Materials and Design 64 (2014) 345–354

Contents lists available at ScienceDirect

Materials and Design

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

The role of 0–2 mm fine recycled concrete aggregate on the compressive

and splitting tensile strengths of recycled concrete aggregate concrete
Adel Gorjinia Khoshkenari a, Payam Shafigh b,⇑, Mahmoud Moghimi c, Hilmi Bin Mahmud b
a
Department of Civil Engineering, Ayatollah Amoli Branch, Islamic Azad University, Amol, Iran
b
Department of Civil Engineering, Faculty of Engineering, University of Malaya, 50603 Kuala Lumpur, Malaysia
c
Faculty of Civil Engineering, Universiti Teknologi Malaysia, Johor Bahru, Malaysia

a r t i c l e i n f o a b s t r a c t

Article history: The aim of this study is to investigate the role of 0–2 mm fine aggregate on the compressive and splitting
Received 1 January 2014 tensile strengths of recycled concrete aggregate (RCA) concrete with normal and high strengths. Normal
Accepted 23 July 2014 coarse and fine aggregates were substituted with the same grading of RCAs in two normal and high
Available online 1 August 2014
strength concrete mixtures. In addition, to keep the same slump value for all mixes, additional water
or superplasticizer were used in the RCA concretes. The compressive and splitting tensile strengths were
Keywords: measured at 3, 7 and 28 days. Test results show that coarse and fine RCAs, which were achieved from a
Recycled concrete aggregate
parent concrete with 30 MPa compressive strength, have about 11.5 and 3.5 times higher water absorp-
Compressive strength
Splitting tensile strength
tion than normal coarse and fine aggregates, respectively. The density of RCAs was about 20% less than
Normal strength concrete normal aggregates, and, hence, the density of RCA concrete was about 8–13.5% less than normal aggre-
High strength concrete gate concrete. The use of RCA instead of normal aggregates reduced the compressive and splitting tensile
strengths in both normal and high strength concrete. The reduction in the splitting tensile strength was
more pronounced than for the compressive strength. However, both strengths could be improved by
incorporating silica fume and/or normal fine aggregates of 0–2 mm size in the RCA concrete mixture.
The positive effect of the contribution of normal sand of 0–2 mm in RCA concrete is more pronounced
in the compressive strength of a normal strength concrete and in the splitting tensile strength of high
strength concrete. In addition, some equation predictions of the splitting tensile strength from compres-
sive strength are recommended for both normal and RCA concretes.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction for concrete will grow to approximately 18 billion tons (about

Concrete is the most widely used construction material due to
its many advantages, such as low cost, availability of raw constitu-
ent materials, workability and its ability to be cast into many
shapes, good fire resistance and durability. The many advantages
of concrete are weighed against the severe energy consumption
and pollution resulting from the manufacture of cement [1]. In
addition, a study conducted by Struble and Godfrey [2] revealed
that between a reinforced concrete beam and a steel I-beam with
the same moment capacity, the concrete beam required much less
energy and had a lower net environment impact than the steel
beam.
It was reported that about 2.7 billion m3 of concrete was pro-
duced in 2002 worldwide, which is more than 0.4 m3 of concrete
produced per person annually [3]. It is expected that the demand
⇑ Corresponding author. Tel.: +60 172437383; fax: +60 379675318.
E-mail addresses: pshafigh@um.edu.my, pshafigh@gmail.com (P. Shafigh).
7.5 billion m3) a year by 2050 [4]. Due to at least three-quarters
of the volume of concrete being occupied by aggregate [5], such
a huge usage of concrete increases the use of natural aggregates,
which creates an ecological imbalance and has a significant effect
on the environment. Consuming such enormous quantities of
natural resources for making concrete and cement means that
the current concrete construction practice is unsustainable [6].
Meyer [7] stated that there are five ways to achieve sustainability
of the concrete industry: (1) increase the use of supplementary
cementitious materials, (2) increase the use of recycled materials,
(3) improve durability, (4) improve mechanical properties, and
(5) reuse of wash water. Among these five ways, the use of recycled
aggregates saves natural resources, dumping space and supports
the environment [8].
Concrete waste constitutes the major proportion of construc-
tion waste totalling about fifty per cent of the total waste gener-
ated [9]. Researchers have shown that among the many types of
solid waste generated from the construction industry recycled

