Академический Документы
Профессиональный Документы
Культура Документы
net/publication/273759752
CITATIONS READS
52 1,633
3 authors:
Ravindra Dhir
University of Birmingham/ Trinity of College Dublin/ University of Dundee
483 PUBLICATIONS 4,718 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
external factors affecting building project performance during construction stage? View project
All content following this page was uploaded by R.V. Silva on 23 October 2017.
1
PhD student in Civil Engineering, ICIST, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco
Pais, 1049-001 Lisbon, Portugal; e-mail: rvpssilva@gmail.com
2
Full Professor, ICIST, DECivil, Instituto Superior Técnico, Universidade de Lisboa, Av. Rovisco Pais, 1049-
001 Lisbon, Portugal; Phone: (351) 218 419 709, e-mail: jb@civil.ist.utl.pt, Corresponding author
3
Professor of Concrete Engineering, School of Civil Engineering, University of Birmingham, B15 2TT, UK;
Phone: +44 121 4145075, email: r.k.dhir@bham.ac.uk
4
Director, Applying Concrete Knowledge, 1A Blakeney Avenue, Birmingham, B17 8AP, UK; Phone: +44 121
4278108/8187, email: r.k.dhir@bham.ac.uk
ABSTRACT
This paper provides a systematic literature review, based on the identification, appraisal, se-
lection and synthesis of publications relating to the effect of incorporating recycled aggre-
gates, sourced from construction and demolition wastes, on the tensile strength of concrete. It
identifies various influencing aspects related to the use of recycled aggregates such as re-
placement level, size and origin, as well as mixing procedure, chemical admixtures, additions
and strength development over time. In this paper, estimated values, using the Eurocode 2
method to determine the tensile strength over time, were compared with the actual measured
values, showing little correlation and therefore new coefficients are proposed. This paper also
presents the relationship between the tensile and compressive strengths according to Euro-
code 2. The results suggest that, regardless of the replacement level, type, and quality of the
recycled aggregate used, the resulting recycled concrete tends to exhibit a similar relationship
Keywords: Recycled aggregates, construction and demolition waste, tensile strength, concrete.
1
1 Introduction
Vast amounts of waste materials are produced by the construction and demolition industry eve-
ry year [1]. The volume of these materials has reached an unacceptable point for environmen-
tal, economic and social reasons. These issues may be addressed by means of more proactive
approaches, which include recovery, reuse and recycling techniques, and facilities.
It takes many years for a waste management system to develop into a sustainable, reliable,
skilful and marketable industry, encouraging the reuse and recycling of components and ma-
terials. It is necessary that all parties involved (i.e. clients, contractors, planners and manufac-
tures) play their role in achieving a more sustainable approach. This can be done by extend-
ing the life cycle of materials, components and resources. The use of recycled materials in
high rather than low-grade applications must also be a priority in the near future. In order to
be successful in this approach, the correct choice of materials, recycling procedures and man-
1.1 Background
The characteristics of construction and demolition waste (CDW), such as quantity, potential
quality and true cost, are often overlooked. This lack of awareness leads to the disposal of
great amounts of potentially recyclable materials, only to be replaced with similar conven-
tional components. Applying recycling and beneficiation procedures to CDW is considered one
of the most effective measures to reduce the ecological footprint caused by the construction in-
Professionals in the construction and demolition industry still often see the concept of selec-
tive demolition as being of debateable economic benefit and little practical value. Neverthe-
less, a detailed economic analysis of conventional versus selective demolition [2] showed
that, in spite of the economic viability of selective demolition depending largely on labour
costs, tipping fees, and market prices for recovered materials, it may ultimately be more prof-
2
itable than the conventional demolition approach. Furthermore, from an environmental point
of view, there are clear benefits from using selective demolition, namely in a direct reduction
in the material sent to landfill [3, 4] as well as other environmental impacts (i.e. climatic
change, acidification, summer smog, nitrification and amount of heavy metals [5]).
The use of recycled aggregates (RA), as replacement for NA, in construction applications has
been considered as one of the cleanest approaches for recycling given materials from CDW,
search and development work on this subject has been carried out over the last 40 years,
which has been becoming increasingly complex, introducing several new variables, in which
After being subjected to proper beneficiation processes in certified recycling plants, accord-
ing to existing specifications [6-21], there are three main types of RA arising from CDW that
are suitable for the production of structural concrete. These materials are derived from
Recycled concrete aggregates (RCA), which is the most commonly studied recycled material in
fragments and NA, according to some of the aforementioned specifications [10, 15, 16, 18].
RA sourced from crushed masonry may be classified as recycled masonry aggregates (RMA)
and are composed of a minimum of 90%, by mass, of the summation of the following materi-
als: aerated and lightweight concrete blocks; ceramic bricks; blast-furnace slag bricks and
blocks; ceramic roofing tiles and shingles; and sand-lime bricks [22].
Mixed recycled aggregates, or MRA, are a mixture of the two aforementioned components,
acquired from mixed demolition debris, after the beneficiation process. Some specifications [8,
16] state that they are composed of less than 90%, by mass, of Portland cement-based frag-
3
ments and NA. In other words, they may contain other common CDW materials such as ma-
sonry-based materials.
