Study on the Properties and Morphology of Nano-ZnO Modified Asphalt Based on Molecular Dynamics and Experiments

Abstract

Plenty of research has verified that nano-ZnO particles could improve the properties of asphalt, but studies on nano-ZnO-modified asphalt haven’t been conducted at the molecular level. Therefore, to investigate the effect of ZnO particles on the properties, structure and morphology of asphalt, the molecular dynamics (MD) methods were carried out. In this study, the models of asphalt, ZnO cluster and ZnO/asphalt blending systems with different particle sizes were built using Materials Studio software. Then, the interaction energies of ZnO/asphalt blending systems under different temperatures were calculated, and the effect of ZnO particles on the modulus and glass transition temperature of matrix asphalt was simulated. The results indicated that the bulk modulus of asphalt increased by ZnO with particle size at 4 Å, 6 Å, 8 Å and 10 Å increased by 15.09%, 12.46%, 10.06% and 8.51%, respectively, which can illustrate that the shear resistance ability and low-temperature properties of asphalt were enhanced. Compared with matrix asphalt, the glass transition temperature of the ZnO/asphalt system decreased by less than 0.1 K, indicating that ZnO’s effect on the low temperature of asphalt was not apparent. With the increase of ZnO particle size, the diffusion coefficient decreased sharply. Compared to matrix asphalt, when the particle size increased to 8 Å and 10 Å, the diffusion coefficient decreased by 13% and 22%, respectively. So, in practice projects, to achieve better dispersion of particle materials in base bitumen, a smaller particle size would be recommended. The results of the radial distribution function (RDF) and AFM simulation indicated that ZnO particles changed the micro-structure of asphalt and increased the roughness of the asphalt surface. As a result, ZnO particles bring matrix asphalt better physical properties.

1. Introduction

Asphalt, the key material of flexible pavement, is used as the binder to mix with the aggregate forming the asphalt mixture and is paved on the surface layer of asphalt pavement. During the decades of service time, the asphalt would suffer from repeat vehicle load, temperature changes and water erosion, and exposure to the sun for a long time. To pursue an excellent performance of asphalt pavement, plenty of studies have been focused on the improvement of the properties of asphalt. The polymers, including styrene-butadiene-styrene (SBS), styrene-butadiene-rubber, and polyethene, have been effective modifiers to improve the stability under high temperatures and the cracking resistance ability under low temperatures of asphalt, especially the styrene-butadiene-styrene material has been widely used in the construction of a highway in China for decades [1,2,3,4,5]. In recent years, the waste polymers such as waste plastic bottles and waste tires have been used as a sustainable additive to improve asphalt properties and pave the trial road [6]. Nano-particle materials with high surface energy and a large fraction of surface atoms also attracted the researcher’s attention, and nano-clay, nano-silica, carbon nanotubes, nano-ZnO and nano-TiO₂ were used to modify base asphalt. The research showed that these nano-particle materials positively affected base asphalt adhesion, temperature sensitivity, and aging resistance [7,8,9,10,11,12,13]. To achieve a better performance of asphalt, nano-particle and polymer composite modification is another approach to improve the properties of asphalt. The ZnO/SBS, carbon nano-tubes/SBS, crumb rubber/SBS, nano-organic palyorskite/SBS et al. modification methods were conducted on asphalt, and the effect on the properties of asphalt had been verified by laboratory tests [14,15,16,17,18,19]. Chen investigated the impact of nano-ZnO on modified asphalt’s adhesion characteristics by preparing mixtures with different modifier contents, and the results showed that nano-ZnO enhanced the moisture sensitivity and impacted the tensile strength and cracks resistance of asphalt mixtures [20]. Sina considered the influence of acidic contaminants on asphalt pavement, prepared porous asphalt mixtures with nano-ZnO and nano-silica and conducted mechanical performance tests on mixtures. The results showed that nano-particles could remove the acidic contaminants from water and prolong the fatigue lives of porous asphalt mixture [21].