http://dx.doi.org/10.1016/j.matdes.2014.07.048
0261-3069/Ó 2014 Elsevier Ltd. All rights reserved.
346 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354
concrete aggregate (RCA) is a good substitute for aggregate in pro-
ducing structural concrete [10,11]. Tam et al. [9] reported that a
concrete made of recycled aggregates fulfils green requirements.
These requirements are: (i) it can recycle and reduce natural
resources and energy consumption; (ii) it will not affect the envi-
ronment; and (iii) it can maintain sustainable development. It
has been reported [12] that a concrete containing recycled aggre-
gate has a lower compressive strength compared to that of normal
aggregate (NA) concrete. This is due to the lack of bonding between
recycled aggregate and cement matrix and also the high water
absorption capacity of recycled aggregate [13]. The quality of con-
crete containing RCA strongly depends on the amount of RCA in the
concrete mixture. It has been reported [6] that for manufacturing
structural grade concrete by the partial substitution of the natural
aggregate with RCA by up to 50% is feasible. In addition, reports
[14–16] have shown that RCA can be used in high strength, high
performance and durable concretes.
To produce a RCA concrete of good quality, factors such as
remaining micro cracks in RCA after the crushing process, high
water absorption, smaller specific gravity of RCA as well as a pos-
sible reduction in quality and durability need to receive proper
consideration in advance [17]. To produce good quality RCA con-
crete, using coarse and fine RCAs, the use of fine RCA in concrete
is not recommended; alternatively, the replacement of natural
sand by recycled fines was recommended to be limited to a maxi-
mum of 20% [12]. Sima and Park [18] reported that at any age (up
to 28 days) of a mortar, the compressive strength reduces as the
natural sand is replaced with the fine RCA. They also reported that
for making structural concrete members, coarse RCA can be fully
used instead of normal coarse aggregate. However, for such a pur-
pose in a concrete containing 100% coarse RCA, the fine RCA can be
used for up to 60% of the total fine aggregate. The reports of many
researchers have revealed that the compressive strength of RCA
concrete decreases as the amount of RCA increases; this is because
of the existence of the cement paste residue in the aggregate par-
ticles [19]. Fig. 1 shows a schematic of a RCA surrounded with new
mortar in RCA concrete [20]. Indeed, the use of RCA in concrete
involves an increase in the mortar content in the resulting con-
crete, which, in turn, potentially increases the shrinkage strain.
The drying shrinkage strain of RCA concrete may range from 20%
to 70% more than the reference concrete, which, together with
the 100% replacement of NA with RCA, might reach up to 263%
[21].
Although, much research has been conducted on the use of
recycled aggregates in structural concrete, a large amount of these
materials are used in non-structural concretes or used as road
bases [22]. Due to the wide variation in the properties of concrete
containing recycled aggregate, engineers still prefer the use of vir-
gin materials, or the use of recycled material is limited to a low
level replacement of coarse aggregate. In 2012, from a review of
the use of RCA in concrete [12] it was concluded that more
Fig. 1. A recycled aggregate surrounded with new mortar in a recycled aggregate
concrete [20].
research is required to investigate the effect of RCA on the proper-
ties of concrete in terms of the splitting tensile and flexural
strengths, the bond between the aggregate and paste, porosity,
chloride penetration, as well as the effect of quality of the parent
concrete of RCA. In addition, it was recommended that further
research is needed to investigate the potential use of RCA in the
production of high strength, high performance, lightweight and
self-compacting concretes. Therefore, there are still many ques-
tions concerning RCAs, RCA concretes and their properties which
are still unanswered.
The aim of this study is to investigate the effects of using
0–2 mm fine RCA instead of NA on the compressive and tensile
strengths of normal and high strength concretes. In this study,
the quality of the parent concrete is known. In addition, to achieve
a better and more accurate comparison between normal aggregate
and RCA concretes, similar grading was used for normal and recy-
cled concrete aggregates.
2. Experimental work
2.1. Materials used
2.1.1. Binder
Type II Portland cement and silica fume were used as the
binder. The cement used had a specific gravity, Blain surface area,
initial and final setting times of 3.15, 3.06 m2/g, 115 min and
195 min, respectively. The cement had 3, 7 and 28 days compres-
sive strength of about 18.5, 30.0 and 40.0 MPa, respectively. The
chemical composition of the cement and silica fume is shown in
Table 1.
2.1.2. Superplasticizer
A sulphonated naphthalene formaldehyde superplasticizer (SP)
was used for mixes containing silica fume. The amount of SP used
was to achieve a slump value in the range of 80–100 mm.
2.1.3. Aggregates
The recycled concrete aggregates (RCAs) were produced from
crushing tested reinforced concrete beams (RCBs) in the laboratory
(Fig. 2), which can be called the parent concrete (PC). The mix pro-
portion of the PC is shown in Table 2. On average, the compressive
strength of the PC was about 30 MPa. The concrete rubble obtained
from the primary crushing of the RC beams was crushed using a
crushing machine to achieve aggregate of different sizes.
Natural coarse and fine aggregates, which were used the manu-
facture of the PC, were also used in this study. From the different
sizes of RCA, coarse and fine RCA with the same grading as the nat-
ural aggregates were achieved. Therefore, the grading of sand used
in all mixes (normal sand, combined sand and RCA sand) is the
same. The grading of sand along with the ASTM: C33 sand grading
curves are shown in Fig. 3. The densities of the coarse and fine RCAs
were 2.31 and 1.97 g/cm3, respectively, while the density of natu-
ral coarse and fine aggregates in the SSD condition were 2.86 and
Table 1
The chemical composition of binder.
Oxide composition OPC Silica fume
SiO2 21.9 94.6
Fe2O3 3.3 0.9
CaO 63.3 0.5
MgO 1.2 1.0
Al2O3 4.9 1.3
SO3 2.1 0.1
K2O 0.6 1.0
Na2O 0.4 0.3
LOI 2.4 –
 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354 347

Fig. 2. (a) Waste concrete for recycling from tested concrete beams and (b) recycled concrete aggregates after primary crushing.

Table 2
Mix proportions of parent concrete (kg/m3).
Content Cement Water Water/cement ratio Aggregate
0–4 mm 4–12.5 mm
Parent concrete 410 225 0.55 947
Fig. 3. Grading curve of fine natural and recycled concrete aggregates.
2.47 g/cm3, respectively. The fineness modulus of the sand was
3.04. The aggregate crushing value of the coarse RCA and normal
coarse aggregate were 40 and 120 MPa, respectively. The water
absorption of the coarse and fine RCA, as well as the normal coarse
and fine aggregates were measured as being approximately 6.87%,
14.05%, 0.60% and 4.16%, respectively.
Fig. 4 clearly shows the porous texture of the coarse RCA, how-
ever, no porosity could be visually observed on the normal coarse
aggregates.
2.1.4. Mix proportions
Two groups of mix proportions were considered in this study.
Group A includes four normal strength mixes. In this group, all
the mixes had the same cement content. The slump value was kept
735 constant in the range of 80–100 mm for all mixes. Mix N was made
of normal aggregate and is the control concrete mix. The other
mixes had fine and coarse RCAs. Due to the higher water absorp-
tion of RCAs as well as porous surface texture, when natural aggre-
gates were substituted with RCAs in mix N, the slump decreased.
For keeping a constant slump value in the range of 80–100 mm,
the slump value can be increased by increasing the water or adding
SP into the concrete mixture. Therefore, N-RW and N-RS mixes,
which only contain RCAs, had additional water and SP, respec-
tively, compared to mix N.
To investigate the effectiveness of very fine RCA ranging in size
from 0 to 2 mm on the fresh and hardened properties of concrete,
RCAs, except fine aggregates in the range of 0–2 mm, were used
instead of normal aggregate in mix N (mix N-RP). Safiuddin et al.
[12] reported that, typically, RCA particles contain 30–60% old
cement paste/mortar, depending on the aggregate size. A smaller
size includes a greater amount of old cement paste/mortar.
Another report [23] showed that about 20% of coarse RCA of
20–30 mm in size includes old cement/mortar while fine RCA of
up to 0.3 mm size includes 45–65% old cement/mortar. Therefore,
to avoid too great a reduction in the compressive strength of RCA
concrete, the use of fine RCA in concrete is generally not recom-
mended [22,24]. In addition, Meyer [25] reported that most of
the strength loss of RCA concrete compared to normal aggregate
concrete is due to the portion of the RCA that is smaller than
2 mm. In this study, to maximize the usage of RCAs in concrete,
instead of ignoring all the fine RCA, the effect of one portion of fine
RCA in the range of 0–2 mm in size on the fresh and hardened
properties of concrete was investigated. The first advantage of
using natural sand of 0–2 mm instead of RCA in RCA concrete
was observed in the workability of the concrete. In mix N-RP, it
was observed that the slump value without additional water or

