Evaluation of Low-Temperature and Elastic Properties of
Crumb Rubber– and SBS-Modified Bitumen and Mixtures
Baha Vural Kök
1
; Mehmet Yilmaz
2
; and Alaaddin Geçkil
3
Abstract: In this study, the performances of bitumen and asphalt mixtures modified by crumb rubber (CR) were compared with those
modified by styrene-butadiene-styrene (SBS). The resultant mixtures were evaluated for their rheological and mechanical performances
by different experimental techniques such as rheological bitumen tests, i.e., dynamic shear rheometer (DSR), bendin g beam rheometer
(BBR), and hot mixture performance tests, that is, indirect tensile stiffness modulus, fatigue, semicircular bending, and toughness index.
Experimental studies show that it is necessary to use twice as much CR as SBS to reach the same performance attained by SBS. CR modi-
fication at high additive content exhibits higher elastic response, i.e., recoverable strain, than the SBS-modified mixture . While the resistance
to crack initiation of CR-modified mixtures increases with increasing additive content, the resistance to crack propagation decreases dra-
matically according to fatigue and semicircular bending tests. DOI: 10.1061/(ASCE)MT.1943-5533.0000590. © 2013 American Society of
Civil Engineers.
CE Database subject headings: Rubber; Rheology; Cracki ng; Mixtures; Elasticity; Temperature effects; Asphalts.
Author keywords: Crumb rubber; Styrene-butadiene-styrene; Rheological properties; Performance tests; Crack propagation.
Introduction
Highways produced with base asphalt cannot sustain low speeds
and heavy loads due to the asphalt’s drawbacks, such as low-
temperature cracking, poor rutting, and low fatigue resistance.
Previous studies showed that polymers improve rutting perfor-
mance, adhesion, and cohesion of an asphalt binder (Kanitpong
and Bahia 2005; Chen et al. 2002). Currently, the most commonly
employed polymer used for bitumen modification is styrene-
butadiene-styrene (SBS). However, it was detected that this modi-
fier not only decreases the workability of hot-mix asphalt (HMA)
but also fails to provide a cost-effective solution. The use of waste
materials has thus become an important issue in this respect.
Around the world, millions of tons of waste tires are generated
and discarded each year. The tire stockpiles and landfills cause
a number of problems to local communities, such as fire hazards
and environmental concerns. The use of these scrap tires to modify
asphalt is considered a significant issue because it contributes sig-
nificantly to environmental conservation and facilitates economic
sustainability by reducing the construction costs of modified roads
(Yildirim 2007). For example, it was discovered that the addition of
9% by weight ground tire rubber to bitumen increased both the lin-
ear viscoelastic modulus and the viscosity at high in-service tem-
peratures (Navarro et al. 2004). It was also reported that the
addition of recycled tire rubber to asphalt mixtures using a dry
process can improve the engineering properties of asphalt mixtures
(Cao and Chen 2008).
Crumb-rubber-modified (CR-modified) asphalt can be produced
by either wet or dry processes, though the former is more popular.
This is attributed to problems associated with the compatibility of
mixtures. The addition of crumb rubber (CR) has proven helpful in
increasing the voids in mineral aggregate in Superpave mix design
and improving the rutting resistance of asphalt mixtures regardless
of rubber size and type (Xiao et al. 2009). Chiu (2008) demon-
strated, by a 4-year field evaluation, the satisfactory performance
of asphalt-rubber-modified mixture and its potential to replace con-
ventional dense-graded mixes. Studies show that the rubber content
of asphalt-rubber mixtures has a significant effect on their perfor-
mance with respect to the resistance to permanent deformation
and cracking (Cao and Chen 2008). CR-modified asphalt was
determined to have the best low-temperature anticracking perfor-
mance at a CR content of between 5 and 20% according to bending
beam rheometer (BBR) tests (Liu et al. 2009 ). Rubber-modified
bitumen shows improved viscoelastic characteristics and, therefore,
higher viscosity than unmodified binders, indicating an improved
resistance to permanent deformation or rutting and low-temperature
cracking. It was also reported that rubber-modified bitumen (9% by
weight) shows very similar linear viscoelastic properties to SBS-
modified bitumen having 3% by weight SBS at 10°C and 7%
by weight SBS at 75°C (Navarro et al. 2002). Kok and Çolak
(2011) showed that to achieve the same performance with SBS
modification, the CR content must be higher than the SBS content.
