Xuewen Zheng1, Wenyuan Xu1, Haoping Xu1, Suxin Wu1, Kai Cao2. 1. School of Civil Engineering, Northeast Forestry University, Harbin 150040, China. 2. College of Civil Engineering and Architecture, Zhejiang University, Hangzhou 310058, China.
Abstract
A real rejuvenator must have the ability to disaggregate oxidized asphaltene nanoclusters. However, few studies pay attention to the topic, and there is a lack of comparison of the disaggregation ability of different rejuvenators. Thus, the disaggregation ability and regeneration mechanism of three bio-rejuvenators (waste cooking oil (WCO), waste wood oil (WWO), and straw liquefied residue oil (SLRO)) on oxidized asphaltene nanoclusters were studied in this paper. Laboratory tests and molecular dynamics (MD) simulation were used to compare the effectiveness of the three bio-rejuvenators and reveal its corresponding mechanism. It is found that these bio-rejuvenators have a softening effect on aged asphalt binder, but not all of them can disaggregate oxidized asphaltene nanoclusters. The introduction of WWO and WCO can effectively disturb the nanoclusters caused by the increase of polar functional groups during the oxidation process. The effect of WWO is more significant, but neither of them can restore the asphaltene dispersion to the virgin asphalt binder. SLRO has an adverse effect on the disaggregation of oxidized asphaltene nanoclusters. WCO, WWO, and SLRO showed different disaggregation mechanisms, including ″pull-out, intercalation, and compression″, respectively. WCO and WWO can increase the activation energy reduced by aging in a short aging time, and SLRO makes the activation energy lower. Such findings can help enterprises screen more reasonable rejuvenators to facilitate the recycling of reclaimed asphalt pavement (RAP) materials and promote the sustainable development of the construction industry.
A real rejuvenator must have the ability to disaggregate oxidized asphaltene nanoclusters. However, few studies pay attention to the topic, and there is a lack of comparison of the disaggregation ability of different rejuvenators. Thus, the disaggregation ability and regeneration mechanism of three bio-rejuvenators (waste cooking oil (WCO), waste wood oil (WWO), and straw liquefied residue oil (SLRO)) on oxidized asphaltene nanoclusters were studied in this paper. Laboratory tests and molecular dynamics (MD) simulation were used to compare the effectiveness of the three bio-rejuvenators and reveal its corresponding mechanism. It is found that these bio-rejuvenators have a softening effect on aged asphalt binder, but not all of them can disaggregate oxidized asphaltene nanoclusters. The introduction of WWO and WCO can effectively disturb the nanoclusters caused by the increase of polar functional groups during the oxidation process. The effect of WWO is more significant, but neither of them can restore the asphaltene dispersion to the virgin asphalt binder. SLRO has an adverse effect on the disaggregation of oxidized asphaltene nanoclusters. WCO, WWO, and SLRO showed different disaggregation mechanisms, including ″pull-out, intercalation, and compression″, respectively. WCO and WWO can increase the activation energy reduced by aging in a short aging time, and SLRO makes the activation energy lower. Such findings can help enterprises screen more reasonable rejuvenators to facilitate the recycling of reclaimed asphalt pavement (RAP) materials and promote the sustainable development of the construction industry.
Reclaimed asphalt pavement
(RAP) contains significant amounts of
asphalt binder, and its application in new construction has been promoted
as a sustainable practice in the pavement industry and to promote
a circular economy. However, this effort has faced challenges such
as the aged asphalt’s inferior properties compared to the virgin
counterpart and its unknown contribution to new pavements. The aged
asphalt binder is highly oxidized during the service life. The oxidation
of asphaltene molecules is one of the critical factors contributing
to the hardening of asphalt and the consequent pavement embrittlement
that eventually leads to the deterioration of asphalt’s mechanical
properties and performance. The major consequence of the irreversible
oxidation process is mainly attributed to two factors:[1] first, the evaporation of light asphalt components and
the reduction of the maltene/asphaltene phase ratio and second, the
oxidation of highly reactive hydrocarbons and polar species that leads
to a change in the functional group composition and increased concentration
of asphalt’s polar components.[2−4] Due to the introduction
of polar functional groups in asphaltene molecules, the polar binding
forces such as hydrogen bond, van der Waals force, and Coulomb interaction
increase, resulting in further molecular agglomeration and the formation
of nano-aggregates.[5−7] The nano-aggregation has been implicated in the undesirable
high stiffness and brittleness of aged asphalt binder.[8]To facilitate the recycling and possibly upcycling
of aged asphalt
binder, there is a need to restore its properties with the help of
modifiers referred to as rejuvenators. Rejuvenators are usually divided
into two categories: petroleum based and biomass based. Petroleum-based
agents are limited in application due to the shortage of petroleum
resources.[9] However, a wide range of bio-sources
with either animal or vegetal origin can be used to produce bio-oils
as renewable rejuvenators. Such bio-rejuvenators can be obtained from
various sources such as waste cooking oil (WCO),[10] corn or corn stover,[11−13] castor,[14] wood,[15,16] cashew,[17,18] and swine
manure.[19−21] Many studies have evaluated the regeneration performance
of bio-rejuvenators from the macroscopic and microscopic perspectives,
including mechanical properties,[22,23] rheological
properties, fatigue properties, and chemical properties.[24,25] However, little attention has been paid to the deagglomeration of
oxidized asphaltene nanoclusters by bio-rejuvenators. It is proved
that the ability to properly dissolve and disperse macromolecules
and microstructures in aged asphalt binder is the key to the durability
and reaging resistance of recycled asphalt.[26,27] Actually, a true rejuvenator should be a kind of asphalt additive
that can deagglomerate large asphaltene assemblies into smaller nanoaggregates
while replenishing the lost compounds.[28] It is reported that swine manure was effective in restoring oxidized
asphalt’s thermomechanical properties as amide groups in it
interact with asphaltene molecules, increasing the stacking distance
of asphaltene dimers while changing their conformational packing.[29] From this perspective, the current mainstream
of bio-rejuvenators pays little attention to the investigation of
disaggregating oxidized asphaltene nanoclusters. And few previous
studies compare the disaggregation ability and rejuvenation mechanism
of different bio-rejuvenators. Research on the disaggregation behavior
of oxidized asphaltene nanoclusters with different rejuvenators can
help manufacturers develop and screen rejuvenators that are more suitable
for regeneration and eliminate those that are not conducive to regeneration.