Fig. 4. Surface texture of coarse natural aggregate (right) and RCA (left).
348 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354
SP was still similar to the control mix (mix N). This showed the
importance and the role of 0–2 mm fine RCA on the fresh proper-
ties of the RCA concrete.
The aim of making group B mixes was to investigate the effect
of using silica fume and the reduction of water to binder ratio on
the quality of RCA concrete. In this group, mixes of H, H-R and
H-RP are the same as the mixes of N, N-RS and N-RP, respectively,
but with a different type of binder, as well as water and SP con-
tents. The mix proportions of all mixes are shown in Table 3.
2.1.5. Test methods
For mixing of the constituent materials, first, all the solid mate-
rials were blended in a mixer for one minute. Then mixing water
was added to the mixture and mixing continued for three minutes.
Then the slump test was performed. In the RCA concretes, the SP or
additional water was added to the mixture to achieve the desired
slump value of 80–100 mm. The concrete specimens were cast in
100-mm cubes for compressive strength testing and cylinders of
100-mm diameter and 200-mm height for the splitting tensile
strength. The specimens were compacted using a vibrating table
and were demoulded one day after casting. The specimens were
then put in water with a temperature of 20 ± 2 °C until the date
of testing. The tests were conducted using a compression testing
machine of 2000 kN capacity in the environmental lab with
temperature of 22–23 °C and 80 ± 5% relative humidity. For each
tests of compressive and splitting tensile strengths sixty-three
specimens were tested. The results reported in this study are the
average of three tests.
3. Results and discussion
3.1. Properties of RCAs
The test results showed that both coarse and fine RCAs are
about 20% lighter than normal aggregates. The lower density of
RCAs was due to the old mortar being attached to the normal
aggregates. The mortar was from a concrete with a compressive
strength of 30 MPa. Therefore, it had more porosity than a natural
aggregate. Some of these porosities can be visually observed on the
surface of the coarse RCA in Fig. 4. Evangelista and de Brito [26]
reported that due to the higher porosity of the recycled aggregates,
these aggregates have lower specific densities. They have recog-
nized the higher porosity of the recycled aggregates from the
greater water absorption.
Test results showed that the water absorption of coarse and fine
RCAs was about 11.5 and 3.5 times more than the natural coarse
and fine aggregates, respectively. Such high level of water absorp-
tion for both coarse and fine RCAs is noticeable. The main reason
for the high water absorption of RCAs is due to the porosity of
these aggregates. The porosity in these aggregates not only can
be observed on the surface (Fig. 4) but also in microscopic scale.
Figs. 5 and 6 show microscopic images of a RCA and normal
Table 3
Mix proportions of normal aggregate and RCA concretes (kg/m3).
Mix group Mix No. Cement Silica fume Water Additional SP Normal aggregates
water
0–2 mm 2–4 mm 4–9.5 mm 9.5–12.5 mm 0–2 mm 2–4 mm 4–9.5 mm 9.5–12.5 mm
A N 410 – 195 – – 675
N-RW 410 – 195 30 – –
N-RS 410 – 195 – 1.4 –
N-RP 410 – 195 – – 675
B H 369 41 100 – 12.3 675
H-R 369 41 100 – 12.3 –
H-RP 369 41 100 – 12.3 675
aggregate in different scales, respectively. It is clear from these fig-
ures that the texture surface of RCA is like a sponge and many
pores can be observed on the surface while these are not observed
in normal aggregates. Kou and Poon [27] reported that water
absorption is the most distinctive difference between the proper-
ties of the RCAs and normal aggregates. They demonstrated that
the high water absorption of RCA is due to the presence of the por-
ous cement mortar. de Juan and Gutiérrez [28] showed that there is
a relationship between the attached mortar and water absorption
of RCA. Previous studies [17,29–35] showed that coarse and fine
aggregates have water absorption in the range of 3.5–7.7% and
4.6–13.1%, respectively.
In addition, the aggregate crushing value test results revealed
that the coarse RCA used in this study had a strength value of about
one third of the normal coarse aggregate. Ismail and Ramli [36]
reported that due to the presence of mortars with porous nature
remains in the RCA as well as the presence of numerous micro-
cracks, the quality of RCA is lower than normal aggregates.
3.2. Density
The density of RCAs is about 20% less than that of normal aggre-
gates. Therefore, it was expected that by substituting the natural
aggregate with such RCAs the density of concrete would reduce.
The densities of group A mixes were 2468, 2132, 2148 and
2273 kg/m3 for the N, N-RW, N-RS and NRP mixes, respectively.
The densities of group B mixes were 2546, 2266 and 2276 for the
H, H-R and H-RP mixes, respectively. In groups A and B, the con-
cretes containing RCAs showed 7.9–13.6% and 10.6–11% lower
density, respectively, compared to the control concrete mix. Saf-
iuddin et al. [12] reported that the density of RCA concrete is typ-
ically 5–15% lower than normal aggregate concrete. They stated
that this is due to the old mortar that adheres to the normal aggre-
gate, which has a much lower density than normal aggregates.
A concrete with a density in the range of 1440–1840 kg/m3 is a
lightweight concrete, while the density of ordinary concrete is in
the range of 2240–2400 kg/m3 [37]. Therefore, a concrete with a
density in the range of 1840–2240 kg/m3 can be called semi-light-
weight concrete. In addition, in practice, a concrete with a density
less than 2160 kg/m3 is considered as lightweight concrete [38].
Therefore, two concrete mixes N-RW and N-RS can be categorized
as lightweight concrete. The immediate benefit of using light-
weight concrete is the reduction of reinforcement and the founda-
tions due to saving in dead load.
3.3. Compressive strength
The compressive strength test results of all mixes at 3, 7 and
28 days are shown in Table 4. The analysis of the results for each
group, as well as a comparison of the results of groups A and B,
are as follows.
Recycled concrete aggregate
272 184 551 – – – –
– – – 675 272 184 551
– – – 675 272 184 551
– – – – 272 184 551
272 184 551 – – – –
– – – 675 272 184 551
– – – – 272 184 551
 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354 349