The scope of this investigation was to bring together, analyse and evaluate the published in-
formation on the effect of several factors related to the use of RA on the tensile strength of con-
crete. A statistical analysis was also performed on the collated data from several studies, in or-
der to understand the effect of introducing an increasing amount of RA on this property. Fur-
thermore, the authors also sought to understand the tensile strength development of recycled
aggregate concrete (RAC) over time by using the model proposed in EC2 [23] and assess its
potential enhancement, by means of a multiple linear regression. Finally, the relationship be-
tween the tensile and compressive strengths was also studied, since it constitutes the main ap-
proach for concrete producers and designers to estimate the tensile strength of concrete.
2 Influencing factors on the flexural and splitting tensile strength of recycled aggre-
gate concrete
According to the brittle fracture theory, failure of a specimen is initiated through the largest crack
oriented in the direction normal to the applied load. This makes the occurrence of such a crack a
stochastic problem, in which the size and shape of the specimen are factors that affect strength,
since there is a higher probability of a larger specimen containing a greater number of critical
cracks, which can initiate failure [24]. In the case of concrete, the energy released at the onset of
cracking may not be sufficient to continue the propagation of a crack because it may be blocked
by a large pore or a more ductile material, which requires more energy to fracture.
The influence of the aggregate shape is more apparent in the flexural strength test than in the
compressive or splitting tensile strength tests, probably because of a stress gradient that de-
lays the progress of cracking leading to ultimate failure. Therefore, concrete with angular-
shaped aggregates will normally exhibit higher flexural strength than when round-shaped
4
aggregates are used, especially in mixes with low w/c ratios. However, to achieve the same
workability, round-shaped aggregates require less water than angular-shaped aggregates, and
thus the flexural strengths of the two concretes are similar [24].
It is suggested that the type of aggregate has little influence on the direct and splitting tensile
strengths, but the flexural strength of concrete is greater when angular crushed aggregate is
used than with rounded natural gravel [24]. This can be explained by the improved bond be-
tween the angular-shaped crushed aggregate and the cement paste. This, however, does not
occur with crushed flint gravels, which normally result in a low tensile strength due to poor
bond with the glassy flint surfaces [25]. The interfacial transition zone (ITZ) between this
type of aggregate and hydrated cement paste may exhibit few mechanical bonds, thus leading
The aggregates’ size also affect the tensile strength concrete, which tends to be higher when
using finer aggregate fractions, due to the increase in aggregate’s surface area and hence re-
Generally, RAC tends to exhibit lower tensile strength when compared to that of the corre-
sponding natural aggregate concrete (NAC), and the magnitude of the difference depends on
several factors related to the use of RA that are discussed in the following sections.
The literature review shows a consensus in that, as the replacement level increases, the tensile
strength decreases. This was observed in all publications studied [26-66]. Figure 1 presents the
relative splitting tensile strength of concrete with increasing amounts of coarse and fine RA,
regardless of type and quality. Although there is a clear trend that the inclusion of RA leads to
lower splitting tensile strengths when compared to that of the control concrete, in a few cases
RAC exhibited similar or even slightly greater strength. Since this was observed mostly for
specimens containing RCA, it is possible that the bond strength in the ITZ, between the old
5
adhered mortar and new cement paste, improved due to the rough nature of this type of RA.
A 95% confidence interval can also be seen in Figure 1. It shows that there is a probability of
95% that the RAC specimens containing 100% coarse or fine RA show a splitting tensile
strength loss below 60% and 54%, respectively, when compared to NAC mixes. However, it
is possible that these values are in fact closer to 40% since there is greater than expected
splitting tensile strength loss in lower replacement levels, which led to a propagation of error
To further understand the effect of adding increasing amounts of RA on the splitting tensile
strength, the data depicted in Figure 1 were separated by type, as shown in Figure 2. The results
clearly show that the addition of increasing amounts of MRA leads to a greater splitting tensile
strength loss when compared to that of RAC containing RCA. This was to be expected since
MRA contain RMA, which normally present greater porosity and lower strength [34, 35, 55].
In their previous study [67], the authors produced a performance-based classification for RA,
using their physical properties (i.e. water absorption and oven-dried density). The use of this
simple methodology intends to fill a gap concerning the variable nature of RA and therefore
provide a means to understand their possible effect on the properties of concrete. This classifi-
cation system, which allowed categorizing RA based on their quality, showed strong correla-
tions with the mean compressive strength of concrete containing them [68]. For this reason, the
data in Figure 1 were also separated by class, as shown in Figure 3. The results show that, as
the quality of RA decreases, concrete tends to exhibit greater losses of splitting tensile strength.
By removing the variability of data, it was possible to observe the effect of increasing amount
of RA of different classes to this property at a more consistent rate (Figure 4). Yang et al. [69]
studied the effect of introducing RCA with different quality and size on the properties of con-
crete. Figure 4 shows that concrete made with high quality RCA (class A) resulted in marginal
tensile strength losses (10% when 100% coarse RCA were used). At the same time, the use of
6
either coarse or fine RCA, which exhibited very similar water absorption and oven-dried densi-
ty values thus belonging in the same category (class B), also resulted in very similar tensile
strength losses. This further reinforces the theory that, regardless of size and type of RA and
provided that they exhibit similar physical properties, when applied at a given replacement lev-
el, it is likely that the resulting concrete exhibits similar performance [67, 68].
