Cement and Concrete Research 124 (2019) 105832

Contents lists available at ScienceDirect

Cement and Concrete Research

journal homepage: www.elsevier.com/locate/cemconres

Rethinking cement standards: Opportunities for a better future T

⁎
Vanderley M. John , Marco Quattrone, Pedro C.R.A. Abrão, Fábio A. Cardoso
National Institute on Advanced Eco-efficient Cement-based Technologies, Brazil
Department of Construction Engineering, Escola Politécnica, University of São Paulo, 05508-070 São Paulo, Brazil

A B S T R A C T

This paper discusses possible changes in cement standards to cope with megatrends: climate change and industry 4.0. Most standards are prescriptive, defining
cements by composition of SCMs, hence the number of types increase, and composition limits widen. Cements are classified by compressive strength of fixed water/
cement mortars. This pragmatic approach ignores physical effects, that results in variable mixing water demand. It hinders the development of cements with low
water demand, products of optimized particle engineering and dispersants, which allows formulate cements with lower clinker fraction. Also, most standard tests are
difficult to automate to generate large datasets crucial to train artificial intelligence. Performance-based standards are an alternative, but simple and progressive
approach is recommended to ease transition. Cement types should not be solely defined by composition but rather classified by performance characteristics including
durability and environmental. Combined water fraction, cwf, may be a good parameter to replace strength class.

1. Introduction digital electronics have allowed large improvements on existing de-

vices, not only in terms of better accuracy and resolution, but also in
Standards are ubiquitous in the industrial society. They provide speed, reduced size and reduced initial and operational costs. Today,
obvious benefits in consumer protection and simplify business, helping electronics control most of the cement plants allowing real time auto-
to create large markets, being particularly useful for commodities like matic adjustment of the process. New knowledge and experimental
cement [1]. They help society to manage risks. They increase pro- techniques that enable measurements and design at the nanoscale have
ductivity replacing an in-depth technical description of the desirable been introduced. The general advance of knowledge allows us to use
cement with quotation of the standard number and a cement type. sophisticated science-based models to better understand hydration. The
In his keynote to the XII ICCC in 2007 Hooton [2] registered the existing thermodynamics model of hydrated solids is now a mature tool,
1995 ASTM Workshop “Cement and Concrete Standards Of The Future” supported by a database that has been continuously improved and
– an event which dare imagine a better future (now our past, 2010) for enriched [6], and allows forecasting the produced mineral phases and
cement industry. In the report from the workshop, Geoffrey Frohnsdorff their chemical composition. Coupled models of hydration, mass transfer
and James Clifton [3] stated that cement standards should improve and chemical materials degradation are being developed. If we succeed
considering “computing and information technology, high-performance to coordinate the research efforts, in the future it may become possible
concrete (HPC) and material science in general, international trade agree- to simulate, at low-cost and within a short period of time, the long-term
ments, and the needs for sustainable technologies.” behaviour of almost any mixture between clinker Portland and any
Nowadays, 23 years later, this message is even more contemporary. potential supplementary cementitious materials. The research commu-
Very few professionals in academy or cement industry will disagree nity has also grown and established much stronger ties with the in-
about the urgency of change to increase the chances of a more sus- dustry and succeeded to establish networks, such as Nanocem, elevating
tainable human society. The main challenge is related to the mitigation to another level the quality of research and work. Average citations of
of anthropogenic CO2 emissions, of which the cement value chain has papers published on CCR had grown twelve times, from < 0.5 to 6 and
an ever-growing share [4]. The large amount of natural resources used the total number of citations of paper's journal grown by a factor of
– about 1/3 of total consumption – and waste generated by the cement seven between 1999 and 2017 [7], exemplifying at least one dimension
industry are also being recognized as an emerging problem that will of this growth.
require the development of a new circular economy model [5]. The information technology, that impressed the authors in 1995, is
If the amount of knowledge and practical experience accumulated in exponentially growing. To put the advance in perspective, in 1995 the
1995 was considerable beyond the one registered in the cement stan- ASTM president informed the audience that organization had an e-mail
dards, this gap certainly has grown during the last 25 years. Modern address. This was the year that Microsoft launched Windows 95 and an

⁎
Corresponding author at: Department of Construction Engineering, Escola Politécnica, University of São Paulo, 05508-070 São Paulo, Brazil.
E-mail address: vmjohn@usp.br (V.M. John).

https://doi.org/10.1016/j.cemconres.2019.105832
Received 25 March 2019; Received in revised form 25 July 2019; Accepted 25 July 2019
Available online 19 August 2019
0008-8846/ © 2019 Elsevier Ltd. All rights reserved.
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

excellent desktop computer had 8 Mb of RAM, 1Gb hard disk and the cement-based industry. Nowadays a team of building designers usually
best internet (or BBS) connection was 28.8 Kbps modem over a copper define strength, durability and other aspects as a function of the most
telephone line. Global processing power has been growing ex- critical situation in the component or in the building. However, it is
ponentially and computing and data storage cost decreasing [8,9]. We already possible to implement a much more optimized design, varying
are living the “digitalization of everything” era [9] and entering in a strength, mass transport or any other property of interest inside a single
new industrial revolution [10]. These technologies are multi-purpose component, using existing models to functionally grade materials
and are going to impact all industries [9,11], including the construction [20–22]. Digital production platform will then automatically produce a
sector [12–15] and, therefore, the cement industry. The 4th industrial component with a variable formulation, matching the required speci-
revolution or industry 4.0 includes the digitalization of the value chain fication at every point as it has been already demonstrated [21]. A
– from design to industrial floor –, not only by robots and 3D printing beam or column will therefore be built by a range of formulations lo-
machines but also by ubiquitously present advanced smart sensors cally optimized to fulfil global performance requirements with
embedded in the Internet of Things (IoT). In fields which enough data minimum life-cycle environmental loads and costs. More sophisticated
are available, the power and speed of artificial intelligent (AI) algo- compositions can be developed, varying mix proportions, actual che-
rithms are already visible, as it is on the fast progress of self-driving cars mical composition of cement, admixtures or even applying protective
and digital assistants. Only when those digital devices begin to generate surface coatings. The digital production platform can provide more
enough data - big data [16]- it will be possible to develop and feed self- adequate conditions for the successful employment of functionally
improving AI algorithms. AI will enable on-the-fly finetuning of pro- graded materials creating finetuned hybrid products.
cesses, controlling and progressively optimizing day-to-day industrial Designers may be free to create shapes without the limits of mould,
operations, ensuring better performance with tremendous potential of changing not only the way of construction, but also the way construc-
positive impacts on economy and environment. The 4.0 industry will tion will be conceived [23]. Design may be defined either by architects'
ease the adaptation of cement to actual client needs and probably will imagination or by mathematically-based topology optimization con-
create new cements tailored to suit digital (3D printing) production cepts [24,25] that consider localized demands and can be combined
processes. with functional gradients of properties to optimize materials perfor-
It seems that Portland cement will probably remain the dominant mance.
low-cost mineral binder in the foreseeable future [4] because it is Mass customization can be also applied to the cement plant. It is
produced with a variety of natural resources available worldwide that possible that a future cement plant will be equipped with multiple silos
are capable to cope with the enormous and increasing demand of (or multi-chamber silo), allowing separate grinding and storage of
building and construction materials. However, in a business-as-usual clinker, gypsum and SCMs, including inert fillers and chemical ad-
scenario, cement materials industry can certainly loose competitivity mixtures. A robotized dry blending system fed by the silos might be able
due to a combination of (a) growing costs related to environmental to deliver custom-made cements for different markets or large con-
mitigation, such as installing and operating Carbon Capture and Storage sumers without requiring any human intervention and error. The po-
(CCS) technologies, environmental taxation and new regulations, which tential gains embedded in this scenario seem to be sizable not only in
are expected to grow in parallel to tangible effects of global warming; terms of technical performance, but also in cost and environment. This
(b) advent of innovative lightweight construction solutions produced in solution will require some CAPEX but the need of knowledge devel-
more robotized environments. opment appears to be the greatest barrier. The current cement types
This paper discusses possible impacts of two global megatrends: (1) based on composition may not be the best options to suit the needs of
the environmental sustainability and (2) digitalization of production – such advanced market.
the industry 4.0 - in the cement industry. It also explores possible Apart from that, the 3D printing technology will probably benefit
course of action in terms of changing a few key aspects of cement from cements different than those available today. For example, 3D
standards and material testing methods, not only to cope with the printing requirements in terms of setting time and rheological beha-
challenges, but also to better explore opportunities. The discussion is viour are rather different than those of today's cementitious materials
limited to aspects that, in our opinion, seem to be more directly affected [26]. Therefore, stricter control of chemical reactions and rheological
by the megatrends; thus, key performance aspects such as those related behaviour over time is required [23].
to durability are not encompassed. This work is mostly based on lit- In a full digital supply value chain, digital machines and omnipre-
erature review, but it does include a few new original ideas and data. sent sensors must generate large amounts of data to allow the formation
We have no intention of making predictions. Rather than presenting of reliable AI models. In a cement plant, an AI model integrating all
definitive answers, the aim is to encourage discussion and to incentive digital data generated from plant machines and quality control may
research and innovation, which certainly will influence the future, help minimize the risks associated to the long-time between production
hopefully for the better. and the actual compressive strength and performance test results,
which a colleague has described as being equivalent to “driving a car
2. Digitalization and the cement industry combining the previous experiences with a video feed of the drive delayed by
1 min”. If AI identifies the risk of non-compliance on clinker that may
An all-digital supply chain driven by low-cost almost unlimited impact strength development, for example, it will immediately fix the
computing power will probably change cement-based materials in- production process. Moreover, even a faulty clinker may be used if the
dustry in unknown ways. For the sake of the argument, we dare to information arrives on time to reduce the filler content before grinding
present a possible scenario. or blending stages. Another option is to electronically inform industrial
Projects will be expressed in standardized digital building in- clients to adjust cement dosage or even to allocate it to markets in
formation models (BIM) with unprecedent levels of detail and seam- which its performance is still suitable. The gains in terms of economy,
lessly digitally transferred to the production. Robotic machines will efficiency and environmental impact are evident. But, certainly, this
mass produce custom-made designs, allowing to reach large scale pro- will be a long transition.
duction without full standardization. The mass customization con- However, standard tests of the cement industry somehow defy the
struction is already a reality in the prefabricated housing market of digital revolution: Vicat's needle is 201 years old being originally de-
Japan, dominated by lightweight steel and wooden frame panelised veloped to hydraulic lime; Le Chatelier soundness is younger, from
construction [17,18]. This concept is slowly being expanded to other 1870; compressive strength was developed in 1851 [27] and the 1:~3
regions [19]. mortar was already in use around 1900 in the USA [28]. These tests are
Mass customization construction can similarly be applied to the not digital, are intensive in labour, time consuming, and have not been

2
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

significantly improved by digitalization. If the multitude of machine 4. Cement standards

sensors can generate thousands or millions of measurements per second
and robotized analytics dozen measurement per hour, these tests cannot Cement standards are among the oldest standards in many coun-
be easily and cheaply scaled up to generate a stream of data large en- tries. Cement Committee of ASTM is C1, which was inaugurated in
ough to allow creating and validating the AI models. The near future 1902 [40]; the Brazilian cement standard EB 1 was the first standard
will be a period when it will no longer be possible to avoid changing specification published in the country, in 1935 [41]. All these standards
those tests, among other things. were prescriptive, meaning they specify, among other aspects, the
composition of Portland cement.
With the progressive introduction of supplementary cementitious
3. Environmental mitigation materials – blast furnace slag, fly ashes or other pozzolans and, in due
time, limestone fillers - the number of standardized cements has grown,
Environmental problems require new supplementary cementitious including an increasing variety of composite cements due to new
materials, since available blast furnace slag and fly ash are insufficient combination of SCMs. The SCMs' compositions are also more or less
to supply the cement demand [4]. Apart from fillers – with plural, since clearly defined in standards.
limestone is by no means the only option [29,30] – there is no other On one side, the prescriptive specification of the cement's in-
source of traditional secondary SCMs as large as those ones. On the gredients has the advantage of simplifying the entire process, including
other side, there are myriad local industrial wastes that can be explored quality control. On the other side, introducing innovation in cement
as SCMs and artificial pozzolans as calcined clays [31,32] and others. composition demands modification in prescriptive standards. This
The diversity of raw materials for cement will grow even more, because modification is time-consuming, requires consensus and delays the ef-
local availability and the potential benefits of blending various raw fective introduction of new products in the market; this is conflicting
materials that may complement each other [33,34], something that is with the concept and rhythm of 4.0 industry. The EN 197-1 had 27
already a fact. Since industry is already accepting inert materials as cement products in 2000. In 2006 the CEN/TC 51 committee started to
cement constituents, it seems inevitable to start accepting SCMs with discuss 10 new composite cements, with clinker factor varying from 35
low degree of reactivity. All considered, by today's standard definition, to 64% [42] using the traditional constituents (limestone, granulated
the number of cement types would increase continuously, or composi- blast furnace slag, fly ash and natural pozzolans) and finally, in 2019,
tional types would be defined loosely and, therefore, become less and the prEN 197-1:2019 was issued. The latest proposed version of Feb-
less capable to describe the actual cement composition and inform the ruary 2019 included provision for the LC3-type cements [42] com-
user of its expected performance. bining calcined clay pozzolans and limestone fillers, allowing the
Another requirement is the increase of resource use efficiency, re- number of cement type notations increased up to 39, being 27 types of
ducing simultaneously the environmental impact [35]. Increasing the CEM II. The large number of cement types is certainly confusing. Al-
strength to reduce the amount of materials is certainly an option with lowing a broad range of mixtures in a single cement type, the increasing
limits related to aspects of acoustical performance of buildings, buck- number of SCMs in a same cement, and the differences on reactivity and
ling, etc. that digital production might help overcoming. Another ne- fineness of each SCM, are rendering progressively clients to estimate the
cessary strategy is the reduction of the binder (clinker and other che- actual composition of the reactive fraction of cement and forecast the
mically reactive materials) content required to produce cementitious expected performance in terms of heat of hydration, CO2 footprint,
materials; this can be better achieved by a multiscale approach. In the resistance to alkali-aggregates reaction, carbonation, etc. The classifi-
millimetre-scale, particle size distribution of aggregates must be opti- cation is somehow losing its capacity to describe the technical perfor-
mized considering the actual particles' shape to ensure flowability with mance.
minimum paste volume. On the micrometre scale, optimizing particle Despite the great deal of experience accumulated internationally
size distribution and morphology of fines, plus ensuring dispersion to and the impressive advance of knowledge on blended Portland cement,
achieve the optimal packing density to make a paste with reduced there are significant differences among countries. Some differences are
amount of water and adequate rheological behaviour [29,36]. This related to local availability of raw materials and other technical rea-
water reduction will allow replacing the binder fraction by filler, within sons. Other differences are harder to justify. For example, despite the
the limits allowed by durability and other requirements. There is evi- overwhelming success and solid technical data, there are still large
dence, proved in relevant industrial application, that it is possible to differences on the maximum amount of limestone filler [4]: it took
replace up to 70% of binder by filler [29,37–39]. Because it compen- decades to introduce it in ASTM standards [2]. Another example is the
sates dilution by reducing the amount of mixing water, it can be per- use of calcined clay as pozzolan, which is limited in many countries,
formed without extra-grinding of the binders, even at the mortar and/or despite the successful use in the USA of about 100 years ago [43], the
concrete plants, but only if the producer has the technical skills and the continuous commercial production in Brazil for nearly five decades and
location has the required supply chain, which reduces dramatically the recent developments of LC3 formulation. Globally, these arbitrary
scalability of the technology and its mitigation potential. Therefore, in standards limitations may imply in unnecessary higher clinker factor,
the best of interest of humanity, binder dilution should be performed by and higher CO2 emissions, a loss for the environment and for society.
cement producers. Besides the large experience with these constituents, good under-
Another dimension is the need of a more circular economy of ce- standing of their chemical reaction with clinker and water and the
ment-based materials, which implies to increase recycling rates of availability of thermodynamic models that allow forecasting the mi-
construction and demolition wastes. This will probably demand in- neralogical composition, understanding the different effects on perfor-
novative technologies to improve the separation of cement paste from mance associated to the combination of SCMs and on in-use perfor-
aggregates. Cement industry, which so far has been little active in this mance is becoming difficult. This is an area that certainly requires
subject, will probably have to deal with its low-carbon waste, and in- further research [44].
corporating the recycled cement paste in the cement production process
can become the best available option in many places. 4.1. The limits imposed by current prescriptive standardization
The forecast is of a competitive future; hence the cement industry
must be flexible and capable to seize the best available raw material. In Apart from specialized cements, such as sulphate resistant, current
other words, cement industry must make easy the introduction of in- standards classify cements based on the content and combination of a
novative products. limit set of maximum eight SCMs according to the prEN 197-1:2019.
Because the traditional reactive SCMs, slag and fly ash are practically

