Article

pubs.acs.org/IECR

Gelation-Stabilized Functional Composite-Modied Bitumen for

Anti-icing Purposes
Derya Aydn,, Riza Kizilel, Ramazan O. Caniaz,,, and Seda Kizilel*,,

Department of Chemical and Biological Engineering and Koc University-TUPRAS Energy Center (KUTEM), Koc University,
Sariyer, Istanbul, Turkey, 34450

TUPRAS Turkish Petroleum Reneries Co., Kocaeli, Turkey, 41790

S Supporting Information
*

ABSTRACT: Ionic salts as anti-icing agents have been extensively used to eliminate accumulation of ice on asphalt surfaces.
However, salt can be easily removed by rain or automobiles and requires frequent application on roads. Besides this economic
consideration, anti-icing agents compromise the mechanical properties of asphalt and have a negative impact on living organisms
and the environment when used in large amounts. Incorporation of hydrophilic salts into bitumen, a hydrophobic asphalt binder,
and controlled release of specic molecules from this hydrophobic medium can provide an eective solution for reducing ice
formation on pavements. Bitumen has previously been modied by various polymers, including styrene-butadiene-styrene (SBS)
for improved strength and thermomechanical properties. However, an anti-icing function was not considered in those previous
designs. In a previous study, we developed a functional polymer composite consisting of potassium formate (HCOOK) salt
pockets dissolved in a hydrophilic gel medium and dispersed in a hydrophobic SBS polymer matrix. Here, we developed an
innovative method to obtain polymer composite-modied bitumen and investigated further the anti-icing properties of the
functional bitumen. We improved incorporation of this polymer composite into bitumen and demonstrated proper distribution
of the composite within bitumen through morphological and rheological analysis. We characterized the anti-icing properties of
modied bitumen surfaces and demonstrated signicant increases in freezing delay of composite-modied bitumen compared to
base bitumen in a temperature- and humidity-controlled chamber. In addition, we characterized the release of HCOOK salt from
polymer composite-modied bitumen and observed salt release within the range of 1.0710.8% (w/w) in 67 days, depending on
the composite content. The results demonstrate the potential of this polymer composite-modied bitumen for anti-icing
functionality and for industrially relevant applications.

1. INTRODUCTION

Sodium chloride, calcium chloride, and sodium acetate have

been commonly used as anti-icing agents; however, these salts
have a negative impact on the stiness and viscoelastic
properties of asphalt pavements.1114 The inuence of
conventional use of potassium formate on asphalt concrete
was also investigated.1517 The deterioration eect of deicers
based on acetate17 and formate16 was observed in these studies
when asphalt concrete was exposed to deicer solutions.
However, polymer modication of bitumen was also suggested
in these studies to counterbalance the negative eects of deicers
in terms of mechanical properties. Various solutions have been
suggested as alternatives to the conventional usage of these
agents. One of those strategies involves the addition of levulinic
salts into the asphalt mixture to eliminate the need for the
common anti-icing agents and to obtain pavements with an
anti-icing property.18 In another study, researchers evaluated
the potential of synthetic llers such as sodium chloride,
calcium carbonate, and magnesium carbonate in an asphalt
mixture to promote an anti-icing property. A delay in frost
formation and a decrease in adhesion between ice and the
pavement were reported.19

Anti-icing functionality is desired in many outdoor systems in

order to delay ice formation and to prevent adhesion of ice on
surfaces, so that mechanical removal of ice from the surface is
easier. In industrial applications, such as for wind turbines and
on pavements for vehicle transportation during the winter, antiicing is critical. According to 10 year averages from 2002 to
2012, 43% of crashes that happen on wet pavement occur
during snow or on slippery pavements.1,2
Until recently, studies on anti-icing surfaces were mainly
geared toward achieving superhydrophobicity on surfaces.38
However, superhydrophobicity is not necessarily appealing on
all surfaces that require anti-icing function. Upon exposure to
water, superhydrophobic surfaces promote slip because they
reduce drag by trapping gas bubbles between the surface and
the liquid using a microscopic texture. Furthermore, surface
geometry modication at a nanolevel is dicult and in most
cases not economically feasible.35 Another approach to
eliminate ice formation on surfaces is the use of anti-icing
agents. Ionic salts are the most commonly employed agents for
this purpose. As ions dissolve in water, they localize between
water molecules and retard the crystallization process. As a
result of the delay in crystallization, water remains in the liquid
phase at lower temperatures than the freezing point of pure
water.9,10
2015 American Chemical Society

Received:
Revised:
Accepted:
Published:
12587

August 17, 2015

November 7, 2015
November 27, 2015
November 27, 2015
DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

Figure 1. Schematic of the preparation of composite polymer-modied bitumen. First, the functional emulsion is prepared. Next, the solvent is
evaporated and the composite membrane in dry form is obtained. Dry composite is then mixed with bitumen in the subsequent step, and nally
dried composite is uniformly distributed into bitumen structure at 170 C using low-shear mixer for 2.53 h.