In a study carried out by Pedro et al. [59], the influence of the RA’s crushing procedure on
the properties of RAC was studied. Two types of coarse RCA were studied, one subjected
only to a primary crushing stage (RCA-1), whilst the other was subjected to primary plus
secondary crushing stages (RCA-2). The second procedure allowed producing rounder RCA
with less old mortar adhered to its surface and thus with greater quality (lower water absorp-
tion and higher oven-dried density). Mixes with 100% RCA-2 showed improvements in the
splitting tensile strength between 4% and 12% in comparison with mixes made with 100%
RCA-1. Similar results were witnessed by other researchers [89, 90], who also produced con-
crete with RCA sourced from materials subjected to different crushing stages. This suggests
that the crushing procedure is a vital part of the recycling process, significantly affecting the
state. This prevents them from absorbing the free mixing water, thus maintaining similar lev-
els of consistency [22]. One way to achieve such state is by pre-saturating RA 24 hours prior to
mixing. It is also possible to obtain saturated and surface-dried RA by using a water compensa-
tion method during mixing [70], which is known to produce RAC with minimum slump and
strength losses [39, 41, 71-73]. This method consists of adding extra mixing water that corre-
sponds to the amount absorbed by RA, with the aim of keeping the free water content constant.
7
Naturally, the additional water and time to absorb it depend on the aggregate’s size and absorp-
tion capacity. This increase in performance, in comparison to that resulting from the pre-
saturation method, was explained by a “nailing effect” caused by the cement paste filling the
surface pores of aggregate particles, thus achieving an improved bond strength [73].
In recent studies [74-77], the “two-stage mixing approach” was proposed, which consists of
pre-wetting the RA before adding the cement, in order to strengthen the weak bond of RA with
the new cement paste. The concept behind this mixing procedure, which is similar to the one
previously mentioned, is that it allows the cement slurry to coat the RA, providing a stronger
ITZ by filling the cracks and pores within them, therefore improving the bond strength [77].
Other authors [78, 79] have also witnessed that the use of a water compensation method dur-
ing the mixing procedure allows the production of mixes with similar workability and residu-
al strength loss.
Another method capable of maintaining the workability of RAC is the use of superplasticiz-
ers. These admixtures allow producing mixes with the same total w/c ratio as that of the con-
trol NAC and offset part of the loss of tensile strength from using RA [80].
Juan and Gutiérrez [81] performed a study on the effect of superplasticizers on the strength of
RAC with increasing coarse RCA content. The test results showed marginal splitting tensile
strength losses with increasing RCA content. The comparison of concrete mixes with the
same cement content but with increasing superplasticizer content showed little change as the
replacement level increased. This suggests that increasing RCA content had marginal influ-
In another research, Pereira et al. [60] studied the effects of incorporating two types of water-
reducing admixtures (i.e. a regular one, WRA, and a high-range water-reducing one, HRWRA)
on the mechanical performance of concrete containing fine RCA. As expected, the results
8
showed that increasing fine RCA led to lower splitting tensile strength. The authors also ob-
served that this decrease was greater when HRWRA was used, which suggests that the effect of
superplasticizers may decrease with the introduction of increasing fine RCA content. However,
further research is required to ascertain whether this is a generalized trend or an isolated case.
Poon and Kou [82] performed a long-term study on the mechanical properties of RAC with
increasing coarse RCA and fly ash content. Figure 5 plots the splitting tensile strength devel-
opment over 10 years of mixes with increasing coarse RCA content. The 28-day tensile
strength values of all concrete mixes were between 60% and 70% of the 10-year tensile
strength, showing that there was a significant strength increase. It is possible that the carbon-
enriched environment (i.e. Hong Kong) may have facilitated reactions between CO2 and the
calcium hydroxide (Ca(OH)2), present in the cement, to produce calcium carbonate (CaCO3).
This process results in a slightly increased strength and a lower permeability, as a result of
the deposition of CaCO3 in the voids within the cement matrix [24].
All mixes showed decreasing splitting tensile strength with increasing replacement levels, at
a relatively young age of 28 days. However, after 1 year and regardless of fly ash content, all
mixes with 100% coarse RCA showed similar or even greater splitting tensile strength than
that of the control mixes. Since coarse RCA have great superficial porosity, which may have
been filled with the new cement paste, it is possible that the increased performance over time
may have been due to an improvement of the ITZ’s microstructure between coarse RCA and
It is well known that using fly ash as cement replacement will cause a decrease in the early-
strength of concrete. A study by Kou et al. [47], on the effect of fly ash on the properties of
RAC, showed that, as the content of this addition increased, the splitting tensile strength de-
9
creased. It was also observed that, for a water to binder (w/b) ratio of 0.45, specimens with
increasing coarse RCA content showed lower strength losses when the fly ash content in-
creased, whilst this effect was not as obvious for mixes with a higher w/b ratio of 0.55. This
Following the study mentioned in section 3.4, Figures 5b, c and d show the splitting tensile
strength of concrete mixes with varying RCA and fly ash content, over a 10-year span. The
effects, over time, of adding fly ash in concrete caused, as expected, a reduction in the split-
ting tensile strength, except for mixes with 25% fly ash, whose strength development was
similar to that of mixes without additions. In fact, as opposed to what happens to compressive
strength, where mixes with fly ash tend to exhibit over time a greater strength increase than
mixes without it, these results suggest a similar splitting tensile strength gain development
over time regardless of fly ash content. For the same time period (10 years) all mixes showed
average splitting tensile strength increases of 1.50 MPa for water-cured specimens and 1.19
MPa for air-cured specimens. This suggests that the pozzolanic reactions of the addition used
did not contribute to the strength gain over time as it happens for compressive strength, where
mixes containing 0%, 25%, 35% and 55% fly ash showed average compressive strength in-
creases of 22.1 MPa, 29.2 MPa, 30.9 MPa and 32.7 MPa, respectively. For air-cured speci-
mens these values were 15.5 MPa, 23.6 MPa, 24.7 MPa and 24.9 MPa, respectively.
