- Email: [email protected]
Mechanical properties of concrete containing a high volume of tire–rubber particles
Available online at www.sciencedirect.com
Waste Management 28 (2008) 2472–2482 www.elsevier.com/locate/wasman
Mechanical properties of concrete containing a high volume of tire–rubber particles Ali R. Khaloo *, M. Dehestani, P. Rahmatabadi Civil Engineering Department, Sharif University of Technology, Center of Excellence in Structures and Earthquake Engineering, Tehran, Iran Accepted 5 January 2008 Available online 26 March 2008
Abstract Due to the increasingly serious environmental problems presented by waste tires, the feasibility of using elastic and flexible tire–rubber particles as aggregate in concrete is investigated in this study. Tire–rubber particles composed of tire chips, crumb rubber, and a combination of tire chips and crumb rubber, were used to replace mineral aggregates in concrete. These particles were used to replace 12.5%, 25%, 37.5%, and 50% of the total mineral aggregate’s volume in concrete. Cylindrical shape concrete specimens 15 cm in diameter and 30 cm in height were fabricated and cured. The fresh rubberized concrete exhibited lower unit weight and acceptable workability compared to plain concrete. The results of a uniaxial compressive strain control test conducted on hardened concrete specimens indicate large reductions in the strength and tangential modulus of elasticity. A significant decrease in the brittle behavior of concrete with increasing rubber content is also demonstrated using nonlinearity indices. The maximum toughness index, indicating the post failure strength of concrete, occurs in concretes with 25% rubber content. Unlike plain concrete, the failure state in rubberized concrete occurs gently and uniformly, and does not cause any separation in the specimen. Crack width and its propagation velocity in rubberized concrete are lower than those of plain concrete. Ultrasonic analysis reveals large reductions in the ultrasonic modulus and high sound absorption for tire–rubber concrete. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Modification of concrete properties by the addition of appropriate materials is a popular field of concrete research. The brittle nature of concrete and its low loading toughness compared to other materials, has prompted the use of waste tire particles as a concrete aggregate to possibly remedy or reduce these negative attributes. Elastic and deformable tire–rubber particles could improve concrete properties. Waste tire management and disposal is a major environmental concern in many countries. Stockpiling is dangerous, not only due to a potential negative environmental impact, but also because it presents a fire hazard and provides a breeding ground for rats, mice, vermin, and mosquitoes (Khatib and Bayomy, 1999; Guneyisi et al., 2004; *
Corresponding author. Fax: +98 21 6601 4828. E-mail address: [email protected] (A.R. Khaloo).
0956-053X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2008.01.015
Eldin and Senouci, 1993, 1994; Toutanji, 1996; Fedroff et al., 1996; Topcu, 1995; Siddiquel and Naik, 2004; Hernandez-Olivares et al., 2002; Ghaly and Cahill, 2005; Li et al., 2004). Waste tire management is increasingly becoming a significant environmental, health, and aesthetic problem that is not easily solved. The use of waste tires as a concrete additive is a possible disposal solution. The importance of recycling of waste tires coupled with the interest in overcoming the aforementioned concrete defects have motivated a significant body of research pertaining to rubberized concrete. Properties, testing, and design of rubber as an engineering material were investigated in 1960 (Eldin and Senouci, 1993). Eldin and Senouci (1993, 1994) used tire–rubber particles as concrete aggregates, elucidating rubberized concrete properties, and proposed an analytical approach to predict the strength in rubberized concrete. Khatib and Bayomy (1999) studied rubberized Portland cement concrete and offered some practical uses of rubberized concrete, including reduction
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
factors. Their paper contains limitations and concerns of using tire–rubber concrete as well. Li et al. (2004) used waste tires in the form of fibers and developed waste tire fiber modified concrete. The static and dynamic behavior of recycled tire–rubber-filled concrete was investigated by Hernandez-Olivares et al. (2002). Siddiquel and Naik (2004) presented an overview of research published on the use of scrap tires in Portland cement concrete. Guneyisi et al. (2004) investigated the properties of rubberized concretes containing silica fume through six designated rubber contents. These previous findings reveal that the properties of rubberized concrete are affected by type, size, content, and the procedure of incorporating the rubber into the concrete. In this paper, tire–rubber concrete properties are investigated using mechanical and non-destructive testing for different sizes of tire particles. The experimental observations and subsequent explanations of tire–rubber concrete behavior under compressive strain are presented. Ultrasonic analysis investigates sound absorption and the ultrasonic modulus of tire–rubber concrete.
2473
Table 2 Rubber specification (ANSI tests) Reclaimed rubber specification: American National Standards Institute Chemical test MLT+4 at 100 c0 Ash Content Acetone Extract Polymer Content Sp. Gravity at 25 c0 Hardness Tensile Elongation
Unit
Actual
Standard
Test method
37.5
Max 70
JIS K 6313
% %
5 5
Max 70 Max 70
JIS K 6313 JIS K 6313
%
46
Min 40
JIS K 6313
g/cm3
1.14
JIS K 6313
SHA kg/ cm2 %
53 41
1.17+/ 0.02 55+/5 Min 70
JIS K 6313 JIS K 6313
200
Min 300
JIS K 6313
Standard recipe reclaim ZNO
ST. ACID
ACC.CBC
Sulfur
Curing condition
Packing
5
1
1
3
140 c0 , 20 min
Rolls of 15 kg
2. Experimental design In order to investigate the mechanical properties of tire– rubber concrete, specimens of a cylindrical shape 15 cm in diameter 30 cm in height were fabricated. These specimens were different in the content and type of tire particles as a portion of total aggregates in concrete. 2.1. Materials Constituent materials for concrete mixes included a Type I Portland cement meeting ASTM C150 requirements, crushed stone gravel with a maximum size of 20 mm as a coarse aggregate, natural sand with a 4.75 mm maximum size as fine aggregate, and tire–rubber particles provided by the Yazd Tire Company in Iran. Tire particle specifications are summarized in Tables 1 and 2. These specifications were provided by tire manufacturers according to ANSI (American National Standard Institute) tests. Two types of scrap tire–rubber particles were used: crumb rubber, which was a fine material with grading close to that of the aforementioned sand, and coarse tire chips produced by mechanical shredding. Tire particle
groups are shown in Fig. 1. Particles of tire finer than 0.15 mm may disturb the cement paste reaction (Neville, 1995); thus these particles were removed from the tire aggregate source using a sieve # 100 based on the ASTM C136 method. Tire particles were not pretreated before their incorporation into the concrete mixture. The properties of fine and coarse aggregates were determined according to ASTM standard test methods C127, C128, C129, and C136. The grading of tire–rubber materials was determined based on the ASTM C136 method. The grading curve of rubber materials was determined by using crushed stones in each sieve in order to provide adequate pressure on tire–rubber particles to pass the sieves. Grading curves are presented in Fig. 2. Data regarding the properties of the aggregates and the rubber particles are given in Table 3. The specific gravity of the cement was evaluated to be 3.15 g/cm3.
Table 1 Rubber–tire specification Rubber powder specification 1 2 3 4 5 6 7
Specific gravity Ash content Plasticizer Carbon black Polymer Sieve residue on mesh 40 Sieve residue on mesh 60 Packing: bags of 30 kg
1.16 g/cm3 5% 10% 29% 50% 3.36% 80% Fig. 1. Type of tire–rubber particles.
2474
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
2.2. Concrete mixtures The experimental setup and specimen fabrication are summarized in Tables 4 and 5, respectively. To unify the rubber content, a designated percentage for each mix type was converted to a total aggregate volume percentage. The equivalent values of rubber content by total aggregate volume are given in Table 4. Specimens were demoulded 24 h after casting, and were then water cured for 7 days. Thereafter, the specimens were kept in a curing room at a temperature of 30 °C, with a relative humidity of 60%, until the time of testing. A normal, non-air-entrained, Portland cement concrete, with a 30 MPa targeted compressive strength, was designed as the control mix following ACI Standard 211.1-81 (American Concrete Institute, 2002). The mix required a 0.45 water–cement ratio. Other constituents are given in Table 5 for P specimens. This control mix was used as the basis for preparing three rubberized concrete mixes specified by C, F, and CF mixes. In the C mixes, the coarse aggregates of the control mix were replaced by rubber chips, and in the F mixes, the sand in the control mix was replaced by crumb rubber. In the CF mixes, chip and crumb rubber particles were used as replacements for gravel and sand, respectively. For a C50F50 mix, crumb rubber replaced 50% of the sand volume and tire chips replaced 50% of the coarse aggregate volume.
Fig. 2. Grading of mineral aggregates and tire particles.