Previous studies are important for the development of antiicing strategies; however, the inuence of these synthetic llers
on the thermomechanical properties of the binder in asphalt
mixtures still needs to be investigated.19 The modication of
bitumen with a styrene-butadiene-styrene (SBS) polymer for
improved thermomechanical properties has also been considered separately in other studies.2029 The hydrophobicity and
thermoplastic nature of the SBS polymer allows for
incorporation of this composite into bitumen, opening up the
possibility of delivering anti-icing agents through bitumen. SBS
has been widely used in polymer-modied bitumen (PMB) for
the construction of asphalt binders and constitutes the
continuous phase of the parent emulsion and the membrane
base of the composite developed here.30 Bitumen is a semisolid
hydrocarbon material, generally produced from certain crude
oils by distillation (atmospheric or vacuum) units in oil
reneries. The majority of bitumen has been used as a binder in
asphalt road pavements.31 Due to high trac density and
increases in loading and pressure, plus insucient maintenance,
asphalt pavements can deteriorate with time. Distress signs
often take the form of rutting and cracking. To minimize the
damage and increase the lifetime of asphalt roads, polymermodied bitumens, including those modied with SBS, have
been used.2023,26,32
To our knowledge, no study exists that considers the antiicing property along with thermomechanical properties for
polymer-modied bitumen. The formation of a functional
polymer-modied bitumen system is based on the ne
dispersion of a polymerionic salt composite in bitumen and
on the compatibility of the polymer bitumen system.19,20,23,25
In our previous study, we incorporated an ionic salt, potassium
formate (HCOOK) into SBS through a particle-stabilized
emulsion templating approach. Our design consists of a
composite membrane coating, which integrates a hydrophobic
polymer adhesive base and a dispersed hydrophilic phase. The
hydrophilic phase includes agarose (agar) hydrogel domains to
store the anti-icing agent HCOOK. Controlled release of
HCOOK from this composite was characterized, and a freezing
delay of about 70 min was observed on this functional
composite surface in a temperature- and humidity-controlled
chamber.33 In this study, we modied bitumen with this
functional polymer composite to obtain polymer compositemodied bitumen with anti-icing surface properties. The
compatibility of SBS in bitumen opens up the possibility of
incorporating this composite system into bitumen for anti-icing
functionality as a result of eective incorporation of functional
packages. Here we accomplish proper incorporation of the
polymer composite into bitumen and evaluate the morphology,
rheology, salt release, and contribution of this modication to
anti-icing properties of bitumen surfaces. The eect of the salt
content in the composite and the presence of agar on the anti-

icing properties of composite-modied bitumen are investigated for the rst time.

2. MATERIALS AND METHODS

2.1. Materials. Styrene-butadiene-styrene block copolymer
(SBS; S:B weight fraction, 70:30) (Kraton D1101), potassium
formate (HCOOK) salt (99%) (Sigma-Aldrich), silica nanoparticles (AEROSIL 816), and cyclohexane (99.9%) (Merck)
were used for the preparation of particle-stabilized emulsions.
Agar powder was purchased from Sigma-Aldrich. The 50/70
penetration grade bitumen was kindly supplied by TUPRAS.
Deionized water was used in the experiments (Purelab Option,
ELGA). Glass Petri dishes and glass microscope slides were
purchased from Nunc.
2.2. Preparation of Emulsions. Particle-stabilized emulsions with 0.25 (v/v) internal phase fraction ( = 0.25), 1.0%
(w/v) nanoparticle concentration with or without gelation in
the internal phase were prepared as described in our previous
study.33 Briey, 0.5 mL of HCOOK (0.5 g/mL) salt solution
with or without agar was added to 0.5 mL of 1.0% (w/v)
nanoparticle-cyclohexanone stock solution under high shear
applied by a vortex mixer at the maximum speed for 5 s. After
stabilization of the aqueous droplets by nanoparticles in the
stock solution, 1 mL of SBS polymer in cyclohexane (110 mg/
mL) was added dropwise to 1 mL of the HCOOK, agar, and
nanoparticle solution by mixing slowly.
2.3. Preparation of Functional Composite Membranes
and Incorporation into Bitumen. Prepared emulsions were
cast in Petri dishes to dry, and functional composite
membranes were obtained. The emulsions were dried in Petri
dishes for 24 h at room temperature. The SBS polymer and the
composites were chopped up and then incorporated into
bitumen using a laboratory scale mixing device. The bitumen
composite blends were prepared with a three-blade impeller
low-shear mixer (RW20, IKA Germany). In the literature,
polymer modication of bitumen was generally carried out at
180190 C. These experiments were also performed with
altered temperatures, such as 160163 C20,24,34 and 150170
C,22 both with low- and high-shear mixers. In this study, our
goal was to control the agent release, and we applied these
mixing conditions to obtain sucient dispersion of composite
throughout the bitumen. We mixed SBS to bitumen with the
same conditions to be able to compare with compositemodied bitumen. The composite was mixed into bitumen at a
temperature of 170 C and a rotation speed of 125 rpm for
2.53 h. We chose these mixing conditions to obtain sucient
dispersion of composite throughout the bitumen. After
processing, the samples were stored at room temperature
(Figure 1).
2.4. Surface Characterization of Composite-Modied
Bitumen. 2.4.1. Fluorescence Microscopy. Epiuorescence
12588

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

has been commonly used.20,22,2426,34 For this reason, 3%

weight fraction was used here in the beginning, and composite
concentration was increased to 5%. Therefore, ve dierent
samples were examined, as listed in Table 1.

