Journal of Cleaner Production 99 (2015) 59e74

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Mechanical performance of concrete made with aggregates

from construction and demolition waste recycling plants
Miguel Bravo a, Jorge de Brito b, *, Jorge Pontes a, Lus Evangelista c
a

ICIST, IST, Universidade de Lisboa, Portugal, Lisbon, Portugal

ICIST, Department of Civil Engineering, Architecture and Georresources, Instituto Superior T
ecnico (IST), Universidade de Lisboa, Lisbon, Portugal
c
ICIST, Lisbon's Polytechnic Engineering Institute (ISEL-IPL), Lisbon, Portugal
b

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 11 June 2014
Received in revised form
24 February 2015
Accepted 3 March 2015
Available online 11 March 2015

This research aims at analysing the mechanical performance of concrete with recycled aggregates (RA)
from construction and demolition waste (CDW) from various locations in Portugal.
First the characteristics of the various aggregates (natural and recycled) used in the production of
concrete were thoroughly analysed. The composition of the RA was determined and several physical and
chemical tests of the aggregates were performed.
In order to evaluate the mechanical performance of concrete, compressive strength (in cubes and
cylinders), splitting tensile strength, modulus of elasticity and abrasion resistance tests were performed.
Concrete mixes with RA from CDW from several recycling plants were evaluated, in order to understand the inuence that the RA's collection point, and consequently their composition, has on the
characteristics of the mixes produced.
The analysis of the mechanical performance allowed concluding that the use of RA worsens most of
the properties tested, especially when ne RA are used. On the other hand, there was an increase in
abrasion resistance when coarse RA were used. In global terms, the use of this type of aggregates, in
limited contents, is viable from a mechanical viewpoint.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Recycled aggregates
Construction and demolition waste
Recycling plants
Concrete
Mechanical performance

1. Introduction
Every year 3000 Mt of waste are produced in the European
Union, of which 90 million are considered hazardous. The construction industry generates in the EU around 900 million tons per
year of wastes. Therefore, this waste ow represents around 25%e
30% of all wastes produced (Eurostat, 2010).
CDW have a very heterogeneous composition. The most
important fraction corresponds to inert material, i.e. between 40%
and 85% of the overall waste volume discounting excavation soils
(Eurostat, 2010). The main sources of inert material are concrete
and ceramic materials. Pereira et al. (2004) and Bergsdal et al.
(2007) determined the amount of concrete, masonry and

List of abbreviations: CDW, construction and demolition waste; CRA, coarse

recycled aggregates; FRA, ne recycled aggregates; ITZ, interfacial transition zone;
NA, natural aggregates; RA, recycled aggregates; w/c, water cement ratio.
* Corresponding author. Tel.: 351 218419709.
E-mail addresses: miguelnbravo@gmail.com (M. Bravo), jb@civil.ist.utl.pt (J. de
Brito),
jorgempontes@gmail.com
(J.
Pontes),
evangelista@dec.isel.ipl.pt
(L. Evangelista).
http://dx.doi.org/10.1016/j.jclepro.2015.03.012
0959-6526/ 2015 Elsevier Ltd. All rights reserved.

mortar in inert material, which accounts for 58% and 67% in

Portugal and Norway, respectively. Costa and Ursella (2003) and
Reixach et al. (2000) obtained values of approximately 85% in Italy
and Spain, respectively.
The amount of waste from the construction industry used as
lling material or illegally dumped in vacant lots has been
increasing over time. This has led to an increasing lack of landll
areas, useful lands becoming dumping yards and highly increased
dumping costs at landll sites. So handling wastes has become one
of the most important environmental issues in developed countries
(Behera et al., 2014) In 2010, around 75% of all CDW produced in the
EU were dumped (Ortiz et al., 2010). However, reuse ratios over 80%
have already been reached by countries such as the Netherlands,
Denmark and Germany (Eurostat, 2010). The Community Directive
2008/98/EC establishes that the EU state members must take the
necessary measures to reach until 2020 a minimum of reuse ratio
70% (in weight) of the CDW produced.
The use of natural aggregates in concrete leads to high environmental impacts, both because of the amount of emissions of CO2
produced during their extraction and of the depletion of natural
resources that this activity implies. Therefore, the incorporation of

60

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

recycled aggregates from construction and demolition waste

signicantly improves concrete in terms of its ecological footprint,
saving ecosystems for generations to come. Considering that a
substantial amount of the environmental impacts of using RA from
CDW comes from their production and that such impacts are
intrinsic to the construction and demolition activities, it is possible
to conclude that the use of RA in concrete production has a positive
environmental effect.
Adequate waste management and recycling lead to a decrease of
the consumption of natural resources and of the volume of waste
sent to dumping grounds, two measures with effects benecial to
the environment (Cochran et al., 2007). To that purpose, the use of
waste in the production of concrete has been analysed. There are
various researches that evaluate the use of a wide variety of wastes
in concrete, such as used tyres (Bravo and de Brito, 2012). Another
example is the use of wastes from the quarrying of marble in the
production of concrete. The quarrying of marble, a well-known
ornamental stone, has a substantial positive impact on Portugal's
economy, but it also generates large environmental impacts. The
amount of waste produced during quarrying can be as much as 80%
 et al., 2014). However, CDW are
of all stone/soil extracted (Andre
the waste with the greatest potential of reuse in concrete. The
European Commission identied CDW as a priority waste ow for
treatment and recycling, due to the high amounts produced and its
great potential of reuse as raw material. Therefore it is not surprising that in the latest years there were various researches with
the objective of evaluating the use of CDW in concrete. The properties of concrete made with RA from concrete, ceramic materials,
glass, plastics, among other materials, were analysed in order to
fully understand their potential and limitations. However, the researches that evaluate the use in concrete of CDW from recycling
plants are still scarce. Most of the previous studies analyse the use
of a single RA type from CDW, with no comparisons between CDW
types with different compositions. On the other hand, there are but
a few studies that exhaustively analyse the RA used in concrete,
through the characterization of their composition and physical and
chemical tests. Furthermore, as described below, studies in which
ne RA from CDW were used are even scarcer.
In this research program, the rst step was to collect CDW from
ve recycling plants in different locations in Portugal. Then the
characteristics of all the aggregates (natural and recycled) used in
the production of concrete were examined. To that purpose, the
composition of the RA was determined and various physical and
chemical tests of the aggregates were performed. In order to
evaluate the mechanical performance of concrete, compressive
strength (in cubes and cylinders), splitting tensile strength,
modulus of elasticity and abrasion resistance tests were performed.
The main objective of this work is the analysis of concrete
mixes with RA from various CDW recycling plants, with compositions representative of this type of waste in industrialized
countries. The aim is to analyse the inuence of the RA's collection
location, and consequently of their composition, on the characteristics of the mixes produced. Five types of coarse CDW (CRA)
and three types of ne CDW (FRA) were collected, with the
objective of also analysing the inuence of the size of the CDW RA
on their use in concrete.
2. Literature review
There are various researches aimed at analysing the use of
recycled aggregates (RA) in concrete. These studies evaluate the
properties of concrete with aggregates recycled from concrete,
ceramic materials, glass, plastics, among others. However, studies
evaluating the use in concrete production of CDW from recycling
plants are still scarce.