Kou et al. [84] studied the effect of adding 10% silica fume, 15% metakaolin, 35% fly ash
and 55% ground granulated blast furnace slag, by weight, to the properties of concrete with
increasing coarse RCA content. As expected, the increasing amount of RCA caused a slight
reduction in splitting tensile strength. However, the results suggest that the relative effect of
using a given type of addition was not hindered by the increasing amount of RCA.
The simultaneous use of ground granulated blast furnace slag (GGBS), as cement replace-
ment, and RA in the production of concrete, was studied by Berndt [85]. The use of 50% of
10
GGBS caused tensile strength gains around 20%, whilst, when using 70%, the tensile strength
remained the same. Nevertheless, the use of RA in conjunction with GGBS caused the same
strength loss as in the corresponding NAC specimens. This suggests that the effect of adding
GGBS to RAC mixes is the same as when it is added to corresponding NAC mixes.
3 Proposal of new coefficients for the prediction of tensile strength over time using EC2
Using the “Statistical Package for the Social Sciences” (SPSS) software, a comprehensive
statistical analysis was performed on the formulae proposed by EC2 [23] for the prediction of
the tensile strength of concrete at different ages, which are the following:
Where βcc(t) is defined by formula (2); α is equal to 1 for t < 28 days and; α is equal to 2/3 for t ≥
28 days. Up to 28 days, the development of the tensile strength is the same as that of compressive
strength. However, beyond 28 days the tensile strength is assumed to develop to a lesser extent.
28 0.5
�𝑠𝑠�1−� � ��
𝛽𝛽𝑐𝑐𝑐𝑐 (𝑡𝑡) = 𝑒𝑒 𝑡𝑡 (2)
Where s is a coefficient that depends on the type of cement and is equal to: 0.20 for cement of
strength classes CEM 42.5R, CEM 52.5N and CEM 52.5R (Class R - high early strength);
0.25 for cement of strength classes CEM 32.5R and CEM 42.5N (Class N - normal early
strength) and; 0.38 for cement of strength classes CEM 32.5N (Class S - slow early strength).
Generally, when the cement strength class is unknown, class R should be assumed, unless the
following alternatives apply: class N may be assumed if GGBS or fly ash contents exceed
35% or 20%, respectively, of the cement combination; class S may be assumed if GGBS or
fly ash contents exceed 65% or 35%, respectively, of the cement combination.
Figure 6 presents the relationship between measured and calculated splitting tensile strengths
using the EC2 formulae and results obtained by Poon and Kou [82]. The results show that the
Pearson’s r and standard error of the estimate obtained for this relationship were equal to 0.766
and 0.39, respectively. Although the Pearson correlation test produced a p-value of 2.33∙10-24,
11
thus showing that the null hypothesis is rejected and that the explanatory variables are not inde-
The EC2 formulae were created specifically for NAC. However, after isolating concrete mixes by
replacement level, an analysis of the accuracy of the EC2 formulae regarding NAC showed that
there was a greater skewness with increasing concrete age. The measured values of the 10-year
tensile strength of concrete were 17% to 35% greater than those of the predicted values. Since
similar results were obtained for mixes with increasing RCA, these findings suggest that there is a
similar discrepancy between predicted and measured values, regardless of replacement level.
After setting an origin equal to zero in the linear regression, a very low coefficient of determina-
tion (R2 = 0.07) was observed. Furthermore, an analysis of the coefficient concerning the use of
additions (in this case, fly ash) revealed that it does not have a significant effect on the tensile
strength development over time, albeit its obvious effect on the 28-day splitting tensile strength.
Therefore, it can be said that, for this data sample, the use of additions had marginal influence on
the splitting tensile strength development over time and that the constant concerning the type of
cement may be removed. Additionally, in section 3.5, it was established that the type of curing
has a greater effect on the tensile strength development over time than the use of additions.
Following this line of thought, by means of a non-linear regression analysis, the authors pro-
pose the use of different coefficients, which can be applied in the formula proposed by EC2.
