International Journal of Pavement Engineering

ISSN: 1029-8436 (Print) 1477-268X (Online) Journal homepage: [Link]/journals/gpav20

A comprehensive evaluation of mixture and

binder properties to explore the use of low-
density polyethylene (LDPE) as an asphalt
modifier and co-modifier

Angelo Filonzi, Satyavati Komaragiri, K. Lakshmi Roja, Anand Sreeram, Amit

Bhasin & Eyad Masad

To cite this article: Angelo Filonzi, Satyavati Komaragiri, K. Lakshmi Roja, Anand Sreeram,
Amit Bhasin & Eyad Masad (2023) A comprehensive evaluation of mixture and binder
properties to explore the use of low-density polyethylene (LDPE) as an asphalt modifier
and co-modifier, International Journal of Pavement Engineering, 24:2, 2120988, DOI:
10.1080/10298436.2022.2120988

To link to this article: [Link]

Published online: 20 Sep 2022.

Submit your article to this journal

Article views: 444

View related articles

View Crossmark data

Citing articles: 12 View citing articles

Full Terms & Conditions of access and use can be found at

[Link]
INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING
2023, VOL. 24, NO. 2, 2120988
[Link]

A comprehensive evaluation of mixture and binder properties to explore the use of

low-density polyethylene (LDPE) as an asphalt modifier and co-modifier*
Angelo Filonzia, Satyavati Komaragiri a
, K. Lakshmi Rojab, Anand Sreerama, Amit Bhasina and Eyad Masadb
a
Department of Civil, Architectural and Environmental Engineering, The University of Texas at Austin, Austin, TX, USA; bDepartment of Mechanical
Engineering, Texas A&M University at Qatar, Doha, Qatar

ABSTRACT ARTICLE HISTORY

There has been growing interest in the use of polyethylene (PE) in the asphalt industry as one way to Received 15 October 2021
mitigate the adverse environmental impact of waste plastics. The main goal of this study was to Accepted 29 August 2022
evaluate the potential use of low-density polyethylene (LDPE) as a modifier for use in asphalt
KEYWORDS
mixtures. In addition, this study also explored the potential synergistic impact of using elastomeric Waste plastics; low-density
polymers typically used with asphalt binders as a co-modifier along with LDPE. This study was based polyethylene; environmental
on a comprehensive evaluation of the rheological and performance-related properties of asphalt impact; rheological
binder from two very different sources as well as asphalt mixtures produced using a subset of these properties; performance
binders. Typical methods that are standard practice in the industry and methods that are being properties
developed to evaluate the performance metrics of asphalt binders and mixtures were used. The two
chemically diverse base binders and redundancy in the evaluation process were introduced to ensure
a more robust evaluation of the expected properties. Overall, results demonstrate that although the
impact of LDPE on the expected performance of the binder may vary with the source of binder, LDPE
can be successfully used to enhance the properties of the asphalt binder in most respects, particularly
with the synergistic use of a conventional elastomeric modifier.

1. Introduction and background

Zhang 2013; Ma 2016; Liang 2019) have investigated the use
High-density polyethylene (HDPE) and low-density polyethy- of chemical, organic or clay modifiers to improve the compat-
lene (LDPE) represent two of the seven broad classes of plastics ibility between bitumen and polymers. For example, Liang
identified by the Society of Plastics Industry (SPI). The et al. (1993; 2000) used a steric stabilisation technique using
increased use of PE combined with its very short life cycle a styrene–butadiene–styrene (SBS) co-polymer and an ethyl-
and inability to naturally degrade over very long time has led ene–vinyl acetate co-polymer to produce an asphalt mixture
to massive waste accumulation and environmental concerns. known as Polyphalt. Another technology called Ditecpesa
There has been growing interest in the use of PE in the asphalt was introduced that added silica gel and fine carbon black to
industry as one way to mitigate this concern. Over the last few stabilise the low-density PE (LDPE)-modified binder (Gil
decades, several studies to evaluate the use of PE modified Rodondo and Caldeiro Jimenez 2012). There are also commer-
asphalt mixtures have been undertaken. Some of the findings cially available asphalt binders modified with recycled PE
from previous studies that are relevant to this study are sum- (Vargas 2013; White and Reid 2018; White 2019), but the
marised here, and a complete summary of these studies can be specific technologies for these products are proprietary.
found elsewhere (Masad 2020). One approach to enhance dispersion and stability of PE
One of the early works related to PE-modified asphalt was when blended with asphalt binders is to use elastomers such
referred to as Novophalt (Denning and Carswell 1983). Several as Styrene–Butadiene–Styrene (SBS) or reactive ethylene ter-
test sections were built with Novophalt in Canada and the Uni- polymer (RET). Such a combination can also provide synergis-
ted States (Jew 1986; Little 1991; Button and Phillips 1993; tic benefits between the PE and the elastomer. Previous studies
Daly et al. 1993; Maupin Jr 1993). Research has also evaluated have shown that grafting glycidyl methacrylate or GMA (one
the micro-scale interaction between PE and the binder, and of the polymers that is co-polymerised with other polymers
reported phase separation and storage stability issues with for use with asphalt binders) to PE can improve PE polarity
the PE-modified binder (Denning and Carswell 1983; Ho by forming chemical networks through ester linkage. The
2006; Ahmedzade 2016; Liang 2019; Kakar 2021; Tušar epoxy group of GMA-g-PE reacts with the functional group
2022). Subsequently, proprietary techniques to produce a (such as carboxylic acid, acid anhydride) in asphaltene of the
stable blend of PE and asphalt binder were developed by bitumen to form an intermolecular crosslink. This network
other researchers. As a primary concern is the phase separation formation can prevent or reduce phase separation as this addi-
of polymer with asphalt binders, several researchers (Yousefi tive chemically reacts with the binder rather than interacting
2003; Ho 2006; Ouyang 2006; Fang 2012, 2013a 2013b; physically (Selvavathi 2002; Polacco 2004; Pérez-Lepe 2006;