Rubber particles of multiple sizes were also believed to have a bet-
ter sound-absorbing effect in spra y applications (Zhu and Carlson
1999). Several researchers demonstrated the improved performance
of bituminous mixtures with CR (Lalwani et al. 1982; McGennis
1995; Bahia and Davis 1994). Increased fatigue life, reduced reflec-
tive cracking and low-temperature cracking, and improved tensile
strength were cited as the advantages of CR-modified mixtures
(Oliver 2000).
In this study, the low-temperature performance and elastic prop-
erties of CR- and SBS-modified bitumen and bituminous mixtures
1
Fırat Univ., Dept. of Civil Engineering, Elazığ, Turkey (corresponding
author). E-mail: bvural@firat.edu.tr
2
Fırat Univ., Dept. of Civil Engineering, Elazığ, Turkey. E-mail:
mehmetyilmaz@firat.edu.tr
3
Fırat Univ., Dept. of Civil Engineering, Elazığ, Turkey. E-mail:
alaaddingeckil@hotmail.com
Note. This manuscript was submitted on November 29, 2011; approved
on May 23, 2012; published online on May 26, 2012. Discussion period
open until July 1, 2013; separate discussions must be submitted for indi-
vidual papers. This paper is part of the Journal of Materials in Civil En-
gineering, Vol. 25, No. 2, February 1, 2013. © ASCE, ISSN 0899-1561/
2013/2-257-265/$25.00.
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 257
J. Mater. Civ. Eng. 2013.25:257-265.
Downloaded from ascelibrary.org by Istanbul Teknik Universitesi on 02/18/13. Copyright ASCE. For personal use only; all rights reserved.
produced by wet process were compared. In the first stage,
the rheological properties of bitumen modified either with SBS
or CR were examined by dynamic shear rheometer (DSR)
and BBR tests. Next, the mechanical properties of SBS- or CR-
modified hot bituminous mixtures were investigated by indirect
tensile stiffness modulus, indirect tensile fatigue, toughness index
(TI), and semicircular bending tests.
Materials and Methods
Materials and Sample Preparation
Asphalt cement, B 160–220, obtained from Turkish Petroleum
Refineries was used as bitumen for mixture preparation. An SBS
polymer, Kraton D-1101, used in the study was supplied by Shell
Chemical Company. Five SBS-polymer-modified bitumen sam-
ples, (denoted as PMBs), were produced. The polymer con-
tents of these PBMs ranged from 2 to 6% by weight, wi th 1%
increments.
Similar to PBMs, five CR-modified bitumen samples were
obtained whose rubber content ranged from 2 to 10% by weight,
with 2% increments. Processing of scrap tires into CR can be ac-
complished through either ambient grinding or cryogenic grinding
technologies. The ambient processes was found to be more effec-
tive at producing CR-modified binders that are more viscous and
less susceptible to rutting and cracking. The use of rubber-particle
sizes of less than 0.35 mm and high shear rates during manufac-
turing operations is highly recommended (Navarro et al. 2004). In
this study, CR particle s having sizes between 0.30 and 0.60 mm
were obtained via ambient grinding processes.
Modified bitumens were produced with a laboratory-scale
mixing device with a four-blade impeller (IKA) at a temperature
of 180°C for 1 h at a rotation speed of 1,000 rpm.
Limestone aggregate was used in the asphalt concrete mixture.
The properties of the aggregate are given in Table 1. A crushed
coarse and fine aggregate, with a maximum size of 19 mm, was
selected as the dense-graded asphalt mixture. The gradation of
the aggregate mixtures is given in Table 2. The asphalt mixture
was designed in accordance with the standard Marshall mix design
procedure. The physical properties of the mixtures such as opti-
mum bitumen content (OBC), bulk specific gravity (Gmb), maxi-
mum specific gravity (Gmm), air void (Va), voids filled with
asphalt (VFA), and voids in mineral aggregates (VMA) are given
in Table 3.