With the development of the study on modified asphalt, the research methods on the modification of asphalt have experienced great changes based on the review of related literature, and more and more researchers try to investigate the micro-structure of asphalt and the modification mechanism using different methods. By scanning electron microscopy, Liu and Wang observed the morphology of activated carbon powder-modified asphalt and the infiltration interface between carbon powder and asphalt [22]. Ji and Xu studied the interaction between asphalt and Sasobit using infrared spectra. They checked the molecular weight distributions of Sasobit-modified asphalt using a gel permeation chromatograph, and the correlation between microstructure and properties of asphalt was analyzed qualitatively [23]. Sun and Xu used dynamic Fourier transform infrared spectroscopy to analyze the functional group compositions of nano polyamide-modified asphalt. They reported the thermophysical properties of polyamide-modified asphalt according to thermogravimetric analysis [24]. Li and Zeng tested the micro-structure of styrene-butadiene-styrene-modified asphalt using atomic force microscopy [25]. Lv and Yao carried out the Fourier transform infrared spectroscopy to analyze the modification mechanism of Button rock-modified asphalt [26,27]. Obviously, in recent years, the microstructure and morphology of modified asphalt were investigated and captured by microscope techniques, and the effect of modifiers on the structure of base asphalt and the dispersion of nano-particles material or polymer modifiers in base asphalt were studied based on the microscope results.

Molecular dynamics simulations are another important approach to study the properties and structure of materials. In road engineering, the molecular dynamics method has been conducted as an effective way to study the performance of asphalt and select the appropriate modifiers to achieve the desired properties. Yao and Dai built the crystal quartz model, asphalt model and interface model of two materials. They determined the fundamental factors of moisture damage in asphalt mixture by the MD method [28]. Hu and Yu calculated the mean square displacement of SBS in matrix asphalt and the binding energies of the asphalt system and analyzed the interaction mechanism between modifier and asphalt [29]. Yu and Hu conducted MD on carbon nanotube/recycled polyethylene-modified asphalt. The radial distribution function, glass transition temperature and binding energies of asphalt systems were calculated, and it concluded that carbon nanotubes weaken the repulsion of recycled polyethylene with resin in asphalt. The storage stability of recycled polyethylene-modified asphalt was enhanced by carbon nanotubes [30]. Guo and Zhang studied the variation of the agglomeration of rubber asphalt by simulation the radial distribution function of asphalt molecules and revealed the interaction mechanism between asphalt and rubber [31]. To understand the effects of modifiers on the properties of matrix asphalt, we have conducted laboratory tests to study the influence of modifiers on the high and low-temperature properties of asphalt, and the changes of SBS on the microstructure and morphology of base asphalt were studied using MD method [32,33,34].

In this study, we try to reveal the effect of ZnO on the physical properties, microstructure, and morphology of matrix asphalt by molecular dynamics method and laboratory experiments. The asphalt molecule model, ZnO cluster model and ZnO/asphalt blending system were built. Then the interaction between ZnO and asphalt and the influence of ZnO on the physical properties of asphalt were investigated. Then combined with the atomic force microscope technique, the morphology and micro-structure of asphalt were studied.

2. Materials

2.1. Asphalt

In this study, SK-70 matrix asphalt imported from SK HOLDINGS which located at Seoul of South Korea was selected to conduct the laboratory tests and establish the asphalt model. The basic physical properties of SK-70 are shown in

Table 1. Properties of matrix asphalt.

Properties		Testing Results
Penetration (25°C, 100 g, 5 s) (0.1 mm)		74.4
Softening point (°C)		49.8
Ductility	5 cm/min, 10 °C	68.3
Thin film oven test (TFOT)	Mass loss (%)	0.64
	Penetration ratio (%)	72.7
	Ductility (5 cm/min, 10 °C) (cm)	9.6

.

2.2. Nano-ZnO

Nano-ZnO is of typical nano-particle material with a large surface area, high catalytic activity, and the ability to absorb ultraviolet light. Considering its cheap market price, nano-ZnO has been widely used in developing high physical property materials, and our previous study has verified that nano-ZnO can improve the high-temperature properties of asphalt binders. Nano-ZnO used in this study was brought from Shaanxi Sino Academy Nano Materials Co., Ltd. The properties are shown in

Table 2. Properties of nano-ZnO.