CONTACT Satyavati Komaragiri [Link]@[Link]

*Preprint submitted to International Journal of Pavement Engineering
© 2022 Informa UK Limited, trading as Taylor & Francis Group
2 A. FILONZI ET AL.

Jun et al. 2008). In other studies, researchers used SBS and consumer recycled plastic. However, such studies would
crumb rubber as additives to improve the properties of PE- have to focus on the variability in the waste recycling stream
blended binders. For instance, studies demonstrated that the and sensitivity of the binder and mixture performance to
addition of SBS and natural rubber helps in meeting the spe- this variability, which was beyond the scope of the current
cification of 50% elastic recovery at 15◦ C. In addition, it was study.
noticed (Fang 2008) that the addition of rubber powder to Two different asphalt binders from very different crude
the LDPE blend improved low-temperature performance. sources were used in this study. Although both binders had
However, a drawback of adding crumb rubber is that the a PG grade of 64–22, one of the binders was obtained from a
LDPE blend may exhibit compatibility issues due to the rub- local supplier in Qatar and the other was obtained from a
ber’s high molecular weight (Fang 2008; Singh 2013; Yu 2014). local supplier in Texas, USA.
Several studies in the past have examined the impact of PE Two conventional elastomeric modifiers were used to
modification on the binder and mixture properties (Yousefi evaluate the influence of using LDPE as a co-modifier on the
2003; Ho 2006; Punith and Veeraragavan 2007; Jun et al. performance-related properties of asphalt binders. One of
2008; Fuentes-Audén 2008; Casey 2008; Habib 2011; Singh these modifiers was a reactive ter-polymer and the other was
2013; Zhang 2013; Vargas 2013; Fang 2013a; Nuñez et al. an SBS elastomer. It should be noted here that the additives
2014; Dalhat and Al-Abdul Wahhab 2017). A summary of as well as the quantity of additives were same in the case of
findings from these studies can be found elsewhere (Masad Qatar binder blends as well as Texas binder blends. The follow-
2020). Some of these studies have focused on the performance ing are the combinations of materials along with a short nota-
characteristics of asphalt mixtures (Punith and Veeraragavan tion that were used directly or blended in the laboratory and
2007; Dalhat and Al-Abdul Wahhab 2017). However, many used for further evaluation:
of these studies focused either on the properties of the asphalt
binder only or basic stiffness related characteristics of the
asphalt mixtures. (1) Control binder used in Qatar (Q Control)
(2) Q Control + 3% LDPE by weight of the binder (Q +
LDPE)
2. Scope
(3) Q Control + 3% LDPE by weight of the binder + ter-poly-
The main goal of this study was to evaluate the potential use of mer (Q + LDPE + E1)
LDPE as a modifier or co-modifier for use in asphalt mixtures. (4) Q Control + 3% LDPE by weight of the binder + SBS elas-
In the context of this study, the term co-modifier refers to the tomer (Q + LDPE + E2)
use of the LDPE as a modifier along with a traditional elasto- (5) Control binder used in Texas (T Control)
meric polymer as a secondary modifier. The intent was to (6) T Control + 3% LDPE by weight of the binder (T + LDPE)
explore whether different modifiers could be combined to (7) T Control + 3% LDPE by weight of the binder + ter-poly-
achieve improvements in the performance-related properties mer (T + LDPE + E1)
of the resulting asphalt binders and mixtures. This study was (8) T Control + 3% LDPE by weight of the binder + SBS elas-
based on a comprehensive evaluation of the rheological and tomer (T + LDPE + E2).
performance-related properties of asphalt binder and asphalt
mixtures produced using a subset of these binders. This evalu-
ation was based on typical methods that are standard practice For the blended binders, a high shear mixer was used at
in the industry as well as methods that have been recently 2000 rpm for 2 h at 180◦ C based on the manufacturer’s rec-
developed to evaluate the performance metrics of asphalt bin- ommendations. In the case of SBS modifier, 1.5% of SBS and
ders and mixtures. This redundancy in the evaluation process 0.225% of sulfur were added to the binder (by weight of the
was introduced by design to ensure a more robust evaluation binder). Sulfur was added after initial blending of the SBS
of the expected properties. Two base binders were also used with the binder. In the case of the ter-polymer, 1.25% of the
to examine the differences in outcomes due to variation in polymer with 0.25% of polyphosphoric acid (PPA) was
the source of the binder. In summary, this study examines added to the binder (by weight of the binder). It should be
the synergy between LDPE and traditional elastomeric poly- noted here that the quantity of additives used in this study is
mers in affecting the performance of two chemically different based on manufacturer’s recommendations.
asphalt binders through conducting characterisation tests both A subset of binders was also used to fabricate asphalt mix-
at the binder and mixture length scales. tures to evaluate mixture properties. These mixtures were
designed using a non-absorptive aggregate with a maximum
nominal aggregate size of 12.5 mm following a dense grada-
3. Materials
tion. All mixtures were produced using a binder content of
The plastic used in this study was a low-density polyethylene 5.3%. Authors recognise that the modification of the asphalt
(LDPE) obtained from a production facility in Qatar. The binders may result in a grade change and consequently a slight
use of a controlled sample of LDPE from the producer in change in the optimum binder content. However, to avoid
lieu of a waste recycling stream was to ensure that the LDPE differences in binder content as a confounding variable, the
was homogeneous and consistent throughout the study. same binder content of 5.3% was used for all mixtures. All
Authors recognise that eventually it is important to expand asphalt mixture samples were fabricated using the Superpave
the findings from this and other similar studies to post- Gyratory Compactor (SGC).
 INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING 3