Dynamic Shear Rheom eter and Bending Beam
Rheometer Tests
The principal viscoelastic parameters obtained from DSR are the
magnitude of the complex shear modulus (G
) and that of phase
angle (δ). G
is defined as the ratio of maximum (shear) stress
to maximum strain, and it provides a measure of the total resistance
to deformation when the bitumen is subjected to shear loading
(Airey et al. 2002). Permanent deformation is controlled by limiting
G
= sin δ to values greater than 1.0 kPa (before aging) and 2.2 kPa
[after rolling thin film oven (RTFO) aging]. Fatigue cracking is
controlled by limiting the G
sin δ value of the pressure-aged
(PAV) material to values less than 5000 kPa. The RTFO test is
assumed to simulate short-term aging by heating a moving film
of asphalt binder in an oven for 85 min at 163°C. The PAV is an
oven-pressure vessel combinat ion that takes RTFO-aged samples
and exposes them to high air pressure (2070 kPa) and temperature.
PAV is assumed to simulate the effects of long-term asphalt binder
aging that occurs as a result of 5 to 10 years of HMA pavement
service.
The DSR test was performed on base bitumen and SBS- and
CR-modified binders using a Bohlin DSRII rheometer for unaged,
RTFO-aged, and PAV-aged samples. The test was carried out unde r
controlled-stress loading conditions using 1.59 Hz frequency at five
different temperatures—52, 58, 64, 70, and 76°C—to determine the
high-temperature performance grades and also to compa re the elas-
tic components of the SBS- and CR-modified binders by evaluating
the phase angles.
Low-temperature cracking, commonly referred to as thermal
cracking, is the most recognized non-load-associated distress
(Mihai et al. 2004). Thermal cracking is caused by thermal-
shrinkage-induced stresses resulting from environmental cooling
Table 2. Aggregate Gradation
Sieve size (mm) Passing (%)
19 100
12.5 95
9.5 88
4.75 65
2.36 35
1.18 23
0.6 15
0.3 11
0.15 8
0.075 6
Table 3. Physical Properties of Mixtures
Mixture
type
OBC
(%)
Gb
(g=cm
3
) Gmb Gmm
Va
(%)
VMA
(%)
VFA
(%)
Base 4.90 1.0282 2.348 2.447 4.07 14.78 71.90
2%CR 5.05 1.0310 2.346 2.444 4.01 14.89 73.06
4%CR 5.20 1.0339 2.344 2.439 3.93 15.10 73.99
6%CR 5.31 1.0359 2.337 2.437 4.07 15.42 73.58
8%CR 5.44 1.0377 2.337 2.433 3.96 15.55 74.51
10%CR 5.57 1.0390 2.332 2.429 4.01 15.82 74.64
2%SBS 5.02 1.0268 2.345 2.443 4.00 14.89 73.13
3%SBS 5.14 1.0259 2.339 2.439 4.11 15.24 72.93
4%SBS 5.26 1.0253 2.336 2.435 4.07 15.42 73.62
5%SBS 5.38 1.0232 2.331 2.431 4.10 15.71 73.82
6%SBS 5.50 1.0226 2.328 2.427 4.08 15.91 74.37
Note: OBC = optimum bitumen content; Gmb = bulk specific gravity;
Gmm = maximum specific gravity; Va = air void; VMA = voids in
mineral aggregates; VFA = voids filled with asphalt; Gb = the specific
gravity of bitumen.
Table 1. Physical Properties of Aggregate
Property Standard
Specific
limit Coarse Fine Filler
Abrasion loss (%)
(Los Angeles)
ASTM C131
(ASTM 2006)
Max 30 29 ——
Frost action (%)
(with Na
2
SO
4
)
ASTM C88
(ASTM 2005)
Max 10 4.5 ——
Flat and elongated
particles (%)
ASTM D4791
(ASTM 2010a)
Max 10 4
Water absorption (%) ASTM C127
(ASTM 2012a)
Max 2 1.37
Specific gravity (g=cm
3
) ASTM C127
(ASTM 2012a)
2.613 ——
Specific gravity (g=cm
3
) ASTM C128
(ASTM 2012b)
— 2.622 —
Specific gravity (g=cm
3
) ASTM D854
(ASTM 2010b)
——2.711
258 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013
J. Mater. Civ. Eng. 2013.25:257-265.
Downloaded from ascelibrary.org by Istanbul Teknik Universitesi on 02/18/13. Copyright ASCE. For personal use only; all rights reserved.