3
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

exhausted in most regions [4], the cement industry needs to start ex- generic terms how a cement must behave when exposed to certain
ploring a great variety of locally existing materials in order to mitigate conditions of use, and what is reasonable to be expected over its ser-
CO2 emissions within the budget. Naturally, this requires changes in the vice-life in order to fulfil the needs of its various users.
cement standards to accommodate these local materials. The role of performance-based standards in facilitating the in-
The number of possible future SCMs is much larger than the 5–10 troduction of innovation is recognized in the literature [57,58] by
currently allowed by most standards. Except for inert filler minerals, various organizations such as the European Conference of Ministers of
these materials will not be available globally in large amounts as well as Transport [59], the OECD [60], particularly to overcome environ-
slag, fly ash and natural pozzolans. They will be relevant only in a few mental problems. The potential of the application of performance-based
parts of the world, where availability and logistics favour. Individually standards and codes to allow innovation in construction is known for
their contribution to mitigation may be considered small. But, adding long time. In 1972 the RILEM president, Dr. Wright [61], reported that
up the small individual contributions from the use of many SCMs, the the proposition of performance-based construction dated back to the
total mitigation may become relevant. 1930's. France was the pioneer on adopting a national performance-
Today's standards usually specify fillers to be limestone only, which based technical approval, “Agrément” system for innovative building
is actually slightly reactive [45], specially combined with aluminium technology around 1950, issued by the CSTB [62]. Dr. Frederick Lea,
rich additional SCMs such as calcined clay [46]. Most of standards go author of the famous Lea's Cement Chemistry Handbook reedited since
further, requiring clinker grade limestone. However, science evidences 1935, was the first to present a comprehensive view of performance
and practical experience in some markets show that almost any inert or concept applied to buildings in 1962 during a CIB conference.1 In 1982,
quasi-inert material can be filler [29] and that even the limitation of the ISO 6241 Performance standards in building – Principles for their
chemical composition, such as maximum MgO content in limestone is preparation and factors to be considered - was published under the
not necessary. The number of minerals that can be explored as fillers, leadership of Gérard Blachère, one of the pioneers of performance
some of them presenting a useful low-degree of chemical reactivity thinking in construction [63]. In 1989 the first version of the Con-
[47,48] is unlimited in practical terms and may include mining re- struction Products Directive institutionalized the performance concept
sidues. It is not possible to include all of them in prescriptive standards. in the European Union. For construction products and systems, per-
However, requirements related to aspects such as occupational health formance-based standards have been a reality for > 30 years. Apart
and environmental contamination as well as long-term stability need to from European Union, other countries have adopted the same scheme.
be addressed. It includes Brazil (National System of Technical Approvals), Canada
There are also a variety of Si-Al-Ca-Mg minerals which can present (Canadian Construction Materials Centre), South Africa (Board of
various degrees of chemical reactivity when mixed with clinker Agrément South Africa), Australia (CSIRO), Israel (Building Systems
Portland, including pozzolanic reaction consuming portlandite. It seems Evaluation & Approval), Japan (Center for Better Living), USA (ICC),
reasonable to explore all of them, including SCMs with rather low re- Russian Federation (FCC). It means that, in most countries, the con-
activity such as the already accepted case of some fly ashes [49]. The struction sector, particularly the building sector which consumes most
metallurgical industry is certainly an interesting source, and literature of materials, is already familiar with the performance concept and its
reports potential utilization of slags such as ferro‑manganese [50], application.
stainless steel [51], ladle slag [52,53], electric arc-furnace steel slags The technical approval system has succeeded because it reduces the
both ordinary [53] and one optimized for cement production [54], inherent risk related to innovation [64], increasing the acceptance of
copper slag [55]. Some of these slags – manganese, nickel, EAF steel approved new technologies among private and commercial users, but
slag, are already in use by the Brazilian cement industry as “pozzolanic also policy makers [65]. It also reduces the risk from the producers'
materials”, because the Brazilian standard does not specify the source of point of view by giving a validated model to submit new technologies.
pozzolans. Ashes from a large variety of biomass are silicon-based and, However, currently the technical approvals are usually issued for pro-
when properly produced, can be used as pozzolans [56]. duct that has defined constituents and composition, an aspect that can
None of these potentially new SCMs isolated can solve the CO2 be overcame by better data able to link composition of reactive phases
problem. Some of them may not perform well in special markets due to as well as reaction kinetics to various aspects of performance.
leaching of chemical species that can contaminate water. The summa- The potential of performance standards in cement has been ac-
tion of all contributions of Si-Al-Ca-Mg residues, combined with a knowledge for almost 40 years. Frohnsdorff et al. [66] in 1983 stated:
variety of fillers are yet to be estimated, but it will probably be far from “it is generally agreed that performance standards should at least be avail-
insignificant. The problem is that including all these (new) minerals, able as an option”. Twelve years later, 1995, in the Workshop “Cement
and all possible combinations of SCMs rising from local availability, in and Concrete Standards for the Future”, Bryant Mather, ASTM C1
the cement specification is not practical. leader at that time stated “performance tests are necessary so that cements
The collective effort required to change prescriptive standard spe- and concretes with novel compositions can compete and innovative materials
cifications to include a single new SCM is large. Considering the variety, or systems can be evaluated.” [67]. The final report included the view of
it may become a never-ending task, probably making the effort not Frohnsdorff & Clifton in 1995 that predicted cement and concrete
justified. Market players that have no interest or access to a SCM will products standards for 2010: “Performance-based standards will be com-
not be keen to support. In a voluntary and consensus-based standardi- monplace and will co-exist with prescriptive standards; they will facilitate
zation process, a market player can take active measures to block the evaluation of new materials and materials to be used under unusual con-
inclusion of a given SCM on cement specification to prevent a compe- ditions.”
titor from gaining a competitive advantage. In addition, keeping the In the last 30 years, some progress has been observed in the adop-
current way of classifying cements will imply in a large growing tion of performance-based standardization for cement and perfor-
number of cements types, which may generate confusion and no ben- mance-based specification for concrete. In 1992, ASTM C1157 was
efits for the cement users. From this point of view, a new approach to published, a performance-based cement standard meanwhile the pre-
standardization seems unavoidable. scriptive ones were maintained. However, its market penetration was
low and, four years later, cements produced accordingly to this
4.2. Performance-based standard: an old proposal

Performance based standards specify goals to be reached or func- 1

It is worth noting that P.C. Hewlett, the current lead author of the book was
tions to be delivered, leaving the means to reach them to the market. In for long time director of the British Board of Agreement (BBA) in charge of
other words, the composition will not prescribe but rather specify in performance evaluation in the UK.

4
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

standard were not accepted by the ACI 318 and 301 concrete codes fulfilment of purpose and reduce risk of failures [85]. Various appli-
neither by the ready-mix concrete [68]. In 2000, only 6 out 35 de- cations of cement – reinforced concrete and rendering mortar, for ex-
partments of transportation from the USA did accept cements based in ample – have different requirements and the standard must encompass
the performance-based standard, but 32 of them approved blended all of those. A “pure” cement performance standard, as advocated by
cements accordingly to the ASTM C595, a prescriptive specification Frohnsdorff and Skanly [66], must include provisions for mineral ce-
[69]. Authors identify several reasons for that, including: (a) the lack of ments whatever its chemical and mineralogical composition, when used
minimum values on relevant requirements; (b) the complexity of the in any conceivable application and environment and various applica-
specification which offering too many options - five options for speci- tions, ensuring minimum service life. This will require a very complex,
fying compressive strength, which combined with the 6 cements types, time consuming and expensive evaluation. Despite the advance of
resulted in 1730 possible cement varieties; (c) the minimum strength knowledge, it will require a great effort of the research community.
required was lower than the minimum in the prescriptive ASTM C150 Simplification is, therefore, advisable.
and C595 and did not include 28 days requirements. Between 2005 and
2012, cement with up to 15% of interground limestone was commer- 4.3. Ideas for a simplified performance approach
cialized under the C1157 standard reaching in 2012 < 0.5 Mt. per year
– little more the 0.5% of the total production. Market acceptance A simplification without increasing risk can be achieved limiting the
started to growth after the Portland limestone cement was introduced in scope for systems based on clinker Portland with reactive SCMs from Si-
the prescriptive ASTM C595 in 2012, reaching about 1% of the cement Al-Ca-Mg system. For mostly inert mineral materials - fillers - require-
market by 2016 [70]. The peculiarity of the North-American cement ments must include dimensional stability and low water solubility. For
market to blended cements, which represents < 2% of the market [71] these systems, reliable thermodynamic simulations [6] that allow esti-
certainly have a role in the acceptance of the performance standard. mating some of the long-term performances are required; however, in
Almost 20 years ago, the EN 206-1 (Concrete Specification perfor- many cases, we fail in forecasting aspects related to durability and
mance production and conformity) introduced a flexibility on concrete environmental interaction. Limiting the approach to ordinary applica-
formulation by accepting concretes that do not comply with pre- tions – reinforced concrete, precast components, mortars – will further
scriptive specifications but have equivalent performance to a reference simplify without increasing the risk. In-use performance requirements
concrete that complies with the prescriptive standard. The demonstra- can also be considered by allowing technical approval for limited
tion of equivalent performance or, more precisely, equivalent durability markets, which further simplify durability studies. As soon as the
can be done as a European Technical Approval Document or through knowledge and experience accumulate, the scope could be safely wi-
national standard [72]. At least Portugal [73], Belgium [74,75] and UK dened.
[76] had put standards in place. According to VDZ [77], in Germany a Brazilian cement standard NBR 16697:2018 introduced a large
technical approval scheme was already in place in 2005. At least the flexibility in terms of pozzolana content (6–50%) and composition
Belgium standard applies the rule for non-standardized cements [74]. (SiO2 + Al2O3 + Fe2O3 > 50%, SO3 < 5%, total alkali content <
In 2015, both FIB [78] and ACI [79] issued reports in the subject and, in 1,5% Na2O eq., and loss on ignition < 6%) as defined by NBR
2016, RILEM published a more comprehensive report on Performance- 12653:2014. Additionally, a Brazilian standard (ABNT NBR 5752:2014)
Based specification [44]. Literature is becoming more and more abun- for determining the performance index of pozzolanic materials states
dant [75,80–83]. Nevertheless, almost all cements worldwide still are that a standard mortar formulated with 25% of Portland cement sub-
probably produced under prescriptive standards. stituted by the pozzolan must deliver 75% of the 28 days compressive
The building industry has already accepted the performance ap- strength of the reference mortar made with the same Portland cement.
proach in construction in a significant part of the world. Therefore, Water is fixed, but to overcome workability problems, superplasticizer
cement industry's clients probably are not a barrier to performance- can be added in the pozzolanic mortar to achieve the same flow-table
based cement in the markets where performance-based concept is al- consistence index of the reference mortar (ABNT NBR 5752:2014).
ready established for construction products. More than that, perfor- Although still based on broadly defined chemical composition, the
mance evaluation of cements is already present in some markets. At the Brazilian cement standard NBR 16697:2018 specification offers a very
same time, the prescriptive standardization regime is generating a large degree of freedom for the market in terms of origin, chemical and
growing number of new cement types which is becoming difficult to mineralogical compositions. There are no requirements of chemical
understand and forecast the expected performance. The growing pozzolanic activity, neither requirements of long-term dimensional
number of cement types combined with the accumulation of knowledge stability of all other phases eventually present (e.g. magnesium oxide or
is eroding the confidence on traditional prescriptive criteria, such as allotropic transformations). The minimum relative strength can be
water/cement ratio [74,79] and minimum cement content [84] in en- reached without significant chemical reaction, as simple side effect of
suring service life. As a result, countries with similar environment are porosity reduction resulting from the lower density product combined
adopting divergent criteria regarding the acceptability of cement types with fixed water/cement test and finer grinding of clinker. This flex-
for a given application [78] among others. ibility allowed industry to explore new SCMs as ferro‑manganese and
It is not a surprise that the introduction of an increasing variety of nickel slags “pozzolans”, in an unprecedent innovative activity. Despite
cement compositions, which has been described [64] as “no funda- the lack of detailed information regarding cement actual composition
mental innovations… but many variations”, are simple combinations of and actual Portlandite consumption, which may affect durability in
known materials and used for almost a century, resulted in technical specific environments, so far, no systematic problems associated to this
inconsistencies on complex subject such as durability. The reason is standard have been detected in the Brazilian market. Probably, this
inherent to the consensus-based elaboration process of prescriptive outcome is the combination of good performance of those cements in
standards, which is done most of times without throughout assessment the dominant application and environment, with a side effect of tech-
[1], fuelled by the confidence and familiarity resulting from years of nically careful adoption of the solution by cement producers aiming to
continuous use [1], with an empirical approach that still prevails in protect their market reputation. The fact that Brazilian regulation does
construction [64]. not limit leaching of dangerous chemical species can also play a role.
One reason for the fact that industry is still working on prescriptive Therefore, an excessively complex standard can inhibit innovation,
cement compositions is probably the complexity of the pure perfor- whereas a very flexible standard may increase the risks for the society.
mance approach [1,66,69], whose difficulties cannot be under- A performance-based standard must balance caution with the need of
estimated. Pure performance specification must be very comprehensive innovation [64].
to measure in detail all properties and performance indicators to ensure Requiring evidence of a minimum level of technology readiness as a