Fig. 6. Microscopic images from the surface of normal aggregate.

compressive strength of mix N-RW compared to mix N at 3, 7

and 28 days was about 38%, 34% and 27%, respectively. It can be
seen that with increasing age, the difference in compressive
strength between these two mixes reduced.
In mix N-RS, all aggregates were RCA. In this mix, to achieve the
same workability as mix N, superplasticizer was used in the
mixture instead of adding water. It can be seen from Table 4 that
the compressive strength of this mix was much better than mix
N-RW. Mix N-RS showed 61%, 39% and 18% higher compressive
strength than mix N-RW at 3, 7 and 28 days, respectively. There-
fore, it can be concluded that the use of superplasticizer in concrete
containing wholly RCA is mandatory.

Table 4
Compressive strength of concretes (MPa).

Mix group Mix No. Compressive strength (MPa)

Fig. 5. Microscopic images from the surface of RCA. 3 days 7 days 28 days
A N 15.5 26.1 38.0
N-RW 9.6 17.3 27.9
3.3.1. Group A N-RS 15.5 24.1 32.8
Mix N was a normal strength concrete. In this concrete when all N-RP 13.9 22.8 34.3
the conventional aggregates were substituted with the recycled B H 39.1 63.9 81.3
concrete aggregates, for similar workability (mix N-RW), the H-R 21.9 35.4 52.2
compressive strength reduced significantly. The reduction of the H-RP 22.1 35.5 47.9
350 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354
From the slump test result for mix N-RP, it was observed that if
in a whole RCA concrete, just fine RCA ranging in size from 0 to
2 mm be substituted with normal sand, the slump of RCA would
be similar to normal concrete. Indeed, the performance of normal
fine aggregate of 0–2 mm on the fresh property of RCA concrete
was similar to the use of superplasticizer or additional water into
the mixture. In addition, the 28-day compressive strength of mix
N-RP was about 23% and 5% more than the N-RW and N-RS mixes,
respectively. However, mix N-RP had a 28-day compressive
strength of about 10% lower than the control mix (mix N). In gen-
eral, from the test results of group A mixes, it can be concluded that
if the parent concrete is grade 30, it is possible to produce the same
grade concrete containing wholly RCA by using a small amount of
superplasticizer or by substituting fine RCA of 0–2 mm with nor-
mal fine aggregate.
It is worth noting that the volume of slabs is generally some-
where between 70% and 90% of the total concrete volume of a
building [39]. In addition, strength is not a major consideration
in floor slabs [4]. Therefore, the use of concretes, such as N-RW
and N-RS with a density of about 13% less than normal weight con-
crete in the floors of buildings, particularly in high-rise buildings,
could result in a saving on the material used, such as steel.
3.3.2. Group B
The aim of producing this group of mixes was to investigate the
effectiveness of using silica fume along with a reduction of water
content by using superplasticizer on the properties of RCA con-
crete. In this group, the H, H-R and H-RP mixes were similar to
the N, N-RS and N-RP mixes from group A, respectively, except in
the amount of silica fume, water and superplasticizer contents.
The compressive strength test results are shown in Table 4. The
test results indicated that in concrete containing wholly RCA, if sil-
ica fume is used as a cement replacement together with a reduced
of water content (by using a superplasticizer), with the same
slump value, the compressive strength could be significantly
increased. Mix H-R had a compressive strength of about 42%, 47%
and 59% more than mix N-RS at 3, 7 and 28 days, respectively.
These results showed that the increase in rate of strength gain
for mix H-R was more than mix N-RS. This was due to the existing
silica fume in the H-R mixture, which had a significant effect in
terms of filling and pozzolanic reaction.
Elhakam et al. [40] reported that in a RCA concrete containing
coarse RCAs and natural sand as fine aggregate, adding 10% silica
fume as cement addition enhanced the properties of concrete. They
attributed the improvement to the ultra-fine particles, which
decreases the porosity of recycled aggregate concrete; the pozzola-
nic action of the silica fume, which enhances the cement matrix;
and the improvement of the transition zone between the recycled
aggregates and cement mortar. The 28-day compressive strength
of mix H-R was also significantly more than the compressive
strength of the parent concrete. This showed that if RCAs obtained
Table 5
Splitting tensile strength of concretes (MPa).
Mix group Mix No. Splitting tensile strength (MPa)
3 days 7 days
A N 1.8 2.4
N-RW 1.1 2.1
N-RS 1.7 2.3
N-RP 1.6 2.2
B H 2.9 4.0
H-R 2.1 2.7
H-RP 2.0 2.8
from a weak concrete, it is possible to produce concrete containing
wholly RCA with greater strength.
A comparison between the compressive strength test results of
the H-R and H-RP mixes showed that substituting the fine RCAs of
0–2 mm with normal sand aggregate does not have a significant
effect on the compressive strength, while it was observed that this
size range of aggregate may have a significant effect on the fresh
properties and compressive strength of a RCA concrete with nor-
mal strength.
A comparison between the compressive strength of N and H as
well as N-RS and H-R mixes showed that the contribution of the
silica fume to improve the compressive strength of the RCA con-
crete was lower than that for the normal aggregate concrete. The
improvement in the compressive strength of RCA concrete at early
ages and 28-day age was about one third and half that of the nor-
mal aggregate concrete, respectively. This is in contrast to the
results reported by Kou et al. [41] in the case of recycled aggregate
(RA) concrete in which the RA contained more than 90% crushed
recycled rubble. They studied the effect of different mineral admix-
tures, such as silica fume, metakaolin, fly ash and ground granu-
lated blast furnace slag as cement replacement in percentages of
10%, 15%, 35% and 55%, respectively, on some engineering proper-
ties of RA concrete. They found that silica fume and metakaolin
improve the short and long-term properties of the concrete.
Moreover, they concluded that the contribution of the mineral
admixtures to improve the performance of RA concrete is higher
compared to that for normal aggregate concrete.
It should be noted that most general-use concrete has a com-
pressive strength of between 20 and 40 MPa [42]. The compressive
strength test results showed that in group A, all RCA concretes had
a 28-day compressive strength in this range. However, in group B,
the 28-day compressive strength of RCA concretes were signifi-
cantly higher than 40 MPa, which could be considered as high
strength concrete.
3.4. Splitting tensile strength
Compressive and tensile strengths are both required for the
design of structures. However, in many practical applications, such
as unreinforced concrete structures, structures under seismic load-
ings, pavement, slabs, mass concrete dams, airfield runways, and
segmental pre-stressed concrete bridges, the tensile strength value
is even more important than the compressive strength in the
design of these structures [43,45].
A review of the available information concerning the effect of
RCA on the splitting tensile strength of concrete conducted by Saf-
iuddin et al. [12] in 2012 showed that limited literature is available
