Materials and Design 61 (2014) 8–15

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

Effect of waste plastic bottles on the stiffness and fatigue properties

of modified asphalt mixes
Amir Modarres ⇑, Hamidreza Hamedi
Department of Civil Engineering, Babol Noshirvani University of Technology, Babol, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Nowadays, the use of recycled waste materials as modifier additives in asphalt mixes could have several
Received 22 October 2013 economic and environmental benefits. The main purpose of this research was to investigate the effect of
Accepted 15 April 2014 waste plastic bottles (Polyethylene Terephthalate (PET)) on the stiffness and specially fatigue properties
Available online 29 April 2014
of asphalt mixes at two different temperatures of 5 and 20 °C. Likewise, the effect of PET was compared to
styrene butadiene styrene (SBS) which is a conventional polymer additive which has been vastly used to
Keywords: modify asphalt mixes. Different PET contents (2–10% by weight of bitumen) were added directly to mix-
Waste plastic bottles
ture as the method of dry process. Then the resilient modulus and fatigue tests were performed on cylin-
Stiffness
Fatigue
drical specimens with indirect tensile loading procedure. Overall, the mix stiffness reduced by increasing
Asphalt mixes the PET content. Although stiffness of asphalt mix initially increased by adding lower amount of PET.
Based on the results of resilient modulus test, the stiffness of PET modified mix was acceptable and war-
ranted the proper deformation characteristics of these mixes at heavy loading conditions. At both tem-
peratures, PET improved the fatigue behavior of studied mixes. PET modified mixes revealed
comparable stiffness and fatigue behavior to SBS at 20 °C. However, at 5 °C the fatigue life of SBS modified
mixes was to some extent higher than that of PET modified ones especially at higher strain levels of
200 microstrain.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction The main reason of incorporating polymer modifiers in bitu-

During the recent years, engineers have been looking for new
environmental friendly techniques in construction of roads pave-
ment and much studies have been devoted to this research field
(e.g. utilizing recycled asphalt pavement (RAP) materials, crumb
rubber, construction debris, etc.) [1–3]. During the service life,
many external factors might deteriorate the integrity of pavement.
Among these factors, traffic loading is considered as the main fac-
tor which finally leads to fatigue cracking and permanent deforma-
tions especially in upper pavement layers. There are vast majority
of cases which addressed the fatigue properties of conventional or
modified asphalt mixes. Effects of many parameters and additives
have been studied in this regard [4,5]. Different additive materials
including fibers and polymers have been used to improve the fati-
gue resistance of asphalt mixes. Most of these materials were
found to be effective with beneficial effects on the fatigue behavior
of asphalt mixes [6–10].
⇑ Corresponding author. Tel.: +98 9111163215.
E-mail addresses: a.modarres@nit.ac.ir, amirmodarres2003@yahoo.com
(A. Modarres).
mens is to extend the range of service temperature or reduce the
temperature sensitivity of them. These binders are visco-elastic
materials. The degree to which their behavior is viscous or elastic
is a function of temperature, loading period and loading duration.
At high temperatures or long loading times, they behave like vis-
cous liquids whereas at low temperatures or short loading times
they behave as elastic (brittle) solids. Under intermediate condi-
tions of the service period, they exhibit viscoelastic behavior in
which the material’s response will be dependent upon tempera-
ture or loading velocity. For a polymer to be effective in road appli-
cations, it should be blend with bitumen and improve its efficiency
at service temperatures without making it too viscous at mixing
temperatures or too brittle at low temperatures. In other words,
it must extent the range of service temperature while it improves
the overall performance of pavement.
Polymers that have been used for asphalt mix modification can
be divided into three groups including thermoplastic elastomers
(e.g. styrene butadiene styrene (SBS) and crumb rubber (CR)), plas-
tomers (e.g. ethylene vinyl acetate (EVA) and polyethylene (PE))
and polymers with chemical reaction [11–13].
Thermoplastic elastomers such as SBS are usually used to
extend both minimum and maximum service temperatures of

