Analysis of Factors Influencing the Low-Temperature Behavior of Recycled Asphalt Mixtures in Seasonal Freeze-Thaw Regions

Abstract

The use of recycled asphalt mixtures in regions with seasonal freeze-thaw cycles is becoming more popular. However, strict requirements for low-temperature cracking resistance limit their widespread application. This study designed thirteen types of recycled asphalt mixtures to explore factors affecting low-temperature performance in seasonal freeze-thaw regions and potential optimization methods. The three-point bending test assessed the low-temperature cracking performance of mixtures with varying recycled asphalt pavement (RAP) content, asphalt-aggregate ratios, asphalt types, and gradations under both conventional and freeze-thaw conditions. Results show that destructive strain and fracture energy decrease with higher RAP content, but increase with higher asphalt-aggregate ratios and 4.75 mm sieve passing rates. Adding rubber powder significantly enhances both destructive strain and fracture energy. Furthermore, the destructive strain remains insensitive to factors under both and freeze-thaw conditions, while fracture energy effectively distinguishes differences in low-temperature crack resistance. Analysis of variance reveals that RAP content, asphalt-aggregate ratio, asphalt type, and gradation significantly affect fracture energy after freeze-thaw cycles. Optimizing gradation is recommended to improve low-temperature performance of recycled asphalt mixtures in freeze-thaw regions.

1. Introduction

Seasonal pavement distress is a special consideration in road use. Rutting resistance of pavements during high summer temperatures is a commonly studied issue [1,2,3]. However, the low-temperature anti-cracking of asphalt pavement is urgently needed in seasonal freeze-thaw regions [4]. In seasonal freeze-thaw regions, winter temperatures are low and the duration of low temperatures is long. Asphalt pavement often needs to be sprayed with deicing salt to melt accumulated snow or black ice on the surface. However, this melted snow or black ice can seep into the voids of the asphalt mixture. Subsequently, it may refreeze in the early morning or at low temperatures, causing frost damage to the asphalt mixture. In spring, daytime temperatures are relatively high, causing the ice inside the asphalt mixture to melt, but as temperatures drop at night, the moisture inside the mixture may refreeze, leading to frost damage to the asphalt mixture. Therefore, it is important to study how to improve the low-temperature cracking resistance of asphalt pavements in seasonal freezing and thawing areas and how to mitigate the effects of freezing and thawing cycles in order to safeguard the service life of pavements and reduce the maintenance costs.

The use of recycled asphalt mixtures can not only reduce the accumulation of solid waste but also save a significant amount of road construction costs [5]. However, due to the severe aging of asphalt in RAP, the low-temperature cracking resistance performance of recycled asphalt mixtures is significantly insufficient, leading to a significant challenge in using recycled asphalt mixtures in seasonal freeze-thaw regions with higher requirements for low-temperature cracking resistance [6]. Sun found that geological rock materials in cold regions are prone to freeze-thaw damage, potentially impacting the cracking failure behaviors [7]. Most studies report the adverse impact of the addition of RAP on asphalt mixture behavior in low temperatures and increased potential for thermal cracking [8]. The stress value at cracking failure is decreased by the addition of RAP to the asphalt mixture [9]. Jaczewski et al. observed that the decrease in asphalt mixture strength evaluated in beam bending tests at −20 °C reached 30–40% for RAP content of 20% and 40% [10]. Therefore, it is necessary to further study the low-temperature performance of recycled asphalt mixtures and optimize them to meet the low-temperature performance requirements of seasonal freeze-thaw areas.

Behnia et al. conducted acoustic emission tests and disk-shaped compact tension tests to investigate the low-temperature performance of recycled asphalt mixtures [11]. The results showed that the embrittlement temperature of recycled asphalt mixtures increased with the increase in RAP content. Guo et al. investigated the low-temperature crack resistance and freeze-thaw cycle resistance of warm-mix recycled asphalt mixtures [12]. The study found that the freeze-thaw splitting ratio of warm-mix recycled asphalt mixtures decreased, and there was a decreasing trend in the freeze-thaw splitting ratio with the number of cycles. Hill et al. conducted disk-shaped compact tension tests and indirect tensile tests to study the low-temperature performance of recycled asphalt mixtures with biologically modified asphalt [13]. Indirect tensile tests showed that the addition of RAP reduced the creep performance of asphalt mixtures, but after adding biologically modified asphalt, the loss of creep performance of asphalt mixtures caused by RAP could be recovered, and even the creep performance could be better. Stimilli et al. utilized a temperature cracking analyzer to investigate the limit fracture temperature, shrinkage coefficient, and glass transition temperature of recycled asphalt mixtures [14]. Their findings indicate that optimizing the mixture’s gradation to improve the aggregate skeleton structure, along with appropriately increasing the asphalt content, can significantly lower the limit fracture temperature and shrinkage coefficient of recycled asphalt mixtures. This optimization enhances the low-temperature crack resistance of the mixtures. Moreover, the study found that a recycled asphalt mixture containing 40% RAP, after optimization, demonstrated better low-temperature performance than a mixture containing 25% RAP, requiring less compacting effort.

