value as per
No Value 75 µ 0.01 1 99 0.4
IS:1489-1991(Part-I)

1. Normal Consistency 32.50% -

Pan 0.004 0.4 100 0
Specific Gravity 3.05 3.15
2. Not less than 30
Setting Time (minutes)
minutes
Not more than 600
Initial Setting Time 96 Table 3
3. minutes
Physical properties of sand
Final setting time 187 Not more than 10
Fineness (Sieve Analysis) 2.25 Sr. No. Characteristics Observed Value
4. Compressive Strength 1. Specific Gravity 2.65
(N/mm2)
2. Fineness Modulus 2.7
1) 168 hours 35 Not less than33
5. 3. Water Absorption 2.30%
2) 672 hours 45.08 Not less than 43

Fine Aggregates Coarse Aggregates

Natural river sand obtained from and “UltraTech”. The Crushed aggregates, angular in shape, obtained from
sand is having maximum particle size of 4.75 mm as such “UltraTech”. Two nominal size 10 mm and 20mm shall be
it is sieved through 4.75mm, IS sieve before using it. The collected and combined to have a 20mm graded sample
sieve analysis of fine aggregate is given in Table 2and the as per IS: 383-1970. Physical properties of the combined
physical properties of sand is given below in Table 3.

Organised by
India Chapter of American Concrete Institute 483
Session 4 C - Paper 4

Water
Table 4
Physical properties of coarse aggregates As prescribed in IS: 456-2000, the potable water free
from deleterious material and fit for drinking purpose is
used for mixing as well as curing of concrete was used in
Sr. No. Characteristics Observed Value
experimental work.
1. Specific Gravity 2.7
Steel Fibers
2. Water Absorption 1.20% Steel fibers obtained from ‘Vinayaka Shot Pvt. Ltd. Indore’.
Figure 2 shows steel fibers were round straight type
3. Flakiness index 17.66%
having length 10mm and diameter 0.3mm and having
4. Elongation Index 23.33% aspect ratio of 33.33 was used. They shall be added in
the proportion of 3% to 15% (by weight of cementitious
5. Crushing Value 24.94% material) to the concrete mixes and mixed in the machine.
The tensile strength, carbon content and density is 1920
6. Impact Value 23.21% N/mm2, 0.82 % and7.85 g/cm3 respectively.

Table 5
Sieve Analysis of Graded Coarse Aggregate
(20 mm and 10 mm by 60:40 )

%
Weight % Cumulative Passing
Sieve Grading
Retained Weight % Wt. Zone
Size Zone
(KG) Retained Retained

40 mm 0 0 0 100

20 mm 0 0 0 100
Fig. 2: Steel Fibers 0.3mm and diameter 10mm length
16mm 0.026 0 0.87 0 8.7 99.13

Conformin Super Plasticizer

12.5mm 0.932 31.07 31.94 68.06
g to 20mm The trade name of the super plasticizer is ‘Auromix
graded as
10mm 0.842 28.07 60.01 39.99
per IS.383
400’ where as its brand name is “Fosroc”. Auromix 400
is a unique combination of the latest generation super
4.75mm 1.054 35.13 95.14 4.86 plasticizer, based on a polycarboxylic ether polymer with
long lateral chains having relative density of 1.02 at 25˚C.
2.36 0.114 3.8 98.94 1.06
This greatly improves cement dispersion. At the start of
Pan 0.032 1.06 100 0 the mixing process an electrostatic dispersion occurs but
the cement particle’s capacity to separate and disperse.
This mechanism considerably reduces the water demand
aggregate are given above. Aggregates used in saturated in flow able concrete.
surface dry (SSD) condition for this project work. The
physical property of coarse aggregates is given in Table
4. and the sieve analysis of coarse aggregates is given in
Methodolgoy
Table 5. Concrete mix design
High performance concrete mix was designed to
Silica Fumes
achieve M70 grade concrete using admixture as per ACI
Silica fume with specific surface area of 2200 m2/kg and Committee 211.4-08. Several trial mixes were carried
specific gravity of 2.02 was used in the experimental out to achieve HPC using high range water reducing
study. This silica fume is confirming to IS: 15388-2003. It agent (HRWR) and silica fume. Water-cement ratio
is supplied by ‘Elkem’ having grade 920 D. It is a dry silica was adjusted to have slump 0f 100 ± 5 mm. The details
fume powder. It improves the performance of concrete of the trial mixes are tabulated in Table 6. Concrete mix
and mortar formulations. It acts physically to optimise proportion corresponding to 1:1.36:1.87 (by weight) with
particle packing of the concrete or mortar mixture and water –cement ratio 0.24 %, HRWR dosage of 1.4% by
chemically as a highly reactive pozzolana. weight of cement and silica fume dosage of 12% by weight

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

484 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fiber and Silica Fume on High Performance Concrete

Table 6
Mix Proportion of Different Trial Mixes

Silica Coarse Water

Mix Cement Fine Agg Average Strength
Fume Fly Ash Aggregate (kg) Cement HRWR %
Design (Kg) (kg)
(kg) 10 mm 20 mm Ratio 7 Days 28 Days

M1 450 45 90 398 597 742 0.27 0.7 46.59 62

M2 465 93 - 460 690 800 0.27 1.2 54.24 65

M3 470 47 94 393 589 732 0.27 0.8 58.79 68

M4 485 53 - 1055 - 700 0.28 1.2 61.11 71

M5 500 50 100 385 577 717 0.27 0.7 59.22 72.23

M6 525 63 - 440 660 800 0.24 1.4 60.85 75.66

M7 563 63 - 1100 - 800 0-28 1.4 56.35 73.45

of cement, provided a concrete mix with compressive

strength 84 MPa, under normal water curing after 28 days
without fiber. Straight steel fiber in varying fiber volume
fraction corresponding to 3%, 6%, 9%, 12% and 15% were
used to produce steel fiber reinforced high performance
concrete of higher grades using same mix proportion.

Casting of concrete cubes, cylinders and beams

The test moulds are kept ready before preparing the
mix,moulds are cleaned and oiled on all contact surfaces
then placed on vibrating table. The moulds are filled in
layers and vibrated. For compressive strength 150 mm X
150 mm X 150 mm, for split tensile strength 100 mm X 200 Fig. 3: Testing Compressive strength of test specimen
mm cylinder and for flexural strength 100 mm X 100 mm
X 500 mm size moulds are used. The number and date of
ll The concrete cubes are casted according to IS:
casting are put on the top surface of the cubes.
516,”Methods of tests for strength of concrete”. The
specimens are left to stand in their mould at (27±2) ˚C
Curing
placed having at least 90 % humidity for 23 hours ± 15
The test specimen were stored in a place free from minutes. The moulds are then placed in curing tank at
vibration and covered with the wet gunny bags for 24 hours a temperature of (27±2) ˚C.
from the time of addition of water to the dry ingredient.
After this period, specimen are removed from moulds and ll The variations of compressive strength of the
immediately submerged in curing tank and kept there for specimens were tested as shown in Fig.2 at different
specified days. They are taken out just prior to test. The ages 7 days, 14 days and 28 days.
water of the curing tank was renewed for every seven days
and maintained at temperature of (27±2) °C. Split Tensile Strength
This test method measures the splitting tensile strength of
Test for Workability of Fresh Concrete by Slump Test concrete by the application of a diametrical compressive
Slump test is used to determine the workability of fresh force on a cylindrical concrete specimen placed with its
concrete. Slump test as per IS: 1199 – 1959 is followed. The axis horizontal between the platens of a testing machine.
apparatus used for doing slump test are Slump cone and Figure 4 shows the failure of cylinder by split tensile
tamping rod. strength test, normally testing of specimen without the
fibers the specimen at failure breaks in pieces completing
Compressive Strength of Concrete Mixes but by addition of fibers at failure the specimen does not
The compressive strength test as shown in Figure 3 is break completely at failure load but the there is a big
done by compressive testing machine as below. crack as shown in the figure without the whole specimen
breaking completely, making structure of fiber concrete
ll Normal curing method. more safe at failure load.
ll Normal Curing for 7 days, 14 days and 28 days.

Organised by
India Chapter of American Concrete Institute 485
Session 4 C - Paper 4

be made on each beam. Therefore, for the first test,

position the beam with one end about 30 mm from the
support.
3. The points of support and loading should be marked on
the beam.
4. The test should be carried out at a rate of loading
indicated by the instructor.
5. After the load test, the average depth and width of the
specimen at the failure section must be measured to
the nearest mm.

Fig. 4: Split tensile strength of cylinder Experimental Investigation

Mix design is carried out according to ACI Committee 211.4-
Flexural Strength 08. Recommended guidelines for concrete mix design to
Flexural strength is one measure of the tensile strength get the designed compressive strength and workability.
of concrete. It is a measure of an unreinforced concrete The procedure for design of micro silica in concrete mixes is
beam or slab to resist failure in bending as shown in according to IS: 15388-2003. The trial mix M6 was selected
Figure 5. from various trial mixes and same proportion was used
because of maximum strength and economy, henceM6 was
used forecasting of concrete cubes, cylinders and beams
with different percentages of steel fibers varying from 3%
to 12% of cementitious material.
ll Materials are weighted by weigh batching.
ll Steel fibers were added 0%, 3%, 6%, 9% and 12% to the
total cementitious material.
ll Concrete cubes of size 150 mm X 150 mm X 150 mm
and cylinders of size 100 mm X 200 mm and beams of
size 100 mm X 100 mm X 500 mm of “Forty five” each
were casted for compressive strength, split tensile
strength and flexural strength respectively.
Fig. 5: Flexure Strength test on beam
ll Method of test for determining the compressive strength
Flexure strength of specimen is tested as as below. of samples is according to IS: 516-1959,”Methods of
1. The beam will be tested on its side relative to the position test for strength of concrete”.
in which it was cast. ll Each set consist of 3 cubes, 3 cylinders, 3 beams and
2. The specimen should not be removed from the curing the test strength of the samples has been taken as the
tank until just before testing. Even a small amount of average strength of three specimens.
drying can adversely affect the results. Two tests will

Table 7
Strength of Concrete Mix with Different Percentage of Fibers

Avg. Compressive Strength Avg. Split Tensile Strength Avg. Flexural Strength
No Fiber
7 Days 14 Days 28 Days 7 Days 14 Days 28 Days 7 Days 14 Days 28 Days

1 0% 60.85 69.55 75.66 4.21 4.55 5.26 9.14 9.66 10.29

2 3% 63.67 71.15 83.78 4.42 4.9 5.73 10.73 11.48 12.57

3 6% 70.32 85.66 91.9 4.56 5.02 6.33 11.03 13.79 14.23

4 9% 76.18 88.15 94.54 5.12 5.67 6.68 12.72 13.95 15.58

5 12% 81.12 88.67 95.72 5.32 5.2 6.77 12.86 14.09 15.78

6 15% 82.56 91.7 96.23 5.55 6.24 6.82 12.98 14.15 15.86

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

486 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fiber and Silica Fume on High Performance Concrete

ll It is observed that the individual variation of specimen

strength with in ± 15 percent of average strength. Thus
the test results are quite consistent and reasonable.
ll Workability of fresh concrete with and without steel
fibers is measured by slump test apparatus.

Strength of Concrete Mixes

The compressive strength of sample is carried out
according to IS: 516, “Method of test for strength of
concrete”. The result of specimen strength under normal
water curing at ages 7 days, 14 days and 28 days have been
reported in Table 7. and plotted in graphs from figure 6 to
Fig. 6: Showing 28 days Strength verses Fibre percentage
Figure 9.

Conclusion
1. It is observed that compressive strength, split tensile
strength and flexural strength are on higher side for
9% fibers as compared to that produced from 0%, 3%
and 6% fibers.
2. Fibers percentages are taken with respect to
cementitious material (i.e Cement and Micro silica).
3. All the strength properties are observed with fiber of
diameter 0.3mm and length 10mm with aspect ratio of
33.33.
Fig. 7: Showing Compressive Strength verses Fiber % 4. It is observed that compressive strength increases
from 10.73% to 27.19% with addition of steel fibers.
5. It is observed that split tensile strength increases from
8.94% to 18.63% with addition of steel fibers.
6. It is observed that flexural strength increases from
22.16% to 54.13% with addition of steel fibers.
7. With Increase in fiber percentage after 9% to 15% there
is no appreciable increase in compressive split tensile
as well in flexural strength.
8. From 12% to 15% concrete is not workable and hence
9% is considered.
With reference to literature fiber volume percentage
Fig. 8: Showing Split Tensile Strength verses Fiber %
of concrete can be used in fine aggregate concrete to
achieve concrete above 95MPa. Strengthening of beam –
column joins can be done by high strength concrete during
construction to prevent failure and increase load carrying
capacity.[13]

Acknowledgments
I specially acknowledge ULTRA TECH Ready Mix Plant
staff, NITTTR Chandigarh Director, staff, ME students
and Sushil Thakur. Director of Technical Education,
Maharashtra, Government Polytechnic Nashik, Principal,
staff, students, family and all well-wishers who motivated,
guided, helped me for the successful completion of this
work.
Fig. 9: Showing Flexural Strength verses Fiber %

Organised by
India Chapter of American Concrete Institute 487
Session 4 C - Paper 4

References 8. Palanisamay, T. and Meenambal, T. 2008. Effect of GGBS and silica

fume on mechanical properties of concrete composites. Indian
1. Ashour, S. A., Wafa, F. F. and Kamal, M. I., 2000. “Effect of the
Concrete Institute Journal 9(1), 7-12.
concrete compressive strength and tensile reinforcement ratio
on the flexural behavior of fibrous concrete beams”,engineering 9. Paskova, T. and Meyer, C. 1997, “Low cycle fatigue of plain and fiber
structure 22(9), 1145–1158. reinforced oncrete”, ACI Materials Journal 94.
2. Dhakal, Rajesh P. Peng, Brian H. H. Fenwick, Richard C.; Carr, 10. Shannag M.J., Baradat,S., and Kareem, M.A., 2005. Behaviour of
Athol J. Bull, Des K. Copyright July 1, 2014.ACI Structural Journal Reinforced Concrete Beam-Column Joint-A Review”, Journal of
Cyclic Loading Test of Reinforced Concrete Frame with Precast- Structural Engineering, ASCE,26(3), 207-213.
Prestressed Flooring System. Page 777. 11. Sidney Diamond and SadanandaSahu2006 School of Civil
3. Durai S1, Boobalan SC2*, Muthupriya, P., Venkatasubramani, R., Engineering, Purdue University, U.S.A. Densified silica fume:
2013.Indian Journal of Engineering Volume 3, Number 6, April particle sizes and dispersion in concrete Materials and Structures
2013. Behaviour of high performance concrete in exterior beam Journal39: 849–859
column joint. 12. S.M.Swar, S.K.Sharma, H.K.SharmaJune2015. Performance
4. Fischer, G.., and Li, V.C., 2007. “Effect of Fiber Reinforcement Characteristics of HPDSP Concrete: An Overview “Proceedings of
on the Response of Structural Members” , Engineering Fracture RILEM Workshop HPFRCC-7,Stuttgart Germany”,79-86.
Mechanics, 74,258-272. 13. S.M.Swar, S.K.Sharma, H.K.Sharma, SushilKumar September
5. Ganesan, N., Indira, P.V., and Abraham, R.,2007. “Steel Fiber 2015“Structural Characteristics of HPDSP Concrete on Beam
Reinforced High Performance Concrete Beam-Column Joints column Joints”Proceedings International Conference.by WASET
Subjected to Cyclic Loading” .Journal of Earthquake Technology, in Los Angeles,USA.. International Science Index eISSN: 1307-
44(3-4), 445-456. 6892,waste.org.3044-3049.
6. Liberato Ferrara, Nilufer Ozyurt, Marco di Prisco, 2010 “High 14. Tsonos, A.G. Tegos, I.A., and Penelis, G.G., 2008. “Seismic Resistance
Mechanical Performance of Fiber Reinforced Cememntitious of Type 2 Exterior Beam Column Joints Reinforced with Incline
composites”: the role of “Casting-flow induced fiber orientation”, Bars”. ACI Structural Journal, 89(1), 3-12.
Material and Structures, April 2010, p 109-128 15. Xilin Lu1, Tonny H. Urukap2, Sen Li2 and Fangshu Lin* 2011 State
7. Naaman, A.E.,2006, “High Performance Fiber Reinforced Cement Key Laboratory of Disaster Reduction in Civil Engineering, Tongji
Composites”. Concrete Structures for the Future, IABSE Symposium, University, Shanghai, China Seismic behaviour of interior RC beam-
Paris-Versailles, 371-376. column joints with additional bars under cyclic loading.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

488 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fiber and Silica Fume on High Performance Concrete

Mr. Sushil Kumar Swar

BE (COEP), ME (Thapar),
Research Scholar, Department of Civil Engineering,
National Institute of Technical Teachers Training and Research. (NITTTR), Chandigarh. India.
E mail: swarsushil@rediffmail.com. In teaching from 1986, deputed by Govt of Maharashtra for
Ph-d under QIP (Poly) Scheme.

Dr. Sanjay Kumar Sharma

Ph.D (Panjab University), M.Tech, B.Tech
Professor, Department of Civil Engineering,
National Institute of Technical Teachers Training and Research. (NITTTR), Chandigarh. India.
Email.sanjaysharmachd@gmail.com.

Dr. Hari Krishan Sharma

B.Sc. Engg. M.Sc. Engg. Ph.D
Professor, Civil Engineering Department,
National Institute of Technology (NIT), Kurukshetra. Haryana. India.
Email: hksharma1010@yahoo.co.in

Mr. Paaras Gupta

BE Civil, ME CTM, Department of Civil Engineering,
National Institute of Technical Teachers Training and Research.(NITTTR),
Chandigarh. India
Email: rckgupta2@gmail.com

Organised by
India Chapter of American Concrete Institute 489
Session 4 C - Paper 5

Performance Based Concrete for Residential and

Commercial Structures

Kalahasti Srikanth Raajesh Ladhad Siddappa A Hasbi

Borregaard South Asia Pvt Ltd, Structural Concept Designs Pvt Ltd, Corniche India Pvt Ltd,
Navi Mumbai Navi Mumbai Navi Mumbai

Abstract evaluated or produced based on strength not giving any

credit to durability aspects. This has led to the study we
Performance Based Concrete is not all about target
have done and are still doing since late November 2014
strength; it’s about the optimization of Coarse Aggregate,
and have taken forward these mixes to many ongoing
Fine Aggregate, Cement content and Water content in
projects. The objective of this study was to design concrete
the Concrete. To achieve this either GGBS (Slag) or PFA
mixes using local aggregate sources, ggbs (slag), silica
(Pulverized Fly Ash) is blended with pure OPC 53 grade
fume that increases durability, by optimizing the coarse
Cement along with Silica Fume. This triple blend concrete
and fine aggregate amount, gradation, minimize cement
is significantly more resistant to the ingress of chloride
and water content while maintaining workability.
ions, water permeability and reduced temperature rise
in concrete lowering the chances of thermal cracks. The solutions lie in provision of well-balanced, consistent
Performance Based Concretes have higher resistance mixes with good rheological properties, control and
to Sulphate attacks and Alkali aggregate reactivity maintenance of an adequate slump, and provision
than those made with pure OPC. A good quality mid- of adequate mortar for workability, pumpability and
high range water reducing chemical admixture with finishability. Concrete mixtures produced with a well-
Lignosulphonates is used to keep the water/cement ratio graded aggregate combination tend to reduce the
less than or equal to 0.40. Performance Based Concrete need for excess water, provide and maintain adequate
is a Techno-Commercial solution to the ongoing and workability, require minimal finishing, and consolidate
future construction projects. without segregation. In order to understand concrete as a
material and why some of the “old concrete” continues to
Keywords: Performance Based Concrete (PBC), Ground
perform well, had never been problem in years and some
Granulated blast slag (GGBS), Pulverized Fuel Ash (PFA),
of the “new concrete” which starts to deteriorates soon
Silica Fume (SF), Rapid Chloride Permeability (RCPT),
after construction is a concern and challenge.
Water Permeability (WP).

Introduction Industry Challenges Today

Through years of practice and experience, concrete has Fine aggregate
proven to be a suitable material for construction. It has been Demand for crushed fine aggregates for producing
implemented in numerous residential and commercial concrete is increasing drastically over natural sand for
projects. However, deteriorating condition of the concrete construction industry. For example Pune district alone
is often a worry or never taken into consideration, forcing consumes 30000 to 35000 tonnes of fine aggregates per
authorities and researchers to enhance the properties of day. Natural sand availability is limited; on the other hand
this material. PBC with its improved resistance to loads crush rock fine is inevitable. Crush rock fine aggregates
and environmental conditions has become an increasingly have been regularly used to make concrete for decades in
promising material to assist with the problem. PBC is India. Both natural sand and crush rock fine aggregates
interpreted in many ways but often in manners that differ varies in terms of shape, silt content and fines content.
from the accepted definition developed by the American To have full control on the concrete produced and to have
Concrete Institute. Mehta and Aitcin introduced the term less uncertainty, addition of GGBS (Slag) and SF becomes
“High Performance Concrete” in 1990. For example, necessary.
addition of slag/fly ash and silica fume to concrete reduces
the porosity of the concrete and increases it’s durability.
Fly ash Concrete
High strength is not a PBC criterion. The greatest need for
performance based concrete is in construction in which Strength development in concrete using a given fly
the normal concrete strength requirement is 35Mpa or ash may vary with different cements just as strength
less. Unfortunately, the trend has been that concrete is development of different fly ashes will be different with

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

490 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Performance Based Concrete for Residential and Commercial Structures

the same cement. For a given good quality fly ash also, fly per ASTM C 1202 (low Permeable concrete) and WP is
ash concrete is more sensitive to curing conditions than less than 10 mm as per DIN 1048, Part 5.
plain concrete and needs relatively longer curing more
than 21 days. In practice, most of the concretes in India Why GGBS (Slag)?
are never cured more than 21 days. Addition of slag and silica fume helps in more C-S-H
gel contributing to higher compressive strengths and
Control of water content in mix: low heat of hydration. The replacement level of GGBS
The water content in the concrete should be properly can be as high as 60% of cement, which is about twice
controlled. Variation in mixing water demand due to as much of PFA (typically replacement level is 30%).
temperature and or free water in aggregates is usually Hence, partial replacement of GGBS can enable higher
not done accurately. reduction of cement content. As the manufacture of one
tonne of cement generates about 1 tonne of carbon dioxide
and it is considered more environmentally friendly to
Performance Based Concrete
adopt GGBS owing to its potential higher level of cement
A performance based concrete includes ground granulated replacement. In terms of cost consideration, the current
blast furnace slag, silica fume and mid-high range water market price of GGBS is similar to that of PFA. As the
reducer. Performance based concrete properly designed potential replacement of GGBS is much higher than PFA,
will give typical results of RCPT below 1000 Coulombs as substantial cost savings can be made by using GGBS.

Table 1
Trial Concrete Mixes

Mix ID Source Mix-1 Mix-2 Mix-3 Mix-4 Mix-5 Mix-6

Cement OPC 53 Grade Ultratech Cement 412.5 300 300 300 300 300

Fly Ash (PFA) Dirk India Pvt Ltd 0 100 0 0 0 0

GGBS (Slag) JSW Cement 0 0 100 100 100 100

Silica Fume Corniche India Pvt Ltd 0 12.5 12.5 12.5 12.5 12.5

Aggregate Content
20 mm stone Swastik Quarry-Turbhe 628 628 628 628 628 628

10 mm Stone Swastik Quarry-Turbhe 430 430 430 430 430 430

Crush Rock Sand Swastik Quarry-Turbhe 820 820 820 820 820 820

Water MIDC-Koper Khairane 185 185 185 177 171 165

W/B Ratio - 0.45 0.43 0.43 0.43 0.42 0.40

Wet Density (Kgs/m3) - 2475 2475 2475 2475 2475 2475

Admixture Dosage (% lbwc) CAC Pvt Ltd,

1.2 1.2 1.1 1.1 1.1 1.1
PCE based with Lignosulfonate H & R Johnson (India)

Slump Measurement
10 min - 200 185 200 205 205 200

30 min - 175 190 200 195 170 145

60 min - 140 120 180 180 145 140

90 min - 120 100 170 100 120 125

Compressive Strength in Mpa

3 Day - 22.3 16.0 19.2 22.7 21.2 22.8

7 Day - 31.3 23.2 31.2 36.7 37.4 38.9

28 Day - 43.2 45.6 49.7 57.2 57.2 56.7

90 Day - 53.1 56.2 65.3 71.3 71.0 72.4

Rapid Chloride Permeability (Coulombs) 2400 235 165 210 165 105

(ASTM C1202) Water Permeability in mm (DIN 1048 Pt.5) 28 8 7 3 3 3

Organised by
India Chapter of American Concrete Institute 491
Session 4 C - Paper 5

Why Silica Fume?

Silica fume is an ultrafine material with spherical particles
less than 1 μm in diameter, the average being about 0.15
μm. This makes it approximately 100 times smaller than
the average cement particle. The bulk density of silica
fume depends on the degree of densification in the silo
and varies from 180 to 700 kg/m3. The specific gravity of
silica fume is generally in the range of 2.2 to 2.3.
Generally speaking the bulk density of silica fume used
in the laboratory studies and field trials were in order
of above 600 kg/m3. Silica fume is fantastic Pozzolanic
material (super Pozzolana) which contributes to strength
of concrete, rheology of cementitious material when Fig. 1: Laboratory Concrete mixer
fresh and decreases porosity of concrete. The SiO2
content of silica fume used in laboratory work, trials and
ongoing projects is greater than 88.0%.

Why mid-high range water reducers with

Lignosulphonates?
Superplasticizers influence the ratio of water to
cementitious material. By reducing the amount of water,
the cement paste will have higher density, which results
in higher paste quality. An increase in paste quality will
yield higher compressive and flexural strengths, lower
permeability, increase resistance to weathering, improve
the bond of concrete and reinforcement, reduce the volume
change from drying and wetting, reduced shrinkage
cracking. To ensure workability of concrete at low w/c a
Fig. 2: Performance Based Concrete in the mixer
retarder is definitely needed. A robust Lignosulphonates
helps concrete to keep its workability up to 2 hrs and
keeps the concrete in cohesive state.

Experimental Work
Trial mixes were done for the optimized aggregate
gradation with and without fly ash/slag and silica fume by
varying the water-cement ratio and the cement content.
Sieve analysis was done for the various sizes of the
aggregates to determine their individual gradations. Then
combined gradation was obtained by blending two coarse
aggregates and crush rock sand. The best possible blend
with the available aggregate sizes that matched the target
gradation was obtained by trial and error. The aim was to
obtain a gradation that would satisfy as nearly as possible
to well gradation or maximum density line. Fig. 3: Comparison of Chloride permeability values

All the concrete mixes were done in horizontal shaft

laboratory mixer specially developed, see Figure 1 & 2.
Mix-1 is a control mix with cementitious replacement or
additive. Mix-2 is with 100 Kgs of fly ash and Mix-3 to Mix-6
is with 100 Kgs of slag as cement replacement. 12.5 Kgs of
silica fume is added per m3 of concrete.
It was evident and as expected the control mix showed
high RCPT value as well as WP up to 28 mm. All the other
concrete mixes which include either PFA or GGBS and
SF showed very low chloride, see Figure 3, Figure 4 and
Fig. 4: RCPT Samples

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

492 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Performance Based Concrete for Residential and Commercial Structures

Fig. 5: WP Samples Fig. 6: Batching Plant, Bhagwati Imperia, Bhagwati Developers

Figure 5, RCPT & WP values making the concretes more

durable.
Compressive strengths for mixes Mix-2 to Mix-6 were in
the range of 45 to 60 Mpa at 28 days and 55 to 75 Mpa
at 90 days. Concrete mixes were also easily workable
up to 90 minutes with slump measuring 125 +/- 25 mm.
The concrete had no bleeding, no segregation and was
a cohesive mix. Use of mid-high range water reducer
with Lignosulphonates made the mixes more robust for
handling. Fig. 7: Bhagwati Imperia, G+12, Ulwe, Bhagwati Developers

Ongoing Projects
From the experience and results obtained from laboratory
concrete trials we have now successfully implemented
these mixes in many new and ongoing projects in Navi
Mumbai. The concretes produced here are of M40-
50 grade with low RCPT and WP values. See Figure 6
batching plant installed at Bhagwati Developers at project
site Bhagwati Imperia at Ulwe, Navi Mumbai. Thanks
to advances in concrete production equipment, many
construction projects are today equipped and willing to
Fig. 8: Niharika Mirage, G+12, Kharghar, Juhi Developers
install on-site batching plants. Batching plants have
helped to produce concrete to the exact specifications and
quantities required. Typical concrete mix designs showed
in Table 2.

Table 2
TG/DTA of NC and concrete modified with CDA

Constituents Kgs/m3 Comments

Cement OPC 53 Fig. 9: Today Genesis, G+13, Ulwe, Today Global Group
300-325 Depending upon raw materials
Grade
Conclusions
GGBS (Slag) 100-125 Depending upon raw materials The PBC produced is very promising and an eye opener
because of below techno-commercial reasons:
Silica Fume 12.5 -
1. Longer service life of the RCC Structure.

Chemical Admixture 1.0-1.2 % Depending upon raw materials 2. Nil repair and maintenance costs of the structure.
3. Techno-commercial concrete costing around 3800-
Water/Binder Ratio 0.40-0.41 Depending upon raw materials 4200 Rs/m3 produced at site.

Organised by
India Chapter of American Concrete Institute 493
Session 4 C - Paper 5

4. PBC will allow RCC consultants to reduce the amount Additives & Chemicals Pvt ltd and Mr. Arup Dana from H
of steel in the building. & R Johnson for providing us samples of superplasticizers
for laboratory and field trials.
5. Problems like inadequate workability, bleeding and
segregation can easily be controlled to larger extent
using PBC concept. References
1. Aykut Cetin and Ramon L. Carrasquillo, “High-Performance
6. We are able to give one common quality concrete for Concrete: Influence of Coarse Aggregates on Mechanical
all structural elements namely the raft, retaining wall, Properties”, ACI Materials Journal, May-June 1998, pp. 252-261
footings, columns, beams and slabs. This will minimise 2. Combinations of Poly Carboxylate ethers and Lignosulfonates in
the confusion relating to the concrete mix at site. chemical admixtures for special performance, American Concrete
Institute (ACI), spring 2012
3. Silica Fume Association (SFA) - User Manual, US Department of
Acknowledgement Transportation, Federal Highway Association document number
FHWA-IF-05-016
We would like to thank Prajapati Constructions, Bhagwati
Developers, Today Prachar Realty, Juhi Habitat and Varun 4. Standard Test Method for Electrical Indication of Ability of concretes
to resist chloride ion penetration, ASTM C 1202-2000
Enterprises to be a part of this assignment. We would
5. DIN 1048-Part 5 “Testing of concrete”-Testing of Hardened concrete
also like to thank Mr. Sandeep Khedekar from Concrete
(Specimens prepared in mould)

Srikanth Kalahasti
Srikanth Kalahasti is a Civil Engineer. He has done his graduation from Osmania University and post-
graduation from South Dakota School of Mines & Technology, South Dakota. He has worked for 8 years at
the “All State Engineering & Construction Services” located in Virginia as an Civil/Structural Engineer
and Designer and has overseen many projects onsite as Quality Manager in Washington DC Metro area.
Over the last 6 years he is working with Borregaard, India overseeing the R&D activities for cement and
concrete application.

Raajesh K. Ladhad
Raajesh K. Ladhad is a Civil Engineer, graduated from B. V. B. College of Engineering and Technology. He
started his career as a Design Engineer with M/s. P. T. Gala Consulting Engineers located in Mumbai and
was associated with the company for five and a half years. He started his own firm Structural Concept
Designs Pvt. Ltd located in Navi Mumbai, India, in the year 2001. Since then he has worked on various
types of projects like residential, commercial, public buildings like hospitals, stadium, retail buildings,
institutional buildings, industrial buildings for pharma, automobiles, etc. The company has achieved the
mile stone of 1000th project in the year 2014.

Siddappa A. Hasbi
Siddappa A. Hasbi is a Metallurgical Engineer, graduated from Karnataka Regional Engg College (KREC),
Surathkal now called NITK, Surathkal. He started his career as a Production Engineer with Indian Metals
and Ferro Alloys Ltd (IMFA) and was associated with this company for two years. Since then, over the past
25 years he has been associated with the product development and the technical marketing and sale of
Silica Fume (Microsilica) to the Concrete, Refractory, Oil Well and the Fiber Cement Industries in India and
the other SAARC Countries. He is presently working as “Managing Director” of an European owned Multi-
National Company “Corniche India (P) Ltd” heading their business activities in India and the other SAARC
Countries plus the Middle-East Countries.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

494 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Features and Effects Gorkha Earthquake

Features and Effects Gorkha Earthquake

Tuk Lal Adhikari
President, Nepal Geotechnical Society and Managing Director, ITECO Nepal

Abstract country. The devastating earthquake was felt across the

region from India to China and left immense destruction
A major earthquake happened on 25th April 2015 with
in 14 districts of Nepal and triggered avalanches in Mount
epicenter at Barpak village of Gorkha district. The
Everest region.
main shock was followed by a series of more than 367
aftershocks during the first 100 days. About 9,000
people died and more than 19,000 people got injuries Geology of Nepal
in this earthquake. This paper deals with the Gorkha Geologically, Himalayan range was formed as a result of
earthquake, its features, aftershocks trend analysis and the collision of the Indian and Eurasian continental plates.
typical damage etc. This collision process still continues and makes the
Key words: Nepal, Seismicity, Earthquake, Aftershock, Himalayan range, seismically one of the most active belts
Reverse Fault. on the earth.
Because of very high relief height ranging from 70 m to
Introduction 8848 m within a distance of 120 km and several rivers
dissecting the topography, Nepal is one of the most rugged
Nepal, situated between India and China in south Asia,
terrains on the earth. Figure 2 shows the geological map
has an area of 147,181 square kilometers and a population
of Nepal showing Trans Himalaya, High Himalaya, Fore
of approximately 27 million. It is located in the southern
Himalaya, Midland, Mahabharat Range, Siwalik Range,
lap of the central Himalayas and bordered to the north
Doon Valleys and Terai belts after Dahal and Hasegawa
by China, and to the south, east and west, by India. Figure
(2008).
1 shows the location map of Nepal showing epicenter of
Gorkha earthquake together with Kathmandu, Pokhara
and Mount Everest. Kathmandu is the capital city of Nepal
and the largest metropolis. Kathmandu metropolis is at
an elevation of approximately 1400 meters in the bowl­
shaped valley in central Nepal surrounded by mountains.
The valley had experienced frequent earthquakes. About
once in a century, the valley gets disastrous earthquake
of higher magnitudes up to ML 8.4. On Saturday, 25th
April a major earthquake of local magnitude 7.6 (Mw 7.8)
struck Nepal and more than 367 aftershocks (of ML >
4.0) have also struck the region subsequently. In 81 years
since 1934, it was the biggest earthquake to strike the
Fig. 2: Geological Map of Nepal (Dahal and Hasegawa, 2008)

Natural disasters are common in Nepal. The country is

geologically young and still evolving. So, landslides and
earthquakes are frequent. Because of its mountainous
topography and the fact that the country comes under the
spell of the monsoon every summer, flash floods, regular
floods and flood­and earthquake­triggered landslides are
also quite common.

Seismicity of Nepal
The earthquake activity in Nepal is caused by the continued
Fig. 1: Map of Nepal (ChannelAsia.com) continental collision between Indian and Eurasian plates

Organised by
India Chapter of American Concrete Institute 495
Session 4 C - Paper 6

Northward under thrusting of Indo­ Australian plate

beneath Eurasian plate generates numerous earthquakes
that consequently make this area one of the most
seismically hazardous on the earth.

Gorkha Earthquake
On 25th April 2015, a magnitude of Mw 7.8 earthquake
occurred with an epicenter 77 km northwest of Kathmandu.
The quake hit at 11:56 am local time.
The salient features of Gorkha earthquake are as follows:
Epicenter: Barpak village, Gorkha district (Latitude 28.240
deg, Longitude 84.750 deg)
Strong shaking time: 40 seconds at Department of Mining
and Geology, Kathmandu
Mechanism: low angle thrust, reverse fault
Fig. 3: Collisions between Indian and Eurasian Plates (South Number of aftershocks: 367 (during first 100 days)
China Morning Post)
Largest four aftershocks: ML 6.6 Gorkha, ML 6.9
Sindhupalchowk, ML 6.8 Dolakha and ML 6.2 Dolakha
(Figure 3). The Himalayan range was uplifted from earlier
Tethys Sea due to the collision. The collision zone is about Magnitude: ML 7.6 (local magnitude), Mw 7.8 (moment
2400 km east­west out of which middle third falls in Nepal. magnitude)
The movement of Indian plate into Eurasian plate is nearly
The earthquake destroyed homes, buildings and temples,
orthogonal in the Himalayas in Nepal. So, the earthquakes
causing widespread damage across the region and killing
from reverse thrust faulting are the most common kind of
more than 9000 and injuring more than 19,000 people. The
earthquakes in Nepal.
earthquake epicenter at Barpak, Gorkha, was the worst to
hit Nepal in over 81 years. The most affected district was
Past Earthquake History
Sindhupalchok which lies directly above the middle part
Nepal is seismically vulnerable as shown by frequent of the rupture zone. Fourteen districts severely­affected
occurrence of earthquakes in the history. The map in by the earthquake are Gorkha, Kathmandu, Bhaktapur,
Figure 5 shows epicenters of earthquakes since 1990 (ML > Lalitpur, Sindhupalchowk, Sindhuli, Ramechhap, Dolakha,
4.0) within the collision zone. Note the belt of earthquakes Nuwakot, Dhading, Rasuwa Solukhumbu, Okhaldhunga
along and south of the Himalayan mountains. Four and Kavre Palanchok districts. Many buildings in
earthquakes with ML > 6.0 have occurred within 250 km Kathmandu valley have collapsed, including historical
of the April 25 Gorkha earthquake over the past century. landmarks such as UNESCO World Heritage temples at
The largest ones included a Mw 6.9 in August 1988 and Basantapur Durbar Square and the historic nine storey
a Mw 8.0 in 1934 which severely damaged Kathmandu. Dharahara tower in Kathmandu by the disaster. Mount
Kathmandu has been struck by frequent earthquakes Everest base camp 1 and Mount Everest base camp 2
of moderate intensities and about once in a century by were severely damaged as a result of avalanches in the
a disastrous earthquake of higher magnitude. Figure 4 Himalayas. In the Langtang valley located in Langtang
shows information of some historical earthquake in and National Park around 250 people were reported missing
around Nepal. after an avalanche hit the village of Ghodatabel and the
village of Langtang.
The aftershock sequence follows a predictable pattern,
although the individual earthquakes are not predictable.
The graph shows in Figure 8 how the number of
aftershocks and the magnitude of aftershocks decay with
increasing time since the main shock. The number of
aftershocks also decreases with distance from the main
shock.
The peak ground accelerations recorded at Kantipath,
Kathmandu by USGS KATNP station is presented in Figure
Fig. 4: Major Earthquakes Recorded in Nepal (Aljajeera, 25th 6. The spectral acceleration shows interesting feature of
April 2015) peaking between 3 to 7 seconds. This low frequency peak

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

496 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Features and Effects Gorkha Earthquake

most damages had occurred in Sindhupalchowk district

which is midway of the epicentral zone. This indicates that
Gorkha earthquake showed strong directivity. The strong
directivity of rupture progression of Gorkha earthquake
has puzzled seismologists because it follows east of
the epicentre even though the seismic gap (since 1505
earthquake) is much longer to west of it which is believed
to accumulate much larger strain energy with likelihood
of creating 8.5+ earthquake. More elaborate discussion
on rupture mechanism is made in twin papers published
recently (6th August 2015) in Science Express and Nature
Geoscience (Galetza et. el. and Avouac et. al.).

Fig. 5: Aftershocks Magnitude of Gorkha Earthquake

in spectral acceleration has surprised seismologists

and earthquake geotechnical engineers. One reason
speculated is due to oscillation of the soft Kathmandu
sediment in its bowl­shaped rock base at a depth of about
500 m. This low frequency peak will have resonating
effect for high rise buildings. Structures of 30­70 stories
would be worst affected as their natural frequencies tally
with ground frequency.

Fig. 7: Zone of Epicenters of Gorkha Earthquake

The hypocenters of the series of Gorkha earthquake are

presented in Figure 8 in the geological cross secton of
Nepal Himalayas. It is noted that the locations of foci of
aftershocks are mostly located in the sub horizontal fault
surface of the Main Himalayan Thrust.

Fig. 6: Peak Ground Acceleration and Corresponding Spectral

Acceleration of Main Shock of Gorkha Earthquake (USGS
KATNP)

Aftershocks Trends
A total of 367 shocks including main shock of local
magnitude greater than 4.0 in Richter scale were recorded Fig. 8: Location of Foci of Gorkha Earthquake in the Cross
and published during first 100 days of the main shock by Section of Nepal Himalayas (Courtesy of GEER Report No. 040,
National Seismological Center, Department of Mines and Version 1.0, 29th July 2015)
Geology, Kathmandu, Nepal. Figure 11 depicts the trend
According to three moment tensor solution reported by
of aftershocks starting from the main shock of ML 7.6 of
USGS, scalar seismic moment of Gorkha earthquake
Gorkha earthquake. The chart shows that the magnitude
is very well constrained within 5.44 to 7.76x1020 Nm
of aftershocks reduces gradually.
corresponding to Mw 7.8 to 7.9. The centroid depth
The epicentres of all the aftershocks of Gorkha earthquake estimates range from 10 to 24 km and the mechanism
were densely distributed within geographical region of shows a low angle NNE dipping fault plane with strike of
84.6°E to 86.5 °E and 27.4°N to 28.4° as depicted in he 290 to 295 degrees, dip of 7 to 11 deg and raking angle of
Figure 7. Though the epicentre of the main shock was 101 to 108 degrees. More details on the features of Gorkha
at the western end of the ruptured zone, all subsequent earthquake are presented in recently published GEER
epicentres were recorded towards east of Barpak and Report No. 040, 29th July 2015.

Organised by
India Chapter of American Concrete Institute 497
Session 4 C - Paper 6

The seismic energy released by Gorkha earthquake was

plotted with dates during the first 100 days of the main
shock as shown in Figure 16. The trend of of seismic
energy release in logarithmic scale (y­axis) is quite similar
to that of magnitude of aftershocks shown in Figure 11.
The total moment of the earthquake was howver in the
range of 3x1020 Joules as reported by USGS.

Fig. 9a: Recorded Aftershocks with Hours of the Day

Fig. 11: Seismic Energy Release in Gorkha Earthquake Sequence

All the shocks of Gorkha earthquake were shallow with

hypocentral depths between 5 km and 30 km as per
published data from the USGS as plotted in Figure 12.
This is understood as one of the major cause for larger
scale damages to ground and structures seriously in 14
districts.
Fig. 9b: Japane Earthquakes by Hours of Day

Another interesting feature of Gorkha earthquake was

that most of the shocks above ML 6.0 had occurred within
two hours after midday as shown in Figure 9a. Similar
trend was observed in Japan earthquakes as shown in
Figure 9b. More researches are to be made to explain the
reason for this behaviour.
Another observation on number of aftershocks during
next 24 hours experienced in Gorkha earthquake is shown
in Figure 15. If a significant aftershock (largest during
the last 24 hours) occurs, the number of aftershocks of
smaller magnitude during the next 24 hours can very well
Fig. 12: Hypocentral Depths of Gorkha Earthquake Sequence
be estimated using this figure with 89% correlation. This
predictive model was found highly useful for the public
information during three months of Gorkha earthquake. Typical Destructions
The earthquake of 25th April of 2015 was highly destructive
due to both the shallow depth (15 km), and the fact that
Kathmandu lies in a basin filled with about 500 m of soft
sediment. The increase in the amplitude of the earthquake
waves within the basin fill increases destructions. In
addition, the sharp density contrast of the soft basin rocks
with surrounding material can cause waves to reflect,
trapping energy in the basin for a period of time. This
extends the duration of shaking during the earthquake.
The ground motion during the earthquake was strong.
A strong motion record from a station KATNP (USGS)
Fig. 10: Number of Aftershocks in Next 24 Hour plotted in Figure 13, the upper plot shows acceleration,

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

498 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Features and Effects Gorkha Earthquake

middle shows velocity, and bottom shows displacement of the Churiyamai in Makwanpur district, the Dolakha
the ground, revealing motion due to elastic seismic waves, Bhimsensthan in Dolakha district, Janaki temple in
but also the net southern shift of 1.5 m. Dhanusha district and the Nuwakot Durbar at Nuwakot
district were partially destroyed.
The damages to houses have occurred in many districts.
In general adobe and brick / stone in mud have collapsed,
while properly constructed RCC buildings have mostly
survived even in Barpak area, the epicenter of Gorkha
earthquake of 25th April 2015.

Fig. 13: Strong motion record of Nepal earthquake (PEER

Berkeley)

Damages in the earthquakes

Across many districts of the country, about 500,000
houses were destroyed and another 300,000 houses Fig. 14: Damages of some historical places (The Hans India)
partially destroyed, especially in the zone above fault
rupture. Several of the temples and monasteries in the
Kathmandu valley were destroyed. Several pagodas on
Kathmandu Durbar Square, a UNESCO World Heritage
Site, collapsed, as did the Dharahara tower, built in 1832.
Many temples, including Kasthamandap, Panchtale
temple, the top levels of the nine­story Basantapur Durbar,
the Dasa Avtar temple and two dewals located behind the
Shiva Pārbati temple were destroyed by the earthquake. Fig. 15: Damages of superstructures (globalnews.ca)
Some other monuments, including the Kumari Temple
and the Taleju Bhawani Temple got partially collapsed.
The top of the Jaya Bageshwari Temple in Gaushala and
some parts of the Pashupatinath Temple, Swyambhunath,
Boudhanath Stupa, Ratna Mandir, Rani Pokhari, and
Durbar High School have been destroyed. In Patan, the
Char Narayan Mandir, the statue of Yog Narendra Malla,
a pati inside Patan Durbar Square, the Taleju Temple,
the Hari Shankar, Uma Maheshwar Temple and the Fig. 16: Damages of Transport system (Author's photos)
Machhindranath Temple in Bungamati were destroyed
by the quake. In Tripureshwar, the Kal Mochan Ghat, a
temple inspired by Mughal architecture, was completely
destroyed and the nearby Tripura Sundari also suffered
significant damage. In Bhaktapur, several monuments,
including the Fasi Deva temple, the Chardham temple and
the Vatsala Durga Temple were fully or partially destroyed.
Outside the valley, the Manakamana Temple in Gorkha,
the Gorkha Durbar, the Palanchok Bhagwati, in Kabhre
Palanchok district, the Rani Mahal in Palpa district, Fig. 17: Damages in Everest (www.telegraph.co.uk)

Organised by
India Chapter of American Concrete Institute 499
Session 4 C - Paper 6

Conclusion 2. “Earthquake” ­Information bulletin Nepal, International Federation

of Red Cross and Red Crescent Societies.
Nepal is situated in a very active earthquake prone area of
Asia and Kathmandu is classified as a highly earthquake 3. “Nepal and Natural Disaster”­United Nations Human Settlement
prone region of the world. The earthquake activity in Nepal Program.
is caused by the ongoing continental collision between 4. Ranjan Kumar Dahal­“Geology of Nepal”­(Lecture Note), Department
Indian and Eurasian plates. On Saturday 25th April, a of Geology, Tribhuvan University, Ghantagar, Kathmundu, Nepal.
magnitude Mw 7.8 earthquake struck from Gorkha district
5. National Seismological Center, Department of Mines and Geology,
of Nepal. Over 367 aftershocks have also struck the region
including a Mw 5.2 in the mountains causing a landslide. National, Kathmundu, Nepal, http://www.seismonepal.gov.np/
More than 9,000 people died and more than 19,000 people Geotechnical Extreme Event Reconnaissance, GEER Association
wounded in this earthquake. The disaster caused many Report No. GEER­040, Version 1.0, 29th July 2015
buildings in Kathmandu valley to collapse. About 500,000 6. Sujan Malla, Dr. Eng., Structural Engineer, Zurich, Switzerland,
houses were destroyed and another 300,000 houses were Version 0, 20th June 2015
partially destroyed in many districts of Nepal, especially
7. Avouac et. al., Nature Geoscience, 6th August 2015, "Lower edge
those near the epicentral zone. Several of the temples and
of locked Main Himalayan Thrust unzipped by the 2015 Gorkha
pagodas in the Kathmandu valley were destroyed.
earthquake"

References 8. Galetza et. el., Science Express, 6th August 2015, "Slip pulse and
1. “M 7.8 Nepal Earthquake of 25 April 201 5”­U.S. Department of the resonance of Kathmandu basin during the 2015 Mw 7.8 Gorkha
Interior U.S. Geological Survey. earthquake, Nepal imaged with geodesy"s

Tuk Lal Adhikari

President, Nepal Geotechnical Society and Managing Director, ITECO Nepal

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

500 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Selecting Concrete Smartly for Making Sustainable Buildings in Smart Cities

Selecting Concrete Smartly for Making Sustainable Buildings

in Smart Cities

Prof. Vinod Vanvari Dr. Sumedh Mhaske Neelam S.Varpe Zeeral Jadav
Head, Civil Engineering Associate Professor, and Former Faculty, Civil Engineering Graduate Engineer,
Department, Shri Bhagubhai Head of Department, Civil Department, Shri Bhagubhai Mumbai
Mafatlal Polytechnic, Vile Parle and Environment Engineering Mafatlal Polytechnic, Vile Parle
Mumbai – 400056. Department, V.J.T.I. Matunga (East) (West) Mumbai – 400056
Mumbai – 400019

Abstract Massachusetts Institute of Technology). Discussions as

smart cities are gaining momentum worldwide. However
“Sustainability” is an issue of global importance and
despite its wide spread awareness, it is difficult to describe
“Smart Cities” is a order of the day. Smart cities must
the nuances behind labeling of the so-called “smart city”.
have sustainable buildings and this can happen only
when materials like concretes are used for construction Simply put, it’s a city that leverages technology to its
smartly. optimum to improve the lives of its citizens. In last ten
years, our cities have been blanketed with, if not consumed
Building has two types of members, one structural
by digital technologies, thus facilitating the creation of an
elements like foundation, frame work of slabs, beams
over-reaching, a smart infrastructure.
and columns, other nonstructural elements like partition
walls, pavements, etc. Lot of importance is laid on It’s vital to understand that all buildings must be sustainable
selecting material for structural elements but selecting for a city to be smart. Advances in technologies related
concrete and other materials for nonstructural elements to vital material like concrete must be adopted to make
takes a back seat due to lack of awareness about the building sustainable ones. Before proceeding further, let
recent advances made in material technology. This paper us attempt to understand sustainability with respect to
attempts to evolving a system for facilitating selection of concrete.
concrete for both structural and nonstructural elements,
which will not only lead to cost saving but takes us one
step ahead toward sustainability. This is based on field
Understanding Sustainability with Respect to
work on project sites in the city of Mumbai. Concrete
“As its essence, sustainability means ensuring prosperity
Information about good work at projects carried out is
and environmental protection without compromising the
gathered in structured questionnaire. Data is analyzed to
ability of future generations to meet their needs.” (Ban Ki-
take stock of present scenario w.r.t selection and usage of
moon, 2013)
concrete. Further illustration is made by three small case
studies which very clearly depict the advantages gained by Sustainable Housing would include Environment, Energy,
selecting and using concrete smartly. Water, Materials, Health, Wellbeing and Affordability.
The conclusion would be suggestive about which type of
Sustainability and Concrete
concrete to use for which component of building for better
gains. We are at a stage, where everything on the earth
seems to be growing at an alarming rate. In the Indian
Key Words: Sustainability, Concrete, Smart Cities, Urban scenario, where development is taking place at a rapid
Renewal. pace with increased construction activity, infrastructure
establishment and economic growth. Rapid development
Introduction increases the demand for natural resources and services.
The world is passing through difficult and troubled times, Concrete which is one of the basic raw materials used in
and we live in a rapidly changing world. construction is adding up to the impact on the atmosphere.
A single industry accounts for around 5% of global
The cities of the developing world are growing at a fast carbon dioxide (CO2) emissions. It produces a material
pace, with the potential to add to it, the huge influx of so ubiquitous it is nearly invisible: cement. Cement is the
people, amidst the physical, social, political and cultural primary ingredient in concrete, which in turn forms the
pressure. The challenge for most developing world cities foundations and structures of the buildings we live and
is formidable. work in, and the roads and bridges we drive on. Concrete
“Sustainability is a prerequisite for a city to be smart” is the second most consumed substance on Earth after
(Carlo Ratti, Leading Architect and professor at the water. On average, each year, three tons of concrete are

Organised by
India Chapter of American Concrete Institute 501
Session 4 C - Paper 7

consumed by every person on the planet. Concrete is

used globally to build buildings, bridges, roads, runways,
sidewalks, and dams. Cement is indispensable for
construction activity, so it is tightly linked to the global
economy. Cement production is growing by 2.5% annually,
and is expected to rise from 2.55 billion tons in 2006 to
3.7- 4.4 billion tons by 2050.

Resaerch Significance
This research work on selecting concrete for various
components of buildings in smart cities has immense
significance in the following aspects.
Fig. 1: Depicts cross section of S.W.D.
ll Accomplish sustainability by facilitating selection of
right type of concrete for a particular component to to be designed, and constructed along 237m long storm
meet specific functional requirements. water pavement edge drain. It was plumbing contractor
suggesting conventional grating in mild steel costing about
ll Accomplish affordability by selecting appropriate type
Rs.1092 per running meter. However due to presence of
of concrete to meet design parameters desired in that
mid of smart and alert site engineer, hydro media product
particular component of meeting.
of one renowned RMC company was projected before
ll Besides accomplishing sustainability and affordability, PMC, Architect and Client costing Rs,491 per running
other advantages gained are speed of construction, meter including provision, installation and curing. That
space saving, better aesthetics, utilization of waste etc. company was already doing a job of stemcrete inside
basement of the same project.
Field Investigation
With the support and assistance of technical cell RMC
Construction activity pertaining to buildings in city of Company 300x300x100 panels of pervious concrete in the
Mumbai has a particular approach and pattern due to desired colour with the required strength were developed
many not technical persons in the trade. On the other and installed above storm water drain all along the three
hand, some building projects are coming up with high sides of the plot. (Figure No. 1 depicts cross section of
technical and international standards, as project delivery S.W.D. at this project where panels of Hydromedia used
is properly constituted. Buildings are under construction, as lids for edge drains.)
repair and reconstruction approach varying from
conventional to highly non-conventional and modern. To ll Space saving to the user/client, as space over SWD can
assess, consolidate and make awareness about approach be merged with car parking space due to its designed
in selecting concrete various components of buildings, compressive strength.
questionnaire was developed and put to field investigation ll Environmental friendly, as it taps rain water for Rain
through a team of research assistants. Outcome obtained Water Harvesting.
is both encouraging and discouraging as well. However,
in order that people are made aware, this outcome is ll More appealing aesthetically.
compiled and tabulated. It would be appropriate to discuss Small cost saving of Rs.1, 50,021/- (2420 USD)
selecting of concrete to accomplish sustainability with few
case studies before we come to outcome. Small cost saving advantage is mentioned here in particular
for those who have wrong notion that unconventional and
latest practices always cost more than conventional ones.
Case Studies
As an illustration to advantages gained by smartly Case Study No.: 2
selecting concrete for a particular component of building, One educational building Gr. + 2 in suburbs of Mumbai
three case studies based on actual construction activities was undergoing vertical expansion. This was being
at the projects in Mumbai, are cited below. One is an use carried out by having cast in situ piles 800 dia on external
of pervious concrete, second one is on using concrete as periphery and 300mm dia. micropiles in inner areas.
composite material with steel, third one is on using fibre- Both old and new parts though appearing together as
reinforced concrete for top-most slab of building i.e. roof. one part of building, were independent structures w.r.t.
structural load sharing. New structure was planned for
Case study No.1
3rd to 6th stories. However after converting Gr + 2 to Gr+4,
One building basement plus podium plus 12 stories was management realized the need of having Gr+8 structure
on verge of ready for occupation. It is during when there instead of Gr+6. Design was reviewed. Original extension
was a situation when grating 300mm wide was required above Gr+4. i.e. upper four floors, steel beams were

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

502 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Selecting Concrete Smartly for Making Sustainable Buildings in Smart Cities

adopted to utilize dead load of concrete into live load of To arrive at suggestive recommendations the following
additional two more floors. Slab thickness was optimized tests were carried out in the laboratory:
by having post tensioning technique. Thus by adopting
(i) Pervious concrete which is just sufficient in
appropriate technology of composite construction and
compressive strength, was tested for these properties.
post-tensioning, much needed purpose of user was
Concrete was designed, prepared and cubes were
served.
casted. Same were tested for compressive strength.
Case Study No.: 3 The results are encouraging. ( Refer Table No 2)
One Hi-end residential building was under construction in Panels of pervious concrete were casted and tested for
suburbs of city of Mumbai. For landscaping at the terrace flexural strength. Here also results are encouraging.
it was not feasible to go for conventional brickbat coba and (Refer Table No. 3)
usual support of chemical based waterproofing. While
(ii) Panels of M20 grade fiber-reinforced concrete were
slab were casted, concrete was mixed with appropriate
casted. After necessary curing the same were test
polypropylene fibers at the rate of 1kg/cum of concrete.
for flexural strength at 28 days. The results are much
Thus topmost slabs (terrace, OHWT and LMR top slabs)
more than satisfactory.
were constructed with concrete which does not shrink at
later date and give cracks to become permeable. Thus
Table 2
small but smart step of advance in technology made Compressive strength of M15 Grade Pervious concrete
possible for engineering team at the project site to meet
the demand of architectural team. Cube Weight 7 day Strength 28 day Expected Strength
Strength at 28 days
Methodology and Testing
1 7.54 13.77 N/Sq 17.55 N/Sq 15 N /Sq mm
From the data gathered through field investigation it is Kg mm. mm
learned that good practices and advances in concrete
technology are being adopted to sufficient extent but in 2 7.65 13.77 N/Sq 17.55 N/sq 15 N/Sq mm
an scattered manner. Case studies reinforced the belief Kg mm mm
that advances in technology in concrete can be best 3 7.65 13.77 N/Sq 17.55 N/sq 15 N/sq mm
adopted accomplish sustainability. To ascertain further Kg mm mm
the capabilities of specific types of concrete , testing of
pervious concrete panels and fiber-reinforced concrete
panels was carried out in the laboratory. Following mix Table 3
Flexural Strength of M15 Grade pervious concrete
proportion was adopted for pervious concrete.
Panel No. 28 day Actual Expected Strength as per
Table 1 750x300x125 strength relationship*
Mix proportion for Pervious Concrete
1 1.4 N/Sq mm 0.5027 N/Sq mm
Void content 20%
2 1.4 N/Sq 0.5027 N/Sq mm
W/C.M 0.32 mm
Dry-rodded density 1750 3 1.4 N/Sq mm 0.5027 N/Sq mm
Sp. Gravity 2.71 * Relationship established by Ahmed and Shah (1985).

Absorption 1.20%
Results and Discussion
Bulk Density 1450
ll The analysis of data received through field investigation
b/bo (Bulk Density / Dry rodded Density) 0.83 reveals that by and large field engineers and contractors
are adopting advances in technology or smart material
Wa (Weight of aggregates) 1450 kg/m2
baring manufactured sand which is yet to find place in
Wssd (Wt. of saturated sand) 1467 kg/m2 construction industry in this part of the world. These
are being used more as general practice rather than
Vol. of paste 0.08
need based for achieving functional or durability criteria
Paste of cement 8 to 10% or meeting sustainability requirements. Education and
training through such seminars is need of hour.
Cement 126 kg
ll Three case studies cited, clearly depict that advances
Water 40.32 lit in technology of concrete has huge potential to meet
specific functional requirement of user and that to
Aggregate of vol. 541.32 kg
keeping in new sustainability.

Organised by
India Chapter of American Concrete Institute 503
Session 4 C - Paper 7

Sustainability By Selecting Right Concrete considerations often referred to as “the three pillars” of
sustainability i.e. social, environmental and Economic.
Concrete is an essential material with a worldwide
estimated consumption of between 21 and 31 billion As knowledge and materials technology develop, it
tonnes of concrete in 20061, concrete is the second most is possible to increase the compressive strength of
consumed substance on Earth after water. A world without concrete. The average tensile strength in low strength
concrete is almost inconceivable! Concrete is made grades is about 10% of the compressive strength, and
from coarse aggregates (gravel or crushed stone), fine in the higher strength grades about 6%. By using high-
aggregates (sand), water, cement and admixtures. These strength concrete (over 60MPa), the dimensions of the
constituents are mostly available locally and in virtually structure can be reduced. As part of the “high-strength
unlimited quantities. concrete development project”, it was estimated that
doubling the strength of the columns reduces the cost/
Primary materials can be replaced by aggregates made
load-bearing ratio by about 25%. A significant proportion
from recycled concrete. Waste materials from other
of this comes from reducing the use of materials, so
industries can be used to produce additions like fly
that as far as environmental impact is concerned it is
ash, slag and silica fumes. Concrete is one of the more
advantageous to use high-strength concrete. Moreover,
sustainable building materials when both the energy
it has the advantage of improving the service life of the
consumed during its manufacture and its inherent
structure.
properties in-use are taken into account. Sustainable
development is commonly defined as “the development
that meets the needs of the present without compromising Outcome
the ability of future generations to meet their own needs”. The outcome of this research work carried out is the
It incorporates the environmental, economic and social following suggestive recommendations:

Sr
Type of Concrete Properties Component of Building Advantages
No.

1. Pervious Concrete Permeability parking areas, areas with light allow air or water to move through the
Compressive strength traffic, concrete
Flexural strength greenhouses

2. Light Weight Concrete / Reduced dead weight. Partition Walls (Concrete Blocks) (i) Reduces dead load
Aerated Concrete (ii) Lowers hauling and handling cost
(iii) Faster progress of construction.

3. Vacuum Concrete • Decrease in Slabs (i)quicker strength gain

permeability of concrete (ii)removal of excess water required for
• High density concrete workability.

4. Self Compacting Concrete Flow under own weight It is used in location unreachable (i) no need for vibrators to compact the
No need of vibration for vibrations. (junction of beam concrete.
and column) (ii) placement being easier.
(iii) no bleed water, or aggregate segregation.

5. High Strength Concrete Load carrying capability Basement Flooring, Factory, (i) Compressive strength of high strength
per unit high Warehouse Flooring concrete mix is usually greater than 6,000
pounds per square inch.

6. High Performance High durability Precast Constructions (i)High strength.

Concrete High strength (ii)High workability.
(iii)High durability.

7. Air Entrained Concrete Barrier to extreme Structural members in extreme It lowers the surface tension of water and thus
climatic conditions hot and cold weathers bubbles are created.

8. Ferro Cement Concrete Ductile Water tank Economical and faster construction.
Impact resistive
Stiffness

9. Fibre Reinforced Concrete Intrinsic Stresses are Runways, surfaces subjected to Durable & Functional
catered. wear & tear, top slabs and ramps

10. Self Consolidating Extreme fluid Mass concrete Detrimental to safety and workmanship
Concrete No need of vibration

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

504 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Selecting Concrete Smartly for Making Sustainable Buildings in Smart Cities

Conclusion 6. D.L Desai , “Smart cities”, Indian construction of BAI, June 2015.
Page nos 9-13.
Whatever activity human being undertakes for 7. Efficient buildings with Lafarge Aggregate & Concrete India Pvt.
development, for survival of mankind, it is imperative Ltd.
to take into account sustainability. New cities are being 8. Efficient buildings with Lafarge.com/…./1659_Lafarge_Hydromedia_
developed as smart cities while old ones are undergoing EBS_VB.
urban renewal so as to be smart and sustainable. 9. Guide for design and Construction concrete Parking Lots, ACI 330
Electronic and computers related applications are being R.
adopted for better communication matching needs of 10. Kartikeyan Nachiappan, “Infotopia – A study on smarter sustainable
smart cities. cities”, The Indian Concrete Journal, July 2015.

Smart city is a concept of the day, while concrete is a very 11. Marceau, J. (2008). Introduction: Innovation in the city and innovative
cities. Innovation: Management, Policy& Practice, 136-145.
smart material. Let us make this an opportunity and use
12. Pervious Concrete – American Concrete Institute ACI-May 2006.
this smart material smartly to accomplish sustainability.
13. Pwan Pandey and J.umaMaheshwari , “Work Study For Sustainable
Construction: A case Study”, Journal of Construction management,
References NICMAR, Pune, Page nos 19-29.
1. Ahmad.s.H and Shah.S.P,1985”Structural properties of High
Strenth Concrete and its implications for Precast Presstressed 14. R.Narayan Swamy, University of Sheffield, England. “Sustainable
Concrete,”PCI Journal,V.30 No 6, Nov /Dec. Concrete for the 21st Century, Concept of strength through
Durability”
2. Al-Hader, M., & Rodzi, A. (2009). The smart city infrastructure
development &monitoring. Theoretical and Empirical Researches 15. Report on Pervious Concrete – American Concrete Institute ACI
in Urban Management, 87-94. 522R-10-March 2010.

3. Anna Gorge Nellicks and Shivkumarpalaniappan, “Built environment 16. Rohini Nilekani, “Invisible water, Visible crises” , INDIA TODAY,
Sustainability Review of Key Concepts, Journal of Construction Independence Day Special, Aug 24,2015.
Management, NICMAR, Pune, Jan-march 2015. Page nos 1-17. 17. Specification for Pervious Concrete Pavement ACI, 522.1.
4. Ashley, E, 2008, “Using Pervious concrete to achieve LEED”. 18. Vinod Vanvari and Dr.Sumedh Mhaske, “Use of Pervious Concrete
5. Carlo Ratti, “Sustainability is a pre-requisite for a city to be smart”, in storm water Drain Construction in Redevelopment Building
Times Property, Sept 05,2015. Projects”, R.N. Raikar Memorial International Conference and
Dr.Suru Shah symposium on “Advances in Science & Technology
of Concrete”, Mumbai, Dec 20-21, 2013.

Vinod Vanvari
Vinod Vanvari, Professor and Consultant, holds a Bachelor of Civil Engineering from University of Poona,
Master of Construction Management from Mumbai University. He is research scholar at VJTI and Head,
Civil Engineering Dept. at Shri Bhagubhai Mafatlal Polytechnic. His expertise includes optimum use
of systems and materials to expedite and economize construction projects. His areas of operation are
teaching, training, testing, consultancy and research.

Dr. Sumedh Mhaske

Dr. Sumedh Mhaske, holds a Bachelor of Civil Engineering from University of Mumbai, ME with specialization
in Construction Management and a Doctorate from IIT Bombay. His research
• Spatial Technology: Geographic Information System (GIS) and its application in Civil and Environmental
Engineering such as for site layouts, construction waste management, Disaster management etc.
• Space Technology: Global Positioning System (GPS) and its application in Civil Engineering such as land
mapping, Contour mapping etc.
• Precise levelling for machine foundations up to micron level accuracy.
• Advanced Geotechnical Earthquake Engineering: Soil Liquefaction, Reclaimed land development,
Liquefaction susceptibility mapping with GIS and GPS etc.
• Integrated approach of GIS, GPS and GPR ( Ground Penetration Radar) for underground Utilities mapping
• Integrated approach of GIS and GPS for study of soil and rock properties in Geotechnical Engineering.

Organised by
India Chapter of American Concrete Institute 505
Session 4 C - Paper 7

Ms. Neelam Varpe

Ms. Neelam Varpe, holds the Bachelor degree from Mumbai University, Masters in Structural
Engineering from Mumbai University. Lecturer in Shri Bhagubhai Mafatlal Polytechnic, Mumbai.

Dr. Sumedh Mhaske

Ms. Zeeral Jadav, holds a Diploma in Civil Engineering from University of Mumbai and a Bachelor of Civil
Engineering from University of Mumbai

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

506 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
SESSION 5 A
Session 5 A - Paper 1

Outcomes of a Major Research on Full Scale Testing of RC Frames in

Post Earthquake Fire

Asif H. Shah, Umesh K. Sharma, G. R. Reddy, Tarvinder Singh

Pradeep Bhargava Reactor Safety Division, Bhabha
Department of Civil Engineering, Indian Institute Atomic Research Centre, Mumbai,
of Technology Roorkee, Roorkee India India

Abstract structures exposed to elevated temperatures with an

aggressive fire or heat source. Particularly important is
Post earthquake fire can result in an expeditious collapse
the behaviour of reinforced concrete structures during
of buildings that have been damaged as a result of a prior
fires following earthquakes (FFE). In the recent past, FFE
earthquake. This paper represents the results of a major
scenarios have been increasingly recognized as a credible
research carried on four full scale RC frames. A novel test
extreme loading case, though not much work, analytical
setup for fire following earthquake has been developed and
or experimental has been reported on the behaviour of
tested in the current exercise. A number of sensors were
structures exposed to fire following earthquake events.
used to monitor the kinematic and thermal fields during
Famous for its ubiquitous influence, concrete is most
the test. The results (in terms of strains, displacements
widely used material in the construction arena thereby
and temperatures) provide valuable information about
making the research on fire resistance more and more
earthquake- damaged RC structures in fire in fully
important.
developed and decay stages. The study investigated
the effect of the pre-damage level, the reinforcement Concrete is a non-homogeneous two phase material at
detailing and the infill on the performance of RC frames the macroscopic level, consisting of hardened cement
subjected to post earthquake fire. The results show paste and aggregate. The heterogeneous nature of porous
significant variation in structural temperatures within medium of concrete is very sensitive to unwanted fire or
a compartment, demonstrating that the assumption of high temperatures conditions. Due to the composition of
uniform temperature, which is implicit in many design concrete and the acute thermal conditions that exist in
fires, is incorrect. The RC frames designed with ductile a fire, concrete and fire have a multifarious interaction.
detailing of reinforcement showed better lateral load Each of the constituents of concrete reacts differently to
carrying capacities before and after fire. The pre-damage thermal exposures and the behavior in fire of the composite
level affects the rise in temperatures in the structural system is not easy to describe or model (Khoury 2000).
elements and also affects the spalling of the members. Lea (1920), and Lea and Strandling (1922) pioneered
This paper describes the experimental investigation and investigations into the influence of high temperature on
serves as a vehicle for dissemination of the key findings and the behaviour of concrete in RC structures. In the early
all the important test data to the engineering community stages, researchers gave consideration to the physical
which could be used for validating numerical simulations and chemical changes within the concrete composition
for further advancing the knowledge and understanding in due to exposure to elevated temperatures (Malhotra
this relatively poorly researched area. 1956, Saemann and Washa 1957, Zoldners 1960, Purkiss
and Dougill 1973). They have shown that when concrete
Keywords: Post earthquake fire, Reinforced concrete
is exposed to elevated temperatures, its chemical
frames, Structural performance, Full scale testing,
composition and physical properties change significantly
Spalling.
which occur primarily in the hardened cement paste and
simultaneously it may also occur in aggregates.
Introduction
A number of chemical and physical changes occurring in
Concerns related to material integrity of concrete exposed concrete due to exposure to elevated temperature have
to elevated temperatures have always been there, in spite been discussed by Khoury (Khoury 2000), Bazant and
of concrete being an incombustible material. The thermal Kaplan (Bazant and Kaplan 1996), Schneider and Herbst
properties and the behaviour of concrete exposed to (Schneider and Herbst 2002) and Carvel (Carvel 2005).
elevated temperatures have received considerable They have noticed that depending on the temperature
experimental and analytical attention in the past. However, attained, some of these changes get reversed upon
the behavior of concrete subjected to fire is understood to cooling, while others are non-reversible and may
a lesser extent as compared to that of steel reinforcement deteriorate concrete significantly after a fire.
bars. It is important to scrutinize the behavior of concrete

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

508 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Outcomes of a Major Research on Full Scale Testing of RC Frames in Post Earthquake Fire

Fire Following Earthquakes However the behaviour of structural systems in fire is

Fire following earthquakes, though rare to occur, have proved to be quite complex due to interaction effects
resulted in damage more pronounced than caused by the between different structural elements of the structural
original shaking. Earthquakes more often are regarded as system. This simplification in design procedure was a
low probability but high consequence events and are difficult result of standard fire tests of simple building elements or
to plan for. Fire conflagrations can develop from situations isolated structural assemblies in testing furnaces which
other than earthquake, although both local and international subject the loaded elements to a standard temperature-
experience confirms that earthquake is a major initiator of time curve. The issues of the inherent problems, which
large-scale urban fires. The threat posed by fire following are associated with using simplified single element based
earthquake has been highlighted by a number overseas laboratory tests subjected to standard temperature- time
earthquakes, notably San Francisco (1906), Northridge, Los curves, are being solved by performing large scale non
Angeles USA (1994), and Kobe, Japan (1995). Devastating standard fire tests using real fires. This shift in testing
fires which broke out as the aftermath of an earthquake philosophy from prescriptive standard fire testing to large
and lasted for several days completely destroyed 80% of scale non standard fire testing using real fires has been
San Francisco, leaving more than 3000 people dead and aimed at understanding the global behaviour of structures
the property damage was estimated at $500,000,000 in in fire which has received relatively little attention in the
1906 dollars (Hansen, 2015). Most of the major earthquakes past. The significance of performing the large scale tests
which have struck Californian belt have been followed by is also reflected by the lack of experimental data on the
varying number of ignitions, the famous being the 1971 San performance of complete concrete structures in fire.
Fernando and 1994 Northridge earthquakes, which were The objective of this paper is to report on the preparation
followed by more than 100 ignitions. Recent earthquakes and methodology employed and to discuss the outcomes
including 1989 Loma Prieta earthquake in California, of a major research programme that investigated
1995 Hanshin (Kobe) earthquake in Japan, 1999 Marmara the performance of full scale RC frames under post
earthquake in Turkey were all followed by major incidents earthquake fire. This paper discusses the results of four
of fire, the intensity of which has been reported to have tests carried on full scale RC frames in Civil Engineering
varied from few hours to three days (Chen and Scawthorn, department of IIT Roorkee, India.
2003). Scenario studies of future large-scale earthquakes
in San Francisco and Tokyo indicate that Fire Following The Experimental Programme
Earthquake will be a major factor in subsequent property The research investigated the effect of structural and
damage and lives lost (Wellington Lifeline Group, 2002). loading parameters on the performance of RC frames
in post-earthquake fire. Table 1 gives the overview of the
Full Scale Fire Testing tests carried out.
Structural fire testing is undergoing rejuvenation with
The parameters involved include:
full scale tests being performed on various structural
systems. The conventional, and widely used, method (a) Simulated seismic loading: Two level types of
for fire testing, where in single structural elements cyclic lateral loads were applied on the test frames
are subjected to a standard fire test (ISO, 1975) and the corresponding to 2% and 4% of the storey drift as per
thereby obtaining a fire resistance rating which is mainly FEMA 356 code (FEMA, 2000).
in the form of a time to failure, has of late been noticed
(b) Reinforcement detailing: Two levels of reinforcement
to have a number of drawbacks (Drysdale, 1999) though
were used in the frames. While as two frames were
the method is still widely used. The perception that the
detailed as per the IS 456:2000 (BIS, 2000), the other
concrete structures behave well in fire has led to overly
two were designed as per the clauses of IS 13920:1993
simplified design guidelines in the form of concrete cover.
(BIS 1993).

Table 1
Test Matrix of the Experimental programme

Frame Simulated seismic damage Fire loading Aftermath

 life safety structural performance level of Residual lateral capacity
 A. (Kumar, 2012) 900oC -1000oC for 1 hr
FEMA 356:2000 test
Collapse Prevention structural performance level Residual lateral capacity
B. (Kamath, 2014) 900oC -1000oC for 1 hr
of FEMA 356:2000 test
 Collapse Prevention structural performance Residual lateral capacity
C. 900oC -1000oC for 1 hr
level of FEMA 356:2000 without Ductile Detailing test
 Collapse Prevention structural performance Residual lateral capacity
D. 900oC -1000oC for 1 hr
level of FEMA 356:2000 with infill wall test

Organised by
India Chapter of American Concrete Institute 509
Session 5 A - Paper 1

(c) Infill: Three frames were tested as bare frames without the plinth beams. The columns were constructed in two
any infill material while as one frame was masonry stages of 1500 mm each. The construction culminated
infilled. with the casting of the roof beams, slab and the top
400 mm extensions of the columns. Extensive efforts
In order to evaluate the behaviour of Reinforced Concrete
were done to avoid formation of cold joints with quality
(RC) frames in post earthquake fire, full scale RC frame
control during the construction stages. During each
assemblages were constructed before subjecting them to
stage of construction sufficient number of cubes as per
pre-determined earthquake damage and then exposing
the guidelines provided in IS 456 2000 (BIS 2000) were
it to a compartment fire of one hour duration. After fire
collected to monitor the strength of concrete.
the frames were again subjected to cyclic lateral loads to
determine their residual lateral load capacities. The four
frames were tested over a span of four years.

Construction Details of RC Frames

The structure subjected to the tests was a symmetric RC
frame assembly. Figure 1 shows the test frames after
construction. The test frame consisted of four columns
(300 mm × 300 mm), four plinth beams (230 mm × 230
mm), four roof beams (230 mm × 230 mm) and a roof Fig. 1: RC test frames before testing. (Left: without infill (A, B,
slab (120 mm thick). All the elements of the test frame C), Right: with infill (D)
were cast monolithically and the column fixity at the base
was provided by the termination of all the four columns Instrumentation
into a 900 mm thick RC raft foundation of plan size 6900 Figures 2 and 3 give an insight into the instrumentation
x 8700 mm. The RC frames “A” and “B” were designed developed for the testing. The main aim of testing the
as per the guidelines of Indian standard code of practise, RC frame sub assemblages was to derive an in depth
IS 456:2000 (BIS, 2000) and IS 13920: 1993 (BIS, 1993), understand of the behaviour of structures in post
while as the frames “C” and “D” were designed as per IS earthquake fire. Data which was planned to be recorded
456:2000 (BIS, 2000) only without providing the ductile during all the phases of the tests for the structural
detailing. However the minimum shear reinforcement aspects consisted of temperatures within the roof slab,
was provided as per the recommendations as prescribed columns, beams, and the compartment ; strains on the
by IS 456:2000 (BIS, 2000). This was done so as to test rebars embedded in all the structural elements of the RC
the frames “A” and “B” as ductile and frames “C” and frames; deflections of the roof slab at different points both
“D” as non ductile structures and hence study the effect vertical as well as horizontal. Extensive instrumentation
of the reinforcement detailing on the behavior of the RC was planned for the test frame and a number of sensors
frames in post-earthquake fire. Frames “A”, “B” and “C” were installed to record the data.
were designed and tested as bare frames while as the
The temperature measurement was taken by means
frame “D” was tested as infilled frame with 230 mm
of thermocouples embedded in concrete at a number
masonry units used as the infill material. Equivalent
of points and at different levels. The temperature build
gravity load, the wall load and the live load on the roof
up inside the concrete of various structural elements
slab were imposed as per the recommendations of the
of the RC frame was recorded using 0.5 mm dia K-type
Indian seismic design code, IS 1893 (Part-1):2002 (BIS,
thermocouples embedded along the depth and width
2002). An M30 concrete was designed, and utilized for the
of the elements. Thermocouples were installed at five
construction of the RC frames. The frame was designed
different sections in the various members i.e. near the
using the reinforcement bars of grade Fe500 having the
end supports, at the midspan and at the section between
yield strength of 500 MPa. The RC frame sub assemblage
these two. Each section in case of columns and top beams
was constructed on a strong floor in an outdoor testing
consisted of five thermocouples with three thermocouples
facility. The construction of the test frame was carried
along the depth of the cross section and two along the
out in a number of stages starting from the footings. The
width of the cross section. In case of plinth beams the
test frame was based on 4 footings 500 mm in depth
thermocouples were embedded only along the depth and
and each footing was constructed in a manner that 8
in case of slab the three thermocouples were placed along
No’s 1200 mm long protruding bolts extended in raft
the depth at nine sections.
foundation passed through it. This was done to avoid any
uplift of the footings later during the seismic testing by To record the buildup of the gas temperature inside the
holding the footings down with the use of girders. The frame compartment thermocouple trees were placed at
construction of footings was followed by casting of 800 five different plan locations, each at the four corners (near
mm height of columns, the top of which formed the base four columns, TMIC1, TMIC2, TMIC3, TMIC4) and one near
of plinth beams. Next stage of the construction involved the centre (TMIC). Each thermocouple tree consisted of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

510 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Outcomes of a Major Research on Full Scale Testing of RC Frames in Post Earthquake Fire

five 6 mm diameter metal insulated (MI) thermocouples

positioned at different heights, of 0.20 m, 0.90 m, 1.60 m,
2.30 m and 2.90 m from the floor level of the compartment,
to obtain the thermal profiles inside the compartment. A
total of 312 thermocouples, which included 287 K- type
and 25 MI type thermocouples, were utilized for the
measurement of the temperatures. Figure 3 shows the
MI thermocouple positions for compartment temperature
measurements.

Fig. 3: Position of thermocouples in compartment

The stains were recorded by pasting strain gauges on

the steel rebars at different locations in the RC frames.
Electrical resistance-type surface mounted strain gauges
were mounted on rebars at three sections in plinth beams,
columns and top beams. Each location had two strain
gauges, one on the exposed rebar and the other on the
opposite unexposed rebar. Similarly steel rebars at four
key locations in plan of the roof slab were instrumented
with strain gauges.
Extensive instrumentation using linearly varying
displacement transducers (LVDT’s) of different ranges
and stroke lengths was done in order to measure
displacements of the different structural elements during
all the test phases. During the first phase of test i.e. the
simulated seismic loading, seven LVDT’s were mounted
to measure the horizontal displacements at roof level
in different directions while as four LVDT’s were used
to measure horizontal displacements along different
directions at plinth level. In order to check if the footings
of the RC test frame undergoes any uplift during the
seismic loading, each footing was instrumented with
LVDT’s to measure any vertical displacement. An array
of nine displacement transducers was used to record the
deflections of roof slab during the fire loading.

Experimental Setup and Test Procedure

The gravity loads applied on the frames were based on
the Indian standards for general loading, IS: 875 (Part
1 & part 2): 1987 and the Indian seismic design code, IS
1893 (Part-1):2002. The gravity load and the imposed
load applied were 1kN/m2 and 2kN/m2 respectively. The
design loads applied for earthquake forces was however
0.5kN/m2 (25% of 2kN/m2) as per the guidelines of IS 1893
(Part-1):2002. In order to simulate gravity loads of the
Fig. 2: Position of thermocouples in (a) Columns (b) beams and upper floors of the G+3 structure on the columns of the
(c) slab test frame a self equilibrating loading arrangement was

Organised by
India Chapter of American Concrete Institute 511
Session 5 A - Paper 1

developed as shown in the Fig 4. The vertical loading from cycles starting with 10 mm cycle and then each successive
the top stories was simulated by applying the calculated cycle was increased with an increment of 10 mm till the
equivalent load using 4 hydraulic jacks positioned end of the lateral load test.
centrally on each column. These jacks were made to rest
After damaging the frames by inducing a predefined level
on a specially designed ball bearing assembly in order
of damage, the frames were subjected to a one hour
to keep the verticality of the jacks maintained during the
designed compartment fire to assess the global structural
simulated seismic load so that the actual simulation of
behavior in a simulated fire following earthquake event. A
the applied loads is achieved. The live load on the slab of
series of mock tests were conducted, using kerosene as
the test frame was calculated and simulated by putting
the fire load, to arrive at the fire design with an objective
the equivalent load using the sand bags placed uniformly
to maintain the temperatures within the compartment
throughout the area of the slab. Similarly the wall load on
at an average of 900-1000 °C for a period of one hour.
the top beams coming from the walls of the immediate top
The fire was designed as per the Thomas and Heseldon
floor was calculated and simulated by putting the sand
(Thomas and Heseldon, 1972) and modified to allow rapid
bags along the four top beams. The sand bags were put
accumulation of smoke in order to enhance the flashover.
in four layers and their stability during the load tests was
With these modifications a compartment size of 3m ×
ensured by putting a steel mesh all around the sand bags.
3m × 3m was arrived at with an opening of 1 m height
Figure 4 shows the actual test arrangement before the
by 3 m length. The opening was kept low to enhance the
test.
accumulation of smoke for the rapid flashover. The peak
burning rate of kerosene oil was kept approximately 0.117
kg/m2/s for maintaining a post flashover temperature of
900°C to 1000°C. A peak fuel flow rate of 1.43 x 10-4 m3/s
which corresponds to 9 litres/minute was maintained
using a fixed head. Hence, a reasonably simple technique
was adopted to attain gas temperature of about 1000°C
within 5 minutes after ignition to simulate a realistic
compartment fire. Before exposing the frame to fire, the
envisaged fire design was given the practical shape by
compartmentalization of the RC test frame assemblage.
Fig. 4: Test Setup: (Left) Cyclic load test; (Right) Fire Test The necessary compartmentalization was provided by
four fire proof panels. On one of the panel a 1 m height by
A three phase test procedure was adopted and followed in 3 m length opening was provided at the bottom to account
testing the RC frame which consisted of: for the ventilation. The panels were coated with a lining of
(a) Subjecting the frame to a simulated cyclic lateral load fire proof glass wool material which was followed by the
in a quasi-static fashion. blocks of same material tied to the panels with the help
of stainless steel rods. Extra care was taken to ensure
(b) Subjecting the damaged frame to one hour the fire tightness of the panels which was manifested
compartment fire and during the test with no smoke or flame leaking out of the
(c) Residualloadtest. compartment from any side except the opening.
The simulated cyclic lateral load was applied onto the RC Following the fire test the fire panels were removed and
frame using two displacement controlled double acting the RC frame was again tested under simulated lateral
hydraulic actuators having a capacity of producing the seismic load for measuring the residual lateral load
displacement of 300 mm in either direction acting in capacities of the frame.
tandem with each other against a strong reaction wall.
The initial seismic damage was achieved by inducing Results and Discussions
a pre-planned lateral displacement through applying
lateral cyclic load corresponding to the “Life Safety” Simulated Seismic Load Test
structural performance level (S-3), for frame “A”, and Figure 5 shows the load displacement response of all the
“Collapse Prevention” structural performance level (S- four frames. The test frames were observed after each
5), for frames “B”, “C” and “D”, of FEMA 356:2000. The cycle of displacement for detecting the formation and
damage level in frame “A” was ensured by subjecting the growth of cracks, if any.
RC frame to a maximum roof level lateral displacement of
In case of frame “A” no visually observable cracks were
72 mm corresponding to a roof drift ratio of 2 % as given
seen upto a lateral displacement of 20 mm corresponding
in FEMA 356. Similarly the damage level in frames “B”,
to a roof drift ratio of 0.53 %. The first crack was seen at
“C” and “D” was ensured by subjecting the RC frame to
the ends of the roof level beams oriented along the N-S
a maximum roof level lateral displacement of 150 mm
direction at a lateral displacement of 35 mm (roof drift
corresponding to a roof drift ratio of 4 % as given in FEMA
356.The test frames were subjected to a number push pull
2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on
512 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Outcomes of a Major Research on Full Scale Testing of RC Frames in Post Earthquake Fire

ratio of 0.92 %) corresponding to a base shear of 185

kN. The next crack was observed at the ends of the N-S
direction oriented plinth beam B1 at the measured lateral
displacement of 44 mm corresponding to a base shear of
212 kN. As expected, flexural cracking was seen to occur
earlier and was more pronounced in the beams oriented
parallel to the loading direction. Flexural cracking in the
columns was observed near their junction with the plinth
beams at a lateral displacement of 52 mm (roof drift ratio
of 1.37 %). Spalling of concrete at the ends of the roof
beams oriented along the N-S direction was observed
at a roof drift ratio of about 2 % along with widening of
cracks in the columns and the plinth beams. The test was
terminated at a roof displacement of 80 mm (roof drift
ratio of 2.11 %) corresponding to a base shear of 267 kN.
Stiffness degradation and pinching of the hysteresis loops
between successive cycles of loading is seen in Fig. 5”A”
though the hysteresis loops continued to remain stable
without any significant degradation of strength.
In case of frame “B” which was subjected to a drift of
4% numerous cracks developed near beam- column
intersection, both at floor as well as roof level during push
and pull cycles. Cracks on columns were initiated during
20 mm push cycle whereas those on beams were initiated
during 30 mm push cycle. Cracks were initiated on the
plane perpendicular to the loading direction and these
cracks further propagated in the subsequent cycles. The
crack widths in different members were measured using
a crack micro-scope and they varied from 0.1 mm to 2.4
mm with no spalling at the joints as shown in Figure 6 (a).
Most severe cracks were developed at the beam-column
junctions at 130 mm and 140 mm displacement cycles.
The average growth rate of crack width was found to be
0.1 mm per 20 mm cyclic displacement. Figure 5(c) shows
the load displacement history of the frame.
The RC frame “C” with non-ductile detailing showed the
signs of cracking from the initial 10 mm cycle with the
cracks getting wider with each successive loading cycle.
Cracks were mostly localised with wider cracks at the ends
of beam column joints. At the end of 50 mm cycle, cracks
as wider as .4 mm were observed on the columns. After
80 mm cycle a number of diagonal shear cracks started
forming at the ends of plinth and top beams, with these
shear cracks running into the slab offsets. With the start
of 100 mm cycles the cracking noise was quite audible and
the spalling of cover concrete started at the beam column
joints. The cracks in the beams started running into
slabs near top beams, symptomatic of composite action
between the RC slab of the test frame and the roof beams.
Cracks with a width of 4 mm started appearing in beams
near joints. Extensive spalling was observed in the top
beams in the direction of loading. A spall volume of 230
mm × 60 mm × 16 mm was registered at one of the beam
column joints with the resulting exposure of underlying
rebars. In one of the top beams all the three bars at the
Fig. 5: Measured Hysteretic curve of the test frame
beam column joint buckled between the shear stirrups

Organised by
India Chapter of American Concrete Institute 513
Session 5 A - Paper 1

Fig. 6: Beam column joint after seismic loading in (a) Frame “B” (b) Frame “C” (c) Frame “D”

as in Figure 6 (b). At the end of mechanical loading it was widening in this displacement cycle and spalling of mortar
seen that the number of fine cracks the as less in the was also observed. More number of bricks cracked in
RC frame with non-ductile detailing with more number 50 mm (1.67 % drift) displacement cycle with chunks of
of wider cracks than in the frame with ductile detailing. mortar coming out from the bonds of masonry in the in-
A plastic residual displacement of 41 mm was recorded plane walls. The out of plane walls were still intact with
after unloading in the RC frame with non-ductile detailing. no signs of distress and cracks. The diagonal cracks in the
two opposite in-plane walls widened and extended more
Figure 5 “D” shows the load displacement history of the
towards the corners of the walls. A new 1.2 mm wide
masonry infilled frame. In this frame no sign of cracks
shear crack was measured in beam B3 at column C3.
or distress was observed in any structural member
This crack originated from the bottom cornor of the beam
or masonry infill till the displacement of 6 mm. At the
and met the top section at 100 mm from the column C3.
displacement of 8 mm the cracks started getting developed
The first crack in slab appeared in this cycle and 0.06 mm
in the masonry- beam interface. At the displacement of
wide crack was measured in slab parallel to beam B8. At
10 mm, which corresponds to .3 % drift, cracks started
a displacement of 100 mm shear failure of beams B1 and
developing on infill walls. A brick in the in-plane wall was
B3 took place as shown in figure 6(c). At a displacement of
observed to be cracked in 10 mm (pull) displacement
120 mm buckling of reinforcement and shear stirrups in
cycle. The mortar also started to get loosened up in 10
beam B1 was observed.
mm cycle. Cracks in in-plane walls started progressing
in diagonal fashion towards the corners in 12 mm cycle.
Spalling of mortar also started with 12 mm displacement Fire Test
which corresponds to .4 % drift ratio. However no signs of Figure 7 (a) - (d) depict the compartment time-temperature
distress were found in any of the structural member of the history at one of the key locations of the RC frames “A”, “B”,
RC frame. The cracks in the masonry started widening in “C” and “D” respectively. The temperature history shows
12 mm cycle and progressed in diagonal fashion towards an 18 hour log with a complete heating and cooling cycle.
the corners. No cracks were observed in out of plane walls The maximum temperature recorded during the course
and structural elements. With a displacement of 25 mm a of fire was 1421 °C in RC frame “A”, 1370 °C in RC frame
number of bricks cracked with wider cracks. Separation “B”, 1369 °C in RC frame “C” and 1356 °C in RC frame “D”.
of the bottom and top layers of the bricks in the in-plane Though the compartment temperature developed in
masonry wall started in 30 mm cycle (1 % drift). With all the frames are almost comparable, the uniformity
this displacement cycle the cracks in the masonry walls of maximum compartment temperature should not be
started widening. The cracks were formed in a diagonal mistakenly taken as uniformity in the exposure levels
fashion with symmetry followed in push and pull events of for different structural elements as the fire dynamics
the cycle. At a drift of 1 %, cracks were observed in beams is dependent on a number of factors which cannot be
and columns also. Hairline cracks were formed in beam controlled during fire at such a large scale e.g. the
B1, B5, B3 and B7 at beam column joints. Flexure cracks direction of wind blowing at the time of test.
were also observed in column C3 at the secondary beam
joint.40 mm (1.33% drift) displacement cycle caused the In the test frame “A”, the spalling of concrete from the roof
shear cracks to get developed in beam B1 and B3 which slab was noticed immediately after about 4 to 5 minutes
were placed along the direction of force. The width of of initial ignition, which continued further for almost 15
these shear cracks in beam B1 at beam column joints minutes. The spalling was observed to begin when the
near column C1 and C2 was measured to be 0.1 mm. A 0.2 compartment temperature was between 300 °C to 400
mm wide flexural crack was also observed at the centre °C. Considerable damage was noticed in the roof slab in
of beam B1and B3. At column C3 a 1 mm wide shear crack terms of spalling of concrete and resultant exposure of
was measured in beam B3. Cracks in the masonry started reinforcement. This was mainly due to the movement of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

514 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Outcomes of a Major Research on Full Scale Testing of RC Frames in Post Earthquake Fire

fire plume towards the ceiling causing a rapid flow of hot

gases in the radial direction termed as a ceiling jet. Further,
the portions of the roof slab, which had cracked during
the lateral load test, indicated more spallling compared
to the other uncracked portions of the slab. The extent of
spalling, which exposed the bottom reinforcement in some
portions of the roof slab is shown in Fig. 8. The examination
of the spalled area and the explosive nature of the spalling
suggest high compressive forces induced in the roof slab
due to the restraint to the thermal expansion. The roof
slab did not collapse even after severe spalling which
means that the effect of compressive membrane would
considerably enhance the load-carrying capacity of the
roof slab above the calculated flexural action. In the later
stages of test, the cracking and spalling of concrete was
observed in beams and columns as well. It was observed
that in general, the fire induced surface cracks on the
various columns and beams initiated at a temperature
of about 600 °C. These cracks further widened at higher
temperatures. The portions of the columns and beams,
where the cracking and spalling was noticed during the
lateral simulated earthquake load test, experienced more
thermal damage compared to the other portions where
relatively negligible damage was noted during the initial
lateral load test.

Fig. 8: Spalled slab in Frame “A”

In the RC frame “C” with non-ductile detailing spalling

was observed in all the structural elements with massive
spalling in the slab. During the course of fire it was seen
that the slab started to spall within 4 minutes of the
fire with intense bullet shot sound like noise heard at 5
minutes. The spalling was intense at 6 minutes which
corresponded to a slab temperature of 300°C. Post
fire inspection of the structure revealed the damage
in various structural elements. As seen in figure 9 (a),
slab spalled extensively, attributed to high temperature
gradients across the slab cross section, with most of the
cover concrete lost exposing the reinforcement. At some
sections the reinforcement cross section was reduced by
melting of the steel. Sintering of aggregate was also seen
Fig. 7: Compartment temperature buildup in all the frames
in most of the slab area. A maximum temperature of 845
during the fire test

Organised by
India Chapter of American Concrete Institute 515
Session 5 A - Paper 1

°C was recorded in the slab section near column 1 of the

RC frame with Non-ductile detailing while a temperature
of about 450 °C was recorded at the point in RC frame
“B” with ductile detailing. Figure 9(b) shows the post
fire picture of the roof slab of RC frame “B” with ductile
detailing. This high temperature build up in the non-
ductile detailed RC frame can be attributed to extensive
wide cracks developed during the seismic load test. Figure
10 shows the variation of temperature with time in a slab
cross section near column 1 of both the RC frames. It can
be seen that the temperatures at a given time at a given
depth at a slab section is higher in RC frame with non
ductile detailing (NDTSC1D1-D3) than those in RC frame
with ductile detailing (DTSC1D1- D3). This can attributed
to higher amount of damage in the RC frame with Non-
ductile detailing.