http://dx.doi.org/10.1016/j.matdes.2014.04.046
0261-3069/Ó 2014 Elsevier Ltd. All rights reserved.
 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15
bitumen, whereas plastomers are well known as effective additives
at high service temperatures [11,12].
Although, the use of polymer modifiers has been recognized as
an appropriate solution for promoting the engineering properties
of bitumen and asphalt mixes, but it is relatively a costly procedure
for paving roads [14,15]. From an environmental and economic
point of view, the use of recycled instead of virgin materials could
have several advantages such as help easing landfill pressures and
reducing demands of extraction from natural quarries. Further-
more, this would be an alternative solution for environmental pol-
lution by utilizing waste materials as secondary materials in road
construction projects. As published in the literature, the waste of
glass, rubbers, plastics and mineral productions were some popular
materials used to modify the properties of bitumens and asphalt
mixes [16]. Most researches have focused on using waste additives
to improve the deformation and fatigue characteristics of asphalt
mixes. According to research results, waste glass and waste rubber
had a considerable contribution to fatigue resistance of these
mixes [17,18].
Nowadays, many countries are seriously encountered with
problems related to waste plastic materials. Plastic materials such
as plastic bottles are mainly composed of Polyethylene Terephthal-
ate (PET) polymer. PET is a thermoplastic polymer resin of the
polyester family and is used in synthetic fibers, beverage, food
and other liquid containers, thermoforming applications and engi-
neering resins often in combination with glass fiber [19]. PET is
produced by the polymerization of ethylene glycol and tere-
phthalic acid. Ethylene glycol is a colorless liquid obtained from
ethylene, and terephthalic acid is a crystalline solid obtained from
xylene. When heated together under the influence of chemical cat-
alysts, ethylene glycol and terephthalic acid produce PET in the
form of a molten, viscous mass that can be spun directly to fibers
or solidified for later processing as a plastic.
Based on previous studies PET has a great potential to be reused
as modifier in asphalt mixture. Results indicated that adding PET to
asphalt raised the mix resistance against permanent deformation
and rutting [20,21]. During a laboratory study Mahrez & Karim
examined the effect of different PET contents on the rheological
properties of modified bitumen. They found that addition of PET
to bitumen will increase the viscosity and reduce the temperature
susceptibility of modified bitumen. Furthermore, the PET modified
bitumen showed preferable elastic properties than the original one
(i.e. higher complex modulus and lower phase angle) [22]. During a
laboratory study about stone matrix asphalt (SMA) mixes the effect
of PET was investigated using the cylindrical specimens. It was
inferred that incorporating PET will reduce the bitumen loss which
is one of the main SMA deficiencies. Furthermore, the effect of PET
on the moisture susceptibility of these mixes was found to be neg-
ligible [23]. In a 2012 study Moghaddam et al. compared the stiff-
ness and fatigue properties of PET modified mixes with
conventional asphalt. Based on their report the fatigue life of mod-
ified mix containing 1% PET (by weight of aggregate) was twice
than that of unmodified mix. However, the stiffness of modified
mix was to some extent lower than conventional mix. The out-
comes of this research indicated that the application of PET in
SMA mixes could meet the various requirements of different envi-
ronmental and loading conditions. Especially the results of stiff-
ness test warranted the proper deformation characteristics of
modified mixes at heavy loading conditions [24].
The addition of thermoplastic polymers (e.g. PET) to bitumen or
asphalt mix enhances the material rigidity and restricts the perma-
nent deformations under heavy loading conditions especially in
upper pavement layers at higher temperatures [25]. The beneficial
effects of PET on such high temperature characteristics of asphalt
mixtures have been proved elsewhere [22,24]. However their per-
formance in increasing the bitumen elasticity during drastic and
9
sudden temperature drops is not always satisfactory. In fact they
might deteriorate the intermediate and low temperature charac-
teristics of bitumen and asphalt mix (i.e. increasing the cracking
potential of mix) [25].
Apart from abovementioned investigations, there is not enough
information regarding to fatigue properties of PET modified mixes.
For example the fatigue response of these mixes at various temper-
atures has not been well established. Since extending the range of
service temperature is the main purpose of bitumen and asphalt
modification it will be interesting to investigate the fatigue proper-
ties of PET modified mixes at various temperatures.
Hence, in this study, the fatigue and stiffness properties of PET
modified mixes have been investigated at intermediate and low
temperatures. In this regard the effect of PET was compared with
SBS which is a conventional polymer modifier in asphalt mixes
and most of earlier researches have proved the beneficial effects
of this additive on the technical characteristics of asphalt mixes
[12,21,25].
The main objectives of this research were as follows:
 To investigate the effects of PET on stiffness properties of
modified mixes at two testing temperatures.
 To evaluate the fatigue behavior of PET modified mixes in
comparison with unmodified asphalt mixes.
 To compare the stiffness and fatigue properties of PET mod-
ified mixes with that of modified with SBS.
2. Materials and mix design
2.1. Bitumen and aggregate
The original binder used in this study was 60/70 penetration
grade bitumen that produced in Tehran oil refinery. Table 1, pre-
sents the basic properties of this bitumen. Also, as shown in
Fig. 1 a 0–12.5 mm aggregate gradation was selected which was
approximately in the middle limit of specifications. Table 2 sum-
marizes the specifications of coarse and fine aggregate fractions
and filler materials which were blended to achieve the final
gradation.
2.2. Additives
2.2.1. PET
In this study, waste plastic bottle (PET) was used as modifier
additive in hot mix asphalt. To this end, PET bottles were cut into
small pieces and crushed by a special crusher. Finally crushed par-
ticles were sieved to obtain the needed gradation. As indicated by
previous researches, desired results were obtained by single size
PET particles between the range of 0.425–1.18 mm [23,25]. Hence,
in this research, PET chips were crushed and sieved to obtain the
above-mentioned dimensions. Fig. 2 shows the image of the PET
crumbs after the crushing and sieving process. PET consists of poly-
merized units of the monomer ethylene terephthalate, with
repeating C10H8O4 units. The related components of studied PET
were terephthalic and ethyleneglycol monomers. The physical
Table 1
Technical properties of original bitumen.
Property (unit) Standard Value
Specific gravity ASTM: D70 1.013
Penetration (0.1 mm) ASTM: D5 65
Softening point (°C) ASTM: D36 50
Viscosity at 120 °C (cSt) ASTM: D2170 966
Viscosity at 135 °C (cSt) ASTM: D2170 467
Viscosity at 160 °C (cSt) ASTM: D2170 168
10 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15