Kou et al. (2004) produced concrete with coarse RA from CDW,

at 0%, 20%, 50% and 100% of the overall mass of the coarse NA. They
analysed the concrete compressive strength at 7, 28 and 90 days
and, for full replacement, obtained losses of 33%, 37% and 31%.
Oliveira et al. (2004) studied the compressive strength of concrete with replacement of coarse NA with coarse RA at 10%, 20%,
30%, 40% and 100%. The RA came from CDW, mostly from crushed
concrete. In order to produce a mix with 100% of coarse RA with the
same strength as the reference concrete, the authors had to increase the cement content from 378 kg/m3 to 475 kg/m3.
Poon et al. (2007) studied the inuence of the replacement of
coarse NA with coarse RA on the mechanical strength. They used RA
from CDW from a recycling plant and produced concrete mixes
with 10%, 20%, 50%, 80% and 100% replacement ratios, keeping the
w/c ratio constant. They found that full replacement of the coarse
NA led to compressive strength losses of 24%, 16%, 19% and 10% at 3,
7, 28 and 90 days.
Medina et al. (2014) analysed the viability of producing 30 MPa
concrete using coarse RA from CDW with a high content of asphalt
and oating matter. They found that the use of 50% of these RA
caused a 28-day compressive strength loss up to 18%.
Kou et al. (2004) evaluated the tensile strength at 1, 4, 7, 28 and
90 days of concrete with coarse RA from CDW. For full replacement
of the coarse NA, they obtained losses of 34%, 26%, 11%, 17% and 8%.
They concluded that the negative inuence of the use of these RA
partially fades over time.
Oliveira et al. (2004) evaluated the inuence on concrete's
modulus of elasticity of replacing coarse NA with coarse RA at 10%,
20%, 30%, 40% and 100%. These RA came from CDW and were mostly
made of crushed concrete. The authors found that full replacement
of the coarse NA caused an 18% decrease of the modulus of
elasticity.
Kou et al. (2004) also evaluated the 28- and 90-day modulus of
elasticity of concrete with coarse RA from CDW at 0%, 20%, 50% and
100% of the overall mass of the coarse aggregates. For full
replacement of the coarse NA they obtained losses of 40% and 28%
at 28 and 91 days.
The Waste & Resources Action Programme (WRAP) developed
in the UK to facilitate the use of RA from CDW has concluded that
coarse RA from UK washing plants are suitable for use in concrete
and ne RA are also suitable for use in concrete, with the exception
of RA with sulphate content above limits of 1.0% (Dhir et al., 2008).
Regarding RA concrete properties, the same programme
concluded that Eurocode 2's (EC2) relations for concrete properties
can be used to determine the RA's mechanical properties in terms
of compressive strength, with the exception of secant modulus of
elasticity. Regarding that property, the use of the EC2 relation to
determine the secant modulus of elasticity for RA concrete resulted
in poorer deformation predictions than those made for standard
concretes. The use of RA resulted in an increase by 14% in the elastic
deformations (Waleed and Canisius, 2007).
In terms of the use of RA from concrete in the production of new
concrete, it is consensual that the replacement of the part of the
coarse natural aggregates (NA) with coarse RA does not signicantly damage the concrete's characteristics. However, there is no
consensus concerning the replacement of ne aggregates. The main
factor singled out by some researchers against the use of ne RA in
concrete is their high water absorption, which may lead to concrete
with worse performance. Nevertheless, recent researches indicate
that the use of ne RA in concrete may be viable, since it does not
lead to a signicant loss of its properties, both in mechanical
(Evangelista and de Brito, 2010) and durability terms (Evangelista
and de Brito, 2014).
Concrete's compressive strength is one of the most important
properties to evaluate the performance of a structure. Merlet and

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Pimienta (1994) evaluated the compressive strength of various

concrete mixes with RA from concrete. They used ne and/or
coarse RA at 0%, 20%, 50% and 100% of the global aggregates'
volume and cement contents of 250 kg/m3, 300 kg/m3 and 350 kg/
m3. All the mixes analysed showed losses in compressive strength,
between 18% and 39%. According to Sanchez and Alaejos (2004),
these losses are due to the following factors: less mechanical
strength of the RA; greater water absorption of the RA; increase of
the fragile areas within concrete, i.e. the interfacial transition zone
(ITZ) between the old cement paste and the original NA and also
the one between the RA and the new cement paste. Poon et al.
(2004) evaluated the microstructure of concrete produced with
RA from current concrete and from high-performance concrete
using scanning electron microscopy (SEM) and compared it with
the microstructure of NA concrete. They found that the ITZ between the RA from current concrete and the cement paste had a
high porosity. The authors refer that the high porosity and water
absorption of these aggregates, together with their low initial
water content during the constituents mixing process, may have
produced signicant water absorption at the initial stage of the
mixing. Consequently, this process may have caused the reported
ITZ0 high porosity. On the other hand, the ITZ between the RA
from high-performance concrete and the cement paste is much
denser. Xiao et al. (2013) also analysed the microstructure of
concrete with RA from concrete and compared the old and new
ITZ between aggregates and the cement paste. They concluded
that the average modulus of elasticity of the old ITZ is around
70%e80% that of the old cement paste. On the other hand, the
average modulus of elasticity of the new ITZ is around 70%e80%
that of the new cement paste.
Barra (1996) found that concrete with coarse RA from concrete
need greater cement content to reach the compressive strength of a
conventional concrete. For mixes with strengths of 45 MPa and
57.5 MPa, increases of 7.2% and 17.3% of the cement content were
necessary.
Etxeberria et al. (2007) obtained a loss of 20%e25% of the 28-day
compressive strength of concrete for full replacement of coarse NA
with coarse RA from concrete, keeping constant the effective water/
cement (w/c) ratio and cement content. When 25% of the aggregates were replaced there were no signicant effects on the
compressive strength.
Evangelista and de Brito (2007) analysed the compressive
strength of various concrete mixes with ne RA from concrete and
found similar values to that of the reference concrete (60 MPa).
They justify these good results with the presence of non-hydrated
cement in the ne RA and the better bond between the cement
paste and the ne RA, due to their higher porosity.
Pereira et al. (2012) evaluated the inuence of superplasticizers
(SP) on the compressive strength of concrete with ne RA. Three
families of concrete were produced (without SP, with a current SP
and with a high-performance SP) at ve replacement ratios of ne
RA with ne RA from concrete (0%, 10%, 30%, 50% and 100%),
maintaining constant the slump in all the mixes. The authors obtained an increase in compressive strength due to the SP's incorporation up to 34.8% and 69.5%, for mixes with current SP and highperformance. As expected, the strength gain increased as the water
reduction power of the admixtures improved. The authors also
found that the SP had a greater inuence on the compressive
strength of conventional concrete than of concrete with ne RA
from concrete. The loss of efciency of the admixtures was justied
by the increase of the specic surface of the aggregates, due to the
incorporation of the RA. Barbudo et al. (2013), in a research parallel
to the previous one, analysed the inuence of the use of SP in
concrete with coarse RA. They reached similar conclusions to those
described in the Pereira et al. (2012) study.

61

The results from researches concerning other mechanical

properties, such as tensile strength and modulus of elasticity, are
similar to the ones on compressive strength. However, the use of RA
from concrete seems to allow an improvement of the new concrete's abrasion resistance.
Matias et al. (2013) performed wear by abrasion tests and
detected a decrease of the worn thickness when coarse RA were
used. They justied this result with a better bond between these RA
and the cement paste.
Evangelista and de Brito (2007), even though with results not
completely conclusive, also found that use of ne RA from concrete
tends to improve the concrete performance against wear actions.
Finally Pereira et al. (2012) evaluated the inuence of the use of
SP on the wear by abrasion resistance of concrete with various
contents of ne RA from concrete (0%, 10%, 30%, 50% and 100%). The
use of RA caused increases of the worn thickness up to 21.7%, 395%
and 51.3%, for mixes without SP, with current SP and with highperformance SP.
The researches that analysed the use of ceramic RA in concrete
are not as consensual as those that evaluated the use of RA from
concrete.
Gomes and de Brito (2009) analysed the performance of two
concrete types: with coarse RA from concrete only (full replacement) and with coarse RA from concrete and ceramic materials (up
to 75% replacement). None of these mixes showed a signicant loss
in compressive strength. However, the mixes with ceramic RA
displayed a slightly descending trend of strength as the replacement ratio increased.
Medina et al. (2013) evaluated the compressive strength of
concrete with ceramic RA and found that it increased by 11% in
mixes with 25% of RA. Medina et al. (2012) found that the ITZ between the ceramic sanitary ware RA and the cement paste is more
compact and stable than the ITZ between the NA and the paste.
Other studies also hint at an increase of compressive strength when
30% of NA are replaced by ceramic RA (http://www.sciencedirect.com/science/article/pii/S0959652612004660
Pacheco-Torgal
and Jalali, 2010).
Alves et al. (2014) analysed the incorporation in concrete of
coarse ceramic RA from bricks and sanitary ware. They found that
the decrease of 28-day compressive strength is less than 10% for full
replacement of NA with coarse brick RA. Full use of coarse sanitary
ware RA caused a compressive strength loss of 42.5%.
Zong et al. (2014) refer that the use of brick RA in concrete
jeopardizes its mechanical performance and increases its permeability. They found that this is due to the increase in porosity of the
aggregates and their effects on the concrete's microstructure.
In terms of the other mechanical properties of concrete, results
are similar to the ones relative to compressive strength, except for
wear by abrasion resistance.
De Brito et al. (2005) found there is an approximately linear
increase of the abrasion resistance with the replacement ratio of NA
with ceramic brick RA. They justied this trend with the better
bond between the ceramic aggregates and the cement paste,
because of the greater porosity of the RA.
Alves et al. (2014) found an increase in abrasion resistance of
31.4% and 49.8% from the use of 100% of coarse ceramic RA from
bricks and sanitary ware.
There are also some researches on the use of glass RA in concrete. Castro and de Brito (2013) analysed the use of coarse and ne
glass RA in concrete and found that the incorporation of 20% of
glass RA only led to a decrease in compressive strength of 3% and
14% for coarse and ne aggregates.
Park et al. (2004) evaluated the mechanical performance of
concrete with ne glass RA. They found that the use of 30%, 50% and
70% of these RA caused a decrease of the 28-day compressive

62

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

strength of 0.6%, 9.8% and 13.6%. The results were justied by the
poorer adherence between the glass RA's surface and the cement
paste.

