Microsoft Word - CET--006.docx
CHEMICAL ENGINEERING TRANSACTIONS
VOL. 59, 2017
A publication of
The Italian Association
of Chemical Engineering
Online at www.aidic.it/cet
Guest Editors: Zhuo Yang, Junjie Ba, Jing Pan
Copyright © 2017, AIDIC Servizi S.r.l.
ISBN 978-88-95608- 49-5; ISSN 2283-9216
A Statistical Study on the Compressive Strength of Recycled
Aggregate Concrete
Qin Lia, Chunhong Zhangb*
a Chemical Engineering Faculty of Yunan Open University, Kunming 650093, China
b School of Ecology and Environmental Science, Yunnan University, Kunming 650091, China
shelly20@163.com
In order to explore the effect of RCA on the compressive strength of concrete, two RCAs from different
sources are utilized to take experiments in this study. Four groups of concrete, including three RACs
incorporated with various RCAs and another one made with natural aggregate (i.e. normal aggregate concrete
(NAC)) are tested under an identical laboratory condition. The influence of RCA on the statistical parameters
of the compressive strength, including the mean value, the standard deviation and the coefficient of variation
are examined in detail. In addition, the fitted normal distribution modelto is used to describe the probability
distribution characteristics of the compressive strength of the RACs, and the Monte-Carlo simulation technique
is applied to verify the suitability of the normal distribution model for the probability distribution of the
compressive strength of RAC. At last, it is summed up that the RCA does not have a significant effect on the
variation and probability distribution of the compressive strength of concrete.
1. Introduction
Recycled aggregate concrete (RAC), was introduced into engineering practice many years ago. The
application of RAC is beneficial for solving the increased waste storage problems, for reusing the large
amount of waste concrete form the demolition of old concrete structures, and also for protecting the limited
natural aggregate sources. Numerous studies covering the processing of the demolished rubble, the mix
design, the material and structural behavior, and the economic and sustainability aspects of RAC have been
carried out (Xiao et al., 2006). These findings have been summarized in some review papers and reports.
Since 1900s, due to the accelerated urbanization, numerous researchers in China have devoted themselves
to the research and applications of RAC in practice. The most important achievements obtained in China in
the past years on this topic have been extensively reviewed. It is revealed from the previous studies in China
and other countries that RAC exhibits generally lower mechanical properties in comparison to normal
aggregate concrete (NAC)(i.e. concrete produced with natural aggregates) (Li et al., 2006). However, until
now there is no one test result indicating that RAC is not suitable for use in engineering practice. In the recent
years, Significant studies have been carried out on the stiffness, deformation, strength, and ductility of
structural components and structures made of RAC. A review of existing work on the structural behavior of
RAC has recently been given. It is found that the cracking patterns and failure of RAC structural elements and
structures are generally similar to that of NAC structures.
Although tremendous advances in understanding the material and structural behavior of RAC, there still exist
some significant barriers in the practical application of RAC up to now. These barriers include the lack of
appropriately located recycling facilities, the absence of appropriate technology, the lack of public awareness
and government support, and the lack of proper standards, as pointed out (Hendriks, 1985). In addition, too
little knowledge about the safety of structures made of RAC is another very significant barrier. It is often
assumed that the coefficient of variation of the compressive strength of RAC increases in comparison to NAC,
owing to the larger scatter in the quality and the composition of demolished concretes in practice (Rao et al.,
2007). The increased scattering of the compressive strength of RAC reduces the safety of structural
components and structures made of this material. This decreased structural safety obviously introduces
additional worry when RAC is used in engineering practice. Therefore, the safety of structural elements and
DOI: 10.3303/CET1759071
Please cite this article as: Qin Li, Chunhong Zhang, 2017, A statistical study on the compressive strength of recycled aggregate concrete,
Chemical Engineering Transactions, 59, 421-426 DOI:10.3303/CET1759071
421
mailto:shelly20@163.com
structures made of RAC should be carefully studied before this material can be used in practice with sufficient
confidence.
The statistical characteristics of the compressive strength of concrete are of great importance for the reliable
design and safety assessment of concrete structural elements and structures. For RAC, the use of RCA might
increase the variations or uncertainties in the strength properties of such concrete due to the impurities
introduced when producing RCA from waste concretes. Keeping this in view, the objective of the present work
is to investigate the effect of RCA on the variations in the compressive strength of concrete (Xiao et al., 2005).