Fig. 1. The aggregate gradation of studied hot mix asphalt.

Table 2 Fig. 3. The viscosity–temperature diagram of bitumen used to select the mixing
Properties of coarse and fine aggregate fractions and filler materials. and compaction temperature ranges.

Property (unit) Standard Unit Value 2.3. Mix design

Coarse aggregate
Water absorption ASTM: C127 (%) 2.2 Modifier additives are usually added to mixture under wet or
Bulk specific gravity ASTM: C127 (gr/cm3) 2.498 dry conditions. During the wet process, additive first mixed with
Apparent specific gravity ASTM: C127 (gr/cm3) 2.663
bitumen with a proper mixer until achieving a homogenous blend.
Then blended materials are added to aggregates. In dry method,
Fine aggregate
Water absorption ASTM: C128 (%) 2.4
according to additive’s type and nature this material is mixed with
Bulk specific gravity ASTM: C128 (gr/cm3) 2.467 aggregates before adding bitumen or added after mixing the bitu-
Apparent specific gravity ASTM: C128 (gr/cm3) 2.623 men and aggregates as a part of solid materials. Due to high melt-
ing point, it was not possible to mix the PET particles with bitumen
Filler in wet process. In fact it was impossible to achieve a homogeneous
Specific gravity ASTM: D854 (gr/cm3) 2.665 mixture in this process. The impossibility of achieving a desired
blend through adding PET in wet process has been also mentioned
in literature [26]. Therefore, in this study, dry method was fol-
lowed and PET was added to mix with various quantities of 2%,
4%, 6%, 8% and 10% by the weight of bitumen. The optimum bitu-
men content for unmodified mix was equal to 5.7% which was
determined with Marshall method. Based on previous studies the
optimum bitumen content for PET modified mixes was almost
equal to unmodified mix [24]. Therefore, in this research the same
bitumen content was selected for modified and unmodified mixes.
The mixing and compaction temperatures were determined by
viscosity–temperature diagram. This diagram has been shown in
Fig. 3. The viscosity of bitumen during the mixing and compaction
process has been recommended between 150–190 and 250–
310 cst, respectively. On the basis of this criterion the mixing and
compaction temperatures were kept constant between 157–162
and 145–150 °C, respectively. For each composition first aggregate
and bitumen were mixed at abovementioned temperature range
and then PET particles were added directly to mixture. Previous
studies revealed that the viscosity of PET modified bitumen is to
some extent higher than that of unmodified mixes. However, even
for PET contents as high as 8% (by the weight of bitumen) the dif-
ference between the viscosity of the original and modified bitumen
Fig. 2. Image of crushed PET particles.
was negligible especially at mixing and compaction temperatures
properties of this additive were analyzed using related standard (i.e. higher than 130 °C) [22]. Therefore, PET modified mixes were
methods. Based on performed analysis the density of PET was mixed and compacted at similar temperature ranges of unmodified
equal to 1.08 gr/cm3. Also the glass transition and melting point ones.
temperatures were equal to 250 and 70 °C, respectively. SBS modified mixes were prepared by wet process. First, by the
means of a high shear mixer, SBS was added to bitumen with the
values of 4%, 5% and 6% by the weight of bitumen. Then, the pre-
2.2.2. SBS pared SBS-bitumen blend was added to aggregate. After the mixing
The optimum SBS content was determined based on the results and compaction process, prepared mixes were tested to measure
of resilient modulus test. Then asphalt mixes were prepared with the resilient modulus. Based on the criterion of maximum resilient
the optimum SBS content and the fatigue tests were performed. modulus the optimum SBS content was selected equal to 5%.
Finally the fatigue response of SBS modified mixes were compared Finally, indirect tensile fatigue test was performed for mixes con-
to PET modified ones. taining 5% SBS.
 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15
3. Experimental design
The main laboratory program of this research consisted of the
resilient modulus and fatigue tests. These tests were performed
on cylindrical specimens using indirect tensile method, according
to ASTM: D4123 & EN 12697-24, respectively [27]. In order to
determine the stress level in above-mentioned tests indirect ten-
sile strength (ITS) was measured according to ASTM: C496. All tests
were accomplished at two testing temperatures of 5 and 20 °C. A
universal testing machine (UTM-14) apparatus was used which
had been equipped with a temperature control chamber. Chamber
contained a reference specimen with two linear variable differen-
tial transducers (LVDTs) that measured and recorded the skin
and core temperatures during the test. In order to achieve the
intended temperature, specimens were put inside the chamber at
least 5 h before testing. Before starting the test, the chamber, skin
and core temperatures were controlled by related software. Test
was started when the coefficient of variation of these three tem-
peratures which automatically calculated by the controlling sys-
tem software was less than 5%.
3.1. Resilient modulus (Mr)
During the Mr test a haversine load was applied with the load-
ing frequency of 1 Hz including 0.1 s loading and 0.9 s recovery
times. Horizontal deformations were measured by two LDVTs that
were installed along the sample’s diameter. At 5 °C resilient mod-
ulus test was done at two stress levels of 15% and 20% of ITS. More-
over, at 20 °C stress levels were fixed to 20% and 40% of ITS. For a
dynamic load of P, resilient modulus was calculated by Eq. (1):
Pðc þ 0:27Þ
Mr ¼
td
where P: maximum dynamic load (N), c: poisson’s ratio, t: sample
height (mm), d: horizontal deformation (mm).
Possion’s ratio was calculated according to Eq. (2) [28]:
0:35
c ¼ 0:15 þ
1 þ eð3:18490:04233tÞ
where t: temperature (°F)
3.2. Fatigue
At each temperature fatigue test was performed at two stress
levels utilizing indirect tensile loading method with a haversine
loading [27]. Each load pulse consisted of 0.25 s loading and
1.25 s recovery times. Loading continued until complete splitting
of samples. For indirect tensile fatigue test the maximum tensile
stress and strain at the center of sample were calculated by
Eqs. (3) and (4), respectively.
2p
St ¼
ptd
  