microscopy was used to infer SBS morphology in a polymermodied bitumen. The morphologies of SBS in polymermodied bitumen and in composite-modied bitumen were
investigated under uorescence microscopy with 520 nm
emission and 450490 nm excitation wavelength range.
2.4.2. Atomic Force Microscopy. Atomic force microscopy
(AFM) experiments were performed in a Bruker Dimension
Icon AFM instrument with MPP21100-10 REFSP tips. Typical
scans covered areas of 20 20 m2 in bitumen samples.
Imaging was performed in soft tapping mode (intermittent
contact) at room temperature. Both topographic and phase
dierence microscopy (PDM) images were recorded and
analyzed.
2.5. Dynamic Mechanical Analysis. The most commonly
used method of fundamental rheological testing of bitumen is a
dynamic mechanical method using oscillatory-type testing,
generally conducted within the region of the linear viscoelastic
(LVE) response. The temperature sweeps were carried out
under controlled strain. The oscillatory tests were performed
using a Discovery Hybrid Series-2 Rheometer (TA), which
applies oscillating shear stress and strain to a bitumen sample
sandwiched between parallel plates, at dierent loading
frequencies and temperatures. The dynamic hybrid rheometer
tests reported in this study were performed at 0.02 Hz constant
frequency and temperature sweeps between 20 and 100 C.
The tests were undertaken with a 20 mm diameter, 1 mm gap,
and parallel plate testing geometry.
2.6. Salt Release Characterization. The release of
HCOOK from the polymer composite-modied bitumen was
investigated by immersing samples in water and measuring the
K+ ions released into water using an ion selective electrode (ISE
K800). Dry composite-modied bitumen samples were
incubated in 20 mL of deionized water for dierent time
periods without changing the incubation water medium. K+
ions were measured using the ion selective electrode (WTW
Potassium Combination Electrode K 800) at specic time
points. The amount of HCOOK released was calculated using
the measured K+ concentrations.
2.7. Anti-icing Property Characterization. The anti-icing
properties of base bitumen and composite-modied bitumen
were investigated with respect to the freezing rate of a water
droplet on the bitumen surface. Freezing of water droplets on
the bitumen surfaces was induced using a temperature- and
humidity-controlled chamber (Teknol Co.). A 100 L sample
of water was dropped onto each membrane surface and
incubated in the chamber with a 14 C plate and 56 C
ambient temperature. The durations of the freezing of water
droplets were recorded based on the visual observation for
disappearance of transparency of the droplets. Incubation of the
surfaces with water was repeated three times. For the rst part
of our experiments, water droplets were removed after each
exposure and fresh drops were added after recording freezing
times. In the second part, water droplets were kept on the
surface after each exposure and subsequently evaporated in
place so that the salt would not be lost with the water, and the
fresh drops of water were then added onto the surface for the
repeat measurements.

Table 1. Description of the Sets Studied for Characterization

of Mechanical and Anti-icing Properties
sample set
base bitumen
bitumen + 5% salt
bitumen + 5% solid SBS
bitumen + 5% dry
composite (no agar)
bitumen + 5% dry
composite (with agar)

percent polymer in
bitumen (w/w)

presence
of agar

presence of
KCOOH

0
0
5
5

no
no
no
no

no
yes
no
no

yes

no

Dynamic mechanical tests were conducted to investigate the

viscoelastic properties of bitumen. The viscoelastic parameters
measured in the oscillatory tests are the complex modulus (G*)
and the phase angle (). G* is dened as the ratio of maximum
stress to maximum strain and provides a measure of the total
resistance to deformation under shear loading. It contains
elastic and viscous components, which are the storage modulus
(G) and loss modulus (G), respectively. These two
parameters are related to each other through the phase angle
(), which is the phase lag between the applied shear stress and
the shear strain during an oscillatory test. The phase angle is a
measure of the viscoelastic balance of the material behavior and
is equal to the arctangent of the ratio of loss modulus (G) to
storage modulus (G).20 Isochronal plots of the complex
modulus G* versus temperature at 0.02 Hz for all samples
dened above are shown in Figure 2. Measurements were done
three times for these experiments, and the average of data
points was plotted. The measurement error of these devices is
less than 0.1%.
Modication of the bitumen with the SBS polymer resulted
in only a minor increase in G* at lower temperatures, but there
was a signicant increase at high temperatures compared to
base bitumen. The G* prole reached a plateu region,
consistent with a polymer network dominant structure
observed in previous studies with SBS polymer-modied

3. RESULTS AND DISCUSSION

3.1. Dynamic Mechanical Analysis. Bitumen was
modied by the SBS polymer in the dry form as a control. A
mixture of HCOOK and bitumen was prepared as an additional
control. In previous studies, 310% weight fraction of polymer