(Zaniewski and Pumphrey 2004). The BBR is used to measure the
stiffness of binders at very low temperatures. The test uses engi-
neering beam theory to measure the stiffness of a small asphalt
beam sample under a creep load used to simulate the stresses that
gradually build up in a pavement when the temperature drops.
Creep stiffness and m-value are the two parameters evaluated with
BBR. The former measures how asphalt resists constant loading
and the latter measures how the asphalt stiffness changes as loads
are applied (Roberts et al. 1996). The creep stiffness of the asphalt
beam sample at any time of loading (t) is determined by
S ¼ PL
3
=ð4bh
3
δ
t
Þð1Þ
where S = creep stiffness (MPa); P = applie d constant load (N);
L = span length of beam sample (102 mm); b = beam width
(12.7 mm); h = beam thickness (6.35 mm); and δ
t
= deflection
(mm) at time t.
To prevent thermal cracking in Superpave, creep stiffness has a
maximum limit of 300 MPa, and the m-value has a minimum limit
of 0.300. For grading asphalt binders, the required performance
characteristics are determined, and the temperatures at which these
characteristics are satisfied establish the grade of the binder
(Malpass 2003). It was reported that a decrease in creep stiffness
leads to smaller tensile stresses in asphalt binders and reduced
chances for low-temperature cracking (Asphalt Institute 2003). Liu
et al. (2009) define the S=m-value ratio as coefficient λ; the smaller
its value, the better the low-temperature performance. In this study,
the test was performed at three different temperatures: −18, −24,
and −30°C. Hence, the low-temperature performance grades of the
binders were also determined. The binders are represented as
performance grade (PG) X-Y in Superpave specification, where
X is the highest pavement temperature rating, in degrees Celsius
(°C), Y is the lowest pavement temperature rating, in minus degrees
Celsius (°C), for which the PG binder was tested and expected to
perform.
Indirect Tensile-Stiffness-Modulus Test
The stiffness-modulus test of asphalt mixtures measured in
indirect-tensile mode is the most popular form of the stres s—strain
measurement methods used to evaluate the elastic properties of
these mixtures. The indirect tensile-stiffness-modulus (ITSM) test
defined by BS DD 213 is a nondestructive test. ITMS, which is
considered a very important performance characteristic for pave-
ment formulation, is defined as
S
m
¼ PðR þ 0.27Þ= tH ð2Þ
where S
m
= stiffness modulus in MPa; P = peak value of applied
vertical load (repeated load) (N); H = mean amplitude of horizontal
deformation (mm); t = mean thickness (mm); and R = Poisson ratio
(here assumed to be 0.35). The test was performed with deforma-
tion controlled using a universal testing machine (UTM). Target
deformation was selected as 5 μm. During testing, the rise time,
which is the time that passes for the applied load to increase from
zero to a maximum value, was set at 124 ms. The load-pulse
application was set to 3.0 s. The test was performed at 5°C.
Indirect Tensile-Fatigue Test
Fatigue is one of the most significant distress modes in pavements
associated with repeated traffic loads (Ye et al. 2009). In this study,
a constant-stress indirect tensile-fatigue test was conducted by ap-
plying a cyclic constant load of 500 kPa for 0.1 s followed by a rest
period of 1.4 s. The measurements were carried out using a UTM
at 5°C. The deformation of the specimen was monitored through
linear variable-differential transducers clamped vertically onto the
diametrical side of the specimen. A repeated dynamic compressive
load was applied to specimens across the vertical cross section
along the depth of the specimen using two loading strips 12.5 mm
in width. The resulting total defo rmation parallel to the applied
force was measured.
Toughness Index Test
The toughness index (TI) calculated from the indirect tensile test is
a parameter describing the toughening characteristics in the post-
peak region. The indirect tensile strength (ITS) test was used to
determine tensile strength and strain of the sample. Cylindrical
specimens were monotonically loaded to failure along the vertical
diametric axis at a constant rate of 50.8 mm=min. Base d upon the
maximum load at failure, the ITS in kilopascals was calculated
from the following equation:
ITS ¼ 2P=πtD ð3Þ
where P = peak value of applied vertical load (kN); t = mean
thickness of test specimen (m); and D = specimen diameter (m).
