Research for Development

Joaquim A.O. Barros

Liberato Ferrara
Enzo Martinelli Editors

Recent Advances
on Green Concrete
for Structural
Purposes
The Contribution of the EU-FP7 Project
EnCoRe
Chapter 1
State of Knowledge on Green Concrete
with Recycled Aggregates and Cement
Replacement

Enzo Martinelli, Eduardus A.B. Koenders and Marco Pepe

Abstract Since the construction industry is characterized by a huge demand for

both energy and raw materials, it is fully concerned by the need for enhancing
sustainability, which is certainly the main challenge for all industrial sectors in the
twenty-first century. Therefore, several solutions are nowadays under investigation
to reduce the environmental impact of concrete production. They often consist of
partially replacing the ordinary concrete constituents with recycled ones, in view of
the objective of reducing both the demand of raw materials and the amount of waste
to be disposed in landfills. The most recent advances in this field are summarized in
this chapter, which is intended at drawing the line of the current state of knowledge
on “sustainable” structural concrete.

Concrete is the most widely used construction material and, hence, the reduction of
the environmental impact induced by its production processes is a relevant and
timely challenge for modern science and technology.
As a matter of fact, the production of concrete is characterised by a considerable
demand for energy and raw materials and results in significant emission of
Greenhouse Gases (GHG). Specifically, the cement production industry is deemed
responsible for about 5% of the total CO2 emissions, whereas the whole concrete
production leads to almost double this share (Moya et al. 2010).
Moreover, the construction of new buildings, as well as the maintenance and/or
demolition of existing ones, is responsible for the production of large amount of
waste, commonly referred to as Construction & Demolition Waste (CDW), which
generally require environment-sensitive and expensive disposal procedures (Moll
et al. 2005).
Therefore, recycling these waste to replace part of ordinary aggregates (OA) is a
straightforward and rational solution to produce more sustainable and

E. Martinelli (&) ! M. Pepe

University of Salerno, Fisciano, Italy
e-mail: e.martinelli@unisa.it
E.A.B. Koenders
TU Darmstadt, Darmstadt, Germany

© Springer International Publishing AG 2017 3

J.A.O. Barros et al. (eds.), Recent Advances on Green Concrete for Structural
Purposes, Research for Development, DOI 10.1007/978-3-319-56797-6_1
4 E. Martinelli et al.

environmentally-friendly concrete, also for structural applications (USDT 2004).

Furthermore, replacing part of the Portland cement needed in ordinary concrete
mixtures with alternative binders, often obtained by recycling industrial
by-products, is a viable option for reducing the emissions of GHG due to the
production of concrete (Lothenbach et al. 2011). These and other possibilities are
currently under investigation for enhancing sustainability in the concrete industry
by reducing the environmental impact due to producing and supplying the afore-
mentioned ordinary constituents and the water always needed in concrete mixtures
(Sandrolini and Franzoni 2001).
This chapter is intended at providing an overview of the current state of
knowledge about the physical and mechanical properties of a wide class of mate-
rials, often referred to as “green concretes” (fib 2013).
First of all, Sect. 1.1 analyses the most promising solutions for producing and
supplying alternative constituents whose use in concrete production can result in a
significant reduction of the environmental impact. Specifically, this close exami-
nation is subdivided into three main parts dealing with recycled aggregates, alter-
native binders and further solutions for the other concrete constituents. The main
physical properties, also connected to the most industrially feasible processing
solutions, are examined in these subsections that are intended at describing the main
differences between these “sustainable” concrete constituents and the ordinary ones,
in terms of both physical and mechanical properties.
Although several options have been explored for producing concrete and other
cementitious composites with recycled constituents, the following sections focus on
the ones intended at obtaining structural concrete by employing Recycled Concrete
Aggregates (RCAs), possibly in conjunction with industrial by-products, such as
Fly-Ash (FA) or Silica Fume (SF). Therefore, Sect. 1.2 outlines the most recent
findings about the fresh-state behaviour of concrete made with RCAs, generally
referred to as Recycled Aggregate Concrete (RAC), and the influence of the pos-
sible use of the aforementioned mineral additions. Moreover, the current state of
knowledge on the physical and mechanical properties of RAC and their correlation
with the relevant engineering properties of the constituents is outlined. Unveiling
these correlations is a key step for making predictable the resulting physical
characteristics and mechanical properties of these “green” concretes. However, no
general well-established theory has been formulated and validated so far. In this
respect, the current codes and guidelines generally provide strict limitations on the
use of recycled constituents: a wide overview of these codes and guidelines is
proposed in Sect. 1.4.
Moreover, Sect. 1.5 proposes an overview of the main contributions about the
environmental implications of both producing concrete with the aforementioned
components and quantifying the possible beneficial effects on sustainability.
It is worth highlighting that the following sections focus on the most recent
advances in this broad field of research and, hence, it refers to theoretical and
experimental contribution appeared in the last decade in the international scientific
literature. For a further discussion about the first studies on the mechanical char-
acterisation of cementitious composites made out by replacing part of the ordinary
1 State of Knowledge on Green Concrete with Recycled Aggregates … 5

constituents with recycled ones, the Reader may refer to well-established texts
(Hansen and Narud 1983; Hendriks et al. 2005).

1.1 Sustainable Concrete Constituents

Sustainable concrete constituents can be obtained by recycling various classes of

waste and by-products (Pacheco-Torgal et al. 2013). Since aggregates and binders
are generally the main ingredients in any concrete mixtures, the following sub-
sections only deal with the most recent advanced in classifying, producing, pro-
cessing and employing these two main constituents for producing structural
concrete. Although some studies address the effect of recycling water (CCAA
2007), this aspect is not covered in the following subsections.
It is worth highlighting that the definitions adopted hereinafter are inspired to the
classical “terminology” adopted by Hansen (1986a, b): therefore, the notation
adopted by other authors is sometimes modified for being consistent with the
aforementioned work.