2 DH 1 þ 3c
e¼
D 4 þ pc  p
where p: maximum dynamic load (N), t: sample height (mm),
D: sample diameter (mm), e: tensile strain at the center of samples,
DH: horizontal deformation which measured by two LDVTs.
During the fatigue testing horizontal deformations were auto-
matically recorded and deformation–loading curve was plotted
for each specimen by related software. Fig. 4 shows an example
of horizontal deformation curve that obtained in this research. In
this figure two fatigue life definitions were compared together
(N1 & N2). For the first definition, according to EN12697-24, fatigue
life is equal to the total number of cycles which leads to complete
11
break of sample. As seen in this figure for this definition fracture
life corresponds to the point of vertical asymptote [27].
Likewise, Fig. 4 depicts the second definition of fracture life (N2)
in indirect tensile method [29]. As it can be seen, the diagram of
horizontal deformation is generally defined by the three zones.
The accumulated permanent deformations rapidly increase in the
primary zone. In the second zone, the rate of deformation incre-
ment gets stabilized and the fatigue curve has a linear trend. In
the third zone, microcracks which formed in the second stage will
progress. Finally the progress and combination of these cracks
leads to complete splitting of specimen. According to Fig. 4 in the
second definition (N2) the start point of crack progression is
defined as the fracture life. As shown this point corresponds to
the intersection point of the second and third zones slope.
4. Results and discussion
4.1. Indirect tensile strength (ITS) and resilient modulus (Mr)
Fig. 5 shows the results of ITS test. As seen temperature had
considerable effects on the ITS of specimens. Addition of 2% PET
led to increase of ITS at both testing temperatures. After that, ITS
continuously decreased by adding the PET content. At higher PET
contents, bitumen accumulates on the surface of the PET particles.
This issue results in the reduction of the bitumen film thickness
around the aggregate particles and reduces the aggregate–bitumen
adhesion and finally the tensile strength of the modified mix. How-
ever, at all PET contents the ITS values were in acceptable limit.
Based on previous studies due to reduction of the bitumen film
thickness, excessive amounts of PET will also reduce the moisture
ð1Þ resistance of the modified mix [23]. Results of resilient modulus
test at 5 and 20 °C, have been shown in Fig. 6. As seen, the stiffness
of studied mixes reduced by increasing the stress level. However,
similar to ITS test, at a constant stress level the highest stiffness
quantity achieved at 2% PET content. At higher PET contents aggre-
gates will be replaced by these particles which have less stiffness.
ð2Þ Moreover, the reduction of the bitumen film around the aggregate
particles might be another reason of stiffness reduction especially
at higher PET contents.
According to Fig. 6 the resilient modulus of studied mixes
increased to twice by reducing the temperature from 20 °C to
5 °C. Similar to obtained results at 20 °C, at higher PET contents
(i.e. more than 2%) there was a drop in the resilient modulus of
studied mixes at 5 °C. However, due to noticeable stiffening of
bitumen the dispersion of results were higher at this temperature.
The interaction between bitumen and additive materials in
modified mixes could have considerable effect on the behavior of
ð3Þ
ð4Þ
Fig. 4. The load cycle–displacement curve and fatigue life definitions in ITFT
method.
12 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15
Fig. 5. Results of ITS test at 5 and 20 °C.
mixture. Studies indicated that this interaction changes with
increasing the amount of additive. At high polymer contents the
polymer phase becomes dominant, whereas, at optimum content
of additive there will be an effective interaction which improves
the mechanical properties of asphalt mix [30].
In order to evaluate the temperature sensitivity of studied
mixes, the rate of resilient modulus (Mr) changes with an increase
in temperature was investigated. Fig. 7 shows the effect of temper-
ature on Mr of studied mixes. It should be noted that this test was
performed at two temperatures and two stress levels at each tem-
perature. Also at each condition (i.e. each temperature and stress
level) the test was repeated twice. Therefore each line in Fig. 7
was drawn based on the results of 8 tests (i.e. 4 tests at each tem-
perature). In this figure the slope values represent the temperature
sensitivity of compared mixes. As seen higher slope values
obtained for unmodified mixes and the mixes containing lower
PET contents (e.g. 440 MPa/°C for unmodified and 445 MPa/°C for