A dimensionless indirect tensile TI, the TI is defined as follows:
TI ¼ðA
e
− A
p
Þ=ðε − ε
p
Þð4Þ
where A
e
= area under normalized stress—strain curve up to strain
ε; A
p
= area under normalized stress—strain curve up to strain
ε
p
; ε = strain at point of interest; and ε
p
= strain corresponding
to peak stress. The TI compares the elastic performance of a speci-
men with that of a perfectly elastic reference material for which the
TI remains constant at 1.0. On the other hand, for an ideal brittle
material without postpeak load carrying capacity, the value of TI
equals zero (Kabir 2008). In this study, the values of the indirect
tensile TI were calculated up to tensile strains of 1, 2, and 3%. The
test was performed at 25°C.
Semicircular Bending Test
This test method determines the tensile strength or fracture tough-
ness of an asphalt mixture for the assessment of the potential for
crack propagation. The crack propagation phase describes the sec-
ond part of a failure mechanism during dynamic loading. The test
was conducted according to EN 12697-44 at 0°C. The specimens
were prepared with a gyratory compactor 150 mm in diameter and
50 mm thick. They were cut into two equal semicircular parts from
the middle. A single notch 10 mm deep was cut in the middle of the
specimens loaded by applying a constant cross-head deformation
rate of 5.0 mm/min. The load and deformation were recorded
continuously, and the frac ture toughness (K
Ic
N=mm
3=2
) was de-
termined using the following equations:
σ
max
¼ 4.263.F
max
=D:t ð5Þ
K
Ic
¼ σ
max
.5.956 ð6Þ
where σ
max
= maximum stress at failure; F
max
is the maximum
force in newtons; D = diameter of specimen in millimeters;
t = thickness of specimen in millimeters. The fracture energy
was also calculated according to the TC-50-FMC specification,
which was used for asphalt mixtures by Li and Marasteanu
(2010). The fracture energy (G
f
) can be obtained by the following
formula:
G
f
¼ W
0
=A
lig
ð7Þ
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 259
J. Mater. Civ. Eng. 2013.25:257-265.
Downloaded from ascelibrary.org by Istanbul Teknik Universitesi on 02/18/13. Copyright ASCE. For personal use only; all rights reserved.
where W
0
= fracture work, the area below the measured load-
deformation curve; A
lig
= area of ligament, which is the product
of the notch depth and the thickness of the specimen.
Results and Discussion
DSR and BBR Test Results
The complex modulus G
and phase angle δ of the base and
modified binders were measured at different temperatures. The
G
= sin δ of neat and RTFO-aged binders and G
. sin δ of PAV-aged
binders were also determined at the desired temperature to specify
the high-temperature performance grades. DSR test results are
given in Table 4. The base and 2% CR-modified binders were
graded as PG 58. A grade of PG 64 was assigned to 2 and 3%
SBS-modified and 4 and 6% CR-modified binders. A grade
of PG 70 was assigned to 4% SBS-modified and 8 and 10%
CR-modified binders. A grade of PG 76 was assigned to 5 and
6% SBS-modified binders. The rutting parameter (G
= sin δ) values
of SBS modification were higher for all temperatures than those of
CR modification (Fig. 1). This can result from either a higher com-
plex modulus or lower phase angles or the combined effects of fac-
tors. Obviously, high CR content is required to reach the same
rutting parameter with SBS modification. The variation of phase
angles with G
= sin δ is plotted in Fig. 2 to evaluate the elastic
behavior of binders at the same rutting parameter level.