Therefore, three common bio-oils used in previous studies from WCO,[30,31] straw liquefaction residue (SLRO),[32−34] and waste wood (WWO)[35−37] are selected in this paper to study their disaggregation ability
of asphaltene nanoclusters in aged asphalt binder. Computational and
experimental methods are used to examine the ability of the three
rejuvenators to disaggregate asphaltene clusters. The effect of each
bio-rejuvenator on the rheological properties of aged asphalt can
be observed through experiments. MD simulation can observe the structural
changes of the asphaltene dimer and reveal the regeneration mechanism.
In addition, the MD simulation results can be verified by experiments.
Materials and Methods
Preparation of Aged Asphalt
Binder
In this study, aged 90# asphalt binder is rejuvenated
using three
bio-rejuvenators: WCO, WWO, and SLRO. The basic properties of the
virgin asphalt binder are shown in Table .
Table 1
Basic Properties
of the Virgin Asphalt
Binder
properties
temperature
results
penetration (0.1 mm)
25 °C
82.5
ductility (cm)
15 °C
122
softening point (°C)
47.8
To prepare
the aged asphalt binder, the unaged binder was aged
initially using a rolling thin-film oven (RTFO) followed by 20 h of
aging in a pressure aging vessel (PAV). The aging in the PAV was conducted
at 2.10 MPa pressure and at 100 °C temperature for 20 h. And
the 20 h PAV aged samples were used to represent long-term aged asphalt.[38] Recycled asphalt was prepared by adding bio-rejuvenators
to aged asphalt. The mass of bio-rejuvenators was 10% of the mass
of aged asphalt. There are two reasons for choosing the 10% dosage:
first, the ability of the three bio-oils to disaggregate asphaltene
nanoclusters in aged asphalt is compared under the condition of controlling
the same dosage, and second, in previous literature researches, the
optimal dosage of the three bio-oils has been mentioned to be about
10%. Although this dosage may not be the optimal dosage of the three
bio-rejuvenators in this study, to compare the disaggregation effects
of the three rejuvenators under the same research conditions, the
dosage was set as 10%. The aged asphalt and bio-rejuvenators were
stirred in a 135 °C agitator for 15 min at a stirring rate of
300 rpm to preliminarily achieve the purpose of predispersion. The
mixture was then sheared at a high speed of 4000 rpm for 20 min to
fully disperse the bio-rejuvenators. After the preparation of the
mixture, the samples required for the test were prepared, and the
remaining mixture was put into a closed container to prevent pollution
for future use.
Dynamic Shear Rheometer
(DSR)
An
Anton Paar MCR 302 rheometer was used to measure the elastic behavior
and viscous behavior of each sample following ASTM D7175. Tests were
conducted at 10 °C with frequencies of 0.1–100 rad/s and
at a strain rate of 0.1%.[28] For this study,
an 8 mm parallel plate spindle was used. Using the corresponding storage
modulus (G′) and loss modulus (G″) results, the crossover modulus Gc (at which G′ is equal to G″) was determined. Additionally, the complex shear modulus
(G*), which is the material deformation resistance
with repeated shear, and phase angle (δ), the time lag between
stress and strain, were calculated from the measured stress and strain
data using eq .[28]in whichwhere γmax = maximum strain,
τmax = maximum stress (Pa), T =
maximum applied torque (mN·m), r = radius of
the sample (mm), θ = deflection (rotational) angle (°),
and h = height of the sample (mm).