Fig. 11: Temperature profiles in the beam cross sections in

frame “D”

load carrying capacity was only 5% at the displacement

of 150 mm in RC frame “B” with ductile detailing, the
reduction in load carrying capacity was about 35 % at the
Fig. 9: Post fire picture of slab for RC frame with (a) Non-ductile displacement of 150 mm in RC frame “C” with non ductile
detailing (b) ductile detailing
detailing. In the RC frame “D” with non ductile detailing
and masonry infills the loss in the residual capacity was
about 37%.

Concluding Remarks
A report on a full scale fire test on a damaged RC frame
has been presented. A novel test setup has been developed
and tested which gives an understanding of the global
behavior of a concrete structure in post earthquake fire.
The following can be concluded from the study:
1. The higher levels of initial damage cause more
number of wide cracks which leads to development
of higher temperatures in the structural elements of
the frame.
2. The set of data as extracted from the tests shows the
Fig. 10: Temperature profiles in the slab cross sections in frames
“B” and “C” localized nature of compartment fires which results in
highly non uniform heating of structures which is not
in line with the codal assumptions. While the results
The masonry infill walls in the frame “D” provided
from the test show maximum temperatures that are
insulation to the structural members. It was seen that
comparable to those that are predicted by standard
the measured temperatures in the structural elements of
procedures, the location of these maxima is very much
this RC frame were much lesser than the other frames.
localised.
Whereas the temperatures in the members of frame “C”
were very high with similar reinforcement detailing and 3. The cracks that appeared on the different elements
initial damage, the structural elements of the frame “D” of the test structure suggest that detailing of
witnessed very low temperatures as show in figure 11. reinforcement may have implications on the global
behaviour of the structure when exposed to fire.
Residual Test 4. Masonry provides insulation to the structural members
Residual test was carried out to measure the residual of the RC frames which results in less temperature
capacities of the frame after fire. While reduction in buildup in them resulting in higher residual strength.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

516 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Outcomes of a Major Research on Full Scale Testing of RC Frames in Post Earthquake Fire

Acknowledgement 11. Kamath, P. (2014), “Response of RC framed structures subject to

post - earthquake fire”, PhD Thesis, Department of Civil Engineering,
The authors thankfully acknowledge the financial support IIT Roorkee, Roorkee India.
received from Board of Research in Nuclear Sciences 12. Khoury, G.A. (2000), "Effect of Fire on Concrete and Concrete
(BRNS)- Mumbai and the U.K.- India Educational and Structures", Progress in Structural Engineering and Materials,
Research Initiative (UKIERI) for conducting this study. 2(4) pp. 429-447.
13. Kumar Virendra (2012), Fire performance of earthquake damaged
reinforced concrete structures. PhD Thesis, Department of Civil
References
Engineering, IIT Roorkee, Roorkee India
1. Bazant, Z.P. and Kaplan, M.F. (1996), “Concrete at High Temperature”,
Material Properties and Mathematical Models. Longman Group 14. Lea, F. C. (1920), “The effect of temperature on some of the
Limited, London, pp. 196. properties of materials”, Engineering London, 110, pp. 293-298.

2. BIS, Indian Standard IS 456 (2000). “Code of practice for plain and 15. Lea, F. C. and Stradling, R. E. (1922), “The resistance to fire of
reinforced concrete” Bureau of Indian Standards, New Delhi, India. concrete and reinforced concrete”, Engineering London, 114, pp.341-
344.pp.380-382.
3. BIS, Indian Standard, IS 1893(Part 1) (2002). “General provisions and
buildings: Criteria for earthquake resistant design of structures.” 16. Malhotra, H. L. (1956), “The effect of on the compressive strength
Bureau of Indian Standards, New Delhi, India. of concrete”, Magazine of Concrete Research (London) 8, pp. 85-94.

4. BIS, Indian Standard, Code IS 875 (Part 1) (1987). “Code of practice for 17. Purkiss, J. A. and Dougill, J. W. (1973), “Apparatus for compression
design loads (other than earthquake) for buildings and structures” tests on concrete at high temperatures”, Magazine of Concrete
Bureau of Indian Standards, New Delhi, India, 2003. Research, 25(83), pp. 102-108.

5. BIS, Indian Standard, Code IS IS 13920:1993 (1993), (Reaffirmed 18. Saemann, J. C. and Washa, G. W. (1957), “Variation of mortar and
2008) “Ductile detailing of reinforced concrete structures subjected concrete properties with temperature”, ACI Journal Proceedings,
to seismic forces-Code of practice, (Reaffirmed 2008)”, BIS New 54(5), pp. 385-395.
Delhi. 19. Schneider, U., and Herbst, H. (2002), “Theoretical considerations
6. Carvel, R. (2005), “Fire protection in concrete tunnels”, The about spalling in tunnels at high temperatures”, Technical Report,
Handbook of Tunnel Fire Safety (Eds. Beard, A. & Carvel, R.) Thomas Technical University Vienna, Austria.
Telford, London. 20. Thomas PH, Heselden AJM (1972) “Fully developed fires in single
7. Chen WF, Scawthorn C (2003) New directions in civil engineering, compartments. A Co-Operative Research Programme of the Conseil
Earthquake Engineering Handbook. CRC Press, Boca Raton, FL International Du Bâtiment, Building Research Establishment”, CIB
Report No. 20, Fire Research Note 923, Fire Research Station, UK.
8. Drysdale D (1998) An Introduction to Fire Dynamics, 2nd edn. Wiley,
Chichester, UK 21. Wellington Lifeline Group (2002) Fire following earthquake:
Identifying key issues for New Zealand, Report on a project
9. FEMA, FEMA 356 (2000) “Prestandard and commentary for the undertaken for the New Zealand fire service contestable research
seismic rehabilitation of buildings”, Building Seismic Safety Council, fund. Fire Research Report. New Zealand Fire Service Commission,
Federal Emergency Management Agency, Washington DC Wellington, NZ, 2002.
10. ISO (International Organisation for Standardization) (1975), ISO 22. Zoldners, N. G. (1960), “Effects of high temperature on concrete
834: Fire resistance tests. Elements of building construction. ISO, incorporating different aggregate”, ASTM Proceedings, 60, pp.
Geneva, Switzerland. 1087-1108.

Asif Hussain Shah

Asif Hussain Shah graduated from National Institute of Technology Srinagar, Kashmir, with a B-Tech in Civil
Engineering in 2010 and with an M-Tech in 2012 in structural Engineering from the same university with
a Gold Medal. He submitted his Doctoral thesis titled “An Experimental Investigation of Fire Performance
of Earthquake Damaged Structures” in September 2015. Currently he is a Research associate at Indian
Institute of Technology Roorkee, India working on the behaviour of concrete structures in Fire. His main
research interests in Civil Engineering include Behaviour of structures in extreme conditions, Durability of
concrete structures, condition assessment and Non destructive testing of Concrete structures, traditional
constructions. The paper is based on a part of his PhD research.

Dr. Umesh Kumar Sharma

Dr. Umesh Kumar Sharma is an Associate Professor in the Department of Civil Engineering, Indian
Institute of Technology Roorkee. His areas of research interest include degradation mechanisms of
reinforced concrete, confinement of concrete, fire effects on concrete structures and evaluation and repair
of concrete structures.

Organised by
India Chapter of American Concrete Institute 517
Session 5 A - Paper 2

Using calorimetry to understand fly ash reactivity in high volume fly

ash concrete
Dr Paul J Sandberg and Pratik Bhayani, Calmetrix

Abstract activity in a well-defined chemical environment, excluding

any interference from Portland cement clinker.
Isothermal calorimetry is a well-established technique
for directly measuring the hydration of Portland and While the pozzolanic activity is normally referred to as
blended cements. Isothermal calorimetry is gaining the ability of the pozzolanic material to react with calcium
popularity because of its superior repeatability and ease hydroxide to form calcium silicate hydrates, the presence
of use compared to traditional methods, e.g, compressive of highly soluble alkalis, e.g. sodium and/or potassium,
strength development. However, it is difficult to isolate is critical. After a few hours of cement hydration, the
the hydration activity of fly ash in mixtures of fly ash and soluble sodium and potassium are balanced by hydroxide
Portland cement, since the heat evolved from Portland ions in the pore water, thereby largely controlling the pH
cement hydration is much larger and easily overwhelms of the pore water. This is important because the pH in
the heat evolution from low calcium fly ash, which is turn strongly impacts the dissolution of the amorphous
predominantly found in India. phase in the fly ash, which is a necessary step before the
pozzolanic reaction can take place.
This paper describes a novel use of isothermal calorimetry
to measure the hydration activity of fly ash directly
in a simulated Portland cement environment, i.e. an Experimental
environment that contains the relevant chemistry provided Two samples of Indian fly ash labeled “NTPC” and
by a Portland cement, but without the actual hydration of “TANDA” were tested for pozzolanic activity with calcium
Portland cement. The results are very promising and open hydroxide and various concentrations of sodium hydroxide
the way for the development of a direct pozzolanic activity in the mix water at a fixed water-to-solids ratio of 1.0 for 3
test of fly ash using isothermal calorimetry, which can be weeks at 20 °C without the presence of Portland cement
performed at a wide range of temperatures and without in an I-Cal 8000 HPC isothermal calorimeter.
the variability or noise provided by Portland cement
Keywords: Isothermal calorimetry, heat of hydration, fly Test Procedure
ash, pozzolanic activity, temperatures. In this test, sodium hydroxide was added to the mix water
at zero, 0.1, 0.5 and 1.0 M, representing cement pore water
Introduction of pH 13 – 14. Typically, most cements will generate a pore
water in the range of pH 13-14 depending on the soluble
The purpose of this study was to explore if isothermal
alkali content of the cement and the water-cement ratio.
calorimetry can be used to measure directly the
Most fly ashes themselves contain soluble alkali, such that
pozzolanic activity of fly ash in a “simulated Portland
the pH driving the dissolution of the amorphous phase is
cement environment” without actual Portland cement.
typically above pH 13.
Normally, fly ash activity is tested in a mixture with
Portland cement, which makes the resulting “activity” 31 g calcium hydroxide was pre-mixed with 50 g mix water
result heavily dependent on the chemistry and activity of in the calorimetry sample cup, together with sodium
the Portland cement itself. Furthermore, the heat from hydroxide, if any, and left inside the calorimeter to pre-
fly ash hydration is very low compared to that of Portland condition the temperature to 20.0 °C. Once the mix water
cement, thus making it difficult to measure the fly ash – calcium hydroxide slurry was at thermal equilibrium
activity directly with calorimetry in a mixture of Portland with the calorimeter, the sample cup was quickly removed
cement and fly ash. Therefore, in this novel approach we from the calorimeter and 50 g fly ash was mixed into the
use heat of fly ash hydration as a test method for fly ash mix water – calcium hydroxide slurry for 60 seconds,

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

518 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Using calorimetry to understand fly ash reactivity in high volume fly ash concrete

concentration of sodium hydroxide

(0, 0.1, 0.5 and 1.0 M, respectively).
Figures 3-4 show a close up of the
same results during the first two
days.

Discussion
1) The results obtained show that
it is clearly possible to measure
the fly ash activity in the absence of
Portland cement using isothermal
calorimetry, but one has to be careful
when designing the test matrix for
Fig. 1: Power (rate of reaction) by weight of NTPC fly ash during 3 weeks of hydration results that are both representative
and within the measureable range
of the calorimeter.
2) Since the heat of activity is
generally very low for low calcium
fly ash, it is important to use as
large a quantity of fly ash in the
sample as possible, to achieve the
highest possible signal for long
term studies. In this regard it is often
best to prepare two samples for
each calorimetry test: one sample
with a slightly smaller mass of fly
ash using internally equilibrated
mix water and a disposable spoon
left inside the sample cup as for the
Fig. 2: Power (rate of reaction) by weight of TANDA fly ash during 3 weeks of hydration examples above (as above), in order
to accurately capture the initial
reactivity. A second calorimetry
sample with the maximum possible
fly ash mass in the sample cup,
which implies that the fly ash –
calcium hydroxide mixture be
prepared externally in a larger
mixing bowl to facilitate proper
mixing.
3) The I-Cal 8000 HPC is clearly
capable of measuring differences
in fly ash activity as shown above.
However, for long term studies it
is critical to minimize any noise as
the signal is likely to be very small,
Fig. 3: Power (rate of reaction) by weight of NTPC fly ash during the first two days of
even with a full sample vial of fly
hydration.
ash-calcium hydroxide mixture. In
such cases, the proper alternative
using a disposable spoon that was left inside the sample is to rely on a calorimeter with completely separate
cup. The sample cup was loaded into the calorimeter and sample cells, such as the I-Cal 2000 HPC model.
the signal was logged for 3 weeks at 20.0 °C. The separation of sample cells minimizes corss-talk
between adjacent samples and guarantees the lowest
Results noise and best baseline stability.
Figures 1-2 show the activity of the two fly ash samples 4) It is probably of interest to explore the effect of
in presence of excess calcium hydroxide and a variable temperature on the pozzolanic activity. Isothermal

Organised by
India Chapter of American Concrete Institute 519
Session 5 A - Paper 2

Conclusion
Isothermal calorimetry appears to
be very well suited for measuring
the pozzolanic activity of fly ash
as a function of temperature in
a “simulated Portland cement
environment” emulating the
fundamental elements normally
provided by Portland cement.
The use of a “simulated Portland
cement environment” without
actual Portland cement greatly
improves the usefulness of the
calorimetry test method as it
Fig. 4: Power (rate of reaction) by weight of TANDA fly ash during the first two days of
excludes any thermal noise that
hydration.
would result from the hydration of
the Portland cement.
calorimetry is well suited for this, given its built-
in superior temperature control that effectively References
eliminates the need for a larger climate control 1. Zhang, T., Yu, Q., Wei, J., Gao, P., Zhang, P., 2012. Study on
chamber for tests at different temperatures. optimization of hydration process of blended cement. J Therm
Anal Calorim, 107, 489–498
5) As for the detailed results, the NTPC fly ash was
considerably more active compared to the TANDA fly 2. Kocaba, V., 2009. Development and evaluation of methods to follow
microstructural development of cementitious systems including
ash at all conditions tested. slags. PhD thesis No. 4523, École Polytechnique Fédérale de
Lausanne, Switzerland, pp. 239.
6) For both fly ashes, the difference in activity between
zero and 0.1 M sodium hydroxide was quite small. 3. Zhang, Y., Sun, W., Liu, S., 2002. Study on the hydration heat of
binder paste in high performance concrete. Cem. Concr. Res. 32,
For the more reactive NTPC fly ash, the effect of 1483-1488.
higher sodium hydroxide concentrations was quite
4. Snellings R, Mertens G, Elsen J., 2010. Calorimetric evolution of
substantial. However, for the much less reactive the early pozzolanic reaction of natural zeolites. J Therm Anal
TANDA fly ash the sodium hydroxide concentration did Calorim. 101, 97–105.
not impact reactivity – probably because the TANDA fly 5. Gruyaert E, Robeyst N, De Belie N., 2010. Study of the hydration of
ash contains less of the pozzolanic amorphous phase. Portland cement blended with blast-furnace slag by calorimetry
and thermogravimetry. J Therm Anal Calorim. 102, 941–51.

Dr Paul Sandberg
Dr Paul Sandberg has a background in cement chemistry and concrete durability from both industry and
academia in Sweden and USA. He worked 9 years for Cementa, a cement producer in Sweden and then 12
years for W.R. Grace, and admixture producer headquartered in Cambridge, MA, USA. Paul’s research
has focused on the interaction between Portland cement, SCM and admixtures, as well as on durability
of cementitious binders. Paul has 6 patents and many papers published on durability and the use of
calorimetry for sulfate optimization and studying cement-SCM-admixture compatibility.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

520 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Tensile Performance of SHCC Exposed to Low and High Temperatures

Tensile Performance of SHCC Exposed to Low and High Temperatures

Keitetsu ROKUGO, Koichi KOBAYASHI, Daichi HAYASHI and Yukio ASANO
Department of Civil Engineering, Gifu University, JAPAN
Mitsuo OZAWA,
Department of Environmental Engineering Science, Gunma University, JAPAN
Hyundo YUN
Department of Architectural Engineering, Chungnam National University, KOREA

Abstract has not been fully elucidated (RILEM, 2011). Wu, R. et al.
(2014) reported that the strain capacity of SHCC made
This study aims to elucidate the tensile performance of
with quartz sand decreased at temperatures above
Strain-Hardening Cement Composite (SHCC) exposed
to low and high temperatures. Uniaxial tension testing 50 °C and that the composite material was less
was conducted on SHCC dumbbell-shaped specimens, temperature sensitive, if natural river sand was used.
immediately after removal from cooled or heated Oliveira, A.M. et al. (2014) reported that the tensile strength
conditions ranging from -20 ° to 180 °C and after being of SHCC decreased and the strain capacity increased with
allowed to return to room temperatures. The cracking an increase in temperature and that the strain capacity
strength and tensile strength of the SHCC decreased as was reduced at elevated temperatures by reducing the
the specimen temperature increased. These strengths internal humidity of the specimens (from 95% to 0%). The
clearly recovered after returning to room temperatures. effects of temperature and strain rate on the behavior of
The degree of such recovery became greater with a longer SHCC subjected to tensile loading have been investigated
period after returning to room temperatures. The ultimate (Mechtcherine et al., 2013). Physical and mechanical
strain of SHCC specimens increased by approximately properties of SHCC after exposure to high temperature
1% as the tested temperature of specimens increased have been studied (Magalhães et al., 2009). The authors
from 40 °C to 100 °C in 20 °C steps. The ultimate strain investigated the influence of freeze-thaw actions on the
of specimens after returning to room temperatures was mechanical properties of SHCC (Yun and Rokugo, 2012,
nearly the same as that of 20 °C specimens. Jang et al., 2014).

Keywords: Tensile performance, fiber reinforced mortar, In this study, the authors investigated the tensile
SHCC, thermal history, high temperature performance, including cracking strength, tensile
strength, and ultimate strain, of SHCC containing PE
fibers immediately after removal from cooled or heated
Introduction
conditions ranging from -20°C to 180°C and after allowing
Strain-Hardening Cement Composites (SHCCs) are high to return to room temperatures, by uniaxial tension tests
toughness fiber-reinforced mortars that show pseudo on dumbbell-shaped specimens. The tests also included
strain-hardening behavior and multiple fine cracking uniaxial tension testing on the matrix mortar (equally
behavior under tensile forces. SHCC contains high density proportioned to the SHCC but with no fibers) under the
polyvinyl alcohol (PVA) fibers or high density polyethylene same temperature conditions and compression testing
(PE) fibers measuring around 10 mm in length and 0.01- on the SHCC and the matrix mortar under the same
0.04 mm in diameter at a ratio of 1-2% by volume. These conditions.
fibers are inferior in heat resistance as compared with
the cement matrix. This material has increasingly been
Experiment Overview
used for repair of the surfaces of agricultural waterways
and repair of such structures as roads and railways. The Test Types and Conditions
surfaces of such concrete structures are subjected to low
Uniaxial tension testing and compression testing
and high temperatures ranging from -20°C in winter in
were conducted on dumbbell-shaped and cylindrical
cold climates to 40 to 60°C on the south-facing sides in
specimens, which were made of the SHCC and the matrix
summer. The temperature can be higher in the event of
mortar, immediately after a temperature history to a
fire.
high or low temperature (hereinbelow in a “heated” or
Tension testing on SHCCs is generally conducted at “cooled” condition) and after being allowed to return to
normal room temperatures of around 20°C. Their tensile room temperatures of 20°C (hereinbelow a “reverted”
performance including tensile strength and ultimate condition) to investigate their mechanical performance.
strain (strain capacity) at such low temperatures as Table 1 gives the test conditions. The nine levels of thermal
-20°C and such high temperatures as 100°C or higher conditions ranged from -20°C to 180°C.

Organised by
India Chapter of American Concrete Institute 521
Session 5 A - Paper 3

Table 1. Test conditions

Heating Cement Specimen temperature (drying oven was used at heating) (°C) Specimen temperature (electric furnace
history composites was used at heating) (°C)
-20 0 20 40 60 80 100 100 140 180
Heated, cooled SHCC ¡ ¡ ¡ ¡ ¡ ¡ ¡
or room Mortar ¡ ¡ ¡ ∆ ¡
conditions
Reverted SHCC ¡ (¡) ¡ ∆ ¡ ¡ ¡ ¡
conditions Mortar ¡ (¡) ¡ ∆ ¡ ¡ ¡ ¡
∆:Only compression tests, (¡): Same with those in room condition (20 °C)

Materials and Mixture Proportions specimens protected from wetting by polyethylene bags
were immersed in ice water. A drying oven was used for
Table 2 gives the mixture proportions of the SHCC. Silica
40°C or higher. Specimens were kept in the specified
sand No. 7 and high-early-strength portland cement were
ambient temperatures for 2h, and subjected to loading
used as the fine aggregate and cement, respectively. In
tests immediately after removal from each environment
order to reduce the shrinkage of the SHCC, 25% of the
(heated or cooled conditions) and 24 h later (reverted
cement content was replaced with limestone powder (LP).
conditions).
PE fibers (diameter: 0.012 mm, length: 12 mm, tensile
strength: 2.6 GPa, elastic modulus: 88 GPa, density: 0.97 For the three thermal conditions on the right side of the
g/cm3, coefficient of thermal expansion: -0.000012/°C and header row of Table 1 (100°C, 140°C, and 180°C tested at
melting point: 145°C) were included at a ratio of 1.25% by Gunma University), specimens were heated at a constant
volume. It should be noted that the length of these PE fibers heating rate (5°C/min) to the specified temperatures using
decreases as the temperature increases and increases as an electric furnace, kept at the temperatures for 2 h, and
the temperature decreases (negative expansiveness in slowly allowed to cool down in the furnace for a day. These
the fiber longitudinal direction). The matrix mortar was specimens were then transferred to Gifu University and
equally proportioned to the SHCC, excluding the fibers, subjected to loading tests at 20°C 1 week (6 or 7 days) after
with a reduced dosage of air-entraining and high-range the cooling phase.
water-reducing agent (SP, an ether polycaboxylate-type
compound).

Specimens and Thermal Conditions

Dumbbell-shaped specimens (specimen length: 330 mm,
cross-sectional area of the gauge zone: 30 by 15 mm, and
gauge length: 80 mm) as shown in Fig. 1 were fabricated
for uniaxial tension testing. Steel molds for the dumbbell
shape were used to fabricate 60 specimens at a maximum
from each batch. Cylindrical specimens (diameter: 50 mm
and height: 100 mm) were also fabricated for compression
testing using plastic molds. Fig. 1: Dumbbell-shaped specimen
Mixtures placed in molds were left for one day in a
Loading Test
chamber at 20°C, demolded, water-cured for 1 week, and
then air-cured for more than 1 week. The test ages ranged Five dumbbell-shaped specimens and three (six only in the
from 32 to 42 days. case of 20°C) cylindrical specimens were used for each
set of conditions in uniaxial tension tests and compression
Seven thermal conditions ranging from -20°C to 100°C, in
tests, respectively. The strengths and ultimate strain are
steps of 20°C as given on the left side of the header row of
expressed as averages of the test results.
Table 1, were applied to the specimens at Gifu University.
A freezer was used for cooling down to -20°C. For 0°C, Figure 2 shows the tension test apparatus used for

Table 2. Mixture proportions of SHCC

SHCC W/C W/P Unit mass (kg/m3)
PE: 1.25% (%) (%) Water Powder Silica sand SP Methyl-cellulose PE
Cement LP
42.9 30.0 380 887 380 352 19.00 1.76 12.1

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

522 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Tensile Performance of SHCC Exposed to Low and High Temperatures

as the specimen temperature increases from 20°C to

60°C, and to 100°C. The number of cracks increased as
the specimen temperature increased from 20°C to 60°C,
and to 100°C, as found from the repeated stress drops on
the tensile SSCs corresponding to microcracking.
Figure 4 compares the tensile SSCs of SHCC dumbbell
specimens after being allowed to cool down from 100°C
with those at room temperatures (20°C) and those in the
heated condition at 100°C. The shapes of the tensile SSCs
of SHCC specimens that slowly cooled down from 100°C is
close to those of 20°C specimens, with the ultimate strain
being reduced.

Fig. 2: Tension test apparatus

uniaxial tension testing (JSCE, 2008). Tensile loads were

measured using a load cell with a capacity of 10 kN, and
the deformation of the gauge zone was measured using
displacement gauges with a capacity of 25 mm attached
Fig. 3: Typical tensile stress-strain curves of SHCC
to the top and bottom ends of the zone. The test for each
specimen took 3 to 10 min. According to measurement
using a non-contact surface thermometer, the surface of
a 100°C SHCC specimen, for instance, was around 95°C
in the drying oven and 45°C at the end of the test (at most
10 min after removal from the drying oven). The internal
temperature of specimens presumably remained high.
Based on the results of uniaxial tension tests on the
SHCC, the cracking strength, tensile strength, and
ultimate strain were determined from the point at which
the stress significantly decreases for the first time, the
point of maximum stress, and the strain at the point of
tensile strength, respectively, on the tensile stress-strain
curve (SSC). Only tensile strength (= cracking strength)
was determined in the uniaxial tension tests on the matrix
Fig. 4: Tensile stress-strain curves of SHCC in heated, reverted
mortar.
and room conditions
Uniaxial Tension Test Results and Discussion Cracking Strength and Tensile Strength
Tensile Behavior of SHCC Exposed to Different Figures 5 and 6 show the average tensile strength of
Temperatures the SHCC and matrix mortar specimens in the heated
Figure 3 shows typical tensile SSCs of SHCC dumbbell- (or cooled) and reverted conditions, respectively. Figure
shaped specimens in heated or cooled conditions to 5 demonstrates that, in the temperature range of the
-20°C, 60°C, and 100°C, as well as at 20°C. These figures present tests of -20°C to 100°C, the tensile strength of
reveal that no appreciable difference is observed between the SHCC and matrix mortar tends to decrease as the
the shapes of the tensile SSCs of SHCC specimens of the specimen temperature increases. The tensile strengths
cooled condition (-20°C) and those of the room condition of the SHCC and matrix mortar in the heated condition at
(20°C). In contrast, the ultimate strain is found to increase 60°C and 100°C are around 70% and 30%, respectively, of
and the variability of the shape of tensile SSCs to decrease those at 20°C.

Organised by
India Chapter of American Concrete Institute 523
Session 5 A - Paper 3

The tensile strengths of both the SHCC and matrix mortar

after being allowed to cool down (Fig. 6) clearly recovered
from those in the heated condition at 60°C or 100°C
(Fig. 5). When the SHCC and matrix mortar specimens
were heated in an electric furnace to 100°C, 140°C, and
180°C and allowed to cool down for a week, their tensile
strengths recovered more significantly than the case
of heating in a drying oven and cooling for 1 day, nearly
achieving the strength values of 20°C specimens. The
causes of such significant recovery in tensile strength
may include, as will be described later, the effects of the
long period from the end of the cooling phase to loading
testing and moisture absorption during the period, as well
as the effects of heating and cooling rates. These remain
problems for future study.
Figure 7 shows the average cracking strength of the
SHCC in the heated (or cooled) and reverted conditions. Fig. 6: Tensile strength of SHCC and mortar in reverted
Note that the tensile strength of mortar equals the conditions
cracking strength, as stated above. In the temperature
range of the present tests, the cracking strength of the
SHCC also increases as the temperature decreases and
vice versa. The cracking strength after returning to room
temperatures significantly recovers, nearly returning to
the strength level at 20°C.

Ultimate Strain
Figure 8 shows the average ultimate strain of the SHCC in
the heated (or cooled) and reverted conditions. As found
from this figure, the ultimate strains of SHCC specimens at
0°C and -20°C do not appreciably differ from that at 20°C.
However, in the range from 20°C up to 100°C, the ultimate
strain linearly increases as the specimen temperature
increases, becoming greater by approximately 1% as the
temperature increases in steps of 20°C from 40°C to
100°C. In regard to specimens heated in a drying oven
to 60°C and 100°C and then allowed to cool down, the Fig. 7: Cracking strength of SHCC in heated, cooled, reverted
ultimate strain is as small as around 25% of that in the and room conditions
heated condition.

Fig. 5: Tensile strength of SHCC and mortar in heated, cooled Fig. 8: Ultimate strain of SHCC in heated, cooled, reverted and
or room conditions room conditions

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

524 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Tensile Performance of SHCC Exposed to Low and High Temperatures

As to specimens heated to 100°C, 140°C, and 180°C in an

electric furnace, allowed to cool down, and tested 1 week
later, the ultimate strains are similar to that at 20°C.

Causes of Large Ultimate Strain of SHCC in Heated

Conditions
Table 3 summarizes the tension test results of the SHCC
and matrix mortar subjected to thermal condition up to
20°C, 60°C, and 100°C. As stated above, the cracking
strength decreases as the tested temperature of SHCC
specimens increases, but the tensile strength does not
decrease to that extent in the range of 20°C to 100°C. The
widened difference between both presumably increases
the region of pseudo strain-hardening, increasing the
ultimate strain. Figure 9 shows the difference between
the tensile strength and cracking strength related to the Fig. 9: Ultimate strain and difference between tensile and
ultimate strain. cracking strengths of SHCC
When combining the tensile strength test results of matrix
mortar specimens (= cracking strength), the following are
possible causes of the reduction in the cracking strength
of the SHCC:
1. Drying and dehydration due to heating caused
reduction in the cohesive strength in the mortar matrix
of the SHCC to cause voids and microcracks, reducing
the cracking strength.
2. Temperature gradients within specimens during
heating and loading contributed to the loss in the
cracking strength.
Meanwhile, specimens that returned to room
temperatures may have absorbed moisture during
the time before loading testing, restoring the cohesive
strength within the mortar matrix, which may have caused Fig. 10: Compressive strength of SHCC and mortar in heated,
recovery of cracking strength. These possibilities need to cooled and room conditions
be investigated henceforth. Note that no cracks (wider
than 0.01 mm) were identified by a microscope, before
loading testing, on the surfaces of SHCC specimens that
were allowed to cool down from 100°C.

Table 3. Tension test results of SHCC and mortar subjected to temperature histories up to 20, 60 and 100 °C
Heated conditions Reverted conditions
SHCC Mortar SHCC Mortar
Cracking strength (N/mm2) 5.83 (5.83)
5.37 (5.37)
Tensile strength (N/mm2) 8.58 (8.58)
20°C
Strength difference (N/mm2) 2.75 (2.75)
Ultimate Strain (%) 0.72 (0.72)
Cracking strength (N/mm ) 2
4.19 5.58
Specimen 1.87 3.76
temperature Tensile strength (N/mm2) 6.21 8.12
60°C
(heated in drying Strength difference (N/mm2) 2.02 2.54
oven)
Ultimate Strain (%) 3.9 0.78
Cracking strength (N/mm2) 3.35 5.13
1.68 2.52
Tensile strength (N/mm )
2
6.12 7.08
100°C
Strength difference (N/mm2) 2.77 1.95
Ultimate Strain (%) 5.98 1.61

Organised by
India Chapter of American Concrete Institute 525
Session 5 A - Paper 3

specimens, respectively, immediately after subjecting the

specimens to seven levels of temperatures ranging from
-20°C to 100°C in 20°C steps and after allowing them to
cool down or heat up. Separate tests were also conducted
at three temperature levels of 100°C, 140°C, and 180°C
for strength testing after allowing the specimens to cool
down. The findings within the range of this study are
summarized as follows:
1. The cracking strength and tensile strength of the
SHCC and the tensile strength of the matrix mortar
decreased as the specimen temperature increased.
These strengths clearly recovered after returning to
room temperatures when compared with those in the
Fig. 11: Compressive strength of SHCC and mortar in reverted heated condition. The degree of such recovery became
conditions greater with a longer period after returning to room
temperatures.
Compression Test Results and Discussion
2. The ultimate strain of SHCC specimens increased
Figures 10 and 11 show the results of compression tests by approximately 1% as the tested temperature of
on cylindrical SHCC and matrix mortar specimens in the specimens increased from 40°C to 100°C in 20°C
heated (or cooled) and reverted conditions, respectively. steps. The ultimate strain of specimens after returning
The compressive strength of the SHCC is 15% to 20% to room temperatures was nearly the same as that of
lower than that of the matrix mortar, both in the heated 20°C specimens.
(or cooled) (Fig. 10) and reverted (Fig. 11) conditions. This
is presumably because the included PE fibers form a 3. The compressive strength of the SHCC and matrix
weakness under compressive forces. mortar decreased as the specimen temperature
increased. The compressive strength slightly
In both figures, the compressive strength of both the recovered after allowing the specimens to cool down.
SHCC and matrix mortar decreases as the specimen
temperature increases, in the temperature range of 4. The compressive strength of the SHCC was 15% to
-20°C to 100°C. When comparing the results of the 20% lower than that of the matrix mortar.
uniaxial tension tests and compression tests as a whole, The authors intend to carry out various tests to investigate
the strength reduction with the increase in the applied the causes of the phenomenon in which a higher
temperature was most evident in the tensile strength of temperature of specimens leads to a larger ultimate
the matrix mortar. strain. The tests will include the following: Tests in which
Similarly to tensile strength, the compressive strengths specimens are immersed in hot water beforehand to
of both the SHCC and matrix mortar specimens after avoid drying, tests in which specimens are dried at room
returning to room temperatures from 60°C or 100°C (Fig. temperatures beforehand, and tests in which PVA fibers
11) slightly recovered from those in the heated condition are used instead of PE fibers.
at 60°C or 100°C (Fig. 10). The compressive strength of References
concrete subjected to high temperatures is well known 1. Jang, S.J., Rokugo, K., Park, W.S., and Yun, H.D., 2014. Influence
to recover over time, but the recovery of tensile strength of rapid freeze-thaw cycling on the mechanical properties of
has scarcely been known, partly because uniaxial tension sustainable strain-hardening cement composite (2SHCC). Materials,
tests have scarcely been conducted. 7(2), 1422-1440.
2. JSCE Concrete Committee, 2008. Recommendations for design and
As found from Fig. 11, the compressive strength of both construction of high performance fiber reinforced cement composites
SHCC and matrix mortar specimens heated to 100°C in a with multiple fine cracks (HPFRCC). Concrete Engineering Series,
drying oven is nearly the same as the case of heating in an (82). http://www.jsce.or.jp/committee/concrete/e/hpfrcc JSCE.pdf
electric furnace. It is therefore found that the difference 3. Magalhães, M.S., Toledo Filho, R.D., and Fairbairn, E.M.R., 2009.
between these heating apparatuses scarcely affect the Physical and mechanical properties of strain-hardening cement-
based composites (SHCC) after exposure to elevated temperatures.
compressive strength. Proceedings of the International Conference on Advanced Concrete
Materials (ACM), 203-207.
Conclusions 4. Mechtcherine, V., Silva, F.A., Muller, S., Jun, P. and Toledo Filaho,
This study aims to elucidate the tensile performance R.D., 2013. Effects of temperature and strain rate on the behavior
of strain-hardening cement-based composites (SHCC) subjected
of SHCC subjected to low and high thermal conditions. to tensile loading. Proceedings of the 8th International Conference
Uniaxial tension tests and compression tests were on Fracture Mechanics of Concrete and Concrete Structures
conducted on dumbbell-shaped and cylindrical (FraMCoS-8).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

526 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Tensile Performance of SHCC Exposed to Low and High Temperatures

5. Oliveira, A.M., Silva, F.A., Fairbairn, E.M., and Toledo Filho, R.D., 7. Wu, R., Wittmann, F.H., Wang, P. and Zhao, T., 2014. Influence of
2014. Temperature and internal moisture effects on the tensile elevated and low temperature on properties of SHCC. Proceedings
behavior of strain-hardening cement-based composites (SHCC) of the 3rd International RILEM Conference on SHCC (SHCC3-Delft),
reinforced with PVA fibers. Proceedings of the 3rd International 3-8.
RILEM Conference on SHCC (SHCC3-Delft), 51-60.
8. Yun, H.D. and Rokugo, K., 2012. Freeze-thaw influence on the flexural
6. RILEM TC 208-HFC SC2, 2011. Chapter 4-Durability under thermal properties of ductile fiber-reinforced cementitious composites
loads, Durability of Strain-Hardening Fiber-Reinforced Cement- (DFRCCs) for durable infrastructures. Cold Regions Science and
Based Composites (SHCC), RILEM State of the Art Report, Springer. Technology, 78, 82-88.

Professor Keitetsu ROKUGO

Professor Keitetsu ROKUGO earned his Doctor degrees from Kyoto University in 1980, Japan. He also
studied in the University of Illinois at Urbana–Champaign from 1978 to 1979. Then he worked in Kyoto
University and moved to Gifu University in 1980. He became a professor in 1993. He was also a visiting
scholar of EPFL, Switzerland in 1986 for a year. From 2012 to present, he has been a dean of Faculty of
Engineering, Gifu University.
He contributed to publish the State-of-the-Art Report of the RILEM Technical Committee 208-HFC, SC3,
“Strain Hardening Cement Composites: Structural Design and Performance”, as a chair of SC3. He was
a chair of the committee on design and construction of HPFRCC in JSCE, and he contributed to establish
“Recommendations for Design and Construction of HPFRCC”. His research interests span a wide variety
of concrete such as “fracture mechanics of concrete and concrete structures” and “fiber reinforced
cementitious composites and their applications”.

Organised by
India Chapter of American Concrete Institute 527
Session 5 A - Paper 4

Effect of Slab on Strength and Behaviour of

Exterior RC Beam Column Joint
N. Ganesan, Nidhi M. and P.V. Indira
Department of Civil Engineering, National Institute of Technology Calicut

ABSTRACT positive cycle and 52-65% in the negative cycle due to the
presence of slab (Ahmed et al, 2012). Thus, ignoring the
This paper investigates the effect of slab on the strength
effect of slab can lead to underestimation of beam capacity
and behaviour of exterior reinforced concrete (RC) beam
and violation of strong column-weak beam philosophy.
column joint under reverse cyclic loading. Beam column
The extensive research works carried out by several
joints with and without slab were tested to investigate
investigators support that the effect of slab participation
the role of slab in the overall response of the joint. The
improves flexural strength, structural stiffness, energy
results indicate that the beam column joint with slab
dissipation capacity of the joint (Durrani and Wight, 1987,
showed better performance with respect to the first crack
Pantazopoulou and French, 2001, Shin and LaFave, 2004,
load, strength, ductility and energy absorption capacity
Fenwick et al., 2006 and Canbolat and Wight 2008)
of the joint when compared to joint without slab. Hence,
neglecting the slab effect significantly underestimates the Large number of researches have been conducted on
strength and ductility of the beam column joint. high strength concrete (HSC) (Shah and Ribakov, 2009,
Ishakov and Ribakov, 2009) fibre reinforced high strength
Keywords: Beam column joint, reverse cyclic loading,
concrete (Holschemacher et al 2010), high performance
ductility, energy absorption capacity, stiffness degradation
concrete (HPC) (Elahi et al, 2010) and also fibre reinforced
HPC (Ganesan et al., 2007 and Ganesan et al., 2013).
Introduction However research on the effect of slab on the strength
Beam column joints are the most vulnerable regions in and behaviour of HPC beam column joints are limited.
RC moment resisting frames designed to endure severe Considering this gap in the literature, an attempt has been
earthquakes. The current practice for the design of made to investigate the effect of slab on the strength and
connections has evolved from several experimental tests behaviour of HPC exterior beam column joint subjected to
conducted on beam column joints. The first series of tests reverse cyclic loading.
on reinforced concrete beam column joints were conducted
in the early 1960’s by Hanson and Conner (1967) and have Experimental Programme
been used as a benchmark for later studies. In most of the
previous investigations, (Park and Paulay, 1973, Ehsani Materials used
and Alameddine, 1991, Raffaelle and Wight, 1995, Lee et Portland Pozzolana Cement conforming to IS 1489 (Part
al. 2009, Tsonos, 2010, Yen and Chien, 2010, Ganesan et 1):1991, crushed stone aggregate passing through 4.75
al., 2013, Singh et al., 2014, Robert et al. 2014) the test mm IS sieve conforming to grading zone II of IS: 383-
specimens did not include a floor slab. However, in real 1970 (reaffirmed 2002) with fineness modulus 2.92 and
RC buildings, slabs are monolithically cast with the beam specific gravity 2.39 and crushed stone with a maximum
so that the slabs interact structurally with the members size 12.5 mm with specific gravity 2.78 were used for this
framing into a joint. The slab alters various performance investigation. Mineral admixture silica fume supplied by
aspects such as the first crack load, strength, ductility and Elkem Micro Silica were used and a naphthalene based
energy absorption capacity of the joint. superplasticizer Conplast SP430 was used to obtain the
Since 1980’s, some concern has been shown on the effect required workability. HPC of M60 grade is used in this
of the presence of slab on the behaviour of beam column investigation.
joints. Ehsani and Wight,1985 have reported that due to
Details of specimens
the contribution of longitudinal reinforcement of slab, the
flexural strength ratio is significantly reduced. Also, it is The experimental programme consisted of casting and
reported that the presence of slab in the beam column testing of exterior beam-column-joints with and without
connections increased the negative flexural capacity slab subjected to reverse cyclic loading. The mechanical
of beams by 70% (Durrani and Zerbe, 1987). The beam properties of the reinforcements are as given in Table 1.
strength was found to be increased by 16-19% in the The details of reinforcement and the overall dimensions of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

528 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Slab on Strength and Behaviour of Exterior RC Beam Column Joint

exterior beam column joint (BC) and exterior beam column simulated by a steel ball placed between two steel
joint with slab (BCS) specimens are shown in Figure 1. plates provided with spherical grooves. The bottom
of the column was firmly resting on top of the I-beam
Table 1. Mechanical Properties of Steel Reinforcement fixed to the test floor. An axial compressive load of
Diameter of Yield Ultimate Modulus of Elasticity 25% of the axial capacity of the column was applied
bar Strength Strength (N/mm2) on the column by means of a hydraulic jack so that the
(mm) (N/mm2) (N/mm2) specimen was just firmly supported in position without
12 419 580 2.28x105 toppling (Ganesan et al. 2013).
10 426 570 2.32x105
The specimen represents an exterior beam-column-
6 431 660 2.44x105 joint which is isolated at inflection points. A hydraulic
jack of 500 kN capacity which was connected to a 50
kN load cell through a plunger was used to apply the
loads. The load was transferred to the tip of the beam
through the arrangement of two channel sections and
two rods. The speci­mens were loaded up to 2 kN then
unloaded to zero, and then reloaded to 2 kN in the
negative direction and again unloaded to zero, which
was continued till the failure of the specimen. At each
loading cycle the displacements were measured by
using linear variable differential transducers (LVDTs) at
the top and bottom of the beam. The loading sequence
is given in Figure 2. The schematic diagram of the test
setup is shown in Figure 3 and typical test arrangement
of the specimen is shown in Figure 4.

Fig. 1(a): Reinforcement detailing of beam-column-joint

Fig. 2: Loading sequence

Fig. 1(b): Reinforcement detailing of beam-column-joint with slab

Testing
The test setup consisted of a steel loading frame of 300
kN capacity. After 28 days of curing, the specimens
were tested in an upright position in the loading frame.
The top support of the column was a hinged support, Fig. 3: Schematic diagram of test setup

Organised by
India Chapter of American Concrete Institute 529
Session 5 A - Paper 4

Results and Discussion

Load deflection behaviour

Figure 6 shows the typical load displacement plots of BC
and BCS specimens tested subjected to reverse cyclic
loading. For comparison and better representation the
envelopes of the hysteresis of all the specimens plotted
in a single graph, as shown in Figure 7. Envelope curves
were obtained by joining the peak points of all the cycles
(Ganesan et al. 2013). The envelopes curves were used to
obtain the first crack load, energy absorption capacity and
displacement ductility factor for the specimens and are
given in Table 2 and Table 3.

Fig. 4: Typical test arrangement

Behaviour of specimens
Figure 5 shows the crack pattern of the specimens
at failure. In both the specimens, the first crack was
observed in the beam portion. Upon further increase in
loading additional cracks were formed in the beam and
the initial cracks propagated. Finally the cracks widened
leading to the failure of the joint. Since the columns were
designed to be stronger than the beams, most of the Fig. 6:Typical load displacement plots
cracks were concentrated in the beam portion near the
column. In specimen with slab (BCS), joint shear cracks
were observed and the crack from beam propagated to
the slab during the last stages of loading.

Fig. 7: Envelope of load displacement plots

First crack load and ultimate load

Fig. 5: (a) Crack pattern of beam-column-joint specimen (BC) The first crack load, the ultimate load and the displacement
at ultimate load of the specimens are listed in Table 2.
First crack load was determined from the envelope curve
of the load displacement plot corresponding to the point
at which the curve deviated from linearity (Ganesan et al.
2013). From the Table 2 it can be observed that due to the
presence of slab, the first crack load increased by 4.3%,
ultimate load in forward cycle by 38% and reverse cycle
by 67% when compared to the specimen without slab.
This may be because, the slab reinforcement within the
effective width of the slab acts along with the longitudinal
reinforcement of the beam increasing the flexural strength
Fig. 5: (b) Crack pattern of beam-column-slab joint specimen of the beam. The strength increase, however, was smaller
(BCS) in the forward cycle loading (slab in compression) then the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

530 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Slab on Strength and Behaviour of Exterior RC Beam Column Joint

reverse cycle loading (slab in tension) direction due to the summing up the energy dissipated in consecutive load
action of slab as a tension flange of the beam. displacement loops of the specimens throughout the
test. The cumulative energy dissipation of the specimens
Table 2.Test results during each cycle is given in Figure 8. It is evident from the
Designation First crack Ultimate load Displacement
Figure that the presence of slab enhanced the cumulative
of specimen load (kN) (kN) at ultimate load energy dissipation than the specimen without slab.
(mm)
Forward Reverse Forward Reverse
cycle cycle cycle cycle
BC 11.5 14.53 16.16 11.52 13.02
BCS 12.0 20.12 26.97 15.67 20.43

Energy absorption capacity and displacement ductility

The area under the load deflection curve gives the energy
absorption capacity. Energy absorption capacity was
calculated and the values obtained are listed in Table 3.
From the Table 3, it can be seen that due to the contribution
of slab, the energy absorption capacity increases by 294%
in forward cycle and by 313% for reverse loading cycles
when compared to the specimen without slab.
Fig. 8: Cumulative energy dissipation
Ductility of a structure can be defined as its ability to
undergo deformation beyond the initial yield deformation Stiffness degradation
without significant reduction in its load carrying capacity. The application of cyclic or repeated loading on the beam
The ductility factor which is a measure of ductility of a column joint reduces its stiffness significantly. This
structure is defined as the ratio of maximum deflection reduction in stiffness of the specimens can be assessed by
(δu) to the deflection at yield (δy) (Ganesan et al. 2013). The computing the secant stiffness which provides a measure
ductility factors were calculated and the results obtained of the stiffness degradation in the specimens (Ganesan
are given in Table 3. The details of the procedure adopted et al., 2013, Ganesan et al., 2007, Shannag et al. 2005).
are described elsewhere (Ganesan et al., 2007). The The secant stiffness in each cycle was calculated using a
values in Table 3 show that the presence of slab influences line drawn between the maximum positive displacement
the energy absorption capacity and ductility. Compared to point in one half of the cycle and the maximum negative
the specimen without slab, the ductility factor increases displacement point in the other half of the cycle (Shannag
by 1.12 times for specimen with slab. et al. 2005). The stiffness degradation trend for the beam
column joint specimens with and without slab is given in
Table 3. Energy absorption capacity and displacement ductility Figure 9. It is evident that as the number of loading cycles
increases, the stiffness decreases and the specimen
Designation Energy δy δu Displacement
of specimen absorption (mm) (mm) ductility factor without slab (BC) has lower initial stiffness and also,
capacity shows a quick degradation in secant stiffness than the
Forward Reverse Absolute Relative specimen with slab (BCS). It can be observed from Figure
cycle cycle 9 that the initial stiffness increased by 21.4% due to the
BC 0.093 0.135 4.47 13.02 2.91 1 presence of slab.
BCS 0.366 0.558 6.25 20.43 3.27 1.12

Energy dissipation capacity

Energy dissipation capacity is an important indicator
of the seismic properties of a structure. The structures
can withstand strong ground earthquake motions only
if they have sufficient ability to dissipate seismic energy.
This energy dissipation is provided mainly by inelastic
deformations in critical regions of the structural system
and requires adequate ductility of the elements and their
connections (Paulay and Priestly, 1992 and Ganesan et
al. 2013). It was estimated from the area within the load
displacement hysteretic loop for every cycle of load.
The cumulative energy dissipated was calculated by Fig. 9: Stiffness degradation plots

Organised by
India Chapter of American Concrete Institute 531
Session 5 A - Paper 4

Engineers India J., 88: 20–3 .

Conclusions
14. Ganesan N., Indira P.V., Sabeena M.V., 2013. Behaviour of hybrid
The experimental results lead to the following conclusions: fibre reinforced concrete beam-column joints under reverse cyclic
The strength of the exterior beam column joint was found loads. Materials and Design, 54: 686-693.
to increase by 38% and 67% in positive and negative 15. Ganesan N., Indira P.V., Sabeena M.V., 2013.Tension stiffening and
directions of loading respectively due to the presence of cracking of hybrid fiber-reinforced concrete. ACI Materials Journal,
110-66:715-722.
slab.
16. Hanson, N.W., Conner H.W., 1967. Seismic resistance of reinforced
The energy absorption capacity increased by 294% in concrete beam-column joints. proceedings ASCE 93: 533-560.
forward cycle and by 313% for reverse loading cycles 17. Holschemacher K., Mueller T., Ribakov Y., 2010. Effect of steel fibres
when compared to the specimen without slab. on mechanical properties of high strength concrete. Materials and
Design, 31: 2604-2615.
When compared to the specimen without slab, the ductility
18. IS 1489 (Part I): 1991, Portland pozzolona cement specifications, Part
factor increased by 1.12 times for specimen with slab. I, Fly ash based, Bureau of Indian Standards, New Delhi.
The initial stiffness increased by 21.4% due to the presence 19. IS 383: 1970 (reaffirmed 2002), Specification for coarse and fine
of slab. aggregates from natural sources for concrete, Bureau of Indian
Standards, New Delhi.
References
20. Iskhakov I., Ribakov Y., Shah A., 2009. Experimental and theoretical
1. ACI 211.1-91. Standard Practice for Selecting Proportions for investigation of column flat slab joint ductility. Materials and
Normal, Heavyweight and Mass Concrete. Farmington Hills: Design,30: 3158-3164.
American Concrete Institute.
21. Lee J.Y., Kim J.Y., Oh G.J., 2009. Strength deterioration of
2. Ahmed S.M. Uma G. MacRae G.A., 2012. A parametric study of reinforced concrete beam-column joints subjected to cyclic loading.
RC slab in beam column connection under cyclic loading. NZSEE Engineering Structures , 31: 2070-2085.
Conference; 58.
22. Pantazopoulou S.J., French C.W., 2001. Slab participation in practical
3. Aïtcin PC. 1992. High Performance Concrete. E&FN Spon London;
earthquake design of reinforced concrete frames. ACI Structural
1998.A. Eisenberg, Guide to technical editing, Oxford University,
Journal, 98-4: 1-11.
New York.
23. Park, R., Paulay T.,1973. Behaviour of reinforced concrete external
4. Canbolat B.B., Wight J.K., 2008. Experimental investigation on
beam-column joints under cyclic loading. Proceeedings. Fifth World
seismic behaviour of eccentric reinforced concrete beam-column-
Conference on Earthquake Engineering, 88: 772-781.
slab connections. ACI Structural Journal, 105-2:154-162.
24. Paulay T., Priestley M.J.N., 1992. Seismic design of reinforced
5. Durrani A.J. and Wight J.K.,1987. Earthquake Resistance of
concrete and masonry buildings. NewYork: John Wiley & Sons.
Connections Including Slabs. ACI Structural Journal, 85-5:400-406.
25. Raffaelle G.S., Wight J.K., 1995. Reinforced concrete eccentric beam
6. Durrani A.J. and Zerbe H.E., 1987. Seismic Resistance of R/C
column connections subjected to earthquake-type loading. ACI
Exterior Connections with Floor Slab. Journal of Structural
Engineering, 113:1850-1864. Structural Journal, 92(1): 45-55.

7. Ehsani M., Wight J.K., 1985. Effect of transverse beams and slab on 26. Roberto R., Napoli A ., Pinilla JGR., 2014. Cyclic behavior of RC
beam-to-column connections. ACI Journal, 82-2: 188-195. beam-column joints strengthened with FRP systems Constr Build
Mater, 54: 282–297.
8. Ehsani M.R., Alameddine F., 1991. Design recommendations for type
2 high- strength reinforced concrete connections. ACI Structural 27. Shah S.A.A., Ribakov Y., 2005. Experimental and analytical study
Journal, 88(3):277-291. of flat-plate floor confinement. Materials and Design, 26: 655-669.

9. Elahi A., Basheer P.A.M., Nanukuttan S.V., Khan Q.U.Z., 2010. 28. Shannag M., Abu-Dyya N., Abu-Farsakh G, 2005. Lateral load
Mechanical and durability properties of high performance concrete response of high performance fibre reinforced concrete beam
containing supplementary cementitious materials. Construction and column joints. Constr Build Mater, 19:500–8.
Building Materials, 24:292–299. 29. Shin M., LaFave J.M., 2004. Seismic performance of reinforced
10. Fenwick R., Bull D.K., Macpherson and Lindsay R., 2006. The concrete eccentric beam-column connections with floor slabs. ACI
influence of diaphragms on strength of beam. NZSEE Conference: Structural Journal, 101-3: 403-412.
paper no. 21 30. Singh V., Bansal P.P., Kumar M., Kaushik S.K., 2014. Experimental
11. Ganesan N., Bharati Raj, Shashikala A. P., 2013. Behavior of studies on strength and ductility of CFRP jacketed reinforced
self-consolidating rubberized concrete beam-column joints. ACI concrete beam-column joints. Constr Build Mater, 55: 194–201.
Materials Journal, 110-64: 697-704. 31. Tsonos A.D.G., 2010. Performance enhancement of R/C building
12. Ganesan N., Indira P.V., Ruby A., 2007. Steel fibre reinforced high columns and beam–column joints through shotcrete jacketing.
performance concrete beam-column joints subjected to cyclic Engineering Structures ,32:726–740.
loading. ISET Journal of Earthquake Technology, 44: 445–56. 32. Yen J.Y.R., Chien H.K., 2010. Steel plates rehabilitated RC beam–
13. Ganesan N., Indira P.V., Ruby A., 2007.Behaviour of reinforced column joints subjected to vertical cyclic loads. Constr Build Mater,
high performance concrete members under flexure. Institution 332–339.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

532 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Slab on Strength and Behaviour of Exterior RC Beam Column Joint

Dr. N. Ganesan
Dr. N. Ganesan holds an M.E. and PhD from IISc., Bangalore. He was formerly the Dean (Planning &
Development) and currently Professor of Civil Engineering at the National Institute of Technology, Calicut,
India. His research interest includes reinforced concrete, ferrocement, fibre reinforced concrete, polymer
modified concrete and self compacting concrete. He is a fellow of The Institution of Engineers, India and
IFIC consultant. He was a visiting professor at the Asian Institute of Technology, Bangkok and King Khalid
University, Kingdom of Saudi Arabia. He had visited University of Dundee, Scotland, Queens University,
Belfast, National University of Singapore, University of Stuttgart, Germany and University of Michigan, USA.

Nidhi M.
Nidhi M. received her B.Tech (Civil Engineering) from Calicut University, Kerala and M.Tech (Structural
Engineering) from College of Engineering Trivandrum, Kerala. Her area of research includes fibre
reinforced concrete, high performance concrete and beam-column-slab joints. At present she is a PhD
scholar in NIT Calicut, Kerala.

Dr. P.V Indira

Dr. P.V Indira holds an M.Tech from IIT Madras and PhD degree from University of Calicut. She is a Professor
of Civil Engineering at the National Institute of Technology, Calicut. Her research interest includes
reinforced concrete, fibre reinforced concrete, polymer modified concrete and self compacting concrete.

Organised by
India Chapter of American Concrete Institute 533
SESSION 5 B
Session 5 B - Paper 1

Nano Technology of Tomorrow Made Useful Today - Effective and Cost

Effective Stabilization of Various Soils by Using Subnano Molecules
Sourabh Manjrekar and Ishita Manjrekar
Sunanda Speciality Coatings Pvt. Ltd.

Abstract foundations due to the continuous erosion/damage by

the Yamuna river water. Delhi’s floodplains are located
Rains, floods, landslides and other natural disasters
in seismic zone IV, the second highest earthquake hazard
result in the loss of lives and property. Buildable stable
zone in India. Even medium intensity tremors can lead
land with natural load bearing capacity is scarce in
to liquefaction of soil, a condition resembling quicksand,
most cities and construction on poor soils results in
causing the sinking of all the structures resting on water-
failures and devastation of structures. Soil stabilization
saturated grounds along the Yamuna river.
is the process of maximizing the suitability of soil for
construction purposes. Chemical Grouting is the process To solve this problem, in many cities across the world,
of injecting a solution or a mix of solutions into the soil the idea of freeing rivers is gaining ground. In the US,
for in situ soil stabilization. It improves stability, strength, New Orleans is breaking its floodwalls to allow the storm
compressibility, permeability, durability and load carrying water to come into the city by building new canals and
capacities of the soil. ponds. (Figure 1).
By adopting appropriate chemical engineering and
chemistry, grouts with nano particle sizes in the range of
5 to 10 nanometers have been prepared in our laboratory.
These nano particles behave as a fluid and penetrate deep
into the soil and then later react to form a solid, semisolid
or a gel at a pre- determined time. This gel is a precipitated
binder of high quality at lower concentrations of reaction
components. These nano particles are amorphous and
since they have higher specific areas (120 – 260 m2/g) Fig. 1: Building of new canals in the city of New Orleans
they distribute themselves evenly and bind the entire soil
matrix. Laboratory tests of various combinations of the
After building dykes to keep water from entering
grout have been done, followed by field tests to determine
the country for 800 years, the Netherlands is now
the best engineering properties for soil stabilization.
implementing the ‘Room for the River’ project. (Figure 2)
These nano particle grouts enhance the stability of the
soils. This saves land, structures and human lives as well.
Keywords: Soil Stablization, water control, tunneling, soil
bearing capacity, nanotechnology, nano grouts.

Introduction
In July 2014, the collapse of a building in Chennai raised
suspicions about the soil conditions not being favourable
for a high-rise building as the site where the building was
coming up was a huge wetland and water catchment area
for the Porur Lake until a few years ago.
In Delhi, the Yamuna floodplains and wetlands are home to
almost one-fifth of the city’s population. Structures (legal Fig. 2: Steps taken in the Netherlands for the ‘Room for the
and illegal) in these high-density population areas (from River’ project
northeast Delhi to Noida, Okhla, Badarpur and Faridabad The Netherlands government also demolished numerous
in Haryana) are highly vulnerable because they have been houses and shops to make room for their rivers. (Figure 3).
built on soft alluvial soil. These buildings virtually float
on a high groundwater table that keeps weakening their However, demolishing houses or relocating a fifth of its
population is not an option for Delhi. Therefore, effective

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

536 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Nano Technology of Tomorrow Made Useful Today - Effective and Cost Effective Stabilization of Various Soils by Using Subnano Molecules

Fig. 3: Demolition and relocation of shops and houses in the Fig. 4: Particle Sizes of Standard cement and ultrafine cement
Netherlands to make room for their rivers grouts

stabilization of the suspect soil can be a very good solution. grouts could not percolate into these pores. Cements
Soil Stabilization is extremely important in enhancing typically have a particle size of 15 microns or 15,000
the engineering properties like hydraulic conductivity, nanometers and microfine/ultrafine cements can have
compressibility, strength, and density. particle sizes in the range of 3 to 7 microns or 3000 to
7000 nanometers. (Figure 4)
Soil Stabilization can also be very useful for prevention
of the collapse of soil while carrying out tunneling Hence, Nanotechnology can be a novel approach to
applications, increasing the soil bearing capacity and conduct soil manipulation at the atomic or molecular
reducing settlement in foundations of existing buildings.(1) scale, which is facilitated by introducing the concept of
nano particles as an external factor to soil. These nano
Conventionally stabilization of soil is done using lime,
particles have a particle size of 5 to 10 nanometers which
cement or micronized cements as grouts; however, these
is almost 3000 times smaller than cement.
methods are inefficient and give only partial results.
This paper highlights the current research at Sunanda In chemical grouting two chemically reactive solutions
Speciality Coatings Pvt. Ltd. for the development of nano are mixed together to react and form a precipitate after a
grouts to increase the efficiency and performance of soil predetermined time which fills the voids/pores in the soil,
stabilization operations.(2) consolidates the soil making it more dense which in turn
increases its bearing capacity and reduces water flow.(3,4)
Conventional Methods of Soil Stabilization This reaction is analogous to the formation of natural
Suspect soil is conventionally stabilized by methods such sandstone, which is formed by gluing together loose sand.
as: The subsequent important geotechnical properties of soil
can be determined in the laboratory after completion of
ll Replacement of the entire treacherous soil with good the grouting process. Grouted samples can be subjected
quality soil imported from elsewhere to alternate freeze-thaw and wet-dry cycles to determine
ll Compaction of the soil using mechanical means their durability properties. The chemical compositions
and microstructure of soils can be analysed using x-ray
ll Using cement and / or lime
diffraction and scanning electron microscopes.
ll Using cement grouts
However, nano grouts have the limitation in that they
ll Using micronized cementitous grouts are often more expensive than particulate grouts. To
All of the above measures though useful are seen to compensate for this, large voids are typically grouted with
give partial results. Also, they are labour intensive and cementitious grout first and then chemical grouting is
very time consuming processes. In cementitious / lime done as needed.
or mixed grouts, the suspended particulate matter i.e.
cementitious matter has a well-defined size and as a Case Studies
result cannot penetrate deep enough to reach till the last
void. Most times this cementitious material simply goes Case Study 1: Increasing the soil bearing capacity
as a filler material (wherever it reaches). In addition it has (by using SUNGEOGROUT – nano grout) of an existing
limited binding capacity due to overhydration. (3) five star hotel in Florida, USA in order to construct 2
additional floors
Chemical Grouts This was a case of an existing 5 star hotel in Florida, USA
Chemicals grouts were developed as pore sizes in soils where the client wanted to construct 2 additional floors.
were very small and as a result conventional cement The Soil composed of 62% sand, 16% silt and 22% clay

Organised by
India Chapter of American Concrete Institute 537
Session 5 B - Paper 1

and had a soil bearing capacity of 14.16 T/m2. In order to Case Study 2: Prevention of collapse of sandy strata
construct 2 additional floors, the required soil bearing while constructing an underground tunnel for a metro
capacity needed to be 19.63 T/m2. railway project in India
This was achieved by injecting an engineered liquid nano The tunnel was to be constructed 24 m below the street
grout (SUNGEOGROUT) through vertical and inclined level (Figure 8). The soil was composed of 80% sand,
injection ports under each of the foundations (Figure 5 & 15% silt and 5% clay. There was a tremendous amount
6). The soil bearing capacity was checked after 72 hours of ground water pressure due to which the sandy strata
and 14 days respectively where it showed an increase of would collapse making excavation very difficult (Figure 9).
38 %, reaching the required soil bearing capacity of 19.63
T/m2. (Figure 7)

Fig. 8: Location of tunnel 24 m below street level

Fig. 5: Chemical Grouting Injection Points under each foundation

Fig. 9: Sandy strata highly saturated with water

Fig. 6: Vertical & Inclined grouting detail under each foundation

Fig. 10: Ease of excavation due to grouting (pink in colour) car-

ried out prior to excavation

Cement grouting and micro-fine cement grouting was

not very useful as it could not percolate into the pores of
Fig. 7: Increase in Unconfined Compressive Strength Results the sandy strata. In this case a nano grout was used for
by using the nano grout water control and was successful as it could percolate

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

538 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Nano Technology of Tomorrow Made Useful Today - Effective and Cost Effective Stabilization of Various Soils by Using Subnano Molecules

much deeper than conventional cementitious grouts,

after which it would solidify to fill the pores and arrest
the flow of ground water which made excavation a lot
more manageable and efficient saving a lot of time and
money for the contracting company (Figure 10). The nano
grout was grouted from the top street level via numerous
grouting ports using Tube-à-Manchette (TAM) grouting.
(Figure 11)

Fig. 13: Typical installation of micro piles

Fig. 11: A nano grout was grouted from the top (street level)
via numerous grouting ports using Tube-à-Manchette (TAM)
grouting
Fig. 14: Typical structural connection of micro piles to foundations
Case 3: Settlement of a ‘side arm charger in a coal
handling plant’ in INDIA Case Study 4: Stabilizing the soil along the boundary
This was a case of the settlement of foundations of a coal walls of a plot in North India, during the construction
handling plant under construction due to which further of the retaining wall
construction was put on hold (Figure 12). The remedial This was a case where the sandy strata along the
options under consideration were: boundary walls of a plot was stabilized in order to hold
i) Installation of vertical and inclined micro piles to the soil in place and prevent collapse of the soil during the
transfer the load from the foundations to micropiles construction of the retaining wall. (Figure 15)
by structurally connecting them with the pile caps.
(Figure 13 & 14)
ii) Demolishing and rebuilding the partially completed
foundation.
iii) Sungeogrout - Chemical grout engineered with nano
particles.
Option i & ii were very expensive and time-consuming,
hence rejected. Sungeogrout (nano grout) was the solution
selected.

Fig. 15: Collapse of soil during construction of retaining wall

Criteria While Selecting a Soil Stablisation

Grout
1. Nature of the soil to be treated
2. Initial grout viscosity and the viscosity profile during
gelation
3. Setting time of the grout
4. Particle size (Use of the grout):
Fig. 12: Typical settlement patterns

Organised by
India Chapter of American Concrete Institute 539
Session 5 B - Paper 1

Viscosity
i. Reducing water ingress
ii. Short term or long term consolidation
iii. Increasing of Soil Bearing Capacity
Viscosity is an important parameter of successful
penetration of the grout and among many other factors,
viscosity is dependent on the particle / molecule size of
the reactive chemicals as well as that of the vehicle. Hence
lesser the molecule size of the grout, the more penetrative
it will be and hence more effective. (Table 1)

Table 1
Relationship between Viscosity and Hydraulic Conductivity

Viscosity Hydraulic Conductivity Relationship Fig. 17: Variation in setting time of the grout at constant tem-
perature of 20° C due to change in activating nano particle

 Less than 2 cP 10-4 cm/sec

two chemicals is the particle or molecular dimensions.
Thus molecular size alongwith optimum concentration
10 cP Greater than 10-3 cm/sec become two important factors in the success of chemical
grouting of the soil. The nano level dimensions of the
5cP Above 10-2 cm/sec
molecules in grout can be achieved by proper selection
of components of the grout. By adopting appropriate
chemical engineering and chemistry, it is possible to reach
Setting Time in the Nano range where the average particle size of the
The ‘working time’ or the ‘gel time’ of the grout can be grout fluid will be in the range of 5 to 10 nanometers. Such
engineered based on the type of application as per the chemical reactions which are made to happen at the nano
need of the job. The grout setting time depends on: level can give a precipitated binder product of high quality
at lower concentrations of reaction components. With the
ll Nature of grout right chemistry, these precipitated binders can be made of
ll Grout and soil temperature - the higher the amorphous character and have higher specific areas (120-
temperature, the shorter the setting time (Figure 16) 260 m2/g) so that they can evenly distribute themselves
while binding the soil matrix.(5)
ll Type of activating nano molecules (Figure 17)
ll Nature of the soil
Conclusions
Particle Size i) Chemical grouting is an established effective technique
of soil stabilization
One of the important factors in the forward reaction of any
ii) Nano particles of the grout improve the penetration
capability of the grout
iii) It is possible to explore various combinations of
chemical grout reactants to enter into micron, nano
and subnano zone.
i) Entering in nano zone may enhance the performance
of the grout in terms of unconfined compressive
strength and economics. As a result chemical grouts
can be injected into soils containing voids that are too
small to be penetrated by cementitious or other grouts
containing suspended solid particles.
ii) Chemical Grouts once reacted become a material
analogous to sandstone and hence have no detrimental
effect on the soil and ground water.
Fig. 16: Temperature versus Setting time of the grout

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

540 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Nano Technology of Tomorrow Made Useful Today - Effective and Cost Effective Stabilization of Various Soils by Using Subnano Molecules

iii) Chemical grouts have an adaptability over a wide range and Geosynthetics, American Society of Civil Engineers, Geotechnical
Special Publication 30(1), 725-736.
of applications
3. Siwula and Krizek 1992. Siwula, J. M., and Krizak, R. J. 1992.
iv) Nano grouts are the strongest and least toxic of the "Permanence of Grouted Sand Exposed to Various Water
existing range of chemical grouts Chemistries," Grouting, Soil Improvement and Geosynthetics,
American Society of Civil Engineers, Geotechnical Special Publi-
cation 30(1), 1403-1419.
References
4. Mori, Tamura, and Fuki 1990. Mori, A., Tamura, M., and Fuki, Y. 1990.
1. Polivka, Witte, and Gnaedinger 1957 Polivka, M., Witte, L. P., and
"Fracturing Pressure of Soil Ground by Viscous Materials," Soils
Gnaedinger, J. P. 1957. "Field Experiences with Chemical Grouting,"
and Foundations 30, 129-136.
American Society of Civil Engineers, Soil Mechanics and Foun-
dations Division Journal 83 (SM2), Paper 1204, 1-31. 5. Mori, et al. 1992. Mori, A., Tamura, M., Shibata, H., and Hayashi,
H. 1992. "Some Factors Related to Injected Shape in Grouting,"
2. Yonekura and Kaga 1992 Yonekura, R., and Kaga, M. 1992. "Current
Grouting, Soil Improvement and Geosynthetic, American Society
Chemical Grout Engineering in Japan," Grouting, Soil Improvement
of Civil Engineers, Geotechnical Special Publication 30(1), 313-324.

Mr. Sourabh Manjrekar

Sourabh Manjrekar is Director at Sunanda Speciality Coatings Pvt. Ltd and oversees Sunanda’s
International Operations, with a specific focus on developing sustainable solutions for large
industrial Oil & Gas projects in the middle-east and Africa.
Sourabh serves on the Board of Direction of the India Chapter of American Concrete Institute.
Sourabh has collaboratively worked with some of the most reputed global structural engineering
firms to provide some very innovative materials solutions for projects of international repute,
including:
i) Hotel Cabana, Florida, USA
ii) Jumeirah Lake Towers, Dubai, UAE
iii) Al Yaquob Tower, Dubai, UAE (World’s tallest clock tower)
iv) Palais Royale, Mumbai (India’s tallest tower)
vi) Royal Residency, Tanzania
Sourabh has completed BS from Illinois Institute of Technology, Chicago, USA and his MBA from
S.P.Jain Institute of Management & Research, Mumbai.

Ms. Ishita Manjrekar

Ishita Manjrekar is a Director at Sunanda Speciality Coatings Pvt. Ltd., and oversees Sunanda’ s R
& D, with a specific focus on developing and marketing Sunanda’s line of sustainable construction
chemicals.
In this role, Ishita draws on 4 years of rich experience in sustainability at Primary Global Res
earch in San Francisco and New York. While at Primary Global Research Ishita led the “Cleantech
and Green Technologies” business unit. She has been invited to feature on Bloomberg TV as well
as Bloomberg Radio numerous times as a subject matter expert on sustainability and green
technologies. Ishita expertise has
also been sought by multiple print media including Reuters, Financial Times, Bloomberg, Forb
es, BBC News and Marketwatch in USA.
Ishita is member of ACI International and works actively on various international board appointed
as well as technical committees. She currently serves on the ACI International Board Committees
for International Advisory Committee, ACI Membership Committee, ACI Marketing Committee,
International Project Awards Committee, IPAC Judging Subcommittee and Student & Young
Professionals Activities Committee.
Ishita server as Honorary Secretary on the Board of Direction of the India Chapter of American
Concrete Institute.

Organised by
India Chapter of American Concrete Institute 541
Session 5 B - Paper 2

Precast Industry Contribution toward Green Construction

Dr. Ekasit Limsuwan
Professor Emeritus Chulalongkorn University, Bangkok Thailand

Abstract Some research work has conducted a life cycle

assessment of an office building and found that the
Current construction industries have developed toward
operational phase of the building accounted for 52% of
sustainable development as which the technologies are
global warming potential, 71% of photo- oxidant potential,
intended to optimum use of natural resources, minimize
and 66% of total acidification potential. The report also
energy consumption, reduce waste and encourage possible
showed that for the embodied impacts concrete and steel
recycle. Precast construction would be an alternative for
were responsible for 74% and 24% of global warming
civil infrastructure projects. This paper will introduce the
potential, 30% and 41% of photo-oxidant formation
building process for project execution in planning, design,
potential, and 42% and 37% of acidification potential,
construction, operation and maintenance. Some examples
respectively. A similar study in Japan, the carbon dioxide
of precast construction of elevated highways and high rise
emission in building stage by construction materials as
buildings in Thailand will be presented. Some aspects
shown in Figure 1 to indicate at about the same amount.
toward green technologies to demonstrate improvement
The carbon dioxide discharge in construction stage is
in quality control to accelerate speed of construction, to
said to be about half of the practical use stage, anyway
reduce energy consumption and to minimize environment
in concerning with this subject, quantitative investigation
impact assessment will be presented. The life cycle
and the presentation of definite scenario is yet to be done.
management will also be synthesized for green rating as
A systematic approach for carbon dioxide reduction
far as the low carbon strategy can be implemented toward
can be done through sustainable engineering of which
sustainable engineering.
the concept of sustainability focused on ecological
issues environmental degradation dangerous depletion
Introduction of resources and in conflict with economic growth.
It is apparently evident that global warming have been However the new concept of sustainability shall be as
induced from accumulative discharge of carbon dioxide the interaction of environmental, economic and social
into the environment . Then it has deteriorated ozone domains with a positive outcome for all three as shown
layer of atmosphere by means of green house effect. in Figure 2. Measurement of sustainability objectives
Low Carbon Model (LCM) have been studied by various
organizations to quantify investigations in order to reduce
carbon dioxide emission in practical use stage of buildings
with definite scenarios showing the possibility to actualize
of large volume reduction in 2050. Green concept is
another approach for carbon dioxide reduction for
buildings, industries, and town or city. For buildings and
industries, emphasis should be laid on energy sources
of the materials process, out-come products; while
town or city should have concentrated on city planning,
environmental planning, traffic and transportation
system, energy areas and management. Balancing the
variety of matrix and goals across the three spectra
requires a departure from purely quantitative metrics as
used in life cycle assessments, then green construction
concept would take the major role for newly construction
buildings this paper has introduced the criteria of green
construction on planning phase, in execution phase and
in-use phase during operation and maintenance. Fig. 1: Carbon dioxide emission in building stage by construction
materials[1]

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

542 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Precast Industry Contribution toward Green Construction

Fig. 2: Inter-relation of sustainability low carbon concept and

green construction
Fig. 3: Green construction concept
within the construction industry is increasingly achieved
through green rating system. The system has offered
emphasis their planning toward sustainable design
a framework to evaluate built structures based on
concepts and strategic engagement with engineer
numerous environmental, social, and economic criteria.
architect consultants for design, construction, and
It is often said that buildings are responsible for carbon supervision. In execution phase ; the green building design
dioxide emission to about half of the total emission can be an environment friendly approach for design work,
of most developed countries. A holistic view of the tender document, specification and costing in tender bid.
emission from the whole life cycle of building process Green construction in this phase should have concentrated
has shown significant attribution along the building in construction materials, manufacturing, equipments,
process of planning, design, construction, operation and construction techniques and some other associated
maintenance. However, the prefabricated construction facilities for the design. For construction, the green building
especially concentrated at precast industries should have index (GBI) and the environment management system
been remarkably significant for carbon dioxide reduction (EMS) should be considered along the process. Some
and greatly contribute toward green construction. alternatives for materials, manufacturing, equipment and
This paper will give some implementation examples to the techniques may be changed for better performance.
demonstrate the environmental friendly approach for On-site environmental management practices, would
green construction in Thailand emphasis on the precast also be very important index for green construction as
industries. site plan, on-site housekeeping, waste management, dust
and mud control, noise and vibration control, workforce
management, energy efficiency, and water efficiency.
Green Construction Project environment plan and project monitoring have to
Green construction may involved materials, be carried out during construction prior to the completion
manufacturing, distribution, equipment, techniques, and and handover sequence. In operation and maintenance
miscellaneous. However, it would require green technology (in-use) phase ; the green rating would have been as the
in development and application of products, equipment monitoring sequence to conceived some key performance
and system used to conserve the natural environment of building management, water and energy conservation,
and resources, which minimize and reduces the negative indoor environmental quality and some environmental
impact of human activities. The green construction then protections. It should be noticed that the good planning
should be satisfied to the following criteria : would result on satisfactory conditions of which green
ll Minimize the degradation of the environment construction have served the sustainable development as
the whole.
ll Zero or low greenhouse gas emission (safe for use,
promote healthy and improve environment)
Precast Industry
ll Conserve the use of energy and natural resource
Prefabricated construction becomes a more important
ll Promote the use of renewable resources and reduce factor in building industry every day. With cost of labor
waste disposal and materials constantly rising, precast construction can
ll Energy efficient, cost effective and low maintenance probably are considered an evolutionary development
especially toward green construction and sustainable
Green construction may be categorized into 3 major development. The method requires less forming and
groups of buildings of residential, commercial or industrial placing material than conventional building methods. It
and infra-structure. The process will also concern with permits strong contemporary architectural expression
planning, execution and in-use as shown in Figure 3. In and is becoming more versatile in this area. The most
planning phase; some legal requirements of Environmental important tasks for precast industry go into planning,
Impact Assessment (EIA) and Environmental Management close co-ordination and timing in advance. Actual on-
Plan, (EMP) have to be carried out for approval by the site construction proceeds with remarkable speed and
Environmental Commission. Then the developer or the efficiency. Achievement of precast construction shall be
owner by means of the project management should filling as provision for

Organised by
India Chapter of American Concrete Institute 543
Session 5 B - Paper 2

ll Sound structures
ll Ease construction
ll Fast operation
ll Economic execution
ll Quality assurance and high performance construction
ll Environmental friendly approach
ll Development toward green construction for
sustainability
In dealing with precast industry, key factors should have
been considered not only for technological development Fig. 4: Precast and Prestressed Structure
but also some professional practices such as :
ll Loads, weigh, actions to the structural behaviors for strength, serviceability
and durability. The key performance indicator shall be
ll Structural system monitored and accomudated the design service life. The
ll Formwork technology composite members of high performance concrete or
structural steel can be advantages in combination actions
ll Concrete technology
of high strengths in compression, tension, shear, bond and
ll Prefabrication, lifting, handling, transportation some others. Some other means of composite actions of
high performance concrete with prestressing tendons
ll Construction sequences and erections
or fiber reinforcement may be categorized as mixed
ll Structural performance in strength, serviceability, and structures of thus specific application and innovation to
durability support the concept of green construction.
Even now, code of practice in this area may not applicable
in all cases. But best practice learned from past Implementation Example
experiences and some execution or operation skills can Some mega-projects of precast industry for building
be leaded toward green construction and as the whole of construction and infra-structure projects. There is no
sustainable development. detail for analytical evaluation of green rating and further
The high strength and high performance concrete can study still be made especially during operation and
contribute a great deal for precast construction of having maintenance thru-out their service life of the projects.
remarkably perceived by ready mixed industry as which However, the planning phase have been done for project
they can supply for precast or prefabricated industry financing and budgeting approval. The execution phase
as well as the prestressed concrete especially for for design and construction should be done along with
segmental construction with post-tensioning on the job the bidding and evaluation process which should be made
site. Concrete strength of these types normally be higher corresponding to the same green rating. Anyway the
than 30 Mpa at 18 – 24 hrs. of age, and the characteristic official records still not be in the same criteria, and still
strength would be higher than 60 Mpa. In many cases, the be anchoraged to integrate in the system for complete life
characteristic strengths would be expected at 90 – 100 cycle management.
Mpa. However, initial strength at stripping the formwork I. Building Construction Projects. Some examples as
or prior to stressing sequence should be at least 50% of shown in Figure 5 always involved large building
those specified strengths. Constructivity of prefabricated and highrise building where structural system and
components or precast members may concentrate typical components can be implied to utilize precast
flowability in placing, but high early strength may be construction to the most benefit. The projects should
requiring to enhance strengths and stiffness for removing be concerned with public building, highrise building,
shutters, lifting and handling prior to the erection and and industrial building.
fabrication. Key parameters of thus concrete such as
II. Infrastructure Projects. Precast construction in mega
flowability, stripping strength, dimension stability and
projects of civil infra-structure as shown in Figure 6
surface texture can take an important role for precast
may have been applied for bridge or elevated highways,
member.
tunneling construction and mass rapid transit system.
The prestressing beams and girders can be significantly
These projects can be categorized to contribute
improved by means of strengths and stiffnesses.
sustainable engineering and green construction as
Pretensioning or postensioning precast components
which life cycle assessment and public utilization can be
have been developed to accomplish the most benefit
achieved.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

544 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Precast Industry Contribution toward Green Construction

(a) Public Building (a) Bridges and Via-duct

(b) Highrise Building (b) tunneling Construction

(c) Industrial Building (c) Mass Rapid Transit System

Fig. 5: Building Construction Projects Fig. 6: Infrastructure Projects

Conclusion 2) Green construction in Thailand is still pre-mature for

construction industry. It requires a national agenda to
In accordance with developments in precast industry, the
advocate for need of changing laws and regulations related
contribution toward green construction can be concluded
to energy conservation and environmental management
as follow :
for residential buildings, industrial buildings, and related
1) Precast construction has been exercised in to planning for infra-structure projects..
construction industry in Thailand due to benefit of
3) In execution and in-use phases are on voluntary
time saving, and economic point of views. It required
basis, then the quantitative investigations of carbon
sustainability consciousness, and also requires to take
dioxide reduction can not be estimated along life cycle
into account all individual concerns.
assessment.

Organised by
India Chapter of American Concrete Institute 545
Session 5 B - Paper 2

4) Green construction by means of green technology IABSE Symposium, Bangkok 2009.

in construction stage concentrated on materials, 2. John Anderson, “Measuring Sustainability and Life-cycle
manufacturing, equipments, and techniques as which Assessment”, IABSE Working Commission 7 – Sustainable
Engineering, Structural Engineering Documents – Design for
monitoring and supervising to be assessed. Sustainability.
5) Professional institution may need for development of 3. Ekasit Limsuwan, “Integration Concept of Sustainable Engineering”,
policy by sharing best practice, advising on effective IABSE Working Commission 7 – Sustainable Engineering, Structural
and risk of putative plans on green construction. Engineering Documents – Design for Sustainability.
4. Kofoworola OF, Gheewala SH., “Environmental Life Cycle
Assessment of a Commerical Office Building in Thailand”,
Reference International Journal of Life Cycle Assessment. 2008: p. 498-511.
1. Tatsuo Inada., “A Study concerning the way building construction
should be in the era of Low Carbon Emission”, Sustainable 5. TGBI TREES – NC, “Thai’s Rating of Energy and Environmental
Infrastructure Environment Friendly, Safe and Resources Efficient, Sustainability for New Construction and Major Renovation” January
2010.

Dr. Ekasit Limsuwan

Dr. Ekasit Limsuwan is a Professor Emeritus of Chulalongkorn University, Thailand. He is a registered
professional engineer in Texas, USA. His area of specialization include Structural Concrete including
reinforced concrete structures, prestressed concrete structures, partially and fully prestressed concrete
structures; Structural Reliability including structural evaluation, and structural rehabilitation;
Construction Materials / Techniques including high strength concrete, high performance concrete and
precast construction; Fire Endurance, Fire Safety and Formwork Technology. Dr. Limsuwan has many years
of industry experience in Thailand, USA, and ASEAN Community before he has joined the Chulalongkorn
University since 1979. He is an active member of national and international professional organizations
including the Engineering Institute of Thailand (EIT),Thailand; International, Association for Bridge and
Structural Engineer, (IABSE), Switzerland; Federation Internationale du Beton (gib); American Concrete
Institute (ACI), USA.; Prestressed Concrete Institute (PCI), USA, International Committee on Concrete Model
Code (ICCMC), and Asia Concrete Federation (ACF).
Dr. Limsuwan has many consultancy projects in Thailand as structural designer and engineer of large
construction projects. He was affiliated with construction projects in Thailand in various capacities
including as project manager. Recently he is a JICA-Expert for the Mega-Project Construction Supervision
for ASEAN-Engineers, USIAD-Experts in Rural Industrial Promotion (Concrete Products) for Thai Industrial
Ministry and UNDP/TCDC-Specialist in the Workshop on High Performance Concrete for Asia, in Madras,
India. He obtained his Bachelor in Engineering from Chulalongkorn University, Bangkok, Thailand, his MSCE
from University of Houston, Houston, Texas and his Ph.D. (Structural Engineering), from University of Texas,
Austin. He had been appointed ad a member of Professional Engineer Board since 1992 until the Council of
Engineers have been setting up under the Engineer Act. B.E. 2542 (1999). Dr.Limsuwan has been elected
as member of the Council of Engineers for 2 terms and serve his position as Secretary General. Now he
is being active as member of International Committee, and also served as secretarite of the committee as
well.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

546 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of Steel Reinforcement in Chloride and Combined Chloride-Sulfate Contaminated Concrete Powder Solution Extracts

Behaviour of Steel Reinforcement in Chloride and Combined Chloride-

Sulfate Contaminated Concrete Powder Solution Extracts

Fouzia Shaheena Bulu Pradhan

Research Scholar, Department of Civil Associate Professor, department of Civil
Engineering, Indian Institute of Technology Engineering, Indian Institute of Technology
Guwahati, Guwahati – 781039. Guwahati, Guwahati – 781039.

Abstract or by internal agents within the concrete. There is a need

to consider all potential deterioration mechanisms at the
Degradation of reinforced concrete structures due to
design stage in order to select and specify an appropriate
corrosion of reinforcing steel is one of the major durability
concrete mixture from a durability perspective[2].
problems. This paper reports the influence of chloride
Concrete normally provides excellent resistance to
and sulfate ions on steel reinforcement corrosion. In
corrosion of steel reinforcement. The protective action
the present work, concrete mixes were admixed with
of concrete is mainly ascribed to its high alkalinity (pH of
sodium chloride and magnesium sulfate. The admixed
about 13) provided by calcium hydroxide that is formed
concentrations of sodium chloride were 3% and 5%,
during hydration of Portland cement and this high
whereas those of magnesium sulfate were 3%, 6% and
alkalinity passivates the steel by forming a chemically and
12%. Thermex TMT steel, ordinary Portland cement (OPC)
electrically inactive layer (passive film) of ferric oxide[3].
and w/c ratio of 0.5 were used in the present experimental
However, if the concrete is permeable to the extent that
investigation. Concrete powder was obtained by crushing
carbonation reaches the concrete in contact with steel
concrete specimens those were admixed with different
and lowers the pH to about 9 or soluble chlorides can
concentrations of chloride and sulfate salts. The behaviour
penetrate right up to the steel reinforcement, and water
of steel reinforcement was studied by conducting
and oxygen are present, this passivity gets disrupted and
potentiodynamic polarization test and linear polarization
corrosion initiation of steel reinforcement takes place[1].
resistance (LPR) test in concrete powder solution extracts.
Corrosion of steel reinforcement is the major cause of
On the basis of the results obtained, ranges of potential
premature degradation of concrete structures and has
for different zones of corrosion of steel reinforcement
become a serious durability problem throughout the
in concrete powder solution extracts contaminated
world[4]. The accumulation of corrosion products (oxides/
with chloride and composite chloride-sulfate ions have
hydroxides) in the concrete pore space near the steel rebar
been obtained. From the results it is observed that, the
can build up hoop stresses around the rebar and results in
half-cell potential values of steel reinforcement were
cracking and spalling of concrete, which in turn facilitates
more negative than -270 mV (SCE) in contaminated (with
the ingress of moisture and oxygen, and aggressive
chloride and composite chloride-sulfate ions) concrete
ions to the rebar and accelerates the corrosion of steel
powder solution extracts. The corrosion current density
reinforcement[5]. This results in loss of bond at steel/
of steel reinforcement increased with increase in chloride
concrete interface and thus reduces the serviceability of
ion and sulfate ion concentration in the concrete powder
reinforced concrete structures[6].
solution extracts. It is also observed that the corrosion
current density of steel reinforcement in concrete powder The deterioration of reinforced concrete structures
solution extracts contaminated with sodium chloride plus exposed to marine environment and located in
magnesium sulfate was more as compared to that in contaminated soil and ground water is mainly due to
concrete powder solution extracts contaminated with only chloride-induced reinforcement corrosion and sulfate
sodium chloride. attack[7, 8]. Chloride ions may be introduced in to the
concrete by mix ingredients namely aggregates, mixing
Keywords: Concrete; Steel reinforcement; Chloride ion,
water and admixtures. Alternatively, they may penetrate
Sulfate ion; Corrosion.
the hardened concrete from the service environment[9].
Chloride ions depassivate the steel reinforcement
Introduction and are considered as the primary cause of rebar
The durability of concrete is one of its important properties corrosion[10]. Sulfate ions react with hydrated C3A and
because it is essential that concrete should be capable of Ca(OH)2, to produce expansive and/or softening types of
withstanding the conditions for which it has been designed deterioration[11]. From durability viewpoint, the behaviour
throughout the life of a structure[1]. Lack of durability can of steel reinforcement in chloride-sulfate contaminated
be caused by external agents arising from the environment concrete needs to be studied. The behaviour of steel

Organised by
India Chapter of American Concrete Institute 547
Sessiion 5 B - Paper 3

reinforcement can be described by means of polarization

curves, which relate the potential with anodic or cathodic
current density[12]. Determination of polarization curves
for reinforcing steel in concrete is more complicated than
that in simulated concrete pore solutions[12]. Different
researchers[13, 14, 15] have studied the corrosion behaviour
of steel in simulated pore solutions either in the presence
of chloride ions or in the presence of sulfate ions. Further
some work has been carried out on corrosion behaviour of
steel in concrete powder solution extracts contaminated
with only chloride salts[9]. However, the study on
Fig. 1: Chart showing composition of admixed chloride and
corrosion behaviour of steel reinforcement in concrete
sulfate salts in concrete mixes
powder solution extracts contaminated with composite
chloride-sulfate ions is scanty. Therefore in the present concrete powder. Thermex TMT (Thermomechanically
experimental investigation, anodic polarization curves treated) steel of 70 mm length and 12 mm diameter
of steel reinforcement were obtained by conducting was used as bare steel specimen and it was drilled and
potentiodynamic linear sweep test on steel in concrete threaded at one end. The steel specimens were cleaned
powder solution extracts contaminated chloride ions and with wire brush and coated with epoxy leaving a length of
composite chloride-sulfate ions. Further half-cell potential 5 mm exposed, as shown in Figure 2.
and linear polarization resistance measurements were
also carried out in concrete powder solution extracts.

Experimental Program
Materials used and Specimen Preparation
Ordinary Portland cement (OPC) satisfying IS: 12269-
1987 [16] was used to prepare different concrete
mixtures. The concrete mixtures were prepared with
a water-cement ratio of 0.5 and cement content of 390
kg/m3. Fine aggregate with specific gravity of 2.61 and
Fig. 2: Schematic diagram of bare steel specimen
conforming to grading zone II as per IS: 383-1970 [17]
and coarse aggregates of size 20 mm MSA (maximum
Preparation of Concrete Powder Solution Extracts
size aggregate) and 10 mm MSA in proportion of 66% and
34% respectively of the total mass of coarse aggregate The concrete cubes made from OPC and admixed with
were used in the preparation of concrete mixtures. The different concentrations of chloride and sulfate salts were
specific gravity of 20 mm MSA and 10 mm MSA were crushed at the age of 56 days from the day of preparation
2.64 and 2.63 respectively. Water content of 195 kg/m3 in the compression testing machine and after that further
was used in the preparation of concrete mixtures for a crushing was carried out in abrasion testing machine. The
slump range of 20 mm – 50 mm. Cube specimens of size crushed concrete powder was then sieved through a sieve
150 mm were prepared from different concrete mixtures with square opening of size 150 μm. The collected powder
i.e. from control mix and from mixes contaminated with was then stored in air tight bags. For the preparation of
chloride and sulfate ions. Sodium chloride was used to concrete powder solution extracts, the stored concrete
provide chloride ions while magnesium sulfate was used powder was mixed with distilled water in 1:1 proportion
to provide sulfate ions in the concrete mixes. The admixed by mass and stirred for half an hour. The solution was
dosages of sodium chloride were 3% and 5% (by mass of then boiled for 15 - 20 minutes and then allowed to settle
cement content) and similarly those of magnesium sulfate down and cooled to ambient temperature. After that, the
were 3%, 6% and 12% (by mass of cement content). The solution was filtered through Whatman no. 1 filter paper.
required quantities of salts were dissolved in the mixing This filtered solution was then used as concrete powder
water. The details of varying concentrations of chloride solution extracts for conducting corrosion tests on bare
and sulfate ions are shown in Figure 1. steel specimen.

