Construction and Building Materials 145 (2017) 326–335

Contents lists available at ScienceDirect

Construction and Building Materials

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

Effect of storage stability on chemical and rheological properties of

polymer-modified asphalt binders for road pavement construction
L. Zani a, F. Giustozzi b,⇑, J. Harvey c
a
Department of Civil and Environmental Engineering, Transport Infrastructures Section, Polytechnic University of Milan (Politecnico di Milano), P.zza Leonardo da Vinci 32,
20133 Milan, Italy
b
Civil and Infrastructure Engineering, Royal Melbourne Institute of Technology (RMIT University), 376-392 Swanston St., Melbourne, VIC 3000, Australia
c
Department of Civil and Environmental Engineering, 3153 Ghausi Hall, University of California Davis, One Shields Avenue, Davis, CA 95616, USA

h i g h l i g h t s

Storage stability of five polymer-modified binders was evaluated.

Advanced indices were developed to identify storage stability properties of PMBs.

Physical, rheological, and chemical properties were analyzed.

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

Article history: Polymer modified asphalt binders (PMB) are being commonly used instead of neat binders when superior
Received 2 January 2017 performance is needed during the road service life. The addition of polymer improves the binder’s prop-
Received in revised form 4 April 2017 erties at high, low and mid-range temperatures. However, the effectiveness of polymer modification in
Accepted 5 April 2017
improving road performance depends on a chemical-structural modification in the binder’s composition
Available online 12 April 2017
to be effective and durable. The phase inversion phenomenon between asphalt components and the
added polymer leads the binder to assimilate the polymer characteristics. PMBs can have stability prob-
Keywords:
lems and separation; in fact, polymers and bitumen components can lose most of the benefits from the
Asphalt binder
Bitumen
modification due to separation during the storage phase. Common testing for storage stability assessment
Storage stability includes the cigar tuben test (EN 13399, 2010) which provides an indication of whether separation has
Rheology occurred. To evaluate the effects of separation on performance-related properties of the binder, the pre-
Polymer modification sent study analyzed the storage stability and separation phenomena of five PMBs by performing an
Asphaltene advanced-rheological storage stability test, based on frequency sweep tests on samples taken from the
top and bottom part of 3, 5 and 7-day stored cigar tuben specimens. Master curves were developed for
newly mixed and several storage conditions and were compared to identify possible separation phenom-
ena between top and bottom samples. Indices were also developed to quantify separation of samples.
Results showed that storage stability problems commonly start at the third day of storage. Binders which
were seriously affected by separation, seemed to incrementally lose their homogeneity (top and bottom
part) between 3 and 7 days. Other binders preserved their rheological characteristics during storage,
maintaining equal values between the top and bottom samples. All stored binders were affected by an
aging factor.
Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction
Changes in the asphalt binder structure due to polymer modifi-
cation procedures can cause several problems during the storage
phase. Binder in the asphalt storage tank, before being mixed with
the aggregates at the plant, can suffer from separation between
⇑ Corresponding author.
E-mail address: filippo.giustozzi@rmit.edu.au (F. Giustozzi).
polymer and bitumen phase, for instance. Low density polymers
float to the top of the tank, while asphaltenes segregate and move
to the bottom part of the tank. The separation process leads to a
consequent failure of the polymer modified binder’s properties
and loss of all the benefits coming from the modification. Binders
weakened by separation can therefore suffer from both poor rut-
ting performance at high temperatures and poor fatigue and ther-
mal cracking performance at medium-low temperatures. To
preserve binders from these flaws several precautions seem to be

http://dx.doi.org/10.1016/j.conbuildmat.2017.04.014
0950-0618/Ó 2017 Elsevier Ltd. All rights reserved.
 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335 327

PMBs

Physical tests

Softening point [7] Chemical tests

Penetration grade [14]
SARA index [19]
Elastic recovery [15]
Viscosity [16]
Cigar Tuben Test [1]

Rheological Tests

Frequency sweep [17, 18]

Advanced Storage Stability analysis:
frequency sweep test on top cigar tuben parts after 3, 5, and 7 days;
frequency sweep test on bottom cigar tuben parts after 3, 5, and 7 days.
storage stability assessment
Linear Amplitude Sweep test [20]
6

Fig. 1. Storage stability of polymer-modified asphalt binders – Experimental plan.