Two RCAs from different sources were employed in this study. The effect of the RCA on the statistical
parameters of the compressive strength, such as mean value, standard deviation, and coefficient of variation
as well as probability distribution is evaluated. The normal distribution probabilistic model is examined whether
it can be used to fit the experimental results for the compressive strength of RAC. The model is verified by
Pearson goodness of fit test at 95% confidence level. Finally, the Monte-Carlo simulation technique is
performed to obtain the theoretical strength distribution. The simulated distribution is compared with the
experimental data.
2. Previous Studies
Very few investigations on the effect of RCA on the statistical parameters of the compressive strength of
concrete have been carried out in previous studies. Early studies in Japan implied that the coefficient of
variation of the compressive strength of RAC, which was produced in the laboratory, was not much different
from NAC when one and the same RCA was used throughout the whole production process (Nihibayashi and
Yamura, 1988). This finding was later confirmed and reported that the quality of RCA from crushed concrete
specimens produced in the laboratory was relatively uniform, which resulted in a similar variability of the
compressive strength of RAC and NAC (Larranaga, 2004). Similar results were also observed by some
researches who found the compressive strength of RAC does not vary much from that of normal concrete with
similar strength. In all the above experiments, RCAs crushed from laboratory specimens were used.
To more realistically represent the RCA from demolished concrete structures, RCA that was crushed from
waste concretes from a concrete pavement was used. On the basis of a large experimental program, the
statistical parameters of compressive strength of RAC are investigated. The experimental results indicated
that the coefficient of variation of the compressive strength of RAC is not much different from that of NAC
under their test conditions, regardless of the RCA replacement percentage. Similar findings were also reported
by a scholar who used RCA produced from a concrete plant.
It should be noted that in most of the above tests, only RCA from one single source were used. However, in
practice, RCA may come from distinct different sources. For RAC made with RCAs from different sources, the
probabilistic distribution characteristics for the compressive strength of the concretes need to be further
investigated.
3. Experimental Program
Ordinary Portland cement (OPC) with a 28-day compressive strength of 42.5 MPa and local river sand
(fineness modulus 2.75), and water through campus water supply systems in Lishui city are used. Two RCAs
were utilized in this study. The first RCA, denoted as RCA1, was prepared by crushing laboratory cast
specimens made from original concrete, which has the compressive strength about 43MPa. The second, that
is, RCA2 was produced from demolished concrete rubble in Lishui city. The drilled core tests showed that the
compressive strength of the concrete rubble is about 37MPa (Xiao, et al., 2008). The natural aggregate is
crushed granite. The RCAs and natural aggregate have the similar grading curves, and they meet the
corresponding requirement for normal concrete. The basic properties of both natural aggregate and RCA are
presented in Table 1. From Table 1, it can be seen that the RCAs have the following characteristics in
comparison to natural aggregates: lower bulk and apparent density, higher water absorption ratio and lower
strength characterized by larger crushing index (de Paw, 1981). The characteristics are attributed to old
mortar adheres to the RCAs.
Due to the high water absorption ratio of RCAs, the mix design method for conventional concrete, cannot be
directly used to RAC since it may result in poor workability of the concrete mixtures. In this work, the mixing
design procedure proposed for RAC in is used. The mixtures were divided into four groups, namely NAC,
RAC1, RAC2 and RAC3. The mainly difference among various mixtures is the used coarse aggregate. For the
mixture NAC, only natural aggregate was used, while for mixtures RAC1 and RAC2, the RCA1 and RCA2
were used, respectively. In the last mixture RAC3, both RAC1 and RAC2 were used with a volume ratio of 1:1
(Hansen, et al., 1983). For each mixture, a total of 45 cubes of 150 x150x150mm were prepared considering
422
the need for Pearson goodness of fit test. Table 2 shows the mix proportions and slump for all concrete
mixtures.