1.1.1 Recycled Aggregates

Recycled aggregates can be produced by using various types of waste, often

deriving from Construction and Demolition Waste (CDW), but also obtained
through other types of waste that are not strictly connected with the construction
section and the concrete production (Kuosa 2012). Other sources of waste, not only
belonging to the class of CDW, that may be recycled and employed as aggregate in
concrete are listed and discussed into details by de Brito and Saikia (2013). Further
classifications are available both in national pre-standard regulation documents
(Kreijger 1981) and in the international scientific literature (Butler et al. 2014).
Other proposals deals with the use of such aggregates for specific purposes (Zhu
et al. 2011; Tebaldi et al. 2012). However, the present discussion is restricted to
CDW and the expression “Recycled Aggregates” (RAs) identifies aggregates pro-
duced by crushing and processing any kinds of CDW. Specifically, Recycled
Concrete Aggregates (RCAs) are those obtained by selecting, crushing and pro-
cessing concrete members coming from different sources, such as the demolition of
existing buildings or the recovery of residuals in pre-cast concrete factories (Pedro
et al. 2014) and unused concrete returned to plant (Ferrari et al. 2014).
Although RCAs are the most relevant option for the subject under discussion in
this book, RAs can be further classified by considering their original materials and,
hence, the following main classes can be recognised within CDW:
• Recycled Masonry Aggregates (RMAs), obtained by crushing masonry bricks
(Corinaldesi 2012);
6 E. Martinelli et al.

• Recycled Ceramic Aggregates (RCerAs), obtained by crushing waste ceramic

tiles and sanitary ware (Alves et al. 2014);
• Recycled Glass Aggregates (RGAs), mainly intended at replacing the fine fraction
of natural concrete aggregates (Adaway and Wang 2015; Mardani-Aghabaglou
et al. 2015).
Moreover, RAs derived by industrial activities that are connected with con-
struction and demolition (such as the extraction and transformation of marble
stones) are also considered in the scientific literature as a viable source of recycled
aggregate for concrete (Corinaldesi et al. 2010).
However, demolition and processing generally imply that the various types of
aggregates get mixed and, hence, some classifications adopted within the scientific
literature take into account the possibility that recycled aggregates do not belong to
a unique waste source (Yang et al. 2011). To this extent, the following classification
was recently proposed (Agrela et al. 2011):
• Concrete Recycled Aggregate (CRAs) in which concrete content is at least 90%
(in weight);
• Mixed Recycled Aggregate (MixRAs) in which the ceramic content ranges
between 10 and 30%;
• Ceramic Recycled Aggregate (CerRAs), containing more than 30% of ceramic
particles.
Since concrete and ceramic (sometimes using the latter term in a broader sense
including also masonry) are the main sources of waste produced in construction and
demolition of buildings, a further classification criterion intended at a visual
selection of these two main waste streams can be based on their colors and, hence,
defines the two following “fraction” (Toledo Filho et al. 2013):
• the grey fraction, consisting of particles mainly made of structural concrete (and,
in a minor portion, mortar) debris
• the red fraction, including clay bricks and ceramic-based (i.e. tiles) materials.
Classifications based on the original source of RAs are certainly useful, as they
can drive the selection and separation processes, either during demolition or in
dedicated recycling plants (Mas et al. 2012). However, a performance related
approach (WRAP 2007), intended at classifying recycled aggregates in terms of
their relevant engineering properties, would be more useful for the design of con-
crete mixtures employing these sustainable constituents. Experimental evidences
highlighted that the actual content of red particles controls the main engineering
properties of aggregate (Agrela et al. 2011). However, the mere classification by
origin does not lead to CDW aggregates with homogeneous properties: as a matter of
fact, water absorption measured in concrete aggregates, albeit collected in the same
geographic area, can be extremely variable (Angulo et al. 2010).
Therefore, a more accurate performance-based classification, based on the tests
usually carried out for concrete aggregates (CEN 2013) and intended at determining
the main geometric (such as grain size distribution and shape parameters), physical
1 State of Knowledge on Green Concrete with Recycled Aggregates … 7

(such as density and water absorption) and chemical properties (such as total sulfur,
acid-soluble sulfate and chlorides contents) is needed (McNeil and Kang 2013;
Rodríguez-Robles et al. 2014). For instance, the sink–float technique is a reasonably
feasible tool for separating CDW in different density classes (Angulo et al. 2010).
Based on the results of these tests, a close correlation between water absorption
(WA) and oven-dried density (ODD) was recently unveiled by analyzing the data
regarding 589 aggregates of different types, sizes, origins, and sourced drawn from
116 publications. The statistical processing of these data and, particularly, a
regression analysis carried on the two aforementioned quantities led to the fol-
lowing polynomial relationship (Silva et al. 2014):

WA ¼ A3 ! ODD3 þ A2 ! ODD2 þ A1 ! ODD + A0 ð1:1Þ

where WA is expressed in percentage and ODD in kg/m3, and the constants assume
the following values: A3 = 2.9373 10−9, A2 = −9.4014 10−6, A1 = 1.8977 10−2
and A0 = 65.745. Silva et al. (2014) also reported further statistical information
intended at describing the distribution of the experimental results with respect to the
relationship (1.1); particularly, a coefficient of determination R2 = 0.878 was
estimated.
The correlation described by relationship (1.1) is the basis for a classification
proposed by the same Authors for CDW waste in view of their use as RCAs (Silva
et al. 2014): four classes, denoted with A, B, C and D, ranging from the better to the
worse one, were defined.
A similar idea is presented herein, around the definition of a Quality Index QI of
RCAs:

WAOA
QIRCA ¼ ð1:2Þ
WARCA

where WAOA is the water absorption of ordinary aggregate that is going to be

replaced, and WARCA is the one of the recycled aggregate under consideration. In
principle, the QI should be defined for each relevant size range considered in the
mixture.
Both criteria and any other classification approach based on either of the
aforementioned quantities is based on the physical observation that, especially in
RCA (Pepe et al. 2014a, b), but also in other types of RAs (Corinaldesi and
Moriconi 2009a; García-González et al. 2014), porosity is mainly related to
attached mortar (AM). Since AM is one of the key parameters affecting the relevant
properties of concrete at both fresh and hardened states (Duan and Poon, 2014), it
should be carefully considered in rational mix-design rules for RAC (Fathifazl et al.
2010). As a matter of fact, these quantities can also be controlled by means of
various techniques intended at “cleaning” the outer part of AM and reducing the
particle of crushed concrete as close as desired to the original properties of the
original OA. In fact, the “autogenous cleaning” process, based on treating crushed
concrete particles in a rotating mill for a certain time, is a viable solution for
8 E. Martinelli et al.

reducing the amount of AM and, hence, controlling all the related engineering
properties of RCAs, such as density, water absorption and porosity (Pepe et al.
2014b). As it can be expected, the processing procedure plays a key role on the
resulting quality of RCAs and, hence, of RAC. In fact, experimental results
demonstrated that RCAs processed by coupling a primary plus secondary crushing
(PSC), using a jaw crusher followed by a hammer mill, performed significantly
better than the case in which a simple primary crushing process is executed (Pedro
et al. 2014).
The results show that the PSC process leads to higher performance, especially in
terms of durability. The experimental evidence highlights that higher contents of
AM with scarce presence of natural aggregates characterizes the smaller size
fractions of RAC (Evangelista et al. 2015). This is the reason why, as it will be
reported in Chap. 2, the use of RCA for replacing the fine fraction of is not gen-
erally allowed for producing structural concrete (NTC 2008). However, several
studies have demonstrated that the weaknesses induced by the higher porosity of
RCA can be somehow balanced by means of mineral additions often derived by
recycling environmentally harmful industrial by-products, such as FA and SF.
Particularly, the former, when employed in partial replacement of the finer fraction
of aggregates, has been shown to be able to enhance workability and strength of
RAC (Kou et al. 2008; Lima et al. 2013). Moreover, the addition of fly ash has been
also very effective in reducing carbonation and chloride ion penetration depths in
concrete, even in RAC (Corinaldesi and Moriconi 2009b). The aforementioned
by-products are also characterised by pozzolanic properties (Wang et al. 2013;
Dilbas et al. 2014) and, hence, they can be employed in partial substitution of
Portland cement, in view of an even more sustainable structural concrete.
Finally, RCA have been proved to be suited also for producing High-Performance
concrete: however, the quality (i.e. the mechanical properties) of the original concrete
plays a decisive role in limiting the maximum strength of RAC (Gonzalez and
Etxeberria 2014).

1.1.2 Alternative Binders

Portland Cement (PC) is an essential constituent of concrete as it has been produced

and utilized so far. As already said, the production of PC is deemed responsible for
a significant share of the global GHG emissions. Therefore, reducing the amount of
PC needed for producing structural concrete of a given quality (in terms of resulting
physical and mechanical properties) would result in a straightforward reduction in
the environmental impact of the concrete industry and the construction sector as a
whole.
The hydration reaction of PC developing in concrete mixtures during the phase
generally referred to as “setting” and “hardening” are characterized by a number of
coupled chemical processes whose kinetics is determined by both the nature of the
process and the state of the system at that instant. A macroscopic interpretation of
1 State of Knowledge on Green Concrete with Recycled Aggregates … 9

these reactions based upon combining two well-established physical tools, such as
the heat transfer theory and the Arrhenius principle, is capable of reproducing the
influence of thermal boundary conditions on the resulting rate of hydration and the
time development of the reaction heat and temperature throughout concrete
(Azenha 2009; Ventura-Gouveia 2011; Martinelli et al. 2013).
However, a more fundamental analysis of the hydration reaction highlights six
main chemical processes (generally referred to as: dissolution/dissociation, diffu-
sion, growth, nucleation, complexation, adsorption) that may develop either in
series or parallel, the latter case resulting in a further complication of this complex
chemical system (Bullard et al. 2011).
The growing interest for developing more sustainable concretes and cementitious
composite materials is leading to considering more and more the possible employ-
ment of secondary mineral additions, generally referred to as supplementary
cementitious materials (SCMs, Lothenbach et al. 2011) and often originating as
by-products of other industrial activities. Therefore, a more complete knowledge of
the fundamental mechanisms of hydration is needed to provide a rational basis for
selecting the most effective constituents and designing more sustainable concrete
mixtures. Furthermore, PC is not the ideal binder for all construction applications, as
it suffers from durability problems in particularly aggressive environments. Several
alternative binders have been available for almost as long as Portland cement but,
despite this, they have not been yet extensively used (Lollini et al. 2014).
The most promising alternative binders currently available can be classified as
follows (Juenger et al. 2011):
• Calcium aluminate cements featuring rapid strength development and good
durability in high sulfate environments;
• Calcium sulfoaluminate cements characterized by low CO2 emissions and
energy demand, but with several unknown aspects on the time development of
mechanical properties and long-term durability;
• Alkali-activated binders often obtained by recycling from waste materials and
industrial by-product and, hence, exposed to the natural variability about
physical compositions and chemical properties of these waste and by-products;
• Supersulfated cements almost entirely made from waste materials, coupled with
low heat production and good durability in aggressive environments such as
seawater.
Based on the above short descriptions, it is clear that the last two classes have
potential to be employed in “green concrete” for reducing the demand of PC. As
regards the alkali-activated binders (Pacheco-Torgal et al. 2008), the following five
categories can be introduced and considered in the following classification (Shi
et al. 2011):
• Alkali-activated slag-based cements, including blast furnace slag cement,
phosphorus slag cement, blast furnace slag-fly ash cement, blast furnace
slag-steel slag cement, blast furnace slag-MgO cement, blast furnace slag-based
multiple component cement
10 E. Martinelli et al.

