- Email: [email protected]
Uni-axial compressive stress-strain relation of recycled coarse aggregate concrete after freezing and thawing cycles
Construction and Building Materials 134 (2017) 210–219
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Uni-axial compressive stress-strain relation of recycled coarse aggregate concrete after freezing and thawing cycles Jin Wu ⇑, Xianhang Jing, Zhe Wang Department of Civil Engineering, Nanjing University of Aeronautics & Astronautics, Nanjing 210016, China
h i g h l i g h t s The complete compressive stress-strain curves of RAC after freezing and thawing cycles were measured in this study. Theoretic stress-strain curve of RAC after freezing and thawing cycles was presented. Parameters of descending branch in stress-strain curve of RAC were given after freezing and thawing cycles.
a r t i c l e
i n f o
Article history: Received 1 February 2016 Received in revised form 17 October 2016 Accepted 21 December 2016
Keywords: Recycled aggregate concrete Freezing and thawing Mass loss Dynamic elastic modulus Compressive strength Stress-strain curve
a b s t r a c t The mass loss, compressive strength, dynamic elastic modulus and stress-strain relationship of recycled coarse aggregate concrete under different cycles of freezing and thawing were investigated by comparison with normal concrete in the research reported in this paper. Forty-eight prism specimens with the size of 100 mm 100 mm 300 mm and ninety-six cubic specimens with the size of 100 mm 100 mm 100 mm were fabricated and tested. Test results show that, the mass loss of recycled coarse aggregate concrete decreased first then increased with the increase of freezing and thawing cycles; the relative cubic compressive strength and the relative dynamic elastic modulus of recycled aggregate coarse concrete decreased linearly with the increase of freezing and thawing cycles. The stress-strain curve of recycled aggregate concrete first ascended rapidly and then descended rapidly after exceeding the peak stress till to 1.5 times of the peak strain, finally turned into a slow decrease process. The ductility of recycled coarse aggregate concrete became worse with the increase of freezing and thawing cycles. Theoretic stress-strain curve of recycled coarse aggregate concrete after different cycles of freezing and thawing was presented and it was in good agreement with the experimental results from the other researcher. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction With the development of construction industry, a lot of waste demolished concrete is generated every year. In 2010, the amount of construction and demolition (CD) waste in China is about 1.55 billion tons, accounting for 30–40% of the total waste [1]. If this trend will continue, landfills will be saturated and environment will be polluted. Therefore, how to dispose of the huge amount of waste has become interests of many researchers. It has been reported that this huge amount of CD waste can be potentially used as recycled aggregates (RAs) for the production of recycled aggregate concrete (RAC). And the use of such CD waste as RAs will not only reduce landfills use but also help to minimize the use of natural aggregates (NAs) [2]. ⇑ Corresponding author. E-mail address: [email protected] (J. Wu). http://dx.doi.org/10.1016/j.conbuildmat.2016.12.142 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
The behavior of RAC is significantly affected by its durability, and the freezing-thawing durability is one of the major concerns associated with the application of RAC. Some studies on the freezing-thawing durability of RAC have been investigated. Zaharieva et al. [3] pointed out that the freezing-thawing resistance of RAC is lower than that of natural concrete (NC). The main reason seems to be higher porosity and lower the freezing-thawing resistance of RAs themselves. Kasai et al. [4] found that a high replacement ratio of RAs declines the freezing-thawing resistance of RAC. Gokce et al. [5] investigated the freezing-thawing resistance of RAC. Test results showed that RAC made with an air-entrained admixture had better performance than RAC made with non-airentrained admixture. Salem et al. [6] reported that RAs originated from concrete made with an air-entrained admixture produced high-quality freezing-thawing resistance of RAC. Haitao et al. [7] conducted a research on the freezing-thawing resistance of high strength (50 MPa) RAC made with 100% replacement of RAs. Test
211
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
results showed that, compressive strength, splitting tensile strength and bending strength of RAC decreased with the increase of freezing and thawing cycles. After 50 cycles of freezing and thawing, the bending strength decreased by 40%. Richardson et al. [8] conducted an experiment on the freezing-thawing resistance of RAC based on the ASTM 666 standards. The test results showed that RAs with the addition of additives can be used for applications where freezing and thawing of concrete occur whilst still providing the durability that NAs offer. The stress-strain relationship of RAC is important in theoretical and numerical analysis as well as engineering design of RAC structures. In recent years, several investigations have been performed for the stress-strain relation of RAC. Xiao et al. [9] investigated the compressive strength and stressstrain relation of RAC with different replacement ratio of RAs, and obtained the approximate stress-strain curves of RAC according to the analytical expression in Chinese Code GB50010 [10] for uniaxial compression of NC. Du et al. [11] investigated the complete stress-strain curve of RAC with 100% replacement ratio of RAs, and constituted the model of stress-strain of RAC under uniaxial compression loading. Huda et al. [12] conducted a research on the stress-strain curves at the age of 120 days of RAC with different RAs replacement ratio, and found that the pattern of the stress-strain curves was similar for all RAC mixes but the value of the strain corresponding to the peak stress was higher for RAC compared to that of NC. Belén et al. [13] also developed an analytical expression of the stress–strain curve of RAC using the experimental results, and verified the proposed model equation by comparing it to the experimental data. The results showed that the proposed model equation satisfactorily describes the effect of RAs on the stress–strain curve. However, very limited information is available on the stressstrain relations of RAC after freezing and thawing. A previous research study by Shang [14] investigated the compressive stress-strain relations of RAC after freezing and thawing, and presented a calculation model based on NC stress-strain relation, but freezing and thawing cycles were not considered in the model, and a larger deviation between the theoretical values and experimental values can be seen in the descending branch of stress-strain curves. The objective of this study is to investigate the influence of freezing and thawing cycles on the mass loss, the compressive strength, dynamic elastic modulus and stress-strain curve of RAC with 100% replacement ratio of RAs, and to present an analytical expression for stress-strain relationship of RAC after different freezing and thawing cycles. The results presented in the study are significant for using the RAC in cold areas. 2. Experimental program 2.1. Materials 2.1.1. Recycled coarse aggregate (RCA) There is great randomness and variability for RCA made of different sources of original concrete. To ensure the unification of source of RCA, the waste concrete in this test was totally from waste concrete blocks demolished from a cement road in Nanjing university Aeronautics and Astronautics. After crushing, washing and grading into 5–31.5 mm continuous gradation, RCA conforming the standard was prepared. Table 1 shows the grading profile of RCA. Water absorption and crushing index of RCA are 5.7% and 10.4%, respectively, being measured according to a method in JGJ 52-2006 [15]. The RCA used in the study conformed to the requirements of Grade Ⅱ [16].
Table 1 Grading profile of RCA. Sieve Size (mm)
Grader Retained (%)
Accumulated Retained (%)
Continuous Gradation (%)
31.5 25.0 20.0 16.0 10.0 5.0
2.25 18.21 15.76 14.38 32.52 10.41
2.25 20.46 36.22 50.60 83.12 93.53
5–0 45–15 90–70 100–90
2.1.2. Natural coarse aggregate (NCA) The NCA used in this experiment was made of natural stone, the size of particles was 5 mm to 40 mm. 2.1.3. Fine aggregate The natural river sand used in this test was medium sand. The performance indicators conform the requirements in JGJ 52-2006 [15], grading profiles and performance of sand are shown in Tables 2 and 3, respectively. 2.1.4. Cement The cement used in this test was ASTM Type Ⅱ Portland cement produced by the China Jiangnan Cement Co., LTD. Its specific surface area was 385 m2/kg. 2.1.5. Admixtures The admixtures used in this test were GYQÒ-III concrete airentraining agent and PCAÒ (I) carboxylic acid high range water reducer provided by Jiangsu Bote New Materials Co., LTD. The quality of admixtures conformed to GB 50119-2013 [17]. Before casting specimens, twelve groups of proportion tests were conducted in order to obtain the optimum dosage of air-entraining agent and water reducer. 2.1.6. Release agent GBT 50082-2009 [18] specifies that it is forbidden to use hydrophilic release agent molding specimens for freezing and thawing tests. The DL-J04 hydrophilic release agent used in this test came from Shenzhen Xindeli chemical Co., LTD. The main components of the release agent are polymer organic matters and can be diluted with water in any proportion conforming to the requirements of the test.