Figure 2. Isochronal plots of the complex modulus, G*, at 0.02 Hz for

SBS and composite-modied bitumen.
12589

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

bitumen.20 In previous studies, similar plateau regions are
reported in 37 wt % SBS polymer-modied bitumen, and they
are conned in linear viscoelastic region.20,34 G* values
obtained for 5% SBS-modied bitumen within this temperature
range (20100 C) is also consistent with other studies.20,34
We also modied bitumen with an SBS solution (SBS dispersed
in cyclohexane) and a wet composite (SBS and HCOOK
dispersed in cyclohexane) and observed a signicant decrease in
G* compared with base bitumen at all temperatures
(Supporting Information, Figure S1), which may be associated
with the damage of the binder structure due to the presence of
cyclohexane. Because there is no such decrease in the G* versus
temperature prole for the dry composite-modied bitumen,
we continued to use dry composites rather than wet composites
in our experiments. It was observed from the G* proles that
the composite-modied bitumen had G* proles that were very
similar to those of the base bitumen and did not have higher
values of G*, as was found with the dry SBS modication. This
lower G* was due to the lower amount of SBS in the bitumen
dry composite sample compared to the amount in bitumen
solid SBS sample. This result was further supported by the G*
values of the salt-modied bitumen sample, where there was a
minor decrease in G*. These results suggest that dry
composite-modied bitumen has similar mechanical strength
despite the presence of the anti-icing agent. In previous studies,
the deterioration eect of acetate- and formate-based deicers
was demonstrated.15,17 In those studies, deicers were applied on
asphalt pavements directly, which could cause this kind of
deterioration. In our design, we incorporated potassium
formate into bitumen through SBS polymer that would
counterbalance the deterioration eects of formate-based
deicers through formation of a polymer network which
increased the resistance of bitumen against deformation. This
eect was also discussed in a previous study in which asphalt
concrete was stored in potassim formate solution at 19 C for
48 h. In that study, authors reported that polymer mix
modication including SBS polymer counterbalanced the
negative eects of potassium formate solution in terms of
mechanical properties, such as indirect tensile strength and
fracture, and resulted in reduced susceptibility to the anti-icing
chemical.16 However, SBS polymer modication does not
contribute to binderaggregate binding anity, and binder
aggregate anity decreases in potassium formate solutions. We
observed similar results from dynamic mechanical analysis in
terms of mechanical properties (Figure 2). Although salt
addition causes a minor decrease in the complex modulus of
bitumen, salts including composite-modied bitumen have
slightly increased complex modulus values compared to base
bitumen, which shows that adding salt in composite structure
does not decrease mechanical strength and resistance of
bitumen against deformation. In addition, tests used in the
previous study imitate conventional application of salts on
asphalt pavements by storing them in salt solutions. However,
in this study, we considered encapsulation of salt in SBS
polymer and controlled release of salt through bitumen for antiicing property to replace the conventional use of salt on asphalt
for the deicing property.
Phase angle () versus temperature values for the SBS
polymer and composite-modied bitumen groups are shown in
Figure 3. It can be observed from Figure 3 that measurements
are also sensitive to the modication of bitumen. Tan is the
ratio of the loss modulus to storage modulus, which
corresponds to the ratio of the viscous to elastic nature of

Figure 3. Isochronal plots of the phase angle at 0.02 Hz for SBS and
composite-modied bitumen.

the material. Figure 3 shows that the values of bitumen

samples modied with the dry composite are lower than the
base bitumen at low temperatures, but then they increase and
reach values similar that of the base bitumen at high
temperatures. This result suggests that the composite-modied
bitumen is more elastic than the base bitumen at temperatures
up to about 60 C. values of solid SBS-modied bitumen has
a decreasing trend with temperature, which is consistent with
previous studies and is due to an increase in the elastic nature
of the binder from the distribution of polymer.20 Lower
values were observed for bitumen samples modied by the wet
composite compared to bitumen modied by the dry
composite for the entire temperature range (Supporting
Information, Figure S2). The low values can be explained
by decreases in viscosity due to the presence of cyclohexane in
the medium. This is undesirable for pavement applications,
further supporting our decision to focus on dry composites for
this work. The measurements together with G* measurements also suggest that the dry composite-modied bitumen
increases the elasticity at low temperatures, and composite
modication does not aect the viscoelastic balance and the
strength of bitumen at high temperatures. Hence, it was
concluded that, from a viscoelastic property standpoint, the
functional composite in the dry form provides sucient
mechanical performance for pavement applications.
3.2. Morphological Characterization of Modied
Bitumen. 3.2.1. Fluorescence Microscopy. The resins and
aromatics present in bitumen swell the SBS polymer, and this
swelling results in uorescence of SBS under uorescence
microscopy with 520 nm emission and 450490 nm excitation
wavelength range. The uorescence behavior due to swelling
has been used to infer the morphology of the SBS and
composite in bitumen.22 Bitumen samples modied with SBS
and composite were investigated under uorescence microscopy to visualize the distribution of the additive throughout the
bitumen. Fluorescence microscope images for composite- or
SBS-modied bitumen samples are shown in Figure 4. It can be
observed that solid SBS can form better polymer distribution
than composite-modied bitumen (Figure 4a). This dierence
in morphology is consistent with the observed dynamic
mechanical properties of solid SBS-modied bitumen, as a
uniform distribution of polymer would have a positive inuence
on the complex modulus and elastic properties of bitumen,
12590

