Comprehensive Study on Dynamic Modulus and Road Performance of High-Performance Asphalt Mixture

Abstract

Asphalt pavement durability significantly impacts the service life of roads, and hence, understanding the performance of high-performance asphalt mixtures is crucial. This study investigates the performance of four high-performance asphalt mixtures: heavy-load AC-20, high-viscosity AC-20, heavy-load SMA-13, and heavy-load SMA-10. Linear Amplitude Sweep (LAS) tests revealed that heavy-load asphalt mixtures exhibit superior fatigue resistances, with the fatigue life of heavy-load SMA-13 exceeding 1.5 times that of high-viscosity AC-20 under similar stress levels. Bending Beam Rheometer (BBR) tests at −6 °C, −12 °C, and −18 °C demonstrated that both heavy-load and high-viscosity asphalts had comparable low-temperature crack resistance, with heavy-load SMA-13 showing a stiffness modulus of 627 MPa at −18 °C. Marshall tests indicated that heavy-load AC-20 had the highest stability (14.3 kN) among the tested mixtures, while heavy-load SMA-13 exhibited the highest density (2.603 g/cm³). Dynamic modulus tests spanning a frequency range of 10⁻⁴ Hz to 10⁵ Hz at various temperatures showed that heavy-load SMA-13 had a higher dynamic modulus than heavy-load SMA-10, particularly at lower frequencies (higher temperatures). Rutting tests at 60 °C indicated that heavy-load SMA-13 had the lowest rut depth (18.5 mm), outperforming other mixtures by up to 25%. The heavy-load SMA-13 asphalt mixture demonstrated the best overall performance, especially in terms of high-temperature stability, fatigue resistance, and rutting resistance. This study provides essential material performance parameters for the development of durable high-performance asphalt pavement structures.

1. Introduction

Asphalt is one of the primary materials used in China’s road construction, with a designed service life of up to 15 years or more. However, due to its lack of durability, the actual service life is much shorter than the expected design life, often leading to early-stage diseases that seriously affect driving safety and comfort. This also results in China needing to bear huge road maintenance costs annually [1,2]. To reduce maintenance costs and enhance pavement durability, researchers worldwide have dedicated themselves to researching and developing high-performance asphalt mixtures to meet the increasing demands of traffic loads and environmental protection requirements. Over the past decade, with rises in axle load levels, many highways have experienced premature damage, affecting the lifespan of pavement structures. In response to the heightened traffic demands and requirements of pavement service life, the research and development of high-performance asphalt mixtures have become important research directions [3]. China has made progress in promoting new high-performance asphalt mixtures, aiming to improve pavement load capacity, durability, and comfort. Therefore, research on the design of high-performance asphalt pavement composite structures has emerged [4].

High-viscosity modified asphalt is primarily used as a binder for porous asphalt in China and Japan. To meet the demand for porous asphalt under high-temperature conditions in Africa, Li et al. [5] developed high-viscosity asphalt. Based on simulations of the local climate of Africa, the appropriate range of high-viscosity additive dosages for different base asphalts was determined by analyzing the dynamic viscosity of the asphalt. Luo et al. [2] investigated the feasibility of warm mix technology in high-viscosity asphalt mixtures. They modified the base asphalt using AR-HVA (a high-viscosity modifier) and further blended the high-viscosity modified asphalt with two warm mix materials (Evotherm M1 and Retherm). The results indicated that the high-viscosity modifier significantly reduced the penetration of the base asphalt and increased its softening point and ductility, while both warm mix materials enhanced the penetration and ductility of the high-viscosity modified asphalt and reduced its softening point [5]. Zhang et al. [6] tested the changes in functional groups of high-viscosity modified asphalt under the influences of ultraviolet radiation intensity, high temperature, and other conditions, using infrared spectroscopy. The results showed that under the influence of various environmental factors, the high-viscosity modified asphalt underwent physical changes rather than chemical reactions. From the perspective of functional group indicators, the carbonyl index is more suitable for evaluating the degree of ultraviolet aging, while the sulfoxide index is more suitable for evaluating the impact of temperature on aging. Cai et al. [7] studied the preparation of high-viscosity asphalt using engineering waste such as recycled rubber powder and rejuvenator oil, developing an environmentally friendly asphalt binder. They prepared three types of waste-based high-viscosity asphalt binders and studied pavement performance, including viscosity, viscoelasticity, high-temperature, low-temperature, and storage stability properties. The study indicated that rubber powder, rejuvenator oil, and appropriate additives could enhance the performance of high-viscosity asphalt binders. Qin et al. [8] investigated the high-temperature performance of high-viscosity asphalt through temperature sweep and frequency sweep tests, demonstrating significant improvements in rutting resistance. Geng et al.’s experimental results [9] showed that compared with traditional asphalt and styrene–butadiene–styrene (SBS) modified asphalt, high-viscosity asphalt exhibited significantly improved resistances to high-temperature rutting and low-temperature cracking [8]. Delft University in the Netherlands used the Lintrack device to study the rutting resistance of different asphalt pavements, including hard asphalt dense concrete, modified asphalt dense concrete, and ordinary asphalt dense concrete. Rutting tests revealed that the resistances to permanent deformation of hard asphalt dense concrete and modified asphalt dense concrete were significantly better than that of ordinary asphalt dense concrete. The results of triaxial compression tests indicated that the deformation resistance of dense concrete made from hard asphalt was significantly better than that of ordinary asphalt dense concrete, confirming the superiority of hard asphalt concrete in asphalt pavement performance. The research by Zhao et al. [10] found that incorporating the EME modifier could enhance the high-temperature performance of asphalt mixtures but would reduce their low-temperature performance.