Table 2 Grading profile of sand. Sieve Size (mm)
Grader Retained (%)
Accumulated Retained (%)
Continuous Gradation (%)
5.00 2.50 1.25 0.630 0.315 0.160
3.20 9.16 8.13 25.75 45.70 29.70
3.33 12.49 20.62 46.37 92.07 93.30
10–0 25–0 50–10 70–41 92–70 100–90
Table 3 Performance of sand. Fineness Module
Mud Content (%)
Particle Diameter (mm)
2.6
2.1
<5
212
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
2.2. Mix proportion Forty-eight prism specimens (100 mm 100 mm 300 mm) and ninety-six cubic specimens (100 mm 100 mm 100 mm) were cast. Details of the mix proportions are listed in Table 4. All specimens were fabricated in a laboratory and cured for 24 h at room temperature, then demolded and cured for 28 days. 2.3. Test method (1) Specimens for freezing and thawing were placed in 20 ± 2 °C water for 4 days before the test. (2) After soaking water for enough time, specimens were removed out of the water, wiping moisture on the surface. The initial mass of the specimens was weighed and the initial transverse fundamental frequency was collected and recorded. (3) The concrete specimens were placed in the freeze-thaw boxes and filled with water level about 5 mm higher than the top of the specimens. (4) The specimen boxes were placed in the freezing and thawing machine. It began to freeze and thaw after the temperature sensors were placed. 2.4. Test equipment 2.4.1. Equipment of collecting the data of relative dynamic elastic modulus To collect the data of relative dynamic elastic modulus, a nonmetal ultrasonic detection analyzer was used in this study to measure the ultrasonic velocity of ultrasonic wave. The test set-up is shown in Fig. 1. 2.4.2. Equipment of collecting the data of stress and strain The stress state of the concrete will be changed due to the effect of the friction on the end of the concrete specimen. According to
Fig. 2. Load sensor and displacement sensor arrangement.
Saint & Venant’s principle, non-uniform vertical compressive stress will only have a significant impact on the stress state of the end of the specimen with a height approximately equal to the width of the specimen. In order to obtain a more accurate stress state under uniaxial compression, the strain in the middle part of the specimen (170–180 mm) was measured in this test as Fig. 2 showed. The hydraulic servo test system was used in the test of the axial compression of concrete after freezing and thawing. Dynamic data acquisition system was used to collect the displacements and loads. The axial loads applied on the prism specimens were controlled by strain, and the strain rate was 400 le/min. There was a load sensor on the end of loading device which can directly measure the magnitude of applied loads. A displacement sensor was installed on the two opposite sides to measure the longitudinal deformation. The dynamic data acquisition system was used to collect the test data.
Table 4 Mixture proportions. Types
RAC NAC *
Specimen Number
Water* (kg/mt)
Cement (kg/m3)
Sand (kg/m3)
1–8 9–16
186.0 150.0
404.7 404.7
646.5 646.5
Coarse Aggregate RCA (kg/m3)
NCA (kg/m3)
1011.6 0
0 1011.6
Pre-water for RCA is included.
Fig. 1. Test set-up for dynamic elastic modulus.
Water Reducing Agent (kg/m3)
Air-entraining Agent (kg/m3)
2.430 2.430
0.243 0.243
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
213
3. Experimental results 3.1. Specimens failure pattern With the increase of freezing and thawing cycles, the surface of RAC specimens became rough. It could be observed that lots of pits appeared, mortar spalled, micro cracks formed and extended gradually, some corners of specimens spalled and fine aggregates and coarse aggregates exposed. Fig. 3 shows the appearances of RAC after different cycles of freezing and thawing. As Fig. 3 shows, after 100 cycles of freezing and thawing, a lot of micro cracks appeared and some cracks were radial and cracks formed and extended based on a certain spot as the center. Fig. 4 shows the partial radial cracks on the end of the specimens after 100 cycles of freezing and thawing. Old mortar is adhered on the surface of RCA and the interfacial transition zone (ITZ) is formed in RAC. The interfacial transition zone (ITZ) is a weak part in concrete. Due to the higher porosity of RCA, the absorbed water in RCA easily gets saturated and upon freezing, and it develops internal stress. If this internal stress exceeds the tensile strength of RCA or the interfacial transition zone (ITZ), then microcracks will generate inside the RAC with the increase of freezing and thawing cycles. As Fig. 5 shows, the damage is a process from inside to outside specimens. Fig. 6 shows the surface of RAC specimens after 100 cycles of freezing and thawing and NAC specimens after 125 cycles of freezing and thawing. Though NAC specimens were subjected to more cycles of freezing and thawing, the surface of NAC specimens appeared less rough compared with RAC specimens. 3.2. Compressive strength The freezing and thawing durability performance of RAC and NAC was measured in terms of the compressive strength, the mass loss and the relative dynamic elastic modulus. As Fig. 7 shows, the relative cubic compressive strength of RAC decreased linearly with the increase of freezing and thawing cycles. The decline was approximate 5.5% every 25 cycles of freezing and thawing. The relative cubic compressive strength of NAC remained constant and even increased slightly within 50 cycles of freezing and thawing, but it began to decrease after 50 cycles of freezing and thawing. The reason may be that 0.2‰ air-entraining agent used in this test produced 4.7% air content in the concrete approximately, and independent closed pores can release the pressure of water freezing to some extent, thus slow down the internal cracks formation and development process. As a result, when the freezing and thawing cycles are less, the loss of compressive strength caused by the internal micro cracks is not obvious. The relative cubic compressive strength of NAC decreased more slowly than that of RAC within 125 cycles of freezing and thawing.
Fig. 4. Partial damage after 100 cycles of freezing and thawing.
The relative cubic compressive strength of NAC was larger than that of RAC after the same cycles of freezing and thawing, which shows that the freezing-thawing resistance of RAC with 100% replacement ratio was lower than that of NAC with the same mixture proportions. 3.3. Mass loss Based on GBT50082-2009, the mass loss can be calculated according to Eq. (3-1):
DW ni ¼
W oi W ni 100% W oi
ð3-1Þ
where DW ni – mass loss (%) of No.i specimen after n cycles of freezing and thawing; W oi – mass (g) of No.i specimen before n cycles of freezing and thawing; W ni – mass (g) of No.i specimen after n cycles of freezing and thawing. The average mass loss of a set of specimens can be calculated according to Eq. (3-2):
P3
DW n ¼
i¼1 DW ni
3
100%
ð3-2Þ
where DW n – average mass loss of a set of specimens after n cycles of freezing and thawing. Fig. 8 shows the mass loss of RAC and NAC with different cycles of freezing and thawing. Taking No. 6 group as an example, it can be observed that the mass of RAC specimens had been increasing with the increase of freezing and thawing cycles. From 0 to 25 cycles of freezing and thawing, it increased most rapidly nearly reaching 50% of the peak mass, but the mass of specimens did not continue to increase after 75 to 100 cycles of freezing and thawing.
(a) 0 cycles of freezing and thawing (b) 100 cycles of freezing and thawing (c) 150 cycles of freezing and thawing Fig. 3. Appearances of RAC after different cycles of freezing and thawing.
214
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
Fig. 5. Internal damage after 75 cycles of freezing and thawing.
(a) RAC100 freeze-thaw cycles
(b) NAC125 freeze-thaw cycles
Fig. 6. Comparison of surface damage between RAC and NAC.
Fig. 8. Comparison on mass loss between RAC and NAC.
Fig. 7. Comparison on relative cubic compressive strength between RAC and NAC.
There are two reasons for the change of mass of RAC. One is that, mortar spalls after freezing and thawing and internal cracks caused by the pressure of water freezing formed. Concrete became loose and spalled with the development and extend of the cracks which led the mass of specimens to decrease. On the other hand, those independent closed pores in concrete became connected by the power of water freezing after more cycles of freezing and thawing which made the water permeate into the concrete. This process
led to the increase in mass of specimens. The two aspects interacted with each other. When the absorption of water was greater than the loss of concrete spalled, the mass of specimens increased, on the contrary, it decreased. The mass of NAC specimens first increased and then decreased over freezing and thawing cycle duration. The mass of almost all of the specimens increased to some extent around 25 cycles of freezing and thawing. This is because the water absorption was greater than the loss of concrete spalled. But it started to decrease after 25 cycles of freezing and thawing which meant that the water absorption was smaller than the loss of concrete spalled. Due to the higher porosity and water absorption of RCA, the ability to absorb water of RAC was much higher than that of NAC. Although the specimens had been soaking for four days before the freezing and thawing test, the water could not totally permeate into the RCA due to the introduction of independent, uniform and closed pores produced by air-entraining agent. With the increase of freezing and thawing cycles, internal cracks formed and the pores became connected which made the water available to the RCA, and then led to the increase in the mass of specimens of RAC. The water absorption was greater than the loss of concrete spalled. So, the mass of RAC specimens increased until 75 cycles of freezing and thawing. But the surface of RAC spalled seriously and the loss of concrete spalled was higher than the water absorption after 75 cycles of freezing and thawing, which resulted in the decrease in the mass of RAC specimens.