Delta Ratio Index-calculation example @100000hz after 5 day storage

100.000
|G*top-G*bot|

|G*top-G*un-stored|
Complex Modulus(MPa)

10.000
|G*bot-G*un-stored|

1.000

0.100
1 10 100 1000 10000 100000
Frequency(Hz)
Un-Stored 5d_BOT 5d_TOP

Fig. 2. Delta Ratio index.

necessary: a stabilizing agent can be added into the binder for Table 1
building a durable bitumen-polymer network that is difficult to Standard physical characterization of PMBs.

break and a continuous mixing action in the storage tank can be Binder Softening Point [°C] Penetration [dmm] Elastic recovery [%]
provided to avoid the physical separation. Storage stability prob- [7] [14] [15]
lems are commonly evaluated through a simple physical test PMB_1 77.5 51.5 92.5
[1,2] which measures the difference between the softening point PMB_2 62.1 62.0 94.2
of samples taken from the top and bottom parts of a standard cigar PMB_3 72.5 56.0 93.7
PMB_4 75.5 47.0 96.4
tube after vertical storage in an oven at high temperature (i.e.; PMB_5 65.6 64.0 96.7
150–180 °C). However, the softening point test itself is not suffi-
cient to evaluate the storage stability effects on performance
related properties and, in this study, rheological modifications
to consider the difference between separation issues and unavoid-
due to storage were also taken into account.
able aging effects, which affect the binder by showing increased
PMBs were stored vertically in cigars for 3, 5 and 7 days in an
complex modulus and reduced phase angle. Several separation
oven kept at 180 °C. After each period, samples were taken from
indices were therefore adopted to evaluate the storage stability
the top and bottom sections and frequency sweep tests were per-
of PMBs.
formed at multiple temperatures. This test provided information
about complex modulus (G⁄) and phase angle (d) trends with
increasing storage time. A comparison between master curves 2. Background
from stored and un-stored binders of top and bottom sections eval-
uated the storage effect. However, when studying the change in Polymer Modified Binders have superior performance-related
binder properties over time at high temperature, it is important properties compared with neat bitumen after a chemical process
328 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335

Dynamic Viscosity of PMBs

100.0

Viscosity [Pas]
10.0

1.0

0.1
100 120 140 160 180 200
Temperature [° C]
PMB_1 PMB_2 PMB_3 PMB_4 PMB_5

Fig. 3. Dynamic viscosity of PMBs.

between the bitumen’s components and the added polymer, of any
kind. The durability of the phase inversion chemical process is
strictly connected with the storage procedure used for the newly
modified binder. The storage stability of a PMB plays a key part
in the binder’s life, providing the binder with the enhanced proper-
ties it was created for; i.e., greater elasticity, higher softening point
temperature, lower penetration, reduced viscosity at mixing and
compaction temperature and, primarily, complex modulus values
sufficiently low at medium-low temperatures and sufficiently high
at high temperatures for better resisting to thermal and fatigue
cracking, and rutting, respectively [3].
Hanyu et al. [4] studied the chemical bonding of Styrene-
Butadiene-Styrene (SBS) block copolymer and asphalt binder.
According to the study, by observing a picture of the SBS-
modified asphalt binder, its structure can be broadly divided into
two types. In one the asphalt forms a continuous phase and in
the other the SBS forms a continuous phase. However, trying to
join binder and polymer properties to improve durability in pave-
ment design seemed to be highly dependent on the storage phase
of the modified binder. Lu et al. [5] studied the effects of compat-
ibility problems between the chosen polymer and the neat binder
produced problems in the pavement.
During the storage process, a phase separation can occur in
which the asphaltenes in the bitumen settle to the bottom of the
storage tank, while the swollen SBS polymers move to the top of
the tank. The phase separation may be determined by storage tests.
Lu et al. [6], working on both rheological and physical features of
stored binders, noticed that the ratio of complex modulus between
bottom and top portions of a stored cigar tuben sample provided a
better indication of grading separation problems than the common
softening point test [7,8]. In Ref. [3], the LAST (Laboratory Stability
Test) device was developed in which separation and degradation
ratios are calculated from the complex shear modulus (G⁄) and
the shear phase angle (d) responses. Bahia et al. [9], working with
the LAST technology, indicated how separation and degradation
phenomena appeared to have influences on future rheological life
of bitumen. These parameters highlighted the need for use of an
agitation process during storage to prevent separation.
Gonzalez [10] developed two parameters based on the trend of
tanðdÞ of bottom and top samples to characterize the storage stabil-
ity phenomenon, obtaining reliable results. Lu et al. [6] showed
that the compatibility and storage stability of SBS modified binders
were influenced by the chemical nature of the neat bitumen and
the characteristics and content of the polymer. At a given SBS con-
tent, modified binders produced from bitumens with a higher con-
tent of aromatics exhibited better compatibility and greater
storage stability, while an increase in asphaltenes may adversely
affect the storage stability [5]. On the other hand, the storage sta-
bility of modified binders may decrease with increased SBS con-
tent. Hanyu et al. [4] related the SBS-binder network, studied
with the fluorescence microscope, with binder properties and dis-
covered that better dispersion and higher fineness of polymer can
increase the binder stability during storage. From a chemical
standpoint, using different types of SBS (i.e.; radial or linear elas-
tomers) or adding other components (i.e.; sulfur) can improve bin-
der stability, as stated by Fu et al. [11] and Wen et al. [12]. Soenen
et al. [13] demonstrated that for a given PMB, morphology and
storage stability are influenced by the temperature and the ther-
mal history of the binder.
3. Materials and methods
In the present study five SBS-modified asphalt binders were provided by differ-
ent suppliers in ‘‘as- sold” condition. SBS contents ranged between 4 and 6% by
weight of the asphalt binder. The binder Performance Grade (PG) after modification
was either 64-28 or 76-22, depending on the supplier. A neat bitumen PG 58-22
was also tested before and after SBS-modification for comparison purposes.
The following tests were conducted:

Physical characterization – Softening Point [7], Penetration grade [14], Elastic
Recovery [15] and Brookfield Viscosity [16] to provide a general overview of
the physical properties.

Standard Storage Stability test – Cigar Tuben Test [1].

Table 2
Cigar tuben test results.

Binder Softening point [°C] Acceptability [1] Average top-bottom softening Increase respect to un-stored binder
point @3 days [°C] before storage [%]
Bottom Top
PMB_1 81.5 82.8 Yes 82.2 6.03
PMB_2 68.7 69.4 Yes 69.1 11.23
PMB_3 80.5 81.1 Yes 80.8 11.48
PMB_4 83.4 85.1 Yes 84.3 11.69
PMB_5 71.3 72.9 Yes 72.1 9.90
 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335 329

PMB_1 @different Storage Time

100.000

10.000

Complex Modulus(MPa)
1.000

0.100

0.010

0.001

0.000
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000
Frequency(Hz)
Un-Stored 3days_Bot 3days_Top 5days_Bot
5days_Top 7days_Bot 7days_Top
a)
PMB_2 @different Storage Time
100.000

10.000
Complex Modulus(MPa)

1.000

0.100

0.010

0.001

0.000
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000
Frequency(Hz)
Un-Stored 3days_Bot 3days_Top 5days_Bot
5days_Top 7days_Bot 7days_Top
b)
Fig. 4. Master Curves of Un-stored and Stored PMBs: a) PMB_1; b) PMB_2; c) PMB_3; d) PMB_4; e) PMB_5.

Rheological evaluation – Frequency Sweep tests [17,18] using a dynamic shear
rheometer (DSR) from 0.2 Hz to 30 Hz at T = 6 °C, 15 °C, 25 °C, 40 °C, 55 °C, 70 °C
on fresh binders. The 8 mm plate configuration of the DSR was used from
T = 6 °C to 25 °C, while the 25 mm configuration was used in the 40 °C - 70 °C
temperature interval.

Advanced Storage Stability analysis – DSR Frequency Sweep tests (same exper-
iment design) on bottom and top samples of cigar tuben samples stored for 3, 5
and 7 days (180 °C), respectively. The same plate configuration was adopted
depending on the testing temperature.

Chemical analysis – Saturate, Aromatic, Resin and Asphaltene (SARA) index [19]
to evaluate the asphaltene content was also performed to relate storage stabil-
ity performance to chemical characterization of the binders.
A summary of the experimental plan is reported in Fig. 1.
The advanced storage stability analysis included the pouring of each binder into
several aluminum foil cigars (32 mm in diameter and 160 mm in height), which
jGtop Gbot j
- Delta Ratio index: DRatio Index½0;00001;1;100000 Hz ¼ jðG    (Fig. 2).
top Gunstored ÞjþjðGbot Gunstored Þj

