Construction and Building Materials 27 (2012) 439–449

Contents lists available at ScienceDirect

Construction and Building Materials

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

Use of recycled concrete aggregate in fly-ash concrete

Mukesh Limbachiya, Mohammed Seddik Meddah ⇑, Youssef Ouchagour
School of Civil Engineering & Construction, Kingston University, London KT1 2EE, UK

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

Article history: Nowadays, environmentally friendly building is becoming a crucial issue in construction industry. The
Received 15 October 2010 course towards sustainable concrete involves mainly minimizing the environmental impact of concrete
Received in revised form 28 June 2011 production by substituting virgin mineral materials by recycled ones as well as reducing the global
Accepted 18 July 2011
CO2 emissions. The approach adopted here includes a large substitution of natural coarse aggregates
Available online 9 September 2011
(NA) by recycled concrete aggregates (RCA) obtained from crushed concrete debris, as well as the use
of 30% fly ash (FA) as a partial substitute of Portland cement for FA concrete production.
Keywords:
Previous study by the authors has revealed the potential of using coarse RCA to produce concrete with a
Carbonation
Drying shrinkage
similar 28-day design strength to that obtained when using natural aggregates. This paper discusses the
Durability effect of both partial and full replacement of natural coarse aggregates by coarse RCA in a fly ash concrete.
Fly ash Engineering properties and durability performance have been examined on both concrete types (Portland
Mechanical properties cement and fly ash) for mixes designed with various proportions of the RCA (0%, 30%, 50% and 100%) by
Portland cement mass. The results obtained showed that while embedding high amount of the RCA could lower the resis-
Permeability tance to chloride penetration and carbonation of concrete still comparable design strength to that of the
Recycled concrete aggregates control mix might be achieved.
Sustainable concrete
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction tonnes of construction and demolition residues are generated in

the UK; around 60 million tonnes of this is derived from concrete.
Concrete industry, especially Portland cement manufacture is Therefore, as concrete is still the material the most used in civil
known to be a heavy contributor to the environmental damage and industrial infrastructure and is also the major absorbing of nat-
and CO2 emissions. The less cement and natural aggregates that ural mineral resources, recycling rubble concrete gains importance.
are used in concrete production, the lower the impact the concrete It preserves natural resources and eliminates the need for disposal
industry has on the environment. Using various types of supple- by using the demolished concrete as an alternative aggregates for
mentary cementing materials (SCMs), especially FA as a cement new concrete production.
replacement could result in a substantial contribution to reduce Embedding the maximum possible amount of recycled materials
the overall CO2 footprint of the final concrete product. Over the in concrete matrix is the most effective and a promising policy
years, the use of FA in concrete production has become a common toward sustainable concrete material. In fact, a sustainable concrete
practice worldwide not only to reduce environmental charges but design includes minimizing the global CO2 released and energy con-
also due to the several benefits. The use of FA in concrete has pro- sumed to produce concrete as well as the various components
ven to improve workability and long term strength, reduce perme- needed. Indeed, extracting virgin aggregates is causing huge dam-
ability, minimize risk of alkali silica reaction, lowering heat of age to the environment and considerable energy is required for both
hydration in mass concrete, and enhancing durability performance extraction as well as crushing processes. Thus, a growing interest in
(resistance to chloride and sulphate attack) [1–4]. substituting natural aggregates with alternative recycled aggre-
Aggregates, in terms of volume, are the major component of con- gates derived from different constructions and demolitions wastes.
crete and may have significant effect on both engineering properties On the other hand, aggregate in fact, is known to play a substan-
and the final cost of concrete mixture. Moreover, natural resources tial role in determining workability, strength, dimensional stabil-
remarkably decline due to extensive use generated by high demand ity, and durability of the concrete. Due to their bonded mortar,
of new buildings and constructions. Every year, more than 165 mil- recycled concrete aggregates have a lower specific gravity and a
lion tonnes of natural aggregates are used in different civil and higher water absorption capacity compared to natural aggregates.
industrial constructions. Meanwhile, approximately 109 million The compressive strength varies with the compressive strength of
the old concrete and the water-cementing materials ratio (w/c) of
the new concrete. While recycling old concrete into aggregate is a
⇑ Corresponding author. Tel.: +44 (0)20 8417 2179. relatively simple process which involves breaking, removing, and
E-mail address: [email protected] (M.S. Meddah). crushing existing concrete into a material with a specified size

0950-0618/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.conbuildmat.2011.07.023
440 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449

and quality; the properties of concrete made with RCA are strongly Table 1
dependent on the quality of the recycled materials used as well as Chemical composition of Portland cement and fly ash used in this study.

the primary concrete crushed. Constituents Percentage (%)

Although the potential for the use of coarse RCA has now been Portland cement Fly ash
widely acknowledged and promoted, however RCA shall conform
SiO2 21.40 50.4
to the requirements specified in BS 8500-2 and the resulted com- Al2O3 4.70 28
posites with RCA shall perform quite similarly to NA concrete. Lack Fe2O3 2.70 9
of widespread reliable data on RCA aggregate characteristics and CaO 65.20 6
its influence on concrete performance can restrict it use to full MgO 1.00 1.5
SO3 2.90 0.4
potential. For durable RCA concrete design, a wide range of test
K2O 0.64 2.5
data are need on concrete made with various cements and combi- Na2O 0.13 0.90
nations, and different replacement levels of RCA. LOI 0.90 3.8
The use of various industrial by-products and recycled materials Density, g/cm3 3.14 2.4
Fineness, m2/kg 350 280
[5,6] offers multiple environmental advantages by offering poten-
tial diversion of useful materials from the waste streams, reducing
the energy investment in processing virgin materials, conserving
natural resources, and allaying pollution. Extensive research work Table 2
has already been performed on the use of fly ash [7–15] and other Mix proportions (for 1 m3 concrete).