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
215
3.4. Relative dynamic elastic modulus According to GB/T 50082-2009 [19], the relative dynamic elastic modulus of concrete can be calculated by Eqs. (3-3) and (3-4). 2
Pi ¼
f ni 2
f 0i
100
ð3-3Þ
100
ð3-4Þ
and
Pi ¼
V 2ni V 20I
where Pi – the relative dynamic elastic modulus of No. i specimen after n cycles of freezing and thawing (%); f 0i , V 0i – the fundamental transverse frequency and the ultrasonic pulse velocity No. i specimen before the freezing and thawing test, respectively; f ni , V ni – the fundamental transverse frequency and the ultrasonic pulse velocity No. i specimen after n cycles of freezing and thawing, respectively.
Fig. 9. Relative dynamic elastic modulus of RAC and NAC.
The average relative dynamic elastic modulus of a set of specimens can be calculated according to Eq. (3-5):
P¼
3 1X Pi 100% 3 1
ð3-5Þ
Fig. 9 shows the relative dynamic elastic modulus of RAC and NAC specimens after different cycles of freezing and thawing. As shown in Fig. 8, the relative dynamic elastic modulus decreased linearly with the increase of freezing and thawing cycles. The relative dynamic elastic modulus of RAC-4, RAC-6 and RAC-7 decreased to 60% after 75–100 cycles of freezing and thawing. The freezing and thawing resistance of concrete was determined by the air content when the concrete strength was low. The air content of the concrete in this test was 4.7%, but the freezing thawing resistance of RAC was still poor, which might be related to the low air content of RCA. Gokce [5] pointed out that RCAs produced from no-air-entrained concrete caused poor freezing and thawing resistance of RAC even when proper air content was introduced in the new mixed concrete using air entraining agent. So the relative dynamic elastic modulus of RAC specimens decreased faster than that of NAC specimens after same cycles of freezing and thawing. In order to verify the research conclusion of Gokce [5], the electronic microscope produced by Shenzhen Supereyes Technology Co., Ltd. was used to observe the internal structure of concrete. Figs. 10 and 11 are the pictures enlarged 200 times. Fig. 10 shows that the new mixed concrete contained more pores, which became connected with the accumulation of damage caused by freezing and thawing. This is the reason of the high water absorption of RAC and the change of mass loss. From Fig. 11 it can be seen that, the air content of RCA in the left was very low. At the same time, the adhered mortar on the surface of the RCAs became a weak link with the accumulation of freezing and thawing damage, which led to a low freezing and thawing resistance for RAC compared with that of NAC, especially when the air content of RCAs was extremely low or no air content exists inside at all as Gokce A [5] pointed out. The relative dynamic elastic modulus of NAC did not decrease in the range of 0–125 cycles of freezing and thawing. This was related to the air content in the concrete specimens. Those independent, uniform and closed pores could release the pressure generated by water freezing. The relative dynamic elastic modulus of NAC was larger than that of RAC under the same cycles of freezing and thawing, which
Fig. 10. Internal porosity of mortar.
Fig. 11. Interface between RCA and new mortar.
showed that the frost resistance of RAC was lower than that of NAC with the same mixture proportions. This was mainly due to the quality of RCAs. And there were also great differences in the freezing and thawing resistance for different RAC specimens, because of the randomness and inhomogeneity of RCAs.
216
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
When the strain reached 1–1.35 times of the peak strain, the first crack was visible on the surface at the mid-height of specimen. This crack was fine and short, and parallelled to the loading direction. As the strain of the specimen increased continuously, several short cracks occurred in the loading direction. The adhesive cracks on the aggregate boundary and the cracks in the interior of cement mortar in concrete gradually expanded, extended and were connected together. Eventually, a macroscopic inclined crack was formed along the weakest plane and run through the whole section. When the strain of the specimen increased further, the inclined crack was gradually widened under the actions of normal stress and shear stress, and became a damaged belt, but the cracks on the other parts of the specimen didn’t develop again. The load on the specimen was resisted by the friction and residual adhesive strength on the inclined plane, so the residual strength of concrete decreased slowly. When the failed specimen was separated into two piece, it was found that the failure plane passed through the interfacial transition zone (ITZ) and interior of the cement mortar in RAC, but no natural coarse aggregate (NAC) was broken. The typical failure patterns of RAC were shown in Fig. 12. 3.5.2. Measured stress-strain curves The complete compressive stress strain curve of RAC was converted into the dimensionless coordinates:
Fig. 12. Typical failure patterns of RAC.
RAC 4-2 aer 75 freeze/thaw cycles RAC 4-3 aer 75 freeze/thaw cycles RAC 5-1 aer 65 freeze/thaw cycles RAC 7-2 aer 100 freeze/thaw cycles
1.0
RAC 6-1 aer 125 freeze/thaw cycles 0.9
RAC 1-1 aer 0 freeze/thaw cycles
0.8
Stress/Peank stress
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
e r ; y¼ rc ec where rc is the peak stress (maximum stress), ec
x¼
is the peak strain (the strain corresponding to peak stress). The stress strain curves of RAC after freezing and thawing were shown in Fig. 13. The curves included the ascending branch and descending branch no matter how many the cycles of freezing and thawing are. The ascending branch could be divided into two parts. The first part represented the linear portion and the second one represented the nonlinear portion. When the stress was less than 60% of peak stress, the stress-strain curve was approximately linear, which illustrated that the concrete was still at the elastic stage. With the load increasing till to peak stress, the slope of curve decreased, which meant that the plastic deformation of concrete had occurred. For the descending branch, the stress decreased rapidly before the strain reached 1.5 times of peak strain, after which it became a process of slow decrease. Compared with the stress-strain curves after 0 and 125 cycles of freezing and thawing, the descending branch of the stress-strain curve of RAC became steeper, which meant the fragility of RAC increased with the increase of freezing and thawing cycles. The enveloped area of the curve represents the ductility of concrete specimens. From Fig. 13 it can be seen that the ductility of RAC specimens became lower with the increase of freezing and thawing cycles.
Strain/Peak strain Fig. 13. Stress strain curves of RAC after freezing and thawing.
3.5. Stress-strain curves of RAC
3.5.3. Theoretic stress-strain curves Referring to the conclusion obtained by Zhenhai Guo [19], the boundary conditions of stress-strain curve are following: (1) x ¼ 0; y ¼ 0; 2
3.5.1. Failure mode The stress-strain curve was almost linear up to about 40% of the peak stress. As the stress increased further, the plastic deformation and microcrack of concrete slightly developed and the strain rate of concrete gradually increased, so the slope of the stress-stain curve gradually decreased.
(2) When 0 6 x < 1; ddx2y < 0; that is the slope of the ascending branch (dy=dx) decreases and no point of inflection appears; ¼ 0, y ¼ 1, there is only one peak value in the (3) When x ¼ 1, dy dx curve; 2
(4) When ddx2y ¼ 0, x > 1, and there is one inflection point on the descending branch;
217
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219 Table 5 Parameters of the ascending branch and the descending branch. Specimen No.
Freeze thaw Cycles
Zhenghai Guo [19] model Ascending branch
1-1 2-1 2-2 2-3 3-1 3-2 3-3 5-1 4-2 4-3 7-2 6-1 6-2 *
0 25 25 25 50 50 50 65 75 75 100 125 125
Descending branch
a
R2
b
R2
2.704 1.626 –* 1.688 1.524 2.156 2.013 1.816 2.419 0.948 3.007 –* 0.303
0.999 0.997 –* 0.997 0.999 0.999 0.979 0.998 0.994 0.996 0.995 –* 0.997
0.171 –* 2.037 0.834 13.040 0.468 1.549 0.359 0.913 2.318 176.900 5.026 –*
0.997 –* 0.968 0.885 0.997 0.865 0.992 0.989 0.963 0.716 0.925 0.796 –*
– means the data was not collected.
Table 6 Parameter of the descending branch based on equation (3-6).
1.0 0.9
1-1 2-1 2-2 2-3 3-1 3-2 3-3 5-1 4-2 4-3 7-2 6-1 6-2
0 25 25 25 50 50 50 65 75 75 100 125 125
experimental value
Descending branch B
R2
0.769 –* 2.253 1.429 1.468 0.272 0.425 1.029 1.295 2.247 6.429 4.190 –*
0.949 –* 0.997 0.990 0.993 0.988 0.957 0.970 0.987 0.960 0.981 0.980 –*
0.8
theorecal value
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
– means the data was not collected.
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 15. Complete stress strain curve of RAC after 25 cycles of freezing and thawing.
1.0 0.9
experimental value
0.8
Stress/Peak stress
*
Freeze Thaw Cycles
Stress/Peak stress
Specimen Number
theorecal value
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 14. Relationship of B and n.
3
(5) When ddx3y ¼ 0, x > 1, the point with the maximum slope appears on the descending branch; (6) When x ! 1; y ! 0; dy ! 0, the descending branch of the dx curve can be infinitely extended;
Fig. 16. Complete stress strain curve of RAC after 75 cycles of freezing and thawing.