Gas Chromatography–Mass Spectrometry
(GC–MS) Test
Gas chromatography–mass spectrometry
(GC–MS) analysis was performed using a GCMS-240 instrument
from PHARMACIA, Sweden. The chemical constituents of bio-rejuvenators
were identified by GC–MS, respectively. For gas chromatography
analysis conditions, the initial temperature of the sample was 40
°C and was maintained for 2 min. The temperature was then increased
to 290 °C at a heating rate of 4 °C/min and kept for 20
min. Helium gas was selected as the carrier gas at a flow rate of
1.2 mL/min. Under the condition of mass spectrometry, the temperature
of the ion source was 260 °C and the temperature of the transmission
line was 280 °C. The electron beam energy was 70 eV.
Molecular Dynamics (MD) Simulation
It has been documented
that the asphaltene component increases in
most cases after oxidation.[30] The increase
of the asphaltene portion is often accompanied by a decrease in the
resin and aromatic portion.[16] Besides,
due to the reaction of small molecules such as oxygen absorption polymerization
and the generation of asphaltene with stronger polarity and higher
molecular weight than the virgin asphaltene, which greatly increases
the solubility parameters of the total asphaltenes, while the solubility
parameters of soft asphalt do not change much, the difference between
the solubility parameters of asphaltenes and soft asphalt becomes
larger, and the solubility of soft asphalt to asphaltenes becomes
weaker. The main structure of asphaltene is polycyclic aromatic hydrocarbons
(PAHs), and the planar aromatic ring system has a strong π–π
force. Due to the aromaticity resulting from the delocalization of
π electrons as the responsible factor for the π–π
interaction of asphaltene molecules, an asphaltene layer accumulation
structure can be formed and the result is the formation of clusters
at the nanoscale.The continuous oxidation reaction in the long-term
aged asphalt binder is considered. More oxygen and sulfur atoms were
added to the long-term aged asphalt compared to the virgin asphalt
binder.[29] The asphaltene molecule used
in this study is the molecular structure published by Martín-Martínez
et al.,[39] which has a chemical structure
with lower energy and thus provides a more stable and probable model
for the asphaltene compared with the original molecule proposed by
Li and Greenfield.[40] To develop the model,
a system of 15 asphaltene pyrrole molecules shown as ″VirAsP″
and 15 oxidized asphaltenes[41] shown as
″OxAsP″ was separately built and optimized (Figure ).
(a) Virgin asphaltene
pyrrole (VirAsP) structure and (b) oxidized
asphaltene pyrrole (OxAsP) structure.As can be seen from Tables –4, acids,
alcohols, and phenols contributed the most to WCO, SLRO, and WWO components,
respectively. WCO had the most 9,12-octadecadienoic acid (Z,Z)- (53.4%), SLRO had the most glycerin
(10.4%), and WWO had the most 4-ethyl-2-methoxyphenol (9.8%). In this
paper, the molecular model with the highest content of three kinds
of bio-oils was used as the source oil model when constructing the
bio-rejuvenator MD models. Molecular models of the three bio-rejuvenators
are shown in Figure .
Table 2
The Main Individual Compounds of WCO
Identified by GC–MS
WCO
formula
area (%)
1
9,12-octadecadienoic acid
(Z,Z)-
C18H32O2
53.4
2
oleic acid
C18H34O2
22.9
3
n-hexadecanoic acid
C16H32O2
9.4
4
7-hexadecyn-1-ol
C16H30O
1.9
5
9,17-octadecadienal, (Z)-
C18H32O
1.8
6
1,5,9-undecatriene, 2,6,10-trimethyl-,
(Z)-
C14H24
1.8
7
dibutyl phthalate
C16H22O4
1.7
8
6-octadecenoic acid, methyl ester, (Z)-
C19H36O2
1.6
9
3,4-octadiene, 7-methyl-
C9H16
1.6
10
cyclopropaneoctanal, 2-octyl-
C19H36O
1.4
97.4%
Table 4
The Main Individual Compounds of WWO
Identified by GC–MS
Molecular models of the bio-rejuvenators.MD simulation
was used to investigate the stacking and aggregation
pattern of each model in the presence of the bio-rejuvenator molecules.
To do so, each subset of bio-rejuvenators molecules was added to a
cell of oxidized asphaltenes at a 10% weight ratio and was solvated
in the n-heptane solvent.[20] Material Studio was used to carry out the simulation in this study.