4. Test methods and parameters of interest ductility based on a derivation of a mechanical analog to rep-
resent the ductility test consisting of springs and dashpots.
In the case of asphalt binders, it was of interest to evaluate the
They also demonstrated that this parameter is directly corre-
impact of the aforementioned modifications on the stiffness,
lated to measured ductility for unmodified binders. Rowe
strength and ductility of the asphalt binder. In addition, the
and Sharrock (2011) re-defined the Glover parameter in
stability and microstructure, i.e. dispersion of the modifiers
terms of G∗ and δ based on analysis of a black space diagram
in the blend was also of interest, as these parameters would
and suggested use of the parameter described in Equation (1).
ultimately impact the performance of the asphalt binders
and mixtures. It should also be noted here all the tests were G′ |G∗ |( cos d)2 )
run using two replicates. The following methods and metrics = (1)
h′ /G′ sin d
were used to evaluate these properties of the asphalt binders.
In this study, the G-R parameter from Equation (1) was com-
puted based on the construction of a master curve from fre-
4.1. Binder stiffness
quency sweep testing at 5◦ C, 15◦ C and 25◦ C in the DSR
The stiffness of the asphalt binder was assessed using a num- using 8 mm plate and interpolating to find the value of the par-
ber of different rheological parameters used with the Per- ameter at 15◦ C and 0.005 rad/s (Glover 2005). A higher value
formance Grading (PG) of asphalt binders. In addition, the of this parameter indicates increased brittleness and suscepti-
Multiple Stress Creep and Recovery (MSCR) test was also bility to cracking.
used to evaluate the high temperature grade of the asphalt It is also important to emphasise that the correlation pro-
binder based on the non-recoverable compliance (Jnr) cut- posed by Glover (2005) was not universal when including
off value of 4.5 kPa−1 . The MSCR test also provides the elas- both modified and unmodified binders. Hajj et al. (2017)
tic recovery of the asphalt binder, which was also used as a demonstrated the use of the poker-chip test to identify differ-
measure of the binder’s rheological properties. All of these ences in the tensile strength of asphalt binders with the same
measurements were carried out using relevant AASHTO performance grade and as a direct measure of ductility. Diak
standards. et al. (2020) used a similar approach with a DSR (albeit a
thicker binder film) to correlate the results from a butt-joint
test to the fundamental parameters from a double etched
4.2. Binder strength and ductility
notch test that had previously been shown to strongly correlate
One of the shortcomings of the current PG framework is the with field performance. Filonzi (2020) demonstrated that the
inability to accurately reflect the cracking resistance of asphalt results from the poker-chip test correlated much better than
binders at intermediate temperatures. This was particularly other binder metrics with mixture cracking performance
important in this study to evaluate the impact of LDPE on measured in the laboratory using a variety of different asphalt
the strength and ductility of the asphalt binder. To assess the mixtures. Therefore, as an additional means of assessment, the
strength and ductility of the binders, two different methods poker-chip test was used to evaluate the tensile strength and
and metrics were used. ductility of control and modified binders. The test was con-
First, the Glover–Rowe parameter, that is considered as a ducted by testing 1.59 mm thin film of asphalt binder confined
surrogate for ductility, was measured. Glover (2005) proposed between two metal plates of 50 mm diameter in tension at
the rheological parameter, G′ /(h′ /G′ ), as an indicator of loading rate of 2 N/s (Figure 1).

Figure 1. Images showing the main steps in sample preparation (top row) and testing (bottom row) using the poker-chip method.
4 A. FILONZI ET AL.