supplementary cementitious materials to enhance sustainability Mix code RCA (%) Component (kg/m3)
and durability of concrete material; whereas, most of the existing PC FA Water Aggregates w/c
work is mainly carried out on concrete with natural aggregates.
NA RCA Sand
The current investigation attempts to strengthen the concept of sus-
tainability in concrete industry by combining the use of FA with var- PC20 0 275 0 180 1260 0 625 0.66
30 275 0 180 882 378 625 0.66
ious proportions of coarse RCA in concrete. 50 295 0 180 635 635 595 0.61
Previous research by the authors [16,17] has proven that RCA 100 310 0 180 0 1240 610 0.58
could successfully be used as substitute of natural aggregates to pro- PCFA20 0 215 90 160 1311 0 589 0.52
duce normal and high strength concrete, meeting required design 30 215 90 160 918 393 589 0.52
strength and desire durability performance. The results led to a gen- 50 230 100 160 657 657 562 0.48
eral conclusion that up to 30% coarse RCA, used as natural aggregate 100 245 105 160 0 1280 575 0.45
replacement, has no negative effect on fresh and hardened concrete PC30 0 330 0 180 1245 0 585 0.55
properties of Portland cement and silica fume concretes. 30 330 0 180 872 373 585 0.55
50 355 0 180 623 623 560 0.51
The reported study was undertaken to examine suitability of
100 372 0 180 0 1252 536 0.48
coarse RCA for use in fly ash concrete, produced using a single
PCFA30 0 260 110 160 1285 0 550 0.43
batch FA conforming to BS EN 450-2; 2005 as PC replacement.
30 260 110 160 900 386 550 0.43
The fly ash cement concrete used in this research therefore was a 50 280 120 160 641 641 523 0.40
combination of PC and 30% FA by mass, as PC replacement, com- 100 295 126 160 0 1267 517 0.38
bined in the concrete mixer as per BS 8500-2; 2002. RCA concrete PC35 0 355 0 180 1245 0 560 0.50
mixes were proportioned using natural (gravel and sand) and recy- 30 355 0 180 872 373 560 0.50
cled aggregate blends with 30%, 50% and 100% coarse RCA. The PC 50 385 0 180 613 613 550 0.47
concrete mixes were proportioned for 28 day design cube strength 100 409 0 180 0 1226 525 0.44

of 20, 30 and 35 MPa. The mixes were tested for basic mechanical PCFA35 0 280 120 160 1282 0 523 0.40
and engineering properties (compressive and flexural strengths, 30 280 120 160 898 384 523 0.40
50 300 130 160 631 631 514 0.37
elastic modulus and drying shrinkage) as well as durability perfor- 100 322 138 160 0 1257 488 0.34
mance in aggressive environments (sulphate attack, carbonation
and chloride ions).

2. Experimental testing programme

to be coarser, porous and rougher compared to natural gravel. The grading of RCA
obtained was within the current limits for crushed-rock aggregate to BS 882
2.1. Materials and concrete mix design
[16,17]. The RCA obtained has a density of 2.43, a water absorption capacity of
5.3%, and a crushing rate value of 23.4%.
2.1.1. Binders
Both PC and FA concrete mixes were proportioned following a conventional mix
A Portland cement (PC) CEM I 42.5 N confirming to BS EN 197-1 was used
design method, and are given in Table 2. In this, proportions of fine aggregate were
throughout this study for PC and FA concretes production. A single source of fly
established based on the percentage passing a 600 lm BS sieve and the average rela-
ash conforming to BS EN 450-2; 2005 was used, as 30% replacement of PC in all
tive density of aggregates. For PCFA concrete, total cement, free water and coarse
FA concrete production. Physical properties and chemical composition of Portland
aggregate contents were kept the same as the corresponding PC concrete mixes. This
cement and fly ash are given in Table 1.
mix contained 30% FA level by mass of cement. Whilst RCA concrete mixes were pro-
portioned using natural and recycled concrete aggregate blends with 30%, 50% and
2.1.2. Aggregates 100% coarse RCA used as NA substitute. All concrete mixes were prepared in a labora-
Natural river sand having a fineness modulus of 2.6 and a specific gravity of 2.68 tory pan mixer. Immediately after mixing, fresh concrete properties (slump and com-
was used as fine aggregates. Natural gravel with a maximum size of 20 mm and a spe- pacting factor) were determined. Batching and mixing of all the mixes were
cific gravity of 2.60 was employed as coarse aggregates. The water absorption of nat- performed according to the current procedures described in BS 1881: Part 125: 1986.
ural coarse and fine aggregates was 0.61% and 0.81%, respectively. It is worth
mentioning that no chemical admixtures were used for all concrete mixtures studied. 2.2. Concrete specimens preparation and curing regimes
The RCA used in this study was graded 20–5 mm aggregates, meeting the
requirements of BS 882, and were produced at a commercial recycling plant, com- Each mould was filled with concrete, consolidated on a vibrating table and cov-
prising primary jaw and secondary cone crushers and screens. Construction and ered with plastic film for the first 24 h to preserve the initial moisture conditions.
demolition waste, principally comprising concrete debris, were used to produce The samples were then demoulded and water-cured up to 28 days. Mixes are clas-
coarse RCA. The debris selected were clean and free from detrimental levels of sified into three grades according to their 28 day design compressive strength.
chemical impurities and harmful constituents. In general, RCA samples were found Labels C20, C30 and C35 refer to concrete with a 28 day design compressive
 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449 441

strength of 20, 30 and 35 MPa, respectively. For identification, PC denotes mixes reported. The specimens were coated with a bituminous coating on five of the six
made with Portland cement and mixes containing FA were labelled as PCFA. The faces, leaving only the top face uncoated to simulate unidirectional chloride ingress.
first numerical value after ‘PC’ and ‘PCFA’ represents the 28 day design strength. The top side of the cube was exposed to 1 mol of sodium chloride solution (58.4 g of
Depending on test procedure, three main curing regimes were adopted in this NaCl per litre of distilled water). Following exposure period of 42 days, the specimens
study. A conventional curing in water at 20 ± 2 °C for 28 days, curing under 95% were dried and drilled from the top side to the inner section at different depths. The
of RH at 20 ± 2 °C; and a standard dry conditions curing under 20 ± 2 °C and concrete specimen powder was collected from the exposed surface to chloride at
60 ± 5% of RH. regular intervals of 5 mm by drilling. The powder samples obtained was analysed
for acid soluble chloride by potentiometric titration as described in BS 1881-124.
2.3. Testing procedures Chloride concentrations are quoted in terms of percent by mass of cement.