After a number of iterations, the following formula is capable of predicting more accurately
Where a is the coefficient that depends on the type of curing (10 for water curing and 9 for air
By means of the T-test in SPSS, in which the null hypothesis considered the means of the meas-
ured and calculated values being equal, it was found that the p-value was equal to 0.374 in both
12
water and air-cured specimens, thus validating the null hypothesis. Furthermore, the correlation
test between the two variables produced a p-value equal to 4.05∙10-45, for both water and air-
cured specimens, which means that the null hypothesis is rejected and that the measured and cal-
culated values are not independent. Indeed, this test of correlation also led to a Pearson’s coeffi-
cient r of 0.903, which means that there is a very strong correlation between the two variables
[86] (see Figure 7). By comparing the Pearson’s r obtained using EC2’s formulae and the coeffi-
cients proposed in this paper (i.e. 0.766 and 0.903, respectively), it is clear that the latter method
is capable of predicting more accurately the tensile strength of concrete over time (see example in
Figure 8). Nevertheless, it should be mentioned that, since the proposed coefficients were deter-
mined using the results of a study conducted over the span of 10 years, it is only natural that these
be used only to predict the 10-year tensile strength of concrete. Further research is required to
4 Relationship between the tensile and compressive strength of recycled aggregate con-
In the design of concrete structures, tensile strength is used in both serviceability (SLS) and ulti-
mate limit state (ULS) calculations, e.g. considerations of cracking, shear, punching shear, bond
and anchorage; evaluation of the cracking moment for prestressed elements; design of reinforce-
ment to control crack width and spacing resulting from restrained early-age thermal contraction;
Increasing the tensile strength of concrete may not necessarily be advantageous in some cir-
cumstances. For example, in the case of early thermal cracking, higher tensile strength re-
quires an increased minimum steel ratio to accommodate the higher stress transferred to the
steel when a crack occurs. In addition higher strength normally requires concrete with a high-
er binder content and thus higher temperature rise and thermal strain due to hydration reac-
13
tions [25]. Therefore, depending upon specific requirements, it may be desirable to either
Tensile strength of concrete is commonly defined in one of three ways: direct tensile strength,
splitting tensile strength or flexural strength, determined through different testing methods. The
direct tensile strength is a value that is rarely determined by testing, but may be acquired using
the splitting tensile strength test, as per EN 12390-6 [87], by using the following formula:
Where, fctm is the mean tensile strength and; fctm,sp is the mean splitting tensile strength.
The flexural strength can be measured using the four-point method test procedure according
to EN 12390-5 [88]. The mean tensile strength can also be calculated by using the following
equation:
Where, fctm,fl is the mean flexural strength, and; h, the total member depth, in mm. It is possi-
ble to rearrange formula (2) considering that the flexural strength is, in most cases, measured
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,𝑓𝑓𝑓𝑓
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = (3)
1.5
Generally, tensile strength varies in proportion to the compressive strength. For normal structural
uses, the mean tensile strength is related to the cylinder strength using the following equations:
𝑓𝑓𝑐𝑐𝑐𝑐
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = 2.12 ∙ 𝑙𝑙𝑙𝑙(1 + ), for concrete classes > C50/60 (5)
10
Where, fck is the specified characteristic cylinder 28-day compressive strength, and; fcm, the
mean concrete cylinder 28-day compressive strength. Their relationship, proposed in the EC2
14
Using the aforementioned formulae, Figure 9 is drawn, presenting the relationship between
the mean 28-day tensile strength and characteristic 28-day compressive strength of 630 con-
crete mixes with a strength class above C12/15, produced with RA of different size, type and
quality, sourced from 41 publications [26-66]. It also contains the 95% confidence interval
for the characteristic axial tensile strength of concrete according to the EC2 [23], which can
A statistical analysis to Figure 9 showed that, of the 590 mixes, 575 are above the lower limit
(fctk;0.05), which corresponds to 97.5% of the total amount. This facilitates the wider use of these
materials in the production of structural concrete, since the design tensile strength of concrete,
α𝑐𝑐𝑐𝑐 ×𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐,0.05
𝑓𝑓𝑐𝑐𝑐𝑐𝑐𝑐 = (9)
𝛾𝛾𝐶𝐶
Where, fctd is the design tensile strength; fctk,0.05, the characteristic tensile strength of concrete at
28 days; αct, the coefficient that takes into account the long term effects on the tensile strength,
with a recommended value of 1.0, and; γC, the partial safety factor for concrete of 1.50.
Of the 590 mixes, 59 (10% of the total) are above the upper limit of the 95% confidence inter-
val. These mainly belong to mixes with relatively low fck (average value of 20.2 MPa), which
In order to understand the effect of RA incorporation on the tensile strength of concrete, the
values plotted in Figure 9 were separated by RA size and content. These results, which can be
seen in Figure 10, suggest that, regardless of size, type, quality, and replacement level of RA,
the relationship between the mean 28-day tensile strength and characteristic cylinder 28-day
compressive strength remains the same. The authors, in their previous study [68], found that the
compressive strength is affected, not only by the RA content, size, and type, but also by its
15
quality [67]. The statistical analysis presented in section 3.1 showed that the tensile strength is
also affected by the same factors. It is possible that the inclusion of such materials causes a
proportional decrease in both tensile and compressive strengths, at a rate that follows the same
relationship as observed in conventional concrete. This means that the use of RA in concrete
does not influence the relationship between the tensile strength and compressive strength and
5 Conclusions
Several conclusions were drawn from studying the various effects on the tensile strength of
it is possible to control this effect by carefully selecting the RA when producing con-
crete. Indeed, depending on the RA’s quantity, size, type and quality, there is a higher
or lower relative tensile strength loss between the NAC and RAC;
• Despite the likelihood of tensile strength loss between NAC and RAC with increasing
RA content, this difference can beat least partially offset by the use of proper mixing
state to maintain workability. This can best be achieved using a water compensation
method during mixing. Besides being the most practical and affordable method to satu-
rate RA, it has also proved to have the best results in terms of concrete performance;
• The use of superplasticizers is an effective way of offsetting the strength loss of RAC
with increasing replacement levels. These admixtures may show less effectiveness in
mixes with progressively higher replacement levels of non-saturated RA, which absorb
part of the mixing water with superplasticizers. To balance this, it is vital that RA are
used in nearly or fully saturated state to ensure that the superplasticizers are not ab-
16
• Contrary to that observed for compressive strength, the use of RA in concrete appeared
to have positive effects on the tensile strength gain over time. Concrete with increasing
coarse RCA content may achieve equal or higher tensile strength one year after casting
when compared to that of the control NAC. Nevertheless, this should be examined and
• When using mineral additions, the expected tensile strength gain or loss when using
additions is not generally affected by the increasing RA content in the mix. In other
words, the final tensile strength of concrete will result from the overlapping effects of
• Although there is an obvious effect on the 28-day tensile strength when using fly ash, it
appears to have a marginal effect on the tensile strength development over time, despite
• The EC2’s formulae for tensile strength at different ages were not capable of predicting
very accurately the actual tensile strength values of concrete from the sample of data
considered in this study. For this reason, a set of new coefficients, compatible with the
formula proposed in EC2, were proposed in this paper, which are capable of predicting
• The relationship between tensile and compressive strengths appears to have been unaf-
fected by the use of RA. This is very important as many of the existing standards and
specifications for conventional concrete use this relationship for estimating the tensile
strength by means of a given compressive strength. This will facilitate the wider use of
6 Acknowledgements
The authors are grateful for the support of the ICIST Research Institute, IST, University of
17
Lisbon and FCT (Foundation for Science and Technology).