(7) x P 0; 0 < y 6 1. The equation of the complete stress strain curves of normal concrete, which was proposed by Prof. Zhenghai Guo [21], is shown as follows:
218
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
(
1.0
y¼
0.9
Stress/Peak stress
theorecal value
0.7
x¼
0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 17. Complete stress strain curve of RAC after 100 cycles of freezing and thawing.
1.0 0.9
experimental… theorecal value
Stress/Peak stress
0.8 0.7 0.6
0.4
x¼
0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 18. Complete stress-strain curve of RAC after 125 cycles of freezing and thawing.
1.0 0.9
experimental value from Yongkang Shang[14]
0.8
theorecal value
0.7 0.6
ð3-6Þ
where ec is the peak strain; rc is the peak stress. a; b are parameters to be determined in the ascending branch and descending branch, respectively. The First Optimization software developed by 7D-Soft High Technology Inc. was used to find the optimum parameters of the ascending branch and the descending branch according to the test data based on Zhenghai Guo [19] model. The calculation results are shown in Table 5, and it can be seen that, for the ascending branch, the correlation coefficients are all greater than 0.97, which illustrates the equation (3-6) is suitable for the calculation of RAC after more cycles of freezing and thawing. However, with the increase of freezing and thawing cycles, the brittleness of RAC increases, the slope of the descending branch decreases rapidly, the correlation coefficients becomes smaller, which illustrates the equation (3-6) is not suitable for the calculation of RAC after more cycles of freezing and thawing. So the descending branch of the stress-strain curve was addressed in the research. According to the characteristics that the descending branch slope of the complete stress strain curves decreases rapidly for RAC after freezing and thawing, the calculation formula of Zhenhai Guo is modified as equation (3-7) shows.
y¼
0.3
P1
e r ; y¼ rc ec
(
0.5
0.2
Stress/Peak stress
x ;x bðx1Þ2 þx
experimental value
0.8
ax þ ð3 2aÞx2 þ ða 2Þx3 ; x 6 1
ax þ ð3 2aÞx2 þ ða 2Þx3 ; x 6 1 x ;x Bðx1Þþx
P1
ð3-7Þ
e r ;y ¼ rc ec
where ec is the peak strain; rc is the peak stress. a; B are parameters to be determined in the ascending branch and descending branch, respectively. The First Optimization software was also used to find the optimum parameter of the descending branch according to the test data based on equation (3-7). The calculation results are shown in Table 6. Comparing with the parameters calculated with Zhenhai Guo [19] model, the correlation coefficients of parameters calculated with equation (3-7) are all greater than 0.95, which indicates that the calculation results are agreement with the test data and the model can well represent the descending branch of the complete stress strain curves of RAC after freezing and thawing cycles. Taking the parameter B for the descending branch as the dependent variable, the freezing and thawing cycles n as the independent variable, coordinates are given and the mathematical calculation model with B and n is fitted as shown in Fig. 14.
B ¼ 9 106 n3 þ 0:0024n2 0:1689n þ 4:2712
0.5
R2 ¼ 0:9918
0.4 0.3 0.2 0.1 0.0
Strain/Peak strain Fig. 19. Verification of theoretical stress-strain curves of RAC after freezing and thawing.
where B is the parameter for the descending branch, n is cycles of freezing and thawing of RAC with 100% RCA replacement ratio. Figs. 15–18 are the complete stress-strain curves of RAC with 100% RCA replacement ratio after different cycles of freezing and thawing, and it can be seen that the theoretical curves are in good agreement with the measured curves. After correlation calculation, the correlation coefficient R of Figs. 15–18 are 0.996, 0.989, 0.991 and 0.768, respectively. The correlation coefficient of the complete stress-strain curve of RAC after 125 cycles of freezing and thawing is only 0.768, much smaller than that of the other curves. The reason may be that the discreteness of RAC is stronger than that of NAC due to the discreteness of RCA, especially when the cycles of
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
freezing and thawing is larger. From the calculation results, it is shown that the theoretic stress-strain curve presented in this paper has a good correlation with the experimental results. 3.5.4. Verification of theoretical stress-strain curves The experimental data of group NRC31.5-D100 from Yongkang Shang [14] was collected, for which, the replacement rate of RCA is 100%, RCA has no air inside, the maximum particle size of aggregate is 31.5 mm, and the freeze-thaw cycles is 100. The calculated and experimental results were shown in Fig. 19. It was found that the theoretical curves using the model proposed in this paper are in good agreement with the experimental results. 4. Conclusions (1) Mass loss of RAC decreased first then increased with the increase of freezing and thawing cycles. (2) The relative cubic compressive strength of RAC is lower than that of NAC under the same freezing and thawing cycles. (3) The relative dynamic elastic modulus of RAC decreased linearly with the increase of freezing and thawing cycles. The relative dynamic elastic modulus of RAC is lower than that of NAC after the same cycles of freezing and thawing. (4) The stress strain curve of RAC after freezing and thawing first ascends rapidly and then descends rapidly after exceeding the peak stress, finally turns into a slow decrease process. Theoretical stress-strain curve of RAC with 100% RCA replacement ratio after different freezing and thawing cycles was presented, which are in good agreement with the experimental results from other researcher.
References [1] Hong Zhang, The Wonderful and the Helpless of the Concrete Industry in China, China Building Materials Network, 2014. http://www.cbmd.cn/sxy/Info5620.html (in Chinese).
219
[2] Jingpeng Hou, Wei Shi, Yupu Song, Development and application promoting of recycled concrete, Archit. Technol. 33 (l) (2002) 15–17 (in Chinese). [3] R. Zaharieva, F. Buyle Bodin, E. Wirquin, Frost resistance of recycled aggregate concrete, Cem. Concr. Res. 34 (10) (2004) 1927–1932. [4] Y. Kasai, M. Hisaka, K. Yanagi, Durability of concrete using recycled concrete, in: Proc. Int. RILEM Symp. on Demolition and reuse of concrete and Masonary, Taylor and Francis, London, 1988, pp. 623–632. [5] A. Gokce, S. Nagataki, T. Saeki, M. Hisada, Freezing and thawing resistance of air-entrained concrete incorporating recycled coarse aggregate: The role of air content in demolished concrete, Cem. Concr. Res. 34 (5) (2004) 799–806. [6] R.M. Salem, E.G. Burdette, N.M. Jackson, Resistance to freezing and thawing of recycled aggregate concrete, ACI Mater. J. 100 (3) (2003) 216–221. [7] Y. Haitao, T. Shizhu, Preparation and properties of high-strength recycled concrete in cold areas, Materiales de Construcción 65 (318) (2015) 1–6. [8] A. Richardson, K. Coventry, J. Bacon, Freeze/thaw durability of concrete with recycled demolition aggregate compared to virgin aggregate concrete, J. Cleaner Prod. 19 (2) (2011) 272–277. [9] J. Xiao, J. Li, C. Zhang, Mechanical properties of recycled aggregate concrete under uniaxial loading, Cem. Concr. Res. 35 (6) (2005) 1187–1194. [10] Ting Du, Weihong Wang, Zhongxin Liu, The complete stress-strain curve of recycled aggregate concrete under uniaxial compression loading, J. Wuhan Univ. Technol.-Mater. Sci. Ed. 25 (5) (2010) 862–865. [11] Code for design of concrete structures (GB50010-2010), China Architecture & Building Press, Beijing, 2010 (in Chinese). [12] S.B. Huda, M. Shahria Alam, Mechanical and freeze-thaw durability properties of recycled aggregate concrete made with recycled coarse aggregate, J. Mater. Civ. Eng. (2015). 04015003-19. [13] González-Fonteboa Belén, Martínez-Abella Fernando, Carro López Diego, et al., Stress-strain relationship in axial compression for concrete using recycled saturated coarse aggregate, Constr. Build. Mater. 25 (5) (2011) 2335–2342. [14] Yongkang Shang, The Experimental Study on the Mechanical Properties and Frost Resistance of Recycled Aggregate Concrete, Harbin Institute of Technology, 2010 (in Chinese). [15] Standard for Technical Requirements and Test Method of Gravel and Crushed Stone for Ordinary Concrete (JGJ 52-2006), China Architecture & Building Press, Beijing, 2006 (in Chinese). [16] Recycled Aggregate for Concrete (GB/T25177-2012), China Architecture & Building Press, Beijing, 2010 (in Chinese). [17] Code for Concrete Admixture Application (GB 50119-2013), China Architecture & Building Press, Beijing, 2014 (in Chinese). [18] Standard for Test Method of Long-term Performance and Durability of Ordinary Concrete (GB/T 50082-2009), China Architecture & Building Press, Beijing, 2009 (in Chinese). [19] Zhenhai Guo, Concrete Strength and Constitutive Relation-Principle and Application, China Building Industry Press, Beijing, 2004. 17–18 (in Chinese).