For aged asphalt and virgin asphalt, when the molecular systems were
established, the systems’ energy was minimized by the smart
method. Then, the systems were run at 800 K for 100 ps in an NVT (constant
value for the number of particles, volume, and temperature) followed
by an NPT ensemble (constant value for the number of particles, pressure,
and temperature) at a pressure of 200 atm and 800 K for 500 ps to
shake the system. This will prevent the simulation from sinking into
a local minimum energy state. The next procedure began with an NVT
ensemble at 408.15 K for 2 ns to reach equilibrium and correct any
probable overlaps of molecules. Following an NPT ensemble, the systems
were run for 2 ns at a pressure of 1 atm. After the equilibration
of oxidized asphaltenes in heptane, the rejuvenator molecules were
added to the system. Then, the systems were run at 408.15 K to simulate
the mixing temperature of the binder and aggregates in asphalt plants
for a typical bio-modified asphalt mix.[20] Although the selected temperature was higher than the boiling point
of heptane, the authors deliberately chose it to dilute the solvent
enough to reduce its effect on asphaltene aggregation.[41] Then, the systems continued to run in a two-step
procedure: a canonical NVT ensemble for 2 ns followed by an isothermal–isobaric
NPT ensemble for 2 ns to reach equilibrium. The number of asphaltenes
aggregates as well as the average size of aggregates was calculated
for each scenario after the systems reached equilibrium.To
have a quantitative measure for the aggregation of asphaltenes,
two indices were defined: the number of aggregates and the average
size of aggregates. An aggregate was considered as two or more parallel
(or near-parallel) sheets being at a distance of less than 4 Å
while having at least 50% overlap.[20,42] After the
number of aggregates was determined, the average number of aggregates
was calculated using the following formula:[20,42]where N is the number of aggregates
containing m asphaltene molecules.
Results and Discussion
MD Simulation Results
Because aging
will increase the proportion of asphaltene, asphaltene tends to aggregate
and form nanoclusters under the influence of the π–π
bond. This paper first analyzes the distribution of asphaltene molecules
in the virgin asphalt binder and aged asphalt binder and then analyzes
the distribution of OxAsp in the presence of bio-rejuvenator molecules
so as to analyze the disaggregation ability of the three bio-rejuvenators
on OxAsp nanoclusters. To see the relative position of asphaltenes
more clearly, after the simulation was completed, the bio-rejuvenator
molecules and n-heptane molecules were removed, and
the side chain of asphaltene molecules was cut off.Figure shows the distribution
state of asphaltenes after the simulation. To qualitatively describe
the interaction degree between asphaltenes, the asphaltenes’
aggregation in the system is covered with a red shadow. When the red
shadow area is larger and the quantity is fewer, it indicates that
the asphaltenes are more aggregated. Comparing Figure a,b, it can be clearly seen that VirAsp in
the virgin asphalt binder system is distributed in an island shape,
the number of red shadows is scattered, and the area is smaller. Most
areas are below 100 Å2, and the largest is about 176
Å2. In contrast, in the aged asphalt binder system,
the red shadow area is larger and the number is fewer. The areas are
up to more than 300 Å2, indicating that the VirAsp
molecules are evenly distributed, and the interaction between them
is weak. Thus, large-area aggregation is not formed. Meanwhile, the
OxAsp molecules in the aged asphalt binder are concentrated, indicating
that the interaction between them is more intense, and nanoclusters
are formed.
Figure 3
Asphaltene distribution state: (a) virgin asphalt and (b) aged
asphalt.
Asphaltene distribution state: (a) virgin asphalt and (b) aged
asphalt.Figure a–c
respectively shows the distribution of OxAsp in the presence of bio-rejuvenators.
Among the three systems, the OxAsp distribution with WWO is the most
dispersed, presenting a multipoint dispersion state. However, compared
with the virgin asphalt, there is still a large area of red shadow
in the system. The value is basically the same as that of aged asphalt,
with an area of more than 300 Å2, indicating that
there is still a relatively dense asphaltene aggregation in the system.
WCO followed with good dispersion, and the worst is the asphalt binder
system with SLRO. Similarly, the maximum shadow area in the two systems
is almost the same as that of aged asphalt. However, after the addition
of SLRO, the OxAsp interaction almost has no difference from that
of the aged asphalt system, indicating that SLRO has a limited ability
to disaggregate OxAsp nanoclusters.
Figure 4
Asphaltene distribution state: (a) WCO,
(b) WWO, and (c) SLRO.
Asphaltene distribution state: (a) WCO,
(b) WWO, and (c) SLRO.To see the distribution
of asphaltene more clearly, the snapshots
after the simulation are shown in Figures and 6. Figure a shows that due to the interaction
between VirAsp molecules, some VirAsp molecules are stacked in parallel,
which is caused by the π–π interaction and hydrogen
bonding caused by the structural characteristics of the large aromatic
ring system and heteroatoms of asphaltene molecules.[43] The stacking structure is more dense in Figure b. The OxAsp molecular sheet
is changed from the original two- or three-sheet stacking to the stacking
form of three or more sheets, and the maximum number of stacked sheets
can reach five. It is reported the π–π interaction
between asphaltene molecules reaches the maximum when they are parallel
to each other,[44] indicating that the distribution
is more concentrated than that in the virgin asphalt binder. This
is due to the increase of oxygen-containing polar functional groups
after asphalt aging, which promotes the interaction of asphaltene
and forms nano-aggregation. Figure shows the OxAsp stacking structure after adding bio-rejuvenators.