7 References
July, 2014.
[3] ACWMA. Building 802 deconstruction project - Deconstruction results report. Oakland:
[5] Coelho A, de Brito J. Influence of construction and demolition waste management on the
[6] BCSJ. Proposed standard for the use of recycled aggregate and recycled aggregate
concrete (in Japanese). Japan: Committee on Disposal and Reuse of Construction Waste,
[7] BRE. Recycled aggregates. Watford, UK: Building Research Establishment; 1998. 6 p.
Specification for constituent materials and concrete. United Kingdom: British Standards
[9] DAfStb. Richtlinie - Beton mit rezykliertem zuschlag (in German). Germany: Deutscher
[10] DIN-4226. Aggregates for mortar and concrete, Part 100: Recycled aggregates. Berlin,
18
[12] JIS-5021. Recycled aggregate for concrete-class H (in Japanese). Japan: Japan Standards
Association; 2011. 30 p.
[13] JIS-5022. Recycled aggregate for concrete-class M (in Japanese). Japan: Japan
[14] JIS-5023. Recycled aggregate for concrete-class L (in Japanese). Japan: Japan Standards
Association; 2012. 68 p.
[15] LNEC-E471. Guide for the use of coarse recycled aggregates in concrete (in
Requirements and methodologies (in Portuguese). Rio de Janeiro, Brasil: Brazilian Association
[18] PTV-406. Recycled aggregates from construction and demolition wastes (in French).
[20] RILEM. Specifications for concrete with recycled aggregates. Mater Struct
1994;27(173):557-9.
[21] TFSCCS. Draft of Spanish regulations for the use of recycled aggregate in the
Spain). In: Vázquez E, Hendriks C, Janssen GMT, editors. International RILEM Conference
on the Use of Recycled Materials in Building and Structures. Barcelona, Spain: RILEM
19
Publications SARL; 2004. pp. 511-25.
[22] Hansen TC. Recycling of demolished concrete and masonry. London, UK: E & FN
Spon; 1992.
[24] Neville AM, Brooks JJ. Concrete technology. 2nd ed. United Kingdom: Pearson; 2010.
[27] Ajdukiewicz AB, Kliszczewicz AT. Comparative tests of beams and columns made of
recycled aggregate concrete and natural aggregate concrete. J Adv Concr Technol
2007;5(2):259-73.
[28] Akbarnezhad A, Ong KCG, Zhang MH, Tam CT, Foo TWJ. Microwave-assisted
[29] Alves AV, Vieira TF, de Brito J, Correia JR. Mechanical properties of structural
concrete with fine recycled ceramic aggregates. Constr Build Mater 2014;64:103-13.
[30] Arezoumandi M, Smith A, Volz JS, Khayat KH. An experimental study on shear
strength of reinforced concrete beams with 100% recycled concrete aggregate. Constr Build
Mater 2014;53:612-20.
performance of concrete made with recycled aggregates sourced from construction and
demolition wastes of Portuguese recycling plants (in Portuguese). Project FCT - PTDC /
ECM / 108682 / 2008: Instituto Superior Técnico, University of Lisbon, Portugal; 2013. 55 p.
[32] Cachim PB. Mechanical properties of brick aggregate concrete. Constr Build Mater
20
2009;23(3):1292-7.
[34] Correia JR, de Brito J, Pereira AS. Effects on concrete durability of using recycled
[35] Dhir RK, Paine KA. Performance related approach to the use of recycled aggregates.
Banbury, Oxon, UK: Waste and Resources Action Programme (WRAP) Aggregates Research
Programme; 2007. 77 p.
recycled aggregate concrete (RAC) with and without silica fume. Constr Build Mater
2014;61:50-9.
[37] Duan ZH, Poon CS. Properties of recycled aggregate concrete made with recycled
aggregates with different amounts of old adhered mortars. Mater Des 2014;58:19-29.
[38] Evangelista L. Behaviour of concrete with fine recycled concrete aggregates (in
[39] Evangelista L, de Brito J. Mechanical behaviour of concrete made with fine recycled
[40] Folino P, Xargay H. Recycled aggregate concrete - Mechanical behavior under uniaxial
mechanical performance of concrete made with recycled concrete waste. Cem Concr Compos
2011;33(6):637-43.
21
with recycled fine ceramic and coarse mixed aggregates. Constr Build Mater 2014;68:618-26.