in this regard. In addition, there has not been enough studies to
develop an empirical equation for predicting the splitting tensile
strength for RCA concretes. According to the available information,
it has been reported that the splitting tensile strength of RCA con-
crete is 0–10% lower than that for normal aggregate concrete. It can
Splitting tensile strength/compressive
strength ratio at 28 days (%)
28 days
4.2 11.1
2.8 10.0
3.1 9.5
3.4 9.9
5.7 7.0
4.0 7.7
4.2 8.8
 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354
be seen that the reduction in the splitting tensile strength is not
significant.
Table 5 shows the splitting tensile strengths for all mixes at 3, 7
and 28 days. In group A, all RCA concretes had a lower splitting
tensile strength than the control mix (mix N). The reduction varied
from 4% to 40% for all ages. The lowest splitting tensile strength
was for mix N-RW. Compared to mix N-RW, mix N-RS showed
about 55%, 10% and 11% higher splitting tensile strength at the
3-, 7- and 28-day ages. Such an improvement in the tensile
strength was due to the use of a small amount of superplasticizer
instead of increasing the water content in the RCA concrete mix-
ture. Therefore, according to the test results of the compressive
and tensile strengths, it is strongly recommended that a superp-
lasticizer be used in concrete containing wholly RCA mixtures.
The best result for the tensile strength of the RCA concrete in
group A was achieved for mix N-RP at the 28-day age. The splitting
tensile strength of the RCA concrete containing normal 0–2 mm
sand (mix N-RP) was about 10% more than for the RCA concrete
containing superplasticizer (mix N-RS) at 28-day age. This shows
that the performance of the normal sand in the 0–2 mm size range
is similar or better than with the superplasticizer.
It should be noted that the minimum splitting tensile strength
for a structural lightweight aggregate concrete is 2.0 MPa [46].
Mix N-RW has the lowest density among all the mixes. This mix
and mix N-RS are considered as semi-lightweight concretes. The
splitting tensile strength of both mixes at 7 and 28 days is more
than the minimum requirement for structural lightweight aggre-
gate concrete.
In group B, similar to group A, both RCA concretes showed a
lower splitting tensile strength in the range of 26–32% compared
to the normal aggregate concrete at all ages. However, a compari-
son between the splitting tensile strength of RCA concretes in
groups A and B showed that group B RCA concretes had much
higher splitting tensile strength, and, generally, they had a better
tensile strength than the group A control concrete (mix N). It can
be seen that if silica fume is used in the RCA concrete mixture
along with reduced water to binder ratio, the tensile strength of
RCA concrete improved to about 14% at early ages and is equivalent
to the tensile strength of normal aggregate concrete.
Table 5 shows the splitting tensile strength to compressive
strength ratio for all mixes at 28-day age. In group A, the ratio of
all RCA concretes was lower than the control concrete. This showed
that although the contribution of RCA in the concrete mixture
reduced both the compressive and splitting tensile strengths, the
reduction in the splitting tensile strength was more pronounced
than for the compressive strength. Another reason that proves this
conclusion is that, as reported by Neville [5], when the compres-
sive strength of concrete increases, the tensile strength also
increases but at a decreasing rate. For example, an investigation
Fig. 7. Relationship between splitting tensile strength and compressive strength at
28 days.
351
on the splitting tensile and compressive strengths of concretes
with a 28-day compressive strength up to 84 MPa showed that at
low strengths, the splitting tensile strength may be as high as
10% of the compressive strength; however, at higher strengths, it
may reduce to 5% [47]. Therefore, normally, the splitting tensile
strength/compressive strength ratio of N-RW, N-RS and N-RP
mixes would be expected to be higher than mix N due to their
compressive strength being lower than mix N, while the ratio
was lower than mix N in these concretes. Therefore, it can be seen
that the contribution of RCA in a concrete with normal strength has
a greater negative effect on the splitting tensile strength compared
to the compressive strength. However, the test results of group B
concretes showed that reducing the water content and the use of
silica fume in the RCA concrete mixture improved the splitting ten-
sile strength/compressive strength ratio significantly. In this group,
it can be seen that, generally, concrete with lower strength had a
higher splitting tensile strength/compressive strength ratio. A
comparison between the splitting tensile strength ratios of the
N-RP and N-RS mixes showed that the reduction rate of the split-
ting tensile strength of concrete containing wholly RCA reduced
if the 0–2 mm RCAs was substituted with the same size of normal
aggregates. This positive effect of 0–2 mm normal sand was more
pronounced in the group B RCA compared to group A.
Fig. 7 shows the relationship between the splitting tensile and
compressive strengths for all mixes at 28 days. The relationship
is considered for all concretes as well as for concretes containing
RCA. It can be seen that in both cases, there is a nonlinear relation-
ship with a strong correlation between the two strengths. Zain
et al. [43] stated that the relationship between the splitting tensile
strength and compressive strength is nonlinear. This is because
when the compressive strength of concrete increases, the tensile
strength also increases but at a much smaller rate when compared
to the increase in compressive strength.
There are a number of empirical relationships between the
splitting tensile and compressive strengths proposed by various
codes of practice and researchers. In most cases, the relationships
are in the following format:
ft ¼ k  fca ð1Þ
where ft is the splitting tensile strength, fc is the compressive
strength and k and a are constant coefficients.
However, there are other forms of equation predictions, such as
Eqs. (2) and (3), which were reported by Zain et al. [43] and Eq. (4)
reported by Parrott [48].
ft ¼ 0:54fcy0:5 =ðw=bÞ ð2Þ
ft ¼ fcy =ð0:1fcy þ 7:11Þ ð3Þ
ft ¼ 0:226fcy0:705 þ 1 ð4Þ
where ft is the splitting tensile strength (MPa), fcy is the cylinder
compressive strength (MPa), w/b is water to binder ratio and k
and a are constant coefficients.
A comparison between the experimental results obtained for
wholly RCA concretes in this study and those estimated by Eq.
(1) reported by [43,48–58] showed that the equations proposed
by ACI 318-99 [50] (Eq. (5)), Arioglu et al. [54] (Eq. (6)), and Bhanja
and Sengupta [56] (Eq. (7)) give a closer estimated value to the real
value measured for mixes of N-RW, N-RS and H-R. The average
underestimate error of these equations was about 3%. In these con-
cretes, Eq. (2) also gives a good estimation with an average under-
estimate error of about 6%. In the general form, if all concretes
containing RCA are considered, the best proposed equations are
those reported by Arioglu et al. [54] (Eq. (6)) and Bhanja and Seng-
upta [56] (Eq. (7)) with, on average, underestimate error of about
352 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354