Islam et al. evaluated the effect of freeze-thaw cycles on the flexural strength modulus and tensile strength of recycled asphalt mixtures [15]. It was found that as the number of freeze-thaw cycles increased, the modulus of recycled asphalt mixtures decreased sharply in the first 20 freeze-thaw cycles and then leveled off; the indirect tensile strength changed very little. Ma et al. found that increasing the heating temperature of both virgin and reclaimed materials can enhance the low-temperature crack resistance performance of recycled asphalt mixtures [16]. Researchers have proposed many methods to improve the low-temperature performance of recycled asphalt mixtures, such as using recycling agents, adopting lower or softer new asphalt with lower temperature grades, using modified asphalt, increasing asphalt content, improving production processes, and optimizing gradation designs [17,18,19,20,21,22].

However, many current methods for improving the low-temperature performance of recycled asphalt mixtures involve using more expensive materials or adding extra production processes. These approaches increase the cost of recycled asphalt mixtures, which diverts from the original goal of using recycled asphalt pavement (RAP) to reduce costs. In addition, when exploring the low-temperature cracking resistance of recycled asphalt mixtures, the effects of different material compositions and degrees of aging on performance are usually ignored. The main focus is on the effects of conventional freeze-thaw cycling on water stability, while a systematic assessment of low-temperature cracking resistance is lacking. Therefore, it is crucial to better understand the various factors that impact the low-temperature performance of recycled asphalt mixtures. This understanding can help identify cost-effective ways to enhance low-temperature crack resistance, especially in regions with seasonal freeze-thaw cycles. Existing research primarily focuses on how freeze-thaw cycles affect the water stability of recycled asphalt mixtures. However, there is limited knowledge about the mixtures’ performance in low-temperature conditions, particularly concerning their cracking resistance after being subjected to freeze-thaw cycles. In these regions, the damage from freeze-thaw cycles is a major cause of low-temperature cracking in asphalt pavements.

This study investigates the low-temperature performance of recycled asphalt mixtures, considering the specific challenges posed by seasonal freeze-thaw cycles. The research involves experiments designed to explore the effects of different RAP contents, asphalt-aggregate ratios, asphalt types, gradations, and the aging of reclaimed materials on low-temperature crack resistance. Additionally, it evaluates how these factors influence performance under freeze-thaw conditions. The findings aim to provide practical recommendations for optimizing the low-temperature performance of recycled asphalt mixtures in regions that experience seasonal freeze-thaw cycles.

2. Materials and Methods

2.1. Materials

Following the standard test methods of bitumen and bituminous mixtures for highway engineering and technical specifications for highway asphalt pavement recycling [23,24], the recycled material was crushed and screened to obtain gradation and reduce variability. Asphalt was also extracted from the recycled material. The RAP used in the experiment was categorized into three grades, denoted as R1#, R2#, and R3#. The gradation and asphalt content of each grade of RAP after extraction are shown in Figure 1 and

Table 1. Asphalt content of RAP.

RAP	Asphalt Content (%)
R1#	3.8
R2#	3.2
R3#	5.8

, respectively. The properties of the virgin and recycled asphalt are represented in

Table 2. Properties of virgin and recycled asphalt.

	RAP Asphalt	Virgin Asphalt	Test Methods [23]
Softening point (°C)	64	48	T0606-2011
Penetration (25 °C, 0.1 mm)	34	63	T0604-2011
Ductility (10 °C, cm)	-	29	T0605-2011
Rotational viscosity (135 °C, mPa·s)	2270	498	T0625-2011

. The new aggregate used is limestone, categorized into four grades denoted as N1#, N2#, N3#, and N4#. The gradation of the new aggregate and mineral powder is illustrated in Figure 1, while the densities of the new aggregate, old aggregate, and mineral powder are provided in

Table 3. Density of recycled aggregate, virgin aggregate, and mineral powder.

Material	Apparent Relative Density	Gross Volume Relative Density
R1#	2.699	2.659
R2#	2.758	2.702
R3#	2.755	-
N1#	2.725	2.690
N2#	2.732	2.671
N3#	2.905	-
N4#	2.722	-
Mineral powder	2.650	-

.