Zheng et al. [11] aimed to investigate the viscoelastic behaviors of rubber-modified asphalt mixtures with high rubber contents through dynamic modulus tests using MTS equipment under unconfined conditions. The tests were conducted at different loading rates, temperatures, and types of rubber-modified asphalt mixtures to elucidate the influence of rubber particle content on the elastic deformation and recovery capability. The results demonstrate the pronounced viscoelastic behaviors of these mixtures, with enhanced elasticity at low temperatures and high frequencies, and increased viscosity at high temperatures and low frequencies. Furthermore, the rutting resistance factors of the asphalt mixtures increase with loading frequency but decrease with temperature, suggesting that rutting is most likely when vehicles drive at low speeds in hot weather conditions. Zhang et al. [12] focuses on inter-particle interaction, a key reinforcement mechanism for aggregates in asphalt mixtures, which is a classic example of high-volume fraction particulate composites. A two-step approach using the elastic–viscoelastic correspondence principle is proposed to predict the effective dynamic modulus of asphalt mixtures at different frequencies.

The research findings suggest that the dynamic modulus of asphalt mixtures can be accurately predicted using appropriate micromechanics methods, based on laboratory tests conducted at the fine aggregate mix level, considering the void ratio and gradation of coarse aggregates. This study addresses the lack of research on predicting the dynamic moduli of semi-flexible materials. The results show that high-performance cementitious paste semi-flexible material has a higher dynamic modulus and lower phase angle compared to SMA-16, with better temperature sensitivity and deformation resistance. The dynamic modulus is positively correlated with porosity and grouting mass ratio, and the phase angle shows strong elastic properties [13,14].

Huang et al. [15] evaluated the effect of three HMAMs and two SBS modifiers on asphalt mixtures’ rutting resistances through dynamic modulus and wheel track tests, with the results simulated and analyzed via ABAQUS. The findings indicate that the dynamic moduli of the mixtures vary with loading frequency and testing temperature, and two indicators—the absolute value of the modulus and the ratio of 0.1 Hz dynamic modulus to 25 Hz dynamic modulus at 55 °C—are recommended for evaluating rutting resistance.

Zeiada et al. [16] explored the impact of confinement on the dynamic modulus |E| of asphalt concrete (AC) mixtures, a critical indicator of pavement performance. Confined |E| testing is essential for evaluating both conventional and modified AC mixtures. The findings revealed significant performance variations, particularly in gap-graded AC mixtures, with service life differences up to 45% and notable changes in resistance to rutting, fatigue cracking, and International Roughness Index (IRI). Dense-graded AC mixtures showed smaller sensitivity to confinement.

He et al. [17] addressed the inadequacy of conventional dynamic modulus parameters, determined through uniaxial compression testing, for the design and performance analysis of steel bridge deck pavement layers, which primarily fail due to flexural–tensile damage. To simulate actual pavement stress under wheel load, both four-point bending fatigue and uniaxial compression test methods were used to measure the dynamic modulus of an epoxy asphalt mixture, and the differences between the two methods were analyzed. The study further investigated the dynamic modulus and phase angle properties across varying temperatures and frequencies using the four-point bending fatigue test, constructing master curves and utilizing Sigmoidal models for correlation. The influence of epoxy resin content, mixture composition, and aging on the dynamic modulus was also examined.

Zhang et al. [18] focuses on a novel composite high-modulus agent (CHMA)-modified asphalt binder and mixture, aiming to comprehensively evaluate its high-temperature stability through various dynamic mechanical tests. The research seeks to provide insights into addressing rutting problems and enhancing the lifespan of asphalt pavement.

Viscoelastic properties and mechanical properties are the core aspects in studying the performance of high-rutting-resistance asphalt [19,20]. Therefore, we need to delve deeper into these aspects to gain a better understanding of the nature of high-rutting-resistance asphalt performance. Current research primarily focuses on analyzing the performance of mixtures after adding modifiers, and these performance improvements are often accompanied by changes in material properties under high- and low-temperature conditions. Therefore, it is of great significance to explore the viscoelastic variation patterns of these properties in depth and establish new durability evaluation criteria after modulus enhancement. This study tests the high-temperature rutting resistance, fatigue resistance, and other basic pavement performance and dynamic modulus of four types of high-performance asphalt mixtures (AC-20 high viscosity, AC-20 heavy load, SMA-10 heavy load, and SMA-13 heavy load) formed using the Marshall method. The results indicate that the SMA-13 heavy-load asphalt mixture exhibits corresponding advantages in both high-temperature performance and fatigue resistance. This study provides a foundation of material performance parameters for future exploration of high-performance asphalt pavement structures. As experts in the field of road engineering, we believe that this research holds significant value.

2. Materials and Methods

2.1. Materials

2.1.1. Asphalt

The high-viscosity asphalt and heavy-duty modified asphalt used in this study were provided by a certain asphalt preparation plant. The asphalt used in this study was produced by Yantai Jinbo Asphalt Co., Ltd., Shandong province, China.

Table 1. Technical indicators of high-viscosity asphalt performance.

Experimental Indicators	Test Result [21]
Penetration (25 °C, 100 g, 5 s)/0.1 mm	66.1
Softening point/°C	87.6
Ductility (5 °C, 5 cm/min)/cm	39.3
Dynamic viscosity (60 °C)/Pa·s	62,375
Flash point/°C	287
Elastic recovery (25 °C)/%	89
Film heating (residual penetration ratio)/%	91.9
Film heating (residual ductility) (5 °C/cm)	36.1

and

Table 2. Heavy-duty asphalt performance technical indicators [22].

Experimental Indicators	Test Result
Penetration (25 °C, 100 g, 5 s)/0.1 mm	52.3
Softening point/°C	97.2
Ductility (5 °C, 5 cm/min)/cm	38.2
Dynamic viscosity (60 °C)/Pa·s	341
Flash point/°C	97
Elastic recovery (25 °C)/%	73.2
Film heating (residual penetration ratio)/%	20.4

present the conventional performance parameter indicators for high-viscosity asphalt and heavy-duty modified asphalt.