Items	Values
SSA (m2/g)	>35
Purity (%)	>98
Morphology	Spherical
Color	Yellowish
Density (g/m3)	5.642
Melting point (°C)	1975
Mohs hardness	4.5
Space group	P63mc (186)
The binding energy of exciton (mev)	60
Thermal conductivity (W/(cm·K))	1.16 ± 0.08
Resistivity(Ω·cm)	1012

, and its image was captured by scanning electron microscope and was given in Figure 1.

3. Methodology

The objective of this study was to try to investigate the interaction between ZnO particles and asphalt and to reveal the effect of ZnO on the structure and physical properties of asphalt. First, we will build the models of asphalt, ZnO cluster and ZnO/asphalt blending systems, then, the MD of these models were conducted using the Forcite tool of Materials Studio software, and the simulation tasks were given as follows.

• The interaction energy between ZnO and asphalt

In the Forcite tool of Materials Studio software, when the molecular dynamics of the model system was conducted, the system’s van der Waals energy and electrostatics energy was calculated directly. Therefore, in this study, three interaction energies between ZnO (particle size at 4 Å, 6 Å, 8 Å and 10 Å)and asphalt under five temperatures, including 383.15 K, 393.15 K, 403.15 K, 413.15 K, 423.15 K, 433.15 K, 443.15 K and 453.15 K would be obtained.

2.

Physical properties of asphalt

The mechanical properties calculation can be performed on either a single structure or a series of structures in the Forcite module. For each configuration structure of the asphalt model, several strains were applied. As a result, the elastic stiffness matrix (C) and elastic compliance matrix (S) were calculated. The matrix of C and S were given as follows. According to Equations (1) and (2), Lame’s constants ($\lambda$ and $\mu$) were obtained, and then the elastic modulus of the system can be calculated by Equation (3). The bulk modulus and shear modulus were calculated by Equations (4)–(9) and the results of mechanical properties [35].

$$C=\left[\begin{array}{cccccc}{c}_{11}& c_{12}& c_{13}& c_{14}& c_{15}& c_{16}\\ c_{21}& c_{22}& c_{23}& c_{24}& c_{25}& c_{26}\\ c_{31}& c_{32}& c_{33}& c_{34}& c_{35}& c_{36}\\ c_{41}& c_{42}& c_{43}& c_{44}& c_{45}& c_{46}\\ c_{51}& c_{52}& c_{53}& c_{54}& c_{55}& c_{56}\\ c_{61}& c_{62}& c_{63}& c_{64}& c_{65}& c_{66}\end{array}\right]S=\left[\begin{array}{cccccc}{s}_{11}& s_{12}& s_{13}& s_{14}& s_{15}& s_{16}\\ s_{21}& s_{22}& s_{23}& s_{24}& s_{25}& s_{26}\\ s_{31}& s_{32}& s_{33}& s_{34}& s_{35}& s_{36}\\ s_{41}& s_{42}& s_{43}& s_{44}& s_{45}& s_{46}\\ s_{51}& s_{52}& s_{53}& s_{54}& s_{55}& s_{56}\\ s_{61}& s_{62}& s_{63}& s_{64}& s_{65}& s_{66}\end{array}\right]$$

$$\lambda =\frac{1}{3}(c_{11}+c_{22}+c_{33})-\frac{2}{3}(c_{44}+c_{55}+c_{66})$$

(1)

$$u=\frac{1}{3}(c_{44}+c_{55}+c_{66})$$

(2)

$$E=u\left(\frac{3\lambda +2u}{\lambda +u}\right)$$

(3)

$$K=(K_V+K_R)/2$$

(4)

$$K_V=\left[(c_{11}+c_{22}+c_{33})+2(c_{12}+c_{23}+c_{13})\right]/9$$

(5)

$$K_R=1/\left[(s_{11}+s_{22}+s_{33})+2(s_{12}+s_{23}+s_{13})\right]$$

(6)

$$G=(G_V+G_R)/2$$

(7)

$$G_V=\left[(c_{11}+c_{22}+c_{33})-(c_{12}+c_{23}+c_{31})+3(c_{44}+c_{55}+c_{66})\right]/15$$

(8)

$$G_R=15/\left[4(s_{11}+s_{22}+s_{33})-4(s_{12}+s_{23}+s_{31})+3(s_{44}+s_{55}+s_{66})\right]$$

(9)

where E is the elastic modulus of the system, $K$ is the bulk modulus, $K_V$ is the approximate upper limit of bulk modulus, $K_R$ is the approximate lower limit of bulk modulus, $c_{ij}$ is component of elastic stiffness matrix, $s_{ij}$ is component of elastic compliance matrix, $G$ is the shear modulus, $G_V$ is the approximate upper limit of shear modulus, $G_R$ is the approximate lower limit of shear modulus.