Mechanical properties investigated included compressive and flexural strengths, 2.3.4. Carbonation
and elastic modulus. Compressive strength was determined on 100 mm cubes at 3, 7, Cubic specimens measuring 100 mm were used for carbonation test. The spec-
14, 28, 56, 91 and 365 days in accordance to EN 12390-3. Flexural strength under four- imens were first cured in water for 28 days and then conditioned under dry air for
point loading was measured at 28 days using beams of 100  100  500 mm, while 14 days. Thereafter, specimens were coated on all except the top side using a bitu-
elastic modulus was determined at 28 days on 150  300 mm cylinders according minous coating. The specimens were stored in a carbonation chamber at 20 ± 2 °C
to EN 1390-5 and BS EN 13412:2006, respectively. The ends of the cylinders were and 60 ± 5% of RH and 3.5% of CO2. Thus, top surface of test specimens was exposed
capped with a sulphur compound to ensure a flat and parallel surface and a better con- to a CO2 enriched atmosphere. After exposure periods of 2, 4, 8, 12 and 20 weeks,
tact with the loading device. All specimens (cubes, beams and cylinders) were cured in carbonation depth in concrete was measured by spraying a phenolphthalein indica-
water tank at 20 ± 2 °C until the time of testing which is carried out on three samples tor solution onto the surface of a split of freshly broken pieces of the concrete spec-
per mix and the average values were reported. imens. This turns pink when the concrete is alkaline (pH about 9) but remains
Drying shrinkage was measured on 75  75  300 mm prism specimens in one colourless where the concrete is carbonated. The carbonation depth indicated by
direction; using stainless steel pins fixed over 200 mm gauge length on two oppo- the colourless zone of the concrete, was then measured at five points along the ex-
site long sides. Test samples were stored in upright position under dry conditions posed surface and the average carbonation depth of the five measurements is
(20 ± 2 °C and 60 ± 5% of RH) and measurements were taken up to 90 days. The reported.
length change of the specimens was then measured using a comparator and peri-
odic measurements were taken.
To establish durability performance, a range of tests that include sulphate 3. Results and discussion
attack, initial surface absorption, chloride ions penetration, and carbonation were
carried out, as briefly described in the following section.
3.1. Workability
2.3.1. Sulphate attack
The resistance of concrete with and without the coarse RCA to sulphate attack
As abovementioned, embedding various industrial by-products
was determined by submerging concrete beams measuring 75  75  300 mm in a in cement paste as a partial replacement of Portland cement com-
0.3 g/l solution of Na2SO4 at 20 ± 2 °C cured under water for 28 days prior to expo- bined with a partial and/or full replacement of natural coarse
sure to the sulphate solution. The sodium sulphate solution was renewed on a aggregates by the RCA could affect the fresh concrete properties
monthly basis and the total test duration was 2 months. The length change mea-
of the concrete mix.
surements of specimens were taken throughout the whole period of exposure to
the sulphate solution (60 days). Slump and compacting factor values of each mix investigated
are given in Table 3. Whilst for both concretes (PC and PCFA), the
2.3.2. Initial surface absorption average slump values were generally within the limits specified
This test consists of the measurement of the rate at which water flows into the by the EN 206-1 for the designated concrete grade, the slump of
capillary pore network of concrete through a known surface area. Estimation of the fly ash concretes was generally lower than that of the PC concretes.
volume flow is obtained by measurement of the length of flow along a capillary of
This is due to the fact that fly ash concretes had higher binder con-
known dimension. The initial surface absorption of the various mixes was deter-
mined on 150 mm cubes after the concrete was water-cured for 28 days as per tent (approximately 10%) and lower water content (11%) compared
BS 1881-208. The specimens were oven-dried to constant weight prior to the test to the PC concretes.
and left to cool to the laboratory temperature. The contact area is defined by a plas- It can be observed that, increasing the amount of the coarse RCA
tic cell sealed onto the concrete surface and should not be less than 5000 mm2. in the PC mixes has generally resulted in quite a significant in-
Water is introduced into the cell via a connecting point and maintained at a head
of 200 mm using a filter funnel. A second connection point to the cap leads to a
crease of the slump value. A similar trend was also observed with
horizontal capillary tube. The connection to the reservoir is closed and the absorp- fly ash concrete mixes. This could be explained by the moisture
tion is measured by observing the movement of the end of the water line in the provided by the RCA added to the mix as these aggregates are more
capillary tube with an affixed scale at 10 min. The setup used is shown in Fig. 1. porous and have a higher water absorption capacity compared to
the natural coarse aggregate (NA). This initial moisture of the
2.3.3. Chloride ions penetration
Cubes of 150  150  150 mm cured, for 3 days, under wet curing regime (95% of
coarse RCA has contributed in the improvement of the initial val-
RH at 20 ± 2 °C) and then exposed to dry conditions for 14 days. Chloride ions pene- ues of the slump for the mixes made with the RCA compared to
tration was measured on three specimens per mix and the average values were the mixes made with the NA.

Table 3
Fresh concrete properties.

Mixture code RCA (%) Slump (mm) Compacting factor

PC PCFA PC PCFA
C20 0 70 40 0.97 0.96
30 70 20 0.98 0.95
50 60 50 0.98 0.97
100 40 30 0.96 0.91
C30 0 70 30 0.97 0.91
30 60 20 0.98 0.86
50 80 60 0.98 0.97
100 120 70 0.98 0.94
C35 0 40 10 0.95 0.83
30 20 30 0.95 0.92
50 30 20 0.98 0.86
100 50 20 0.95 0.92
Fig. 1. Initial surface absorption (ISA) measurement setup.
442 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449

3.2. Mechanical performance While the general trend suggest that, compressive strength of
both concrete types (PC and PCFA) decreases as the replacement le-
Compressive strength and elastic modulus of concrete are vel of natural aggregates by the RCA increases, still this tendency is
commonly considered in structural design, but for some purposes influenced by the w/c and the design strength of concrete. The
and specific project, the flexural or tensile strength is of interest, PC35 has showed quite a unique tendency with the highest com-
particularly to enhance resistance to cracking. Substituting natural pressive strength achieved by the concrete made with a solely
aggregates, at different replacement levels, by recycled aggregates RCA, whereas the lowest compressive strength was obtained by
would have influence on the mechanical behaviour of the concrete. the mix containing 30% of the RCA (Fig. 4). This inconsistency in
The results of the compressive and flexural strengths and modulus the behaviour of concrete mixes designed with the RCA could
of elasticity are presented in Figs. 2–9. It could be seen, as pointed mainly be attributed to the various intrinsic parameters of the
out above, that the targeted design strength was achieved for all RCA that are of a great influence on the concrete properties. The
the concrete mixes made with coarse RCA. amount and thickness of old mortar attached on the RCA are one

60
40 0% PCFA20

Compressive strength (MPa)

0% PC20 30%
Compressive strength (MPa)

50
35
30% 50%
30 50% 40 100%

25 100%
30
20

15 20

10 10
5
0
0 3 7 14 28 56 90 180 365
3 7 14 28 56 90 180 365 Age (days)
Age (days)
Fig. 5. Compressive strength versus time for PCFA20 concrete.
Fig. 2. Compressive strength versus time for PC20 concrete.