5
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

Fig. 1. - Life cycle of a cement that must be covered by a performance-based standard. By rigour, the raw materials used, the chemical composition of the cement will
be outside of the scope of a pure performance-based standard. The physical characteristics of cement powder will probably not be prescribed either but will be an
additional information provided.

precondition for the technical approval application [64] can also make condition. They should be described by their basic properties (e.g. al-
the system simpler and robust. Technology readiness level allows un- kaline reserve, gas and water permeability, rheological curves for ty-
derstand the limits of pure laboratory research, making evident further pical water/solid ratio). Resistance to chloride diffusion is relevant only
developments such as demonstration of good performance in relevant for steel reinforced applications with risk of corrosion. It should not be
environments integrated in the whole functional system [86]. relevant for mortar and unreinforced concrete. It may become a special
Performance requirements must include all life cycle of the product, type of cement, like the sulphate resistant.
from storage to the end of service life (Fig. 1). This encompasses fre- Carbonation is usually seen as a problem due to the depassivation of
quently neglected aspects such the shelf life of the stored cement and steel reinforcement and the consequent induced corrosion. However,
post-use “waste” impact. However, the major focus is the performance for applications such as mortars, unreinforced concrete, concrete in
of the cement paste, both in fresh and hardened states, perhaps as- indoor dry environments [89], carbonation acts a CO2 capture strategy.
suming aggregates chemically inert. In the fresh state, cement paste It not only can improve the microstructure of cementitious materials,
must provide a workable and stable (against phase separation) sus- but, most importantly, it may also reduce cement industry carbon
pension, ideally with a minimum amount of water. In this stage, re- footprint [90]. Of course, most of the reinforced concrete applications
quirements may consider the time-resolved rheological behaviour – need to be protected against carbonation, thus, cements designed to
today's limited to setting time - stability of the suspension regarding accelerate carbonation could be a special class of low CO2 footprint
phase separation under dynamic or static conditions (bleeding, ag- materials, exploring carbon capture and becoming attractive in many
gregate settlement, particle segregation, etc.), reaction kinetics, heat of markets.
hydration, chemical shrinkage, etc. The hardened cement paste must Minimum initial setting time is a simple method to establish the
provide adequate mechanical properties – a result of the dissolution- rheological behaviour variation over time, which is a performance re-
precipitation of hydrated phases increasing the solids volume – and quirement for fresh cement paste. But the 3D printing and shotcrete
dimensional stability, something strongly related to the microstructure, [91] applications require shorter setting times than ordinary concretes
particularly its pore structure. Durability requirements will depend on or mortars. More than that, each digital production route – from dry
the markets and the selected environments, because not all cements cast to 3D printing by extrusion – will benefit from cements with par-
must perform well in all possible environments. For hydraulic cement, ticular rheological behaviour over time, demanding greater control of
water resistance of the hydrates is mandatory. It also should include rheological behaviour than the one employed for today's technology.
requirements concerning the environmental impacts in cradle to cradle A digitalized cement plant and supply chain can become flexible
basis. The top agenda includes footprints of CO2, energy, water, waste enough to supply tailor-made cements and concretes to suit client's
and natural resources use. Another relevant topic is the risk of con- demand at minimum cost and environmental impact. This will require
tamination due to leaching of chemical species during the use and post- substantial advance of the knowledge and give rise to knowledge-based
use phases [52,87,88]. An appropriate assessment method, with ac- competition in the cement industry. By the other side, it may help in-
ceptable repeatability and reproducibility is needed for each require- dustry to remain competitive.
ment, which is not simple to be achieved without extensive research, as
RILEM workgroup report clearly demonstrates [44]. 4.5. Performance assessment

The application of performance-based standards is not straightfor-

4.4. Performance-based cement classification criteria
ward, requires deep and comprehensive knowledge of materials and its
interaction with possible environments. In building products, perfor-
A classification of cements based on their raw-materials composi-
mance assessment is a domain of specialized organizations, that in
tion is not coherent with performance-based approach. A performance-
Europe are united in the European Organization for Technical
based cement classification can be multidimensional, emphasizing re-
Assessment (EOTA) which issues Technical Approval for innovative
levant differences for the market's point of view, both professional as
products without standards, based in clearly stated methodology for
well as retail. For retail market, classification accordingly to application
each product class (a.k.a European Assessment Document). The German
– e.g. masonry cements – probably will be the best option. For the
Institute for Building and Technology provides this service for the in-
professional market, with bulk cement, a more sophisticated approach
dustry at least since 2005 [77], with several special cement products
may help cement industry maintain and even increase competitiveness,
being approved until today,2 including some containing new SCMs,
because it will allow industry to select the cheapest composition that
such as the Durabilo3 from Lafarge-Holcim. Still, the composition of
fulfil users needs. Nevertheless, experience related to ASTM 1157 [69]
those cement is clearly described.
recommends to keeping the classification system easily understandable
by professionals.
Performance requirements related to aspects such as durability, 2
For a list of technical approvals https://www.dibt.de/fileadmin/
mechanical performance, environmental or rheological behaviour are verzeichnisse/NAT_n/vSVA_3.htm.
3
relevant and different for each application and environmental https://www.holcimpartner.de/produkte/cement/durabilo-4.

6
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

The digital cement value chain scenario, as described in the pre- where admixtures dominate.
vious item, probably can be better explored if the industry embraces the Over the decades cement composition and fineness have changed.
performance concept: an accurate quantitative description how the Construction industry demanded shorter production cycles, requiring a
cement will perform over time before setting and after moulded will be cement with faster strength growth rate and higher final strengths. By
needed. This will require that researchers from industry and academy the other side, environmental pressures caused an increase of SCMs
collaboratively develop performance models that could be translated in content, reducing chemical reactivity and strength gain speed. To an-
common set of relevant performance indicators for cement. swer both demands simultaneously, cement surface area became higher
[92]. Increasing fineness becomes a tool to partially compensate the
5. Ideas for cement standardized tests for the future dilution by SCMs: the less reactive is the SCM and the higher its frac-
tion, the finer the cement tends to be. Additionally, due to the re-
Digitalization will make possible and desirable to reinvent most of placement of clinker by SCMs the cement density has been reduced,
the test methods currently used for all materials, including cement, in increasing the volumetric concentration of solids in a fixed 1:3 weight
order to reduce cost and labour, making viable the increase of test bases mix proportion mortar.
frequency. Discussing all the already existing opportunities, many of As a result of fineness and lower density, without packing optimi-
which in use at research labs, is not possible. Therefore, few tests we zation and dispersion the water demand for an acceptable rheological
consider important as cement performance indicators and particularly behaviour does increase. In this scenario, the fixed water/cement is
crucial to scale up tests' frequency for developing an AI-based cement becoming a problem: for many contemporary cements the ~0.5 water/
industry will be the focus of the discussion. cement is not enough to produce a workable mortar. In these cases, the
mortar is dry, and proper casting following the standard procedure
5.1. Mechanical performance becomes difficult, often impossible. Consequently, the test variability
increases due to higher frequency of large and random defects from
Worldwide, cements of all compositions, are classified mainly by inadequate workability. It has been found that sometimes variability
composition and 28 days minimum compressive strength4 with a fixed increases to the point of making the test statistically inadequate [93].
water/cement ratio. Compressive strength is not a performance char- When the amount of mixing water is adequate to ensure good work-
acteristic of a cement, since the actual strength depends on the amount ability, a coefficient of variation from 5% [94] to 7% [95] is expected in
of water in excess to the needed for chemical reaction; i.e. the water interlaboratory programs. On the other hand, for SCMs whose particle
amount added to provide acceptable workability. Cement compressive features (particle size distribution, surface area, morphology) have
strength test performed with constant water/cement ratio as done positive effects in flowability, which would allow the reduction of
today, is the prime indicator of the cement's chemical reactivity and mixing water could even be reduced to provide adequate workability,
binder capacity. as for fly ash (round particles) and some metallurgical slags (smooth
Despite today's importance, compressive strength requirement was vitreous surface) [96,97].
not universal in the past. For a long time, tensile strength was far more In both opposite cases, the fixed water/cement ratio is impairing an
common requirement. In the USA compressive strength became man- adequate evaluation of the cements: either causing an imperfect pro-
datory for all cements only in 1953 [28]. Its importance for markets duction of test specimens or penalising the strength values of cements
derives from the fact that stronger means better for most users, since with lower mixing water demand.
stronger construction is generally associated to a safer construction. To circumvent the moulding problems associated with fixed water
Technical users generally expect that selecting a cement with higher content, it is not uncommon to technicians to add dispersant admixture,
strength class will reduce the amount of cement needed in a product something that generally is not specified in standard test protocols.
formulation, therefore, a savings opportunity. ABNT NBR 5752:2014 for assessing the performance of pozzolanic
Today's typical standard tests are based on a mortar with mix pro- materials with Portland cement is an exception as mentioned in Section
portions of 1 part of cement to around 3 parts of sand, being the mixing 4.3. Despite being generally non-compliant with nowadays standards,
water constant around 0.5. It is 0.5 in European EN 197; 0.48 in this practice controls variability and may help to detect eventual in-
Brazilian NBR 7215. In the USA ASTM C109 specifies 0.46 for air-en- compatibility between cement and dispersants [98]. Fig. 2 exemplifies
trained cements, 0.485 for Portland cement and for blended cements the mentioned problems of a fixed water/cement showing the final
the value is defined by flowability. spread and appearance of mortars after flow-table test: images (a) OPC,
The amount of mixing water has been changed over time in almost (b) 50FA are Brazilian commercial cements, and (c) LC3 – lab mixed
all countries. The ASTM Portland cement specification used the normal cement using OPC, calcined clay and limestone filler. These mortars
paste consistency until 1934, fixed to 0.53 between 1934 and 1944 and had no dispersant and, therefore, comply with Brazilian standard with
defined by flowability 100–115% from 1944 to 1970 [28]. From 1937 fixed w/c of 0.48. Images (d) OPC_dis and (e) 50FA_dis are of mortars
to 1978 the Brazilian standard established strength with constant with w/c of 0.48 and the amount of dispersant to guarantee the full
flowability; from 1978 it was fixed to 0.48, despite that the standard dispersion, as determined by rotational rheometry [96,97]. The dry and
consistency test remained in the standards as a non-mandatory test. In non-cohesive behaviour of standard mortars with OPC and LC3 is evi-
Belgium, repeated changes on mixing water have also been reported dent – more intensely for the latter –, which is not favourable for a good
[92]. moulding. For the composite cement with fly ash (50FA), however, this
amount of water provided a fairly workable mortar with a cohesive and
5.1.1. Constant w/c is becoming a problem continuous paste. Contrastingly, the w/c value of 0.48 became too high
It is understandable the decision of the standardization bodies in when the suspensions were effectively dispersed (images d, e, f), re-
keeping the mixing water amount constant. Adjusting mixing water for sulting in extremely segregated mortars due to the very low paste
workability is time consuming and introduces another variable in the viscosity. Mortars with such behaviour are not suitable for proper
system: small changes due to experimental variation in water demand moulding either, since a very heterogeneous microstructure tends to
have large consequences on compressive strength. More than that, the form, higher concentration of sand at the bottom, a progressively
water demand with no dispersant admixture has limited correlation thinner paste at higher layers, and bleeding at the top. Optimum test
with actual industrial water demand in concretes, a major market conditions will require the water content to be adjusted in order to
achieve good rheological behaviour as it is in practical applications,
since proper workability is essential to ensure quality and productivity.
4
In the USA minimum strength is present but do not defines a cement class. The compressive strength obtained with fixed mixing water is an

7
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

Fig. 2. – Images, after flow-table test, of Brazilian

standard cement mortars prepared with fixed w/c
(0.48): (a) OPC, similar to CEM 1, (b) 50FA, com-
posite cement with 50% of fly ash and (c) LC3,
Portland limestone calcined clay cement; (d) OPC_dis
and (e) 50FA_dis with optimized content of disper-
sing admixture and fixed w/c of 0.48. The w/c of
0.48 was adequate only for 50FA without admixture,
whereas for all the other cements, this value was
inadequate: OPC and LC3 mortars were dry and with
no cohesion; OPC_dis, 50FA_dis and LC3_dis segre-
gated intensely, since 0.48 was too high when the
mortars were dispersed. Images (a), (b), (d) and (e)
from [96,97].