4.3. Binder stability SGC compacted cylindrical asphalt mixture specimen (150
mm in diameter and 64 mm thick) along its diametric axis at
The high temperature stability of the asphalt binder was eval-
a displacement rate of 50.8 mm/minute. The air void content
uated using the ASTM D7173 (2011) test method. In sum-
of each specimen was 7.0 %. The stress and displacement of
mary, this test involves placing the asphalt binder blend in
the sample were recorded during the process. This approach
an aluminium tube 25 mm in diameter and 125 mm in height
is typically used to evaluate a cracking index, referred to as
at 163◦ C for a duration of 48 h in a vertical position. After this
the CT-Index calculated using Equation (2), where Gf is the
time, the tube is kept in a freezer for a minimum of 4 h and the
fracture energy, m75 is the slope of the load versus displace-
binder from the top and bottom third of the tube was evaluated
ment curve as the load drops from 85% to 65% of the peak
using a 25-mm plate in DSR for G∗ and compared to serve as
load, l75 is the post-peak displacement at 75% of the peak
an indicator of separation.
load and D is the diameter of the sample. In the context of
this study, it was important to examine the individual par-
4.4. Dispersion of modifiers in the binder ameters rather than consolidated index.
The extent of homogenisation of any modifier in the binder is Gf l75
CTindex = (2)
an important consideration. Heterogeneities can often intro- |m75 | D
duce stress localisation at interfaces and act as sources for
The IDT test has been critiqued for not being an accurate indi-
damage initiation. Heterogenous dispersions are also likely
cator on account of the high compressive stresses involved at
to separate over time. In this study, a fluorescence trinocular
the ends of the test specimen. Therefore, to evaluate the crack-
compound laboratory microscope manufactured by OmaxTM
ing resistance from an additional perspective, the asphalt mix-
was used to examine the dispersion of the modifiers in the
tures were also evaluated using the Semi-Circular Bending
asphalt binder. The test involved placing a very thin amount
(SCB) test following AASHTO TP124 (2018). SGC compacted
of heated binder on a glass slide and then covering it with
asphalt mixture specimens, 50 mm thick, were cut along their
another glass slide. The low thickness of the samples prepared
diameter into semi-circular test specimens. The target air voids
in this manner allowed for light to freely pass through and is
content of each disc was 7.0%. A notch 15±1 mm deep was cut
critical for proper characterisation using this method. The
perpendicular to the diametrical length of the specimen. The
samples were illuminated using a blue light for excitation (in
specimen was then subjected to a displacement-controlled
the range 390–490 nm) and the fluorescent light re-emitted
load until failure at a rate of 50 mm/ min. The test was con-
by the sample was observed under the microscope. A magnifi-
ducted at 25◦ C. Figure 2 shows the typical set up for the
cation of 100x was used to view the samples, which allowed for
SCB test.
the dispersion of the modifiers in the binder to be studied in
the micrometre (μm) range.
5. Results
4.5. Stiffness of asphalt mixtures 5.1. Binder stiffness
The stiffness of the asphalt mixtures produced using the control Figures 3 and 4 present the high continuous grade based on
and modified binders was compared based on their respective G∗ /sind parameter at unaged condition and RTFO aged con-
complex moduli master curves. The master curves were devel- dition and low continuous grade using stiffness and m-value
oped by testing at least two replicates of 150 mm diameter and parameters at PAV condition, respectively, for the eight differ-
170 mm height SGC compacted samples that were cored and ent asphalt binders. An increase in the high temperature grade
cut to produce 100 mm diameter and 150 mm tall finished indicates improvement in rutting resistance, while a reduction
specimens and had 6–7% air voids content. The compressive in the low temperature grade indicates improvement in
complex modulus of the test specimens was measured by con-
ducting a temperature-frequency sweep at −10◦ C , 4◦ C , 25◦ C ,
38.8 ◦ C and 54.4◦ C at 25, 10, 5, 1 and 0.1 Hz. This test was in
accordance with AASHTO TP62 (2003).

4.6. Rutting resistance of asphalt mixtures

The rutting resistance of asphalt mixtures was evaluated using a
Hamburg Wheel Tracking Device (HWTD) under wet con-
dition at 50◦ C in accordance with AASHTO T324 (2011).
Samples with an air void content of 7.0% were used for this test.

4.7. Cracking resistance of asphalt mixtures

The cracking resistance of asphalt mixtures was evaluated
using two different approaches. The first approach was to
use the indirect tensile strength test (IDT) by loading an Figure 2. Setup showing a typical SCB test.
 INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING 5

Figure 5. Continuous high grade based on RTFO aged Jnr values and using as cut
Figure 3. High temperature PG based on G∗ /sind parameter using DSR. off 4.5 kPa−1 .

Figure 6. Elastic recovery values at 64◦ C from MSCR test using the DSR.
Figure 4. Low temperature continuous grade of PAV aged samples using S and
m-values.

the high temperature PG, i.e. the use of LDPE substantially

increases the high temperature grade or rutting resistance of
resistance to thermal cracking. Two important observations
the asphalt binders. This increase is further augmented by
here are as follows. First, the addition of LDPE raised the
the use of an elastomeric polymer.
high temperature grade of the binder by almost one full
One additional and important observation from these
grade; this increase was further enhanced with the addition
results is that although LDPE helped improve the high temp-
of elastomers. Second, although the stiffness-based low-temp-
erature rutting resistance, it did not contribute to the elastic
erature grade of the binder did not change with the introduc-
recovery of binders. The addition of elastomer E1 substantially
tion of the LDPE or other elastomers, the m-value-based low-
enhanced the elastic recovery of the binder, whereas elastomer
temperature grade of the binder substantially deteriorated with
E2 did not demonstrate the same extent of benefit. It is specu-
the addition of LDPE. The addition of elastomers helped the
lated that this is likely because the concentration of E2 was just
LDPE-modified binder partially recover the m-value low-
below the threshold required for network formation and dem-
temperature grade while keeping the stiffness based low temp-
onstration of its relevant benefits. In fact, this observation is
erature grade almost the same.
substantiated by other results included subsequently in this
In addition to the PG parameters, the MSCR test at RTFO
report.
condition is also now being frequently used as a more accurate
indicator for rutting resistance of asphalt binders. Figures 5
and 6 show the high temperature grade based on the non-reco-
5.2. Binder strength and ductility
verable compliance or Jnr (using the cut-off for the standard
grade) and the elastic recovery of the binders, respectively. As described earlier, the strength and ductility of the binders
These results are consistent with the inferences drawn from were evaluated using two different approaches and metrics:
6 A. FILONZI ET AL.