Table 1
Composition of the reference concrete (RC) (l/l).
Cement
Fine aggregates

3. Experimental program
In this chapter the materials used in the production of concrete
are presented. The different mixes' composition and the different
processes of production are also identied and explained. Finally
the several tests performed with the objective of characterizing the
mechanical performance of concrete with RA from CDW are
presented.
3.1. Materials
In this research the aggregates used were NA and RA from ve
CDW Portuguese plants (Valnor, Vimajas, Ambilei, Europontal and
Retria). In three of them (Vimajas, Ambilei and Europontal) coarse
and ne RA were analysed and in the remaining ones only coarse
RA. These recycling plants were selected in order to represent
various geographical areas, leading to a wide range of construction
materials and processes, as well as geological backgrounds and a
variety of recycling procedures. The material used came from the
output of the plants. Vimajas yielded two products that contain
material smaller than 4 mm, which were both sampled. The
Ambilei plant processes separately CDW made mostly of concrete
and CDW that is a mixture of concrete and ceramics. As for the NA,
limestone gravel and alluvial rolled sand were used. In the production of concrete cement CEM I 42.5 R and tap water were used.
3.2. Mixes' composition
33 concrete mixes were produced: a reference concrete (RC),
mixes with replacement ratios of 10%, 25%, 50% and 100% of the
overall volume of coarse NA (with coarse RA from ve recycling
plants) and mixes with replacement ratios of 10%, 25%, 50% and
100% of the overall volume of ne NA (with ne RA from three
plants).
The aggregates were considered ne when their particles
passed through a 4 mm sieve and coarse when they were retained
in that sieve. The maximum particle size used was 22.4 mm.
The NA were replaced by the RA in volume and by size fraction,
in order to keep constant in all the mixes with RA the aggregates
size distribution of the RC. No admixtures or additions were used in
this research.
Every mix was produced with a 125 15 mm slump, for a better
comparison between them. For that purpose in a preliminary stage
the water content was adjusted, when necessary, to each mix in
order to comply with that requirement.
The RC composition was determined using the Faury's method
with a C30/37 target strength class. The materials content, in volume, are provided in Table 1. All other mixes were designed based
on this composition with just slight changes in solid volume in
order to comply with the different w/c ratios required to keep of all
mixes within target slump.
3.3. Tests
EN 12620:2008 Aggregates for concrete species the properties required from natural aggregates, mechanically-processed aggregates, recycled aggregates and mixtures of aggregates for use in
concrete. It covers the aggregates with dry density higher than
2000 kg/m3, to be used in every concrete type, including concrete
in conformity with NP EN 206-1 (2013).

Coarse aggregates

0.115
0.000
0.016
0.044
0.050
0.057
0.066
0.076
0.041
0.046
0.047
0.121
0.122
0.182
0.017
1.000

0e0.063
0.063e0.125
0.125e0.25
0.25e0.5
0.5e1
1e2
2e4
4e5.6
5.6e8
8e11.2
11.2e16
16e22.4

Water
Voids
Total

The characteristics of the aggregates were determined according to the requirements of the standards and specications listed in
Table 2. Besides these tests, an analysis of the composition of the
various RC was performed.
With the objective of characterizing the mechanical performance of concrete with RA from CDW, several tests were performed. In the fresh state, the following properties were measured:
slump using the Abrams cone based on EN 12350-2 (2002) and
density according to EN 12350-6 (2002). The characterization of the
hardened state included the following properties: compressive
strength in cubic and cylindrical specimens (according to EN
12390-3, 2003), splitting tensile strength (according to EN 123906, 2003), modulus of elasticity (according to LNEC E-397, 1993)
and abrasion resistance (according to DIN 52108, 2010).
4. Results and discussion
4.1. Aggregates' properties
4.1.1. Composition
The NA used in the production of concrete were crushed limestone gravel and rolled river sand.
In order to better know the various RA used in the production of
concrete, their composition was determined by visual observation
(Table 3). The RA all have a high content of concrete, mortar and
natural sand (between 68% and 86%). The content of masonry e
clay materials varies between 1% and 29%. The RA from Ambilei
and Vimajas stand out in terms of glass and bituminous materials
content, respectively.

Table 2
Tests performed to determine the aggregates' properties.
Properties
Physical tests
Particles density and water absorption
Bulk density and voids volume
Shape index
Fragmentation resistance (Los Angeles)
Chemical tests
Water soluble chlorides content
Water soluble sulphates content
Acid soluble sulphates content
Sulphur global content
Light contaminants content
Humus content
Water solubility

Standard
NP EN 1097-6 (2003) and
Rodrigues et al. (2013a) patent
NP EN 1097-3 (2003)
NP EN 933-4 (2002)
NP EN 1097-2 (2002)
EN
EN
EN
EN
EN
EN
EN

1744-1
1744-1
1744-1
1744-1
1744-1
1744-1
1744-1

(2009)
(2009)
(2009)
(2009)
(2009)
(2009)
(2009)

Section
Section
Section
Section
Section
Section
Section

7
10
12
11
14.2
15.1
16

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

63

Table 3
Composition of the recycled aggregates (% in mass).
Composition (in %)

CRA Valnor

CRA Retria

CRA Ambilei

CRA Vimajas

CRA Europontal

FRA Ambilei

FRA Vimajas

FRA Europontal

Concrete, mortar and natural sand

Masonry e clay materials
Glass
Bituminous materials
Others
Total

70.8
28.6
0.5
0.0
0.1
100.0

69.1
28.6
2.1
0.0
0.2
100.0

85.6
4.2
10.2
0.0
0.0
100.0

79.8
11.1
1.0
6.2
2.0
100.0

79.9
17.1
0.2
2.8
0.0
100.0

83.7
0.9
15.4
0.0
0.0
100.0

75.2
11.6
1.0
10.5
1.7
100.0

68.8
26.5
3.4
1.0
0.3
100.0

4.1.2. Physical properties

Tables 4 and 5 show the results of the tests of characterization of
the physical properties of all aggregates used. The RA, by comparison with the NA, have lower particles density and higher water
absorption. This results from the nature and porosity of the RA. It
demonstrates that there is a propensity of the RA mixes to require
more mixing water than the NA mixes. The greater water absorption of the FRA relative to the CRA stands out. In order to determine
the water absorption of the FRA the method dened in the patent of
Rodrigues et al. (2013a) was used. This method solves the problems
that occur when using NP EN 1097-6 in FRA, namely those associated to the formation of air bubbles in the sample during the
immersion of the pycnometer and to the particles agglomeration
during the saturation stage, because of the cohesiveness of the FRA.
When comparing aggregates of similar size, it is found that the
RA have a lower bulk density than the NA. This is due to the more
porous nature and sharper shape of the RA. The low bulk density of
the CRA Valnor stands out.
With the exception of the Ambilei aggregates, every RA has a
higher shape index than the NA, which may cause a lower slump of
the mixes with RA relative to the RC.
It is also found that the RA have a lower fragmentation resistance than the NA, possibly due to their composition and crushing
process, which creates micro-cracks in the particles, making them
more unstable and fragile.
4.1.3. Chemical properties
Table 6 presents the results of the chemical tests performed on
the RA showing that the light contaminants content in the various
RA is their main problem, in terms of the chemical tests results. The
threshold value for theses contaminants is 0.5% in mass, according
to EN 12620 (2008). This value drops to 0.25% when there is special
concern with the surface nish. Rodrigues et al. (2013b) chemically
analysed eight types of RA from CDW recycling plants and also
found very high contents (between 0.3% and 17.1%).
These chemical tests also indicate that the RA Vimajas have
higher water soluble chlorides content than the threshold. The
chlorides in aggregates that most affect concrete may be found in
alkaline salts, namely sodium chloride and potassium chloride,
decisively contributing to the overall chlorides content inside
concrete. This limitation is needed to minimize the risk of corrosion
of the reinforcement within concrete. It is also important to
consider the chlorides content of the remaining constituent materials of concrete when designing its composition. According to

APEB (2007), the chlorides content in concrete is expressed in

percentage of chloride ions in cement mass, which means that the
chlorides threshold content may not be exceeded if the cement
content used is high.
All the other results of the chemical tests of the RA fall within
the thresholds imposed, except for the sulphates content in the FRA
Vimajas that equal the limit. Internal sulphates present a risk of
chemical attack of concrete (Mehta and Monteiro, 2006).
4.2. Fresh-state concrete properties
4.2.1. Workability
All mixes were produced with a 125 15 mm slump, in order to
be more fairly compared. This involved a preliminary stage in which
the mixing water of each mix was adjusted, whenever necessary, in
order to comply with this requirement.
As Table 7 shows, it was necessary to increase the effective w/c
ratio as the RA ratio increased. However, this increase was not
identical in all the families of mixes with RA. It was concluded that
the shape and composition of the various RA inuenced the mixes'
workability. As referred by De Brito and Robles (2010), the rougher
shape of the RA, relative to the NA, may contribute to this change of
workability.
The signicant increase of the w/c ratio of the mixes with FRA
from Vimajas and Europontal stands out. This change is due to the
high clay content of these RA (Rodrigues et al., 2013b). These ne
particles absorb a great amount of water, forcing an increase of the
w/c ratio in order to maintain the slump of the mixes with these RA.
Out of the 33 mixes produced, only one (that with 25% of CRA
Valnor) had a cone Abrams slump slightly outside the target range
(110 mme140 mm). Since the incompliance was only by 4 mm, this
was not considered relevant for the analysis of the hardened state
concrete tests.