The concrete cubes prepared from different concrete

mixtures were moist cured for a period of 27 days after Test Methods
demoulding. The demoulding of cube specimens was
carried out after 24 hours of preparation. After completion Potentiodynamic linear sweep test
of moist curing, the specimens were kept under the The corrosion behaviour of bare steel was evaluated
laboratory condition till the age of crushing for obtaining by conducting potentiodynamic linear sweep test using

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

548 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of Steel Reinforcement in Chloride and Combined Chloride-Sulfate Contaminated Concrete Powder Solution Extracts

the corrosion monitoring instrument (make ACM, Gill Where βa and βc are anodic and cathodic Tafel constants
AC serial no. 1542). Bare steel specimen was used as respectively. In the present investigation the value of B was
working electrode. Saturated calomel electrode (SCE) taken as 26 mV, considering the steel in active condition[21].
was used as the reference electrode. The filtered solution The experimental setup for potentiodynamic linear sweep
extract was poured into the electrochemical cell. Working test and LPR test is shown in Figure 3.
electrode (WE), auxiliary electrode (AE), and reference
electrode (RE) placed in the electrochemical cell were
Results and Discussions
connected to the corrosion monitoring instrument. The
potentiodynamic linear sweep test was carried out on the Corrosion Zones of Steel Reinforcement
bare steel specimen immersed in the solution extracts
The anodic polarization curves were obtained from
by applying the potential scan from 0 mV to 1500 mV with
potentiodynamic linear sweep test conducted on bare
offset from equilibrium potential at a sweep rate of 50 mV
steel specimens in concrete powder solution extracts.
per minute.
The anodic polarization curve of steel in concrete powder
solution extracts made from control mix i.e. without any
Linear polarization resistance (LPR) test
admixed salt is shown in Figure 4.
Reinforcement corrosion was monitored by measuring
corrosion potential and corrosion current density of bare In Figure 4, potential ranges for different zones of corrosion
steel specimen in concrete powder solution extracts. The namely semi-immune zone, active zone, passive zone and
corrosion current density was determined by conducting pitting zone are identified. The zone of corrosion below the
linear polarization resistance (LPR) test. The LPR test rest potential is termed as semi-immune zone. In semi-
was performed on bare steel specimen using the same immune zone the steel is thermodynamically unable to
experimental set up, which was used in potentiodynamic undergo anodic reactions and thus immune to dissolution.
linear sweep test. Concrete powder solution extracts was The zone above the semi-immune zone is termed as
poured in to the electrochemical cell and three electrodes active zone. In this zone there is significant increase in
as stated earlier were immersed in the solution and corrosion current density with small change in potential.
connected to the corrosion monitoring instrument. It is to The zone above the active zone is known as passive zone
be noted that separate bare steel specimens were used and in this zone the change in current density is very less
in this test. The half-cell potential of steel specimen was with significant increase in potential. The zone above the
measured with reference to saturated calomel electrode passive zone is called as pitting zone. In pitting zone the
(SCE). LPR test was conducted by polarizing the steel anodic current density increases significantly leading to
specimen in the potential range of ± 20 mV from the localized dissolution of steel reinforcement. The potential
equilibrium potential at a scan rate of 0.1 mV/sec. The values at boundaries of different zones such as, active/
corrosion current density was then calculated using passive zone boundary (Act/Pass) potential and passive/
Stern-Geary equation [18, 19, 20], which is stated below. pitting zone boundary (Pass/Pitt) potential are shown
in Figure 4. From this figure, it is observed that the rest
I corr = R .......................................................................(1)
B
P
potential value of steel is -325.69 mV (SCE), active/passive
boundary potential and passive/pitting potential values
Where Icorr = Corrosion current density, B = Stern-Geary are -225.14 mV (SCE) and +565.3 mV (SCE) respectively in
constant and Rp = Polarization resistance of steel. uncontaminated concrete mix.
The Stern-Geary constant ‘B’ is given by the following
equation.
ba # bc
23 Q b a + b c V
B= ............................................................(2)

Fig. 3: Experimental setup for potentiodynamic linear sweep Fig. 4: Anodic polarization curve of steel in concrete powder
test and linear polarization resistance (LPR) test solution extracts made from control mix

Organised by
India Chapter of American Concrete Institute 549
Sessiion 5 B - Paper 3

The anodic polarization curves of steel reinforcement in

concrete powder solution extracts admixed with different
concentrations of sodium chloride are shown in Figure 5
and those in concrete powder solution extracts admixed
with different concentrations of sodium chloride plus
magnesium sulfate are shown in Figure 6 and Figure 7.
From Figure 5, it is observed that the rest potential values
of steel in concrete powder solution extracts contaminated
with 3% NaCl and 5% NaCl are -354.9 mV and -365.79
mV respectively. The active/passive boundary potential
values are -276.28 mV and –260.51 mV and passive/pitting
boundary values are -145.31 mV and -161.59 mV for 3%
NaCl and 5% NaCl concentrations respectively.
From Figure 6, it is observed that the rest potential values
of steel in concrete powder solution extracts contaminated
Fig. 5: Anodic polarization curve of steel in concrete powder with 3% sodium chloride concentration at magnesium
solution extracts contaminated with 3% NaCl and 5% NaCl
sulfate concentrations of 3%, 6% and 12% are -376.23
mV, -379.0 mV and -389.15 mV respectively, while the
active/passive boundary potential values are -248.68 mV,
-252.69 mV and -291.93 mV; and passive/pitting boundary
potential values are -127.74 mV, -161.46 mV and -207.61
mV respectively. Similarly From Figure 7, it is observed
that the rest potential values of steel in concrete powder
solution extracts contaminated with 5% sodium chloride
concentration at 3%, 6% and 12% magnesium sulfate
concentration are -356.46 mV, –383.85 mV and -430.36
mV respectively whereas, the active/passive boundary
potential values are -274.95 mV, -330.62 mV and -340.94
mV; and passive/pitting boundary values are -184.12 mV,
-262.92 mV and -289.41 mV respectively.
From the results it is observed that, for steel in
uncontaminated concrete powder solution extracts, the
passive/pitting boundary potential value (i.e. +528.83 mV)
Fig. 6: Anodic polarization curve of steel in concrete powder in Figure 4 represents the beginning of transpassivity
solution extracts contaminated with 3% NaCl and MgSO 4 zone characterized by oxygen evolution. However for steel
concentrations of 3%, 6% and 12% in chloride and composite chloride-sulfate contaminated
concrete powder solution extracts, much lower passive/
pitting boundary potential values were observed. This
indicates that the range of passive zone is more in
uncontaminated concrete mixes whereas, that in concrete
mixes contaminated with chloride ions and composite
chloride-sulfate ions is very small. Thus the chance of
pitting corrosion (or localized corrosion) is more in the
presence of chloride ions and composite chloride-sulfate
ions.

Potential values
The measured half-cell potential values of steel
reinforcement in concrete powder solution extracts
contaminated with chloride and composite chloride-
sulfate ions are shown in Figure 8. The measured half-cell
potential value of steel specimen in the uncontaminated
concrete powder solution extracts is -338.62 mV (SCE).
Fig. 7: Anodic polarization curve of steel in concrete powder
solution extracts contaminated with 5% NaCl and MgSO 4 The corrosion potential provides a qualitative indication of
concentrations of 3%, 6% and 12% occurrence steel reinforcement corrosion. As per ASTM

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

550 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Behaviour of Steel Reinforcement in Chloride and Combined Chloride-Sulfate Contaminated Concrete Powder Solution Extracts

C 876 [ ], the threshold potential is taken as -350 mV (Cu/

CuSO4 electrode)/-270 mV (SCE) and the potential values
more negative than -270 mV (SCE) correspond to greater
than 90% probability of occurrence of steel reinforcement
corrosion. From Figure 8, it is observed that all the
potential values are more negative than -270 mV (SCE),
which indicates greater probability of occurrence of
steel reinforcement corrosion in both chloride and
composite chloride-sulfate contaminated concrete mixes.
Further from Figure 8, it is observed that the potential
values decreased (more negative) with increase in the
concentrations of both admixed sodium chloride and
magnesium sulfate. In addition the potential values of steel
in concrete powder solution extracts contaminated with
only chloride ions are less negative than that in concrete
powder solution extracts contaminated with composite
Fig. 9: Corrosion current density of steel in concrete powder
chloride-sulfate ions.
solution extracts contaminated with different concentrations of
chloride ions and composite chloride-sulfate ions

current density of steel in concrete powder solution

extracts contaminated with composite chloride-sulfate
ions may be due to the presence of higher free chloride
content near the steel reinforcement. In addition from the
results, it is observed that the corrosion current density
increased with increase in sulfate ion concentration at
all admixed concentrations of chloride ions. The increase
in corrosion current density with increase in sulfate ion
concentration may ascribed to the preferential reaction of
sulfate ions as compared to chloride ions, with hydrated
C3A, thereby resulting in lower chloride binding and higher
free chloride content near the steel reinforcement.

Fig. 8: Potential value of steel in concrete powder solution Conclusions

extracts contaminated with different concentrations chloride From the potentiodynamic polarization study on steel in
ions and composite chloride-sulfate ions
uncontaminated and contaminated (with chloride ions
and composite chloride-sulfate ions) concrete powder
Corrosion Current Density (Icorr) solution extracts, the potential ranges for different zones
The corrosion current density (Icorr) of steel reinforcement of corrosion, namely semi-immune zone, active zone,
was determined by LPR technique in uncontaminated passive zone, and pitting zone were obtained. These
and contaminated concrete powder solution extracts zones of corrosion provide information about the state of
(with chloride and mixed chloride-sulfate ions). The corrosion of steel reinforcement.
obtained value of corrosion current density of steel in
uncontaminated concrete powder solution extracts is 1.08 The half-cell potential values of steel reinforcement
μA/cm2. The values of corrosion current density (Icorr) of were more negative than -270 mV (SCE) in concrete
steel reinforcement in contaminated concrete powder powder solution extracts contaminated with chloride and
solution extracts are shown in Figure 9. composite chloride-sulfate ions. The corrosion current
From Figure 9, it is observed that the corrosion current density of steel in concrete powder solution extracts
density of steel increased with increase in chloride ion increased with increase in chloride ion and sulfate ion
concentration. This may be attributed to increase in concentration. Further the corrosion current density of
conductivity of concrete in the presence of higher amount steel in concrete powder solution extracts contaminated
of chloride ions. Further from this figure, it is observed that with NaCl+MgSO4 was more as compared to that in
the corrosion current density of steel in concrete powder concrete powder solution extracts contaminated with
solution extracts contaminated with NaCl+MgSO4 is more only NaCl. Thus the conjoint presence of chloride-sulfate
as compared to that in concrete powder solution extracts ions has increased the effect of chloride ion on steel
contaminated with only NaCl. The higher corrosion reinforcement corrosion.

Organised by
India Chapter of American Concrete Institute 551
Sessiion 5 B - Paper 3

Acknowledgement 10. Pradhan, B., Bhattacharjee, B., 2011. Rebar corrosion in chloride
environment. Construction and Building Material, 25:2565-2575.
The authors wish to express their gratitude to Department 11. Al-Amoudi, O.S.B., Maslehuddin, M., Abdul-Al, Y.A.B., 1995. Role
of Science and Technology, Government of India for of chloride ions on expansion and strength reduction in plain and
funding the corrosion monitoring instrument (used in this blended cements in sulfate environmemnts. Construction and
study) through a project under Fast Track Scheme for Building Materials, 9:25-33.
Young Scientists. 12. Bertolini, L., Elsener, B., Pedeferri, P., Radaelli, E., Polder, R., 2013.
Corrosion of steel in concrete. 2nd edition,WILEY-VCH Verlag GmbH
& Co. KGaA: Weinheim.
References
13. Al-Tayyib, A.J., Somuah, S.K., Boah, J.K., Leblanc, P., Al-Manna,
1. Neville, A.M., Brooks, J.J., 2004. Concrete Technology. Pearson
A.I., 1988. Laboratory study on the effect of sulfate ions on rebar
Education, Fourth Indian Reprint.
corrosion. Cement and Concrete Research, 18:774-82.
2. Soutsos, M., 2010. Concrete durability. Thomas Telford Ltd, London.
14. Cheng, T.Y., Lee, J.T., Tsai, W.T., 1990. Corrosion of reinforcement in
3. Sakr, K., 2005. Effect of cement type on the corrosion of reinforcing artificial sea water and concentrated sulfate solution. Cement and
steel bars exposed to acidic media using electrochemical techniques. Concrete Research, 20:243-52.
Cement and Concrete Research, 35:1820-1826.
15. Mammoliti, L., Hansson, C.M., 2005. Influence of cation on corrosion
4. Chen, W., Du, R., Ye, C., Zhu, Y., Lin, C., 2010. Study on the corrosion behaviour of reinforcing steel in high-pH sulfate solutions. ACI
behaviour of reinforcing steel using in situ raman spectroscopy Material Journal, 102:279-85.
assisted by electrochemical techniques. Electrochimica Acta,
16. IS 12269-1987. (reaffirmed 2004) Specifications for 53 grade
55:5677-5682.
ordinary Portland cement. New Delhi: Bureau of Indian Standards.
5. Shi, X., Xie, N., Fortune, K., Gong, J., 2012. Durability of steel
17. IS: 383-1970 (reaffirmed 2002), Specification for coarse and fine
reinforced concrete in chloride environments: an overview.
aggregate from natural sources for concrete. New Delhi: Bureau
Construction and Building Materials, 30:125-138.
of Indian Standards.
6. Abosrra, L., Ashour, A.F., Youseffi, M., 2011. Corrosion of steel
18. Al-Amoudi, O.S.B., 1995. Performance of 15 reinforced concrete
reinforcement in concrete of different strengths. Construction and
mixtures in magnesium-sodium sulphate environments.
Building Materials, 25:3915-3925.
Construction and Building Materials, 9:149-158.
7. Al-Amoudi, O.S.B., 1998. Sulphate attack and reinforcement
19. Pradhan, B., Bhattacharjee, B., 2009. Performance evaluation
corrosion in plain and blended cements exposed to sulfate
of rebar in chloride contaminated concrete by corrosion rate.
environments. Building and Environment, 33:53-61.
Construction and Building Material, 23:2346-2356.
8. Dehwah, H.A.F., Austin, S.A., Maslehuddin, M., 2002. Chloride-
20. Pradhan, B., Bhattacharjee. B., 2007. Role of steel and cement type
induced reinforcement corrosion in blended cement concretes
on chloride-induced corrosion in concrete. ACI Material Journal,
exposed to chloride-sulphate environments. Magazine of Concrete
104:612-619.
Research, 54:355-364.
21. Dehwah, H.A.F., Maslehuddin, M., Austin, S.A., 2002. Long-term
9. Pradhan, B., Bhattacharjee, B., 2007. Corrosion zones of rebar in
effect of sulfate ions and associated cation type on chloride-induced
chloride contaminated concrete through potentiostatic study in
reinforcement corrosion in Portland cement concretes. Cement and
concrete powder solution extracts. Corrosion Science, 49:3935-52.
Concrete Composites, 24:17-25.

Mrs. Fouzia Shaheen

Department: Civil Engineering
Research Scholar
Mobile: 09085894798
E-mail: fouzia@iitg.ernet.in
Address: Fouzia Shaheen, w/o Syed Omar, # 206, B-Block, MSH, IIT Guwahati Campus, Guwahati- 781039
Alternate e-mail ID: syedafouziashaheen@gmail.com
Biographical Sketch: Mrs. Fouzia Shaheen did her Bachelor of Engineering in Civil from Osmania University.
She did her Master of Engineering in Structural Engineering from Osmania University and pursuing Doctor
of Philosophy from Indian Institute of Technology Guwahati (IITG) in the area of Concrete Technology.
Experience Description: Assistant professor at Muffakham Jah College of Engineering and Technology:
2008-2010.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

552 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Paraffin in Drywalls as Sensible Heat Storage Material for Temperature Moderation

Use of Paraffin in Drywalls as Sensible Heat Storage Material for

Temperature Moderation
Rampradheep G. S.,
Assistant Professor, Department of Civil Engineering, Kongu Engineering College, Erode, Tamilnadu, India – 638 052.
Dr. Sivaraja M.
Principal, N.S.N College of Engineering and Technology, Karur, Tamil Nadu, India – 639 603.
Namratha, Ariram Prasath M., Surya K., Saranya R., Arunkumar N., Lekha G., Manissa N.
UG Student, Department of Civil Engineering, Kongu Engineering College, Erode, Tamilnadu, India – 638 052.

Abstract Phase Change Materials

Phase Change materials (CnH2n+2, n=24~36) is a smart Phase change materials are those materials which
material having high heat of fusion. It is used in stabilizing can store and release heat. The main property of phase
the room temperature at all seasons. The most commonly change material is the storage of heat energy in its latent
used PCMs are salt hydrates, fatty acids and esters and form which leads to great capacity per volume of heat
various paraffin. They have been used in refrigerated storage than that of normal building materials[2]. When the
transportation, for rail and road applications. Our idea is temperature of the room rises to the ambient value, the
to use these, Phase-change materials in drywall board chemical bonds of the material will break up and thus the
to store and release heat to save power. The phase solid will change its form to liquid. This change of phase
change material, we used here is paraffin. Phase change from solid to liquid is an endothermic process which will
materials are prepared and plastered along with the walls result in absorption of heat energy from the surrounding.
of the buildings. During daytime, melting of paraffin takes During the night when the temperature falls, the phase
place, which is endothermic process and hence absorbs change material will return to the solid state and release
the heat and keeps the room cool. Similarly during night the absorbed heat.
time solidification of paraffin takes place, which is an
exothermic process and hence keeps the room warm.
Types of Phase Change Materials
Thus, by using phase change materials in drywalls the
human comfort zone temperature (20° to 30°C) can be Phase change materials commonly used are classified as
maintained in buildings thereby reducing the usage of air- organic and inorganic phase change materials.
conditioners. In this paper we have derived four different
mix ratios for gypsum boards and have found one optimum Building Applications
mix based on the test results. The cost analysis has been
Phase Change Materials (PCM’s) have been considered
done for this optimum mix alone.
for thermal storage in buildings since 1980[4]. With the
Keywords: Smart material, Phase Change Material advent of PCM implemented in gypsum board, plaster,
(PCM), Paraffin, Temperature, Drywall, Gypsum. concrete or other wall covering materials, thermal
storage can be part of the building structure even for
light weight buildings [5]. In this project, development
Introduction and testing were conducted for prototypes of PCM
According to the Indian Energy Information Administration, wallboards to enhance the thermal energy storage (TES)
buildings consume about 40% of the electricity generated capacity of standard gypsum wallboard [6]. The phase-
in India, and about 25% of that is used for air conditioning in change materials incorporated in drywalls keep a room
homes and offices. Due to the highest potential for energy
and environmental saving energy efficiency has become a
priority in buildings. PCM are the materials which change
phase in a narrow range of temperature. This substance
on melting and solidifying at a certain temperature, is
capable of storing and releasing large amounts of energy
correspondingly (i.e. this PCM absorbs the heat and keeps
the room cool during daytime and gives out the stored
latent heat during night time and keeps the room warm)[1].
This property is used in stabilizing the room temperature
by plastering PCMs along with the walls. These types of
plasters are called drywalls.
Fig. 1: Working principle of phase change materials

Organised by
India Chapter of American Concrete Institute 553
Session 5 B - Paper 4

of fusion is the change in enthalpy resulting from heating

Table 1
Types of PCM
a given quantity of a substance to change its state from a
solid to a liquid. The temperature at which this occurs is
the melting point. The 'enthalpy' of fusion is a latent heat,
ORGANIC PCM INORGANIC PCM
because during melting the introduction of heat cannot be
observed as a temperature change, as the temperature
Organic PCM are chemically Inorganic PCM remains constant during the process. The latent heat of
stable, safe and non- have high thermal fusion is the enthalpy change of any amount of substance
reactive. They also have conductivity, high
DEFINITION an ability to melt without heat of fusion, when it melts. The liquid phase has a higher internal
segregation. They are are relatively energy than the solid phase. This means energy must be
further classified into cheap and non- supplied to a solid in order to melt it and energy is released
paraffins and non paraffins. flammable. from a liquid when it freezes, because the molecules in
the liquid experience weaker intermolecular forces and so
Paraffin wax – have a higher potential energy[2].
Molecular formula- Hydrated salts- ll It should have high specific heat for sensible heat
CnH2n+2, n= 24~36. which have a high storage effects – Specific heat is nothing but the heat
Melting point- 20° to 70°C. storage density of
Depending upon the number capacity per unit mass of the material. It is generally
about 240 k J/kg.
of C atoms present in the Glauber’s salt or denoted as Joules per kilogram per Kelvin.
chain, the melting point Na2SO4.H2O- one
of the wax varies. Higher of the cheapest
ll The melting point can be selected to suit the
MOST USED PCM the number of carbon desired operating temperature range. Generally
materials available
atoms, higher will be the for thermal about 27°C is considered as optimum temperature
compounds melting point. storage but its for human comfort. Chemically stable with no
Non Paraffins – use is restricted
because of super
chemical decomposition and corrosion resistance to
Molecular formula cooling and phase construction materials.
-CH3(CH2)2nCOOH. Melting segregation
point- similar to those of ll It should be non-explosive, non-poisonous and non-
paraffin PCMs. flammable. It should be easily available in large
quantities
Non Paraffins are not Corrosive to ll It should have low cost.
preferred because they most metals and
DISADVANTAGE are much more expensive undergo super
than paraffin Phase change cooling and phase Incorporation of PCM in Dry Walls
materials. decomposition.
Utilizing phase change material integrated into modern
drywall (2cm), the heat storage capacity becomes that
of 24 cm of concrete, 36 cm of brick masonry, 38 cm of
cool in much the same way that ice cubes chill a drink: timber, or 226 cm of light construction. The integration of
by absorbing heat as they melt. Each polymer capsule phase change materials allows for a high degree of heat
contains paraffin waxes that melt at around room storage capacity in a thin and lightweight structure[9, 10].
temperature, enabling them to keep the temperature of
a room constant throughout the day. The paraffin’s work
best in climates that cool down at night, allowing the
materials inside the capsules to solidify and release the
heat they've stored during the day [7]. The work is part
of a push in the construction industry toward greener
building materials that help maintain comfortable
temperatures without using electricity [8].
Fig. 2: Heat storage capacity of different materials
Choosing the Right Phase Change Material
In order to effectively use phase change material for Dry Walls or Gypsum Boards
thermal energy storage systems, we must choose the apt The internal surfaces of walls and ceilings of most
phase change material. The most important criteria for buildings are finished internally by applying plaster in
selecting the phase change materials are: one or more coats[11]. In order to reduce the demand
of site labour, the use of building board as covering
It should have high latent heat of fusion per unit mass so
for walls and ceiling is increasing steadily. Gypsum
as to store large amount of energy in smaller amount of
boards have the specific advantage of being lighter
material.The enthalpy of fusion also known as (latent) heat

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

554 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Paraffin in Drywalls as Sensible Heat Storage Material for Temperature Moderation

than the boards of similar nature such as fibre hard

Table 4
boards and asbestos cement building boards. Gypsum
Properties of Coconut Fibre
boards also possess better fire resisting, thermal and
sound insulating properties. Gypsum boards may be
Physical Numerical
manufactured as plain, laminated and reinforced boards. S.No Unit
property value
Reinforcing materials generally used are glass, paper,
vegetable fibres, etc. The boards may be used to provide 1 Density 1.4 g/cc
lining finishes to masonry walls, to ceilings, to steel or
timber framed partitions or as claddings to structural 2 Diameter 16 Micron
steel columns and beams, or in the manufacture of
3 Aspect ratio 300 -
prefabricated partition panels[12].

Materials and Their Properties

Paraffin as a Phase Change Material

Table 2
Properties of paraffin wax

Property of paraffin Numerical Value Unit Fig. 5: Coconut fibre

Density 900 kg/m3

Material Testing
Heat of combustion 42 kJ/g
Test for Melting Point of Wax
Specific heat capacity 2.14 – 2.9 Jg−1K−1 The beaker is filled with water up to 500ml mark. 50
gm of Paraffin wax is taken in a test tube. A mercury
Heat of fusion 200-220 J/g thermometer is inserted inside the test tube. The test
tube is held using a stand and is placed inside the 500 ml
beaker such that the bottom of the test tube is just above
Materials required for Gypsum Board
the base of the beaker. The whole apparatus is placed
Manufacturing over a hot plate or an electric heater and allowed to heat.
On heating the wax in the test tube starts to melt. The
Gypsum Powder
temperature is allowed to rise till the entire wax becomes
liquid. Once the entire wax in the test tube is in liquid form
Table 3 it is allowed to cool. The stop watch is then started. The
Physical properties of gypsum powder thermometer readings are taken corresponding to every
2 minutes. A graph is plotted against time taken on x-axis
Property of paraffin Numerical Value Unit and temperature reading on y-axis. The temperature
corresponding to the straight line in the plotted graph
Density 2.3-2.8 g/cc gives the melting point of wax.
Molecular Weight 172.2 g

Specific heat capacity 1.09 Jg-1k-1

Solubility in water 0.21 g/100 ml at 200°C

Fig. 5: Paraffin wax Fig. 4: Gypsum powder Fig. 6: Melting point apparatus Fig. 7: Temperature reading taken

Organised by
India Chapter of American Concrete Institute 555
Session 5 B - Paper 4

paraffin runs off, leaving no buildup of wax. Immersion

Table 5
Temperature readings for determining melting point of wax
times vary depending on the amount of PCM uptake
desired, however, they rarely exceed ten minutes. PCM
content ranges up to 30% of the composite weight of
Time in Thermometer Time in Thermometer
minutes in °C minutes in °C 1/2 inch drywall. Drywall dipped in paraffin becomes
water resistant. While common gypsum drywall is
0 50 26 27
fire-resistant, PCM-drywall is quite flammable unless
treated with fire-retardant chemicals. Immersion is the
simplest, lowest cost method for making PCM drywall.
2 47 28 27
Polyethylene pellets, saturated with melted paraffin,
then mixed with wet gypsum and compressed in sheet
4 46 30 26
form, also yield production quality drywall. Relative to
immersed drywall, this material is more fire- resistant,
6 43 32 26
less water-resistant, and conforms to the current
gypsum drywall manufacturing process. Both versions
8 40 34 25
work well for heat transfer and storage, and the paraffin
remains permanently in the drywall.
10 36 36 25

12 34 38 25 Microencapsulation Of PCM
Microencapsulation is a process in which tiny particles or
14 32 40 25 droplets are surrounded by polymeric material to form
capsules[14]. In a relatively simplistic form, a microcapsule
16 30 42 25 is asmall sphere with a uniform wall around it. The
material inside the microcapsule is referred to as the
18 29 44 25 core, whereas the wall around the core material is called a
shell. Most microcapsules have diameters between a few
20 27 46 25
micrometers and a few millimeters. The main reasons
for microencapsulation of PCMs are to isolate core
material from its surroundings (External agency), Control
22 27 48 25
the rate of release of core material and to keep the core
material intact within the desired boundary, so that oozing
24 27 50 25
out during its transition (from solid to liquid) is avoided.
Microencapsulation processes are usually categorized
into two groups: chemical processes and physical
The melting point of wax is thus found from this test as
processes. The use of some methods has been limited
the temperature which remains constant for the longest
to the high cost of processing, regulatory affairs, and the
duration. From figure 10, a graph of temperature reading
use of organic solvents, which are a concern for heath
versus time is plotted. The flat portion of the graph
and the environment. Physical methods are mainly spray
indicates the temperature which remains constant.
drying or centrifugal and fluidized bed processes, which
Thus, from the above graph, we can obtain the melting
are inherently not capable of producing microcapsules
point of the sample paraffin wax as 27°C. To increase the
smaller than 100 μm. The chemical processes include
melting point of paraffin, the number of carbon atoms of
the interfacial polymerization, in situ polymerization,
the phase change material must be increased. More the
the sample or complex coacervation, phase separation,
carbon atoms higher will be its melting point. Similarly
suspension-like polymerization and other fabrication
lesser the carbon atoms lesser will be the melting point.
methods. MEPCM are mainly fabricated by chemical
Hence we can choose a wax which melts in the desired
methods due to its properties and applications.
range of temperature.
In our project encapsulation of paraffin is done using
Poly Vinyl alcohol. The required quantity of wax is first
Fabrication weighed and heated so that it melts completely. Then
Methods of Incorporation of PCM in Dry Walls a suitable quantity of poly vinyl alcohol is weighed and
added to boiling water so that PVA solution is obtained.
Paraffin can be incorporated into drywall in two ways, Molten paraffin wax is then added to the PVA solution
by direct immersion and by adding micro-encapsulated and it is stirred continuously. The water in the solution
pellets to the drywall mixture during the manufacturing repels the wax to forms capsules and also drives the
process[13]. Since drywall is a porous material, it can PVA to coat the paraffin. At the end of this process tiny
absorb melted paraffin when immersed in it. Extra spheres of paraffin are encapsulated in polymer shells.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

556 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Paraffin in Drywalls as Sensible Heat Storage Material for Temperature Moderation

This is then cooled in water so that it hardens. Our idea Manufacture of Gypsum Boards
is to add this resulting wet mixture to the powder that
A wall board panel generally consists of a layer of gypsum
is used to make the dry walls. The following is a step by
plaster sandwiched between two layers of paper. The raw
step procedure followed for the encapsulation of phase
gypsum, (CaSO4·2H2O), is heated to drive off the water
change material.
then slightly re-hydrated to produce the hemihydrate
STEP 1: Weight the amount of wax required for of calcium sulfate (CaSO4·1⁄2 H2O). The plaster is mixed
encapsulation. with fiber (typically paper and/or fiberglass). The board
is then formed by sandwiching a core of the wet mixture
STEP 2: Melt the wax so that paraffin in its liquid form is
between two sheets of heavy paper or fibreglass mats.
obtained.
The face and back papers should be securely bonded to
STEP 3: Weigh the amount of Poly vinyl alcohol required. the core. The paper surface is varied according to the
use of the particular type of board, and the core may
STEP 4: Add Poly Vinyl Alcohol to boiling water in a beaker
contain additive to impart additional properties. The
and allow it to melt completely forming a PVA
longitudinal edges are paper covered and profiled to
Solution.
suit the application. When the core sets it is then dried
STEP 5: Add molten paraffin to the mixture and stir in a large drying chamber, and the sandwich becomes
continuously until capsules are found. rigid and strong enough for use as a building material.
STEP 6: The PVA coated paraffin capsules are then cooled The following chemical reaction takes places in the
using water. manufacture of gypsum board:
i. When mineral gypsum is heated to about 150oC, it
loses water and produces gypsum powder also known
as plaster of paris.
CaSO4·2H2O + heat ---> 2CaSO4·1⁄2H2O + 3H2O .... (1)
Gypsum + heat ------------> Plaster of Paris + steam
ii. When water is added to this gypsum powder it
rehydrates and quickly hardens.
2CaSO4·1⁄2H2O + 3H2O ---> 2CaSO4·2H2O+heat..(2)
Plaster of Paris + water ----- > Gypsum + heat
Fig. 8: Weighing wax Fig. 9: Melting wax using hot plate

Table 6
Mix ratios adapted for testing of PCM wallboards

Gypsum Coconut
Specimen  Water (%) PCM (%)
powder (%) fibre (%)

1 60 30 5 5

2 55 30 10 5

3 50 30 15 5
Fig. 10: Weighing poly vinyl Fig. 11: Addition of PVA to 4 45 30 20 5
alcohol boiling water

Table 7
Quantity of material required for each mix ratio

Gypsum Coconut
Specimen  Water (%) PCM (%)
powder (%) fibre (%)

1 5.94 2.973 0.495 0.495

2 5.45 2.973 0.991 0.495

3 4.95 2.793 1.48 0.495

Fig. 12: Stirring molten paraffin F i g . 1 3 : F o r m a t i o n o f
4 45 30 20 5
and PVA solution encapsulated paraffin on cooling

Organised by
India Chapter of American Concrete Institute 557
Session 5 B - Paper 4

lower the content of gypsum powder, lesser is the water

absorption.
This may be because there is more than adequate water
for setting of gypsum. Percentage absorption = [(W 1 – W2)
/ W2] X 100.

Table 9
Percentage absorption of water

Fig. 14: Wall board mould and Fig. 15: Paste poured in to wall Percentage
Specimen W1(g) W2(g)
Gypsum board paste board mould Absorption (%)
1 355 330 7.5
Results and Discussion 2 353 333 6
3 353 335 5.4
Transverse Strength of Gypsum Board Test Results
4 352 335 5.1

Table 8
Mix ratios adapted for testing of PCM wallboards Test Results for Temperature Moderation

Sr. No. Specimen

Transverse Strength Experiment carried out with Specimen 1 on 16.2.2015
(N/mm2)

1 Specimen 1 0.7
Table 10
Temperature readings taken in the cubicles (specimen 1)
2 Specimen 2 0.8

3 Specimen 3 0.7  Temperature in Temperature in

Time in hours
room without PCM room with PCM
4 Specimen 4 0.6 06.30 hrs 23°C 23°C
08.30 hrs 27°C 27°C
10.30 hrs 31°C 31°C
12.30 hrs 36°C 36°C
14.30 hrs 38°C 38°C
16.30 hrs 36°C 36°C
18.30 hrs 32°C 32°C
20.30 hrs 30°C 30°C
22.30 hrs 27°C 27°C

It is seen from Figure 19, that there is no effect in the

Fig. 16: Transverse strength test Fig. 17: Water absorption test temperature of the cubicle when Specimen 1 is used.
Hence, the percentage of Phase Change Material is
Water Absorption Test Results increased by 5 percent and tested again [15].
The average percentage
water absorption must
not be greater than 10
percent for gypsum
boards according to
Indian Standard codes.
The average water
absorption percentage
obtained from the tested
specimens is 6%.Thus
the water absorption
percentage is within
limits. It is also seen that Fig. 18: Temperature test Fig. 19: Graph showing hourly temperature variation with specimen 1

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

558 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Paraffin in Drywalls as Sensible Heat Storage Material for Temperature Moderation

Experiment carried out with Specimen 2 on 17.2.2015 temperature. This means at this composition the phase
change materials start working and shows its effect in
temperature moderation.
Table 11
Temperature readings taken in the cubicles (Specimen 2)

Temperature in Temperature in
Time in hours
room without PCM room with PCM
06.30 hrs 23°C 23°C
08.30 hrs 25°C 25°C
10.30 hrs 31°C 31°C
12.30 hrs 36°C 36°C
14.30 hrs 38°C 38°C
16.30 hrs 35°C 35°C
18.30 hrs 31°C 31°C
20.30 hrs 29°C 29°C Fig. 21: Graph showing hourly temperature variation with specimen 3
22.30 hrs 27°C 27°C
Experiment carried out with Specimen 4 on 19.2.2015
It is seen from Figure 20, that there is no effect in the
temperature of the cubicle when Specimen 2 is used. Table 13
Hence, the percentage of Phase Change Material is further Temperature readings taken in the cubicles (specimen 4)
increased by 5 percent and tested again.
Temperature in Temperature in
Time in hours
room without PCM room with PCM
06.30 hrs 24°C 25°C
08.30 hrs 28°C 27°C
10.30 hrs 33°C 31°C
12.30 hrs 36°C 35°C
14.30 hrs 38°C 37°C
16.30 hrs 35°C 34°C
18.30 hrs 31°C 31°C
20.30 hrs 28°C 27°C
22.30 hrs 25°C 26°C

Fig. 20: Graph showing hourly temperature variation with specimen It is seen from Figure 22, that when Specimen 4 with
20 percent phase change material is used there is a
Experiment carried out with Specimen 3 on 18.2.2015 temperature difference of 1oC is obtained in peak hour
temperature. When compared to the previous trial this
Specimen 4 has lesser effect. This is because as the
Table 12
Temperature readings taken in the cubicles (specimen 3) percentage of Phase change material is increased the
encapsulation efficiency decreases. Hence there is lesser
Temperature in Temperature in moderating effect by the Specimen 4 when compared to
Time in hours
room without PCM room with PCM Specimen 3.
 06.30 hrs 24°C 26°C
 08.30 hrs 29°C 28°C
 10.30 hrs 33°C 32°C
 12.30 hrs 36°C 34°C
 14.30 hrs 37°C 35°C
 16.30 hrs 34°C 32°C
 18.30 hrs 32°C 30°C
 20.30 hrs 27°C 28°C
22.30 hrs 25°C 27°C

It is seen from Figure 21, that when Specimen 3 with

15 percent phase change material is used there is a Fig. 22: Graph showing hourly temperature variation with specimen 4
temperature difference of 2oC is obtained in peak hour
Organised by
India Chapter of American Concrete Institute 559
Session 5 B - Paper 4

Cost Analysis Acknowledgments

The general cost of 1kg of gypsum in the form of plaster We, the authors heartfully thank the Management,
of paris available in market is Rs.12. The cost of paraffin Principal, Deans and HOD's of various departments,
is Rs. 250 per kg. The cost of a 0.76 x 0.46 m board will Teaching, Non - teaching staffs and student friends for
be around Rs.430per m2 (approx.). The standard size their support in successful completion of the project.
of gypsum boards available for plastering in market is
1.219m x 1.819m costs about Rs. 250 per square metre. References
Thus the micro encapsulated gypsum board of a standard 1. Babich,M.W., Benrashid,R., and Mounts,R.D., “DSC studies of
energy storage materials”, Thermal and flammability studies Act ,
market size would cost around Rs. 1500 (approx.) per V.3,No.243, 1994, pp.193-200.
square metre. More over the cost of running a 1.5 tonne 2. Khudhair.A.M., Farid.M.M., “A review on energy conservation in
A/C for around7hours at 23°C during summer costs about building applications with thermal storage by latent heat using
Rs. 2500 per month. The cost of the gypsum encapsulated phase change materials”, No.45, 2004, pp.263-275.
board is little higher compared to normal gypsum 3. Mohammed M.Farid and SaidAl-Hallaj., “A review on phase
boards. But the cost due to air conditioning and electricity change energy storage: materials and application” Conversion and
Management, No.45, 2004, pp.1597–1615.
consumed for using fans can be reduced as the gypsum
4. Cabeza,L.F., Castellón,C.M., and Nogués.M., “Use of
board is a one-time investment. In the long run, this will microencapsulated PCM in concrete walls for energy saving”,
prove to be more cost effective and environment friendly. Energy and Buildings, No.39, 2007, pp.113-119.
5. Jamekhorshi.A., Sadrameli.S.M., Farid.M., ”A review of
microencapsulation methods of phase change materials (PCMs) as a
Conclusions thermal energy storage (TES) medium”, Renewable and Sustainable
From the results taken from the above conducted Energy Reviews , No.31, 2014, pp.531-542.
research, the following conclusions can be derived, 6. IS 2095: 2011 (PART I) – Plain Gypsum plaster boards.
7. Felix Regin,A., Solanki,S.C., and Saini,J.S., “An analysis of packed
1. Phase change materials are capable of storing and bed latent heat thermal energy storage system using PCM capsules:
releasing heat at the desired temperature Numerical Investigation”, Renewable energy, No.34,2009, pp.1765-
1773.
2. They can be incorporated in dry walls by encapsulation
8. Ruben Baetens., Bjorn Petter Jelle., Arild Gustavsen.,“Phase change
to moderate the temperature of rooms in a building to materials for building applications: A state-of-the-art-review”,
an extent. Energy and Buildings, No.42,2010, pp.1361-1368.
9. Shapiro.M., Feldman.D., Hawes.D., and Banu.D., “PCM thermal
3. Paraffin wax can be used as a Phase change material
storage in drywall using organic phase change material”, No.4,1987,
due to its high latent heat capacity pp.419-438.
4. Paraffin wax when encapsulated in poly vinyl alcohol 10. Zalba.B., Marín.J.M., Cabeza.L.F., Mehling.H.,“Review on thermal
energy storage with phase change materials”, No.23,2003, pp.251.
shells shows good temperature moderation
11. IS 2547:1976 (Part I) – Gypsum building plaster.
5. Gypsum boards with 15% PCM are able to bring out a 12. IS 12679: 1989 – By product gypsum for use in plaster, blocks and
temperature moderation of about 2oC boards.

6. The efficiency of the encapsulation decreases if the 13. Zhao.C.Y., and Zhang.G.H., ”Review on Micro-encapsulated Phase
Change Materials (MEPCMs): Fabrication, Characterization and
PCM content is increased beyond 15% Applications”, Renewable and Sustainable Energy Reviews,No.1,
2011, pp.3813-3832.
7. The cost of air conditioning and heating are reduced
drastically. 14. http://www.smitha-iitd.com/research-highlights/encapsulation.
15. Leppers.R., “Development of Smart Microencapsulated Phase-
8. The PCM incorporated gypsum boards are environment Change-Materials for Enhancing Heat Storage (concrete
friendly. compositions)” 2005.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

560 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Paraffin in Drywalls as Sensible Heat Storage Material for Temperature Moderation

Rampradheep G.S.
Rampradheep G.S. has completed his graduation in Civil Engineering from Kongu Engineering College,
Perundurai, Erode, Tamil Nadu, India in the year April 2008 and completed his Masters in Structural
Engineering from Bannari Amman Institute of Technology on May 2010. He was awarded two Gold Medals
for his outstanding performance in Structural Engineering. He pursuing his Ph.D in the field of Concrete
under the Anna University, Chennai. He has published more than 14 papers in reputed conferences and a
paper in reputed journal. His project Electricity Generation from cement matrix incorporated with self-
curing agent has been awarded the second best project by the IEEE, Consumer Society, Bangalore. He
was the recipient of various awards and his team has obtained two national awards in the year 2014 and
2015. He has successfully awarded and completed a research and seminar grant over a lakh from various
agencies like AICTE, CSIR, TBI-KEC etc., At present, he is working as an Assistant Professor in Kongu
Engineering College, Perundurai, Erode Tamil Nadu, India.

Dr.Sivaraja M.
Dr.Sivaraja M. has completed his graduation in Civil Engineering from Madurai Kamaraj University and
Masters in Structural Engineering from Periyar University. He completed his Ph.D from Anna University,
Chennai during 2008. He has done his Post Doctoral Research at “University at Buffalo”, The State
University of New York, USA on “Multifunctional Cementitious Composites” during 2009 under a prestigious
scholarship called “BOYSCAST Fellowship” given by DST, New Delhi. He has published more than 25
papers in reputed journals and 60 papers in conferences. He has successfully awarded and completed a
research grants over 50 lakhs from various agencies like DST, AICTE, IEI, Coir Board of India etc., He is the
reviewer of reputed journals. At present he is working as a principal at N.S.N College of Engineering and
Technology, Karur, Tamil Nadu.

Organised by
India Chapter of American Concrete Institute 561
SESSION 5 C
Session 5 C - Paper 1

Assessment of Fresh Properties of Cementitious Grouts Used for

Post-Tensioning Applications in India

Suruthi Kamalakkannan, Radhakrishna G.Pillai Manu Santhanam

Ramya Thirunavukkarasu Assistant Professor Professor
Former Graduate Student Department of Civil Engineering, IIT, Department of Civil Engineering, IIT,
Department of Civil Engineering, IIT, Madras, Chennai-36, India. Madras, Chennai-36, India.
Madras, Chennai-36, India.

Abstract 1950’s and 60’s, pre-stressed concrete, especially post

tensioning, has found wide application in the infrastructure
Post-tensioning is one of the most widely used
development of the world. The lack of insight into the
technologies in many major infrastructure projects. The
potential problems associated with these structures in the
tendons are the ‘lifeline of Post-Tensioned structures’
early ages had led to premature deterioration issues. This
and grouts act as their ‘last line of defense system’, filling
becomes more critical in the Indian scenario, where major
the interstitial spaces between the ducts and strands.
infrastructure development is being planned currently.
Unfortunately, significant voids were found in the ducts
For instance, metro rail projects covering 528 km and
of many bridges worldwide, leading to the premature
involving nearly 3750 tonnes of cementitious grout have
failure of tendons. To ensure complete filling of ducts,
been initiated in India (Kamalakkannan, 2015; NDM&CW
the grouts should have excellent fluidity and pumpability
59, 2015; MOSPI, 2014). In the United Kingdom (UK), a ban
with no segregation, retention, bleeding, and dimensional
on the construction of PT bridges was imposed following
stability. Such pre-packaged, high performance grouts
the failure of Bickton Meadows Footbridge, UK in 1967 and
are widely available in the market abroad (say, developed
Yny-y-Gwas Bridge, UK in 1985 (FHWA, 2013). The post-
world) and not in India. In addition, the many design/
collapse investigations revealed that many post-tensioned
construction engineers in India are unaware of such high
bridges had corroded tendons and severe wire fractures.
performance grout materials, leading to possibly poor
The reason for corrosion was the ingress of water and
quality construction in India. Therefore, there is a dire need
chloride ions (from de-icing salts) into the tendon system
to evaluate (1) the presence of voids in existing tendons
through air voids resulting from poor grout and grouting
and (2) the performance of currently used grouts in India.
practices (Clark, 2013). The ban brought to light the
This paper focuses on the latter need on, improving
potential problems and triggered investigations in other
grouting practices and grout materials by analysing the
countries. The most dominant failure mechanism was
performance of commercial grouts and the influence of
corrosion of the stressed tendons inside poorly grouted
site-dependent parameters on it. A two stage test program
ducts. A history of problems has been reported, due to
was carried out to evaluate seven types of commercially
poor workmanship or quality control during construction.
available grouts, which includes Pre-Packaged Grout
These left the ducts containing pre-stressing tendons not
(PPG) and Plasticized Expansive Admixture (PEA) grout.
fully filled, resulting in voids in the grout where the steel
The study serves as strong evidence in substantiating
is unprotected (VSL, 2002). Entry of contaminated water
that, the most commonly used materials for PT system
or water containing de-icing salts into these partially
grouting in India fail to meet the stringent requirements
grouted tendon ducts can lead to chloride and/or moisture
of standards. They also do not meet the specifications
induced corrosion. Thus, improvements in design details,
given by the manufacturers themselves. It is evident that
materials, and grouting practices become critical in
the use of these materials based on the reliance of data
decreasing the likelihood of corrosion in PT tendons and
sheets are unwarranted and requires careful testing and
anchorages in future projects.
evaluation based on local conditions. The properties and
performance of the grout mix prepared are influenced by
a combination of site-dependent parameters like ambient Limitations of PT Grout Materials
temperature and speed of mixing. Grout is the last layer of protection system in direct contact
Key words: Cementitious grout, post-tensioning, fluidity, with the stressed steel. Thus, poor grout material can be
bleed resistance, mixing speed, pre-packaged grouts, a significant weakness of PT system. Efforts are focused
plasticized expansive grout admixture. on improving it so as to ensure a grout system without any
discontinuities (i.e., voids). Presence of voids in the duct
system can have significant influence on corrosion rate
Introduction (Pillai, 2009). The reasons for the formation of these voids
Beginning with the construction of many long span include bleed-water evaporation, poor grouting, materials
bridges in the Europe and United States (US) during and construction practices, or a combination of these

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

564 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Assessment of Fresh Properties of Cementitious Grouts Used for Post-Tensioning Applications in India

(Schupack, 2004). In India, Ordinary Portland Cement Limitations of Existing Code

(OPC) grouts with water-binder ratios (w/b) ranging from
MORTH (Ministry of Road Transportation and Highways)
0.40 to 0.44 and site-mixed plasticized expansive grout
specifications are widely followed in India for all post
admixtures (PEA) are still being used commonly. Past
tensioning works. It can be observed that other than
failure experiences have led to the development and
laying down basic grout requirements, the performance
use of high-performance pre-packaged (HPG) grouts in
parameters are not specified in MORTH. Overall, most
developed countries as a norm (FHWA, 2013). However,
of the specifications followed throughout the world, are
they are not available in India. Further, these HPGs are also
not stringent enough to ensure high performance. For
found to have various quality issues that widely affect their
instance, there are no specific considerations for vertical
performance and, performance of some are no better than
grouting where high static water pressure would be
Ordinary Portland cement grout without any admixtures
developed leading to segregation and bleeding. Also, there
(Tritthart et al., 2006). There are major concerns
is no proper correlation between the mixing times given,
regarding the variations encountered in manufacturing
with the mixing methodology and mixing speed. These are
and during field applications, as contamination can occur
major influencing factors on cement rheology and hence,
during various stages of operations (Emmons et al., 2014).
the grout’s fresh property performance. Further, the test
Also, there are not much clear descriptions in the product
methods and acceptance criteria used for grouts are not
specifications and many do not even comply with all the
representative of the real performance of grout inside
internationally accepted test standards (Table 2). Hence,
a tendon duct and are not able to correctly differentiate
it is necessary to test even these products that are sold
between good and poor quality grouts. All these emphasize
exclusively for PT grout preparation.
the need for a holistic specification requirement, to ensure
high quality grout system.
Site Issues
Although many improvements could be made through Critical Fresh Properties of Grouts for PT
research in materials, design and technology, site Applications
practices are critical factors for actual implementation.
This is very important considering that many of the Flowability or fluidity - fluidity controls the ability of grout to
techniques developed in laboratories under ideal situations penetrate fine interstitial spaces and can be determined by
are different from those encountered at sites. Ambient Marsh funnel, Mini-slump, etc. (Hatem et al., 2014). Fresh
temperature, mixing water and grout temperature has grouts should be easily pumpable, and hence should have
great impact especially on the fresh grout properties (Bras low viscosity. Flowability is dependent on the rheology of
et al., 2013; Mirza et al., 2013) and hence is stressed widely cement paste and is important to ensure complete filling
in codal provisions (EN 445:2007). Grout mixes subject to of tendon ducts containing congested strands.
elevated temperature show rapid changes especially in Bleed - Generally, water is added to grout in excess of
fresh property behaviour (Kamalakkannan, 2015, Mirza that needed for hydration, to provide sufficient viscosity
et al., 2013). Grout has to be optimized for the expected for injection. In such situations, the cement particles
range of temperatures and minimum water quantity. Grout flocculate and settle under gravity, but, the lighter water
mixing procedure is another important factor affecting its moves up and get collected at the top called bleed water.
rheological properties as, sometimes, the failure in the This results in anisotropic behaviour of the grout material
performance of grout is not attributed to the grout itself (Tan et al., 2005). This excess water gets collected at
but the mixing procedure (Yahia, 2011; Tritthart et al., 2006; high points of tendon profiles, and upon evaporation or
Toumbakari et al., 1999). If the grout is mixed properly, absorption by surrounding grout material, exposes the
the number and size of cement particles agglomerating stressed steel strands without alkaline grout protection.
decrease and hence the grout becomes more fluid. It also leads to weakness, porosity and lack of grout
However, longer mixing times do not significantly improve durability (Jamal, 2002). It is usually tested by Standard
the grout fluidity (Nguyen et al., 2011). Dispersion of Bleed test (SB) and wick-induced bleed test (WB), which
particles depends on time or duration of exposure, power takes into consideration, the capillary action of water
and speed (i.e., frequency) of mixing. There exists an through interstitial spaces in the tendons.
optimal frequency for which dispersion is the best and, if
Grouting of long vertical tendons is associated with the
the particle size increases, the frequency corresponding
risk of excessive bleed, segregation and sedimentation
to the maximum dispersion is moved towards lower
due to considerable pressure differential between top and
frequencies (Toumbakari et al., 1999). The influence of
bottom elevations (Schokker et al., 2002). Pressure Bleed
mixing time, type and speed on the grout properties is
test (PB) is carried out to evaluate suitability of grouts
evident from the emphasis laid in codal specifications. The
for such applications. (refer Kamalakkannan, 2015 for
use of similar mixing equipment and techniques that are
detailed information on SB, WB and PB)
representative of the site for enabling satisfactory on-site
performance are stressed (EN 445: 2007).

Organised by
India Chapter of American Concrete Institute 565
Session 5 C - Paper 1

Research Significance grout were mixed at speeds of 500, 1500 and 2500 rpm,
were prepared in this phase. These were studied under
In the past, corrosion induced failures of tendons have
two controlled temperature conditions of 15°C and 30°C
been attributed to the formation of voids, resulting from
(60% ± 10% RH). Here, test parameters corresponding to
inadequate grouting and poor grout materials and grout
Table 1 were measured for these mixes and compared, to
formulation. PPGs were developed to counter these
investigate the effect of mixing speed and temperature.
problems in the developed world. However, these are not
Three batches of grouts were tested in every phase. In
even accessible in India. But, the drastic infrastructure
Phase 2, a standard procedure was laid down based on
development planned in India, demands a large volume of
recommendations from ASTM C305, literature studies
grouting (nearly 3750 tonnes). On these lines, the present
(Bras et al., 2013; Sonebi, 2010; Nguyen et al., 2007)
study indicates that the grouts widely used in India for
and site practices in India. A custom- made high shear
post-tensioned systems do not perform satisfactorily,
mixer was designed, fabricated and used for the study
to ensure durable PT systems. This serves as a strong
(Kamalakkannan, 2015).
indicator to the grout manufacturers and PT industry
regarding the potential disaster and the quality of systems
in place and also the need for better PPGs. Table 1
Phases of testing

Experimental Program Phase Test Test parameter

Reference
standard adopted
This study involved two phases of testing. The first phase Fluidity or
evaluated the commercially available materials used Flow cone EN 445 : 2007
flowability
for PT system grouting in India. For this study, plain
Mini slump cone Viscosity EN 445 : 2007
OPC 53 Grade + 0.4w/c grout, three PEA grouts (used in
major infrastructure projects in India), and three PPG’s Phase 1 Standard bleed Bleed ASTM C940 : 2010
(including high performance grouts used in developed Wick-induced EN 445 : 2007
Bleed
countries and those that have been recently developed bleed (modified)
in India) were used. The results obtained were checked Pressure bleed
Bleed in vertical ASTM 1741 : 2012
for their compliance with specifications laid by both grout applications (modified)
manufacturers and codal standards. The second phase One PEA grout and one PPG grout were selected from
evaluated the effect of site-dependent parameters, such Phase 1 and evaluated for performance under 3 mixing
Phase 2
speeds and 2 temperature conditions (fresh property tests
as mixing speed and temperature, on the critical fresh are same as in Phase 1)
properties of grout. In the first phase, seven commercially
available grout materials were subjected to fresh property
testing as listed in Table 1 and their compliance with the Performance of Commercial Grouts – Results
manufacturers’ specifications and codal standards were from Phase 1 tests
checked. Grout mixes were prepared under controlled The results obtained were compared with the specifications
temperature conditions of 25°C, and manufacturers’ given by the corresponding grout manufacturers and the
recommendations on mixing procedure, duration and standard recommendations listed in Table 2. Most of the
speed were followed. materials failed in the bleed resistance parameter, which
Based on the performance in Phase 1, two comparatively is critical from the durability point of view, as it is an
better grout materials, one each from PEA and PPG indication of probability of void formation. PPG 1 and PPG
were selected for the second phase testing. Three 3 were found to have better performance as per Table
batches of grout mixes, in which the constituents of 2. However, their efflux time was very high. Currently,

Table 2
Status of compliance of commercial grouts with standards and manufacturer’s specifications on fresh properties

Manufacturers' Specifications Compliance***

Specifications from
Property / Test PPG OPC PEA PPG
standards*
1 2 3 1 2 3 1 2 3
Flow / efflux time ≤ 25 s (After 10 min.) 7 to 20 s - -     X  X
Mini slump > 140 mm - - -       
Standard bleed 0% at 3 hrs 2% at 3 hrs - - X X X X  X 
Wick induced bleed ≤ 2% 0 % at 4 hrs - - X X X X  X X
Pressure bleed  ≤ 2% at 50 psi 0% at 100 psi - - X X X X *** - X
Values correspond to those standards identified to be stringent among those specified in EN 445:2007, PTI, FIP, JSC and MORTH (Mishra et. al, 2004). Results comply
* *

with standards but not manufacturer specifications. ***Compliance of test results observed in this study with both standard and manufacturers’ specification

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

566 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Assessment of Fresh Properties of Cementitious Grouts Used for Post-Tensioning Applications in India

efflux time and spread by mini-slump are the main or only happens at a faster rate at higher temperatures.
parameters measured in Indian construction sites. The Signature patterns of variation were followed in PEA and
results show that it is not appropriate to use efflux time as PPG at different temperatures for all speed variations
the only deciding parameter in the selection of grouts for as evident from Fig.1 (a) and Fig.1 (b). Understanding the
post tensioning. reason behind this pattern requires greater depth of study
in areas of rheology and cement chemistry. In general, a
steep increase in efflux time is observed one hour after
Influence of Mixing Speed and Temperature
the preparation of grout mix in both categories of material.
on Fresh Properties (Results from Phase 2 This is of great importance in deciding the pot life of the
tests) grout mix at a particular environmental condition and
PEA 2 and PPG 1 that were better performers from their mixing speed. The PPG at 30°C stiffens and stops flowing
respective categories in Phase 1 were selected for further through the Marsh cone by the third hour of time elapsed,
detailed analysis in Phase 2. at all mixing speeds.
At a mixing speed of 2500 rpm, the flow at third hour
Effect on fluidity stopped for both PEA and PPG at 30°C and 15°C
The influence of mixing speed and temperature on the respectively.
efflux time as a function of time is shown in Fig.1 (a)
The variation of flow spread by mini slump test, as a
and Fig.1 (b). It indicates that, the efflux time decreases
function of mixing speed for the mixes prepared under
with increase in mixing speed for both PEA and the PPG
15°C and 30°C is shown in Fig.1 (c). It is in correlation
selected for study. While the change is significant when
with the results on efflux time (Fig.1 (a) and Fig.1 (b)), but
the speed is increased from 500 to 1500 rpm, it is not
the spread of PPG is significantly higher than that of the
much appreciable when increased further from 1500 to
PEA due to its thixotropic behaviour. While the variation
2500 rpm. This may be attributed to the better dispersion
in flow spread at higher temperature is not much until
of particles at high speed. Also, as a common trend, the
mixing speed of 1500rpm for both materials, the spread
initial efflux time at higher temperature (i.e., 30°C) is
decreases significantly at 2500rpm at 30°C for PPG. The
lesser than that at 15°C. However, as time elapses, the
reason for this behaviour could not be identified within the
efflux time at 30°C becomes higher because hydration
scope of this research and hence requires further study.

a. Influence of mixing speed and temperature on b. Influence of mixing speed and temperature on c. Influence of mixing speed and temperature
efflux time of PPG on flow spread

d. Influence of mixing speed and temperature on e. Influence of mixing speed and temperature on WB f. Influence of mixing speed and temperature
cumulative SB at 3 hours of elapsed time at 3 hours of elapsed time on pressure sustained (solid line) by the
grout and pressure bleed (PB) (dashed line)

Fig. 1: Influence of mixing speed and temperature on critical fresh properties of grout

Organised by
India Chapter of American Concrete Institute 567
Session 5 C - Paper 1

Effect on bleed resistance The important conclusions from the research are,
The cumulative bleed at three hours as a percentage of 1. The most commonly used PT grout materials used
initial grout volume, increases with increasing mixing in India and some of the PPGs that are recently being
speed at 15°C for the PEA grout, under both standard and manufactured in India, fail to meet the requirements
wick- induced conditions as shown in Fig.1 (d) and Fig.1 of standards, and even the specifications given by the
(e) respectively (this may be because of the slow reaction manufacturers themselves.
rate at lower temperature). While this increase is not very
appreciable, it is observed that, there is a drastic decrease 2. Efflux time shall act as a quality control parameter to
in bleed (both SB and WB) with increasing speed at 30°C assess pumpability and fluidity but not as a screening
which shall be attributed to the increased reaction rates factor in the final selection of grout materials.
with increase in temperature. In case of the PPG tested, 3. The performance of any grout material can be highly
the cumulative bleed for three hours (both SB and WB), influenced by site-dependent parameters like ambient
in general, decreases with increasing mixing speed under temperature and speed of mixing.
both the temperature conditions, which, may be due to
its higher fineness when compared to the cement used 4. The optimal mixing speeds for the PEA and PPG
in PEA mix. However, the reduction is drastic at 30°C, studied is 1500 and 2500 rpm, respectively. This clearly
which is attributed to the faster reaction rate at higher shows that every grout material has optimal values for
ambient temperature. There is no bleed at 30°C for the various mixing variables leading to enhanced/desired
PPG mixed at a speed of 2500 rpm under both standard performance.
and wick-induced conditions. The reason for zero bleed of
the material at this temperature when mixed at 500 rpm Future Work
is not known. Further investigations on the influence of other variables
In the context of vertical grouting application, Fig.1 (f) like, mixing time, mix volume, mixer type, mixing
shows the trend of pressure sustained (represented by procedure and mechanism, etc., on the properties of
solid line) and the corresponding bleed (represented grout is in progress. A matrix can be developed based
by dashed line) at different mixing speeds for the two on this for arriving at optimal parameters (mixing speed,
different temperatures under study. It could be observed time, temperature, volume) to attain optimal grout
that, in general, the pressure sustained for the particular performance (in terms of critical properties like bleed
grout mix remains almost the same across varying mixing resistance). Further, real time prototype testing needs
speeds and temperatures, but there is a change observed to be conducted to substantiate the results of this study
in the bleed values. Bleed of PPG is very less when - ultimately leading to the development of guidelines and
compared to that of PEA, irrespective of temperature revisions to the codal specifications.
and mixing speeds. Also, it is observed that there is no
significant change in the bleed of PPG with speed and Acknowledgements
temperature variations. However, bleeding is the lowest
at 2500 rpm. In the case of PEA, as observed from Fig.1 This research was performed at the Department of Civil
(f), there are significant changes in the bleed with respect Engineering, Indian Institute of Technology Madras,
to temperature and mixing speed. At 15°C, the bleed Chennai, India. The authors acknowledge all the lab
increases with increasing mixing speed. The bleed at assistants and technicians of the BTCM Division, Dept.
30°C is higher than that at 15°C for 500 rpm. However, of Civil Engineering, in IIT Madras for helping with all the
it gradually reduces at higher speeds with no significant experimental works and the Central fabrication Facility
difference between 1500 rpm and 2500 rpm and finally for the fabrication of grout mixer. The authors also extend
falls below that observed for 15°C. This may be owed to their gratitude towards Ministry of Human Resources
the temperature stabilisation and difference in reaction Development, Govt. of India and Larson and Toubro,
rates as discussed in previous sections. ECC division for financially supporting this study and all
the other organisations for providing with the materials
required for the research.
Summary and Conclusions
Many durability issues dealing with the failure of PT References
tendons have been witnessed in the last 20 years which 1. Bras A., Giao, R., Lúcio, V., Chastre, C., 2013, Development of an
has led to realisation over the quality of grout system in injectable grout for concrete repair and strengthening, Cement
& Concrete Composites 37, pp.185–195. British Standards EN
place. This research provides an insight into this issue 445:2007, Grout for pre-stressing tendons,-Test methods, Part 3.
especially with respect to the Indian scenario by assessing
2. Clark, G.M., 2013, Post-Tensioned Structures – Improved Standards
the fresh properties of seven commonly used PT grouts in Based on Lessons Learnt, Fédération Internationale Du Béton (fib)
India. Further, it also emphasises the importance of site conference, Chennai.
parameters that could possibly affect these properties. 3. Corven J., P.E., Alan Moreton, P.E, FHWA 2013, Post-Tensioning

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

568 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Assessment of Fresh Properties of Cementitious Grouts Used for Post-Tensioning Applications in India

Tendon Installation and Grouting Manual”, Federal Highway 11. Schokker, A.J., Breen, J.E. & Kreger, M.E., 2002, Simulated Field
Administration (FHWA), US Department of Transportation, Testing of High Performance Grouts for Post-Tensioning, Journal
URL:http://www.fhwa.dot.gov/bridge/construction/pubs/hif13026. of Bridge Engineering-7, pp.127-133
pdf, FHWA-NHI-13-026.
12. Schokker, A. J., Hamilton, H. R., & Schupack, M., 2002, Estimating
4. Jamal, S.M., 2002, High-performance cementitious grouts for Post-Tensioning Grout Bleed Resistance Using a Pressure-Filter
structural repair, Cement and Concrete Research 32, pp. 803–808 Test, Pre-stressed Concrete Institute (PCI) Journal, 47(2), pp.32-39.
5. Kamalakkannan, S., 2015, Evaluation of commercially available 13. Sonebi, M., 2010, Optimization of Cement Grouts Containing Silica
post-tensioning grouts and assessment of mixing variables, M.Tech Fume and Viscosity Modifying Admixture, Journal of Materials in
thesis, BTCM division, Department of civil engineering, IIT Madras, Civil Engineering, Vol. 22, No. 4, American Society of Civil Engineers
India
(ASCE), ISSN 0899-1561/ 2010/4-332–342.
6. Mirza, J., Saleh, K., Langevin, M.A., Mirza, S., Bhutta, M.A.R.,
14. Tan, O., Zaimoglu, A.S., Hinislioglu, S., Altun S., 2005, Taguchi
Mahmood M. Tahi, 2013, Properties of microfine cement grouts
approach for optimization of the bleeding on cement-based grouts,
at 4°C, 10°C and 20°C, Construction and Building Materials 47,
Tunnelling and Underground Space Technology 20, pp.167–173.
pp.1145–1153.
7. Mishra, S., Kumar, R., and Kumar, S., 2004, A review of test methods 15. Toumbakari, E.E., Van Gemert, D., Tassios, T.P., Tenoutasse, N., 1999,
and specifications for grouts in post-tensioned construction, The Effect of mixing procedure on injectability of cementitious grouts,
Indian Concrete Journal, pp.39-46. Cement and Concrete Research 29, pp.867–872.

8. Nguyen, V.H., Remond, S., Gallias, J.L., 2011, Influence of cement 16. Tritthart, J., Stipanovic, I., Banfill, P.F.G., 2006, Grouts for Pre-
grouts composition on the rheological behaviour, Cement and stressed structures and Ground anchors: A critique of European
Concrete Research, pp.281- 292. Standards and Guidelines, Proceedings of the International
Conference on Bridges-2, pp.639-646.
9. Emmons, P.H., Goodwin, F.R., and Sprinkel, M.M., 2014, Quality of
Prepackaged Powdered Materials Used in Construction - What can 17. VSL International Ltd., 2002, Grouting of Post - Tensioning Tendons,
be done to improve it?, Concrete International, pp.43-47. VSL Report Series-5, Switzerland
10. Pillai, R.G., 2009, Electrochemical Characterization and Time-Variant 18. Yahia, A., 2011, Shear-thickening behavior of high-performance
Structural Reliability Assessment of Post-Tensioned, Segmental cement grouts - Influencing mix-design parameters, Cement and
Concrete Bridges, PhD thesis, Texas A&M University, Texas. Concrete Research 41, pp.230–235.

Radhakrishna G. Pillai, Ph.D.

Current Position: Assistant Professor, Building Technology and Construction Management (BTCM)
Division, Department of Civil Engineering, Indian Institute of Technology Madras, Chennai – 600 036, India
Dr. Radhakrishna G. Pillai earned his Ph.D. in Civil Engineering from Texas A&M University, College
Station, Texas, USA. For the past 5 years he is serving as an Assistant Professor in the Department of
Civil Engineering at IIT Madras, Chennai, India. His current research interests include corrosion control
in concrete structures, service life estimation, repair and rehabilitation of structures, and grouting of post
tensioned systems.

Organised by
India Chapter of American Concrete Institute 569
Session 5 C - Paper 2

Study on Mechnical and Durability Properties of Recycled Aggregate

Concrete Incorporated with Silica Fume and Mineral Quartz

Anand K. Darji Dr. Indrajit N Patel Mrs. Jagruti Shah

Structural Engineering Dept., Professor, Structural Engineering Project Coordinator, Gujarat
BVM Engg. College, V.V. Nagar, Dept., BVM Engg. College, V.V. Nagar, Technological University,
Gujarat, India. Gujarat, India. Gujarat.

Abstract concrete will decreases but at the same time if we are

Disposal of construction waste is now new challenge for add the SCMs like silica fume and mineral quartz the
the construction industry in this era. This is peak time strength will increases.
to use Construction waste as recycled aggregate (RA) in Silica fume is an amorphous type of silica dust collected
concrete manufacturing for sustainable development. in bughouse filters as a by-product from manufacturing
Recycled aggregate concrete (RAC) is the future to save silicon metal or ferrosilicon alloys[3]. The smoke that
environment from the waste. Supplementary Cementing results from furnace operation is collected and sold as
Materials (SCMs) are widely used these days to improve silica fume, rather than being land filled. Silica fume is
the durability of concrete. Silica fume has gained widely used in concrete and refractory application. One
worldwide acceptance due to its high pozzolanic reactivity of the most beneficial uses for silica fume is in concrete.
compared to other SCMs. Using mineral admixtures Due to its unique chemical and physical properties, silica
as cement replacement substance in concrete has a fume has become a versatile mineral admixture for a
tendency to increase by the future in order to provide multitude of applications. Silica fume consists primarily of
greater sustainability in construction industry. On the amorphous (non-crystalline) silicon dioxide (SiO2). Silica
other hand Quartz is the second most abundant mineral fume is an ultrafine material with spherical particles
in the Earth's continental crust, after feldspar. It is used
less than 1 μm in diameter, the average being about
as SCM in concrete. In this study, replacements of cement
0.15 μm[2]. The individual particles are extremely small,
with silica fume 4%, 8% and12% & mineral quartz 5%
approximately 1/100th the size of an average cement
for concrete mix ofM35 and M40 grade. The natural
particle silica fume. Because of its chemical & physical
aggregate is replaced by recycled aggregate (RA) with
properties, like high silica content & extreme fineness,
30%. This paper study conducted on water absorption and
silica fume is a very effective pozzolanic material. It has
sorptivity test on recycled aggregate concrete (RAC). The
experiment result analysis shows that durability of 8% SF High early compressive strength, high tensile flexural
and 5% Quartz are better than other replacements. strength, low permeability to chloride and water intrusion,
enhanced durability, Increased toughness, higher bond
Keywords: Silica Fume (SF), Mineral Quartz, Recycled strength.
Aggregate Concrete (RAC), Supplementary cementing
materials (SCMs). Quartz is the second most abundant mineral in the
Earth's continental crust, after feldspar. It is made
up of a continuous framework of SiO4 silicon–oxygen
Introduction tetrahedral, with each oxygen being shared between two
RA is derived from the processing of materials tetrahedral, giving an overall formula SiO2. There are
previously used in a product and/or in construction. many different varieties of quartz, several of which are
Examples include recycled concrete from construction semi-precious gemstones. Especially in Europe and the
and demolition waste material (C&D). In order to reduce Middle East, varieties of quartz have been since antiquity
the usage of natural aggregate, RA can be used as the the most commonly used minerals in the making of
replacement materials. RA is comprised of crushed, jewelry and hard stone carvings. Quartz is an essential
graded inorganic particles processed from the materials constituent of granite and other felsic igneous rocks. It is
that have been used in the constructions and demolition very common in sedimentary rocks such as sandstone
debris. These materials are generally from buildings, and shale and is also present in variable amounts as an
roads, bridges, and sometimes even from catastrophes, accessory mineral in most carbonate rocks. It is also
such as wars and earthquakes. RA concrete is generally a common constituent of schist, gneiss, quartzite and
reported to be between 15% and 40% weaker than other metamorphic rocks. Because of its resistance to
natural aggregate concrete[1].as the replacement of weathering it is very common in stream sediments and
RA with natural aggregate increases the strength of
in residual soil.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

570 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Study on Mechnical and Durability Properties of Recycled Aggregate Concrete Incorporated with Silica Fume and Mineral Quartz

Literature Review Mineral Quartz

L. Evangelista, J. de Brito[8] investigated that use of fine
recycled concrete aggregates to partially or globally Table 2
Properties of Mineral Quartz
replace natural fine aggregates in the production of
structural concrete. The compressive strength does not Chemical Characteristics Units Micro Silica
affected by the fine aggregate replacement ratio, at least
for up to 30% replacement ratios.Tensile splitting and Mineral quartz % 99%
modulus of elasticity are reduced with the increase of the
SiO2 % <1
replacement ratio but both properties are still acceptable
up to 30% replacement. Specific Surface Area Micron 400

Ozgur Cakır, Omer Ozkan Sofyanlı studied the effects of

[9]

incorporating silica fume (SF) in the concrete mix design Cement

to improve the quality of RA in concrete. Portland cement
was replaced with SF at 0%, 5% and 10%. Specimens were Brand Name: Hati
manufactured by replacing natural aggregates with RA. Grade of cement: 53 Grades (IS12269:1987)
The compressive strength decreased with increase in the Type of cement: Ordinary Portland cement
silica fume content. The concrete with 10% SF and having
4/12 mm fraction RA showed a better performance than
the other concrete series in terms of the physical and the Table 3
Physical properties of cement
mechanical properties.
R. Sathish Kumar [10] conducted experimental study to Sr. Test
Test Name Method of Test Specification
No Results
find the suitability of the alternate construction materials
such as, rice husk ash, sawdust, recycled aggregate 1 Consistency (%) IS 4031: Part-4 30% --
and brickbats as a partial replacement for cement and
conventional aggregates. The strength of rice husk Initial Setting Shall not < 30
2 IS 4031: Part-5 90min
Time Min
ash concrete was found to be in the range of 70-80% of
conventional concrete for a replacement of cement up Final Setting Shall not >600
3 IS 4031: Part-5 178min
to 20%. The study shows that the early strength of rice Time Minutes
husk ash concrete was found to be less but the strength
4 Specific Gravity IS 4031 : Part-11 3.15 --
increased with age.
S.C. Kou, C.S. Poon[11] conducted study on use of RA in Fine Aggregate
concrete would reduce its compressive strength and render
the concrete less durable. At the same recycled aggregate Locally available river sand was used as fine aggregate.
replacement level, the use of fly ash as addition of cement The properties of fine aggregate, confirming to IS: 383 –
with 25% at 90 days increased the compressive strength. 1970, are shown in Table 4.

Materials Table 4
Physical properties of fine aggregate
Silica Fume
Sr. No. Particulars Sand
Table 1
Physical & chemical properties of SF 1 Source Bodeli, Gujarat

Chemical & Physical Characteristics Units Micro Silica 2 Zone Zone II (IS: 383-1970)
SiO2 % 85+
3 Sp. gravity 2.6
CaO % 0.94
AL2O3 % 0.61 4 Fineness modulus 3.05
Fe2O3 % 0.31
Loss on Ignition % 2.00
Moisture % 1 Coarse aggregate
Bulk Density Kg/m 3
450-600 Natural aggregate of maximum size 20 mm are taken in
Pozzolanic Activity Index (7 days) % 105+ the study. The physical properties of course aggregate are
Coarse Particles > 45 Microns % 09% shown in table. The sieve analysis of coarse aggregate is
Coarse Particles < 45 Microns % 91% shown in Table. The aggregate were tested as per IS 2386
Specific Surface Area M2/gm 16 (Part: 1, 2, 3) – 1963 and IS: 383– 1970.

Organised by
India Chapter of American Concrete Institute 571
Session 5 C - Paper 2

Table 5 Results Analysis

Properties of Natural Aggregate and RA
Compression test
Sr. No. Particulars Natural Aggregate RA
Test results for the compression test for the different
1 Source Sevalia, Gujarat Anand, Gujarat grade of concrete with the fix proportion of quartz and
different proportion of SF in 150mm * 150mm * 150mm
2 Max. aggregate 20mm 20mm specimens are as follow.
3 Specific gravity 2.86 2.75
Flexural strength test
4 Fineness modulus 6.94 7.31
Flexural strength
is determined on a
Water beam size of 100 mm
Water is an important ingredient of concrete as it actually X 100 mm X 500 mm.
participates in the chemical reaction with cement. Since it In which single point
helps to from the strength giving cement gel, the quantity load method is used
and quality of water is required to be looked into very to determine flexural
carefully. Water cement ratio used is 0.37 for M35 and strength.
Fig. 1: Compression test
0.35 for M40 concrete.

Chemical Admixture
To decide the chemical dosage in the concrete Mix it is
require doing the Marsh Cone test.

Table 6
Properties of Chemical Admixture

Admixture FF-T (VC)

Produced by Sikament

Physical state liquid

Fig. 2: Compression test result for M35 & M40 grade concrete
Colour Transparent mix at 7 days

Density (kg/l) 1.03

Compatibility Use for all cement

Dosage 0.5%-3% of cement weight

Experiment Methodology
Mix design
Based on is 10262:2009 concrete mix proportioning –
guidelines trial mix design with different proportion of
ingredients has been designed. Table 3.1 presents the Fig. 3: Compression test result for M35 & M40 grade concrete
mix at 28 days
design mix proportion for M35 and M40 grade

Table 7
Mix Design

Coarse aggregate
Cement Fine aggregate RA Admixture
(CA)
Mix w/c Ratio
Kg/m3 Kg/m3 Kg/m3 30% of (CA) liter

M35 0.37 399.65 757.31 869.74 372.74 5.994

M40 0.35 422.48 742.04 857.05 367.30 6.297

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

572 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Study on Mechnical and Durability Properties of Recycled Aggregate Concrete Incorporated with Silica Fume and Mineral Quartz

Fig. 4: Compression test result for M35 & M40 grade concrete Fig. 8: Split tensile test result for M35 & M40 grade concrete
mix at 7 days mix at 28 days

Sorptivity Test
In sorptivity test 100mm
x 100mm x 100mm cube
is used. Test results
of the Sorptivity test
for the different grade
of concrete with the
different proportion of
Fig. 5: Flexural strength test silica fume and mineral
Fig. 9: Water Absorption quartz at 28, 91 days
are as follow

Fig. 6: Flexural strength test result for M35 & M40 grade
concrete mix at 28 days
Fig. 10: Water absorption test result for M35 & M40 grade
Split tensile test concrete mix at 28 days
Split tensile strength
determined on cylinder
of size of 150 mm dia.
and 300 mm height.

Water absorption test

For water Absorption
100mm X 100mm X
100mm cube is used
Fig. 7: Split tensile test
to find out water
absorption result,
test result shows with
different grade of concrete with different Percentage of
silica fume and mineral quartz are as follows. Fig. 11: Water absorption test result for M35 & M40 grade
concrete mix at 91 days

Organised by
India Chapter of American Concrete Institute 573
Session 5 C - Paper 2

ll At 12% SF and 5% Qu, Compressive strength is

decreasing 2.69% than 08% of SF and 5% Qu.
ll For the Design mix M35 and M40 Flexural and Split
Tensile test 8% SF and 5% Q gives better performance.
ll Average loss of flexural strength is observed 1.25% for
design mix M35 and M40.
ll In Water Absorption test, water absorb is max 0.58%
and min 0.24% at 28 days.
ll In sorptivity test as the percentage of SF increases
Fig. 12: Sorptivity test sorptivity decreases.
ll The concrete with inclusion of recycled aggregates
can be used for the high value application as it has both
improved engineering as well as durability parameter.
ll Mechanical properties increases with percentage
increase of Silica Fume compare to normal convention
concrete.
ll All the specimen shows an ideal property of impervious
concrete with minimum water absorption.
ll The percentage water absorption decreases as the
grade of concrete increases.
Fig. 13: Sorptivity test test result for M35 & M40 grade concrete
ll Durability of RAC with 8% & 12% Silica Fume and
mix at 28 days
5%Mineral Quartz is higher with compare to normal
RAC.
ll As the % of SF increases the cost will increase.
ll It is observed that the optimum value is achieved for
30% RAC is 8% Silica Fume and 5% Mineral Quartz.

References
1. F. Tittarelli ,M. Carsana” Effect of hydrophobic admixture and
recycled aggregate on physical–mechanical properties and
durability aspects of no-fines concrete” Construction and Building
Materials 66 (2014) 30–37
2. González-Fonteboa Belén” Structural shear behaviour of recycled
Fig. 14: Sorptivity test test result for M35 & M40 grade concrete concrete with silica fume” Construction and Building Materials 23
mix at 28 days (2009) 3406–3410
3. H. Dilbas, M. Simsek” An investigation on mechanical and physical
Conclusion properties of recycled aggregate concrete (RAC) with and without
silica fume” Construction and Building Materials 61 (2014) 50–59
Based on experimental investigation of compressive
4. L. Evangelista, J. de Brito, “ Mechanical behavior of concrete
strength, flexural strength, split tensile strength, water
made with fine recycled concrete aggregates” Cement & Concrete
absorption and sorptivity of concrete, the following Composites 29 (2007) 397–401.
observations are made partially replacement of SF and
5. Ozgur Cakır, Omer Ozkan Sofyanlı, “Influence of silica fume on
Mineral quartz for M35 and M40 grade in RAC:
mechanical and physical properties of recycled aggregate concrete”
ll For design mix M35 and M40 grade concrete all the Housing and Building National Research Center June 2014.
replacements with RAC gains early strength up to 78% 6. R. Sathish Kumar, “Experimental study on the properties of concrete
to 83% at 7 days. made with alternate construction materials” International Journal of
Modern Engineering Research (IJMER) ISSN: 2249-6645 oct 2012.
ll In Compressive strength, 8% Silica Fume(SF) and 5%
Mineral Quartz(Qu) increases strength (9.17% in M35) and 7. S.C. Kou, C.S. Poon, “Enhancing the durability properties of concrete
prepared with coarse recycled aggregate” Construction and Building
(12.41% in M40) than the target mean strength 28 days.
Materials 35 (2012) 69–76.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

574 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Study on Mechnical and Durability Properties of Recycled Aggregate Concrete Incorporated with Silica Fume and Mineral Quartz

Anand K.Darji
Perusing his Master's Degree in Structural Engineering from Birla Vishvakarma Mahavidyalaya Vallabh
vidhyanagar. He completed Bachelor of Engineering degree in Civil Engineering from the Bharati
Vidhyapeeth Deemed University in 2013. Currently he is doing research work in the field of RAC.
andi2891@yahoo.com

Prof. (Dr.) Indrajit Patel

He has over two and half decades of academic & Professional experience in the field of Civil & Structural
Engineering & Technology. He has successfully worked on many national and international projects and
activities in context to academic excellence, quality assurance and improvement, exchange program,
department and institute development program. To his credit there are more than 40 research publication
at national and international level conference and journals. He is recipient of young designer’s award, Best
Teacher Award and Promising Engineering Teacher award. Currently He is associated with S.P. University,
Gujarat Technological University (GTU) and Indian Society for Technical Education in various capacities.
Email: inpatel34@gmail.com

Ms. Jagruti Shah

Ms. Jagruti Shah, an Urban Planner (Gold Medalist), currently holds the position of Project coordinator
in Gujarat Technological University (GTU), Gujarat for the Project initiated by Government of Gujarat
“Vishwakarma Yojana: an approach towards Rurbanization” under which she is preparing Village
development Plan for scaling up village life with Urban amenities with the help of young Technocrats.
She hails five years of teaching and research experience in multi-disciplinary field of Urban and Regional
planning, Building Planning concrete technology and AHP based Urban Modelling. To her credit more than
30 research publication at national and international level. She is instrumental in preparation of Different
research project related to Urban Infrastructure Planning and Management, Affordable Housing, Building
and Town Planning, Urban Development Plan and Village development Plan. Email: jagruti@gtu.edu.in

Organised by
India Chapter of American Concrete Institute 575
Session 5 C - Paper 3

Importance and Comparison of Various Micro Materials in High

Performance Concretes
Mr. Yatin Joshi
(BE Civil, MBA). Ambuja Cements Limited

Abstract Study is based on various independent research works

A common and basic factor in all the construction carried out in institutions like IIT Madras, IIT Mumbai and
activities is concrete and concrete structures. Looking at other laboratories across the country.
a huge investment required in building up infrastructure Key words: HPC, ultra-fine GGBS, duarability, rheology,
/ industrial / residential facilities, durability of concrete PSD, packing effect
structures will be a focus point. Over the period of time,
concrete technology has also undergone a tremendous
change. Term High Strength is being replaced by High
Introduction
Durability. Concrete satisfying all the durability parameters Durability has become a key parameter for modern
can also be high strength but reverse is not true! structures. It extends the life of the structure and hence
benefit to the owner. Most important parameter to get
In practice, high performance concrete, are generally durable structure is to have concrete with least porosity and
characterized by high cement factors and very low W/C dense structure. Usage of supplementary cementitious
ratios. Such concrete suffer from two major weaknesses. materials helps in designing durable concrete. SCM now a
It is extremely difficult to obtained proper workability, and days being used are fly ash, GGBFS, densified silica fume,
to retain the workability for sufficiently long period of time. etc. Reaction of these SCMs is mainly pozzolanic.
High dosage of high range water reducing agents (HRWR)
then become a necessity. Resulting thixotropic and sticky Aim of this study was to understand the behaviour of
mixes are equally difficult to pump / place and compact various micro-fine / ultra-fine additives and utility of
fully and efficiently. ultra-fine GGBS in producing durable concrete.

The usage of micro fine materials like silica fume, meta

kaolin, rice husk ash, etc. are being used to obtain high Hydration Reaction
strength concretes. These micro fine materials work OPC + H2O → C – S - H + Ca(OH)2 (Hydraulic Reaction)
pozzolanically. They are used to fill up the inter particle SCM + Ca(OH)2 → C–S–H (Pozzolanic reaction)
spaces of cement grains and react with Ca(OH)2 produced
during cement hydration process. Majority of the times, Ordinary Portland Cement reacts with water to produce
fly ash, GGBFS present in concrete react with Ca(OH)2 calcium silicate hydrate (C-S-H) gel which binds all the
and micro fine materials work only as pore fillers. Micro aggregates together to develop the strength of concrete.
fine materials, due to their very high specific surface This exothermic reaction is known as hydraulic reaction.
area, tend to increase the water demand. This results in CaO content of the cement is main contributor to this
increased admixture dosage for low w/c ratio, increased reaction. Calcium hydroxide Ca(OH)2 which is a weak
heat of hydration and shrinkage cracks. Ultra-fine GGBS component; gets released as a bi product of hydraulic
(average particle size < 5 microns) is found to react reaction. SCM like fly ash & GGBS reacts with this
both; pozzolanic & hydraulic way. It does not increase the Ca(OH)2 to produce further C-S-H gel. This reaction is
water demand rather admixture dosage required are less categorised as pozzolanic reaction. SiO2 content of SCM
compared to that of other micro fine additives. governs this phase of the reaction. Usage of SCMs like fly
ash, GGBS in combination with micro fine additives like
This paper presents the discussion on comparison of
meta kaoiline, micro silica, rice husk ash are being used
various micro fine materials in terms of chemical and
to produce HPC. These micro fine additives also react
physical parameter, reactivity index, impact on strength,
due to their SiO2 content. Majority of the times, fly ash,
rheology and durability of various concrete mixes.
GGBFS present in concrete react with Ca(OH)2 and micro
Importantly, this paper explains the difficulties / fine materials work only as pore fillers. Replacement of
challenges faced by concrete producers while using OPC by these micro fines has a limit over which concrete
micro fine materials and solutions derived. It explains in becomes sticky and non pumpable. Micro fine materials,
details that why certain type of micro fine additives work due to their very high specific surface area, tend to
in superior fashion. increase the water demand. This results in increased

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

576 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Importance and Comparison of Various Micro Materials in High Performance Concretes

admixture dosage for low w/c ratio, increased heat of added together; combined grading of all the particles play
hydration and shrinkage cracks. Moreover, high level of an important role regarding flow ability, compactness &
replacements has derogatory effect over early strength strength of binder paste.
of the concrete.
If we use materials having very narrow PSD (undensified
micro silica has all the particle sizes between 0.1 micron
Recent Developments in Micro fine SCM to 1 micron); paste becomes less workable or sticky.
Recent advent of micro fine materials has helped Single size particles generally tend to “coagulate”. If we
address many challenges in producing highly durable, use well graded “broader” PSD material then it results in
rheologically friendly concrete mixes with very high level smooth flow of paste and concrete with good rheological
of OPC replacements. Ultra-fine GGBS has very hifh properties. It also results in reduced water demand.
fineness of more than 12000 cm2/gm and average particle
size (d50) less than 5 microns.

Sem Micrographs (1000 X Magnification) of Micro Fine

Materials

Fig. 1: Par ticle size distr ibution for var ious binder s
From left hand side (finer) to right hand side (coarse)

From above discussion, it is clear that micro fine material

(a) GGBS (b) Densified Silica fume due to their fine particle size and pozzolanic reaction
helps in designing durable high strength concrete. Some
of these materials play a dual role: pozzolonic as well as
hydraulic. Let us see the chemical composition of these
materials.

Chemical Composition of micro fine materials

Table 1
Comparative Chemical Composition of Micro Fine SCM

Component
(c) Metakaolin (d) Ultra-Fine Slag
CaO SiO2 Al2O3 Fe2O3
Material

The micro fine natures of the particles of these materials Silica Fume 0.2 - 0.4% 92 - 94% 0.04 - 0.06% 0.4 - 0.6%
help fill minute inter particle cavities in cement paste.
Metakaolin 0.4 - 0.8% 50 - 54% 41 - 45% 0.5 - 1%
This improves the “packing density” of cement paste. It
results in increased compactness and hence refined pore Micro-fine
32 - 34% 28 - 32% 18 - 20% 1.8 - 2%
structure. As the porosity gets reduced, durability of the Slag

concrete improves. Cement OPC 60 - 67% 18 - 22% 3 - 6% 1 - 4%

PSD Fly Ash 1 – 2% 55 – 65% 10 – 16% 10%

It has been observed that along with fineness; one more

parameter has major impact on the overall rheology of the It has been seen that micro-fine slag has similar
concrete. It is “particle size distribution” (PSD). Importance composition as that of OPC. Unlike other micro fine
of well graded combined aggregates in the concrete is materials, it has all major three components (CaO, SiO2,
well known. The same principal applies to binder content Al2O3). This enables the latent hydraulic properties in
of the concrete. When there are more than two materials these materials. This can be represented as in Figure 2.

Organised by
India Chapter of American Concrete Institute 577
Session 5 C - Paper 3

Fig. 2: Fig. 3: Slump & Compressive Strength

Because of the combined qualities (fineness + PSD + dual
reaction), micro fine slag holds a bright future. Also, as From the graph, it can be concluded that ground micro fine
it is manufactured in India (as against imported silica slag performs better in terms of slump, slump retention
fume from abroad), timely availability of the materials is and compressive strength over that of densified micro
warranted. silica. It can be attributed to the glassy surface of micro
fine slag. More over micro fine slag acquires –ve surface
charges when ground below 10 microns of the size. This
Table 2 result in small electrostatic repulsion, liberating water
Effects of micro fine SCMs in the concrete
trapped in two particles and hence improved slump and
or reduced water demand (improved strength).
Silica Micro-fine slag
Property Metakaoline
Fume (ALCCOFINE 1203) Assessing Chloride Permeability using ASTM 1543
test
Dispersion in
Critical Critical Proper Reference mix of M40 grade with OPC 400 kg / Cu M, 160
concrete
Ltr water & 2 kg admixture was compared with 360 Kg
Replacement to
Cement
10% max Up to 20% Up to 70% OPC, 40 Kg Ultra-fine slag (UFS), 160 Ltr water & 2 kg
admixture for chloride permeability test.
Water Demand High High Low

Table 3
Workability Reduced Reduced Improved
Chloride Content
Super plasticizer
Increased Increased Reduced Depth of Sample
dosage Results (% by mass)
Extraction (mm)
Rate of slump loss Fast Fast Slow
Chloride content for Chloride content
Initial Heat Of OPC concrete with UFS
Increased Increased Lower
hydration
Top Surface 0.4 o.3900
Alkalinity of Pore
Reduced Reduced Maintained
Solution
5 mm depth 0.31 0.045

10 mm depth 0.29 0.0056

Comparative performance in concrete
A comparative study on slump and strength development 15 mm depth 0.2 0.0051

of M60 grade of concrete containing densified micro silica 20 mm depth 0.18 0.0051
and micro fine slag was done. Binder content and dosage
25 mm depth 0.15 0.0048
of micro fine additives was kept same. (OPC 53: 430 kg/
cu m, fly ash: 80 kg/cu m, micro silica / ultra-fine slag: 30 mm depth 0.15 0.0048
40 kg/cu m). In the first experiment, water to binder ratio
35 mm depth 0.11 0.0048
(0.28) and PC admixture dosage (1.2%) was kept constant.
Comparison of slump readings was observed at every 30
minutes interval till two hours. In second experiment, Results sows that UFS could produce concrete with very
water to binder ratio and admixture dosage were adjusted low permeability i.e. very high durability performance.
to achieve constant slump values of 180 to 200 mm.
Compression strength test results were recorded at 1, 3, Conclusions
7 and 28 days for both the mixes.
Due to growing urbanisation & heavy cost of infrastructure;
Results are represented in graphical manner as below: durable concrete structures is need of the hour. Use of

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

578 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Importance and Comparison of Various Micro Materials in High Performance Concretes

various micro-fine materials to produce highly durable Ambuja Cements Ltd.

concrete will be a necessity than a choice. A judicious
Ambuja Knowledge Centre
approach has to be adopted while selecting the right
type of micro-fine additives to the concrete. Indigenously Technical Laboratory – CMPPL
“manufactured” micro-fine mineral additives hold a
Independent laboratories from Mumbai, Bengaluru &
promising future over imported “waste material / bi-
Delhi.
products” as techno commercial viable performance
improver for durable concrete.
References
1. Micro-fine / ultra-fine slag is a credible solution for 1. Darren, T. Y. Limda, DA. XU, Divsholi, B. Sabet, Kondraivendhan, B.
high strength high performance concrete and Susanto Tong (2011), Effect of ultra fine slag replacement on
durability and mechanical properties of high strength concrete, 36th
2. Particle Size Distribution (PSD) is an important conference on ‘Our world in concrete and structures’, Singapore.
parameter for designing paste component with 2. Huiwen Wan, Zhonghe Shui and Zongshou Lin, (2004), Analysis of
optimum packing density Geometric Characteristic of GGBS particles and their influences
on Cement Properties, Cement and Concrete Research, V. 34, No.
3. Ultra-fine slag does not increase the water demand as 1, 133-137
for other micro fine materials 3. Khayat, K. H., Yahia, A. and Sayed, M. (2008), Effect of supplementary
cementitious materials on rheological properties, bleeding and
4. Unlike densified micro silica, all the particles of ultra- strength of structural grout, Material Journal, V.105
fine slag are available for immediate reaction
4. Nakamura, N. M. Sakai and Swamy, R. N. (1993), Effect of
5. Ultra-fine slag is a right solution for high rise / long slag fineness on the development of concrete strength and
microstructure, ACI SP 132-2, 1343-66
distance pumping
5. Swamy R N. Sustainability concrete for the 21st century-concept
of strength through durability, The Indian concrete journal,V. 84,
Acknowledgement No.12, 2010, 7-15

IIT Chennai 6. Dr. P Dinakar - IIT Bhubaneshwar. Effect of ultr-fine slag on ultra
high strength concrete. ICI AECOCON Proceedings at IIT Chennai
IIT Mumbai December 2010

Mr. Yatin Joshi

Mr. Yatin Joshi is graduate in civil engineering and post graduate in management studies. He has vast and
multi-faceted experience in the field of specialized mortars, concrete products and additives for more than
16 years.
He was involved in various land mark projects such as national highway (NH4), Dabhol Power Project,
Ghatgar Dam, Dubai Metro, Burj Khalifa (world’s tallest tower), Palm Jumeirah (world’s biggest man-made
island), Dubai airport, Hodariyat Cable stayed post tensioned segmental construction bridge, concrete raft
foundation of 12,000 cubic meter in one go, etc.
He specializes in concrete technology for various types of concrete like high strength, durable, self-
compacting, roller compacting, pre-cast, light weight, temperature controlled, semi dry concrete mixes,
structural repairs & strengthening, water tight structures, specialized floorings, grouting for machine
foundations and post tensioned structures.
He was personally involved in developing “zero bleed” grout for post tension cable duct grouts complying
latest ASTM C 940, BSEN 445 & 447 standards. He has presented various technical papers on national and
international forums.
Currently he is working as a Head for Alccofine (range of ultra-fine additives for concrete & cementitious
micro-fine injection grouts) business with Ambuja Cements Ltd.

Organised by
India Chapter of American Concrete Institute 579
Session 5 C - Paper 4

Comparative Evaluation of Constitutive Models for Concrete Under

Cyclic Compression

Mayank Tripathia Devia, Saptarshi Sasmal

A. Kanchana and K. Ramanjaneyulu
Academy of Scientific and Innovative Research, CSIR-Structural Engineering Research Centre,
Chennai, India Chennai-600113, India

Abstract curve. From the available knowledge in the literature, it is

clear that envelope curve for concrete under axial cyclic
Number of constitutive models is available in the literature
compression can be approximated by monotonic stress-
for concrete under cyclic compression. Some of the models
strain curve of concrete.
are based on experimental data fitting and neglecting the
material parameters and the others are based on material
parameters. In the present study, constitutive models
for concrete under cyclic compression are reviewed and
limitations of these models are brought out. To evaluate
their efficacy, cyclic constitutive models reported in the
literature for concrete under compression are implemented
in a MATLAB code and stress-strain curves obtained using
these models are compared with the available experimental
results.