2.2. Test Methods

A total of 13 types of recycled asphalt mixtures were designed, as shown in

Table 4. Types of recycled asphalt mixtures.

Groups	RAM	RAP Content (%)	Asphalt-Aggregate Ratio (%)	Asphalt
Different RAP content	AC20	0	3.92	Base asphalt
	AC20R10	10	4.04	Base asphalt
	AC20R20	20	4.09	Base asphalt
	AC20R30	30	4.1	Base asphalt
	AC20R40	40	4.3	Base asphalt
Different asphalt-aggregate ratio	AC20B40	20	4.09	Base asphalt
	AC20B43	20	4.39	Base asphalt
	AC20B46	20	4.69	Base asphalt
Different asphalt type	AC20B40	20	4.09	Base asphalt
	AC20T49	20	4.9	Rubber-modified asphalt
Different gradation	AC20G27	20	4.08	Base asphalt
	AC20G42	20	4.09	Base asphalt
	AC20G55	20	4.15	Base asphalt

, targeting the median value of AC-20 specification as the target gradation. Except for those with varying asphalt-aggregate ratios, all recycled mixtures were formulated using the Marshall design method to determine the optimal asphalt-aggregate ratio. Initially, recycled asphalt mixtures with RAP dosages of 0%, 10%, 20%, 30%, and 40% were devised, with corresponding optimal asphalt-aggregate ratios of 3.92%, 4.04%, 4.09%, 4.1%, and 4.3%, respectively. These mixtures were denoted as AC20, AC20R10, AC20R20, AC20R30, and AC20R40, respectively. Subsequently, using the 20% RAP dosage as a reference, recycled asphalt mixtures with oil-to-stone ratios of 4.09%, 4.39%, and 4.69% were formulated and named AC20B40, AC20B43, and AC20B46, respectively. The virgin asphalt and rubber-modified asphalt were used in the mixture, with the best oil-to-stone ratios identified as 4.09% and 4.9%, respectively, named AC20B40 and AC20T49. The proportions of graded 4.75 mm sieve holes in the recycled asphalt mixtures were set at 27%, 42.5%, and 55%, with the corresponding optimal oil-to-stone ratios determined as 4.08%, 4.09%, and 4.15%, respectively. These mixtures were named AC20G27, AC20G42, and AC20G55.

The specimens of 300 mm × 300 mm × 50 mm were firstly compacted by the wheel rolling method according to the specific material mixing ratio, and the compaction process was in strict accordance with the requirements of standard test methods of bitumen and bituminous mixtures for highway engineering (JTG E20-2011) [23]. The prepared asphalt mixture was loaded into preheated molds, spread evenly, and compacted using standard compaction methods. After compaction, the rutted slab was cooled in the mold to room temperature. Then the specimens were cut into 250 mm × 30 mm × 35 mm beams, and their low-temperature cracking performance was tested by beam bending test. In order to simulate the actual environment of the freeze-thaw cycle in the monsoon freezing area, the low-temperature cracking performance of the mixture after a freeze-thaw cycle was also tested by using the trabecular bending test.

The specimen was pre-cooled in a cryogenic chamber to the low temperature required for the test. It is usually necessary to stabilize the specimen at this temperature for 1–2 h to ensure that the entire thickness of the specimen reaches the target temperature. For the low-temperature cracking, performance was evaluated using the beam bending test at −10 °C with a loading rate of 50 mm/min. The test was conducted using a Universal Testing Machine (UTM-25). Next, the specimen was removed and placed on the equipment support. The pre-cooled specimens were placed on the supports of the three-point bending tester with a typical support span of 200 mm. The displacement measuring device was positioned at the center of the lower edge of the beam span, and both the load cell and displacement meter were connected. The press was activated and a concentrated load was applied at the center of the span at a specified rate until the specimen fractured. The curve of load versus deflection was recorded during the test. The ultimate bending stress and fracture energy of the specimen from the curve were calculated.

Under freeze-thaw cycle conditions, the trabecular specimens were initially immersed in a water bath at 20 °C for 30 min. Subsequently, the specimens were placed in plastic bags containing 1 L water and then stored in a refrigerator at −18 °C for 16 h. Following this, the specimens were immersed in water at 60 °C for 24 h. After completing three such cycles, the low-temperature cracking performance of the trabeculae was reassessed. The bending destructive strain and fracture energy of the trabeculae were used as evaluation indices for the low-temperature cracking resistance of asphalt mixtures.