2.1.2. Aggregates

The mineral powder, lignin fibers, and other materials used in the laboratory tests of this study were provided by a company in Nanjing, while the aggregates were supplied by a company in Shanghai. The laboratory tests encompassed SMA-10 and SMA-13 gradations. For SMA and fine-grained AC gradations, the coarse aggregates were basalt materials. For AC-20, the coarse aggregates were limestone materials, and the fine aggregates were limestone materials. The SMA-13 heavy-load modified mixture and AC-20 mixture were provided by a company in Shanghai, using hot bin aggregates. The relevant physical and technical indicators of coarse aggregates are presented in

Table 3. Physical and technical indicators of coarse aggregates.

Technical Index	Apparent Relative Density (g/cm3)	Relative Density of Gross Volume (g/cm3)	Crushing Value (%)	Needle Shaped (%)	Abrasion Value (%)
Basalt	2.856	2.803	25.6	5.3	17.8
Limestone	2.745	2.712	24.6	4.8	18.1
SMA heavy-duty aggregate	2.905	2.865	25.9	5.5	18.4
AC aggregate	2.738	2.678	26.3	6.1	18.2

.

The filler used in this study is limestone mineral powder. The technical indicators of the mineral powder used in the asphalt mixture are shown in

Table 4. Mineral powder technology for asphalt mixture.

	Pilot Project		Measured Value	Technical Requirements	Test Method
Laboratory test of mineral powder	Performance density (t/m3)		2.675	≥2.50	T0352
	Water content (%)		0.3	≤1.0	T0352
	Particle size range	<0.6 mm (%)	100	100	T0351
		<0.15 mm (%)	97.5	90~100
		<0.075 mm (%)	87	70~100
	Hydrophilic coefficient		0.73	<1	T0353
Ore powder for heavy duty	Performance density (t/m3)		2.735	≥2.50	T0352
	Water content (%)		0.4	≤1.0	T0352
	Particle size range	<0.6 mm (%)	100	100	T0351
		<0.15 mm (%)	97.0	90~100	T0353
		<0.075 mm (%)	88.4	70~100	T0352
	Hydrophilic coefficient		0.81	<1	T0352

. Additionally, lignin fibers were incorporated into the SMA asphalt mixture during the laboratory tests, with a dosage of 0.3% of the asphalt mixture’s mass. The technical indicators of the lignin fibers are presented in

Table 5. Technical indexes of lignin fibers.

Technical Index	Lignin Fiber
Appearance and color	Gray
Fiber length (mm)	<6
Ash content (%)	18.7
pH value	7.3
Oil absorption (%)	6.8
Moisture content (%)	4
Density (g/cm3)	0.87

.

In this experiment, AC-20 high viscosity, AC-20 heavy load, SMA-10 heavy load, and SMA-13 heavy load were selected as the gradings of asphalt mixtures in accordance with the specification requirements and previous engineering examples. The specific grading ranges are shown in

Table 6. AC-20 high-viscosity asphalt mixture grading range.

Screen Size		26.5	19	16	13.2	9.5	4.75	2.36	1.18	0.6	0.3	0.15	0.075
Standard grading	Lower limiting value	100	90	78	62	50	26	16	12	8	5	4	3
	Upper limit value	100	100	92	80	72	56	44	33	24	17	13	7
	Mid-value	100	95	85	71	61	41	30	22.5	16	11	8.5	5
	Target grading	100.0	92.2	83.5	75.2	61.1	37.7	26.4	18.0	10.0	6.8	5.6	5.2

,

Table 7. AC-20 heavy-load asphalt mixture grading range.

Screen Size		26.5	19	16	13.2	9.5	4.75	2.36	1.18	0.6	0.3	0.15	0.075
Standard grading	Lower limiting value	100	90	78	62	50	26	16	12	8	5	4	3
	Upper limit value	100	100	92	80	72	56	44	33	24	17	13	7
	Mid-value	100	95	85	71	61	41	30	22.5	16	11	8.5	5
	Target grading	100.0	92.2	83.5	75.0	60.2	36.7	25.7	17.5	9.8	6.7	5.6	5.2

,

Table 8. SMA-10 heavy-load asphalt mixture grading range.

Screen Size		16	13.2	9.5	4.75	2.36	1.18	0.6	0.3	0.15	0.075
Standard grading	Lower limiting value	100	100	90	28	20	14	12	10	9	8
	Upper limit value	100	100	100	60	32	26	22	18	16	13
	Mid-value	100	100	95	44	26	20	17	14	12.5	10.5
	Target grading	100.0	100.0	99.6	37.8	28.7	22.1	17.1	12.9	11.2	9.1

and

Table 9. SMA-13 heavy-load asphalt mixture grading range.

Screen Size		16	13.2	9.5	4.75	2.36	1.18	0.6	0.3	0.15	0.075
Standard grading	Lower limiting value	100	90	45	22	18	14	12	10	9	8
	Upper limit value	100	100	65	34	27	22	19	16	14	12
	Mid-value	100	95	55	28	22.5	18	15.5	13	11.5	10
	Target grading	100.0	91.1	63.1	27.0	22.3	18.4	15.4	12.9	11.8	9.5

.