specimen containing 4% PET). By increasing the PET content up
to 6% the slope reduced to 404 MPa/°C. After that increasing the
amount of PET resulted in higher slope. At 10% the slope value
was equal to 445 MPa/°C which was even higher than the unmod-
ified one. Results indicated that for controlling the temperature
susceptibility of asphalt mix if PET added in dry method there will
Fig. 6. Results of resilient modulus test for various PET contents at 5 and 20 °C.
Fig. 7. Relationship between Mr and temperature.
be an optimum content which in this research was equal to 6%. It is
recommended to consider the temperature susceptibility as a
design criterion during the production process of PET modified
mixes especially in projects in which the modifier is added to
mix via dry method.
4.2. Fatigue
Fig. 8 shows the horizontal deformation curve of modified spec-
imen containing 10% PET which tested at 20 °C. The initial strain
level in this test was equal to 535 microstrain. Based on the frac-
ture life definitions that presented in Section 3.2, the fracture life
for the first and second definitions will be N1 = 65,664 and
N2 = 58,500 cycles, respectively.
Results of fatigue tests that performed at 5 and 20 °C have been
shown in Fig. 9. It must be noted that in this figure the initial stress
level at 5 and 20 °C were equal to 20% and 15% of ITS, respectively.
Furthermore, this figure shows the results of fatigue test based on
both fracture life definitions. Results indicated the beneficial
effects of PET on the fatigue behavior of studied mixes. Therewith,
at aforementioned stress levels, fatigue life increased upon reduc-
ing the test temperature. Unlike the results of ITS and Mr tests,
adding the PET content even as high as 10% had profitable effects
on the fatigue response of modified mixes.
 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15
Fig. 8. Example of displacement curve obtained during the indirect tensile fatigue
test.
Fig. 10 compares the fatigue curves of three specimens contain-
ing 0%, 8% and 10% PET which tested at 20 °C. With regard to this
figure, for modified specimens the slope of deformation curve in
the second zone (i.e. the zone with the constant rate of deforma-
tion increment) was less than unmodified one. Therefore, it could
be concluded that PET modified mixes exhibited higher cracking
resistance and flexibility than unmodified mixes.
Finding a meaningful relationship between the mix stiffness
and fatigue life has been a challenge for pavement scientists. Much
research proved that this relationship is completely dependent on
the method of fatigue testing. In controlled stress method, usually
stiffer mixes revealed higher fatigue life whereas, in strain constant
method the reverse was true [4,31–33]. Based on the findings of
this research increasing the PET content resulted in lower stiffness
and higher fatigue life. It might be due to higher energy absorbency
of the PET particles than the bitumen phase which resulted in
superior behavior against repeated loadings [7]. This phenomenon
will postpone the crack propagation throughout the specimen
diameter.
Fig. 9. Results of fatigue tests at 20 °C (at 20% of ITS) and 5 °C (at 15% of ITS).
13
Fig. 10. Comparison between the fatigue curves at 20 °C .
Fig. 11 shows the fatigue curves of both modified and unmodi-
fied specimens which attained at 5 and 20 °C. It should be noted
that in all presented conditions there was a proper correlation
between the initial strain and fatigue life (R2 values were more
than 0.8).
Fig. 11A compares the fatigue curves of modified and unmodi-
fied specimens based on the first fatigue life definition (N1). Simi-
larly, Fig. 11B shows the fatigue laws of studied mixes based on
the second definition (N2). According to this figure, at a constant
strain level on average the fatigue life of PET modified mix was
about 20% higher than unmodified one.
Fig. 11 signifies the considerable effect of temperature on the
fatigue response of studied mixes. As seen in this figure, the slope
of fatigue curves noticeably reduced by decreasing the test temper-
ature. As a result, the fatigue curves intersected each other at initial
strain levels of 160–210 microstrain. It means that at higher strain
levels of 210 microstrain the fatigue life of studied mixes reduced
upon reducing the temperature. In contrast at lower strain levels
of 160 microstrain specimens that tested at 5 °C revealed superior
fatigue response than those tested at 20 °C. Between the initial
strain levels of 160–210 microstrain there was an interference
zone in which no meaningful relationship could be found between
the testing temperature and fatigue life. Therefore it could be
14 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15