Introduction
Computational modelling and simulation of any reinforced Fig. 1: Typical response of concrete under cyclic compression
concrete structure subjected to cyclic loading requires (Sinha et. al.(1964)
realistic cyclic constitutive models or stress-strain models
that can accurately simulate the behaviour of structure
under cyclic loading. For performance evaluation of
any structure both experimental as well as analytical
investigations are required but under certain conditions
it’s difficult to conduct experimental investigations. In
such conditions, analytical investigations can be carried
out using realistic constitutive models.
Research was carried out in the past for the development
of constitutive model for concrete under cyclic
compression with first published model by Sinha et al.
(1964) to characterise the behaviour of concrete under
cyclic compression. Afterwards, number of models was Fig. 2: Key points in cyclic response of concrete (Otter and
developed for characterising the cyclic behaviour of Naaman, 1989)
concrete. Figure 1 shows typical response of concrete
under cyclic compression. The key points identified According to CEB [1996], cyclic constitutive models can
in the stress-strain plane to define the response of be broadly classified into three categories: (i) models
concrete under cyclic compression are: (i) envelope based on theory of elasticity, (ii) models based on theory
curve, (ii) unloading strain, (iii) plastic strain/residual of plasticity, and (iii) models based on continuum damage
strain, (iv) common point, and (v) reloading strain. Figure theory or combination of above models. In this paper, only
2 shows the key points that are required to define the the models that are essentially mathematical formulations
stress-strain response of concrete. Points C and I are are discussed.
termed as unloading points. Points F and K define the
corresponding plastic strains at zero stress level. Point Sinha et al. (1964) proposed first model for cyclic stress
H is the reloading point where the loading curve meets strain relationship of concrete under compression.
the envelope curve. Point D is called as the common point The model was developed based on the experimental
i.e. the intersection of reloading curve with the unloading investigations carried out on plain concrete cylinders
under cyclic loading. Polynomial expression was adopted

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

580 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Comparative Evaluation of Constitutive Models for Concrete Under Cyclic Compression

for defining the envelope curve. Unloading and reloading final unloading stiffness. Other major disadvantage of
curves were defined by second order parabola and linear this model is that the unloading curves show too much
curve respectively. Karsan and Jirsa (1969) proposed a softening.
model based on the extensive experimental investigations
Aslani and Jowkarmeimandi (2012) proposed hysteretic
carried out. Based on the experimental results, analytical
stress-strain model for unconfined concrete for both full
expressions for envelope curve, unloading-reloading
unloading-reloading and partial-unloading-reloading
curves and upper and lower shakedown limits, plastic
cases. Their model addresses both cyclic compression
strain, and the reloading strain were developed.
as well as cyclic tension. Expressions were proposed for
Yankelevsky and Reinhardt (1987) proposed a simple envelope curve, unloading curve, reloading curve and
model based on the location of fixed geometric loci transition curve i.e. the curve when transition takes place
called ‘focal points’ in stress-strain plane. Unloading and from compression to tension regime or vice-versa. The
reloading curves were defined as piecewise linear. The reliability of this model was checked by implementing
complete model was based on the geometrical location it into non-linear finite element analysis and the results
of focal points rather than understanding of the material were compared with the experimental data. Mayank
behaviour. Otter and Naaman (1989) published the first (2014) proposed model for defining the cyclic stress-strain
model considering the case of random loading histories. response of concrete. Damage parameter was introduced
The model required definition of experimental coefficients to define the envelope, unloading and reloading curves.
for implementation. For defining the unloading curve, a The model tried to consider both strength as well as
polynomial expression was adopted whereas for defining stiffness degradation caused due to load cycling. The
the reloading curve linear expression was adopted. model proposed was validated with the experimental
Martinez-Rueda and Elnashai (1989) proposed a model results of Karsan and Jirsa(1969) and Sinha et. al. (1964).
that was the modification of model proposed by Mander Experimental investigation by Karsan and Jirsa (1969)
et al. (1988) so as to take into account the strength and and Sinha et. al. (1964) were adopted over experimental
stiffness degradation. Locus of common point was used investigations by Bahn and Hsu (1998) as the experiment
for defining the damage caused due to load cycling. For conducted by Bahn and Hsu was on variable minimum
defining the unloading curves, a second degree parabola stress level which was difficult to be validated by some of
was adopted and for defining reloading curves a linear the model from literature. In addition to this most of the
expression was adopted. Bahn and Hsu (1998) performed models in literature have been developed and validated
parametric study as well as experimental investigations using the experimental investigation by Karsan and Jirsa
on the behaviour of concrete under random cyclic load (1969) and Sinha et. al. (1964), which serves as benchmark
history. Experimental investigations were carried out to experiment results.
determine the major parameters that define the stress- From the review of literature, it has been understood
strain response of concrete under cyclic loading. For that significant number of cyclic constitutive models
defining the unloading curve, a polynomial of degree for concrete under compression were proposed. Some
‘n’ was adopted and to define reloading curve a linear of these models are highly dependent on experimental
expression was adopted. coefficients and the other models are based on the
Palermo and Vecchio (2003) developed constitutive model material parameters. These models differ from each
for concrete based on modified compression field theory other in terms of formulations. In this paper, the models
(Vecchio and Collins, 1986). Model was developed by them available in the literature are reviewed, implemented and
for both cyclic compression as well as cyclic tension. the stress-strain curves obtained using these models are
Ramberg-Osgood formulations were used to define non- compared with that of the tests reported in the literature.
linear unloading curve whereas linear reloading curve
that incorporates degradation in reloading stiffness based Evaluation of Cyclic Constitutive Models for
upon the amount of strain recovered during unloading
phase was proposed. Polynomial expression was adopted Concrete
for defining the unloading curve whereas, reloading curve In this paper, some of the cyclic constitutive models
was defined by a linear expression. Sima et al. (2008) available in the literature are identified and implemented
proposed a model that took into account strength as well for their evaluation. Due to incompleteness of the model
as stiffness degradation caused due to load cycling. Their proposed by Sima et. al. (2008) and due to numerical
model tries to consider the complete hysteretic effect of instability in the model proposed by Palermo and Vecchio
the stress-strain response so that full as well as partial (2003), these models are not implemented. Mayank (2014)
unloading and reloading can be taken into account very proposed constitutive model for concrete under cyclic
easily. Major advantage of this model is that the initial and compression to consider both stiffness and strength
final unloading stiffness is not constant and hence it can degradation due to load cycling. Damage parameters were
simulate the accurate response of concrete. But, it lacked defined for evaluating the damage accumulation due to
in defining the relation between the initial stiffness and load cycling. Mayank (2014) is based on the understanding

Organised by
India Chapter of American Concrete Institute 581
Session 5 C - Paper 4

of the material behaviour under cyclic compression rather For defining the unloading curve (as shown in Figure 4),
than on the approach of curve fitting. Salient features of the following analytical expression is adopted (Sima et.
the Mayank model are presented below. For defining the al. 2008), but the expression for coefficient E is modified,
envelope curve, the expression proposed by Mazars and (Mayank 2014).
v = De E 1 - f - f E o Q f - f p V
Pijaudier cabot (1989) was adopted. Figure 3 shows the S f-fp
X
.................................................(7)
generalised stress-strain envelope curve for concrete.
u p

The regions of envelope curve are defined by: The identified models are implemented using MATLAB
v = E o f f < f o ....................................................(1) codes and stress-strain curves obtained using these

v = Q f o + K 1 V E o + K 2 fe S
fo - f
X
models are compared with that of the available
f cl Eo f o # f # f cl ............(2) experimental results. The flow-chart for implementation
v = ! B + Cfe $E o of the model proposed by Mayank (2014) is shown in
fo - f
f cl f $ f cl ..................................(3)
Figure 5.

Fig. 3: Envelope curve for concrete

Fig. 4: Unloading and Reloading Curves

The value of damage parameter for pre-peak as well as Fig. 5: Flowchart for implementation of models
post-peak regions of stress strain curve is given as:
d = 1 - fo Q 1 - K 2 V - K 2 e f l
f f -f o
Otter and Naaman Model
f o # f # f cl ..........(4)
c

Otter and Naaman (1989) proposed model for cyclic

B f -f
f $ f cl ..................................(5)
o
d = 1 - f - Ce f l c
compression behaviour of concrete that takes into account
the case of random loading histories. Model proposed by
For defining the plastic strain level from the envelope Otter and Naaman (1989) was implemented and compared
unloading strain value, the following expression is with the experimental results of Karsan and Jirsa (1969)
proposed. and is shown in Figure 6.
= 0.1478 S ul X + 0.1665 S ul X ...............................(6)
fp f 2 f
From Figure 6, it is seen that this model behaves well
f cl fc fc
corroborated with the experimental data with slight
Expression proposed for defining the plastic strain is able deviation in near peak region but the overall behaviour
to accurately predict the plastic strain level as obtained was found to be good. The major disadvantage of their
from experimental data. Though having this advantage formulation is definition of the experimental coefficients.
the main limitation of this expression is its dependency on Therefore, this model requires determining the
the experimental data. coefficients based on experimental data.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

582 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Comparative Evaluation of Constitutive Models for Concrete Under Cyclic Compression

with the experimental results by Karsan and Jirsa (1969)

is shown in Figure 9. From the Figure 9, it is clear that
the model proposed by Mayank (2014) is able to predict
the cyclic behaviour of concrete at near peak region as
well as post peak region. This Model predicts the plastic
strain level as well as the stiffness of the unloading and
reloading curves with sufficient accuracy.

Fig. 6: Comparison of model proposed by Otter and Naaman

(1989) with cyclic compression test by Karsan and Jirsa (1969)

Fig. 8: Comparison of model proposed Aslan and Jowkarmeimandi

(2012) with cyclic compression test by Karsan and Jirsa (1969)

Fig. 7: Comparison of model proposed by Bahn and Hsu (1998)

with cyclic compression test by Karsan and Jirsa (1969)

Bahn and Hsu Model

Bahn and Hsu (1998) proposed a model based on the
parametric study carried out on response of concrete to
cyclic compression. Model is implemented and compared
with the experimental results by Karsan and Jirsa (1969)
Fig. 9: Comparison of model proposed Mayank (2014) with cyclic
and is shown in Figure 7. From Figure 7, it is clear the
compression test by Karsan and Jirsa (1969)
model proposed by Bahn and Hsu (1998) shows stiffer
behaviour as compared to the experimental result.
Conclusions
Aslani Model Constitutive models for concrete under cyclic compression
Aslani and Jowkarmeimandi (2012) proposed constitutive are reviewed and implemented using MATLAB code
model for unconfined concrete. The main advantage of to evaluate their efficacy. The stress-strain curves
the model was its non-dependency on the experimental obtained using these models are compared with that of
coefficient and considers both stiffness as well as strength the experimental results. From this study, it has been
degradation. Figure 8 shows the comparison of model observed that the model proposed by Mayank (2014) is
proposed by Aslani and Jowkarmeimandi (2012) with the able to accurately predict the behaviour of plain concrete
experimental results of Karsan and Jirsa (1969). Figure 8 under cyclic compression. Some of the conclusions drawn
clearly indicates that though this model is able to predict from the present study are:
the behaviour of concrete under cyclic loading, it is unable
to exactly define the cyclic response of concrete. From 1. The model proposed by Otter and Naaman (1989)
the Figure 8 it is also seen that the model under predicts and Mayank (2014) are able to predict the behaviour
the plastic strain ratio which gives softening behaviour to of concrete under cyclic compression but the major
unloading curve after peak. For the region near the peak, disadvantage of Otter and Naaman model is its
the model is able to predict with sufficient accuracy. requirement of experimental coefficients.
2. From the cyclic response of concrete, it has been seen
Mayank Model that as the strain increases there is reduction in both
The comparison of the model proposed by Mayank (2014) load as well as stiffness of the unloading and reloading

Organised by
India Chapter of American Concrete Institute 583
Session 5 C - Paper 4

curves which becomes almost flat at high strain levels. 6. Martinez-Rueda, J.E., & Elnashai, A.S., 1997, Confined concrete
under cyclic load. Materials and Structures, Vol. 30(3), 139-147.
3. Though the model proposed by Aslani and
7. Mayank Tripathi, 2014, Behaviour of gravity load designed beam-
Jowkarmeimandi (2012) is a recent constitutive
column sub- assemblages under seismic loading, M.Tech Thesis
model, it deviates from the experimental result in Phase-I, Academy of Scientific and Innovative Research, CSIR-SERC.
determination of plastic strain level.
8. Mazars, J., & Pijaudier-Cabot, G., 1989, Continuum damage theory.
Application to concrete Journal of Engineering Mechanics ASCE,
Acknowledgement 115(2), 345–365.

This paper is being published with the kind permission of 9. Otter, D.E., & Naaman, A.E., 1989, Model for response of concrete
the director, CSIR-SERC, Chennai to random compressive loads. Journal of Structural Engineering
ASCE, Vol.115 (11), 2794-2809.

References 10. Palermo, D., & Vecchio, F.J., 2003, Compression field modelling of
1. Aslani, F., & Jowkarmeimandi, R., 2012, Stress-strain model for reinforced concrete subjected to reversed loading: Formulation,
concrete under cyclic loading. Magazine of Concrete Research, ACI Structural Journal, and Vol.100 (5), 616- 625.
Vol. 64(8), 673-685. 11. Sima, J.F., Roca, P., & Molins, C., 2008, Cyclic constitutive model for
2. Bahn, B.Y., & Hsu, C-T. T., 1998, Stress-strain behavior of concrete concrete. Engineering Structures, Vol.30 (3), 695-706.
under cyclic loading, ACI Material Journal, Vol. 95(2), 178-193.
12. Sinha, B.P., Gerstle, K.H., and Tulin, L.G., 1964, Stress-strain
3. Comite Euro-International du Beton (CEB), 1996, R.C. Elements relations for concrete under cyclic loading. Journal of the ACI, Vol.
under cyclic loading- State of the Art Report. London: Thomas 61 (2), 195-212.
Telford.
13. Vecchio, F. J., and Collins, M. P., 1986, The modified compression-
4. Karsan, D.I., & Jirsa, J.O., 1969, Behavior of concrete under
field theory for reinforced concrete elements subjected to shear,
compressive loadings, Journal of the Structural Division ASCE,
Vol. 95(12), 2543-2564. ACI Journal Proceedings, Vol. 83(2), 219-231.

5. Mander, J.B., Priestley J.N. & Park, R., 1988, Theoretical stress- 14. Yankelevsky, D.Z, & Reinhardt HW., 1987, Model for cyclic
strain model for confined concrete, Journal of Structural Division compressive behavior of concrete. Journal of Structural Engineering
ASCE, Vol. 114(12), 1804–1826. ASCE, Vol.113 (2), 228–240.

Mayank Tripathi
Current Position: Senior Research Fellow, CSIR-Structural Engineering Research Centre, Chennai
Correspondance Address:- CSIR-SERC Guest house, CSIR-SERC, Taramani, Chennai-600113
I was born in small city of Uttarakhand, on 28th of January 1993, and currently lives at Chennai.
I completed my Bachelor degree in Civil Engineering from Uttarakhand Technical University, Dehradun
with a focus on development of stability charts for analysing the stability of slopes in Himalayan region that
will benefit the construction industry especially the state highway departments carrying out construction
in hilly regions.
After the completing my Bachelor’s degree I had joined CSIR-Structural Engineering Research Centre,
Chennai for pursuing higher degree and joined Academy of Scientific and Innovative Research at CSIR-
SERC to complete my Master’s degree. For past 2.5 Years I have been involved in various research projects
with keen focus on seismic behaviour of deficient beam-column joints and development of upgradation
strategies for the same.
My keen interest involves working in the field of seismic behaviour of structures, Constitutive modelling
of concrete, seismic retrofitting of degraded/damaged structures. I wish to pursue my career as an
academician, researcher and industry consultant.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

584 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Waste Marble Powder as Partial Replacement in Cement Sand Mix

Use of Waste Marble Powder as Partial Replacement in

Cement Sand Mix
Nitisha Sharma and Ravi Kumar
P.G. Research Scholar, Civil Engineering , Swami Devi Dyal Institute of Engineering Technology, Haryana, India

Abstract Usually this type of waste can be utilized by using it as

a raw material or as constituent in a material because
In this research work, the effects of using waste marble
they had a different chemicals present in it that causes a
powder (WMP) as a partial replacement of cement and
harmful effects on the environment.
sand on the mechanical properties of the concrete have
been investigated. For this purpose seven different series Nowadays, concrete has a great advancement in concrete
of concrete-mixtures were prepared by partially replacing technology in which it can reduce the consumption of
cement, sand with WMP at proportions of 0, 10 and 15% natural resources as well as the energy sources and
by weight separately and in combined form. In order that can further reduce the impact of pollutants on the
to determine the effect of the WMP on the compressive surroundings. Due to hike in price, waste should be used
strength, split tensile strength and flexural strength, in the constituents to decrease the cost and make the
strengths of the samples were recorded at the curing ages project cost effective.
of 7 and 28 days. Finally, all of the data were compared
In this experimental study we had experimental effect
with each other. It was observed that the addition of WMP
of marble waste powder on the concrete mix by partially
such that would partially replace the sand and cement
replacing cement and sand with the marble powder waste.
separately at particular proportions has displayed an
In this project, we check the effect on mechanical and
enhancing effect on the mechanical properties of concrete
physical properties of concrete mix with varying marble
mix. It is also useful because marble waste is a by-product
powder waste partially replaced in concrete mix.
of marble production and also creates environmental
pollution in large scale. Therefore, it could be possible to
prevent the environmental pollution. Research Significance
Key words: Waste marble powder, concrete, compressive In this experimental study fine marble powder dust were
strength, flexure strength, split-tensile strength. collected from the nearby source for the investigation.
Different concrete mixtures were prepared by using
different percentages of marble powder like 0%,
Introduction 10%(sand), 10%(cement), 15%(sand), 15%(cement),
Marble is obtained from the transformation of pure lime 20%(cement(10%) + sand(10%)) and 30%(cement(15%)
stone. The purity of marble depends upon the colour of + sand(15%)) as a partial replacement of cement and
the marble. Since the ancient times marble is widely used sand mix. The mechanical and physical properties were
in monuments and historical buildings for decorative checked on the 7 & 28 days.
purpose. The various types of constituents present in
marble, some of which varies from origin to origin. There
are some chemical as well as mineral impurities which
Experimental Methodology & Investigation
are associated with marble like quartz, muscovite, SiO2, Concrete Mix Constituents
limonite, Fe2O3. But some impurities like magnesia,
phosphate, leads, zinc, alkalis and sulfides affect the Cement
properties of cement. In general, large amount of marble The cement use for the experimental studies was 43
powder dust are obtained during the cutting and forging grade OPC conforming to the specifications of Indian
process. In India, tons of waste has been produced from Standard Code IS: 8112-1989 shows in table 1. It was fresh
the industries. But there are some impurities present in and without any lumps.
the waste that cannot be easily disposed off. Such type of
impurities mixed with soil and water. When they mix with Aggregate
soil, it reduces the porosity and permeability of soil and Normal river sand which is locally available in the market
also reduces the fertility of soil. Also if it mixes with water and confirming to Zone II as per IS 383 1970 as shown
it pollute the water and make the water unfit for use. So in table 2 and specific gravity of fine sand is 2.614 and
it is necessary to use the waste in functional manner. coarse aggregates were used in this experiment whose

Organised by
India Chapter of American Concrete Institute 585
Session 5 C - Paper 5

finess modulus is 2.65. Coarse aggregate used as 20 mm Supplementary Cementitious Materials

down size. The lumps of clay and other foreign materials The marble powder was obtained by crushing marble
were separated out carefully. Sand was washed and dried powder forms in a marble industry. The bulk density
before testing. The coarse aggregates were washed to was 1118.01 kg/m3 and fineness modulus is 2.03 and has
remove dirt, dust and then dried to surface dry conditions. specific gravity of 2.21.

Table 1
Characterstics Properties of Cement
Concrete Mixture Proportion
In this experimental study, the mix design is taken as
Specified value as Experimental M30. Water binder ratio is taken as 0.43. Different mixes
Characteristics was prepared by using a different percentage of marble
per IS:8112-1989 value
powder (0%, 10%, 10%, 15%,15%, 20% and 30%) namely
Consistency of cement (%) - 31.5 MX0,MX1,MX2, MX4,MX5 & MX6 as a partial replacement
in the cement sand mix, where MX0 is control mix with
Specific gravity 3.15 3.01 no marble powder dust,MX1with 10% marble powder
as partial replacement of sand, MX2 with 10% marble
Initial setting time (minutes) > 30 40 powder dust as partial replacement of cement and MX5
with 20% marble powder dust as partial replacement of
Final setting time (minutes) < 600 380 cement and sand together and also MX3 with 15% marble
powder as partial replacement of sand, MX4 with 15%
Compressive strength (N/mm2)
marble powder dust as partial replacement of cement and
(i) 3 days > 23 25.10 MX6 with 30% marble powder dust as partial replacement
(ii) 7 days of cement and sand together.
> 33 36
(iii)28days > 43 48.10

Soundness (mm) 10 1.05

Fineness of Cement (gm) 10 1.5

Table 2
Physical Properties of Fine and Coarse Aggregates

Fine Coarse
Sr.No Physical Properties
Aggregates Aggregates Fig. 1: Marble Powder Fig. 2: Curing of Specimen

1 Specific Gravity 2.614 2.66

Casting Detail
2 Free Moisture Content 2% - Size of standard cubical moulds for the casting of specimen
is 150mm x 150mm x 150mm were casted according to the
3 Water Absorption 1.80% 0.13% mix proportion. The size of specimens of 150 mm diameter
and 300 mm height of the cylinder size and the size of
4. Finess Modulus 2.86 2.65 prism 100 x 100 x 500 mm were also casted according to
the mix proportion and also by partial replacing of cement
and sand with marble powder in different proportions.
Table 3
Physical Properties of Marble Powder Curing of Specimen
After the hardened of specimen in about 24 hours then
Sr.No Physical Properties Values the casted concrete specimens were cured under water
which is free from chlorides and sulphates are placed for
1 Specific Gravity 2.210 curing and tested after required curing.

2 Dry Moisture Content 1.58% Testing the Specimen Details

Concrete specimens were tested using compression
3 Bulk Density(kg/m3) 1118
testing machine (CTM) of capacity 200 tones and with a
4. Finess Modulus 2.03
constant rate of load is 14 N/min for all specimens and
were tested at different curing ages for 7days and 28

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

586 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Use of Waste Marble Powder as Partial Replacement in Cement Sand Mix

days. Split tensile strength was testing on the 200 tones

capacity machine and constant rate of load is 2.4 N/mm²/
minute. Flexural strength testing was also conducted
by using a 100kN capacity electrically operated flexural
testing machine at a displacement rate of 0.05 mm/sec.

Experimental Test Result & Discussion

Workability
Slump values of concrete sample have been tested for
different sample of mix with different percentages of
marble powder as replacement of cement and sand in a Fig. 3: Graph Between Different Mixes and Compressive
mix. The result showed that the workability of a concrete Strength of 7 & 28 Days
mix was decreases with increase in the marble powder
dust content. Split Tensile Strength
It can be noted that the split tensile strength for concrete
Strength mix increased with 14% & 12% when partially replaced by
10% marble powder dust against cement in 7 & 28 days
Compressive Strength respectively and also increased by 7% & 13% as in the
It can be noted that when cement is partially replaced partial replacement with sand having 10% marble dust
by the marble powder upto 10% then the compressive powder in 7 & 28 days respectively. But the split tensile
strength of the mix after 7 days increased upto 11.6% and strength decreased for the mix which contains 15 %
after 28 days it increase upto 10% and when partially marble powder dust against cement by 4.6%& 1.89% in 7
replace it with sand then again compressive strength & 28 days respectively and 1.4% & 2.43% when partially
after 7 days increased upto 10.8% after 28 days it increase replaces with 15% sand. in 7 & 28 days respectively. Also
upto 11.2% but when marble powder dust is partially when sand and cement together partially replaces upto
replaced by cement (15%), compressive strength after 7 20%and 30% they have low split strength as compare
& 28 days decreases slightly and when partially replaces to the replacement of marble waste in cement and sand
it with sand then it decreases upto 6.4% & 4.4% after 7 individually.
& 28 days respectively also when together replaces by
20%(10%+10%) by marble powder then its compressive
strength after 7 & 28 days decreased upto 9.3% & 18% Table 5
Comparison of Split Tensile strength after 7 & 28 days
respectively and 9.7% & 19% when partially replaces both
cement and sand by 30%(15%+15%) marble waste. Hence
% Increase Average % Increase Average
result shows that marble powder when mixes with sand S.No.
Mix
Split Tensile Strength Split Tensile
and cement upto 10% has high compressive strength and Designation
in 7 Days Strength in 28Days
thereafter its strength decreases. 1 MX0
2 MX1 7.0% 13%
Table 4 3 MX2 14% 12%
Comparison of Compressive Strength after 7 & 28 Days
4 MX3 -1.40% -2.43%
% Increase Average % Increase Average
Mix 5 MX4 -4.6% -1.89%
S.No. Compressive Strength Compressive
Designation
in 7 Days Strength in 28 Days 6 MX5 -8.41% -10%

1 MX0 7 MX6 -10.7% -13.7%

2 MX1 10.8% 11.2% Flexural Strength

3 MX2 11.6% 10%
It can be observed that the flexure strength for the
concrete mix containing 10% of marble powder dust in
4 MX3 -6.4% -4.4% cement got increased by the value of 7.9% & 1.5% in 7 &
28 days respectively and for 10% replacement with sand
5 MX4 -1.1% -1.0% the flexure strength also got increased about 7.2% & 4.8%
in 7 & 28 days respectively but it decreased when the mix
6 MX5 -9.3% -18% contains 15% marble powder dust against cement about
4.1% & 2% in 7 & 28 days respectively and 2.4% & 1.1%
7 MX6 -9.7% -19% when mix contains 15% marble powder dust against sand

Organised by
India Chapter of American Concrete Institute 587
Session 5 C - Paper 5

1. When cement is replaced with marble powder upto

10% weight a high strength concrete was achieved but
decreased when replaces it with 15%.
2. Increasing the amount of marble powder decreases
the workability of concrete.
3. Based on the experiment result it showed that
replacement of cement and sand by marble powder
upto 10% increases the compressive strength but
above 10% content of marble powder decreases the
compressive strength.
Fig. 4: Graph Between Different Mixes and Split Tensile Strength
of 7 & 28 Days 4. Split tensile strength increases with increase in marble
powder dust upto some proportion.
Table 6 5. Compared to the control concrete flexural strength is
Comparison of Flexural Strength After 7 Days & 28 Days
maximum when replace with fine aggregate upto 10%.
% Increase Average % Increase Average
Mix References
S.No. Flexural Tensile Flexural Tensile
Designation
Strength in 7 Days Strength in 28Days 1. Er: Raj.p.singh kushwah, Prof (Dr.) Ishwar Chand Sharma,
1 MX0 Prof (Dr.) PBL Chaurasia(2015) Utilization of “Marble Slurry” In
Cement Concrete Replacing Fine Aggregate. American Journal
2 MX1 7.2% 4.8% of Engineering Research (AJER) e-ISSN : 2320-0847 p-ISSN :
3 MX2 7.9% 1.5% 2320-0936Volume-04, Issue-1, pp-55-58.
2. Bahar Demirel, The Effect of the using Waste Marble Dust as Fine
4 MX3 -2.4% -1.1%
Sand on the Mechanical Properties of the Concrete ISSN 1992 - 1950
5 MX4 -4.1% -2.0% ©2010, International Journal of the Physical Sciences Vol. 5(9), pp.
1372-1380, 18 August, 2010.
6 MX5 -3.8% -4.0%
3. Baboo Rai, Khan Naushad H , Abhishek Kr, Tabin Rushad S, Duggal
7 MX6 -6.0% -4.6%
S.K, The effect of using marble powder and granules as constituents
of fines in mortar or concrete INTERNATIONAL JOURNAL OF CIVIL
AND STRUCTURAL ENGINEERING Volume 1, No 4, 2011.
4. Hassan A. Mohamadien, The effect of marble powder and silica
fume as partial replacement for cement on mortar INTERNATIONAL
JOURNAL OF CIVIL AND STRUCTURAL ENGINEERING, Volume 3,
No 2, 2012.
5. Noha M. Soliman, Effect of using Marble Powder in Concrete Mixes
on the Behavior and Strength of R.C. Slabs, International Journal
of Current Engineering and Technology ISSN 2277 - 4106 Vol.3,
No.5 (December 2013).
6. V. M. Sounthararajan and A. Sivakumar, Effect of The Lime Content
in Marble Powder for Producing High Strength Concrete. ARPN
Fig. 5: Graph Between Different Mixes and Flexural Strength Journal of Engineering and Applied Sciences, Vol. 8, No. 4, APRIL
of 7 & 28 Days 2013 ISSN 1819-6608.
7. Animesh Mishra, Abhishek Pandey, Prateek Maheshwari, Abhishek
Chouhan, S. Suresh*, Shaktinath Das, Green Cement For Sustainable
in 7 & 28 days respectively. Also when sand (10%) and Concrete Using Marble Dust, Department of Chemical Engineering,
cement (10%) were partially replaced with marble powder Maulana Azad National Institute of Technology, (MANIT) Bhopal,
dust it decreases the strength about 3.8% & 4% in 7 & 28 Madhya Pradesh India, Research CODEN( USA): IJCRGG ISSN
: 0974-4290 Vol.5, No.2, pp616-622, April-June2013.
days respectively and 6% & 4.6% when sand (15%) and
cement (15%) were partially replaced with marble powder 8. Prof. Veena G. Pathan, Prof. Md. Gulfam Pathan, Feasibility and need
of use of waste marble powder in concrete production IOSR Journal
dust in 7 & 28 days respectively. Hence result shows
of Mechanical and Civil Engineering (IOSR-JMCE)
that marble powder when mixes with sand and cement
together has low flexural strength, and individually it 9. Nitisha Sharma, Ravi Kumar “Review on Use of Waste Marble
Powder as Partial Replacement in Concrete Mix” SSRG International
enhances the strength upto some proportion. Journal of Civil (SSRG-IJCE), ISSN: 2348-8352, EFES April, 10-11,
2015.
Conclusion 10. IS: 8112-1989, Indian standard of ordinary Portland cement, 43
grade-specification (second revision)
Experimental investigation showed the following
conclusions: 11. IS: 383- 1970, Indian standard of specification for coarse and fine
aggregates from natural sources for concrete (second revision)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

588 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Optimizing the Structural and Foundation Systems of the 151 Story Inchon Tower : The Development of New Generation of Tall Building System

Optimizing the Structural and Foundation Systems of the

151 Story Inchon Tower : The Development of New Generation of
Tall Building System

Ahmad Abdelrazaq Moonsook-Jeong and Soogon-Lee Taeyoung-Kim

Sr Executive Vice President and Head of PH.D. & Senior Structure Engineer, Highrsie Structure Engineer, Highrsie and
Highrsie and Complexe Building Division, and Complexe Building Division, Complexe Building Division,
Samsung C&T Samsung C&T Samsung C&T

Abstract the principal design Architect (Portman and Associate),

several tower massing and corner treatments were
The 151 story super high-rise building located in an area
introduced and included 1) softening of the tower
of reclaimed land constructed over soft marine clay in
corners, introducing openings along the building height
Songdo, Korea. The focus of this paper will provide 1) the
with different configurations, and edge treatments were
structural engineering techniques utilized to optimize
considered. The final tower massing and geometry are
the structural and foundation systems of the tower, 2)
shown in Figure 1.
the different floor framing system considered, 3) the
wind engineering approach to tame the wind behavior of The structural system of the tower in the east-west
the tower to reduce the overall wind forces and to tame direction consist of reinforced concrete core wall linked
the dynamic response of the tower, 4) a description to the exterior mega columns with reinforced concrete
of an additional reliable damping system that is well or composite panels to maximize the effect structural
integrated with the structural and the architectural depth of the tower. However, the lateral load resisting
design concepts to improve the tower overall behavior system of the tower in the north-south direction
under service and extreme lateral load conditions, 5) consists of mega-frame structure, where the reinforced
and finally describe the impact of foundation flexibility concrete core walls are linked by 4-story structural
on the overall behavior of the tower, through soil steel trusses at 3 levels at approximately every 30
structure interaction. The introduction of the damped floors. Alternatively, these trusses were augmented with
mega-frame structural system for the 151 story Inchon supplemental damping system, see figure 11, to reduce
will be a catalyst in utilizing the latest Damping systems the overall vibration response, enhance the overall
and technologies for a “New Generation of Tall Building strength and performance of the tower under extreme
System”. events, and ultimately provide reliable damping. The
tower superstructure is founded on pile supported
Keywords: Super high-rise building, Structural
raft foundation consisting of 5.5 meter thick reinforced
optimization, floor framing design, foundation design,
concrete raft over 172-2.5m diameter bored piles with
foundation stiffness, piled raft
variable lengths and anchored a minimum of 5 meters
into soft rock. The vertical and lateral pile testing
Introduction programs have already been successfully completed
The proposed 151 story Multi-use Inchon Tower is utilizing the “O-Cell Method”.
located in the Songdo Inchon Free Economic Zone and
founded on new reclaimed land. The 600m tall tower is
composed of approximately Thirty (30) stories of office
floors, seventeen (17) stories of hotel & other supporting
facilities, 100 stories of residential floors, and several
levels of mechanical plant floors. The base of the tower
consists of retail, future subway station, and several
levels of parking below grade. See figure 1 for the final
rendering of the tower and typical office, residential and
mechanical plan arrangements. Several tower massing
were studied in details in an effort to improve the overall
building response to the overall dynamic excitations. The
original tower shape was trapezoidal with very sharp
corner. This shape was extremely sensitive to cross
wind response and subject to significant lift forces and
Fig. 1: 151 story Inchon Tower rendering and typical office,
dynamic excitation. Therefore and in coordination with
residential and mechanical floor plans

Organised by
India Chapter of American Concrete Institute 589
Technical Papers

Tower’s Mega-Frame Structural Concept and outline of

the structural system behavior. Alternatively, the Mega-
Frame concept is augmented with supplemental damping
to tune the dynamic behavior of the tower under service
and extreme loading conditions and to provide reliable
damping (See figure 11).

Fig. 2: Mega-Frame Structural Concept

Structural System Description

The lateral load resisting system of the tower consists of
a central reinforced concrete core walls up to level 40 that
splits into two cores above level 40. Because of the high
aspect ratio of the tower and the sensitivity of the floor
Fig. 3: Lateral Load System organization, and Exterior Steel
airfoil geometry to wind excitation, extensive gravity and
Column Transfer (gravity Columns)
wind engineering management was considered during
the early development of the structural concept to control
Floor Framing System
the wind forces, building serviceability, and to prevent
tension at the extremities of the tower. In addition, the While, several floor framing systems were considered
relative column shortening between the vertical elements for the tower, only two options were evaluated in details
was also carefully considered to reduce its impact on the for time and cost implications. Option 1 consists of a
structural member performance, architectural detailing, composite floor framing system at both the office zone
and building services. The gravity load support structure and at the residential and hotel zones. The composite
(core walls and columns) are strategically located and option shown in Figure 3 comprise of 135mm trussed deck
proportioned at the building extremities to attract the over composite floor framing. However, option 2 consists
maximum gravity load to maximize their resistance to of 200 to 250 mm two way reinforced concrete slab with
lateral loads. In addition, the reinforced concrete core beams. Considering the height of the building and wind
walls and mega columns are designed on equal stress conditions, option 2 was selected for the final design as
basis under gravity loads. See figure 3 below depicts it is more economical, less sensitive to wind effects and
typical floor framing plans. its impact on work efficiency, constructability, and overall
delivery schedule. Figure 3 also provides the column and
The lateral load resisting system of the tower in the core wall thicknesses throughout the building height for
east-west direction consists of reinforced concrete core option 2. The exterior steel columns in the office zone,
walls that are rigidly connected to the mega-columns at between the mega interior columns, are transferred at
the mechanical levels (every 30 floors) through 4-story level 5 to allow for column free lobby. See figure 5 for the
reinforced concrete or composite shear wall panels as configuration of the column transfer system at and above
shown in Figures 3 & 4. The lateral load resisting system in level 5.
the north-south direction consists of reinforced concrete
core wall that are rigidly connected by 4-story structural 390mm tow way hollow core (bubble deck) system was
steel bridging trusses, provided at the mechanical levels, also considered as it provided flat ceiling as shown in
to complete the mega-frame structural system. Multi- Figure 4. This option required placing solid concrete
story reinforced concrete wall panels between the core beam along the mega column sizes for bracing and
walls and the mega reinforced columns are provided at stability requirements. Note that Figure 4 also provides
the mechanical levels to provide continuity of the 4-Story the concrete strength used for the vertical elements and
structural steel bridging trusses and to integrate the mega their sizes along the building height.
columns in the lateral the lateral system to 1) maximize Detailed finite element analysis model for the floor
the stiffness of the tower 2) prevent tension along the framing system was performed to validate the reinforced
entire building height under extreme loading conditions, concrete floor framing system described above that took
and 3) reduce the relative column shortening between the into construction sequence, cracking, long term and short
core wall and the mega-columns. Figures 2 to 4 depict the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

590 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Optimizing the Structural and Foundation Systems of the 151 Story Inchon Tower : The Development of New Generation of Tall Building System

Wind Engineering Approach

The initial architectural design features of the tower and its
airfoil shape when combined with very sharp edges are not
suitable for a 600m tall and slender tower with 9:1 aspect
ratio. The principal author requested collaboration from
John Portman and Associates (JPA), principal designers,
to modify the design by considering edge treatment and
shape variation along the building height by introducing
corner slots with varying shapes and sizes, which was
later optimized with RWDI for optimum performance.
See Figure 7 for the original design concept and history
Mega Column Size Wall Thickness Size
of the architectural design modification for aerodynamic
Criteria
Concrete
MC1 MC2 MC3 Flange Web
shaping and corner slot treatment. Initial wind tunnel
(Mpa)
(m2) (m2) (m2) (mm) (mm) studies revealed that the tower 1) original design concept,
F127~F15 1 80 4.00 4.75 2.10 600 600 with no slots, exhibited excessive accelerations and wind
F94~F126 80 7.00 4.75 2.90 600 600 forces that resulted in significant impact on the overall
F66~F93 80 9.40 6.5 4.10 800 800 space efficiency and excessive construction cost, 2)
F43~F65 100 11.92 7.75 4.75 1000 1000 overall response is sensitive to the shape and size of slots
F21~F42 100 14.08 11.55 7.20 1200 1000 along the building height, and 3) is extremely sensitive to
B2~F20 100 15.2 11.55 7.20 1400 1000 corner slot shape, size, and orientation. After extensive
studies, the final design of the tower resulted in 40%
Fig. 4: Typical Office and Residential floor framing plans, wall reduction in wind forces and overall dynamic response
thicknesses, and mega column sizes of the tower. The maximum expected acceleration at the
highest occupied floor for one and ten year return periods
were 14.9 and 29.9 milli-g’s respectively, thus requiring
damping system at the top of the building.
Because of the shape of the building, split and slender
tower, the overall response of the tower needs to be
combined with each of the tower legs above the last
bridges, thus complicating the prediction of the tower
response for each leg. During the developing of the
Fig. 5: Typical Office and Residential floor framing plans with hollow tower design, extensive wind tunnel testing force balance
(green Ball) flat plate testing regimes were considered. However, the additional
test needed to conclude the wind tunnel testing programs
and selection of the damping would include 1) aeroelastic
model studies, pedestrian wind studies, cladding wind
load study (including pressure integration), auxiliary
damping system selection and studies, and falling ice and
snow studies.
The wind climate studies include a statistical analysis of
local wind conditions based on measurements from the
Inchon Meteorological Station with due consideration

Fig. 6: Typical Office and Residential floor framing plans with hollow
(green Ball) flat plate

term deformation. Figure 5 provides summary of the

loading criteria, strength of concrete and the expected Fig. 7: Architectural Massing History for Aerodynamic Shaping
slab deflection for the different concrete framing options. with Corner Slot Shaping

Organised by
India Chapter of American Concrete Institute 591
Technical Papers

to the influence of typhoons based on a Monte-Carlo 2) Sit classification = Sd (Stiff Soil), 3) Design spectral
simulation for the area. Deriving the wind loads for acceleration (Seismic Design Category D): SDS=0.425;
the Tower were based on 2005 Korean Building Code SD1=0.246, 4) Occupancy Category = IE = 1.2, 5) Response
and Standards (KBCS 2005) with 30m/sec (basic wind Modification Coefficients = R = 5.5, 6) Periodic Parameter
speed) x 1.1 (importance factor), which resulted in 33m/ (Factored design Seismic Force) = CT= 0.049, and 7)
sec design wind speed. Performance based design was Minimum Design base Shear =Csmin = 0.01. Considering
also considered for wind loads and included the following that the dynamic characteristics of the tower (long period)
criteria: and the fact that the tower is founded on reclaimed land
with approximately 20 meter of soft to firm marine silty
Deflection : 50 yr wind loads, damping = 1.5%, I = 1.0
clay and in close proximity to Japan with high and severe
Accelerations: 1yr and 10 yr predictions at 1.0% and and seismic activities, a site specific seismic hazard
1.5% damping analysis was commissioned that takes into account the
Strength : 100 yr wind loads, damping = 2.0%, I = 1.0 regional tectonic environment, historic seismicity of the
Strength : 300 yr wind loads, damping = 2.5%, I = 1.0 region, effect of near and rare, but far earthquake.
Strength : 1772 yr wind loads, damping = 5.0%, I = 1.0 Site specific response spectrum curves and time history
records were provided for frequent and rare seismic
events. Figure 9 depicts some of the historic earthquakes,
near the Korea Peninsula, that were used for determining
the site specific seismicity. Because of the tower long
period, the lateral load resisting system of the tower is
controlled by wind forces.

Fig. 8: Summary of Story Shear and Overturning Moment along

the building height
Fig. 9: Summary of Story Shear and Overturning Moment along
In addition to the wind design criteria established above, the building height
the wind loads were calculated using the KBCS 2005
using exposures C and D. Comparison of wind loads,
Verification of the Lateral Load Resisting System
story shears and overturning moments are shown in 8.
Note that the wind loads per wind tunnel testing program Detailed three-Dimensional Finite Analysis Models
were significantly higher than those predicted by the code. (FEAM) for both the composite framing, option 1, and for
The design of the tower for strength designed used the the concrete framing option, option 2, were prepared that
worst case conditions. Rational wind loads for 1, 10 and included the floor framing arrangement described above,
50 year with different damping were used as a base for 1) exterior column transfer systems, raft foundation, and
predicting the building displacement under different load the pile foundation flexibility, and the P-delta effects. The
conditions and their combinations and 20 predicting the complete soil structure analysis models as shown in figure
maximum building resultant accelerations and torsional 10. The analysis was also use to simulate the construction
velocities. Taming the dynamic response of the tower
under wind excitation for both frequent and rare wind
events was very challenging as they have to be done within
the constraint of the architectural design parameters.
However, supplementing the tower with reliable damping
system will no doubt tune the dynamic excitation and
reduce significantly the design wind forces. Due to the
limitation of the paper size, the wind engineering works
cannot be fully covered here.

Seismic Engineering
The 151 Inchon Tower is located in Seismic Zone 1
according to the Korean Building Code (KBCS2008) Fig. 10: Structural System Summary for option 1 & option 2
and with 1) Seismic Acceleration Parameter = S = 0.22, Structural concepts And their Corresponding three (3) Dimensional
Finite Element Analysis Models

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

592 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Optimizing the Structural and Foundation Systems of the 151 Story Inchon Tower : The Development of New Generation of Tall Building System

sequence of the tower and its impact on the design of the system provided substantial increases damping and
multi-story outrigger/shear panels. reduction in the dynamic response of the building. The
preliminary analysis results also showed that this type
The analysis also includes a detailed analysis of the soil-
of viscous damper can be applied to control low and high
structure interaction, where the raft foundation and pile
level of dynamic input. The principal author believes
stiffness are modeled in details to expedite the system
that integrating a reliable damping systems in the
studies comprehensively. Because of the tower piles
fundamental design concept of building structure must
are founded in more 20 meters of soft marine silty clay,
be considered to continue to improve the reliability of
the tower lateral movement and dynamic behavior are
building structure for sustainable cities that must work
sensitive the soil subgrade modulus. Therefore, extensive
under extreme events. The principal author believes that
studies and pile testing regimes were put in place by
integrating supplemental damping system into super tall
the principal author to ensure the overall response and
building structure will no doubt provide a catalyst for new
behavior of the tower under extreme loading conditions
generation of tall building systems.
are captured, especially as it related to selecting
appropriate pile stiffness under gravity loads, dynamic
wind and seismic loads, dynamic lateral stiffness of the
pile, and the mechanism of dissipating the lateral load to
the foundation. The tower lateral displacements and inter-
story drifts were limited to H/500 and H/350 for 50 year
return period respectively. The results of the dynamic
frequency analyses are depicted in Figure 10, with and
without foundation flexibility.

Fig. 11: Mega -Toggle Brace Damper A new generation for Tall
Building System

Fig. 10: Modal Period and Participations with foundation flexibility Foundation System: Pile Supported Pile Raft
and (without foundation flexibility) Foundation
The tower superstructure is founded on 5.5 meter thick,
Supplemental Damping System for the Tower high performance reinforced concrete raft foundation
Extensive wind engineering studies and wind engineering over 172-2.5 diameter reinforced concrete piles. The piles
treatment were considered to control the dynamic are designed for 6000MTonnes service capacity and are
excitation of the tower under frequent and sever wind intended to anchor into competent soft rock at least 5
events. However, working within the limitation of the meters. The final pile lengths vary significantly, from 45
design concept, the dynamic response for the building to 76 meters, because of the variability of the geotechnical
was reasonably controlled, but could not meet the weathered and soft rock formation. The selection of the
habitability and service requirements for this high end final pile toe location was also influenced by the presence
residential tower. Thus, a supplemental damping system rock fractures in the soft rock formation. The All piles will
was considered for the tower. Considering the latest be anchored to a minimum of 5 meters into the soft rock..
technological advances in damping devices, the author Selection of the optimum pile size, number of piles,
suggested the toggle damping system at the structural and pile layout were determined from a series of trial
steel coupling trusses as shown in Figure 11. The intent analyses undertaken collaboratively by the geotechnical
of this damping system is to provide a reliable damping designers and the structural designers. To evaluate the
system to control vibration induced by wind and seismic foundation settlement, the geotechnical engineering
excitations for both serviceability and ultimate loading consultants used Plaxis 3D and other relevant programs.
conditions. See figure 11 for preliminary selection of the Based on these analyses, the pile stiffness values were
toggle frame geometry brace geometry and configuration. provide and used to finalize the foundation design. An
Detailed and extensive numerical analyses were independent 3-dimenational finite element analysis,
performed to test the efficacy of the mega-toggled using general analysis programs (MIDAS and ETABS),
brace system geometry and configurations. The was also performed, as shown in figure 10, to include
numerical results demonstrated that meg-toggle brace the soil structure interaction and the stiffening effects

Organised by
India Chapter of American Concrete Institute 593
Technical Papers

meter raft foundation geometry superimposed over the

pile layout, the raft foundation detailed analysis model, and
a summary of the foundation analysis results, including
foundation settlement, behavior of the foundation under
wind loads, the pile axial load distribution summary.
Note that 1) the final pile layout is optimized and result in
equal load conditions under gravity and lateral loads, 2)
the maximum tower predicted settlement and differential
settlement are 42mm and 20mm respectively, and 3) the
expected foundation behavior is linear under lateral loads.
During the main design stage, the pile design was generally
based on theoretical solutions and previous experience
in similar conditions at adjacent sites. However, because
of the complexity of the 151 the Inchon tower, a detailed
pile load testing program was put in place to confirm the
Fig. 12: Raft Foundation Plan superimposed over Pile Layout, and design assumptions and finessing the foundation design.
results of the Foundation Analysis The piles were instrumented so that detailed information
can be derived on the distributions of shaft friction and soil
of the superstructure. This analysis also included the stiffness at various depths along the pile shaft. The following
construction sequence of the tower to allow for additional comprehensive vertical, lateral, and cyclic pile load testing
load redistribution between the piles because of the large programme was developed for the tower foundation piles,
stiffness of the superstructure. Figure 12 shows the 5.5 as shown in table 1 to achieve the following objectives:

Table 1
Summary of Pile Load Test Programs

Test Type Purpose Loading Method Monitoring Items

Vertical (4 Estimation of the end bearing and shaft friction Bi-directional load cells Pile movement of shaft and toe -
No. test piles) capacities within weathered/soft rock. (O-cells) embedded at two Stress, strain along piles.
locations in pile (1 in upper
Evaluation of the vertical pile stiffness - Check of pile shaft and 1 close to pile toe) Pile stiffness underrepetitive/cyclic
response and stiffness to due to static and dynamic/ loading due to wind and seismic loads
repetitive/cyclic loading such as wind and seismic loads

Horizontal (1 Evaluation of the lateral pile stiffness Lateral Loading of the test pile against Lateral load and displacement
No. test & 1 deformation characteristics of UMD around pile head a reaction pile (static &
No. reaction dynamic loading) Pile deflections along the shaft
pile) Check of pile response and stiffens due to static and
dynamic/repetitive/cyclic to loading such as wind and Pile stiffness under cyclic/repetitive
seismic load loading.

Fig. 13: a - e) O-Cell Pile Vertical Load Test and instrumentations, f) Lateral Load Test

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

594 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Optimizing the Structural and Foundation Systems of the 151 Story Inchon Tower : The Development of New Generation of Tall Building System

ll Assess and confirm the constructability and integrity of The vertical load test were carried to ultimate load with
the piles using the proposed construction techniques overall safety factor in access to 3. The lateral load test
(reverse circulation drilled piling techniques). was also subject to dynamic load test to confirm the lateral
soil stiffness under dynamic loads. The results of the pile
ll Allow comparison of measured pile performance
testing program and regimes were covered in separate
with design expectations and refinement of the
papers and has detailed descriptions of the geotechnical
geotechnical parameters adopted in design (e.g.
engineering works performed for the 151 Inchon Tower
ultimate skin friction and end bearing values, pile
and cannot be covered here.
foundation stiffness, effect of dynamic loading on the
pile stiffness, both vertical and lateral, etc.)
References
ll Assess possible variability of pile performance in 1. Ahmad Abdelrazaq, Frances Badelow, SungHo-Kim and Harry G.
relation to variations in ground conditions across the Poulos, 2011. Foundation Design of the 151 Story Incheon Tower in a
foundation footprint. Reclamation Area. Geotechnical Engineering Journal of the SEAGS
& AGSSEA Vol 42 No.2 June 2011.
The pile load test shown in Figure 13 shows the vertical 2. Halla Eng., 2008. Geotechnical Investigation Report : Geotechnical
and lateral pile load arrangement and instrumentation. Investigation on Inchon Tower Area.

Ahmad Abdelrazaq
Mr. Ahmad Abdelrazaq, who earned his B.S. and M.S. in Civil Engineering from the University of Texas at
Austin in 1984 and 1986 respectively, is Senior Executive Vice President, the Head of Highrise/Healthcare/
Aviation and Complex Building Division, and Head of Building Business Development (Marketing & Sales)
Division at Samsung C & T Corporation, Seoul, Korea.
Among many projects at Samsung, Mr. Abdelrazaq has been involved in all aspects of construction planning,
pre-construction services, and structural design of the Burj Khalifa, Jumeriah Gardens in Dubai, Samsung
HQ, Seoul, the 151-story Inchon Tower, and the Yongsan Landmark Tower (620m Tall, 111-Story Tower) in
Seoul. Presently, Mr Abdelrazaq is directly involved in the design and construction of several mixed-use
highrise and complex building projects in Asia and the Middle East, including the Worli development project,
Mumbai, the Tanjong Pagar Tower complex, slated to be the tallest Building in Singapore, the UIC project in
Singapore, and the Rasuna Tower, Jakarta, Dhirubhai Ambani International Convention & Exhibition Center
(DAICEC) , Mumbai, India.
In addition to presenting at several International professional conferences and workshops, Mr. Abdelrazaq
serves also as a lecturer at Seoul National University where he teaches graduate classes for the structural
design of high-rise buildings and spatial structures. Mr. Abdelrazaq served as an adjunct professor at
the Illinois Institute of Technology’s School of Architecture, where his research interest included the
development of innovative structural systems in concrete/steel/composite structures, and in aerodynamic
shaping of super-tall buildings to mitigate wind effects to reduce the dynamic wind forces and resonant
vibration; these mitigation measures were later incorporated in some of the real projects mentioned above.
Mr. Abdelrazaq is also an investigator in a National Science Foundation-sponsored project with the University
of Notre Dame, The University of Western Ontario/ Boundary Layer Wind Tunnel Laboratory and Skidmore,
Owings & Merrill LLP that seeks to improve correlations between actual wind load responses of tall
buildings and those predicted by computer models and wind tunnel studies. Mr. Abdelrazaq continued this
research by sponsoring and financing (through Samsung) real time Structural Health Monitoring Programs
(SHMP) at Tower Palace III and Burj Khalifa Tower. Through this research, Mr. Abdelrazaq conceptualized
and executed one of the most comprehensive research programs in the history of super-tall buildings. Mr.
Abdelrazaq developed several real time SHMP for Burj Khalifa to correlate the predicted structural and
foundation behavior to the actual measured response during construction and the tower lifetime, including
wind and seismic events. Mr. Abdelrazaq is also involved in the CTBUH activities worldwide.
He is member of ASCE, ACI, SEAOI, AISC and IABSE and has received many recognitions and special project
awards, including ‘2015 ASCE Ernest E. Howard Award’ for major contribution to the design, construction,
and full scale monitoring of Signature Tall Buildings and other structures, including Burj Khalifa; ‘State of
the Arts Civil Engineering Award’ for a paper on ‘Validating Wind Induced Response of Tall Buildings”’and
‘SEAOI 2010 Most Innovative Structural Engineering awards’ in 1998, 2003, 2004.

Organised by
India Chapter of American Concrete Institute 595
Technical Papers

Effect of Fly Ash Utilization for the Treatment of Entrapped Air in

Copper Slag Concrete

Koji Sakai Yuya Tanaka Kunpei Watanabe

Japan Sustainability Institute, Goda Komuten Co., Ltd, Takamatsu, Mitsubishi Materials, Naoshima,
Sapporo, Hokkaido, Japan Kagawa, Japan Kagawa, Japan

Abstract The use of copper slag, with a current annual production

of approximately 3 million tons in Japan, has actually
At present, it is required to form a resource-circulation
been directed to sandblasting and cement materials for
society by utilizing industrial byproducts as resources.
the most part. With the declining demand for these uses,
Copper slag is one of them. However, the use of copper
however, the utilization of copper slag as fine aggregate
slag as a concrete ingredient causes the problems such
for concrete has come to be sought.
as entrapped air and bleeding. This paper describes
the effect of fly ash on the dissipation of entrapped air Studies have been conducted on using copper slag as a
in copper slag concrete. The fundamental properties of substitute for fine aggregate, but this entails problems of
concrete using copper slag and fly ash were also studied. air entrapment and bleeding.
It was found that the entrapped air in copper slag concrete Research has revealed that the increase in the spacing
was dispersed by the use of fly ash and the freeze-thaw factor due to the use of copper slag is caused by an
resistance was obtained with the appropriate use of AE increase in the amount of entrapped air due to the shape
Agent. The compressive strength, length change and of copper slag(3). The occurrence of air entrapment is also
carbonation of concrete showed the different combination identified from the fact that an increase in the copper slag
effect according to the replacement ratio of copper slag replacement ratio reduces the air-entraining admixture
and fly ash. dosage(4),(5). It has also been revealed that sufficient
Keywords: Copper slag, fly ash, fine aggregate resistance to freezing and thawing action can be obtained
replacement, compressive strength, length change, by adding a defoamer to eliminate air entrapped by copper
carbonation, freeze-thaw resistance. slag and an air-entraining admixture to entrain air of good
quality(5).

Introduction As to the increase in the amount of bleeding water due to

the use of copper slag, the following are pointed out as the
Modern society, which has achieved economic
causes: the smooth surface and large density (6), low water
development through the consumption of vast resources
absorption(7), small fine particle fraction(8), and vitreous
and energy, is now pressed to shift to sustainable
surface with low water retentivity(4) of copper slag. The
development in the economic, social, and environmental use of fly ash is one solution to increased bleeding, as it
aspects. To this end, the formation of a recycling-based increases the paste viscosity due to the increase in the
society in which resources are effectively utilized is powder content(6) and improves the water retentivity (7),
sought. The construction industry has contributed to thereby reducing the amount of bleeding water.
the formation of such a recycling-based society by using
concrete containing industrial byproducts, such as As to the strength of concrete containing copper slag,
various slags and fly ash. These byproducts have been the compressive strength of concrete containing 5-0.3
actively studied, with various guidelines for their use being mm copper slag decreases due to the increased bleeding
published. However, it cannot be said that these industrial water, which adversely affects the bond interfaces
byproducts have so far been fully utilized as aggregate between copper slag and cement paste (4), as well as to
for concrete. Copper slag and fly ash are two examples of the low percentage of fine particles, which leads to void
such byproducts. formation by bleeding(5). When copper slag 2.5 mm or
less in size is used, the compressive strength of concrete
Copper slag was standardized in August 1997 as a new becomes higher than the case of using 5-0.3 mm copper
fine aggregate material for concrete in JIS A 5011-3(1). slag, due to the lesser amount of bleeding water (4). When
The Japan Society of Civil Engineers also conducted fly ash is used in combination, the compressive strength
investigation to use copper slag for concrete, publishing increases due to the pozzolanic reaction and filling effect
recommendations for its use in 1998 (2) . of fly ash(4), 5), 6), 7).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

596 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fly Ash Utilization for the Treatment of Entrapped Air in Copper Slag Concrete

The length change of concrete containing copper slag is

Table 1
inhibited due to the large density of copper slag and to Properties of materials
bleeding, which reduces internal water (4),(5),(8). It should
be noted that, when the density of copper slag is high, no Material Property
drying-induced water escape is considered to occur. When Density:3.15g/cm3
fly ash is used in combination with copper slag, length Ordinary Portland cement
Blain:3430cm2/g
change is inhibited by pozzolanic reaction, which densifies
the microstsructure(5), (7), and the filling effect of fly ash(5). Surface-drying density:2.60g/cm3
On the other hand, water escape from the microstructure, Andesite crushed sand
Absorption:2.16%
which is densified by pozzolanic reaction, can increase the
drying shrinkage(4). Surface-drying density:3.49g/cm3
Copper slag CUS5-0.3
Absorption:0.73%
Carbonation of concrete is accelerated by the use of
copper slag, as the internal structure becomes less dense Density:2.31g/cm3
due to the increase in the amount of bleeding water4),(9). Fly ash(FA) Blain:4030cm2/g
When fly ash is used in combination, the carbonation
depth decreases at early ages due to the filling effect of Ignition loss: 1.80%
fly ash(4), but carbonation is accelerated at later ages due Blain:4170cm2/g
to the consumption of calcium hydroxide by pozzolanic Low carbon fly ash(nc)
Ignition loss: 0.70%
reaction(4),(6).
Surface-drying density:2.57g/cm3
In this manner, the use of copper slag for concrete with Crushed sand stone 1505
or without fly ash causes complicated effects on the basic Absorption:1.83%
properties as concrete. The treatment of entrapped air Surface-drying density:2.58g/cm3
particularly poses a major concern. Crushed sand stone 2015
Absorption:1.56%
With this as a background, the authors investigated the Highly-functional AE water- Modified lignin-sulfonic acid
effect of fly ash as a measure against air entrapment as reducing agent compound
an alternative for the use of a defoamer. Additionally, the
High-range AE water-
basic properties of concrete containing both copper slag reducing agent
Polycarboxylate ether type
and fly ash were investigated. The use of fly ash with a
AE agent Modified rosin acid compound
small ignition loss was investigated as well.
Antifoam agent (D) polyalkylene glycol derivative

Experiments
Table 2
Materials Mixture proportions of concrete
Table 1 gives the types and qualities of concrete materials (W/C=0.55, s/a=46%, Water=170kg/m3, C=309kg/m3)
used in this study. The grading of crushed sand conforms
Cru-
to JIS A 5005. Though the grading of copper slag alone Copper Fly
Cru- Copper
Fly
Cru-
shed
Anti-
does not meet the JIS requirements, its use in combination shed slag shed foam
No slag ash ash coarse
sand (CUS5 sand agent
with crushed sand satisfies the requirements. Production (Cu) (FA)
(S) - 0.3)
(FA)
(S)
aggre-
(D)
and proportioning of concrete gate (G)

Concrete was produced using a forced mixing-type twin- (Volume %) (kg/m3) (C×%)
axis mixer with a capacity of 100 liters in a laboratory 1 0 0 100 0 0 822 957 0
at 20°C and 60% R.H. The batch size was 65 liters. The
2 15 0 85 165 0 698 957 0
mixing procedure was as follows: Dry mix cement, fine
aggregate, and coarse aggregate in the mixer for 15 sec; 3 15 0 85 165 0 698 957 0.008
add water and the chemical admixture; and mix for 120 4 15 10 75 165 73 616 957 0
sec.
5 15 20 65 165 146 534 957 0
Table 2 gives the mixture proportions of concrete. The
6 30 0 70 331 0 575 957 0
target slump and air content were adjusted by the type
and dosage of the chemical admixture. In this study, an 7 30 10 60 331 73 493 957 0
air-entraining and high-range water-reducing (AEHRWR)
8 30 10 60 331 73 493 957 0.004
admixture was used for all mixtures excepting mixture
No. 1, as these mixtures did not attain the target slump 9 30 20 50 331 146 411 957 0
with a high function air-entraining admixture. Also, air 20
10 30 50 331 148 411 957 0
was entrained without using a defoamer in this study, (nc)

Organised by
India Chapter of American Concrete Institute 597
Technical Papers

(5) Accelerated carbonation

Table 3
Slump and air Accelerated carbonation tests were conducted in
accordance with JIS A 1153 on beam specimens 10
No Cace Slump(cm) Air(%)
by 10 by 40 cm in size. These were water-cured for
1 Cu0FA0 8.1 5.8 4 weeks at 20C as precuring and then left to stand
for 4 weeks in a thermohygrostatic chamber at 20C
2 Cu15FA0 7.3 4.8 and 60% R.H. The carbonation depth was measured as
3 Cu15FA0(D) 7.2 5.8
follows: Cut the specimens in the transverse direction;
immediately spray a 1% phenolphthalein solution onto
4 Cu15FA10 10.3 5.9 the cleft surfaces; measure the distance from the
surface to the area colored in purple at 5 points on
5 Cu15FA20 10.8 6
each side (10 in total); and determine the carbonation
6 Cu30FA0 7.4 4.3 depth by averaging the measurements.

7 Cu30FA10 7.7 4.3 (6) Freezing and thawing

8 Cu30FA10(D) 7.5 3.4

Freezing and thawing tests were conducted in
accordance with JIS A 1148. Beam specimens
9 Cu30FA20 9.5 5.7 measuring 10 by 10 by 40 cm were demolded on the
day following the placing day and water-cured for 4
10 Cu30FA20(nc) 10.9 4.5
weeks at 20C. Specimens were then placed in testing
equipment and subjected to 6 freezing and thawing
excepting two mixtures for examining the effect of the cycles every day. The relative dynamic modulus (RDM)
defoamer. When the target air content range was achieved was measured up to 300 cycles at intervals of less
without using the air-entraining admixture, specimens than 36 cycles.
were fabricated with that air content.
The water-cement ratio (W/C) and sand percentage were Results and Discussion
kept constant at 55% and 46%, respectively, for all mixtures
in this study. Tests were conducted on ten mixtures: one Slump and air content
in which fine aggregate consisted only of crushed sand; Table 3 gives the slump and air content test results.
those in which 15% or 30% of fine aggregate was replaced Though all measurements fell in the target air content
with copper slag, with or without fly ash further replacing range of this study of 4.5  1.5%, the measured air
10% or 20% of fine aggregate; one containing a defoamer; content values relatively widely scattered from 3.4% to
and one which contained fly ash with a small ignition loss. 6.0%, as the entrapped air made it difficult to adjust the
Note that the specimens were compacted using a tamping air content.
rod.
Admixture dosage
Test items and procedures Figure 1 shows the dosage of the AEHRWR admixture
(1) Slump and air content for each replacement ratio of copper slag and fly ash.
The AEHRWR admixture dosage increases as the copper
Slump and air content tests were conducted in
slag replacement ratio increases, presumably because
accordance with JIS A 1101 and JIS A 1128, respectively
interlocking between copper slag particles and between
(2) Bleeding copper slag and aggregate causes reduction in the
consistency4). Also, the AEHRWR admixture dosage of
Bleeding tests were conducted in accordance with JIS
the mixture with a copper slag replacement ratio of 30%
A 1123.
decreases with the use of fly ash in combination, due to
(3) Compressive strength the water-reducing effect resulting from fly ash’s ball-
Specimens for compression tests were demolded on bearing action4). With a copper slag replacement ratio
the day following the placing day, water-cured at 20C of 15%, however, an increase in the fly ash replacement
until the specified ages, and subjected to testing in ratio increases the AEHRWR admixture dosage. This is
accordance with JIS A 1108. presumably because the effect of low mobility due to the
increase in the powder content is greater than the ball-
(4) Length change bearing effect.
Length change tests were conducted on beam Figure 2 shows the AEHRWR admixture dosage and air
specimens measuring 10 by 10 by 40 cm by the dial content for each replacement ratio of copper slag and
gauge procedure in accordance with JIS A 1129. fly ash. When copper slag is the only replacement, the

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

598 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fly Ash Utilization for the Treatment of Entrapped Air in Copper Slag Concrete

specified air content is attained by the entrapped air

even without the AE admixture. When fly ash is used in
combination, the AE admixture dosage increases, as
unburnt carbon contained in fly ash adsorbs the entrapped
air. The mixture with 15% copper slag and 10% fly ash
attained an air content equivalent to the mixture with no
copper slag or fly ash by the same AE admixture dosage as
this mixture. On the other hand, the AE admixture dosage
of the mixture containing fly ash with a small ignition loss
is smaller than that of mixtures containing fly ash with a
high unburnt carbon content. This is presumably because
the small unburnt carbon content of fly ash with a small
ignition loss leads to a small amount of entrapped air
adsorption. Fig. 3: Bleeding ratio
These results demonstrate that the use of fly ash is
effective in eliminating the air entrapped by copper slag. Compressive strength
Figures 4, 5, 6, and 7 compare the compressive strengths
Bleeding of mixtures with different copper slag replacement ratios,
Figure 3 shows the bleeding ratios. The bleeding with different fly ash replacement ratios, with and without
ratio increases as the copper slag replacement ratio the defoamer, and with different fly ash types, respectively.
increases but is reduced by the addition of fly ash. A fly Figure 4 reveals that the 91-day compressive strength
ash replacement ratio of 10% halves the bleeding ratio increases as the copper slag replacement ratio increases,
when the copper slag replacement ratio is 15%. When the presumably due to the high density of copper slag.
copper slag replacement ratio is 30%, however, 10% fly The 28-day compressive strength of 30% copper slag
ash is not sufficient to control the bleeding ratio. specimens is lower than 15% copper slag specimens.
This low strength can be attributed to the adverse effect
of increased bleeding on the bond interfaces between
copper slag and cement paste(4) and the small fine particle
fraction and bleeding causing voids in the structure(5).
When comparing the mixtures containing copper slag
with and without fly ash in Fig. 5, the 91-day compressive
strength of the mixture containing both copper slag and
fly ash is found to be higher. This is presumable due to the
pozzolanic reaction and filling effect of fly ash(4),(5),(6),(7). The
strength behavior of mixtures at other ages is complicated
and difficult to explain in a uniform manner.
Figure 6 reveals no marked difference between the
compressive strengths of mixtures with and without the
Fig. 1: Amount of high-range AE water reducing agent defoamer. It is therefore considered that entrapped air
scarcely affects the compressive strength.

Fig. 2: Amount of AE agent and air content Fig. 4: Compressive strength (Effect of copper slag content)

Organised by
India Chapter of American Concrete Institute 599
Technical Papers

Fig. 5: Compressive strength (Effect of fly ash content) Fig. 8: Length change (Effect of copper slag content)

Fig. 6: Compressive strength (Effect of defoamer agent) Fig. 9: Length change (Effect of fly ash content)

Figure 9 demonstrates that, with a copper slag

replacement ratio of 15%, the length change significantly
decreases by using 10% fly ash. However, the ratio of
length change reduction decreases when the fly ash
replacement ratio increases to 20%. This can be explained
as follows: Fly ash with a replacement ratio of 20% inhibits
bleeding, and the retained water escapes, increasing the
length change. With 30% copper slag, however, the water-
retaining effect of 10% fly ash is insufficient for producing
a significant effect. Water retention is achieved more
effectively with 20% fly ash, but unretained water escapes
to cause shrinkage. This is offset by the expansion due
to pozzolanic reaction of fly ash, resulting in a reduced
Fig. 7: Compressive strength (Effect of carbon content)
length change ratio.
Figure 7 reveals that the use of fly ash with a small ignition
Carbonation depth
loss increases the compressive strength. This is not
because of the use of this type of fly ash but because the Figure 10 shows the carbonation test results. The
air content of the mixture containing fly ash with a small carbonation depth increases by replacing part of fine
ignition loss turns out to be smaller than the mixture aggregate with copper slag, presumably because the
containing normal fly ash. increased amount of bleeding water makes the internal
structure less dense(4),(9). With copper slag with a
Length change replacement ratio of 15%, the use of 20% fly ash tends to
slightly reduce the carbonation depth, but the difference is
Figures 8 and 9 show the length change ratios by copper
not significant. With 30% copper slag, however, 10% fly ash
slag replacement ratio and fly ash replacement ratio,
leads to a reduction in the carbonation depth. In this case,
respectively. Figure 8 reveals that, without fly ash, no
the bleeding-inhibiting effect of fly ash is not sufficient.
appreciable difference is observed between the length
This reduction in the carbonation depth is presumably
change ratios of mixtures with and without copper slag.
due to the effect of the reduced W/C. With 20% fly ash, its

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

600 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fly Ash Utilization for the Treatment of Entrapped Air in Copper Slag Concrete

Fig. 10: Carbonation depth Fig. 12: Relative dynamic modulus of elasticity (Effect of fly ash
content
bleeding-inhibiting effect increases, but the denseness of
the microstructure is reduced in return, increasing the
carbonation depth to a degree similar to the case of no fly
ash, where the W/C decreases by bleeding. Actually, these
are mixed with the effect of pozzolanic reaction of fly ash.

Resistance to freezing and thawing

Figures 11, 12, 13, and 14 compare the relative dynamic
moduli of mixtures with different copper slag replacement
ratios, with different fly ash replacement ratios, with and
without the defoamer, and with different fly ash types,
respectively.

Fig. 13: Relative dynamic modulus of elasticity (Effect of antifoam

agent)

Fig. 11: Relative dynamic modulus of elasticity (Effect of copper

slag content)

Figure 11 reveals that the relative dynamic modulus Fig. 14: Relative dynamic modulus of elasticity(Low carbon Fly
(RDM) of mixtures with fine aggregate partly replaced ash effect)
only with copper slag is extremely low. This is due to
the absence of good quality entrained air. The fact that the increased copper slag content increases entrapped
entrapped air lowers the resistance to freezing and air and the unburnt carbon to adsorb the entrapped air is
thawing is demonstrated also by Fig. 13, in which the RDM insufficient in 10% fly ash. When the replacement ratio of
of mixtures solely replaced with copper slag is improved fly ash is increased to 20%, this problem is solved, with
by eliminating the entrapped air by the defoamer and sufficient air being entrained by the ARHRWR admixture.
entraining good quality air by the AE admixture. Thus the RDM increases, achieving sufficient resistance to
frost damage. Figure 13 demonstrates that, even with 30%
Figure 12 reveals that the mixture with 15% copper slag copper slag and 10% fly ash, resistance to frost damage
and 10% fly ash achieves sufficient resistance to freezing can be improved by using a defoamer and entraining
and thawing. However, the RDM significantly decreases sufficient air using an AE admixture. As found from Fig. 14,
with 30% copper slag and 10% fly ash. This implies that fly ash with a small ignition loss leads to the RDM similar

Organised by
India Chapter of American Concrete Institute 601
Technical Papers

to normal fly ash. It is therefore inferred that this type of (7) While the entrapped air content varies depending on
fly ash also has the effect of adsorbing entrapped air even the replacement ratio of copper slag, it is necessary to
with a small ignition loss. eliminate the entrapped air using a sufficient amount
of fly ash, in order to achieve resistance to freezing and
thawing. Air entrainment is also necessary using an
Conclusions
air-entraining admixture.
The results obtained in this study are summarized as
follows: (8) Fly ash with a small ignition loss also has an effect of
adsorbing entrapped air.
(1) It was reconfirmed that copper slag in place of part
of fine aggregate entraps air, as the specified air References
content was attained without using an air-entraining 1. JIS A 5011-3:2003, Slag aggregate for concrete – Part 3 Copper
admixture. slag aggregate
(2) Unburnt carbon in fly ash was found to be effective in 2. JSCE, Guide for Construction of concrete using copper slag
aggregate, Concrete Library 92, Japan Society of Civil Engineers,
eliminating air entrapped by copper slag.
1998
(3) An increase in the copper slag replacement ratio 3. Tetsuya Murakami, Yasuhiro Ido, Nobuharu Gomi, and Yoshihiro
increases the amount of bleeding water. Masuda: Mixture and fundamental properties of high-density
concrete using byproduct – Part 2 Length change and freeze-thaw
When fly ash is used in combination with copper slag, resistance, Summaries of technical papers of annual meeting,
bleeding is inhibited, but its degree depends on the Architectural Institute of Japan, 2006
replacement ratio of fly ash. 4. Yoshitaka Okawa and Koji Sakai: Properties of fresh and hardened
concrete using copper slag and fly ash as a part of fine aggregate,
(4) An increase in the copper slag replacement ratio Proceedings of the Japan Concrete Institute, Vl.33, No1, pp125-
increases the 91-day compressive strength. 130, 2011
5. Mitsuhiro Ishii, Kota Deguchi, Koji Sakai, and Shinji Abe: Performance
The use of fly ash in combination with copper slag
improvement of copper slag concrete by use of fly ash, Cement
increases or reduces the compressive strength Science and Concrete technology, Japan Cement Association, Vol.66,
depending on the conditions. Also, entrapped air No.1,2012
scarcely affects the compressive strength. 6. Takashi Kaji, Mitsuhiro Ishii, and Hirohiko Iwahara: Properties of
concrete using fly ash and copper slag fine aggregate, Proceedings
(5) The use of copper slag in combination with fly ash of the Japan Concrete Institute, Vl.26, No1, 2004
increases or reduces the length change, depending
7. Keisuke Ishimaru, Hiroyuki Mizuguchi, Chikanori Hashimoto,
on the degree of water escape, which depends on the Takao Ueda, Kazuhiro Fujita, and Masaaki Ohmi: Properties of
degree of water retention by fly ash, and the degree of concrete using copper slag and second class fly ash as a part of
expansion due to pozzolanic reaction of fly ash. fine aggregate, Journal of the Society of Materials Science, Vol.54,
No.8, 2005
(6) Copper slag in place of part of fine aggregate generally 8. Jun-ichi Watanabe, Hiromi Tamura, Takashi Fujii, and Katsunori
increases the carbonation depth due to the effect Ayano: The utilization of copper slag for concrete materials,
of bleeding. When fly ash is used in combination, Proceedings of the Japan Concrete Institute, Vol.32, No.1, 2010
the degree of carbonation varies depending on the 9. Kentaro Kyo, Daisuke Komori, Yoshitaka Kato, and Taketo Uomoto:
denseness of the microstructure, which depends on Effect of mix proportion and condition of construction on cold joint
the degree of water retention by fly ash, ultimate W/C, in concrete, Proceedings of the Japan Concrete Institute, Vol.22,
No.1, 2000.
and degree of pozzolanic reaction.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

602 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effect of Fly Ash Utilization for the Treatment of Entrapped Air in Copper Slag Concrete

Prof. Koji Sakai

Prof. Koji Sakai is the representative of the Japan Sustainability Institute, Sapporo, Japan. He was a
Professor at Kagawa University until March 31, 2014. He also worked for Hokkaido University, Civil
Engineering Research Institute of Hokkaido Development Bureau, the University of Canterbury, Houston
University. Currently, he has some professorships including Honorary Professor of the University of British
Columbia University (Canada), Guest Professor at Southeast University (China), and Special Professor at
King Monkut’s University of Technology – North Bangkok (Thailand).
He has authored or coauthored over 700 technical papers, reports, and articles. He founded ISO/TC71/
SC8, Environmental Management for Concrete and Concrete Structures, in 2007. He also founded two
international conferences: International Conference on Concrete under Severe Conditions (CONSEC) in
1995 and International Conference on Concrete Sustainability (ICCS) in 2013. Sakai has organized many
workshops in the world and forums including ACI Concrete Sustainability Forum and ACF Sustainability
Forum. He was the chair of fib Commission 3 on environmental aspects, JCI Sustainability Committee, and
other many committees.
He has received many awards, including the JSCE Yoshida Prize in 1997, CANMET/ACI Award in 2006, the
JSCE Yearbook award in 2013, ACI Concrete Sustainability Award in 2014, and fib Medal of Merit in 2015.
He has been invited to more than 90 international conferences/seminars as the keynote/invited speaker on
environmental and sustainability issues, which are his main research interests.

Organised by
India Chapter of American Concrete Institute 603
Technical Papers

Self-Compacting Concretes Using Fly Ash and Fine Limestone Powder:

Shrinkage and Surface Electrical Resistivity of Equivalent Mortars

A. Durán-Herrera J. De-León-Esquivel D.P. Bentz Pedro Valdez-Tamez

Professor and Head of Civil Engineer (2012) and MSc Chemical engineer in the Professor and Principal of the
Concrete Technology at the in Construction Materials Engineering Laboratory at the Civil Engineering School and
School of Civil Engineering, (2014) from Universidad National Institute of Standards Institute, Universidad Autónoma
Universidad Autónoma de Autónoma de Nuevo León and Technology (NIST), de Nuevo León, Monterrey,
Nuevo León, Monterrey, (UANL), Monterrey, Nuevo Gaithersburg, MD, USA Nuevo León, Mexico.
Nuevo León, Mexico. León, Mexico.

Abstract leads to a decrease of the concrete’s permeability and

Self-compacting concrete (SCC) has become a consequently to an increase of its durability2.
preferred option for many projects that should satisfy The low water-to-cementitious materials ratios commonly
strict fresh state properties that are of major concern used for SCC range from 0.3 to 0.403, and in combination
in quality assurance. To ensure stable and robust fresh with a higher powder content, often lead to an increase
state properties, typically a significant amount of fine of the total shrinkage of the concrete4, which is typically
materials is incorporated, but this often increases the combination of two processes nominally present
shrinkage. For this purpose, fly ash (FA) has been used, at different stages of the life of concrete: autogenous
but because it can induce delays in times of setting, shrinkage and drying shrinkage. The first one is more
it is not extensively used. Under this scenario, micro significant at early ages, and concerns volumetric changes
limestone powders (L) have been effectively used to that involve self-desiccation inducing capillary stresses
counteract the delays in the times of setting of concrete and leading consequently, to cracking. The second
with high volumes of fly ash. one occurs at a later age along with the contribution of
For a fixed water/powder ratio equivalent to a water-to- temperature and concrete restraint and is due to moisture
cement ratio of 0.40 in a 100 % cement mixture, a total of loss to the surrounding environment by evaporation5, 6.
thirteen mortars were produced to evaluate the synergetic Shrinkage can cause cracking that could lead to the failure
effects of twelve portland cement substitutions by FA+L of the concrete structure, because cracks in concrete will
on times of setting, compressive strength, shrinkage and reduce mechanical properties and durability7.
electrical resistivity. Results indicate appropriate FA+L Due to the advantages offered by the incorporation of
combinations to counteract delays in times of setting and industrial byproducts into concrete, fly ash (FA) obtained
to significantly improve electrical resistivity and volume from coal combustion is one of the most commonly
stability.
used SCMs for concrete, chiefly because of the benefits
Keywords: Fly ash, micro limestone powder, self- it provides to the rheological, mechanical and durability
compacting concrete, setting times, shrinkage, surface performance, and for its contributions to concrete industry
electrical resistivity. sustainability. However, the early-age reactivity of FA is
generally lower than that of portland cement, such that
it contributes to delays in times of setting and to slower
Introduction
strength development. In order to reduce these effects,
Since its first use in the late 1980’s, Self-Consolidating some researchers have studied limestone additions with
Concrete (SCC) has been used as a replacement a mean particle size of about one micrometer as an option
of conventional Vibrated Concrete (VC) because of to diminish these delays in setting 8-10.
its numerous advantages, including labor savings,
construction cost reductions, and ease of placing in
heavily reinforced members, among others1. One of the Research Significance
most important differences between SCC and VC is the Limestone powder has been used previously in order to
incorporation of higher amounts of fillers in the SCC, as counteract the delays in setting times encountered in
this type of concrete demands a higher powder content for most concretes made with FA; however, only a few of these
its formulation. These fillers often include supplementary works have evaluated the effects of the combination of
cementitious materials (SCMs) in replacement of portland limestone powder and FA on the shrinkage and durability
cement and to improve the fresh and hardened state of concrete. Results of shrinkage, both autogenous and
properties of the SCC, since these powders can improve drying shrinkage, as well as a durability index (surface
the densification of the cementitious matrix, which electrical resistivity) are presented in this study in order

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

604 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Self-Compacting Concretes Using Fly Ash and Fine Limestone Powder: Shrinkage and Surface Electrical Resistivity of Equivalent Mortars

to demonstrate the beneficial synergistic effects of a

Table 1
FA produced in Northeast Mexico and a commercial Chemical composition and specific gravity of PC, FA and L
micro limestone powder (L), regarding times of setting,
compressive strength, volume stability and surface Component (%) PC FA L
electrical resistivity.
SiO2 17.66 67.48 0.2
Procedure Al2O3 4.72 23.94
Fe2O3 2.26 4.63
Materials and Mixture Proportions
CaO 63.68 1.76
The materials used in this study are: an ASTM C15011 Type
I Portland cement (PC); an ASTM C61812 type F fly ash (FA) MgO 1.2 1.16
and a commercial micro limestone powder (L), the latter SO3 5.12 -
with an average particle size of 0.7 μm (according to the Na2O 0.66 1.14
manufacturer’s technical data sheet). The physical and
K2O 0.74 0.99
chemical characteristics of these powders are given in
Table 1. Also, the corresponding particle size distributions TiO2 0.24 0.94
(PSDs) as measured using laser diffraction are shown in P 2 O5 0.09 0.03
Figure 1. Table 2 presents the parameters associated with MnO 0.08 -
the PSDs of PC, FA and L; each parameter indicates the
(SiO2+Al 2O3+Fe2O3) %  96.05
particle size of the materials for which 10 %, 50 % and 90
% of the total sample fall below and are referenced as D10, CaCO3, %  > 99
D50 (also known as the median), and D90, respectively. From Specific gravity, g/cm3 3.10 2.03 2.68
these results, it can be observed that the FA particles are
4 times coarser than the portland cement particles; while
the micro limestone particles are up to 11 times finer than
the portland cement particles.
A limestone from Nuevo León, Mexico that meets
ASTM C33 requirements13 was used as coarse and fine
aggregates. To obtain the target flowability and stability,
two ASTM C49414 admixtures were used: a Type F water
reducer and a Type S anti-wash out admixture as a
viscosity modifier.
Reference mortar proportions were established from
a Self-Consolidating Concrete (SCC) with a water-to-
cement (w/c) ratio of 0.4 by mass and with optimized
proportions to meet the following target properties: slump Fig. 1: Particle size distributions of PC, FA and L
flow of 56 cm ± 1 cm according to ASTM C161115, static
segregation of 2.6 % ± 0.2 % using the segregation column
Table 2
according to ASTM C161016, and a flow of 51 cm ± 1 cm for D10, D50 and D90 values in μm of PC, FA and L
the J-Ring passing ability test determined according to
ASTM C162117. Reference mortar proportions kept a w/c Material D10 (μm) D50 (μm) D90 (μm)
of 0.4, and were obtained by the Concrete Equivalent PA 3.0 18.4 46.2
Mortar (CEM) method18. Self-Consolidating Concrete (Ref
FA 14.0 71.2 242.1
SCC) and reference concrete equivalent mortar (Ref CEM)
proportions are given in Table 3. L 0.6 1.7 4.7

In addition to Ref CEM made just with cement, twelve (12)

Experimental Methods
mortars, distributed in three groups, were produced to
evaluate different combinations of PC+FA+L. In order to Fresh state characterization consisted of: flow as
eliminate volumetric variations attributed to the significant a measure of the mortar workability with the cone
differences in the specific gravities of the powders (PC, specified in ASTM C1437 19; times of setting according
FA and L), based on the Ref CEM mixture, volumes of to ASTM C40320; unit weight and air content obtained
water and powders were maintained constant in all of the through the procedures described in ASTM C185201.
mixtures. The composition of the paste fraction of these When determining the unit weight and air content, the
mortars varies according to the proportions presented in mold was filled in one layer and 12 strokes were applied
the experimental matrix in Table 4. around the perimeter of the mold with a rubber head
mallet weighing 0.6 kg ± 0.2 kg.

Organised by
India Chapter of American Concrete Institute 605
Technical Papers

Table 3 Table 5
Reference concrete equivalent mortar proportions Measured fresh stage properties of mortars
(Ref CEM) per cubic meter of concrete/mortar, obtained from
SF1 Self Consolidating Concrete (Ref SCC) Unit weight Air content
Group Identification Flow (cm)
(kg/m3) (%)
Materials Ref SCC Ref CEM
REF REFERENCE 17.0 2188 3.0
Reaction, L/m 3
166.2 221.7
Tap water 1PC-FA 16.0 2070 9.9
Absorption, L/m3 25.1 33.4
1PC-FA-5L 17.0 2099 8.9
Type I Portland cement (kg/m3) 420.0 560.2 A
1PC-FA-10L 17.0 2122 8.1
Amount (L/m3) 2.7 3.6
HRWA 1PC-FA-15L 16.5 2130 8.0
Dosage (ml/kg) PC) 6.5 6.5
2PC-FA 17.0 2061 9.1
Amount (L/m3) 0.10 0.13
VMA 2 PC-FA-5L 17.0 2091 8.1
Dosage (ml/m3) 100.0 133.4 B
2 PC-FA-10L 17.0 2111 7.4
Coarse aggregate (kg/m3) 696.7 -
2 PC-FA-15L 16.5 2130 6.8
Fine aggregate (kg/m3) 1136.3 1532.2
3PC-FA 18.0 2036 9.1

3PC-FA-5L 18.0 2081 7.3

Table 4 C
Experimental matrix, % in volume 3PC-FA-10L 17.0 2124 5.7

3PC-FA-15L 17.5 2147 4.9

Group Identification PC FA L

REF REFERENCE 100 0 0

mortar, all of the specimens were cast with only
1PC-FA 70 30 0 one layer of mortar without any compaction energy;
1PC-FA-5L 70 25 5 afterwards, the standard procedure described in
A ASTM C10922 was followed for storage, curing and for
1PC-FA-10L 70 20 10
the determination of compressive strength.
1PC-FA-15L 70 15 15
2) Autogenous strain: measured on triplicate specimens
2PC-FA 55 45 0 according to the standard procedure from ASTM
2 PC-FA-5L 55 40 5 C169823. For each mortar mixture, the first reading
B was taken at the final time of setting, and thereafter
2 PC-FA-10L 55 35 10
up to an age of 28 d. Because the standard procedure
2 PC-FA-15L 55 30 15 of this test requires significant care, precision and
3PC-FA 40 60 0 permanent monitoring, intermediate mixtures
1PC- FA-10L, 2PC-FA-10L and 3PC-FA-10L were not
3PC-FA-5L 40 50 5 considered for this test since their results should be
C
3PC-FA-10L 40 50 10 between those exhibited by the corresponding 5L and
3PC-FA-15L 40 45 15
15L mixtures.
3) Drying shrinkage: Measurements were made on
Results of single tests are reported in Table 5; no triplicate specimens and ultimate drying shrinkage
uncertainty of measurements could be calculated. All of was determined according to the procedure described
the flow results meet the targeted consistency of 17 cm in ASTM C59624. Average and ultimate drying shrinkage
± 1 cm. For the air content, the precision and bias section were calculated and estimated for ages of 28 d and 52
of ASTM C185 reports a single-operator within laboratory w, respectively, according to the procedures described
standard deviation of 0.56 % for air contents in the range in the standard test method (Section 12, note 2 and
of 8 % to 19 %. No uncertainty is reported for unit weight Figure 1).
measurements. 4) Surface electrical resistivity: was obtained with a four-
After fresh state characterization, mortar test specimens point resistivity meter operating on the principle of the
were cast and cured for the determination of the following Wenner probe and with 38 mm inter probe spacing
hardened state properties: according to the AASHTO TP 95-11 specification25. For
this test, 3 cylindrical specimens of 10 cm in diameter
1) Compressive strength :measured on triplicate and 20 cm in length were made for each mortar
specimens at 7 d, 28 d and 56 d using 50 mm standard mixture and were stored in water saturated with
cubes. Due to the self-consolidating nature of the calcium hydroxide. Measurements were taken at ages

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

606 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Self-Compacting Concretes Using Fly Ash and Fine Limestone Powder: Shrinkage and Surface Electrical Resistivity of Equivalent Mortars

of (1, 2, 3, 4, 5, 7, 14, 21, 28, 35, 42, 49 and 56) d. The

measured electrical resistance (Ω) was obtained as the
average of 12 measurements (with 4 measurements
per cylinder). Thereafter, surface electrical resistivity
was calculated with Eq. (1):
t = R $ k .......................................................................(1)

where R is the measured resistance and k is the

geometric correction factor used for cylindrical
specimens as shown in Eq. (2):
2r a
k= 0.730 7.82
1.10 - + ........................................(2)
d /a (d/a) 2
Fig. 3: Compressive strength of mortar mixtures
This formula is only valid when cylinder (mm), is the
inter probe distance of the Wenner apparatus and is fineness of the L, because it has a smaller particle size
the length of the cylinder (mm)25. than the portland cement, representing a greater surface
area available as nucleation centers that promote the
Results and Discussion formation of portland cement hydration products28, 29.

Fresh Properties Compressive Strength

The measured fresh properties for the thirteen mortar The compressive strength results at ages of 7 d, 28 d, and
mixtures are provided in Table 5. While the flows of the 56 d are given in Figure 3. Groups A, B and C show results
mortars were similar, the air contents of all of the mortars from five binary blends (PC+FA), and ternary blends with
containing FA or FA and L were significantly higher than varying dosages of L that go to 15 % in substitution of the
that of the reference mortar (3 %). The replacement of FA volume in the cementitious system.
FA with L decreased the measured air content in all three Compared to the reference mortar mixture, the
series of mortars, but not back to the level of the reference. substitutions of PC by FA evaluated in groups A, B and C
(30 %, 45 %, and 60 % FA by volume) produced reductions
Times of Setting of the compressive strength of 31 %, 61 %, and 63 % at 7 d,
FA causes delays in the times of setting as reported of 30 %, 37 %, and 54 % at 28 d, and of 23 %, 30 %, and 48
previously27, a characteristic observed in Figure 2, where % at 56 d, respectively.
the results of the initial and final times of setting are
On the other hand, the effect of L as a filler agent and
reported for the thirteen mortar mixtures evaluated in
nucleation center, combined with the pozzolanic effect
this study.
of FA, led to higher compressive strengths at 56 d for
In Figure 2, it is evident that the incorporation of L reduced mixtures 1PC-FA-5L, 2PC- FA-15L and 3PC-FA-10L in
the retardation caused by FA. It can be observed that for comparison to their respective reference (1PC-FA, 2PC-
a 5 % substitution of FA by L, times of setting similar to FA and 3PC- FA), specifically leading to increases of 10 %,
those of the reference mortar were obtained. Further 4 %, and 4 %, respectively.
substitutions of FA by L lead to greater reductions in
The standard deviations for the compressive strengths
times of setting. The acceleration can be attributed to the
reported in Figure 3 were generally within the range
of variations permitted by ASTM C109 (8.7 % for three
cubes and 7.6 % for two cubes). For each average
result, the standard errors calculated as an estimate
of the uncertainty with regard to the estimation of the
compressive strength were within the following ranges:
1.09 % to 3.98 % for the reference mixture, 0.11 % to 4.83
% for group A, 0.10 % to 14.74 % for group B, and 0.10 % to
11.13 % for group C.

Autogenous Strain
To obtain reliable results, only the ten mixtures identified
in Table 6 were evaluated. The total autogenous strains at
28 d are given in Table 6, the negative sign corresponds
to shrinkage. For groups A, B and C, autogenous strains
Fig. 2: Measured times of setting of mortars

Organised by
India Chapter of American Concrete Institute 607
Technical Papers

systems of PC+FA or PC+FA+L that led to a reduction

Table 6
Measured fresh state properties of mortars in self-desiccation. Substitutions of FA by L also have
a significant influence, with the mixtures with 5 % of L
Autogenous Strain Standard having the higher reductions in autogenous shrinkage for
Group Identification strain at reduction deviation
28 d (μm/m)* %* (μm/m) groups A, B and C.
REF REFERENCE -378 0 26 The mortar mixtures with FA contents of 45 % and 60 % by
volume (2PC-FA and 3PC-FA) showed a slight expansion
1PC-FA -230 39 12
after the final time of setting, which was more noticeable in
A 1PC-FA-5L -146 61 15 mixture 3PC-FA. Previous research works on the subject
1PC-FA-15L -192 49 13 have attributed this behavior to the formation of ettringite
2PC-FA -135 64 38
and/or to the reabsorption of the bleed water30, 31. The
reduction of the total shrinkage can be also attributed
D 2 PC-FA-5L -82 78 11
to the larger particle size of the FA used in this study, in
2 PC-FA-15L -157 58 41 comparison to the portland cement32, 33.
3PC-FA -60 84 23 The results of mixture 3PC-FA-5L confirm that L can
C 3PC-FA-5L -53 86 4 also contribute to a slight autogenous expansion after the
final time of setting, which can be caused by the fineness
3PC-FA-15L -123 67 17
of L, as it can absorb and/or adsorb water. However,
shrinkage will appear when this water is subsequently (re)
absorbed by the anhydrous and hydrated cement phases
because of the capillary tensions inside of the paste
pores34, 35. Other authors attribute this expansion to the
formation of calcium hydroxide crystals and/or ettringite
and carboaluminate phases6,36. In this regard, it is worth
noting that the presence of the fine limestone powder
will stabilize the ettringite formed at early ages37. This in
turn will maximize the contribution of ettringite (needles)
to the early-age autogenous expansion and minimize
the contribution to autogenous shrinkage that may
subsequently occur when these needle- like structures
dissolve (removing their expansive restraint) during their
Fig. 4: Autogenous strain of Group A conversion to monosulfoaluminate phases.
The calculated standard deviations of the autogenous
strain measurements presented in Figures 4, 5 and 6
were within the following ranges: 3 μm/m to 41 μm/m for
the reference mixture, 3 μm/m to 31 μm/m for group A
(Figure 4), 1 μm/m to 41 μm/m for group B (Figure 5), and 1
μm/m to 47 μm/m for group C (Figure 6).

Fig. 5: Autogenous strain of Group B

measured up to an age of 28 d are also presented in

Figures 4, 5, and 6. A greater substitution of PC by FA
or FA+L produces a greater reduction in the measured
autogenous strain. Reductions were between 39 % and
86 % in comparison with the reference mixture, which
can be attributed to the reduction of Portland cement (the
most reactive component of the three) in the cementitious Fig. 6: Autogenous strain of Group C

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

608 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Self-Compacting Concretes Using Fly Ash and Fine Limestone Powder: Shrinkage and Surface Electrical Resistivity of Equivalent Mortars

Fig. 7: Drying shrinkage of group A Fig. 9: Drying shrinkage of Group C

Drying Shrinkage 77 %, and 55 % as a result of the substitutions of PC by FA

Results of drying shrinkage up to an age of 28 d are in series A, B, and C, respectively. In comparison with the
presented in Figures 7, 8 and 9 for Groups A, B and C, potential increments estimated between 28 d and 52 w for
respectively. Average results at 28 d and estimations at mixtures R, 1PC-FA, 2PC-FA and 3PC- FA, substitutions of
52 weeks (52 w) are presented in Figure 10; a summary FA by L lead to lower drying shrinkage estimations, with
of the results at 28 d and 52 w is presented in Table 7. values in the range of 55 % to 82 %, 50 % to 69 %, and 32
Results show that drying shrinkage is reduced by 17 % to 56 % for series A, B, and C respectively. For groups A
%, 32 %, and 45 % in groups A, B, and C, respectively and B, figure 10 indicates that drying shrinkage decreases
when the substitutions of PC by FA or FA+L increases. as L increases and evidence an inverse trend for results
This reduction can be attributed to the larger particle of group C.
size of the FA, which is 4 times larger than the portland The standard deviations of the 28 d average drying
cement particles. This physical characteristic could lead shrinkage presented in Figures 7, 8 and 9 were within the
to average pore sizes larger than 50 nm in the mortar following ranges: 16 μm/m to 21 μm/m for the reference
mixtures, originating a reduction in shrinkage, since the
pore solution within pores smaller than 50 nm will induce
significant capillary stresses38, 39.
It is also observed that drying shrinkage slightly increased
when FA was substituted by L. This increase was more
significant for group C, when greater amounts of FA or
FA+L were used. This increase was attributed to the
resulting finer pore structure due to the filler effect of L.
For groups A and B, at 28 d, substitutions of FA by L did not
significantly influence drying shrinkage.
Comparing 28 d drying shrinkage for each single mix
with the estimations at 52 w, results indicate that drying
shrinkage could increase 74 % for the reference and 82 %,
Fig. 10: Average and ultimate drying shrinkage calculated and
estimated at 28 d and 52 w, respectively

mixture, 5 μm/m to 20 μm/m for group A (Figure 7), 5

μm/m to 32 μm/m for group B (Figure 8), and 6 μm/m to
24 μm/m for group C (Figure 8).

Surface Electrical Resistivity

The surface electrical resistivity measured at different
ages (between 1 d and 56 d) is presented in Figures 11,
12, and 13 for specimens corresponding to groups A, B
and C, respectively. Electrical resistivity at early ages
decreases as PC by FA substitution increases. At later
Fig. 8: Drying shrinkage of Group B
ages, the electrical resistivity of the FA and FA and L

Organised by
India Chapter of American Concrete Institute 609
Technical Papers

Moreover, for groups A, B, and C, it can be observed

that the resistivity decreases when FA is substituted by
L. The presence of L increases the concentration of OH-
ions that will act as an electrolyte in the pore solution of
the mortars/concrete and will decrease the electrical
resistivity because these ions are the more conductive
ions within the pore solution45-46.
The standard errors calculated as an estimate of the
uncertainty with regard to the estimation of the surface
electrical resistivity were within the following ranges: 0.12
% to 1.26 % for the reference mixture, 0.19 % to 2.45 % for
group A (Figure 11), 0.16 % to 3.06 % for group B (Figure
Fig. 11: Surface electrical resistivity vs curing time of Group A 12), and 0.11 % to 2.43 % for group C (Figure 13).