2.2. Experimental Methods

2.2.1. Medium-Temperature Fatigue Resistance LAS Test

The Linear Amplitude Sweep (LAS) test is a rapid test method used to evaluate the fatigue performance of asphalt. Compared with the traditional Time Sweep (TS) test, it can more efficiently overcome time constraints. This test is conducted based on the principle of Viscoelastic Continuum Damage (VECD). By assessing the resistance to damage of asphalt binders under cyclic loading, it can predict fatigue life [23,24,25,26,27]. The damage parameter D (Damage Intensity) represents the cumulative sum of damage increments for each asphalt binder, as shown in Equation (1) [28].

$$\mathrm{D}(\mathrm{t})\cong {\sum }_{\mathrm{i}=1}^{\mathrm{N}}\left[\mathsf{\pi }{\mathsf{\gamma }}_0^2\right({\mathrm{C}}_{\mathrm{i}-1}-{\mathrm{C}}_{\mathrm{i}}\left){]}^{{\displaystyle \frac{\mathsf{\alpha }}{1+\mathsf{\alpha }}}}\right({\mathrm{t}}_{\mathrm{i}}-{\mathrm{t}}_{\mathrm{i}-1}{)}^{{\displaystyle \frac{1}{1+\mathsf{\alpha }}}}$$

(1)

where $\mathsf{\alpha }$ represents the damage parameter.

The fatigue factor is related to the inherent fatigue properties of asphalt. The new product obtained by multiplying the dynamic modulus $\mid {\mathrm{G}}^{*}\mid$ by the sine value of the phase angle $\mathrm{s}\mathrm{i}\mathrm{n}\mathsf{\delta }$ is called the fatigue factor, as shown in Equation (2):

$$\mid {\mathrm{G}}^{*}\mid \mathrm{s}\mathrm{i}\mathrm{n}\mathsf{\delta }={\mathrm{C}}_0-{\mathrm{C}}_1(\mathrm{D}{)}^{{\mathrm{C}}_2}$$

(2)

The calculation of fatigue life adopts Equation (3):

$${\mathrm{N}}_{\mathrm{f}}=\mathrm{A}({\mathsf{\gamma }}_{\mathrm{m}\mathrm{a}\mathrm{x}}{)}^{\mathrm{B}}$$

(3)

where $\mathrm{C}$ represents the model parameter, ${\mathrm{N}}_{\mathrm{f}}$ represents the number of fatigue cycles when the fatigue factor decreases to 35% of its initial value, ${\mathsf{\gamma }}_{\mathrm{m}\mathrm{a}\mathrm{x}}$ represents the maximum strain of the asphalt binder, and $\mathrm{A} \mathrm{a}\mathrm{n}\mathrm{d} \mathrm{B}$ represent the fitting parameters of the fatigue model.

2.2.2. Low-Temperature Crack Resistance BBR Test

The Bending Beam Rheometer (BBR) can apply continuous external loading to asphalt specimens. Stiffness modulus and creep rate indicate the ability to resist cracking at low temperatures. The stiffness modulus (S) should be less than 300 MPa, and the m value should be greater than 0.3. Physically, a higher S value indicates better crack resistance of asphalt at low temperatures, while the creep rate (m) reflects the sensitivity to strain, with a higher m value being more favorable for asphalt’s crack resistance [29].

2.2.3. Marshall Test

With the introduction of new modified materials such as high-performance asphalt and heavy-load asphalt, the high-temperature performances of asphalt mixtures have been significantly improved, which can be directly reflected in the Marshall test through changes in various test indicators. Therefore, it is very worthwhile to study and analyze the impact of high-performance asphalt and heavy-load asphalt on the indicators of the Marshall test. In this study, after determining the optimal gradation of the mixture, Marshall specimens of four types of high-performance asphalts were prepared, and the indicators of individual Marshall specimens were obtained through testing.

2.2.4. Dynamic Modulus Test

Before the test, to ensure that the whole specimen reaches the required temperature, it needs to be placed in a temperature-controlled box for 5 h. The dynamic modulus specimen with sensors installed is then tested. During the test, a sinusoidal load is applied to measure the stress and strain responses of the asphalt mixtures, thereby obtaining dynamic complex modulus data. These data are then used to plot the master curve of modulus. The establishment of the master curve of the modulus is based on the time–temperature superposition principle [30]. This principle suggests that the dynamic response characteristics of materials are similar when changes occur in temperature and frequency under test conditions. The most commonly used W.L.F equation is adopted in the study. The parameter α(T) at different temperatures is determined by obtaining the complex modulus and phase angle corresponding to different temperatures and frequencies. By using nonlinear fitting to process the data based on the Sigmoid anti-curvature function model, the master curve of the complex modulus and the master curve of the phase angle can be obtained, as shown in Equations (4) and (5), respectively [22]:

$$\mathrm{l}\mathrm{o}\mathrm{g} f_r=\mathrm{l}\mathrm{o}\mathrm{g} f+\mathrm{l}\mathrm{o}\mathrm{g} a\left(T\right)$$

(4)

$$\mathrm{l}\mathrm{o}\mathrm{g} E^{*}=f_r+{\displaystyle \frac{\alpha }{1+e^{\beta +\gamma \mathrm{l}\mathrm{o}\mathrm{g}\left(f_r\right)}}}$$

(5)

where $f_r$ is the reduced frequency in Hz, $E^{*}$ is the minimum modulus in MPa, $\beta \mathrm{a}\mathrm{n}\mathrm{d} \gamma$ are the shape parameters, and $\alpha$ is the amplitude of the dynamic modulus.

2.2.5. Fatigue Test

In high-performance pavement layers, AC-20 with a larger texture depth and wear-resistant properties, as well as modified asphalts SMA-10 and SMA-13, are selected as the pavement materials. Three parallel specimens are prepared for each scenario. According to the method T0719-2011 in the “Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering” (JTG E20-2011) [31], plate-shaped specimens with dimensions of 300 mm × 300 mm × 60 mm (80 mm and 100 mm) are formed by wheel rolling. The standard test temperature is 60 °C, and the wheel pressure is 0.7 MPa. Before the test, the specimens are first placed in an environmental box at the test temperature or at 15 °C for at least 4 h to maintain a constant temperature. Then, the small beam specimens for the test are installed in the fatigue loading control system and clamped securely. The four-point bending fatigue test is conducted by setting the test parameters on the software interface [32]. The required parameters for the test (such as load size, loading frequency, and loading waveform) are set in the computer program. The small beam specimens are tested with a loading rate of 50 mm/min to obtain their ultimate load. During the test, changes in load displacements and deformations of the small beam specimens after loading are observed. Finally, the test results are recorded and analyzed, as shown in Figure 1.