• Alkali-activated pozzolan cements, including fly ash cement, natural pozzolan

cement, metakaolin cement, soda lime glass cement;
• Alkali-activated lime-pozzolan/slag cements, lime–natural pozzolan cement,
lime–fly ash cement, lime–metakaolin cement, lime–blast furnace slag cement;
• Alkali-activated calcium aluminate blended cement, including combinations of
calcium aluminate cement (CAC) with metakaolin, pozzolan and fly-ash;
• Alkali-activated Portland blended cement (hybrid cements), including Portland
blast furnace slag cement, Portland phosphorus slag cement, Portland Fly ash
cement, Portland blast furnace slag–steel slag cement, Portland blast furnace
slag–fly ash cement, multiple components blended cements.
The use of Fly Ash and Silica Fume in partial replacement of cement has been
also employed in junction with RCA (Corinaldesi and Moriconi 2009b; Mahmoud
et al. 2013; Lima et al., 2013).
More recent solutions for a partial replacement of cement in concrete were
developed by considering the ash obtained by burning municipal solid waste of
agricultural waste. Three relevant examples of these emerging supplemental
cementing materials are reported below:
• Rice-Husk Ash (RHA);
• Sugar Cane Bagasse (SCB) ash;
• Municipal Solid Waste Incinerator (MSWI) ash.
Rice husk is an agricultural residue that accounts for 20% of the 649.7 million
tons of rice produced annually worldwide. The chemical composition of rice husk is
found to vary from one sample to another due to the differences in the type of
paddy, crop year, climate and geographical conditions. Burning the husk under
controlled temperature below 800 °C can produce ash with silica mainly in
amorphous form Ghassan and Hilmi (2010). The performance of RHA as a sup-
plemental cementing materials was even investigated in Ultra-High Performance
Concrete (UHPC) and the obtained results highlighted that RHA acts both as highly
pozzolanic admixture and internal curing agent in UHPC (Van et al. 2014).
SCB ash is a by-product of the sugar/ethanol agro-industry abundantly available
in some regions of the world and has cementitious properties indicating that it can
be used together with cement (Fairbairn et al. 2010).
MSWI ashes have several applications. As regards the applications of relevance
in cement and concrete industry, on the one hand, the addition of MSWI ash for
clinker production has been demonstrated to shorten the setting time and decrease
workability. On the other hand, experimental results demonstrate that addition of up
to 50% treated MSWI fly ash does not significantly affect the mechanical properties
(Lam et al. 2010).
Finally, nanotechnology offers further options for reducing the amount of cement
needed in concrete. Specifically, incorporating colloidal Nano-Silica in concrete
with 100% coarse RCAs led to similar results, in terms of mechanical properties,
with respect to a reference concrete mixture (Mukharjee and Barai 2015).
1 State of Knowledge on Green Concrete with Recycled Aggregates … 11

1.2 Fresh-State Behaviour

The peculiar features characterising RCAs have a significant influence in affecting

the properties of concrete at the fresh-state. Particularly, they generally affect
workability due to the two main reasons explained below:
• the higher porosity and water absorption capacity of RCAs have an interaction
on the water actually available in the mixture: initially dry RCAs result in a
reduction of the free water and, hence, a reduction of workability, whereas, part
of the water absorbed in initially saturated ones can have the opposite effect
(Pepe et al. 2014a);
• the higher irregularity and roughness of RCA particles with respect to ordinary
aggregates generally leads to reduced workability (Safiuddin et al. 2011).
On the contrary, the addition of FA to RAC mixtures generally enhances
workability: this is mainly due to the higher regularity and fineness of FA particles,
which contribute to reducing friction interactions among aggregates in both normal
(Lima et al. 2013) and self-compacting (Revathi et al. 2013) RAC.
Although the slump test (CEN 2009) is the most common methodology for
quantifying workability in fresh concrete mixture, a systematic investigation carried
out on concrete mixtures characterised by variable aggregate replacement ratio and
water compensation rate demonstrated that the VeBe time and flow table tests are
more suitable to determine workability of RAC. In fact, these testing techniques
result are more capable to detecting the influence of relevant quantities, such as free
water content and aggregates’ shape (Leite et al. 2013). As a general observation, as
it is easy to expect that, by increasing the water compensation rate results in
enhancing workability (as a result of the well-known Lyse’s rule) and reducing
compressive strength at the hardened state (as a result of the well-known Abrams’
law).
Therefore, chemical admixtures are required to guarantee the target workability
without inducing detrimental effects on the resulting mechanical properties of RAC.
Particularly, a new generation of superplasticizers containing some copolymer
polycarboxylate makes it possible to significantly improve the fluidity of the RAC
(Braymand et al. 2015). To a certain extent this superplasticizers can also com-
pensate the detrimental effect induced by replacing the fine fraction of aggregate
with recycled sand (Pereira et al. 2012).
An alternative option, intended at reducing the interaction of RCAs with free
water, is based on tailored polymer-based treatments which consist in soaking these
aggregates in polymer solutions and developing a water-repellent film on the
aggregate outer surface. A recent study compared the effect of various polymers and
variable concentrations of the aforementioned solutions on both the fresh and
hardened state properties of RAC (Spaeth and Tegguer 2013).
Finally, due to the same reasons affecting workability, the presence of RCA has
been proved to play a role in the initiation and development of the hydration
reaction of concrete at early age: the aggregate replacement ratio modifies the
12 E. Martinelli et al.

reaction kinetics (Koenders et al. 2014) and the initial moisture condition of
aggregates modifies the initial rate of reaction and the time development of cement
hydration (Pepe et al. 2014a).