CONCLUSIONS
From this study, the following conclusions can be drawn:
1. The results for electrical resistivity indicate a
significant improvement in the cementitious matrix
imperviousness due to the replacement of PC by
FA. As the substitution of PC by FA increases, the
electrical resistivity results at later ages demonstrate
an increasing benefit.
2. The results confirm that the addition of L is an effective
option to counteract the delays in the times of setting
that result from replacing PC by FA and identify
Fig. 12: Surface electrical resistivity vs curing time of Group B the more appropriate combinations of FA+L for this
purpose. As for the compressive strength, the results
show that some of the tested combinations can lead
to a better performance at advanced ages, which
is attributable to the effects of the micro limestone
powder acting as a filler in the cementitious matrix and
as a nucleation center for the hydration products from
the surrounding cement particles.
3. In all cases, the FA substitution by L produced a
decrease in the electrical resistivity; as the substitution
levels increased, the electrical resistivity further
decreased.
4. Substitutions of PC by FA led to significant reductions
in autogenous shrinkage ranging between 39 % and 84
Fig. 13: Surface electrical resistivity vs curing time of Group C % for substitutions between 30 % and 60 %. In relation
to mixtures based on the binary blend (PC+FA),
systems increases significantly above that measured for substitutions of FA by L resulted in either slight
the reference. This later age behavior is attributed chiefly reductions or increases of autogenous shrinkage,
to the densification of the cementitious matrix and to the which were not greater than 84 μm/m and 63 μm/m,
finer pores produced by the pozzolanic reaction of FA40- respectively. These reductions were more evident for
42. Surface electrical resistivity is an indirect measure PC by FA substitutions of 30 %; for substitutions of 45
of both porosity and diffusivity. The electrical current % and 60 %, substitutions higher than 5 % of FA by L
flowing through the hydrated paste is due to an electrolytic resulted in increased autogenous shrinkage.
process mainly resulting from the flow of ions present 5. Substitution of PC by FA reduced 28 d drying shrinkage
in the pore solution (Na+, K+, Ca2+, SO42+, OH-).43 When between 17 % and 45 %, the reduction becoming more
portland cement is replaced by FA, the concentration of significant as the substitution of PC by FA increased.
the alkali ions generally decreases (Na+ and K+), leading For series A, substitutions of FA by L produced less
to a lower ionic conductivity and therefore to a greater significant reductions, 35 % on average. In series B and
electrical resistivity44.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

610 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Self-Compacting Concretes Using Fly Ash and Fine Limestone Powder: Shrinkage and Surface Electrical Resistivity of Equivalent Mortars

C, substitutions of FA by L showed quite similar trends 10. Bentz, D. P., Sato, T. de la Varga, I., Weiss, W. J. “Fine limestone
additions to regulate setting in high volume fly ash mixtures”
to those of series A for monitoring times between 3
Cement & Concrete Composites. No. 34. (2012) Pages: 11-17.
d and 15 d. However, at later ages, in both cases, the
11. ASTM Standard C150/C150M - 12. In Standard specification for
PC+FA mixtures as well as the PC+FA+L mixtures Portland cement ASTM International. Vol. 04.01. (2012).
showed practically the same shrinkages at 28 d. At
12. ASTM Standard C618-05. In Standard specification for coal fly ash
this age, average reductions for series B and C were and raw or calcined natural pozzolan for use in concrete. ASTM
18 % and 31 %, respectively. Regarding projected 52 w International. Vol. 04.02. (2005).
results, substitutions of PC by FA indicate that ultimate 13. ASTM Standard C33-13. In Standard specification for concrete
drying shrinkage could increase between 55 % to 82 aggregates. ASTM International. Vol. 04.02. (2013).
% and that substitutions of FA by L increase drying 14. ASTM Standard C494-09. In Standard specification for chemical
shrinkage between 32 % and 81 %. Estimations at 52 w admixtures for concrete. ASTM International. Vol. 04.02. (2009).
tend to be lower as the substitutions of PC by FA or PC 15. ASTM Standard C1611-14. In Standard test method for slump flow of
by FA+L increases. self-consolidating concrete. ASTM International. Vol. 04.02. (2014).
16. ASTM Standard C1610-14. In Standard test method for static
segregation on self- consolidating concrete using column technique.
Acknowledgements ASTM International. Vol. 04.02. (2014).
The authors wish to express their gratitude to the Consejo 17. ASTM Standard C1621-14. In Standard test method for passing ability
Nacional de Ciencia y Tecnología Mexico (CONACYT) for of self- consolidating concrete by J-Ring. ASTM International. Vol.
providing the scholarship to Juan De-León for his MSc 04.02. (2014).
studies. Also to Dave Carr and Francisco Pazos from OMYA 18. Schwartzentruber, A., Catherine, C.“Method of Concrete Equivalent
and Israel Olivares from Representaciones Técnicas S.A. Mortar–Anoveltool to help in formulation of concrete with
admixtures”. Materials and Structures. Volume 33. Issue 8. (2000).
de C.V., for providing the micro limestone powder used in
Pages: 475-482.
this study. As well as to Filiberto Marín from EUCOMEX
S.A. de C.V., for providing the high range water reducer 19. ASTM Standard C1437-13. In Standard test method for flow of
hydraulic cement mortar. ASTM International. Vol. 04.01. (2013).
and viscosity modifier admixtures, and to Ramón Álvarez
20. ASTM Standard C403-08. In Standard test method for time of setting
from Concretos La Silla, S.A. de C.V., for providing the
of concrete mixtures by penetration resistance. ASTM International.
portland cement. Vol. 04.02. (2008).
21. ASTM Standard C185-08. In Standard test method for air content
References of hydraulic cement mortar. ASTM International. Vol. 04.01. (2008).

1. Okamura, H., Ouchi, M. “Self-compacting concrete – Development, 22. ASTM Standard C109-13. In Standard test method for compressive
present and future”. First International RILEM Symposium on Self- strength of hydraulic cement mortars (using 2-in. or [50-mm] cube
Compacting Concrete. (1999). Pages: 3-14. specimens). ASTM International. Vol. 04.01. (2013).

2. Usyal, M., Yilmaz, K. “Effect of mineral admixtures on properties of 23. ASTM Standard C1698-09. In Standard test method for autogenous
self-compacting concrete”. Cement & Concrete Composites. Volume strain of cement paste and mortar. ASTM International. Vol. 04.02.
33. Issue 7. (2011). Pages: 771-776. (2014).

3. Khayat, K., Mitchell, D. “Self-consolidating concrete for precast, 24. ASTM Standard C596-09. In Standard test method for drying
prestressed concrete bridge elements”. NCHRP REPORT 628. shrinkage of mortar containing hydraulic cement. ASTM
National Cooperative Highway Research Program. (2009). International. Vol. 04.01. (2009).

4. Aslani, F., Maia, L. “Creep and shrinkage of high-strength self- 25. A ASHTOTP95-11.StandardMethodofTestforSurfaceResistivityo
compacting concrete: experimental and analytical analysis”. fConcrete’sAbility to Resist Chloride Ion Penetration. American
Magazine of Concrete Research. Institution of Civil Engineers. Association of State Highway and Transportation Officials,
Volume 65. Issue 17. (2013). Pages: 1044-1058. Washington, DC.

5. Loser, R., Leeman, A. “Shrinkage and restrained shrinkage cracking 26. Morris, W., Moreno, E., Sagues, A. “Practical evaluation of resistivity
of self-compacting concrete compared to conventionally vibrated of concrete in test cylinders using a Wenner array probe”. Cement
concrete”. Materials and Structures. Volume 42. Issue 1. (2008). and Concrete Research. Volume 26. (1996). Pages 1779-1787.
Pages: 71-82. 27. Durán-Herrera, A., Juárez, C.A., Valdez, P., Bentz, D.P. “Evaluation
6. Bentz, D. P., Peltz, M. A. “Reducing thermal and autogenous of sustainable high- volume fly ash concretes”. Cement & Concrete
shrinkage contributions to early-age cracking”. ACI Materials Composites. Volume 33. Issue 1. (2011). Pages: 39-45.
Journal. Volume 105. Issue 4. (2008) Pages: 414-420. 28. Bentz, D.P. “Modeling the influence of limestone filler on cement
7. Zhutovsky, S., Kovler, K. “Effect of internal curing on durability- hydration using CEMHYD3D”. Cement & Concrete Composites.
related properties of high performance concrete”. Cement and Volume 28. Issue 2. (2005). Pages: 124- 129.
Concrete Research. Volume 42. Issue 1. (2012). Pages: 20-26. 29. Yilmaz, B., Olgun, A. “Studies on cement and mortar containing
8. Bentz, D. P., Tanesi, J., Ardani, A. “Ternary blends for controlling low-calcium fly ash, limestone, and dolomitic limestone”. Cement
cost and carbon content”. Concrete International. Volume 35. Issue & Concrete Composites. Volume 30. Issue 3. (2008). Pages: 194-201.
8. (2013) Pages: 51-59. 30. Institute Japan Concrete. “Technical committee report on
9. Gurney, L. Bentz, D. P., Sato, T., Weiss, W. J. “Using limestone to autogenous shrinkage” Tazawa, editor. Proc. Of int. workshop.
reduce set retardation in high volume fly ash mixtures: improving London: E&FN. (1998) Pages: 1-63
constructability for sustainability”. Transportation Research Record. 31. Lura, P. “Autogenous deformation and internal curing of concrete”
Journal of the Transportation Research Board. No. 2290. Concrete PhD Thesis, Delft University of Technology, Delft. (2003).
Materials. (2012) Pages: 139-146.

Organised by
India Chapter of American Concrete Institute 611
Technical Papers

32. Jiang, C., Yang, Y., Wang, Y., Zhou, Y., Ma, C. “Autogenous shrinkage 39. Chindaprasirt, P., Homwuttiwong, S., Sirivivatnanon, V. “Influence of
of high performance concrete containing mineral admixtures fly ash fineness on strength, drying shrinkage and sulfate resistance
under different curing temperatures”. Construction and Building of blended cement mortar”. Cement and Concrete Research. Volume
Materials. Volume: 61. (2014). Pages: 260-269. 34. (2004). Pages: 1087-1092.
33. Bentz, D.P., Jensen, O.M., Hansen, K.K., Olsen, J.F., Stang, H., 40. Polder, R.B., Peelen, W.H.A. “Characterization of chloride transport
Haecker, C-J. “Influence of cement particle size-distribution on early and reinforcement corrosion in concrete under cyclic wetting and
age autogenous strains and stresses in cement- based materials”. drying by electrical resistivity”. Cement & Concrete Composites.
Journal of the American Ceramic Society. Volume 84. No. 1. (2001) Volume 24. Issue 5. (2002). Pages: 427-435.
Pages: 129-135.
41. De la Varga, I., Spragg, R.P., Di Bella, C., Castro, J., Bentz, D.P.,
34. Esping,O.“EffectoflimestonefillerBET(H2O)-areaonthefreshandhard Weiss, J. “Fluid transport in high volume fly ash mixtures with and
enedproperties of self-compacting concrete”. Cement and Concrete without internal curing”. Cement & Concrete Composites. Volume
Research. Volume 38. Issue 7. (2008). Pages 938-944. 45. (2014). Pages: 102-110.
35. Craeye, B., De-Schutter. G., Desmet, B., Vantomme, J., Heirman, G., 42. Raif Boga, A., Bekir Topçu, I. “Influence of fly ash on corrosion
Vandewalle, L., Cizer, O., Aggoun, S., Kadri, E.H. “Effect of mineral resistance and chloride ion permeability of concrete”. Construction
filler type on autogenous shrinkage of self-compacting concrete”. and Building Materials. Volume 31. (2012). Pages: 258-264.
Cement and Concrete Research. Vol. 40. Issue 6. (2010). Pages:
43. Ampadu, K.O., Torii, K., Kawamura, M. “Beneficial effect of fly ash
908-913.
on chloride diffusivity of hardened cement paste”. Cement and
36. Sato, T., Beaudoin, J.J. “The effect of nano-sized CaCO3 addition on Concrete Research. Volume 29. Issue 4. (1999). Pages: 585-590.
the hydration of cement paste containing high volumes of fly ash”
44. Shi,C.“Effect of mixing proportions of concrete on its electrical
Proceeding of the 12th international congress on the chemistry of
conductivity and the rapid permeability test (ASTM C1202 or
cement. (2007) Page: 12.
ASSHTO T277) results”. Cement and Concrete Research. Volume
37. Zajac, M., Rossberg, A., Le Saout, G., Lothenbach, B. “Influence of 34. Issue 3. (2004). Pages: 537-545.
limestone and anhydrite on the hydration of portland cements”.
45. Ghrici, M., Kenai, S., Said Mansour, M. “Mechanical properties and
Cement and Concrete Composites, Volume 46. (2014). Pages: 99-108.
durability of mortar and concrete containing natural pozzolana
38. Wongkeo, W., Thongsanitgarn, P., Chaipanich, A. “Compressive and limestone blended cements”. Cement & Concrete Composites.
strength and drying shrinkage of fly ash-bottom ash-silica fume Volume 29. Issue 7. (2007). Pages 542-549.
multi-blended cement mortars”. Materials and Design. Volume 36.
46. Andrade, C. “Calculation of chloride diffusion in concrete from ionic
(2012). Pages: 655-662.
migration measurements”. Cement and Concrete Research. Volume
23. No. 5. (1998). Pages: 1655- 1667.

Alejandro Durán-Herrera
Affiliation:
Facultad de Ingeniería Civil
Universidad Autónoma de Nuevo León
Av. Universidad S/N
Cd. Universitaria
San Nicolás de Los Garza N.L.
México 66455
Phone: +52-81-14424420 Fax: +52-81-14424443 Email: alejandro.duranhr@uanl.edu.mx
Current Position:
ACI Fellow and Board of Directors (2013-1017) Member, Alejandro Durán-Herrera is Professor and
Head of the Concrete Technology Department at the School of Civil Engineering of the Universidad
Autónoma de Nuevo León in Monterrey, Mexico, as well as member of several ACI - Board, Educational
and Certification Committees and of ASTM and RILEM. Recognized by the Mexican National Council of
Science and Technology as National Researcher Level I and as Full Time Professor with Desirable Profile
by The Mexican Secretariat of Public Education. Research interests: internal curing of high-performance
concrete, self-consolidating concrete, use of by-products in concrete, concrete volume stability, etc.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

612 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

Affordable Prefabricated Modular Houses using cement and polymer

based materials and advanced design tools

J.A.O. Barros, R.M. Lameiras, A.

Abrishambaf, C.M.V. Frazão, V.M.C.F.
Cunha, M.A.D. Azenha, I.M.B. Valente D.M.F. Gonçalves,
ISISE, Department of Civil Engineering, L.A.P. Lourenço
University of Minho, Portugal CiviTest Company, Vila Nova de Famalicão, Portugal

Abstract to flow in the interior of the formwork, filling it in a natural

By taking advantage of the appropriate use of cement and manner and passing through the reinforcing bars (in case
polymer based materials and advanced computational of partial reinforcement) and other obstacles, flowing and
tools, a pre-fabricated affordable house was built in a consolidating under the action of its own weight without
modular system. Modular system refers to the complete the occurrence segregation of its constituents.
structure that is built-up by assembling pre-fabricated Due to its relatively high tensile strength, lightness, good
sandwich panels composed of steel fibre reinforced self- insolation properties and excellent durability in corrosive
compacting concrete (SFRSCC) outer layers that are environments, glass fibre reinforced polymer (GFRP)
connected by innovative glass fibre reinforced polymer materials have also significant advantages on their use
(GFRP) connectors, resulting in a panel with adequate on the prefabrication of high performance construction
structural, acoustic, and thermal insulation properties. elements. By using advanced numerical tools with
The modular house was prepared for a typical family of constitutive models capable of modeling appropriately
six members, but its living area can be easily increased the relevant properties of SFRSCC and GFRP materials,
by assembling other pre-fabricated elements. The speed innovative prefabricated construction systems can be
of construction and the cost of the constructive elements optimized not only in terms of structural performance but
make these houses competitive when compared to also in terms of thermal and acoustic comfort (Lameiras
traditional solutions. et al., 2013).
In this paper the relevant research subjacent to this The LEGOUSE project was dedicated to the development
project (LEGOUSE) is briefly described, as well as the of cost competitive pre-fabricated modular buildings.
construction process of the built real scale prototype. It aimed to take advantage of the appropriate use of
Keywords: SFRSCC; GFRP connectors; sandwich panels; advanced materials and computational tools in the concept
pre-fabrication; thermal and acoustic comfort. phase, design and construction of modular housing
with competitive costs. In this respect the expression
“modular system” refers to the complete structure that
Introduction is built-up by assembling of pre-fabricated elements. On
The use of fibre reinforced concrete (FRC) presents, their term, these elements are pre-fabricated sandwich
actually, a significant increase in the construction structures, with SFRSCC outer layers, that are connected
industry. The main areas of the application of FRC are by lightweight and cost-effective GFRP materials with
the slabs on soil and piles (Mobasher and Destrée, 2010), sufficient mechanical properties as to fulfill the structural
pre-fabrication of structural elements (Barros and di requirements of each structural element, resulting in
Prisco, 2009), structural strengthening (Baghi, 2015) a panel with adequate structural, acoustic and thermal
and tunneling (Lourenço et al., 2011). The partial or total insulation properties. Within this project, a real scale
replacement of conventional reinforcements (steel bars modular housing was built for a typical family of six
and grids) by discrete fibers can have technical and members, but which living area can be easily increased
economic advantages in several applications, by taking by assembling new pre-fabricated elements. The speed
advantage of the enhancement provided by fibers in of construction and the cost of the constructive elements
terms of durability, ductility, and resistance to impact make these houses competitive as compared to traditional
and high temperatures, as well as its easier and fast solutions.
application for the reinforcement of brittle cement based
materials (Barros et al., 2014). The steel fibre reinforced This project has involved the development and the
self-compacting concrete (SFRSCC) is a cement based characterization of physical and mechanical properties of
material that merges the inherent advantages provided the materials that compose the structural elements of this
by steel fibre reinforcement with the self-compacting housing concept, the optimization of structural systems,
nature in its fresh state, resulting a composite that is able the building and testing of the structural elements of the

Organised by
India Chapter of American Concrete Institute 613
Technical Papers

modular system, the full-scale construction and testing foundations assure quick and stable fixing conditions for
of a family modular house, and the development of the the panels, and the load transference to the soil. Smaller
technical specifications, and design rules. This paper number of phase constructions is required, reducing
presents a resume of part of the extensive experimental significantly the time to have a typical house completely
involved in the development of this project, as well as the operational, being a competitive solution considering the
construction technology adopted to build the real scale costs and the global quality.
prototype.
The use of sandwich panels in the construction industry
is quite common. The main advantages associated to
The Legouse Concept the use of this construction element are the thermal
The modular construction system developed in the and structural efficiency. However, for structural uses,
scope of this project is applicable to several types of current sandwich panels require relatively thick external
buildings, namely, single-family houses (Figure 1), multi- layers, since they are reinforced with conventional steel
family houses, commercial and/or industrial buildings. reinforcements that demand at least 30 mm of concrete
The construction system is based on the assembling of cover thickness for corrosion protection. The sandwich
structural prefabricated sandwich panels, both for the panel for the facades of the LEGOUSE is based on the
facades and for the floors and roofing. The panels are use of thin outer layers of SFRSCC and an insulation core
connected to each other in situ with innovative systems layer, all of them of 60 mm thick, but smaller thickness of
in order to assure the aimed structural and comfort the SFRSCC layers would have been possible if a ribbed
performance. This construction system allows embedding SFRSCC layer was adopted (Lameiras et al., 2013b).
all the current infrastructures for housing functionalities
When used for facade sandwich panels, the SFRSCC has
into the sandwich panels, like water supply, wastewater
uncracked and post-cracking mechanical performance
disposal, electricity and communications, without
that does not require being strengthened with any further
compromise the thermal and acoustic performance,
reinforcement, apart the discrete steel fibers. To assure
as well as the finishing appearance of these elements.
proper stress transfer between both outer SFRSCC
The sandwich panels are built under typical conditions
assured by prefabrication industry, and then transported layers, innovative GFRP connectors of reduced thermal
for the local of construction, where prefabricated blocks of bridge effect are used. By eliminating the conventional
reinforcement, the weight and the production time of
the sandwich panels are significantly decreased, while
the durability is increased and the costs maintenance
are decreased, since available research evidences that
surface corrosion can be prevented having conjointly
W/C ≤ 0.5 and a minimum cover of the fibers of 0.2 mm
(Balouch et al., 2010).
The sandwich panel concept is described in Figure 2. The
GFRP connectors can be continuous or discontinuous,
depending on the geometry of the panel and loading
and supporting conditions. The sandwich panel used
for the roofing of the LEGOUSE is similar to the panel
for the facades, but since the span length and actuating
loads are generally higher, the GFRP connectors have
superior mechanical performance, and the bottom
SFRSCC layer is reinforced at its middle surface with
conventional steel bars in the direction of the span length
(fiber reinforcement assures the transversal flexural
demands). The sandwich panels of the roof are supported
on the internal layer of the sandwich facade panels,
and special elements were developed for assuring the
connection between these panels, as well as between
facade panels.
The in-plane and outer-plane load carrying capacity
of these sandwich panels do not require any extra
structural element (columns and beams) for the
Fig. 1: Representative single-family houses based on the LE- structural stability of this type of construction, apart the
GOUSE concept block of foundations.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

614 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

homogeneity and cohesion. Tests to achieve density and air

content were carried out according to EN 12350-6 (2009)
and EN 12350-7 (2009), respectively, and the obtained
values are indicated in Table 2. Due to the relatively low
percentage of fibers in the composition (2.5% by weight),
the density of SFRSCC is similar to the corresponding SCC
(of identical strength class), while the use of high amounts
of fines and the good quality of aggregates in order to
obtain a self-compacting concrete justify the relatively low
air content obtained.

Mechanicalpropertiesathardenedstate
The elasticity modulus and the concrete compressive
Fig. 2: Proposed building system: (a) Components of the devised strength of SFRSCC were assessed according to prEN
load-bearing sandwich wall panel; (b) System cross section 12390-13 (2012) and EN 12390-3 (2011), respectively, at
(Lameiras et al., 2013a) 7, 28 and 90 days, and the results are indicated in Table
3. As expected, the Ecm and the fcm increased with age,
The Relevant Properties of the Constituents more pronouncedly up to 28 days of age. Frazão et al.
(2015) verified that the evolution of the elasticity modulus
of the Sandwich Panel
and the compressive strength after 28 days of age can be
SFRSCC estimated according to Eurocode 2 (EN 1992-1-1, 2004).
However some deviations exist at early ages, due to the
Mixcomposition high volume of ultra-thin material, such as calcareous
The constituent materials used in the composition of filler, included in these compositions. Cunha et al. (2008)
the SFRSCC were: Portland cement CEM 42.5R, water, have adjusted the equations proposed by Eurocode 2 in
superplasticizer Sika® 3005 (SP), limestone filler, order to be capable of simulating the time evolution of the
crushed granite aggregate, fine and coarse sand, and Young’s modulus and compressive strength of SCCs.
hooked-end steel fibres (length, lf, of 33 mm; diameter,
The flexural behavior of SFRSCC was characterized
df, of 0.55 mm; aspect ratio, lf/df, of 60 and a yield stress
of 1100 MPa). The adopted mix proportions are shown in according to the recommendations of RILEM TC 162 TDF
Table 1, where W/C is the water/cement ratio. (2003) and CEB-FIP Model Code (2011). The bending tests
were performed following the proposal of RILEM TC 162
Properties at fresh state TDF (2003) in terms of curing procedures, position and
To evaluate the properties of SFRSCC in the fresh state, dimensions of the notch sawn into the specimen, load and
the inverted Abrams cone slump test and L- Box test were specimen support conditions, characteristics for both the
performed according to EN 12350-8 (2010) and EN 12350- equipment and measuring devices, and test procedures.
10 (2010) recommendations, respectively, and the results From the obtained force-deflection relationship, the limit
are presented in Table 2. The composition has verified of proportionality ( fct,L), the equivalent ( feq,2 and feq,3 )
the self-compacting requirements, no visual sign of and the residual ( fR,1 and fR,4 ) flexural tensile strength
segregation was detected, and the mixture showed good parameters were calculated, and the results are indicated

Table 1
Mix proportions of steel fibre reinforced self-compacting concrete per m3

Coarse Coarse sand

Fine sand
Cement [kg] Water W/C SP Filler [kg] aggregate Fibres (kg)
[kg] [kg]
[kg]

413 127.8 0.31 7.83 353 640 195 713 60

Table 2
Fresh properties of SFRSCC (Frazão et al., 2015)

Slump flow L-Box

Density (g/cm3) Air content (%)
spread (mm) T500 (s) H2/H1 T200 (s) T400 (s)

667 15.6 0.81 5.3 10.1 2.40 0.80

Organised by
India Chapter of American Concrete Institute 615
Technical Papers

Table 3
Relevant results of compression tests (Frazão et al., 2015)

7 days 28 days 90 days

Ecm (MPa) 31.58 36.88 37.80

CoV (%) 7.62 6.71 6.38

fcm (MPa) 50.17 61.90 66.13

CoV (%) 6.92 6.34 9.99

Table 4
Relevant results of flexural tests

RILEM TC 162 TDF CEB-FIP MODEL CODE

fct,L (MPa) feq,2 (MPa) feq,3 (MPa)
fR,1 (MPa) fR,4 (MPa) fR,1 (MPa) fR,4 (MPa)

AVG (7d) 5.09 8.34 8.51 8.25 7.09 8.39 6.56

CoV (%) 5.27 20.28 19.41 19.00 27.29 18.26 30.14

AVG (28d) 6.39 10.12 9.72 9.92 7.63 10.05 7.05

CoV (%) 7.58 19.73 15.25 19.37 13.03 18.72 11.86

AVG (90d) 7.01 4.50 7.94 3.62 4.97 18.78 20.35

CoV (%) 5.75 10.11 12.85 9.82 7.55 9.99 6.90

in Table 4. It is verified that both feq and fR have increased resistivity of concrete in 63%;
up to 28 days, and for 90 days a decrease was registered,
- Determining the diffusion coefficient from the chloride
mainly for the parameters evaluated at larger deflection/
migration test under non-steady state may not be
crack width, which means that due to the relatively high
feasible for a SFRSCC, since the test methodology can
strength of the matrix, some fibers would have failed by
cause significant corrosion of steel fibers and chlorides
tensile rupture.
may tend to settle in steel fibers. The determination of
Durability properties the diffusion coefficient for a SFRSCC is more feasible
by natural immersion test in salt solution. The results
Durability test indicators were also assessed by performing
obtained in both concretes (SCC and SFRSCC) were
durability tests on specimens of SFRSCC at 28 days of
similar;
age. These tests were focused on the determination of
water absorption by immersion (LNEC E394, 1993) and by - In conditions of extreme aggressiveness, corrosion of
capillarity (LNEC E393, 1993); permeability to air (Cabrera, steel fibers can induce cracking in concrete, leading
1999); electrical resistivity (RILEM TC 154-EMC, 2004); to a decrease of tensile strength for the SFRSCC.
chloride diffusion by migration under non-steady state However, it should be noted that this damage was
(LNEC E463, 2004); resistance to chloride penetration by obtained in conditions of extreme aggressiveness,
immersion (LNEC E390, 1993); and carbonation (FprCEN/ which is not expected to occur in real environmental
TS 12390-12, 2010). The description of these tests are conditions;
provided elsewhere (Frazão et al., 2015). Based on the
- Due to the relatively high compactness of SCC mixes,
obtained results, and when compared to the durability
they presented good resistance to carbonation.
indicators of the SCC of similar strength class, the following
conclusions were extracted (Frazão et al., 2015): Fibre distribution effect on the post-cracking tensile
- Adding steel fibers did not change significantly the behaviour of SFRSSC panel’s layer
water absorption by capillarity of SCC, indicating that The dispersion and orientation of fibres in the hardened-
the capillarity pore size was not substantially changed; state results from a series of stages that SFRC passes
from mixing to hardening state, namely (Laranjeira, 2010):
- The air penetrability was not substantially affected by
fresh-state properties after mixing; casting conditions
the presence of steel fibers, although a slight reduction
into the formwork; flowability characteristics; vibration
in SFRSCC was observed;
and wall-effect introduced by the formwork. Among
- The presence of steel fibers has reduced the electrical these factors, wall-effects introduced by the moulds,

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

616 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

and the properties of SFRC in the fresh state, especially

its flowability, are the most important ones (Laranjeira
(2010), Dupont and Vandewalle (2005), Martinie et al.
(2010)). Having in mind that mechanical properties are
significantly related to the fibre orientation and dispersion,
which are affected by concrete’s flow in the fresh state, it
is important to control both those parameters (flowability
and wall-effect) (Ferrara and Meda (2006), Pansuk et al.
(2008), Kim et al. (2008)). For the SFRSCC layers of the
developed sandwich panels the connection of the influence
of the distribution/orientation of fibres to the post-cracking
behaviour of SFRSCC was assessed experimentally by
Abrishambaf et al. (2015). For this purpose SFRSCC
panels were cast from their centre point, and for each
SFRSCC panel cylindrical specimens were extracted and
notched either parallel or perpendicular to the concrete
flow direction, in order to evaluate the influence of fibre
dispersion and orientation on the tensile performance
(Figure 3). In this figure the pale dash lines with arrows
represent the supposed concrete flow directions. The
post-cracking behaviour was assessed by both splitting
tensile tests (Brazilian type) and uniaxial tensile tests.
The hatched cores were used for executing splitting
tensile tests (24), while the rest were used for uniaxial
tensile tests (22). In the splitting tensile test, to localize
the specimen’s fracture, two notches with a 5 mm depth Fig. 3: Core extracting plan from the SFRSCC panel: (a) Panel
were executed on cores’ opposite sides. The influence A; (b) Panel B (Abrishambaf et al., 2013)
of the crack orientation towards the concrete flow was
assessed in two distinct directions. By assuming θ as the
angle between the notched plane and the direction of the
concrete flow, the notch plane is designated parallel for
θ = 0° or perpendicular for θ = 90°. Since the core scheme
was maintained for both panels, for each core location
there are two cores with perpendicular notch direction.
This enables to evaluate the influence of fibre orientation
at a certain distance from the casting position on the
stress-crack width (σ-w) relationship. For instance, θ of
A1 specimen is 90° and 0° in panels A and B, respectively.
The remaining cores extracted from the cast panels were
sawn out from cylinders of 150 mm diameter and 60 mm
thickness according to the schematic representation
shown in Figure 3. Twenty two prismatic specimens with
dimensions of 110×102×60 mm3 were produced for the
uniaxial tensile test program. Following the same notching
procedure for the splitting test specimens, the prismatic
specimens were notched according to parallel (θ = 0°)
and perpendicular (θ = 90°) directions to the expected
concrete flow. The notch was executed in the four lateral
faces of the specimen, at its mid-height, with a thickness
of 2 mm and a depth of 5 mm. The detailed description of
these tests can be found in Abrishambaf et al. (2015).
Figure 4 depicts the average and the envelope nominal
indirect tensile stress versus the crack opening mouth
displacement relationship, θ-w, for specimens extracted
from distinct panels’ locations, while Figure 5 presents
Fig. 4: Nominal indirect tensile stress – crack opening width
the average and envelope tensile stress-crack width relationship, σ–w, obtained from splitting tensile test for: (a) θ
= 0° and (b) θ = 90° (Abrishambaf et al., 2013)

Organised by
India Chapter of American Concrete Institute 617
Technical Papers

velocity that diffuses outwards radially from the centre

of the panel (Figure 6). Hence, the total number of the
effective fibres intersecting the parallel crack plane
(θ = 0°) was higher than the one registered in the orthogonal
crack plane (θ = 90°) (Abrishambaf et al., 2013).

GFRP connectors
Four different types of GFRP were chosen as material
candidates for the adopted connectors (a detailed
description can be found in Lameiras et al., 2013a): C1) a
quasi-isotropic laminate (hereinafter referred to as Quasi
laminate), with 5 mm of thickness, obtained by stacking 6
sheets of a mat consisting of continuous E- glass fibres.
Each sheet of this laminate comprises: 300 g/m2 in the
direction 0° (longitudinal direction of connector), 297 g/
m2 of fibres in the direction 90° (transversal direction of
connector), 303 g/m2 of fibres in the direction +45° and
303 g/m2 of fibres in the direction -45°; C2) a laminate with
2.5 mm of thickness, comprising 6 layers of a chopped
strand mat (CSM) consisting of 500 g/m2 of E-glass
fibres (hereinafter referred to as CSM laminate); C3) a
sandwich composite consisting of 5 mm of polyurethane
foam (PU foam) of density 35 kg/m3 as core material, and
2 outer skins laminates, each one produced by stacking
2 sheets of the same mat adopted in the production of
Fig. 5: Uniaxial tensile stress – crack width relationship, σ-w: the Quasi laminate, providing a sandwich with 1.4 mm
(a) θ = 0° and (b) θ = 90° (Abrishambaf et al. 2013) + 5 mm + 1.4mm = 7.8 mm of thickness (this material is
hereinafter referred to as Quasi sandwich composite); and
C4) a sandwich composite with the same arrangement of
the Quasi sandwich composite, but replacing the quasi-
isotropic mat consisted of continuous fibres by the same
CSM mat adopted in the manufacturing of the CSM
laminate, totalling a laminate with: 2 mm + 5 mm + 2mm
= 9 mm of thickness (in this paper it is referred to as CSM
sandwich composite).
All these GFRPs were manufactured by Vacuum Assisted
Resin Transfer Moulding (VARTM) process (Williams et
al., 1996). All the GFRPs have a polyester resin matrix, due
to its low cost and because it is one of the most common
thermosetting resins used in the reinforced plastics.
For the determination of the mechanical properties of
the composites, direct tensile tests under constant head-
Fig. 6: Explanation for fibre alignment in flowing concrete of a speed rate of 2 mm/min were executed. In the case of
panel casting from the centre (Abrishambaf et al. 2013). the laminate C1, in which the content of fibre is slightly
different in the different directions, tests were carried
(θ-w) curves regarding to each series from direct tensile out loading the laminate in tension along the 0° and
tests. The results obtained in both types of tests show that 90° fibre directions (hereafter called C1-0° and C1-90°,
fibre distribution and orientation have a strong impact respectively). On the other hand, the laminate C3 was
on the tensile behaviour of specimens drilled from the tested only along the 90° direction.
panels. In the case of the series with crack plane parallel
Plates were manufactured using the same process that
to the concrete flow direction (θ = 0°), specimens shown
was chosen for the connectors’ production, and test
significantly higher post-cracking parameters than the
specimens were cut from the flat plate using a diamond
other studied case with a perpendicular crack plane to
saw wheel. The specimens were 25 mm wide, and with
the flow direction (θ = 90°). When a panel is cast from the
the exception of the quasi laminates, the specimens were
centre, fibres have a tendency to line up perpendicularly
250 mm long. In the case of the quasi laminates, the
to the radial flow, mainly due to the uniform flow profile

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

618 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

Table 5
Mechanical properties of different GFRPs under direct tensile tests

Ultimate tensile strength Ultimate tensile strain Tensile Modulus of Elasticity

GERP Number of ( σpt ) [MPa] (∈pu) [μm/m] (Ept ) [GPa]
type specimens
Average St. Dev. Average St. Dev. Average St. Dev.

Quasi laminate 0° 5 363.4 17.2 26185.1 1982.1 14.30 0.32

Quasi laminate 90° 6 350.1 12.8 25912.3 1501.2 13.52 0.27

CSM laminate 8 202.0 8.8 17881.4 969.0 12.65 0.47

Quasi sandwich 4 194.7 5.4 21910.9 2732.1 9.51 0.24

CSM sandwich 4 100.8 2.9 16462.4 1912.5 6.89 0.41

specimens were 280 mm long for the 0° direction and

205 mm long for the 90° direction. The test procedures
described in ASTM D3039 were followed, and tensile
strength, stiffness and stress-strain relationship up
to failure were determined. The tensile stresses of the
laminate specimens were calculated for each data point
simply dividing the registered force by the average cross-
sectional area of specimen. For the sandwich composites
the tensile stresses were computed only considering
the contribution of the laminate skins, disregarding the
sectional area of the foam. The direct tensile test results
are presented in Table 5, and the average stress-strain
curves are depicted in Figure 7.
As shown in Figure 7 the responses of all laminates are
not exactly linear in the loaded directions. This behaviour
was expected due the high content of fibres transversally Fig. 7: Average tensile stress-strain curves from direct tensile
positioned to the load direction. From the results, it should tests with GFRP specimens
be also noted that the laminates comprising longer fibres
(C1 and C3) presented higher ultimate tensile strength The types of GFFP connectors represented in Figure 8
and strain (see Table 5) than the similar composites were investigated for assessing the one that assures the
comprising CSM laminates (C2 and C4, respectively). required stress transfer between both SFRSCC layer,
considering also aspects like application feasibility and
Among the types of GFRP investigated, the CSM laminate costs.
(C2) was the material chosen to produce the connectors
used in the pull-out experimental program due to its For this purposes pull-out tests were carried out using
relative low cost, production simplicity and satisfactory the test setup represented in Figure 9a. Due to the lack
mechanical properties. of space, only the results corresponding to the perforated
plates with 4 circular holes (L4C) are presented (3
Pin-bearing tests were also carried out to characterize specimens, Figure 9b) but the results of the complete
the failure modes associated with GFRP perforated experimental program are available elsewhere (Lameiras
connectors used in concrete sandwich panels (Lameiras et al., 2013a). The designation PERFOFRP was attributed
et al., 2013). The results show that for the same type of to this connector.
specimen and the same thickness, the maximum load
values are similar, regardless the fibres orientation. The failure mechanism observed in the perforated plate
By increasing the thicknesses of the specimen, a connectors was associated to the rupture of the GFRP in
proportional increase in maximum load is observed. In all the vicinity of the holes (Figure 10a). For these connectors,
types of GFRP specimens, it is verified that the maximum the rupture often happened in a sequence of abrupt decays
load values associated with lateral failure are higher than of load, indicating that the failure of the GFRP in the vicinity
the maximum load values registered in specimens with of each hole occurred in different stages of the test (see
failure bellow the hole. Figure 10b). This may be a consequence of different stiffness
amongst the concrete dowels, localized imperfections in

GFRP-SFRSCC Connections
Organised by
India Chapter of American Concrete Institute 619
Technical Papers

Fig. 8: Cross-sectional scheme of investigated connections for sandwich panels: (a) embedded – simply perforated plate; (b) em-
bedded – profiled and (c) adhesively bonded

Fig. 9: Pull-out tests: (a) test setup; (b) and (c) perforated plates (Units in millimeters)

the connector and/or non-uniform distribution of stresses friction between these materials. Moreover, most of the
within the GFRP. However, all the perforated connectors specimens with perforated connectors presented fracture
show a significant residual load capacity after maximum surfaces in the SFRSCC, formed by the tensile forces
load. Inspecting the GFRP/SFRSCC contact surfaces transferred from the GFRP connector to the surrounding
after the tests, it was noticed that these surfaces were SFRSCC (Figure 10c).
scratched, indicating the existence of some adherence/
Push-out tests were also conducted on connections

Fig. 10: (a) Representative rupture of L4C GFRP connector, (b) load-slip response (L4C 01 specimen), and (c) fracture failure modes
in the SFRSCC layer

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

620 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

comprising PERFOFRP connectors and SFRSCC layers

(Lameiras et al., 2012). To evaluate the SFRSCC dowel
effect, tests were also conducted with PERFOFRP
connectors with 3 evenly spaced holes and also with GFRP
plates without holes. Moreover, the end-bearing effect
was isolated comparing the results of specimens with and
without concrete in front of the PERFOFRP rib. Based on
the obtained results, it was verified that the load carrying
capacity of the investigated connections is limited by the
strength of the SFRSCC. It was also observed that the
SFRSCC dowels have provided a contribution of 21% for
the load carrying capacity of the connection.
Fig. 11: Casting the bottom SFRSCC layer

Production of the Sandwich Panels in

Prefabrication Industry Conditions
The construction process is initiated by casting the bottom
SFRSCC layer (Figure 11), of 60 mm thick, followed by the
application of the EPS insulation core material (60 mm
thick) together with continuous (Figure 12) or discontinuous
(Figure 13) GFRP connectors, and finally by casting the top
SFRSCCC layer (Figure 14). After 24 hours of curing, the
sandwich panels could be demoulded, and transported
for the storage area. Figure 15 shows the construction
process of panel “G” of the built prototype (Figure 16) that
includes an opening for a door.

Fig. 12: Application of the EPS insolation core material and

Construction of the Real Scale Legouse continuous GFRP connectors
Prototype
For the construction of the designed prototype (Figure
1, Lameiras et al., 2013b), a set of sandwich panels was
prepared according to the plan described in Figure 16.
These panels were built in the prefabrication factory
of the Mota-Engil Company, in Rio Maior city, Portugal,
according to the procedures described in the previous
section, and then assembled in a selected location of this
factory in order that the LEGOUSE real scale prototype can
constitute an open laboratory for continuous monitoring
purposes in terms of material durability, and thermal-
acoustic performance assessment.
Figures 17 to 20 show a set of photos of the relevant phases
of the construction of this prototype. For the sandwich
panels of the roof (Figure 19), the innovative GFRP and
conventional steel truss connectors were used in order Fig. 13: Application of discontinuous GFRPconnectors
to make a comparison of the technical and cost attributes
of these two types of connectors in real construction and
Preliminary thermal and acoustic tests were carried
environmental conditions. In these sandwich panels the
out in the prototype, and comfort indexes similar, and
top SFRSCC layer was cast in place in order to avoid joints
even higher, to those registered in traditional good
between consecutive panels. The finishing of the external
constructions made by two leafs of clay masonry bricks of
surface of the roof is constituted by an impermeable
11 cm thickness separated by 4/5 cm thick EPS (or mineral
membrane. The construction of the prototype has ended
wool) insulation material (Barros et al., 2015).
with the installation of the infrastructures for water
supply, sewage flow, electricity and communications, and This type of prototype has 100 m2 of area, and according
painting (facultative due to the good finishing quality of the to the company that built it, two weeks are necessary to
external faces of the panels). conclude this process, with a final cost of about 400€/m2
in September of 2013.

Organised by
India Chapter of American Concrete Institute 621
Technical Papers

Fig. 14: Casting the top SFRSCC layer

Fig. 17: Casting the RC block foundations for the sandwich panels

Fig. 15: Casting the “G” panel (Figure 16)

Fig. 16: Plan of assembling of sandwich panels Fig. 18: Installation of the sandwich panels

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

622 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

Conclusions
The present paper describes relevant research carried out
to develop a real scale prototype built by the assemblage
of prefabricated panels formed by 60 mm thick outer
layers of SFRSCC, connected by innovative perforated
GFRP plate connectors, and 60 mm thick EPS insulation
core layer. Extensive experimental programs were
carried out to optimize the SFRSCC at fresh and hardened
state, to characterize the mechanical and durability
performance of SFRSCC, and also to assess the effect
of the fibre orientation and distribution on SFRSCC post-
cracking behaviour. The most technical and cost effective
GFRP connectors were obtained by executing tensile
tests with several types of GFRP composite systems,
as well as pull-out and push-out tests with prototypes
representing the SFRSCC-GFRP real connection. A real
scale prototype of LEGOUSE concept of 100 m2 area,
capable of hosting 6 persons, was built to evaluate the
construction time and costs, as well as for serving as an
open laboratory for assessing the durability, thermal and
acoustic performance. Two weeks are required to build
completely this modular construction, with a final cost per
m2 of 400∈.

Fig. 19: Installation of the roof sandwich panels Acknowledgement

This work is part of the research project QREN number
5387, LEGOUSE, involving the companies Mota- Engil,
CiviTest, the ISISE/University of Minho and PIEP. The
second author would like to thank the FCT for the financial
support through the PhD Grant SFRH/BD/64415/2009, and
the third and fourth authors acknowledge the research
grant provided by LEGOUSE project.

References
1. Abrishambaf, A., Barros, J.A.O., Cunha, V.M.C.F., 2013. Relation
between fibre distribution and post-cracking behaviour in steel fibre
reinforced self-compacting concrete panels. Cement & Concrete
Research, V.51, pp.57-66.
2. ASTM standard D 3039/D 3039M, 2008. Tensile properties of
polymer matrix composite materials. ASTM International, West
Conshohocken, PA.
3. Cabrera, J.G., 1999. Design and production of high performance
concrete. Proceedings of International Conference: Infrastructure
Regeneration and Rehabilitation Improving the Quality of Life
Through Better Construction, Sheffield, pp.1-14.
4. Baghi, H., 2015. Shear strengthening of reinforced concrete beams
with SHCC-FRP panels. PhD Thesis, University of Minho.
5. Balouch, S., Forth, J., Granju, J., 2010. Surface corrosion of steel
fibre reinforced concrete. Cement & Concrete Research, V.17,
pp.410-414.
6. Barros, J.A.O., Lourenço, L., Oliveira, L., Silva, S., Lopes, P., 2015.
LEGOUSE – Modular pre-fabricated house: concept, construction
and tests. IV Conference Ibero-American on the self-compacting
concrete – BAC2015, FEUP.
7. Barros, J.A.O., Lourenço, L.A.P., Soltanzadeh, F., Taheri, M., 2014.
Steel fibre reinforced concrete for elements failing in bending and
in shear. European Journal of Environmental and Civil Engineering,
Fig. 20: Different views of the built prototype V.18(1), pp.33-65.

Organised by
India Chapter of American Concrete Institute 623
Technical Papers

8. Barros, J.A.O., di Prisco, M., 2009. Assessing the possibilities of 24. Lameiras, R.M., Gonçalves, C., Valente, I.M.B., Barros, J.A.O.,
fibre reinforced concrete for underground prefabricated structures. Azenha, M.A.D., 2013. Failure modes and load capacities in filled
Technical report 09-DEC/E-12, Dep. Civil Eng., School Eng., hole GFRP laminates for FRP vs. concrete embedded connections.
University of Minho. FRPRCS11, Guimarães, Portugal.
9. CEB FIP Model Code, 2011. V.1, pp.350. 25. Lameiras, R.M., Santos, T., Barros, J.A.O., Azenha, M.A.D., Valente,
I.M.B., 2012. Development of structural sandwich panels made
10. Cunha, V.M.C.F., Barros, J.A.O., Sena-Cruz, J.M., 2008. Modelling the
by ribbed SFRSCC layers and GFRP connectors. Portuguese
influence of age of steel fibre reinforced self-compacting concrete
Conference on Structural Concrete, BE2012, Paper 143, 10pp. (in
on its compressive behaviour. RILEM Materials and Structures
Portuguese).
Journal, V.41, pp.465-478.
26. Laranjeira, F., 2010. Design-oriented constitutive model for steel
11. Dupont, D., Vandewalle, L., 2005. Distribution of steel fibres in
fiber reinforced concrete. PhD thesis, Universitat Politècnica de
rectangular sections. Cement and Concrete Composites, V.27,
Catalunya.
pp.391-398.
27. LNEC E390, 1993. Concrete. Determination of resistance to chloride
12. EN 12390-3, 2009. Testing hardened concrete – Part 3: compressive
penetration – immersion test. Lisbon, LNEC, p.2.
strength of specimens.
28. LNEC E393, 1993. Concrete. Determination of water absorption by
13. EN 12350-6, 2009. Testing fresh concrete – Part 6: Density.
capillarity. Lisbon, LNEC, p.2.
14. EN 12350-7, 2009. Testing fresh concrete – Part 7: Determination
29. LNEC E394, 1993. Concrete. Determination of water absorption
of air content - Pressuremeter methods.
by immersion – test at atmospheric pressure. Lisbon, LNEC, p.2.
15. EN 12350-8, 2010. Testing fresh concrete – Part 8: Self-compacting
30. LNEC E463, 2004. Determination of diffusion coefficient of chlorides
concrete – Slump-flow test.
by migration under non-steady state. Lisbon, LNEC, p.8.
16. EN 12350-10, 2010. Testing fresh concrete – Part 10: Self-compacting
31. Lourenço, L.A.P., Barros, J.A.O., Alves, J.G.A., 2011. Fiber reinforced
concrete - L-Box test.
concrete of enhanced fire resistance for tunnel segments. ACI SP-
17. EN 1992-1-1, 2004. Eurocode 2: Design of concrete structures - Part 276-4, Durability enhancements in concrete with fiber reinforcement,
1-1: General rules and rules for buildings, European Standard, CEN. Editors: Corina-Maria Aldea and Nur Yazdani.
18. Ferrara, L., Meda, A., 2006. Relationships between fibre distribution, 32. Martinie, L., Rossi, P., Roussel, N., 2010, “Rheology of fiber
workability and the mechanical properties of SFRC applied to reinforced cementitious materials: classification and prediction.
precast roof elements. Materials and Structures, V.39, pp.411-420. Cement & Concrete Research,.V.40, pp.226-234.
19. FprCEN/TS 12390-12, 2010. Testing hardened concrete – Part 12: 33. Mobasher, B., Destrée, X. 2010. Design and Construction Aspects
Determination of the potential carbonation resistance of concrete: of Steel Fiber-Reinforced Concrete Elevated Slabs. In: Aldea C-M,
Accelerated carbonation method. Technical Specification, European Ferrara L, editors. SP-274 Fiber Reinforced Self-Consolidating
Committee for standardization, Brussels. Concrete: Research and Applications CD: ACI, pp. 95-107.
20. Frazão, C.M.V., Camões, A.F.L.L., Barros, J.A.O., Gonçalves, D.M.F., 34. Pansuk, W., Sato, H., Sato, Y., Shionaga, R., 2008. Tensile behaviours
2015. Durability of steel fiber reinforced self-compacting concrete. and fibre orientation of UHPC. Proceedings of Second International
Construction & Building Materials, V.80, pp.155-166. Symposium on Ultra High Performance Concrete, Kassel, Germany
21. Kim, S.W., Kang, S.T., Park, J.J., Ryu, G.S., 2008. Effect of filling (Kassel University Press), pp.161-168.
method on fibre orientation and dispersion and mechanical 35. prEN 12390-13, 2012. Testing hardened concrete – Part
properties of UHPC. Proceedings of Second International 13:Determination of secant modulus of elasticity in compression.
Symposium on Ultra High Performance Concrete, Kassel, Germany Austrian Standards Institute/ Österreichisches Normungsinstitut
(Kassel University Press), pp.185-192. (ON), Vienna.
22. Lameiras, R.M., Barros, J.A.O., Azenha, M.A.D., Valente, I.M.B., 36. RILEM TC 162-TDF, 2003. Test and design methods for steel fibre
2013b. Development of load-bearing insulated panels combining reinforced concrete. σ-e-design method. Final Recommendation.
steel fibre reinforced self-compacting concrete layers and glass Materials and Structures, V.36, pp. 560-567.
fibre reinforced polymer connectors – part II: numerically evaluation
37. RILEM TC 154-EMC, 2004. Electrochemical Techniques for
of mechanical behaviour. Composite Structures Journal, V.105,
Measuring Metallic Corrosion. Materials and Structures, V.37,
pp.460-470.
pp.623-643.
23. Lameiras, R.M., Barros, J.A.O., Valente, I.M.B., Azenha, M.A.D.,
38. Williams, C., Summerscales, J., Grove, S., 1996. Resin infusion under
2013a. Development of load-bearing insulated panels combining
flexible tooling (rift): a review. Composites Part A: Applied Science
steel fibre reinforced self-compacting concrete layers and glass
and Manufacturing, V.27, pp.517-524.
fibre reinforced polymer connectors – part I: conception and pull-out
tests. Composite Structures Journal, V.105, pp.446-459.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

624 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Affordable Prefabricated Modular Houses using cement and polymer based materials and advanced design tools

Joaquim A. O. Barros
Joaquim A. O. Barros is Full Professor of the Department of Civil Engineering of Minho University and
coordinator of the Structural Composites Group. He is a voting member of ACI Technical Committees:
440-Fiber-Reinforced Polymer Reinforcement and 544-Fiber-Reinforced Concrete. He is a member of the
Technical Committees of the International Federation for Structural Concrete (fib): TG 4.1-Fibre Reinforced
Concrete, TG 5.1-FRP Reinforcement for Concrete Structures, TG 4.3-Structural design with flowable
concrete.
He is member of the International union of laboratories and experts in Construction materials, systems
and Structures – RILEM: TC-234-DUC – Design procedures for the use of composites in strengthening of
reinforced concrete structures.
His research interests include structural strengthening, composite materials, fiber reinforced concrete
and the development of constitutive models for the simulation of the behavior of cement based and polymer
based materials, and their implementation in software based on the finite element method. He is author of
more than 600 papers divided by books, chapter of books, journal papers, conferences, monographs and
educational reports.
He is the co-founder of FEMIX FEM-based computer program for advanced structural analysis. He has
participated in 29 research projects (20 as coordinator). Supervisor of 2 Post-Doc (concluded), 34 PhDs
(20 concluded) and 33 MSc (32 concluded). He is the founder of the CiviTest Company and consultant on the
areas of fiber reinforced concrete (FRC) structures, structural rehabilitation and strengthening, and for the
development of new materials for innovative structures, most notably the design of affordable houses in
FRC for South America Countries, the strengthening of residential/commercial buildings with prestressed
carbon fibre laminates, and the structural design of the Centro de Solidariedade de Braga, which was
named for the Secil 2005 prize. He is co-inventor of the national patent No. 107111.

Organised by
India Chapter of American Concrete Institute 625
Technical Papers

A methodology to quantify the self-healing capacity of HPFRCCs

L. Ferrara and V. Krelani
Department of Civil and Environmental Engineering, Politecnico di Milano, Italy

Abstract (Romualdi and Batson, 1963; Romualdi and Mandel,

1964) and the latter respectively (Okamura et al., 1997).
In this paper a methodology is proposed to quantify the self-
It can now be reliably stated that both FRC and SCC are
healing capacity of High Performance Fiber Reinforced
well matured technologies, already boasting significant
Cementitious Composites (HPFRCCs), as a function of
engineering applications and which have even finally
maximum crack opening and exposure conditions. The
harbored into codified design approaches (di Prisco et
topic has been investigated including the effect of different
al., 2009), widely accepted and incorporated into several
flow-induced alignment of fibers, which can result into an
national and international guidelines and design codes (fib
either a strain hardening or softening behavior, whether
Model Code 2010). In the very last decade, the attention
the material is stressed parallel or perpendicularly
of the research community as well as of the construction
to the fibers. Specimens were pre-cracked in 4-point
industry, has focused on the synergy between the two
bending and up to different values of crack openings,
aforementioned FRC and SCC technologies. Interestingly,
and submitted to different exposure conditions, including
the outcome of this synergy is not merely limited to the
water immersion, exposure to humid or dry air, and wet/
additive combination of either benefits, i.e. faster rate of
dry cycles. After scheduled exposure times, specimens
construction and limited human intervention inborn in the
were tested up to failure according to the same test set-up
SCC one, and enhanced toughness and energy absorption
and outcomes of the self-healing, if any, were quantified
capacity, standing as a distinctive feature of the FRC one
in terms of recovery of stiffness, strength and ductility. In
(Grünewald, 2004; Ferrara and Meda, 2006; Ferrara
a durability-based design framework, self-healing indices
et al., 2007). Neither the added value coming from the
to quantify the recovery of mechanical properties were
aforementioned synergy can be “limited” to the possibility
also defined and their significance checked.
of positively addressing and solving the drawbacks, in
Keywords: self-healing, delayed hydration, HPFRCCs, terms of fresh state performance, implied by the addition
deflection hardening, deflection softening. of dispersed wirelike reinforcement into a fluid-like fresh
concrete mix (Bayasi and Soroushian, 1992).
Introduction One of the latest and highly cutting edge developments
The construction industry is challenged by the fast in the field of FRCCs with self compacting properties is
changing and continuously evolving societal needs, and, surely represented by the so-called High Performance
with increasing thoughtfulness, is asking the research Fiber Reinforced Cementitious Composites (HPFRCCs).
community new solutions, in terms of construction Their composition and performance is designed through
materials, building products and technologies. Cutting micromechanics based concepts in such a way that,
edge solutions are required aiming to high-end upon first cracking, the fibers, restraining further
engineering applications able to provide a multifold crack opening, redistribute the stresses to the bulk
tailored performance, even in extremely demanding material and the energy required to pull out the fibers
service conditions could be built faster, and, in cases, at the first crack location is higher than the one needed
likewise adapted fast to change of use or even demolished to form a crack at another location. This process can be
and recycled for new purposes. The concrete materials iterated up to crack spacing saturation and gives rise to
and structures research community worldwide has a stable multi-cracking propagation, which results into
produced, in the last fifty years, two highly significant a strain hardening behavior featuring high strain, energy
achievements which can be encompassed under the absorption and damage tolerance capacity, before the
broad categories of Fiber Reinforced Concrete/Fiber onset of unstable crack localization, accompanied by
Reinforced Cementitious Composites (FRC/FRCCs) and strain softening. The mix composition which makes the
Self Compacting/Consolidating Concrete (SCC). Five aforementioned behavior possible actually features high
and two decades have been respectively elapsed since cement and binder content, low water/cement (or water/
the earliest structured research works on the former binder) ratios and high dosage of superplasticizers, low

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

626 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 A methodology to quantify the self-healing capacity of HPFRCCs

maximum aggregate size for high compactness of the bearing and deformation capacity.
matrix, besides a moderately high quantity of fibers,
The new concept underlying engineering application
surely higher than 1%, or even 2% by volume. As a matter
of this category of advanced cement based materials
of fact such a composition is highly conducive to exhibit
is thus further enriched by a new additional durability
superior performance in the fresh state: this makes related value. On the one hand, in fact, the material is
it possible to align the fibers along the casting flow intrinsically more durable, because of the compactness
direction (Ferrara et al., 2011; 2012a; di Prisco et al., 2013). of its cementitious matrix and of its capacity to spread an
Nevertheless, it is worth remarking that a flow induced otherwise localized damage into several tightly spaced
alignment of the fibers also resulted into a strong material tiny cracks. On the other hand, the material, because of
anisotropy. On the one hand, when the direction of the flow its autogenic self-healing capacity, will also be able to
aligned fibers matches with that of the applied stress, self-promote the healing of the aforementioned damage,
stable multi-cracking behavior is obtained, associated consequently, recovering its pristine level of performance
with deflection and even strain hardening behavior. On and thus automatically extending its service life and the
the other hand when the two aforementioned fiber and service life of structures made of or retrofitted with it.
stress directions are orthogonal to each other, strain and The great sustainability potential of this outcome in the
deflection softening behavior are obtained (Ferrara et al., framework of a holistic design approach is also evident.
2012b). The same material will hence show a significantly
different behavior, whether stressed in either direction. In this study, a thorough investigation of the autogeneous
The resulting material orthotropy hence implies a truly healing capacity of a typical HPFRCC mix (Table 1),
ground-breaking conceptual approach when deciding to containing 100 kg/m3 (1.28% by volume) of short straight
employ HPFRCCs structural applications. In either case, steel fibers has been performed, considering different
in fact, the multi-cracking process which characterizes exposure conditions. As a distinctive feature of this study,
the behavior of the material in the case of favorable fiber the influence has been investigated of the flow induced
orientation reflects the ability of the same material to alignment of the fibers on the material behavior, either
spread into several tiny and tightly cracks, induced by deflection hardening or softening, and on the related self-
healing capacity of the cementitious composite. By means
mechanical stress as well as by any other cause. This
of 4-point bending tests performed on specimens in the
makes HPFRCCs highly suitable for applications where
pre-cracked and post-conditioning stages, the recovery
energy dissipation associated with deformation capacity
of load bearing capacity, ductility and stiffness has
are both required at high levels, such as new earthquake-
been evaluated, in the deflection softening or hardening
resistant structures (Canbolat et al., 2005) or retrofitting
behavior. Suitable healing indicators for the recovery of
of damaged and upgrading of poorly designed ones
the mechanical properties have been defined (Ferrara et
(Martinola et al. 2010; Meda et al. 2014; Muhaxheri et al.
al., 2014) and correlated, in a design oriented durability
2015).
based framework, to an index of crack healing, also herein
The capacity of spreading the localized damage into several defined through a tailored methodology.
tightly spaced but tiny cracks has also a twofold benefit on
durability. On the one hand, in fact, it is well known that Experimental Programme
the driving factor for any degradation mechanism, is the Slabs 30 mm thick, 1m long and 0.5 m wide were casted.
maximum opening of a single crack, which, on the other Fiber reinforced material was poured directly from the
hand, is also a relevant parameter when the capacity of mixer onto a chute pouring the material at one short edge
the material to heal itself has to be evaluated. As a matter of the molds, and allowing it to flow parallel to the long
of fact, the same material composition which makes this sides (Figure 1). From the slabs, once hardened, beam
category of advanced cementitious composites highly specimens 100 mm wide and 500 mm long were cut,
conducive to exhibit a superior performance in the
according to the schematic also shown in Figure 1, to be
fresh state and an enhanced toughness associated with
tested in 4-point bending. The beam specimens were cut
stable multiple cracking and strain/deflection hardening
from the slabs so that their axis, and hence the direction
behavior, also awards the same material the additional
of the principal tensile stresses due to the bending action
benefit of being highly capable of self-heal the cracks
to be applied during the tests, was either parallel or
(Yang et al., 2009; Mihashi and Nishiwaki, 2012). This is
perpendicular to the flow direction of the fresh concrete,
likely to occur because of delayed hydration of cement and
along which the fibers are aligned (Ferrara et al., 2011,
binder particles which remain un-hydrated because of the
2012a-b).
low water/binder ratio and which, upon cracking, can be
exposed to water or atmosphere moisture. The resulting After two or eleven months aging in lab environment,
hydration products, precipitating along the crack faces, specimens were tested in 4-point bending, according
promote the closure of the same crack, in case also to the set-up shown in Figure 2. Tests were performed
associated with the partial or total recovery of the pristine controlling the actuator displacement, at a 5µm/sec rate,
level of mechanical performance, in terms, e.g., of load and measuring the Crack Opening Displacement (COD) at

Organised by
India Chapter of American Concrete Institute 627
Technical Papers

the beam intrados over a gauge length equal to 200 mm. cycles, consisting of one day in water and one day in the
Results of typical tests are shown in Figure 3, in terms 50% RH chamber. Different exposure durations were
of nominal bending stress vs. COD curves: the material scheduled, namely 1, 6 and 12 months for specimens pre-
evidently features a deflection hardening or softening cracked at the age of 2 months, and 1,3 and 6 months for
behavior whether stressed parallel or orthogonal to specimens pre-cracked at 11 months aging. The effects of
the flow induced alignment of the fibers. The deflection age of pre-cracking (two or eleven months) was considered
hardening behavior was the result of a stable multi- only for specimens to be immersed in water, whereas for
cracking process in the constant bending moment region other exposure conditions specimens were pre-cracked
of the specimen (Figure 4a), made possible by the favorable
about two months (minimum 56 days) after casting. In
alignment of the fibers with respect to the applied stress.
this way it is believed that not only cement hydration but
An unfavorable alignment of the fibers resulted in a single
also a significant part of pozzolanic reaction of slag was
crack unstably propagating after the cracking (Figure 4b).
completed, also due to the lower water content in the mix,
It was thus decided to pre-crack specimens featuring a and only exposure to water/atmosphere moisture could
deflection softening response up to a COD value equal to activate the delayed hydration reactions responsible of
0.5 mm; for specimens featuring a deflection hardening self-healing. A synopsis of the experimental program is
response, three different levels of crack opening were given in Table 2.
selected and induced in the specimens: two in the pre-peak
regime, respectively equal to 1 mm and 2 mm, and one in After the scheduled exposure times, specimens were
the post-peak regime equal to (CODpeak + 0.5 mm). CODpeak removed from the conditioning environment, wiped, in
denotes the measured value of COD in correspondence case, and tested up to failure according to the same set-
of the peak stress. It is worth remarking that, because up in Figure 2. “Superposition” between pre-cracking and
of the stable pre-peak multi-cracking process and of the post-conditioning σN-COD curves allowed self-healing
employed test set-up, the measured value of the COD in capacity to be evaluated through recovery of mechanical
the pre-peak regime actually represents the sum of the performance, as a function of the different investigated
opening of all the cracks; on the other hand, the value testing variables, and as it will be described and analyzed
of pre-cracking COD in the post-peak regime has been in detail in forthcoming sections.
selected on the basis of an equivalent opening of the
unstable localized crack, in analogy to the deflection Table 1.
softening/single cracking case, according to Ferrara et al. Mix design of the investigated HPFRCC
(2011, 2012b). Mix constituent Dosage
(kg/m3)
After pre-cracking specimens were submitted to different Cement type I 52.5 N 600
exposure conditions, including: immersion in water, Slag 500
exposure to open air in the lab courtyard (temperature and
Sand 0-2 mm 982
humidity were daily monitored); exposure in a chamber at
Water 200
constant temperature T = 20°C and relative humidity RH
Superplasticizer 33
= 95%, exposure in a chamber at constant temperature T
= 20°C and relative humidity RH = 50% and wet and dry Straight eel fibers (13 mm long – diameter 0.16 mm) 100

Table 2.
Synopsis of experimental programme (number of tested specimens per each testing variable) (in italics tests to be still performed)

Deflection softening Deflection hardening

Pre-crack opening 0.5 mm 1 mm 2 mm CODpeak + 0.5 mm

Exposure condition Exposure duration

1 6 12 1 6 12 1 6 12 1 6 12
Water immersion (2 months pre-cracking) 1 2 2 1 1 1 2 2 2 2 2 2
Water immersion (11 months pre-cracking) 3 2 (3m) 3 (6m) 1 1 = 2 2 2 2 2 2
Air exposure (2 months pre-cracking) 1 2 2 1 1 1 2 2 2 2 2 2
Wet environment 90% RH (2 months pre-cracking) 1 2 2 1 1 1 2 2 2 2 2 2
Dry environment 50% RH (2 months pre-cracking) 1 2 2 1 1 1 2 2 2 2 2 2
Wet/dry cycles (2 months pre-cracking) 1 2 2 1 1 1 2 2 2 2 2 2

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

628 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 A methodology to quantify the self-healing capacity of HPFRCCs

Fig. 1: Slab casting scheme and specimen cutting procedure

Fig. 2: 4-point bending tests set-up for beam specimens obtained as in Figure 1.

Fig. 3: Nominal stress σN vs. COD curves for specimens featuring: (a) stable pre-peak multi-cracking and deflection hardening
behavior (fibers parallel to the bending axis); (b) unstable post-cracking localization and deflection softening behavior (fibers or-
thogonal to the bending axis).

(a) (b)
Fig. 4: Crack patterns in specimens featuring: (a) stable pre-peak multi-cracking and deflection hardening behavior (Figure 3a); (b)
unstable post-cracking localization and deflection softening behavior (Figure 3b).

Organised by
India Chapter of American Concrete Institute 629
Technical Papers

Experimental Results - quite poor, and again as coherently expectable,

intermediate between the one of specimens in dry
The results of the pre-cracking and post-conditioning
environment, was the performance exhibited by the
4-point bending tests have been plotted in terms of
specimens subjected to the wet and dry cycles.
nominal stress σN-COD curves, for the different pre/
crack openings and exposure conditions investigated In the case of deflection hardening specimens pre-
and distinguishing between deflection softening and cracked in the pre-peak regime, i.e. at a COD value
deflection hardening specimens. The experimental equal to either 1 mm or 2 mm, since the pre-cracking
results processed as above will be first of all qualitatively threshold was set before the specimen could attain its
analyzed to capture, if any, the trends of the healing. In a peak and enter into the stage of unstable propagation of
further step, by comparing for each and all the specimens, the localized crack (softening), a deflection hardening
the curves in the pre-cracking and in the post-conditioning (wrongly interpretable as strength gain) would have
regime for the same specimen, suitable indices will be anyway occurred even in instantaneous tests. The
defined to quantify the effects of healing on the recovery of strength gain measured after the conditioning has to
the load bearing capacity, ductility, stiffness and damage be suitably cleansed of the aforementioned deflection
accumulation. A methodology will be finally proposed hardening capacity that specimens do inherently
to estimate the amount of crack closure, to which the possess. In this framework, with reference to the notation
values of the aforementioned healing indices will also be explained in Figure 6a, the Index of Stress Recovery for
correlated. deflection hardening specimens pre-cracked in the pre-
peak regime is defined as follows:
Index of stress recovery
(f peak, post - conditioning - v N unloading pre - crack) v N, unloading virgin ...
In the case of deflection softening specimens the ISR =
(f fpeak, virgin - v N, unloading virgin) v N, unloading pre - crack - 1 2
effectiveness of healing in promoting the recovery of the
“through crack” residual stress bearing capacity can be
evaluated by calculating the amount of strength gained where f fpeak, virgin - v N, unloading virgin v N, unloading pre - crack
v N, unloading virgin
after the conditioning period, with respect to the residual
strength featured at the maximum pre-crack opening, represents the amount of load bearing capacity that the
and comparing it to the stress loss exhibited by the same specimen, due to its deflection hardening behavior, would
specimen when pre-cracked up to the aforementioned have anyway gained if tested monotonically.
crack opening threshold (Figure 5a). With reference to the For deflection hardening specimens pre-cracked up to
notation in the same Figure, the Index of Stress Recovery 0.5 mm after the peak, the ISR is calculated as in Equation
(ISR) is defined as follows: (1), since, similarly to deflection softening specimens,
f peak after conditioning - v N unloading ........................................1 the pre-cracking brought already the specimen into
ISR = f ft.pre - cracking - v N unloading the stage of the unstable localized crack propagation. It
has to be anyway remarked that whereas in the case of
From the computed values of ISR, as plotted in Figure 5b,
deflection softening specimens, in which only one crack
the following statements hold:
formed, the ISR does really represent what is due to the
- specimens immersed in water together with those healing of that same single crack, in the case of deflection
exposed to natural environment (in a quite humid hardening specimens pre-cracked after the peak, the
climate, like the Northern Italy one), featured, in value of the ISR computed as above will incorporate the
average, the highest, and a quite similar, recovery effects of healing both the single localized cracks and
trend, gaining, in the post-conditioning stage a strength also all the other micro-cracks which have been formed
even higher that the virgin cracking strength; up to the peak.
- significantly, specimens pre-cracked at 11 months The trends of ISR are shown in Figures 6b to 6d
and immersed in water, even continued to show some (respectively for deflection hardening specimens pre-
moderate healing capacity, at a moderately increasing cracked up to 1 mm, 2 mm and 0.5 mm after the peak).
trend with prolonged immersion; With some exceptions, attributable only to random
experimental scattering, the trends and influence of
- specimens exposed to a humid environment featuring
exposure conditions, as discussed in detail for deflection
an appreciable healing rate since from earlier
softening specimens, are likely to be confirmed. It can
immersion times, but which did not show any significant
be furthermore observed that the on-going healing of
improvement with prolonged exposure time;
cracks was also instrumental to shadow the damage that,
- as expectable specimens exposed to a dry environment in some cases (see e.g. specimens pre-cracked up to 0.5
featured an almost negligible healing, even if somewhat mm after the peak) was caused by not favorable exposure
increasing with prolonged exposure time; conditions (e.g. dry environment).

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

630 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 A methodology to quantify the self-healing capacity of HPFRCCs

(a) (b)
Fig. 5: Notation and significance of Index of Stress Recovery (ISR) for deflection softening specimens (a) and Index of Stress Recovery
vs. conditioning time for deflection softening specimens (hollow markers refer to values of single tests, solid markers represent
average values of nominally identical tests).

(a) (b)

(c) (d)

Fig. 6: Notation and significance of Index of Stress Recovery (ISR) for deflection hardening specimens pre-cracked in the pre-peak
regime (a); Index of Stress Recovery for deflection hardening specimens pre-cracked up to 1 mm COD (b), 2 mm COD (c) and CODpeak
+ 0.5 mm (d) (hollow markers refer to values of single tests, solid markers represent average values of nominally identical tests).

Index of stiffness/damage recovery Index of Damage Recovery

K reloading postconditioning - K unloading pre - cracking ..........................
Thanks to unloading-reloading cycles performed IDaR = K loading pre - cracking (3)
during both pre-cracking and post-conditioning
tests, the values of secant unloading and tangent whose plots (Figures 7a to 7d) appear to be coherent with
reloading stiffness, respectively denoted as Kunl,j and the previously discussed trends of ISR:
K loading,j at different levels “j” of crack opening were - deflection softening specimens exhibited moderate
evaluated and an Index of Damage Recovery was stiffness recovery, with some improvement over time,
calculated as: except for exposure to dry environment or air (see
above);

Organised by
India Chapter of American Concrete Institute 631
Technical Papers

(a) (b)

(c) (d)

Fig. 7: Index of Damage Recovery for deflection softening specimens (a) and for deflection hardening specimens pre-cracked up
to 1 mm COD (b), 2 mm COD (c) and CODpeak + 0.5 mm (d) (hollow markers refer to values of single tests, solid markers represent
average values of nominally identical tests).

- for deflection hardening specimens, except for

specimens pre-cracked up to 1 mm in the pre-peak
regime under prolonged immersion in water, a
moderate improvement in the recovery was calculated;
- for specimens pre-cracked up to 2 mm up to 0.5 mm
after the peak, the recovery in the performance holds
almost constant with conditioning time, with some
expectable worsening in the case of exposure to dry
environment. Availability of water generally yields to
better results.
When compared to the Index of Stress Recovery (Figure 8),
the Index of Damage Recovery appears to increase with it
as well, even if at a milder pace, the un-cracked specimen
stiffness being never completely got back. Moreover,
despite the unavoidable scattering, deflection softening
Fig. 8: Correlation between IDaR and ISR.
and hardening specimens are likely to feature a common
trend, which also stands as a proof of the significance of
the defined indices. of the damage growth. From the comparison between
the fitted trend in the pre-cracking and in the post-
Index of Crack Healing
conditioning stages the crack closure due to self healing
From the evolution of stiffness all along the load path could be estimated, as graphically explained Figure 9 and
values of the scalar internal damage variable were also an Index of Crack Healing was defined as:
calculated and damage evolution curves were built, COD post - conditioning ............
for different exposure conditions, the effects of healing Index of Crack Healing ICH = 1 - COD 4
pre - cracking
resulting, as shown in Figure 9, in a general “slowering”

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

632 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 A methodology to quantify the self-healing capacity of HPFRCCs

Results (Figure 10a to 10d) provide a further confirmation Comparison between Indices of Mechanical Properties
to the statements exposed above with reference to Recovery and Index of Crack Healing
other indices. The coherence of the whole set of results The different indices of recovery of mechanical properties
garnered so far also stands as a proof of the reliability of have been finally compared with the Index of Crack
the proposed methodology for the evaluation of effects of Healing (Figures 11a-b). Together with visual evidence
self-healing of HPFRCCs. of crack healing (Figures 12 to 14) a high sparsity of the
results is evident, which is likely to be due to the different
significance of each single Index for deflection softening
or hardening specimens, either pre-cracked in the pre-
peak or post-peak regime. Anyway a common trend can
be inferred which is likely to confirm that a remarkable
crack healing (e.g. > 0.6-0.8) is necessary before any
significant recovery of any mechanical property can be
observed. The influence not only of exposure conditions,
as discussed above, but also of the pre-crack opening
can also be estimated from the aforementioned graphs:
deflection softening specimens always perform the
poorest, because of the large opening of the single crack,
followed by deflection hardening pre-cracked in the pre-
peak regime up to 1 mm. On the other hand, it can be got
that the performance of deflection hardening specimens
pre-cracked in the pre-peak regime and up to 2 mm
Fig. 9: Example of damage evolution curves with ICH estimation crack opening always falls in the highest range. It is worth

(a) (b)

(c) (d)

Fig. 10: Index of Damage Recovery for deflection softening specimens (a) and for deflection hardening specimens pre-cracked up
to 1 mm COD (b), 2 mm COD (c) and CODpeak + 0.5 mm (d) (hollow markers refer to values of single tests, solid markers represent
average values of nominally identical tests).

Organised by
India Chapter of American Concrete Institute 633
Technical Papers

once again reminding that for an integral crack opening Conclusions

equal to 2 mm, the opening of each single crack, because
of the obtained multi-cracking pattern and considering High Performance Fiber Reinforced Cementitious
the employed pre-cracking set-up, is roughly equal to Composites (HPFRCCs) do inherently possess an
150 µm. This hence may result in the highest observed autogeneous self-healing capacity due to the positive
crack closure and recovery performance since for the synergy between the crack closure effect provided by the
aforementioned maximum crack openings a synergistic fibers and the peculiar material composition, featuring
compromise between the larger cluster of un-hydrated
high cement/binder content and low water binder ratio.
particles exposed and the sealability and healability of the
crack can be effectively got. This clearly adds new value Cracking expose clusters of binder particles remained
and significance to the already well known high damage un-hydrated because of the mix composition as above
tolerance that HPFRCCs do inherently possess. to outdoor moisture or water and this makes possible

(a) (b)
Fig. 11: Correlation between Index of Stress and Damage Recovery ISR - IDaR and Index of Crack Healing ICH.

Fig. 12: Examples of healed/healing cracks in deflection softening specimens (pre-crack opening 0.5 mm) after one/six months
exposure to different conditioning environments

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

634 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 A methodology to quantify the self-healing capacity of HPFRCCs

Fig. 13: Healed/healing cracks in deflection s deflection hardening specimens pre-cracked at 2 mm (single crack opening about
0.15 mm) after one/six months exposure to different conditioning environments

Fig. 14: Healed/healing cracks for deflection hardening specimens pre-cracked at CODpeak+0.5 mm (localized crack opening about
0.7 mm) after one/six month exposure to different conditioning environments.

Organised by
India Chapter of American Concrete Institute 635
Technical Papers

delayed hydration reactions whose products seal and heal 369-374.

the cracks. 2. Canbolat B.A., Parra-Montesinos G.J., Wight J.K., 2005.
Experimental study on seismic behavior of high performance
In this paper a dedicated “three-stage” experimental fiber-reinforced cement composite coupling beams. ACI Structural
methodology has been employed to investigate the Journal, 102 (1): 159–166.
effects of crack healing on the recovery of the mechanical 3. di Prisco, M., Plizzari, G., Vandewalle, L., 2009. Fiber reinforced
performance, in terms of load bearing capacity and concrete. New design perspectives. Materials and Structures, 42
(9): 1261-1281.
stiffness for both deflection softening and hardening
behavior, as obtainable because of the flow induced 4. di Prisco, M., Ferrara, L., Lamperti, M.G.L. (2013). Double Edge
Wedge Splitting (DEWS): an indirect tension test to identify post-
alignment of the fibers in a tailored casting process. In the cracking behaviour of fibre reinforced cementitious composites.
case of deflection hardening behavior, self-healing has Materials and Structures, 46 (11): 1893-1918.
been assessed with reference to different crack opening 5. Ferrara, L., Meda, A., 2006. Relationships between fibre distribution,
stages, both in the pre- and in the post-peak regime. workability and the mechanical properties of SFRC applied to
Effects of different exposure conditions and age of pre- precast roof elements. Materials and Structures, 39 (4): 411-420
cracking have been investigated as well. A procedure 6. Ferrara, L., Park, Y.D., Shah, S.P., 2007, A method for mix-design
has been also calibrated to estimate the amount of crack of fiber reinforced self-compacting concrete. Cement and Concrete
Research, 37 (6): 957-971.
closure, based on damage comparison between suitably
built damage evolution curves in the pre-cracking and 7. Ferrara, L., Ozyurt, N., di Prisco, M., 2011, High mechanical
performance of fiber reinforced cementitious composites: the role of
post-conditioning regimes. “casting-flow” induced fiber orientation. Materials and Structures,
As from the obtained and analyzed experimental results, 44 (1): 109-128.
thanks to the aforementioned autogeneous mechanisms, 8. Ferrara, L., Faifer, M. and Toscani, S., 2012a, A magnetic method
HPFRCCs are able to seal even wider cracks (up to 0.5 mm for non destructive monitoring of fiber dispersion and orientation in
Steel Fiber Reinforced Cementitious Composites – part 1: method
and sometimes even more) and recovery a load bearing calibration. Materials and Structures, 45 (4): 575-589.
capacity as high as or even higher than the one featured by
9. Ferrara, L., Faifer, M., Muhaxheri, M., Toscani, S., 2012b, A magnetic
the virgin material in its pre-cracking stage. This in case method for non destructive monitoring of fiber dispersion and
of exposure conditions favorable to healing, characterized orientation in Steel Fiber Reinforced Cementitious Composites –
by the presence of water, even discontinuous, such part 2: correlation to tensile fracture toughness. Materials and
Structures, 45 (4): 591-598.
as during wet and dry cycles. On the other hand the
capacity of the material to “capture” the moisture of 10. fib Model Code 2010: (2012): 2 voll. 1st volume: 318 pages – ISBN
978-2-88394-095-6; 2nd volume: 312 pages – ISBN 978-2-88394-
the atmosphere, as in the case of humid environment, 096-3.
appears quite scattered and resulted in a not reliable
11. Grünewald, S., 2004, Performance based design of self-compacting
healing capacity. Interestingly, in the case of immersion in steel fiber reinforced concrete. PhD Thesis, Delft University of
water, even quite old HPFRCCs (pre-cracked about 1 year Technology.
after casting) showed a not negligible capacity to seal the 12. Martinola, G., Meda. A., Plizzari, G.A., Rinaldi, Z., 2010, Strengthening
cracks and partially regain the load bearing capacity lost and repair of RC beams with fiber reinforced concrete. Cement and
upon pre-cracking. This is surely due to the peculiar mix Concrete Composites, 32: 731-739.
composition, as above, but is most likely enhanced by the 13. Meda, A., Mostosi, S., Riva, P., 2014, Shear strengtheneing of
latent activity of the slag, employed in large amounts in reinforced concrete beam with HPFRCC jacketing. ACI Structural
Journal, 111 (5), pp. 1059-1068.
the HPFRCC herein investigated.
14. Mihashi, H., Nishiwaki, T., 2010, Development of engineered self-
The recovery of the load bearing capacity and of the healing and self-repairing concrete. State-of-art report. Journal of
stiffness was consistently correlated to the crack closure, Advanced Concrete Technology, 10: 170-184.
both visually observed and quantitatively estimate through 15. Muhaxheri, M., Spini, A., Ferrara, L., di Prisco, M., Lamperti, M.G.L.,
the aforementioned tailored procedure. 2015, Strengthening/retrofitting of coupling beams using advanced
cement based materials. Accepted for presentation to ICCRRR2015,
The whole set of experimental results herein presented is Leipzig, Germany, 6-7 October 2015.
going to be completed with long term (more than one year) 16. Okamura, H., 1997. Self-compacting high performance concrete.
exposure tests and with further analysis on the recovery Concrete International, 19 (7): 50-54.
of mechanical properties, including the ductility in the 17. Romualdi, J.P., Batson, G.B., 1963. Behavior of Reinforced Concrete
case of deflection hardening specimens. A comprehensive Beams with Closely Spaced Reinforcement. ACI Journal, 60 (6):
database will be provided on the self-healing capacity 775-790.
of High Performance Fiber Reinforced Cementitious 18. Romualdi, J.P., Mandel, J.A., 1964. Tensile Strength of concrete
Composites, adding remarkable value to the whole life- Affected by Uniformly Distributed and Closely Spaced Short Lengths
of wire Reinforcement. ACI Journal, 61 (6): 657-672.
cycle performance of structures made of or retrofitted
19. Yang, Y., Lepech, M.D., Yang, E.H., Li, V.C., 2009. Autogenous healing
with this category of advanced cement based composites.
of ECCs under wet-dry cycles. Cement and Concrete Research, 39:
References 382-390.
1. Bayasi, M.Z., Soroushian, P., 1992. Effect of Steel fiber reinforcement
on fresh mix properties of concrete. ACI Materials Journal, 89 (4):

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

636 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Health & Life Cycle Monitoring of Infrastructure Asset Management- An Emerging Industry and the Role of Long Term Planning

Health & Life Cycle Monitoring of Infrastructure Asset Management-

An Emerging Industry and the Role of Long Term Planning
Shah, Janvi
(MSc Distinction, PhD Researcher) Department of Civil Engineering, University of Birmingham, UK

Abstract Introduction
Asset Management is a fast evolving industry and has India with one of the largest and well-spread INFRA
grown considerably over the past decade. Transportation STRUCTURE ASSETS is further on the verge of expanding
assets are currently facing challenges in the form of its Infrastructure network. Need of the hour is Scientific,
ageing assets, deterioration, underinvestment in its systematic Health Monitoring of not only existing assets
maintenance, exposure to extreme climatic conditions, but also the upcoming ones right at the planning stage is
need for better service within constrained budgets and utmost essential Ground reality is different and needs are
increased accountability. There is a lot of literature and “many faceted”. The experience of Developed Countries
guidance documentation available in the public domain, and their Health Monitoring tools will be definitely useful
which focuses on showcasing the best practices in asset to India and this paper highlights the same .The Institute
management across the world. The asset management of Asset Management (IAM) defines the concept of asset
systems developed and being used currently, mainly management as “management of (primarily) physical
focus on data management systems, deterioration assets (their selection, maintenance, inspection and
modelling, integrated information management systems renewal) plays a key role in determining the operational
and developing decision support frameworks using risk performance and profitability of industries that operate
management and cost analysis. The tools are being assets as part of their core business”. Asset Management
used extensively to develop asset maintenance regimes is the art and science of making the right decisions and
based on its performance and risk within budget and optimising these processes. A common objective is to
time. While this is of paramount importance, knowing minimise the whole life cost of assets but there may be
& predicting how the assets will perform throughout its other critical factors such as risk or business continuity
design & service life, under the influence of changing to be considered objectively in this decision-making.
future conditions, is gaining significant importance over (Institute of Asset Management, https://theiam.org/what-
the recent years due to series of events suggesting a asset-management, accessed November 2012). For the
need for change. purpose of this paper asset management is defined as
‘a process of managing and maintaining assets through
This paper provides background of this industry especially
improved utilisation of resources in order to provide a
in the UK transportation sector. The paper highlights the
better level of service to the customer’.
challenges and benefits of adopting asset management
systems and practices in monitoring maintaining and Infrastructure asset management can operate over a
managing physical infrastructure assets in an efficient range of different levels, within both national and local
and systematic fashion. The paper also discusses the networks. For most infrastructure authorities, it is a key
need to adopt a long-term approach in planning and area of development; however the methodologies differ
decision-making & reviewing to make the most of the vastly, from sophisticated integrated data warehouses,
physical assets on the transportation network for the with incorporated condition modelling and decision
future. The Paper is thus an endeavour to discuss in support tools; to basic spreadsheets containing local
this conference especially in developing country like maintenance and renewal programmes. In both cases, the
INDIA with very large outlay of Infrastructure assets, its chosen method should appropriately support the level at
monitoring & management of infrastructure assets and which the authority is working and the size of the network.
development of long term planning potential. The reason for this is that Asset Management is focused on
organisational strategy and policy. With a strong, defined
Keywords: health monitoring, predicting life cycle
asset management strategy and supporting policies, any
costs, infrastructure asset manegement, international
organisation can deliver an asset management approach
standards, infrastructure, resilient solutions, service life,
to long-term maintenance. (PAS 55 2008, ICE 2001 and
ground reality.
OECD 2001).

Organised by
India Chapter of American Concrete Institute 637
Technical Papers

based within the United Kingdom (e.g. ICE, CSS, Network

Rail, Highways Agency, Department for Transport, etc.);
however some are internationally derived, or developed
from practices undertaken worldwide by international
working groups comprising of professionals and experts
from various countries. These organisations typically
own the assets and have responsibility for operating and
maintaining them on the infrastructure network. Such
documentation provideshelpful insights and guidance
on adopting best practices in asset management. The
multiple objectives of these guidance documents are to:
1. develop a common understanding of asset
management;
2. identify the key components of an asset management
system;
3. to define the strategies and policies for asset
management and
4. to document the state of implementation of asset
management programs in various member countries.
Some of the widely referred guidance documents are
mentioned, for example
ll ISO 55000 series of international standard: comprises
of three standards ISO55000 which provides an
Fig. 1: Generic Asset Management System overview of the subject, ISO 55001 highlights the
requirements for specifications for adopting an
effective asset management system and ISO 55002
Benefits of an Asset Management System provides guidance on implementation of asset
(Theory) management. The series is further development for
An asset management system, when tailored to the needs the existing PAS 55 standard.
of the industry and adopted in a proactive manner,may
ll PAS 55 (2008). Published by the Institute for
provide multiple benefits to an infrastructure asset owner,
Asset Management (IAM) and British Standards
includingthe provision of a ‘better-informed’ decision tool,
Institute (BSI). This is the de-facto specification
improved results and outcome from the modelling process
for asset management across all sectors, hinging
and an organised strategy for delivery. County Surveyors
asset management around a 28-point checklist of
Society ‘Framework for Highway Asset Management’
requirements that demonstrate good practice. Due to
(CSS, 2004) highlights areas where a well- defined asset
the fact that this is not infrastructure specific it might
management system provides specific benefits including:
be perceived as overly generic in nature.
ll Reduced Life Cycle Costs
ll Asset Management for the Roads Sector (2001).
ll Defined level of Service Published by the OECD. This document, now a decade
old, aimed to address some of the key issues in
ll Ability to track performance
undertaking an asset management approach within the
ll Improved transparency in decision making infrastructure sector. It delivers comprehension around
the key focuses of infrastructure asset management;
ll Predicting consequences of funding decision
however, it is limited by its international remit. Whilst
ll Decreased financial, operational and legal risk OECD is a multi-national economic forum that has
ll Ability to discharge to financial reporting provided substantial Asset Management guidance via its
responsibilities and statutory valuation specifically assigned Asset Management Working Group
(OECD, 2001) the working group strives to develop a
common understanding of the goals, scope and definition
Asset Management – Key Guidance Documentation UK
of asset management strategies for implementation
and International
across the world. The international members include
There are a number of key pieces of documentation that Australia, Belgium, Canada, Denmark, France, Sweden,
provide Infrastructure Asset Management guidance. Switzerland, Turkey, UK and USA.
These are largely produced by specific organisations

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

638 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Health & Life Cycle Monitoring of Infrastructure Asset Management- An Emerging Industry and the Role of Long Term Planning

ll International Infrastructure Management Manual on common principles on which an effective and uniform
(2011): Published by New Zealand Asset Management asset management system can be developed. Therefore, a
Support (NAMS). A consortium of companies and standardised asset management system, applicable for an
bodies delivering asset management developed this infrastructure network, should follow the stages outlined
manual. Infrastructure asset management has been in Figure 1, in order to provide a consistent framework
practiced in New Zealand since 1995, and encompasses for infrastructure asset management. The process
a wide range of publically owned physical assets. begins with a clear idea of the goals and objectives of the
organisation and it should be coherent with the policies
ll Framework for Highway Asset Management (2004);
and strategies laid out by the organisation for effectively
Published by the County Surveyors Society (CSS). A
managing the assets; It then moves to identifying the data,
general purpose guide to highway asset management,
condition of the assets and by undertaking a suitable gap
largely aimed at Local Authority providers, this
analysis a comparison can be made between the existing
document has a significant emphasis on Highway
performance of the asset with that expected. This process
Asset Management.
involves undertaking adequate financial and resource
ll Manual of Highway Design and Management (2011); analysis with respect to asset maintenance in order to
Published by the Institute of Civil Engineers (ICE). A output optimised design solutions which take into account
further general purpose guide, updated to incorporate the whole-life cycle of the asset. The process enables
current thinking on asset management provision. breaking bigger tasks into smaller work packages thereby
Useful for both local and national highway authorities. ensuring a clear programme of works is prepared,
monitored and well reported for progress to occur.
ll C667 Whole-life Infrastructure Asset Management,
A good practice guide for civil infrastructure (2009). Organisational of Economic Co-operation and Development
Published by Construction Industry Research and (OECD) (2004) along with other guidance bodies coherently
Information Association (CIRIA). This is a general suggests that for an effective asset management
guidance provided for maintaining physical assets on implementation a clear idea of the goals and objectives of
the infrastructure network in the UK.CIRIA document the organisation is required, not least in terms of the type
shares information and best practice on undertaking of network and types of assets. Modelling the condition of
asset management for physical assets while ensuring the assets therein and determining their performance in
skills are retained for delivering challenging and order to develop implementation strategies with adequate
innovative solutions. feasibility analysis is required in order to determine
the selection criteria of a project and finally to allocate
ll Asset owners in the UK, for example Network Rail and
appropriate funds and budget for its implementation. These
the Highways Agency produce their own specific asset
are the essential elements that underpin and ultimately
management strategies, highlighting key areas of
steer an asset management process.
work and how different aspects of asset management
will deliver business plan objectives. For example, the
Highways Agency, in association with the Department Data Management
for Transport and other UK motorway and truck road Effective data management is crucial in implementing
authorities, produce the Design Manual for Roads and effective asset management systems. Elements include
Bridges (DMRB). the amount of data, how it is held, who has access to it and
how it is managed. These are all fundamental to achieving
optimal performance of the asset and the management
Asset Management Systems (Theory) team within an infrastructure maintenance environment.
It is worthwhile to acknowledge the various components
of an asset management system that will help understand Typically, data sets may be housed in a number of different
its key elements and the associated challenges of ways; however all should be managed with a similar set
amalgamating them into one coherent framework. This of policies, which rigorously address the following (Faiz,
section therefore, discusses what an asset management et.al, 2009):
system is based on the literature review and elaborates ll Network Location Data – with GPS mapping, where
on its key components for better understanding. appropriate
Guidance for the development and implementation of an ll Inventory Data
asset management system can be found in a number of
ll Condition Data
the guidance documents described in the previous section.
They outline the processes that they follow in order to set ll Inspection Data – Last undertaken/next due
up a framework for an effective management system.
ll Maintenance Records
Based on the review of literature from various asset
management guidance bodies discussed in Section X, it ll Reporting – for engineering and business performance
is observed that Asset management concepts are based
ll Quality Assurance

Organised by
India Chapter of American Concrete Institute 639
Technical Papers

By ensuring that datasets are adequately maintained and ll Utilisation: An important element to assess and
kept up-to-date, confidence in the methods chosen to evaluate current state of the asset. For example
allocate maintenance provision can be improved, if not Questions are posed such as ‘are the assets over-
assured. utilised, under-utilised?oris asset utilisation at an
optimum level? Is its utilisation significant enough to
Moreover, decision support systems provide evidentiary
be justified as profitable asset?
support for the selection of project, and can be invaluable
when submitting bids at the beginning of funding cycles.
The following sub-sections discuss key challenges in
Organisational Behaviour
effective adoption of asset management systems in Asset management works best when the emphasis
practice. is placed largely on change management, where
employees of the appropriate competence and seniority
Single Referencing System are taught how to properly manage the assets within
their remit and actively take responsibility for them
ICE (2011) suggests that most of the current data
(ICE, 2011). However this requires care as Kellick (2010)
management systems and practices are often referenced
recognises problems,not least where development and
to separate network models, rather than defining them
initiation of implementing asset management systems
within a single referencing model, which is in accordance
in an organisation has become a responsibility of all,
with CSS (2004) which stipulates that any data should be
ending up being a responsibility of none, resulting in a
referenced to National Grid Co-ordinates and National
lack of ownership of any set actions and non-uniformity
Streets Gazetteer.
in approach. Therefore it is evident that ownership,
accountability and responsibility remain key factors of the
successful implementation of asset management for any
organisation.
Both The Institute of Civil Engineers ‘Manual of Highways
Design and Management’ (ICE, 2011) and Kellick (2010)
suggest that getting the commitment from the organisation
and its senior management is essential for the asset
management practice to get an initial start. Kellick (2010)
also highlights that involvement of senior management
in formulation of asset management system right from
the early stages will initiate and enable accessibility of
financial and human resources. Both The Institute of Civil
Engineers ‘Manual of Highways Design and Management’
(ICE, 2011) and Kellick (2010) agree that there should
be an asset management steering group or a working
Fig. 2: Components of Asset Management (ICE, 2011)
group, which will focus the direction of work to business
and industry objectives, whilst ensuring the interaction
Data Management
of different departments to exchange knowledge and
ICE (2011) further highlights that, the inventory of data resources through effective communications
maybe either (or both) inadequate or obsolete in terms
of the condition data. One cannot underemphasise the
role of adequate and accurate data in implementing Asset Management Systems and Tools –
effective asset management system. Hence undertaking Industry Perspective
gap analysis based on inaccurate data could result in an In summary, all asset management systems aim to identify
overall redundant asset management system due to a and undertake to ensure the following tasks are achieved
cacophony of errors, resulting from lack of understanding (Prescott et al 2013, Faiz et al, 2009, FHWA, 2005):
and inaccurate analysis of asset condition and therefore
inappropriate decision-making. ll Provide an inventory of assets and their ownership.