Table 3 Properties of aggregate and rubber Aggregate type
Specific gravity
Coarse aggregate Fine aggregate Rubber Particles
2.65
Absorption (%)
Fineness modulus
Unit weight (kg/m3)
2.66
NA
1701.3
2.67
5.01
4.35
1716.8
1.16
49.56
NA
1150
Table 4 Experimental program Specimen designation
Tire content (%) by total aggregates
Fine tire aggregate (%)
Coarse tire aggregate (%)
Fine mineral aggregate (%)
Coarse mineral aggregate (%)
Replicates in compressive test
Replicates in ultrasonic test
P C25 C50 C75 C100 F25 F50 F75 F100 C25F25 C50F50
0 12.5 25 37.5 50 12.5 25 37.5 50 25 50
0 0 0 0 0 25 50 75 100 25 50
0 25 50 75 100 0 0 0 0 25 50
100 100 100 100 100 75 50 25 0 75 50
100 75 50 25 0 100 100 100 100 75 50
3 3 3 3 3 3 3 3 3 3 3
2 2 2 2 2 2 2 2 2 2 2
Table 5 Concrete mixture proportions Specimen
Water (lit)
Cement (kg)
Coarse tire aggregate (kg)
Fine tire aggregate (kg)
Gravel (kg)
Sand (kg)
Moisture of gravel (%)
Moisture of sand (%)
P C25 C50 C75 C100 F25 F50 F75 F100 C25F25 C50F50
227 229 299 356 442 215 282 384 453 298 434
350 350 350 350 350 350 350 350 350 350 350
0 152.375 304.18 456.3 609.5 0 0 0 0 152.375 304.2
0 0 0 0 0 149.5 299 452.1 602.8 149.5 301.4
900 675 450 225 0 900 900 900 900 675 450
900 900 900 900 900 675 450 225 0 675 450
0 2.6 2 3 3 3 4 3 3 3 4
0 5.5 6.4 8 7 7 7.5 2.5 7 6 9
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
To evaluate the properties of fresh concrete, slump and unit weight were measured according to ASTM C143 and ASTM C138 (ASTM, 1988), respectively. A compressive strain-control test was conducted for hardened concrete specimens to obtain the stress–strain curves for all of the specimens. The test was performed by a universal testing machine and a sensitive data acquisition system. The machine yielded a loading value variation due to a constant rate of specimen deformation. This rate was chosen to be 0.005 mm/sec. The ends of the cylinders were capped with traditional sulfur mortar following the requirements of ASTM C617 (ASTM, 1988) prior to testing. Compressive strengths of the cylindrical specimens were evaluated after 50 days. The tangential moduli of elasticity at 40% of the ultimate stress on the elastic portion of the stress–strain curves were evaluated. A non-destructive test using an ultrasonic pulse device was conducted for cylindrical specimens to evaluate the velocity of wave transmission in the material, and also, the dynamic (ultrasonic) modulus of elasticity of the hardened concrete. The velocity of wave transmission in the concrete was measured in the longitudinal direction of the specimen. 3. Experimental results and discussion 3.1. Properties of fresh concrete Variations of slump and unit weight of fresh concrete with respect to tire aggregate concentration are presented in Fig. 3. The workability, defined as the ease with which concrete can be mixed, transported, and placed, of fresh concrete is affected by the interactions of tire particles and mineral aggregates. As shown in Fig. 3a, the slump for F mixes increased with tire aggregate concentrations lower than 15%, and reached a maximum value when the tire aggregate concentration was 15%. Tire aggregate concentrations exceeding 15% reduced the slump. The slump for C mixes decreases to a minimum value with tire aggregate concentrations of 15%. The slump fluctuates slightly over the minimum value for tire aggregate concentrations exceeding 15%. Slump reduction for combined mixes was less than that of C mixes. In general, the rubberized concrete specimens have acceptable workability in terms of ease of handling, placement, and finishing. As shown in Fig. 3, the ordinary procedure for evaluating the slump of the investigated mixes does not support the actual state of the mix workability. These findings suggest that another method is required to properly measure the slump of rubberized concrete (Eldin and Senouci, 1994).
8.5
Coarse
7.5
Fine Combined
6.5 slump (cm)
2.3. Test methods
a
5.5 4.5 3.5 2.5 1.5 0
10
20
30
40
50
Tire content (%) of total aggregates
b
2500 Coarse
Unit weight (kg/m3 )
Several specimens were fabricated from each of the C and F mixes, wherein the coarse or fine mineral aggregate was replaced by rubber aggregates in increments of 25% by volume. C25F25 and C50F50 were also fabricated.
2475
Fine
2200
Combined 1900
1600
1300 0
10 20 30 40 50 Tire content (%) of total aggregates
Fig. 3. Properties of fresh concrete (a) slump and (b) unit weight of fresh rubberized concrete mixtures.
The unit weight of the concrete ranged from 2409 to 1324 kg/m3, depending on rubber content. Increasing the rubber content reduces the unit weight of the concrete, resulting in lighter concretes. The unit weights of the C, F, and CF mixes were reduced 45%, 34%, and 33%, respectively, compared to plain concrete. The unit weight reduction is a result of the lower unit weight of tire–rubber particles replacing the much heavier mineral aggregates. Thus, rubber–tire concrete could be used wherever lightweight concrete is required. For example, tire–rubber concrete containing low tire–rubber concentrations can be used in structures to reduce earthquake damage. Due to the high water absorption of tire particles, the ratio of the fresh concrete unit weight to the hardened unit weight in tire–rubber concrete is greater than that of plain concrete. Therefore, tire–rubber concrete is expected to be more porous than plain concrete. A smaller reduction in unit weight, compared to that of the F and CF mixes, was realized for C mixes with rubber concentrations lower than 40%. At higher concentrations, the result is reversed. 3.2. Hardened concrete properties 3.2.1. Effective parameters Due to the non-polar nature of rubber particles and their tendency to entrap air in their rough surfaces, tire– rubber concrete specimens contain a higher air content.
2476
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
Furthermore, due to their tendency to repel water, when tire particles are substituted for mineral aggregates in the control mixture, the tire particles attract air, with the amount depending on the internal pressure in the mixture. Air can then adhere to the tire particles (Siddiquel and Naik, 2004), thus, increasing the tire content results in a higher air content in tire–rubber concrete mixtures, thereby decreasing the unit weight of the mixtures. A higher air content suggests a subsequent strength reduction in concrete specimens. For example, in air-entrained concrete, 8% entrained air reduces the strength of the concrete specimens by 45% (Neville, 1995). Due to a low modulus of elasticity with respect to mineral aggregates, rubber aggregates act as large pores, and do not significantly contribute to the resistance to externally applied loads. Thus, a tire–rubber concrete specimen loses its strength depending on its tire content. For example, the effective cross section of a specimen resisting an external load decreases according to the tire particle concentration of that specimen. Assume a to be the volumetric percentage of the tire particle concentration in the mixture, and k is the relative dimensional length reduction. These parameters satisfy the following condition
k3 ¼
a 100
ð1Þ
and the reduced effective cross section Ae can be expressed as a 23 2 A ð2Þ Ae ¼ ð1 k ÞA ¼ 1 100 where A is the nominal cross section of the specimen. For example, a tire aggregate concentration of 20% of the total concrete volume reduces the effective cross section by 34%. The low elastic modulus of tire particles has these particles behave as weak inclusions in the hardened concrete mass. The effect of stress concentration should be considered for these inclusion-like particles. For a more detailed discussion of this topic, refer to (Eldin and Senouci, 1993). 3.2.2. Visual observation of concrete specimen behavior The failure duration, defined as the duration to concrete failure, for plain concrete is abrupt and explosive. In contrast, the tire–rubber concrete failure duration is more gradual, since the concrete becomes more flexible with increasing tire particle substitution of mineral aggregates.
Fig. 4. Failure types of specimens.
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
Tire–rubber concretes are able to withstand loads beyond the peak load, which is referred to as post-failure strength. Failure states in plain concrete specimens, as shown in Fig. 4, are accompanied with the separation of pieces or slices from the specimen. For concrete containing tire particles, the failure state was not accompanied by any detachment due to the bridging of cracks by rubber particles. Tire–rubber concrete specimens did not exhibit any detachment, despite losing a considerable amount of strength as shown by the F25 and C25 specimens in Fig. 4. Tire–rubber concrete specimens present large deformations compared to plain concrete specimens. During the unloading process, the flexible behavior of tire particles decreases the internal friction among the concrete elements, and recovers extra strain. Failure properties, like discontinuities and cracks, propagated uniformly and gradually in tire–rubber con-
2477
crete specimens. In contrast, the propagation of failure symptoms were abrupt and concentrated for plain concrete. Fig. 4 demonstrates that the failure parameters grow uniformly from the bottom to the top of a C25 of specimen. The lateral deformations of tire–rubber concrete specimens are larger than those of plain concrete specimens; however, because of the porosity due to the substitution of tire particles, Poisson’s ratios for tire–rubber concrete are slightly more than those for plain concrete. It is important to note that the behavior of rubberized concrete is not perfectly elastic, therefore Poisson’s ratio is not constant for the entire loading process. Poisson’s ratio increases and approaches 0.5 as the behavior of rubberized concrete becomes plastic-like. As shown in Fig. 4, considerable lateral deformations are observable in tire–rubber concrete specimens after an entire loading process.