1.3 Hardened-State Behaviour

The properties of aggregates play a significant role also on the resulting behaviour
of RAC at the hardened state, both in terms of mechanical response and physical
durability-related parameters.
The present review of the most recent advances on this topic focuses on RAC
made from RCAs, as it is proved to be the best suited option for producing “green”
structural concretes. However, it is worth mentioning that studies on the mechanical
characterisation of concrete with coarse RMAs or MixRAs (see Sect. 1.1) are
available in the literature (Gomes and de Brito 2009). They demonstrate that limited
replacement of natural aggregates with the aforementioned mixed recycled ones
(lower than 25% in weight) result in concrete characterised by sufficient strength
and durability for housing construction and almost the same physical properties
with respect to the reference mixtures (Medina et al. 2014). Moreover, studies on
the use of RGAs are also available: they are often intended at investigating the
conditions of occurrence and the solutions for suppressing the Alkali-Silica
Reaction (ASR) which can be harmful for the material durability (Rajabipour et al.
2010). The resulting behaviour of RAC made from mixed glass/concrete RAs has
been also investigated for understanding the concurrent influence of the lighter
glass particles and the more porous crushed concrete ones (Mardani-Aghabaglou
et al. 2015). Furthermore, although the focus of the present review is on RAC for
structural purposes, RCA seems particularly suited for some specific non structural
applications, such as the production of pervious concrete (Chen et al. 2012;
Güneyisi et al. 2016), as they are mainly made of coarse particle and should
guarantee a significant water porosity (Sriravindrarajah et al. 2012). In fact,
replacing the fine fraction of aggregates with fine RCA results in several detrimental
effects on the resulting mechanical performance of RAC, among which a significant
increase in shrinkage and creep deformation (Cartuxo et al. 2015).
Since the compressive strength fc (generally determined after 28 days of curing)
is the main design parameter for structural concrete, several studies on RAC focus
on determining the role of RCA on the resulting value of this property. Although
these are often limited to an empirical observations of the role of RCA on fc,
scrutinising the cement reactions developing in RAC unveils the fundamental
influences of the main engineering parameters of both aggregates (i.e. either open
porosity or water absorption capacity, along with their initial moisture condition at
mixing) and mixtures (i.e. the water/cement ratio) on the resulting hydration
kinetics (Koenders et al. 2014). However, it was demonstrated that RAC is affected
by curing conditions (either laboratory conditions, external environment or wet
chamber) roughly in the same way as ordinary concrete (Fonseca et al. 2011).
1 State of Knowledge on Green Concrete with Recycled Aggregates … 13

Between the mere empirical observation and these fundamental modelling

approach, various quantitative relationships have been recently proposed for
expressing the correlation between fc depending on the main parameters describing
the mix composition. These relationship are often formulated in terms of the
fc,RAC/fc,RC ratio between the compressive fc,RAC of RAC and the one, denoted
fc,RC, of the reference concrete (RC) mixture made with only ordinary aggregates
and the same size grading of RAC (de Brito and Alves 2010):
! "
f c;RAC DRAC
¼ 1 & 2:619 ! 1 & ð1:3Þ
f c;RC DRC

where DRAC and DRC denote the weighted density of RAC and RC, respectively,
determined across the various size fraction of aggregates included in the concrete
mixtures under consideration. The relationship (1.3), determined on seven different
series of experimental tests (mainly carried out on mixtures of RCAs) was cali-
brated for weighted density ratios ranging between 0 and 0.10 and exhibited a
reasonably good correlation with the aforementioned experimental results (ex-
pressed by a coefficient of determination R2 = 0.7615). The same authors (de Brito
and Alves 2010) proposed similar relationships intended at expressing the corre-
lation between the same weighted density ratio and other concrete properties, such
as compressive strength at 7 days, splitting and flexural tensile strength, modulus of
elasticity, abrasion resistance, shrinkage, water absorption, carbonation penetration
and chloride penetration: the analytical expressions of these relationships are
omitted herein and the interested Reader can refer to the cited paper for further
details.
The knowledge achieved so far about the influence of RCAs on the resulting
strength of RAC led to the formulation of generalised mix-design rules for the latter
(Fathifazl et al. 2009; Yehia et al. 2015): a physically-based conceptual proposal
based on the findings of a recent Ph.D. thesis (Pepe 2015) will be presented in
Chap. 6. The structural scale response of RAC was investigated under both per-
manent and variable loads. On the one hand, the results obtained on RAC with
100% of coarse RCAs reveal that the ratio between strength determined at low
loading rate and the standard one in compression and in tension is similar for RAC
and ordinary concrete (González-Fonteboa et al. 2012). On the other hand,
beam-to-column joints made of RAC with 30% limited replacement ratio (30% of
the coarse aggregates) and subjected to cyclic-actions highlighted the suitability of
using RAC in seismic zones (Corinaldesi et al. 2011).
Moreover, an empirical correlation was proposed for expressing the time evo-
lution of strength (Malesev et al. 2010)

a!t
f c ðtÞ ¼ ; ð1:4Þ
tþb

where a and b are two constants whose values can be derived through best-fitting of
the available experimental result: as it is clear, the parameter a represents the
14 E. Martinelli et al.

asymptotic value fc,∞ of fc, whereas b (dimensionally a time quantity) controls the
initial rate of growth of compressive strength. The relationship (1.4) will be con-
sidered and calibrated in the following Chap. 3 for the concrete mixtures reported
therein.
Due to the peculiar properties of RAC and their interaction with the other
mixture constituents, the resulting correlation between the compressive strength fc
and the other mechanical parameters cannot generally be expressed by means of the
same analytical expressions adopted in Codes (CEN 2005) and Guidelines (fib
2013) for ordinary structural concrete. As regards the Young’s Modulus Ec, the
following correlation between its mean value and the cubic compressive strength of
concrete fc,cube proposed (Corinaldesi 2010):
rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
3 0:83 ! f c;cube
Ecm ¼ 18800 ! : ð1:5Þ
10