Table 5
Results of the physical tests of the ne aggregates.
Physical tests

Fine
sand

Coarse
sand

FRA
Ambilei

FRA
Vimajas

FRA
Europontal

Oven-dry particles
density (kg/m3)
Water absorption (%)
Bulk density (kg/m3)

2583

2581

2112

2070

2063

0.3
1530

0.7
1540

12.9
1435

10.1
1332

10.4
1358

Table 4
Results of the physical tests of the coarse aggregates.
Physical tests

Gravel 2

Gravel 1

Granule

CRA Valnor

CRA Retria

CRA Ambilei

CRA Vimajas

CRA Europontal

Oven-dry particles density (kg/m3)

Water absorption (%)
Bulk density (kg/m3)
Shape index (%)
Los Angeles wear (%)

2599
1.5
1360
15
26

2609
1.3
1350
17
28

2522
2.7
1348
18
e

2091
8.6
1095
24
52

2137
8.4
1236
24
46

1928
9.9
1288
14
44

2243
6.4
1261
25
39

2262
5.5
1285
21
43

64

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Table 6
Results of the chemical tests of the recycled aggregates.
Chemical tests

CRA Valnor

CRA Retria

CRA Ambilei

CRA Vimajas

CRA Europontal

FRA Ambilei

FRA Vimajas

FRA Europontal

Threshold

Water soluble chlorides content

Water soluble sulphates content
Acid soluble sulphates content
Sulphur global content
Light contaminants content
Humus content
Water solubility

<0.010
0.04
0.2
0.1
1.5
Neg.
1.0

<0.010
0.04
0.3
0.1
<0.1
Neg.
1.2

<0.010
0.04
0.1
<0.1
0.9
Neg.
0.2

0.016
0.13
0.4
0.2
1.5
Neg.
1.4

<0.010
0.06
0.3
<0.1
0.3
Neg.
0.8

<0.010
0.11
0.2
0.1
1.8
Neg.
0.5

0.016
0.18
0.8
0.3
1.7
Neg.
2.1

0.010
0.04
0.1
<0.1
1.0
Neg.
0.8

0.010a
0.20a
0.8a
1.0a
0.5a
Neg.
10.0b

a
b

According to EN 12620 (2008).

According to EN 1744.

Table 7
Slump and effective w/c ratio of the concrete mixes.
Aggregates replacement ratio (%)
0

CRA Valnor
CRA Vimajas
CRA Ambilei
CRA Europontal
CRA Retria
FRA Vimajas
FRA Ambilei
FRA Europontal

10

25

50

100

Slump (mm)

w/c ratio

Slump (mm)

w/c ratio

Slump (mm)

w/c ratio

Slump (mm)

w/c ratio

Slump (mm)

w/c ratio

114

0.51

126
121
125
125
118
119
140
120

0.52
0.52
0.52
0.51
0.52
0.53
0.52
0.53

106
120
110
125
130
121
140
118

0.53
0.53
0.53
0.51
0.52
0.55
0.52
0.54

120
130
120
128
125
124
140
125

0.53
0.53
0.53
0.51
0.53
0.58
0.53
0.57

120
119
126
115
121
121
135
140

0.53
0.53
0.55
0.51
0.53
0.64
0.55
0.63

4.2.2. Density
Figs. 1 and 2 show that the fresh-state density of concrete decreases as the RA incorporation ratio increases. This is justied by
the lower particles density of the RA, relative to the NA. The gures
also show that this reduction changes according to the specic RA
used, because it depends on the constitution of the RA from each
plant.
Finally, it is clear that there is an approximately linear decrease
of the density with aggregates replacement ratio (determination
coefcients between 0.90 and 0.99). Gomes and de Brito (2009)
reached the same conclusion when they investigated concrete
with coarse RA.

4.3. Concrete mechanical properties

4.3.1. Compressive strength in cubes
Table 8 presents the mixes' average compressive strength at
several ages (fcm,7, fcm,28 and fcm,56), as well as the strength variations (D) relative to the RC. It shows that the 28-day compressive
strength of the RC is 53.9 1.8 MPa. This strength level corresponds
to the strength class C35/45, higher than the previously set target.
To better interpret the 28-day compressive strength results,
they were represented graphically in Fig. 3. Independently of the
aggregates' size fraction replaced, the compressive strength is
affected by the incorporation of RA. This is due to the RA's

Fig. 1. Fresh-state density of the mixes with CRA from CDW.

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

65

Fig. 2. Fresh-state density of the mixes with FRA from CDW.

composition and the increase of the effective w/c ratio as the RA

incorporation increases. However, the results demonstrate that the
scale of the compressive strength's decrease varies with several
factors.
One of the factors that most inuenced the results was the RA's
source. Because the RA were collected in different plants, spread
out in the Portuguese territory, their composition varies according
to the source. This was demonstrated in the analysis presented in
Table 3. As expected the results varied considerably. In the mixes
with 100% of CRA Ambilei, there was a reduction of 5.2% of the 28day compressive strength, while in the mixes with RA from Valnor

Table 8
Compressive strength of the RA concrete mixes, in cubes, at 7, 28 and 56 days.
fcm,7 (MPa) D (%)
BR
C10C-Valnor
C25C-Valnor
C50C-Valnor
C100C-Valnor
C10C-Vimajas
C25C-Vimajas
C50C-Vimajas
C100C-Vimajas
C10C-Ambilei
C25C-Ambilei
C50C-Ambilei
C100C-Ambilei
C10C-Europontal
C25C-Europontal
C50C-Europontal
C100C-Europontal
C10C-Retria
C25C-Retria
C50C-Retria
C100C-Retria
C10F-Vimajas
C25F-Vimajas
C50F-Vimajas
C100F-Vimajas
C10F-Ambilei
C25F-Ambilei
C50F-Ambilei
C100F-Ambilei
C10F-Europontal
C25F-Europontal
C50F-Europontal
C100F-Europontal

44.8
46.4
42.8
41.0
29.6
42.6
44.1
38.1
35.7
43.6
39.3
39.6
40.7
41.3
38.7
39.1
32.2
40.4
37.4
36.4
31.8
39.9
38.5
31.1
23.0
41.3
40.0
36.5
29.1
38.9
39.3
36.7
26.1

1.2
0.5
1.5
1.4
1.9
0.7
0.9
0.8
0.2
0.8
0.6
0.8
0.8
1.6
0.6
0.1
0.2
0.8
0.3
1.6
1.3
0.5
0.6
0.1
0.3
0.4
0.4
1.7
0.0
1.4
1.0
1.2
0.5

e
3.4
4.6
8.5
33.9
5.0
1.7
15.0
20.3
2.8
12.4
11.7
9.3
7.9
13.8
12.9
28.3
9.9
16.6
18.8
29.1
11.1
14.2
30.6
48.7
7.9
10.8
18.6
35.1
13.3
12.3
18.1
41.7

fcm,28 (MPa) D (%)

53.9
54.1
48.9
46.2
35.3
52.3
54.1
48.4
42.0
53.7
50.0
48.8
51.1
45.4
46.0
46.6
36.2
48.3
44.6
44.9
40.1
49.2
45.6
37.6
30.2
51.6
47.3
46.8
38.4
50.3
44.7
44.5
29.9

1.8
2.2
1.1
2.7
1.4
1.7
1.3
0.9
1.5
2.1
2.4
2.0
1.5
1.3
1.8
1.1
1.3
2.3
0.8
1.4
1.1
1.1
1.5
1.3
0.5
1.0
1.1
1.2
1.2
1.5
2.0
1.0
0.6

e
0.2
9.4
14.3
34.5
3.0
0.3
10.3
22.1
0.5
7.2
9.6
5.2
15.9
14.8
13.6
32.9
10.4
17.3
16.8
25.6
8.7
15.5
30.2
44.1
4.3
12.3
13.3
28.8
6.8
17.0
17.4
44.6

fcm,56 (MPa) D (%)