indicator of chemical reactivity but is becoming a poor estimator of the particles and dispersion can improve ITZ and bond strength, even with
cement consumption. For instance, the increase of water demand above 30%vol reduction of clinker content [105].
standardized 0.5 can cause a ~15–20% strength reduction in OPC One possible drawback of the technology based on dilution by filler
blended with diatomite and ~8% for pumice [99]. For the client, such is that significant reduction of mixing water demand implies in lower
blended cements will mean higher cement demand in comparison with paste volume for the same amount of cement. Considering the need of
reference OPC cement with the same strength class. minimum cement paste required to provide enough paste to maintain
The use of fixed amount of water can result in the cement being aggregates apart from each other ensuring workability, the amount of
classified with a higher standardized strength than the one justified by cement powder may increase in some practical applications. In this
its predictable in-practice performance. This approach penalises ce- cases, aggregate packing optimization can decrease the paste volume
ments engineered for low-water demand, which can be attained by demand. Nevertheless, even in the cases that an increase of cement is
improved particle features (particle size and morphology) and/or using required to compensate water reduction, filler dilution substantially
dispersing admixtures. With a fixed w/c in the strength test, these ce- reduces cementitious materials CO2 footprint [29] and is one of the
ments are classified at lower strength class than they perform, making most promising low-cost, worldwide-available strategy for emissions
them less appealing for the clients and inhibiting the industry to adopt mitigation [4].
this strategy to reduce clinker factor with fillers and dispersants. Hence, it is evident that both industry and society will benefit from
cement standards that consider mixing water demand as a performance
5.1.2. Environmental implications criterion of primary importance and test cement strength considering
The major consequence of compressive strength determined from its actual water demand.
mortars with fixed mixing water is that it prevents industry to exploit
the mitigation potential of low-water demand cements, which are based 5.1.3. 4.0 cement industry and compressive strength
on the employment of dispersant admixtures or a combination of dis- Cement performance optimization by AI (artificial intelligence) re-
persion, optimized physical characteristics (size distribution and mor- quires massive amounts of data, both from process and about the actual
phology) of particles and dilution by fillers. performance of the cement product. Mechanical performance is a re-
The use of admixtures is allowed in standards such as EN 197-1 levant performance indicator and a unique indicator of the chemical
2012 which limits to a maximum 1% of the mass. The decrease in water activity of the cement. Therefore, it seems to be crucial to properly
demand provided by dispersants makes possible to reduce up to 20% training AI based system-optimization.
the binder content for the same strength, which has obvious benefits However, the compressive strength testing are carried out currently
[100]. However, without adjusting the mixing water to the actual de- with low frequency, typically in a daily basis or even, as required by the
mand during application, the benefit is not evident if the cement EN 196-1 and other cement standards, only twice a week [106]. These
strength class is defined by a fixed water/cement. Apart from that, in a testing frequencies are not enough to provide the amount of data to
fully dispersed cement, the w/c range (0.48–0.5) of the standards de- develop a capable AI. And, because test is carried on “average” sample
finitely produces mortars very prone to segregation (Fig. 2, images d composed from sub-samples collected between tests, the effect of an
and e) to be properly tested [96,97]. eventual problems in a specific moment of production may be com-
The fixed water on the cement strength determination also hinders pensated in a sample with mostly good results. This makes difficult to
the cement industry to explore the technology in which dilution by correlate almost instantaneous process measurements with actual per-
filler is compensated by reduction of mixing water, achieved by packing formance, making AI training difficult. Furthermore, brittle materials
optimized to flowability and full dispersion. It allows reducing water by tend to require a considerable number of specimens to provide statis-
50% making possible substitution levels up to 70% of binder by inert tically reliable results of mechanical behaviour.
filler [29,37]. This strategy can reduce permeability, provide adequate One reason for the low frequency of testing is that, for process
carbonation rate and higher dimensional stability when compared with quality control, it gives answers too late, between 2 and 28 days after
reference Portland cements [101–104]. Interestingly, recent experi- the event. Speedier results will be definitively important. Additionally,
mental results also indicate that the combination of adequate fine filler testing for compressive strength has not been fully automated and

8
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

requires substantial specialized laboratory work. As discussed before, note a decrease of the correlation coefficient (R2 = 0.91); probably the
including the water demand in the standard is mandatory, which will lack of standardized method to determine the combined water has a
imply in additional work. Therefore, it is reasonable to think that di- large impact on the cwf. The typical scatter observed in Fig. 3 (a) CS vs.
gitalization of the cement industry can benefit from a different testing cwf curve is about ± 6 MPa in the worst case. This is a large value, but if
protocol. we compare with interlaboratory coordinated by ATIHL [94], the
compressive strength range for a single cement, with standardized test
5.2. Combined water fraction–a possible cement classification criterion that has been performed for long time is ± 5 MPa. A more detailed
analysis (not shown) of Fig. 3 (a) indicates that each cement formula-
Compressive strength is not a characteristic of cement: changing the tion has a different optimum trendline. If we take one single class of
water/cement ratio is possible to produce a range of strengths. A few cement tested in our lab using the same methodology, for example
years ago, in a discussion with the deceased Dr. Ellis Gartner, we variations around LC3 formulations (filler + calcined clay) or lime-
concluded that a SCM has almost always a double effect on cement: (a) stone filler only, considering combined water between 7 and 91 days,
it changes chemical reactivity; (b) its physical effect influences the R2 results around 0.94–0.98 and the data are typically scatter around
mixing water demand for a flowable paste and cementitious product. the regression line ± 3 MPa, being the differences of the regression line
The natural conclusion was that a simple indicator combining both coefficient significant. Apart of some model deficiencies, the major
dimensions would be useful, since we found the gel-space ratio difficult source of error is the determination of combined water, which max-
to work especially in a multi-composite cement. After some tests, the imum value for OPC varies between 22% up to 32% of the ignited mass
combined water (or bound water) fraction index (cwf) emerged: [112], as discussed in item Section 5.2.2. The results seem promising
wc and we are preparing a larger study on it.
cwf = Because porosity and mass transfer properties are correlated with
wm
strength, they will also be correlated to cwf. Therefore, it is also ex-
where wc is the combined water or bound water at any age and wm is the pected good correlation between cwf and crucial durability-related
amount of mixing water. The term combined water in this work refers performance indicators.
to the water chemically combined by the binder during hydration,
which was also defined in [107]. From data obtained in our laboratory, 5.2.1. Practical use of cwf concept
the determination of combined water followed these steps: (i) pastes The concept seems to be helpful as starting point for the design of
with w/c ratio of 0.5 were mixed with a high-energy rotational mixer, complex cement mixtures (Fig. 4) and their strength characterization
moulded and stored at 23 °C and 90–100% relative humidity (then kept considering both synergic chemical reactivity with possible influence
under water after 24 h). At the age of interest, slices (2 mm-thick) were on water demand. Because it introduces the mixing water demand in
removed and then underwent the hydration stoppage protocol with the equation of cement development, it facilitates the industry to better
isopropanol and diethyl ether. After hydration stoppage, ground sam- explore any locally available SCM. For example, it makes easier esti-
ples were subjected to thermogravimetric test (STA 409PC/PG, mating the reduction of mixing water demand needed to compensate
NETZSCH) up to 1000 °C (10 °C/min) in Nitrogen protective atmo- the low reactivity of SCMs in order to keep the performance. It also
sphere. Further details regarding this methodology can be found else- eases to estimate the variation of clinker factor to keeping under control
where [108,109]. The quantification of the combined water (Cw) is the cement performance when raw materials or process change, e.g.
expressed as follows [110]: unexpected changes in clinker, SCMs or sulphate reactivities. These
w − w550 ⎞ possibilities provided by cwf combine nicely with a digitalized cement
Cw = ⎛ 40
⎜ ⎟
plant, where each cement component can be ground separately, and the
⎝ w550 ⎠
process optimized by a sophisticated AI tool. It also allows estimate
where WX is the percentage of mass loss at temperature x°C. During the cement compressive strength at w/c = 0.5 (or other) from results ob-
data mining effort to test the concept, we found that the bound water- tained from other w/c, provided there is enough water for hydration.
to-mixing water ratio had been quickly presented by Powers and It is possible to classify cements using cwf at 28 days, a classification
Brownyard [111], who decided to adopt the gel-space ratio instead. that will be correlated to the cement's strength classes, providing con-
Gel-space ratio, despite being more precise, requires measurement of ditions for an easy transition between the two systems. Using the lim-
the degree of hydration and densities of each individual phase, which is ited amount of data, we have already (Fig. 4) estimated the 28 days
not easy specially in composite cements. cwf is much more practical, combined water: it resulted in 0.14, 0.18, 0.22 respectively for cement
since requires only two simple measurements: combined water at any classes for 32.5, 42.5 and 52.5 as tested with the fixed w/c of 0.5. Such
age – which can be achieved by TG or in a simple ventilated oven – and classification of cement may be complemented by the actual combined
the mixing water. These measurements can be performed in almost any water data at relevant age, making the cement formulation strategy
reasonably equipped laboratory. clear for the technical user.
Fig. 3 shows the correlation between cwf and compressive strength It also evidences the potential benefits from the development of
of 1:3 mortar cubes. Fig. 3 (a) shows 92 data from our lab with variable binders capable to combine with more water than the today's clinker
w/c ratio, half of them with dispersants, testing accordingly to [107] Portland and the feasibility of “super-cement” class equivalent to 72.5 –
whereas Fig. 3 (b) shows 147 results from literature with ages varying which will be justified only when aggregates are very well packed.
from 1 to 360 days. Data includes a variety of exotic SCMs, including
nano clay, graphene admixture, high amount (70%) of clinker sub- 5.2.2. Measuring combined water
stitution by limestone fillers, variations around LC3 formulation. Combined water is a fundamental property of a cement, directly
Data in Fig. 3 (a) grouped in function of the cement type show that, related to solids fraction increase which is responsible for strength
despite the good correlation (R2 ≥ 0.94) between compressive strength growth. It measures the amount of hydration of the binder fraction in
and cwf, the trendlines seem to be parallel among them and slightly the period between mixing and testing. It is sensitive to relevant and
shifted. For the same cwf the maximum variation among the cements is common situations related to cement hydration such as: clinker dilution
about 10 MPa (Fig. 3-a) and the slope of the curve related to each ce- by filler, partially or slowly reactive SCMs or even the influence of
ment families is slightly different. For instance, acid slag and fly ash admixtures on hydration kinetics and degree.
cements show similar behavior, whereas for filler up to 48% and OPC it Thermogravimetry is the most used research method to measure
is possible to note that for the same cwf the former reach a compressive combined water because it gives more information regarding hydrated
strength ~9 MPa higher. Analysing all data (Fig. 3-b) it is possible to phases, evidencing pozzolans or other SCMs reaction [49,107]. If the

9
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

Fig. 3. – Results of cwf versus compressive strength: (a) 92 results from our own data; (b) 239 samples 1:3 mortar from 10 sources [113–121] including our own data.
Data includes an assortment of SCMs and mixtures (LC3 cement being one of them) with ages from 1 to 360 days.

0.5 equipment. However, pre-drying at 110 °C gives another value, the

32.5 42.5 52.5 62.5 72.5 evaporable water content, which is significantly lower than the actual
chemically-combined water [112,123]. Heating up to 1000 °C will also
0.4
decompose the CO2 and cause some mass gain due to phase oxidation in
the uncontrolled atmosphere, particularly if slag is present. Perhaps a
Combined Water (g/g)

0.3 variation of TG sample treatment with isopropanol presented before,

can overcome the problems associated to drying at 110 °C. It is certainly
possible to develop a standardized reproducible and useful test based
0.2 on this platform. Chemical shrinkage also is a secondary measurement
tool [123,124], possibly correlated with combined water.
Measuring the combined water require careful handling but is not as
0.1 labour intensive as mechanical strength and seems it can be easily fully
automatized. It is not sensitive of moulding procedures and presence of
defects such as small air bubbles, but sampling handling and con-
0
0.3 0.4 0.5 0.6 0.7 0.8 ditioning is critical. It requires much smaller amount of materials and
Water demand (g/g) storage space than compressive strength. Currently available TG ma-
chines are not optimized for the productivity required in the industrial
Fig. 4. –The two routes available to control cement strength class: adjusting com-
environment, but at least one producer already has a sample carrousel
bined water capacity or controlling mixing water/cement (binder + filler) demand.
The green area is higher than the maximum combined water of Portland
that keeps samples in a N2 atmosphere until the test. The integration of
clinker. Model built by the average equation from Fig. 3, considering cement classes sample crushing, milling and treating with isopropanol within such
according to 28 days cement compressive strength in 1:3 mortar. (For interpretation machine will be highly desirable.
of the references to colour in this figure legend, the reader is referred to the web Because the combined water does not depend significantly on the
version of this article.) amount of mixing water - provided enough water and space (w/
c > 0.4) - the method allows determining the mixing water demand for
test is carried up to 1000 °C, the amount of CO2 due to limestone de- constant rheology being carried after starting the combined water tests.
composition is measured. The amount of combined water measurement These tests can be accelerated by increasing curing temperature to 40 °C
error is between ± 5 and ± 10% and it is associated with the proble- [122], therefore reducing curing time by a factor of 6.
matic of properly sampling small amounts, being recommended a
minimum 50 mg of hydrated paste to minimize this issue [107]. How-
5.2.3. Measuring mixing water demand
ever, sample preparation, handling and storage can introduce larger
Struble [125] already had emphasized water demand: “Because
errors. Best guide on cement testing is the one published by Scrivener
water requirement is such an important property of hydraulic cement,
et al. [109] as result of an effort of Nanocem team and chapter 5 is
adding a specification on water requirement for standard con-
dedicated to TGA [107]. Sample crushing and grinding is best to be
sistency...”.
carried quickly on CO2 free atmosphere and hydration being stopped by
Mixing water demand of cement can be measured by consistency
isopropanol followed by washing with diethylene ether and im-
index from the flow-table test in standard mortars, as it has been spe-
mediately transferred to N2 atmosphere of the TG equipment. It can be
cified in the past cement standards as previously mentioned in 5.1. For
accelerated to give just the information needed, by increasing heating
the development of cwf, our own data set was obtained for a fixed
rate from the usual 10 °C/min to 20 °C/min and limiting the testing
spread of 240 ± 10 mm (considering that initial diameter is 125 mm as
temperature to a maximum of 600 °C.
defined by NBR 13276). This value was adopted because resulted in
Simple small electric ovens, commonly used for ceramics in labs,
workable mortars suitable for moulding of a dozen cements, with and
may be good replacement of a TG machine, [122] and have the ad-
without dispersant, without problems due to extreme segregation or
vantage of the possibility of using larger samples in a much cheaper
lack of water and cohesion [96,97]. Fig. 5 shows a continuous and

10
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

Fig. 5. – Images, after flow-table test, of Brazilian

standard 1:3 mortars prepared with commercial ce-
ments and varying w/c for fixed spread diameter of
240 ± 10 mm. Mortars without dispersant: (a) OPC,
similar to CEM 1, (b) 50FA, composite cement with
50% of fly ash and (c) 49DE, composite cement with
49% of diatomaceous earth. Mortars with dispersing
admixture: (d) OPC_dis, (e) 50FA_dis, (f) 49DE_dis.
Values of w/c are indicated in the legends.
Originally from [96,97].