The gap between the two horizontal G asymptotes for top and bottom curves
was used to spot the degree of possible separation between the top and bottom
sections. The distance between the stored and un-stored curves showed the
oxidation process eventually occurring during storage conditioning. The above
index was calculated after 3, 5 and 7 days of storage.
G 
- Separation index: SIf ðTemperatureÞ ¼ av erage GTop
 in the 0.2–10 Hz interval at
Bot
T = 6 °C, 25 °C, and 55 °C. Then, the final Separation Index was calculated at each
storage time as SI[@3,5,7 days] = average(SI6°C + SI25°C + SI55°C).
Binders therefore present good storage stability properties if Delta Ratio is close
to zero and SI is close to one at each storage step. For example, values of SI from 0.9
to 1.1 identify good storage stability. In the following, a normalized complex mod-
G
ulus, Gnormalized ¼ G stored , will be also used to compute G increase after 7-day storage
unstored

were kept in a vertical position in an oven at 180 °C for 3, 5 and 7 days, respectively. period over the entire frequency range as further explained.
After the storage period, all samples were cooled to ambient temperature and then
cut horizontally in three equal sections according to the standard [1]. The bottom
and top portions were used for further testing to evaluate separation phenomena 4. Analysis of results
due to storage. From the frequency sweep results, master curves [21] were built
for un-stored and stored conditions and then compared. Three replicate tests were
4.1. Physical tests
performed at each storage time and top and bottom portions.
Preliminary investigations included a simple comparison of master curves,
observing how the stored and un-stored curves behave at different storage time. Physical properties of PMBs are summarized in Table 1. Physical
In addition, the following parameters were also considered from the DSR analysis. characterization of binders clearly highlighted some major differ-
330 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335

PMB_3 @different Storage Time

100.000

10.000

Complex Modulus(MPa)
1.000

0.100

0.010

0.001

0.000
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000
Frequency(Hz)
Un-Stored 3days_Bot 3days_Top
5days_Bot 5days_Top 7days_Bot
7days_Top
c)
PMB_4 @different Storage Time
100

10
Complex Modulus(MPa)

0.1

0.01

0.001

0.0001
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000
Frequency(Hz)
Un-Stored 3days_Bot 3days_Top 5days_Bot
5days_Top 7days_Bot 7days_Top
d)

PMB_5 @different Storage Time

100.000

10.000
Complex Modulus (MPa)

1.000

0.100

0.010

0.001

0.000
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000
Frequency(Hz)
Un-Stored 3days_Bot 3days_Top 5days_Bot
5days_Top 7days_Bot 7days_Top
e)
Fig. 4 (continued)
 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335 331

ences; although penetration is in the 50–70 pen range for four out Table 3
of the five binders (one binder is slightly below the lower limit), Delta Ratio index at different storage conditions.

the softening point temperature greatly varied and exhibited up Delta Ratio Index
to 15 °C difference between the minimum and maximum value Binder 3-day storage 5-day storage 7-day storage
between the binders. Generally, lower penetration corresponded
PMB_1 0.27 0.22 0.87
to higher softening temperature but a strong linear relationship PMB_2 0.23 0.11 0.26
was not observed. The presence of SBS-polymer was fully con- PMB_3 0.50 0.66 1.00
firmed by the Elastic Recovery results; all binders provided an elas- PMB_4 0.12 0.19 0.10
tic response which was greater than 90 percent. PMB_5 0.26 0.22 0.30