3.

Glass transition temperature of matrix

Asphalt material would appear to have a critical change in the volumetric properties when asphalt experiences the state from high modulus to high elasticity. Hence, the glass transition temperature can be determined through the change from the volumetric-simulation temperature curves. In this study, considering that the glass transition temperature is mainly related to the low-temperature performance of asphalt, the simulation temperature range was set from 150 K to 350 K during the MD simulation, the interval temperature was 25 K, and the specific volume of asphalt and ZnO/asphalt systems would be measured by Atom volumes and Surface module of Materials Studio software.

4.

Diffusion coefficient of ZnO

In the Forcite module of Materials Studio software, the mean square displacement of ZnO in asphalt can be selected from the list of properties on the Forcite Analysis dialog. The mean square displacement (MSD) presents the migration ability of the particle, which can be calculated according to the trajectory of the ZnO molecule after MD simulations. The diffusion coefficient of ZnO is one-sixth of the slope of the mean square displacement curve based on Einstein’s law of diffusion.

5.

Structure and morphology of asphalt

In this study, the radial distribution function was selected to investigate the structure of the asphalt molecular system, and the distances between asphalt molecules were calculated based on the location of each molecule’s center of mass. Atomic force microscopy is an imaging tool to capture surface morphology and micro-mechanical information, such as roughness, without destroying the materials. In this study, the matrix asphalt and ZnO-modified asphalt were examined on AFM-Nanoview 6600, which was made in China, and the morphology, roughness, and micro-structure of these two kinds of asphalt were investigated.

4. Results and Discussion

4.1. Building of Molecular Models of Asphalt and Nano-ZnO

4.1.1. Asphalt Molecular Model

To build four-component molecules of asphalt accurately, the Fourier transform infrared spectroscopy (FTIR) test, separation test, and elemental analysis test of SK-70 basic asphalt were adopted. Figure 2 displays the FTIR spectra of the SK-70 asphalt sample, and the band assignment of vibrations from SK-70 asphalt was achieved, as shown in

Table 3. Band assignment of SK-70 matrix asphalt.

Band Position (cm−1)	Band Assignment
721	(CH2)n in-phase bending vibration
746	aromatic branch bending vibration
810	benzene-ring stretching vibration
1376	C—H symmetric bending vibration of CH3
1459	C-H2 bending vibration connects with thiophene
1616	—C=C— stretching vibration
2852	C-H symmetric stretching vibration
2923	C-H asymmetric stretching vibration
3435	H-O stretching vibration

. Except for the characteristic vibration in Table 3, the weak peaks in the wavenumber range of 600–700 cm⁻¹ and 700–900 cm⁻¹ correspond to pyridine and thiophene heterocyclic type from SK-70 asphalt, respectively. Moreover, the four fractions separation test results and elemental analysis test of SK-70 matrix asphalt were conducted according to the standard test methods of bitumen and bituminous mixtures for highway engineering (JTG E20-2011), and the results were given in

Table 4. Separation test results of SK-70 matrix asphalt.

Parameter	Asphalt	Saturate	Aromatic	Resin	Asphaltene
Weight (g)	0.994	0.223	0.375	0.291	0.071
Percent (wt%)	-	22.43	37.73	29.28	7.14

and

Table 5. Elements percentages of SK-70 matrix asphalt.

Element	C	H	O	S	N	The Rests
Percent (wt%)	86.960	7.628	1.011	3.464	0.910	0.027

, respectively.

According to the recent findings and the FTIR test results of SK-70 matrix asphalt, the representative molecules (see Figure 3) were selected to represent asphalt components, including asphaltene, resin, aromatic and saturate [35,36,37,38,39,40]. Then, the asphalt molecular model was packaged with different typical molecules, and the specific information of the asphalt model was given in

Table 6. The information of asphalt molecular model.