This relationship was calibrated on the results obtained from RAC specimens
with five different water/cement ratios (ranging from 0.40 to 0.60) and replacement
ratio of 30% of the coarse aggregates. The experimental results highlighted that
compressive strength was almost unaffected by RCAs, whereas RAC exhibited
lower static elastic modulus: this reduction was around 10% and justifies the
reduction of the coefficient adopted in Eq. (1.5) with respect to the one currently in
use for ordinary concrete (CEN 2005; fib 2013).
Moreover, analytical relationships were also proposed for generalising the
Sargin curve, adopted by the aforementioned Codes and Guidelines, to take into
account the increase in axial deformability observed in RAC. Specifically, three
rec
coefficients arec rec
c ; bcu and ucm were calibrated by González-Fonteboa et al. (2011)
for modifying the values of ec2, ecu and Ecm defining the stress-strain curve for
ordinary structural concrete (CEN 2005; fib 2013):

arec
c ¼ 0:0021 ! RRCRCA þ 1; ð1:6Þ

brec
c ¼ 0:0022 ! RRCRCA þ 1; ð1:7Þ

urec
cm ¼ &0:0020 ! RRCRCA þ 1; ð1:8Þ

where RRCRCA is the percentage of coarse RCA employed in the concrete mixture.
The relationship (1.8) confirms that the reduction in Ecm expected for RAC with
replacement ratio lower than 30% foresees a reduction in strength lower than 10%,
as already found by the aforementioned Author (Corinaldesi 2010). Moreover,
similar conclusions can be achieved by means of alternative analytical expressions
proposed for the stress-strain curve of RAC under uniaxial compression (Wardeh
et al. 2015). A complete constitutive formulation capable of simulating the response
and predicting the failure mode of members made of RAC and subjected to tri-axial
stress states (Folino and Xargay 2014) would also lead to the same conclusions.
Further mechanical properties of interest for reinforced concrete structures, such as
1 State of Knowledge on Green Concrete with Recycled Aggregates … 15

bond between RAC and deformed steel rebars, were recently investigated for
understanding the influence of RCA and the bond-slip relationships generally
adopted for ordinary concrete were recalibrated for RAC characterised by either
normal (Prince and Singh 2014a) or high strength (Prince and Singh 2014b).
Moreover, the widely adopted non-destructive testing techniques, such as those
based on ultrasonic velocity (Rao et al. 2011) and acoustic emissions (Kencanawati
et al. 2013), were recalibrated for RAC.
The mechanical response under sustained loads and, particularly, the develop-
ment of creep and shrinkage phenomena are other features of relevance in RAC for
structural purposes. Experimental observations confirm that the evolution of these
phenomena in RAC is fairly similar to the one of ordinary structural concrete,
although a certain influence of the aggregate replacement ratio can be detected. An
experimental study available in the literature shows creep increase of around 50%
and shrinkage increase of about 70% in RAC specimens with 100% replacement of
coarse aggregate tested in uniaxial compression (Domingo et al. 2010). However,
the same study highlighted much lower differences for RAC specimens with lower
replacement ratio, the one made from 20% of recycled aggregate being almost
unaffected by the effect of RCA on creep and shrinkage. For this reason, it is
generally accepted that no formal changes are needed to the general models
available in Codes and Guidelines for ordinary concrete. Fathifazl and Razaqpur
(2013) suggested only to introduce a new coefficient for increasing the basic pre-
dictions based on those models. As regards the response under tensile stresses, a
uniaxial restrained shrinkage cracking test was executed to investigate the tensile
creep properties caused by the restraint of drying shrinkage of RAC: the results
highlighted that the tensile creep of RAC caused by the restraint of shrinkage was
about 20–30% higher than that of the corresponding RC (Seo and Lee 2015).
Finally, durability is the other main aspect of concern in RAC. Correlation
between compressive strength and durability-related properties, mainly controlling
carbonation and chloride ingress, are available in the literature. Specifically, the
following relationship between chloride migration coefficient Dnssm and compres-
sive strength fc was recently proposed (Silva et al., 2015a, b):

Dnssm ¼ 47:618 ! e&0:024!f c : ð1:9Þ

Besides the common belief, the presence of fine recycled aggregate (FRA), that
is technically feasible for low replacement ratios (e.g. <30%), does not lead to any
reduction in durability (Evangelista and de Brito 2010). Conversely, the possible
“contamination” of RAC due to the presence of chlorides or sulphates plays a
significant role on mechanical and durability-related properties of RAC (Debieb
et al. 2010).
All the aspects mentioned in this section will be further developed in the fol-
lowing sections of this chapter. For other issues, not addressed in this book, the
Reader may refer to Behera et al. (2014).
16 E. Martinelli et al.

1.4 Codes and Guidelines

Several codes, regulations and guidelines dealing with the use of recycled aggre-
gates in concrete are available worldwide. Particularly, the most comprehensive
documents concern countries, such as Hong Kong and North European countries,
where waste disposal represents a crucial problem due to morphological and
environmental conditions, or industrial countries, which have promoted a new
urban development process (e.g. Germany during the last few decades). This sec-
tion proposes an overview of the most significant documents that are currently in
force in various regions of the world.

1.4.1 Europe

RILEM, among the first concerned Institutions, proposed some specifications for
concrete with recycled coarse aggregates, while the use of recycled sand was not
recommended (RILEM 1994). Therefore, these recommendations deal with
Recycled Coarse Aggregate Concrete (RCAC) and suggest a classification of this
material according to material composition: RCAC Type I is mainly constituted by
crushed brick, RCAC Type II is made of crushed concrete, while RCAC Type III is
recycled material from concrete and brick mixture containing up to 50% brick.
Focusing the attention on RCAC Type II, it is allowed a total replacement of natural
coarse aggregates with recycled ones in concretes up to class C50/60.
Similarly, the structural code currently in force in Italy (NTC 2008) only allows the
use of recycled aggregates for replacing the coarse fraction of aggregates in new
concrete production. A total replacement of natural aggregates, by recycled ones made
of CDW, it is allowed only for concrete produced for non-structural applications. In
the case of structural concrete, the maximum allowed percentage of RCAs is strongly
limited for the usual compressive strength targets related to structural elements.
In Germany, recycled aggregates are classified into four types, depending on the
material composition (DIN 4226-100 2002). Particularly, Type 1 and Type 2 derive
both from demolition of concrete structures, but the minimum content of concrete
plus natural aggregate should be at least 90 and 70% by mass, respectively; the
other part may consist of clinker and calcium silicate bricks. Type 3 shall contain
more than 80% of dense bricks, and it generally is obtained from pure brick
masonry demolition. Finally, Type 4 is a mixture of all mineral building materials
without strict specification of the constituents. In terms of applications and
mechanical requirements, the best reference is the “Guideline of the German
Committee for Reinforced Concrete (DAfStb 1998)”. This document specifies that
only aggregates with equivalent size bigger than 2 mm belonging to Type 1 or Type
2 can be used in producing structural concrete; moreover, it proposes correlations
between replacement percentage and mechanical performance of recycled aggregate
concrete.
1 State of Knowledge on Green Concrete with Recycled Aggregates … 17