Table 1: Basic properties of natural and recycled aggregates
Aggregate
type
Grading
(mm)
Bulk density
(kg/m3)
Apparent density
(kg/m3)
Water absorption
(%)
Crushing index
(%)
NAC 5-31.5 1450 2894 0.25 3.82
RCA-1 5-31.5 1202 2410 6.25 12.6
RCA-2 5-31.5 1232 2480 4.16 9.8
RCA-3 5-31.5 1218 2442 5.22 11.2
Table 2: Mix proportions and slumps of concrete
Mix W/C
Materials used in 1m3 (kg/m3)
Slump
(mm)
Cement Sand
Coarse aggregate
Water
Natural Recycled
NAC 0.39 475 505 1265 - 185 52
RAC1 0.39 521 505 - 1066 185 48
RAC2 0.39 521 505 - 1084 185 50
RAC3 0.39 473 562 - 1093 185 45
The preparation and the cure of all mixtures were conducted in the Laboratory for Building Material Research
at Lishui University, PR China. All mixtures are prepared with a 150L mixing machine. The sand, cement and
coarse aggregates are placed and mixed for about 2 min before water is added. After 3 min of mixing followed,
a slump test is run to determine its workability (Coquillat, et al., 1982). The mixture in each group is cast in
steel moulds and then compacted on a vibration table. They are demolded 24h after casting. All the concrete
specimens were cured in a standard frog room (2020C and 95% relative humidity) for 28 days before the test.
All specimens of NAC and RAC are tested in a universal testing machine to obtain the failure load.
4. Test results and statistical analysis
4.1 Statistical parameters of the tested compressive strength
After testing N number concrete cube specimens under the same condition, a series of order strength values
can be obtained. The statistical parameters, such as the maximum and the minimum value, and the mean
value, the standard deviation and the coefficient of variation for the compressive strength of the concretes are
obtained and shown in Table 3.
Table 3: Statistical parameters for the compressive strength
No max,cuf min,cuf cuf
~
cuf
cuf
cuf
(%)
NAC 45.2 33.8 40.1 39.6 3.34 8.43
RAC1 42.9 29.8 38.3 37.8 3.22 8.52
RAC2 44.4 32.9 39.5 39.4 3.29 8.35
RAC3 43.7 30.2 38.7 38.0 3.43 9.04
It can be seen from Table 3 that the mean compressive strength of RAC is in general lower than that of NAC.
For RAC1 and RAC2, the reductions are 3.62% and 5.08 %, respectively; while for RAC3, the strength
reduction is 3.96% (Lü, 2000). Several reasons may be responsible for the reduction of the compressive
strength, including an increased porosity and a weak aggregate-matrix interfacial bond.
One can see from Table 3 that the standard deviations and the coefficient of variation for the compressive
strength of all the three RACs are somewhat higher than that of normal concrete. In addition, the two
parameters mentioned above for RAC2 and RAC3 are slight higher than that of RAC1 (Li, 2009). This is
because the RCA1 has more uniform quality than RCA2 and RAC3. However, generally speaking, the
variability for the three RACs is not much different from those of the reference NAC.
423
4.2 Probability distribution characteristics of the compressive strength
For conventional concrete, it is generally accepted that the normal distribution model is suitable for describing
the probability distribution of the compressive strength. The histograms, which are primarily used for
illustrating the distributions of the concrete compressive strength of concrete mixtures, are shown in Fig.1.
Figure 1: Histogram for the distribution of the compressive strength
It can be seen that the observed strength varies over a wide range (Yen, et al., 1997). From the histograms, it
is reasonable to assume a normal distribution to fit the measured data for all the concretes, irrespective of the
natural and recycled aggregates. Thus, the normal distribution is tried to fit the data. In order to verify the
normal distribution for RACs compressive strength, a Pearson test and a Kolmogorov-Smirnov test are
performed.
For the Pearson test, we suppose for normal distribution H0
The results of the-test for the distribution of the compressive strength for all the tested concretes are
presented in Tables 4. The results in this Table indicate that H0 for the normal distributions cannot be rejected
(Nixon, et al., 1978). Thus, it can be concluded that the normal distributions is able to applied to fit the
compressive strength results of all the concretes at a 95% confidence level, irrespective of natural and
recycled aggregates.