ICE (2009) classifiesthe challenges in Asset Management ll Obtain information relating to the current condition of
as: the assets and its’ utilisation therefore understand the
asset performance.
ll Inventory: Questions like what, where condition, value,
performance, significance and Impact on the network ll Identify an expected level of performance from
are important. thisasset base, based on organisational objectives,
customer expectations and performance indicators.
ll Impacts: Short term, long term and medium term? Are
the objectives deliverable cost effectively? ll Undertaking gap analysis –i.e. the difference between

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

640 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Health & Life Cycle Monitoring of Infrastructure Asset Management- An Emerging Industry and the Role of Long Term Planning

existing condition of the asset and expected condition asset management systems in order to review the impact
of the Asset. of plausible changing socio-economic, environmental and
regulatory influences, rather the economic implications.
ll Where appropriate include future demand on the
The proposed research framework aims to address these
critical infrastructure which can aid in developing long
changes and enable effective decision-making for long
term and short term action plans.
term asset management.
ll Provision of short term plans that are tied in with the
asset performance and the developed gap analysis. Need for Long Term Planning and Resilience
ll Identify a Long term development plan that look at One of the prime lessons learnt from the Sandy hurricane in
financial plans, risk management plans, intermediate the USA in 2012 wasthat we need to design both redundancy
plans i.e. medium term plans and develop medium and flexibility into all infrastructure systems in order to
term or intermediate financial plan and cash flow create resilience (Lee, 2012). Lee (2012) highlighted that
predictions. out of such natural disasters comes the opportunity to
develop adaptive strategies for the future. Similarly, a
number of recent events in the UK demonstrated the
Decision Support Tools vulnerability of transportation networks and how this can
Decision-making processes require consideration of cause substantial disruption across the whole country.
a broad range of problem areas and require suitable For example, the unusually low temperatures in 2010, high
‘optioneering’ approaches in order to develop effective flood levels in 2007 and the eruption of the Eyjafjallajokull
solutions. A decision support tool is used to aid and volcano in Iceland in 2010, exposed the vulnerability in
improve this process. A typical decision support tool the UK’s national infrastructure (in particular airlines)
comprises of 3 components, to rapid breakdown and to some extent failure. The UK
ll An information database, Cabinet Office (2011) highlighted that these events not only
caused inconvenience to the public but incurred financial
ll a systematic course of action, which interrogates the losses in the form of lost revenues, reputational damage
existing knowledge from the stored data using a tool- and contractual fines and potential for legal action. For
based application which enables user interface (Faizet. example, the 2007 floods alone cost the UK economy
al, 2009). over £4 billion, and the damage specifically to critical
ll Identification of optimal maintenance strategies, which infrastructure was valued at approximately £674 million.
minimise risk of failure along with whole life costs. This substantial economic outlay reinforces a critical need
(Faiz et al, 2009). for related organisations to manage and mitigate ‘risks’
and embed ‘resilience’ into their business processes (UK
Limitations of Current Asset Management Systems – Cabinet Office, 2011). This justifies the need for a paradigm
Industry perspective shift that places emphasis on designing in “Resilience”
rather than designing in “Resistance” as is the case for
Michele (2011) argues that asset management has
many present solutions (Rogers et, al, 2012).
significant influence on the use and growth of infrastructure
development. However, the author points out that without Resilience in the infrastructure industry implies that the
understanding the broader impacts of technological and network (road, rail, utilities, water telecommunications
social evolution, and the associated complexity and diversity etc.) is up and running, even if not in its best shape, but
it brings, the system will invariably waste economic, social, continues to perform for its intended purpose even in
cultural and environmental resources. Lemer (1998) and harsh and unpredictable conditions that it may be exposed
Michele (2011) both highlight the growing challenges in to at any given time. In UK, £30 billion a year is invested
urban infrastructure management when trying to relate in maintaining and managing the infrastructure, which is
reduced funds, higher user’s interest and attention to such set to increase to £50 billion by 2030 (HM Treasury, 2010).
things as: the quality of service; increasing interest in public This justifies the need to embed resilience into planning
health and safety; enhanced focus on water and air quality and design for such an investment to reap benefits for the
and green spaces; reduction in vehicle traffic and noise; long term.
ageing population; and consequent difficulties in accessing Figure 3 identifies the various attributes that constitute a
the town and cities services and obsolescence of structures resilient transport solutions,in order to meet theresilience
as a result of town growth. definition as stated in this thesis, in other words that
Lemer (1998) suggests that the future of asset solutions should ‘offer flexibility and multi-functionality
management is where an embodied version of the capital to meet changing demands and user patterns while
value can consider the significance of cultural, economic, continuing to be fit for purpose and sustainable in the light
environmental, political and social dynamics relating to of changing conditions while continuing to be cost effective
an infrastructure asset. However, the author highlights in its construction, maintenance and operation over its life
that there is still more work needed in order to develop cycle.’

Organised by
India Chapter of American Concrete Institute 641
Technical Papers

factors, changes in demographics, technological reforms

and its influence on transportation industry. However,
there is no significant information available in the public
domain on methodology or framework on how this can
be incorporated in the decision making process. As a
result, although the need is established and discussed,
lack of publically available information on frameworks
and methodologies of incorporating the same in working
practices is evident. This shows the need for further work
in the future, where existing asset management tools while
balancing the ever important cost, risk and performance
aspects of assets also emphasise on considering the long
term performance of these assets beyond the typical
funding regime cycles. And expand the vision to see how
these assets perform in the light of changing conditions
throughout the design life of the asset. In this way we can
truly consider the whole life performance of the asset.
Fig. 3: Resilient Transportation Infrastructure

Therefore a solution which provides most (if not all) of Acknowledgement

these attributes is considered to provide the greatest The paper is published based on the author’s PhD research
resilience. Whilst this could be a relative comparison and findings at the University of Birmingham. The author
between various options, this definition and explanation is thankful to the supervisors Prof Ian Jefferson and Dr
covers the essence of determining a resilient solution Dexter Hunt at the School of Civil Engineering, University of
which a range of stakeholders (i.e. industry, academia Birmingham, UK for the invaluable support and guidance.
geotechnical engineers, strategic planners and asset The author is thankful to sponsors Amey and University
managers) can relate to. of Birmingham for sponsoring the above research and
providing guidance and help.
We do have many studies, data of Developed Nations in
this Engineering Science but the NEED is to have design
Referrences
a monitoring system which has “in built” parameters
1. Cabinet Office, (2010). Strategic Framework and Policy Statement
suitable to conditions, parameters, “tools” for our Country. Strategic Framework and Policy Statement. London: Cabinet Office,
This will surely reduce the “life cycle costs”. Reality is Crown Copyright.
different and “knee jerk” designs , systems may only work 2. County Surveyors Society (Css), 2004.Framework for Highway Asset
for “limited” purpose requiring further “costs burden” of Management. Css.
maintaining assets. 3. Department for Transport (Dft), (2010). Climate Change Adaptation
Plan for Transport 2010-2012: Enhancing Resilience to Climate
Change. London: Dft Publications, Crown Copyright.
Conclusion 4. International Infrastructure Management Manual, 2011. 4th Ed. New
With established that “time proven” tools of predicting Zealand Asset Management Support
through monitoring of service life & health of 5. International Organization For Standardization (Iso) 55000, 2012.
Infrastructure assets adopted by the Developed Nations Asset Management. International Organization for Standardization.
is helpful and guiding force and caters the “NEED” for 6. Lemer, A. 1998. Progress Toward Integrated Infrastructure-Assets-
infrastructure development in emerging economies such Management Systems: Gis and Beyond. In: Innovations in Urban
as India. However, “designing” to suit our Nation is the Infrastructure Seminar of the Apwa International Public Works
Congress, 1998. Citeseer, Pp. 7-24.
“ground reality”. The asset management systems are
7. Michele, D. S. & Daniela, L. 2011.Decision-Support Tools for
required to incorporate long term planning and decision-
Municipal Infrastructure Maintenance Management.procedia
making such that the assets continue to be fit for purpose Computer Science, 3, 36-41.
and serviceable under changing conditions. There is a vast 8. Publically Available Standard (Pas) 55, 2008.Optimal Management
amount of literature focussing on the need to embed long of Physical Assets.british Standards Institution.
term planning in asset management. An equal amount of 9. Shah, J., Jefferson, I. And Hunt, D. (2013). Resilient Infrastructure
importance is given to embedding resilience in planning Asset Management – A Global Perspective and Lessons for
and decision-making. However the existing tools/ Infrastructure in India. In: Advances in Science & Technology of
frameworks, mainly focus on the risk due to vulnerabilities Concrete. Mumbai: India Chapter of American Concrete Institute,
Pp.180-184.
arising from climate change and man made threats such
as accidents and factors affecting national security, etc. 10. Shah, J., Jefferson, I. and Hunt, D. (2014).Resilience Assessment
for Geotechnical Infrastructure Assets.infrastructure Asset
There is also literature available on the need to expand the Management. [Online] Available At: Http://Dx.doi.org/10.1680/
vision to consider changes in socio-economic and political Iasma.14.00007 [Accessed 25 Sep. 2014].

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

642 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Health & Life Cycle Monitoring of Infrastructure Asset Management- An Emerging Industry and the Role of Long Term Planning

11. Theiam.org (2012) What is Asset Management?, Iam - The Institute 13. Kellick, P., 2010. Developing a Strategic Asset Management
of Asset Management. [Online] Available at: Http://Theiam.org/ Framework. Proceedings of the Ice - Municipal Engineer.vol 163.
What-Is-Asset- Management [Accessed: 14 April 2012]. Issue 4. Pp 221 -224
12. Trl.co.uk, (2014). Intelligent Transport Systems - Trl. [Online] 14. Lee, A. S. 1991. Integrating Positivist and Interpretive Approaches to
Available At: Http://Www.trl.co.uk/Solutions/Intelligent-Transport- Organizational Research. Organization Science, 2, 342-365.
Systems/ [Accessed 5 Mar. 2013].

Janvi Shah
Ms. Janvi Shah has over 4 years of experience in project managing and delivering infrastructure projects
for public sector clients such as Highways Agency, Network Rail and Local authorities. Recent experience
includes undertaking project management of over £5 million worth of Infrastructure Projects (Highways
Agency’s Pinch Point Highway Improvement programme). A proven track record of undertaking value
management, whole life costing, risk management, multi-disciplinary project planning including extensive
stakeholder liaison, resource scheduling in a complex environment leading to effective project delivery on
time and budget. Initiated and developed the sustainability assessment process and now championing same
in the current team for project feasibility studies.
Currently also undertaking a PhD in Resilient Infrastructure Asset Management. The research focuses
on developing strategic asset management framework for providing decision support through assessing
resilience of infrastructure plans. The resilience assessment considers the key performance indicators
linked to the organisational objectives in the light of changing socio-economic, environmental, regulatory
and technological conditions of the future. Published papers on Resilience in Asset Management and
presented to large international expert audiences in Asset Management.
She is a winner of the ‘Young Geotechnical Engineer 2014’ Award and the ‘Infrastructure Asset Management
Prize 2014’. Her papers have been accepted in conferences in Australia, Indiaand USA.

Organised by
India Chapter of American Concrete Institute 643
Technical Papers

Disclosing the creep mechanisms of cement paste by

micro- indentation at different relative humidity
L. Sorelli, J. Frech-Baronet, Z. Chen
Department of Civil Engineering, Université Laval

Abstract in bridge pre-stressing cables, and increases the stress

in the steel bars of reinforced concretes). The creep
This work aims at characterizing the long term creep
behavior of concrete depends on several factors, such as:
behavior of hardened cement paste by microindentation
the microstructure composition, the porosity distribution,
tests at uniform hygral equilibrium at different relative
the thermal and hygral conditions, and the age at the
humidity. Recent studies on different kinds of cement pastes
loading time (Neville, 1971). Among all of them, the water
have shown that the asymptotic creep rate measured after
seems to play a critical role on the creep behavior (Acker
few minutes by microindentation well correlates to the long
and Ulm, 2001). Concrete creep is commonly split into
term creep rate of macroscopic samples in compression
basic creep, i.e., the deformation occurring without water
after 15 years. The technique of micro-indentation allows
exchange with the environment, and the drying creep, i.e.,
capturing the creep mechanisms at relatively small time
the deformation in excess to the basic creep due to drying
scale due to the limited dimension of the probed volume
under constant load.
which is of few hundreds of micrometers. In this work, we
further investigate the creep mechanisms of the cement Troxell et al. carried out creep compressive tests on
paste by micro-indentation at different relative humidity in concrete cylinders (100 mm diameter and 150 mm height)
order to better understand the role of water on the creep at different relative humidity for 30 years (Troxell et al.,
mechanisms of the Calcium-Silicate-Hydrates (C-S-H) 1958). Their results showed that, after reaching the hygral
which are still today not well understood. The duality equilibrium at about 1 year, the creep rate seems quite
between creep and relaxation tests has been studied in independent of the relative humidity (50%, 70% and 100%).
order to distinguish different mechanisms. Interestingly, While the applied load is not sufficient to significantly
the results showed that the creep rate increases, when the change the pore volume and the relative humidity in
relative humidity increases from 50% to 85% in the range the pore (Bazant et al., 1997), the applied stress can
of the capillary pressures. The mechanical response cause water micro- diffusion between pore of different
was modeled with simplified visco-elastic models. The sizes, especially between the gel pores (~10-100 nm)
comparison between the relaxation and creep tests hints and the capillary pores (~0.1-10 µm), which accompany
for an additional mechanism to the shear sliding of C-S-H, creep at loading (Powers, 1968). Recently, it was found
e.g., micro-cracks with water micro-diffusion. Finally, experimentally that the loading and unloading causes an
the conclusions of this work are twofold: (i) the relative instantaneous change of about 5% of RELATIVE HUMIDITY
humidity affects the long term creep rate of cement paste (Wyrzykowski and Lura, 2014). Bernardi et al. showed the
from r.h. values of 50% to 85%; (ii) during a creep test short term creep is mainly spherical (i.e. volume change)
at constant relative humidity, capillary pressures may and due to water movement, while the long term creep
induce local force in the contact points of the C-S-H gels, is mainly deviatoric (i.e., shape change) and due to the
which in turn may increase the friction of the C-S-H sliding viscosity of the cement paste (Bernard et al. 2003). They
mechanism. also found that Poisson’s ratio varies from about 0.15-0.2
to about 0.4-0.5 during a creep test.
Keywords: Cement paste, creep, relaxation,
microindentation tests, relative humidity, C-S-H sliding Numerous theories have described the creep mechanisms
mechanism. in concrete, but no one has reached an overall consensus.
Some theories which have considered the role of water in
concrete creep are:
Introduction
Creep is the time dependent deformation due to an external ll Diffusion of hindered water or layers of absorbed
load and it has, from a structural point of view, favorable water. Creep is caused by the diffusion of hindered
effects (e.g., it can relax the stresses induced by restrained or absorbed water under stress concentration state
shrinkage or temperature deformation) and detrimental (Feldman, 1972; Powers, 1965; Sercombe et al., 2000;
effects (e.g., it increases the deflection, reduces the force Wyrzykowski and Lura, 2014). In particular, Feldman

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

644 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Disclosing the creep mechanisms of cement paste by micro- indentation at different relative humidity

and Sereda pointed out that inter-particle bonds can Lately, microindentation techniques have been emerging
be broken and remake under stress concentration for measuring the creep properties of cement- based
states and that the diffusion of interlayer water is materials as it allows significant reduction of the time
responsible for the major portion of volume change scale for observing creep phenomena ((Nguyen et al., 2013;
(Feldman, 1972) Nguyen et al., 2014; Pourbeik et al., 2013; Zhang, 2014)).
ll Micro-diffusion gel theory. Water molecules between Zhang et al. have proven that, for several mix designs, the
the gel pores and the capillary pores due to energy long term creep rate measured by microindentation and
imbalance induced by local stress peaks (Bazant and the ones measured by compressive tests over 30 years
Chern, 1985) are linearly correlated (Zhang et al., 2013). Zhang found
that the creep rate measured by microindentation on a
ll Sliding of C-S-H sheets. The basic creep of cement
cement paste increases by a factor of 5 when the relative
pastes was explained as a sliding mechanism of sheets
humidity level increases from 11% to 95% (Zhang, 2014).
of C-S-H under a shear stress state (Vandamme et al.
2009; Sanahuja et al. 2010). Recently, the inelastic Microindentation appears as an ideal technique to
slit motion was linked to the shear stress accounting characterize the effect of humidity on the viscosity as
for the lubricant role of water in the framework of it takes a limited time to reach the hygral equilibrium
micromechanics (Shahidi et al. 2014); in the material depth under study (as explained in the
ll Rearrangement and compaction of the C−S−H gel example of the next section). The scope of this work is
particles like a sort of secondary consolidation process to better understand the effect of humidity on the creep
(Jennings, 2004) mechanisms of a cement paste by microindentation. How
does the relative humidity affect the creep response of a
ll Micro-diffusion process induced by water micro-
cement paste at hygral equilibrium? Is the effect of relative
diffusion between the micro-cracks and the tip of the
humidity the same on the short term and long term creep?
propagating major crack (Rossi et al. 2012).
How relative humidity affect the dual mechanisms of
Decoupling basic and drying creep is not trivial as reaching creep and relaxation of a cement paste? The key questions
an uniform relative humidity can take up to 1 year for a this work aims at answering with this work.
sample size of 15 cm. Ideally one must use extremely
thin specimens and lower the environmental humidity
gradually and so slowly that the distribution of humidity
Material and Methods
throughout the sample thickness remains almost constant Materials under study
(Bazǎnt and Raftshol, 1982). The few data available on
experimental tests performed at hygral equilibrium (i.e. This study employed a cement paste and a concrete made
no moisture exchange) show that the lower the moisture with ordinary Portland (type I) with water-to- cement ratio
content, the lower the creep deformation (Bazant and (w/c) of 0.4. The mix design of concrete was adapted to
Chern, 1985) so that the creep rate is proportional to the have the same matrix characteristics of the cement paste.
square of the relative humidity as follows: The samples were cast in cylindrical molds of diameter
100 mm and 150 mm for cement paste and concrete,
........................(1) respectively. They were then cured at 100% R.H for 28
days and then stored at 50% RELATIVE HUMIDITY for 2
where the viscosity apparent η(h) is a function ϕh of an months. The samples were tested 90 days after casting
empirical function of pore humidity h, E is the elastic to reduce the effect of aging. A cubic sample of 30 mm
modulus, and τµ is the characteristic time at the reference side was sawn from the core of the cylindrical samples.
humidity. Function ϕh should be directly determined from Secondly, a coarse grinding was performed with the use
creep tests of specimens which were loaded after being of abrasive paper. Thirdly, the surfaces were polished by
dried to thermodynamic equilibrium at various humidifies 1 µm fineness diamond suspension oil-based solution.
and were at constant humidity during creep. Based Finally, a 0.250 µm fineness diamond paste was used to
on data on cement paste specimens pre-dried in oven finalize the polishing. Special attention was paid to keep
and then rewetted , constant αh may be as small as 0.1 the samples levelled since the angle of indentation could
(Wittmann, 1968). However, the pre-drying was made in an influence the results of measurement. After polishing,
oven at 105° to achieve equilibrium soon enough. However, the samples were put in an ultrasonic bath to remove any
heating in an oven involves a thermal strain and possible trace of diamond particles left on the surface. The samples
microcracks. This probably happens at the subsequent were cured for 1 day at the testing relative humidity before
rewetting to a desired relative humidity. Without pre- microindentation testing. The size of the indentation
drying, a larger value of αη of 0.5 (Bažant et al., 1976) was probed volume D can be estimated to be about 5 times the
measured. Recent relaxation three-point bending tests on maximum penetration depth (hmax ~ 40 µm in this study)
T-section beams of millimeter size have also confirmed (Vandamme and Ulm, 2009), which can be estimated
that fully saturated cement paste relax much faster than approximately as D ~ 200 µm. At the same diffusivity,
those at lower humidity of about 11% (Alizadeh et al., 2010). the diffusion half-times are proportional to the square

Organised by
India Chapter of American Concrete Institute 645
Technical Papers

of the distance. For a normal low-porosity concrete, the (Figure 1.a). The contact stiffness S=dP/dh is the slope
diffusion half-time for a wall 0.15 m thickness is about measured during the initial stages of the unloading
1 year (Bazant and Chern, 1985). Thus, for the micro curve (Figure 1.b). The Young modulus E of an isotropic
diffusion, the half time would be about 1 x 365 x 3600 x material is estimated from the indentation modulus M
24 x (200-6) 2/0.152= 56 seconds. Thus, a curing time of as follows : M-1=(1-υ2) E+(1-υ2) Ei, where Ei and υi are the
24 hours would be enough to allow a uniform relative elastic modulus and the Poisson’s ratio of the diamond tip,
humidity within the probed volume which are equal to 1141 GPa and 0.07, respectively; while
The sample surface was carefully prepared for cement E and υ are the Young modulus and the Poisson’s ratio of
paste and concrete specimens according to a well the material, respectively. Furthermore, to account for
established protocol (Miller et al. 2008). The Roughness the tip imperfection, the contact area function, AC=τ2a,
Mean Square (RMS), which is a statistical parameter is calibrated according to a standard procedure on a
was checked by Atomistic Force Microscope (AFM). reference fused silica sample. Following the standard
Topographic images were carried out on different zones. Oliver and Pharr’s method, the contact area can be
As common procedure, a Gaussian filter was applied to estimated as AC =3√3h2Ctan2θ , where hc =hm -ε Pmax /S. The
filter out the wave larger than 8.0 (Miller, Bobko et al. geometry correction factor β and the intercept factor ε are
2008). The RMS value was obtained by averaging three 1.034 and 0.75, respectively, for Berkovich-type indenter
selected topographic area of 50 µm by 50 µm per zone (Fischer-Cripps 2011).
of interest. The RMS was about 65 nm with a standard Figure 1.c shows an experiment where the load is linearly
deviation of 36 nm. applied over a time τL, then held constant over a time τH,
and finally reduced to zero over a time τU. While plastic
Microindentation test set-up with control of relative deformation often occurs during loading, it is generally
humidity assumed that the unloading curve is elastic. Different
In order to control the relative humidity (h), a hermetic parameters are employed to describe the creep behavior
enclosure was built. The system consists of a hermetic and relaxation behavior, such as the creep coefficient C
enclosure which is connected in series to an Erlenmeyer and the relaxation coefficient R:
flask containing a saturated salts solution, which was
employed to reach specific humidity equilibrium. A air ...(3,4)
pump forces the circulation of vaporized air through flask
and the chamber. Within the chamber, the relative humidity Additionally, we define the creep rate coefficients S1 and
and CO2 concentration are continuously monitored by two S2 defined at the beginning and the end of the holding
sensors. In this work, we studied two levels of relative time, respectively, as follows:
humidity, namely 33% and 85%.
.......................(5,6)
Basics of microindentation theory
Microindentation has been successfully employed to Analogously, we can define the relaxation rate coefficients
measure the elasticity and hardness of cement paste and χ1 and χ2 for the relaxation test as follows:
other hydrates products (Nguyen et al., 2013; Nguyen et
al., 2014; Pourbeik et al., 2013). During an indentation test, ....(7,8)
the force (P) and the penetration depth (h) are measured
simultaneously (Figure 1.a). A typical P-h curve (Figure
The above creep/relaxation rates have been calculated
1.a) includes a loading curve up to the maximum load Pmax,
followed by an unloading curve. Note that the penetration by linearly interpolation of the experimental points over a
depth vs. load curve profile is parabolic in the ideal case period of 3 seconds for the initial slope (S1 and χ1) and over
of conical indenters with no friction on the tip-material time 100 seconds for the final slopes (S2 and χ2).
interface. For this study we employed a Berkovich-type
indenter which is three-sided pyramid with an equivalent
cone semi-angle θ of 70.3° as shown in Figure 2a. The
analysis of the P-h curve consists in extracting the
indentation properties, the indentation Modulus and the
indentation Hardness H, as follows

...............(2)

The coefficient β accounts for the non symmetrical Fig. 1: (a) Schematic represe(nata)tion of a an indentation test; (b)
shape of the indenter. The projected area AC of the Typical curve applied load vs. penetration(c) depth curve P-h; (b)
indenter-sample contact depends on the contact depth hc typical time histories of the penetration depth and applied load.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

646 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Disclosing the creep mechanisms of cement paste by micro- indentation at different relative humidity

Analogously for a relaxation test for which the load Pmax

is instantaneously applied at time t, the time dependent
load function reads as shown in equation (10):
For the calculation of those functions, one can read
(Vandamme, 2008).

Model-test #2: logarithmic deviatoric creep

This model relies on the hypotheses the creep is purely
deviatoric (i.e., only variation of shape) and logarithmic in
time, as proposed by Vandamme et al. (Vandamme and
Ulm, 2009), and that the Poisson’s ratio is constant (υ=0.2).
Concrete creep was observed to be rather logarithmic in
Fig. 2: (a) Example of load control test for a creep test; (b) time for several structures as well as for several studies
example of displacement control test for a relaxation test. of microindentation tests. The model which is based on
the assumption that the compliance modulus is:
Simplified visco-elastic model for microindentation
tests ....................................(11)
The scope of this section is to test simplified visco-elastic
models in order to verify different hypotheses.
If we assumes a constant Poisson’s ratio υ = 0.2 , the shear
Model-test #1: split of the reversible and irreversible modulus results K = 1.33G . The penetration depth results
creep
The first model relies on the hypothesis that the visco-elastic (12)
response can be decomposed into two mechanisms acting
at different time, which are: (1) a short term creep governed
by a load-induced micro- diffusion. Based on the diffusion Experimental Results
half time of a typical cement paste, one can estimate the As an example, Figure 3.a shows an example of grid
characteristic time associated to the micro-diffusion for a indentation tests (100 indents) carried on the sample for
water trajectory similar to the probed depth (~200 µm) is the relative humidity of 33%. In addition, the Figure 3.b and
about 60 seconds; (2) a long term creep governed by sliding 2.c show the mapping of the indentation Modulus M and
of the C-S-H sheets or rearrangement of cement paste gel. the Hardness H on the same sample. One can appreciate
For sake the simplicity, we assumed that the volumetric that the cement paste is quite homogeneous.
creep is governed by a Kelvin-Voigt model and the deviatoric
creep is governed by a Maxwell model. The former model Effect of relative humidity on the hardness of a cement
is reversible as associated to water micro-diffusion, while paste
the latter is irreversible as associated to viscous shear
Figure 4.a and b shows the effect of the relative humidity
behaviour of the C-S-H within the cement paste. This model
on the elastic modulus M of the cement paste and the
implies that the Poisson’s ratio may change as the bulk
indentation hardness H, respectively, for both load and
modulus and the shear modulus varies independently.
displacement control tests. The standard deviation for
For a creep test with an instantaneous application of the 100 tests is also indicated. The elastic modulus has been
load at t=0, the penetration depth varies in time as shown calculated from the indentation modulus assuming a
in equation (9):

....................(9)

Where

..................(10)

Where

Organised by
India Chapter of American Concrete Institute 647
Technical Papers

Fig. 3: Example of indentation grid on the cement paste for

r.h.=33% (a); mappingof the M-modulus (b); mapping of the
Hardness (c).

Fig. 5: At hygral equilibrium for different relative humidity: (a)

mean curve penetration depth vs. time h(t) for a creep test; (b)
mean curve load depth vs. time P(t) for a relaxation test.

control tests (red and black curves). The slight difference

between the Hardness values measured by load and
displacement control tests is expected as the contact area
AC can be only estimated in a load control test, while it is
known in a displacement control test.

Effect of the relative humidity on the creep and

relaxation behaviour of a cement paste
Figure 5.a shows the mean creep curves in term of
penetration depth vs. time during the holding time for
different level of relative humidity. When relative humidity
Fig. 3: Effect of the relative humidity on the indentation Modulus reduces, creep deformation decreases. The result
M (a) and on the indentation Hardness H (b) for both force control dispersion was very limited at few percentages.
and displacement control tests.
Figure 5.b shows the mean relaxation curves in terms
constant Poisson’s ratio (υ=0.2) as E=M (1-υ ). While the
2 of load vs. time during the holding time for each relative
relative humidity does not affect significantly the elastic humidity level. The load relaxation curves confirm the
modulus, its effect on the Hardness is more remarkable. observed effect of humidity on the viscous deformation: the
more the relative humidity, the more the load relaxation
The Hardness value can be related to the strength increases. However, one can note that for the creep tests
(cohesion and friction) of the cement paste (Cariou et al., the curves are more far apart from each other, that is, the
2008). The reduction of the strength due to the increase effect of humidity appears more important.
of relative humidity was aslo reported by Wittman
(Wittmann, 1973). This effect can be explained by thinking
of a sponge of which structure becomes stronger when Analysis and Discussion
partially saturated. This was mathematically formulated
within the framework of poro-mechanics of porous media Effect of the relative humidity on the creep coefficients
(Dormieux et al., 2006). One can appreciate the good Figure 6.a shows the creep coefficient C as well the
repeatability the results for the load control and force relaxation coefficient R in function of the relative

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

648 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Disclosing the creep mechanisms of cement paste by micro- indentation at different relative humidity

(a)
(a)

(b)
(b)

Fig. 6: Effect of the relative humidity on the creep coefficient

C and relaxation coefficient R (a) as well as on the normalized
creep rate coefficient S2 and χ2.

Fig. 8: Comparison between the experimental and simulated

(a) curves for creep tests (a) and relaxation tests (b) at different
relative humidity.

(b)

Fig. 9: Effect of the relative humidity on the viscosity model

parameter η2.

humidity. The creep coefficient increases with the

increase of the relative humidity, especially in the
humidity range from 55% to 85%. Interestingly, the
trend of the relaxation coefficient is similar. Figure
6.b shows the creep rate coefficient S1 defined at the
time of the load application in correlation between the
initial slopes S1 measured in the creep testes with those
which were measured in the relaxation tests. Figure 7.b
Fig. 7: Correlation of the initial (S1) and final slope (S2) for the shows the same correlation for the final slope S2. The
creep and relaxation tests correlations are quite linear.

Organised by
India Chapter of American Concrete Institute 649
Technical Papers

Table 1
Estimated needs for global infrastructure in different categories. Period 2013-2030[2]

Creep Relaxation

R.H k1 [GPa] η1 τ1 [sec] k2 [GPa] η2 τ2 [sec] k1 [GPa] η1 τ1 [sec] k2 [GPa] η2 τ2

R.H.33% 0.040 0.003 7.4 24.248 3.291 0.040 0.066 0.001 8.0 227.542 28.510 0.066

R.H.55% 0.042 0.004 7.3 18.237 2.514 0.042 0.165 0.003 7.5 170.268 22.558 0.165

R.H.75% 0.059 0.006 6.0 12.888 2.164 0.059 0.057 0.001 7.2 124.139 17.158 0.057

R.H.85% 0.056 0.006 6.6 10.324 1.561 0.056 0.059 0.001 6.8 95.251 14.059 0.059

(a)

Fig. 11: Schematic representation of the coupling effect among

capillary forces, local forces in the contact points of C-S-H gels,
and the frictional C-S-H sliding mechanism.

Testing hypotheses by simplified models

Table 1 shows the best fitting parameters for the model-
(b)
test #1 for the creep and relaxation tests. Figure 8 shows
the comparison between the model simulation and the
experiment. The model does not capture well the overall
shape of the creep h(t) or relaxation curves P(t). However,
the comparison between the asymptotic curves at the end
of the holding phase (slope of the curve at 300 s) is quite
acceptable. The model seems deficient in capturing the
short term creep/relaxation behaviour (up to 50 s) and the
long term creep/relaxation rate. The change of humidity
from 33% to 85% reduced the long term viscosity η2 of
a factor of about 2. Interestingly, Figure 9 shows that the
relative humidity affects the viscosity coefficient η2, which
models the long term creep, for both load control (creep)
and displacement control (relaxation) tests, but the effect is
much more important for the former case with a difference
Fig. 10: Comparison between the simulated and measured
of about 90%. Since delayed plasticity (or damage) can
penetration depth vs time curves for different humidity (a);
not be present during a relaxation test (Vandamme et al.,
graphical representation of the model parameters C and τ for
each relative humidity (b). 2012), one could postulate that this difference may be due to
micro-cracking under creep test at constant loading. This
Figure 7.a shows the creep rate coefficient χ1(h)/χ1(h18%) hints that two mechanisms may be at stake during a creep
and S1(h)/S1(h18%) in function of the relative humidity tests, a basic creep due to shear sliding of C-S-H sheets
for relaxation and creep tests, respectively. The and a secondary mechanism due to micro-cracking.
normalization at the lowest relative humidity is herein The second model-test is based on the hypothesis of a
considered as the reference state. Figure 7.b shows the deviatoric logarithmic creep and constant Poisson’s ratio
creep rate coefficient χ2(h)/χ2(h18%) and S2(h)/S2(h18%) in (υ=0.2). Figure 10.a shows the comparison between this
function of the relative humidity. model and the creep curves h(t) at different relative humidity.
The initial creep rates (normalized S1 and χ1) increases The simulation fits quite well the shape of the mechanical
slightly with the increases of relative humidity. More response at different relative humidity. Figure 10.b shows
important, the final creep rate (normalized S2 and the model parameters: when relative humidity increases,
χ2) increase significantly when the relative humidity the characteristic time (τ) increases and the constant C
increases from 55% to 85%. linearly decrease by a factor of 2 from 33% to 85%.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

650 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Disclosing the creep mechanisms of cement paste by micro- indentation at different relative humidity

Conclusions 4. Bazant, Z. and Chern, J., 1985. Concrete creep at variable humidity:
constitutive law and mechanism. Materials and structures, 18(1):
The present work presents original results on the effect of 1-20.
relative humidity on the mechanical response of cement 5. Bazant, Z.P., Hauggaard, A.B., Baweja, S. and Ulm, F.-J., 1997.
pastes by means of microindentation techniques. Based Microprestress-solidification theory for concrete creep. I: Aging and
on the presented results, the following conclusions can be drying effects. Journal of Engineering Mechanics, 123(11): 1188-1194.
drawn: 6. Bazǎnt, Z.P. and Raftshol, W.J., 1982. Effect of cracking in drying
and shrinkage specimens. Cement and concrete research, 12(2):
ll The indentation hardness of a cement paste reduces 209-226.
when the relative humidity increases due to poro- 7. Cariou, S., Ulm, F.-J. and Dormieux, L., 2008. Hardness–packing
mechanical effects of capillary forces; density scaling relations for cohesive- frictional porous materials.
Journal of the Mechanics and Physics of Solids, 56(3): 924-952.
ll The greater is the relative humidity, the greater is
8. Dormieux, L., Sanahuja, J. and Maghous, S., 2006. Influence of
the final extent of the viscous deformation and the capillary effects on strength of non-saturated porous media.
deformation rate for both a creep and relaxation test; Comptes Rendus Mécanique, 334(1): 19-24.

ll For each relative humidity, the creep deformation 9. Feldman, R.F., 1972. Mechanism of creep of hydrated Portland
cement paste. Cement and concrete research, 2(5): 521-540.
appears greater than that of the relaxation tests hinting
for possible micro-cracking effects; 10. Jennings, H.M., 2004. Colloid model of C− S− H and implications
to the problem of creep and shrinkage. Materials and structures,
ll The logarithmic model is more suitable to fit the 37(1): 59-70.
kinetics of creep tests than a combined Maxwell- 11. L’Hermite, R., 1959. What do we know about the plastic deformation
Kelvin Voigt model; and creep of concrete?
12. Neville, A.M., 1971. Creep of concrete: plain, reinforced, and
ll The viscosity coefficient of the logarithmic model is prestressed.
approximately linearly proportional to the relative
13. Nguyen, D.-T., Alizadeh, R., Beaudoin, J. and Raki, L., 2013.
humidity between relative humidity of 33% and 85%, Microindentation creep of secondary hydrated cement phases and
which is the range of the capillary pressures in the C–S–H. Materials and structures, 46(9): 1519-1525.
cement paste pores. 14. Nguyen, D.-T., Alizadeh, R., Beaudoin, J.J., Pourbeik, P. and Raki,
L., 2014. Microindentation creep of monophasic calcium–silicate–
The results of this work hints for a possible mechanism hydrates. Cement and Concrete Composites, 48: 118-126.
for explaining the effect of the relative humidity on the
15. Pourbeik, P., Alizadeh, R., Beaudoin, J.J., Nguyen, D.-T. and Raki,
long term creep rate. As expected in poromechanics, the L., 2013. Microindentation creep of 45 year old hydrated Portland
reduction of relative humidity causes an increase of the cement paste. Advances in Cement Research, 25(5): 301-306.
pressure capillary (suction), which in turns increases the 16. Powers, T., 1968. The thermodynamics of volume change and creep.
stress concentration (compression) in the contact points Matériaux et Construction, 1(6): 487- 507.
of the cement paste gels. In those sites, the C-S-H sheets 17. Powers, T.C., 1965. Mechanisms of shrinkage and reversible creep of
are subjected to a high value of normal compression hardened cement paste. The structure of concrete and its behaviour
forces which can increase the friction coefficient between under load: 319-344.
the C-S-H sheets. Finally, the effect is a reduction of the 18. Sercombe, J., Hellmich, C., Ulm, F.-J. and Mang, H., 2000. Modeling
creep rate due to the reduction of relative humidity. of early-age creep of shotcrete. I: model and model parameters.
Journal of Engineering Mechanics, 126(3): 284-291.
On-going works are focusing on the link with macroscopic 19. Troxell, G.E., Raphael, J.M. and Davis, R., 1958. Long-term creep
tests as well as the consideration of possible load induced and shrinkage of plain and reinforced concrete. ASTM, pp. 1101-1120.
drying creep. 20. Vandamme, M., 2008. The nanogranular origin of concrete creep:
a nanoindentation investigation of microstructure and fundamental
properties of calcium-silicate-hydrates, Massachusetts Institute
Acknowledgement of Technology.
We would like to acknowledge the support of NSERC 21. Vandamme, M., Tweedie, C.A., Constantinides, G., Ulm, F.-J. and
Canada for the support of this research through the Van Vliet, K.J., 2012. Quantifying plasticity- independent creep
compliance and relaxation of viscoelastoplastic materials under
Discovery Grant No. 386488-2010. contact loading. Journal of Materials Research, 27(01): 302-312.
22. Vandamme, M. and Ulm, F.-J., 2009. Nanogranular origin of concrete
References creep. Proceedings of the National Academy of Sciences, 106(26):
1. Acker, P. and Ulm, F.-J., 2001. Creep and shrinkage of concrete: 10552-10557.
physical origins and practical measurements. Nuclear Engineering
23. Wittmann, F., 1968. Surface tension skrinkage and strength of
and Design, 203(2): 143-158.
hardened cement paste. Matériaux et Construction, 1(6): 547-552.
2. Alizadeh, R., Beaudoin, J.J. and Raki, L., 2010. Viscoelastic nature
24. Wittmann, F., 1973. Interaction of hardened cement paste and water.
of calcium silicate hydrate. Cement and Concrete Composites,
Journal of the American ceramic society, 56(8): 409-415.
32(5): 369-376.
25. Wyrzykowski, M. and Lura, P., 2014. The effect of external load
3. Bažant, Z., Asghari, A. and Schmidt, J., 1976. Experimental study of
on internal relative humidity in concrete. Cement and concrete
creep of hardened Portland cement paste at variable water content.
research, 65: 58-63.
Matériaux et Construction, 9(4): 279-290.

Organised by
India Chapter of American Concrete Institute 651
Technical Papers

26. Zhang, Q., 2014. Creep properties of cementitious materials: effect 27. Zhang, Q., Le Roy, R., Vandamme, M. and Zuber, B., 2013. Long-term
of water and microstructure: An approach by microindentation, creep properties of cementitious materials–comparing compression
Université Paris-Est. tests on concrete with microindentation tests on cement, American
Society of Civil Engineers, pp. 1596-1604.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

652 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influences of w/c ratio on the properties of lightweight concrete

Influences of w/c ratio on the properties of lightweight concrete

Pengkun Hou, Xin Cheng, Zhaoheng Guo, Maoqiang Fu, Zonghui Zhou, Peng Du, Xiuzhi Zhang, Lina Zhang
School of Materials Science & Engineering, University of Jinan, Jinan, Shandong, China, 250022
Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, Jinan, Shandong, China, 250022

Abstract heat-insulating system, such as polystyrene board,

polyurethane board, polyethylene board. Although
The pore structures of lightweight cementitious materials
these materials significantly improve the energy-saving
play a vital role in both the mechanical and thermal
property of the buildings, their shortages, such as the
insulating properties. Water was used as an “air-
low aging resistivity, the instability of the volume, and
entraining agent (AEA)” for making porous cementitious
more importantly, the weak fire-resistivity, block their
materials. Characteristics of the water-introduced pores
applications in real constructions[3]. Although some
were studied and compared with those obtained from
newly-developed organic heat insulating building
the traditional AEA. SEM images revealed that a more
materials have been explored, their performance is far
homogenous pore size distribution was acquired when
from satisfactory and many fire accidents occur due to the
water other than the traditional AEA was used. A superior
use of the organic heat-insulating building materials[4].
thermal insulating property was shown in the water-
The other type of heat-insulating building material used is
induced porous system: reductions of 58% and 50% of
the in-organic material, and it possesses the advantages
the thermal-insulation efficiency (thermal conductivity
such as the excellent fire-resistance and anti-aging
coefficient per volume of porosity) were seen in the w/c
behaviors. However, the main drawback of this type of
ratio=0.9 and 0.4% AEA-added samples to that of the w/c
insulating material, mainly cementitious material, when
ratio=0.5 paste sample, respectively. In addition, porous
compared to the organic materials is the relatively high
cement pastes of higher w/c ratio showed a superior
heat conductivity, which has been regarded as the key
compressive strength to those obtained from the AEA-
property of this type of material. To handle this problem,
added samples. These advantages were made good use
techniques such as using lightweight aggregate, air-
of to prepare the thermal insulating lightweight aggregate
entraining agent, were used to increase the porosity
concrete, and a better performance of the heat insulation
of the cementitious material, as well as to lower the
was acquired.
density, so as to increase the heat insulation behavior
Key words: water-induced pores, thermal insulation, of the resulting materials. The key issue of producing
lightweight aggregate concrete. cementitious material with good performance in heat
insulation and strength is to coordinate relationship
Introduction between the porosity, the pore size distribution and the
mechanical property. The normally used technique of
It is known that great amount of energy has been improving the heat-insulation property of cementitious
consumed by constructions/buildings in the world, and materials is by adding lightweight aggregate into high-
this figure can be as high as 30% of the total consumed strength cementitious binder, and results showed that the
energy in China[1] annually. With the growth of the energy contradictory relationship between the heat-insulation
demand, as well as the growth of the awareness of and mechanical property can be partially resolved to
energy-saving among normal citizens, energy-saving produce a low-density and high-strength composite. But
in construction and building materials has been paid other results suggested that the key factor governing
great attention to. It has been reported that more than the mechanical property development was the original
80% of the newly-built constructions/buildings in China compressive strength of the lightweight aggregate,
have been grouped into “the high energy-consuming” as well as the ITZ property. So the enhancement of the
constructions/buildings. More than 95% constructions/ mechanical property of the base matrix contributed little
buildings are those “high energy-consuming” ones, which to that of the entire body[6-7], and a high w/c ratio of the
means that constructions/buildings consume more than cementitious material has been proposed.
35% of the energy built in 1980[2].
In this work, on studying the effects of high w/c ratio on the
One of the mostly used techniques of lowing consuming mechanical and heat-insulating property of cement paste,
energy in constructions and buildings is to use heat- the performances were compared with those obtained
insulating building materials, including the organic from traditional air-entraining agent.

Organised by
India Chapter of American Concrete Institute 653
Technical Papers

Table 1
Physiochemical properties of cement

SiO2 Al2O3 Fe2O3 SO3 CaO MgO LOI Density, g/cm3 Surface area, m2/kg 28d Compressive strength/MPa

21.1 4.7 3.5 3.3 62.9 2.8 1.1 3.1 390 50.1

Experiments w/c ratio of 0.5 was used and SDS dosage of 0%, 0.2%
and 0.4% were used. The mechanical properties of pastes
Materials and proportions were evaluated using the 70.7x70.7x70.7 mm3 molds.
In this work, ordinary Portland cement OPC42.5 was For the concrete study, cement of 400kg/m3 and 450kg/
used and its physiochemical properties are listed in m3, w/c ratio of 0.3, 0.5 and 0.7, sand to aggregate volume
Table 1. Air-entraining agent (AEA), of Sodium dodecyl ratios of 0.45 and 0.60 were used. Mix proportions of
sulfonate, SDS, and thickening agent (TA) of hydroxypropyl concrete are listed in Table 3.
methylcellulose (HPMC) with the molecular weight of 200
000 were used to adjust the workability of the mixture of
high w/c ratio. Polycarboxylate water reducer was used to Methods
a mixture of relatively low w/c ratio. The preparation techniques of the pastes and concretes
followed the processes described in Chinese standards.
To prepare the lightweight concrete, porcelain granule
When a high w/c ratio was used, half water was added into
was used as coarse aggregate and its properties are
the mixture for the achievement of a homogeneous mixing
listed in Table 2. Crushed porcelain granule was used as
before another half was added. The fluidity of cement
fine aggregate for concrete mixing.
paste was measured according to Chinese Standard GB/
T8077-2000; the compressive strength at was measured
Table 2
Properties of porcelain granule
following JGJ70-1990; apparent dry density and absolute
density of cement paste were determined by following
water-absorption ratio Compressive JGJ51-2002 and GB/T8077-2000, respectively. The
Packing density thermal conductivity coefficients of the cement paste and
1-hour 1-day strength/MPA
concrete was measured using the thermal conductivity
389 13.7 18.9 1.35 measuring apparatus following GB/T10294-2008. Before
measurement, samples were completely dried at 85 oC.
Sample proportions and preparation
SEM (Quanta FEG-250) technique was used to observe the
In this study, paste and concrete samples were prepared. morphology of hardened cement pastes of different w/c
For the paste study, w/c ratios of 0.5, 0.7 and 0.9 were ratios.
used and 0%, 0.3% and 0.5% HPMC were used to avoid
bleeding. For cement paste with the addition of AEA, the

Table 3
Concrete mix proportions/kg/m3

No. Cement w/c ratio Water reducer TA Vsand/Vaggregate Fine aggregate Coarse aggregate  Dry density
1 400 0.3 0.4 / 0.45 294 304 1197

2 400 0.5 / 0.8 0.45 263 271 1029

3 400 0.7 / 2.2 0.45 231 239 1005

4 400 0.3 0.4 / 0.6 392 221 1213

5 400 0.5 / 0.8 0.6 350 197 1019

6 400 0.7 / 2.2 0.6 308 174 1027

7 450 0.3 0.45 / 0.45 282 291 1167

8 450 0.5 / 0.9 0.45 247 255 1048

9 450 0.7 / 2.48 0.45 211 218 1073

10 450 0.3 0.45 / 0.60 376 212 1306

11 450 0.5 / 0.9 0.60 329 185 1129

12 450 0.7 / 2.48 0.60 282 159 1060

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

654 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influences of w/c ratio on the properties of lightweight concrete

Results and Discussions the slump and fluidity of the concrete mixtures increase
with the increase of w/c ratio, and this easies their in-
Physical properties of the fresh and hardened high w/c situ placement. It was observed from the experiments
ratio cementitious materials that when HPMC was added, segregation of the concrete
The basic physical properties of cement pastes are mixture can be entirely mitigated.
listed in Table 4. Results show that the incorporation of
air-entraining agent and the use of high w/c ratio help to Compressive strength
acquire a good state of the mix. The main concern of the heat-insulating building
material is the contradictory relationship between the
Table 4 porosity and the compressive strength. The compressive
Physical properties of high w/c ratio cement paste [8] strength of cementitious materials of different pore-
introduction techniques were tested and compared, and
w/c ratio 0.5 0.7 0.9
the results are shown in Figs. 2 and 3. It shows in Fig. 2
AE/ TA/% 0 0.2AE 0.4AE 0.3TA 0.5TA that the compressive strength of hardened cement pastes
Bleeding little no no no no decreases with the increase of w/c ratio and the content
of AEA. When comparing the compressive strength of the
Dry apparent density/g/cm 3
0.92 1.19 0.99
AEA-added sample and the high w/c ratio sample that
Density/g/cm3 1.5 1.14 2.11 possessing the same range of density, it can be concluded
Porosity/% 28.8 46.1 56.4 43.6 52.8
that the latter has a higher compressive strength: the
compressive strength of the w/c ratio=0.7 and 0.9 paste
sample at 28 days are 36% and 58% higher than samples
Comparable apparent dry densities of the 0.2%/0.4% with 0.2% and 0.4% AEA. Moreover, a higher compressive
AEA-added and the 0.3%/0.5% TA-added are seen, which strength growth is seen in the high w/c ratio samples, and
can be helpful for the macro-scale property comparison these could be due to the microstructures of the paste
of them. samples.
The influences of w/c ratio on the properties of fresh The influences of high w/c ratio on the compressive
concrete mixtures are shown in Fig. 1. As expected that strength of hardened lightweight concrete are shown in

Fig. 1: Slump flow and fluidity of fresh concrete mixtures

Organised by
India Chapter of American Concrete Institute 655
Technical Papers

that of conventional AE-concrete, and this has be ascribed

to the variation of the pore structure. In this section, the
heat conductivity coefficients of hardened cement pates/
concrete were measured to reflect the feature of the
pores on the heat insulation property. For a quantitative
comparison, index of the heat conductivity/porosity of
cement paste was used. It can be seen in Fig. 4(a) that at
a comparable porosity, the reduction efficiency of water
induced pores on the heat conductivity coefficient is
superior than that caused by the air-entraining agent: it
shows in the w/c ratio=0.9 sample that a reduction of the
index of 58% to that of control is observed compared to the
value of 50% of the AE-added sample.
Fig. 2: Compressive strength of hardened cement paste[8]
The effect of w/c ratio on the heat conductivity of lightweight
Fig. 3. As expected, the compressive strength decreases concrete is shown in Fig. 4(b), and the heat conductivity
with the increase of w/c ratio. Generally the later age efficiency, i.e., the heat conductivity coefficient per unit
strength increment of higher w/c ratio sample is higher, weight of the lightweight concrete, is also shown. It can
which is consistent with the results shown in the paste be seen that both values decrease with the increase of w/c
sample. It also demonstrates in Fig. 3 that a comparable ratio, implying the increase of the heat insulating capability.
later age compressive strength of the w/c ratio=0.5 and The increase of the heat insulating efficiency of the high
0.7 samples at 28 days can be observed. w/c ratio cement concrete could be due to the higher heat
insulating capability of the water-induced pore system as
Heat insulation property indicated in the paste system. For a comparison study, the
heat insulating efficiency of samples investigated in this
As shown in the compressive strength section, the work was compared with those published, and the results
incorporation of pores into cementitious materials with are listed in Fig. 5.
water contributes to higher mechanical properties than

Fig. 3: Compressive strength of lightweight concrete

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

656 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influences of w/c ratio on the properties of lightweight concrete

Fig. 4: Heat conductivity of cement paste/concrete

Fig. 6: Morphology of hardened cement pastes[8]

It can be seen in Fig. 5 that superior heat insulation

efficiency is obtained in the present work, i.e., lower heat
Fig. 5: Relationship between the bulk density/compressive insulating coefficients were acquired at the comparable
strength and the heat insulating coefficient of lightweight density and 28-day compressive strength levels. Thus
aggregate concrete[9-13] it can be concluded that a high w/c ratio will benefit the

Organised by
India Chapter of American Concrete Institute 657
Technical Papers

improvement of the heat insulation performance of Acknowledgements

lightweight cement-based composites.
Fund on New Board Materials from the Housing and
Urban-Rural Construction Bureau of Shandong Province
Morphology and Program for Scientific Research Innovation Team in
The morphology of hardened cement paste of high w/c Colleges and Universities of Shandong Province are greatly
ratio was captured and compared with those having a acknowledged. We also thank financial support from
comparable apparent dry density but achieved by adding National High Technology Research and Development
AE agent, and the results are shown in Fig. 6. Program (“863 Program”, 2015AA034701).
It shows in Fig. 6 that the microstructure of cement paste
becomes more porous with the addition of SDS and the References
increase of w/c ratio. Comparatively speaking, pores 1. Wang Xi. Study on the residential building energy saving and
consumption reduction level evaluation index system of Jilin
introduced by SDS are the bigger (with the most probable Province, 2012, thesis, University of Jilin, Jilin. (in Chinese)
pore size of about 100 microns) and isolated ones, but in
2. Pinshi Zhang. Development of the energy- and land-saving society.
the high w/c ratio paste the pores are more homogeneous Building Industry, 2006(6): 96-98.
distributed. In regard of the hydration products connection
3. Haitao Bo. Durability and its evaluation of the heat-insulation system
degree, it can be observed that a more tightly compaction of the external wall. 2009, thesis, Huazhong University of Science
morphology is observed in the SDS-added pastes. A & Technology, Wuhan.
more isolated morphology, which is separated by the 4. Zongkun Yang, Yunan Yang, Xiaosheng Hua. Misunderstanding of
homogeneously distributed pores, is seen in the high w/c fire-prevention reflected from from the fire-disaster of the CCTV
ratio sample, and this can be more clearly seen in the w/c tower. Building Energy Saving. 2009, 37(5):1-7.
ratio=0.9 paste sample. The difference in the pore features 5. Wei Chen, Synergic action of lightweight aggregate-matrix on
could be accounted for the variance of the properties of performance of concrete. 2013, Ph.D Thesis, Chongqing University,
Chongqing.
the samples.
6. Wei Chen, Jueshi Qian, Jun Liu, et al. Preparation and properties
of lightweight aggregate concrete by using higher water-to-cement
Conclusions ratio. Journal of Building materials, 2014, 17(2): 298-302, 335.

In this work, lightweight cementitious materials were 7. Fazhou Wang, Preparation and application of high performance
lightweight aggregate concrete. 2003, Ph.D thesis, Wuhan University
prepared by using high w/c ratios, and superior heat of Science & Technology, Wuhan.
insulating performance was obtained when compared
8. Zhaoheng Guo, Maoqiang Fu, Pengkun Hou, et al., Characteristics
with traditional lightweight cementitious materials of water-induced pores and their influences on the properties of
prepared by using air-entraining agent, and the following cementitious materials. Non-metallic materials, in press.
conclusions could be drawn: 9. Weixin Hu, Wei Sun, Honggen Qin.Development and application of
high efficient light concrete. 2012, Concrete, 273: 1-3.
1) At a comparable bulk density, a higher compressive
strength of cementitious material with high w/c ratio 10. Handong Yan, Xiufeng Chen. Study on effects of waste clay brick
recycled aggregate on concrete properties. Sichuan Building
can be acquired when compared to the air-entraining Science. 2009, 35(5): 179-182.
agent-added cementitious material;
11. Min Zhou. Experimental study on preparation of light-weight and
2) A better performance of heat insulating can be high-strength concrete made by kaolin tailings-gangue-fly ash
haydite. New Building Materials. 2014, 9, 67-70.
achieved by the high w/c ratio technique;
12. Jinzhong Fan. Performance and utilization of thermal insulation
3) A more homogeneous micro-structure is obtained of refractory ceramisite concrete and products. Brick and Tile. 2010.
the high w/c ratio sample when compared with air- 10, 41-44.
entraining agent-added cementitious material; 13. Xuan Wang, Lei Wang. Shale ceramsite concrete mix design and
performance experimental study. Sichuan Building Science. 2012,
4) Lightweight aggregate concrete with a superior 38(5): 16-17
performance of heat insulation can be obtained using
the high w/c ratio technique.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

658 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influences of w/c ratio on the properties of lightweight concrete

Pengkun Hou, Ph.D

Lecturer at School of Materials Science & Engineering, University of Jinan, Shandong, China
Add.: No.336 Nanxinzhuang West Rd., Jinan City, Shandong, China
Phone: +86 531 82767655
One of the main research interests of Dr. Hou is to make cement-based building materials greener through
the improvement of the durability by the adoption of nanotechnology. Synthesis of nanomaterials of novel
functions, the influences and mechanisms of nanoparticles on the physical and chemical properties of
cementitious materials, the interaction of nanomaterials with cementitious materials, and the real usage of
nanotechnology of producing green cementitious materials are his major work.

Organised by
India Chapter of American Concrete Institute 659
Technical Papers

Thermal Strain and Strength Development of Cement-based Materials

at Cryogenic Temperatures
JIANG Zhengwu, DENG Zilong, LI Wenting
Key Laboratory of Advanced Civil Engineering Materials of Ministry of Education, Tongji University, Shanghai 200092, China

Abstract to monitor the strain of concrete tank and to investigate

cryogenic properties of concrete.
The properties of cement-based materials at cryogenic
temperatures are quite different from those at room However, it is difficult to measure the strain of concrete, since
temperature or low temperature. The compressive the resistance strain gauge fail at cryogenic temperatures.
strength, flexural strength and thermal strain of Fiber bragg grating sensors has been widely used in concrete
cement paste, mortars and fiber mortars modified with structures to monitor the strain at ambient and elevated
polypropylene fiber and cellulose fiber at cryogenic temperatures (Lin, Chen et al., 2004; Slowik, Schlattner
temperature were measured. A new method was et al., 2004; Wong, Childs et al., 2007). In addition, optical
developed to measure the thermal strain of mortars at fibers are physically and chemically stable with concrete at
cryogenic temperatures using fiber Bragg grating (FBG) cryogenic temperatures. Thus, it is possible to develop FBG
sensors. The influences of polypropylene fibers and sensor to measure thermal strain of concrete.
cellulose fibers on the cryogenic strength of mortars
A great increase in both compressive and tensile strength
were discussed. The results show that FBG sensors can
of concrete at cryogenic temperatures has been reported
be used to measure the strain of cement paste. Cement
(Miura, 1989; Rostasy, 1979; Jiang and Li, 2010). Miura
paste expands from -30 °C to -50 °C in the cooling period
(Miura, 1989) indicated that both compressive strength and
and contracts from -50 °C to -7 °C in the heating period
tensile strength increase with temperature decreasing
because of the volume change of pore water freezing. The
and stay constant after -120 °C. While cryogenic strength
compressive and flexural strength increase by 1~2 times
development of fiber modified cement based materials
from room temperatures to cryogenic temperatures,
and the effects of fibers on cryogenic concrete are still
growing rapidly from 20 °C to -80 °C and then reaching
unknown. Consequently, thermal strain and cryogenic
a limit at the temperature range of -80~-140°C. The
strength of cement-based materials are investigated in
addition of fibers in mortars does not work in improving
this paper. And a new method to measure the strain of
the cryogenic strength alone. Water content is more
concrete by FBG sensors is utilized in cryogenic conditions.
important for cryogenic strength.
Keywords: Cryogenic, freeze-thaw, thermal strain, fiber
Bragg grating sensors, mortars.
Theoretical Principles of FBG Sensors
Ambient Temperature
Introduction It is well known that the wavelength, λb of fiber grating, is
Concrete has been the most widely used civil engineering given as
materials for about a hundred years and has been used to m b = 2nd ......................................................................(1)
construct buildings under a variety of severe conditions.
Natural gas are mainly stored in a liquid state at -165 °C where n is the refractive index of the core and d is the
in tanks made of concrete and 9% nickel steel. It was the grating period. A change in temperature and strain causes
steel wall inside that acts as the primary containment the shift in the Bragg reflecting wavelength. The refractive
tank before. By 2004, it has been reported that nine tanks index of the core changes due to the themo-optic effect.
using post-tensioned reinforced concrete as the primary Therefore, the shift in bragg wave length is given by
dm Q
= 1 - peV l + n
containment of LNG have been constructed (Jackson et dl dn
al., 2004; Krstulovic-Opara, 2007). Since the direct use of m ...............................................(2)
concrete for primary containment of LNG will substantially where pe is the photoelastic constant of fiber, is the
reduce the construction costs and period, there may be an
increasing impetus for such tanks. The LNG tank is a large fractional length change of FBG and is the fractional
structure which has a high requirement on safety. And the index change. At ambient temperature range, both
large temperature gradient may lead to large thermal and are proportional to temperature change, so the
strain and stress in concrete. Therefore, it is significant equation above can be written as
2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on
660 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Strain and Strength Development of Cement-based Materials at Cryogenic Temperatures

= !Q 1 - p e V a + p$ TT .............................................(3) solid content of 30%, was added. Length of cellulose fiber

dm
m is about 6mm. Table 1 shows different types of cement
where α is the coefficient of thermal expansion of FBG and based materials prepared and the mixture proportions.
ξ is the thermo-optic coefficient of the fiber. At ambient
temperature, (1 − pe) α is one order smaller than ξ, so the Sample Preparation
equation can be further changed to Cement paste and mortar samples, P, M and MP (see
dm Table 1), were casted to characterize the thermal strain
= pTT ..................................................................(4) by FBG sensors. Firstly, A core cylinder with diameter of
m
16mm was casted. Then, two FBG strain sensors were
This is the fundamental equation the FBG temperature stuck parallel to the axis on both sides of the core cylinder
sensors. The temperature change is proportional to the with a thin layer of cryogenic adhesives. Finally, this core
wavelength shift. For FBG strain sensors, the shift in cylinder was embedded in the center of thermal strain
Bragg wavelength is affected by both temperature and testing cylinder samples with a diameter of 75mm. The
strain, which can be given by mixture proportion of core cylinder and testing cylinder
dm Q
= 1 - p e V f + pTT ................................................(5)
were the same. FBG temperature sensors were embedded
m in the center of another testing cylinder together with T
type thermocouple, to eliminating the effects of FBG
where ε is the strain measured.
sensor and thermocouple on the thermal strain. Each
thermocouple was recalibrated at the temperature range
Cryogenic Temperature
of -1°C~80°C.
At cryogenic temperature, however, both α and ξ are
temperature dependent. Equation (3) cannot be used
directly. Equation (2) can be written in the form of

= "Q 1 - p e V l dT + n dT % ....................................(6)
dm 1 dl 1 dn
m
It has been reported that the fiber bragg grating temperature
response is nonlinear at cryogenic temperature. Reid (a) The core cylinder and FBG strain (b) The testing cylinder with thermocouple
sensors and FBG temperature sensor embedded
and Ozcan (Reid 1998) approximate this temperature in the center
dependence with a third order polynomial. For FBG strain
sensor, the wavelength shift resulted from strain can be Fig. 1: The core cylinder and testing cylinder
calculated by subtracting the wavelength shift resulted
from temperature. Therefore, in order to apply FBG sensors M, MC and MP were casted to the size of
in concrete at cryogenic temperatures, it is important to 40mm*40mm*160mm to study the strength development
correlate the wavelength with temperature first. In this at cryogenic temperatures. All samples were cured in
paper, the FBG temperature sensors are calibrated by water for 28 days. Water content of hardened mortars
thermocouples and the relationship equation between were tested after curing and the results were included in
wavelength and temperature has been established. table 1. Two thermocouples were embedded respectively
on the surface and in the center.
Experiment Program
Procedures
Raw Materials and Mixture Proportion The cryogenic freeze thaw experiment is completed in
Cement used is Xiaoyetian ordinary Portland cement a cryogenic refrigerator (see figure 2a), in which the
P.O.52.5. Sand used is natural sand. And the fineness temperature can be controlled. Liquid nitrogen is sprayed
modulus is 2.49. Polycarboxylate superplasticizer with a to cool the temperature down to -170 °C and 3 resistance

Table 1
Mixture proportion of mortars and water content after water curing for 28 days (by mass)

Cellulous Polypropylene Super Water

Type Cement Water Sand
fiber fiber plasticizer content

Cement Paste (P) 1 0.33 - - - - -

Mortar (M) 1 0.4 2.5 - - - 6.0%

Mortar containing cellulous fibers (MC) 1 0.4 2.5 0.5% - 1.09% 9.3%

Mortar containing polypropylene fibers (MP) 1 0.4 2.5 - 0.5% 0.87% 10.3%

Organised by
India Chapter of American Concrete Institute 661
Technical Papers

(a) Cryogenic refrigerator (b) Thermal insulation box

Fig. 2: Cryogenic refrigerator and thermal insulation box

heaters are used to raise the temperature to 15 °C. A

magnetic valve is used to control the amount of liquid
nitrogen sprayed and make sure practical temperature Fig. 3: Testing results of FBG temperature sensor and
follow the temperature program we set. thermocouple
The cooling/heating rate of cryogenic refrigerator is set
at 0.5°C/min. samples were kept for 30 minutes at 15°C order polynomial can also be used in this experimental
before cooling down, maintained 60 minutes at -170 °C temperature range to calculate the temperature
before heating and maintained for 30 minutes at 15°C at (see figure 5). Same results can be observed in other
last. Thus, a whole freeze thaw cycle takes 490 minutes. three different samples MP, M1 and M2 in figure 6. The
wavelength can be given by
The strength testing samples were cool down to the .......................................................(7)
m b = aT 2 + bT + m b0
set temperature at a rate of 1°C/min and maintained
30 minutes after the surface temperature and center where a and b are constant and are related to material
temperature reached the same. Then samples were constitute of FBG sensor. λb0 is the Bragg wavelength at
transported in a thermal insulation box (see figure 2b) 0 °C. For the FBG sensors used in this experiment, a is
before strength testing with ice at the set temperature 0.00926 and b is 1.71×10-5. Therefore, it can be concluded
maintaining temperature. All strength tests were that FBG temperature sensor can be used for cryogenic
completed in 3 minutes. temperature monitoring after calibration.

Results and Discussions

Temperature Measurement
The testing results of thermocouple are used to compare
and calibrate those of FBG temperature sensor. Figure 3
indicates a same tendency of results from Thermocouple
and FBG temperature sensor. Above -120 °C, the
wavelength is proportional to temperature measured
by thermocouple. As the temperature goes down, the
absolute value of slope in wavelength curve become
smaller, which means that FBG temperature sensor (a) The 1st cycle
become less sensitive at cryogenic temperature. This
result can be shown in a coordinate system of Bragg
wavelength versus temperature (see figure.4a). In figure
4a, the cooling curve and heating curve overlap almost
completely. After 5 cooling and heating cycles (See figure
4b), the results coincide with each cycle, which suggest
a strong relationship between the bragg wavelength
and temperature and a good repeatability. Although the
sensitivity decrease at cryogenic temperature, it is still
sensitive enough to characterize the temperature above
-170 °C.
(b) after 5 cycles
Both second-order (R2=99.994%) and third-order
polynomial (R2=99.994%) are fitted to the results. Second- Fig. 4: Wavelength of FBG temperature sensor versus
temperature after 5 cycles

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

662 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Strain and Strength Development of Cement-based Materials at Cryogenic Temperatures

(a) Thermal strain versus time

Fig. 5: Second-order and third order polynomial fitting

(b) Thermal strain versus temperature

Fig. 7: Thermal strain of cement paste at cryogenic temperature

by FBG strain sensor

flow or a much higher pressure is required to push water

Fig. 6: Wavelength of different FBG temperature sensors after to faraway pores. In this case, the growth of ice will lead to
5 cooling and heating cycles the pressure increasing within pore water between pore
wall and ice, which then slow down the contraction and
Thermal Strain Measurement finally cause the expansion of pores. At around -50 °C,
By substracting the wavelength shift obtained by FBG almost all the water has frozen the volume of the sample
temperature sensor, wavelength shift caused by thermal continuously contracts.
strain, as well as thermal strain, can be determined by In the thawing period of figure 7b, samples expanses
equation (5). Figure 7 shows the thermal strain in the first proportional to temperature until -50 °C. Above -50
freeze thaw cycle. Thermal strain of cooling and heating °C, pore water begins to melt, which result in sample
period overlap below -50 °C. Above this temperature, expanding slowly and contracting finally. Above -7 °C, the
the thermal straiin curve of in the cooling period is quite sample resumes to expand proportional to temperature.
different from that in the heating period. And there is It indicates that ice melt after -7°C do not impede the
residual strain within cement paste after a freeze thaw expansion of the sample. As temperature goes back to
cycle. ambient temperature, residual strain about 225 με exists
As shown in figure 7b, in the cooling period, cement paste in the sample. The reason might be expanded pore cannot
sample contracts slowly or even stops contracting from resume to the original status.
-25 °C to -35 °C and expands from -35 °C to – 50 °C. This The difference of thermal strain curves in cooling and
is because the freezing of pore water. As temperature heating period is resulted from the difference in freezing
goes down, water freezes in larger pores first and the and melting point in pores at same temperature. The
volume expansion of ice will push water around to flow freezing point in pores is highly related to the curvature
into smaller pores around. Above -25 °C, ice grows freely of ice/liquid interface. And the curvature of this interface
in the pore and does not influence the thermal shrinkage in a cylinder pore in melting period would be half as large
of the hardened cement paste. Below -25 °C, most of pore as that in cooling period, resulting in the hysteresis of
water has frozen and there is no enough space for water to freezing point. For pure pore water, Brun et al(Brun,

Organised by
India Chapter of American Concrete Institute 663
Technical Papers

Lallemand et al. 1977). gives the relationships equations adhesives and the mortar matrix lead to non-symmetric
between the freezing temperature depression and the stress in FBG, which would also influence the period of
pore radius. The radius can be calculated by following Bragg grating. Other FBG sensor embedded methods
equations. Based on these equations, thermoporometry should be investigated in the future research.
is developed to characterize the pore size distribution of
Some researchers indicates that ice crystals transformed
porous materials. (Brun, Lallemand et al., 1977; Sun and
from hexagonal system to orthorhombic system around
Scherer, 2010; Jiang, Li et al., 2013)
-120 °C, which may lead to 20% decreasing in ice volume.
64.67 They use the volume reduction to explain the decrease
R p = - TT + 0.57 (freezing)(0 2 TT 2 -40 o C) ......(8)
in strength around -120 °C. Hansen et al. observe low
In the freezing period, water in pores with radius of 3.2 nm temperature (around -70 °C) water freezing with NMR
starts to freeze at -25 °C. And in the heating period, water and indicate that part of the unfrozen water are also able
in 5.3 nm pores initial to melt at -7 °C. For the pore system to freeze at low temperature. These results cannot be
in this cement paste, it seems that sample continues observed in the thermal strain curves obviously. Or the
contracting until the freeze of pore water in 3.2nm pores crystal phase change and low temperature water freezing
in the cooling period; and continues expanding until the has no influence on the thermal strain of cement-based
melting of pore ice in 5.3nm pores in the heating period materials.
at a heating rate of 0.5°C/min. It should be noted that the
freezing point for pore solutions depress further. And the Cryogenic Strength of Fiber Modified Mortars
expansion in cooling period or contraction in the heating
period is an accumulating result of water freezing or ice Figure 9 shows the strength development of M, MP and MC
melting. at cryogenic temperatures. Both compressive and flexural
strength increase with decreasing temperature firstly and
Figure 8 shows thermal strain of cement paste at cryogenic then reaches a limit at temperature range of -80 ~-140 °C.
temperature by FBG strain sensor after 5 cooling and
heating cycles. The residual strains are clearly presented
by the thermal strain value at -170 °C. In the first two
cycles, residual strain increase obviously and then
reaches a limit. The similar tendency in thermal strain
curves of cement-based materials by dilatometer can be
found in the work of Rostary et al(Rostasy, Schneider et
al. 1979).

(a) Compressive strength

Fig. 8: Thermal strain of cement paste at cryogenic temperature

by FBG strain sensor after 5 cooling and heating cycles

Therefore, the FBG strain sensor can be used to

characterize the strain of cement paste. However,
for Mortar samples, multiple peaks in the reflection
spectrum have been observed at cryogenic temperature
range (-130~-170 °C). In this case, thermal strain in this (b) Flexural strength

temperature range cannot be obtained. This might be Fig. 9: Compressive and flexural strength of different mortars
because the thermal incompatible of the cryogenic at cryogenic temperatures

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

664 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Thermal Strain and Strength Development of Cement-based Materials at Cryogenic Temperatures

The difference is the flexural strength reaches the limit Conclusions

at -80~-100°C and the compressive strength reaches the
Properties, such as compressive strength, flexural
limit at -120~-140 °C. Compared with strength at room
strength and thermal strain of cement paste and
temperature, the compressive strength of M, MP and MC
mortars containing fibers at cryogenic temperatures, are
increases by 62%, 100% and 176% respectively at -170 °C.
investigated. FBG strain sensor is developed to measure
The flexural strength increases by 188%, 173% and 217%
the thermal strain of cement based materials. It can be
respectively. The water content of M, MP and MC, included
concluded that
in table 1, are 6%, 9.3% and 10.3% respectively. Thus,
higher water content leads to higher strength increase 1. FBG temperature sensors are less sensitive at
at cryogenic temperatures. This result is the same with cryogenic temperatures than those at ambient sensors,
other researchers. but acceptable in the experimental temperature range.
The increasing of strength are contributed to the freezing 2. FBG strain sensors can be used to characterize the
of pore water. While most pore water has frozen at -50 strain of cement paste at cryogenic temperatures.
°C, which can also be indicated in figure 7b, The strength A new embedded method should be developed
increasing below -50 °C might be explained by the to measure the strain of mortars and concrete at
freezing of ‘unfrozen water’, which is a layer of water cryogenic temperatures.
existing between pore wall and ice crystals. The thickness 3. Cement paste expands from -30 °C to -50 °C in the
of this layer is about 0.8 nm for cement based materials cooling period and contracts from -50 °C to -7 °C in
(Brun, Andre et al., 1977; Sun and Scherer, 2010). Hansen the thawing period because of the volume change of
et al (Hansen, Stocker et al., 1996) found that this unfrozen pore water freezing.
water freezes around -73 °C and accounts for more than 4. A positive residual strain exists within cement paste
65% of water content in the porous materials tested. after a cryogenic freeze thaw cycle. The residual strain
These results can well explain the reason of the strength increase with cycles firstly and then reaches a limit.
increasing between -60 and -100 °C. 5. Both compressive strength and flexural strength
Figure 10 shows the compressive strength to flexural of mortars increase by 1~2 times as temperature
strength ratio of mortars at cryogenic temperature. It can decreasing from -20 °C to -170 °C.
be indicated that flexural strength increase much more 6. The addition of fibers in mortars does not work in
quickly than compressive strength before the strength improving the cryogenic strength alone. Water content
limits. This is because both crushing strength and adhesive is more important for cryogenic strength.
strength of ice increase as temperature decreases. In
addition, the adhesive strength under tension is much
greater than that under shear (Chatterji 1999). The flexural Acknowledgement
strength of MP and MC increase slowly compared to that The authors gratefully acknowledge the financial
of M, although the water content is higher. The reason is supports provided by National Basic Research Program
that the adhesive bond between ice and matrix is much of China (973 Program: 2011CB013805), National Natural
stronger than that between fiber and matrix. It seems that Science Foundation of China (51478348, 51278360,
the existing of fibers may weak the adhesive bond of ice 51308407), National Key Project of Scientific and Technical
and cement paste. Supporting Programs of China(No. 2014BAL03B02), the
Specialized Research Fund for the Doctoral Program
of Higher Education of China (No. 20130072110047),
Key project of the Shanghai Committee of Science and
Technology(No.14DZ1202302) and the Fundamental
Research Funds for the Central Universities.

References
1. Brun, M., Lallemand, A., Quinson, J.F. and Eyraud C., 1977. A new
method for the simultaneous determination of the size and shape
of pores: the thermoporometry. Thermochimica Acta, 21(1): 59-88.
2. Chatterji, S., 1999. Aspects of freezing process in porous material-
water system: Part 2. Freezing and properties of frozen porous
materials. Cement and Concrete Research, 29(5): 781-784.
3. Chatterji, S. 1999. Aspects of the freezing process in a porous
material–water system: part 1. Freezing and the properties of water
and ice. Cement and concrete research, 29(4): 627-630.
4. Hansen, E. W., Stocker M., and Schimidt, R., 1996. Low temperature
phase transition of water confined in mesopores probes by NMR.
The Journal of Physical Chemistry, 100(6): 2195-2200.
Fig. 10: Compressive strength to flexural strength ratio

Organised by
India Chapter of American Concrete Institute 665
Technical Papers

5. Jackson, G., Powell, J.,Vucinic, K. and Harwood D., 2004. Delivering 10. Miura, T., 1989. The properties of concrete at very low temperatures.
LNG tanks more quickly using unlined concrete for primary Materials and Structures, 22(4): 243-254.
containment. PO-10, LNG 14.
11. Reid, M. B., 1998. Temperature dependence of fiber optic Bragg
6. Jiang, Z. and Li, X., 2010. Experiment study on mechanical properties gratings at low temperatures. Optical Engineering, 37(1): 237-240.
of mortars at ultra low temperature. Journal of the Chinese Ceramic
12. Rostasy, F., Schneider, U., and Wiedemann G., 1979. Behaviour of
Society, 38(4): 602-607. (In Chinese)
mortar and concrete at extremely low temperatures. Cement and
7. Jiang, Z., Li, W., Deng, Z. and Yan, Z., 2013. Experimental Concrete Research, 9(3): 365-376.
investigation of the factors affecting accuracy and resolution of
13. Slowik, V., Schlattner, E., and Klink T., 2004. Experimental
the pore structure of cement-based materials by thermoporometry.
investigation into early age shrinkage of cement paste by using fibre
Journal of Zhejiang University SCIENCE A, 14(10): 720-730.
Bragg gratings. Cement and Concrete Composites, 26(5): 473-479.
8. Krstulovic-Opara, N., 2007. Liquefied natural gas storage: material
14. Sun, Z., Scherer, G. W., 2010. Pore size and shape in mortar by
behavior of concrete at cryogenic temperatures. ACI Materials
thermoporometry. Cement and Concrete Research, 40(5): 740-751.
Journal, 297-306.
15. Wong, A. C., Childs, P.A., Berndt, R., Macken, T., Peng, G.D. and
9. Lin, Y. B.,.Chen J.C., Chang K.C., Chan Y.W. and Wang L.A., 2004.
Gowripalan, N., 2007. Simultaneous measurement of shrinkage
The utilization of fiber Bragg grating sensors to monitor high
and temperature of reactive powder concrete at early-age using
performance concrete at elevated temperature. Smart materials
fibre Bragg grating sensors. Cement and Concrete Composites,
and structures, 13(4): 784.
29(6): 490-497.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

666 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating Different Mineral Admixtures

Microstructures and Elastic Properties of Interfacial Transition Zones

in Mortars Incorporating Different Mineral Admixtures
Yueyi Gao, Chuanlin Hu, Yamei Zhang*, Zongjin Li, Jinlong Pan
Jiangsu Key Laboratory of Construction Materials, Southeast University
School of Civil Engineering, Southeast University
Jiangsu Academy of Safety Science and Technology
Department of Civil and Environmental Engineering, the Hong Kong University of Science and Technology

Abstract fraction and preferential orientation are more and more

significant between ITZ and bulk cement paste matrix[8].
Interfacial transition zone (ITZ) is a key phase to the elastic
The characteristics of ITZ considering the influences of
behavior and strength of mortar within the framework of
aggregate size and other factors have been investigated
micromechanics. Some researchers have tried various
experimentally by some researchers[8,12-­16]. However,
microstructural assessment methods to evaluate the
the research on the characterization of ITZ based on
effective thickness of the ITZ. However, little attention was
microscopic mechanical properties is very limited[17].
paid to the method of determining the effective thickness
of the ITZ based on the microstructural features which are The effective thickness of ITZ is the key factor in
directly related to the elastic properties. In this study, the determination of the volume fraction of ITZ[18], and
ITZs in mortars, with and without the addition of mineral furthermore the macroscopic performances of
admixtures, were analyzed by the grid­nanoindentation mortar and concrete. Previously, some researchers
coupled with scanning electron microscope (SEM) image have conducted various microstructure assessments
analysis. In mortar at high water­to­cement ratio, it is found to evaluate the effective thickness of the ITZ, and
that the ITZ thickness decreases with the hydration age consensually accepted that the thickness of the ITZ is 15­
from 3 days to 28 days, while the indentation modulus ratio 50 μm depending on many factors[19]. Scrivener found that
of ITZ and cement matrix increases from 55.3% to 77.9%. the packing of cement grains resulted in microstructure
In contrast to mortar at high water­to­cement ratio, the gradients extending some 50 μm from the interface
effective thickness of the ITZ in the mortar at low water­ into the cement paste, as the amount of anhydrous
to­cement ratio vary little with hydration age. It is revealed cement increased with the distance from the interface[10];
that porosity is the dominant microstructure influencing Zimbelmann suggested that the thickness of ITZ ranged
the elastic properties of the ITZ in the mortar. Microscopic from 10μm to 30 μm according to the porosity variation[20];
characterization shows that the ITZs between the sands Ollivier et al. suggested that the ITZ ranged from 15μm
and the pastes disappeared at late age when mineral to 20 μm according to the variations of ettringite or CH
admixtures were incorporated. concentrations[8]. Thus, the effective thickness of the ITZ
varies with the microstructural features being concerned.
Keywords: Mortar; Mineral admixtures; Elastic
In order to calculate the elastic properties of mortar
properties; Microstructure; Interfacial transition zone.
more accurately, it is essential to determine the effective
thickness of ITZ based on the microstructural features
Introduction which are directly related to the elastic properties.
Mortar and concrete are usually modeled as composites in Besides, Mineral admixtures (e.g., silica fume, fly ash and
studying their mechanical properties and durability, where slag) have been widely used to partially replace Portland
a three­phase model is prevalent[1­-7]. The three phases cement in normal and high­strength mortar or concrete.
referred to cement paste matrix, aggregate and interfacial Zhan et al.[8, 21] showed that ultrafine active admixtures
transition zone (ITZ), respectively. In a broad sense, ITZ can could improve the ITZ through their positive influences
be defined as a zone where its microstructure is different on packing density and hydration process. Paulon et al.[22]
from the surrounding bulk cement paste, for mortar and reported experimental results that the incorporation
concrete in fresh, young or hardened state. It has been of silica fume and fly ash increased the strength of the
widely accepted that the microstructures in the vicinity of ITZ. However, the information about the modification of
the aggregates is due to the so­called “wall effect”[8-­11]. A mineral admixtures on the elastic properties of the ITZ
natural phenomenon in the ITZ around aggregates is size was rarely reported.
segregation which leads to different gradients at different
water to cement ratios and aggregate sizes[7, 9]. With the Of various experimental methods used to study the
development of hydration process, the differences in microstructure of the ITZ, SEM using back­scattered
chemical composition, porosity, portlandite crystal size, imaging is undoubtedly the most effective technique[9,­10,23].

Organised by
India Chapter of American Concrete Institute 667
Technical Papers

On the other hand, the application of nanoindentation

Table 2
makes it possible to quantitatively characterize the Chemical composition and specific surface area of cement,
mechanical properties of ITZ[14, 16­17, 24]. Although the fly ash and slag
research that effectively combined nanoindentation and
SEM has been successfully applied on the cement pastes Sand/
Name w/b FA % S% PCA %
binder
[25­29]
, it has not been used to study ITZ in depth due to the
difficulties: (1)the difficulty in mortar sample preparation M23 0.23 ­- - 1 1.2
for nanoindentation testing; and (2) precise positioning of
M35 0.35 -­ - 0.3 1.5
the two testing methods[30].
M53 0.53 ­- - - 2.0
In this paper, grid­ nanoindentation coupled with SEM
image analysis method was applied to investigate the M35FA 0.35 30 - 0.3 1.5
elastic properties of paste matrix and the ITZ in mortars M35S 0.35 - 50 0.3 1.5
with and without mineral admixtures. Influences of the
water­to­cement ratio, mineral admixtures and hydration In this table, columns 3 and 4 show the contents of the fly
age on the elastic properties and the microstructures of ash and slag as weight percentages of the replaced cement,
the ITZ were investigated. respectively. Column 2 and 6 show the water to binder mass
ratio and sand to binder mass ratio, respectively. Column 7
shows the dosage of PCA by the mass percentage of the
Materials and Experiments binder. In the name of the specimens, M represents mortar,
Specimen Preparation the following two numbers represents water­ to­
binder
ratio, S and FA represents the addition of slag and fly ash,
Three binders were used in the preparation of mortars in respectively.
this study. Table 1 shows the chemical composition and
specific surface area of each binder. River sands with the During the preparation of samples, the raw materials
size of 0­2.36mm and the fineness modulus of 2.5 were were firstly mixed and then cast in 40 mm × 40 mm × 160
used. Polycarboxylate type superplasticizer PCA with 30% mm steel moulds. The specimens were demolded after 24
solids content was used to adjust the workability of mortars hours, and then cured in a curing room (temperature: 20
at low water­to­cement ratios. For the sake of comparison, ± 2 °C, relative humidity: 95%) to an age of 3, 7, 28 and
five mortars were prepared in this study; details are given 180 days. After the specimens were moved out from the
in Table 2 . curing room, they were immediately immersed in alcohol
for the subsequent samples preparation procedure.
Table 1 The thin samples with the thickness of approximately 7
Chemical composition and specific surface area of cement,
fly ash and slag
mm were cut out from the middle of the specimens and
impregnated in epoxy till the epoxy got hardened. After
Cement Fly ash Slag that, samples were polished by using Lapping Films down
to 80μm, and then using oil and corundum paste down
Oxide mass fraction
to 1μm, finally using Flat Ion Milling System IM­3000 to
Calcium oxide, CaO 62.60 4.77 34.54 obtain a flat and smooth surface finish. At each step, the
polishing was done in absence of water, and an optical
Silicon dioxide, SiO2 21.35 54.88 28.15
microscope was used to check the effectiveness. At the
Aluminum oxide, Al2O3 4.67 26.89 16.00 end of each step, the samples were placed in an ultrasonic
Ferric oxide, Fe2O3 3.31 6.49 1.10
bath to remove the dust and corundum particles left on
the surface.
Magnesium oxide, MgO 3.08 1.31 6.00

Sulfur trioxide, SO3 2.25 1.16 0.32 Testing Methods

A NanoTest Vantage system was used to study the
Potassium oxide, K 2O 0.54 1.05 0.45
nanomechanical properties of the samples. The
Sodium oxide, Na2O 0.21 0.88 0.46 basic theory, methodology and application for the
Mass fraction Bogue composition
nanoindentation technique are reviewed and presented
in detail[31­34]. Usually, the aim of such testing is to extract
Alite 55.5 - - the elastic modulus (E) and hardness (H) of the sample
Belite 19.1 - - from continuous recordings of indenter load (P) and
penetration depth (h). According to the conventional
Aluminate 6.5 - - Oliver­Pharr procedure[31­32], once the contact area (A) is
Ferrite 10.1 - - determined from penetration depth, the hardness (H) and
the indentation modulus (Er) can be computed by
Specific surface (m /kg)
2
370 454 416

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

668 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating Different Mineral Admixtures

P ......................................................................(1)
H = Amax
dP
r ............................................................(2)
E r = b 2 dh
A

where Pmax is the peak load at the maximum of the curve, A

is contact area, and β is a correction factor (approximately
equal to 1.034 for the Berkovich indenter [1]). Eeff is the
effective elastic modulus, which is defined by
1 1 - n 2 1 - n 2i ................................................(3)
E eff = E + Ei

Which is related to elastic modulus (E) and Poisson’s ratio

(μ) of the indented material and those of indenter tip (Ei =
1141 GPa, μi = 0.07).
In this work, for each sample, two­c ycle nanoindentation
loading process with different maximum indentation
depths was performed in eight grids. Each grid consists
of 10 × 10 nanoindentation tests, and the interval distance
between two adjacent tests was 20 μm. The first cycle
(maximum indentation depth = 300nm) was used to
determine the mechanical parameters of material
microstructure, while the second cycle (maximum
indentation depth = 1500nm) was used to achieve a visible
footprint for further observation under SEM.
The indented areas of the sample were coated with carbon,
and then studied using JEOL JSM­6390 SEM. The qualitative
information about microstructure and phase composition
of the indented area was collected by secondary electron
and backscattered electron imaging analysis, while the
chemical composition of the indented microstructures
was determined by energy dispersive X­r ay spectroscopy.
The operating conditions of SEM were given as follows:
accelerating voltage, 15 kV; working distance, 15 mm; spot
size, 50 nm. The above experimental set­ups enable the
determination of the nanomechanical properties of ITZ
and the correlation between microstructure, composition
and mechanical properties.

Results and Discussion

Influence of Water­to­cement Ratio on ITZs Fig. 1: SEM­BSE images of the areas around sands in cement
mortars with different w/c after 3 days curing period: (a) w/
As Scrivener and Pratt[10] suggested, the common
c=0.23, (b) w/c=0.35, (c) w/c=0.53
constituents of mortar could be distinguished by the grey
scale of the BSE imaging. Fig. 1 presents the representative the age of 3 days. Thus, the results verify the existence
BSE images of area around sand in Portland cement of ITZ in mortar based on the distribution of porosity and
mortars with different water­to­cement ratio at the same unhydrated clinker. The results also show the influence of
hydration age of 3 days. It is observed that the cement water­to­cement ratio on ITZs at early age qualitatively.
paste of M23 is relatively dense, while the cement pastes
of M35 and M53 are loose with more pores. That is to Using the same method proposed by Diamond[35], the
say, the average porosity of mortar increases with the porosity profile with the distance from sand surface was
water­to­cement ratio significantly. More importantly, it is calculated quantitatively. Fig. 2 presents the porosity
observed that the microstructures around sands have analysis results of Portland cement mortars at the age of
significantly higher porosity and less unhydrated clinker 3 days, 7 days and 28days. It is revealed that the porosity
than the outer cement paste matrix in M35 and M53 at decreases with the distance from sand surface firstly, and
then remains stable.

Organised by
India Chapter of American Concrete Institute 669
Technical Papers

the ITZs thickness in Table 3, which are the calculating

results by observing continuous grey level variation with
the distance from sand surface. It is obvious in Table 3 that
the standard deviation of the result is relatively higher,
which is mainly due to the feature of the average grey
approach. This approach not only reflects the influence of
porosity, but also provides the distribution information of
unhydrated clinker and hydration products.

Table 3
The thickness of ITZ obtained by BSE image analysis (m)

3d 7d 28d

M23 M35 M53 M23 M35 M53 M23 M35 M53

Mean 15.3 25.5 30.7 14.3 24.3 29.7 13.9 16.7 21.1

Standard
3.4 4.0 3.2 3.1 3.3 3.0 3.2 3.3 3.7
deviation

Fig. 3 to Fig. 5 show the procedure of determining the

effective thickness of ITZ directly based on the variation
of mean indentation modulus with distance from the sand
surface. Fig. 3a and 3b present the representative BSE
images of two different regions after nanoindentation
testing. Since the location of each indent could be found
in the secondary electron image, the BSE images were
coupled with the secondary electron images to determine
the phase composition and the distance from sand surface
of each indent. Fig. 3c shows the indentation modulus and
phase distribution of each indent with distance from the
sand surface. Here, "OP" represents outer products, "IP"
represents inner products, "Interface" represents the
phase indented very close to the clinker particles, and
"P" represents pores. It is revealed that the random grid­
indents fell into structures of pore and outer hydration
product more frequently in the vicinity of the sand. Fig. 3d
shows the mean indentation modulus distribution with the
distance from sand surface. It is obvious that the fluctuation
of the average indentation modulus is in a relatively
low level within 30 m from the sands surface, and then
the mean value increases to fluctuate in a higher level.
Thus, according to the variation of indentation modulus,
the effective thickness of ITZ is supposed to be 30 m for
M53 at the age of 7 days. The reasons of the relatively
low indentation modulus in the ITZ can be expected and
explained by Fig. 3c, which presents a low emergence
probability of unhydrated clinker and a high probability of
appearance of pores in the ITZ. The outer products with
a relatively lower indentation modulus in the ITZ is also
Fig. 2: Average detectable porosity in the vicinity of the sands
affected by the relatively high porosity. Here, the thickness
after 3days (a), 7days (b), 28days (c) curing period
of ITZ determined by indentation modulus variation is
close to the value 29.7 m in Table 3, which was calculated
Fig. 2a shows that, at the age of 3 days, the inflection
by BSE image analysis. In order to get the effective elastic
points which represent the width of ITZ are 33 m, 27
properties of ITZ, the test points were classified into two
m, 18 m in M53, M35, M23, respectively. Fig. 2b and 2c
groups according to their distance from the sand surface.
present the results at 7 days and 28 days, accordingly.
The mean value of the testing points located within 5~30
These results of the inflection points are coordinated with

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

670 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating Different Mineral Admixtures

m distance were considered to represent the average Fig. 4 and Fig. 5 show the similar phenomenon as Fig.
indentation modulus of ITZ, whereas the results of the 3 in the aspects of phase distribution and average
testing points located within 30~100 m distance might modulus with distance from the sand surface. However,
provide an indication of elastic properties of the cement at hydration age of 7 days, the obtained thickness of
paste matrix. As the calculation results demonstrated, ITZ decreases while the calculated ratio of the average
the ratio between the average indentation modulus of ITZ indentation modulus of ITZ to that of cement paste matrix
and that of cement paste matrix was 75.4% for M53 at the increases with the water­to­cement ratio. It is revealed in
age of 7 days. Therefore, the ITZ cannot be ignored for Fig. 4d and Fig. 5d that the obtained thickness of ITZ of
conducting multi­scale prediction of the elastic properties M35 and M23 at 7 days is 25 m and 15 m, respectively.
of the cement mortar with high water­ to­
cement ratio. It can also be calculated that the ratio of the average
Otherwise the results may be seriously overestimated. indentation modulus of ITZ to that of cement paste matrix
of M35 and M23 is 75.6% and 80.3%, correspondingly. In
summary, the influence of water­to­cement ratio on ITZs is
mainly reflected in terms of porosity. That is the reason
why the ITZ thickness determined by BSE image analysis
agrees well with that obtained by examining the variation
of indentation modulus.

Fig. 3: Grid indentation modulus on ITZs of cement mortar (w/ Fig. 4: Grid indentation modulus on ITZs of cement mortar (w/
c=0.53) after 7days curing period:(a)(b) SEM­BSE images of the c=0.35) after 7days curing period:(a)(b) SEM­BSE images of the
indents after NI, (c) indentation modulus and phase distribution, indents after NI, (c) indentation modulus and phase distribution,
(d) mean indentation modulus distribution across ITZ (d) mean indentation modulus distribution across ITZ

Organised by
India Chapter of American Concrete Institute 671
Technical Papers

Fig. 6: Effect of hydration age on the porosity in the ITZ of

cement mortar: (a) w/c=0.53, (b) w/c=0.23

respectively. It is revealed that the maximum drop of the

ITZ thickness is 29.0% from 7 days to 28 days. Thus, there
is a corresponding relationship between the development
of porosity and variation of the ITZ thickness. Fig. 6b
shows that the porosity development of cement paste in
Fig. 5: Grid indentation modulus on ITZs of cement mortar (w/
M23 has the similar trend as that in M53, but the variation
c=0.23) after 7 days curing period:(a)(b) SEM­BSE images of the
indents after NI, (c) indentation modulus and phase distribution, is limited. It is observed that the development of ITZ
(d) mean indentation modulus distribution across ITZ thickness follows the same rule as porosity development
in M23. The thickness of ITZ decreases from 15.3 m to 13.9
Effect of Hydration Age on ITZs m from 3 days to 28 days. Based on the obtained rules of
ITZ development, it is supposed that the thickness of ITZ
Fig. 6 and Fig. 7 show the effect of hydration age on
in M23 will decrease continuously after 28 days, but with
the porosity based on BSE image analysis of the ITZ
a limited variation.
of Portland cement mortars. The dotted vertical line
represents the width of ITZ calculated by average grey Fig. 7 presents the entire development process of porosity
approach. The folding line with the same colour as the and ITZ thickness in M35 from 3 days to 210 days. It is
vertical line represents the porosity analysis results of observed that the porosity decreases with the hydration
the same sample. age continuously, while the change of ITZ thickness
follows the same rule as that in M53 from 3 days to 28
Fig. 6a shows that the porosity of cement paste in M53
days. Specifically, the green arrows in Fig. 7 clearly
decreases with the hydration age. It can also be found that
identify the variation of the obtained thickness of ITZ in
the porosity dropped sharply from 7 days to 28 days. On the
M35. It is found that there are two periods with sharp
other hand, the obtained thickness of ITZ in M53 is 30.7 m,
decline of ITZ thickness. One period is from 7 days to 28
29.7 m and 21.1 m at the hydration age of 3, 7 and 28 days,
days with the decline of 31.3%, the other period is from 28

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

672 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating Different Mineral Admixtures

With the development of hydration process, it is obvious

that the microstructure variation of ITZ is more intense
in mortars at higher water­to­cement ratio. In order to
correlate the elastic properties of ITZ with characteristic
microstructure directly, the nanoindentation testing
coupled with BSE image analysis was conducted on M53
at 3, 7, 28 days respectively. Fig. 8b shows that the ITZ
thickness of M53 is around 30 m at the hydration age
of 3 days. The ratio between the average indentation
modulus of ITZ and that of cement paste matrix is 55.3%
for this sample. Fig. 9b shows that the variation of average
indentation modulus of M53 at 28 days is relatively
smaller than that at 3 days and 7 days. The ITZ thickness
of M53 is around 20 m at 28 days. The ratio between the
average indentation modulus of ITZ and that of cement
Fig. 7: Effect of hydration age on the porosity in the ITZ of cement
paste matrix is 77.9% for M53 at 28 days.
mortar (w/b=0.35)

days to 90 days with the drop of 38.9%. It is also shown

that the change of ITZ thickness is limited after 90 days.
By comparing the ITZs in M23, M35 and M53, it is revealed
that the development of ITZs follows different rules for
mortars at different water­to­cement ratio.