In the United Kingdom, the BS 8500-2 (2006) provides general requirements for
coarse recycled aggregate. In accordance with the use of recycled concrete aggre-
gate in new concrete production, a maximum of 20% replacement of coarse
aggregate is allowed and the corresponding compressive strength is limited between
20 and 40 MPa. Moreover, it is specified that RAC can be used for unreinforced
members, internal elements or external elements not exposed to chlorides or subject
to de-icing salts. RAC also cannot be used in foundations or paving elements.
Finally, it is useful to note that no provisions are given in BS 8500-2 for the use of
fine recycled aggregates, but their use is not it precluded in principle.
The Spanish Code on Structural Concrete EHE-08 (2008), in annex 15
“Recommendations for using recycled concrete”, specifies that the use of coarse
recycled concrete aggregates is allowed in structural concrete for replacing up to
20% (by weight) of the total amount of coarse aggregates. However, this is only
allowed for concrete with cylindrical compressive strength up to 40 MPa and the
recycled aggregates should be characterised by a water absorption capacity lower
than 7%.
The same limitation in terms of replacement ratio is provided by the French
standard NF EN 206-1/CN (2012) for concrete classes up to C35/45 to be employed
in the exposure classes XC1, XC2, XC3, XC4 or XF1, provided that the origin of
demolished/deconstructed concrete is traceable.

1.4.2 United States, Hong Kong and Australia

The American Concrete Institute (ACI) highlights that possible sources of RAs can
be identified in concrete pavements, structures, sidewalks, curbs and gutters that
when are removed can be used in concrete production. Particularly, ACI E-701
(2007) specifies that new concrete mixtures can contain both fine and coarse
recycled aggregate. Although up to 100% of the coarse aggregates can be made of
recycled materials, the percentage of fine aggregate replacement is usually limited
to 10–20%.
The Buildings Department of Hong Kong proposed one of the most detailed
Guidelines about the use of recycled concrete aggregates (HKBD 2009). These
Technical Guidelines specify that concrete with 100% of recycled coarse aggregates
shall only be used for non-structural works. Both 100 and 20% recycled coarse
aggregates analysed in these guidelines, shall be produced by crushing old concrete.
The Cement Concrete & Aggregates Australia (CCAA), that is the main national
body in Australia (representing the interests of six billion dollar a year heavy
construction materials industry), recently published an interesting document
reporting the current knowledge about the use of recycled aggregates in new
concrete production (CCAA 2008). The Commonwealth Scientific and Industrial
Research Organisation (CSIRO 2002) and the Standards of the Concrete Institute of
Australia are the most important references for this document. Five types of
recycled aggregates are identified and classified: Recycled Concrete Aggregate
18 E. Martinelli et al.

(RCA), Recycled Concrete and Masonry (RCM), Reclaimed Aggregate (RA),

Reclaimed Asphalt Pavement (RAP) and Reclaimed Asphalt Aggregate (RAA).
As CSIRO reported, Class 1A RCA (which is a good quality RCA with no more
than 0.5% brick content) has the potential for being used in a wide range of
applications. Applications include partial replacement of ordinary material in
concrete production for non-structural components, such as kerbs and gutters.
Although CSIRO (2002) emphasizes that the current field experience with the use
of RCAs for structural applications is scarce, it contributes to clarify this matter and
defines two different grades of RAC both made by using Class 1A RCAs:
• Grade 1 RAC, characterized by a maximum 30% replacement ratio with Class
1A coarse RCAs, has a maximum specified compressive strength limit of
40 MPa;
• Grade 2 RAC, made of up to 100% Class 1A coarse RCAs recycled, has a
maximum specified compressive strength limit of 25 MPa.

1.4.3 Some Remarks About Existing Regulations

and Standards

As the aforementioned recommendations show, the use of RCAs is intended mostly

to replace the coarse fraction of ordinary aggregates. In fact, recycled fine aggregate
from concrete exhibit deleterious characteristics that might affect performance and
workability of recycled concrete, especially if the concrete mixture is not accurately
designed.
The requirements that aggregate shall meet in order to be used as RCA, seem to
be almost the same for the documents proposed from different institutions: they are
outlined in Table 1.1. Although it is required that RCAs mainly derive from
demolished concrete, a certain limited content of other “alien” materials, such as
metals, plastics, clay lumps and glass, is allowed. In this respect, only ACI 555R-01
(2001), among the current regulations, provides clear indications about a selective
process of demolition intended at safeguarding the “purity” of RCAs. Conversely,
the processing procedure implemented for producing recycled aggregates, is gen-
erally designed by operator companies, according to their specific practices.
A total replacement of coarse natural aggregates is allowed only for
non-structural concrete, due to the decrease of compressive strength that generally
occurs considering the recycled material source.
Moreover, the use of recycled coarse concrete aggregate is still limited in
structural applications, as several international standards define an upper limit
between 20 and 30% for their replacement ratio (in volume). Table 1.2 summarizes
the main requirements and limitations provided by the regulations and guidelines
considered herein.
1 State of Knowledge on Green Concrete with Recycled Aggregates … 19