Table 4: Test results on the normal distribution of compressive strength
No -Statistic -Critical (at 0.05 level, 2-tailed) Accept / refuse H0
NAC 4.2379 7.815 Accept
RAC1 7.6089 7.815 Accept
RAC2 4.9288 7.815 Accept
RAC3 6.7227 7.815 Accept
4.3 The Monte-Carlo simulation
Due to the limited test results, a Monte-Carlo simulation study was also conducted. Such simulations have
also been carried out in previous studies. The Monte-Carlo method refers to the practice of randomly
assigning a value to an underlying random variable or vector, and observing the outcome of the process of
424
interest (Mukherjee and Chakraborty, 2003). Here, the underlying random variable is the compressive strength
of the concrete and the process of interest is the probability of failure. The random concrete strength (Xi) is
obtained by assigning the cumulative distribution function of the normal model, namely, F(Xi) to a random
value ri between 0 and 1. The process is repeated several times to obtain the failure probability, i.e., the
number of specimens failed at any stage of loading to the total number of simulations at that stage of loading
(Hansen, et al., 1992). The probabilities of the failure, that is, the cumulative distribution function at different
stage of loading obtained by experiment and simulation for all the tested concretes are calculated and
presented in Figs. 2(a)-2(d). The results from the Monte-Carlo simulation are verified again that the normal
distribution can be adopted to describe the compressive strength of the RACs.
Figure 2: Comparison of the test and simulated compressive strength distribution
5. Conclusions
This paper investigates the probability distribution characteristics for the compressive strength of concrete with
RCA from two different sources. Within the scope of this study, the following conclusions can be drawn:
The standard deviation and coefficient of variation of the compressive strength of concrete with different RCAs
are only slight higher than that of normal concrete.
The normal distribution can be utilized to fit the compressive strength of RAC at 95% confidence level.
The distributions for the compressive strength of RAC do not vary much from that of normal concrete with
similar strength, inspection of the RCA from different sources.
References
Coquillat G., 1982, Recyclage de materiaux de demolition dans la confection de Beton. CEB -Service d’Etude
des Materiaiaux Unite, Technology des Beton.
De Paw, C., 1981, Fragmentation and recycling reinforced concrete. Some research results. Proceedings of
the NATO Conference on Adhesion Problems in the Recycling of Concrete, 311-317.
Hansen T.C., 1983, Strength of recycled aggregate concrete made from crushed concrete coarse aggregate.
Concrete International, 79-83.
Hansen T.C., 1992. Recycling of Demolished Concrete and Masonry. E&FN SPON, London.
425
Hendriks C.F., 1985, The use of concrete and masonry waste as aggregates for concrete production in the
Netherlands. Proceedings of the EDA/RILEM Conference on Reuse of Concrete and Bricks Materials,
Rotterdam, the Netherlands.
Larranaga M., 2004, Experimental study on microstructure and structural behavior of recycled aggregate
concrete. Ph.D thesis, University of Politecnica de Catalunya.
Li J., Xiao J., Huang J., 2006, Influence of recycled coarse aggregate replacement percentages on
compressive strength of concrete. Chinese Journal of Building Materials, 9: 297–301.
Li X., 2009, Recycling and reuse of waste concrete in China. Part II: Structural behavior of recycled aggregate
concrete and applications. Resources, Conservation and Recycling, 53: 107-112.
Lü S., 2000, Shear Behavior of Reinforced Concrete Elements with Recycled Aggregates. DAfStb, Heft 504.
Berlin.
Mukherjee D, Chakraborty S., 2003, A statistical study on compressive strength of recycled concrete, Journal
of the Institution of Engineer, 83: 219-220.
Nihibayashi S., Yamura K., 1988, Mechanical properties and durability of concrete from recycled coarse
aggregate prepared by crushing concrete. Proceedings of the Second International RILEM Symposium on
Demolition and Reuse of Concrete and Masonry. Tokyo, Japan, 652-659.
Nixon P.J., 1978. Recycled Concrete as an Aggregate for Concrete- A Review. Materials and Structures,
11:371-378.
Rao A., Jha K.N., Misra S., 2007, Use of aggregates from recycled construction and demolition waste in
concrete. Resources, Conservation and Recycling, 50: 71-81.
Xiao J., Lei B., Yuan B., 2008, Compressive strength distribution of recycled aggregate concr ete derived from
different origins. Chinese Journal of Building Structures, 29: 94–100.
Xiao J., Li J., Zhang C., 2005, On statistical characteristics of the compressive strength of recycled aggregate
concrete. Structural Concrete, 6: 149–153.
Xiao J., Sun Y., Falkner H., 2006, Seismic performance of frame structures with recycled aggregate concrete.
Engineering Structures, 28: 1-8.
Yen T, Huang Y.L., Chen H.J., 1997, Reuse of deserted concrete, Research Report, ABRI, Taiwan.
426