Fig. 9: Grid indentation modulus on ITZs of cement mortar (w/

c=0.53) after 28 days curing period: (a) indentation modulus and
phase distribution, (b) mean indentation modulus distribution
across ITZ

The estimating results of ITZ thickness of M53 at different

Fig. 8: Grid indentation modulus on ITZs of cement mortar (w/ hydration age, based on average grey approach (AGA) and
c=0.53) after 3 days curing period: (a) indentation modulus and indentation modulus distribution (IMD) are illustrated in
phase distribution, (b) mean indentation modulus distribution Fig. 10. It is shown that the ITZ thickness decreases with
across ITZ the hydration age constantly, from 30 m at 3 days to 20

Organised by
India Chapter of American Concrete Institute 673
Technical Papers

m at 28 days. The ratio between the indentation modulus

of ITZ and that of cement paste matrix increases from
55.3% to 77.9% at the corresponding age. By comparing
the results from AGA and IMD, it is revealed that the two
methods get a good agreement, while AGA is more precise
and efficient for operation.

Effect of Fly Ash on ITZs

As presented in Table 1, specific surface of fly ash is
much larger than that of cement. So, fly ash grains may
be considered as micro­ fillers which are supposed to
reduce the porosity and water­to­cement ratio gradients
in the ITZ. Moreover, the pozzolanic reaction of fly ash
may change the composition and microstructure of
ITZ at late age[8,22]. However, these facts do not provide
quantitative information on how much they influence the
elastic properties of ITZ as compared to the bulk cement
paste. In order to compare Portland cement mortar
and fly ash blended mortar, Fig.11 illustrates the typical
microstructures, phase distribution and elastic modulus
variation with the distance from sand surface of the two
samples.

Fig. 10: Estimating results of ITZ thickness for M53 at 3, 7, 28

days

Fig. 11a and 11b show two SEM­ BSE images of the
representative zones of M35 and M35FA after
nanoindentation testing at 180 days. As mentioned before,
each indent (marked with a red circle) could be located
and visually classified into several phases through simple
identification[30]. Fig. 11c shows the testing results of
elastic modulus, location and phase identification of each
indent. Fig. 11c and d shows the total testing results of 8
areas (800 indents) in M35 and M35FA respectively. For
M35FA, it is difficult to distinguish the indents located in
the areas around unhydrated clinkers because there are
much less unhydrated clinkers in ITZ than in hardened
cement mortar. So a few of indents located around clinker Fig. 11: Grid indentation modulus on ITZs of mortars (M35, M35FA)
are ignored in Fig. 11d. after 180 days curing period: (a) SEM­BSE image of the indents in
M35 after NI, (b) SEM­BSE image of the indents in M35FA after
Fig. 11e and 11f show the distribution of statistic elastic NI, (c) indentation modulus and phase distribution in M35, (d)
modulus across the ITZ in M35 and M35FA at 180 days indentation modulus and phase distribution in M35FA, (e) mean
respectively. For M35, Fig. 11e shows that the statistic indentation modulus distribution across ITZ in M35, (f) mean
modulus profiles has a trough within the ITZ, with the indentation modulus distribution across ITZ in M35FA

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

674 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating Different Mineral Admixtures

minimum values occurring at 10­ 15μm from the sand

interface. The statistic modulus then increases when
test points move towards the paste region and becomes
roughly constant at distances greater than 50μm. It is also
noted that the results show moderate variations even at
the same distance from the sand, which is caused by the
heterogeneity of the cement­based materials. However,
for M35FA, there is no obvious elbow region of modulus
value adjacent to sands and the statistic modulus always
fluctuates around a mean value at the distances greater
than 5μm. The high modulus value at the distance of
2.5μm, which is attributed to the influence of sands with
high elastic modulus, should be ignored.
The results in Fig. 11e clearly indicate that the average
elastic modulus of the ITZ is significantly lower than the
value of the paste matrix for M35. For instance, in the
surrounding areas of sands (5­30μm), the average elastic
modulus is only 84% of the value in the paste matrix of
M35. However, the corresponding result for M35FA is
99%, which seems to suggest that the ITZ properties
are more uniform in the fly ash blended mortar than in
the pure cement mortar at 180 days. This may be due
to the morphological effect, micro­ aggregate effect
and pozzolanic effect of fly ash, which results in more
homogeneous structure of the ITZ.

Effect of Slag on ITZs

Fig. 12 shows the total testing results of 6 areas (600
indents) in M35S at 180 days. Fig. 12a shows the testing
results of elastic modulus, location and phase identification
of each indent. Fig. 12b shows SEM­BSE image of the
representative zone of M35S after nanoindentation
testing. As shown in Fig. 12b, it is difficult to distinguish the
indents located in OP, IP or secondary hydration products.
So all the hydration products are denoted as HP in Fig.
12a. The phase “Interface” marked with green pentagram
includes the indents located around clinker particles and
slag grains. Fig. 12c shows the distribution of statistic Fig. 12: Grid indentation modulus on ITZs of slag blended mortar
elastic modulus across the ITZ in M35S at 180 days. (M35S) after 180 days curing period: (a) SEM­BSE image of the
As shown in Table 1, specific surface of slag is a little larger indents after NI, (b) indentation modulus and phase distribution,
(c) distribution of average indentation modulus across ITZ
than that of cement, that is to say, the size of slag is in the
same level of cement grains. It could be assumed that the
the same level as outer product[26]. In summary, the
packing around sand particles is not affected. However, it
disappearance of ITZ in slag blended mortar is mainly due
can be seen in Fig. 12c that the average elastic modulus of
to its pozzolanic reaction during the hydration progress
the vicinity of sand is close to that of the area away from
rather than the particles filling effect in the initial state.
sand surface. Thus, it is supposed that the effect of slag
on ITZs is implemented in the hydration process by the
pozzolanic reaction between Ca(OH)2 from the hydration Conclusions
of Portland cement clinkers and glassy phase in slag. Grid­
nanoindentation coupled with scanning electron
Compared with Fig. 11e, it is revealed that the average microscope image analysis was employed to characterize
elastic modulus of paste matrix in M35S is much lower the microstructure and elastic properties of mortars at
than that in reference Portland cement mortar. So, it is microscale. The variation of ITZ with water­to­cement ratio,
believed that a large number of secondary hydration hydration age and mineral admixtures were investigated.
products have played a role in filling and homogenizing The following conclusions can be drawn from the results
the original structures, as it has been verified that the obtained so far.
elastic modulus of secondary hydration products is in

Organised by
India Chapter of American Concrete Institute 675
Technical Papers

1. The thickness and elastic properties of ITZ depend 6. Lee KM, Park JH. A numerical model for elastic modulus of concrete
considering interfacial transition zone [J]. Cement Concrete Res,
on its microstructure, such as porosity and phase
2008, 38(3): 396­402.
distribution. The average grey variation with the
7. Hu J, Stroeven P. Properties of the interfacial transition zone in
distance from sand surface of BSE image, which model concrete [J]. Interface Science, 2004, 12(4): 389­397.
reflects the microstructural features directly related
8. Ollivier JP, Maso JC, Bourdette B. INTERFACIAL TRANSITION ZONE
to the elastic properties, can be used to determine the IN CONCRETE [J]. Advanced Cement Based Materials, 1995, 2(1):
effective thickness of ITZ. 30­38.
2. The effect of water­to­cement ratio on ITZs is mainly 9. Scrivener KL, Crumbie AK, Laugesen P. The interfacial transition
zone (ITZ) between cement paste and aggregate in concrete [J].
reflected in terms of porosity. At the same hydration
Interface Science, 2004, 12(4): 411­421.
age, the effective thickness of ITZ increases with
10. Scrivener KL, Bentur A, Pratt P. Quantitative characterization of the
water­to­
cement ratio, while the ratio between the transition zone in high strength concretes [J]. Advances in cement
elastic modulus of ITZ and that of cement paste matrix research, 1988, 1(4): 230­237.
decreases correspondingly. 11. Bentur A, Alexander MG, Comm RT. A review of the work of the
RILEM TC 159­E TC: Engineering of the interfacial transition zone
3. The effect of hydration age on ITZs of mortars is
in cementitious composites [J]. Mater Struct, 2000, 33(226): 82­87.
related to water­to­cement ratio. At low water­to­cement
12. Hussin A, Poole C. The geochemical features of the interfacial
ratio 0.23, the ITZ thickness decreases from 15.3 m to
transition zone [J]. Advances in cement research, 2010, 22(1): 21­28.
13.9 m from 3 days to 28 days. At high water­to­cement
13. Hussin A, Poole C. Petrography evidence of the interfacial
ratio 0.53, the ITZ thickness decreases from 30.7 transition zone (ITZ) in the normal strength concrete containing
m to 21.1 m from 3 days to 28 days. The indentation granitic and limestone aggregates [J]. Constr Build Mater, 2011,
modulus ratio of ITZ to cement paste matrix increases 25(5): 2298­2303.
from 55.3% to 77.9% during the corresponding age. 14. Xie YT, Corr DJ, Jin F, Zhou H, Shah SP. Experimental study of the
At medium water­to­cement ratio 0.35, there are two interfacial transition zone (ITZ) of model rock­filled concrete (RFC)
periods with sharp decline of the ITZ thickness. One [J]. Cement Concrete Comp, 2015, 55: 223­231.
period is from 7 days to 28 days with the decline of 15. Kong LJ, Du YB, Gao LX. Effect of coarse aggregate on the interfacial
31.3%, the other period is from 28 days to 90 days with transition zone of concrete based on grey correlation [J]. Magazine
of Concrete Research, 2014, 66(7): 339­347.
the drop of 38.9%.
16. Xiao JZ, Li WG, Sun ZH, Lange DA, Shah SP. Properties of
4. Microscopic characterization shows that the ITZs interfacial transition zones in recycled aggregate concrete tested
between the sands and the pastes disappeared at late by nanoindentation [J]. Cement Concrete Comp, 2013, 37: 276­292.
age when mineral admixtures were incorporated. 17. Mondal P, Shah SR, Marks LD. Nanoscale characterization of
The reasons are somewhat different between fly ash cementitious materials [J]. Aci Mater J, 2008, 105(2): 174­179.
blended mortar and slag blended mortar. For the 18. Sun GW, Sun W, Zhang YS, Liu ZY. Numerical calculation and
former, it is mainly attributed to the filling effect of fly influencing factors of the volume fraction of interfacial transition
zone in concrete [J]. Sci China­Technol Sci, 2012, 55(6): 1515­1522.
ash particles; for the later, it is supposed to be due to
19. Machovic V, Kopecky L, Nemecek J, Kolar F, Svitilova J, Bittnar Z,
the pozzolanic reaction of slag.
et al. Raman micro­spectroscopy mapping and microstructural and
micromechanical study of interfacial transition zone in concrete
reinforced by poly(ethylene terephthalate) fibres [J]. Ceram­Silikaty,
Acknowledgements 2008, 52(1): 54­60.
The support from NSFC under 51178105 is gratefully 20. Zimbelmann R. A contribution to the problem of cement­aggregate
acknowledged. bond [J]. Cement Concrete Res, 1985, 15(5): 801­808.
21. Zhan BG, Wang Q, Zhou WL, Sheng HY. Mechanism of Silica fume in
References suppressing alkali­silica reaction of concrete: interfacial transition
1. Ulm FJ, Constantinides G, Heukamp FH. Is concrete a poromechanics zone concerns. In: Zheng JJ, Du XL, Yan W, Li Y, Zhang JW, editors.
material? ­A multiscale investigation of poroelastic properties [J]. Trends in Building Materials Research, Pts 1 and 2. Stafa­Zurich:
Mater Struct, 2004, 37(265): 43­58. Trans Tech Publications Ltd; 2012. p. 676­682.

2. Sun Z, Garboczi EJ, Shah SP. Modeling the elastic properties of 22. Paulon VA, Dal Molin D, Monteiro PJM. Statistical analysis of the
concrete composites: Experiment, differential effective medium effect of mineral admixtures on the strength of the interfacial
theory, and numerical simulation [J]. Cement and Concrete transition zone [J]. Interface Science, 2004, 12(4): 399­410.
Composites, 2007, 29(1): 22­38. 23. Diamond S. The microstructure of cement paste and concrete––a
3. Ma HY, Li ZJ. Multi­Aggregate Approach for Modeling Interfacial visual primer [J]. Cement and Concrete Composites, 2004, 26(8):
Transition Zone in Concrete [J]. Aci Mater J, 2014, 111(2): 189­199. 919­933.

4. Grondin F, Matallah M. How to consider the Interfacial Transition 24. Xiao JZ, Li WG, Corr DJ, Shah SP. Effects of interfacial transition
Zones in the finite element modelling of concrete? [J]. Cement zones on the stress­strain behavior of modeled recycled aggregate
Concrete Res, 2014, 58: 67­75. concrete [J]. Cement Concrete Res, 2013, 52: 82­99.

5. Jiang JY, Sun GW, Wang CH. Numerical calculation on the porosity 25. Chen JJ, Sorelli L, Vandamme M, Ulm FJ, Chanvillard G. A
distribution and diffusion coefficient of interfacial transition zone Coupled Nanoindentation/SEM­EDS Study on Low Water/Cement
in cement­based composite materials [J]. Constr Build Mater, 2013, Ratio Portland Cement Paste: Evidence for C­ S­H /Ca(OH)(2)
39: 134­138. Nanocomposites [J]. J Am Ceram Soc, 2010, 93(5): 1484­1493.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

676 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Microstructures and Elastic Properties of Interfacial Transition Zones in Mortars Incorporating Different Mineral Admixtures

26. Hu CL, Li ZJ, Gao Y Y, Han YG, Zhang YM. Investigation on Ceramic Society, 2012, 40(11): 1559­1563.
microstructures of cementitious composites incorporating slag
31. Oliver WC, Pharr GM. An Improved Technique for Determining
[J]. Advances in cement research, 2014, 26(4): 222­232.
Hardness and Elastic­Modulus Using Load and Displacement Sensing
27. Hu CL, Han YG, Gao YY, Zhang YM, Li ZJ. Property investigation of Indentation Experiments [J]. J Mater Res, 1992, 7(6): 1564­1583.
calcium­silicate­hydrate (C­S­H) gel in cementitious composites [J].
32. Oliver WC, Pharr GM. Measurement of hardness and elastic
Mater Charact, 2014, 95: 129­139.
modulus by instrumented indentation: Advances in understanding
28. Hu C, Li Z. Micromechanical investigation of Portland cement paste and refinements to methodology [J]. J Mater Res, 2004, 19(1): 3­20.
[J]. Constr Build Mater, 2014, 71(0): 44­52.
33. Fischer­Cripps AC. Nanoindentation: Springer; 2011.
29. Hu C. Nanoindentation as a tool to measure and map mechanical
34. Hu C, Li Z. A review on the mechanical properties of cement­based
properties of hardened cement pastes [J]. MRS Communications,
materials measured by nanoindentation [J]. Constr Build Mater,
2015, 5(01): 83­87.
2015, 90: 80­90.
30. Gao Y, Zhang Y, Hu C, Li Z. Nanomechanical Properties of Individual
35. Diamond S. Considerations in image analysis as applied to
Phases in Cement Mortar Analyzed Using Nanoindentation Coupled
investigations of the ITZ in concrete [J]. Cement and Concrete
with Scanning Electron Microscopy [J]. Journal of The Chinese
Composites, 2001, 23(2): 171­178.

Dr. Yamei Zhang

Dr. Yamei Zhang is a professor in school of materials science and engineering, Southeast University, China.
Her research interests include hydration of cement, characterization of the microstructure and properties
of cement based composites, utilization of solid wastes as construction material. She is now on the editorial
board of Journal of Cement and Concrete Composite, guest research fellow of Japan Sustainability
Institute, member of the International Federation for Structural Concrete fib Com 9 and TG3.10. She is also
the vice chair of the Committee for Regenerated Concrete of China Civil Engineering Society.

Organised by
India Chapter of American Concrete Institute 677
Technical Papers

Sustainable Production of Fiber Reinforced Cementitious Composites

B.Y. Pekmezci
Department of Civil Engineering, Istanbul Technical University, Maslak, Istanbul, Turkey

Abstract and wide composite panels. The developing technique

should improve the production speed with low labor for
Properties of fiber reinforced cementitious panels
a sustainable production. The product should have high
produced with varying production techniques studied
durability and satisfactory strength, low maintenance,
experimentally in the scope of this study. Calender
and very stable quality at each series of production.
extrusion proposed as sustainable production technique
for fiber reinforced cementitious composites. A successful Conventional cement composite processing techniques,
industrial production of calendar extrusion applied. such as casting and spraying, are either expensive or
Mechanical properties as well as the water absorption very slow compared to cement-bonded fiberboard panel
and expansion under different humidity conditions were production techniques. On the other hand, technical
discussed. The experimental results showed that calender properties of the cement composites produced with
extrusion might be a promising method for sustainable conventional methods are significantly better than those
production of thin and wide cementitious composites. of cement-bonded fiberboards produced using various
As the main results of the study: the mechanical mass production techniques. Considering their high
properties of the calendered cementitious composites volume of usage, efficient mass production techniques
(CCC) were higher than that of the materials produced for fiberboards gain importance in the building materials
with conventional particle boards (CBPB) and hatschek industry. The most widely used technique, the Hatschek
fiberboards (FC) as well as the dimensional stability and process, was developed in the early 1900s for production
water absorption values. of flat and corrugated cement-based panels for cladding
and roofing. In the Hatschek process, the dewatering
Keywords: Sustainability, Fiberboard, cementitious
slurry of fiber, cement, and water forms thin laminated
composites, fiber, fiber reinforced concrete. layers. These layers are then bonded together while they
are still in the plastic stage. This laminar structure shows
Introduction comparatively low strength and durability (Kuder and
The most important elements of the sustainability are, Shah 2010; Mohr et al. 2004).
economy, environment and society. There is a close link Investigations on possible alternative production
between economy and environmental sustainability. Since techniques to Hatschek production are being done
the life of the structure is directly related to sustainability, for thin and wide cementitious composites. Extrusion
high durable materials can be accepted as more sustainable method is one of the alternatives. It is recommended
(Struble and Godfrey 2004). Portland cement is the widely as a highly efficient method for the production of short
used element of construction elements, and has the fiber-reinforced cementitious composites (SFRCCs).
highest responsibility of releasing CO2 to the environment Extrusion also improves the mechanical properties of
(Malhotra 2004). A sustainable concrete should has very cement-based materials (Shao et al. 1995; Shao et al.
low societal impact during its entire life (Naik 2008). For 1996; Peled et al. 2000; Burke and Shah 1999; Qian et
sustainability in construction industry, researchers and al. 2003). In the extrusion process, stiff dough is passed
producers focused on structural concrete due to its mass through a die and shaped with a desired cross-section
consumption in building industry. There are big amount (Kuder and Shah 2010). Extrusion has also been proposed
of cement based non-structural construction materials, as a suitable method for industrial production since it
which can be produced with techniques and materials is a continuous process. Extrusion is more efficient to
allowing sustainable production. For a sustainable produce products with complicated shapes (Aldea et al.
production, conserving of energy and natural resources 1998). Since the wall of the extrusion die is effective in
and economic viability are the most important parameters. generating slip resistance against the extrusion direction,
Production speed and cost of cement-based materials are it is used for relatively thicker products than Hatchek
strongly influenced by the processing method. There is a products. Excessive extrusion pressure is required when
need for a production technique of fiber reinforced thin die cross- section is very narrow. Processing becomes

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

678 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainable Production of Fiber Reinforced Cementitious Composites

difficult due to generated high wall resistance. Moreover,

the production is limited to thin and short fibers. Mortars
having comparatively longer fibers cannot be processed
through narrow dies of extrusion.
In the scope of this experimental work calender extrusion is
applied as a sustainable method that provides continuous,
rapid, and economic production of durable thin and wide
cementitious composites. Allowing mass production Fig. 1: Schematic description of calender extrusion-section
of fiber-reinforced composites in relatively smaller
plants shows that calender extrusion is very efficient In the laboratory study, a pilot scale calender set was used
method for production of fiber reinforced cementitious to produce thin fiberboard panel cement composites.
composites providing an economic production. A general Panels with a 1.0 cm to 1.5 cm thicknesses with a pilot
overview of the novel production technique was done, scale calendaring system were successfully produced.
and the properties of the produced glass fiber reinforced
cementitious composites were discussed in this paper. Materials
The performance of the composites produced in pilot The cement composite consisted of glass fiber, cement,
scale was compared to other commercial industrial sand, calcite dust, superplasticizer, and a synthetic
fiber reinforced cementitious composite products. Two polymer-based water-retaining agent. The experiment
commercial panel products tested other than calendered employed long glass fibers with a length of 50 mm.
cementitious composite (CCC) panel were Cement bonded
One cubic meter matrix contained 780 kg of White Cement
particle board (CBPB) and fibercement board (FC).
52.5, 345 kg of calcite dust, and 360 kg of siliceous
sand. Glass fiber amount was 3.7 % of total weight of the
Experimental Study composite. The maximum size of sand was 400μm, and
that of calcite dust was 100 μm. The specific gravities of
Calender Extrusion Process
cement and siliceous materials were 3.16 and 2.64 g/cm3,
Calender extrusion is a process widely used in the respectively. The specific gravity of calcite was 2.70 g/cm3.
production of plastic sheets. This technology has been
adapted to cement-based materials to produce fiber- Flexural tests
reinforced cement composite panels. Figure 1 shows
Four-point bending tests were carried out in a close loop
schematic representation of the system. In this method,
Instron 5500R testing machine, with a maximum load
first the dry substances, except the fibers, are weighed
capacity of 100 kN, at a loading speed of 0.8 mm/ min.
according to a proper mix design, and poured into a
Sample dimension was 35x500x10 mm. Span length
mixer. The dry mix is then prepared. Next, half of the total was kept at 300 mm. Flexural strength test setup is
amount of water to be introduced into the mix is added to shown in Figure 2a. Tests were performed on the 28th day
moisten the mix and mixed for 5 minutes. Meanwhile, the after casting. The tests for the determination of specific
plasticizer, liquid materials, and other chemical additives, fracture energy were performed in accordance with the
as well as the remaining portion of water, are mixed in a recommendation of RILEM 50-FMC Technical Committee
dosing unit and taken into the mixer. Finally, the fibers are using a closed- loop testing machine (Instron 5500R). The
added to the mix. specific fracture energy (Wf) was calculated based on the
A feeding unit drops the fresh material to a pouring area under the load-deflection curve of the specimens.
band. An auger squeezes the material through a die. Flexural tests were finalized when the deflection at
The first band takes the fiber included mortar from the midpoint had approximately reached 8 mm.
feeding unit and delivers it to the first roller set, which
then compresses the cementitious mortar to a 3-5 Tensile tests
cm thickness. The material is then passed to a second Direct tensile tests of the composites were performed on
band, and the second band transfers the material to a an MTS 5000N testing machine using an elongation rate
second roller set. The thickness of the material is further of 1.0 mm/min. Load was applied in long dimension of the
decreased in the second roller set to 1-3 cm. The thinned composites. Tensile testing setup is shown in Figure 2b.
material is taken from the rollers, and passed to a third
band, and the third band transfers the material to a third Expansion and water absorption tests
roller set. The third and the fourth roller sets decrease Expansion tests were carried out according to EN 12746.
the fresh material thickness to 0.6-1.0 cm and 0.2-0.4 cm, Composite panels having 75x350x10 mm dimensions
respectively. The thickness of the product is adjustable were conditioned in 30% and 90% relative humidity (RH).
with the distance between the rollers. The dimensional differences between the environmental

Organised by
India Chapter of American Concrete Institute 679
Technical Papers

(a) flexural strength test (b) tensile strength test

Fig. 2: Flexural and tensile test setups Fig. 4: Flexural strength displacement at mid-span relations
of CCC panels
conditions of 30% and 90 % RH were recorded.
was showing the lowest value. Modulus of elasticity
Specimens dried in oven and then transferred in water for
values at tensile test was parallel to the tensile strength
determination of water absorption. The weight difference
test results. The modulus of elasticity value of CBPB was
in between was recorded as water absorption.
approximately 4.5 GPa while the others were around 15
GPa. Lower modulus of elasticity decreases the crack risk
Test results and discussion of the composite, but the deformation properties under
Flexural strength and displacement at mid-span relations different humidity rates, and the tensile strengths should
of different cement based composite panels are shown be taken into consideration as well when evaluating crack
in Figure 3. Fibercement board (FC) produced with risk. Since the MOE of CBPB is low, it is not easy to say the
Hatschek process showed the highest flexural strength crack risk of this composite is the lowest among others
(17.2 MPa) while the cement-bonded particleboard (CBPB) due to its high expansion, water absorption and low tensile
showed the lowest flexural strength (7.6 MPa). Glass strength.
fiber reinforced CCC panel exhibited 14.1 MPa of flexural Water absorption test results are presented in Figure 6a.
strength. The specific fracture energy values for FC, Fibercement panels showed the highest water absorption
CBPB and CCC panels were 0.355J, 0.144J and 1.466 J value. Water absorption of calendered cementitious
respectively. The fracture energy values showed that CCC composite (CCC) was much lower than that of fiber cement
panels exhibited 10 times higher fracture energy than
CBPB. CCC board had the highest modulus of elasticitiy
value at flexure, which shows that CCC panel has the
highest crack risk among the others. Flexural strength
and displacement at mid-span relations of CCC panels are
given in Figure 4, which shows that the distribution of the
results is very low.
Tensile strength and modulus of elasticity values are given
in Figure 5. FC showed the highest tensile strength values
among others while the cement-bonded particleboard

Tensile strength

Modulus of elasticity
Fig. 3: Flexural strength-displacement at mid-span relations
of different cement based panels Fig. 5: Tensile strength and modulus of elasticity values

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

680 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainable Production of Fiber Reinforced Cementitious Composites

(a) Inside panel (b) fracture and panel surface

Fig. 7: Micrographs of cement bonded particleboard

(a) water absorption
composite surface. The fibers in the surface lamina
causes very rough surface for finishing side of the panel.
Micrographs of the CCC panel are shown in Figure 9.
Although there are some fibers present on the surface of
the panel, the surface is quite smooth as an architectural

(b) expansion

Fig. 6: Water absorption and expansion values

(FC) composite. The difference in water absorption values (a) Cracks inside panel (b) Surface of the panel

was partially reflected to the results of expansion test.

Fig. 8: Micrographs of fibercement board
CBPB composite showed the highest expansion value of
2 mm/m between 30% and 90% of relative humidity while
finishing surface. There are some micro cracks inside
the FC composite and CCC composite were showing 1.30
panel along cross-section occurred during production.
and 0.70 mm/m respectively. This difference is attributed
The crack length was approximately 2 mm. This crack
to the wood particles and cellulose fiber in CBPB and FC
may behave as a defect during flexural strength or tension
composites respectively. Since there is a solid structure
test. Some of entrapped air voids in the composite were
along the cross section of glass fiber reinforced CCC
determined as well. These entrapped airs introduced into
there is a significant improvement of expansion value in
the composite and are not able to get out of the sample
comparison to other composites.
due to the consistency of the matrix. This entrapped air
Micrographs of the cement-bonded particleboard (CBPB) can behave as defect during tests and can decrease the
are shown in Figure 7. Wood particles inside the CBPB strength of the sample if there is not a crack bridging
panel, which causes low dimensional stability, can be activity of fibers crossing the crack.
easily distinguished. Moreover, there is a cement matrix
According to Pekmezci’s studies (Pekmezci 2014, Pekmezci
layer on the surface of the panel having approximately
2015) done, the parameters influences the performance
1 mm thickness, which can easily be seen in Figure 7b.
of CCC panels can be summarized as follows. Properties
This layer prevents the panel from durability problems
of CCC panels vary with curing conditions, type of the
and makes less permeable. Although there is a thin layer
fiber and volume fraction of the fiber. Curing condition
at the surface of the composite, water absorption of the
CBPB panel is 20%. Due to high water absorption and
wood particles inside, the expansion of the composite is
very high.
Micrographs of the fibercement panel are shown in Figure
8. Interlaminar cracks between the laminar structures
occurred during the production of Hatschek process is
given in figure 8a. It is stated that in the literature that
these cracks causes serious durability problems during
(a) Surface of the panel (b) Inside panel
wetting drying and expansion shrinkage cycles (Kuder
and Shah 2010; Mohr et al. 2004). Figure 8b shows the Fig. 9: Micrographs of calendered cementitious composite (CCC)

Organised by
India Chapter of American Concrete Institute 681
Technical Papers

is an effective parameter on mechanical performance of FC and CCC panel showed much more higher elasticity
CCC panels. Water cured panels show higher strength modulus and tensile strength than CBPB panel.
values at both longitudinal and transverse directions. This
CCC panels showed the lowest water absorption and
performance increase at water curing can be attributed
expansion values in comparison to CBPB and FC panels.
to denser microstructure formed during hydration. The
performance of the CCC panel depends on the direction
of the sampling. Samples taken from longitudinal Acknowledgement
directions have higher performance than samples taken Author acknowledges Ticem Advanced Structural
along transverse direction due to alignment of the fibers Technologies for financial support to the experimental
during processing. Behaviour of the composite might vary
study.
due to the direction of sampling as well. As an example,
PVA fiber reinforced cement composites exhibit strain-
References
hardening behavior longitudinally while they do not
1. Aldea, C., Marikunte, S., Shah S. P. 1998. Extruded fiber reinforced
exhibit the same behavior transversely. Volume fraction
cement pressure pipe, Advn. Cem. Bas. Mat. (8) 47-55.
of fiber is another effective parameter on mechanical
2. ASTM C1185-08 (2012) “Standard Test Methods for Sampling and
performance. Increasing fiber volume fraction for glass,
Testing Non-Asbestos Fiber-Cement Flat Sheet, Roofing and Siding
polypropylene and polyvinyl alcohol fibers, causes an Shingles, and Clapboards.”
increase of mechanical performance of panel. In previous
3. Burke, P. L., Shah, S. P., 1999. Durability of extruded thin sheet PVA
studies, it is stated that following accelerated freeze-thaw fiber-reinforced cement composites.” ACI SP-190 high performance
tests showed no noticeable deterioration of the exposed fiber reinforced concrete thin sheet products. 133-164.
surfaces following 100 freeze-thaw cycles. 4. Kuder, K. G., Shah, S. P., 2010. Processing of high performance
fiber reinforced cement based composites.” Construction Build.
Mater., (24) 181-186.
Conclusions
5. Malhotra, V. M., 2004. Role of supplementary cementing materials
According to the results obtained within the scope of this and superplasticizers in reducing greenhouse gas emissions. Proc.,
experimental work, the following conclusions can be ICFRC Int. Conf. on Fiber Composites, High-Performance Concrete,
drawn. and Smart Materials, Indian Institute of Technology, Chennai, India,
489–499
Calender extrusion is a promising processing method for
6. Mohr, B. J., El-Ashkar, N. H. Kurtis, K. E., 2004. “Fiber-cement
thin and versatile calendered cementitious composites composites for housing construction: state- of-the-art review.”
(CCC). This method aims to produce sustainable cement Proc. of the NSF Housing Research Agenda Workshop.,Orlando,
composite panels and architectural materials that have FL. 112-128.
adequate strength and are durable against the external 7. Naik, T., 2008. Sustainability of Concrete Construction.” Pract.
environment. Very low labour of production contributes to Period. Struct. Des. Constr., 13 (2), 98–103.
this methods sustainability concept. 8. Pekmezci, B.Y., 2014. Properties of PVA-reinforced cement-bonded
fiberboards processed with calender extrusion, Science and
CCC composites have a solid cross section of cementitious
Engineering of Composite Materials, DOI 10.1515/secm-2014-0030.
matrix and fibers while CBPB and FC composites have
9. Pekmezci B.Y., 2015. Calender Extrusion of Cement Mixtures for
wood particles and interlaminar cracks which can be
Sustainable Composite Production, ACI SP-299 Fiber Reinforced
accepted as structural defects influencing mechanical Concrete for Sustainable Structures.
and durability performance.
10. Peled, A., Cyr, M., Shah, S. P., 2000. High content of fly ash (Class F)
CCC composites have internal defects such as micro in extruded cementitious composites. ACI Mater. J. 97 (5) 509-517.
cracks having 2 cm of length and entrapped air due to 11. Qian, X., Zhou, X., Mu, B., Li, Z., 2003. Fiber alignment and property
the processing technique. These defects might affect the direction dependency of FRC extrudate. Cem. Conc. Res. (33) 1575–
mechanical performance negatively. 1581.
12. Shao, Y., Marikunte, S., Shah, S.P., 1995. Extruded fiber-reinforced
Among three types of panels industrially produced and composites” Concrete Int. 17 (4) 48-52.
tested in the scope of the study. Fibercement (FC) panel
13. Shao, Y., Shah, S.P., 1996. High perfromance fiber-cement
produced with Hatschek process showed the highest composites by extrusion processing. Proc. 4th Mat. Eng. Conf.
strength value (17.2 MPa) while the cement bonded Washington DC.251-260.
particleboard (CBPB) showed the lowest value (7.6 MPa). 14. Struble, L. and Godfrey, J., 2004. How sustainable is concrete?"
The fracture energy of CCC panel was 10 times higher than International Workshop on Sustainable Development and Concrete
CBPB panel while it was 5 times higher than FC panel. Technology, Beijing. 201-211.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

682 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Sustainable Production of Fiber Reinforced Cementitious Composites

Dr. Bekir Pekmezci

Dr. Pekmezci is working as Associate Professor at Istanbul Technical University, Turkey. He has been
working for Istanbul Technical University since 1998. He graduated from the same University. He had been
to ACBM-Northwestern University, USA in 2004-2005. He is directing Research & Development Company
TICEM Advanced Structural Technologies since 2010 parallel to his Professor position. He has completed
numbers of industrial projects and published papers related to, application of nano technology on building
materials, innovative processing of building materials, high performance lime and cement based materials.

Organised by
India Chapter of American Concrete Institute 683
Technical Papers

Effects of anionic and non-ionic anti-washout admixtures on the

performance of underwater concrete
Bo Pang1, Zonghui Zhou1, Xin Cheng1
1
Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, Engineering Center of Advanced
Building Materials of Ministry of Education, University of Jinan, Jinan 250022, China

Abstract soluble polymers used to enhance the washout resistance

and cohesion of underwater concrete. Different AWAs
An investigation has been carried out to study the
have diversities in molecular, chain, chain length and
different influence on the performances, in particular
flocculated structure, cause a large difference in washout
the workability and washout-resistance of underwater
resistance, viscosity, fluidity loss and cement hydration of
concrete with two kinds of different anti-washout
concrete.
admixtures (AWAs), the Hydroxy propyl methyl cellulose
(HPMC) and Polyacrylamide (PAM) are non-ionic AWAs However, K.H. Khayat al. found that adjacent polymer
while the carboxymethyl starch (CMS) and polyanionic chains (HPMC) can also develop attractive bonds among
cellulose (PAC) are anionic. Rheological properties (slump one another blocking the motion of water and increasing
and slump flow), washout resistance and mechanical further the apparent viscosity, especially at low rates of
properties are tested to measure the fresh and hardened shear10-13. Mixing energy can easily destroy such bond, and
properties of underwater concrete. the polymer chains can be aligned in the direction of flow,
which results in a reduction in viscosity, hence leading to
The results reveal that, anionic AWAs are more effective
a shear-thinning or pseudoplastic behavior14-18.Thus, there
than the non-ionic ones in improving the workability and
are still many problems in conventional non-ionic AWAs.
strength of the anti-washout underwater concrete. When
the dosage of non-ionic AWAs over 0.3% (w/c = 0.45), the The present works establish the flocculation slurry
non-ionic AWAs have a negative impact on the fluidity models of the UWC with non-ionic and anionic AWAs,
and air content of concrete. Besides, the morphological and compare their different effects on fluidity, stability,
structures of cement paste, from the SEM, are easier to compressive strength and morphology of UWC.
be changed by non-ionic AWAs, thereby they will affect Additionally, explanations are given according to the AWAs’
the hydration of cement particles and the properties of different molecular structures and chemical bonds.
underwater concrete adversely. In mechanical properties
In the 1970s AWAs were first used in Germany and then
aspects, the compressive strength (28d) of concrete with
used widely in Asian countries in the 1980s.In north
anionic AWAs can reach up to 37Mpa, increased by 151%
America, AWAs have been used in cement-based material
and 131% to the ones with no AWA and non-ionic AWAs.
since the late 1980s19. Mailvaganam20 categorized AWAs
The adaptability of concrete on anionic AWAs is more
into five classes. The AWAs used in the experiments are
stable than the non-ionic ones. Key terms: anti-washout
HPMC, PAM, PAC and CMS.HPMC and PAM are the most
admixture; underwater concrete; anionic; non-ionic.
commonly applied non-ionic AWAs in building materials
industry in China.PAC and CMS are anionic AWAs for
Introduction comparison with the former.
Nowadays, with the increase of underwater engineering, HPMC is a kind of nonionic alkyl hydroxyalkyl cellulose
the using of concrete used in underwater construction mixed ether which is obtained from alkalization and
operations is increasingly frequent. However, the etherification of cellulose. Its basic raw material is refined
constructions in water, especially in deepwater areas, are cotton, wood pulp, etc.
complex and difficult. In addition to higher requirements
for concrete quality, the convenient construction process PAM is a kind of water-soluble acrylamide polymer formed
and harmlessness to environment are also required1-3. in the polymerization of acrylamide monomer via initiators.
In order to improve the adverse effects of admixtures on PAM macromolecules overlap each other to form a cluster
flow properties, the charge of cement particles and the or curve structure which result in increasing the viscosity
flocculants molecules are needed to be considered4-9. of the solution. Anionic polymer AWAs will be adsorbed to
the suspended cement particles by charge neutralization
Anti-washout admixtures (AWAs), also known as to form a larger clusters, then the larger clusters, then
flocculants or viscosity-enhancing admixtures, are water- the adjacent clusters in unstable state are adsorbed to

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

684 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effects of anionic and non-ionic anti-washout admixtures on the performance of underwater concrete

each other by bridging and precipitate eventually. Mixing procedures and concrete mix
Compared with non-ionic AWAs, the anionic AWA has The mixing protocols and processes have a huge influence
some advantages as following: lower dosage, higher on the fluidity and workability of UWC ,particularly with
clarification of treated water, wider range of pH value in the addition of AWAs and superplasticizers. Potential
applications and well suitability with inorganic cementing chemical and physical interaction such as cement
agent etc. hydrates, adsorption of superplasticizers and AWA-
crosslinking in UWC will be greatly affected22-29. When
PAC is an uniformly substituted carboxymethyl
cellulose(CMC)essentially. It is formed after alkalization AWAs and superplasticizer are mixed up with cement
and carboxymethyl etherification of natural cotton.PAC is mortar slurry,cement particles will be wrapped by AWA,
soluble in cold water and the fluidity of solution is better so the amount of superplasticizer adsorption will decrease
than other carboxymethyl celluloses with the similar which result in slow setting.
molecular weight. However, if PAC produces conglobation In order to maximize the synergy and combination of
phenomenon, the water molecules cannot penetrate admixtures, and limit the experimental errors, the mixing
the network structure of the conglobation and result in program is made as following:
reducing dissolution rate of PAC. Besides, air bubbles are
easily formed in the mixing process of the UWC. Step1: All the cement, coarse and fine aggregates are dry
mixed for 1 minute.
CMS is formed after alkalization and etherification of
common starch, a portion of the hydroxyl groups are Step2: Water and superplasticizer have been mixed into
substituted by carboxyl groups in the alkaline condition. solution already. Adding the solution into the mixtures
Crops widely distributed such as corn, wheat, potatoes, uniformly in 30s while stirring and keep stirring for 2
etc. can be the raw materials of CMS. minutes.
Limited of studies have been carried out to evaluate the Step3: Adding the AWAs into mixtures and stirring for 1
effect of washout loss of UWC between non-ionic and minute.
anionic AWAs. Also, no design provisions have been made
available to descript the flocculating situation in the water The stirring speed is 42r·min-1 and the UWC is mixing
of UWC of different AWAs which result in the inaccuracy without stopping.
and limitations of pH factor test. This paper presents In the mixing process, non-ionic AWAs generate a lot of
useful information to estimate the effect of casting UWC bubbles in concrete which are not conducive to the sinking
with different AWAs on and workability and Macro and of UWC30 .
micro flocculating conditions21.
Table 2
The mix proportions of UWC
Experimental program
Cement Water Sand Aggregate(kg/m3) Super-
Materials W/C (kg/ (kg/ (kg/ plasticizer
2/16 16/26.5
All the AWAs are received from Yan Xing Chemical Industry m3) m3) m3) mm mm (kg/m3)
Co. Ltd, China. The degree of substitution(DS) is above 0.2
0.45 467 210 690 618 412 8.41
for CMS. The apparent viscosity of PAC is 15 mPa·s. The
surface tension of 2% HPMC solution is 42 to 56 dyn·cm-1.
The average molecular weight of PAM is 5,000,000 to The test with cement mortar
6,000,000. All chemicals used are analytical grade. Fluidity and fluidity loss
The Portland cement having blaine fineness of 311m2/kg The fluidity of cement mortar can not only reflect the
was utilized for preparing the concrete test specimens viscosity and washout-resistance of the UWC, but also
used in determination of concrete properties. The reflect the workability and fluidity loss as a reference.
chemical composition of the cement is shown in Table 1. Thus, in order to study the fluidity and fluidity loss of
The water reducing rate is 40% for the polycarboxylate different UWC, cement mortar text has been made to
superplasticizer. determine the scope of the dosage of AWAs.

Table 1
Mixture proportion of mortars and water content after water curing for 28 days (by mass)

CaO SiO2 Al2O3 Fe2O3 MgO SO3 K 2O Na2O P2O5 CaO free Ignition loss

55.85% 22.91% 7.12% 3.36% 3.28% 2.30% 0.69% 0.22% 0.19% 1.25% 1.44%

Organised by
India Chapter of American Concrete Institute 685
Technical Papers

mortar. A sample of 100g of cement mortar is dropped

into a beaker containing 300ml water. After 3 minutes
precipitation, most sediments sink to the bottom. The pH
factor test has been proposed in the recommendation
for washout-resistance of UWC in Japan31.However, it
is found that the flocculating states with different AWAs
or without AWAs are significant different. The interfaces
between supernatant and suspension are extremely not
obvious, and the concentrations of the suspension are also
different (fig3). Thus, it is hard to separate supernatant
solution from suspension and this phenomenon leads to
instability and inaccuracy of pH factor test. Finally, pH is
Fig. 1: Fluidit y and fluidit y loss of cement paste with measured in the supernatant solution which is relatively
CMS,PAC,HPMC and PAM clear. The higher the pH, the greater the washout. When
AWAs dosage increases, there is a relative decline in pH of
The standard sand manufactured according to supernatant solution. And before the dosage of 0.6%, test
ISO679:1989,EN196-1 is used in test and W/C is 0.5. results indicated that the washout resistance is affected,
in order of importance, by the dosage of HPMC,PAM,PAC
With the increase of dosage of AWAs from 0 to 1.2%,
and CMS.(fig4)
there is a sharp decline in fluidity of the cement mortar,
specially, at low dosage. However, the decrease in fluidity
of the mortar with non-ionic AWAs is significantly greater
than ionic one. (Fig1) Besides, it can be seen from the
fluidity loss after two hours that there is an opposite
tendency in fluidity loss between two kinds AWAs before
the dosage of 0.9%, and the fluidity loss with CMS is lower
and more stable relatively. (Fig2)Additionally, the fluidity
loss of UWC with AWAs is even less than the non-AWA one
in some dosages, this indicates that AWAs retain water in
UWC and reduce the fluidity loss, or the fluidity of UWC is
so little that it reach the lower limit. In practical test, when
the dosage of non-ionic AWAs reach to 0.9%,the cement
mortar is too sticky to meet a fine workability of UWC, thus,
Fig. 3: Flocculating states of cement paste with different AWAs
the dosage of AWAs is limited between 0.1-0.6% in UWC. In
fluidity test, the average fluidity is decreased by 45.1% and
37.5% with non-ionic and anionic AWAs respectively which
shows that the anionic AWAs have less negative impact on
flow properties of concrete.

pH factor test and stream test

Considering the greater impact of coarse aggregate on
the test result, the anti-washout resistance is tested with

Fig. 4: pH of supernatant solution with different AWAs

Stream test has been developed in Belgium firstly32.

This method simulates the water flow and reduces the
impact of the experimental error, thus it is closer to actual
situation.
The test procedure uses a 70 mm diameter guttering
channel 800 mm long set at a slope of 15° to the
horizontal(fig 5). The sample of cement mortar is placed
300 mm from the raised end of the channel and 100 ml
Fig. 2: The fluidity loss after two hours

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

686 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effects of anionic and non-ionic anti-washout admixtures on the performance of underwater concrete

HPMC and PAM. The anionic groups in anionic AWAs will

adsorb cement particles to enhance flocculation while the
ionic AWAs can only fasten cement particles by adhesion
of a mass of spatial mesh structure. Therefore, the ionic
flocculation is poor and it will impede cement hydration.
Overall, the linear correlation between pH and mass loss
of the sample with CMS is better than others, and anionic
AWAs are better and stable than ionic AWAs.
In comparison of pH factor test and steam test, we find
that though they both have the same tendency on dosage
of AWAs and anti-dispersion effect, the pH factor test is
Fig. 5: Stream test
more unstable. In the horizontal comparison, the results
of steam test reflect the actual mass loss of cement paste
water is poured down the pipe for 3 times. Then the mass
and it is more closed to reflect the actual anti-dispersing
loss of the mortar is measured. Besides, the pH of the
ability of AWAs. The mass loss of paste with HPMC and
turbid
PAM is bigger than CMS and PAC. However, the results of
With the increasing of dosage of AWAs from 0 to 1.2%,there pH factor test are completely opposite to steam test (Fig
is a sharp decline in fluidity of the cement mortar, 5). And the results of pH of the turbid liquids improve a
especially, at low dosage . The curves are leveling off little but still have deviation with actual values.
when the AWAs dosage is over 0.6% (fig6) and the cement
mortars with these three kinds AWAs have the similar
performance in washout resistance. But before dosage of Slump and slump flow of UWC
0.6%, the HPMC perform better relatively. Then the slump and slump flow of the UWC is tested and
the results are arranged in Table3. (slump is the vertical
drop length while the slump flow is horizontal expansion
diameter of the UWC)
At the AWA dosage of 0.6%, concrete with AWA can hardly
flow. Even at the dosage of 0.1%, the fluidity of concrete
is tremendous impacted by AWA, while, without bleeding
or segregation . The comprehensive results show that the
slump and slump flow are affected, in order of importance,
by PAM, HPMC, PAC and CMS. Additionally, PAM is difficult
to adapt to the mixing of concrete.

Table 3
Fig. 6: The mass loss and pH of cement paste with different The slump and slump flow of UWC
AWAs
AWA-Dosage (%) Non-AWA HPMC PAM PAC CMS

With the comprehensive analysis of the results of fluidity 0.1 Slump 205 180 165 195 190
test and washout resistance test, the dosages of AWAs in
Slump flow 345 275 - 290 300
UWC are set at 0.1%,0.3% and 0.6% respectively and the
mix proportions adopted of UWC are arranged in Table2. 0.3 Slump 205 150 145 165 185

The change trends of the pH are similar to mass loss, Slump flow 345 - - 250 290
however, they are more unstable, this is mainly due to
the different flocculated situations. (Fig6)Even though the 0.6 Slump 205 - - 140 175
supernatant of cement slurry without AWA is relatively Slump flow 345 - - - -
clear, the pH is almost over 12. Since there is no binding
effect of AWA, Ca2+ and OH- are more likely to spread
SEM analysis of UWC
easily from cement paste to water and the suspension
layer which is low-strength after hardened is also thicker. A certain amount of cement paste are taken during the
The supernatants with AWAs are more turbid but because process of the UWC mixing, then the samples are dried
of the flocculation of AWAs, those cement particles are immediately and observed by SEM. It can be seen that
bound in flocculation structure so the ion migration is more the sample without AWAs has started hydrating. There
difficult. Further, the amount of those suspended particles are thin layers of Aft and C-S-H gel on the surfaces of
is small, thus the pH of supernatant is decreased. There cement particles (Fig7.a). The sample with PAM has also
are relatively large fluctuations in pH of samples with started hydrating but slower than the one without AWA,

Organised by
India Chapter of American Concrete Institute 687
Technical Papers

because PAM network structures have not fully covered Compressive strength
cement particles and only the boundaries have been Fig 8 shows the compressive strength of the UWC with
covered which result in hydration happened in the middle different dosage of AWAs. Additionally, the Non and the
areas of cement particles(Fig 7.c). The samples with Non(normal) are the controlled trials without AWAs
HPMC, CMS and PAC are almost no hydrated. Most part which are formed under water and the ordinary method
of cement particles have been covered by AWAs ,which respectively. The early strengths (3 day) of the UWC with
causes slow setting of the UWC. Besides, a large number AWAs are generally lower than that without AWAs(Fig
of communication pores have been formed in the sample 8 a.b.c.), but the later strengths are higher. With the
of paste with HPMC(Fig 7.b), and this may be the reason of increasing of the dosages of AWAs, the strengths are
lower compressive strength. The influence of the CMS and decreasing because of the negative effect of AWAs on
PAC on pore structures is less than HPMC relatively (Fig early hydration of cement.[33]
7.d & e).The main adhesion parts are the edges and the
contact points between cement particles, therefore, they From the perspective of reactive groups of AWAs, there
play positive roles for bridging effects in cement paste. To are a large number of hydroxyl groups which can form
sum up, the anionic AWAs can cause slow setting of UWC hydrogen bonds and slow down the hydration rate in
for the wrapping effects, but they have less effect on the CMS. Moreover, carboxyl groups will form film layers
later hydration and pore structure. The ionic AWAs fibrous of complex compounds with Ca2+, and wrap in cement
are more dense and thicker than the anionic one, and the particles’ surfaces, result in further slowing down
fibrous overlap each other to form well-like structures, a
large number of communication pores are presented in
these structures. Besides, this well-like structures will
reduce the surface energy and formation energies of
crystals of cement hydration products, thereby inducing
the growth of crystals along these structures. In the end,
the inside structures of UWC are changed and lots of
harmful pores are formed.

Fig. 7: Scanning electron micrograph of the dried cement paste

with different AWAs and the chemical formulars Fig. 8: The compressive strength of the concrete

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

688 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Effects of anionic and non-ionic anti-washout admixtures on the performance of underwater concrete

the cement hydration. However, when the continuous surfaces of cement. Non-ionic AWAs flocculate mainly
hydration going on, the complex layers is collapsing, and by adsorption, so they have less effective components
hydration products break the layers to make the further and greater negative impact on pore structure and
reaction continue . Although hydroxyl groups are also hydration of cement.
existing in HPMC, the effects of covering and bridging on
2. In the anti-dispersion test, the stream test is more
hydration of cement are too intense to compensate for
stable than pH factor test but pH factor test is more
strength. Besides, too much pores are formed and result
convenient and faster, thus, pH factor test is more
in decreasing of strength of concrete with HPMC. The
suitable for longitudinal comparison on one kind of
compressive strength (28d) of concrete with anionic AWAs
AWA.
can reach up to 37Mpa, increased by 151% and 131% to the
ones with no AWA and non-ionic AWAs. 3. Although the non-ionic AWAs reinforce the concrete
on the viscosity, the UWC is more likely to break up by
To sum up, the CMS is better than the PAM and HPMC on
flow water, and the concrete is difficult to subsidence
workability and strength of UWC.
without electric charge. Accordingly, the water around
The impact of AWA molecular structures and polar will get contaminated easily.
groups on flocculation is quite different.(fig 7.e)This four
4. For the anionic AWAs, the main chain is short and
kinds of AWAs are all linear but the length and shapes of
branched chain is longer, so the molecules are
the molecular chains and functional groups dissociated in
relatively trick, the anionic AWAs are better than non-
water are different. The length of amide group in PAM is
ionic AWAs on workability and strength of UWC.
about 0.15 nm, the main chain is short and branched chain
is longer, so the molecules are relatively trick. Moreover, 5. In the mixing process, non-ionic AWAs generate a lot
there are a content of –COOH and –NH2 groups which have a of bubbles in concrete which are not conducive to the
greater rigidity and toughness, therefore, PAM molecules sinking of UWC.
have a larger influence in fluidity of concrete. HPMC and
6. In fluidity test, the average fluidity is decreased by
PAC have the same main chain but the functional groups
45.1% and 37.5% with non-ionic and anionic AWAs
are different. Molecular weight of HPMC is approximately
respectively which shows that the anionic AWAs have
10,000 to 1,500,000 and the molecular chain can easily
less negative impact on flow properties of concrete.
reunite to a ring-form and its appearance is beam-like
structures with many monofilament fiber collecting. 7. The compressive strength (28d) of concrete with
Thus, the proportion of effective flocculation of HPCM is anionic AWAs can reach up to 37Mpa, increased by
really small while, instead, these structures will affect the 151% and 131% to the ones with no AWA and non-ionic
internal pore structures of concrete. Although PAC has AWAs. 4. Acknowledgements
the same main chain with HPMC, the molecular weight This research work has been made possible thanks to
is only about 17,000 (when the n is about 100). So PAC financing from National Natural Science Foundation of
molecular chains are not easy to be intertwined and the China (No.51172098). It could not have gone ahead without
–CH2COONa on the glucose ring will make PAC molecules the materials supplied by Zonghui Zhou.
with the same charge to repel each other.
The molecular structure of CMS is different with the Reference
others. Firstly, the main chain is short but the branched 1. M. Sonebi , P.J. M. Bartos I , K. H. Khayat,Assessment of washout
chains are longer and softer because they are connected resistance of underwater concrete: a comparison between CRD
C61 and new MC-1 tests. Materials and Structures/Mat6riaux et
by single bonds so the chains are easier to rotate. Besides, Constructions, Vol. 32, May 1999, pp 273-281.
the negatively charges promote dispersion of CMS, the
2. Joseph J. Assaad, Camille A. Issa, Bond strength of epoxy-coated
flocculation structures are more likely to be net-form bars in underwater concrete. Construction and Building Materials,
rather than linear one. Therefore, the proportion of the 30 (2012) 667–674.
effective flocculation is larger. 3. El-Hawary MM. Evaluation of bond strength of epoxy-coated bars
in concrete exposed to marine environment. Constr Build Mater,
1999,13:357–62.
Conclusions
4. Erdogdu S, Bremner TW, Kondratova IL. Accelerated testing of plain
In this paper, we study the effects of different AWAs on and epoxycoated reinforcement in simulated seawater and chloride
the performance of UWC and cement mortar. In the range solutions. Cem Concr Res ,2001;31:86,1867.
of investigation of AWAs, used admixtures and received 5. P.J. Tumidajski, AS. Schumacher, S. Perron, P. Gu and J.J. Beaudoin,
research results it was indicated that: ON THE RELATIONSHIP BETWEEN POROSITY AND ELECTRICAL
RESISTIVITY IN CEMENTITIOUS SYSTEMS ,Cement and Concrete
1. The anionic AWAs have a smaller negative impact on Research, Vol. 26, No. 4, pp. 539-544, 1996
workability of UWC than non-ionic AWAs because the 6. Yeih W, Chang JJ, Tsai CL. Enhancement of the bond strength of
force points of the anionic AWAs are at the junction epoxy coated steel by the addition of fly ash. Cem Concr Compos,
of cement particles while the other one are on the 2004;26:315–21.

Organised by
India Chapter of American Concrete Institute 689
Technical Papers

7. Saric-Coric, M., K. H. Khayat, and A. Tagnit-Hamou. "Performance Park Ridge, NJ, 1995,Chapter 1.5, pp. 994-995.Kawai, T., Non-
characteristics of cement grouts made with various combinations dispersible.
of high-range water reducer and cellulose-based viscosity
21. V. Fernon, A. Vichot, N. Le Goanvic, P. Colombet, F. Corazza, U. Costa,
modifier."Cement and Concrete Research,33.12 (2003): 1999-2008.
Interaction between Portland cement hydrates and polynapthalene
8. Yahia, A., and K. H. Khayat. "Experiment design to evaluate interaction sulfonates, ACI Spec. Publ.173 (1997) 225-248.
of high-range water-reducer and antiwashout admixture in high-
22. C. Jolicoeur, M.A. Simard, Chemical admixture–cement interactions:
performance cement grout." Cement and Concrete Research, 31.5
phenomenology and physico-chemical concepts, Cem. Concr.
(2001): 749-757.
Compos. 20 (1998) 87-101.
9. Sonebi, Mohammed. "Factorial design modelling of mix proportion
23. R.J. Flatt, Y.F. Houst, A simplified view on chemical effects
parameters of underwater composite cement grouts." Cement and
perturbing the action of superplasticizers, Cem. Concr. Res. 31
Concrete Research 31.11 (2001): 1553-1560.
(2001) 1169-1176.
10. Khayat, Kamal H., and Mohammed Sonebi. "Effect of mixture
24. J. Cheung, A. Jeknavorian, L. Roberts, D. Silva, Impact of admixtures
composition on washout resistance of highly flowable underwater
on the hydrationkinetics of Portland cement, Cem. Concr. Res. 41
concrete."ACI Materials Journal, 98.4 (2001).
(2011) 1289-1309.
11. Peng, Ya, and Stefan Jacobsen. "Influence of water/cement ratio,
25. W. Prince, M. Edwards-Lajnef, P.C. Aitcin, Interaction between
admixtures and filler on sedimentation and bleeding of cement
ettringite and a polynaphthalene sulfonate superplasticizer in a
paste." Cement and Concrete Research, 54 (2013): 133-142.
cementitious paste, Cem. Concr. Res. 32 (2002) 79-85.
12. Sonebi, Mohammed. "Rheological properties of grouts with viscosity
26. W. Prince, M. Espagne, P.C. Aitcin, Ettringite formation: a crucial
modifying agents as diutan gum and welan gum incorporating
step in cement superplasticizer compatibility, Cem. Concr. Res. 33
pulverised fly ash." Cement and concrete research 36.9 (2006):
(2003) 635-641.
1609-1618.
27. A. Zingg, F. Winnefeld, L. Holzer, J. Pakusch, S. Becker, L. Gauckler,
13. Łaźniewska-Piekarczyk, Beata. "Effect of viscosity type modifying
Adsorption ofpolyelectrolytes and its influence on the rheology, zeta
admixture on porosity, compressive strength and water penetration
potential, and microstructureof various cement and hydrate phases,
of high performance self-compacting concrete." Construction and
J. Colloid Interface Sci. 323 (2008)301-312.
Building Materials 48 (2013): 1035-1044.
28. Giergiczny Z, Glinicki MA, Sokolowski M, et al. Air void system
14. Shuanfa Chen , Rui He. "Influence of Thickeners on Cement
and frost-salt scaling of concrete containing slag-blended cement.
Paste Structure and Performance of Engineered Cementitious
Constr Build Mater 2009;6(23):2451–6
Composites." Journal of Wuhan University of Technology (Materials
Science Edition) 2 (2013): 016. 29. J. Plank, D. Zhimin, H. Keller, F.V. Hössle, W. Seidl, Fundamental
mechanisms for polycarboxylate intercalation into C3A hydrate
15. K.H. Khayat, Effect of antiwashout admixtures on fresh concrete
phases and the role of sulfate present in cement, Cem. Concr. Res.
properties, ACI Mater. J. 92 (2) (1995) 164– 171.
40 (2010) 45-57.
16. LI, Yong-peng, et al. "Effect of Polyacrylamide and Hydroxyethyl
30. C. Giraudeau, J. D'Espinose De Lacaillerie, Z. Souguir, A. Nonat,
Cellulose Ether on the Performance of the Cement Mortar." Journal
R.J. Flatt, Surface and intercalation chemistry of polycarboxylate
of Wuhan University of Technology 7 (2012): 007.
copolymers in cementitious systems, J.Am. Ceram. Soc. 92 (2009)
17. Rouis, Mohamed Jamel, and Afef Makni. "Influence of adjuvants on 2471-2488.
the properties of underwater cast concrete on base of cement (HRS
31. Takeshi Ohtomo,Yasunori Matsuoka,Yashitaka Nakagawa,etal.
32.5 N)."MATEC Web of Conferences. Vol. 11. EDP Sciences, 2014.
Influence of Materials on tne Action ot Admaxture in Antiwashout
18. Assaad, Joseph J., Yehia Daou, and Hadi Salman. "Correlating Underwater Concrete. ACI Material Joumal. 1995,92(3):315-320.
washout to strength loss of underwater concrete." Proceedings of
32. Japan Society of Civil Engineers, JSCE, 'Recommendations for
the ICE-Construction Materials 164.3 (2011): 153-162.
design and construction in antiwashout underwater concrete,
19. Kamal H. Khayat, Viscosity-Enhancing Admixtures for Cement- Concrete library of JSCE (67) (May 1991) 89 p.
Based Materials - An Overview, Cement and Concrete Composites
33. Davies, B. A., 'Laboratory methods of testing concrete for placement
20 (1998) 171-188.
underwater', Proceeding, Marine concrete 86-International
20. Mailvaganam, N., Miscellaneous admixtures. Concrete Admixtures Conference on Concrete in 'the Marine Environment', London, Sept.
Handbook, 2nd edition, ed. V. S. Ramachandran.Noyes Publications, 1986 (Concrete Society, London,1986) 279-286.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

690 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Basics of Corrosion in Reinforced Concrete

Basics of Corrosion in Reinforced Concrete

F.R. Goodwin
BASF Construction Chemicals, Beachwood, OH USA

Abstract cost of concrete reinforcement corrosion is $125 billion

US per year.
Concrete is the second most common man-made
material after potable water yet it is a complex material
that is poorly understood (1). Steel reinforcement is added Concrete Properties
to improve the concrete's tensile strength and ductility There are two truisms about reinforced concrete:
and is initially protected by the high concrete pH and Concrete cracks and steel rusts. There is no escaping the
depth of cover concrete. Eventually, due to the ingress of inevitability of these facts; eventually both will happen.
deleterious ions, lowering of the pH from carbonation, or Concrete cracks because it is comparatively weak in
electrical potential changes within the steel, corrosion tensile strength. Expansive forces developing within the
will occur. Steps can be taken throughout the concrete concrete or a relatively low tensile stress imposed on the
life cycle to minimize this damage through prevention, concrete will cause cracking. Steel is made from iron ore
protection, or mitigation of corrosion. This paper will by imparting a great deal of energy to refine to iron and
discuss the properties of concrete, the causes of concrete then to steel. Steel will eventually convert to a lower energy
damage and deterioration, issues related to corrosion of state by rusting. The combination of concrete and steel is
reinforcing steel in concrete, and options to reduce the synergistic. A superior composite material is produced
effects of concrete deterioration. using reinforced concrete with the steel providing tensile
Keywords: concrete, corrosion, mitigation, reinforcing strength and the concrete providing fire and corrosion
steel, rust. protection to the steel.
Combining steel reinforcement and concrete creates the
Introduction common reinforced concrete used in the myriad forms
seen today. When steel rusts, the oxide formation creates
Concrete in some fashion has existed for thousands of
an expansive force within the concrete that causes
years based on natural or pozzolanic cement. About 1824
cracking, spalling, and eventual section loss. When
Joseph Aspdin is credited with a patent for “Portland
concrete cracks, the steel is exposed to a combination
Cement” which was a synthetic version of what nature
of factors that accelerate its corrosion leading to further
had already created. Portland cement was named after
cracking of the concrete that if allowed to continue will
the popular building stone from the Portland area of
lead to destruction of the structure.
England. Reinforced concrete was introduced at the Paris
Exhibition and patented in 1867 (). That introduction of steel
into concrete is also when many of our current problems Concrete Composition
with concrete also began. Steel used for reinforcing wants Concrete can be thought of as a hard wet sponge.
to corrode, called “rusting”. Steel begins as iron ore and Technically, it is a complicated composite material
after a great deal of energy is expended, the ore is reduced consisting of aggregates, water and a mixture of hydraulic
to produce steel. Steel is a material at an elevated energy cement hydration products that are mainly calcium
state that will eventually corrode to its original lower silicates, calcium aluminates, and other trace materials.
energy state. Fortunately, concrete has a high pH creating The hydration products consist primarily of calcium
a passivating layer on steel in contact with concrete to silicate hydrate (called “gel” located in capillary pores),
prevent corrosion from occurring. Unfortunately, this calcium hydroxide (called “portlandite”) and ettringite.
passivating layer is eventually destroyed by a reduction in Water is chemically bound in the gel as well as within
pH of concrete, the presence of chloride or other ions, an larger pores around and between the portlandite and
electrical charge imposed on the steel, or a combination aggregate particles. This pore water is saturated with
of these factors. NACE estimates that the total cost of calcium hydroxide and other soluble minerals such as
corrosion is approximately 3% of the USA’s gross domestic sodium and potassium salts and is free to evaporate or
product (NACE 1). This estimate from NACE estimates the move within the concrete. Both the size and quality of the

Organised by
India Chapter of American Concrete Institute 691
Technical Papers

pore structure depends on the amount of water added

to the cementitious binder system (W/CM=cement +
pozzolan+ water) with desirable smaller pores formed
as the mixing water content decreases provided there is
sufficient binder to coat and fill the spaces between the
aggregate particles. At extremely low W/CM (below about
0.3) continued hydration of the cementitious material
may cause autogenous shrinkage from self-desiccation
where capillary gel pore water is consumed as the cement
continues to hydrate.
Concrete needs water for the cement to hydrate to
form these minerals; however water in excess of that
required for proper hydration dilutes the performance
properties of the hardened material. Admixtures are
additives that are added to the freshly mixed concrete
Fig. 1: Carbonation induced generalized corrosion
to modify both the plastic and hardened concrete. These
can include pozzolans that react with the portlandite
slow compared to the movement of gaseous carbon
to reduce the porosity of the concrete, water reducers
dioxide through air filled pores. In water saturated
and superplasticizers that reduce porosity through
concrete, the hydrated cement paste is dissolved at a
improvement of the dispersion of ingredients, and
rate depending on the type and concentration of the
corrosion inhibitors that help stabilize the passivating
aggressive acid present and the removal of reaction
layer between the reinforcing steel and the concrete to
products through abrasion or flow of the liquid to expose
extend the time until corrosion begins(3).
fresh unreacted concrete to further acid attack. In
In addition to these improvements in the concrete extremely dry concrete there is insufficient moisture at
properties, how deeply the reinforcing steel is embedded the surface of the concrete pores for the formation of
within the concrete is a significant factor in the amount of carbonic acid and its reaction with portlandite to form
time before corrosion begins. This is called “cover”. Cover calcium carbonate and water at a measureable rate.
is necessary for the steel to create composite action The maximum rate of concrete carbonation occurs at an
within the concrete to properly function as designed. internal relative humidity of about 60%. Wetting a drying
Furthermore, cover insulates the steel from exposure of the concrete may accelerate carbonation; however the
to fire and deterioration from deleterious materials in formation of calcium carbonate may also tend to block
the external environment. Two of the most common some very fine pores reducing this rate. (4) Carbonation
deleterious materials are carbon dioxide and chloride induced corrosion is usually generalized corrosion where
ions. much of the reinforcement begins to corrode at nearly
the same time and is shown by generalized spalling of
Carbonation the concrete cover.
As mentioned above, when cement hydrates, both CSH gel
and portlandite are formed. Both of these materials are Chlorides and Other Aggressive Ions
strongly alkaline. Steel in such an alkaline environment The presence of chloride (or other halide and aggressive
forms an insoluble, protective, passivating oxide layer ions) also destroys the passivating film on the surface of
that is one of the synergistic benefits from combining the steel; however this process is much more localized
concrete with steel. The high pH of concrete will fall as the than the generalized corrosion from carbonation.
concrete ages, mainly due to reaction of the portlandite Chlorides may be present from the materials used to
with carbon dioxide (CO2) in the humid environment of make the concrete or migrate into the concrete from
concrete. This process is known as “carbonation” and applied road salt or air borne chlorides from nearby
occurs either from acid rain or CO2 gas diffusing into salt water. Some chloride is captured by the concrete
the concrete and reacting at the surface of the pores hydration products, and may remain bound until the pH
to form calcium carbonate. When this carbonation or temperature reaches some threshold. ASTM C1218
front approached the reinforcing steel, the protective measures water soluble chlorides, or the concentration of
passivating layer at the steel surface in the concrete to chloride that has not been chemically bound by the cement
dissolves and corrosion initiates. Carbonation is a slow hydration products. Total chloride is measured by using
process depending on the porosity of the concrete, the an acid soluble chloride test such as ASTM C1152 which
thickness of the cover, the concentration of pH reducing measures the worst case scenario. Chloride ions react
materials, and the amount of moisture in the concrete. In as shown below and are not consumed in the corrosion
very wet concrete, the diffusion of acidic materials into reaction:
the concrete occurs within liquid filled pores and is quite

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

692 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Basics of Corrosion in Reinforced Concrete

the anode readily travel through the electrical path to

the cathode; in concrete this is the reinforcing steel. Ions
formed at the cathode migrate through an electrolyte; in
concrete this is the water in the concrete pores. Concrete
may readily conduct ions if it is moist and has many
pores. Concrete is a poor conductor of ions if it has low
internal relative humidity or a very dense pore structure
Fig. 2: Chloride induced corrosion reaction (such as from a low W/CM and the use of pozzolanic
materials). Usually the ability of concrete to conduct ions
By remaining present to react further, a pitting-type of is measured by determining the resistivity of the concrete.
corrosion occurs. As the pit grows deeper, the acid formed Resistivity is a bulk material property where the electrical
by the splitting of water further accelerates corrosion resistance is measured over a given volume of concrete.
causing pits to form deeper and deeper into the reinforcing Drying of concrete once it has properly cured, in addition
steel.(4) to reducing the tendency to reinforcement corrosion can
also minimize deterioration from freezing and thawing
Cracks cycles, alkali aggregate reactions, and sulfate attack.
Most deterioration, damage, or failure of concrete is The anode and cathode in a corrosion cell may be adjacent
seen by cracking. Cracks are both the cause of these to each other, have different surface areas, or be widely
issues as well as the observable effects. Since concrete separated. Both the anode and cathode are free to move
is comparatively weak in tension, expansive forces in the to different locations depending on the conditions of the
concrete or tensile stress imposed on the concrete can reaction. Macrocell corrosion is usually described if
easily exceed its tensile strength. Cracks follow the path the anode and cathode are widely separated so that the
of greatest weakness through the concrete beginning electrical current generated by the corrosion reaction can
at small defects and propagating as stress increases. be measured such as between the two layers of reinforcing
New cracks form once the initial crack meets sufficient steel within a concrete slab. Microcell corrosion is where
resistance, the path of the stress changes, or the ingress of the anode and cathodes are very close to each other (such
deleterious materials causes new stress to develop. Once as on the same reinforcing bar) and corrosion current
a crack forms it becomes a “freeway” for deterioration by between them cannot be directly measured.
funneling water and dissolved minerals deeper into the Like most other chemical reactions, elevated
concrete. When water freezes in the crack, the expansion of temperatures accelerate the reaction and greater
ice formation forces the crack faces to become wider. When concentrations of the reacting materials such as chlorides,
water evaporates from a crack, the dissolved minerals moisture, and oxygen also increase the rate of corrosion.
(such as chlorides) are left behind. Road salt, dust, and Ionic movement essentially ceases during freezing and
debris also accumulate within the crack keeping it open for corrosion therefore stops.(5) The ratio of the anode to the
further infiltration. When the crack inevitably reaches the cathode areas are also very important in the rate of the
reinforcing steel, corrosion begins in a localized area. The corrosion reaction, with small anodes and large cathodes
expansive forces from the reinforcement corrosion further creating a much more intense oxidation of the steel in a
widen the crack as well as create new cracks. small area. In cracked concrete, when the crack intersects
a reinforcing bar the corrosion is very localized by this
Corrosion Principles difference in anode to cathode area. Furthermore, the
Corrosion is an electrochemical process. This means concentrating of chlorides, increased oxygen availability,
that both chemical reactions and electrical processes wet to dry cycles, and susceptibility to carbonation within
occur simultaneously. Usually, the electrical process the crack create circumstances favoring accelerated
involves ions changing their valence state (gaining and corrosion at the crack to reinforcing steel intersection.
losing electrons also known as oxidation and reduction or The expansion of corrosion product formation from the
redox reactions). Chemical reactions form new chemical corroding steel causes further cracking and the cycle
compounds. continues until either the structure disintegrates or action
is taken to intervene in the corrosion cycle.
For corrosion to occur, four components must be present:
an anode, a cathode, an electrical path, and an ionic path External sources of current such as from dissimilar metal
(or electrolyte). In reinforcing steel, rust forms at the anode (or galvanic) corrosion of more noble metals in contact
where electrons are generated and move away towards with the steel or stray current leakage can also drive
the cathode and oxidation occurs. An equal reaction must reinforcing steel corrosion reactions. More noble metals
also be present at the cathode where the electrons are such as copper pipes will cause the steel to corrode. Less
consumed by forming ions (in concrete hydroxyl ions are noble metals such as aluminum railings or conduit in
formed) and reduction occurs. The electrons formed at contact with reinforcing steel will corrode expansively as
well leading to cracking.

Organised by
India Chapter of American Concrete Institute 693
Technical Papers

Fig. 3: The four parts of a corrosion cell and methods to address them in concrete

Anti-Corrosion Treatments Alternative Reinforcement

Interruption or restriction of any of these the four corrosion The most effective method to prevent reinforcing steel
components (anode, cathode, electrical continuity path, corrosion is to substitute a reinforcing material that is
and ionic path) will slow down the rate of corrosion as less prone to corrosion, such as stainless steel, other
shown in Fig. 3. This is the basis for all treatments to specialty alloys, epoxy coated reinforcement, or fiber
address corrosion of reinforcing steel in concrete. The reinforced polymer materials (FRP). With no material to
corrosion treatment methods described in this paper corrode (changing the anode material), the root cause
address one or more of these treatments. of corrosion is removed. Far too often, the steel that is
used for reinforcing concrete is already significantly
Another way of looking at corrosion treatments is corroded or has residual mill scale from manufacturing
described in figure 4 below. Prevention of corrosion to create cathodic and anodic areas before installation.
either changes the reinforcement to more corrosion While a light surface rusting on conventional reinforcing
resistant materials or creates an environment within steel is usually not detrimental, heavy rusting gives
the concrete where corrosion is much less likely to corrosion a head start. Corrosion products that are not
occur through the use of admixtures. Protection creates removed prior to repairs also will provide opportunities
a barrier between the reinforcement and the corrosive for corrosion to resume. Caution should be exercised
environment through the use of coatings or increased using alternative reinforcing materials as the structural
cover. Mitigation treatments attempt to address development and ductility of the reinforcement may be
corrosion after its initiation. Cathodic protection can different or the alternative material may be sufficiently
be used either preventatively, protectively, or as a electrochemically dissimilar to conventional steel to
mitigation technique. Fig. 4 describes this classification initiate galvanic corrosion. Galvanized reinforcement
according to the stage of corrosion occurrence. The may develop expansive corrosion products. Epoxy coated
following sections briefly discuss many of these reinforcement works well in laboratory testing, however
corrosion treatment technologies. construction damage frequently causes small defects
in the coating creating anodic hot spots surrounded by
large areas of reinforcement that are cathodic due to the
intact coating. FRP materials may be susceptible to fire
damage, trap moisture within the concrete and produce
brittle failure. Often alternative reinforcing materials
are significantly more expensive than conventional
reinforcement.

Corrosion Inhibitors
Inhibitors may be added to the concrete either as
admixtures when the concrete is placed or surface
applied to make the passivating film on the steel surface
more resistance to chloride attack, dissipation from
carbonation, or otherwise more durable in adverse
conditions. The inhibitor concentration required for
effectiveness is often dependent on the extent of corrosive
conditions for the steel. Corrosion inhibitors that are used
as admixtures can be precisely controlled for dosage,
pretested for compatibility in the material and once used
are in intimate contact with the steel to be protected.
Fig. 4: Corrosion treatments for concrete classified according Surface applied corrosion inhibitors for hardened concrete
to the stage where they are applied are intended to penetrate into the concrete to reduce

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

694 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Basics of Corrosion in Reinforced Concrete

corrosion activity. Corrosion inhibitors are concentration conversion layer formers, and scavengers (). Adsorbed
dependent. The amount of corrosion inhibitor depends on layer formers are organic inhibitors which strongly
the level of corrosion activity. Some of these inhibitors can adsorb onto the metal surface and interfere with the
increase the rate of attack in unprotected areas, similar anodic or cathodic reactions in the area of adsorption.
to the haloing effect which is sometimes produced around Nitrogen is usually the active atom in an adsorbed layer
concrete repairs due to the increase in the cathodic area inhibitor acting in a non-acid electrolyte on steel. Typical
after repair. Because of this concentration dependency and compounds of nitrogen used as inhibitors are organic
effects on adjacent areas, inhibitors with these properties nitrate and amines. Vapor phase inhibitors (VPI or Volatile
are sometimes called “dangerous”. ASTM C1582 is a Corrosion Inhibitor VCI) are similar to adsorbed layer
specification for admixtures used as corrosion inhibitors. inhibitors. Volatile Corrosion Inhibitors are secondary
ASTM G109 is a widely used test method for evaluating electrolyte layer inhibitors that possess appreciable
the effectiveness of corrosion inhibiting admixtures and vapor pressure under atmospheric conditions, thus
various modifications have been used for SACI and VCI allowing significant vapor phase transport of the inhibitive
types. No standards exist to specifically address SACI substance. Aliphatic and cyclic amines and nitrites with a
and VCI corrosion inhibitors for concrete with much of the high vapor pressure typically make up these inhibitors.
published literature presenting differing opinions about VPI/VCI inhibitors can be added to cementitious materials
their effectiveness. as admixtures as well being used as Surface Applied
Corrosion Inhibitors (SACI). VCI and VPI ideally provide
Most corrosion inhibitors are water soluble with some
inhibition rapidly while lasting for long periods. Both
inhibitors also being volatile so the concentration of
qualities depend on the volatility of these compounds, fast
inhibitor may change with time. The corrosion activity will
action wanting high volatility while enduring protection
also change as concrete deterioration occurs, chloride
requires low volatility. Oxidizing inhibitors or passivators
ions ingress, and carbonation occurs which may require
are another form of barrier inhibitor which act by shifting
a higher concentration of inhibitor than what is within
the electrochemical potential of the corroding metal such
the new concrete. Corrosion is a complex series of
that an insoluble oxide or hydroxide forms on the metal
electrochemical processes in a changing environment
surface. Sodium nitrite and chromates are examples of
that can produce varying results with inhibitors in field
this inhibitor. Another form of passivators are metal soaps
conditions. Several classification systems of corrosion
which are a form of the basic pigments (metal oxides)
inhibitors are discussed in the following paragraphs.
and oxidation products of oil which form passivating
ASTM C125 defines an admixture as “a material other films on the metal surface. Conversion layer inhibitors
than water, aggregates, hydraulic cementitious material, form insoluble compounds on metal surfaces without
and fiber reinforcement that is used as an ingredient oxidation. In neutral or basic solution, the presence of
of a cementitious mixture to modify its freshly mixed, calcium and magnesium ions inhibits corrosion by the
setting, or hardened properties and that is added to formation of an insoluble calcareous scale on the metal
the batch before or during its mixing”(). ICRI Concrete surface. Finally, scavengers act as neutralizing inhibitors
Repair Terminology defines a corrosion inhibitor as a by removing concentrations of corrosive materials such
chemical compound which, when used as an admixture as chloride ion.
in fresh concrete or as a topical application to hardened
NACE classifies corrosion inhibitors (all types, not just
concrete, inhibits corrosion of embedded metal (). Several
those specific to concrete applications) as passivating
classification systems are discussed below.
inhibitors, cathodic inhibitors, organic inhibitors,
The simplest classification of corrosion inhibitors divides precipitation inhibitors, and volatile corrosion inhibitors
them into three types: cathodic, anodic, and ambiotic (Nathan). Passivating inhibitors (passivators) cause
(or mixed) inhibitors depending on their area of activity a large anodic shift of the corrosion potential, forcing
on the corroding steel. Cathodic inhibitors act indirectly the metallic surface into the passivation range and are
to prevent the corrosion reaction at the cathode rather further divided into oxidizing anions and non- oxidizing
than metal dissolution. Anodic inhibitors reduce the rate ion types. Cathodic inhibitors either slow the cathodic
of metal dissolution reactions at the anode. They usually reaction itself or selectively precipitate on cathodic areas
react with the corrosion products to form a protective to increase the surface impedance and limit the diffusion
coating on the metal surface. Nitrite based corrosion of reducible species to these areas and are further
inhibitors are one common type of anodic inhibitor used classified as cathodic poisons and oxygen scavengers.
in concrete. Mixed (or ambiotic) inhibitors influence both Organic inhibitors can show both anodic and cathodic
the anodic and the cathodic sites which are claimed to be effects but, as a general rule, affect the entire surface of a
especially advantageous in reinforced concrete due to the corroding metal when present in sufficient concentration.
prominence of micro-cell corrosion. Organic inhibitors usually designated as 'film-forming',
to protect the metal by forming a hydrophobic film on the
Inhibitors are classified by SHRP 666 as follows:
metal surface. Precipitation inducing inhibitors are film
adsorbed layer formers, oxidizing inhibitors, passivators,

Organised by
India Chapter of American Concrete Institute 695
Technical Papers

forming compounds that also have a general action over ring anode effect. Typical reinforcing steel coatings (“rebar
the metal surface, blocking both anodic and cathodic sites coatings”) are pure epoxy, zinc rich, epoxy/cementitious
indirectly by forming a precipitated deposit on the surface hybrids, and dispersion polymer (“latex”) cement blends.
of the metal to provide a protective film. Volatile corrosion Most information on these materials is provided by
inhibitors (VCI), also called Vapor phase inhibitors (VPI), material suppliers for their individual products. There are
are compounds transported in a closed environment to no known industry standards.
the site of corrosion by volatilization from a source. On
contact with the metal surface, the vapor of VCI salts Oxygen Availability
condenses and is hydrolyzed by any moisture to liberate Oxygen availability can also inhibit corrosion such as in
protective ions. deeply immersed structures or where treatments are
applied to the reinforcement surface or to completely
Drying and Barrier Treatments encapsulate the concrete. Once the available oxygen
Most deterioration mechanisms of concrete involve the is consumed by corrosion, the rate of the reaction will
presence of water (freeze/thaw deterioration, alkali reduce and be controlled by the ingress of oxygen.
aggregate reaction, sulfate attack, as well as corrosion). However, areas immediately adjacent to these inhibited
Materials may be applied to dry the concrete and increase areas may experience accelerated corrosion, since the
the resistance to ion flow (characterized as resistivity) oxygen depleted areas are cathodic compared to the
such as penetrating sealers or breathable coatings. exposed sections that become anodic.
Since chlorides move through the concrete with liquid
water, ingress of these deleterious ions also slows Electrochemical Treatments (Galvanic and Impressed
as the concrete dries and further water penetration Current Cathodic Protection, Electrochemical Chloride
is inhibited. Penetrating sealers based on silane and Extraction, and Electrochemical Realkalization)
siloxane technologies are simple to apply, do not change Al source of electrical power can reverse the corrosion
the appearance of the concrete, and still allow the escape reaction by changing the direction of electron flow
of water vapor. Barrier materials such as waterproofing through the steel to reverse the flow of ions from the
membranes and anti-carbonation coatings may also be anode to the cathode. This is the principle of cathodic
used to stop water or carbon dioxide from penetrating protection. There are two types of cathodic protection:
into the concrete, but may also be sensitive to the amount galvanic and impressed current. (Whitmore) In galvanic
of moisture present in the concrete during application. cathodic protection, a competing corrosion reaction is
Non-breathable barrier materials may trap water in the used to outpace the oxidation of the steel through the use
concrete leading to deterioration from other causes (such of a sacrificial anode material that then makes the steel
as freezing and thawing damage and alkali aggregate the cathode as it corrodes. Galvanic cathodic protection
reaction). In addition may coatings and membranes comes in two types. Discrete galvanic anodes provide
experience difficulties in adhesion and durability due to the localized protection from changes in the anode and
effects of moisture within the concrete as discussed in ACI cathode locations caused by repairs (also known as the
302.2R and other sources (Kanare). Any sealer, coating, or incipient anode, ring anode, or halo effect). Distributed
membrane is subject to deterioration and should have an galvanic anode systems use a mesh (or other forms) of
inspection and maintenance program to ensure continued galvanic anode material installed next to the reinforcing
effectiveness. In addition, barrier coatings can provide steel and metallization where a galvanic material is
protection to substances that chemically attack concrete. sprayed onto the surface of the concrete to sacrificially
Use of protective systems for concrete is discussed in ACI corrode. In all cases the galvanic material must be in
515.2R (). electrical contact with the reinforcing steel as well as
have an electrolyte with the ability to permit the passage
Reinforcing Steel Coatings of ions between the anode and the cathodically protected
Coatings may also be applied to the reinforcing steel to steel. With galvanic systems producing their own power
create a barrier between the steel and the concrete. and corroding due to many of the same factors influencing
If perfectly applied over the entire steel surface these the corrosion rate of the reinforcing steel, these systems,
materials can be quite effective, however even a small although consumable, are relatively maintenance free
defect in the coating can create strongly concentrated throughout their lifecycle and are self- regulating. The
corrosion from the formation of a small anode to large oxidation products from galvanic cathodic protection
cathode ratio causing rapid deterioration. The same systems need to be considered as they can deposit on the
situation exists with reinforcement coatings applied during anode surface to passivate the anode or create expansive
repairs where the reinforcement continues through the forces if embedded within the concrete to cause cracking.
repair into the original concrete. Corrosion in this case Most discrete galvanic anodes embedded into concrete
may be accelerated at or near the bond line of the repair are therefore encased in a porous mortar containing
material to the original concrete further enhancing the proprietary additives to help prevent passivation and

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

696 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Basics of Corrosion in Reinforced Concrete

allow dissolution of the corrosion products. In addition,

the repair material surrounding the embedded anode
influences the effective protection of reinforcing steel,
or “throwing power.” This “throwing power” is controlled
by the resistivity of the repair mortar and encasement
mortar as well as the distance the anode is located from
the reinforcing steel as it depends on the resistance to
ionic flow.
It is a common practice to repair spalled and delaminated
areas of concrete, often referred to as the “chip and
patch” technique. In this technique, the damaged concrete
is removed, the reinforcing steel cleaned and the area
patched with a cementitious repair mortar, either with or
without a coating on the steel reinforcement. Repairs like
this often result in heightened corrosion in the reinforcing
steel adjacent to the repair area. This phenomenon is Fig. 6: An example of an incipient anode influenced repair on
often referred to as the ring anode, halo or incipient anode a bridge deck. The area being removed is between two previ-
effect. The incipient anode is manifest as an accelerated ous repairs and adjacent to several previously repaired areas.
corrosion of the reinforcing steel immediately adjacent
to newly patched concrete, resulting in cracking and In impressed current cathodic protection (ICCP), an
spalling around the new repair. This gives the appearance external electrical source of direct current is applied
of a “halo” surrounding the area previously repaired. to embedded or surface mounted anodes to also force
When reinforcing steel corrodes in concrete, some areas the steel to become cathodic. Acidic conditions form at
corrode more rapidly than others owing to differences in the anode so the materials for the anodes and in their
steel composition, chloride content in the surrounding proximity require acid resistance with good electrical
concrete, cracks in the concrete and other factors. These conductivity. Just like galvanic anodes, the resistance
corroding areas cause the adjacent reinforcing steel that is to ionic flow determines the “throwing power” of the
nearly ready to corrode to be cathodically protected by the impressed current system, however controlling the
actively corroding anodes. The corrosion soon causes the reinforcement polarization or current may be controlled
embedding concrete to crack and repairs are made to the from the impressed current power supply.
damaged area. The reinforcing steel that was corroding The voltage and current of this impressed current may
is now encased in a comparatively chloride free and high also be optimized to repel chlorides from the steel (known
pH environment and therefore becomes cathodic. The as electrochemical chloride extraction) or generate
areas of the reinforcing steel that previously were being alkaline conditions at the steel to reform the naturally
protected by the corrosion of the area prior to repair now occurring passivating layer (known as electrochemical
begin to corrode, since they are anodic compared to the realkalization) typically using temporary surface
newly embedded reinforcing steel in the concrete. This mounted anodes connected with a conductive pulp().
causes accelerated corrosion immediately next to the ICCP systems must be connected to all metallic objects
newly repaired area as shown in Figure 5 and Figure 6. In in the concrete to prevent them from corroding. They
Figure 6, the area being removed is between two previous must be designed to produce a relatively uniform current
repairs and adjacent to several previously repaired areas. distribution throughout the concrete structure in changing
Differences in chloride content, pH, reinforcing steel environmental conditions or otherwise some areas of
composition, permeability and ionic conductivity between steel may corrode and other areas experience issues
the base concrete and the new repair create electrical from excessive current flow (such as anode consumption,
potentials that drive these new corrosion cells at the excessive alkalinity, etc.). The ICCP systems must be
interface. robust to remain operational throughout the lifetime of
the structure. For these reasons, ICCP systems must be
designed, installed, maintained and monitored correctly
to ensure that corrosion is prevented or controlled and are
somewhat expensive throughout their operation.