Glover–Rowe parameter and the poker-chip test. Figure 7 pre- of the binder. An increase in tensile strength with little to no
sents the results for the Glover–Rowe parameter for the bin- loss in ductility is generally a good indicator because it reflects
ders at PAV condition. Recall that higher value of the better resistance to crack nucleation and growth. Second, the
parameter is indicative of increased susceptibility to cracking. addition of elastomer E1 helped increase the ductility of the
Based on this parameter alone, the use of LDPE appears to binder. The elastomer E2 did not show a similar enhancement,
increase the cracking susceptibility of the asphalt binders, but as discussed with the results from the MSCR elastic recov-
which appears to be mitigated to some extent by the use of ery, it is speculated that this could be because the percentage
the elastomers. The influence of elastomers varied depending used was below the threshold required to create and benefit
on the source of the binder and the type of elastomer. This par- from a ‘network’ effect formed due to the creation of intermo-
ameter is primarily a surrogate for the measurement of duct- lecular crosslinks from the chemical reaction of the functional
ility and therefore the poker-chip test was also used to groups in asphalt binder with the functional groups in grafted
evaluate the strength and ductility of the binder. LDPE to RET.
Figure 8 presents the tensile strength and failure strain as a Figure 9 presents typical results showing the stress–strain
measure of ductility from the poker-chip test at RTFO con- behaviour from the poker-chip test. The enhanced ductility
dition. In this case, the failure strain is defined as the post- and slight to no change in tensile strength due to the addition
peak strain that corresponds to 80% of the tensile strength of of E1 are more clearly visible from these figures.
the binder. The two main observations from these results are
as follows. First, the addition of LDPE resulted in an increase
5.3. Dispersion of modifiers in the binder
in the tensile strength of the binder, however, this was also
accompanied by a slight loss in the failure strain or ductility The ASTM D7173 separation test and Fluorescence
microscopy were used to evaluate the homogenisation and
tendency for the LDPE to separate from the binder matrix.
Based on the separation test, the binder from Qatar with
LDPE showed a variation of 23%. The use of E1 and E2
changed this variation to 30% and 9%, respectively. Simi-
larly, the binder from Texas showed a variation of 25%
with LDPE. The use of E1 and E2 changes this to 4% and
9%, respectively. Note that the acceptable range of two test
results (d2s%) for a single operator evaluating unaged binder
with the DSR is 4.6%. These results suggest that the efficacy
of E1 to prevent separation depends on the type of binder
whereas E2 presented similar results with both binders.
Figure 10 presents images from the fluorescence microscope
for the different binders. These images show that the LDPE
is relatively well dispersed with E2 as a modifier. The dis-
persion was also better with E1 modifier, particularly for
the binder from Texas. These results were consistent with
the results from the polymer separation test.
Figure 7. Glover–Rowe parameter of PAV aged samples.

5.4. Stiffness of asphalt mixtures

A subset of six binders was used to evaluate the performance
characteristics of asphalt mixtures. The complex moduli of
the mixtures were measured using frequency-temperature
sweeps to develop master curves for the selected materials.
The master curves are not shown for brevity, but Figure
11 compares the results for the LDPE and LDPE+E1
modified mixes to the control mix for the two different bin-
der sources. The key observation here is that the stiffness of
the mixes, particularly in the low frequency range, increases
compared to the control mixes due to the addition of LDPE.
Further addition of an elastomer reduced the E∗ compared
to the use of LDPE alone. This is somewhat expected
because the use of elastomers in some cases can increase
resilient strain and corresponding reduced E∗ of the asphalt
mixtures (Bhasin et al. 2004). Although E∗ is a basic measure
of stiffness, the performance characteristics of the modified
Figure 8. Tensile strength and failure strain (defined as the post-peak strain at mixtures is of greater interest and is discussed in the follow-
80% of peak stress) of the selected asphalt binders. ing sections.
 INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING 7

Figure 9. Typical stress–strain curves for the binders at RTFO condition from the poker-chip test: (a) base binder from Qatar and (b) base binder from Texas.

Figure 10. Fluorescence microscopy images showing dispersion of the LDPE with and without the elastomer as a co-modifier.

5.5. Rutting resistance of asphalt mixtures testing (high PG and Jnr), thus reinforcing the finding that
the use of LDPE improves rutting resistance of the asphalt
Figure 12 shows the rut depth at 20,000 passes from the
mixture. The use of co-modifiers (LDPE + Elastomer) further
HWTD on the selected subset of mixtures. These results are
improves the rutting resistance of the asphalt mixture.
very consistent with the results reported based on binder

Figure 11. Relative change in E ∗ due to addition of LDPE and elastomer. Figure 12. Results from the Hamburg Wheel Tracking Device.
8 A. FILONZI ET AL.

Figure 13. Displacement post-peak at 75% of peak load. Figure 15. Normalised fracture energy by the absolute value of post peak slope
from IDT.