Name	Chemical Formula	Molecular Number
Asphaltene	C42H54O	2
	C36H39ONS	2
	C72H100S	1
Resin	C44H65O2N	3
	C39H61ON	6
	C28H49ON	6
	C18H10S2	4
	C48H75ONS	9
Aromatic	C44H44	10
	C24H38	16
	C24H30S	15
Saturate	C30H49	10
	C30H62	13

.

To verify the accuracy of the asphalt model, the four components fraction, and the element contents of asphalt were compared, and the results are given in

Table 7. Components and element contents comparison of asphalt model and sample.

Items	Component (%)				Element Content (%)
	Saturate	Asphaltene	Resin	Aromatic	C	H	S	O	N
Sample	22.43	7.14	29.28	37.73	86.96	7.63	3.46	1.01	0.91
Model	22.84	7.61	30.95	38.61	86.84	8.20	3.13	0.96	0.87

. The four component fractions and element contents of the asphalt model were close to the SK-70 asphalt sample, which means the asphalt model we built in this study was reasonable. Then, the asphalt three- dimension model was constructed using Materials Studio software and shown in Figure 4.

4.1.2. Nano-ZnO Model

ZnO belongs to crystal, and its space group is P63mc (186). The length and the angle of the ZnO crystal can be obtained from its space group, and the lengths of the crystal are 3.2485 Å × 3.2485 Å × 5.20662 Å, and the angles of the crystal are 90° and 120°. Hence, we can set the parameters by using the crystal build module of Materials Studio software to build the ZnO crystal. Figure 5 shows the process of the parameter setting interface, and the ZnO crystal model is given in Figure 6. Then using the nano-structure building module of Materials Studio software, the shape of the nano-ZnO cluster was set as a sphere, the diameters were sat at 4 Å, 6 Å, 8 Å and 10 Å, respectively, and then the nano-ZnO clusters with different diameters were constructed and shown in Figure 7.

4.1.3. Nano-ZnO/Asphalt Model

Our previous research proved that when the nano-ZnO content was at 4%–5%, the nano-ZnO modified asphalt had better physical properties than other content. Therefore, the nano-ZnO content in the nano-ZnO/asphalt model was set around 4%–5%, and the specific information of nano-ZnO/asphalt models was given in

Table 8. Information of nano-ZnO/asphalt models.

Diameter of the ZnO (Å)	Number of ZnO Atom	Number of Atom in Blending System	ZnO Content (%)
4	8	7340	4.6
6	7	7355	4.7
8	2	7340	4.6
10	1	7343	4.5

. Then, four ZnO/asphalt models were built using the amorphous module in Materials Studio software and given in Figure 8.

4.2. The Interaction Energy of ZnO/Asphalt Blending System

The interaction energy represents the interaction between molecules and is related to the stable state of the blending system—the interaction energy, including bond and non-bond energy. Bond energy is determined by the distance and angle between atoms, and non-bond energy consists of van der Waals energy and electrostatics energy of the molecular system. The van der Waals interaction energy, non-bond interaction energy and electrostatics interaction energy of modified asphalt with nano-ZnO particle sizes at 4 Å, 6 Å, 8 Å and 10 Å under temperatures of 383.15 K, 393.15 K, 403.15 K, 413.15 K, 423.15 K, 433.15 K, 443.15 K and 453.15 K were calculated and given in Figure 9. It is apparent from Figure 9 that for ZnO/asphalt blending systems with different nano-ZnO particle sizes. The simulation temperature slightly affected the electrostatic interaction energy when the temperature was below 410 K. The Van deer Waals and non-bond interaction energies of ZnO/asphalt blending systems with particle sizes at 8 Å and 10 Å fluctuated markedly with the increase of temperature compared to the blending systems with particle sizes at 4 Å and 6 Å. When the simulation temperature was above 410 K, the van deer Waals and non-bond interaction energies of the ZnO/asphalt blending systems waved dramatically along with the increase in temperature. Besides, the three kinds of interaction energies of blending of asphalt and different ZnO particle sizes (4 Å, 6 Å, 8 Å and 10 Å) had reached a peak at temperatures of 423.3 K, 422.5 K, 418.3 K and 423.2 K, respectively. This result illustrated that the ZnO/asphalt blending systems achieved a more stable state at these temperatures.