6%. In addition, Eq. (4) also gives an average error of about 5% for all
five mixes containing RCA.

ft ¼ 0:59fcy0:5 ð5Þ

ft ¼ 0:387fcy0:63 ð6Þ

0:717
ft ¼ 0:248fcu
Among the many existing proposed equations that were inves-
tigated by the authors it was found that a few equations could be
used for all mixes with an error of less than 10%. Eq. (4) with an
absolute error of about 7% as well as Eqs. (6) and (7) with an under-
estimate error of about 8% and 7%, respectively, can be used for all
mixes. Overall, the use of Eqs. (4) and (7) are recommended for
predicting splitting tensile strength from compressive strength
for all concrete mixes at the 28-day age.
Although most prediction equations for the splitting tensile
strength are a function of the 28-day compressive strength, some
of the proposed equations were developed based on the function
of the compressive strength and concrete age. Eq. (8) shows that
there is a relationship between the tensile strength, compressive
strength and concrete age, which was proposed by Zain et al. [43].
0:5
fts;t ¼ 0:59fcy;t =ðt=t 28 Þ0:04
where fts,t and fcy,t are the splitting tensile strength and compressive
strength, respectively, in MPa at the age of t days, where t is age of
concrete at t days.
In addition, Eq. (9), as proposed by De Larrard and Malier [44],
was developed for predicting the splitting tensile strength from the
compressive strength at any age.
ftj ¼ 0:6 þ 0:06fcj
ð7Þ
Fig. 8. Relationship between splitting tensile strength and compressive strength of
concretes at all ages.
1.13. These results indicate the good accuracy of Eqs. (8) and (9),
particularly in the case of RCA concretes.
Fig. 8 shows relationship between splitting tensile strength and
compressive strength at all ages. It can be seen that there is a rela-
tionship with strong correlation between these two properties for
concretes containing RCA as well as for all concretes. From this fig-
ure, it can be concluded that irrespective of the type of aggregate
ð8Þ
used in the concrete, Eq. (10) can be used to predict the splitting
tensile strength from compressive strength with good accuracy.
This equation is very similar to the equation reported by Bhanja
and Sengupta [56].
0:71
ft ¼ 0:24fcu ð10Þ
4. Conclusions
ð9Þ

where ftj and fcj are average values of the splitting tensile strength
and the compressive strength, both in MPa, respectively, at j days.
Table 6 presents the results of the predicted splitting tensile
strength derived from Eqs. (8) and (9), as well as the experimen-
tal/predicted splitting tensile strength ratio of these equations. As
can be seen in this table, generally, Eq. (8) gives an overestimate
prediction at early ages and a conservative estimation at the
28-day age. However, Eq. (9) gives a conservative prediction at
all ages. The experimental/predicted ratio of Eq. (8) ranged
between 0.73 and 1.27 with an average value of 0.98. The average
experimental/predicted ratio value for wholly RCA concretes
(mixes of N-RW, N-RS and H-R) was 0.94 and for all concretes
containing RCA it was 0.95. The experimental/predicted ratio of
Eq. (9) ranged between 0.91 and 1.45 with an average value of
1.12. The average experimental/predicted ratio value for wholly
RCA concretes was 1.12 and for all concretes containing RCA was
From the test results of this experimental study the following
conclusions can be drawn:
(1) Coarse and fine RCAs obtained from a parent concrete with
30 MPa compressive strength may have significantly higher
water absorption than normal coarse and fine aggregates of
about 11.5 and 3.5 times, respectively. In addition, the
aggregate crushing value of the coarse RCA is about one
third of the normal coarse aggregate.
(2) The density of RCAs is about 20% less than normal aggre-
gate. Concretes containing RCAs have 8–13.5% less density
than normal aggregate concrete and can be categorized as
semi-lightweight aggregate concrete.
(3) Due to the higher water absorption value of RCAs, as well as
the porous surface texture of these aggregates, substituting
normal coarse and fine aggregates with RCAs reduces the

Table 6
Predicted splitting tensile strength of concretes for different ages (MPa).

Mix group Mix no. Predicted splitting tensile strength (MPa)

Eq. (8) Eq. (9)
3d 7d 28d 3d 7d 28d
P E/P P E/P P E/P P E/P P E/P P E/P
A N 1.9 0.95 2.6 0.92 3.3 1.27 1.5 1.20 2.2 1.09 2.9 1.45
N-RW 1.5 0.73 2.1 1.0 2.8 1.00 1.2 0.92 1.6 1.31 2.3 1.22
N-RS 1.9 0.89 2.5 0.92 3.0 1.03 1.5 1.13 2.0 1.15 2.6 1.19
N-RP 1.8 0.89 2.4 0.92 3.1 1.10 1.4 1.14 2.0 1.10 2.7 1.26
B H 3.0 0.97 4.0 1.0 4.8 1.19 2.9 1.00 4.4 0.91 5.5 1.04
H-R 2.3 0.91 3.0 0.90 3.8 1.05 1.9 1.11 2.7 1.00 3.7 1.08
H-RP 2.3 0.87 3.0 0.93 3.7 1.14 1.9 1.05 2.7 1.04 3.5 1.20