70
60
0% PCFA30
Compressive strength (MPa)

0% PC30 60
Compressive strength (MPa)

30%

50 30% 50%
50
50% 100%
40 100% 40

30 30

20 20

10 10

0 0
3 7 14 28 56 90 180 365 3 7 14 28 56 90 180 365
Age (days) Age (days)

Fig. 3. Compressive strength versus time for PC30 concrete. Fig. 6. Compressive strength versus time for PCFA30 concrete.

70
0% PCFA35
Compressive strength (MPa)

70 60 30%
Compressive strength (MPa)

0% PC35
60 50%
30% 50
100%
50 50%
40
100%
40
30
30
20
20

10 10

0 0
3 7 14 28 56 90 180 365 3 7 14 28 56 90 180 365
Age (days) Age (days)

Fig. 4. Compressive strength versus time for PC35 concrete. Fig. 7. Compressive strength versus time for PCFA35 concrete.
 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449 443

7 On the whole, the use of up to 30% coarse RCA, as natural aggre-

PC20 PC30 PC35 PCFA20 PCFA30 PCFA35
gate substitute, could be considered as the optimum content, as
6
Flexural strength (MPa)

beyond this level, a negative effect on compressive and flexural

5 strengths as well as the elastic modulus was observed.

3
3.3. Drying shrinkage

2 Results of drying shrinkage and swelling of the PC and FA con-

cretes designed with and without the coarse RCA are presented in
1
Table 4. As it could be seen, all fly ash concretes have exhibited a
0 lower magnitude of drying shrinkage strains while they showed
0 30 50 100 a higher swelling compared to their corresponding PC concretes.
RCA content (%) In fact, ashes are known by their capability to reduce shrinkage
strains whether due to an internal drying (self-desiccation) or an
Fig. 8. Effect of the RCA content on the 28 day flexural strength. external drying and hence, a significant reduce of the internal
stresses could be achieved.
of the major factors affecting the final concrete product perfor- As is well recognized, water content in concrete mixture is the
mance. In fact, the RCA has also been used in concrete as received largest contributor to drying shrinkage. The lubricating action of fly
from the manufacturer and may have different initial moisture ash would reduce the amount of water needed and consequently,
which could affect the cement hydration process. High initial mois- the drying shrinkage magnitude. In addition, ashes are known for
ture may lower the early-age compressive strength but it would be their high capacity of retention of water [18,19] during the earliest
beneficial for a continuous wet conditions and long term cement stage of cement hydration which could contribute to minimize the
hydration, especially for PCFA concrete and hence, strength gain. water available in the pore network for any external drying.
The results obtained for the flexural strength (Fig. 8) and the The inclusion of 30% FA in NA concrete with various w/c
elastic modulus (Fig. 9) of concretes made with the RCA follow a resulted in a reduction in the drying shrinkage magnitude of up
quite similar trend to that observed in the compressive strength to 37% when compared to the PC concretes. On the other hand,
development. For a given concrete mix and at a given age, increas- embedding different proportions of the RCA in both concretes
ing the amount of the RCA in the mixture has led to a slight de- (PC and PCFA) resulted in a significant increase of the drying
crease of both the flexural strength and the elastic modulus. In shrinkage. This increase was proportional to the amount of the
fact, the flexural strength loss could be attributed to the weak RCA incorporated. The higher the replacement level of RCA, the
bonding strength between the hydrated cement paste and the higher the magnitude of drying shrinkage exhibited. Whereas,
RCA due to the old mortar attached on it. Additionally, elastic mod- the effect of the increased amount of RCA in concrete on drying
ulus of concrete is known to be intimately linked to the aggregate shrinkage strains was more pronounced in the PC concretes com-
skeleton rather than to the cement paste. The low values of the pared to the fly ash concretes. It should also be noted that no sig-
elastic modulus of concrete made with the RCA might be explained nificant increase of shrinkage strains was observed when using 30%
by corresponding low strength characteristics of the RCA when RCA in both concrete types. This may support the trend previously
compared to the NA. Moreover, the natural gravel used in this observed that 30% replacement level would be the optimum con-
study may also contribute in reducing the elastic modulus of con- tent of the RCA in terms of concrete engineering properties.
crete as compared to crushed aggregate. In fact, the increase in shrinkage strains of concrete made with
For fly ash concretes, as expected and shown in Figs. 5–7, RCA could be attributed to the supplementary water provided by
extending the humid curing period evidently benefits compressive these recycled aggregate used. In the meantime, the old mortar at-
strength development, which is attributed to the fact that the poz- tached on the RCA may also contribute in the increased shrinkage
zolanic reaction between FA and portlandite is slow; and hence, strains as the volume of cement paste increases as well. The lowest
the beneficial effects were exhibited at later ages. The results re- shrinkage strains magnitude was observed with the control PCFA
vealed that for the three concrete grades examined and beyond whilst the largest shrinkage was obtained with the PC concrete
14 days curing period, the compressive strength of fly ash con- made with solely RCA. Obviously, reducing drying shrinkage by
cretes was higher than that of the PC concretes.
Table 4
25 Drying shrinkage and expansion of concrete specimens after 91 days of curing.
PC20 PC30 PC35 PCFA20 PCFA30 PCFA35
Mixture code RCA (%) Strains (le)
Elastic modulus (Gpa)

20 Drying shrinkage Expansion

PC PCFA PC PCFA
15 C20 0 290 190 125 145
30 320 260 110 190
50 450 250 100 190
10 100 650 450 60 220
C30 0 340 215 100 90
5 30 340 240 120 145
50 520 430 80 110
100 630 550 80 135
0
C35 0 280 195 120 140
0 30 50 100
30 320 250 130 140
RCA content (%) 50 425 425 130 65
100 810 695 140 100
Fig. 9. Effect of the RCA content on the 28 day elastic modulus.
444 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449