Fig. 5. compressive stress–strain response of rubber–tire concrete, (a) responses of concrete types shown all together, response of concrete with (b) 12.5%, (c) 25%, (d) 37.5% and (e) 50% tire content by total aggregate volume.
2478
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
line connecting the origin to the ultimate stress. A higher nonlinearity index implies a more nonlinear stress–strain curve. Fig. 6a illustrates how the nonlinearity index is determined. The nonlinearity indices for different concentrations of rubber and various types of mixtures are summarized in Fig. 6b. As shown in Fig. 6b, the nonlinearity index increases as the rubber content increases for all mixtures. A comparison between the investigated mixtures reveals that the C mixture behavior is more nonlinear than the F and CF mixtures behavior. The increased nonlinearity for tire–rubber concrete can explain the observations of the more global and well dispersed failures of the tire– rubber concrete specimens. The substitution of rubber for mineral aggregates appear to permit more uniform crack development and provide gentler crack propagation, compared to plain concrete. Considering the stress–strain curves, tire–rubber concrete specimens experience larger deformations compared to plain concrete specimens under the same loading conditions. Hence, these curves support the assertion that tire particle usage in concrete results in concrete failures with larger deformations and higher energy dissipation. Et is defined as the slope of the line tangent to the stress– strain curve at the point with 40% of the ultimate stress and can be obtained using Eq. (3), wherein drt and det are stress changes and the corresponding strain changes at the point with 40% of the ultimate stress Et ¼
Fig. 5 (continued)
3.2.3. Stress–strain response The stress–strain curves for the investigated rubber types and concentrations are shown in Fig. 5. The curves indicate that the behavior of tire–rubber concrete is more nonlinear compared to that of plain concrete, implying a different failure type for tire–rubber concrete. The nonlinear behavior for tire–rubber concrete mixtures may also be due to the lower compressive strength of these mixtures. To compare the nonlinearity between the plain control concrete and tire–rubber concrete, a nonlinearity index was defined as the ratio of the slope of the line connecting the origin to 40% of the ultimate stress, to the slope of the
drt @ 40% of ultimate stress det
ð3Þ
Et can be a suitable indicator of the tire–rubber concrete stiffness attributed to elastic deformation. The values of Et for different mixtures and tire concentrations are given in Table 6. Significant decreases in Et for tire–rubber concrete specimens imply large deformations of those specimens. A comparison between C and F specimens shows that, for specimens having less than 25% rubber concentration, C specimens have higher Et values compared to F specimens. Hence, larger elastic deformations are observed for F specimens at the same elastic loading condition. For specimens having more than 25% rubber concentration, C and F specimens have identical Et values, which results in equal elastic deformations for the same elastic loading condition. Failure state or softening phases of C specimens were accompanied by a larger deformation compared to F specimens at the same loading condition. C specimens exhibited higher strength compared to F specimens at the failure state for the same strain condition. This extra strength is due to the existence of fibers in coarse tire–rubber particles. As shown in Fig. 5, the CF specimen stress–strain curves located between the C and F specimens stress–strain curves contain the same tire concentration. The combined stress– strain curve demonstrated stress values which were close to those of the F mixture, but the curve’s shape dominantly resembled the C mixture curve. This suggests that the ultimate stress of CF tire–rubber concrete depends on the fine
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
a 0.6
Stress (MPa)
Ultimate Stress
0.4
β
α 40% of Ultimate Stress
0.2
Nonlinearity Index= tg β / tgα
0 0
0.04
0.08
0.12
0.16
Strain (mm/mm)
b
8
Nonlinearity Index
7
Coarse Fine
6
Combined
5 4 3 2 1 0 0
15
30
45
60
Tire rubber content by total aggregate (%)
Fig. 6. (a) Evaluation of nonlinearity index and (b) nonlinearity index for various types of rubberized concrete.
aggregate concentration, while the shape of the stress– strain curve is primarily affected by the coarse aggregate concentration. These observations imply the stress–strain response for CF tire–rubber concrete is located between the corresponding C and F stress–strain curves equaling the total respective tire particles concentration. Hence any desired stress–strain curve between the C and F curves can be obtained by tuning coarse and fine tire concentrations. Table 6 Ultimate strengths and tangential moduli of elasticity for various concrete types containing different tire–rubber contents Concrete mixture
Stressmax (MPa)
Et (GPa)
0
30.77
7.41
Total rubber content (%)
Control
P
C-type
C25 C50 C75 C100
12.5 25 37.5 50
6.52 1.49 0.65 0.37
2.47 0.31 0.12 0.03
CF-type
C25F25 C50F50
25 50
1.17 0.53
0.42 0.04
F-type
F25 F50 F75 F100
12.5 25 37.5 50
6.36 1.22 0.81 0.55
1.15 0.31 0.11 0.04
2479
3.2.4. Compressive strength As shown in Table 6, the ultimate compressive strength reduces significantly with increasing amounts of rubber concentration, and the strengths for all mixes approach a minimum of 1 MPa. The ultimate strength for C specimens are slightly more than that of F specimens for total rubber concentrations lower than 25%; however for total rubber concentrations greater than 25%, the ultimate strength results are reversed. Ultimate strengths of combined specimens were nearly between the ultimate strengths of the C and F specimens, but closer to F specimen strengths. The systematic reduction of ultimate strength in tire– rubber concrete might restrict the use of tire–rubber concrete, with tire–rubber concentrations exceeding 25%, in structural applications. Reduction in tire–rubber concentration and rubber particle pretreatments are required to enhance the ultimate strength and other mechanical properties of rubber-particle concrete. Pretreatment is available, ranging from an inexpensive and easy treatment with water, to complicated and expensive physical, chemical, and mechanical treatments. A typical treatment for tire– rubber particles is a NaOH solution to improve rubber adhesion with the cement paste (Segre and Joekes, 2000). Another report, in contrast, states that a NaOH and silone pretreatment of rubber does not significantly change the compressive strength and splitting tensile strength of rubber-concrete composites, when compared to untreated composites (Albano et al., 2005). 3.2.5. Toughness Toughness of tire–rubber concrete was determined by calculating the area under the stress–strain curve up to 80% of the ultimate stress in the post-peak region. The toughness value is defined as a ratio between the area under the stress–strain curve up to 80% of the ultimate stress, to the area under the stress–strain curve up to the ultimate stress. Fig. 7a illustrates how each of the areas was determined. It should be noted that the 80% factor of ultimate stress was selected to limit further reductions in strength level. Thus, the toughness index (Ti) is expressed as follows Ti ¼
T 80% T 100%
ð4Þ
The toughness indices for different rubber concentrations and different mixtures are presented in Fig. 7b. Tire–rubber concrete exhibited greater toughness as compared to the plain concrete. Toughness indices maximize as rubber concentration approaches 25% of the total aggregate volume. Beyond rubber concentrations of 25%, toughness indices decrease due to the systematic reduction in strength. The toughness index for a combined C25F25 mixture is higher than that of the F50 and C50 mixtures. 3.2.6. Reduction factors (RF) for hardened concrete properties Mechanical property results of rubberized concrete showed that property values were primarily dependent on
2480
a
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
a
7 Ultimate Stress
6
All Data
F25
5
T 100% =A T 80% =A+B Toughness Index=(A+B)/B
4 3
SRF
Stress (MPa)
0.8
80% of Ultimate Stress
Poly. (All Data)
0.6 0.4
A
2
1.0
B
0.2
1 0
0.0 0
0.01
0.02
0.03
0.04
0.05
0
b
7
b
30
60
All Data
0.8
Fine
5
45
1.0
Coarse
6
Combined
Poly. (All Data)
4
ERF
Toughness Index
15
Rubber content by total aggregate volume (%)
Strain (mm/mm)
3
0.6 0.4
2 0.2
1 0 0
0.0
10 20 30 40 50 60 Rubber content by total aggregate volume (%)
0
Fig. 7. (a) Evaluation of toughness index and (b) toughness index values for various amounts of rubber contents.
the total rubber concentration. Therefore, it is convenient to establish a function to investigate the influence total rubber concentration has on the mechanical properties of concrete. Hence, the reduction factor (RF) is defined for a mechanical property as the ratio of the compressive strength or tangential modulus of elasticity in rubber-concretes containing a rubber concentration, R, to the value of the control plain concrete. The variation of the RF-determined compressive strength and tangential modulus of elasticity corresponding to rubber concentration are presented in Fig. 8a and b, respectively. The RF was unity for a 0% of rubber concentration (control specimen), and gradually decreased with increasing rubber concentration from 0% to 50% by total aggregate volume. Khatib and Bayomy (1999), in a similar study, examined several mathematical functions, including various degrees of polynomial functions, and proposed the following equation to quantify the reduction in the compressive strength SRF ¼ a þ bð1 RÞ
m
ð5Þ
Since the strength reduction factor (SRF) equals unity for 0% of R, the following condition can be satisfied aþb¼1
ð6Þ
In Eqs. (5) and (6), the SRF varies from 1 to 0; R is the rubber content in a volumetric ratio of total aggregate volume, and a, b, and m are the functional parameters. Khatib and Bayomy (1999) found the functional parameters a, b, and m to be 0.1, 0.9, and 7 in concrete aged 28 days, respectively. A similar procedure was conducted in the present study to examine the reduction factors for ultimate stress
15
30
45
60
Rubber content by total aggregate volume (%)
Fig. 8. (a) Strength reduction factor and (b) elastic modulus’ reduction factor of all types of rubberized concrete containing different rubber content.