Contents lists available at ScienceDirect
Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat
Uni-axial compressive stress-strain relation of recycled coarse aggregate concrete after freezing and thawing cycles Jin Wu ⇑, Xianhang Jing, Zhe Wang Department of Civil Engineering, Nanjing University of Aeronautics & Astronautics, Nanjing 210016, China
h i g h l i g h t s The complete compressive stress-strain curves of RAC after freezing and thawing cycles were measured in this study. Theoretic stress-strain curve of RAC after freezing and thawing cycles was presented. Parameters of descending branch in stress-strain curve of RAC were given after freezing and thawing cycles.
a r t i c l e
i n f o
Article history: Received 1 February 2016 Received in revised form 17 October 2016 Accepted 21 December 2016
Keywords: Recycled aggregate concrete Freezing and thawing Mass loss Dynamic elastic modulus Compressive strength Stress-strain curve
a b s t r a c t The mass loss, compressive strength, dynamic elastic modulus and stress-strain relationship of recycled coarse aggregate concrete under different cycles of freezing and thawing were investigated by comparison with normal concrete in the research reported in this paper. Forty-eight prism specimens with the size of 100 mm 100 mm 300 mm and ninety-six cubic specimens with the size of 100 mm 100 mm 100 mm were fabricated and tested. Test results show that, the mass loss of recycled coarse aggregate concrete decreased first then increased with the increase of freezing and thawing cycles; the relative cubic compressive strength and the relative dynamic elastic modulus of recycled aggregate coarse concrete decreased linearly with the increase of freezing and thawing cycles. The stress-strain curve of recycled aggregate concrete first ascended rapidly and then descended rapidly after exceeding the peak stress till to 1.5 times of the peak strain, finally turned into a slow decrease process. The ductility of recycled coarse aggregate concrete became worse with the increase of freezing and thawing cycles. Theoretic stress-strain curve of recycled coarse aggregate concrete after different cycles of freezing and thawing was presented and it was in good agreement with the experimental results from the other researcher. Ó 2016 Elsevier Ltd. All rights reserved.
1. Introduction With the development of construction industry, a lot of waste demolished concrete is generated every year. In 2010, the amount of construction and demolition (CD) waste in China is about 1.55 billion tons, accounting for 30–40% of the total waste [1]. If this trend will continue, landfills will be saturated and environment will be polluted. Therefore, how to dispose of the huge amount of waste has become interests of many researchers. It has been reported that this huge amount of CD waste can be potentially used as recycled aggregates (RAs) for the production of recycled aggregate concrete (RAC). And the use of such CD waste as RAs will not only reduce landfills use but also help to minimize the use of natural aggregates (NAs) [2]. ⇑ Corresponding author. E-mail address: [email protected] (J. Wu). http://dx.doi.org/10.1016/j.conbuildmat.2016.12.142 0950-0618/Ó 2016 Elsevier Ltd. All rights reserved.
The behavior of RAC is significantly affected by its durability, and the freezing-thawing durability is one of the major concerns associated with the application of RAC. Some studies on the freezing-thawing durability of RAC have been investigated. Zaharieva et al. [3] pointed out that the freezing-thawing resistance of RAC is lower than that of natural concrete (NC). The main reason seems to be higher porosity and lower the freezing-thawing resistance of RAs themselves. Kasai et al. [4] found that a high replacement ratio of RAs declines the freezing-thawing resistance of RAC. Gokce et al. [5] investigated the freezing-thawing resistance of RAC. Test results showed that RAC made with an air-entrained admixture had better performance than RAC made with non-airentrained admixture. Salem et al. [6] reported that RAs originated from concrete made with an air-entrained admixture produced high-quality freezing-thawing resistance of RAC. Haitao et al. [7] conducted a research on the freezing-thawing resistance of high strength (50 MPa) RAC made with 100% replacement of RAs. Test
211
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
results showed that, compressive strength, splitting tensile strength and bending strength of RAC decreased with the increase of freezing and thawing cycles. After 50 cycles of freezing and thawing, the bending strength decreased by 40%. Richardson et al. [8] conducted an experiment on the freezing-thawing resistance of RAC based on the ASTM 666 standards. The test results showed that RAs with the addition of additives can be used for applications where freezing and thawing of concrete occur whilst still providing the durability that NAs offer. The stress-strain relationship of RAC is important in theoretical and numerical analysis as well as engineering design of RAC structures. In recent years, several investigations have been performed for the stress-strain relation of RAC. Xiao et al. [9] investigated the compressive strength and stressstrain relation of RAC with different replacement ratio of RAs, and obtained the approximate stress-strain curves of RAC according to the analytical expression in Chinese Code GB50010 [10] for uniaxial compression of NC. Du et al. [11] investigated the complete stress-strain curve of RAC with 100% replacement ratio of RAs, and constituted the model of stress-strain of RAC under uniaxial compression loading. Huda et al. [12] conducted a research on the stress-strain curves at the age of 120 days of RAC with different RAs replacement ratio, and found that the pattern of the stress-strain curves was similar for all RAC mixes but the value of the strain corresponding to the peak stress was higher for RAC compared to that of NC. Belén et al. [13] also developed an analytical expression of the stress–strain curve of RAC using the experimental results, and verified the proposed model equation by comparing it to the experimental data. The results showed that the proposed model equation satisfactorily describes the effect of RAs on the stress–strain curve. However, very limited information is available on the stressstrain relations of RAC after freezing and thawing. A previous research study by Shang [14] investigated the compressive stress-strain relations of RAC after freezing and thawing, and presented a calculation model based on NC stress-strain relation, but freezing and thawing cycles were not considered in the model, and a larger deviation between the theoretical values and experimental values can be seen in the descending branch of stress-strain curves. The objective of this study is to investigate the influence of freezing and thawing cycles on the mass loss, the compressive strength, dynamic elastic modulus and stress-strain curve of RAC with 100% replacement ratio of RAs, and to present an analytical expression for stress-strain relationship of RAC after different freezing and thawing cycles. The results presented in the study are significant for using the RAC in cold areas. 2. Experimental program 2.1. Materials 2.1.1. Recycled coarse aggregate (RCA) There is great randomness and variability for RCA made of different sources of original concrete. To ensure the unification of source of RCA, the waste concrete in this test was totally from waste concrete blocks demolished from a cement road in Nanjing university Aeronautics and Astronautics. After crushing, washing and grading into 5–31.5 mm continuous gradation, RCA conforming the standard was prepared. Table 1 shows the grading profile of RCA. Water absorption and crushing index of RCA are 5.7% and 10.4%, respectively, being measured according to a method in JGJ 52-2006 [15]. The RCA used in the study conformed to the requirements of Grade Ⅱ [16].
Table 1 Grading profile of RCA. Sieve Size (mm)
Grader Retained (%)
Accumulated Retained (%)
Continuous Gradation (%)
31.5 25.0 20.0 16.0 10.0 5.0
2.25 18.21 15.76 14.38 32.52 10.41
2.25 20.46 36.22 50.60 83.12 93.53
5–0 45–15 90–70 100–90
2.1.2. Natural coarse aggregate (NCA) The NCA used in this experiment was made of natural stone, the size of particles was 5 mm to 40 mm. 2.1.3. Fine aggregate The natural river sand used in this test was medium sand. The performance indicators conform the requirements in JGJ 52-2006 [15], grading profiles and performance of sand are shown in Tables 2 and 3, respectively. 2.1.4. Cement The cement used in this test was ASTM Type Ⅱ Portland cement produced by the China Jiangnan Cement Co., LTD. Its specific surface area was 385 m2/kg. 2.1.5. Admixtures The admixtures used in this test were GYQÒ-III concrete airentraining agent and PCAÒ (I) carboxylic acid high range water reducer provided by Jiangsu Bote New Materials Co., LTD. The quality of admixtures conformed to GB 50119-2013 [17]. Before casting specimens, twelve groups of proportion tests were conducted in order to obtain the optimum dosage of air-entraining agent and water reducer. 2.1.6. Release agent GBT 50082-2009 [18] specifies that it is forbidden to use hydrophilic release agent molding specimens for freezing and thawing tests. The DL-J04 hydrophilic release agent used in this test came from Shenzhen Xindeli chemical Co., LTD. The main components of the release agent are polymer organic matters and can be diluted with water in any proportion conforming to the requirements of the test.