Conclusions
In summary, reinforcement corrosion will be with us for
the foreseeable future. Although technologies to prevent
reinforcement corrosion in concrete exist, they are rarely
Fig. 5: Incipient anode, ring anode, halo effect considered as economical alternatives compared to
ignoring the problem until rust is coming out of cracks
Organised by
India Chapter of American Concrete Institute 697
Technical Papers

in the concrete. Sound practices using low permeability 2. The principle causes of reinforcing steel corrosion
concrete and proper cover in new construction and in concrete are carbonation, chloride ingress, and
maintaining existing structures to mitigate corrosion polarization effects.
before significant deterioration occurs are well known,
3. Concrete deteriorates mainly by cracking. Cracking of
however since corrosion develops years after a structure
concrete increases corrosion activity.
goes into service ignoring the problem until it is too late is
likely to continue. 4. There are many alternatives for addressing concrete
reinforcement corrosion, none of which is perfect.
1. Concrete reinforcement corrosion is an increasing
problem with significant economic and societal impact. 5. Ignoring the problem of concrete reinforcement
corrosion does not solve the problem.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

698 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Bi-Directional Electro-Migration Rehabilitation for RC durability improving and its application

Bi-Directional Electro-Migration Rehabilitation

for RC durability improving and its application
Weiliang Jin, You Zuo, Jiayun Chen
Institute of Structural Engineering, Zhejiang University, Hangzhou 310058, P.R. China Ningbo
Institute of Technology, Zhejiang University, Ningbo 315100, P.R. China
Jianghong Mao
Institute of Technology, Zhejiang University, Ningbo 315100, P.R. China
Chen Xu, Jin Xia
Institute of Structural Engineering, Zhejiang University, Hangzhou 310058, P.R. China Ningbo

Abstract and mine，the chloride attack is the primary cause of

The reinforced concrete structures in marine environment reinforcement corrosion.
are influenced by chloride attack, of which the durability Currently, there are four methods to repair and extend
problem is prominent. To improve the durability of the the life of the reinforced concrete structure: traditional
existed reinforcement concrete structures, kinds of repair method, migratory corrosion inhibitor, electro-
methods are widely used worldwide. Researches have chemical desalination technology and electro-migration
been done to approve that electro-chemical desalination technology.
technologies can remove the harmful irons from concrete
For the concrete structure where rust cracks and cracking
effectively. The main methods of electro-chemical
spall appear, the traditional repair methods is chiseling
desalination technologies include cathodic protection,
the deteriorated concrete cover, and clean the rust on the
electro-chemical chloride extraction, re-alkalization and
steel. However, over the last decade, corrosion inhibitor
bi-directional electro­chemistry. In this research, short-
is widely used as a simple, cost-effective steel corrosion
term and long-term effects of BIEM with TETA corrosion
protection method,mostly in incorporation-type projects.
inhibitor were studied. The results demonstrated that
after BIEM the electro-chemical parameters were stable With increasing work to be repaired, migratory corrosion
and the specimens showed strong resistance to chloride inhibitor has developed and begun to be applied in
ion penetration. The BIEM has been applied in Zhoushan concrete structures. Bavarian studied the representative
Cross-sea Bridge successfully. The practical test results migratory corrosion inhibitor such as MCI2020 and
showed the high efficiency of chloride emigration and MCI2020M and found that after coating the surface, the
corrosion inhibitor immigration, while the durability corrosion rate declined and the presence of N in the
evaluation showed the life of the bridge could be improved steel surface(Bavarian et al, 2004) could be detected.
15-20 years. Söylev found that the amino alcohol and esters corrosion
inhibitor coating on the surface do not adversely affect
Keywords: chloride attack, BIEM, corrosion
the performance of concrete in the study (Söylev et al,
inhibitor,migration efficiency,durability evaluation
2007). As the pores of the concrete surface are sealed, the
water absorption and the frost resistance of concrete are
Introduction improved. Nmai used the methods such as the electro-
The reinforced concrete structure is widely used in chemical impedance spectroscopy, Fourier infrared
the industrial and civil construction because of its spectroscopy and linear polarization to confirm the
well performance. However, as time went by, under presence of the steel surface protection film which can
the environment effect which can’t be ignored, the reduce the corrosion rate effectively (Nmai, 2004).
durability issue of reinforced concrete structure
But research indicates the penetration depth of the
is increasingly prominent. In the 2 international
nd

migratory corrosion inhibitor is closely related to the

conference of concrete durability in 1991, Pro. Mehta thickness of the concrete cover, and the compaction degree
gave a report. The report pointed that if we ranked of concrete. When the concrete cover is thick or dense,
the reasons for concrete damage in the descending the corrosion inhibitor can not reach the steel surface or
order of their importance, it would be: reinforcement the inhibitor can not play the desired effect for the lack
corrosion，freeze-thaw damage，physical and chemical of the inhibitor concentration nearby. In addition, studies
reactions under erosional environment. For reinforced have found that it takes almost a year for the penetration
concrete structure under chloride attack, such as the depth of migratory corrosion inhibitor to reach 50 mm
harbor wharf，cross-sea bridge，offshore platform in marine concrete structures. Thus, for thick concrete

Organised by
India Chapter of American Concrete Institute 699
Technical Papers

cover and the high density of marine concrete structures, how to prolong the service life of concrete structures
a migratory corrosion inhibitor is not ideal technology for under chloride erosion has become one of the important
repairing and life extension. research topics.
Electro-chemical desalination technology is a fast, Although the chloride ions in concrete can be excluded
effective, cost-friendly and non-destructive repairing by electrochemical desalination technology effectively,
and life extension technology for reinforced concrete however, only external factors can be eliminated. In
structure. The basic principle is to remove the chloride other words, the steel corrosion will continue after the
ions in the concrete cover outward in an applied electric electrochemical desalination,if the steel is still in an active
field, so as to achieve the removal of chlorine ions. state,
Fan’s study found that electro-chemical desalination Inspired by corrosion inhibitor electro-migration
technology can effectively remove the chlorine ions in researches in the carbonation concrete, researches on
concrete, but experimental parameters have a great the immigration of corrosion inhibitors with the emigration
impact on the desalination effect (Fan et al, 2005). Jiang of chloride ion are studied. It will not only eliminate the
found that, under the same conditions, the desalination external factors that cause steel corrosion, but also
effect is different as the water-cement ratio of concrete prevent corrosion of steel continuing. This method named
changes: it takes a long time for concrete with low
bi-direction electro-migration technology (BIEM) showed
water-cement ratio to achieve the desired effect (Jiang
high efficiency of the improvement of concrete structure
et al, 2008). Fajardo and other studies have found that
durability under chloride attack.
the efficiency of electro-chemical desalination and the
thickness of concrete cover are related; thicker the
concrete cover, lower the efficiency. (Fajardo et al, 2006).
Nzeribe found that the efficiency of desalination is related
to the total reinforcement rate in the concrete specimens
(Ihekwaba et al, 1996).
Electro-chemical desalination have an impact on all
aspects of concrete performance. Guo found that, after
the electro-chemical desalination, density and resistance
to chloride ion permeability increase, but the steel and
concrete bond strength decrease slightly (Guo et al, 2008).
Using the Ca(OH)2 as the electrolyte, Wang found that
internal total porosity of the concrete and the small holes Fig. 1: Technical Mechanism of BIEM
near the steel increased, while the outer porosity and the The technical mechanism of BIEM is shown in figure. 1.
large holes reduced. (Wang el al, 2005). Lu’s experiment The internal structure of reinforced concrete can be
showed that, if the concrete itself had a certain amount treated as a cathode, while the laying stainless steel
of alkali aggregate activity, it would accelerate the alkali
mesh on the external surface of the concrete structure
aggregate reaction; but if lithium ion was added to the
is the anode. On the outside of the stainless steel mesh
electrolyte, AAR could be suppressed (Lu et al, 2002). In
a sponge layer containing corrosion inhibitors is set to
2009, Tang applied this technology in naval pier tentatively
apply the DC power. It can form an electric field between
and achieved the desired effect (Tanget al, 2009).
the steel and stainless steel mesh. Under the electric
As can be seen, the technology can effectively remove the field, the chloride ion with negative charge will emigrate
harmful chlorine ions, increase the pH of concrete pore outward the concrete while the corrosion inhibitors ions
solution, and solve the problem of corrosion of steel well. with positive charge in the sponge layer will immigrate
It will not damage the concrete cover; the processing into the concrete. When the concentration of the corrosion
time is short; and no long-term maintenance is needed. inhibitors of the steel surface reach a certain value, a
But when the chloride content is high or concrete cover is dense protective film, of which the main components are
thick, it takes longer time and it needs larger energizing corrosion inhibitors, will form around the surface of the
current density,which causes negative impact on concrete steel. This film will isolate the chloride ion, oxygen and
structure. other corrosive media from the steel, so as to prevent the
steel from corrosion.
Technical Mechanism of BIEM
During electricity,the anode reactions are as followed:
For the reinforced concrete structures in the marine
environment, the chloride ions would accumulate around 2H2O → O2 + 4H+ + 4e- ...................................................(1)
the steel surface. Corrosion of steel would happen since ............................................................(2)
2Cl- → Cl 2 + 2e-
the concentration of chloride ions reached to the threshold
value resulting in the early failure of structures. Hence, 4OH- → 2H2O + O2 + 4e- ..............................................(3)

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

700 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Bi-Directional Electro-Migration Rehabilitation for RC durability improving and its application

The cathode reaction are as followed• calciumhydroxide, oxygen etc.;

2H2O + 4e- → 2OH- + H2 ...............................................(4) (4). Remaining long enough in the concrete under normal
..............................................(5) environmental;
O2 + 2H2O + 4e → 4OH
- -

(5). Economic, environmental-friendly and can be

As it can be seen from the anode reaction equations, the
applied in engineering widely.
hydroxide ions in the anolyte reduce with the increase of
power-on time, while hydrogen ions increase with the Thus, according to the requirements above, we can put
decrease of pH of the anolyte. Corrosion in the concrete alcohol amines apiary as the target of electro-migration
surface will occur and affect negatively if the pH value type corrosion inhibitor.
is too low. Therefore, the pH value should be observed
during the power-on time. When the pH drops below 7, the Primary selection of amine corrosion inhibitors
electrolyte should be replaced. Besides the basic requirements, solubility, volatility,
As can be seen from the cathode reaction equations, the toxicity, stability and dissociation constants should be
oxygen near the steel will reduce with the increase of taken into consideration.
the power-on time and a large amount of hydroxyl ions Based on the above requirements, six amine-based
and hydrogen will generate. Because of the formation corrosion inhibitors were selected for further study:
of hydroxyl ions, the pH of concrete pore fluid near the Triethylentetramine:(CH3)2-NH, Dimethylamine: NH2-
reinforcement rises, contributing to the passivity of the CH2CH2-(NHCH2)2-NH2 N, N-dimethylethanolamine
steel. Meanwhile, the hydroxyl ions migrate outward in (CH3)2(OH-CH2CH2)-N, Ethanolamine: (CH3)2-NH,
the electric field. During the migration process, calcium 1,6-hexanediamine: H2NCH2(CH2)4CH2NH2, Guanidine:
hydroxide will form with calcium ions in the concrete pore HN=C(NH2)2.
solution. The calcium hydroxide will not only attach in the
concrete pore wall but also narrow the pores in concrete The six inhibitors were studied both in chloride-
or even block them. However, if the power-on time is too contaminated simulation concrete pore solutions and
long or the current density is too large, numerous calcium chloride-contaminated reinforced concrete specimens.
hydroxides will deposit in the vicinity of the steel. It may The permeability of six amine-based inhibitors,
affect the bond strength of concrete and steel, and induce effectiveness of chloride removal and the change of
the risk of alkali aggregate reaction. At the same time, with carbon steel were tested by organic elemental analyzer,
the generation of hydrogen, the hydrogen embrittlement rapid chloride test (RCT) , electrochemical impedance
of steel may be caused. And the generation of hydrogen spectroscopy (EIS), weak polarization method and
will increase the number of holes near reinforcement. scanning electron microscopy (SEM).

When selecting the power-on time and the current Screening of electro-migration type of corrosion
density during BIEM technology. In addition to the factors inhibitor in simulated concrete pore solutions
of the concrete structure itself, the negative impact on The resistance to rust of six amine-inhibitors in chloride
the concrete structure caused by electricity should be environment was studied. The influence on corrosion
considered as well. While ensuring the penetration of the of steel was studied by potential dynamic polarization
corrosion inhibitors and the exclusion of the chloride ion, and EIS in simulated concrete pore solution of chlorine-
negative effects caused by electricity should be minimized containing salt considering chlorine ion concentration,
corrosion inhibitor concentration and pH value. The
Choice of the Inhibitors anodic polarization curve of steel in simulation solution
One of the key points of BIEM technology is to find a is shown in Fig 2. Compared to the solution without
corrosion inhibitor which can resistant to rust and can corrosion inhibitors, the stable passive region of steel
be electro-migrated. From the basic principle of BIEM anodic polarization curves became long with them. Pitting
technology, migratory corrosion inhibitor applied to the potential increased defiantly,indicating that they had a
technology should accord with the following conditions: certain resistance to rust.
(1). In the chloride environment, it is able to effectively Fig.3 Shows how the pitting potential of reinforcement
prevent or delay steel corrosion, lower the corrosion varies with the concentration of chlorine ion. From Fig.5,
rate of corroded steel significantly; when the concentration of chloride ion was 0.10 mol / L,
the dimethylamine, guanidine and 1, 6-diamine improved
(2). Easily soluble in water, positively charged soluble in
the pitting potential to more than 530 mV, which showed a
water, and the solution is alkaline. When immigrating
significant effect. Ethanolamine and triethylenetetramine
to the concrete, it will not degrade of the performance
increased to about 200 mV, which showed a well effect N,
of concrete nor react with concrete;
N-dimethylethanolamine did not make the pitting potential
(3). Stable in an alkaline environment, maintaining increased significantly, which showed poor effect. Data
the ability of resistance to rust after react with showed that with chloride concentration increasing,

Organised by
India Chapter of American Concrete Institute 701
Technical Papers

degree of the same corrosion inhibitor to improve pitting With the reduction of pH, the corrosion resistance of
potential decline in steel. For better effect, corrosion steel decreased, and the effect of the same inhibitor was
inhibitor concentration should be close to the chloride ion reduced as well.
concentration.
Screening of electro-migration type of corrosion
inhibitor in concrete shipments
According to the basic principles of BIEM technology,
except the configuration of electrolytes, other factors were
controlled and a control group of the electro-chemical
desalination was set.

Fig. 2: The anodic polarization curve of steel in simulation

solution

Fig. 4: Molar content of sixamine-based

Fig. 3: Changes of the pitting potential in steel varies with the

concentration of chlorine ion

Combined with three factors: chlorine ion concentration,

corrosion inhibitor concentration, pH value, according to
the study,the following conclusions can be obtained:
With chloride concentration increasing, ability of the
same corrosion inhibitor to resist steel corrosion is Fig. 5: The content of remaining chloride corrosion inhibitors
declined. Under the same chloride ion concentration, in concrete ion in concrete specimens at different electrolyte
different amine corrosion inhibitor varies:
dimethylamine1, 6-hexamethylenediamine, guanidine, Molar content of six amine-based corrosion inhibitors
triethylenetetramine were better while the ethanolamine in concrete is shown as Fig 4. The content of N,
and N, N-dimethylethanolamine were weak. N-dimethylethanolamine, ethanolamine were the highest
1,6-diamine center, triethylenetetramine, guanidine were
With the concentration of the corrosion inhibitors
the second dimethylamine was the least.
increasing, the ability of same inhibitor to improve the
corrosion resistance of steel was increased firstly and Fig. 5 shows the content of remaining chloride ion in
decreased then. concrete specimens at different electrolyte. The remaining

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

702 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Bi-Directional Electro-Migration Rehabilitation for RC durability improving and its application

chloride ion in concrete specimens decreased with depth.

Compared with the initial chloride ions in concrete, the
chloride ions near steel decreased while the chloride ion
content in the concrete surface did not decrease or even
more than the original content. It shows that under the
influence of the electric field applied, the chloride ion near
the steel migrate outward. However, in the surface of the
concrete cover, where the chloride ions were excluded
outward while the internal ones were moved, the content
of chloride ion dropped little. Compared with the exact
electro-chemical desalination, the effect was weak based
on amine corrosion inhibitor. In conjunction with Fig 4 and
Fig 5, content of corrosion inhibitor near the concrete pore
fluid is 2 to 7 times than that of the chloride ions.
Fig. 6(b): The effect of energization time on chloride ion
Based on the experimental study on the corrosion of
discharge
steel under different amine corrosion inhibitor in chloride
environment, 1,6-diamine, triethylenetetramine and
guanidine are suitable for BIEM technology. What’s more,
long-term experiment should be taken.

Optimum Design of Parameter in BIEM

According to research results, we designed related
studies. In these studies, we assessed the short-term
effect of different energizing current density and power-
on time of dimethylamine on BIEM: exaction of chloride
ion, immigration of corrosion inhibitors and change of
reinforced electro-chemical parameter. Based on the
laws obtained from the researches, the best power-on
time and the electric current density were selected as
well. Fig. 6(c): The effect of the amount of electricity on chloride ion
We took the power-on time t=15d and the electric current discharge
density i=3A/m2. Different studies on impact of water-
cement ratio, the initial chloride ion content and the
thickness of concrete cover were applied.
The effects of W-C ratio on BIEM

Effects of electrical parameters on BIEM

Fig. 7(a): The effect of W-C ratio on chloride ion

Fig.6(a): The effect of current density on chloride ion discharge; discharge;

Organised by
India Chapter of American Concrete Institute 703
Technical Papers

The effects of thickness of concrete cover on BIEM

Fig. 7(b):The effect of W-C ratio on immigration of corrosion

inhibitors Fig.9: The effects of thickness of concrete cover

The effects of initial chloride ion on BIEM

Fig.10: The effects of thickness of concrete cover on the dis-

charge of chloride ion on the immigration of corrosion inhibitor

BIEM in Practical Engineering

Fig.8(a): The effect of initial chloride ion on chloride ion discharge With the rapid economic development in coastal areas,
construction of coastal infrastructure has become more
and more important. Zhoushan cross-sea Bridge is not
only costly, but also concerning people’s livelihood. As
the project is located in severe corrosive environment like
sea water and marine climate, so whether the project life
can meet the design requirement of people is a serious
problem. Hence, improving the durability of concrete
structures is the key to resolve this problem. BIEM
technique is adopted to enhance the durability of Cengang
Bridge by the research group.

Initial chloride concentration

For determine the degree of chloride attack to Cengang
Bridge, drilling holes and obtaining the concrete inside
was used to take samples of cushion caps and piers.
The rapid chloride tester (TR-ClA 2501B) is employed to
Fig. 8(b): The effect of initial chloride ion on immigration of cor- determine the initial chloride concentration, of which the
rosion inhibitor. data was shown in the following figures.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

704 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Bi-Directional Electro-Migration Rehabilitation for RC durability improving and its application

Fig.11: Chloride concentration in cushion caps of Cengang Bridge

before BIEM (the central part and the lower part)

Fig. 13: Installation of electro-migration devices in Cengang

Bridge

The equipment in the cushion caps was divided into three

layers, each of which had three sets of devices. They were
located in the tide zone, splash zone and atmosphere zone
respectively. Two-layer devices were equipped around
the pier. After parameters setting, pump water from the
devices, inject corrosion inhibitor and use rubber stopper
or glass cement to seal solution filler opening and conduit
outlet for preventing water from entering into the device.

Key parameters after BIEM

Application of BIEM in the project was started in January,
Fig.12: Chloride concentration in piers of Cengang Bridge before 2014. The piers and cushion caps in the electro-migration
BIEM
area were sampled to evaluate the periodic effects of
BIEM. Experimental setup was removed and the solution
on the concrete surface in the test area was cleaned up.
Different electricity parameters could be settled to achieve
the purpose of independent work. Leakage protector Table. 1 Comparison of chloride concentration in the cushion
switch and power off alarm were installed in output caps before and after BIEM
voltage to master the current condition of the electro Depth Content of chloride Content of Exclusion of
osmosis experiment in real time. As seen in the above /mm /% inhibitors/% chloride/%
fig, concrete electric resistance is a important factor that before after
affects the duration of BIEM. And 150 days should be taken 5 1.20 0.29 0.24 76
to accomplish BIEM in the cushion caps under the voltage 10 1.95 0.35 0.18 82
of 48V. 15 1.56 0.22 0.10 86
20 0.74 0.16 0.09 78
BIEM process
25 0.46 0.10 0.06 79
The cushion cap with the height of 3.0 m could be divided 30 0.27 0.05 0.02 84
into three zones, which were tide zone, splash zone 35 0.11 0.03 0.02 69
and atmosphere zone. Electro-migration anticorrosion
40 0.08 0.03 0.02 56
devices were arranged in three different zones to verify
45 0.07 0.04 0.02 50
the effect of BIEM anticorrosion test on the different zones.
50 0.08 0.03 0.02 57
As a result of uniform chloride attack to different parts
55 0.07 0.03 0.02 50
of piers, the convenient part could be chosen to conduct
the electro-migration tst. Electro-migration devices were As indicated in the test data, chloride concentration was
installed on the basis (Figure 13). reduced remarkably after BIEM. Particularly, chloride

Organised by
India Chapter of American Concrete Institute 705
Technical Papers

Table.2 Evaluation of the effect of BIEM on Cengang Bridge

Location (mm) Concentration of Corrosion Concentration of Chloride ion Effect after BIEM
inhibitor (%) inhibitor (mol/g) chloride (%) (mol/g) N/Cl-
10 0.0707 1.263E-05 1.1436 3.221E-04 0.04
30 0.0049 8.750E-07 0.1476 4.158E-05 0.02
45 0 0.000E+00 0.0456 1.285E-05 0.00
55 0.0004 7.143E-08 0.0444 1.251E-05 0.01

exclusion efficiency could reach to approximately 80% (4) Concept design criteria and testing indexes of
for the location within 35mm distance from concrete BIEM control system were determined based on
surface. As a result of increase of concrete depth and the experiment research results of parameters
low chloride concentration, chloride exclusion efficiency optimization and controlled during BIEM. In addition,
could reduce to approximately 50%. But residual chloride the testing on chloride concentration and steel
concentration was about 0.03%, meanwhile, corrosion potential were optimized in electro osmosis control
inhibitor migrated into concrete. system for built building structures.

As seen in the above table, migration quantity of corrosion (5) The concrete surface treatment technology and the
inhibitor was relatively large despite of low chloride reusable external anode complete sets equipment
concentration, resulting in the good anti-corrosion effect. and laying technology during the field construction of
And due to large electric resistance of cushion caps, the BIEM were studied based on the research results of
ratio of N/Cl- was low. Seen from the effect of chloride lad and field test. The BIEM device had been applied
removal, the purpose to protect reinforcement is attained. to some built structures, which verified the economy
and effectiveness of BIEM device.
Conclusions (6) The BIEM technology was applied in Zhoushan
Chloride erosion is the main cause of reinforces concrete cross-sea Bridge. The results showed that the
structure durability problems. It is one of the important BIEM technology could excrete 80% chloride and
research topics when discuss about how to extend the transport corrosion inhibitor to the surface of steel
in about a month, which could improve 20 years life
durability of reinforced concrete structures in chloride
of the concrete durability.
environment. A systematic investigation of BIEM
technology, which is a new extended durability technology,
had been performed in this paper. The main research Acknowledgements
results are shown as follows. This paper was supported by the National
Natural Science Foundation of China (Grant No.
(1) The effect of six amine organics on steel corrosion 51278459,51408544,51408534), Natural Science
in simulated concrete pore solution with chloride Foundation of Zhejiang Province (Grant No. LQ14E080007),
had been studied. Moreover, researchers had been Service Project of Ningbo (2014F10016) and Ningbo
carried out on the effect of six amine organics on Scientific and Technological Innovation Team (Grant No.
electronic migration, chloride exclusion, variation of 2011B81005).
steel corrosion pre and post electrifying in concrete
References
specimen. Then the appropriate corrosion inhibitor
1. Bavarain B, Reiner L. The efficacy of using migrating corrosion
was selected. inhibitors for reinforced concrete[R]. Report-1137,Northridge:
California State University,2004:2-17.
(2) The repairing effect of BIEM technology on chloride
erosion reinforced concrete structures was verified. 2. Söylev T A, Mcnally C, Richardson M G. The effect of a new generation
surface-applied organic inhibitor on concrete properties [J]. Cement
The BIEM technology showed its superiority when and Concrete Composites, 2007, 29(5): 357-364.
compared with electrochemical chloride removing
3. Nmai C K. Multi-functional organic corrosion inhibitor [J]. Cement
technology. The effect of key factors on the efficiency and Concrete Composites, 2004,26(3): 199-207.
and results of bidirectional electro osmosis 4. Eydelnart A., Miksic B, Gelner L. Migrating corrosion inhibitors for
technology had been studied by experiment. reinforced concrete [J]. Con Chenm Journal,1993.

(3) The performance of concrete structure after BIEM 5. Montemnor M F,Simöes A M P, Ferreira M G S. Chloride-induced
corrosion on reinforcing steel: from the fundamentals to the
restoration had been evaluated. The variation of monitoring techniques [J]. Cement and Concrete Composites,
surface strength of concrete cover, bond behavior of 2003(25):491-502.
concrete and steel as well as the pore characteristics 6. Fajardo G, Escadeillas G, Arliguie G. Electrochemical chloride
of concrete cover had been studied by experiment. extraction (ECE) from steel-reinforced concrete specimens

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

706 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Bi-Directional Electro-Migration Rehabilitation for RC durability improving and its application

contaminated by “artificial” sea-water [J], Corrosion Science. 2006, State-of-the-art report[J]. Construction and Building Materials.
48(1):110-125. 2008, 22(4):609-622.
7. Ihekwaba N M, Hope B B, Hansson C M. Structural shape effect on 15. Zhang D-Q, Gao L-X, Zhou G-D. Synergistic effect of 2-Mercapto
rehabilitation of vertical concrete structures by ECE technique[J]. benzimidazole and KI on copper corrosion inhibition in aerated
Cement and Concrete Composites, 1996, 1(26): 165-172. sulfuric acid solution[J]. Journal of Applied Electrochemistry, 2003,
5(33): 361-366.
8. Miranda J M, Gonzalez J A, Cobo A, et al. Several questions about
electrochemical rehabilitation methods for reinforced concrete 16. Saraswathy V,Muralidharan S, Kalyanasundaram R M, et al.
structures [J].Corrosion Science. 2006, 48(8): 2172•2188. Evaluation of a composite corrosion-inhibiting admixture and its
performance in concrete under macrocell corrosion conditions [J].
9. Gaidis J M. Chemistry of corrosion inhibitors[J]. Cement and Cement & Concrete Research,2001(31): 789-794.
Concrete Composites. 2004, 26(3):181-189.
17. Batis G, Pantazopoulou P,Routoulas A. Corrosion protection
10. Welle A, Liao J D, Kaiser K, et al. Interactions of N,N- investigation of reinforcement by inorganic coating in the presence
dimethylaminoethanol with steel surfaces in alkaline and chlorine of alkanolamine-based inhibitor[J]. Cement and Concrete
containing solutions[J]. Applied Surface Science. 1997, 119(3-4): Composites,2003(25): 371-377.
185-190.
18. Sawada S, Page C L, Page M M. Electrochemical injection of organic
11. Elsener B, Büchler M, Stalder F, et al. Migrating corrosion inhibitor corrosion inhibitors into concrete [J]. Corrosion Science. 2005, 47(8):
blend for reinforced concrete: Part1-Prevention of corrosion[J]. 2063-2078.
Corrosion, 1999. 12(55):1155-1163.
19. Lide,David R. Handbook of Chemistry and Physics [M].Boca Raton:
12. Jamil H E, Montemor M F, Boulif R,et al. An electrochemical and CRC Press, 2009.
analytical approach to the inhibition mechanism of an amino-
20. Sawada S, Kubo J, Page C L, et al. Electrochemical injection
alcohol-based corrosion inhibitor for reinforced concrete[J].
of organic corrosion inhibitors into carbonated cementitious
Electrochim Acta, 2003(48): 3509-3518.
materials: Part 1. Effects on pore solution chemistry [J]. Corrosion
13. Jamil H E, Shriri A, Boulif R, et al. Electrochemical behaviour Science,2007, 49(3): 1186-1204.
of amino-alcohol based inhibitor used to control corrosion of
21. Ormellese M, Lazzari L, Goidanich S, et al. A study of organic
reinforcing steel[J]. Electrochim Acta, 2004(49): 2753-2760.
substances as inhibitors for chloride-induced corrosion in
14. Söylev TA, Richardson M G. Corrosion inhibitors for steel in concrete: concrete[J].Corrosion Science. 2009, 51(12):2959-2968.

Organised by
India Chapter of American Concrete Institute 707
Technical Papers

Assessment of Minimum Flexural Steel Reinforcement Ratio for

Concrete Beams
Azad A. Mohammed
Department of Civil Engineering, Faculty of Engineering, University of Sulaimani, Sulaimani, Iraq

Abstract A s,min = minimum area of flexural reinforcement,

Reinforced concrete beams usually designed for flexural bw = breadth of rectangular beam,
steel reinforcement larger than the minimum limit. In
d = effective depth of beam, and
this paper an assessment was made for the minimum
flexural steel reinforcement, recommended by the well- f y = characteristic strength of steel reinforcement, MPa.
known codes, and their applicability when applied on
The reason for including this ratio is merely for historical
concrete of different strengths. The influence of concrete
reasons, it happens to give the same minimum steel ratio
and steel strengths on the minimum steel ratio was fairly
of 0.005 recommended in earlier codes (Nilson, Darwin,
investigated. Efforts were made to check the accuracy of
and Dolan, 2010).
the limit given by ACI 318 Code. Results indicate that the
limit of minimum steel ratio is safe for rectangular beam, Since 1995, the code added the following limit for minimum
but for T-beam it is critical and not safe. New equations for steel ratio(ACI 318 Committee, 2011)
minimum steel ratio for a rectangular beam, T-beam of fcl
flanges in compression and T-beam of flanges in tension A s, min = 4f b W d ..........................................................(2)
y

were derived. The proposed equations were found to be

more accurate when applied on wide range of concrete in which
compressive strengths, and can be utilized in flexural
fc’ = cylinder compressive strength of concrete.
design of different types of concrete beams.
The larger of the two limits given by the code is
Keywords: Compressive strength, Flexure, Modulus of
recommended for minimum steel ratio. For T- beam of
rupture, Rectangular beam, Steel ratio, T-beam.
flanges subjected to tension instead of bw, 2bw or flange
width (bf ) is used whichever is smaller.
Introduction The limit given by Eq. 2 was driven based on equating
Conventional reinforced concrete beams are classified to flexural moment capacity of a reinforced concrete beam
rectangular, T and L beams and usually designed to resist and the moment just causing first crack in the extreme
flexure, shear, torsion, and serviceability requirements. fiber in tension. The limit essentially depends on the
Regarding the flexural strength, concrete beams usually modulus of rupture of normal strength concrete which is
designed for flexural steel reinforcement larger than the 0.62√fc' given by the code. The two limits provided by the
minimum ratio recommended by standards. Different code are identical for concrete of compressive strength
ratios for minimum steel ratios were recommended equal to 31 MPa. For the concrete compressive strength
by codes of practice and they depend on the strength larger than 31 MPa the limit of √fc'/4f y controls the
properties of concrete and steel bars. In the following minimum steel ratio. So, in the case of medium strength
paragraphs limits of minimum flexural steel ratio for and high strength concretes the classical limit of 1.4bd/f y is
beams recommended by the well-known codes or not used and the only remainder value for minimum steel
standards are presented. ratio is that given by Eq. 2. It was found(Nilson, Darwin,
and Dolan, 2010) that the limit is highly conservative.
ACI 318 Code Derivation of the basic equation is made in the present
Section 10.5 of this code recommends minimum steel study from which the safety factor can be determined.
ratio for rectangular and T-beams. Early versions of ACI
318 code till 1989 (ACI 318 Committee, 1989) contained a British Code
limit for minimum flexural steel area given by The 1985 version of British Code (The Institution of
1. 4 ..........................................................(1) Structural Engineers,1985) provided two ratios for
A s, min = f b W d
y minimum area of steel for rectangular beams, based
on the whole section dimensions. For steel yield stress
where
equal to 250 MPa the area is equal to 0.0024bh and equal

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

708 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Assessment of Minimum Flexural Steel Reinforcement Ratio for Concrete Beams

to 0.002bh for steel of 460 MPa yield stress. After 1997 0.0013 recommended by the code. For T- section bw is
the later limit was changed, instead of 0.002bh the limit taken as mean breadth. It is observed that this limit also
became 0.0013bh(BS 8110-1,1997). For T- and L-beams depends on both steel and concrete strengths.
the state of stress in the flange and the ratio of bw/b were
In author’s opinion the limit of minimum steel ratio should
included, and the ratios are not less than 0.0013. All limits
be based on both concrete and steel materials properties.
given by the code purely depend on steel reinforcement
The limit of ACI 318 code given by Eq. 2 seem to be useful,
strength and the concrete compressive strength has
however, such limit needs to be carefully investigated,
no influence on the minimum flexural steel ratio. For a
because it’s derivation was based on an equation for
concrete beam of a given steel reinforcement the limit is
modulus of rupture of concrete suitable for a concrete of
identical if the beam made from a concrete of 30 MPa or
normal strength, and may not be correct for concretes of
90 MPa. So, the limits given by this code are considerably
compressive strength considerably larger than 31 MPa. If
different from that given by ACI 318 code.
such equation for modulus of rupture is used for deriving an
equation for minimum steel ratio (as done by ACI 318 Code)
Canadian Code it is accurate only for those beams made from normal
The minimum flexural steel reinforcement given in strength concrete with compressive strength, in general,
the 1984 Canadian standards (Canadian Standard lower than 31 MPa. The recommendation of the code is
Association, 1994) was the same as ACI 318-89 code. The quite in contrast, because for such type of concrete the
expression in the Canadian standardfor minimum flexural classical limit of 1.4bd/f y governs the minimum steel ratio.
reinforcement was revised in 1994 to include the influence
of concrete strength. The expression of the Canadian code In this paper the accuracy of the ACI 318 Code equation
(Canadian Standard Association, 2004) is for minimum steel ratio is investigated. First, the factor
of safety included in the minimum steel ratio provided by
0.2 fcl the code is calculated. Later, the suitability of the equation
A s, min = .....................................................(3)
fy b W d is carefully investigated when applied on beams with
variable concrete compressive strengths. Alternative
The limit of validity for concrete strength is given as 20
equations for minimum steel ratio for rectangular and
MPa< fc’< 80 MPa. The ratio of this expression to the one
T-beams were derived. The equation can be applied on
given in ACI 318-11 (ACI 318 Committee, 2011)is constant
wide range of concrete strengths and can utilized in any
and equal to 80%, indicating that the expression is less
revision of the existed codes or standards.
conservative compared with the ACI 318 one.

Indian Code Limit of Minimum Steel Ratio

Indian Code(IS456, 2000) recommended the following In the following paragraphs, a comparison among some
equation for minimum tensile steel reinforcement equations proposed for calculating modulus of rupture of
0.85 concrete is made. Suitable equation for modulus of rupture
A s, min = .............................................................(4)
fy bd of concrete that can apply on wide range of compressive
strengths is used deriving a new limitsof minimum steel
where ratio, for the case of rectangular and T-concrete beams.
b = breadth of beam or the breadth of the web of T-beam.
The limit is only 60% of that given in early versions of ACI Modulus of Rupture
318 code, given by Eq. 1. This limit of steel area in general The limit of minimum amount of flexural reinforcement
is similar to the British code’s limit in which the two limits given by ACI 318 Code is based on the modulus of rupture
depend on steel reinforcement strength and neglect the of 0.62√fc' recommended by the code. Such modulus of
influence of concrete strength. rupture was basically derived for the normal concrete
and may not be accurate when applied on wide range of
Eurocode concrete strengths. Different equations were derived by
According to Eurocode2 (Eurocode 2, 1992) the limit of researchers for modulus of rupture of concretes with
minimum flexural steel ratio is wide range of concrete strengths. In this investigation two
equations have been chosen for the sake of comparison
0.26fctm
A s, min = .............................................................(5) sake, which are fr =0.42fc’0.68 proposed by Rashid et al
fy
(Rashid et al, 2002) and fr=0.94√fc' proposed by ACI 363
fctm is mean axial tensile strength given by Committee (ACI Committee 363, 1996).Now, variation of
modulus of rupture with concrete compressive strength
fctm = 0.3 fck0.666 ..............................................................(6)
calculated using the these equations in addition to that
in which fck is cube concrete strength. Cylinder given by ACI 318 code can be made. Figure 1 shows
compressive strength is assumed as equal to 0.8 times modulus of rupture variation with concrete compressive
the cube strength. The above value should be larger than strength. It is shown that the modulus given by Rashid et

Organised by
India Chapter of American Concrete Institute 709
Technical Papers

al and ACI 363 are close to each other and identical for where Mcr is cracking moment, fr is modulus of rupture
compressive strength close to 90 MPa. In contrast, values and S is the section modulus.
of modulus of rupture based on the equation of ACI 318 The equation of modulus of rupture given by ACI 318 Code
Code is low for all values of concrete compressive strength is
and the difference becomes larger when compressive
strength increase. It follows that any expression derived fr = 0.62√fc' ..................................................................(9)
for minimum steel ratio based on ACI 318 equation is For a rectangular beam the section modulus is given by
not accurate and not safe when applied on high strength
concrete, because it depends on the modulus of rupture S = bh2/6 ....................................................................(10a)
considerably lower than the actual value. Unfortunately, On substituting value of h, the modulus will be
that limit of ACI 318 code which depends on concrete
strength governs for those concretes having compressive S = 0.202bd2 ..............................................................(10b)
strength larger than 31 MPa. Substituting Eq. 9 and Eq. 10b into Eq. 8 lead to
Mcr =0.125√fc'bd2 .........................................................(11)
Equation of Minimum Steel Ratio
Before developing equation of minimum flexural steel Combining Eq. 11 and Eq. 7b and rearranging lead to the
ratio for high strength concrete beam, it is necessary following relationship for A s, which is the area of steel for
to check the basic expression given by ACI 318 Code for a beam of ultimate moment capacity equal to cracking
minimum steel ratio and illustrate the factor of safety, if moment.
any, to obtain a good view about the accuracy when applied 0.1318 fcl
A s min . = fy bd ..............................................(12a)
on wide ranges of concrete strengths.
or
fcl
A s min . = 7.6f bd .........................................................(12b)
y

Therefore, minimum steel ratio is given by

fcl
t min = 7.6f .................................................................(13)
y

in which ρmin is the minimum steel ratio.

Comparing the above equation with the equation given by
the code (i.e √fc^'/4f y ) one can find that the code accepts
90% factor of safety. The proposed equation is suitable for
normal strength concrete but not essentially be accurate
for high strength concrete.

Fig. 1: Variation of modulus of rupture with concrete compressive Proposed Equation for Rectangular Beam
strength
Now, it is a time to derive a suitable expression based on
modulus of rupture equation suitable to be applied on
Ultimate moment capacity for singly reinforced concrete wide range of concrete strengths. Accordingly, the derived
section is the tensile force acting on the section multiplied equation for minimum steel ratio can be applied on wide
by the lever arm, or range of beams with regard the concrete strength.
Mn = A s f yr ...................................................................(7a) Based on the procedure presented in this study for
Mn is the nominal moment capacity, A s is the total steel area minimum steel ratio, another equation for such ratio can
in the section, f y is the yield stress of steel reinforcement, be derived. The main difference between this new limit
and r is the moment leverarm. Simplifications were made for minimum steel ratio is in the use of an equation for
to derive the equation for minimum steel ratio given by modulus of rupture suitable for wide range of concrete
ACI 318 Code. Such simplifications are r = 0.95d and h/d strengths, including high strength concrete. Instead of
= 1.1(Nilson, Darwin, and Dolan, 2010). On substituting, the modulus of rupture given by Eq. 9 the equation proposed
nominal moment capacity will be by ACI 363 (ACI Committee 363, 1996) shown below is used
Mn = A s f yr ...................................................................(7b) fr = 0.94√fc' .................................................................(14)
The moment just causing cracking in the extreme fiber in Following the same procedure discussed in sec. 2.2, the
tension is given by value of minimum steel ratio will be
Mcr = fr S .......................................................................(8) fcl
A s min . = 5f bd ...........................................................(15)
y

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

710 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Assessment of Minimum Flexural Steel Reinforcement Ratio for Concrete Beams

From the nature of the obtained equation one can observe Minimum steel ratio for T- beam
that this basic equation gives higher value of minimum The essential difference between rectangular and
steel ratio (compared with Eq. 12b) by 52%, but lower than T-beam is the geometry, and hence the moment of inertia.
that given by ACI 318 code ( including safety factor) by 20%. Accordingly the cracking moment is different for the two
The proposed limit is exactly that given by the Canadian cases. For the T-beam, equating the cracking moment
code. Therefore, there is no safety factor in using the and the ultimate moment capacity lead to an equation of
equation of Canadian Code when applied of high strength minimum flexural steel ratio.
concrete, and there is a chance for sudden failure to occur
if the beam is reinforced with steel reinforced close to the Case 1: Flanges in Compression
minimum value recommended by the code.
Figure 3 shows the case of T-beam of flanges subjected to
Minimum steel ratio given by Eq. 15 can be used for the compression. This case is available in slab-beam system
design of reinforced concrete beams, made from a concrete subjected to a pisitive moment. In order to derive the
up to 100 MPa without problem, because rational procedure equation of minimum steel ratio for this case value of the
of analysis was followed and they based on an accepted effective flange width is required. The limit given by ACI
modulus of rupture equation, proposed by researchers Code below usually governs and used here
based on large amount of test data. However, there is a need
bE = bw + 16hf ..................................................................(17)
for safety factor. The safety factor used by ACI 318 code may
be too high and lower value can be used here. If a safety Ultimate moment capacity for this type of T-beam is not
factor of 67% which is lower than that used by the code is differe from that of rectangular sectio because the amount of
used, the new equation for minimum steel ratio will be steel is low and the compression zone locates in the flange.
The ultimate moment capacity can be calculated from Eq. 7a.
fcl
A s min . = 3f bd ...........................................................(16)
y The next stage is the calculation of the centroid of area (y)
and moment of inertia for the gross section (Ig). Cracking
The later value can be used for design of rectangular
moment is given by
beams as a limit for minimum steel ratio. The proposed
equation can be used safely for the design and can be frIg
Mcr = ................................................................(16)
y{
utilized for the revsision of the current codes of practice.
Figure 2 shows variation of minimum steel ratio with Equating the ultimate moment capacity given by Eq. 7b
concrete compressive strength using different methods. to the cracking moment lead to the area of steel which is
The yield stress of steel is assumed to be 460 MPa. It is the minimum steel ratio for this case. For a typical beam
shown that the final design proposed equation is more of dimensions frequently used in slab-beam system the
safe, and the safety is increased with increase in concrete accuracy of the ACI 318 code equation can be studied
strength. Kept in mind is that the equation is based on a here, for h/tf ratios 2,3,4,and 5. Figure 4 shows variation of
safety factor lower than that used by ACI 318 code. steel reinforcement area with the concrete compressive
strength. It is clearly shown that the value of steel area
based on the derived euation is slightly higher than that
given by ACI 318 code for low a depth beams (h/tf=2),
indicating that there is no safety related to using the code’s
equation. Figure 5 through 7 show variation of flexural
steel area with the total beam depth for different concrete
strengthes. It is shown that the variation based on the
exact equation has no overall linear trend, and the safety
of the code’s equation is reduced when the depth of the
beam (or h/tf ratio) is reduced.

Fig. 2: Variation of minimum steel ratio using different methods Fig. 3: T-beam of flanges subjected to compression

Organised by
India Chapter of American Concrete Institute 711
Technical Papers

Fig. 4: Variation of steel area with concrete compressive Fig. 7: Variation of steel reinforcement area with the beam
strength for T-beam depth (fc’ = 90 MPa)

Case 2: Flanges in Tension

Figure 8 shows the case of T-beam of flanges subjected
to tension. This case is available in slab-beam system
subjected to a negative moment. Cracking moment for
this case is calculated from Eq. 18 with a different value
of calculated based on section arrangement shown in
Figure 8. Equating the cracking moment and the ultimate
moment capacity leads to the minimum area of steel
ratio. The obtained steel area in this way is called here
to be exact to make a comparison with the ACI 318 Code
limits. Figure 9 through 12 show the variation of flexural
steel area for T-beam of flanges subjecetd to tension with
the depth ratio (hf) for a selected beam of web and flanges
normally used in slab-beam system. It is clearly shown
Fig. 5: Variation of steel reinforcement area with the beam that there is a difference between the area of steel based
depth (fc’ = 50 MPa) on the analysis (exact) and those given by the ACI 318 Code.
Based on the code’s recommendation the smaller value
(Limit 1) governs. Accordingly, the comparison should be
made with this limit.
In general, the limit provided by the code for a T-beam of
flanges in tension is critical and there is no safety in using
the code’s recommendation for minimum steel ratio. The
safety is reduced with the increase in the beam depth (or
h/tfratio ) as illustrated from making a comparison among
figures 9 through 12. When h/tf for a beam is close to 5 the
safety related to using the ACI 318 equation for minimum
steel ratio is quite low. This fact should make designers
to be aware when they use such equation, and encourage
the researchers to develop other equations for this type of
beams. Figure 13 show the percentage of steel using the
exact analysis equation to that based on the equation given
Fig. 6: Variation of steel reinforcement area with the beam by ACI 318 code. One can observe that if the depth ratio
depth (fc’ = 70 MPa) (h/tf) is more than 3 the steel area percentage becomes
higher and the safety of the code’s equation be lower. In
A new safe equation suitable for a wide range of concrete the best of author’s idea to develop a new equation for
strengths for this case can be derived. For the case of T-beam of flanges in tension two corrections should be
T-beam of flanges subjected to compression the area made on the basic equation. First, the derived equation
of steel based on Eq. 16 can be used but the depth b is should be obtained from the analysis based on modulus
replaced with 2bw or bE whichever is smaller. of rupture other than 0.62√fc' used by the code. Modulus

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

712 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Assessment of Minimum Flexural Steel Reinforcement Ratio for Concrete Beams

of rupture given by Eq. 14 is suitable and accordingly the Figure 14 shows the relationship between predicted and
derived equation can be applied on wide range of concrete calculated percentage of steel area.
strengths. Second, correction factor based on the results
Using such correction factor combined with the area given
of Figure 13 should be determined and combined with the
by Eq. 16 lead to the final form of the steel area for T-beam
derived equation for the final limit of steel area.
of flanges subjected to tension as follows

A s min . = 4.08f Q h/t f V b W d .......................................(19b)

fcl 1.02

Such equation can be used safely for minimum steel area

for T-beams of flanges subjected to tensile force.

Fig. 8: T-beam of flanges subjected to compression

Fig. 11: Variation of steel area with concrete compressive

strength for T-beam (h=4tf)

Fig. 9: Variation of steel area with concrete compressive

strength for T-beam (h=2tf)

Fig. 12: Variation of steel area with concrete compressive

strength for T-beam (h=5tf)

Conclusion
The following conclusions can be drawn
1. The expressions given by codes depend on steel yield
stress or based on both steel yield stress and concrete
Fig. 10: Variation of steel area with concrete compressive
compressive strength. Some expressions are critical
strength for T-beam (h=3tf)
and others are conservative with a high safety factor.
Based on regression analysis the percentage of exact 2. Using the equation of modulus of rupture given by
steel area to that given by ACI 318 equation is ACI 363 for deriving minimum steel ratio leads to
an equation for the limit similar to that given by the
R= 0.736(h/tf)1.02 ...................................................(19a)
Canadian code.

Organised by
India Chapter of American Concrete Institute 713
Technical Papers

3. The equation of ACI 318 Code is critical and not safe for
T-beams, especially when the flanges are subjected
to tension. New equations were derived for minimum
flexural steel ratio for rectangular beams, T-beams
of flanges in compression and T-beams of flanges in
tension. Such equations can be applied on wide range
of concrete strengths and were found more accurate
and have a higher safety compared with the equations
of ACI 318 code, and can be utilized in flexural design of
concrete beams.

References
1. ACI 318 Committee, 1989.Building code requirements for structural
concrete and commentary. Detroit, USA.
2. ACI 318 Committee, 2011.Building code requirements for structural
Fig. 13: Variation of the perectage of flexural steel area with concrete and commentary. Detroit, USA.
depth / flange thickness ratio
3. ACI Committee 363, 1996.State- of- the- Art Report on High-
Strength Concrete,”(ACI 363R-84), American Concrete Institute,
Detroit, 1996.
4. BS 8110-1,1997, Structural use of concrete, Part 1: Code of practice
for design and construction, ” pp.88.
5. Canadian Standard Association, 1994, Design of concrete structures
for building, CSA-A23.3-94, Rexdale, Ontario, Canada.
6. Canadian Standard Association, 2004, Concrete structures,”
CSA-S474-04, Mississauga, Ontario, Canada.
7. Eurocode 2 (1992). Design of concrete structures Part 1-1: General
rules and rules for building, EN 1992-1-1, Europiancommittee for
standardization (CEN), Brussels.
8. IS456 (2000),Plain and reinforced concrete – Code of practice,
Fourth Revision, Bureau of Indian Standards,46-47.
9. Nilson, A. H., Darwin, D. and Dolan, C. W. 2010.Design of concrete
structures.McGraw Hill-Higher Education. Fourteenth Ed.,pp.126.
10. Rashid, M. A., Mansur, M. A. and Paramasivam, P., 2002. Correlation
Fig. 14: Relationship between predicted and calculated between mechanical properties of high strength concrete, Journal
percentage of steel area of Materials in Civil Engineering, ASCE, 14(3),230-238.
11. The Institution of Structural Engineers, (1985).Manual for the design
of reinforced concrete building structures, pp.46.

Dr. Azad A. Mohammed

Assistant Professor in Civil Engineering Department, Faculty of Engineering, University of Sulaimani.
Kurdistan Region, Iraq.
Obtained BSc. degree in civil engineering department, College of engineering, university of Baghdad in
1994, MSc degree in Structural engineering, university of Technology- Baghdad in 1997 and Ph.D. degree in
structural engineering, university of Technology in 2004.
Published 20 Scientific papers in the field of concrete materials and structure.
His research has focused on strengthening concrete structures, ferrocement structures, mechanical
properties and nonlinear behavior of concrete material in addition to the properties of new types of concrete.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

714 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of limestone powder on hydration properties of Portland cement

Influence of limestone powder on hydration properties of

Portland cement
Xiuzhi Zhang, Mingle Liu, Guodong Zhang, Xin Cheng, Zonghui Zhou
School of Materials Science & Engineering, University of Jinan, 250022, China
Shandong provincial key laboratory of preparation and measurement of building material,
250022, Jinan, China.

Abstract Limestone powder (LP) as byproduct of limestone

crushers is a big problem from the aspects of disposal,
Limestone powder (LP) as a by-products of stone crushers
environmental pollution and health hazards. The
is a big problem for disposal, environmental pollution and
introduction of limestone quarry dust to concrete mixes
health hazards. As a kind of mineral admixture, LP was
could turn it into a valuable resource. Most investigations
introduced in concrete so that not only it could add in make
have mainly concentrated the influence of the content of
concrete good properties but also yield economic and
limestone powder on compressive strength as micro-
environmental advantages.
filler. When the powder dosage less than 10%, the
In order to study the effect of LP on the mechanical concretes or motar own the best compressive strength
properties and hydration process of portland cement, and impermeability[2,3]. Compared with chalk powders as
the series cement mixtures are prepared by different fillers, the strength of the SCC mixes containing the lime-
percentage. The mechanical properties, hydration and stone and chalk powders was significantly greater than
hardened performance and micro-structure of cement- that of the conventional vibrated reference concrete at the
based materials added the LP were explored by means same water/cement ratio[4].
of the modern analytical tools such as X--ray diffraction
As a kind of filler, it is possible to successfully utilise certain
instrument (XRD), scanning electron microscopy (SEM),
dosage of quarry waste limestone powder in producing
and so on. The results showed that the cementitious
self-compacting concrete (SCC)[5,], and the cumulative
system heat release decreases gradually with the increase
heat release in self-compacting cement paste containing
of dust content; CaCO3 particles of limestone powder can
limestone powder is higher and the rate of heat release
accelerate the early hydration of portland cement and
is also higher than for traditional cement pastes and high
improve the early strength. In the hydration late, CaCO3
performance cement pastes. It is believed that the powder
will reacts with aluminate to generate single-carbon
does not participate in the hydration reaction[7,8].
hydrated calcium aluminate (CA) so that ettringite was
reduced, and the late strength correspondingly reduced. However, LP has also been reported to have a chemical
Powder in the hardened cement paste acts as a kind of effect, in which carboaluminate hydrate is formed
filler and plays the roles of activation and acceleration.
during the hydration of Portland limestone cement. In a
Keywords: Limestone powder (LP), hydration rate, normal Portland cement without limestone powder,the
microstructure. dissolved sulfate ions will interact with C3A and/or C 4 AF
to form stable ettringite ,Upon depletion of the
Introduction sulphates,the rest of C3A and C 4 AF will continue to react
With the development of the construction, the amount with the ettringite to form unstable monosulfoaluminate
of sand usage grows rapidly, as the main raw material (C 4 A H 12 ). In the presence of LS,calcium carbonate
in mortar and concrete. However, as the natural sand
(C ) can react with (C 4 A H 12 ) to form calcium
currently does not meet the requirements for engineering
construction; the use of manufactured sand (MS) has carboaluminate (C4 A H11) and ettringite [9-11].This lead to
become a trend. However, compared with natural sand, the the stabilization of ettringite and result in an increase of
particles of MS are rough and angular with unsatisfactory the total volume of hydration products[12].
grading, and there are fewer particles with sizes between
0.315 and 0.630 mm. The most remarkable difference 3C4 A H 12 + 2C + H18 → C6 A 3 H 32 + 2C4 A H11 ........(1)
between MS and natural sand is that MS contains a large Burak Felekoğlu et al [13] compared the effect of
amount of stone powder,whose physical and chemical limestone filler and fly ash filler on the performance of
properties are completely the same with the source rock, self-compacting repair mortars. The results derived from
and which has a particle size of less than 0.075 mm[1]. compressive strength tests showed that limestone filler

Organised by
India Chapter of American Concrete Institute 715
Technical Papers

Table 1
Chemical Compositions of cement and Limestone powder

Component CaO SiO2 MgO Fe2O3 Al2O3 SO3 TiO2 Loss Total

Cement 65.3 21.5 1.05 3.42 5.12 2.03 0.68 0.46 99.56

LP 48.66 6.93 1.31 0.63 2.17 0.21 0.18 39.56 99.65

was more effective than fly ash in terms of early strength

gain. However, beyond 28 days, mixes incorporating fly
ash gave higher strength values than the control mixtures
due to the pozzolanic effect of fly ash.
The present work is to investigate the influences of the
content of LP on the hydrate rate of cement, microstructure
and hydrate product of the hardened past. The study aims
to determine the hydration and hardening properties of
cement system containing different content of limestone
powder.

Experimental Program
Raw materials Fig. 2: Morphology of limestone power
Ordinary Portland cement (P.O42.5), limestone powder,
and superplasticizer were used in this study. The used the limestone powder. The scanning electron microscope
Portland cement provided by Shandong Shanshui Cement (SEM) photograph of limestone power is presented in Fig.
Group satisfied all the requirements of the “Ordinary 2. The limestone powder showed a random geometric
Portland Cement” (GB175-2007). The limestone powder structure and a certain gradation .
is a by-product of quarry crushers. The powder was
collected from the filtration system (0.08 mm) of a quarry Mixture design and specimen casting
crusher. The chemical composition of Portland cement
and limestone powder used are presented in Table 1. Mixture proportion
The particle size distributions of Portland cement and Mortar experimental was intended to analyse the effect
limestone powder measured by laser diffraction are of manufactured sand containing lime stone on the
illustrated in Fig.1. The average particle size of the strength of the cement cementitious system. The mixture
limestone powder is slightly finer than that of the Portland proportions of cement mortar are summarized in Table 2.
cement. By the particle size analysis test, the average The content of limestone powder in MS is 5%, 10%, 15%,
particle size and median diameter were 19.30 μm and 20%, 25%, respectively(% by mass of manufactured sand).
12.98 μm for the cement, while 17.53 μm and 11.39 μm for
Table 2
Mix proportions of cement mortar

Mass/g
Specimen
C LP S W/C

M0 450 0 1350 0.5

M1 450 67.5(5%) 1282.5 0.5

M2 450 135(10%) 1215 0.5

M3 450 202.5(15%) 1147.5 0.5

M4 450 270(20%) 1080 0.5

Fig. 1: Particle size distributions of cement and limestone M5 450 337.5(25%) 1012.5 0.5
powder

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

716 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of limestone powder on hydration properties of Portland cement

Table 3
Mix proportions of cement paste

Mass/g
Specimen
C LP W/C
S0 300 0 0.30
S1 270 30 0.30
S2 240 60 0.30

The mixture proportions of cement paste are summarized

in Table 3, in which the limestone powder is as a kind
of mineral filler. The mass percentages of limestone
powder replacing cement in this research are 0%, 10% (a) Compressive strength
and 20%, respectively(% by mass of cement). The content
superplasticizer is 0.8 % and w/c ratio is 0.30.

Testing method
The hydration heats and rates of heat evolution of all
pastes were measured in a thermometric isothermal
conduction calorimeter (TAM Air Isothermal Calorimeter).
All measurements were performed at a temperature
of 20°C. XRD was used to analyse the different phases
in hydration products of all pastes. SEM was used to
determine the microstructure of hydration products. MIP
is used to determine the pore size distribution of mortars.

Results and discussion (b) Flexural strength

Effect of limestone powder on strength of the cement Fig. 3: Effect of LP on strength of the mortar
cementitious system
The mortar specimens were formed according to the strength of specimens decreased gradually, the degree
mix proportions. The flexural strength and compressive of reduction in strength of each sample for 28 d was
strength were tested after standard curing according significantly slower than that for 3d. This showed that
to the GB/T 17671-1999. Method of testing cements- the effect of limestone powder on the later strength of
Determination of strength". The results are shown in the cement cementitious system reduced gradually. The
Fig. 3. lower degree of early hydration of limestone powder
The results from Fig.3 clearly show that the flexural reduced the effective gel components, so the degree of
and compressive strength have a first increscent, then reduction in 3d strength was higher. With hydration length
decreasing tendency with the increasing of limestone of time increasing, as limestone powder participated
powder. The strength reached the maximum when the in the hydration reaction and interacted with the micro-
content of limestone powder is between 5% and 10%. This aggregate effect of limestone powder, the late strength
is because the limestone powder substituting the sand decreased gently.
fill in the cement and improve the grain composition and
the uniformity of the gelling materials. At the same time, Effect of limestone powder on exothermic rate during
the proportion of fine particles increases and limestone hydration
powder have a certain filling property which reduce The hydration heat and the rate of heat evolution of
the porosity of mortar and make the structure denser, the specimens S0,S1,S2 are plotted in Fig. 4 and Fig.5
therefore, the strength reaches a peak. But when there respectively.
were more limestone powder, as the excess fine particles
As shown in Fig. 4 and Fig.5,there were obvious
of limestone powder destroyed the particle dense packing
differences between the pure cement and the specimens
structure or made the mortar pasterns ratio deviate from
mixed with limestone powder about the hydration heat
the optimum value, the mortar strength decreased.
and the rate of heat evolution. The exothermic process
It was also concluded from Fig.3 that when limestone of cement hydration was divided into two stages. First,
powder dosage exceeds 10%, although the compressive when the hydration started about 1 h, ettringite formed

Organised by
India Chapter of American Concrete Institute 717
Technical Papers

It can lead the hydration products crystallize and

accelerate the cement hydration at early ages, but
participated in reaction to form calcium carboaluminate
hydrate and blocked AFt from converting into AFm[14].

Effect of limestone powder on hydration products

Mineral phase analysis of hydration products
In order to analyse the effect of limestone powder on
hydration products between early ages and later ages, the
paste specimens having standard curing for 3 d and 28 d
were tested by XRD, as shown in Fig.6 and Fig. 7.

Fig. 4: Effect of LP contents on hydration heat relate of paste

Fig. 6: XRD patterns of the samples hydrated for 3 d

Fig. 5: Effect of LP contents on exothermic rate

and exothermic reactions were severer. The second

process is the hydration period of C3S corresponding
to two exothermic peaks. The first peak is the result
of ettringite formation combined with C3S hydration.
It is controlled mainly by the exothermic reactions of
ettringite formation. The second peak is controlled
mainly by the exothermic reactions of C3S hydration. At
early ages, the rate of heat evolution of the specimens
mixed with 10% or more limestone powder was higher
than that of the pure cement specimen. The rate of
cement hydration is accelerated by limestone powder
at early ages. The amount of C3A in the specimens Fig. 7: XRD patterns of the samples hydrated for 28 d
mixed with limestone powder was less than that in the
pure cement specimen. The limestone powder did not With an analysis and comparison of the Fig.6 and Fig.7, it
participate in the formation of ettringite, so the hydration can be easily concluded that the main hydration products
heat decreased with the increase of limestone powder. for 3 d and 28 d were calcium hydroxide, anhydrated
Due to the more C3S in the pure cement specimen, there clinker minerals, calcite and a small quantity of ettringite.
was also more calcium hydroxide and calcium silicate Choose characteristic peak d=0.263 nm to analyse
hydrate phase in this stage. So its exothermic peak calcium hydroxide. Because there are many overlapping
was higher than other specimens mixed with limestone strong peaks between alite and belite cement clinker, alite
powder. In the process of hydration, limestone powder and belite were analyzed together as unhydrated clinker
can improve the hydration of cement as a nucleating minerals analysed by characteristic peaks d=0.279 nm
effect. and d=0.273 nm. The main mineral in limestone powder

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

718 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of limestone powder on hydration properties of Portland cement

is calcite(d=0.303,0.210,0.386 nm). Its strongest peak and the strength. Fine particles of CaCO3 can increase
superposed with the stronger peak of tricalcium silicate. the early strength of cement paste and accelerate the
When analyzed the calcite in hydration products, its early hydration rate of C3S. During the hydration process
strongest peak was for reference only but we can use the of C3S-CaCO3-H2O system, some CaCO3 participated the
stronger peak for comparative analysis. reaction and combined with the C-S-H. So the surface
state of these CaCO3 particles changed which benefited to
The above two figures indicated that there were some
the bonding with hydration C3S particles and improved the
big differences among specimens S0, S1 and S2. The
early strength of cement paste[16].
diffraction peak in 29° significantly reduced with the
increasing of limestone powder and the diffraction peak
of the pure cement specimen here was relatively weaker. Microstructure analysis of hydration products
The diffraction peak in 29° was the overlapped peaks For better analysis of the microstructure of hydration
of calcite and alite. There were also large amounts of products, scanning electron microscopy was used to
calcium carbonate in cement paste hydrated for 3 d determine the microstructure and the phase distribution
through the characteristic peaks around 23° and 43° and of the samples hydrated for a certain age. Fig.9 and Fig.10
also existed for 28 d age. This phenomenon showed that were the SEM photographs of the samples hydrated for 3
a majority of limestone powder did not participate in the d and 28 d respectively.
hydration reaction.

(a) Specimen S0 (b) Specimen S1

Fig. 9: SEM photographs of the samples hydrated for 3 d

Fig. 8: XRD patterns of the hydration products of specimen S1

Fig.8 is the XRD patterns of the hydration products of

specimen S1. As shown, calcium hydroxide and ettringite
obviously reduced with increase of age. The introduction
of limestone powder involves a decrease in the amount of (a) Specimen S0 (b) Specimen S1
cement and consequently, an increase in the effective w/c
ratio. The filler effect implies a modification in the initial Fig. 9: SEM photographs of the samples hydrated for 3 d
porosity of the mix, and it generally produces a decrease
in the water required to maintain constant workability. The photographs above showed that at 3 days the hydration
Heterogeneous nucleation occurs because limestone products of pure cement specimen were AFt, C-S-H,
particles act as nucleation sites, increasing the early Ca(OH)2 and some unhydrated particles. However, the
hydration of cement and, therefore, producing a more hydration products of the specimen added 10% limestone
disoriented crystallization of CH[15]. powder contained more C-S-H and Ca(OH)2 crystal. CaCO3
crystal was wrapped with Ca(OH)2 crystal nucleated on
In later age some CaCO3 participated in hydration the surface of CaCO3 crystal. So it was difficult to observe
reaction, other CaCO3 had micro aggregate effect and CaCO3 crystal. For the hardened paste of the specimen
micro crystal nucleus effect. Because of the addition S1, the structure was more compact and the porosity
of limestone powder, there were more CaCO3 and new reduced. The hydration products wrapped a small amount
mineral C3A·CaCO3·11H2O formed. CaCO3 can prevent of unhydrated particles. There were more C-S-H and
AFt from converting into AFm due to the reacting with Ca(OH)2 in the hydration products of the specimen S1 which
aluminate mineral to form C3A·CaCO3·11H2O making indicated that appropriate limestone powder can increase
AFt stable. So the content of AFt in pure cement paste the early rate of cement hydration. There was also lots of
reduced relatively decreasing the solid volume content limestone powder filling in the hydration products.

Organised by
India Chapter of American Concrete Institute 719
Technical Papers

One explanation of the above phenomenon was that the (2) Due to the reduction of total amount of cement after
micro crystal nucleus effect of limestone powder made the limestone powder replacing cement, the total
C-S-H and Ca(OH)2 attached to the surface of CaCO3 hydration heat decreased. When the content was less
particles which prevented Ca(OH)2 from growing into than 10%, limestone powder can accelerate the rate of
larger crystals in the paste pores and improve the viscosity cement hydration.
capacity of the paste. As a result, the strength of hardened (3) During early cement hydration, limestone powder can
paste increased. Limestone powder did not act as the increase the density of mortar and the rate of cement
cementitious minerals during the hydration process, hydration because of its filling effect and accelerating
so with the limestone powder increasing, the content effect. At the later stage, small amount of limestone
of cement reduced. The reduction of the cementitious powder reacted with aluminate mineral to form
materials in the paste led to the decrease of the strength. C3A·CaCO3·11H2O blocking AFt from converting into
AFm and reducing the later strength of paste because
Effect of limestone powder on pore structure of of its filling effect and active effect.
cement mortar
(4) The micro aggregate effect of limestone powder
The pore size distribution measurements by MIP of improved the pore structure of cement paste. The
specimens M0,M1,M2,M4 having standard curing for 28 d pore diameter and the total porosity reduced with
are shown in Fig. 11. the increase of limestone powder content. Changes
in microstructure increased the macro-strength of
paste.

Acknowledgement
This work was financed by Chinese government for science
research (Grant NO. 51208227) ,National High Technology
Research and Development Program (“863 Program”,
2015AA034701), and Jinan of Shangdong government for
science research (201303078).

References
1. Bonavetti V L, Donza H, Menéndez G, et al. Limestone filler cement
in low w/c concrete: A rational use of energy. Cem Concr Res, V.33,
No.5,2003,pp.865-871.
2. Ramezanianpour Ali A, Ghiasvand E, Nickseresht I, et al. Influence
of various amounts of limestone powder on performance of
Fig. 11: Pore distribution of mortar sample Portland limestone cement concretes. Cem Concr Comp, V.31,
No.10,2009,pp.715-720.
As can be seen from Fig.11, the appropriate limestone 3. Damidot D, Lothenbach B, Herfort D, Glasser FP. Thermodynamics
powder can significantly improve the pore structure of the and cement science. Cem Concr Res, V.41, No.7,2011,pp.679-95.
hardened cement paste increasing harmless pores and 4. Zhu W Z, John C. Gibbs. Use of different limestone and chalk
decreasing harmful pores. From Fig.1, it can be known powders in self-compacting concrete. Cem Concr Res, V.35,
that the average particle size of the limestone powder is No.8,2005,pp.1457-1462.
finer than that of the cement. Fine particles of different 5. De Weerdt K, Kjellsen KO, Sellevold E, Justnes H. Synergy between fly
sizes in limestone powder had a micro aggregate effect on ash and limestone powder in ternary cements. Cem Concr Compos,
V.33, No.1,2011,pp.30-38.
the hardened cement paste. They filled in the larger pores
to reduce the porosity of the hardened paste making the 6. K. Celik , M.D. Jackson , M. Mancio , C. Meral , A.-H. Emwas , P.K.
Mehta , P.J.M. Monteiro. High-volume natural volcanic pozzolan and
pore size smaller and then improving other properties of limestone powder as partial replacements for portland cement in
the corresponding concrete. self-compacting and sustainable concrete. Cem Concr Comp, V.45,
2014,pp.136-147.
7. P.R. da Silvaa,J.de Brito.Experimental study of the porosity and
Conclusions microstructure of self-compacting concrete (SCC) with binary and
Based on the test results presented in this paper, the ternary mixes of fly ash and limestone filler. Construction and
following conclusions can be drawn: Building Materials, V.86, No.1,2015,pp.101-112.
8. Felekoglu B. Utilisation of high volumes of limestone quarry wastes
(1) Adding limestone powder into cement can increase in concrete industry (self-compacting concrete case). Resour Conserv
the mortar strength and density. The strength reached Recy, V.51, No.4,2007,pp.770-791.
maximum when the limestone powder content was 9. Axel Schöler , Barbara Lothenbach , Frank Winnefeld, Maciej Zajac.
5% and 10%. The enhancement effect was achieved by Hydration of quaternary Portland cement blends containing blast-
micro aggregate effect, micro crystal nucleus effect furnace slag, siliceous fly ash and limestone powder. Cem Concr
Comp, V.55, 2015,pp.374-382.
and active effect.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

720 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Influence of limestone powder on hydration properties of Portland cement

10. Tydlitat V, Matas T, Cerny R. Effect of w/c and temperature on the 14. Barbara L,Gwenn L S, Emmanual G, et al. Influence of limestone
early-stage hydration heat development in Portland-limestone on the hydration of Porland cements, Cem Concr Res, V.38,
cement. Constr Build Mater, V.50, 1,2014,pp.140-147. No.6,2008,pp.848-860.
11. Lothenbach B, Le Saout G, Gallucci E, Scrivener KL. Influence of
15. Irassar E F. Sulfate attack on cementitious materials containing
limestone on the hydration of Portland cements. Cem Concr Res,
V.38, No.6,2008,pp.848-60. limestone filler- A review. Cem Concr Res, V.39, No.3,2009,pp.241-254.

12. De Weerdt K, Ben Haha M, Le Saout G, Kjellsen KO, Justnes H, 16. Kakali G, Tsivilis S, Aggeli E, Bati M. Hydration products of C3A, C3S
Lothenbach B. Hydration mechanisms of ternary Portland cements and Portland cement in the presence of CaCO3. Cement and Concrete
containing limestone powder and fly ash. Cem Concr Res, V.41, Research, V.30, No.7,2000,pp.1073-1077.
2011,pp.279-91.
17. P.R. da Silvaa,J.de Brito.Experimental study of the porosity and
13. Burak Felekoğlu, Kamile Tosun, Bülent Baradan, et al. The effect
of fly ash and limestone fillers on the viscosity and compressive microstructure of self-compacting concrete (SCC) with binary and
strength of self-compacting repair mortars. Cem Concr Res, V.36, ternary mixes of fly ash and limestone filler. Construction and
No.9,2006,pp.1719-1726. Building Materials, V.86, No.1,2015,pp.101-112.

Organised by
India Chapter of American Concrete Institute 721
Technical Papers

Imaging the pH profiles in cement based materials

Engui Liu Masoud Ghandehai, Weihua Christian Bruckner Gamal Khalil

New York University Abu Jin, Alexey Sidelev University of Connecticut, University of Washington,
Dhabi, Abu Dhabi, United Arab New York University, Brooklyn, Storrs, CT 06269 Seattle, WA 98195
Emirates NY 11201

Abstract degradation. The knowledge gaps are due partially to lack

of effective experimental methodologies to fully address
We are reporting a new pH measurement methodology
and understand the extreme complexity of the chemical
for cement based materials. This method is based on an
processes in situ. The determination of temporal and spatial
optical measurement that uses a ratio­metric porphyrin­
fluctuations of internal pH levels and the transport of OH­in
based sensing compound. The method has the potential
concrete in situ are perhaps the most important unresolved
to produce full­field 2D pH mapping of fractured concrete
aspect of ASR and carbonation­induced distresses.
surface. The proposed pH sensor molecule incorporated
into the sensing compound has two optical absorbance Methods for the direct in situ measurement of high pH levels
bands at 575nm, and 700nm. The two bands have (around pH 12.0 ­pH 13.0) have not yet been possible as the
opposite absorption responses to pH changes. The ratio of internal concrete pore solution is not readily accessible.
absorbance of the two bands is used to determine the pH Currently, chemical analysis for pH levels in concrete
value. We will present calibration and optical absorption is carried out via pore solution extraction, developed by
performance of the sensor molecule in simulated concrete Longuet et al. in 1973[13], and subsequently modified by
pore solution showing a dynamic pH range from pH 11.0 Barneyback and Diamond in 1981[14]. In this method, a small
to pH 13.5. The digital imaging calibration of the sensing amount of pore fluid is extracted from hardened cement
compound on fractured concrete surface presents the pH paste or mortar under very high pressures; the collected
versus ratio curve for imaging application. sample is then diluted for chemical analysis. Although the
method provides an accurate measure of constituents of
Keywords: pH, sensor, alkali­silica reaction, carbonation,
the extracted pore fluid, the overall process is tedious,
geopolymer.
destructive and costly. Extracting enough pore fluid for
analysis can be extremely difficult, especially in the case of
Introduction high performance concrete[15] where the water to­cement
Deleterious chemical reactions lead to premature ratio can be as low as 0.16 with crushing strength as high
degradation of concrete and cause deterioration of our as 500 MPa, resulting in minimum amounts of extracted
roads, bridges, dams, and levees. One of the major modes of pore fluid. Finally, the measured chemical composition
distress in concrete is the high pH­dependent degradation. reflects only bulk information, with no information on
This includes Alkali­Silica Reaction (ASR), carbonation and spatial variations.
rebar corrosion in concrete. In such cases, the pH gradient
and fluxes of hydroxyl ion (OH­) play important roles in the It is therefore imperative to develop new, innovative,
initiation and propagation of micro­ cracks in materials, easy­to­use technologies that can indicate the chemical
highly potentially leading to structural damage. signature of cement based materials on the temporal
and spatial scales necessary for understanding the
Over the past decades, ASR has been increasingly phenomena.
recognized as a major problem, as witnessed by the
increasing number of research publications devoted to this The objective of this investigation is to make possible
topic [1­6]. At the same time, numerous laboratory studies the direct measurement of pH profiles in Portland
have been carried out to evaluate the pH of carbonated cement concrete with temporal and spatial resolution.
cement paste and concrete, and to thereby address the For achieving that, this investigation seeks to develop a
chemically­induced corrosion of reinforced concrete[7­12]. full­field optical imaging approach using a robust high
pH sensing compound with a porphyrin­based sensor
Despite extensive worldwide research, large gaps remain molecule[16] incorporated for measurement of pH levels in
in understanding the mechanism of high pH­dependent cement based materials.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

722 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Imaging the pH profiles in cement based materials

Materials and Methods filter 575 is from 550 nm to 600 nm with the maximum
transmittance at 575 nm; and from 600 nm to 800 nm with
Materials the maximum transmittance at 700 nm for filter 700. The
image processing generates a third channel of intensity
Simulated concrete pore solution ratio between channel 575 and 700 which determines the
The simulated concrete pore solution was prepared by pH levels based on the extracted intensities from images
dissolving Ca(OH)2, NaOH, and KOH in deionized water. captured under filters 700 and 575.
The [K+], [Na+] and [Ca2+] was kept at 0.4 M, 0.2 M and
0.001 M, respectively; thus the [OH­] balances the sum of
the alkali ions at about 0.6 M[15]. The solution was then Results and Discussion
diluted with deionized water to various pH levels.
Absorbance of sensor molecule in simulated concrete
pH sensor molecule stock solution pore solution
The sensor molecule was dissolved into dimethyl sulfoxide The sensor molecule exhibits instant color change after
(DMSO) with the concentration 0.4×10­-3 M, and used as the added to the simulated concrete pore solution. The color
sensor molecule stock solution. changes from pink to bright yellow (Figure 2) against
various pH levels in the range of pH 10.0 ­pH 13.5. The
pH sensing compound bright yellow presents at high pH values around pH 13.0;
The pH sensing compound was prepared by mixing the while the pink indicates lower pH levels near pH 11.0 and
sensor molecule stock solution and polyvinyl alcohol below; the intermediate colors between yellow and pink
solution (PVA solution, 30% weight in deionized water) represent the pH levels about pH 12.0.
together with a surfactant.

Methods
UV­vis spectroscopy
UV­ vis spectroscopic measurement was performed
to record the absorbance of pH sensor molecule in Fig. 2: Color change of pH sensor molecule versus pH in
simulated concrete pore solution at different pH levels simulated concrete pore solution
from pH 10.0 to pH 13.5. The measurement was conducted
in a cuvette system in which 1 mL simulated concrete pore The underlying mechanism for the observed color
solution and 0.1 mL pH sensor molecule stock solution change is the nucleophilic addition/attack of OH­onto the
was respectively added. The color change of the solution sensor molecule[16] in which the optical absorbance of
was recorded using a digital camera, and the absorbance the molecule changes after the addition/attack. Figure 3
spectrum was measure by the UV­vis spectrometer. shows the absorbance spectra of the sensor molecule in
simulated concrete pore solution at various pH levels. The
Image collection and processing sensor molecule exhibits two characteristic absorbance
A digital camera was used to collect the images. Xenon peaks, one centered at 575nm and the other at 700nm.
lamps were employed as the illumination source, and The 575nm peak represents the neutral/lower pH state of
two optical bandpass filters were used to capture the sensor, e.g. pH 10.0, pH 11.0. The 700nm peak relates
image intensity at two wavelength ranges centered at to the base/higher pH near pH 13.0. The two channels of
575nm and 700nm (Figure 1). The wavelength range for

Fig. 3: Absorbance spectra of pH sensor molecule versus pH

Fig. 1: Experimental set­up for image collection in simulated concrete pore solution

Organised by
India Chapter of American Concrete Institute 723
Technical Papers

absorbance change inversely towards the change of pH

levels, i.e., the absorbance intensity at 575nm decreases
while that of 700nm increases, as the pH level increases.

Dynamic pH sensing range of the sensor molecule

As stated earlier, the optical absorbance at two bands of
575nm and 700nm represent two forms of the sensor
molecule versus pH levels. These two channels, as a
result, were processed to derive the third channel of
ratios between the two channels. In this way, the ratios
determine the pH levels or the concentration of OH­.
Based on the absorption spectra (Figure 3), the absorbance
intensities at band 575nm and 700nm were determined
Fig. 5: Absorbance spectra showing areas covered by the
with intensity values at 800nm as the reference, in which,
bandpass filters
the absorbance value at 575nm and 700nm were obtained
by subtracting the one at 800nm. The ratios between
two peaks of 575nm and 700nm. This exercise was first
absorbance values at 575nm and 700nm were then
carried out on each spectrum and the same baseline
calculated and expressed in Figure 4. Both of the ratios
was chosen for both peaks. The integration range for the
present the same pH dynamic sensing range from pH 11.0
two peaks was determined by the specifications of the
to pH 13.5.
optical bandpass filters shown as the highlighted yellow
coverage in Figure 5.
Figure 6 shows the ratios between integrated areas
under the 575nm and 700nm peaks based on the spectra.
Both ratios show good linearity from pH 11.0 to pH 13.5
which agree well with the peak value calibrations shown
in Figure 4. This implies that optical bandpass filtering
of images taken by a digital camera has potential for
assessing a wide range of pH levels on concrete surface.

Fig. 4: Ratio of absorbance at 575nm and 700nm versus pH

based on absorbance spectra

Peak area integration from absorbance spectra of the

sensor molecule
The objective of the method is to develop an imaging
process to map the spatial variation of pH levels on
fractured concrete surfaces. Unlike a spectrometer
which records the full absorbance spectrum with
intensity information for each wavelength, the digital Fig. 6: Ratio of Integrated Areas under Peaks 575nm and 700nm
camera captures intensity of all photons that arrive versus pH
at the camera sensor matrix without the wavelength
information. Referring to the spectra shown in Figure 3, pH sensing compound evaluation and calibration
the areas under the two peaks of 575nm and 700nm were After preparation, the pH sensing compound was simply
quantified through integration (with y=0 as the base line), tested on a glass slide and on a piece of fractured un­
and the calculated intensities were used to calculate degraded cement paste. Figure 7 shows its color change
ratios which determine the pH values. This way of on a glass slide after treatment with a drop of pH 13.0
spectra processing is quite similar to the way the camera simulated concrete pore solution and on the un­dergraded
sensor acquires intensities of images. Therefore, during cement paste surface, which has the similar high pH
the images collection, two optical bandpass filters were range. In both cases the sensing compound changed color
outfitted to camera lens for capturing intensities at the from pink to yellow over time. In the un­degraded cement

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

724 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Imaging the pH profiles in cement based materials

Fig. 7: The pH sensing compound on glass slide and treated

with pH 13.0 simulated concrete pore solution (top); process
of sensing compound application on surface of un­degraded
cement paste (bottom)

paste, a drop of acid at pH 1.0 made the already­yellow area

return to pink, indicating the reversible performance of
the sensing compound (bottom right).
The calibration of the sensing compound was applying the
sensing compound on un­degraded cement paste with a
known average pH level of pH 13.0, followed by imaging and
processing to obtain an average intensity ratio which was
assigned to pH 13.0. This ratio for pH 13.0 was then used to
compare with the one obtained from the spectra (Figure 6)
and the further interpolation of the curve generates a new
ratio­pH curve for imaging application. Figure 8 shows the
sequence of sensing compound calibration process for
imaging application.

Fig. 9: Images and processed ratio for regular cement paste

(un­degraded, average pH 13.0) sample at pH sensing compound
calibration process, (a) Color image of Phenolphthalein applied
on cement paste; (b) Color image of pH sensing compound
applied on the companion side of cement paste; (c) processed
image showing ratio of intensities from images taken under filer
700 and 575; (d) processed curve showing average transverse
ratio along the sample

Fig. 8: The flow chart for pH sensing compound calibration

process

In Figure 9, an example of images and processed ratio of

an un­degraded cement paste sample was demonstrated;
and specially in part (c) and (d) of Figure 9, the processed
ratio versus position on the un­degraded cement paste
surface was shown. The ratios fluctuated around 1.5 and
further processing of the curve generated an average
ratio of 1.5, which was then assigned to pH 13.0. The final
calibrated ratio­pH curve for imaging application (Figure
10), as earlier mentioned, was obtained through the
interpolation process between ratios from processing Fig. 10: Calibrated curve of pH versus ratio for imaging
spectra of the sensor molecule and images of sensing application (* calibrated by setting the average ratio of 1.5 (ratio
compound on un­degraded cement paste samples. of image: 700/ 575, Figure 9) to pH 13.0 as determined with un­
degraded cement pastes that are averagely pH 13.0)

Organised by
India Chapter of American Concrete Institute 725
Technical Papers

Conclusions supplementary cementing materials against alkali–silica reaction,

Cement and Concrete Research, 31: 1057­1063.
The Porphyrin­based pH sensor molecule is capable of 7. Alonso C., Andrade C., Gonzalez J.A. (1998) Relation between
sensing high pH with the dynamic range from pH 11.0 to pH resistivity and corrosion rate of reinforcements in carbonated
13.5, suitable for applications to cement based materials, mortar made with several different cement types, Cement and
monitoring pH changes at early stages of degradations Concrete Research, 18: 687­698.
or hydration reactions. The pH sensing compound is 8. Glass G.K., Page C.L., Short N.R. (1991) Factors affecting the
capable of being applied onto fractured concrete surface corrosion rate of steel in carbonated mortars, Corrosion Science,
32: 1283­1294.
for pH imaging application. The pH imaging protocol and
calibrated pH versus ratio curve of sensing compound 9. Gonzalez J.A., Algaba J.S., Andrade C. (1980) Corrosion of reinforcing
bars in carbonated concrete. British Corrosion Journal, 15: 135­139.
for application on fractured concrete surface has been
10. Suryavanshi A.K., Swamy R.N. (1996) Stability of Friedel’s salts in
developed.
carbonated concrete structural elements. Cement and Concrete
Research, 26: 729­741.
References 11. Yeih W., Chang J.J. (2005) A study on the efficiency of electrochemical
1. Chatterji S. (2005) Chemistry of alkali–silica reaction and testing of re­alkalisation of carbonated concrete. Construction and Building
aggregates, Cement and Concrete Composites, 27: 788­795. Materials, 19: 516­524.
2. Sargolzahi M., Kodjo S.A., Rivard P., Rhazi J. (2010) Effectiveness 12. Huet B., L’Hostis V., Miserque F., Idrissi H. (2005) Electrochemical
of nondestructive testing for the evaluation of alkali–silica reaction behavior of mild steel in concrete: influence of pH and carbonate
in concrete, Construction and Building Materials, 24: 1398­1403. content of concrete pore solution. Electrochimica Acta, 51: 172­180.
3. Bleszynski R.F., Thomas M. D.A. (1998) Microstructural studies 13. Longuet P., Burglen L., Zelwer A. (1973) The liquid phase of hydrated
of alkali­silica reaction in fly ash concrete immersed in alkaline cement (in French) Rev. Mater. Constr., 676: 35­4.
solutions, Advanced Cement Based Materials, 7: 66­78.
14. Barneyback R. S., Diamond S. (1981) Expression and analysis of
4. Aquino W., Lange D.A., Olek J. (2001) The influence of metakaolin pore fluids from hardened cement pastes and mortars, Cement
and silica fume on the chemistry of alkali–silica reaction products, and Concrete Research, 11: 279­285.
Cement and Concrete Composites, 23: 485­493.
15. Rasanen V., Penttala V. (2004) The pH measurement of concrete
5. Garcia­Diaz E., Riche J., Bulteel D., Vernet C. (2006) Mechanism and smoothing mortar using a powder suspension, Cement and
of damage for the alkali–silica reaction, Cement and Concrete Concrete Research, 34: 813­820.
Research, 36: 395­400.
16. Khalil G. E., Daddario P., Lau K. S. F., Imtiaz S., et al. (2010) mero­
6. Duchesne J., Bérubé M.­
A . (2001) Long­
term effectiveness of Tetraarylporpholactones as high pH sensors, Analyst, 135: 2125­2131.

Engui Liu, PhD

Engui Liu, PhD
Senior Lecturer of Civil & Environmental Engineering
Engineering Division, New York University Abu Dhabi
Saadiyat Island, Abu Dhabi, United Arab Emirates

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

726 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Damage detection of reinforced concrete beams based on the embedded acceleration sensor of concrete structure

Damage detection of reinforced concrete beams based on the

embedded acceleration sensor of concrete structure
Hongda Geng, Suhui Zhai, Shifeng Huang, Fan Lu, Xiao Yuan, Xin Cheng*
Shandong Provincial Key Laboratory of Preparation and Measurement of Building Materials, Engineering Center of Advanced
Building Materials of Ministry of Education, University of Jinan, Jinan 250022, China.

Abstract Since the first appearance of piezoelectric accelerometers

in 1943, Scientists have done lots of research on damage
A kind of embedded acceleration sensor was fabricated
identification technology based on structural dynamic
by using mixture of cement/polymer (cement: epoxy tests. Back in the 1960s, aerospace and mechanical
resin: hardener=4:4:1) as encapsulation layer, and PZT- engineering industry have already employed vibration
5 piezoelectric ceramic with high voltage constant as response data of dynamic tests to evaluate the integrity
sensing element. Effects of thickness on the sensitivity of the structures. After that the research has paid lots
and amplitude-frequency characteristic of the sensor of attention to civil Engineering[4]. Ricardo et al brought
were investigated. The results showed that as the out the method of using local modal stiffness to locate
thickness of PZT-5 increases, the sensitivity of the sensor and recognize the damage of the RC beams. Then the
increased, the frequency range was gradually decreased. feasibility of the method was verified by conducting
The structure damage law of reinforced concrete vibration test which progressively loaded on the RC
beam was discussed by using the method of Damage beams. Hearn did an identification research on structural
damage localization by using the measured change chart
identification method based on structural vibration. The
of the structure random two order frequencies[5].
results showed that the dynamic characteristics of the
structure were closely related to the structural damage. Nowadays significant progress has been made in the
The methods of additional mass and precast crack had damage identification of structure dynamic mechanics
an effect on the dynamic characteristics of the reinforced based on the piezoelectric acceleration sensor, which was
concrete beams. With the increase of damage degree, mostly encapsulated by metal shell, and so it had a poor
compatibility with concrete material. Poor waterproofness
the natural frequency of the reinforced concrete beam
and durability made it more susceptible to corrosion
was decreased. The method of precast cracks was more
after a long period of using[6]. Based on this fact, a kind
suitable for the damage detection research of reinforced of piezoelectric acceleration sensor embedded into the
concrete beams. concrete which could be considered to be the larger
Keywords: Piezoelectric ceramics; the embedded aggregate particle in the concrete and was well compatible
acceleration sensor; natural frequency; damage with the concrete, was prepared. Good waterproof nature
and corrosion resistance made it more suited for concrete
identification.
structure health monitoring. The damage identification
method based on structural vibration technique was
Introduction adopted to study the structure damage law of reinforced
With the advancement of times and scientific and concrete beam.
technology, reinforced concrete construction is
dominant in civil engineering for the improvement and
Experiments procedure
wide application of cement and steel. However, civil
engineering structures will surely result in damage Preparation of the embedded piezoelectric
accumulation and resistance deterioration in service accelerometer
period due to the internal and external reasons, which
Piezoelectric accelerometer, an electromechanical
causes the structural damage. Therefore more and
transducer, works based on the positive piezoelectric
more attention is paid to the adoption of smart material
effect of piezoelectric crystal, When the piezoelectric
system to health monitoring and damage diagnosis[1].
material is subjected to mechanical load, the surface of
Piezoelectric acceleration sensors are being widely used
the piezoelectric material will produce some electric
in damage identification system of structural dynamic
charge, proportional to the load[7]. The Eq. (1) shows the
tests due to their high measure precision, wide operating
relationship between the electric charge and applied
frequency range, and simple structure[2, 3].
force.

Organised by
India Chapter of American Concrete Institute 727
Technical Papers

q d d d d ..................................(1)
e a = C = 33A F = 33A Fm sin t
a

Where F is the force along the direction of polarization

axis, and A, d, d33, ε are the area, thickness, piezoelectric
constant, and dielectric constant of the piezocrystal,
respectively. Ca, q, ea are the capacitance, quantity of Fig. 1: Fabrication scheme of the embedded Piezoelectric
electric charge, voltage. Fm, ω, t are the amplitude of the Accelerometer
force, frequency and time.
A kind of embedded acceleration sensor was fabricated
using PZT-5 piezoelectric ceramic with high piezoelectric
constant as piezo-element (Zibo Yuhai Electronic Ceramic
Co., Ltd), high density tungsten block as mass, and
the mixture of cement/polymer (cement: epoxy resin:
hardener=4:4:1) as encapsulation layer, The noise-signal
ratio of sensor was improved by using the external Fig. 2: The schematic of reinforced concrete beam with
shield wire The fundamental performance parameter embedded sensors
of PZT-5 piezoelectric element was shown in table 1.
Detail operations are as follows: The PZT-5 piezoelectric Damage identification method based on dynamic
elements with different thickness 1mm, 1.2mm, 1.4mm, characteristics
1.6mm, 1.8mm, respectively were prepared by SYZ- The principle of damage identification method based
400 cutting machine. The performance of piezoelectric on dynamic characteristics shows that: structural
ceramic was tested by using the impedance analyzer which damage can be described by the modal parameters of
named Agilent 4294A. Then the sensor was assembled the structure. The natural frequency was easy to be
according to the preparation process as shown in Figure 1. measured with higher precision among all the modal
The sensitivity and frequency response of the sensor were parameters in practical engineering. The change
tested by using standard vibration table and dynamic data of natural frequency can reflect the change of the
acquisition system[8]. structural physical state (rigidity, quality), thereby
The embedded acceleration sensors were fixed in the judging whether there is a damage and determining
hooping of steel frame by wire bar. Figure 2 shows the the damage degree[9,10]. Force hammer was adopted
reinforced concrete beam with embedded sensors. to stimulate the reinforced concrete beam to generate
It was made sure that vertical direction of the sensor structural vibration response acquired by the embedded
was consistent with the vibration direction. The fresh piezoelectric accelerometer.
concrete was poured into the fixed mould. Artificial The modal parameters of the system could be obtained
vibration was adopted to avoid the sensor offset. The after the time domain/frequency domain waveform of the
reinforced concrete beam was demoulded after placed force hammer and structure responses were analyzed by
in the curing room for one day and then maintained for dynamic data acquisition instrument. In the initial state
28 days. The size of the reinforced concrete beam was of the experiment, the attenuation curves of the sensors
1900mm×100mm×100mm. And the size of the concrete at midspan and 1/4 positions were shown in Figure 3,and
blocks was 100mm×100mm×100mm, the average quality the first two-order natural frequencies and mode shapes
of it was 2.4kg. of the reinforced concrete beams were experimentally
measured in Figure 4.
Table 1 The reinforced concrete beam can be equivalent to a
Fundamental performance parameter of PZT-5 piezoelectric single degree of freedom system, which is concentrated
element in the middle of the beam. The Eq. (2) and (3) show the
influence of damage for the first order natural frequency.
Main d33 / Tanδ Kt Kp thickness
εr Qm
parameters pC/N /% /% /% /mm 1 K ..............................................................(2)
f=
2Y= M
PZT-5 520 2200 0.02 50 60 80 1.0-1.8
1 K
f= ....................................................(3)
2Y= M + 3M
In the table 1. d33, εr, Qm, Tanδ stand for the piezoelectric
strain factor, dielectric constant, mechanical quality
factor and dielectric loss. Kt, Kp are the thickness Where ∆M is the additional mass, and K, M, are rigidity
electromechanical coupling coefficient and planar and quality of the equivalent single degree of freedom
electromechanical coupling coefficient. system, respectively.

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

728 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Damage detection of reinforced concrete beams based on the embedded acceleration sensor of concrete structure

In the experiment, the damage change rule of the reinforced Results and discussion
concrete beam was tested by adding mass method
and prefabricated crack method[10, 11, 12, 13]. The method Performance of embedded piezoelectric
of additional mass was adopted 5 steps to add, and the accelerometer
quality of each concrete block was 2.4kg. Vaseline was as Properties of different thickness of piezoelectric ceramics
a couplant between the mass and the reinforced concrete were investigated by using the impedance analyzer which
beams. The dynamic response of the structure was tested named Agilent 4294A. The thickness of PZT-5 are 1mm,
under each level of additional mass. The method of multi 1.2mm, 1.4mm, 1.6mm, 1.8mm, respectively. It was known
position damage with same damage degree was used in based on the piezoelectric theory when the thickness
prefabricated crack method. The dynamic response of the of the piezoelectric ceramic increases, the resonant
structure was tested after the reinforced concrete beam frequency of the piezoelectric ceramic moved to the low
was cut each time by cutting machine, each cutting depth frequency direction.
20mm. Fig 5 shows the impedance-frequency spectrogram of the
piezoelectric element. It can be seen that the resonance
frequency changed according to a general regularity.
But because of the little change of the thickness of the
piezoelectric element, and the error of the cutting process,
it is not particularly obvious. Planar resonance frequency
is about 110 kHz, and Thickness resonance frequency is
about 570 kHz.
From the Eq.1, it can be seen that the output charge of
piezoelectric element is increased with the increase
of the thickness of PZT-5. The Figure 6 shows that
the sensitivity of the sensor is gradually increasing.
Contrarily, Figure 7 shows the test curve of the sensor
frequency response. Figure 8 shows the sensitivity and
frequency response of sensor changes. It can be found
that it is consistent with theory that the sensitivity of
the acceleration sensor and the frequency limit were
contradictory which had been proved. As the thickness
of the piezoelectric element is increasing, the sensitivity
of the sensor increases 17.6%, and the frequency range
decreases 8.6%. Its frequency range is about 0.1~5 kHz,
which is suitable for the low frequency vibration of the
concrete structure. Its frequency range is about 0.1~5
kHz, which is suitable for the low frequency vibration of
Fig. 3: The attenuation curve of acceleration signal the concrete structure.

Fig. 4: The experimental modal analysis of reinforced concrete beam

Organised by
India Chapter of American Concrete Institute 729
Technical Papers

Fig. 7: The test curve of the sensor frequency response

Fig. 5: The impedance-frequency spectrogram of the PZT-5

Fig. 8: Trend chart for sensitivity and frequency response of

sensor

method of additional mass has relatively little influence

on the natural frequency, or even no change. There
was almost no change when No.1 block was added. To
continue with 2#, the natural frequency showed a small
amount of change, or was almost unchanged with the
additional. When completely attached to No.5, the change
of frequency was still small. The Fig.9 shows the rate
of frequency damage. It can be found that the rate of
frequency change was only 4.5% finally. The main reason
was that the additional mass of the reinforced concrete
beam was equivalent to a weak damage. Just changed
the quality of the structure and the internal damage to the
structure had no effect.
The following study was carried out with the method of
Fig. 6: The sensitivity of the different sensors precast crack instead of the method of additional mass.

Damage identification method based on structural Damage identification method based on structural
vibration by the method of additional mass vibration by the method of precast crack
The measured modal frequencies were shown in Table It can be seen that the natural frequency drifts to low
2. It can be seen that the natural frequency drifts to low frequency with increasing the crack at the different
frequency with increasing the mass at the same position position in the reinforced concrete beam. Figure 9 (b)
in the reinforced concrete beam. But the damage by the for the curve of the first two order frequency variation,

2nd RN Raikar Memorial Intl. Conference & Banthia-Basheer Intl. Symposium on

730 ADVANCES IN SCIENCE & TECHNOLOGY OF CONCRETE
 Damage detection of reinforced concrete beams based on the embedded acceleration sensor of concrete structure

the first order frequency change rates were 4.5793%, Conclusions

11.3924%, 10.201%, two order frequency change rates
4.1 In this paper, a kind of embedded acceleration sensor
were 0.6233%, 2.0125%, 3.6509%. The rate of natural
for health monitoring in concrete structures was
frequency change had good regularity. The reason for
fabricated. Performance change rule of different
the change of frequency was that the increase of the
thickness on the sensitivity and amplitude-frequency
crack in the concrete beam would reduce the rigidity of
characteristic of the sensor was investigated. The
the concrete structure. The natural frequency of the first
existing theory that sensitivity of the acceleration sensor
order was more obvious than that of the two order natural
and the upper frequency limit were contradictory were
frequency.
verified in the experiment.
4.2 The reason why the additional mass has little effect on
Table 2
The measured modal frequencies
the natural frequency of reinforced concrete beams was
that the additional mass of the reinforced concrete beam
The method of additional mass The method of precast crack was equivalent to a weak damage. The simple changing in
f1/Hz f2/Hz f1/Hz f2/Hz
quality of the structure had little influence on the internal
damage.
Health 26.86 56.15 Health 26.86 56.15
4.3 Prefabricated crack method of damage identification
1# 26.86 56.15 Crack 2 25.63 55.80
of structure dynamic test had significant influence on
2# 25.63 56.15 Crack 2 24.12 55.02 damaged changing of reinforced concrete structures.
3# 25.63 55.02 Crack 3 23.80 54.10 It can be a better simulation of the damage state of
reinforced concrete structures during service period.
4# 25.63 55.02
This method has important significance on the study of
5# 25.63 55.02 the structure's remaining life and the identification and
reinforcement of the existing structure.

Reference
1. Li Hong-nan. Structural health monitoring. dalian university of
technology press, 2004, pp. 1-20.
2. Li Ai-qun, Miao Chang-qing. Bridge structure health monitoring.
People's Communications Press, 2009, pp. 49-70.
3. Zhao Yu-bao. The application of piezoelectric acceleration sensor in
bridge health monitoring. North transportation, (4), 2012, pp. 115-117.
4. Jiabin Peng, research on simply-supported beam’s damage
detection based on vibration testing. Hebei University, 2010, pp. 7-14.
5. Dongtao Qin, Zhuo Wang. Test analysis for damage extent
identification of reinforced concrete simple-supported beams. The
first National Conference on engineering safety and protection,
200, pp. 501-505.
6. Xu Dong-yu. Fabrication and Properties of Cement Based
Piezoelectric Sensor and Its Application Researching Civil
Engineering Fields. 2010, pp. 35-46.
7. Lu Zhaofeng, Qin Min. Study on vibration measurement system
of compactor based on piezoelectricity accelerometer. Product &
Technology, 2008, pp. 85-87.
8. National Commission on mechanical vibration and shock. GB/T
20485.21-2007, methods for the calibration of vibration and shock
transducers. Beijing: China standard press, 2007.
9. LI Bin, Zhou Hui-xin. Dynamic monitoring and damage detection
of reinforced concrete beams. Journal of Baotou University of Iron
and Steel Technology, 25(3), 2006, pp. 276-279.
10. Wang Yong-yi. Additional mass method for mechanical vibration
modal analysis. Journal of Inner Mongolia Institute of Technology,
(2), 1987, pp. 85-92.
11. Xie Dong-hai. Research on damage identification of bridge
structures. Jinan: Shandong University, 2006, pp. 30-62.
12. Wang Yan. Bridge damage identification based on natural frequency
change. Nanchang: East China Jiao tong University, 2012, pp. 30-56.
13. Ding Yong, Bu Zhan-yu. Analysis of Low-frequency Noise of Bridge
Fig. 8: Trend chart for sensitivity and frequency response of Considering the Vibration of Bridge Deck. Civil engineering and
sensor environmental engineering, 33(2), 2011, pp. 58-69.

Organised by
India Chapter of American Concrete Institute 731
 Thank You Partners

Solitaire Partner

Thermex Rebar Manufacturers’ Association

Platinum Partners

Gold Partners

Silver Partners

R A IK A R G RO U P

Kit Partners
Dongwoo (Thomas) Lee
Cost & Contract GM
HDC INDIA Pvt. Ltd.
Tel)+91.022.2417. 8317~9
http://www.hyundai-dvp.com/