P: prediction; E: experiment.
 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354
slump value of the concrete. However, in a wholly RCA con-
crete, if fine RCA ranging from 0 to 2 mm in size is substi-
tuted with normal sand, the slump value significantly
improves and could be the same as the control normal
aggregate concrete.
(4) A RCA concrete derived from substituting normal coarse
and fine aggregates with RCAs in a normal strength con-
crete has lower compressive strength than normal concrete
for the same slump value. The reduction in the compressive
strength can be significant or negligible. It depends on the
method for enhancing the workability of RCA concrete to
reach the slump value of the control mix. The reduction
in the compressive strength is significant if additional
water is added, and is negligible if superplasticizer or nor-
mal sand of 0–2 mm is used in the RCA concrete mixture.
(5) A RCA concrete may have much better compressive
strength (more than 70%) than the parent concrete if silica
fume is used as a cement replacement along with a
decrease in the water to binder ratio.
(6) The effectiveness of contribution of silica fume along with a
reduction in the water to binder ratio in the concrete
mixture to improve the compressive strength of concrete
is significantly more in normal aggregate concrete than
RCA concrete.
(7) The use of RCAs instead of normal aggregates in a high-
strength-normal-aggregate concrete mixture significantly
decreases the compressive strength. However, it is negligi-
ble in a normal-strength-normal-aggregate concrete.
(8) The reduction in the splitting tensile strength of concrete
due to the use of RCAs instead of normal aggregates in a
normal-strength-normal-aggregate concrete is more pro-
nounced than the reduction in compressive strength. How-
ever, by using silica fume along with a reduction in the
water to binder ratio, the splitting tensile strength signifi-
cantly improves so that it can be equivalent to the control
normal aggregate concrete.
(9) The use of 0–2 mm normal sand in a RCA concrete reduces
the splitting tensile strength loss. This positive effect is
more pronounced in high strength RCA concrete than for
normal strength.
(10) For predicting the 28-day splitting tensile strength from
the compressive strength of both normal aggregate and
RCA concretes, the use of the equation predictions of
ft ¼ 0:226fcy0:705 þ 1 [48], ft ¼ 0:248fcu
0:717
[56] and ft ¼
0:387fcy0:63 [54] are recommended. In addition, from test
results of this study, the equation prediction of ft ¼
0:71
0:24fcu can be used for all concretes.
References
[1] Calkins M. Materials for sustainable sites: a complete guide to the evaluation,
selection, and use of sustainable construction materials. New York: John
Wiley; 2009.
[2] Struble L, Godfrey J. How sustainable is concrete? In: Proceeding of
international workshop on sustainable development and concrete
technology, Beijing, 20–21 May 2004. p. 201–10.
[3] Naik TR. Sustainability of concrete construction. Pract Periodical Struct Des
Constr 2008;13(2):98–103.
[4] Mehta PK, Monteiro PJM. Concrete: microstructure, properties, and materials.
3rd ed. New York: McGraw-Hill; 2006.
[5] Neville AM. Properties of concrete. 14th ed. Malaysia: Prentice Hall; 2008.
[6] Moriconi G. Recyclable materials in concrete technology: sustainability and
durability. In: Kraus RN, Naik TR, editors. Sustainable construction materials and
technologies, Proc. Special Sessions of First inter. conf. on sustainable
construction materials and technologies, Coventry, UK, 11–13 June 2007. p. 1–12.
[7] Meyer C. Concrete as a green building material. In: Proceedings of the Third Int.
Conference on Construction Materials, Vancouver, BC, Canada, August 22–24,
2005.
353
[8] Roos F, Zlich K. Verification of the dimensioning values for concrete with
recycling concrete aggregates: Symposium on sustainable construction and
use of recycled concrete aggregates, 11–12 November 1998, London.
[9] Tam Vivian WY, Gao XF, Tam CM, Ng KM. Physio-chemical reactions in recycle
aggregate concrete. J Hazard Mater 2009;2009(163):823–8.
[10] Levy SM, Helene P. Durability of recycled aggregates concrete: a safe way to
sustainable development. Cem Concr Res 2004;34(11):1975–80.
[11] Yaprak H, Aruntas HY, Demir I, Simsek O, Durmus G. Effects of the fine recycled
concrete aggregates on the concrete properties. Int J Phys Sci
2011;6(10):2455–61.
[12] Safiuddin M, Alengaram UJ, Rahman MM, Salam MA, Jumaat MZ. Use of
recycled concrete aggregate in concrete: a review. J Civ Eng Manage
2013;19(6):1–15.
[13] Topçu IB, Sarıdemir M. Prediction of mechanical properties of recycled
aggregate concretes containing silica fume using artificial neural networks
and fuzzy logic. Comput Mater Sci 2008;42(1):74–82.
[14] Shayan A, Xu A. Performance and properties of structural concrete made with
recycled concrete aggregate. ACI Mater J 2003;100(5):371–80.
[15] Hendriks CF, Pieterson HS. Concrete: durable but also sustainable. In: Dhir DK,
Henderson NA and Limbachiya MC, editors. Proceedings of international
conference on the use of recycled concrete aggregates, UK, 1998. p. 1–18.
[16] Ajdukiewicz A. Influence of recycled aggregate on mechanical properties of
HS/HPC. Cem Concr Compos 2002;24(2):269–79.
[17] Lin YH, Tyan YY, Chang TP, Chang CY. An assessment of optimal mixture for
concrete made with recycled concrete aggregates. Cem Concr Res
2004;34:1373–80.
[18] Sima J, Park C. Compressive strength and resistance to chloride ion penetration
and carbonation of recycled aggregate concrete with varying amount of fly ash
and fine recycled aggregate. Waste Manage 2011;31:2352–60.
[19] Jian Yanga J, Dub Q, Bao Y. Concrete with recycled concrete aggregate and
crushed clay bricks. Constr Build Mater 2011;25:1935–45.
[20] Nassar RUD, Soroushian P. Strength and durability of recycled aggregate
concrete containing milled glass as partial replacement for cement. Constr