Table 2 Grading profile of sand. Sieve Size (mm)
Grader Retained (%)
Accumulated Retained (%)
Continuous Gradation (%)
5.00 2.50 1.25 0.630 0.315 0.160
3.20 9.16 8.13 25.75 45.70 29.70
3.33 12.49 20.62 46.37 92.07 93.30
10–0 25–0 50–10 70–41 92–70 100–90
Table 3 Performance of sand. Fineness Module
Mud Content (%)
Particle Diameter (mm)
2.6
2.1
<5
212
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
2.2. Mix proportion Forty-eight prism specimens (100 mm 100 mm 300 mm) and ninety-six cubic specimens (100 mm 100 mm 100 mm) were cast. Details of the mix proportions are listed in Table 4. All specimens were fabricated in a laboratory and cured for 24 h at room temperature, then demolded and cured for 28 days. 2.3. Test method (1) Specimens for freezing and thawing were placed in 20 ± 2 °C water for 4 days before the test. (2) After soaking water for enough time, specimens were removed out of the water, wiping moisture on the surface. The initial mass of the specimens was weighed and the initial transverse fundamental frequency was collected and recorded. (3) The concrete specimens were placed in the freeze-thaw boxes and filled with water level about 5 mm higher than the top of the specimens. (4) The specimen boxes were placed in the freezing and thawing machine. It began to freeze and thaw after the temperature sensors were placed. 2.4. Test equipment 2.4.1. Equipment of collecting the data of relative dynamic elastic modulus To collect the data of relative dynamic elastic modulus, a nonmetal ultrasonic detection analyzer was used in this study to measure the ultrasonic velocity of ultrasonic wave. The test set-up is shown in Fig. 1. 2.4.2. Equipment of collecting the data of stress and strain The stress state of the concrete will be changed due to the effect of the friction on the end of the concrete specimen. According to
Fig. 2. Load sensor and displacement sensor arrangement.
Saint & Venant’s principle, non-uniform vertical compressive stress will only have a significant impact on the stress state of the end of the specimen with a height approximately equal to the width of the specimen. In order to obtain a more accurate stress state under uniaxial compression, the strain in the middle part of the specimen (170–180 mm) was measured in this test as Fig. 2 showed. The hydraulic servo test system was used in the test of the axial compression of concrete after freezing and thawing. Dynamic data acquisition system was used to collect the displacements and loads. The axial loads applied on the prism specimens were controlled by strain, and the strain rate was 400 le/min. There was a load sensor on the end of loading device which can directly measure the magnitude of applied loads. A displacement sensor was installed on the two opposite sides to measure the longitudinal deformation. The dynamic data acquisition system was used to collect the test data.
Table 4 Mixture proportions. Types
RAC NAC *
Specimen Number
Water* (kg/mt)
Cement (kg/m3)
Sand (kg/m3)
1–8 9–16
186.0 150.0
404.7 404.7
646.5 646.5
Coarse Aggregate RCA (kg/m3)
NCA (kg/m3)
1011.6 0
0 1011.6
Pre-water for RCA is included.
Fig. 1. Test set-up for dynamic elastic modulus.
Water Reducing Agent (kg/m3)
Air-entraining Agent (kg/m3)
2.430 2.430
0.243 0.243
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
213
3. Experimental results 3.1. Specimens failure pattern With the increase of freezing and thawing cycles, the surface of RAC specimens became rough. It could be observed that lots of pits appeared, mortar spalled, micro cracks formed and extended gradually, some corners of specimens spalled and fine aggregates and coarse aggregates exposed. Fig. 3 shows the appearances of RAC after different cycles of freezing and thawing. As Fig. 3 shows, after 100 cycles of freezing and thawing, a lot of micro cracks appeared and some cracks were radial and cracks formed and extended based on a certain spot as the center. Fig. 4 shows the partial radial cracks on the end of the specimens after 100 cycles of freezing and thawing. Old mortar is adhered on the surface of RCA and the interfacial transition zone (ITZ) is formed in RAC. The interfacial transition zone (ITZ) is a weak part in concrete. Due to the higher porosity of RCA, the absorbed water in RCA easily gets saturated and upon freezing, and it develops internal stress. If this internal stress exceeds the tensile strength of RCA or the interfacial transition zone (ITZ), then microcracks will generate inside the RAC with the increase of freezing and thawing cycles. As Fig. 5 shows, the damage is a process from inside to outside specimens. Fig. 6 shows the surface of RAC specimens after 100 cycles of freezing and thawing and NAC specimens after 125 cycles of freezing and thawing. Though NAC specimens were subjected to more cycles of freezing and thawing, the surface of NAC specimens appeared less rough compared with RAC specimens. 3.2. Compressive strength The freezing and thawing durability performance of RAC and NAC was measured in terms of the compressive strength, the mass loss and the relative dynamic elastic modulus. As Fig. 7 shows, the relative cubic compressive strength of RAC decreased linearly with the increase of freezing and thawing cycles. The decline was approximate 5.5% every 25 cycles of freezing and thawing. The relative cubic compressive strength of NAC remained constant and even increased slightly within 50 cycles of freezing and thawing, but it began to decrease after 50 cycles of freezing and thawing. The reason may be that 0.2‰ air-entraining agent used in this test produced 4.7% air content in the concrete approximately, and independent closed pores can release the pressure of water freezing to some extent, thus slow down the internal cracks formation and development process. As a result, when the freezing and thawing cycles are less, the loss of compressive strength caused by the internal micro cracks is not obvious. The relative cubic compressive strength of NAC decreased more slowly than that of RAC within 125 cycles of freezing and thawing.
Fig. 4. Partial damage after 100 cycles of freezing and thawing.
The relative cubic compressive strength of NAC was larger than that of RAC after the same cycles of freezing and thawing, which shows that the freezing-thawing resistance of RAC with 100% replacement ratio was lower than that of NAC with the same mixture proportions. 3.3. Mass loss Based on GBT50082-2009, the mass loss can be calculated according to Eq. (3-1):
DW ni ¼
W oi W ni 100% W oi
ð3-1Þ
where DW ni – mass loss (%) of No.i specimen after n cycles of freezing and thawing; W oi – mass (g) of No.i specimen before n cycles of freezing and thawing; W ni – mass (g) of No.i specimen after n cycles of freezing and thawing. The average mass loss of a set of specimens can be calculated according to Eq. (3-2):
P3
DW n ¼
i¼1 DW ni
3
100%
ð3-2Þ
where DW n – average mass loss of a set of specimens after n cycles of freezing and thawing. Fig. 8 shows the mass loss of RAC and NAC with different cycles of freezing and thawing. Taking No. 6 group as an example, it can be observed that the mass of RAC specimens had been increasing with the increase of freezing and thawing cycles. From 0 to 25 cycles of freezing and thawing, it increased most rapidly nearly reaching 50% of the peak mass, but the mass of specimens did not continue to increase after 75 to 100 cycles of freezing and thawing.
(a) 0 cycles of freezing and thawing (b) 100 cycles of freezing and thawing (c) 150 cycles of freezing and thawing Fig. 3. Appearances of RAC after different cycles of freezing and thawing.
214
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
Fig. 5. Internal damage after 75 cycles of freezing and thawing.
(a) RAC100 freeze-thaw cycles
(b) NAC125 freeze-thaw cycles
Fig. 6. Comparison of surface damage between RAC and NAC.
Fig. 8. Comparison on mass loss between RAC and NAC.
Fig. 7. Comparison on relative cubic compressive strength between RAC and NAC.
There are two reasons for the change of mass of RAC. One is that, mortar spalls after freezing and thawing and internal cracks caused by the pressure of water freezing formed. Concrete became loose and spalled with the development and extend of the cracks which led the mass of specimens to decrease. On the other hand, those independent closed pores in concrete became connected by the power of water freezing after more cycles of freezing and thawing which made the water permeate into the concrete. This process
led to the increase in mass of specimens. The two aspects interacted with each other. When the absorption of water was greater than the loss of concrete spalled, the mass of specimens increased, on the contrary, it decreased. The mass of NAC specimens first increased and then decreased over freezing and thawing cycle duration. The mass of almost all of the specimens increased to some extent around 25 cycles of freezing and thawing. This is because the water absorption was greater than the loss of concrete spalled. But it started to decrease after 25 cycles of freezing and thawing which meant that the water absorption was smaller than the loss of concrete spalled. Due to the higher porosity and water absorption of RCA, the ability to absorb water of RAC was much higher than that of NAC. Although the specimens had been soaking for four days before the freezing and thawing test, the water could not totally permeate into the RCA due to the introduction of independent, uniform and closed pores produced by air-entraining agent. With the increase of freezing and thawing cycles, internal cracks formed and the pores became connected which made the water available to the RCA, and then led to the increase in the mass of specimens of RAC. The water absorption was greater than the loss of concrete spalled. So, the mass of RAC specimens increased until 75 cycles of freezing and thawing. But the surface of RAC spalled seriously and the loss of concrete spalled was higher than the water absorption after 75 cycles of freezing and thawing, which resulted in the decrease in the mass of RAC specimens.