Dynamic viscosity also underlined major differences at working

temperatures (Fig. 3); although PMB2 was very similar to PMB5 for
penetration and softening point, it had three times greater viscos-
ity reduction in the 130–150 °C temperature interval (i.e.; com-
paction phase) but very similar values at 165–180 °C (i.e.; mixing
phase). PMB2 and PMB3, instead, showed a very similar viscosity
although they were very dissimilar in terms of penetration and
softening point values. Even if sold as ‘‘similar” binders by the sup-
pliers (same type of polymer and polymer content range), PMBs
practically behave in many different ways and therefore need fur-
ther testing to be fully characterized. Physical characterization,
although adopted by many countries, needs to be enhanced with
additional rheological and chemical measurements; this is espe-
cially true when external polymers are included into the asphalt
binder with very poor information about their specific nature or
exact content due to copyright and confidential commercial
information.
The Cigar Tuben test was first conducted following the standard
[1]. According to the standard the maximum tolerable difference in
softening point of the top and bottom portions of the cigar is 3 °C to
consider the binder as not prone to storage instability. Results are
presented in table 2.
All binders passed the acceptance stability test although differ-
ent increases of Softening Point temperature were registered due
to stability issues and combined high-temperature aging process.
Again, ‘‘similar” binders by means of penetration and softening
point evaluation (i.e.; PMB_1 and PMB_3 or PMB_2 and PMB_5)
turned to be very dissimilar after the storage test, providing up
to the double of increase in Softening Point with respect to their
un-stored conditions. Due to the great variability and despite the
acceptance criteria, the softening point measurement itself was
not identified as sufficiently rigorous parameter to characterize
the stability phenomenon of binder storage.
A rheological study was therefore conducted to better identify
differences in storage behaviors between PMBs.
4.2. Rheological and chemical tests
iod except PMB_1 and PMB_3. The first binder showed separation
after 5-day storage while the latter already showed separation
after 3-day storage. In particular, Delta Ratio equal to 1.00
(PMB_3) at 7 days means that samples taken from the bottom
and top portions were completely separated. Outcomes from the
penetration test on the bottom portion were found to be almost
ten times smaller than the top portion sample, meaning a soft
asphalt binder on the top portion of the cigar and hard on the bot-
tom. This behavior confirmed the assumption of heavy-weight
asphaltene molecules settling in the bottom portion making this
section of the tube hard and tough, and low-weight SBS molecules
floating to the top portion and softening the binder (high penetra-
tion value and low softening point).
However, the Delta Ratio index was calculated from master
curve data and is therefore dependent on the specific fitting model
adopted (model-dependent data); this could lead to differences in
results, mainly at very high and very low frequencies depending on
the model. Separation was also assessed by means of the Separa-
tion Index, whose values are directly derived from Dynamic Shear
Rheometer data (raw machine data).
Separation Index results (Table 4) still showed that PMB_1 and
PMB_3 are prone to separation compared to the other binders. In
particular, PMB_1 exhibited poor resistance to storage after 7 days
while PMB_3 revealed separation between 3 and 5 days. PMB_2,
PMB_4, and PMB_5 showed instead a Separation Index close to 1,
therefore providing similar G⁄ values for the top and bottom por-
tions at different testing frequencies and temperatures.
The SARA test [19] was then performed to evaluate the asphal-
tene content of binders as this could be related to storage stability
issues. Commonly, greater asphaltene content binders are more
prone to separate according to Lu et al. [5]. Although this was con-
firmed for PMB_3, which had the greatest asphaltene content and
was heavily prone to separation, the same behavior was not
observed for PMB_1, which had small asphaltene content but

Frequency sweep tests were carried out as described in previ- Table 4

ous sections to evaluate the binder’s complex modulus trend at dif- Separation Index evaluation.
ferent storage conditions; a plot of the master curves is provided in Separation Index
Fig. 4.
Binder 3-day storage 5-day storage 7-day storage
Master curves clearly depict differences in the binders’ storage
PMB_1 1.06 1.02 0.64
behavior. PMB_2, PMB_4 and PMB_5 master curves remained quite
PMB_2 0.96 1.07 1.06
close to each other after storage, thus not showing separation PMB_3 1.21 0.28 0.11
issues between bottom and top portions. On the other hand, PMB_4 1.09 0.97 1.05
PMB_1 and PMB_3 instead suffered separation; variation in stiff- PMB_5 1.10 0.95 1.07
ness after storage is particularly evident for PMB_3. Commonly,
longer storage times caused the greatest variation in binders’ stiff-
ness, as expected. Besides separation due to storage, an aging fac- Table 5
tor due to the long exposure at high temperature can be also Asphaltene Content vs. Separation.

observed as an upward shift of the stored master curves (i.e.; Binder Average [%] St. Dev. [%] Prone to separation
PMB_4) compared with un-stored conditions. PMB_1 27.55 2.03 Yes
In order to quantify the separation phenomenon, some indices PMB_2 34.90 1.14 No
were calculated as described in the previous section. PMB_3 52.20 4.38 Yes
Delta Ratio analysis (Table 3) confirmed master curves trends; PMB_4 23.20 2.21 No
PMB_5 30.50 0.94 No
binders maintained good properties after the longest storage per-
332 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335