the incorporation of FA in concrete made with RCA may subse- due to the presence of pozzolan compared to the PC concretes.
quently reduce the cracking tendency of such a concrete. Replacing 30% of PC by FA would decrease the reactive aluminates
(C3A) content in the PCFA concretes and hence, less ettringite
3.4. Sulphate attack formed which enhances the resistance to sulphate attack.
The use of FA could also bind the free lime of the hydrated
It is generally recognized that some chemicals inherent in the cement paste and prevent reaction with the sodium sulphate solu-
environment could be detrimental for the integrity and durability tion. Previous studies [2–4,21–23] have also pointed out a reduction
of concrete structures. Under sulphate environment, cement paste in the expansion exhibited due to sodium sulphate immersion when
undergoes deterioration resulting from destructive expansion, using blended cement with FA. Furthermore, while the incorpora-
spalling and softening. Soluble sulphates in soils, ground water, tion of 30% and 50% of RCA showed a slight effect on the expansion
and sewage can lead to severe deterioration of concrete unless it induced by sulphate attack, a solely use of RCA instead of natural
is properly designed to provide sulphate resistance commensurate aggregates has led to an accelerating expansion pattern for both
with the severity of the attack. concretes PC and PCFA, especially from the first week of exposure
Figs. 10 and 11 present the patterns of expansion induced by sul- and onward. It has also been reported that sulphate attack resis-
phate solution for the PC and PCFA concretes, respectively. It could tance of concrete containing FA depends on various parameters
be seen that while both concretes have shown a quite similar trend, related mainly to FA type used such as the content of reactive alu-
the PCFA mix with and without RCA performs slightly better than mina and lime. Obviously, longer term exposure to sulphate attack
the PC mix in terms of resistance to sulphate attack. Generally speak- is necessary to establish more realistic prediction of durability of
ing, for a given concrete mix containing FA, the total expansion concrete with coarse RCA while 2 months of exposure showed the
exhibited by sulphate attack was more or less 10 microstrains lower superior performance of the PCFA mixes compared to the PC mixes.
than that of the PC mixes. The fact that fly ash concrete induces less
internal pressure and destructive expansion than plain concrete was 3.5. Initial surface absorption (ISA)
reported in earlier research by several authors [2–4,20].
In fact, as is well known, sulphate attack mechanism includes The initial surface absorption test was basically designed to
mainly two-phased process. Sulphates combine with calcium evaluate the porosity of cement-based composite materials. For a
hydroxide generated during cement hydration to form calcium sul- given concrete mix at a given age, the ISA test could roughly indi-
phate (gypsum). Meanwhile, aluminates compound (C3A) from cate the degree of porosity as well as the connectivity of its pore
Portland cement reacts chemically with sulphates and calcium to system.
form ettringite. Both the formation of gypsum and ettringite expand The test results of the ISA of the concrete to water after ten min-
concrete mixture which generates internal pressure and lead to utes are shown in Fig. 12. It can be seen that the partial replacement
cracks formation and disruption in hardened cement paste. The of Portland cement by pozzolanic materials (FA) can effectively
lower expansion recorded in the PCFA concrete is essentially attrib- reduce the initial surface absorption. For the same concrete design
uted to the relatively low volume of monosulphate and portlandite strength, the fly ash concrete has shown lower ISA compared to
the PC concrete, especially for mixtures made with solely RCA. The
use of coarse RCA as a partial or full replacement of the NA has
90 resulted in significant increase of the ISA. The higher the replace-
PC20 ment levels of RCA, the more perceptible the increasing effect on ISA.
80
Generally, for both concrete types, the use of up to 30% coarse
70
Expansion ( x 10-6)

RCA showed no significant effect on the rate of ISA, whereas,

60
beyond this limit, the ISA increases sharply. A similar behaviour
50
was observed with the resistance to sulphate attack, chloride ions
40 ingress and carbonation. From the results presented in Fig. 12, it
NA
30 could also be concluded that, as expected, concrete strength is the
30% RCA
20
50% RCA
major factor governing its ISA. The higher the compressive strength
10 100% RCA of concrete, the lower is its porosity and hence, lower is the ISA. The
0 low ISA obtained with the fly ash concretes is strongly linked to the
0 10 20 30 40 50 60 pozzolanic reaction results and the refinement of the pore capillary
Immersion time (days) network which lower the water flow rate into concrete.

Fig. 10. Length change of PC20 concrete mixes exposed to sulphate solution.
1.2

1.0
80
PCFA20
ISA-10 ( ml/m 2/s)

70 PC20
0.8
Expansion ( x 10-6)

PC30
60
PC35
50 0.6
PCFA20
40 PCFA30
0.4
PCFA35
30
NA
20 0.2
30% RCA
10 50% RCA
100% RCA 0.0
0 0 20 40 60 80 100 120
0 10 20 30 40 50 60 RCA content (%)
Immersion time (days)
Fig. 12. Initial surface absorption (10 min) versus the RCA content for the various
Fig. 11. Length change of PCFA20 concrete mixes exposed to sulphate solution. mixes investigated.
 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449 445

3.6. Carbonation the carbonation process of the mixes made with the RCA compared
to the control mixes. Moreover, it is well recognized that carbon-
Carbonation is the process that leads to the formation of calcium ation of concrete occurs at a relative humidity from about 40% to
carbonate through the reaction of CO2 (atmospheric carbon dioxide 70%. Due to the high water absorption of the coarse RCA, the con-
or artificial) with some constituents of the concrete, (calcium trol concretes have generally lower moisture content compared to
hydroxide, silicates and aluminates), resulting in a reduction of their corresponding concretes designed with RCA. This may also
the pH of concrete from around 12 to below 9. Under such a value explain the low resistance to carbonation of concretes containing
of pH, the steel rebar become active and the corrosion process is coarse RCA.
initiated which affects the interface steel/concrete and the loading As the progress of carbonation process is governed by the pore
carrying capacity of reinforced concrete element. network characteristics, the results presented in Fig. 13 showed
The results of carbonation depth are presented in Fig. 13. It is that the carbonation becomes limited beyond 20 mm because of
clear that the depth of carbonation is proportional to the duration the increasing difficulty for carbon dioxide to penetrate into the
of exposure. The higher the exposure period, the higher the depth depth of the concrete pore network.
of carbonation is. The results have revealed that adding various For a given design strength, the carbonation depth of all the fly
proportions of coarse RCA as a partial replacement of natural ash concrete mixes was quite greater than the one of the PC con-
aggregates has resulted in a lower resistance to carbonation, espe- crete mixes. These results are in agreement with those reported
cially for the C30 and C35 concrete grades. While no significant in previous works [24–27]. Earlier researches have also found that
difference was observed in the carbonation depth of the C20 series increasing FA content in cementitious system results in a higher
between mixes with and without RCA, the gap between the control rate of carbonation compared to a plain cement mix [28,29]. Due
mix and the mixes containing various proportions of RCA is con- to the pozzolanic reaction between FA and portlandite, and the
siderable for the C30 and C35 series, particularly for the PCFA consumption of CH, the pH of the concrete decreased and leads
mixes. to a higher carbonation as compared to the PC concretes. On
It is also believed that the water stored in the pore system of the the other hand, as hydration proceeds in FA concrete, and due to
RCA and released throughout hydration process may contribute in the secondary C–S–H formed, both the total porosity as well as