2.2.6. High-Temperature Rutting Test

The high-temperature stability of asphalt mixtures refers to the ability of asphalt pavement to resist permanent deformation under long-term repeated vehicle loads at high temperatures in summer (usually 60 °C). According to the method T0719-2011 in the “Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering” (JTG E20-2011), plate-shaped specimens with dimensions of 300 mm × 300 mm × 60 mm (80 mm and 100 mm) are formed by wheel rolling. The standard test temperature is 60 °C, and the wheel pressure is 0.7 MPa. After molding, the rutting board specimens of ordinary asphalt mixtures, along with the test molds, are cooled under normal temperature conditions for no less than 12 h, while the cooling time for rutting board specimens of modified asphalt mixtures such as high-modulus asphalt mixtures needs to exceed 48 h. After reaching the designated cooling time, the specimens and test molds are placed together in the rutting test device and kept warm for 5 h. Then, the high-temperature rutting test is conducted according to the specifications, and the deformation curve and temperature are automatically recorded by the computer. The rutting test is stopped when the test time reaches 1 h or the maximum deformation reaches 25 mm.

3. Results and Discussion

3.1. Rheological Properties of Asphalt

3.1.1. Medium-Temperature Fatigue Resistance LAS Test

The VECD model parameters for heavy-load asphalt and high-viscosity asphalt based on the LAS test are shown in

Table 10. Parameters of the asphalt LAS test VECD model.

Asphalt Type	${\mathbf{C}}_0$	${\mathbf{C}}_1$	${\mathbf{C}}_2$	A	B	${\mathsf{\gamma }}_{\mathbf{m}\mathbf{a}\mathbf{x}}$	${\mathbf{N}}_{\mathbf{f}(\mathsf{\gamma }=2.5\mathbf{\%})}$	${\mathbf{N}}_{\mathbf{f}(\mathsf{\gamma }=7\mathbf{\%})}$	${\mathbf{N}}_{\mathbf{f}(\mathsf{\gamma }=7.5\mathbf{\%})}$	${\mathbf{N}}_{\mathbf{f}(\mathsf{\gamma }=10\mathbf{\%})}$
Heavy-load asphalt	3.952	0.880	0.457	5.82 × 108	5.20	333.8	4,983,097	136,044	16,554	3714
High-viscosity asphalt	4.63	0.076	0.494	3.53 × 108	5.50	323.5	2,299,213	51,037	5502	1132

. Figure 2 presents the fatigue damage curves for heavy-load asphalt and high-viscosity asphalt.

As can be seen from Figure 2, there are data vertices in the stress–strain curves of the two types of asphalt. Specifically, the oscillating stress corresponding to the position of the data vertex is the yield stress, and its strain is the yield strain. After reaching the yield stress, the overall curve begins to decline. The trends of the stress–strain curves of the two types of asphalt are basically similar. Heavy-load asphalt has the highest yield stress, while high-viscosity asphalt has the highest yield strain. Before the oscillating strain reaches 13%, the stress of heavy-load asphalt is greater than that of high-viscosity asphalt. After the oscillating strain exceeds 13%, the stress of heavy-load asphalt becomes less than that of high-viscosity asphalt. Research has shown that a wider peak yield stress platform region indicates better fatigue performance. In Figure 2, the platform region of high-viscosity asphalt is significantly wider than that of heavy-load asphalt, indicating that high-viscosity asphalt has lower strain sensitivity.

According to Table 10, the C₁ value of high-viscosity asphalt is smaller than that of heavy-load asphalt, while the C₂ value is larger. However, through a comprehensive analysis of Figure 3 and Table 10, it can be concluded that the C₁ value has a greater impact on the fatigue performance of asphalt than the C₂ value. This means that a decrease in the C₂ value has a weaker effect on the recovery of fatigue resistance of asphalt materials than an increase in the C₁ value on the degradation of fatigue resistance. The fatigue life data under four stress levels in Table 10 show that heavy-load asphalt exhibits larger fatigue data and has significant differences from the fatigue data of high-viscosity asphalt.

3.1.2. Low-Temperature Crack Resistance BBR Test

This study conducted low-temperature performance tests on 70 # base asphalt, high-viscosity asphalt, and heavy-duty asphalt after PAV aging, including BBR tests at −6 °C, −12 °C, and −18 °C. The BBR test results are shown in

Table 11. Summary of BBR test results.

Asphalt Type	Test Temperature (°C)	Stiffness Modulus S (MPa)	Creep Rate (m)
Base asphalt	−6	70	0.431
	−12	181	0.374
	−18	442	0.212
High-viscosity asphalt	−6	111	0.401
	−12	264	0.315
	−18	597	0.229
Heavy-load asphalt	−6	129	0.364
	−12	270	0.311
	−18	627	0.233

.