and the tangential modulus of elasticity. A regression analysis was utilized to determine the functional parameters. Eqs. (7) and (8) yield the ultimate SRF and tangential elastic modulus reduction factor (ERF) in terms of rubber concentration in a volumetric ratio of total aggregate volume. The R-square values were determined to evaluate the accuracy of the equations 12
R square ¼ 0:99
ð7Þ
11
R square ¼ 0:98
ð8Þ
SRF ¼ 0:02 þ 0:98ð1 RÞ
ERF ¼ 0:01 þ 0:99ð1 RÞ
A comparison with results reported in literature demonstrates that the strength obtained in this study is lower than those reported by the others. For example, for a 7.5% total rubber concentration, the SRF attained by Khatib and Bayomy (1999) was 0.6 after 28 days; Li et al. (2004) obtained a SRF of 0.55 from a 7.5% total rubber concentration. Eq. (7) from this study gives the SRF a value of 0.4 from a concentration of 7.5% of total rubber. This anomaly can be attributed to a lack of pretreatment of the tire–rubber particles, improper calibration of equations based on concretes with higher tire–rubber contents, and tire type. 3.2.7. Sound absorption The ultrasonic echo technique is a valuable tool for testing concrete elements. The ultrasonic test is performed by an instrument composed of a transducer and a receiver. Acoustic waves were sent by the transducer and propagate
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
from the flat surface of cylindrical specimen, through the length of the specimen, and are received at the other flat side of the specimen. The time duration for an acoustic wave to propagate through the longitudinal direction of the specimen was recorded for each specimen in this study. The time duration for a specimen was evaluated several times to obtain a stable count. For all specimens, the velocity of the ultrasonic pulses was evaluated by using Eq. (9)
v¼
L t
2300
Coarse Unit weight (kg/m3 )
2100
Fine Combined
1900 1700 1500 1300 1100 0
10
20
30
40
50
Tire content (%) of total aggregates
Velocity of ultrasonic pulse (m/s)
b Coarse
5500
Fine
ð10Þ
where q and v are the hardened concrete unit weight and the velocity of the ultrasonic pulses, respectively. The variable, t, is Poisson’s ratio for the mixtures, and was assumed to be 0.3. Variation of Poisson’s ratio was neglected in this study. Hardened concrete unit weight, ultrasonic pulse velocity, and the values of the ultrasonic moduli for all types of the specimens are shown in Fig. 9. Fig. 9 indicates that the velocity of the ultrasonic waves reduce significantly with increasing tire–rubber content. Wave velocity transfer through a material is one of the most important factors on which mechanical wave energies depend. Therefore tire–rubber concrete is potentially a suitable material for the dampening of sound and other shaking energies, and can be used in noisy sites to serve as sound insulation. Due to the significant reduction in the ultrasonic modulus with increasing tire–rubber concentration, a porous composition is expected for tire–rubber concrete. 4. Conclusions and recommendations
3500
2500
0
10
20
30
40
50
Tire content (%) of total aggregates
Ultrasonic Modulus (GPa)
ð1 þ tÞð1 2tÞ ð1 tÞ
Combined
4500
1500
c
ð9Þ
where L is the specimen’s longitudinal length of 30 cm, and t is the duration obtained from the ultrasonic test. Ultrasonic moduli were evaluated using the velocities of the ultrasonic pulses. The ultrasonic modulus can be obtained using Eq. (10) (Neville, 1995) EU ¼ qv2
a
2481
60
Coarse
50
Fine Combined
40 30 20 10 0 0
10
20
30
40
50
Tire content (%) of total aggregates
Fig. 9. Results of the ultrasonic test: (a) hardened concrete unit weight, (b) influence of rubber–tire on ultrasonic pulse velocity in concrete and (c) values of the ultrasonic moduli for different concrete types.
1. Fresh rubberized concrete mixtures with increasing rubber concentrations present lower unit weights compared to plain concrete. Workability of rubberized concrete with coarse rubber particles is reduced with increasing rubber concentration; however, rubberized concrete with fine rubber particles exhibits an acceptable workability with respect to plain concrete. 2. The substitution of mineral aggregates with tire–rubber particles in concrete results in large reductions in ultimate strength and the tangential modulus of elasticity. Due to the considerable decrease in ultimate strength, rubber concentrations exceeding 25% are not recommended. Pretreatment of tire particle surfaces should be considered for possible improvement of tire–rubber concrete mechanical properties. An investigation is needed to identify the influence of rubber’s mechanical properties on the ultimate strength of rubberized concrete. 3. More ductile behavior is observed for rubberized concrete compared to plain concrete specimens under compression testing. Unlike plain concrete, the failure state in rubberized concrete does not occur quickly and does not cause any detachment in the specimen’s elements. Crack width in rubberized concrete is smaller than that
2482
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
of plain concrete, and the propagation of failure symptoms is more gradual and uniform. The failure state in tire–rubber concrete compared to plain concrete is characterized by more deformation. 4. The results of the combined mixture of tire particles were between the corresponding C and F specimens with the same total tire concentration. The stress–strain response of the combined mixtures demonstrates that the ultimate stress of the combined tire–rubber concrete is chiefly dependent on the fine aggregate concentration, while the shape of the stress–strain curve depends on the coarse aggregate concentration. These findings reveal that, although the mechanical properties of rubberized concrete are mainly dependent on the total rubber content, the concentrations of coarse or fine particles can be adjusted for different behavior. 5. Rubberized concrete is an effective absorber of sound and shaking energy. A large reduction in the ultrasonic modulus with increasing rubber concentration demonstrates a porous composition for rubberized concrete. This study has exclusively focused on the mechanical and physical properties of tire–rubber concrete for fine, coarse, and combined rubber replacements of mineral aggregates. There is a need for future studies to investigate energy absorption of tire–rubber concrete under dynamic loading, and also the durability of tire–rubber concrete under adverse weathering conditions. Acknowledgements Supports of Research Committee of Sharif University of Technology (SUT) and Center of Excellence in Structures and Earthquake Engineering are greatly appreciated. The authors are indebted to the staff of concrete laboratory at Civil Engineering Department of SUT for their valuable assistance.
References Albano, C. et al., 2005. Influence of scrap rubber addition to Portland I concrete composites: destructive and non-destructive testing. Compos. Struct. 71, 439–446. American Concrete Institute, 2002, Standard practice for selecting proportions for normal, heavyweight and mass concrete, Document No: ACI 211.1-91. ASTM, 1988, Concrete and Aggregate, Annual Book of ASTM Standards, vol. 04.02, Philadelphia. Eldin, N.N., Senouci, A.B., 1993. Rubber–tire particles as concrete aggregate. J. Mater. Civil. Eng. 5 (4), 478–496. Eldin, N.N., Senouci, A.B., 1994. Measurement and prediction of the strength of rubberized concrete. Cement Concrete Compos. 16, 287– 298. Fedroff, D., Ahmad, S., Savas, B.Z., 1996. Mechanical properties of concrete with ground waste tire rubber, Transportation Research, Record No. 1532, Transportation Research Board, Washington DC. Ghaly, A.M., Cahill, J.D., 2005. Correlation of strength, rubber content and water: cement ration in rubberized concrete. Can. J. Civil Eng. 32, 1–7. Guneyisi, E., Gesoglu, M., Ozturan, T., 2004. Properties of rubberized concretes containing silica fume. Cement Concrete Res. V 34, 2309– 2317. Hernandez-Olivares, F., Barluenga, G., Bollati, M., Witoszek, B., 2002. Static and dynamic behavior of recycled tire rubber-filled concrete. Cement Concrete Res. 32, 1587–1596. Khatib, Z.K., Bayomy, F.M., 1999. Rubberized Portland cement concrete. ASCE J. Mater. Civil. Eng. 11 (3), 206–213. Li, G., Garrick, G., Eggers, J., Abadie, C., Stubblefield, M.A., Pang, S., 2004. Waste tire fiber modified concrete. J. Compos. B (35), 305–312. Neville, A.M., 1995. Properties of Concrete, fourth ed. Longman, London. Segre, N., Joekes, I., 2000. Use for tire rubber particles as addition to cement paste. Cement Concrete Res. 30 (9), 1421–1425. Siddiquel, R., Naik, T.R., 2004. Properties of concrete containing scraptire rubber – an overview. Waste Manage. 24, 563–569. Topcu, I.B., 1995. The properties of rubberized concretes. Cement Concrete Res. 25 (2), 304–310. Toutanji, H.A., 1996. The use of rubber tire particles in concrete to replace mineral aggregates. Cement Concrete Compos. 18, 135–139.