5.6. Cracking resistance of asphalt mixtures load versus vertical displacement curve) divided by area of
The cracking resistance of asphalt mixtures was evaluated cracking face of the IDT sample.
using IDT and SCB. This test is typically used to compute a One important observation here is that the use of E1 co-
cracking index that incorporates the fracture energy and modifier did not improve the post-peak behaviour of the
post-peak deformation characteristics. Although a cracking asphalt mixes. This apparent contradiction with the results
index is useful for a routine material qualification or screening from the poker-chip test can be easily explained, because the
purposes, in this case it is more insightful to examine the IDT is conducted in a displacement-controlled mode. As a
different parameters from this index individually. Figure 13 result, any modifier that increases in stiffness and subjected
compares the displacement at 75% of peak load and Figure 14 to the same rate of displacement will result in increased rate
compares the slope of the post-peak failure curve. of stress accumulation during the test, which in turn induces
These two aforementioned parameters provide an indi- a less ductile mode of failure. This is an artefact of displace-
cation of the susceptibility of the mix to brittle failure. These ment controlled test and not necessarily the behaviour of the
results show that the use of LDPE to some extent reduces duct- material when subjected to similar rates of stress.
ility of the mix. In particular, the post-peak slope increases The second test that was conducted to evaluate the cracking
with the use of LDPE, however, the use of an elastomeric co- resistance of the asphalt mixture was SCB. The SCB is in some
modifier does not appear to benefit this behaviour except for ways similar to the IDT, i.e. both tests are conducted by applying
a limited extent with the Qatar binder. However, an increase a constant rate of displacement. Some of the artefacts from such a
in the tensile strength of the binder due to the incorporation mode of loading are expected to be present in the results from the
of LDPE and elastomer does result in an increase in the tensile SCB as well. However, SCB is regarded as a more representative
strength of the mixture and binder, which in turn compensates fracture test on account of the notch that propagates as a crack
for the slight loss in ductility. As a result, Figure 15 shows through the specimen and the absence of high compressive stres-
either no change or a slight increase in the overall fracture ses at the diametric ends of the mixture specimen. Therefore, it
energy calculated by the work of fracture (the area of the was worth using this additional approach to examine the cracking
characteristics of the unmodified and modified asphalt binders.

Figure 14. Post-peak slope of load vs. displacement. Figure 16. Post-peak slope from the SCB test.
 INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING 9

more recent methods at both the binder and the mixture

length scale. Potential synergistic influence of using LDPE
along with conventional elastomeric modifiers was also evalu-
ated. Overall, results demonstrate that although the impact of
LDPE on expected performance of the binder may vary with
source binder, LDPE can be successfully used to enhance the
properties of the asphalt binder in most respects, particularly
with the synergistic use of a conventional elastomeric modifier.
The following are some of the findings from the binder and
mixture testing.

(1) The use of LDPE helps increase resistance to plastic defor-

mation and hence rutting resistance. This observation was
consistent based on binder testing as well as mixture test-
Figure 17. Critical displacement from the SCB test. ing. The use of a co-modifier further improves the rutting
resistance of the binders and mixtures.
(2) The use of LDPE increases the tensile strength of the
As with IDT, it is more insightful to examine individual asphalt binder and mixtures. This finding was consist-
components of this load–deformation response. Figures 16 ent across both binder and mixture tests (SCB and
and 17 present the post-peak behaviour and serve as some IDT). The use of a co-modifier does not always con-
indication of ductility (notwithstanding the limitation of this tribute significantly in terms of improvement in tensile
being a displacement-controlled test). These parameters were strength beyond the gain from the use of LDPE by
computed based on the definitions described in AASHTO itself.
TP124 (2018). (3) The use of LDPE may result in a slight decrease in duct-
Qualitatively, the results are similar to the findings from the ility. The use of an elastomeric co-modifier can mitigate
IDT, although some differences are more amplified. Figure 18 this loss to some extent. The evaluation of asphalt bin-
illustrates fracture energy from the SCB. ders using the Glover–Rowe parameter and the poker-
In this case, the fracture energy decreases for the Qatar bin- chip test indicates that one of the elastomers substan-
der, including when an elastomeric co-modifier is added, tially improved the failure strain of the asphalt binder
whereas the fracture energy is almost constant for the Texas while the other did not have a significant impact. It is
binder. These results are slightly different from the IDT as speculated that this was likely because the concentration
the fracture energy from IDT was normalised by the absolute of the elastomer was below a critical threshold that
value of post-peak slope whereas in the case of SCB the frac- would result in the enhanced benefit and a slightly
ture energy was not normalised. higher concentration of this elastomer could change
the overall response. The results from mixture testing
using the IDT and SCB were not consistent with the
6. Conclusion
findings from binder testing. In particular, neither one
The main goal of this study was to evaluate the potential of of the two elastomers showed a positive impact on the
using LDPE as a modifier for use in asphalt mixtures using failure strain of the asphalt mixture. This apparent dis-
two very different sources of asphalt binders based on a com- crepancy between the binder and mixture tests is likely
prehensive evaluation of binder properties, dispersion and due to the displacement-controlled mode of loading
mixture properties. The evaluation included multiple and used with the mixture tests compared to the load-con-
trolled mode of loading used with the poker-chip test.
The use of elastomer as a co-modifier further increases
the stiffness of the binder potentially resulting in a less
than ductile mode of failure in a displacement-con-
trolled test. Although not directly related to the scope
of this study, this discrepancy must be examined closely
in future work.
(4) The addition of a co-modifier aids the dispersion and at
least in one case also the stability of the asphalt binder
modified using LDPE.
(5) The results obtained clearly show that the source and
composition of the binder have a significant impact on
the overall response of the composite to LDPE or a com-
bination of the LDPE with a co-modifier.