Based on the analysis of Figure 4 and Table 11, it can be observed that as the temperature decreases, the stiffness modulus S of the three types of asphalt exhibits an overall upward trend, while the creep rate m shows a downward trend. At −12 °C, the S values of the three types of asphalt are relatively close, especially for heavy-load asphalt and high-viscosity asphalt. As the temperature continues to drop, the growth rates of S values for high-viscosity asphalt and heavy-load asphalt between −6 °C and −18 °C are greater than that of base asphalt. Compared with high-viscosity asphalt, heavy-load asphalt exhibits a higher growth rate between −12 °C and −18 °C, indicating that heavy-load asphalt has better crack resistance when the temperature is not below −18 °C. The stiffness modulus S of heavy-load asphalt is greater than that of base asphalt at low temperatures of −6 °C, −12 °C, and −18 °C, further demonstrating its superior low-temperature crack resistance. According to Figure 4, when the temperature decreases to −12 °C, both asphalts meet the requirement of having a creep rate m not less than 0.3. However, at −18 °C, the creep rate m values of both do not meet the specified limit, while the stiffness modulus S values of both high-viscosity asphalt and heavy-load asphalt are greater than 300 MPa at −18 °C. In summary, high-viscosity asphalt and heavy-load asphalt have similar low-temperature performances. According to the PG grading requirements, the low-temperature performances of both belong to the −16 °C category.

3.2. Marshall Test Analysis

The best oil–stone ratios and Marshall indexes of four asphalt mixtures obtained in this study are shown in

Table 12. Marshall test results of different asphalt mixtures.

Aggregate Grading Type	Asphalt–Aggregate Ratio (%)	Density (g/cm3)	Voidage (%)	Intermittent Rate (%)	Stability (kN)	Saturability (%)	Flow Value (×0.1 mm)
High-viscosity AC-20	4.3	2.464	3.3	12.1	13.0	73.2	32.3
Heavy-load AC-20	4.2	2.477	2.9	11.6	14.3	74.9	31.0
Reload SMA-10	6.2	2.586	3.0	16.8	13.9	75.7	35.2
Reload SMA-13	5.8	2.603	4.4	17.1	14.2	74.6	37.1

. In the heavy-duty asphalt mixture, the density of heavy-duty SMA-13 is the highest, the stability of heavy-duty AC-20 is the highest, and its flow value is the smallest. In the heavy-duty SMA asphalt mixture, the heavy-duty SMA-13 has the highest density, the largest void ratio, the largest clearance ratio, and stability and flow value compared with SMA-10. In the AC-20 asphalt mixture of the same grade type, the density, voidage, and clearance ratio of the heavy-duty asphalt mixture are lower than those of the high-viscosity asphalt mixture, and the stability and saturation are higher than those of the high-viscosity asphalt mixture.

3.3. Analysis of Viscoelastic Performance

Asphalt pavement needs to bear the traffic load generated by vehicle driving in actual operation, and its frequency range is usually 10⁻³ Hz to 10³ Hz. However, due to the limitation of test conditions, it is difficult to obtain accurate test results at lower or higher frequencies. The mechanical response of asphalt mixtures at higher and lower frequencies can be analyzed by using the time–temperature equivalence principle [29].

The mechanical behavior of asphalt mixture in the whole frequency domain and the whole temperature range can be observed by a Sigmoid inverse function. Through the nonlinear programming solution program in Excel software, five different shift factors of four different asphalt mixtures at different temperatures of −10 °C, 4.4 °C, 21.1 °C, 37.8 °C, and 54.5 °C were obtained, respectively. Then, the complex modulus in different frequency ranges was moved. When the shift factor was greater than 1, The complex modulus in the frequency range shifted to the right, whereas the complex modulus curve shifted to the left. After data processing, the main curve of the complex modulus in the whole temperature range could be obtained. The four important parameters of the Sigmoid inverse function were obtained by nonlinear programming, as shown in

Table 13. Inverse curve function coefficients of the main complex modulus curve.

Type of Mixture	G-AC20	Z-AC20	Z-SMA13	Z-SMA10
δ	2.513950364	1.929009021	2.166833537	2.089978375
α	1.986024096	2.598457452	2.303731693	2.382774627
β	−0.616206593	−1.067416479	−0.763456088	−0.777443172
γ	−0.636053933	−0.543338913	−0.577850231	−0.571573125

.

The fitting parameters δ and α in Table 13 constitute two parallel asymptotes of the main curve of the complex modulus, that is, the maximum and minimum values of the complex modulus. If α is positive, the maximum value of the complex modulus is δ + α, and the minimum value of the complex modulus is δ. Conversely, the minimum of the complex modulus is δ+α and the maximum of the complex modulus is δ. The maximum complex modulus of heavy-duty asphalt and high-viscosity asphalt, as well as different gradation types, is about 4.5, which is also different.

According to the fitting parameters in Table 9, in order to compare the viscoelastic properties of different mixtures in the same heavy-duty asphalt, the main curves of the dynamic modulus of two different SMA mixtures are shown in Figure 5.

According to the complex modulus data obtained, it can be seen from Figure 4 that for heavy-duty asphalt, the dynamic modulus of the SMA-13-grade asphalt mixture is slightly higher than that of the SMA-10-grade asphalt mixture in the frequency range shown in the figure. Combined with Table 13, the maximum value of both gradations is 4.4, and the minimum value of the SMA asphalt mixture is 2.1, indicating that the convergence values of the two gradations are similar, and their high- and low-temperature properties are similar.

The main curves of the dynamic modulus of the mixtures of heavy-duty asphalt and high-viscosity asphalt AC-20 are drawn as shown in Figure 6. The frequency selection of the horizontal coordinate in the figure ranges from 10⁻⁴ to 10⁵ Hz, which can basically reflect the action frequency under the actual traffic load.

As can be seen from Figure 7, for the mixture of AC-20, the maximum δ+α of heavy-duty and high-viscosity asphalt mixture is around 4.5 in the selected frequency range. In the low-frequency range, the dynamic modulus of Z-AC20 is low. At low frequency, that is, high temperature, Z-AC20 shows a convergence trend, which can indicate that the high-temperature resistance is better. At this time, the convergence trend of the high-viscosity asphalt mixture cannot be seen in the figure.