4.3. Influence of ZnO Particle on the Modulus of Asphalt

From the simulation results of interaction energies of ZnO/asphalt blending systems, we can know that at the temperature of 423.3 K, 422.5 K, 418.3 K and 423.2 K, ZnO and asphalt molecules achieved a more stable structure. Therefore, the molecular dynamics simulations of mechanical properties analysis of different ZnO/asphalt blending systems were conducted at these temperatures. The elastic, bulk, and shear modulus were calculated according to Equations (1)–(9). The results data are shown in Figure 10. Noteworthy, nano-ZnO particles increased the elastic modulus, bulk modulus, and shear modulus of asphalt. Specifically, compared to matrix asphalt, the bulk modulus of ZnO/asphalt systems with particle size at 4 Å, 6 Å, 8 Å and 10 Å increased by 15.09%, 12.46%, 10.06% and 8.51%, respectively. The elastic and shear modulus of ZnO/asphalt systems increased by 2.03%, 6.27%, 6.5%, 5.85% and 1.33%, 1.71%, 5.33%, and 2.21%, respectively. It can be predicated that nano-ZnO with small size can cross through the pore and fill the space between asphalt molecules and improve the compressive properties of asphalt. As a result, the physical properties, such as the shear resistance ability of matrix asphalt, were enhanced by ZnO particles, and this conclusion agrees with laboratory tests.

4.4. Glass Transition Temperature of Asphalt and ZnO/Asphalt Blending System

The homogeneity of nano-particle-modified asphalt is closely related to thermal stability. If the nano-particle is dispersed in matrix asphalt thoroughly, the modified asphalt would have excellent thermal stability. Therefore, to predict the dispersion of nano-particle in asphalt, we focused on the approach to obtain the diffusion coefficient of ZnO in asphalt. As we all know, asphalt belongs to pure thermo-rheological material, and glass transition temperature is a critical index in the viscoelastic state changing with temperature. Under glass transition temperature, asphalt behaves as a brittle body with a high modulus. However, above the glass transition temperature, it acts in a highly elastic or rubbery state. The specific volumes of nano-ZnO/asphalt blending systems with different nano-ZnO particle sizes at different temperatures were obtained and given in Figure 11. For one system, two lines are fitted to the volume data, and the temperature of the crossover point was the estimation of the glass transition temperature of system. As can be seen from Figure 11 that the glass transition temperature of matrix asphalt was 261.5 K. After adding nano-ZnO with particles sizes of 4 Å, 6 Å, 8 Å and 10 Å, the glass transition temperature of asphalt became 260.9 K, 261.4 K, 260.5 K, and 261.1 K, respectively. The result showed that the addition of nano-ZnO had a little positive effect on the glass transition temperature of matrix asphalt. It indicates that nano-ZnO has almost no effect on the low-temperature property of asphalt, which is consistent with the results of laboratory experiments.

4.5. Diffusion Coefficient of ZnO Particle in Asphalt System

Materials Studio software conducted the molecular dynamics of different ZnO/asphalt blending systems under five temperatures. Then the mean square displacements of ZnO particles in asphalt at 383.15 K were calculated using the Forcite tool of Materials Studio software. The changes of MSD of ZnO particles in the asphalt, along with simulation time, are shown in Figure 12. As shown in Figure 12, for all ZnO/asphalt blending systems with different particle sizes, the MSD increased steadily along with the simulation time, and the particle size showed a considerable influence on the MSD of ZnO in asphalt. To investigate the diffusion coefficient accurately, the regression equations of MSD and simulation time of different ZnO/asphalt blending systems were established, and the specific information was listed in

Table 9. Information of regression equations of MSD and time.