Overall, the results show that LDPE can be used to modify

Figure 18. Fracture energy from the SCB test. asphalt binders to produce mixtures with improved rutting
10 A. FILONZI ET AL.

resistance. Depending on the source and chemistry of the base Daly, W.H., Qui Chiu, Z., and Negulescu, I., 1993. Preparation and
binder, LDPE is likely to increase the tensile strength of the characterization of asphalt-modified polyethylene blends.
Transportation Research Record, No. 1391.
binder, albeit at some loss of ductility. It appears that it may Denning, J., and Carswell, J., 1983. Assessment of ‘novophalt’ as a binder
be possible to mitigate this loss with the use of an elastomeric for hot rolled asphalt wearing coarse. Rep. No. LR, 1101.
co-modifier. The use of an elastomeric co-modifier also Diak, E., Beier, E., and Hesp, S.A., 2020. Development of a butt joint test
benefits the low temperature properties, dispersion, and stab- for the ductile performance grading of asphalt binders. Construction
ility of the binder matrix and should certainly be considered as and Building Materials, 243, 118195.
Fang, C., et al., 2008. Combined modification of asphalt by waste pe and
a viable approach moving forward. Lastly, this work was done rubber. Polymer Composites, 29 (10), 1183–1187.
using a consistent sample of LDPE and not true waste post- Fang, C., et al., 2012. Combined modification of asphalt with polyethylene
consumer plastic to evaluate viability without introduction of packaging waste and organophilic montmorillonite. Polymer Testing,
variability; real scenario will have to deal with this variability 31 (2), 276–281.
in the future. Fang, C., et al., 2013a. Preparation and characterization of an asphalt-
modifying agent with waste packaging polyethylene and organic mon-
tmorillonite. Polymer Testing, 32 (5), 953–960.
Acknowledgments Fang, C., et al., 2013b. Preparation, characterization and hot storage stab-
ility of asphalt modified by waste polyethylene packaging. Journal of
All the statements are those of the authors. The authors also acknowledge Materials Science & Technology, 29 (5), 434–438.
the (i) co-funding of Qatar Petrochemical Company (QAPCO) for sup- Filonzi, A., et al., 2020. Investigating the use of a binder cohesion test to
plying the LDPE, and RAETEX; (ii) Kraton and Dow chemical for supply- evaluate cracking resistance of asphalt mixtures. In: RILEM inter-
ing the elastomers used in this study; and (iii) Mr. Tyler Seay for his help national symposium on bituminous materials. Lyon, France: Springer.
in laboratory evaluation of the materials. Fuentes-Audén, C., et al., 2008. Evaluation of thermal and mechanical
properties of recycled polyethylene modified bitumen. Polymer
Testing, 27 (8), 1005–1012.
Disclosure statement Gil Rodondo, S., and Caldeiro Jimenez, A., 2012. Betunes modificados con
polietileno reciclado, procedente del contenedor amarillo, estables al
No potential conflict of interest was reported by the author(s).
almacenamiento. UTech. Rep. Communication 21, Spain.
Glover, C.J., et al., 2005. Development of a new method for assessing
asphalt binder durability with field validation.
Funding Habib, N.Z., et al., 2011. Rheological properties of polyethylene and poly-
This work was supported by the Qatar National Research Fund (QNRF): propylene modified bitumen. International Journal Civil and
NPRP11S-1128-170041. Environmental Engineering, 3 (2), 96–100.
Hajj, R., Hure, R., and Bhasin, A., 2017. Evaluation of stiffness, strength,
and ductility of asphalt binders at an intermediate temperature.
ORCID Transportation Research Record: Journal of the Transportation
Research Board, 2632, 44–51.
Satyavati Komaragiri [Link] Ho, S., et al., 2006. Study of recycled polyethylene materials as
asphalt modifiers. Canadian Journal of Civil Engineering, 33 (8),
968–981.
References Jew, P., et al., 1986. Polyethlene-modified bitumen for paving appli-
AASHTO T324, T., 2011. Standard method of test for Hamburg wheel- cations. Journal of Applied Polymer Science, 31 (8), 2685–2704.
track testing of compacted hot mix asphalt (HMA). Washington, DC: Jun, L., Yuxia, Z., and Yuzhen, Z., 2008. The research of gma-g-ldpe
American Association of State Highway and Transportation Officials. modified Qinhuangdao bitumen. Construction and Building
AASHTO TP124, T., 2018. Standard method of test for determining the Materials, 22 (6), 1067–1073.
fracture potential of asphalt mixtures using the flexibility index test Kakar, M.R., et al., 2021. Analysis of waste polyethylene (pe) and its by-
(FIT). Washington, DC: American Association of State Highway and products in asphalt binder. Construction and Building Materials, 280,
Transportation Officials. 122492.
AASHTO TP62, T., 2003. Determining dynamic modulus of hot-mix Liang, Z., et al., 1993. Utilization of recycled polyethylene in the
asphalt concrete mixtures. Washington, DC: American Association of preparation of stabilised, high performance modified asphalt bin-
State Highway and Transportation Officials. ders. In: Use of waste materials in hot-mix asphalt. Baltimore:
Ahmedzade, P., et al., 2016. Use of maleic anhydride grafted recycled ASTM International.
polyethylene treated by irradiation in bitumen modification. In: Liang, M., et al., 2019. Phase behavior and hot storage characteristics of
Proceedings of 6th Eurasphalt & Eurobitume congress. Czech asphalt modified with various polyethylene: experimental and
Technical University in Prague. numerical characterizations. Construction and Building Materials,
ASTM:D7173, D., 2011. Standard practice for determining the separation 203, 608–620.
tendency of polymer from polymer modified asphalt. West Liang, Z., Woodhams, R.T., and Smith, J.W., 2000. Stabilized bitumen
Conshohocken, PA: ASTM International. compositions. U.S. Patent 6, 100, 317, Aug. 8, 2000.
Bhasin, A., Button, J.W., and Chowdhury, A., 2004. Evaluation of simple Little, D.N., 1991. Performance assessment of binder-rich polyethylene-
performance tests on hot-mix asphalt mixtures from south central modified asphalt concrete mixtures (novophalt). Transportation
United States. Transportation Research Record, 1891 (1), 174–181. Research Record, No. 1317.
Button, J.W., and Phillips, S.G., 1993. Effect of asphalt additives on pave- Ma, D., et al., 2016. Functionalization of reclaimed polyethylene with
ment performance. maleic anhydride and its application in improving the high tempera-
Casey, D., et al., 2008. Development of a recycled polymer modified bin- ture stability of asphalt mixtures. Construction and Building
der for use in stone mastic asphalt. Resources, Conservation and Materials, 113, 596–602.
Recycling, 52 (10), 1167–1174. Masad, E., et al., 2020. A review of asphalt modification using plastics: a
Dalhat, M., and Al-Abdul Wahhab, H., 2017. Performance of recycled focus on polyethylene. Qatar, Doha: Texas A&M University.
plastic waste modified asphalt binder in Saudi Arabia. International Maupin Jr, G., 1993. Evaluation of a modified asphalt: Novophalt. Final
Journal of Pavement Engineering, 18 (4), 349–357. Report.
 INTERNATIONAL JOURNAL OF PAVEMENT ENGINEERING 11