It is apparent that modified binders are more flexible than base
bitumen. There was almost no difference between the modified
binders’ curve at low G
= sin δ values. However, the difference be-
gan to increase at high G
= sin δ values. When G
= sin δ values of
the CR-modified binders and unmodified binders equal to each
other CR modification gives low phase angles, indicating an elastic
response. The G
= sin δ indexes were determined by dividing the
value of G
= sin δ obtained at 52°C to that obtained at 76°C
(Fig. 3). The SBS-modified binders had a higher slope than CR-
modified binders, indicating less temperature susceptibility for high
temperatures. Although CR modification can provide the same
Table 4. DSR Test Results
Temperature
(°C)
G
= sin δ (Pa)
Base 2%SBS 3%SBS 4%SBS 5%SBS 6%SBS 2%CR 4%CR 6%CR 8%CR 10%CR
52 2263 5168 7452 9551 15001 22133 3346 4486 6514 8788 11710
58 1182 2387 3690 4735 7551 11803 1597 2158 3254 4495 6182
64 523 1194 1788 2444 3896 6352 815 1084 1660 2359 3239
70 265 615 902 1271 2104 3276 423 559 876 1242 1717
76 333 484 687 1202 1801 228 304 471 659 923
δ
52 79.96 72.66 66.33 65.40 61.70 60.10 76.40 72.71 69.16 65.18 62.68
58 81.43 76.26 71.48 69.98 66.61 61.50 79.34 76.30 72.66 68.61 66.02
64 83.99 78.88 76.15 73.39 70.26 66.30 81.79 78.97 75.81 72.89 69.86
70 84.85 80.68 79.84 75.70 72.70 70.21 83.83 81.43 78.84 76.36 73.69
76 81.75 82.49 78.04 74.80 73.02 84.02 83.26 81.53 79.74 77.25
G
. sin δ (Pa.10
6
) PAV residue
16 2.2050 1.7760
19 1.6270 1.3463 1.2240 1.2874 1.2753 1.5260
22 0.6998 1.1521 0.4947 0.9439 1.0964 0.8792 0.9635
25 0.3590 0.5840 0.6504 0.7265
28 0.3500 0.6310
31 0.5010
G
. sin δ (Pa) RTFOT residue
58 7277 9552
64 10960 11938 6849 10560
70 7602 7701 9294
76 6340 8311
Note: SBS = styrene-butadiene-styrene; CR = crumb rubber; PAV = pressure-aged value.
Fig. 1. Variation in G
= sin δ versus additive content Fig. 2. Relation between G
= sin δ and phase angles
260 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013
J. Mater. Civ. Eng. 2013.25:257-265.
Downloaded from ascelibrary.org by Istanbul Teknik Universitesi on 02/18/13. Copyright ASCE. For personal use only; all rights reserved.
performance as SBS modification at low SBS contents, CR demand
increases significantly to provide the same performance as SBS
modification at high SBS contents.
From the BBR tests at different temperatures, shrinkage resis-
tance parameters, stiffnesses, and m-values of the control, SBS- and
CR-modified binders were calculated, and the results are shown in
Table 5. The m-values of all binders decreased and creep stiffness
values increased with decreases in temperature. There was no
arrangement within the m-values of binders at different additive
contents. Furthermore, the variation of the creep stiffness values
of the SBS-modified binders was not regular with increases in ad-
ditive content. However, the creep stiffness values of CR-modified
binders, except for 2% CR, decreased regularly with increases in
CR content and increased regularly with decreases in temperature.
These results indicate that a more homogeneous blend occurs at
different CR contents than at different SBS contents. The minimum
creep stiffness value of the CR-modified binder belongs to 10%
modification and is 30% lower than that of the base bitumen.
The minimum creep stiffness value of the SBS-modified binder re-
sulted from 6% modification and is 6% greater than that of the base
bitumen. As seen from Table 5, the creep stiffness values of all
binders were no gr eater than 300 MPa, even at the highest additive
content and at the lowest temperature. Hence, the m-values deter-
mined the low-temperature performance grades. It was determined
that all binders satisfied the requirements at −24°C (PG-34) except
for 6% SBS-modified binder, which satisfied the requirements at
−18°C (PG-28). Flexible materials exhibit high deflection at low
temperatures. In this respect, it is obvious that the deflection values
of CR-modified binders increased steadily with additive content
at all temperatures. Moreover, 10% CR-modified binder had
higher deflection values than the base bin der. This also indicates
a potential to dissipate the energy induced by loading. Irregular de-
flection values were attained for the SBS-modified binders. It was
seen that 10% CR modified binder had the largest deflections
among the modified binders, and only this binder had higher de-
flections than base bitumen at all temperatures. With the DSR test
results, the performance grades of the base and modified binders
were determined (Table 5). The performance grade as PG 64-34
is ensured by 2 and 3% SBS modifications and by 4 and 6%
CR modifications. The same grade as PG 70-34 is ensured by
4% SBS modification and by 8 and 10% CR modifications.