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
215
3.4. Relative dynamic elastic modulus According to GB/T 50082-2009 [19], the relative dynamic elastic modulus of concrete can be calculated by Eqs. (3-3) and (3-4). 2
Pi ¼
f ni 2
f 0i
100
ð3-3Þ
100
ð3-4Þ
and
Pi ¼
V 2ni V 20I
where Pi – the relative dynamic elastic modulus of No. i specimen after n cycles of freezing and thawing (%); f 0i , V 0i – the fundamental transverse frequency and the ultrasonic pulse velocity No. i specimen before the freezing and thawing test, respectively; f ni , V ni – the fundamental transverse frequency and the ultrasonic pulse velocity No. i specimen after n cycles of freezing and thawing, respectively.
Fig. 9. Relative dynamic elastic modulus of RAC and NAC.
The average relative dynamic elastic modulus of a set of specimens can be calculated according to Eq. (3-5):
P¼
3 1X Pi 100% 3 1
ð3-5Þ
Fig. 9 shows the relative dynamic elastic modulus of RAC and NAC specimens after different cycles of freezing and thawing. As shown in Fig. 8, the relative dynamic elastic modulus decreased linearly with the increase of freezing and thawing cycles. The relative dynamic elastic modulus of RAC-4, RAC-6 and RAC-7 decreased to 60% after 75–100 cycles of freezing and thawing. The freezing and thawing resistance of concrete was determined by the air content when the concrete strength was low. The air content of the concrete in this test was 4.7%, but the freezing thawing resistance of RAC was still poor, which might be related to the low air content of RCA. Gokce [5] pointed out that RCAs produced from no-air-entrained concrete caused poor freezing and thawing resistance of RAC even when proper air content was introduced in the new mixed concrete using air entraining agent. So the relative dynamic elastic modulus of RAC specimens decreased faster than that of NAC specimens after same cycles of freezing and thawing. In order to verify the research conclusion of Gokce [5], the electronic microscope produced by Shenzhen Supereyes Technology Co., Ltd. was used to observe the internal structure of concrete. Figs. 10 and 11 are the pictures enlarged 200 times. Fig. 10 shows that the new mixed concrete contained more pores, which became connected with the accumulation of damage caused by freezing and thawing. This is the reason of the high water absorption of RAC and the change of mass loss. From Fig. 11 it can be seen that, the air content of RCA in the left was very low. At the same time, the adhered mortar on the surface of the RCAs became a weak link with the accumulation of freezing and thawing damage, which led to a low freezing and thawing resistance for RAC compared with that of NAC, especially when the air content of RCAs was extremely low or no air content exists inside at all as Gokce A [5] pointed out. The relative dynamic elastic modulus of NAC did not decrease in the range of 0–125 cycles of freezing and thawing. This was related to the air content in the concrete specimens. Those independent, uniform and closed pores could release the pressure generated by water freezing. The relative dynamic elastic modulus of NAC was larger than that of RAC under the same cycles of freezing and thawing, which
Fig. 10. Internal porosity of mortar.
Fig. 11. Interface between RCA and new mortar.
showed that the frost resistance of RAC was lower than that of NAC with the same mixture proportions. This was mainly due to the quality of RCAs. And there were also great differences in the freezing and thawing resistance for different RAC specimens, because of the randomness and inhomogeneity of RCAs.
216
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
When the strain reached 1–1.35 times of the peak strain, the first crack was visible on the surface at the mid-height of specimen. This crack was fine and short, and parallelled to the loading direction. As the strain of the specimen increased continuously, several short cracks occurred in the loading direction. The adhesive cracks on the aggregate boundary and the cracks in the interior of cement mortar in concrete gradually expanded, extended and were connected together. Eventually, a macroscopic inclined crack was formed along the weakest plane and run through the whole section. When the strain of the specimen increased further, the inclined crack was gradually widened under the actions of normal stress and shear stress, and became a damaged belt, but the cracks on the other parts of the specimen didn’t develop again. The load on the specimen was resisted by the friction and residual adhesive strength on the inclined plane, so the residual strength of concrete decreased slowly. When the failed specimen was separated into two piece, it was found that the failure plane passed through the interfacial transition zone (ITZ) and interior of the cement mortar in RAC, but no natural coarse aggregate (NAC) was broken. The typical failure patterns of RAC were shown in Fig. 12. 3.5.2. Measured stress-strain curves The complete compressive stress strain curve of RAC was converted into the dimensionless coordinates:
Fig. 12. Typical failure patterns of RAC.
RAC 4-2 aer 75 freeze/thaw cycles RAC 4-3 aer 75 freeze/thaw cycles RAC 5-1 aer 65 freeze/thaw cycles RAC 7-2 aer 100 freeze/thaw cycles
1.0
RAC 6-1 aer 125 freeze/thaw cycles 0.9
RAC 1-1 aer 0 freeze/thaw cycles
0.8
Stress/Peank stress
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
e r ; y¼ rc ec where rc is the peak stress (maximum stress), ec
x¼
is the peak strain (the strain corresponding to peak stress). The stress strain curves of RAC after freezing and thawing were shown in Fig. 13. The curves included the ascending branch and descending branch no matter how many the cycles of freezing and thawing are. The ascending branch could be divided into two parts. The first part represented the linear portion and the second one represented the nonlinear portion. When the stress was less than 60% of peak stress, the stress-strain curve was approximately linear, which illustrated that the concrete was still at the elastic stage. With the load increasing till to peak stress, the slope of curve decreased, which meant that the plastic deformation of concrete had occurred. For the descending branch, the stress decreased rapidly before the strain reached 1.5 times of peak strain, after which it became a process of slow decrease. Compared with the stress-strain curves after 0 and 125 cycles of freezing and thawing, the descending branch of the stress-strain curve of RAC became steeper, which meant the fragility of RAC increased with the increase of freezing and thawing cycles. The enveloped area of the curve represents the ductility of concrete specimens. From Fig. 13 it can be seen that the ductility of RAC specimens became lower with the increase of freezing and thawing cycles.
Strain/Peak strain Fig. 13. Stress strain curves of RAC after freezing and thawing.
3.5. Stress-strain curves of RAC
3.5.3. Theoretic stress-strain curves Referring to the conclusion obtained by Zhenhai Guo [19], the boundary conditions of stress-strain curve are following: (1) x ¼ 0; y ¼ 0; 2
3.5.1. Failure mode The stress-strain curve was almost linear up to about 40% of the peak stress. As the stress increased further, the plastic deformation and microcrack of concrete slightly developed and the strain rate of concrete gradually increased, so the slope of the stress-stain curve gradually decreased.
(2) When 0 6 x < 1; ddx2y < 0; that is the slope of the ascending branch (dy=dx) decreases and no point of inflection appears; ¼ 0, y ¼ 1, there is only one peak value in the (3) When x ¼ 1, dy dx curve; 2
(4) When ddx2y ¼ 0, x > 1, and there is one inflection point on the descending branch;
217
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219 Table 5 Parameters of the ascending branch and the descending branch. Specimen No.
Freeze thaw Cycles
Zhenghai Guo [19] model Ascending branch
1-1 2-1 2-2 2-3 3-1 3-2 3-3 5-1 4-2 4-3 7-2 6-1 6-2 *
0 25 25 25 50 50 50 65 75 75 100 125 125
Descending branch
a
R2
b
R2
2.704 1.626 –* 1.688 1.524 2.156 2.013 1.816 2.419 0.948 3.007 –* 0.303
0.999 0.997 –* 0.997 0.999 0.999 0.979 0.998 0.994 0.996 0.995 –* 0.997
0.171 –* 2.037 0.834 13.040 0.468 1.549 0.359 0.913 2.318 176.900 5.026 –*
0.997 –* 0.968 0.885 0.997 0.865 0.992 0.989 0.963 0.716 0.925 0.796 –*
– means the data was not collected.
Table 6 Parameter of the descending branch based on equation (3-6).
1.0 0.9
1-1 2-1 2-2 2-3 3-1 3-2 3-3 5-1 4-2 4-3 7-2 6-1 6-2
0 25 25 25 50 50 50 65 75 75 100 125 125
experimental value
Descending branch B
R2
0.769 –* 2.253 1.429 1.468 0.272 0.425 1.029 1.295 2.247 6.429 4.190 –*
0.949 –* 0.997 0.990 0.993 0.988 0.957 0.970 0.987 0.960 0.981 0.980 –*
0.8
theorecal value
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
– means the data was not collected.
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 15. Complete stress strain curve of RAC after 25 cycles of freezing and thawing.
1.0 0.9
experimental value
0.8
Stress/Peak stress
*
Freeze Thaw Cycles
Stress/Peak stress
Specimen Number
theorecal value
0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 14. Relationship of B and n.
3
(5) When ddx3y ¼ 0, x > 1, the point with the maximum slope appears on the descending branch; (6) When x ! 1; y ! 0; dy ! 0, the descending branch of the dx curve can be infinitely extended;
Fig. 16. Complete stress strain curve of RAC after 75 cycles of freezing and thawing.