cohesive mortar with no signs of substantial phase separation. Thus, rheometers dedicated for cement pastes can be developed and em-
this method seems promising and could easily enable the im- ployed to determine water demand, either by the adoption of a standard
plementation of water demand and cwf concept for cement classifica- torque (or shear stress) level or by the identification of the fluidity point
tion, especially because it employs already established and widely used [132] in the mixing curve. The mixing rheometers allow measuring a
1:3 mortars and flow-table apparatus are cheap and available in almost more complete characterization of rheological behaviour and even
all labs. measuring mixing energy requirement, which can be a relevant para-
Another option already standardized and commonly used is the meter.
amount of water determined for paste of normal consistency (ABNT In the cement plant of the future with robotics and AI, the de-
NBR 16606, ASTM C187, EN 196-3). The test is cheap, simple and, most termination of water demand will be the first step of the series of tests
importantly, employs the single point “constant rheology” concept, to be performed in the cement paste, and this paste can then be possibly
which seems reasonable when comparing a great diversity of cements used for heat of hydration, automatic Vicat, rheometry, TG/DSC or
with different physical features – mainly particle size and surface area – chemical shrinkage performed by a robotic system integrating these
that greatly affect water demand. However, in the opinion of some different experimental techniques.
professionals from the cement industry, the test may be not sensitive Nevertheless, using the fixed spread on the flow-table, the paste of
enough. Other traditional test is the mini-slump test developed by normal consistency defined by Vicat's plunger, or an automatized so-
Kantro [126] and extensively used for dosage of dispersants; it is fast, phisticated mixing rheometer, the most important thing is that tests can
simple and cheap allowing the estimation of yield stress with good be performed with or without the presence of admixtures, which is
correlation with rheometer results with a coefficient of variation lower crucial to effectively employ cwf and extend the potential of cement
than 3% [127]. The test is currently under standardization by ASTM formulations. Coherent relations between the water demand of pastes
WK63516. and those of mortars and concretes will only be established if employed
Although the consistency at flow-table can be effective and induce a methods have the option to use dispersants. This approach has already
smooth transition from strength class to other concept, probably it is been applied to “provide sufficient workability…and consistent re-
not the best option for a 4.0 industry scenario. The digital production of lationship between paste and mortar/concrete rheology” as mentioned
cement-based materials will require stricter and meaningful control of by Bentz and Ferraris [134].
time-resolved rheological behaviour. Establishing rheological beha-
viour more precisely requires measuring more than one parameter
[125], which may become relevant. Additionally, since automation will 5.3. Setting time
be the rule, it is desirable easily automated methods. Flow table does
not seem simple to automatize. It also requires reference sands, which Cement technology is all based on the capacity of the powder to
increases cost and creates environmental impacts including waste. react with water, then to solidify and develop substantial mechanical
Other options based directly on measuring the paste are probably strength over time. Apart from false setting [125], cement set is a
better. Paste of normal consistency defined by standard test with the progressive change mainly – but not only – controlled by chemical re-
Vicat penetration plunger and Kantro's mini-slump are cheap but not actions of the cement, also affected by particle agglomeration. It aims to
easy to be automated. A best option from technical point of view will be define the workable period or open-time of the cementitious material
assessing the mixing behaviour of the paste as a function of water ad- before the suspension starts to solidify, which is an important perfor-
dition. Mixing rheometers have been used in other areas for a long time mance requirement. Its importance had been already noted by scientists
[128] with equipment available in the market capable of measuring the and engineers in the past. Vicat apparatus completed 200 years in 2018
mixing behaviour of wet granular materials [129] and concentrated as it was devised by L.J Vicat in 1818 for dealing with the setting time
suspensions as mortars and concretes [130,131] as exemplified in of hydraulic limes [27]. Original needle diameter (2.54 mm) and weight
Fig. 6. Hence, it is reasonable to expect that automatic mixing (1360 g) were different (larger and heavier) than the ones of the con-
temporary device, but stress at the needle point remains similar and it

11
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

Fig. 6. – Example of mixing curves to de-

termine water demand, measured by
PHESO rheometer, of mortars (25%vol of
fines and 75% ag) containing different types
of fine raw materials: hydrated lime
(Volumetric Surface Area = 15 m2/cm3,
D50 = 10 μm), lime stone fine filler
(VSA = 8 m2/cm3, D50 = 3 μm), lime stone
medium filler (VSA = 3 m2/cm3,
D50 = 23 μm) . After an initial water addi-
tion, agglomeration, dispersion and homo-
genisation take place resulting in the mixing
kinetics. Additional water contents are in-
troduced, causing immediate torque reduc-
tion, until the pre-established torque level is
reached.
Adapted from [133].

Fig. 7. – Comparison of setting time results measured by Vicat (ASTM C191) and by peak stress rheometry method.
Originally from [134].

still is the widely used standard test method to determine setting times. “Vicat setting times are arbitrary, in that they do not correspond exactly
These are defined by established values of penetration of the needle in a to any specific change in properties or to any specific levels of hydration
cement paste with standard consistency – determined with the same reaction” [125]. However, despite the current diverse high-technology
apparatus but with a plunger of 10 mm in diameter. The adopted measuring techniques available, the 200-year old test still is the rule for
parameter for final setting time is the same in Europe, Brazil and USA technological control of setting by the industry, apart from the less used
(as in the year 1818): 0.5 mm of penetration in the paste surface (EN calorimetry test of ASTM C1679 that provides a “thermal indicator” to
196-3, NBR 16607, ASTM C191). But initial setting time in the USA is estimate initial setting time. For research and development, though,
considered as a needle penetration of 25 mm (stays 15 mm from the different methods are employed to evaluate the physical effects of
bottom plate), while in Europe it is 36 mm of penetration (4 mm from setting like ultrasonic wave transmission [135,136] - a non-destructive
the bottom) and slightly different in Brazil, 34 mm of penetration physically-based test that continuous monitors the microstructural
(6 mm from the bottom) according to the mentioned national standards. changes of the paste; rheometry - rotational shear flow or oscillatory
These values are defined by convention and have changed over time, in [134,137–141]; and calorimetry is also used to investigate the relation
Brazil for example it was 1 mm in 1937 and is currently 6 mm. Alter- of the reaction kinetics with setting [134,138–141]. Furthermore, this
native test using Gillmore needles, as prescribed by ASTM C266, em- equipment can be adapted for industrial use and their cost reduces by
ploys the same principle but with different masses and needle geo- production scale.
metry. Different types of rheological measurements are employed to eval-
Vicat needle is unquestionably useful for determination of setting uate setting of cements. The stress growth technique determines an
time, but the evolution of cement compositions, the use of admixtures approximation of the yield stress at different elapsed times after mixing
and the development of new construction techniques have increased (thus requiring new samples for each measurement) and the inflection
complexity and importance of cement setting. As stated by L. Struble, point of the peak stress vs. time curve is considered as the initial setting

12
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

require constant consistency. A very interesting approach is the asso-

ciation of rheometry and calorimetry, enabling to distinguish physical
and chemical contributions to setting [139–141] by analysing G'/ac-
cumulated heat vs. time curves, which can be useful to evaluate com-
plex cement formulations that contain different types of admixtures
(dispersant and set regulators for example) and a variety of SCM's.
None of these methods are perfectly suited for the 4.0 industry
scenario, which requires automation and the possibility of scaling up
the number of tests, except perhaps the ultrasound test. Nevertheless,
these methods cannot compete with an automated Vicat when only cost
is considered.

5.4. Admixture and cement standards

Most of the cement sold in bulk will be mixed with admixtures,

specially dispersants. In many practical applications, admixtures in-
crease the efficiency of cement reducing the environmental impact, and
the CO2 footprint in particular up to 20% [100]. The new cementitious
products, such as concretes and mortars with high filler and low water,
UHPC and the trend of 3D printing [142] are not possible without a
combination of admixtures. Apart from dispersants as superplasticizers,
which are dominant in concrete, there are plenty of others. For mortars,
another important market, dispersants are employed only in specific
product types (mainly self-levelling compositions), but other ad-
mixtures -some also used in concretes - as air-entraining, viscosity-en-
hancing and redispersible polymers powders are commonly used and
affect rheological behaviour and, therefore, water demand in different
ways. Set regulators are becoming also important for digital production.
Cement characteristics, including but not limited to surface area,
may influence the amount of admixture needed to achieve desirable
effect. A full dispersion of a cement can be achieved with 0.3 to 1.3% of
Fig. 8. – Deflocculation curves. (left) Effect of cement composition for com- admixture expressed on the mass of fines below 100 μm.
mercial Brazilian cements in polycarboxylate ether dosage (11MS is a manga- Incompatibility between cement and admixtures is a relevant market
nese slag-blended cement; 16DE and 49DE are natural pozzolan blended ce- problem [98] and, in the case of superplasticizers, can cause an un-
ments with different content of diatomaceous earth [96,97]). (right) Effect of predictable retardation of cement hydration with many practical con-
admixture type – M is melamine; L is lignosulphonate and modified polyacrylic sequences. In certain cases, this problem is known to be related to
resin for the same Italian OPC cement [146]. Deflocculation is reached when sulphate and aluminates interaction and presence of soluble alkalis
yield stress or viscosity approaches the minimum or becomes relatively in-
[143,144]. However, the problem is considered by Marchon and Flatt
sensitive to dosage increase.
[98], to be one of the least well-understood topic and, considering the
ever-growing importance of dispersants, it certainly needs urgent at-
time; when a substantial change of consolidation kinetics occurs. tention.
Compared to ASTM C191 setting time by Vicat needle, the stress growth Despite being in use for long time in conjunction with cement and
technique resulted in considerable shorter times (as shown in Fig. 7), being crucial for many users, there is no requirement or even standar-
indicating a higher sensitivity of the rheometry-based method dized test to measure cement compatibility with admixtures, particu-
[134,138]. Small amplitude oscillatory shear (SAOS) rheometry does larly with the widely used dispersants. The development of such testing
assess rheological changes continuously (in a single sample) without methods is certainly a priority for the future of the industry.
disrupting the developing paste microstructure and provides informa- Cement composition influences superplasticizers dosage sig-
tion regarding elastic (G') and viscous (G") components of the viscoe- nificantly, by a factor of three as shown in Fig. 8 (left), a fact that has
lastic behaviour of the paste. G' is the most important parameter related important economic consequences. Dispersants are dosed by de-
to the set of cement (transition from fluid to solid) and yield stress vs. flocculation curves usually by a mini-slump test [126] that is being
time curve can also be drawn to point out the inflection point re- currently standardized by ASTM. Deflocculation curves [145] can be
presenting the initial setting time [139]. The use of such tests at in- obtained in a paste rheometer, using apparent viscosity and/or yield
dustrial scale demands further research and cost reduction of the stress data [96,97,146], a test that requires careful paste preparation for
equipment. proper repeatability. The admixture stability over time can be measured
The heat of hydration infers the setting time from chemical activity by flow or oscillatory rheometry as well but also by simple test such as
based on previous correlations with Vicat setting time. However, it may mini-slump. The eventual effect of admixtures on the cement hydration
not be sensitive to estimate the setting time when agglomeration plays and to setting time is best measured by isothermal calorimetry [143].
the major role in that period, particularly for cements with dispersants One problem for developing a standard compatibility test between
in which setting can be strongly affected by the admixture, especially dispersant admixtures and cements is the growing variety of admixture
when it loses efficiency, agglomeration is favoured and the con- products that behaves very differently with a given cement (Fig. 8
sequences on rheology of these suspensions with low water content are right), even under the same generic molecule type. Better under-
intensified. By the other side, calorimetry does not depend on the de- standing of the fundamental phenomena is needed.
termination of water for normal consistency. Rheology-based methods
measure the combined effects of agglomeration and chemical reaction
related to changes on the paste microstructure, but as does Vicat, they

13
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

5.5. Particle physical characterization 5.6. Other tests

5.5.1. Density Besides the ones discussed here, there are many other relevant tests
Density is a property that governs the volume of cement and, in a needed to better describe the cement and cement paste, both fresh and
world dominated by clinker blended with SCMs of various densities, hardened, generating fundamental data that will allow performance-
changing the density has important implications in mix design [125] based mix-design concrete. Struble [125] has conducted a compre-
and on economics, since variations higher than 10% are possible with hensive discussion of all those, like heat of hydration dimensional sta-
some pozzolans and fillers. Therefore, it is reasonable to mandatorily bility and including some crucial durability aspects that are relevant in
inform clients. In most of the current cement standards, density mea- some specific applications and regions. We have to agree with the au-
surement is part of Blaine fineness test. The most common test pre- thor state that “many of the tests described … are empirical and need a
scribed is the Le Chatelier flask, a time-consuming test that requires the better fundamental basis. In some cases, the fundamental knowledge is
use of volatile organic substances, generating vapours and waste. This available but just needs to be applied in standards.” And certainly,
test therefore is not suited to routine measurement [125], neither can many of the available tests are not suited for a more automated in-
be easily automated. Gas pycnometers are almost universal in other dustry, so innovation is welcome.
industries and allowed as alternative method in some standards. They Chemical and mineralogical characterization has been evolving ra-
are more precise than Le Chatelier but pycnometer gives densities 1% pidly with the electronics, including the now ubiquitous XRF to QXRD
higher implying that helium penetrates in pores inaccessible to kero- machines that can describe the material with unprecedent detail in a
sene [147]. very short time and at low cost. Nevertheless, most specifications of
compositional limits are not compatible with performance standards.
5.5.2. Surface area and particle size distribution
Particle size distribution and surface area are relevant information 6. Conclusions
for modern mix-design models based on packing [36,148], and this will
become more important in the future. The current standard method for This paper explored possible course of actions to better adapt the
surface area is the air-permeability Blaine test, invented in 1943 [149], cement standards for the two most important megatrends that may
which assumes spherical particles and gives only comparative results in disrupt society and, therefore, shape Portland cement future: environ-
a same set of materials. In standards, particle sizes are measured solely mental sustainability, more specifically resource efficiency and climate
by sieve residue tests. Both Blaine and sieving tests are limited for use in change and the revolution of the digitalization on industry – the 4.0
the new mix design models based on packing and mobility, besides industry. Rather than presenting definitive answers, the aim was to
being time-consuming and not fully automated. identify opportunities and challenges and encourage discussion, re-
Industry adopted laser diffraction as quality control method for search and innovation.
particle size distribution and in-line LD equipment for process optimi- Industry 4.0 artificial intelligence systems require easily automated
zation. Therefore, it seems to be the natural candidate for standardi- cement testing methods, capable of delivering larger data sets than the
zation [149]. However, for accurate results, full dispersion is needed conventional tests currently do. Most of the existing tests are not easy to
and it is not clear how agglomerated the dry particles are when diverted automatize and are difficult to scale up. Digital construction probably
for the in-line industrial test, but Ferraris and Garboczi [150] consider will demand cements with new performance requirements.
that only wet test can ensure full dispersion of bagged cement. Our own Departing from the traditional prescriptive composition of cement
results show that at least in some cases the use of dispersant is advisable standards to performance-based standards, limited to mixtures of
for wet testing in water [96]. Particle size distribution by laser dif- clinker Portland and any set of SCMs including inert ones, will allow
fraction certainly is helpful to technical costumers if informed by ce- industry to explore local available SCMs without waiting for time-
ment producers and, hence, needs to become a cement standard test consuming changes on prescriptive standards. Such standards must
method. Nevertheless, standardization of laser diffraction of cement cover performance requirements in broad sense, including mixing, en-
would definitely be a challenging project, since many aspects like vironment and durability aspects. This will, thus, help to reduce clinker
sample preparation, dispersing and testing methods, significant varia- factor and environmental impact in a world where blast furnace slag
tion in calculation algorithms of equipments from different manu- and fly ashes are scarce. It will also facilitate performance reliable-
facturers, and the scattering theory applied to analyse the results [151] based concrete design.
must all be considered, studied and specified.. Maximum size, as cur- But perhaps the most important conclusion is that the fixed water/
rently prescribed in standards, may remain as a proxy to control seg- cement, almost universally adopted to classify cements according to
regation risk. compressive strength, is hindering the industry to explore one of the
The BET adsorption test is the most fundamental specific surface most promising and cheap strategies to decreased clinker factor: redu-
area (SSA) test and ideally will be adopted by the industry. However, it cing the mixing water demand whilst maintaining good rheological
may take half-day to get results [149] with nowadays available behaviour, by engineering the physical effect of particles therefore
equipment, that also it is not automated. Therefore, without innovation controlling the physical effect of particles. This technology is better
on BET equipment, it is not going to be the specific surface area test for explored by compensating dilution of binders by inert filler or weakly
the 4.0 cement industry. However, for a same cement there is a linear reactive SCMs by reducing the water demand through combining dis-
strong correlation between BET and surface area estimated from laser persant admixtures with improved particle packing (by particle size and
diffraction or Blaine, but since the last two consider particles as perfect shape). Therefore, there is a benefit of allowing adjusting the mixing
spheres they give smaller areas and are not sensitive to the diversity of water to its actual demand when classifying cement by strength and
particle shapes and textures associated to original minerals character- informing clients about mixing water demand.
istics and their modification during chemical reaction [149]. By as- We had shown that varying combined water – by binder dilution or
sessing both laser diffraction and BET surface areas of large sets of grinding – and mixing water allows to formulate cements with more
cements, it could be established a correlation for each product, being degrees of freedom than the current strategy based only in chemical
useful to shorten testing time. Moreover, the shape factor (SSABET/ reactivity and fineness. The combined water fraction (cwf) – an index
SSALD) [36,96,105] is another important information provided by the that Powers overlooked favouring the more complex gel-space ratio – is
combination of methods to better understand particle morphology and strongly and linearly correlated with compressive strength, and pos-
texture, since dynamic image analysis (DIA) technique [152] is not yet sibly also correlated with other porosity-related relevant properties.
fully resolved for cements as recently faced by the authors. Therefore, combined water divided by the mixing water demand – a