In the presence of WCO and SLRO, there is a common focus; that is,
the maximum number of oxidized asphaltene stacks is five, which is
consistent with the aged asphalt, indicating that the two bio-rejuvenators
have a weak disaggregation ability for larger oxidized asphaltene
nanoclusters. WWO can split the five-sheet stacking into two- and
three-sheet stacking, showing a strong ability to disaggregate larger
oxidized asphaltene nanoclusters. In Figure a, WCO divides the three-sheet stacking in
the system into two- and one-sheet stacking, which alleviates the
aggregation of OxAsp. In Figure b, in addition to decomposing the five-sheet stacking,
there are three three-sheet stacking structures in the system, and
the number of stacking is higher than that of virgin asphalt. In the
presence of SLRO (Figure c), the stacking structures in the system have hardly changed,
indicating that SLRO is hardly effective in alleviating oxidized asphaltene
accumulation.
Figure 5
Asphaltene stacking form: (a) virgin asphalt and (b) aged
asphalt.
Figure 6
OxAsp stacking form: (a) WCO, (b) WWO, and (c)
SLRO.
Asphaltene stacking form: (a) virgin asphalt and (b) aged
asphalt.OxAsp stacking form: (a) WCO, (b) WWO, and (c)
SLRO.To quantitatively describe the
aggregation degree of asphaltene
in the MD model system, the radial distribution function (RDF) between
asphaltenes is introduced for evaluation, as shown in Figure . RDF is an important method
to analyze the internal structure of organic polymer materials. The
basic principle of RDF is to calculate the probability of the occurrence
of other molecules around the reference molecule. As can be seen from Figure , the RDF of all
asphalt systems reaches the maximum value in the range of 0–3
Å, and the number of peaks in this range is the same, indicating
that the relative distribution of oxidized asphaltene in all asphalt
systems is basically unchanged. However, the highest RDF values of
all systems are different, indicating that the aggregation quantity
of oxidized asphaltene is different. The aggregation of oxidized asphaltene
is more serious in the system with a high peak value. The peak values
of RDF in both the aged asphalt binder and SLRO asphalt binder are
above 80. WWO and WCO can reduce the peak values, and WWO reduces
the peak values to about 70, which is the most significant. It shows
that both WCO and WWO can alleviate the agglomeration of OxAsp. The
effect of WWO is better, and SLRO has the worst effect.
Figure 7
RDF between
asphaltenes: (a) OxAsp, (b) WCO, (c) WWO, and (d) SLRO.
RDF between
asphaltenes: (a) OxAsp, (b) WCO, (c) WWO, and (d) SLRO.To further quantitatively evaluate the disaggregation ability
of
the three bio-rejuvenators on oxidized asphaltene nanoclusters, the
average aggregation number of asphaltenes in each system is calculated
according to formula . The high dispersion of asphaltenes can be characterized by the
small average aggregate number. It can be seen from Figure that the average aggregation
number in VirAsp is 2.33, while that of OxAsp is 3.33, which is increased
by 42.9% compared with that of VirAsp. It indicates that asphaltenes
are evenly distributed before asphalt binder aging. After aging, due
to the action of polar functional groups, the original asphaltene
slices aggregate into larger asphaltene clusters. After WCO and WWO
are added, the OxAsp average aggregate numbers are 3 and 2.5, respectively,
which decreased by 9.91 and 24.92% compared with aged asphalt binder,
indicating that WCO and WWO could decompose oxidized asphaltene clusters
into smaller aggregation. The addition of SLRO does not change the
average aggregation amount of OxAsp, which is equivalent to that of
the aged asphalt system, indicating that SLRO has no ability to disaggregate
oxidized asphaltene clusters. The ability to disaggregate oxidized
asphaltene clusters is WWO > WCO > SLRO.
Figure 8
Average aggregate number
of asphaltene slices.
Average aggregate number
of asphaltene slices.
DSR Test
Results
To verify the accuracy
of MD simulation, DSR test results were used for verification in this
section. In fact, when describing the molecular weight of a polymer,
previous studies prefer to use the GPC test for evaluation. However,
according to the report of Wang et al., the GPC test has two defects:[45] First, because GPC samples need to be filtered
before entering the column, some large microstructures formed during
aging may not pass through the filter paper and will not be counted.
However, the large microstructure is the key to evaluate whether the
rejuvenators can effectively dissolve the aged products. Second, the
microstructure of GPC is affected by the solvation ability of the
solvent. Therefore, even through the microstructure of GPC filter,
due to the different solvation capacity of maltenes and GPC solvents,
their size may be different from that existing in the asphalt binder.
In this paper, the frequency scanning mode was used to evaluate the
cross modulus, which is the corresponding value at which shear storage
modulus (G′) and shear loss modulus (G″) have the same magnitude. The crossover point
is defined by two symbols of the crossover frequency and the crossover
modulus. At this point, there is a transition from liquid to solid
materials. The polydispersity of polymers and bitumen has been reported
to be associated with the cross modulus.[46] Polydispersity index (PDI) in rheology is usually defined as 100,000/cross
modulus, where the cross modulus is expressed in pascals. Although
the molecular structure of asphalt differs from that of polymers,
its asphaltene portion (heptane insoluble) is the focus of this paper
and consists of self-assembled nanoaggregates with a molecular weight
of approximately 980 g/mol and size of few nanometers.[21,47] PDI is also associated with the point at which sol–gel transfer
occurs in asphaltene-rich bitumen.[48,49] Farrar et
al. used the Christensen and Anderson model for bitumen, referred
to as the dispersed polar fluid model, for utilizing the crossover
modulus as a parameter related to the asphalt structure and its evolution
with aging. They found an approximate linear relationship between
oxygen uptake and the logarithmic reciprocal of the crossover modulus.[46] Liu et al. studied the evolution of asphalt
during oxidative aging and used the change of crossover value as an
index of the aging process. They found that, with the increase of
aging time, the crossover modulus of virgin asphalt and modified asphalt
continued to decline.[50] As aging progresses,
the number of polar functional groups in bitumen increases, leading
to intermolecular interactions that lead to asphaltene aggregation.