Waste Management 28 (2008) 2472–2482 www.elsevier.com/locate/wasman
Mechanical properties of concrete containing a high volume of tire–rubber particles Ali R. Khaloo *, M. Dehestani, P. Rahmatabadi Civil Engineering Department, Sharif University of Technology, Center of Excellence in Structures and Earthquake Engineering, Tehran, Iran Accepted 5 January 2008 Available online 26 March 2008
Abstract Due to the increasingly serious environmental problems presented by waste tires, the feasibility of using elastic and flexible tire–rubber particles as aggregate in concrete is investigated in this study. Tire–rubber particles composed of tire chips, crumb rubber, and a combination of tire chips and crumb rubber, were used to replace mineral aggregates in concrete. These particles were used to replace 12.5%, 25%, 37.5%, and 50% of the total mineral aggregate’s volume in concrete. Cylindrical shape concrete specimens 15 cm in diameter and 30 cm in height were fabricated and cured. The fresh rubberized concrete exhibited lower unit weight and acceptable workability compared to plain concrete. The results of a uniaxial compressive strain control test conducted on hardened concrete specimens indicate large reductions in the strength and tangential modulus of elasticity. A significant decrease in the brittle behavior of concrete with increasing rubber content is also demonstrated using nonlinearity indices. The maximum toughness index, indicating the post failure strength of concrete, occurs in concretes with 25% rubber content. Unlike plain concrete, the failure state in rubberized concrete occurs gently and uniformly, and does not cause any separation in the specimen. Crack width and its propagation velocity in rubberized concrete are lower than those of plain concrete. Ultrasonic analysis reveals large reductions in the ultrasonic modulus and high sound absorption for tire–rubber concrete. Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction Modification of concrete properties by the addition of appropriate materials is a popular field of concrete research. The brittle nature of concrete and its low loading toughness compared to other materials, has prompted the use of waste tire particles as a concrete aggregate to possibly remedy or reduce these negative attributes. Elastic and deformable tire–rubber particles could improve concrete properties. Waste tire management and disposal is a major environmental concern in many countries. Stockpiling is dangerous, not only due to a potential negative environmental impact, but also because it presents a fire hazard and provides a breeding ground for rats, mice, vermin, and mosquitoes (Khatib and Bayomy, 1999; Guneyisi et al., 2004; *
Corresponding author. Fax: +98 21 6601 4828. E-mail address: [email protected] (A.R. Khaloo).
0956-053X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2008.01.015
Eldin and Senouci, 1993, 1994; Toutanji, 1996; Fedroff et al., 1996; Topcu, 1995; Siddiquel and Naik, 2004; Hernandez-Olivares et al., 2002; Ghaly and Cahill, 2005; Li et al., 2004). Waste tire management is increasingly becoming a significant environmental, health, and aesthetic problem that is not easily solved. The use of waste tires as a concrete additive is a possible disposal solution. The importance of recycling of waste tires coupled with the interest in overcoming the aforementioned concrete defects have motivated a significant body of research pertaining to rubberized concrete. Properties, testing, and design of rubber as an engineering material were investigated in 1960 (Eldin and Senouci, 1993). Eldin and Senouci (1993, 1994) used tire–rubber particles as concrete aggregates, elucidating rubberized concrete properties, and proposed an analytical approach to predict the strength in rubberized concrete. Khatib and Bayomy (1999) studied rubberized Portland cement concrete and offered some practical uses of rubberized concrete, including reduction
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
factors. Their paper contains limitations and concerns of using tire–rubber concrete as well. Li et al. (2004) used waste tires in the form of fibers and developed waste tire fiber modified concrete. The static and dynamic behavior of recycled tire–rubber-filled concrete was investigated by Hernandez-Olivares et al. (2002). Siddiquel and Naik (2004) presented an overview of research published on the use of scrap tires in Portland cement concrete. Guneyisi et al. (2004) investigated the properties of rubberized concretes containing silica fume through six designated rubber contents. These previous findings reveal that the properties of rubberized concrete are affected by type, size, content, and the procedure of incorporating the rubber into the concrete. In this paper, tire–rubber concrete properties are investigated using mechanical and non-destructive testing for different sizes of tire particles. The experimental observations and subsequent explanations of tire–rubber concrete behavior under compressive strain are presented. Ultrasonic analysis investigates sound absorption and the ultrasonic modulus of tire–rubber concrete.
2473
Table 2 Rubber specification (ANSI tests) Reclaimed rubber specification: American National Standards Institute Chemical test MLT+4 at 100 c0 Ash Content Acetone Extract Polymer Content Sp. Gravity at 25 c0 Hardness Tensile Elongation
Unit
Actual
Standard
Test method
37.5
Max 70
JIS K 6313
% %
5 5
Max 70 Max 70
JIS K 6313 JIS K 6313
%
46
Min 40
JIS K 6313
g/cm3
1.14
JIS K 6313
SHA kg/ cm2 %
53 41
1.17+/ 0.02 55+/5 Min 70
JIS K 6313 JIS K 6313
200
Min 300
JIS K 6313
Standard recipe reclaim ZNO
ST. ACID
ACC.CBC
Sulfur
Curing condition
Packing
5
1
1
3
140 c0 , 20 min
Rolls of 15 kg
2. Experimental design In order to investigate the mechanical properties of tire– rubber concrete, specimens of a cylindrical shape 15 cm in diameter 30 cm in height were fabricated. These specimens were different in the content and type of tire particles as a portion of total aggregates in concrete. 2.1. Materials Constituent materials for concrete mixes included a Type I Portland cement meeting ASTM C150 requirements, crushed stone gravel with a maximum size of 20 mm as a coarse aggregate, natural sand with a 4.75 mm maximum size as fine aggregate, and tire–rubber particles provided by the Yazd Tire Company in Iran. Tire particle specifications are summarized in Tables 1 and 2. These specifications were provided by tire manufacturers according to ANSI (American National Standard Institute) tests. Two types of scrap tire–rubber particles were used: crumb rubber, which was a fine material with grading close to that of the aforementioned sand, and coarse tire chips produced by mechanical shredding. Tire particle
groups are shown in Fig. 1. Particles of tire finer than 0.15 mm may disturb the cement paste reaction (Neville, 1995); thus these particles were removed from the tire aggregate source using a sieve # 100 based on the ASTM C136 method. Tire particles were not pretreated before their incorporation into the concrete mixture. The properties of fine and coarse aggregates were determined according to ASTM standard test methods C127, C128, C129, and C136. The grading of tire–rubber materials was determined based on the ASTM C136 method. The grading curve of rubber materials was determined by using crushed stones in each sieve in order to provide adequate pressure on tire–rubber particles to pass the sieves. Grading curves are presented in Fig. 2. Data regarding the properties of the aggregates and the rubber particles are given in Table 3. The specific gravity of the cement was evaluated to be 3.15 g/cm3.
Table 1 Rubber–tire specification Rubber powder specification 1 2 3 4 5 6 7
Specific gravity Ash content Plasticizer Carbon black Polymer Sieve residue on mesh 40 Sieve residue on mesh 60 Packing: bags of 30 kg
1.16 g/cm3 5% 10% 29% 50% 3.36% 80% Fig. 1. Type of tire–rubber particles.
2474
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
2.2. Concrete mixtures The experimental setup and specimen fabrication are summarized in Tables 4 and 5, respectively. To unify the rubber content, a designated percentage for each mix type was converted to a total aggregate volume percentage. The equivalent values of rubber content by total aggregate volume are given in Table 4. Specimens were demoulded 24 h after casting, and were then water cured for 7 days. Thereafter, the specimens were kept in a curing room at a temperature of 30 °C, with a relative humidity of 60%, until the time of testing. A normal, non-air-entrained, Portland cement concrete, with a 30 MPa targeted compressive strength, was designed as the control mix following ACI Standard 211.1-81 (American Concrete Institute, 2002). The mix required a 0.45 water–cement ratio. Other constituents are given in Table 5 for P specimens. This control mix was used as the basis for preparing three rubberized concrete mixes specified by C, F, and CF mixes. In the C mixes, the coarse aggregates of the control mix were replaced by rubber chips, and in the F mixes, the sand in the control mix was replaced by crumb rubber. In the CF mixes, chip and crumb rubber particles were used as replacements for gravel and sand, respectively. For a C50F50 mix, crumb rubber replaced 50% of the sand volume and tire chips replaced 50% of the coarse aggregate volume.
Fig. 2. Grading of mineral aggregates and tire particles.