Build Mater 2012;29:368–77.
[21] Gomez-Soberon JMV. Shrinkage of concrete with replacement of aggregate
with recycled concrete aggregate. ACI Spec Publ 2002;209:475–96.
[22] Malešev M, Radonjanin V, Marinkovic S. Recycled concrete as aggregate for
structural concrete production. Sustainability 2010;2(5):1204–25.
[23] Bhikshma V, Kishore R. Development of stress–strain curves for recycled
aggregate concrete. Asian J Civ Eng (Build Hous) 2010;11(2):253–63.
[24] Anderson K, Uhlmeyer W, Russell M. Use of recycled concrete aggregate in
PCCP: literature Search. Washington State Development of Transportation
(WSDOT). Washington: WSDOT Office of Research – Library Services; 2009.
[25] Meyer C. The greening of the concrete industry. Cem Concr Compos
2009;31(8):601–5.
[26] Evangelista L, de Brito J. Durability performance of concrete made with fine
recycled concrete aggregates. Cem Concr Compos 2010;32:9–14.
[27] Kou SC, Poon CS. Properties of concrete prepared with PVA-impregnated
recycled concrete aggregates. Cem Concr Compos 2010;32:649–54.
[28] de Juan MS, Gutiérrez PA. Study on the influence of attached mortar content on
the properties of recycled concrete aggregate. Constr Build Mater
2009;23:872–7.
[29] Gomez-Soberon JMV. Porosity of recycled concrete with substitution of
recycled concrete aggregate: an experimental study. Cem Concr Res
2002;32:1301–11.
[30] Kou SC, Poon CS. Properties of self-compacting concrete prepared with coarse
and fine recycled concrete aggregates. Cem Concr Compos 2009;31:622–7.
[31] Grdic ZJ, Toplicic-Curcic GA, Despotovic IM, Ristic NS. Properties of self-
compacting concrete prepared with coarse recycled concrete aggregate. Constr
Build Mater 2010;24:1129–33.
[32] Manzi S, Mazzotti C, Bignozzi MC. Short and long-term behavior of structural
concrete with recycled concrete aggregate. Cem Concr Compos 2013;37:312–8.
[33] Pereira P, Evangelista L, de Brito D. The effect of superplasticisers on the
workability and compressive strength of concrete made with fine recycled
concrete aggregates. Constr Build Mater 2012;28:722–9.
[34] Oikonomou ND. Recycled concrete aggregates. Cem Concr Compos
2005;27:315–8.
[35] Tangchirapat W, Khamklai S, Jaturapitakkul C. Use of ground palm oil fuel ash
to improve strength, sulfate resistance, and water permeability of concrete
containing high amount of recycled concrete aggregates. Mater Des
2012;41:150–7.
[36] Ismail S, Ramli M. Engineering properties of treated recycled concrete
aggregate (RCA) for structural applications. Constr Build Mater
2013;44:464–76.
[37] Nowak A, Rakoczy A. Statistical model for compressive strength of lightweight
concrete. Architect Civ Eng Environ 2011;4(4):73–80.
[38] Graybeal B, Lwin MM. Lightweight concrete in highway infrastructure. ASPIRE
2013; Spring. p. 44–5.
[39] CEB/FIP manual of design and technology. Lightweight aggregate
concrete. Great Britain: First pub.; 1977.
[40] Elhakam AA, Mohamed AE, Awad E. Influence of self-healing, mixing method
and adding silica fume on mechanical properties of recycled aggregates
concrete. Constr Build Mater 2012;35:421–7.
[41] Kou S, Poon C, Agrel F. Comparisons of natural and recycled aggregate
concretes prepared with the addition of different mineral admixtures. Cem
Concr Compos 2011;33:788–95.
354 A.G. Khoshkenari et al. / Materials and Design 64 (2014) 345–354
[42] Kostmatka SH, Kerkhoff B, Panarese WC. Design and control of concrete
mixtures. 14th ed. USA: Portland Cement Association; 2002.
[43] Zain MFM, Mahmud HB, Ilham A, Faizal M. Prediction of splitting tensile
strength of high-performance concrete. Cem Concr Res 2002;32(8):1251–8.
[44] De Larrard F, Malier Y. Engineering properties of very high performance
concrete. In: Malier Y, editors. High performance concrete: From material to
structure. London: E & FN Spon, Chapman and Hall; 1992.
[45] Abdul Razak H, Wong HS. Re-evaluation of strength and stiffness relationships
for high-strength concrete. Asian J Civ Eng (Build Hous) 2004;5:85–99.
[46] Kockal NU, Ozturan T. Strength and elastic properties of structural lightweight
concretes. Mater Des 2011;32. 2396-2396.
[47] Caldarone MA. High-strength concrete: a practical guide. Taylor & Francis;
2009 [First published in the USA and Canada].
[48] Parrott LJ. A literature review of high strength concrete properties. British
Cement Association, ISBN 0 7210 13724, 1988.
[49] CEB-FIP model code for concrete structures. Evaluation of the time dependent
behavior of concrete. Bulletin d’ information No. 199, Comite Europe du Beton/
Federation Internationale de Precontrainte, Lausanne, 1991.
[50] ACI Committee 318. Building code requirements for structural concrete (ACI
318-99) and commentary (318R-99). Farmington Hills, MI: American Concrete
Institute; 1999.
[51] Oluokun FA. Prediction of concrete tensile strength from compressive
strength: evaluation of existing relations for normal weight concrete. ACI
Mater J 1991;88(3):302–9.
[52] Carino NJ, Lew HS. Re-examination of the relation between splitting tensile
and compressive strength of normal weight concrete. ACI J 1982;79(3):214–9.
[53] Shah SP, Ahmad SH. Structural properties of high-strength concrete and its
implications for precast prestressed concrete. J Prestr Concr Inst
1985;30(6):92–119.
[54] Arioglu N, Girgin ZC, Arioglu E. Evaluation of ratio between splitting tensile
strength and compressive strength for concrete up to 120 MPa and its
application in strength criterion. ACI Mater J 2006;103(1):18–24.
[55] Oluokun FA, Burdette EG, Deatherage JH. Splitting tensile strength and
compressive strength relationships at early ages. ACI Mater J
1991;88(2):115–21.
[56] Bhanja S, Sengupta B. Influence of silica fume on the tensile strength of
concrete. Cem Concr Res 2005;35:743–7.
[57] Khan MI, Lynsdale CJ. Strength, permeability, and carbonation of high-
performance concrete. Cem Concr Res 2002;32:123–31.
[58] Tazawa E, editor. Autogenous shrinkage of concrete. London: E&FN Spon;
1999.