Particle Size (Å)	Intercept	Slope	Function	Variance
4	0.0787	2.0206 × 10−4	$y=0.0787+2.0206\times {10}^{-4}x$	0.9888
6	0.0704	1.9706 × 10−4	$y=0.0704+1.9706\times {10}^{-4}x$	0.9761
8	0.1073	1.7586 × 10−4	$y=0.1073+1.7586\times {10}^{-4}x$	0.9590
10	0.0672	1.5478 × 10−4	$y=0.0672+1.5478\times {10}^{-4}x$	0.9873

. As the diffusion coefficient is one-sixth of the slope of the regression equation of MSD, so, it is very easy to obtain the diffusion coefficient of ZnO particle size at 4 Å, 6 Å, 8 Å and 10 Å in asphalt, that’s 0.34 × 10⁻⁴, 0.33 × 10⁻⁴, 0.29 × 10⁻⁴, 0.26 × 10⁻⁴, respectively. It is apparent that with the increase of ZnO particle size, the diffusion coefficient of ZnO decreased sharply. When the particle size was at 4 Å, the diffusion coefficient of ZnO was very close to the particle size at 6 Å. But, when the particle size increased to 8 Å and 10 Å, the diffusion coefficient decreased by 13% and 22%, respectively. The data showed that the increasing particle size had a negative effect on the diffusion ability of ZnO in asphalt. So, if we only consider the diffusion ability of nano-materials in asphalt and achieve better dispersion of particle materials in base bitumen, a smaller size would be recommended.

To investigate the influence of temperature on the diffusion coefficient of ZnO particles in matrix asphalt, the MSD of ZnO particles at 8 Å under different simulation temperatures (393.15 K, 413.15 K, 433.15 K and 453.15 K) was conducted, and the results were shown in Figure 13. The information on regression equations of MSD and time under different simulation temperatures is given in

Table 10. Information of regression equations of MSD and time under different temperatures.

Particle Size (Å)	Intercept	Slope	Function	Variance
393.15	0.0939	1.7734 × 10−4	$y=0.9939+1.7734\times {10}^{-4}x$	0.966
413.15	0.1361	1.7916 × 10−4	$y=0.1361+1.7916\times {10}^{-4}x$	0.9520
433.15	0.1133	2.1062 × 10−4	$y=0.1133+2.1062\times {10}^{-4}x$	0.9804
453.15	0.0935	2.3107 × 10−4	$y=0.0935+2.3107\times {10}^{-4}x$	0.9772

. According to the slope of the regression equations, the diffusion coefficient at temperatures of 393.15 K, 413.15 K, 433.15 K and 453.15 K were calculated, that’s 0.2956 × 10⁻⁴, 0.2986 × 10⁻⁴, 0.3510 × 10⁻⁴, 0.3851 × 10⁻⁴, respectively. The diffusion ability of ZnO particles in matrix asphalt increased along with the rising temperature. When the temperature rose to 413.15 K, the diffusion coefficient of ZnO increased by 2.4% compared to the temperature of 393.15. However, when the temperature increased from 413.5 K to 433.15 K and from 433.15 K to 453.15 K, the diffusion coefficient of ZnO particles increased by 17.5% and 9.8%, respectively.

4.6. Influence of ZnO Particle on the Structure of Asphalt Molecule

The radial distribution function (g(r)) is proportional to the probability that a molecule of one specie is at distance r from another specie, and it depicts the density varies along with the distance from a reference point and plays an important role in understanding the molecule structure. To reveal the effect of ZnO particles on the structure of asphalt, the ZnO/asphaltene, ZnO/aromatic, ZnO/saturate and ZnO/resin blending systems were established using Amorphous Cell of Materials Studio software. Then the RDF of ZnO and asphalt component blending systems were analyzed and shown in Figure 14. Compared to the RDF of components systems without ZnO particles, two decided differences occurred in the RDF of blending systems with ZnO particles. One difference was the position of the first peak was changed, and another one is the intensity of the peak was enhanced. Specifically, compared to asphalt components systems without ZnO particles, the position of the first peak of resin and asphaltene moved toward the right by 0.03 Å and 0.02 Å, respectively.

On the contrary, the position of the first peak of saturate moved toward the left, and that of aromatic almost didn’t change. So, from the change of ZnO on the position of the first peak of RDF, we can conclude that ZnO particles strengthened the saturate molecules and relieved aggregation of resin and aspahltene molecules. In addition, the increased intensity of ZnO and asphalt components systems illustrated that ZnO particles increased the packing density of the asphalt, and compared to components without ZnO, the RDF of ZnO/asphalt blending systems exhibited higher peaks, which means ZnO enhanced the orderliness of asphalt component molecules.