25
30
Carbonation depth (mm)

C20 PC
Carbonation depth (mm)

C20 FA
20 25

20
15
15
10
NA
10 NA
30% RCA
5 30% RCA
50% RCA 5 50% RCA
100% RCA 100% RCA
0 0
0 4 8 12 16 20 0 4 8 12 16 20
Period of exposure (weeks) Period of exposure (Weeks)

20 20
Carbonation depth (mm)

C30 PC
Carbonation depth (mm)

C30 FA

15 15

10 10

NA NA
5 30% RCA 5 30% RCA
50% RCA 50% RCA
100% RCA 100% RCA
0 0
0 4 8 12 16 20 0 4 8 12 16 20
Period of exposure (weeks) Period of exposure (Weeks)

20
25
Carbonation depth (mm)

Carbonation depth (mm)

C35 FA
C35 PC
20 15

15
10
10
NA
NA
30% RCA 5 30% RCA
5 50% RCA
50% RCA
100% RCA 100% RCA
0 0
0 4 8 12 16 20 0 4 8 12 16 20
Period of exposure (weeks) Period of exposure (Weeks)

Fig. 13. Carbonation depth versus exposure period for the investigated concretes.
446 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449

the coarser pores volume decreased which results in a reduction of

(a)

Carbonation coefficient (mm/week 0.5)

7 PC 35
the carbonation rate.
In low strength concrete such as the C20 concrete grade, carbon- 6
ation is much more rapid as CO2 can permeate the cement matrix
more easily. However, for a relatively dense concrete such as the 5
C30 and C35, the carbonation rate and depth progress slowly
4
throughout the pore network. Obviously, it is expected that reducing
the w/c and using blended cement with FA could further reduce the 3
total porosity and refine coarser pores into smaller ones which
therefore, decreases the CO2 permeable pores. 2
As carbonation of concrete is dependent upon a number of fac-
tors linked to the mix design such as type and content of cement, 1
mix proportions and porosity of concrete, concentration of CO2 in
0
the environment and/or the exposure solution and relative humid- 0 5 10 15 20 25
ity; concrete made with the RCA that will be exposed to environ- Period of exposure, weeks
ment which may contain high concentration of carbon dioxide
should be designed accordingly. In fact, both carbonation and chlo-

Carbonation coefficient (mm/week 0.5)

PCFA 35
ride ions did not present any risk of deterioration if no reinforcing (b) 6

bar is present in concrete element.

5

3.7. Carbonation coefficient 4

Based on the test results of the carbonation depth of the various 3

concrete specimens, the values of corresponding carbonation coef-
ficient (K) have been calculated. 2
The carbonation coefficient K (Eq. (1)) is expressed according to
square-root theory, adopted by numerous researchers [25–27], 1
and is generally used to compare the carbonation resistance of con
crete. 0
pffiffi 0 5 10 15 20 25
Xc ¼ K t ð1Þ Period of exposure, weeks

where Xc is the tested carbonation depth (mm), t is the exposure Fig. 14. Relationship carbonation coefficient-exposure period for the investigated
duration (week) and K is the corresponding carbonation coefficient concretes.
(mm/week0.5).
Generally, it was observed that for a given RCA replacement le-
vel, the coefficient of carbonation tends to increase for both the PC 30
PC
and PCFA concretes following the same trend of the carbonation Xc = -0.7fc + 43.4
depth discussed above. The carbonation coefficient seems to be R² = 0.90 PCFA
25
Carbonation depth (mm)

proportional to the compressive strength of concrete. The lower Linear

the compressive strength of concrete, the higher the carbonation 20 (PC)
Linear
coefficient is. The C20 concrete grade, for both PC and PCFA mixes, (PCFA)
has exhibited a higher carbonation coefficient than the C30 and 15
C35 and its values drop from 6 to 3.7 (mm/week0.5) and 6.1 to
4.1 (mm/week0.5), respectively. 10
It could be seen in Fig. 14 that the carbonation depth varies lin-
early with the square-root of exposure time, and hence, the
5 Xc = -0.83fc+ 43.9
square-root law could be used to predict the carbonation depth of R² = 0.94
concrete exposed to carbon dioxide as has been stated by some
0
authors [30–32]. Based on the normal concentration of carbon diox- 20 30 40 50 60
ide in the atmosphere, it is supposed that 1 week-time exposure of Compressive strength (MPa)
concrete specimen in the carbonation chamber is somewhat equiv-
alent to 12 months exposition under natural environment. This Fig. 15. Relationship carbonation depth-compressive strength for the investigated
could provide a general prediction on the carbonation depth of con- concretes.
crete at long term.
Figs. 14 and 15 show the linear relationship between the 3.8. Resistance to chloride penetration
carbonation coefficients and depth of carbonation as well as the
carbonation coefficients and compressive strength of concrete. Chloride attack may be due to exposure to seawater, to de-icing
For all concretes investigated, Kc increases as the w/c is increased salts or to other environmental and industrial conditions. Chloride
because of the increase in concrete porosity and the decrease of ions penetrate in concrete by diffusion along water paths or open
the compressive strength. pores. Part of these chlorides can react with the cement hydrate
As aforementioned, the addition of fly ash in partial substitution products, mainly tricalcium-aluminates (C3A), forming stable
of Portland cement decreases the initial content of CaO in the ce- chloro-complexes. If the concentration of chloride ions exceeds
ment matrix and consequently, the amount of CH to be formed the amount that could react with some components of hydrated
as a result of cement hydration reaction. This will also reduce the cement paste, the remaining free chloride ions may lead to the ini-
pH of concrete and contribute to accelerate the rate of carbonation. tiation of corrosion process.
 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449 447

PC30 The chloride concentration–penetration depth profiles for the

(a) 1.6 RCA
0%
selected concretes exposed to chloride ions for 42 days are pre-
1.4 sented in Figs. 16 and 17, for both control and exposed concrete
Chloride content (%)
30%
1.2 50% specimens. As expected, the results showed that the chloride
100 concentration is high near to the exposed surface and it decreased
1.0
ASTM limit with penetration depth.
0.8 On the whole, as could be seen in Figs. 16 and 17, the chloride
0.6 concentration profiles for the investigated mixes is mainly
0.4 influenced by the mixture parameters including, design strength,
w/c, binder type and replacement level of natural coarse aggregate
0.2
by the RCA. For both concrete types (PC and PCFA), increasing the
0.0 content of RCA used as partial/full replacement of natural aggregates
0-5 5-10 10-15 15-20
has resulted in a substantial increase of the chloride concentration,
Depth (mm)
especially in the outer region (0–5 mm) as reported also in previous
studies [33,34]. In fact, the increase of chloride concentration in the
PC35 RCA
(b) 1.2 concrete made with various proportions of the RCA could be attrib-
0%
uted to the higher porosity and lower compressive strength of these
Chloride content (%)