61.1
58.8
55.6
50.1
43.2
56.7
57.9
49.5
44.6
55.7
50.3
51.5
49.5
49.7
48.8
47.4
39.5
53.4
46.2
42.7
33.9
52.8
48.7
40.5
30.6
52.3
48.1
48.1
40.3
53.7
48.1
46.3
31.5

1.6
1.9
2.0
2.1
1.5
1.0
2.8
0.6
0.5
2.7
1.2
2.6
1.3
0.1
1.1
3.2
1.4
0.9
1.8
1.8
2.6
1.9
1.2
0.1
0.9
0.4
1.4
0.9
0.1
2.0
1.9
1.2
0.8

e
3.8
9.0
18.1
29.4
7.2
5.4
19.1
27.0
8.9
17.7
15.7
19.0
18.7
20.1
22.5
35.5
12.6
24.4
30.2
44.5
13.7
20.4
33.7
50.0
14.5
21.3
21.3
34.0
12.2
21.3
24.2
48.5

this reduction reached 34.5%. In the mixes with CRA, those with RA
from Valnor, Europontal and Retria showed the highest compressive strength loss. These results are justied by the high ceramics
content of the RA from these three recycling plants (between 17.1%
and 28.6%). Hansen (1992) also found that the RA with greater
ceramics content cause a greater mechanical strength loss than the
RA from concrete. In the mixes with FRA, there was a much greater
28-day compressive strength loss in those with RA from Vimajas
and Europontal (44.1% and 44.6%), relative to those with RA from
Ambilei (28.8%). This lower compressive strength of the mixes with
FRA from Vimajas and Europontal is essentially due to the great
increase of the w/c ratio of these mixes, caused by their clay content
(Rodrigues et al., 2013b). These ne particles coat the RA grains and
absorb the mixing water, besides hindering an adequate bond between the RA and the cement paste, weakening the cement's inner
structure. Through scanning electron microscopy (SEM), it was
proven that these phenomena caused an increase of the porosity of
these mixes, leading to a decrease in compressive strength. Figs. 4
and 5show that the FRA from Vimajas have greater microporosity and macro-porosity than the FRA from Ambilei, as seen
both in the mixes with 50% of RA (Fig. 4a and b) and those with
100% of RA (Fig. 5a and b).
The results of the mixes with FRA from Europontal may also be
explained by their high content of ceramic materials (26.5%).
The size of the RA was another factor inuencing the results. The
compressive strength drop was signicantly greater in the mixes
with FRA than in those with CRA. For example the use of 100% of RA
from Vimajas led to a decrease in 28-day compressive strength of
22.1% and 44.1%, for 100% CRA and FRA. In the mixes with RA from
Ambilei, the compressive strength dropped by 5.2% and 28.8% for
full incorporation of CRA and FRA. In the mixes with RA from
Europontal the corresponding decreases were 32.9% and 44.6%
respectively. These differences are partly justied by the clay content of some of the FRA (from Vimajas and Europontal). As stated
before, the presence of clay makes it necessary to increase the w/c
ratio in the mixes with these FRA, in order to have a slump constant
in all the mixes. This increase naturally leads to a reduction of the
compressive strength.
These results are similar to those of Zaharieva et al. (2003). They
evaluated the performance of concrete with RA from CDW (ne and
coarse), collected randomly from a recycling plant, and found that
for a constant concrete slump the full replacement of the NA caused
a 28-day compressive strength drop from 54.8 MPa to 39.4 MPa
(around 30%).

66

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Fig. 3. 28-day compressive strength for RA concretes versus aggregates' replacement ratio.

Fig. 4. Comparison of the micro-porosity of the concrete mixes with 50% of ne RA: a) detail of the ITZ of the mix with FRA from Vimajas; b) detail of the ITZ of the mix with FRA
from Ambilei.

Fig. 5. Comparison of the micro-porosity of the concrete mixes with 100% of ne RA: a) detail of the ITZ of the mix with FRA from Vimajas; b) detail of the ITZ of the mix with FRA
from Ambilei.

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

67

Fig. 6. Variation of compressive strength with time for the mixes with coarse RA.

Fig. 7. Variation of compressive strength with time for the mixes with ne RA.

Figs. 6 and 7 demonstrate that the evolution of compressive

strength of the mixes with RA is identical to that of conventional
concrete. Therefore, the older concrete is the lower the increase in
concrete becomes. This general trend is more intense in mixes with
RA than in the RC. In other words the compressive strength of the

RC keeps on increasing after 28 days at a steeper rate than in the

mixes with RA.
Fig. 8 shows a linear decrease of the compressive strength with
the replacement of NA with RA. This is conrmed by the high values
of the determination coefcient of the linear regressions

Fig. 8. 28-day compressive strength in cubes of CRA and FRA concrete mixes versus aggregates' replacement ratio.

68

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Table 9
28-day compressive strength in cylinders.
Aggregates' replacement ratio (%)

CRA Valnor
CRA Vimajas
CRA Ambilei
CRA Europontal
CRA Retria
FRA Vimajas
FRA Ambilei
FRA Europontal

10

25

Compressive strength
in cylinders (MPa)

Compressive strength
in cylinders (MPa)

37.5 1.3

35.3
40.8
37.4
35.7
40.5
36.2
39.1
33.8

2.6
1.8
3.1
2.2
0.5
0.9
3.7
3.5

50

D (%) Compressive strength D (%)

in cylinders (MPa)
5.9
8.9
0.2
4.7
8.1
3.5
4.2
9.9

32.4
33.5
36.0
34.6
34.2
35.7
38.9
33.8

1.2
1.8
1.6
3.2
3.4
0.1
1.4
2.5

concerning the mixes with CRA and FRA tested at 28 days (0.93 and
0.97, respectively).

4.3.2. Compressive strength in cylinders

Table 9 shows that the replacement of NA with RA caused a
decrease of the 28-day compressive strength in cylinders. This results from the RA's composition and the increase in effective w/c
ratio as the RA incorporation in concrete goes up.
Fig. 9 represents the 28-day compressive strength in cylinders of
all mixes. As for strength in cubes, the strength in cylinders loss was
greater when ne aggregates were replaced. The FRA from Vimajas
and Europontal are again the ones with the worst results. This is
mostly due to the need of increasing these mixes' w/c ratio to keep
the slump constant, caused by the presence of clay in the FRA's
composition.
Fig. 9 allows conrming that this property varies considerably
according to the RA's source. For example, full replacement of the
coarse NA caused decreases of the compressive strength in cylinders between 17.1% and 31.6%. This is mostly due to the changes of
the ceramics content in the RA (between 4.2% and 28.6%). On the
other hand, full replacement of the ne NA caused decreases between 17.8% and 42.0%, according to the RA's source. This range of
values is mostly due to two factors: the wide range of ceramics
content in the FRA (between 0.9% and 26.5%); the signicant increase in effective w/c ratio of the mixes with FRA from Vimajas and
Europontal, due to their clay content. These factors led to an increase in porosity of the mixes with RA from these two plants and
consequently a drop in compressive strength in cylinders.

13.6
10.6
4.1
7.8
8.9
4.9
3.8
9.8

100

Compressive strength
in cylinders (MPa)
33.5
30.8
33.6
33.1
32.3
30.2
35.2
33.9

1.9
2.3
1.0
2.1
1.7
1.3
0.4
3.1

D (%)

Compressive strength
in cylinders (MPa)

10.8
17.8
10.5
11.6
13.8
19.5
6.1
9.7

25.6
30.5
31.1
25.7
29.4
21.8
30.8
23.3

0.6
1.7
1.7
0.8
2.5
1.8
0.9
0.1

D (%)
31.6
18.6
17.1
31.4
21.5
42.0
17.8
38.0

Table 10 shows the compressive strength of the mixes with 100%

of RA. For 100% of CRA the mixes reached strength classes C25/30
and C30/37. On the other hand, for 100% of FRA the mixes reached
strength classes C20/25 and C30/37. Four of the mixes from Table 10
would be classied in a higher compressive strength class if only
the characteristic value obtained in cubes were considered.
Fig. 10 shows the determination coefcient (R2) of the linear
regressions concerning the mixes with CRA and FRA (0.95 in both
cases). It is concluded that the loss of compressive strength in
cylinders due to the replacement of NA with RA is linear. Fig. 10 also
conrms that this property's loss in cylinders is greater when FRA
are used than when RCA are used.
4.3.3. Splitting tensile strength
Table 11 shows that the incorporation of CRA and FRA causes a
decrease of the splitting tensile strength. This is due to the increase
in the effective w/c of these mixes and the negative effect of the
composition of some of the RA. Kou et al. (2004) also observed that
the use of 100% of CRA from CDW caused a 28-day tensile strength
loss of the mixes of around 17%.
Table 11 and Fig. 11 show that, similarly to compressive strength,
the loss of tensile strength increases as the ne RA's replacement
ratio goes up. This is due to the great clay content of the FRA
Europontal and Vimajas and the consequent higher w/c ratio of the
mixes with them. The tensile strength loss is also due to the worst
quality of the paste of cement and FRA, which is essential to this
property (Neville, 1995).
It is concluded that the composition of the RA used in these
mixes also had great inuence on the tensile strength results.