Fig. 12. Comparison between the results of Mr test for PET and SBS modified
specimens.

Fig. 11. Fatigue laws of studied mixes at various temperatures.

concluded that at higher strain levels the softer mix which tested
at 20 °C exhibited superior fatigue response than the stiffer mix
which tested at 5 °C. Hence, under heavy loading conditions fatigue
failure becomes critical at lower temperatures. In contrast at lower Fig. 13. Comparison between the fatigue curves of PET and SBS modified mixes at
20 °C.
strain levels of 160 microstrain fatigue failure is expected to occur
sooner under moderate climatic conditions.
However, apart from the initial strain level the addition of PET
to studied mixes led to an increase in fatigue life at both testing
temperatures.

4.3. Comparison of PET and SBS

Fig. 12 compares the results of resilient modulus (Mr) test for

PET and SBS modified mixtures. The stress levels at 5 and 20 °C
were equal to 20% and 40% of ITS, respectively. As it can be seen
at 20 °C both additives had similar effects on the stiffness of stud-
ied mixes. At this temperature, on average the Mr of unmodified
mixes was about 4700 MPa. Incorporating 5% SBS increased the
Mr value for about 9%. However, as previously mentioned at higher
contents of 2%, PET reduced the stiffness of studied mixes. For
specimens which tested at 5 °C, SBS had higher efficiency than
PET on reducing the stiffness of unmodified mixes. At this temper-
ature the Mr of unmodified mix was about 12,700 MPa which
reduced to about 8850 MPa with adding 4% SBS. At low tempera-
tures the asphalt mix tends to behave like brittle material. There-
fore at these temperature ranges the lower stiffness is desired
due to higher deformability and higher resistance against the det-
rimental effects of repeated loadings. Therefore, PET modified
mixes showed intermediate behavior in comparison to other
mixes. SBS is a thermoplastic elastomer which usually could
improve both low and high temperature characteristics of modi-
fied mixes [34]. Although some authors claim that a decrease in
strength and resistance to penetration is observed at higher tem-
peratures, but most of previous documents confirmed the proper
effects of this additive at various testing conditions [34].
Fig. 14. Comparison between the fatigue curves of PET and SBS modified mixes at
5 °C.
The SBS content selected for preparing the fatigue test speci-
mens was 5% which corresponded to the SBS content which
resulted in maximum resilient modulus. Fig. 13 compares between
the obtained fatigue curves of PET and SBS modified mixes at 20 °C.
As it can be seen both additives improved the fatigue response of
studied mixes. However, SBS modified mixes showed to some
extent better fatigue behavior than PET modified ones.
As shown in Fig. 14 overall at 5 °C modified mixes revealed pre-
dominant fatigue behavior than unmodified asphalt. However, at
higher strain levels (i.e. more than 300 microstrain), higher differ-
ences were found between the fatigue life of SBS and PET modified
mixes. At lower strain levels the differences between the fatigue
 A. Modarres, H. Hamedi / Materials and Design 61 (2014) 8–15
curves of modified mixes containing 6% PET and the SBS modified
ones reduced. It could be inferred that at lower strain levels of
200 microstrain modified mixes with 6% PET exhibited longer fati-
gue lives than SBS modified mix.
Finally, it could be concluded that PET had comparable effects