The main curves of the complex dynamic modulus, as illustrated in Figure 7, provide a comprehensive perspective on the viscoelastic behavior of the four asphalt mixtures across a wide range of temperatures and frequencies. Notably, convergence trends are observed in distinct regions, suggesting similarities in the high-temperature and low-frequency responses, as well as the low-temperature and high-frequency behaviors of the materials.

Specifically, in the low-temperature and high-frequency range, all asphalt mixtures exhibit a convergence trend, indicating that their dynamic modulus approaches a common value under these conditions. This suggests that at high frequencies, where the time scales of deformation are very short, the mixtures behave similarly regardless of their composition.

Conversely, in the low-frequency and high-temperature region, only the SMA asphalt mixtures show a clear convergence trend. This indicates that at high temperatures and long deformation time scales, the viscoelastic properties of the SMA mixtures converge, potentially due to their unique aggregate gradation and binder properties. The AC-20 mixture, however, does not exhibit a significant convergence trend in this region, suggesting that its behavior at high temperatures may be more influenced by its specific formulation.

Furthermore, it is noteworthy that the dynamic modulus of the SMA-10 mixture is greater than that of the SMA-13 mixture across the frequency range shown in Figure 7. This indicates that the SMA-10 mixture possesses a higher stiffness, which may be attributed to its different aggregate gradation and binder content. The dynamic modulus range of the SMA mixtures is also wider compared to that of the AC-20 mixture, suggesting a greater variability in their viscoelastic responses.

At the extreme high-temperature condition, corresponding to a frequency of 10⁻⁵ Hz, the dynamic moduli of both the SMA and AC mixtures converge to comparable values. However, the AC-20 mixture exhibits a slightly higher dynamic modulus, indicating its superior high-temperature resistance. This finding is particularly significant as it highlights the potential benefits of using heavy-duty AC-20 mixtures in applications requiring excellent high-temperature stability [33].

Overall, the analysis of the main curve of the complex dynamic modulus provides valuable insights into the viscoelastic behaviors of different asphalt mixtures under varying conditions. It underscores the importance of considering both temperature and frequency in evaluating the performance of asphalt pavements and highlights the unique properties of SMA and AC mixtures that can be leveraged to enhance pavement durability and performance.

3.4. Fatigue Test on High-Performance Asphalt Pavement Structures

To better simulate the effects under different wheel loads, stress ratios of 0.3, 0.4, 0.5, and 0.6 were selected for the fatigue testing. Under these conditions, four-point bending fatigue tests were conducted on composite structures of SMA-13, AC-20, and SMA-10 asphalt mixtures. The test results of the composite structures are shown in

Table 14. Fatigue life of three composite structures.

Structure Type	Stress Ratio	Fatigue Life
		1	2	3	Average Value
Structure 1	0.3	16,524	17,532	16,562	16,873
	0.4	9152	8250	8528	8643
	0.5	5652	5012	5186	5283
	0.6	1896	2521	2350	2256
Structure 2	0.3	17,128	16,925	17,367	17,140
	0.4	10,547	9654	11,683	10,628
	0.5	6781	6149	6368	6432
	0.6	1987	2645	2460	2364
Structure 3	0.3	22,679	23,025	23,529	23,078
	0.4	14,456	13,856	13,552	13,988
	0.5	6925	7362	7856	7381
	0.6	4098	3625	3425	3716

, and the relationships between different stress ratios and fatigue life are illustrated in Figure 8.

For the fatigue life (N_f) and stress ratio (S_i) of asphalt mixtures, clear linear relationships are observed when both are expressed in logarithmic terms. Using the stress-controlled mode, the logarithmic fatigue equation in this mode is expressed in Equation (6) as follows [34]:

$$\lg {\mathrm{N}}_{\mathrm{f}}=\mathrm{K}+\mathrm{n}(\mathsf{\delta }/\mathrm{S})$$

(6)

where N_f represents the fatigue life result, K and n are regression coefficients, with K representing the intercept of the stress curve and n representing the slope of the fatigue curve, and $\mathsf{\delta }/\mathrm{S}$ denotes the stress ratio.

By analyzing the fatigue life curves of the three distinct asphalt mixtures, several insightful observations were made regarding their fatigue performance. Under identical stress ratios, the double logarithmic fatigue equation provided valuable insights into the behavior of the composite structural beam specimens. Specifically, the smaller the value of the parameter n in this equation, the gentler the slope of the fatigue curve. This suggests that the beam specimens exhibit less sensitivity to changes in their fatigue life, indicative of more stable and durable performance. Concurrently, a smaller value of the parameter K, which represents the intercept of the stress curve in the double logarithmic fatigue equation, corresponds to a shorter fatigue life for the asphalt mixture. Consequently, the fatigue life curve for such mixtures is positioned at the lowermost end of the plot.

Table 14 presents a detailed comparison of the parameters K and n for the fatigue equations of the three different composite structures. This comparison enables a quantitative evaluation of their fatigue performance. Notably, Structure 1 exhibited the poorest fatigue resistance, as evidenced by its smallest K values in both the single and double logarithmic equations. Conversely, Structure 3 demonstrated the best fatigue resistance, as indicated by its largest K value in the double logarithmic equation, which translates to a higher number of fatigue life cycles under varying stress ratios.

Figure 9 and Figure 10 further illustrate these findings by graphically representing the fatigue life curves for the three structures under the same test conditions. The analysis of these curves, in conjunction with the fatigue life equations, confirms that Structure 3 consistently outperformed Structures 1 and 2 in terms of fatigue resistance. Specifically, under different stress ratios, the fatigue resistance of the composite structures ranked from highest to lowest as follows: Structure 3 > Structure 2 > Structure 1. This ranking underscores the importance of considering both the slope (as indicated by n) and the intercept (as indicated by K) of the fatigue curves in evaluating the fatigue performance of asphalt mixtures.