Table 1.1 Recycled concrete aggregate requirements: synoptic overview

Code/guideline Material source Maximum Maximum
content of content of
fine (%) “alien”
materials
Italian Ministry of Building demolition (for non – –
Infrastructure and structural concrete) and concrete
Transportation demolition (for structural
(NTC 2008) concrete)
RILEM ( 1994) – 5 1%
DAfStB (1998) Demolished concrete structures – ' 0.2%
(Type 1)
' 0.5%
(Type 2)
British Standard Institution Crushing hard concrete 5 1%
(BS 8500-2 2006)
Building Department Crushing old concrete 4 1%
Hong Kong Government
(HKBD 2009)
American Concrete Removed pavements, structures, – 2 kg/m3
Institute (ACI E-701 2007) sidewalks, curbs, and gutter
Cement Concrete & Demolition waste of at least 95% – –
Aggregates Australia concrete
(CCAA 2008)
Note Metals, plastics, clay lumps and glass are considered as “alien” materials

Finally, selective demolition is generally needed in order to obtain materials that

may be easily turned into RAC with limited need for further screening processes
and decontamination procedures (HKBD 2004).

1.5 Insights into Concrete Sustainability

According to its original and most cited definition, “Sustainable development”

should “meet the needs of the present without compromising the ability of future
generations to meet their own needs” (WCED 1987). Therefore, no further spec-
ulation, that would be out of the scope of this work, is actually needed for
understanding that the construction industry is fully concerned by the challenge of
making its processes “sustainable”. In fact, the building construction sector and the
production of cement and concrete is responsible for a significant share of the
global emissions and raw material demand (van den Heede and De Belie 2012).
Particularly, 40% of anthropogenic GHG global emissions and 40% of raw mate-
rials are attributed to building construction sector, whereas the global annual pro-
duction of concrete is going to approach 25 gigatonnes, namely 3.8 t per person
(Gursel et al. 2014). Moreover, the production of CDW requires more and more
20 E. Martinelli et al.

Table 1.2 Main characteristics of recycled concrete: synoptic overview

Country Application Replaceable Maximum Maximum cylindrical
(guideline) aggregate replacement compressive strength
fraction percentage (%) (28 days) (MPa)
Italy (NTC Non-structural Coarse 100 8
2008) Structural Coarse 30 30
60 20
RILEM Not specified Coarse 100 50
(1994)
Germany Structural Coarse 35 25
(DAfStB 25 35
1998)
UK (BS Not specified Coarse 20 40
8500-2 2006)
Spain Structural Coarse 20 40
(EHE-08
2008)
France (NF Structural Coarse 20 35
EN 206-1/CN
2012)
Hong Kong Non-structural Coarse 100 20
(HKBD 2009) Structural Coarse 20 30
USA (ACI Not specified Coarse fine 100 Not specified
E-701 2007) 20
Australia Grade 2 Coarse 100 25
(CCAA 2008) concrete
Grade 1 Coarse 30 40
concrete

landfilling capacity and this, especially in some countries, implies significant impact
on the environment (Kien et al. 2013). As a matter of fact, cost-benefit analyses
reveal positive results when evaluating CDW recycling solutions and highlight that
the landfill charge is a key factor in determining those results and the achievement
of a breakeven post after the initial investment (Yaun et al. 2011).
Therefore, any action capable of even slightly reducing both GHG emissions and
raw material demand results in a significant global effect on the environment, due to
the abovementioned huge figures. In this light, recycling CDW for transforming
them into sustainable “second raw materials” is the solution to answer the above
requirements and achieve a higher sustainability for the construction industry.
Besides the general understanding of recycling solutions for producing “green
concrete”, quantitative assessment methodologies capable of quantifying and
1 State of Knowledge on Green Concrete with Recycled Aggregates … 21

comparing alternative solutions for sustainable concrete are needed for approaching
this problem in a rational way. Particularly, the six quantification methodologies
listed below can be generally recognised in the studies available in the literature,
some of which being also possibly combined in some specific cases (Wu et al.
2014):
• site visit (SV) method, based on direct or indirect surveys carried out at the
construction or demolition sites by duly skilled and trained personnel;
• generation rate calculation (GRC) method, intended at determining the waste
generation rate for a particular activity unit (i.e. kg/m2, and m3/m2) by means of
alternative approaches (such as per capita multiplier, financial value extrapo-
lation and area-based calculation) determined on similar situations analysed in
the past;
• lifetime analysis (LA) method, mainly implemented for demolition waste, and
based on the principle of material mass balance when turning buildings into
demolition rubbles;
• classification system accumulation (CSA) method, based on GRC method with a
further classification system providing a tool for determining the contribution of
a given material;
• variables modelling (VM) method, consisting in a simulation of the produced
amount of CDW taking into account the variables controlling the production of
waste, such as economic indicators, construction areas, on-site working
conditions;
• other particular methods, such as the assumption of a given percentage of waste
generation or other global parameters (the generation of CDW estimated on the
amount of annual cement production).
As a proof of concept, the economic viability of building a construction and
demolition waste recycling plant in Portugal was analysed by Coelho and de Brito
(2013a). According to the analysis proposed in that study, the break-even point is
around 2 years and, hence, in spite of the significant initial investment, the con-
struction of this kind of plant can be a profitable investment. Moreover, since the
factors affecting this result can be significantly variable even in the short period, a
sensitivity analysis was carried out for investigating the influence of various rele-
vant parameters, such as CDW generation rate and landfilling charges and rejected
materials. In the worst scenario the return on investment was eight years and, hence,
still fairly acceptable (Coelho and de Brito 2013b). A more general “environmental
analysis” and the corresponding sensitivity investigation were carried out by the
same authors by using primary energy consumption and CO2,eq emission impact
factors as environmental impact performance indicators (Coelho and de Brito
2013c, d).