The variations on creep stiffness of the same graded binders are
given in Figs. 4 and 5. Expectedly, there is almost no change in
creep stiffness values at −24°C for the binders (Fig. 4) since they
have the same low-temperature performance grade. However,
CR-modified binders seem more flexible than SBS-modified bind-
ers by exhibiting lower stiffness values at −18°C and −30°C. As
PG 64-34 binders are likely to be subjected to temperatures below
−34°C frequently, CR-modified binders having low stiffness values
guarantee a longer service life of pavement. SBS-modified binders
give higher stiffness at all temperatures (Figs. 4 and 5). Therefore,
it is economically advisable to prefer 6% CR to 3% SBS for
PG 64-34 and 10% CR to 4% SBS for PG 70-34.
An increase in creep stiffness causes an increase in thermal
stresses, leading to thermal cracking. While m-values decrease, the
rate of stress relaxation also decreases and the ability to relieve
thermal stresses by flow decreases in HMA pavement (Roberts
et al. 1996). Since lower stiffness values and higher m-values
indicate a good low-temperature anticracking property, the
S=m-value ratio was also evaluated (Figs. 6 and 7). It is seen from
the figures that the effects of temperature on S=m-values of the
SBS- and CR-modified binders are different. While S=m-values
of SBS-modified binders remain constant, those of CR-modified
binders decrease with increases in additive content at −18°C.
The S=m-values values started to increase smoothly with increasing
additive content for SBS-modified binders and increased up to 4%
CR content and began to decrease at −24°C. A similar trend was
observed at −30°C for all binders. The lowest S=m values were
obtained by 5% SBS-modified and 10% CR-modified binders at
−30°C which is severe condition for roads. Any of the binders
can reach a flexibility of base bitumen at −24 and −30°C. However,
at 4, 6, 8, and 10% CR modifications more flexible behavior was
evident than in B 160/220 base bitumen at −18°C.
Fig. 3. Relation between G
= sin δ indexes and additive content
Table 5. BBR Test Results
Temperature
(°C)
m-values
Base 2%SBS 3%SBS 4%SBS 5%SBS 6%SBS 2%CR 4%CR 6%CR 8%CR 10%CR
−18 0.362 0.355 0.372 0.330 0.338 0.345 0.383 0.359 0.370 0.352 0.358
−24 0.338 0.328 0.313 0.313 0.303 0.274 0.325 0.324 0.316 0.310 0.307
−30 0.270 0.239 0.260 0.216 0.242 0.245 0.257 0.255 0.265 0.269 0.248
Creep stiffness (Mpa)
−18 89.31 97.22 99.41 91.92 100.72 94.74 97.54 77.39 68.56 63.27 62.06
−24 112.45 144.25 149.85 138.97 142.67 158.84 123.94 151.59 140.23 134.90 112.91
−30 129.31 224.55 267.15 207.53 202.55 207.25 134.22 159.71 154.87 148.80 122.45
Deflection (mm)
−18 0.893 0.825 0.801 0.880 0.796 0.843 0.820 1.034 1.161 1.266 1.292
−24 0.709 0.541 0.533 0.583 0.570 0.514 0.643 0.527 0.594 0.591 0.710
−30 0.620 0.362 0.299 0.387 0.398 0.390 0.594 0.493 0.515 0.536 0.651
PG
58-34 64-34 64-34 70-34 76-34 76-28 58-34 64-34 64-34 70-34 70-34
Note: SBS = styrene-butadiene-styrene; CR = crumb rubber.
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / FEBRUARY 2013 / 261
J. Mater. Civ. Eng. 2013.25:257-265.
Downloaded from ascelibrary.org by Istanbul Teknik Universitesi on 02/18/13. Copyright ASCE. For personal use only; all rights reserved.