Table 3 Properties of aggregate and rubber Aggregate type
Specific gravity
Coarse aggregate Fine aggregate Rubber Particles
2.65
Absorption (%)
Fineness modulus
Unit weight (kg/m3)
2.66
NA
1701.3
2.67
5.01
4.35
1716.8
1.16
49.56
NA
1150
Table 4 Experimental program Specimen designation
Tire content (%) by total aggregates
Fine tire aggregate (%)
Coarse tire aggregate (%)
Fine mineral aggregate (%)
Coarse mineral aggregate (%)
Replicates in compressive test
Replicates in ultrasonic test
P C25 C50 C75 C100 F25 F50 F75 F100 C25F25 C50F50
0 12.5 25 37.5 50 12.5 25 37.5 50 25 50
0 0 0 0 0 25 50 75 100 25 50
0 25 50 75 100 0 0 0 0 25 50
100 100 100 100 100 75 50 25 0 75 50
100 75 50 25 0 100 100 100 100 75 50
3 3 3 3 3 3 3 3 3 3 3
2 2 2 2 2 2 2 2 2 2 2
Table 5 Concrete mixture proportions Specimen
Water (lit)
Cement (kg)
Coarse tire aggregate (kg)
Fine tire aggregate (kg)
Gravel (kg)
Sand (kg)
Moisture of gravel (%)
Moisture of sand (%)
P C25 C50 C75 C100 F25 F50 F75 F100 C25F25 C50F50
227 229 299 356 442 215 282 384 453 298 434
350 350 350 350 350 350 350 350 350 350 350
0 152.375 304.18 456.3 609.5 0 0 0 0 152.375 304.2
0 0 0 0 0 149.5 299 452.1 602.8 149.5 301.4
900 675 450 225 0 900 900 900 900 675 450
900 900 900 900 900 675 450 225 0 675 450
0 2.6 2 3 3 3 4 3 3 3 4
0 5.5 6.4 8 7 7 7.5 2.5 7 6 9
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
To evaluate the properties of fresh concrete, slump and unit weight were measured according to ASTM C143 and ASTM C138 (ASTM, 1988), respectively. A compressive strain-control test was conducted for hardened concrete specimens to obtain the stress–strain curves for all of the specimens. The test was performed by a universal testing machine and a sensitive data acquisition system. The machine yielded a loading value variation due to a constant rate of specimen deformation. This rate was chosen to be 0.005 mm/sec. The ends of the cylinders were capped with traditional sulfur mortar following the requirements of ASTM C617 (ASTM, 1988) prior to testing. Compressive strengths of the cylindrical specimens were evaluated after 50 days. The tangential moduli of elasticity at 40% of the ultimate stress on the elastic portion of the stress–strain curves were evaluated. A non-destructive test using an ultrasonic pulse device was conducted for cylindrical specimens to evaluate the velocity of wave transmission in the material, and also, the dynamic (ultrasonic) modulus of elasticity of the hardened concrete. The velocity of wave transmission in the concrete was measured in the longitudinal direction of the specimen. 3. Experimental results and discussion 3.1. Properties of fresh concrete Variations of slump and unit weight of fresh concrete with respect to tire aggregate concentration are presented in Fig. 3. The workability, defined as the ease with which concrete can be mixed, transported, and placed, of fresh concrete is affected by the interactions of tire particles and mineral aggregates. As shown in Fig. 3a, the slump for F mixes increased with tire aggregate concentrations lower than 15%, and reached a maximum value when the tire aggregate concentration was 15%. Tire aggregate concentrations exceeding 15% reduced the slump. The slump for C mixes decreases to a minimum value with tire aggregate concentrations of 15%. The slump fluctuates slightly over the minimum value for tire aggregate concentrations exceeding 15%. Slump reduction for combined mixes was less than that of C mixes. In general, the rubberized concrete specimens have acceptable workability in terms of ease of handling, placement, and finishing. As shown in Fig. 3, the ordinary procedure for evaluating the slump of the investigated mixes does not support the actual state of the mix workability. These findings suggest that another method is required to properly measure the slump of rubberized concrete (Eldin and Senouci, 1994).
8.5
Coarse
7.5
Fine Combined
6.5 slump (cm)
2.3. Test methods
a
5.5 4.5 3.5 2.5 1.5 0
10
20
30
40
50
Tire content (%) of total aggregates
b
2500 Coarse
Unit weight (kg/m3 )
Several specimens were fabricated from each of the C and F mixes, wherein the coarse or fine mineral aggregate was replaced by rubber aggregates in increments of 25% by volume. C25F25 and C50F50 were also fabricated.
2475
Fine
2200
Combined 1900
1600
1300 0
10 20 30 40 50 Tire content (%) of total aggregates
Fig. 3. Properties of fresh concrete (a) slump and (b) unit weight of fresh rubberized concrete mixtures.
The unit weight of the concrete ranged from 2409 to 1324 kg/m3, depending on rubber content. Increasing the rubber content reduces the unit weight of the concrete, resulting in lighter concretes. The unit weights of the C, F, and CF mixes were reduced 45%, 34%, and 33%, respectively, compared to plain concrete. The unit weight reduction is a result of the lower unit weight of tire–rubber particles replacing the much heavier mineral aggregates. Thus, rubber–tire concrete could be used wherever lightweight concrete is required. For example, tire–rubber concrete containing low tire–rubber concentrations can be used in structures to reduce earthquake damage. Due to the high water absorption of tire particles, the ratio of the fresh concrete unit weight to the hardened unit weight in tire–rubber concrete is greater than that of plain concrete. Therefore, tire–rubber concrete is expected to be more porous than plain concrete. A smaller reduction in unit weight, compared to that of the F and CF mixes, was realized for C mixes with rubber concentrations lower than 40%. At higher concentrations, the result is reversed. 3.2. Hardened concrete properties 3.2.1. Effective parameters Due to the non-polar nature of rubber particles and their tendency to entrap air in their rough surfaces, tire– rubber concrete specimens contain a higher air content.
2476
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
Furthermore, due to their tendency to repel water, when tire particles are substituted for mineral aggregates in the control mixture, the tire particles attract air, with the amount depending on the internal pressure in the mixture. Air can then adhere to the tire particles (Siddiquel and Naik, 2004), thus, increasing the tire content results in a higher air content in tire–rubber concrete mixtures, thereby decreasing the unit weight of the mixtures. A higher air content suggests a subsequent strength reduction in concrete specimens. For example, in air-entrained concrete, 8% entrained air reduces the strength of the concrete specimens by 45% (Neville, 1995). Due to a low modulus of elasticity with respect to mineral aggregates, rubber aggregates act as large pores, and do not significantly contribute to the resistance to externally applied loads. Thus, a tire–rubber concrete specimen loses its strength depending on its tire content. For example, the effective cross section of a specimen resisting an external load decreases according to the tire particle concentration of that specimen. Assume a to be the volumetric percentage of the tire particle concentration in the mixture, and k is the relative dimensional length reduction. These parameters satisfy the following condition
k3 ¼
a 100
ð1Þ
and the reduced effective cross section Ae can be expressed as a 23 2 A ð2Þ Ae ¼ ð1 k ÞA ¼ 1 100 where A is the nominal cross section of the specimen. For example, a tire aggregate concentration of 20% of the total concrete volume reduces the effective cross section by 34%. The low elastic modulus of tire particles has these particles behave as weak inclusions in the hardened concrete mass. The effect of stress concentration should be considered for these inclusion-like particles. For a more detailed discussion of this topic, refer to (Eldin and Senouci, 1993). 3.2.2. Visual observation of concrete specimen behavior The failure duration, defined as the duration to concrete failure, for plain concrete is abrupt and explosive. In contrast, the tire–rubber concrete failure duration is more gradual, since the concrete becomes more flexible with increasing tire particle substitution of mineral aggregates.
Fig. 4. Failure types of specimens.
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
Tire–rubber concretes are able to withstand loads beyond the peak load, which is referred to as post-failure strength. Failure states in plain concrete specimens, as shown in Fig. 4, are accompanied with the separation of pieces or slices from the specimen. For concrete containing tire particles, the failure state was not accompanied by any detachment due to the bridging of cracks by rubber particles. Tire–rubber concrete specimens did not exhibit any detachment, despite losing a considerable amount of strength as shown by the F25 and C25 specimens in Fig. 4. Tire–rubber concrete specimens present large deformations compared to plain concrete specimens. During the unloading process, the flexible behavior of tire particles decreases the internal friction among the concrete elements, and recovers extra strain. Failure properties, like discontinuities and cracks, propagated uniformly and gradually in tire–rubber con-
2477
crete specimens. In contrast, the propagation of failure symptoms were abrupt and concentrated for plain concrete. Fig. 4 demonstrates that the failure parameters grow uniformly from the bottom to the top of a C25 of specimen. The lateral deformations of tire–rubber concrete specimens are larger than those of plain concrete specimens; however, because of the porosity due to the substitution of tire particles, Poisson’s ratios for tire–rubber concrete are slightly more than those for plain concrete. It is important to note that the behavior of rubberized concrete is not perfectly elastic, therefore Poisson’s ratio is not constant for the entire loading process. Poisson’s ratio increases and approaches 0.5 as the behavior of rubberized concrete becomes plastic-like. As shown in Fig. 4, considerable lateral deformations are observable in tire–rubber concrete specimens after an entire loading process.
Fig. 5. compressive stress–strain response of rubber–tire concrete, (a) responses of concrete types shown all together, response of concrete with (b) 12.5%, (c) 25%, (d) 37.5% and (e) 50% tire content by total aggregate volume.