was equally subject to separation (Table 5). PMB_4 had the lowest Aging, as expected, resulted in an upward shift of the stored
asphaltene content and did not show much separation. Asphaltene curves compared to un-stored condition. Greater aging rate during
content could be, therefore, one of the causes but separation due to storage is shown by larger gap between stored and un-stored
storage should be better addressed as a multi-mechanism phe- curves.
nomenon as further discussed. The Linear Amplitude Sweep (LAS) test [20] is becoming more
As a comparison and to provide a more comprehensive analysis, commonly used for evaluating binder resistance to fatigue, as evi-
a neat binder was stored in an oven for 7 days following the cigar dent in many literature studies by Anderson et al. [22], Hintz et al.
tuben standard [1] and frequency sweep tests on bottom and top [23] and Botella et al. [24], among others. The LAS test determines
samples were performed. Neat binders should not suffer from sep- the asphalt binders’ resistance to damage by means of cyclic load-
aration as there is no polymer modification. As expected, Fig. 5 ing employing linearly increasing load amplitudes. The sample is
showed perfect overlap between the master curves from the bot- tested using a series of oscillatory load cycles at linearly increasing
tom and top portion sections although a significant upward shift- amplitudes (from 0 to 30 percent strain) at a constant frequency
ing was observed due to aging at high temperature. It should be (10 Hz) to cause accelerated fatigue damage. The LAS was con-
further noted that the neat binder tested was the base binder for ducted using the Dynamic Shear Rheometer.
the modification of PMB_5. The LAS test was performed on both stored neat binder and
Neat binder did not suffer from separation but it was subjected PMB_5 at T = 10 °C, 20 °C and 30 °C and results were then com-
to severe oxidation, while PMB_5 seemed to cope with both sepa- pared to un-stored conditions. Separation and aging effects on bin-
ration and aging quite well, as it could be also inferred from the der fatigue life were thus analyzed (Fig. 7); normalized complex
G modulus was adopted as an evaluation parameter.
Gnormalized trend (Fig. 6), where Gnorm ¼ G stored .
unstored

Neat binder and PMB_5 @7-day storage

100.000

10.000
Complex Modulus (MPa)

1.000

0.100

0.010

0.001

0.000

0.000
0.00001 0.0001 0.001 0.01 0.1 1 10 100 1000 10000 100000
Frequency(Hz)
PMB5_7days_Top PMB5_7days_Bot PMB5_Un-stored
Neat_Un-stored Neat_7days_Bot Neat_7days_Top

Fig. 5. Neat and PMB_5 binder after 7-day storage.

Neat Binder and PMB_5 Normalized Complex Modulus @7-day storage

100
Normalized Complex Modulus

10

0.1
0.00001 0.001 0.1 10 1000 100000
Frequency(Hz)
Un-Stored Neat_7_Bot Neat_7_Top PMB_7_Bot PMB_7_Top

Fig. 6. Comparison between polymer-modified and neat binders at 7-day storage.

 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335 333

LAS @10°C and 7-day storage - Neat

LAS @10°C and 7-day storage - PMB_5

Normalized Complex Modulus

1.2 1.2 binder

Normalized Complex Modulus

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0
0 10 20 30 0 10 20 30
Strain (%) Strain (%)
Un-Stored 7days_ Bot Un-Stored 7days_ Bot
7days_Top Lim_0,5G* 7days_Top Lim_0,5G*

LAS @20°C and 7-day storage - Neat

LAS @20°C and 7-day storage - PMB_5 binder

Normalized Complex Modulus

1.2
Normalized Complex Modulus

1.2

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0
0 10 20 30 0 10 20 30
Strain (%) Strain (%)
Un-Stored 7days_ Bot Un-Stored 7days_ Bot
7days_Top Lim_0,5G* 7days_Top Lim_0,5G*

LAS @30°C and 7-day storage - PMB_5 LAS @30°C and 7-day storage - Neat
1.2
binder
1.2
Normalized Complex Modulus

Normalized Complex Modulus

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2

0 0
0 10 20 30 0 10 20 30
Strain (%) Strain (%)
Un-Stored 7days_ Bot Un-Stored 7days_ Bot
7days_Top Lim_0,5G* 7days_Top Lim_0,5G*

Fig. 7. LAS test for Neat binder and PMB_5 @7-day storage and different test temperature.

Different effects due to long-term storage can be identified in and aging effects and preserved the original resistance to damage
Fig. 7 in terms of a binder’s resistance to damage. Polymer- of the binder. Some polymer modifications (i.e.; PMB_4) instead
modified binder again showed great stability to storage and negli- provided good resistance to storage but increased aging (increased
gible aging due to high-temperature; in fact, normalized complex stiffness and upward shift of the master curve); in this case, the
moduli of the bottom and top portions were very close to the LAS test confirmed the similar behavior between top and bottom
un-stored condition during the entire fatigue test. Neat binder, portions but, again, the faster damage rate attributed to aging
although confirming the lack of separation (bottom and top portion was observed (Fig. 8).
curves were similar), was clearly affected by aging. The effect of the
aging process on fatigue damage can be seen in the LAS graphs as a
faster damage rate (loss of stiffness with increased strain) with 5. Discussion and conclusions
respect to the un-stored condition (reduced resistance to damage)
although it was previously identified as an upward shift of the This paper wanted to shed some light about the great variability
master curve (increased stiffness). The polymer modification, that could be observed in polymer-modified binders in ‘‘as-sold”
whenever correctly applied, proved to be effective against storage conditions providing the reader with a multi-approach experimen-
334 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335