4.7. Influence of ZnO Particle on the Morphology of Asphalt

To investigate the influence of nano-ZnO particles on the morphology of matrix asphalt, the AFM tests of asphalt and ZnO-modified asphalt were conducted, and the elevation images of asphalt and ZnO-modified asphalt are shown in Figure 15. Both matrix and modified asphalt exhibited rough surfaces under an atomic force microscope, and the matrix surface seems flatter than modified asphalt. It was apparent that ZnO particles changed the morphology of matrix asphalt.

The Roughness Analysis tool of Nanoscope software was used to analyze the morphology of matrix asphalt and ZnO-modified asphalt. The number of swell heights on the surface was analyzed, and the results are given in Figure 16. The swell height of matrix asphalt was mainly in the range of 8.5 nm~12 nm, but that of ZnO-modified asphalt was in the range of 1.5 nm~12 nm. The quantity of the biggest swell on the surface of matrix asphalt was more than that of ZnO-modified asphalt, but the total number of the swell of modified asphalt was more than matrix asphalt. In addition, we can know from results of roughness analysis that the roughness average of matrix asphalt was 0.4736, and that of ZnO-modified asphalt increased by 3.6% and was 0.5063. It can be concluded that ZnO changed the structure of asphalt molecules and increased the roughness of the surface of the asphalt.

According to our results from laboratory tests and molecular dynamics, we found that ZnO can improve the high and low-temperature properties of matrix asphalt, and we believe that the change of properties is related to the change of microstructure. In addition, previous research had reported that bee-like structures were kind of string-like structures, and they emerge on the surface of the asphalt. ZnO particles increased the roughness of matrix asphalt and brought a more bee-like structure on its surface. What’s more, the bee-like area with higher stiffness compared to other area of asphalt without a bee-like structure, and the elastic recovery of asphalt depends on the bee-like structure [41]. According to the results of MD in this study and AFM, we thought that ZnO particle changed the microstructure and morphology of asphalt and also increased the roughness of asphalt surface, which resulted in improving elastic modulus, shear modulus and glass transition temperature of the matrix asphalt and bringing matrix asphalt better physical properties.

5. Conclusions

In this study, the molecular dynamics were conducted, the interaction of ZnO and asphalt molecule was investigated, and the effect of ZnO on the modulus and glass transition temperature of asphalt was studied for the first time. Combining with the AFM technique, the microstructure and morphology of asphalt were studied, and the conclusions were given as follows:

• The simulation temperature had a slight effect on the electrostatic interaction energy of the ZnO/asphalt blending system, while it had an evident influence on the van der Waals interaction energy and non-bond interaction energy. The interaction energies of blending of asphalt and different ZnO particle sizes (4 Å, 6 Å, 8 Å and 10 Å) had reached a peak at temperatures of 423.3 K, 422.5 K, 418.3 K and 423.2 K, respectively.
• The elastic, bulk and shear modulus of matrix asphalt increased by nano-ZnO particles at different extents. The increase of bulk modulus was very obvious, and compared to matrix asphalt, the bulk modulus of ZnO/asphalt systems with particle size at 4 Å, 6 Å, 8 Å, and 10 Å increased by 15.09%, 12.46%, 10.06% and 8.51%, respectively.
• The glass transition temperature of nano-ZnO/asphalt systems with particles sizes of 4 Å, 6 Å, 8 Å and 10 Å decreased by 0.06 K, 0.01 K, 0.1 K, 0.04 K, respectively, and this result showed that nano-ZnO particle had a very slight influence on the low-temperature properties of asphalt.
• The MSD of nano-ZnO particles increased steadily along with the simulation time in asphalt, and with the increasing of ZnO particle size, the diffusion coefficient of ZnO decreased sharply. In addition, the diffusion ability of ZnO particles in matrix asphalt increased along with the rising temperature. To achieve better dispersion of particle materials in base bitumen, according to the simulation results, a smaller particle size would be recommended.
• The results of the simulation of RDF and AFM indicated that ZnO particle changed the micro-structure of asphalt and also increased the roughness of asphalt surface, which benefited in improving elastic modulus, shear modulus and glass transition temperature of matrix asphalt and bringing matrix asphalt better physical properties.