[45] Kim K, Shin M, Cha S. Combined effects of recycled aggregate and fly ash towards
[46] Kou SC, Poon CS. Properties of self-compacting concrete prepared with coarse and fine
[47] Kou SC, Poon CS, Chan D. Influence of fly ash as cement replacement on the properties
[48] Kou SC, Poon CS, Chan D. Influence of fly ash as a cement addition on the hardened
[49] Kou SC, Poon CS, Lam L, Chan D. Hardened properties of recycled aggregate concrete
prepared with fly ash. In: Limbachiya MC, Roberts JJ, editors. Proceedings of the
[50] Kou SC, Poon CS. Long-term mechanical and durability properties of recycled aggregate
concrete prepared with the incorporation of fly ash. Cem Concr Compos 2013;37:12-9.
[51] Kou SC, Poon CS, Wan HW. Properties of concrete prepared with low-grade recycled
[52] Kou SC, Poon CS, Etxeberria M. Influence of recycled aggregates on long term mechanical
properties and pore size distribution of concrete. Cem Concr Compos 2011;33(2):286-91.
[53] Manzi S, Mazzotti C, Bignozzi MC. Short and long-term behavior of structural concrete
[54] Mas B, Cladera A, Bestard J, Muntaner D, Elena Lopez C, Pina S, et al. Concrete with
mixed recycled aggregates: Influence of the type of cement. Constr Build Mater 2012;34:430-41.
22
[55] Mas B, Cladera A, Olmo T, Pitarch F. Influence of the amount of mixed recycled aggregates
on the properties of concrete for non-structural use. Constr Build Mater 2012;27(1):612-22.
[56] Matias D, de Brito J, Rosa A, Pedro D. Mechanical properties of concrete produced with
recycled coarse aggregates - Influence of the use of superplasticizers. Constr Build Mater
2013;44:101-9.
[57] Medina C, Frias M, Sanchez de Rojas MI. Microstructure and properties of recycled
concretes using ceramic sanitary ware industry waste as coarse aggregate. Constr Build Mater
2012;31:112-8.
[58] Pedro D, de Brito J, Evangelista L. Influence of the use of recycled concrete aggregates
[59] Pedro D, de Brito J, Evangelista L. Performance of concrete made with aggregates recycled
from precasting industry waste: influence of the crushing process. Mater Struct 2014:1-14.
performance of concrete made with fine recycled concrete aggregates. Cem Concr Compos
2012;34(9):1044-52.
Experimental shear resistance of slabs without shear reinforcement. Eng Struct 2012;41:490-7.
[63] Sheen YN, Wang HY, Juang YP, Le DH. Assessment on the engineering properties of
[64] Thomas C, Cimentada A, Polanco JA, Setién J, Méndez D, Rico J. Influence of recycled
[65] Vaishali GG, Rao HS. Strength and permeability characteristics of fiber reinforced high
23
performance concrete with recycled aggregates. Asian Journal of Civil Engineering (Building
[66] Wang WL, Kou SC, Xing F. Deformation properties and direct shear of medium
strength concrete prepared with 100% recycled coarse aggregates. Constr Build Mater
2013;48:187-93.
[67] Silva RV, de Brito J, Dhir RK. Properties and composition of recycled aggregates from
construction and demolition waste suitable for concrete production. Constr Build Mater
2014;65:201-17.
[68] Silva RV, de Brito J, Dhir RK. The influence of the use of recycled aggregates on the
[69] Yang K, Chung H, Ashour A. Influence of type and replacement level of recycled
[70] Leite MB. Evaluation of the mechanical properties of concrete produced with recycled
aggregates from construction and demolition wastes (in Portuguese). Rio Grande do Sul,
[71] Amorim P, de Brito J, Evangelista L. Concrete made with coarse concrete aggregate:
[72] Evangelista L, de Brito J. Durability performance of concrete made with fine recycled
[74] Tam VWY, Gao XF, Tam CM. Microstructural analysis of recycled aggregate concrete
[75] Tam VWY, Tam CM. Assessment of durability of recycled aggregate concrete produced
24
by two-stage mixing approach. Journal of Materials Science 2007;42(10):3592-602.
[76] Tam VWY, Tam CM. Diversifying two-stage mixing approach (TSMA) for recycled
[77] Tam VWY, Tam CM, Wang Y. Optimization on proportion for recycled aggregate in
[78] Poon CS, Shui ZH, Lam L, Fok H, Kou SC. Influence of moisture states of natural and
recycled aggregates on the slump and compressive strength of concrete. Cem Concr Res
2004;34(1):31-6.
[79] Kong D, Lei T, Zheng J, Ma C, Jiang J, Jiang J. Effect and mechanism of surface-
[81] Juan MS, Gutiérrez PA. Influence of recycled aggregate quality on concrete properties.
In: Vázquez E, Hendriks C, Janssen GMT, editors. International RILEM Conference on the
[82] Poon C, Kou S. Effects of fly ash on mechanical properties of 10-year-old concrete
prepared with recycled concrete aggregates. In: Xiao JZ, Zhang Y, Cheung MS, Chu RPK,
[83] Salem RM, Burdette EG, Jackson NM. Resistance to freezing and thawing of recycled
[84] Kou SC, Poon CS, Agrela F. Comparisons of natural and recycled aggregate concretes
25
prepared with the addition of different mineral admixtures. Cem Concr Compos
2011;33(8):788-95.