This, in turn, promotes the formation of nanoaggregates of different
sizes, as molecular affinity overlays the oxidized asphaltene hydrophilic
side chains that form during aging. The different nano-aggregation
will increase the polydispersity of the asphalt matrix. Previous studies
show that almost all rejuvenators can increase the cross frequency
of aged asphalt, affect the viscous composition of aged asphalt, and
play a role in softening aged asphalt binder, but not all rejuvenators
can increase the cross modulus of aged asphalt binder,[25] which is an indicator to measure the degree
of polydispersity.[51−53] The polydispersity of asphaltenes is reported to
increase with aging, and the polydispersity is inversely proportional
to the cross modulus.[28,51] Thus, the appropriate rejuvenators
to disaggregate or disperse asphaltene clusters will increase the
cross modulus of aged asphalt binder.Through the DSR test results,
the cross modulus and cross frequency of aged asphalt binder in the
presence of the three bio-rejuvenators are compared to analyze the
dispersion ability of the three bio-rejuvenators on oxidized asphaltene
nanoclusters. Figure shows that aging increases the composite shear modulus and reduces
the range of the phase angle. Except for SLRO, the other two kinds
of bio-rejuvenators can reduce the modulus value and enlarge the phase
angle threshold, indicating that WWO and WCO can improve the rheological
properties of aged asphalt. However, SLRO exerts poor regeneration
performance for making the modulus and phase angle equal to those
of aged asphalt. When the storage modulus G′
and the loss modulus G″ are equal, the corresponding
modulus is the cross modulus. Figure shows the cross modulus and cross frequency of several
asphalt binders. The results show that compared with virgin asphalt,
the cross modulus and cross frequency of aged asphalt are reduced
by 58.5 and 96%, respectively. Three bio-rejuvenators can increase
the cross frequency of aged asphalt. The most obvious is WCO followed
by WWO and finally SLRO. It is worth noting that the cross frequency
of aged asphalt with WCO is much higher than that of virgin asphalt
binder, but the cross frequency value after adding WWO and SLRO cannot
reach the value of virgin asphalt, indicating that WCO has an excellent
softening effect. However, not all bio-rejuvenators can increase the
cross modulus. WWO recovers the cross modulus by about 92.3% followed
by WCO, and SLRO is the worst in improving the cross modulus. However,
even WWO cannot restore the cross modulus of aged asphalt. For example,
SLRO even reduces the cross modulus. The above research conclusions
show that WWO is more talented in disaggregating asphaltene clusters
and WCO is more suitable as a softener of aged asphalt. The DSR test
results are consistent with the MD simulation results, which verify
the accuracy of the MD simulation results.
Figure 9
Complex shear modulus G* and phase angle δ.
Figure 10
Crossover
modulus and frequency.
Complex shear modulus G* and phase angle δ.Crossover
modulus and frequency.
Effect
of Bio-rejuvenators on the Micromorphology
of Oxidized Asphaltene
To understand the regeneration mechanism
of three kinds of bio-rejuvenators on aged asphalt, the micromorphology
of oxidized asphaltene in the same position in their respective systems
in the presence or absence of bio-rejuvenators is studied by MD simulation. Figures –13 show the micromorphology of
oxidized asphaltenes under different conditions. Oxidized asphaltenes
mainly have the following stacking structures: face-to-face stacking,
offset face-to-face stacking, T-shaped stacking, and other aggregation
phenomena, which are basically consistent with the previous research
conclusions.[29,41,43,54] It is reported that aging increases the
polar functional groups, which increases the strength of mutual attraction
between molecules and shortens the distance between PAH structures
and makes it easier to stack.
Figure 11
Effect of WCO on the microstructure of
OxAsp: (a) offset face-to-face
stacking and (b) T-shaped stacking. Note: green molecules for WCO.
Figure 13
Effect of SLRO on the microstructure of OxAsp: (a) offset
face-to-face
stacking and (b) T-shaped stacking. Note: purple molecules for SLRO.