(7) x P 0; 0 < y 6 1. The equation of the complete stress strain curves of normal concrete, which was proposed by Prof. Zhenghai Guo [21], is shown as follows:
218
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
(
1.0
y¼
0.9
Stress/Peak stress
theorecal value
0.7
x¼
0.6 0.5 0.4 0.3 0.2 0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 17. Complete stress strain curve of RAC after 100 cycles of freezing and thawing.
1.0 0.9
experimental… theorecal value
Stress/Peak stress
0.8 0.7 0.6
0.4
x¼
0.1 0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Strain/Peak strain Fig. 18. Complete stress-strain curve of RAC after 125 cycles of freezing and thawing.
1.0 0.9
experimental value from Yongkang Shang[14]
0.8
theorecal value
0.7 0.6
ð3-6Þ
where ec is the peak strain; rc is the peak stress. a; b are parameters to be determined in the ascending branch and descending branch, respectively. The First Optimization software developed by 7D-Soft High Technology Inc. was used to find the optimum parameters of the ascending branch and the descending branch according to the test data based on Zhenghai Guo [19] model. The calculation results are shown in Table 5, and it can be seen that, for the ascending branch, the correlation coefficients are all greater than 0.97, which illustrates the equation (3-6) is suitable for the calculation of RAC after more cycles of freezing and thawing. However, with the increase of freezing and thawing cycles, the brittleness of RAC increases, the slope of the descending branch decreases rapidly, the correlation coefficients becomes smaller, which illustrates the equation (3-6) is not suitable for the calculation of RAC after more cycles of freezing and thawing. So the descending branch of the stress-strain curve was addressed in the research. According to the characteristics that the descending branch slope of the complete stress strain curves decreases rapidly for RAC after freezing and thawing, the calculation formula of Zhenhai Guo is modified as equation (3-7) shows.
y¼
0.3
P1
e r ; y¼ rc ec
(
0.5
0.2
Stress/Peak stress
x ;x bðx1Þ2 þx
experimental value
0.8
ax þ ð3 2aÞx2 þ ða 2Þx3 ; x 6 1
ax þ ð3 2aÞx2 þ ða 2Þx3 ; x 6 1 x ;x Bðx1Þþx
P1
ð3-7Þ
e r ;y ¼ rc ec
where ec is the peak strain; rc is the peak stress. a; B are parameters to be determined in the ascending branch and descending branch, respectively. The First Optimization software was also used to find the optimum parameter of the descending branch according to the test data based on equation (3-7). The calculation results are shown in Table 6. Comparing with the parameters calculated with Zhenhai Guo [19] model, the correlation coefficients of parameters calculated with equation (3-7) are all greater than 0.95, which indicates that the calculation results are agreement with the test data and the model can well represent the descending branch of the complete stress strain curves of RAC after freezing and thawing cycles. Taking the parameter B for the descending branch as the dependent variable, the freezing and thawing cycles n as the independent variable, coordinates are given and the mathematical calculation model with B and n is fitted as shown in Fig. 14.
B ¼ 9 106 n3 þ 0:0024n2 0:1689n þ 4:2712
0.5
R2 ¼ 0:9918
0.4 0.3 0.2 0.1 0.0
Strain/Peak strain Fig. 19. Verification of theoretical stress-strain curves of RAC after freezing and thawing.
where B is the parameter for the descending branch, n is cycles of freezing and thawing of RAC with 100% RCA replacement ratio. Figs. 15–18 are the complete stress-strain curves of RAC with 100% RCA replacement ratio after different cycles of freezing and thawing, and it can be seen that the theoretical curves are in good agreement with the measured curves. After correlation calculation, the correlation coefficient R of Figs. 15–18 are 0.996, 0.989, 0.991 and 0.768, respectively. The correlation coefficient of the complete stress-strain curve of RAC after 125 cycles of freezing and thawing is only 0.768, much smaller than that of the other curves. The reason may be that the discreteness of RAC is stronger than that of NAC due to the discreteness of RCA, especially when the cycles of
J. Wu et al. / Construction and Building Materials 134 (2017) 210–219
freezing and thawing is larger. From the calculation results, it is shown that the theoretic stress-strain curve presented in this paper has a good correlation with the experimental results. 3.5.4. Verification of theoretical stress-strain curves The experimental data of group NRC31.5-D100 from Yongkang Shang [14] was collected, for which, the replacement rate of RCA is 100%, RCA has no air inside, the maximum particle size of aggregate is 31.5 mm, and the freeze-thaw cycles is 100. The calculated and experimental results were shown in Fig. 19. It was found that the theoretical curves using the model proposed in this paper are in good agreement with the experimental results. 4. Conclusions (1) Mass loss of RAC decreased first then increased with the increase of freezing and thawing cycles. (2) The relative cubic compressive strength of RAC is lower than that of NAC under the same freezing and thawing cycles. (3) The relative dynamic elastic modulus of RAC decreased linearly with the increase of freezing and thawing cycles. The relative dynamic elastic modulus of RAC is lower than that of NAC after the same cycles of freezing and thawing. (4) The stress strain curve of RAC after freezing and thawing first ascends rapidly and then descends rapidly after exceeding the peak stress, finally turns into a slow decrease process. Theoretical stress-strain curve of RAC with 100% RCA replacement ratio after different freezing and thawing cycles was presented, which are in good agreement with the experimental results from other researcher.
References [1] Hong Zhang, The Wonderful and the Helpless of the Concrete Industry in China, China Building Materials Network, 2014. http://www.cbmd.cn/sxy/Info5620.html (in Chinese).
219
[2] Jingpeng Hou, Wei Shi, Yupu Song, Development and application promoting of recycled concrete, Archit. Technol. 33 (l) (2002) 15–17 (in Chinese). [3] R. Zaharieva, F. Buyle Bodin, E. Wirquin, Frost resistance of recycled aggregate concrete, Cem. Concr. Res. 34 (10) (2004) 1927–1932. [4] Y. Kasai, M. Hisaka, K. Yanagi, Durability of concrete using recycled concrete, in: Proc. Int. RILEM Symp. on Demolition and reuse of concrete and Masonary, Taylor and Francis, London, 1988, pp. 623–632. [5] A. Gokce, S. Nagataki, T. Saeki, M. Hisada, Freezing and thawing resistance of air-entrained concrete incorporating recycled coarse aggregate: The role of air content in demolished concrete, Cem. Concr. Res. 34 (5) (2004) 799–806. [6] R.M. Salem, E.G. Burdette, N.M. Jackson, Resistance to freezing and thawing of recycled aggregate concrete, ACI Mater. J. 100 (3) (2003) 216–221. [7] Y. Haitao, T. Shizhu, Preparation and properties of high-strength recycled concrete in cold areas, Materiales de Construcción 65 (318) (2015) 1–6. [8] A. Richardson, K. Coventry, J. Bacon, Freeze/thaw durability of concrete with recycled demolition aggregate compared to virgin aggregate concrete, J. Cleaner Prod. 19 (2) (2011) 272–277. [9] J. Xiao, J. Li, C. Zhang, Mechanical properties of recycled aggregate concrete under uniaxial loading, Cem. Concr. Res. 35 (6) (2005) 1187–1194. [10] Ting Du, Weihong Wang, Zhongxin Liu, The complete stress-strain curve of recycled aggregate concrete under uniaxial compression loading, J. Wuhan Univ. Technol.-Mater. Sci. Ed. 25 (5) (2010) 862–865. [11] Code for design of concrete structures (GB50010-2010), China Architecture & Building Press, Beijing, 2010 (in Chinese). [12] S.B. Huda, M. Shahria Alam, Mechanical and freeze-thaw durability properties of recycled aggregate concrete made with recycled coarse aggregate, J. Mater. Civ. Eng. (2015). 04015003-19. [13] González-Fonteboa Belén, Martínez-Abella Fernando, Carro López Diego, et al., Stress-strain relationship in axial compression for concrete using recycled saturated coarse aggregate, Constr. Build. Mater. 25 (5) (2011) 2335–2342. [14] Yongkang Shang, The Experimental Study on the Mechanical Properties and Frost Resistance of Recycled Aggregate Concrete, Harbin Institute of Technology, 2010 (in Chinese). [15] Standard for Technical Requirements and Test Method of Gravel and Crushed Stone for Ordinary Concrete (JGJ 52-2006), China Architecture & Building Press, Beijing, 2006 (in Chinese). [16] Recycled Aggregate for Concrete (GB/T25177-2012), China Architecture & Building Press, Beijing, 2010 (in Chinese). [17] Code for Concrete Admixture Application (GB 50119-2013), China Architecture & Building Press, Beijing, 2014 (in Chinese). [18] Standard for Test Method of Long-term Performance and Durability of Ordinary Concrete (GB/T 50082-2009), China Architecture & Building Press, Beijing, 2009 (in Chinese). [19] Zhenhai Guo, Concrete Strength and Constitutive Relation-Principle and Application, China Building Industry Press, Beijing, 2004. 17–18 (in Chinese).