LAS @10°C and 7-day storage PMB_4 LAS @20°C and 7-day storage PMB_4
1.20 1.20

Normalized Complex Modulus

Normalized Complex Modulus

1.00 1.00

0.80 0.80

0.60 0.60

0.40 0.40

0.20 0.20

0.00 0.00
0 10 20 30 0 10 20 30
Strain(%) Strain(%)
Un_Stored 7days_Bot Un_Stored 7days_Bot
7days_Top Lim_0,5G* 7days_Top Lim_0,5G*

LAS @30°C and 7-day storage PMB_4

1.20
Normalized Complex Modulus

1.00

0.80

0.60

0.40

0.20

0.00
0 10 20 30
Strain(%)
Un_Stored 7days_Bot
7days_Top Lim_0,5G*

Fig. 8. LAS G* trend for PMB_4 @7 days.

tal observation-based analysis; polymer modified binders with age; this was also combined with shorter fatigue lives according
similar PG grades does not often guarantee similar performance to amplitude sweep tests. Results highlighted that only three of
after storage. In particular, this paper showed several differences the five binders tested were able to bear storage periods longer
in storage stability and aging susceptibility. A key role is played than three days without losing their initial rheological properties.
by the SBS type and content as added by the suppliers and the neat Delta Ratio and Separation indices were proved to be effective in
binder’s characteristics; since binders were tested in ‘‘as-sold” con- capturing storage stability issues; binders with great resistance
ditions, it was not possible to specifically know the content, type, to storage did not substantially modify the index values stepping
and other chemicals addition in each of them and further explain from three to seven days of storage. Binders prone to separation
the experimental observations. However, it was proved that chem- instead showed modified rheological properties between three
ical incompatibility between the neat binder and polymer could and five days of storage; the migration of polymer phase due to
cause instability during storage and reduced long-term perfor- instability was enhanced by long heating periods at high tempera-
mance even if similar PG is credited for the modified binders prior ture. Finally, a great deal of work in the literature suggested that
to storage. Storage life and thermal history was also demonstrated high asphaltene content is a possible cause for binder separation.
to affect expected properties of binders. This was observed for some of the binders, but not for others.
In addition, the storage stability test as required by the Euro-
pean standard [1] is not sufficient to guarantee effective resistance
Conflict of interest
to separation. Storage conditions in typical storage tanks in actual
practice commonly extend storage temperature and time to more
The authors declare that they have no conflict of interest.
severe conditions compared to what is contemplated by the stan-
dard. Mechanical agitation is also not taken into account by the
standard although it is not easy to simulate in laboratory environ- Funding
ments. Outcomes from this paper emphasized that physical mea-
surements should be supplemented by rheological studies to This research did not receive any specific grant from funding
evaluate binder resistance to storage. Performance-related proper- agencies in the public, commercial, or not-for-profit sectors.
ties were measured at multiple temperatures and frequencies pro-
viding insightful additional information about storage conditions References
of PMBs. An increased storage period (i.e.; more than 3 days) or
storage at higher temperatures, could lead to unexpected perfor- [1] EN 13399, Bitumen and Bituminous Binders – Determination of Storage
Stability of Modified Bitumen, 2010.
mance of the binder. A clear aging effect was shown by stiffness [2] ASTM D6930, Standard Test Method for Settlement and Storage Stability of
analysis together with polymer separation due to prolonged stor- Emulsified Asphalts, 2010.
 L. Zani et al. / Construction and Building Materials 145 (2017) 326–335
[3] NCHRP Report 459, Characterization of Modified Asphalt Binders in Superpave