Effect of WCO on the microstructure of
OxAsp: (a) offset face-to-face
stacking and (b) T-shaped stacking. Note: green molecules for WCO.Effect of WWO on the microstructure of OxAsp: (a) face-to-face
stacking and (b) T-shaped stacking. Note: brown molecules for WWO.Effect of SLRO on the microstructure of OxAsp: (a) offset
face-to-face
stacking and (b) T-shaped stacking. Note: purple molecules for SLRO.Figure shows
the stacking morphology of OxAsp before and after adding WCO. For
the aged asphalt binder, the molecular centroid distances of asphaltenes
in offset face-to-face stacking (Figure a, left side) is 6.745 Å. In T-shaped
stacking (Figure b, left side), the centroid distance between the two OxAsp molecules
is 9.535 Å with an angle of 46.657°. WCO molecules play
a ″pull-out″ role between OxAsp molecules in offset
face-to-face (Figure a, right side) and T-shaped stacking (Figure b, right side). When the WCO molecules approach
the asphaltene dimer, OxAsp molecules move in the opposite direction,
increasing the centroid distance from 6.745 to 6.832 Å. In T-shaped
stacking, the centroid distance and included angle of the two OxAsp
molecules increase from 9.535 Å and 46.657° to 10.137 Å
and 55.991°, respectively. It can be seen that the disaggregation
ability of WCO to OxAsp molecules in offset face-to-face stacking
is not as strong as that in T- shaped stacking. This may be due to
the fact that the OxAsp interactions are stronger in offset face-to-face
stacking structures compared to T-shaped stacking.WWO molecules
show an ″intercalation″ effect in OxAsp
aggregates. The ″intercalation″ effect of WWO is achieved
by interfering with the π–π interaction between
aromatic nuclei in OxAsp molecules. WWO separates aromatic nuclei
from each other to increase interstitial distance. In Figure a (right side), WWO is embedded
in OxAsp aggregates, increasing the gap between the two PAH planes.
In face-to-face stacking, the ″intercalation″ regeneration
mode reduces the agglomeration of OxAsp molecules, and the distance
of the OxAsp centroid increases from 6.724 to 8.384 Å. For T-shaped
stacking (Figure b, right side), when WWO is close to PAHs, the strong repulsion causes
the upper OxAsp molecule to move away from the WWO molecule along
the direction perpendicular to the plane of the WWO molecule. At the
same time, affected by the attraction of the lower OxAsp, the upper
OxAsp tends to approach downward, but affected by the repulsion of
another WWO molecule, the two OxAsp molecules cannot approach. This
explains why the centroid distance and included angle of two OxAsp
molecules range from 9.535 to 11.166 Å and 46.657 to 41.529°,
respectively.
Figure 12
Effect of WWO on the microstructure of OxAsp: (a) face-to-face
stacking and (b) T-shaped stacking. Note: brown molecules for WWO.
Figure shows
that when SLRO is added to the aged asphalt binder, the aggregation
of OxAsp is more serious in offset face-to-face stacking. In this
scenario, SLRO molecules are basically at the periphery of OxAsp aggregates
and show a ″compression″ effect. In the offset face-to-face
stacking (Figure a), the centroid distances of OxAsp molecules decrease from 7.840
to 7.200 Å, which means that when SLRO is close to OxAsp aggregates,
there is a strong repulsion force between SLRO molecules and OxAsp
molecules, which promotes the closer distance of OxAsp molecules.
However, in T-shaped stacking (Figure b), SLRO exerts a ″pull-out″
effect, and the centroid distance and included angle slightly increase
from 9.535 Å and 46.657° to 9.775 Å and 82.090°,
respectively. This means that SLRO has the ability to slightly alleviate
the T-shaped aggregation of OxAsp molecules, but this ability is weak
and only works in the state of T-shaped accumulation. However, there
is less T-shaped accumulation in aged asphalt, so the regeneration
ability of SLRO is poor.
Interaction between the
OxAsp Molecules in
the Presence of Bio-rejuvenators
To understand how bio-rejuvenators
reduce the negative effects of aging on molecular conformation, especially
on oxidized asphaltene aggregates, the interactions between oxidized
asphaltene and the main components of bio-rejuvenators were studied. Figure shows the intermolecular
binding energy, the binding distance (d), and its
variations (Δd) between OxAsp molecules in
the presence and absence of bio-rejuvenators. The results show that
SLRO enhances the force between oxidized asphaltenes and reduces the
binding distance between them in offset face-to-face stacking. However,
SLRO slightly increases the binding distance while enhancing the force
between oxidized asphaltenes. WCO and WWO can weaken the forces between
the oxidized asphaltenes and increase the binding distance. Comparing
the same kind of stacking, for example, in T-shaped stacking, affected
by the ″intercalation″ effect, the force between oxidized
asphaltenes is the weakest, and the binding distance is the largest
in the presence of WWO, indicating that WWO has the best disaggregation
ability for oxidized asphaltene aggregates followed by WCO. This is
consistent with the evaluation results of the DSR test.
Figure 14
Interactions between OxAsp
molecules in the presence of bio-rejuvenators.
Interactions between OxAsp
molecules in the presence of bio-rejuvenators.