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

small-sized saturates and aromatics. Masson et al. reported that

asphalt contains multiphase bitumen, consisting of a nanostructure phase, dened as bee-like, and a dispersed phase
referred to as the catana or catanic phase.37 The name bee
comes from the presence of alternating higher and lower parts
in the surface topography of bitumen.
Figure 5 shows AFM images in three-dimensional phase (left
column) and topography (right column) modes for the base
and modied bitumen samples. Base and bitumen samples
modied by SBS polymer and dry polymer composite with and
without agar were investigated using soft tapping mode AFM.
Both topographic and phase dierence images were recorded
and analyzed. In Figure 5, the phase images in the left panel
demonstrate the soft sections as dark regions, where the rigid
domains in the bitumen structure are shown with a light color.
In the corresponding height images in the right panel (Figure
5), the dark areas correspond to valleys and low areas of the
surface prole, while the light areas correspond to peaks on the
surface.
Our AFM images qualitatively demonstrate the standard
description of bitumen structure. As shown in Figure 5a, two
phases can be identied for base bitumen, but the bee-like
structures are not clearly distinguishable. More distinctive beelike structures were observed for bitumen modied by the
polymer composite (5% (w/w)) and SBS (5% (w/w))
demonstrating that modication of bitumen changes the
surface topography of bitumen, resulting in more phase and
height dierences. The bee-like structures become more
distinct with the addition of other ingredients into the bitumen.
Figure 5b and 5c show that the composite polymers with or
without agar in bitumen are distributed more uniformly than
the SBS polymer, which is also consistent with ouresence
microscopy characterization (Figure 4). Larger bee-like features
are observed in the SBS polymer-modied bitumen. In
addition, a new phase is observed around the bee-like features
in the SBS and composite (with agar)-modied bitumen, which
could be attributed to higher swelling of SBS in those regions
(Figure 5c,d).
3.3. Characterization of Anti-icing Properties. Antiicing properties of glass slides coated with base bitumen and
composite-modied bitumen were investigated in a temperature- and humidity-controlled chamber on a 14 C plate,
with 6 C medium temperature and 50% humidity conditions.
After equilibration at the testing temperature, 100 L of water
was dropped on each sample surface, and the duration of
freezing for water droplets was recorded based on the
disappearance of transparency. Figure 6 shows the images of
bitumen-coated samples at dierent time points in the
chamber. It can be observed from Figure 6 that the freezing
started after 4 min on the base bitumen, whereas it took about
9 min for water droplets to freeze on the functional compositemodied bitumen. At the end of 27 min, all droplets on the
base bitumen were completely frozen, while only half of the
droplets were frozen on the functional composite-modied
bitumen. The average droplet freezing times were calculated as
5 and 15 min on the base bitumen and functional compositemodied bitumen surfaces, respectively.
The eect of increasing the salt content of the composite on
the anti-icing properties of composite-modied bitumen (5%
(w/w)) was also investigated by monitoring the freezing of
water droplets on polymer composite (without agar)-modied
bitumen surfaces with altered salt concentrations. Bitumen was
modied with composites with 1X, 2X, 3X, and 4.5X salt

Figure 4. Morphological analysis of SBS and composite-modied

bitumen with uorescence microscopy (scale bar: 100 m).

which were seen through the higher G* and lower phase angle
of samples modied by solid SBS (Figures 2 and 3). We also
modied bitumen with SBS/cyclohexane solution and wet
composites, and we observed that the polymer formed some
aggregates (Figure S3a,c,e). This morphology is also consistent
with the observed complex modulus and phase angle values,
which showed a decrease in mechanical strength (Figures S1
and S2), supporting our selection of dry composites for further
analysis. It was also observed from the uorescence images that
higher concentrations of the polymer composite in the bitumen
have more uorescing regions, as would be expected because
the SBS component is the material leading to uorescence
(Figure S3c,e).
3.2.2. Atomic Force Microscopy. The microstructure of
bitumen is usually described by the so-called colloidal model.
According to this model, bitumen is considered as a colloidal
system constituted by micelles of asphaltenes dispersed in
bitumen.35,36 Typical AFM images demonstrate that bitumen is
a multiphase material. Asphaltenes and resins behave as solidlike materials, resulting in an increased stiness.31 These largersized molecules seem to form the stier parts observed during
AFM testing. The soft matrix, on the other hand, contains
12591

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

Figure 5. AFM phase dierence (left column) and topographic image (right column) of (a) base bitumen, (b) dry composite (with agar)-modied
bitumen (5% (w/w)), (c) dry composite (without agar)-modied bitumen (5% (w/w)), and (d) SBS polymer-modied bitumen (5% (w/w)).
Samples in panels ad were all prepared under the same conditions including temperature, mixing speed, and storage conditions.
12592

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

4.5X salt contents. The results suggest that the 2X salt content
has the greatest inuence on the anti-icing capability, and this
eect is more signicantly observed after a second exposure.
Interestingly, the 3X and 4.5X salt negatively inuenced the
freezing delay. This result may be due to the low encapsulation
eciency of high salt within the composite structure, where a
signicant amount of salt was lost during sample preparation.
The most promising bitumen sample, the compositemodied bitumen with 2X salt content, delayed freezing by
approximately 10 min longer in the rst exposure, and 25 min
longer in the second exposure compared to base bitumen at the
chosen conditions (plate temperature, 14 ; medium
temperature, 56 C; and relative humidity, 50%). These
results suggest that 2X salt content in bitumen is the optimum
amount of salt that should be used in the polymer composite to
study the anti-icing properties of the composite-modied
bitumen.
Longer freezing delays in second exposures occurred
compared to rst exposure because of high salt releases in
the second exposure. However, salt release was not sustained in
the third exposure because of salt lost during the removal of
water droplets after each exposure and addition of fresh ones.
The eect of agar within the composite on the anti-icing
properties of bitumen was also investigated by monitoring the
base and composite-modied bitumen with or without agar
(5% (w/w)) in a temperature- and humidity-controlled
chamber. The bitumen surfaces were exposed to water three
times. However, the method used in previous experiments was
improved by keeping water on the surface after each exposure,
evaporating the water, and leaving the salt on the surface
between exposures to show only the eect of the compositemodied bitumen, not of salt loss. For these experiments, the
2X salt condition, which corresponded to 0.25 g/mL salt in wet
emulsions, was used. The inuence of agar on freezing delays
using the previous method was also studied, and the results are
presented in Figure S4 (Supporting Information). Average
freezing times of water droplets on samples were observed to
decrease in second and third exposures because of the loss of
salt between exposures. The average freezing times of water
droplets on samples for each exposure using the new method
can be observed in Figure 8. The results show that the addition
of agar into polymer composite increases freezing times of

Figure 6. Surface images of base bitumen and composite (with agar)modied bitumen (7.5% (w/w)) in a humidity- and temperaturecontrolled chamber at dierent time points (plate temperature, 14
C; medium temperature, 56 C; and relative humidity, 50%). Time
(t) is given in minutes.