Mix Design Project 9-10 FY’96, ISSN 0077-5614, ISBN 0-309-06707-3,
Transportation Research Board, National Academy Press, 2001.
[4] A. Hanyu, T. Ito, A. Kasahara, K. Saito, Effect of the morphology of SBS modified
asphalt on mechanical properties of binder and performance of mixture, J.
Pavement Eng., JSCE 9 (2004) 65–72, http://dx.doi.org/10.2208/journalpe.9.65.
[5] X. Lu, P. Redelius, H. Soenen, SBS modified bitumens: does their morphology
and storage stability influence asphalt mix perfomance?, in: The 11th ISAP
International Conference, 2010.
[6] X. Lu, U. Isacsson, Compatibility and storage stability of styrene-butadiene-
styrene copolymer modified bitumens, Mater. Struct. 30 (1997) 618–626,
http://dx.doi.org/10.1007/BF02486904.
[7] EN 1427, Bitumen and Bituminous Binders – Determination of Softening Point
– Ring and Ball Method, 2007.
[8] ASTM D36, Standard Test Method for Softening Point of Bitumen (Ring-and-
Ball Apparatus), 2014.
[9] H. Bahia, H. Zhaib, Storage stability of modified binders using the newly
developed LAST procedure, Road Mater. Pavement Des. 1 (1–2) (2000) 53–73,
http://dx.doi.org/10.1080/14680629.2000.9689884.
[10] M.N. Gonzalez, Rheology and engineering parameters of bitumen modified
with polyolefins, elastomers and reactive polymers, Schriftenreihe Kunststoff-
Forschung, Vol. 70, Univerlagtuberlin, 2010, ISBN 3798322295.
[11] H. Fu, L. Xie, D. Dou, L. Li, M. Yu, S. Yao, Storage stability and compatibility of
asphalt binder modified by SBS graft copolymer, Constr. Build. Mater. 21 (7)
(2007) 1528–1533, http://dx.doi.org/10.1016/j.conbuildmat.2006.03.008.
[12] G. Wen, Y. Zhang, Y. Zhang, K. Sun, Y. Fan, Rheological characterization of
storage-stable SBS-modified asphalts, Polym. Testing 21 (3) (2002) 295–302,
http://dx.doi.org/10.1016/S0142-9418(01)00086-1.
[13] H. Soenen, X. Lu, P. Redelius, The Morphology of SBS Modified Bitumen in
Binders and in Asphalt Mix, Advanced Testing and Characterization of
335
Bituminous Materials, CRC Press, 2009, http://dx.doi.org/10.1201/
9780203092989.ch16. Print ISBN: 978-0-415-55854-9, eBook ISBN: 978-0-
203-09298-9.
[14] EN 1426, Bitumen and Bituminous Binders – Determination of Needle
Penetration, 2007.
[15] EN 13398, Bitumen and Bituminous Binders – Determination of the Elastic
Recovery of Modified Bitumen, 2010.
[16] EN 13302, Bitumen and Bituminous Binders – Determination of Dynamic
Viscosity of Bituminous Binder Using a Rotating Spindle Apparatus, 2010.
[17] EN 14770, Bitumen and Bituminous Binders – Determination of Complex
Shear Modulus and Phase Angle. Dynamic Shear Rheometer (DSR), 2012.
[18] AASHTO T 315, Standard Method of Test for Determining the Rheological
Properties of Asphalt Binder Using a Dynamic Shear Rheometer (DSR),
American Association of State and Highway Transportation Officials, 2012.
[19] ASTM D6560, Standard Test Method for Determination of Asphaltenes
(Heptane Insolubles) in Crude Petroleum and Petroleum Products, 2012.
[20] AASHTO TP 101-12, Standard Method of Test for Estimating Fatigue Resistance
of Asphalt Binders Using the Linear Amplitude Sweep, American Association of
State and Highway Transportation Officials, 2012.
[21] T.O. Medani, M. Huurman, A.A.A. Molenaar, On the Computation of Master
Curves for Bituminous Mixes. in: Proceedings, 2004..
[22] D. Anderson, Y. Hir, M. Marasteanu, J.P. Planche, D. Martin, G. Gauthier,
Evaluation of fatigue criteria for asphalt binders, Transp. Res. Rec. (2001) 48–
56, http://dx.doi.org/10.3141/1766-07. No. 1766, Asphalt Binders.
[23] C. Hintz, R. Velasquez, C. Johnson, H. Bahia, Modification and validation of
linear amplitude sweep test for binder fatigue specification, Transp. Res. Rec.
(2011) 99–106, http://dx.doi.org/10.3141/2207-13. No. 2207, Asphalt Binders.
[24] R. Botella, F.E. Perez-Jimenez, R. Mirò, Application of a strain sweep test to
assess fatigue behavior of asphalt binders, Constr. Build. Mater. 36 (2012) 906–
912, http://dx.doi.org/10.1016/j.conbuildmat.2012.06.059.