Active Energy Results
The concept
of activation energy was first proposed by Eyring in 1936 and is mainly
used to describe flow barriers.[55] Based
on the correlation between the activation energy and the G*/sin δ parameter, asphalt binders with higher activation energy
have greater viscosity at higher temperatures than asphalt binders
with lower activation energy. The activation energy of asphalt binders
containing recycled asphalt binders was investigated by Jamshidi et
al.[56] The results show that compared to
neat asphalt binders, recycled asphalt binders (which are highly aged)
require more energy to overcome intermolecular resistance. In this
paper, the effects of three bio-rejuvenators on the activation energy
of aged asphalt are studied. The activation energy can be calculated
based on the Arrhenius equation (eq ).where ηa =
viscosity (Pa · s), A = pre-exponential
factor (dimensionless constant), Eη = active energy (kJ/mol), R = gas constant [8.314
J/(mol·K)], and T = temperature (K).Figure shows
the activation energy of virgin asphalt, aged asphalt, and aged asphalt
after adding bio-rejuvenators. As can be seen from Figure , different from previous
studies, the activation energy of aged asphalt does not increase.
In this paper, the activation energy of aged asphalt is significantly
lower than that of virgin asphalt, and the activation energy of virgin
asphalt is the highest among several asphalts. The activation energy
of aged asphalt after adding three bio-rejuvenators is also lower
than that of virgin asphalt. Compared with the asphalt with bio-rejuvenators,
the activation energy of the asphalt with SLRO is the smallest, even
lower than that of the aged asphalt, while WWO and WCO can increase
the activation energy of the aged asphalt by basically the same range,
and the improvement degree of WCO is slightly higher than that of
WWO. It should be noted that the decrease of aromatics and resins
during aging and the subsequent increase of asphaltene are related
to the increase of the polarity of the former substance so that they
become insoluble enough to be separated in heptane. This in turn
leads to an increase in the percentage of total weight of so-called
asphaltenes in aged bitumen. However, the average molecular weight
of these new classes is lower than that of the original asphaltenes,
suggesting that some resins and aromatics are added to asphaltenes
due to aging.[57] Generally, the increased
polarity promotes intermolecular interactions and the formation of
asphaltene nano-aggregates, which can easily shear against each other
and may cause a reduction in activation energy at a low aging time.[12] This can explain the reduction in the activation
energy of aged asphalt. It further shows that WCO and WWO can effectively
alleviate the reduction of activation energy and alleviate the aggregation
of oxidized asphaltene, while SLRO has no effect.
Figure 15
Active energy results.
Active energy results.
Conclusions
Asphalt
aging is inevitable, which changes the chemical composition
and molecular structure of the asphalt binder and promotes the formation
of asphaltene clusters. The disaggregation of oxidized asphaltene
nanoclusters by three bio-rejuvenators (WCO, WWO, and SLRO) is studied.
MD simulation and DSR test are used to check whether the rejuvenators
could disaggregate asphaltene nanoclusters, and the regeneration mechanism
is analyzed. The conclusions are as follows:MD simulation shows that WWO significantly
reduces the aggregation size of oxidized asphaltenes by increasing
the dispersion of oxidized asphaltenes. WCO is not as capable as WWO
in this aspect, while SLRO not only fails to solve oxidized asphaltene
nanoclusters but also makes the aggregation of oxidized asphaltenes
more serious.According
to the results of rheological
properties, such as cross modulus and cross frequency, SLRO can only
soften aged asphalt binder, while WWO and WCO activate the viscous
and elastic components of aged asphalt binder, which is manifested
by an increase in the cross modulus and cross frequency. WWO increases
the crossover modulus more significantly, while WCO increases the
crossover frequency more obviously.During the process of disaggregating
oxidized asphaltene clusters, WCO shows a ″pull-out″
effect. This effect is more effective in alleviating T-shaped stacking
and general in offset face-to-face stacking. WWO exerts an ″intercalation″
effect in the process of oxidized asphaltene nanocluster disaggregation.
There is a strong repulsive force between WWO and oxidized asphaltene
molecules, which causes the oxidized asphaltene molecules to move
backward and increase the distance between PAHs so as to achieve the
effect of oxidized asphaltene nanocluster disaggregation. SLRO shows
a ″compression″ effect in offset face-to-face stacking
and relatively weak ″pull-out″ effect in T-shaped stacking.
The ″compression″ effect makes the distance between
oxidized asphaltenes closer.After aging, the polar components of
asphalt increase, which make it easier to shear between aggregated
particles in a short time of aging and reduce the activation energy
of asphalt. Both WCO and WWO can improve the activation energy of
aged asphalt, but SLRO makes the activation energy of aged asphalt
lower, indicating that WCO and WWO can alleviate the aggregation of
asphaltene to a certain extent.The research in this paper is based
on the n-heptane environment. Although it does not
affect the reliability of the results, it is different from the real
asphalt system. The disaggregation effect of rejuvenators on oxidized
asphaltene nanoclusters in the real asphalt system should be further
studied.
Table 3
The Main Individual
Compounds of SLRO
Identified by GC–MS
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7
Figure 8
Figure 9
Figure 10
Figure 11
Figure 13
Figure 12
Figure 14
Figure 15