content, which corresponded to 0.125, 0.25, 0.375, and 0.5625

g/mL salt in wet emulsions, respectively. To test salt release
from dierent spots on the bitumen surfaces, subsequent
incubations of the surfaces with water droplets were repeated at
least three times. The bitumen or composite-modied bitumen
surfaces were exposed to water three times by removing the
frozen droplets and subsequently dropping fresh droplets.
Figure 7 shows the results of average freezing delay measured
on composite-modied bitumen surfaces with 1X, 2X, 3X, and

Figure 7. Freezing delays on surfaces of base and composite (without

agar)-modied bitumen (5% (w/w)) with dierent salt content in a
humidity- and temperature-controlled chamber (plate temperature,
14 C; medium temperature, 56 C; and relative humidity, 50%).

Figure 8. Freezing delays on the surfaces of base, composite (with

agar), and composite (no agar)-modied bitumen (5% (w/w)) in a
humidity- and temperature-controlled chamber (plate temperature,
14 C; medium temperature, 56 C; and relative humidity, 50%).
12593

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

water droplets on bitumen surfaces. This improvement can be
explained by the positive contribution of agar to saltwater
interactions in bitumen. Agar behaves as a bridge in these
interactions and helps with the distribution of salt throughout
the bitumen.
The results also show that the freezing delays in the second
exposure are higher than those of the third exposure, but this
dierence is less severe than in the experiments shown in
Figure 7. This indicates that the change in the experimental
method improved salt retention but did not fully maintain the
activity of the salt, likely due to spreading of salt on surfaces
during sample handling.
The results of anti-icing experiments suggest that the
modication of the bitumen with 5% (w/w) functional
composite delayed frost formation for 2025 min, depending
on the salt content and the presence of agar in the composite at
14 C plate temperature conditions. Previous studies have
shown that asphalt mastics containing 50% (w/w) sodium
chloride ller in the bitumen delayed frost formation by 20 min.
We have shown comparable behavior with composites at a
much lower loading (5% (w/w)). In this work, we observed
signicant delay in freezing on composite-modied bitumen
surfaces compared to base bitumen surfaces using low
composite content (5% (w/w)).
3.4. Characterization of Salt Release from CompositeModied Bitumen. Salt release proles from polymer
composite-modied bitumen were investigated by immersing
the samples in water and measuring the HCOOK concentration in water with time. Concentration of HCOOK in water
was measured by a potassium ion selective electrode. For salt
release experiments, functional composite-modied bitumen
with or without agar in the composite was used. Both 5% and
7.5% (w/w) composite loadings in bitumen were studied
separately. Mixing of potassium formate in bitumen was also
tested at the early stages of this study; however, high amounts
of salt aggregates were present on the surface of bitumen, and
salt could not be incorporated into bitumen properly. Salt could
not be dispersed into bitumen uniformly because of the
dierence of hydrophilichydrophobic nature of salt and
bitumen. This made further characterization of salt release
from bitumen impossible, and hence comparisons of salt release
from bitumen and composite-modied bitumen could not be
made. The salt concentration proles in water are presented in
Figure 9. Experiments were carried out at least three times, and
the average value for all data points was plotted. The
measurement error is below 0.1%. Figure 9 shows that the
salt was continuously released for 67 days from the composite
(with agar)-modied bitumen (5% (w/w)), where for other
samples, the HCOOK concentration reached saturation at
around 47 days. The slow release proles of HCOOK from the
samples demonstrate the potential of this composite for longer
periods of use. After 67 days of incubation of the samples in
water, 35.8, 7.46, and 5.34 mg of total HCOOK were released
from composite (with agar)-modied bitumen (5% (w/w)),
composite (no agar)-modied bitumen (5% (w/w)), and
composite (with agar)-modied bitumen (7.5% (w/w)),
respectively. These values corresponded to 10.8%, 1.07% ,and
1.5% of the initial salt contents of the composite-modied
bitumen samples. Release of that low amount of anti-icing agent
is promising for maintenance of the anti-icing property.
Because there will be mechanical abrasion on real roads, salt
release from underlying layers might provide longer anti-icing
eects in real-life conditions. In a previous study, the release of

Figure 9. Time-dependent release of HCOOK from compositemodied bitumen samples.

an anti-icing ller from asphalt mixtures was investigated, where

36% and 47% of sodium chloride was released in two months
from two dierent samples with altered porosity.19 The salt
release prole of polymer composite-modied bitumen
obtained in the current study is promising for preparation of
pavements with sustainable anti-icing properties.
Even though the composite (with agar)-modied bitumen
(7.5% (w/w)) has the highest amount of salt due to the highest
composite concentration, the highest amount of salt was
released from composite (with agar)-modied bitumen (5%
(w/w)) (Figure 9). This result can be explained by the
increased aggregation of the polymer composite in bitumen
when applied at high concentrations. Aggregation of the
polymer composite results in salt retention within the structure
and makes it dicult for salt transport to take place toward
bitumen surface. In order to explain this result further, the
morphologies of the samples were investigated under
uorescence microscopy. The uorescent microscopy images
of the composite (with agar)-modied bitumen (5% (w/w)),
the composite (no agar)-modied bitumen (5% (w/w)), and
the composite (with agar)-modied bitumen (7.5% (w/w)) are
shown in Figure 10. The aggregation of the polymer composite