2478
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
line connecting the origin to the ultimate stress. A higher nonlinearity index implies a more nonlinear stress–strain curve. Fig. 6a illustrates how the nonlinearity index is determined. The nonlinearity indices for different concentrations of rubber and various types of mixtures are summarized in Fig. 6b. As shown in Fig. 6b, the nonlinearity index increases as the rubber content increases for all mixtures. A comparison between the investigated mixtures reveals that the C mixture behavior is more nonlinear than the F and CF mixtures behavior. The increased nonlinearity for tire–rubber concrete can explain the observations of the more global and well dispersed failures of the tire– rubber concrete specimens. The substitution of rubber for mineral aggregates appear to permit more uniform crack development and provide gentler crack propagation, compared to plain concrete. Considering the stress–strain curves, tire–rubber concrete specimens experience larger deformations compared to plain concrete specimens under the same loading conditions. Hence, these curves support the assertion that tire particle usage in concrete results in concrete failures with larger deformations and higher energy dissipation. Et is defined as the slope of the line tangent to the stress– strain curve at the point with 40% of the ultimate stress and can be obtained using Eq. (3), wherein drt and det are stress changes and the corresponding strain changes at the point with 40% of the ultimate stress Et ¼
Fig. 5 (continued)
3.2.3. Stress–strain response The stress–strain curves for the investigated rubber types and concentrations are shown in Fig. 5. The curves indicate that the behavior of tire–rubber concrete is more nonlinear compared to that of plain concrete, implying a different failure type for tire–rubber concrete. The nonlinear behavior for tire–rubber concrete mixtures may also be due to the lower compressive strength of these mixtures. To compare the nonlinearity between the plain control concrete and tire–rubber concrete, a nonlinearity index was defined as the ratio of the slope of the line connecting the origin to 40% of the ultimate stress, to the slope of the
drt @ 40% of ultimate stress det
ð3Þ
Et can be a suitable indicator of the tire–rubber concrete stiffness attributed to elastic deformation. The values of Et for different mixtures and tire concentrations are given in Table 6. Significant decreases in Et for tire–rubber concrete specimens imply large deformations of those specimens. A comparison between C and F specimens shows that, for specimens having less than 25% rubber concentration, C specimens have higher Et values compared to F specimens. Hence, larger elastic deformations are observed for F specimens at the same elastic loading condition. For specimens having more than 25% rubber concentration, C and F specimens have identical Et values, which results in equal elastic deformations for the same elastic loading condition. Failure state or softening phases of C specimens were accompanied by a larger deformation compared to F specimens at the same loading condition. C specimens exhibited higher strength compared to F specimens at the failure state for the same strain condition. This extra strength is due to the existence of fibers in coarse tire–rubber particles. As shown in Fig. 5, the CF specimen stress–strain curves located between the C and F specimens stress–strain curves contain the same tire concentration. The combined stress– strain curve demonstrated stress values which were close to those of the F mixture, but the curve’s shape dominantly resembled the C mixture curve. This suggests that the ultimate stress of CF tire–rubber concrete depends on the fine
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
a 0.6
Stress (MPa)
Ultimate Stress
0.4
β
α 40% of Ultimate Stress
0.2
Nonlinearity Index= tg β / tgα
0 0
0.04
0.08
0.12
0.16
Strain (mm/mm)
b
8
Nonlinearity Index
7
Coarse Fine
6
Combined
5 4 3 2 1 0 0
15
30
45
60
Tire rubber content by total aggregate (%)
Fig. 6. (a) Evaluation of nonlinearity index and (b) nonlinearity index for various types of rubberized concrete.
aggregate concentration, while the shape of the stress– strain curve is primarily affected by the coarse aggregate concentration. These observations imply the stress–strain response for CF tire–rubber concrete is located between the corresponding C and F stress–strain curves equaling the total respective tire particles concentration. Hence any desired stress–strain curve between the C and F curves can be obtained by tuning coarse and fine tire concentrations. Table 6 Ultimate strengths and tangential moduli of elasticity for various concrete types containing different tire–rubber contents Concrete mixture
Stressmax (MPa)
Et (GPa)
0
30.77
7.41
Total rubber content (%)
Control
P
C-type
C25 C50 C75 C100
12.5 25 37.5 50
6.52 1.49 0.65 0.37
2.47 0.31 0.12 0.03
CF-type
C25F25 C50F50
25 50
1.17 0.53
0.42 0.04
F-type
F25 F50 F75 F100
12.5 25 37.5 50
6.36 1.22 0.81 0.55
1.15 0.31 0.11 0.04
2479
3.2.4. Compressive strength As shown in Table 6, the ultimate compressive strength reduces significantly with increasing amounts of rubber concentration, and the strengths for all mixes approach a minimum of 1 MPa. The ultimate strength for C specimens are slightly more than that of F specimens for total rubber concentrations lower than 25%; however for total rubber concentrations greater than 25%, the ultimate strength results are reversed. Ultimate strengths of combined specimens were nearly between the ultimate strengths of the C and F specimens, but closer to F specimen strengths. The systematic reduction of ultimate strength in tire– rubber concrete might restrict the use of tire–rubber concrete, with tire–rubber concentrations exceeding 25%, in structural applications. Reduction in tire–rubber concentration and rubber particle pretreatments are required to enhance the ultimate strength and other mechanical properties of rubber-particle concrete. Pretreatment is available, ranging from an inexpensive and easy treatment with water, to complicated and expensive physical, chemical, and mechanical treatments. A typical treatment for tire– rubber particles is a NaOH solution to improve rubber adhesion with the cement paste (Segre and Joekes, 2000). Another report, in contrast, states that a NaOH and silone pretreatment of rubber does not significantly change the compressive strength and splitting tensile strength of rubber-concrete composites, when compared to untreated composites (Albano et al., 2005). 3.2.5. Toughness Toughness of tire–rubber concrete was determined by calculating the area under the stress–strain curve up to 80% of the ultimate stress in the post-peak region. The toughness value is defined as a ratio between the area under the stress–strain curve up to 80% of the ultimate stress, to the area under the stress–strain curve up to the ultimate stress. Fig. 7a illustrates how each of the areas was determined. It should be noted that the 80% factor of ultimate stress was selected to limit further reductions in strength level. Thus, the toughness index (Ti) is expressed as follows Ti ¼
T 80% T 100%
ð4Þ
The toughness indices for different rubber concentrations and different mixtures are presented in Fig. 7b. Tire–rubber concrete exhibited greater toughness as compared to the plain concrete. Toughness indices maximize as rubber concentration approaches 25% of the total aggregate volume. Beyond rubber concentrations of 25%, toughness indices decrease due to the systematic reduction in strength. The toughness index for a combined C25F25 mixture is higher than that of the F50 and C50 mixtures. 3.2.6. Reduction factors (RF) for hardened concrete properties Mechanical property results of rubberized concrete showed that property values were primarily dependent on
2480
a
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
a
7 Ultimate Stress
6
All Data
F25
5
T 100% =A T 80% =A+B Toughness Index=(A+B)/B
4 3
SRF
Stress (MPa)
0.8
80% of Ultimate Stress
Poly. (All Data)
0.6 0.4
A
2
1.0
B
0.2
1 0
0.0 0
0.01
0.02
0.03
0.04
0.05
0
b
7
b
30
60
All Data
0.8
Fine
5
45
1.0
Coarse
6
Combined
Poly. (All Data)
4
ERF
Toughness Index
15
Rubber content by total aggregate volume (%)
Strain (mm/mm)
3
0.6 0.4
2 0.2
1 0 0
0.0
10 20 30 40 50 60 Rubber content by total aggregate volume (%)
0
Fig. 7. (a) Evaluation of toughness index and (b) toughness index values for various amounts of rubber contents.
the total rubber concentration. Therefore, it is convenient to establish a function to investigate the influence total rubber concentration has on the mechanical properties of concrete. Hence, the reduction factor (RF) is defined for a mechanical property as the ratio of the compressive strength or tangential modulus of elasticity in rubber-concretes containing a rubber concentration, R, to the value of the control plain concrete. The variation of the RF-determined compressive strength and tangential modulus of elasticity corresponding to rubber concentration are presented in Fig. 8a and b, respectively. The RF was unity for a 0% of rubber concentration (control specimen), and gradually decreased with increasing rubber concentration from 0% to 50% by total aggregate volume. Khatib and Bayomy (1999), in a similar study, examined several mathematical functions, including various degrees of polynomial functions, and proposed the following equation to quantify the reduction in the compressive strength SRF ¼ a þ bð1 RÞ
m
ð5Þ
Since the strength reduction factor (SRF) equals unity for 0% of R, the following condition can be satisfied aþb¼1
ð6Þ
In Eqs. (5) and (6), the SRF varies from 1 to 0; R is the rubber content in a volumetric ratio of total aggregate volume, and a, b, and m are the functional parameters. Khatib and Bayomy (1999) found the functional parameters a, b, and m to be 0.1, 0.9, and 7 in concrete aged 28 days, respectively. A similar procedure was conducted in the present study to examine the reduction factors for ultimate stress
15
30
45
60
Rubber content by total aggregate volume (%)
Fig. 8. (a) Strength reduction factor and (b) elastic modulus’ reduction factor of all types of rubberized concrete containing different rubber content.