to SBS on the stiffness and fatigue behavior of studied asphalt
mixes. Since PET is a recycled material and is cheaper than original
polymer modifiers like SBS, the use of it in asphalt mix is desired in
both economical and environmental points of view.
5. Conclusions
This paper presents the results of a laboratory study about the
effects of waste plastic bottles (PET) on the stiffness and fatigue
properties of asphalt mixes. Likewise, similar tests were performed
on SBS modified mixes and the acquired results were compared.
Based on the obtained results the following conclusions can be
drawn.
(1) Addition of more than 2% PET reduced the resilient modulus
at both testing temperatures of 5 and 20 °C. However, at all
PET contents the resilient modulus quantities were in
acceptable limit.
(2) PET improved the fatigue behavior at both testing tempera-
tures. Unlike the results of ITS and Mr tests the addition of
the PET content up to 10% had beneficial effects on the fati-
gue response of modified mixes.
(3) There were some intersection points between obtained fati-
gue curves at 5 and 20 °C. These points were between the
strain levels of 160–210 microstrain. At higher strain levels
of 210 microstrain, adding temperature resulted in higher
fatigue life. However at lower strain levels of 160 micro-
strain, stiffer mixes which tested at 5 °C showed better fati-
gue response than those tested at 20 °C.
(4) At 20 °C PET and SBS had similar effects on the stiffness of
studied mixes. However, at 5 °C, SBS reduced the mix stiff-
ness. The later outcome is ideal because it improves the
material flexibility at low temperatures.
(5) Both additives improved the fatigue response of studied
mixes. Anyway, SBS modified mixes showed to some extent
better fatigue behavior than PET modified mixes especially
at higher strain levels of 200 microstrain.
References
[1] Aravind K, Das A. Pavement design with central plant hot mix recycled asphalt
mixes. Constr Build Mater 2007;21:928–36.
[2] Chen M, Lin J, Wu S, Liu C. Utilization of recycled brick powder as alternative
filler in asphaltic mixture. Constr Build Mater 2011;25:1532–6.
[3] Rubio MC, Moreno F, Belmonte A, Menéndez A. Reuse of waste material from
decorative quartz solid surfacing in the manufacture of hot bituminous mixes.
Constr Build Mater 2009;24:610–8.
[4] Moghadas Nejad F, Aflaki E, Mohammadi MA. Fatigue behavior of SMA and
HMA mixtures. Constr Build Mater 2010;24:1158–65.
[5] Abo-Qudais S, Shatnawi I. Prediction of bituminous mixture fatigue life based
on accumulated strain. Constr Build Mater 2007;21:1370–6.
15
[6] NCHRP APPENDIX II-1. Calibration of fatigue cracking models for flexible
pavements, USA; 2004.
[7] Ye Q, Wu S, Li N. Investigation of the dynamic and fatigue properties of fiber-
modified asphalt mixtures. Int J Fatigue 2009;31:1598–602.
[8] Tapkın S. The effect of polypropylene fibers on asphalt performance. Build
Environ 2008;43:1065–71.
[9] Jahromi S, Khodaii A. Carbon fiber reinforced asphalt concrete. Arab J Sci Eng
2008;33:355–64.
[10] Attaelmanan M, Pei Feng Ch, Al-Hadidy AI. Laboratory evaluation of HMA with
high density polyethylene as a modifier. Constr Build Mater
2011;25(5):2764–70.
[11] Ameri M, Mansourian A, Sheikhmotevali AH. Laboratory evaluation of ethylene
vinyl acetate modified bitumens and mixtures based upon performance
related parameters. Constr Build Mater 2012;40:438–47.
[12] VuralKök B, Çolak H. Laboratory comparison of the crumb-rubber and SBS
modified bitumen and hot mix asphalt. Constr Build Mater 2011;25:3204–12.