Figure 10. Fluorescent images of (a) composite (no agar)-modied

bitumen (5% (w/w)), (b) composite (with agar)-modied bitumen
(5% (w/w)), and (c) composite (no agar)-modied bitumen (7.5%
(w/w)) (scale bar: 100 m).

in bitumen modied by 7.5% (w/w) composite is clearly

observed from the Figure 10, where there is a more uniform
distribution in bitumen modied by 5% (w/w) composite with
or without agar in the internal phase (Figure 10).
Figure 9 also shows that the presence of agar increases the
salt release from the composite-modied bitumen. Water
swelling of agar increases when placed in salt solutions such
as potassium chloride, sodium chloride, and calcium chloride.
12594

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

Author Contributions

However, this eect decreases when these salts are used in two
or three salt combinations.38 Higher amounts of water can be
absorbed from the bitumen surface, and higher amounts of salt
which is dissolved in water can reach bitumen surface when
agar and potassium coexist. The possibility for the better
transport of salt to the bitumen surface in the presence of agar
was also veried by freezing experiments, in which samples with
agar in the composite had increased freezing delay on the
composite-modied bitumen surface (Figure 8).

D.A. performed the experiments. D.A. and S.K. designed the

experiments and wrote the paper. D.A., R.K., R.O.C., and S.K.
analyzed the data. S.K. contributed reagents, materials, and
analysis tools.
Funding

This study was supported by Turkish Petroleum Reneries

(TUPRAS).
Notes

The authors declare no competing nancial interest.

4. CONCLUSIONS
We used a polymer composite containing SBS and HCOOK
salt to modify bitumen and to obtain anti-icing surface
functionality. In the rst part of the work, we investigated the
inuence of mixing for dierent functional composites and
performed morphology and rheology experiments to evaluate
the distribution of the polymer composite in bitumen and
characterize the mechanical properties of the nal mixture. We
achieved a uniform distribution of the polymer composite in
bitumen without compromising the mechanical properties.
Next, we conducted freezing experiments to characterize the
anti-icing properties of polymer composite-modied bitumen
surfaces, where the composite-modied bitumen delayed the
freezing time of water droplets by 20 min in a temperature- and
humidity-controlled chamber with a 14 C plate, 56 C
medium temperature, and 50% relative humidity. Salt content
was optimized to achieve a maximum delay in freezing, and it
was found that the 2X salt condition, or 0.25 g/mL, resulted in
the best results in all cases. In addition, we demonstrated
controlled salt release over 4060 days from compositemodied bitumen and investigated the salt release proles of
modied bitumen with respect to the presence of agar and
composite content. We found that agar positively contributes to
the release of the salt from the bitumen, potentially due to
enhanced water-salt hydrophilic interactions. High composite
content decreases the salt release because of the formation of
polymer aggregates in bitumen, which prevent salt transport to
the surface. Mechanical abrasion might also improve the antiicing property on bitumen surfaces, which will enable more salt
release from underlying layers. Because this kind of abrasion
occurs naturally on roads because of moving vehicles in real-life
conditions, the expected continuous salt release may allow for a
sustainable anti-icing property on road surfaces.
In summary, we showed the possibility of incorporation and
delivery of anti-icing agents through bitumen using a functional
polymer composite. Modication of bitumen with an anti-icing
functional composite with preserved mechanical properties has
been studied for the rst time. The use of polymer compositemodied bitumen with anti-icing properties is promising for
production of functional pavements in industrial applications.

ACKNOWLEDGMENTS
We thank Dr. Blair Brettmann for useful comments about our
manuscript. AFM analyses were performed at Koc University
Surface Science Center (KUYTAM).

(1) Kraj, A. G.; Bibeau, E. Measurement method and results of ice

adhesion force on the curved surface of a wind turbine blade.
Renewable Energy 2010, 35, 741746.
(2) Cuelho, E.; Harwood, J.; Akin, M.; Adams, E.Establishing best
practices for removing snow and ice from California roadways;
Department of Research and Innovation, California Department of
Transportation, 2010.
(3) Kulinich, S. A.; et al. Superhydrophobic Surfaces: Are They Really
Ice-Repellent? Langmuir 2011, 27 (1), 2529.
(4) He, M.; et al. Super-hydrophobic surfaces to condensed microdroplets at temperatures below the freezing point retard ice/frost
formation. Soft Matter 2011, 7 (8), 39934000.
(5) Feuillebois, F.; Bazant, M. Z.; Vinogradova, O. I. Effective slip
over superhydrophobic surfaces in thin channels. Phys. Rev. Lett. 2009,
102 (2), 026001.
(6) Alizadeh, A.; et al. Dynamics of ice nucleation on water repellent
surfaces. Langmuir 2012, 28 (6), 31806.
(7) Bahadur, V.; et al. Predictive model for ice formation on
superhydrophobic surfaces. Langmuir 2011, 27 (23), 1414350.
(8) He, M.; et al. Super-hydrophobic film retards frost formation. Soft
Matter 2010, 6 (11), 23962399.
(9) Hellsten, P. P.; et al. Use of potassium formate in road winter
deicing can reduce groundwater deterioration. Environ. Sci. Technol.
2005, 39 (13), 50955100.
(10) Hall, D. L.; Sterner, S. M.; Bodnar, R. J. Freezing point
depression of NaCl KCl H20 solutions. Econ. Geol. Bull. Soc. Econ.
Geol. 1988, 83, 197202.
(11) Shi, X, L, Y., Mooney, M, Berry, M, Hubbard, B, Fay, L,
Leonard, A. B. Eect of chloride-based deicers on reinforced concrete
structures; WSDOT Research Report WA-RD 741.1; Washington State
Department of Transportation, 2010.
(12) Starck, P.; Lofgren, B. Effect of de-icing agents on the
viscoelastic properties of asphalt mastics. J. Mater. Sci. 2007, 42, 676
685.
(13) Testa, D. Highway operations: Progress and products update;
Strategic Highway Research Program,National Research Council:
Washington, DC, 1991.
(14) Penn LS, M. Ice-pavement bond prevention: Fundamental study;
SHRP-W/UFR-92-606; National Research Council: Washington, DC,
1992.
(15) Nilsson, F. Durability problems on Nordic airelds-The inuence of
deicing agents on asphalt concrete. in The XXIInd PIARC World Road
Congress, 2003.
(16) Santagata, E.; Baglieri, O.; Riviera, P.P. Effect of anti-icing
chemicals on stripping of asphalt concrete mixtures for airport runway
wearing courses. Airfield and Highway Pavement 2013, 12291239.
(17) Pan, T.; He, X.; Shi, X. Laboratory investigation of acetate-based
deicing/anti-icing agents deteriorating airfield asphalt concrete. Asphalt
Paving Technology-Proceedings 2008, 77, 773.