Fig. 9. 28-day compressive strength in cylinders of RA concrete mixes.

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

69

Table 10
Compressive strength class of the mixes with 100% of RA.

RC
C100C-Valnor
C100C-Vimajas
C100C-Ambilei
C100C-Europontal
C100C-Retria
C100F-Vimajas
C100F-Ambilei
C100F-Europontal

Compressive
strength class

Characteristic value
specied in cylinders (MPa)

Characteristic value
specied in cubes (MPa)

Characteristic value
obtained in cylinders (MPa)

Characteristic value
obtained in cubes (MPa)

C35/45
C25/30
C30/37
C30/37
C25/30
C25/30
C20/25
C30/37
C20/25

35
25
30
30
25
25
20
30
20

45
30
37
37
30
30
25
37
25

37.5
25.6
30.5
31.1
25.7
29.4
21.8
30.8
23.3

53.9
35.3
42.0
51.1
36.2
40.1
30.2
38.4
29.9

Fig. 12 allows nding out that the splitting tensile strength loss
is linearly proportional to the replacement ratio of NA with RA as
proved by the relatively high values of the determination coefcient (R2) of the linear regressions tried (0.85 and 0.89, for coarse
and ne aggregates).

Accordingly the mixes with FRA and CRA from Ambilei are the ones
with the best results out of the mixes with full aggregates'
replacement. When 100% of CRA from Ambilei are incorporated in
the mix, the splitting tensile strength decreases only from 4.0 MPa
to 3.6 MPa (11.1%), while the worst result, from the mix with 100%
CRA from Europontal, represents a reduction of 36.1%. This different
performance as a function of the RA's nature agrees with the observations of Gomes et al. (2014). They found that the splitting
tensile strength is little affected by the use of RA from concrete in
the mix but highly reduced by the use of ceramics' RA.
There is no reasonable explanation for the low splitting tensile
strength of the mixes C25C-Valnor and C25C-Ambilei. Therefore, it
is considered that these values may result from unidentied
experimental errors.

4.3.4. Modulus of elasticity

Table 12 shows the modulus of elasticity results of all mixes
tested. It is clear that the replacement of NA with RA reduces the
concrete's modulus of elasticity.
Fig. 13 shows that the RA's composition is the factor that most
inuences the modulus of elasticity. The mixes with FRA and CRA
from Ambilei have the best results. Contrarily, the mixes with RA
from Valnor have the worst results in terms of this property. This

Fig. 10. 28-day compressive strength in cylinders of CRA and FRA concrete mixes versus aggregates' replacement ratio.

Table 11
Splitting tensile strength.
Aggregates' replacement ratio (%)

RC
C100C-Valnor
C100C-Vimajas
C100C-Ambilei
C100C-Europontal
C100F-Vimajas
C100F-Ambilei
C100F-Europontal

10

Splitting tensile
strength (MPa)

Splitting tensile
strength (MPa)

25

4.0 0.0

3.7
3.9
3.7
3.4
3.9
4.1
3.5
4.0

0.3
0.1
0.1
0.1
0.2
0.3
0.4
0.4

50

D (%)

Splitting tensile
strength (MPa)

8.3
2.2
7.2
14.9
2.8
2.1
13.2
0.4

3.0
3.8
3.2
3.5
3.7
3.1
3.8
3.4

0.1
0.2
0.2
0.2
0.2
0.2
0.1
0.1

100

D (%)

Splitting tensile
strength (MPa)

26.3
5.4
19.5
12.4
8.9
22.8
6.1
15.8

3.2
3.7
3.0
3.4
3.4
3.4
3.6
3.2

0.0
0.3
0.2
0.3
0.5
0.1
0.2
0.3

D (%)

Splitting tensile
strength (MPa)

19.5
8.5
26.2
14.6
16.3
16.7
10.0
20.6

3.1
2.9
2.6
3.2
2.6
3.6
2.6
2.7

0.1
0.0
0.0
0.1
0.1
0.3
0.1
0.2

D (%)
24.3
27.5
34.6
20.7
34.4
11.1
36.1
32.1

70

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Fig. 11. Splitting tensile strength of RA concrete mixes.

Fig. 12. Splitting tensile strength of CRA and FRA concrete mixes versus aggregates' replacement ratio.

CDW caused a 28-day modulus of elasticity loss of 40%. Oliveira

et al. (2004) analysed the inuence of RA from CDW mostly
made of concrete and found a decrease of around 18% of the
modulus of elasticity, lower than the one in our study. This is

difference is due to the great inuence of the stiffness of the RA

from each source on the modulus of elasticity of concrete. These
results are similar to those of Kou et al. (2004). These authors
observed that full replacement of the coarse NA with CRA from

Table 12
Modulus of elasticity.
Aggregates' replacement ratio (%)

CRA Valnor
CRA Vimajas
CRA Ambilei
CRA Europontal
CRA Retria
FRA Vimajas
FRA Ambilei
FRA Europontal

10

Modulus of
elasticity (GPa)

Modulus of
elasticity (GPa)

25

40.5 0.2

39.1
40.8
38.7
39.8
37.7
38.6
40.3
40.2

0.4
0.6
0.3
0.4
0.0
0.9
0.3
0.0

50

D (%)

Modulus of
elasticity (GPa)

3.5
0.7
4.3
1.1
6.9
4.7
0.5
0.8

34.6
38.4
39.2
36.9
35.5
34.9
38.0
37.5

0.0
0.2
0.5
0.0
0.4
0.5
0.2
0.3

100

D (%)

Modulus of
elasticity (GPa)

14.6
5.2
3.2
9.0
12.2
13.8
6.2
7.4

29.2
34.8
35.9
34.3
31.5
31.9
37.4
35.6

0.9
0.3
0.0
0.6
0.2
0.2
0.4
0.2

D (%)

Modulus of
elasticity (GPa)

27.9
14.1
11.3
15.4
22.3
21.2
7.7
12.2

21.1
26.7
29.9
25.2
26.3
23.3
32.5
26.0

0.5
0.4
0.1
1.3
0.0
0.6
0.6
0.6

D (%)
47.9
34.1
26.3
37.9
35.1
42.5
19.8
35.8

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

71

Fig. 13. Modulus of elasticity at 28 days for RA concrete mixes.

justied by the absence of ceramic material in the RA used. As

referred by Gomes et al. (2014), when the RA contain ceramic
material, the loss in modulus of elasticity is greater, because of the
lower particles density of that material. These studies conrm that
the RA's source signicantly inuences concrete's modulus of
elasticity. The results of our research are additionally explained by
the increase of the effective w/c ratio (i.e. less stiff paste) with the
RA's incorporation.
The moduli of elasticity of the mixes with CRA and FRA from the
same source are similar. The maximum difference is of 8.4%, for the
RA from Vimajas and it results from high clay content of the FRA
from this source. Therefore, unlike in the previous properties, the
aggregates' size did not inuence the inuence the modulus of
elasticity of the resulting concrete mixes.
Fig. 14 shows the determination coefcient (R2) values of the
linear regressions between the mixes' modulus of elasticity and the
CRA and FRA incorporation ratio (1.00 and 0.99, respectively). These
very high values demonstrate the validity of the linear relationships. Fig. 14 also conrms that the incorporation of FRA and CRA
affects concrete modulus of elasticity in a similar way.

4.3.5. Abrasion resistance

This test's objective is to dene the concrete capacity of withstanding actions that cause disaggregation or section loss. The test
was performed for four of the RA's types only (CRA from Valnor and
Retria and FRA from Vimajas and Ambilei).
Table 13 and Fig. 15 present the average wear registered for the
mixes tested, measured by the mass loss, as well as its variation
relative to the RC. The abrasion resistance increases with the use of
CRA, in agreement with previous researches. De Brito et al. (2005)
evaluated the abrasion resistance of various concrete mixes with
red-clay ceramic CRA and found that the loss of thickness of the
specimens decreased as the replacement ratio increased. The authors explained these results with the better bond between the
cement paste and the ceramic RA, because the greater porosity of
these RA allowed a better penetration of the cementitious paste
inside them. Matias et al. (2013) also performed the wear test by
abrasion and found a decrease of the thickness loss between 24%
and 28% of the mixes made with CRA from concrete relative to the
RC. The authors justied these results with the better bond between the RA and the new cement paste.

Fig. 14. Modulus of elasticity at 28 days of CRA and FRA concrete mixes versus aggregates' replacement ratio.