14
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

proxy of the physical effect - can become a fundamental criterion for a 11 December 2012.
new classification of cements. These to parameters that form cwf can be [12] Construction Products Association, The Future for Construction Product
Manufacturing Digitalisation, Industry 4.0 and the Circular Economy,
easily measured, using traditional or innovative, easily to automate Construction Products Association, London, 2016https://www.
equipment. constructionproducts.org.uk/publications/corporate-and-industrial/the-future-
Industry can also benefit from rethinking other tests. Specific sur- for-construction-product-manufacturing/.
[13] World Economic Forum, Shaping the Future of Construction a Breakthrough in
face area and particle size distribution are becoming relevant in- Mindset and Technology, World Economic Forum, Geneva, 2016http://www3.
formation for modern mix design tools based on packing. Since laser weforum.org/docs/WEF_Shaping_the_Future_of_Construction_full_report__.pdf.
particle size distribution is already available in most cement plants it [14] PwC, Industry 4.0: Building the Digital Enterprise, https://www.pwc.com/gx/en/
industries/industries-4.0/landing-page/industry-4.0-building-your-digital-
seems reasonable that the method replaces the sieving testing still enterprise-april-2016.pdf, (2016).
present in many standards. Informing cement particle density is also [15] D. Lee, Robots to Affect up to 30% of UK Jobs, Says PwC, BBC News, 2017, http://
important because the differences of density between cements not only www.bbc.com/news/business-39377353 , Accessed date: 14 April 2017.
[16] Y. Roh, G. Heo, S.E. Whang, A Survey on Data Collection for Machine Learning: A
influences packing models, but also economics. Compatibility between
Big Data - AI Integration Perspective, ArXiv:1811.03402 [Cs, Stat] http://arxiv.
admixtures and cement is also a desirable aspect to be considered part org/abs/1811.03402, (2018) , Accessed date: 9 February 2019.
of future standards. [17] T. Linner, T. Bock, Evolution of large-scale industrialisation and service innovation
In summary, society and cement-related industries will have a better in Japanese prefabrication industry, Constr. Innov. 12 (2012) 156–178, https://
doi.org/10.1108/14714171211215921.
future if both, industry and academy, start to prioritize the develop- [18] M. Noguchi, The effect of the quality-oriented production approach on the de-
ment of science-based performance test protocols for cement and its livery of prefabricated homes in Japan, J. Housing Built Environ. 18 (2003)
products. 353–364, https://doi.org/10.1023/B:JOHO.0000005759.07212.00.
[19] P. Martinez, R. Ahmad, M. Al-Hussein, A vision-based system for pre-inspection of
steel frame manufacturing, Autom. Constr. 97 (2019) 151–163, https://doi.org/
Acknowledgments 10.1016/j.autcon.2018.10.021.
[20] M. Herrmann, W. Sobek, Gradientenbeton - Numerische Entwurfsmethoden und
experimentelle Untersuchung gewichtsoptimierter Bauteile (functionally graded
This paper is dedicated to Dr. Ellis Gartner, a bright, inspiring and concrete – numerical design methods and experimental studies on mass minimized
generous person, with whom V.M. John first formulated the funda- components), Beton- und Stahlbetonbau. 110 (2015) 672–686, https://doi.org/10.
mentals of combined water fraction, cwf, as a classification tool for 1002/best.201500035 (in German).
[21] M. Wörner, D. Schmeer, B. Schuler, J. Pfinder, H. Garrecht, O. Sawodny, W. Sobek,
cements. Authors would like to thank Rafael Cecel and Liz Zanchetta for Gradientenbetontechnologie: Von der Mischungsentwicklung über den
their contribution on key experiments; Arnaldo Battagin, from Brazilian Bauteilentwurf bis zur automatisierten Herstellung (the technology of graded
Portland Cement Association (ABCP) for providing historical data on concrete – from the development of concrete mixtures and the conceptual design
to the automatized manufacturing), Beton- und Stahlbetonbau. 111 (2016)
cement standards and test methods. InterCement funded part of the
794–805, https://doi.org/10.1002/best.201600056 (in German).
work. This research is part of the CEMtec project – National Institute on [22] C.M.R. Dias, H. Savastano Jr, V.M. John, Exploring the potential of functionally
Advanced Eco-Efficient Cement Based Technologies, supported by graded materials concept for the development of fiber cement, Constr. Build.
CNPq - Conselho Nacional de Desenvolvimento Científico e Tecnológico Mater. 24 (2010) 140–146.
[23] E. Lloret, A.R. Shahab, M. Linus, R.J. Flatt, F. Gramazio, M. Kohler, S. Langenberg,
- Brazil (Process 485340/2013-5) and FAPESP São Paulo Research Complex concrete structures, Comput. Aided Des. 60 (2015) 40–49, https://doi.
Foundation (Process 14/50948-3). P.C.R.A. Abrão work was partially org/10.1016/j.cad.2014.02.011.
supported by CNPq scholarship. M. Quattrone and F.A. Cardoso work [24] L.L. Stromberg, A. Beghini, W.F. Baker, G.H. Paulino, Application of layout and
topology optimization using pattern gradation for the conceptual design of
was supported by a CAPES/BRAZIL EMBRAPII scholarship. The in- buildings, Struct Multidisc Optim 43 (2010) 165–180, https://doi.org/10.1007/
formation presented in this study are those of the authors and do not s00158-010-0563-1.
necessary reflect the opinion of CNPq, FAPESP, CAPES or EMBRAPII. [25] S.R.M. Almeida, G.H. Paulino, E.C.N. Silva, Layout and material gradation in to-
pology optimization of functionally graded structures: a global–local approach,
Struct Multidisc Optim 42 (2010) 855–868, https://doi.org/10.1007/s00158-010-
References 0514-x.
[26] R.A. Buswell, W.R. Leal de Silva, S.Z. Jones, J. Dirrenberger, 3D printing using
concrete extrusion: a roadmap for research, Cem. Concr. Res. 112 (2018) 37–49,
[1] J.V. Tyrrel, Consensus standards formulation, Research and Innovation in the
https://doi.org/10.1016/j.cemconres.2018.05.006.
Building Regulatory Process: Proceedings of the Second NBS/NCSBCS Joint
[27] R.C. McWilliam, Concrete Testing Instruments, in: Instruments of Science: An
Conference, National Bureau of Standards, National Conference of States on
Historical Encyclopedia (Garland Encyclopedias in the History of Science),
Building Codes and Standards, Bozemen, Montana, 1977, pp. 161–164 https://
Routledge, London, 1997, pp. 145–147.
nvlpubs.nist.gov/nistpubs/Legacy/SP/nbsspecialpublication518.pdf.
[28] A.W. Isberner, P. Klieger, History and status of specifications and tests for strength,
[2] R.D. Hooton, Bridging the gap between research and standards, Cem. Concr. Res.
in: P. Mehta (Ed.), Cement Standards—Evolution and Trends, ASTM International,
38 (2008) 247–258, https://doi.org/10.1016/j.cemconres.2007.09.012.
100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428–2959, 1978,
[3] G. Frohnsdorff, J. Clifton, Cement and Concrete Standards of the Future Report
pp. 61–73, , https://doi.org/10.1520/STP663-EB.
from the Workshop, Gaithersbury (1995).
[29] V.M. John, B.L. Damineli, M. Quattrone, R.G. Pileggi, Fillers in cementitious
[4] K.L. Scrivener, V.M. John, E.M. Gartner, Eco-efficient cements: potential eco-
materials — experience, recent advances and future potential, Cem. Concr. Res.
nomically viable solutions for a low-CO2 cement-based materials industry, Cem.
114 (2018) 65–78, https://doi.org/10.1016/j.cemconres.2017.09.013.
Concr. Res. (2018), https://doi.org/10.1016/j.cemconres.2018.03.015.
[30] H. Moosberg-Bustnes, B. Lagerblad, E. Forssberg, The function of fillers in con-
[5] K. Breene, Can the circular economy transform the world's number one consumer
crete, Mat. Struct. 37 (2004) 74–81, https://doi.org/10.1007/BF02486602.
of raw materials? World Economic Forum Global Agenda, 2016 https://www.
[31] K.-C. Thienel, N. Beuntner, Effects of Calcined Clay as Low Carbon Cementing
weforum.org/agenda/2016/05/can-the-circular-economy-transform-the-world-s-
Materials on the Properties of Concrete, in: Concrete in the Low Carbon Era,
number-one-consumer-of-raw-materials.
University of Dundee - Concrete Technology Unit, Dundee, 2012, pp. 504–517.
[6] B. Lothenbach, D.A. Kulik, T. Matschei, M. Balonis, L. Baquerizo, B. Dilnesa,
[32] R. Fernandez, F. Martirena, K.L. Scrivener, The origin of the pozzolanic activity of
G.D. Miron, R.J. Myers, Cemdata18: a chemical thermodynamic database for hy-
calcined clay minerals: a comparison between kaolinite, illite and montmor-
drated Portland cements and alkali-activated materials, Cem. Concr. Res. 115
illonite, Cem. Concr. Res. 41 (2011) 113–122.
(2019) 472–506, https://doi.org/10.1016/j.cemconres.2018.04.018.
[33] Ch. Müller, K. Severins, B. Hauer, New findings concerning the performance of
[7] SJR, Cement and Concrete Research, Scimago Journal & Country Rank, https://
cements containing limestone, granulated blastfurnace slag and fly ash as main
www.scimagojr.com/journalsearch.php?q=23833&tip=sid&clean=0, (2019).
constituents; part 2, Cement International 8 (2010) 82–93.
[8] M. Hilbert, P. Lopez, The World's technological capacity to store, communicate,
[34] Ch. Müller, Performance of Portland-composite cements, Cement International. 4
and compute information, Science 332 (2011) 60–65, https://doi.org/10.1126/
(2006) 2–9.
science.1200970.
[35] UNEP, Decoupling Natural Resource Use and Environmental Impacts from
[9] E. Brynjolfsson, A. McAfee, The Second Machine Age: Work, Progress, and
Economic Growth, UNEP, Paris, 2011.
Prosperity in a Time of Brilliant Technologies, W.W. Norton & Company, New
[36] B.L. Damineli, V.M. John, B. Lagerblad, R.G. Pileggi, Viscosity prediction of ce-
York London, 2016.
ment-filler suspensions using interference model: a route for binder efficiency
[10] The Economist, A third industrial revolution - special report: manufacturing and
enhancement, Cem. Concr. Res. 84 (2016) 8–19, https://doi.org/10.1016/j.
innovation, The Economist, http://www.economist.com/node/21552901, (2012)
cemconres.2016.02.012.
, Accessed date: 11 December 2012.
[37] S. Palm, T. Proske, M. Rezvani, S. Hainer, C. Müller, C.-A. Graubner, Cements with
[11] A third industrial revolution - special report: manufacturing and innovation, The
a high limestone content – mechanical properties, durability and ecological
Economist, http://www.economist.com/node/21552901, (2012) , Accessed date:
characteristics of the concrete, Constr. Build. Mater. 119 (2016) 308–318, https://

15
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

doi.org/10.1016/j.conbuildmat.2016.05.009. Manchester Institute of Innovation Research Manchester Business School,