ASSOCIATED CONTENT

* Supporting Information
S

The Supporting Information is available free of charge on the

ACS Publications website at DOI: 10.1021/acs.iecr.5b03028.
Additional analysis data about rheology and morphology
of composite-modied bitumen (PDF)

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: skizilel@ku.edu.tr. Phone: 90-212-338-1836. Fax: 90212-338-1548.

12595

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596

Article

Industrial & Engineering Chemistry Research

(18) Ganjyal, G.; Fang, Q.; Hanna, M. A. Freezing points and smallscale deicing tests for salts of levulinic acid made from grain sorghum.
Bioresour. Technol. 2007, 98 (15), 28148.
(19) Giuliani, F.; Merusi, F.; Polacco, G.; Filippi, S.; Paci, M.; et al.
Effectiveness of sodium chloride-based anti-icing filler in asphalt
mixtures. Construction and Building Materials 2012, 30, 174179.
(20) Airey, G. Rheological properties of styrene butadiene styrene
polymer modified road bitumens. Fuel 2003, 82 (14), 17091719.
(21) Perezlepe, A. Influence of the processing conditions on the
rheological behaviour of polymer-modified bitumen. Fuel 2003, 82
(11), 13391348.
(22) Sengoz, B.; Isikyakar, G. Analysis of styrene-butadiene-styrene
polymer modified bitumen using fluorescent microscopy and conventional test methods. J. Hazard. Mater. 2008, 150 (2), 42432.
(23) Yildirim, Y. Polymer modified asphalt binders. Construction and
Building Materials 2007, 21 (1), 6672.
(24) Lu, X.; Isacsson, U. Chemical and rheological evaluation of
ageing properties of SBS polymer modified bitumens. Fuel 1998, 77,
961972.
(25) Zhu, J.; Birgisson, B.; Kringos, N. Polymer modification of
bitumen: Advances and challenges. Eur. Polym. J. 2014, 54, 1838.
(26) Lu, X., Soenen, H.; Redelius, P. Rheological characterization of
polymer modied bitumens. Annu. Trans. - Nord. Rheol. Soc. 2011, 19.
(27) Bulatovic, V. O.; Rek, V.; Markovic, K. J. Effect of polymer
modifiers on the properties of bitumen. J. Elastomers Plast. 2014, 46
(5), 448469.
(28) Bulatovic, V.; Rek, V.; Markovic, K. Mater. Res. Innovations 2012,
16 (1), 16.
(29) Bulatovic, V. O.; Rek, V.; Markovic, K. Influence of polymer
types on bitumen engineering properties. Mater. Res. Innovations 2013,
17 (3), 189194.
(30) Airey, G. D. Styrene butadiene styrene polymer modification of
road bitumens. J. Mater. Sci. 2004, 39 (3), 951959.
(31) Shell-Bitumen-U.K. The Shell Bitumen Handbook; ShellBitumen: Chertsey, 1995; Vol. 1.
(32) Carrera, V.; Cuadri, A. A.; Garca-Morales, M.; Partal, P.
Influence of the prepolymer molecular weight and free isocyanate
content on the rheology of polyurethane modified bitumens. Eur.
Polym. J. 2014, 57, 151159.
(33) Kanyas, S.; et al. Nanoparticle and gelation stabilized functional
composites of an ionic salt in a hydrophobic polymer matrix. PLoS One
2014, 9 (2), e88125.
(34) Wen, G.; et al. Rheological characterization of storage-stable
SBS-modified asphalts. Polym. Test. 2002, 21 (3), 295302.
(35) Lesueur, D. The colloidal structure of bitumen: consequences
on the rheology and on the mechanisms of bitumen modification. Adv.
Colloid Interface Sci. 2009, 145 (12), 4282.
(36) Loeber, L.; Muller, G.; Morel, J.; Sutton, O. Bitumen in colloid
science: a chemical, structural and rheological approach. Fuel 1998, 77,
14431450.
(37) Masson, J. F.; Leblond, V.; Margeson, J. Bitumen morphologies
by phase-detection atomic force microscopy. J. Microsc. 2006, 221,
1729.
(38) Prat, S. Antogonistic salt action in relation to the swelling and
dyeing of powdered agar. Am. J. Bot. 1927, 14, 167186.

12596

DOI: 10.1021/acs.iecr.5b03028
Ind. Eng. Chem. Res. 2015, 54, 1258712596