72

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Table 13
Abrasion resistance (measured as mass loss).
CRA Valnor
Mass
loss (%)
0% RA
10% RA
25% RA
50% RA
100% RA

7.6
7.2
6.7
6.6
6.3

0.4
0.2
0.5
0.1
0.5

CRA Retria

D (%) Mass
loss (%)
e
4.6
11.8
12.2
16.7

7.6
7.5
6.9
7.1
6.4

FRA Vimajas

D (%) Mass
loss (%)

0.4 e
7.6
0.5 1.2 7.2
0.4 8.8 8.9
0.3 5.6 10.1
0.9 15.5 11.5

FRA Ambilei

D (%) Mass
0.4
0.4
1.2
0.1
1.3

D (%)

loss (%)
e
4.6
18.3
34.2
52.4

7.6
8.0
8.0
7.9
8.9

0.4 e
0.3 6.4
1.2 5.9
0.4 4.7
0.8 18.0

However, as seen in Fig. 16, the main inuencing factor on the

abrasion resistance is the size of the RA used. Full replacement of
the coarse NA caused an increase of abrasion resistance higher than
15%, independently of the CRA's source. Contrarily, full replacement
of the ne NA led to an abrasion resistance loss between 18% and
53%.
Fig. 16 shows that the abrasion resistance variation depends
linearly on the replacement ratio of NA with RA, as proved by the
determination coefcient (R2) values of the corresponding linear
regressions: 0.73 and 0.98 for CRA and FRA.
5. Conclusions

Table 13 and Fig. 15 show that the RA's nature strongly inuences the abrasion resistance results. For example, in the mixes
made with 100% of FRA from Ambilei there was a reduction of 18.0%
of the abrasion resistance but in the mixes made with 100% of FRA
from Vimajas this reduction reached 52.4%. This results from two
factors. The rst one is the greater w/c ratio of the mixes with FRA
from Vimajas, because they contain clay. The second factor is the
different composition of these FRA. Those from Ambilei are made
mostly of concrete, natural stone and glass, while those from
Vimajas have 75% of concrete, mortar and natural stone, 12% of
ceramic materials and 11% of bituminous materials.

This research intends to analyse the mechanical performance of

concrete made with recycled aggregates (RA) from CDW from
various Portuguese recycling plants. The objective is to evaluate the
inuence of the collection point of the RA, and consequently of
their composition on the characteristics of the concrete produced.
Five types of coarse CDW RA (from Valnor, Vimajas, Ambilei,
Europontal and Retria) and three types of ne CDW RA (from
Vimajas, Ambilei and Europontal) were collected. The objective of
colleting both coarse and ne CDW RA is to also analyse the inuence of their size on concrete's performance.

Fig. 15. Abrasion resistance at 28 days for RA concrete mixes (measured as mass loss).

Fig. 16. Abrasion resistance at 28 days of CRA and FRA concrete mixes versus aggregates' replacement ratio.

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

73

Table 14
Variation of all concrete mechanical properties with the incorporation of RA.
Tests

Compressive strength
in cubes (at 28 days)
Compressive strength
in cylinders
(at 28 days)
Splitting tensile
strength (at 28 days)
Modulus of
elasticity (at 28 days)
Abrasion resistance
(mass loss)

Quantitative variation of the mixes relative to the RC

C100C-Valnor

C100C-Vimajas

34.5

22.1

31.6

C100C-Ambilei

C100C-Europontal

C100C-Retria

C100F-Vimajas

C100F-Ambilei

C100F-Europontal

5.2

32.9

25.6

44.1

28.8

44.6

18.6

17.1

31.4

21.5

42.0

17.8

38.0

24.3

27.5

11.1

36.1

32.1

34.6

20.7

34.4

47.9

34.1

26.3

37.9

35.1

42.5

19.8

35.8

16.7

15.5

52.4

18.0

In order to better understand the experimental results, the RA's

composition was analysed and several physical and chemical tests
of the aggregates were performed. These tests detected a great
variety in the RA's composition, comprising ceramic materials,
concrete, glass, metals, and others.
The results from the fresh-state concrete tests allowed
concluding that the use of RA caused a decrease of concrete's
density. This reduction reached values between 4.7% and 7.7% in the
mixes with 100% RA (ne and coarse). It was also found that, in
order to maintain the slump, it was necessary to increase the
effective w/c ratio of the mixes as the replacement ratio of NA with
RA increased. This situation was more evident with the use of FRA
from Vimajas and Europontal.
The mechanical tests in hardened concrete showed that in most
cases the use of RA caused a decline of the performance of concrete.
Table 14 summarizes the results obtained, allowing the analysis of
the variations in the mixes with 100% RA, relative to the RC, for all
ve properties tested.
There is a wide scatter of results according to the source of the
RA. This was predictable since each RA type has a different
composition, inuenced by the location of the collection point. This
difference causes a signicant range in the aggregates' characteristics and consequently in the resulting concrete mixes' properties.
Table 14 shows that the mixes with FRA from Vimajas had the worst
results, partly because of their high clay content. These ne particles coat the RA grains and hinder a correct bond between them
and the cement paste, weakening the concrete's structure.
Furthermore, clay absorbs the mixing water and that forced an
increase of the effective w/c ratio in order to obtain the same slump
as in the other mixes.
Finally, the abrasion resistance is the property that presents
better results in terms of the use of RA in concrete. As seen in
Table 14, the abrasion resistance seems to improve when coarse
CDW RA are used.
The durability performance of the same concrete mixes has
been analysed in Bravo et al. (2015).

Acknowledgements
The authors gratefully acknowledge the support of the ICIST
cnico from University of
Research Institute, Instituto Superior Te
Lisbon, and FCT (Foundation for Science and Technology).

References
Alves, A.V., Vieira, T.F., de Brito, J., Correia, J.R., 2014. Mechanical properties of
structural concrete with ne recycled ceramic aggregates. Constr. Build. Mater.
64, 103e113.

, A., de Brito, J., Rosa, A., Pedro, D., 2014. Durability performance of concrete
Andre
incorporating coarse aggregates from marble industry waste. J. Clean. Prod. 65,
389e396.
APEB e Portuguese Association of the Concrete Companies, 2007. Aggregates for
Concrete e from Production to Use. Lisbon, Portugal (in Portuguese).
Barbudo, A., de Brito, J., Evangelista, L., Bravo, M., Agrela, F., 2013. Inuence of waterreducing admixtures on the mechanical performance of recycled concrete.
J. Clean. Prod. 59, 93e98.
Barra, M., 1996. Study of the Durability of Concrete with Recycled Aggregates
Applied as Reinforced Concrete. Ph.D. thesis in Construction Engineering. Catalonia Polytechnic University, Barcelona, Spain, p. 222 (in Spanish).
Behera, M., Bhattacharyya, S., Minocha, A., Deoliya, R., Maiti, S., 2014. Recycled
aggregate from C&D waste & its use in concrete e a breakthrough towards
sustainability in construction sector: a review. Constr. Build. Mater. 68,
501e516.
Bergsdal, H., Bohne, R.A., Bratteb, H., 2007. Projection of construction and demolition waste in Norway. J. Ind. Ecol. 11 (3), 27e39.
Bravo, M., de Brito, J., 2012. Concrete made with used tyre aggregates: durabilityrelated performance. J. Clean. Prod. 29 (1), 42e50.
Bravo, M., de Brito, J., Pontes, J., Evangelista, L., 2015. Durability performance of
concrete made with aggregates from construction and demolition waste recycling plants. Constr. Build. Mater. 77, 357e369.
Castro, S. de, de Brito, J., 2013. Evaluation of the durability of concrete made with
crushed glass aggregates. J. Clean. Prod. 41 (2), 7e14.
Cochran, K., Townsend, T., Reinhart, D., Heck, H., 2007. Estimation of regional
building-related C&D debris generation and composition: case study for Florida. US. J. Waste Manag. 27 (7), 921e931.
Costa, U., Ursella, P., 2003. Construction and demolition waste recycling in Italy. In:
Fifth International Conference on the Environmental and Technical Implications
n,
of Construction with Alternative Materials WASCON 2003, San Sebastia
Spain, pp. 231e239.
De Brito, J., Pereira, A.S., Correia, J., 2005. Mechanical behaviour of non-structural
concrete made with recycled ceramic aggregates. Cem. Concr. Compos. 27 (4),
429e433.
De Brito, J., Robles, R., 2010. Recycled aggregate concrete e methodology for estimating its long-term properties. Indian J. Eng. Mater. Sci. 17 (6), 449e462.
Dhir, R.K., Paine, K.A., Halliday, J.E., 2008. WRAP e facilitating the wider use of
coarse and ne recycled aggregates from washing plants. In: Waste & Resources
Action Programme. Oxon, United Kingdom.
DIN 52108, 2010. Testing of Inorganic Non-metallic Materials e Wear Test Using the
Grinding Wheel According to Boehme. Deutsches Institut Fur Normung E.V.
(German National Standard), Berlin, Germany.
Directive 2008/98/EC, 2007. Directive 2008/98/EC of the European Parliament and
of the Council of 19 November 2008 on Waste, Brussels, Belgium.
EN 1744-1, 2009. Tests for Chemical Properties of Aggregates e Part 1: Chemical
Analysis. Committee European for Normalization, Brussels, Belgium.
EN 12350-2, 2002. Testing Fresh Concrete. Slump Test. Committee European for
Normalization, Brussels, Belgium.
EN 12350-6, 2002. Testing Fresh Concrete. Density. Committee European for
Normalization, Brussels, Belgium.
EN 12390-3, 2003. Testing Hardened Concrete. Compressive Strength of Test
Specimens. Committee European for Normalization, Brussels, Belgium.
EN 12390-6, 2003. Testing Hardened Concrete. Tensile Splitting Strength of Test
Specimens. Committee European for Normalization, Brussels, Belgium.
EN 12620, 2008. Aggregates for Concrete. Committee European for Normalization,
Brussels, Belgium.
Etxeberria, M., V
azquez, E., Mar, A., Barra, M., 2007. Inuence of amount of recycled
coarse aggregates and production process on properties of recycled aggregate
concrete. Cem. Concr. Res. 37 (5), 735e742.
Eurostat, environment and energy, 2010. Generation and Treatment of Waste.
Available in: http://ec.europa.eu/eurostat/ (last accessed in 18.05.14.).
Evangelista, L., de Brito, J., 2007. Mechanical behaviour of concrete made with ne
recycled concrete aggregates. Cem. Concr. Compos. 29 (5), 397e401.