1.0 30% concretes compared to the control concrete. At the same depth, the
50% higher the replacement level of the NA by the RCA, the higher the
0.8
100% chloride concentration found. Meanwhile, it seems that for all mixes
0.6 investigated, substituting 30% of natural aggregates by the RCA has
shown no significant effect on the chloride ions content compared
0.4
to the control mix and beyond this limit, concrete produced with
0.2 RCA become more vulnerable to chloride ions ingress.
On the other hand, for the same concrete design strength and at
0.0 the same penetration depth, fly ash concrete tends generally, to
0-5 5-10 10-15 15-20
slightly increase the chloride content. This effect might be attrib-
Depth (mm) uted to the relative changes in intrinsic diffusivity and chloride
binding capacity exhibited by the binder compositions used
Fig. 16. Chloride content versus depth penetration for the PC30 (a) and PC35 (b).
(PCFA). As pointed out above, part of the chlorides ions in the solu-
tion can react with some component of the cement such as the C3A,
while blended cement contains less C3A than the PC concrete.
PCFA30
(a) 1.0 RCA Therefore, the binding capacity of the fly ash concrete to chloride
0.9 0% tends to decrease and consequently, more free chloride ions re-
Chloride content (%)

0.8 30% main in the pore solution as also reported by Kayyali and Qasrawi
0.7 50% [35]. Nevertheless, it is also worth mentioning that FA contains
0.6 100% higher proportions of active alumina compared to Portland cement
0.5
(around 30% by weight of Al2O3 in FA cf. 5% in PC), which is capable
0.4
of binding and immobilizing chloride ions in solution.
0.3
Moreover, FA is characterized by a slow reaction with portlan-
0.2
0.1
dite to produce a secondary C–S–H that fill the coarser capillary
0.0 pores and refine the pore network which leads to a low porosity
0-5 5-10 10-15 15-20 and a disconnected pore network and hence, improve resistance
Depth (mm) to permeability characteristic. As a result, it is expected that at long
term, the fly ash concretes with and without RCA may perform
(b) 1.8 RCA
PCFA35 0%
1.6
Chloride content (%)

30%
1.4
50%
1.2 100% Durability performance
1.0
ASTM limit
Concrete properties

0.8
0.6
0.4
0.2 Drying shrinkage

0.0
0-5 5-10 10-15 15-20
Mechanical properties
Depth (mm)

Fig. 17. Chloride content versus depth penetration for the PCFA30 (a) and PCFA35
(b).
0 30 50 100
Proportions of RCA
The chloride concentration–penetration depth profiles of the PC
and blended cement with FA produced with natural aggregates are Fig. 18. General relationships between concrete properties and the content of the
compared with those containing various proportions of the RCA. RCA.
448 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449