[38] H.S. Müller, M. Haist, M. Vogel, Assessment of the sustainability potential of University of Manchester, Manchester, 2013http://www.innovation-policy.org.
concrete and concrete structures considering their environmental impact, perfor- uk/share/14_The%20Impact%20of%20Standardization%20and%20Standards
mance and lifetime, Constr. Build. Mater. 67 (2014) 321–337, https://doi.org/10. %20on%20Innovation.pdf.
1016/j.conbuildmat.2014.01.039. [66] G. Frohnsdorff, J. Skalny, Cement in the 1990s: challenges and opportunities [and
[39] M. Haist, J.S. Moffatt, R. Breiner, H.S. Müller, Entwicklungsprinzipien und tech- discussion], Philosophical Transactions of the Royal Society of London. Series A,
nische Grenzen der Herstellung zementarmer Betone, Beton- Und Stahlbetonbau Mathematical and Physical Sciences. 310 (1983) 17–30.
109 (2014) 202–215, https://doi.org/10.1002/best.201300068. [67] B. Mather, Cement and concrete standards of the future, Cement and Concrete
[40] W.S. Weave, Committee C-1 on CementSeventy-five years of achievement, in: Standards of the Future: Report on the International Workshop, NIST,
P.K. Mehta (Ed.), Cement Standards—Evolution and Trends, ASTM International, Gaithersburg, 1997, pp. 19–24.
100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428–2959, 1978, [68] Concrete Construction, Prescriptive Vs. Performance Cement Specifications - a
pp. 3–15, , https://doi.org/10.1520/STP663-EB. Comparison of ASTM C 150, C 595, and C 1157, Concrete Construction, 1996,
[41] A.F. Battagin, F.A.C. Munhoz, I.L.S. Battagin, Evolution of standards, finennes and https://www.concreteconstruction.net/concrete-production-delivery-equipment/
compressive stregth of Brazilian Portland cements, Concreto e Construções XLI prescriptive-vs-performance-cement-specifications_o.
(2014) 39–51 (in Portuguese). [69] P. Tennis, J. Melander, User and producer perspectives on ASTM C 1157, Cement,
[42] European Standards, prEN 197–1 - New Working Document - CEN/TC 51/WG 6 - Concrete and Aggregates 23 (2001) 50, https://doi.org/10.1520/CCA10525J.
Definitions and Terminology of Cement, (2019). [70] A. Innis, Portland-Limestone Cement after 10 Years in the Field, http://www.
[43] R. Mielenz, L. Witte, O. Glantz, Effect of Calcination on Natural Pozzolans, in: T. cproadmap.org/publications/MAPbriefOctober2018.pdf, (2018).
Stanton, R. Blanks (Eds.), Symposium on Use of Pozzolanic Materials in Mortars [71] Portland Cement Association, 2016 U.S. Cement Industry Annual Yearbook, PCA,
and Concretes, ASTM International, 100 Barr Harbor Drive, PO Box C700, West Skokie, 2016http://www2.cement.org/econ/pdf/Yearbook2016_2sided.pdf.
Conshohocken, PA 19428–2959, 1950: pp. 43–43–50. http://www.astm.org/ [72] R. Breitenbücher, Revision of EN 206–1: What Will Be Modified? XVI ERMCO
doiLink.cgi?STP39404S (accessed March 26, 2016). Congress, Verona, 2012.
[44] H. Beushausen, L. Fernandez Luco (Eds.), Performance-Based Specifications and [73] H. Beushausen, M. Alexander, C. Andrade, M. Basheer, V. Baroghel-Bouny,
Control of Concrete Durability, Springer Netherlands, Dordrecht, 2016, , https:// D. Corbett, R. d'Andréa, A. Gonçalves, J. Gulikers, F. Jacobs, A.V. Monteiro,
doi.org/10.1007/978-94-017-7309-6. S.V. Nanukuttan, M. Otieno, R. Polder, R. Torrent, Chapter 8 application examples
[45] B. Lothenbach, G. Le Saout, E. Gallucci, K. Scrivener, Influence of limestone on the of performance-based specification and quality control, in: H. Beushausen,
hydration of Portland cements, Cem. Concr. Res. 38 (2008) 848–860, https://doi. L. Fernandez Luco (Eds.), Performance-Based Specifications and Control of
org/10.1016/j.cemconres.2008.01.002. Concrete Durability, Springer Netherlands, Dordrecht, 2016, , https://doi.org/10.
[46] K. Scrivener, F. Martirena, S. Bishnoi, S. Maity, Calcined clay limestone cements 1007/978-94-017-7309-6.
(LC3), Cem. Concr. Res. 114 (2018) 49–56, https://doi.org/10.1016/j.cemconres. [74] G.D. Schutter, Belgian implementation of the ECPC-concept following EN 206-1,
2017.08.017. Int. Conf. On Durability of Building Materials and Components, CIB, Porto, 2011,
[47] B.L. Damineli, Concepts for Formulation of Low-Binder Concretes, Dr. Eng. Thesis https://www.irbnet.de/daten/iconda/CIB22523.pdf.
Universidade de São Paulo (Escola Politécnica), 2013 (in Portuguese). [75] G. De Schutter, P. De Pauw, N. Reybrouck, R. Caspeele, Toward absolute dur-
[48] C. Vogt, Ultrafine particles in concrete: Influence of ultrafine particles on concrete ability performance criteria: preliminary case study of a sea lock, Journal of
properties and application to concrete mix design, Doctoral thesis KTH, 2010, Sustainable Cement-Based Materials 5 (2016) 23–34, https://doi.org/10.1080/
http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-12161. 21650373.2015.1077755.
[49] M.B. Haha, K. De Weerdt, B. Lothenbach, Quantification of the degree of reaction [76] W.J. McCarter, B. Suryanto, H.M. Taha, S. Nanukuttan, P.A.M. Basheer, A testing
of fly ash, Cem. Concr. Res. 40 (2010) 1620–1629, https://doi.org/10.1016/j. methodology for performance-based specification, Journal of Structural Integrity
cemconres.2010.07.004. and Maintenance 2 (2017) 78–88, https://doi.org/10.1080/24705314.2017.
[50] A. Rai, J. Prabakar, C.B. Raju, R.K. Morchalle, Metallurgical slag as a component 1318040.
in blended cement, Constr. Build. Mater. 16 (2002) 489–494, https://doi.org/10. [77] VDZ, Performance of cement (Chaper III), VDZ Activity Report 2003–2005, VDZ
1016/S0950-0618(02)00046-6. Verein Deutscher Zementwerke e.V, Duesseldorf, 2006, pp. 62–84 https://www.
[51] J. Rosales, M. Cabrera, F. Agrela, Effect of stainless steel slag waste as a replace- vdz-online.de/fileadmin/gruppen/vdz/3LiteraturRecherche/Activity_Report/AR_
ment for cement in mortars. Mechanical and statistical study, Constr. Build. Mater. Kap_III_UK.pdf.
142 (2017) 444–458, https://doi.org/10.1016/j.conbuildmat.2017.03.082. [78] C. Gehlen, Benchmarking of Deemed-to-satisfy Provisions in Standards: Durability
[52] V.Z. Serjun, A. Mladenovič, B. Mirtič, A. Meden, J. Ščančar, R. Milačič, Recycling of Reinforced Concrete Structures Exposed to Chlorides; State-of-the-art Report,
of ladle slag in cement composites: environmental impacts, Waste Manag. 43 Fédération internationale du béton, Lausanne, Switzerland, 2015.
(2015) 376–385, https://doi.org/10.1016/j.wasman.2015.05.006. [79] ACI Committee 329, Report on Performance-Based Requirements for Concrete,
[53] J. García-Cuadrado, A. Rodríguez, I.I. Cuesta, V. Calderón, S. Gutiérrez-González, Farmington Hills, MI, 2014. https://www.concrete.org/store/productdetail.aspx?
Study and analysis by means of surface response to fracture behavior in lime- ItemID=32914&Language=English&Units=US_Units.
cement mortars fabricated with steelmaking slags, Constr. Build. Mater. 138 [80] S. Helland, Design for service life: implementation of fib model code 2010 rules in
(2017) 204–213, https://doi.org/10.1016/j.conbuildmat.2017.01.122. the operational code ISO 16204, Struct. Concr. 14 (2013) 10–18, https://doi.org/
[54] H.-S. Kim, K.-S. Kim, S.S. Jung, J.I. Hwang, J.-S. Choi, I. Sohn, Valorization of 10.1002/suco.201200021.
electric arc furnace primary steelmaking slags for cement applications, Waste [81] S. von Greve-Dierfeld, C. Gehlen, Performance based durability design, carbona-
Manag. 41 (2015) 85–93, https://doi.org/10.1016/j.wasman.2015.03.019. tion part 1 - benchmarking of European present design rules, Struct. Concr. 17
[55] W.A. Moura, J.P. Gonçalves, M.B.L. Lima, Copper slag waste as a supplementary (2016) 309–328, https://doi.org/10.1002/suco.201600066.
cementing material to concrete, J. Mater. Sci. 42 (2007) 2226–2230, https://doi. [82] S. von Greve-Dierfeld, C. Gehlen, Performance-based durability design, carbona-
org/10.1007/s10853-006-0997-4. tion part 2 - classification of concrete, Struct. Concr. 17 (2016) 523–532, https://
[56] F. Martirena, J. Monzó, Vegetable ashes as supplementary cementitious materials, doi.org/10.1002/suco.201600067.
Cem. Concr. Res. 114 (2018) 57–64, https://doi.org/10.1016/j.cemconres.2017. [83] S. von Greve-Dierfeld, C. Gehlen, Performance-based durability design, carbona-
08.015. tion, part 3: PSF approach and a proposal for the revision of deemed-to-satisfy
[57] H.J. de Vries, W.P. Verhagen, Impact of changes in regulatory performance rules, Struct. Concr. 17 (2016) 718–728, https://doi.org/10.1002/suco.
standards on innovation: a case of energy performance standards for newly-built 201600085.
houses, Technovation 48–49 (2016) 56–68, https://doi.org/10.1016/j. [84] R. Wassermann, A. Katz, A. Bentur, Minimum cement content requirements: a
technovation.2016.01.008. must or a myth? Mater. Struct. 42 (2009) 973–982, https://doi.org/10.1617/
[58] A.M. Blayse, K. Manley, Key influences on construction innovation, Constr. Innov. s11527-008-9436-0.
4 (2004) 143–154, https://doi.org/10.1108/14714170410815060. [85] R.E. Price, Specifications and tests for new structural cements, in: P. Mehta (Ed.),
[59] European Conference of Ministers of Transport, Performance-Based Standards for Cement Standards—Evolution and Trends, ASTM International, 100 Barr Harbor
the Road Sector, OECD, 2005, https://doi.org/10.1787/9789282123386-en. Drive, PO Box C700, West Conshohocken, PA 19428–2959, 1978, pp. 87–98, ,
[60] OECD, Better innovation policies for better lives, Global Forum on the Knowledge https://doi.org/10.1520/STP663-EB.
Economy, OECD Publishing, Paris, 2011, p. 23. [86] N.D. Towery, E. Machek, A. Thomas, Technology Readiness Level Guidebook, U.S.
[61] J.R. Wright, The performance approach: History and status, Performance Concept Department of Transportation, Cambridge, 2017https://www.fhwa.dot.gov/
in Buildings: RILEM-ASTM-CIB Symposium Proceedings, U.S. National Bureau of publications/research/ear/17047/17047.pdf.
Standard, Philadelphia, 1972, pp. 847–853 https://nvlpubs.nist.gov/nistpubs/ [87] H.V.D. Sloot, A.V. Zomeren, R. Comans, O. Hjelmar, Harmonisation of Leaching
Legacy/SP/nbsspecialpublication361v2.pdf. Test Methods in Support of EU Regulations Controlling the Beneficial Use of
[62] G. Lambert, V. Nègre, E. Gallo, D. Rodriguez-Tomé, Ensembles urbains, Industrial Slag in Construction, Katholieke Universiteit Leuven, Leuven, 2011, pp.
1940–1977. Les ressorts de l'innovation constructive, CENTRE D'HISTOIRE DES 183–199 http://www.slag-valorisation-symposium.eu/2011/images/papers/s3_2_
TECHNIQUES ET DE L'ENVIRONNEMENT (CNAM), 2009. vandersloot.pdf.
[63] P. Spehl, From construction products to building works: The European input, [88] VDZ, VDZ Activity Report 2009–2012, VDZ Verein Deutscher Zementwerke e.V,
International Conference “Sustainable Construction International Conference Duesseldorf, Duesseldorf, 2013.
“Sustainable Construction”’”, Moscow State University, Moscow, 2011. [89] G. Sergi, A. Dunster, Corrosion of Steel in Concrete: A Review of the Effect of
[64] P.C. Hewlett, Interfacing innovation with best practice, in: R.K. Dhir, M.R. Jones Humidity, Building Research Establishment, Garston, Watford [England, 2004.
(Eds.), Innovation in Concrete Structures: Design and Construction, Thomas [90] F. Xi, S.J. Davis, P. Ciais, D. Crawford-Brown, D. Guan, C. Pade, T. Shi, M. Syddall,
Telford Ltd, Dundee, 1999, pp. 1–14, , https://doi.org/10.1680/iicsdac.28241. J. Lv, L. Ji, L. Bing, J. Wang, W. Wei, K.-H. Yang, B. Lagerblad, I. Galan,
[65] K. Blind, The Impact of Standardization and Standards on Innovation C. Andrade, Y. Zhang, Z. Liu, Substantial global carbon uptake by cement carbo-
Compendium of Evidence on the Effectiveness of Innovation Policy Intervention, nation, Nat. Geosci. 9 (2016) 880–883, https://doi.org/10.1038/ngeo2840.