74

M. Bravo et al. / Journal of Cleaner Production 99 (2015) 59e74

Evangelista, L., de Brito, J., 2010. Durability performance of concrete made with ne
recycled concrete aggregates. Cem. Concr. Compos. 32 (1), 9e14.
Evangelista, L., de Brito, J., 2014. Concrete with ne recycled aggregates: a review.
Eur. J. Environ. Civ. Eng. 18 (2), 129e172.
Gomes, M., de Brito, J., 2009. Structural concrete with incorporation of coarse
recycled concrete and ceramic aggregates: durability performance. Mater.
Struct. 42 (5), 663e675.
Gomes, M., de Brito, J., Bravo, M., 2014. Structural concrete with incorporation of
coarse recycled concrete and ceramic aggregates: mechanical performance.
J. Mater. Civ. Eng. doi.org/10.1061/(ASCE)MT.1943-5533.0000973.
Hansen, T., 1992. Recycling of Demolished Concrete and Masonry. E & FN SPON,
RILEM report 6, London, United Kingdom, p. 305.
Kou, S., Poon, C., Chan, D., 2004. Properties of steam cured recycled aggregate y ash
concrete. Use of recycled materials in buildings and structures. In: International
RILEM Conference, Barcelona, Spain, pp. 590e599.
LNEC E-397, 1993. Determination of Elasticity Modulus in Compression. National
Laboratory of Civil Engineering (LNEC), Lisbon, Portugal (in Portuguese).
Matias, D., Brito, J. de, Rosa, A., Pedro, D., 2013. Mechanical properties of concrete
produced with recycled coarse aggregates e inuence of the use of superplasticizers. Constr. Build. Mater. 44, 101e109.
nchez de Rojas, M.I., 2012. Microstructure and properties of
Medina, C., Fras, M., Sa
recycled concretes using ceramic sanitary ware industry waste as coarse
aggregate. Constr. Build. Mater. 31, 112e118.
Medina, C., Rojas, M., Fras, M., 2013. Freeze-thaw durability of recycled concrete
containing ceramic aggregate. J. Clean. Prod. 40, 151e160.
Medina, C., Zhu, W., Howind, T., Rojas, M., Fras, M., 2014. Inuence of mixed
recycled aggregate on the physical e mechanical properties of recycled concrete. J. Clean. Prod. 68, 216e225.
Mehta, P., Monteiro, P., 2006. Concrete: Microstructure, Properties and Materials,
third ed. McGraw Hill, New York, United States of America.
Merlet, J., Pimienta, P., 1994. Mechanical and physic-chemical properties of concrete
produced with coarse and ne recycled concrete aggregates. In: Demolition and
Reuse of Concrete and Masonry, RILEM Proceedings 23, Odense, Denmark,
pp. 343e353.
Neville, A., 1995. Properties of Concrete, fourth ed. United Kingdom, London, p. 844.
NP EN 206e1, 2013. Concrete. Specication, Performance Production and Conformity. IPQ, Lisbon, Portugal.
NP EN 933e4, 2002. Tests for Geometrical Properties of Aggregates e Part 4:
Determination of Particle Shape e Shape Index. IPQ, Lisbon, Portugal.
NP EN 1097e2, 2002. Tests for Mechanical and Physical Properties of Aggregates e
Part 2: Methods for the Determination of Resistance to Fragmentation. IPQ,
Lisbon, Portugal.
NP EN 1097e3, 2003. Tests for Mechanical and Physical Properties of Aggregates e
Part 3: Determination of Loose Bulk Density and Voids. IPQ, Lisbon, Portugal.
NP EN 1097e6, 2003. Tests for Mechanical and Physical Properties of Aggregates e
Part 6: Determination of Particle Density and Water Absorption. IPQ, Lisbon,
Portugal.

Oliveira, M., Assis, C., Wanderley, A., 2004. Study on compressed stress, water absorption and modulus of elasticity of produced concrete made by recycled
aggregate. In: Use of Recycled Materials in Buildings and Structures, International RILEM Conference, Barcelona, Spain, pp. 636e642.
Ortiz, O., Pasqualino, J.C., Castells, F., 2010. Environmental performance of construction waste: comparing three scenarios from a case study in Catalonia,
Spain. J. Waste Manag. 30 (4), 646e654.
Pacheco-Torgal, F., Jalali, S., 2010. Reusing ceramic wastes in concrete. Constr. Build.
Mater. 24 (5), 832e838.
Park, S., Lee, B., Kim, J., 2004. Studies on mechanical properties of concrete containing waste glass aggregate. Cem. Concr. Res. 34 (12), 2181e2189.
Pereira, L., Jalali, S., Aguiar, J., 2004. Construction and demolition waste management. In: Workshop Integrated Management Systems for Specic Waste
Streams, Lisbon, Portugal.
Pereira, P., Evangelista, L., Brito, J. de, 2012. The effect of superplasticisers on the
workability and compressive strength of concrete made with ne concrete
aggregates. Constr. Build. Mater. 28 (1), 722e729.
Poon, C.S., Shui, Z.H., Lam, L., 2004. Effect of microstructure of ITZ on compressive
strength of concrete prepared with recycled aggregates. Constr. Build. Mater. 18
(6), 461e468.
Poon, C., Kou, S., Lam, L., 2007. Inuence of recycled aggregate on slump and
bleeding of fresh concrete. Mater. Struct. 40 (9), 981e988.
, A., 2000. Waste Management Plan in Construction and
Reixach, F., Barroso, J., Cusco
Demolition Works. General Direction of the Environment, Institute of Construction Technology of Catalonia, Spain (in Spanish).
Rodrigues, F., Evangelista, L., de Brito, J., 2013a. A new method to determine the
density and water absorption of ne recycled aggregates. Mater. Res. J. 16 (5),
1045e1051.
Rodrigues, F., Carvalho, T., Evangelista, L., de Brito, J., 2013b. Physical-chemical and
mineralogical characterization of ne aggregates from construction and demolition waste recycling plants. J. Clean. Prod. 52, 438e445.
Sanchez, M., Alaejos, P., 2004. Inuence of recycled aggregate quality on concrete
properties. In: Use of Recycled Materials in Buildings and Structures, International RILEM Conference, Barcelona, Spain, pp. 545e553.
Waleed, N., Canisius, T., 2007. WRAP e engineering properties of concrete containing recycled aggregates. In: Waste & Resources Action Programme. Oxon,
United Kingdom.
Xiao, J., Li, W., Sun, Z., Lange, D., Shah, S., 2013. Properties of interfacial transition
zones in recycled aggregate concrete tested by nanoindentation. Cem. Concr.
Compos. 37, 276e292.
Zaharieva, R., Buyle-Bodin, F., Skoczylas, F., Wirquin, E., 2003. Assessment of the
surface permeation properties of recycled aggregate concrete. Cem. Concr. Res.
25 (2), 223e232.
Zong, L., Fei, Z., Zhang, S., 2014. Permeability of recycled aggregate concrete containing y ash and clay brick waste. J. Clean. Prod. 70, 175e182.