The fatigue life analysis conducted in this study provides a comprehensive understanding of the fatigue performance of different asphalt mixtures and their composite structures. The findings emphasize the critical role of the parameters K and n in the double logarithmic fatigue equation in assessing the fatigue resistance of asphalt materials, and they offer valuable insights for the design and optimization of asphalt pavement structures to enhance their durability and service life.

3.5. High-Temperature Rutting Test on High-Performance Asphalt Mixtures

The rutting test results are shown in Figure 11. According to relevant specifications, the dynamic stability of the middle layer is required to be at least 1000 cycles/mm for light and medium traffic, at least 3000 cycles/mm for heavy traffic, and at least 5000 cycles/mm for very heavy traffic. By analyzing Figure 11, the dynamic stability of various asphalt mixtures, from highest to lowest, is as follows: heavy-load SMA-13 > heavy-load SMA-10 > heavy-load AC-20 > high-viscosity AC-20, all exceeding 3000 cycles/mm and meeting the dynamic stability requirements for light, medium, heavy, and very heavy traffic. The addition of high-performance asphalt significantly increases dynamic stability, which is six times higher than that of ordinary asphalt mixtures and exceeds the known modification effects of SBS on asphalt. The modification effect of high-performance asphalt is mainly manifested during mixing and transportation. When heated, high-performance asphalt melts onto the surface of the aggregate and dissolves into the asphalt, creating a cementing effect that enhances the adhesion between the aggregate and the asphalt. This results in an increased softening point of the asphalt and reduced thermal sensitivity. The impact of high-performance asphalt on the high-temperature performance of asphalt mixtures is analyzed. As shown in Figure 11, under the standard test temperature of 60 °C, the dynamic stability of asphalt mixtures increases with the addition of high-performance asphalt. A higher dynamic stability indicates better high-temperature stability, with asphalt mixtures represented by heavy-load SMA-13 exhibiting the best high-temperature performance.

In order to further assess the high-temperature performance of the various composite pavement structures, rutting tests were conducted on specimens with layer configurations of 3 + 3, 4 + 4, and 4 + 6, corresponding to the thicknesses of 60 mm, 80 mm, and 100 mm, respectively. Prior to testing, a tack coat was applied to the specimens to ensure adequate bonding between layers. The rutting tests were carried out at a temperature of 60 °C using a wheel-tracking rutting tester, adhering strictly to the test procedures outlined in the Chinese national standard JTGE20-2011 “Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering” [31].

The results of these rutting tests, as presented in Figure 12, revealed a discernible trend in the dynamic stability of the composite pavement structures. Specifically, as the number of layers increased, the dynamic stability exhibited a decreasing trend across the three configurations (3 + 3, 4 + 4, 4 + 6). Nevertheless, it is noteworthy that the 3 + 3 composite pavement structure, consisting of a double-layer SMA composite, demonstrated a superior dynamic stability compared to the other two structures. This observation underscores the effectiveness of the double-layer SMA composite design in enhancing the overall high-temperature performance of the pavement structure.

To elucidate the reasons behind this enhanced performance, it is imperative to consider the inherent properties of SMA mixtures. SMA, characterized by its gap-graded aggregate distribution and the inclusion of fiber modifiers, is renowned for its superior rutting resistance and durability. The double-layer SMA composite structure, by leveraging these attributes, is able to distribute and resist loads more effectively, thereby maintaining a higher dynamic stability under high-temperature conditions [35].

Based on these findings, it can be concluded that the 3 + 3 composite pavement structure, featuring a double-layer SMA composite, represents the optimal choice for high-temperature performance among the configurations tested. This structure not only meets but exceeds the dynamic stability requirements for various traffic loads, making it a highly suitable candidate for use in highway construction where high-temperature resistance is paramount. Future research may further explore the long-term performance and durability of this composite structure under varying environmental and traffic conditions.

4. Conclusions

This study explored the basic pavement performances, including high-temperature rutting resistance, fatigue resistance, and dynamic modulus, of four types of asphalt mixtures. Through comparative analysis, the following conclusions can be drawn.

(1)

The addition of high-performance asphalt significantly improved the high-temperature stability of the mixtures. In high-performance asphalt mixtures, the dynamic stability also increased. Among different gradations, SMA-13 exhibited the best high-temperature performance, and the 3 + 3 composite structure was the best among the studied composite configurations.

(2)

Under the same stress ratio, comparing the fatigue test results of the three composite structures and analyzing them using the single and double logarithmic fatigue life equations, it was found that composite Structure 3 had the best fatigue resistance, followed by Structure 2, while Structure 1 had the worst fatigue resistance.

(3)

After adopting high-performance asphalt materials, the dynamic stability values obtained from the rutting test exceeded the requirements for very heavy traffic specified in the regulations, being more than double the value required for heavy traffic. Among different asphalt mixtures, SMA-13 exhibited the best high-temperature performance. Within the same type of asphalt mixture, heavy-load AC-20 and heavy-load SMA-13 demonstrated good high-temperature performance, indicating the excellent performance of heavy-load asphalt materials in heavy-load traffic conditions.

(4)

SMA asphalt mixtures showed a convergence trend in the low-frequency and high-temperature region, while AC-20 did not exhibit a clear convergence trend. The dynamic modulus of the SMA-10 mixture was greater than the complex modulus of the SMA-13 mixture, and the dynamic modulus range of the SMA mixtures was relatively wide. Under extreme high-temperature conditions, the dynamic moduli of the SMA and AC mixtures were comparable, but the value of AC-20 was slightly higher, indicating better high-temperature resistance.

(5)

Under different strain conditions, the fatigue life of the four mixtures showed a pattern where higher strain levels led to lower fatigue cycles. Under different strain conditions, the fatigue resistance, from highest to lowest, followed this order: heavy-load SMA-13 > heavy-load SMA-10 > heavy-load AC-20 > high-viscosity AC-20.