Nuñez, J.Y.M., Domingos, M.D.I., and Faxina, A.L., 2014. Susceptibility of Tušar, M., et al., 2022. Rilem tc 279 wmr round robin study on waste poly-
low-density polyethylene and polyphosphoric acid-modified asphalt ethylene modified bituminous binders: advantages and challenges.
binders to rutting and fatigue cracking. Construction and Building Road Materials and Pavement Design, 1–29.
Materials, 73, 509–514. Vargas, M.A., et al., 2013. Asphalt/polyethylene blends: rheological prop-
Ouyang, C., et al., 2006. Low-density polyethylene/silica compound erties, microstructure and viscosity modeling. Construction and
modified asphalts with high-temperature storage stability. Journal of Building Materials, 45, 243–250.
Applied Polymer Science, 101 (1), 472–479. White, G., 2019. Evaluating recycled waste plastic modification and exten-
Pérez-Lepe, A., et al., 2006. Destabilization mechanism of polyethylene- sion of bituminous binder for asphalt. In: Eighteenth annual inter-
modified bitumen. Journal of Applied Polymer Science, 100 (1), national conference on pavement engineering, asphalt technology and
260–267. infrastructure. Liverpool, UK.
Polacco, G., et al., 2004. Rheology of asphalts modified with glycidyl- White, G., and Reid, G., 2018. Recycled waste plastic for extending and
methacrylate functionalized polymers. Journal of Colloid and modifying asphalt binders. In: 8th symposium on pavement
Interface Science, 280 (2), 366–373. surface characteristics (SURF 2018). Brisbane, Queensland, Australia,
Punith, V., and Veeraragavan, A., 2007. Behavior of asphalt concrete mix- 2–4.
tures with reclaimed polyethylene as additive. Journal of Materials in Yousefi, A.A., 2003. Polyethylene dispersions in bitumen: the effects of the
Civil Engineering, 19 (6), 500–507. polymer structural parameters. Journal of Applied Polymer Science, 90
Rowe, G., and Sharrock, M., 2011. Alternate shift factor relationship for (12), 3183–3190.
describing temperature dependency of viscoelastic behavior of asphalt Yu, B., et al., 2014. Evaluation of plastic–rubber asphalt: engineering
materials. Transportation Research Record: Journal of the property and environmental concern. Construction and Building
Transportation Research Board, 2207, 125–135. Materials, 71, 416–424.
Selvavathi, V., et al., 2002. Modifications of bitumen by elastomer and Zhang, H., et al., 2013. Effect of linear low density-polyethylene
reactive polymer – a comparative study. Petroleum Science and grafted with maleic anhydride (lldpe-g-mah) on properties of
Technology, 20 (5-6), 535–547. high density-polyethylene/styrene–butadiene–styrene (hdpe/sbs)
Singh, B., et al., 2013. Polymer-modified bitumen of recycled ldpe and modified asphalt. Construction and Building Materials, 47, 192–
maleated bitumen. Journal of Applied Polymer Science, 127 (1), 67–78. 198.