[85] Berndt ML. Properties of sustainable concrete containing fly ash, slag and recycled
[86] Piaw CY. Basic research statistics (Book 2). Kuala Lumpur, Malaysia: McGraw-Hill; 2006.
[87] EN-12390-6. Testing hardened concrete: Tensile splitting strength of test specimens.
[88] EN-12390-5. Testing hardened concrete: Flexural strength of test specimens. Brussels,
[90] Nagataki S, Iida K. Recycling of demolished concrete. In: Malhotra VM, editor. Fifth
Singapore: Canada Centre for Mineral and Energy Technology (CANMET) of Natural
26
FIGURES
Figure captions:
Figure 1 - Influence of increasing (a) coarse and (b) fine RA content on the relative splitting
tensile strength
Figure 2 - Influence of increasing coarse RCA and MRA contents on the relative splitting
tensile strength
Figure 3 - Effect of incorporating increasing amounts of RA with different quality on the relative
Figure 4 - Effect of incorporating increasing amounts of RA with different quality on the (a)
relative splitting tensile and (b) flexural strengths (adapted from Yang et al. [69])
Figure 5 - Strength over time of concrete mixes with increasing coarse RCA content and
varying fly ash content: a) No fly ash; b) 25% fly ash; c) 35% fly ash; 55% fly ash (adapted
Figure 6 - Relationship between the measured and calculated splitting tensile strength using EC2
Figure 7 - Relationship between the measured and calculated splitting tensile strength over
Figure 8 - Measured (M) and calculated (C) splitting tensile strength over time of mixes with
increasing coarse RCA content using (a) the EC2 method and (b) the coefficients proposed in
this paper
Figure 9 - Relationship between the mean tensile strength and characteristic cylinder
compressive strength
Figure 10 - Relationship between fctm and fck of concrete with increasing replacement levels
27
1.6 1.6
1.4 1.4
1.2 1.2
fctm,sp,RAC / fctm,sp,NAC
fctm,sp,RAC / fctm,sp,NAC
1 1
0.8 0.8
0.6 0.6
0.4 0.4
95% LCL 95% LCL
0.2 0.2
95% UCL 95% UCL
0 0
20 0 40 60 80 100 0 20 40 60 80 100
Coarse RA content (%) Fine RA content (%)
a) b)
Figure 1 - Influence of increasing (a) coarse and (b) fine RA content on the relative splitting tensile strength
28
1.6
1.4 y = -0.0012x + 1
1.2
fctm,sp,RAC / fctm,sp,NAC
1
0.8
0.6
0.4 y = -0.0027x + 1
RCA
0.2
MRA
0
0 20 40 60 80 100
Coarse RA content (%)
Figure 2 - Influence of increasing coarse RCA and MRA contents on the relative splitting tensile strength
29
1.6
1.4
1.2
fctm,sp,RAC / fctm,sp,NAC
1
0.8
0.6
Class A y = -0.0009x + 1
0.4
Class B y = -0.0015x + 1
0.2
Class C y = -0.0017x + 1
0
0 20 40 60 80 100
Coarse RA content (%)
Figure 3 - Effect of incorporating increasing amounts of RA with different quality on the relative splitting tensile
strength
30
1.2 1.2
y = -0.001x + 1 y = -0.0011x + 1
1.0 R² = 0.5785 1.0 R² = 0.9575
fctm,sp,RAC / fctm,sp,NAC
fctm,fl,RAC / fctm,fl,NAC
0.8 0.8
31
5.0 5.0
4.5 4.5
4.0 4.0
fctm,sp (MPa)
fctm,sp (MPa)
3.5 3.5
3.0 3.0
4.5 4.5
4.0 4.0
fctm,sp (MPa)
fctm,sp (MPa)
3.5 3.5
3.0 3.0
32
5.0
y = 0.9132x
4.5 R² = 0.0733
3.5
3.0
2.5
2.0
1.5
1.0
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Measured fctm,sp (MPa)
Figure 6 - Relationship between the measured and calculated splitting tensile strength using EC2
33
5.0 5.0
y = 0.9978x y = 1.0075x
4.5 R² = 0.7895 4.5 R² = 0.7989
Calculated fctm,sp (MPa)
3.5 3.5
3.0 3.0
2.5 2.5
2.0 2.0
1.5 1.5
1.0 1.0
1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
Measured fctm,sp (MPa) Measured fctm,sp (MPa)
a) b)
Figure 7 - Relationship between the measured and calculated splitting tensile strength over time for (a) water-
cured and (b) air-cured specimens
34
5.0 5.0
4.5 4.5
4.0 4.0
fctm,sp (MPa)
fctm,sp (MPa)
3.5 3.5
3.0 3.0
35
8
fctm (MPa)
4
2
fctk,0.95
1 fctk,0.05
EC2 - fctm
0
0 20 40 60 80 100
fck (MPa)
Figure 9 - Relationship between the mean tensile strength and characteristic cylinder compressive strength
36
NAC RAC 0-25 RAC 25-50
NAC RAC 0-25 RAC 25-50 RAC 75-100
RAC 50-75 RAC 75-100
8 8
7 7
6 6
5 5
fctm (MPa)
fctm (MPa)
4 4
3 3
2 2
1 1
0 0
0 20
40 60 80 100 0 20 40 60 80 100
fck (MPa) fck (MPa)
a) b)
Figure 10 - Relationship between fctm and fck of concrete with increasing replacement levels of (a) coarse and (b)
fine RA
37