16
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

[91] R.P. Salvador, S.H.P. Cavalaro, I. Segura, A.D. Figueiredo, J. Pérez, Early age [118] A. Schöler, B. Lothenbach, F. Winnefeld, M. Zajac, Hydration of quaternary
hydration of cement pastes with alkaline and alkali-free accelerators for sprayed Portland cement blends containing blast-furnace slag, siliceous fly ash and lime-
concrete, Constr. Build. Mater. 111 (2016) 386–398, https://doi.org/10.1016/j. stone powder, Cem. Concr. Compos. 55 (2015) 374–382, https://doi.org/10.
conbuildmat.2016.02.101. 1016/j.cemconcomp.2014.10.001.
[92] G. De Schutter, Evolution of cement properties in Belgium since 1950, Mag. Concr. [119] R. Snellings, L. Horckmans, C. Van Bunderen, L. Vandewalle, Ö. Cizer, Flash-cal-
Res. 53 (2001) 291–299, https://doi.org/10.1680/macr.2001.53.5.291. cined dredging sediment blended cements: effect on cement hydration and prop-
[93] E. Ülker, Comparison of Compressive Strength Test Procedures for Blended erties, Mater. Struct. 50 (2017) 241, , https://doi.org/10.1617/s11527-017-
Cement, Master Dissertation MIDDLE EAST TECHNICAL UNIVERSITY, 2010. 1108-5.
[94] ATILH, Interlaboratory Testing Programme 2017 Results from the First Statistical [120] X. Wang, Properties prediction of fly ash blended concrete using hydration model,
Analyses, Association Technique de l'industrie des liantshydrauliques, Paris, Sci. China Technol. Sci. 56 (2013) 2317–2325, https://doi.org/10.1007/s11431-
2017https://www.crb-gmbh.com/ringversuch/rfa/rfa-2016-2017/crb- 013-5319-x.
ringveruch-rfa_2017-07_atilh_zement.pdf. [121] K.D. Weerdt, E. Sellevold, K.O. Kjellsen, H. Justnes, Fly ash–limestone ternary
[95] CCRL, Cement and Concrete Reference Laboratory Proficiency Sample Program cements: effect of component fineness, Adv. Cem. Res. 23 (2011) 203–214,
Final REport Portrland Cement Profeficiency Samples Number 193 and Number https://doi.org/10.1680/adcr.2011.23.4.203.
194, Cement and Concrete Reference Laboratory, Frederik, Maryland, 2014http:// [122] F. Avet, R. Snellings, A. Alujas Diaz, M. Ben Haha, K. Scrivener, Development of a
www.ccrl.us/Psp/Reports/Portland%20Cement%20Report%20193.pdf. new rapid, relevant and reliable (R3) test method to evaluate the pozzolanic re-
[96] P.C.R.A. Abrão, F.A. Cardoso, V.M. John, Assessing the efficiency of Portland- activity of calcined kaolinitic clays, Cem. Concr. Res. 85 (2016) 1–11, https://doi.
pozzolana cements: reactivity, water demand and environmental indicators, org/10.1016/j.cemconres.2016.02.015.
Constr. Build. Mater. (2019) Submitted. [123] H. Köster, I. Odler, Investigations on the structure of fully hydrated Portland ce-
[97] P.C.R.A. Abrão, Pozzolan as Supplementary Cementitious Material: Availability, ment and tricalcium silicate pastes. I. Bound water, chemical shrinkage and den-
Reactivity, Water Demand and Environmental Indicator, Thesis (Master's degree) sity of hydrates, Cem. Concr. Res. 16 (1986) 207–214, https://doi.org/10.1016/
University of São Paulo, 2018 (in Portuguese). 0008-8846(86)90137-7.
[98] D. Marchon, R.J. Flatt, Impact of chemical admixtures on cement hydration, [124] M. Geiker, Characterisation of development of cement hydration using chemical
Science and Technology of Concrete Admixtures, Elsevier, 2016, pp. 279–304, , shrinkage, A Practical Guide to Microstructural Analysis of Cementitious
https://doi.org/10.1016/B978-0-08-100693-1.00012-6. Materials, 1st ed., CRC Press, Boca Raton, FL, 2016, pp. 75–102.
[99] W. South, A Study of the Compressive Strength and Drying Shrinkage of [125] L. Struble, 37 - hydraulic cements—Physical properties, in: J. Lamond, J. Pielert
Cementitious Binders Prepared Using Natural Pozzolans, PhD Thesis University of (Eds.), Significance of Tests and Properties of Concrete and Concrete-Making
Wollongong, 2009, https://ro.uow.edu.au/theses/3100. Materials, ASTM International, West Conshohocken, PA, 2006, pp. 435–449, ,
[100] J. Cheung, L. Roberts, J. Liu, Admixtures and sustainability, Cem. Concr. Res. 114 https://doi.org/10.1520/STP169D-EB.
(2018) 79–89, https://doi.org/10.1016/j.cemconres.2017.04.011. [126] D.L. Kantro, Influence of water-reducing admixtures on properties of cement
[101] J.J. Chen, A.K.H. Kwan, Y. Jiang, Adding limestone fines as cement paste re- paste—a miniature slump test, Cement, Concrete and Aggregates 2 (1980) 95,
placement to reduce water permeability and sorptivity of concrete, Constr. Build. https://doi.org/10.1520/CCA10190J.
Mater. 56 (2014) 87–93, https://doi.org/10.1016/j.conbuildmat.2014.01.066. [127] Z. Tan, S.A. Bernal, J.L. Provis, Reproducible mini-slump test procedure for
[102] T. Proske, S. Hainer, M. Rezvani, C.-A. Graubner, Eco-friendly concretes with re- measuring the yield stress of cementitious pastes, Mater. Struct. 50 (2017),
duced water and cement contents — mix design principles and laboratory tests, https://doi.org/10.1617/s11527-017-1103-x.
Cem. Concr. Res. 51 (2013) 38–46, https://doi.org/10.1016/j.cemconres.2013.04. [128] R.C. Rowe, Characterization of wet powder masses using a mixer torque rhe-
011. ometer. 4. Effect of blade orientation, Int. J. Pharm. 133 (1996) 133–138, https://
[103] C. Müller, S. Palm, C.-A. Graubner, T. Proske, S. Hainer, M. Rezvani, W. Neufert, doi.org/10.1016/0378-5173(95)04428-0.
I. Reuken, Cements with a High Limestone Content – Durability and Practicability, [129] Caleva, Mixer torque rheometer, www.caleva.com, (2019).
VDZ Verein Deutscher Zementwerke e.V, Düsseldorf, 2016https://www.vdz- [130] Kniele, Kniele Mixing Systems, www.kniele.de, (2019).
online.de/en/publications/concrete-technology-reports/. [131] Calmetrix, Pheso Rheometer, www.calmetrix.com, (2019).
[104] S. Hainer, T. Proske, C.-A. Graubner, Einfluss der Nachbehandlung auf das [132] B. Cazacliu, N. Roquet, Concrete mixing kinetics by means of power measurement,
Karbonatisierungsverhalten von Beton aus klinkerarmen Zementen, Beton- und Cem. Concr. Res. 39 (2009) 182–194, https://doi.org/10.1016/j.cemconres.2008.
Stahlbetonbau 110 (2015) 41–49, https://doi.org/10.1002/best.201400042. 12.005.
[105] E.B.C. Costa, F.A. Cardoso, V.M. John, Influence of high contents of limestone [133] F.A. Cardoso, A.C. Augusti, R.G. Pileggi, Influence of fine particles and viscosity
fines on rheological behaviour and bond strength of cement-based mortars, Constr. enhancing admixtures on the rheological behavior of mortars, Proceedings of the X
Build. Mater. 156 (2017) 1114–1126, https://doi.org/10.1016/j.conbuildmat. SIMPÓSIO BRASILEIRO DE TECNOLOGIA DAS ARGAMASSAS, ANTAC, Fortaleza,
2017.09.029. 2013, p. 12 (in Portuguese).
[106] D. Knöfel, Cement chemistry — Cement quality, in: O. Labahn, B. Kohlhaas (Eds.), [134] D.P. Bentz, C.F. Ferraris, Rheology and setting of high volume fly ash mixtures,
Cement Engineers' Handbook, 4. Engl. ed, Bauverl, Wiesbaden, 1983, pp. Cem. Concr. Compos. 32 (2010) 265–270, https://doi.org/10.1016/j.
103–170. cemconcomp.2010.01.008.
[107] B. Lothenbach, P. Durdziński, K.D. Weerdt, Chapter 5 thermogravimetric analysis, [135] H.W. Reinhardt, C.U. Große, A.T. Herb, Ultrasonic monitoring of setting and
A Practical Guide to Microstructural Analysis of Cementitious Materials, 1st ed., hardening of cement mortar—a new device, Mater. Struct. 33 (2000) 581–583,
CRC Press, Boca Raton, FL, 2016, pp. 177–206. https://doi.org/10.1007/BF02480539.
[108] M.H. Maciel, G.S. Soares, R.C. de O. Romano, M.A. Cincotto, Monitoring of [136] G. Trtnik, G. Turk, F. Kavčič, V.B. Bosiljkov, Possibilities of using the ultrasonic
Portland cement chemical reaction and quantification of the hydrated products by wave transmission method to estimate initial setting time of cement paste, Cem.
XRD and TG in function of the stoppage hydration technique, J. Therm. Anal. Concr. Res. 38 (2008) 1336–1342, https://doi.org/10.1016/j.cemconres.2008.08.
Calorim. (2018), https://doi.org/10.1007/s10973-018-7734-5. 003.
[109] K. Scrivener, R. Snellings, B. Lothenbach (Eds.), A Practical Guide to [137] S. Amziane, C.F. Ferraris, Cementitious paste setting using rheological and pres-
Microstructural Analysis of Cementitious Materials, CRC Press, Boca Raton, 2016. sure measurements, ACI Mater. J. 104 (2007) 137–145 http://www.concrete.org/
[110] K. De Weerdt, M.B. Haha, G. Le Saout, K.O. Kjellsen, H. Justnes, B. Lothenbach, Publications/InternationalConcreteAbstractsPortal.aspx?m=details&i=18576
Hydration mechanisms of ternary Portland cements containing limestone powder (accessed August 30, 2014).
and fly ash, Cem. Concr. Res. 41 (2011) 279–291, https://doi.org/10.1016/j. [138] D.P. Bentz, C.F. Ferraris, S.Z. Jones, D. Lootens, F. Zunino, Limestone and silica
cemconres.2010.11.014. powder replacements for cement: early-age performance, Cem. Concr. Compos. 78
[111] T.C. Powers, T.L. Brownyard, Studies of the physical properties of hardened (2017) 43–56, https://doi.org/10.1016/j.cemconcomp.2017.01.001.
Portland cement paste - part 6. Relation of physical characteristics of the paste to [139] A.M. Betioli, P.J.P. Gleize, D.A. Silva, V.M. John, R.G. Pileggi, Effect of HMEC on
compressive strength, JOURNAL of the AMERICAN CONCRETE INSTITUTE 18 the consolidation of cement pastes: isothermal calorimetry versus oscillatory
(1947). rheometry, Cem. Concr. Res. 39 (2009) 440–445, https://doi.org/10.1016/j.
[112] H.F.W. Taylor, Bound water in cement pastes and its significance for pore solution cemconres.2009.02.002.
compositions, MRS Proc. 85 (1986), https://doi.org/10.1557/PROC-85-47. [140] R.C.O. Romano, C. Liberato, M. Montini, J.B. Gallo, M.A. Cincotto, R.G. Pileggi,
[113] F. Deschner, F. Winnefeld, B. Lothenbach, S. Seufert, P. Schwesig, S. Dittrich, Evaluation of transition from fluid to elastic solid of cementitious pastes with
F. Goetz-Neunhoeffer, J. Neubauer, Hydration of Portland cement with high re- bauxite residue using oscillation rheometry and isothermal calorimetry, Appl.
placement by siliceous fly ash, Cem. Concr. Res. 42 (2012) 1389–1400, https:// Rheol. 23 (2013) 23830–23838, https://doi.org/10.3933/ApplRheol-23-23830.
doi.org/10.1016/j.cemconres.2012.06.009. [141] R.C.O. Romano, M.A. Cincotto, R.G. Pileggi, Hardening phenomenon of Portland
[114] M. Heikal, N.S. Ibrahim, Hydration, microstructure and phase composition of cement suspensions monitored by Vicat test, isothermal calorimetry and oscilla-
composite cements containing nano-clay, Constr. Build. Mater. 112 (2016) 19–27, tory rheometry, Revista IBRACON de Estruturas e Materiais 11 (2018) 949–959,
https://doi.org/10.1016/j.conbuildmat.2016.02.177. https://doi.org/10.1590/s1983-41952018000500003.
[115] M.S. Hemalatha, M. Santhanam, Characterizing supplementary cementing mate- [142] D. Marchon, S. Kawashima, H. Bessaies-Bey, S. Mantellato, S. Ng, Hydration and
rials in blended mortars, Constr. Build. Mater. 191 (2018) 440–459, https://doi. rheology control of concrete for digital fabrication: potential admixtures and ce-
org/10.1016/j.conbuildmat.2018.09.208. ment chemistry, Cem. Concr. Res. 112 (2018) 96–110, https://doi.org/10.1016/j.
[116] Z. Kledyński, A. Machowska, B. Pacewska, I. Wilińska, Investigation of hydration cemconres.2018.05.014.
products of fly ash–slag pastes, J. Therm. Anal. Calorim. 130 (2017) 351–363, [143] S. Dean, P. Sandberg, L. Roberts, Cement-admixture interactions related to alu-
https://doi.org/10.1007/s10973-017-6233-4. minate control, J. ASTM Int. 2 (2005) 12296, , https://doi.org/10.1520/
[117] R.G. Liu, P.Y. Yan, Study on hydration degree of Portland cement-slag complex JAI12296.
binders, Key Eng. Mater. 539 (2013) 172–177, https://doi.org/10.4028/www. [144] P.-C. Nkinamubanzi, P.-C. Aïtcin, Cement and superplasticizer combinations:
scientific.net/KEM.539.172. compatibility and robustness, Cement, Concrete & Aggregates 26 (2004) 1–8,

17
V.M. John, et al. Cement and Concrete Research 124 (2019) 105832

https://doi.org/10.1520/CCA12329. 1016/0008-8846(94)90022-1.
[145] J.E. Funk, D.R. Dinger, Predictive Process Control of Crowded Particulate [149] C. Ferraris, E. Garboczi, Identifying improved standardized tests for measuring
Suspensions, Springer US, Boston, MA, 1994http://link.springer.com/10.1007/ cement particle size and surface area, Transportation Research Record: Journal of
978-1-4615-3118-0 (accessed December 8, 2016). the Transportation Research Board 2342 (2013) 10–16, https://doi.org/10.3141/
[146] A. Papo, L. Piani, Effect of various superplasticizers on the rheological properties 2342-02.
of Portland cement pastes, Cem. Concr. Res. 34 (2004) 2097–2101, https://doi. [150] C. Ferraris, E. Garboczi, Measuring Cement Particle Size and Surface Area by Laser
org/10.1016/j.cemconres.2004.03.017. Diffraction, Transportation Research Board, Washington, D.C., 2013, https://doi.
[147] M.A. Helsel, C.F. Ferraris, D. Bentz, Comparative study of methods to measure the org/10.17226/22587.
density of cementitious powders, J. Test. Eval. 44 (2016) 20150148, , https://doi. [151] A. Jillavenkatesa, S.J. Dapkunas, L.H. Lum, Particle Size Characterization, NIST
org/10.1520/JTE20150148. Special Publication 960–1, National Institute of Standards and Technology (USA),
[148] F. de Larrard, T. Sedran, Optimization of ultra-high-performance concrete by the Washington DC, 2001.
use of a packing model, Cem. Concr. Res. 24 (1994) 997–1009, https://doi.org/10. [152] Sympatec, Dynamic Image Analysis, www.sympatec.com, (2019).

18