much better than the PC concretes in terms of resistance to chlo- and recycled concrete aggregate using a proper binder type and
ride ions penetration. This was also confirmed in earlier researches w/c could significantly promote sustainability and enhance
by several authors [1,36]. durability in concrete industry.
In the practical use of cement-based materials, not all concrete
Acknowledgement
mixes are designed to be used in reinforced concrete structural
elements. A large part of concrete, especially normal strength
The authors would like to acknowledge the Day Group Ltd., UK
concretes are usually used in nonaggressive environment under
for partial funding of the reported work and supplying recycled
normal exposure conditions for which resistance to aggressive
concrete aggregate sample for the test programme.
environment is not required.
Fig. 18 summarizes the relationship between concrete proper-
ties (Portland cement and PC/FA ash blended cement concrete) References
and the replacement level of the NA by the RCA. As the replace-
[1] Zuquan Jin, Wei Sun, Yunsheng Zhang, Jinyang Jiang, Jianzhong Lai. Interaction
ment proportion increases, the compressive and flexural strengths,
between sulfate and chloride solution attack of concretes with and without fly
and modulus of elasticity decrease as well as resistance to aggres- ash. Cem Concr Res 2007;37(8):1223–32.
sive environment; and on the other hand, drying shrinkage strain [2] Shi C, Stegemann JA. Acid corrosion resistance of different cementing
increases. According to the properties derived from concrete de- materials. Cem Concr Res 2000;30(5):803–8.
[3] Chindaprasirt P, Homwuttiwong S, Sirivivatnanon V. Influence of fly ash
signed with RCA, it is clearly established that replacing 30% of nat- fineness on strength, drying shrinkage and sulphate resistance of blended
ural aggregates by coarse RCA in concrete produced with either PC cement mortar. Cem Concr Res 2004;34(7):1087–92.
or FA has no major negative effect on concrete performance. Fur- [4] Chindaprasirt P, Kanchanda P, Sathonsaowaphak A, Cao HT. Sulfate resistance
of blended cements containing fly ash and rice husk ash. Constr Build Mater
thermore, due to their high water absorption capacity, the RCA 2007;21(6):1356–61.
could be used as an internal curing agent when added to concrete [5] Meddah MS, Bencheikh M. Properties of concrete reinforced with different kinds
mixture in a saturated state, especially for fly ash concrete which of industrial waste fibre materials. Constr Build Mater 2009;23(10):
3196–205.
needs a prolonged wet curing for an improved pozzolanic reaction. [6] Suzuki M, Meddah MS, Sato R. Use of porous ceramic waste aggregates for
The results of current study combined with previous data by the internal curing of high-performance concrete. Cem Concr Res 2009;39(5):
authors [16,17] as well as finding by several researchers prove the 373–81.
[7] Mehta PK. Influence of fly ash characteristics on the strength of Portland-fly
efficiency of using coarse RCA in concrete production as a partial ash mixtures. Cem Concr Res 1985;15(4):669–74.
substitute of natural aggregates. It seems also that there is a gen- [8] Bilodeau A, Sivasundaram V, Painter KE, Mahotra VM. Durability of concrete
eral consensus regarding the optimum amount of coarse RCA incorporating high volumes of fly ash from sources in the US. ACI Mater J
1994;91(1):3–12.
which is set to approximately 30% as a maximum replacement le-
[9] Ravina Dan, Mehta PK. Compressive strength of low cement/high fly ash
vel for concrete with a proper performance. concrete. Cem Concr Res 1988;18(4):571–83.
[10] Malhotra VM. Durability of concrete incorporating high-volume of low-
calcium (ASTM class F) fly ash. Cem Concr Compos 1990;12(4):271–7.
4. Conclusions [11] Mehta PK. Reducing the environmental impact of concrete. ACI Concr Int
2001;23(10):61–6.
[12] Malhotra VM. Making concrete greener with fly ash. ACI Concr Int
Overall, the results of this study could provide the construction 1999;21(5):61–6.
industry with technical information on marketable products FA [13] Malhotra VM. High-performance high-volume fly ash concrete. ACI Concr Int
2002;24(7):30–4.
and RCA, which has a key role to play in continuing implementa- [14] Malhotra VM. Introduction: sustainable development and concrete technology.
tion of sustainability within the sector. ACI Concr Int 2002;24(7):22.
[15] Seiki S, Meddah MS, Nukushina T, Sato R. Improvement of fly ash concrete
properties by internal curing using porous ceramic waste aggregate.
1. The general trends observed indicate that 30% coarse RCA has
Performance, innovations and structural implications. In: Proceedings of the
no major negative effect on a range of mechanical and durabil- 4th international conference on construction materials, August 24th–26th,
ity properties of PC and PC/30% FA concrete. However, reduction Nagoya, Japan; 2009. p. 491–6.
[16] Limbachiya M, Meddah MS, Ouchagour Y. Performance of Portland/silica fume
in resistance to carbonation and chloride ions penetration as
cement concrete produced with recycled concrete aggregate. ACI Mater J,
well as sulphate attack was observed with increase in RCA pro- submitted for publication [M-2010-320R2].
portions beyond 30%. Thus, 30% has been set as the optimum [17] Ouchagour Y. Sustainability of recycled concrete aggregate for use in binary
level for the coarse RCA. cement concrete. PhD thesis, Kingston University London, UK; 2007. p. 340.
[18] Lee HK, Lee KM, Kim BG. Autogenous shrinkage of high-performance concrete
2. Both carbonation depth and coefficient of carbonation of con- containing fly ash. Mag Concr Res 2003;55(6):507–15.
crete increase with increase in RCA content, w/c and the expo- [19] Meddah MS, Tagnit-Hamou A. Pore structure of concrete with mineral admixtures
sure period. The results tend to indicate that combining fly ash and its effect on self-desiccation shrinkage. ACI Mater J 2009 ;106(3):241–50.
[20] Torii K, Kawamura M. Effects of fly ash and silica fume on the resistance of mortar
with coarse RCA in concrete may enhance the long term resis- to sulfuric acid and sulphate attack. Cem Concr Res 1994;24(2):361–70.
tance to carbonation. [21] Shehata MH, Adhikari G, Radomski S. Long-term durability of blended cement
3. A reliable linear relationship between carbonation coefficient against sulphate attack. ACI Mater J 2008;105(6):594–602.
[22] Thomas MDA, Shehata MH, Shashiprakash SG, Hopkins DS, Cail K. Use of
and time of exposure as well as between the carbonation coef- ternary cementitious systems containing silica fume and fly ash in concrete.
ficient and compressive strength was observed. As expected, Cem Concr Res 1999;29(8):1207–14.
higher the compressive strength of concrete, the lower is the [23] Al-Dulaijan SU, Maslehuddin M, Al-Zahrani MM, Sharif AM, Shameem M,
Ibrahim M. Sulfate resistance of plain and blended cements exposed to varying
carbonation coefficient was noted.
concentrations of sodium sulfate. Cem Concr Compos 2003;25(4–5):429–37.
4. The use of fly ash in RCA concrete may significantly improve the [24] Khunthongkeaw J, Tangtermsirikul S, Leelawat T. A study on carbonation
resistance to chloride ingress and resistance to sulphate erosion depth prediction for fly ash concrete. Constr Build Mater 2006;20(9):744–53.
[25] Papadakis VG, Fardis MN, Vayenas CG. Hydration and carbonation of
and consequently, has potential to enhance the service life of
pozzolanic cements. ACI Mater J 1992;89(2):119–30.
concrete exposed to such aggressive environment, providing [26] Sideris KK, Sawa AE, Papayianni J. Sulfate resistance and carbonation of plain
suitable proportion of FA and low w/c is used. and blended cements. Cem Concr Compos 2006;28(1):47–56.
5. It is clear that the main requirement for concrete quality and [27] Khan MI, Lynsdale CJ. Strength, permeability and carbonation of high-
performance concrete. Cem Concr Res 2002;32(1):123–31.
durability in aggressive environment is improved resistance to [28] Byfors K. Carbonation of concrete with silica fume and fly ash. J Nord Concr Res
permeability (low and finer porosity) which could enhance its 1985;4:26–35.
resistance to the penetration of harmful materials and chemical [29] Maage M. Carbonation in concrete made of blended cements. In: McCarthy GJ,
Maage M, Glasser FP, Roy DM, editors. Symposium of materials research
attack. Concrete designed with various industrial sub-products society, vol. 65, Boston, USA; 1986. p. 193–7.
 M. Limbachiya et al. / Construction and Building Materials 27 (2012) 439–449 449

[30] Sulapha P, Wong SF, Wee TH, Swaddiwudhipong S. Carbonation of concrete [34] Berndt ML. Properties of sustainable concrete containing fly ash, slag and
containing mineral admixtures. J Mater Civ Eng 2003;15(2):134–43. recycled concrete aggregate. Constr Build Mater 2009;23(7):2606–13.
[31] Wee TH, Suryavanshi AK, Logendran D. Pore structure controlling the [35] Kayyali OA, Qasrawi MSh. Chloride binding capacity in cement–fly-ash pastes.
carbonation of a hardened cement matrix blended with mineral admixture. J Mater Civ Eng 1992;4(1):16–26.
Adv Cem Res 1999;11(2):81–95. [36] Bai J, Wild S, Sabir BB. Chloride ingress and strength loss in concrete with
[32] Kritsada S, Lutz F. Carbonation rates of concretes containing high volume of different PC–PFA–MK binder compositions exposed to synthetic seawater.
pozzolanic materials. Cem Concr Res 2007;37(12):1647–53. Cem Concr Res 2003;33(3):353–62.
[33] Otsuki N, Miyazato S, Yodsudjai W. Influence of recycled aggregate on
interfacial transition zone, strength, chloride penetration and carbonation of
concrete. J Mater Civ Eng 2003;15(5):443–51.