and the tangential modulus of elasticity. A regression analysis was utilized to determine the functional parameters. Eqs. (7) and (8) yield the ultimate SRF and tangential elastic modulus reduction factor (ERF) in terms of rubber concentration in a volumetric ratio of total aggregate volume. The R-square values were determined to evaluate the accuracy of the equations 12
R square ¼ 0:99
ð7Þ
11
R square ¼ 0:98
ð8Þ
SRF ¼ 0:02 þ 0:98ð1 RÞ
ERF ¼ 0:01 þ 0:99ð1 RÞ
A comparison with results reported in literature demonstrates that the strength obtained in this study is lower than those reported by the others. For example, for a 7.5% total rubber concentration, the SRF attained by Khatib and Bayomy (1999) was 0.6 after 28 days; Li et al. (2004) obtained a SRF of 0.55 from a 7.5% total rubber concentration. Eq. (7) from this study gives the SRF a value of 0.4 from a concentration of 7.5% of total rubber. This anomaly can be attributed to a lack of pretreatment of the tire–rubber particles, improper calibration of equations based on concretes with higher tire–rubber contents, and tire type. 3.2.7. Sound absorption The ultrasonic echo technique is a valuable tool for testing concrete elements. The ultrasonic test is performed by an instrument composed of a transducer and a receiver. Acoustic waves were sent by the transducer and propagate
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
from the flat surface of cylindrical specimen, through the length of the specimen, and are received at the other flat side of the specimen. The time duration for an acoustic wave to propagate through the longitudinal direction of the specimen was recorded for each specimen in this study. The time duration for a specimen was evaluated several times to obtain a stable count. For all specimens, the velocity of the ultrasonic pulses was evaluated by using Eq. (9)
v¼
L t
2300
Coarse Unit weight (kg/m3 )
2100
Fine Combined
1900 1700 1500 1300 1100 0
10
20
30
40
50
Tire content (%) of total aggregates
Velocity of ultrasonic pulse (m/s)
b Coarse
5500
Fine
ð10Þ
where q and v are the hardened concrete unit weight and the velocity of the ultrasonic pulses, respectively. The variable, t, is Poisson’s ratio for the mixtures, and was assumed to be 0.3. Variation of Poisson’s ratio was neglected in this study. Hardened concrete unit weight, ultrasonic pulse velocity, and the values of the ultrasonic moduli for all types of the specimens are shown in Fig. 9. Fig. 9 indicates that the velocity of the ultrasonic waves reduce significantly with increasing tire–rubber content. Wave velocity transfer through a material is one of the most important factors on which mechanical wave energies depend. Therefore tire–rubber concrete is potentially a suitable material for the dampening of sound and other shaking energies, and can be used in noisy sites to serve as sound insulation. Due to the significant reduction in the ultrasonic modulus with increasing tire–rubber concentration, a porous composition is expected for tire–rubber concrete. 4. Conclusions and recommendations
3500
2500
0
10
20
30
40
50
Tire content (%) of total aggregates
Ultrasonic Modulus (GPa)
ð1 þ tÞð1 2tÞ ð1 tÞ
Combined
4500
1500
c
ð9Þ
where L is the specimen’s longitudinal length of 30 cm, and t is the duration obtained from the ultrasonic test. Ultrasonic moduli were evaluated using the velocities of the ultrasonic pulses. The ultrasonic modulus can be obtained using Eq. (10) (Neville, 1995) EU ¼ qv2
a
2481
60
Coarse
50
Fine Combined
40 30 20 10 0 0
10
20
30
40
50
Tire content (%) of total aggregates
Fig. 9. Results of the ultrasonic test: (a) hardened concrete unit weight, (b) influence of rubber–tire on ultrasonic pulse velocity in concrete and (c) values of the ultrasonic moduli for different concrete types.
1. Fresh rubberized concrete mixtures with increasing rubber concentrations present lower unit weights compared to plain concrete. Workability of rubberized concrete with coarse rubber particles is reduced with increasing rubber concentration; however, rubberized concrete with fine rubber particles exhibits an acceptable workability with respect to plain concrete. 2. The substitution of mineral aggregates with tire–rubber particles in concrete results in large reductions in ultimate strength and the tangential modulus of elasticity. Due to the considerable decrease in ultimate strength, rubber concentrations exceeding 25% are not recommended. Pretreatment of tire particle surfaces should be considered for possible improvement of tire–rubber concrete mechanical properties. An investigation is needed to identify the influence of rubber’s mechanical properties on the ultimate strength of rubberized concrete. 3. More ductile behavior is observed for rubberized concrete compared to plain concrete specimens under compression testing. Unlike plain concrete, the failure state in rubberized concrete does not occur quickly and does not cause any detachment in the specimen’s elements. Crack width in rubberized concrete is smaller than that
2482
A.R. Khaloo et al. / Waste Management 28 (2008) 2472–2482
of plain concrete, and the propagation of failure symptoms is more gradual and uniform. The failure state in tire–rubber concrete compared to plain concrete is characterized by more deformation. 4. The results of the combined mixture of tire particles were between the corresponding C and F specimens with the same total tire concentration. The stress–strain response of the combined mixtures demonstrates that the ultimate stress of the combined tire–rubber concrete is chiefly dependent on the fine aggregate concentration, while the shape of the stress–strain curve depends on the coarse aggregate concentration. These findings reveal that, although the mechanical properties of rubberized concrete are mainly dependent on the total rubber content, the concentrations of coarse or fine particles can be adjusted for different behavior. 5. Rubberized concrete is an effective absorber of sound and shaking energy. A large reduction in the ultrasonic modulus with increasing rubber concentration demonstrates a porous composition for rubberized concrete. This study has exclusively focused on the mechanical and physical properties of tire–rubber concrete for fine, coarse, and combined rubber replacements of mineral aggregates. There is a need for future studies to investigate energy absorption of tire–rubber concrete under dynamic loading, and also the durability of tire–rubber concrete under adverse weathering conditions. Acknowledgements Supports of Research Committee of Sharif University of Technology (SUT) and Center of Excellence in Structures and Earthquake Engineering are greatly appreciated. The authors are indebted to the staff of concrete laboratory at Civil Engineering Department of SUT for their valuable assistance.
References Albano, C. et al., 2005. Influence of scrap rubber addition to Portland I concrete composites: destructive and non-destructive testing. Compos. Struct. 71, 439–446. American Concrete Institute, 2002, Standard practice for selecting proportions for normal, heavyweight and mass concrete, Document No: ACI 211.1-91. ASTM, 1988, Concrete and Aggregate, Annual Book of ASTM Standards, vol. 04.02, Philadelphia. Eldin, N.N., Senouci, A.B., 1993. Rubber–tire particles as concrete aggregate. J. Mater. Civil. Eng. 5 (4), 478–496. Eldin, N.N., Senouci, A.B., 1994. Measurement and prediction of the strength of rubberized concrete. Cement Concrete Compos. 16, 287– 298. Fedroff, D., Ahmad, S., Savas, B.Z., 1996. Mechanical properties of concrete with ground waste tire rubber, Transportation Research, Record No. 1532, Transportation Research Board, Washington DC. Ghaly, A.M., Cahill, J.D., 2005. Correlation of strength, rubber content and water: cement ration in rubberized concrete. Can. J. Civil Eng. 32, 1–7. Guneyisi, E., Gesoglu, M., Ozturan, T., 2004. Properties of rubberized concretes containing silica fume. Cement Concrete Res. V 34, 2309– 2317. Hernandez-Olivares, F., Barluenga, G., Bollati, M., Witoszek, B., 2002. Static and dynamic behavior of recycled tire rubber-filled concrete. Cement Concrete Res. 32, 1587–1596. Khatib, Z.K., Bayomy, F.M., 1999. Rubberized Portland cement concrete. ASCE J. Mater. Civil. Eng. 11 (3), 206–213. Li, G., Garrick, G., Eggers, J., Abadie, C., Stubblefield, M.A., Pang, S., 2004. Waste tire fiber modified concrete. J. Compos. B (35), 305–312. Neville, A.M., 1995. Properties of Concrete, fourth ed. Longman, London. Segre, N., Joekes, I., 2000. Use for tire rubber particles as addition to cement paste. Cement Concrete Res. 30 (9), 1421–1425. Siddiquel, R., Naik, T.R., 2004. Properties of concrete containing scraptire rubber – an overview. Waste Manage. 24, 563–569. Topcu, I.B., 1995. The properties of rubberized concretes. Cement Concrete Res. 25 (2), 304–310. Toutanji, H.A., 1996. The use of rubber tire particles in concrete to replace mineral aggregates. Cement Concrete Compos. 18, 135–139.