[13] Al-Hadidy AI, Yi-qiu T. Effect of polyethylene on life of flexible pavements.
Constr Build Mater 2009;23(1):1456–64.
[14] Chui-Te Chiu, Li-Cheng Lu. A laboratory study on stone matrix asphalt using
ground tire rubber. Constr Build Mater 2007;21:1027–33.
[15] Ahmed LA. Improvement of Marshall properties of the asphalt concrete
mixtures using the polyethylene as additive. Eng Technol 2007;25(3):383–94.
[16] Huang Y, Bird RN, Heidrich O. A review of the use of recycled solid waste
materials in asphalt pavements. Resour Conserv Recycle 2007;52:58–73.
[17] Arabani M, Mirabdolazimi SM, Sasani AR. The effect of waste tire thread mesh
on the dynamic behaviour of asphalt mixtures. Constr Build Mater
2010;24:1060–8.
[18] Arabani M, Mirabdolazimi SM, Ferdowsi B. Modeling the fatigue behaviors of
glasphalt. Sci Iranica A 2012;19(3):341–5.
[19] Li JM, Liu C, Qiao HC, Zhu LY, Chen G, Dai XD. Hot embossing/bonding of a
poly(ethylene terephthalate) (PET) microfluidic chip. J Micromech Microeng
2008;18-015008:10.
[20] Wan Mohd Nazmi, Wan Abdul Rahman, AchmadFauzi Abdul Wahab. Green
pavement using recycled polyethylene terephthalate (PET) as partial fine
aggregate replacement in modified asphalt. Proc Eng 2013;53:124–8.
[21] Baghaee Moghaddam T, Soltani M, RehanKarim M. Evaluation of permanent
deformation characteristics of unmodified and poly-ethylene terephthalate
modified asphalt mixtures using dynamic creep test. Mater Des
2013;53:317–24.
[22] Mahrez A, Karim MR. Rheological evaluation of bituminous binder modified
with waste plastic material. In: 5th International Symposium on Hydrocarbons
& Chemistry (ISHC5), Sidi Fredj, Algiers, May 23–25; 2010.
[23] Ahmadinia E, Zargar M, RehanKarim M, Abdelaziz M. Performance evaluation
of utilization of waste polyethylene terephthalate (PET) in stone mastic
asphalt. Constr Build Mater 2012;36:984–9.
[24] Baghaee Moghaddam T, RehanKarim M, Syammaun T. Dynamic properties of
stone mastic asphalt mixtures containing waste plastic bottles. Constr Build
Mater 2012;34:236–42.
[25] Ahmadinia E, Zargar M, Karim MR, Mahrez A, Shafigh P. Using waste plastic
bottles as additive for stone mastic asphalt. Mater Des 2011;32:4844–9.
[26] Casey D, McNally C, Gibney A, Gilchrist MD. Development of a recycled
polymer modified binder for use in stone mastic asphalt. Resour Conserv
Recycle 2008;52:1167–74.
[27] EN 12697–24. Bituminous mixtures-test methods for hot mix asphalt – Part
24: resistance to fatigue. European committee for standardization; July 2004.
[28] Witczak, MW., Kaloush, KT. pellinen, K. El-basyouny, M. Simple performance
test for superpave mix design. NCHRP 465; 2002.
[29] Thiago F, Kim YR, Lee J. Research on fatigue of asphalt mixtures and pavements
in Nebraska. Report No. P579, University of Nebraska Lincoln; February 25;
2008.
[30] Sengoz B, Isikyakar G. Evaluation of the properties and microstructure of SBS
and EVA polymer modified bitumen. Constr Build Mater 2008;22:1897–905.
[31] Modarres A. Investigating the toughness and fatigue behavior of conventional
and SBS modified asphalt mixes. Constr Build Mater 2013;47:218–22.
[32] Mahrez A, Karim MR. Fatigue characteristics of stone mastic asphalt mix
reinforced with fiber glass. Int J Phys Sci 2010;5(12):1840–7.
[33] Tayebali A, Rowe AGM, Sousa JB. Fatigue response of asphalt
aggregatemixtures. J Assoc Asphalt Paving Technol 2010;61:333–60.
[34] Yildirim Y. Polymer modified asphalt binders. Constr Build mater
2007;21:66–72.