The destructive strain is the maximum strain a material undergoes when subjected to an external force that causes damage. The destructive strain of asphalt mixtures is directly related to cracking resistance at low temperatures. It is shown in Equation (1).

$${\epsilon }_B={\displaystyle \frac{6\times h\times d}{L^2}}$$

(1)

where ${\epsilon }_B$ represents the maximum bending and tensile strains of the specimen at the time of damage; the larger the value, the better the low-temperature performance of the asphalt mixture. h denotes the height of the specimen at span break (mm). d denotes the mid-span deflection of the specimen at break (mm). L denotes the span diameter of the specimen (mm).

Fracture energy refers to the energy required for a material to expand a crack per unit area. It is also an important parameter for measuring the cracking resistance of asphalt mixtures. The fracture energy for beam bending tests is calculated by dividing the area enclosed by the load-displacement curve by the fracture surface, as shown in Equation (2). The physical interpretation of the area under the load-displacement curve represents the work performed on the asphalt mixture by external loads, such as temperature or mechanical loads. This work can also be understood as the stored energy within the asphalt mixture due to the influence of external loads. In other words, under identical conditions, a greater amount of work performed via external loads results in a larger amount of stored energy, leading to increased fracture energy and improved low-temperature crack resistance.

$$G_f={\displaystyle \frac{W}{A}}$$

(2)

where $G_f$, W and A denote fracture energy, work of fracture and area of fracture, respectively.

3. Results and Discussion

3.1. RAP Content Effects

The destructive strain and fracture energy of recycled asphalt mixtures with different RAP contents are shown in Figure 2.

In Figure 2, it is evident that the destructive strain and fracture energy of recycled asphalt mixtures progressively diminish with increasing RAP content. This trend signifies that augmenting the RAP content in recycled asphalt mixtures leads to a decline in their low-temperature performance. It is worth noting that the degree of influence of RAP content on the low-temperature performance of recycled asphalt mixtures expressed by these two indicators is significantly different. The reduction in destructive strain reached 42% with 10% RAP addition, whereas at 40% RAP addition, the reduction in fracture energy was only 9.4%. This discrepancy can be attributed to the fact that while destructive strain primarily reflects the ductility of the asphalt mixture, fracture energy accounts for both ductility and strength. Aging of the asphalt in RAP significantly reduces ductility, leading to a notable decrease in destructive strain. Conversely, increased asphalt viscosity and enhanced cohesion due to aging improve the strength of the recycled asphalt mixture. Consequently, while the reduction in ductility results in a significant decrease in destructive strain, the increase in viscosity and cohesion contributes to the improved strength of the mixture. As a result, the reduction in fracture energy is not as pronounced as the reduction in destructive strain. In addition, the asphalt in RAP is severely aged and deteriorates, but the addition of recycling agents restores its performance to a level close to or even improved from that of ordinary mixtures. With the increase in RAP content, the effect is gradually prominent. Although the amount of regeneration agent is also increased proportionally, but the improvement of its performance is still not as large as the impact of the old material, so that the performance of mixture decreases.

Figure 3 shows the destructive strain and fracture energy of recycled asphalt mixtures with different RAP contents after freeze-thaw cycles.

In Figure 3, after being subjected to freeze-thaw cycle damage, changes in destructive strain and fracture energy are the same as those not subjected to freeze-thaw cycles, which decrease with the increase in RAP content. When the doping of RAP is less than 30%, the decrease in destructive strain and fracture energy is small, and the decrease in fracture energy is greatly increased when the doping of RAP is up to 30% or 40%, which indicates that the evaluation indexes of fracture energy are more sensitive to RAP doping after a freeze-thaw cycle. When RAP doping is small, RAP in the aging asphalt and aggregate adhesion is good, moisture is not easy to soak into, so the effect of the freeze-thaw cycle on the recycled asphalt mixture damage is not big. When RAP doping is larger, the attached new aggregate asphalt reduces asphalt film thickness due to the high RAP asphalt content. Therefore, moisture is more likely to invade. The damage to recycled asphalt mixture increased and the damage strain and fracture energy decreased more.

Comparing Figure 2 and Figure 3, it can be found that the freeze-thaw cycle reduces the fracture energy of recycled asphalt mixture. Compared with the pre-freeze-thaw cycle, the fracture energy of each RAP dosage decreased by 23.9%, 22.5%, 21.2%, 27.1%, and 32.1%, respectively. The low-temperature cracking performance was severely damaged by the freeze-thaw cycle under the high RAP dosage, indicating that the use of recycled asphalt mixtures with high RAP dosage in the seasonal freezing zone requires a special design.

3.2. Asphalt-Aggregate Ratios

The aging asphalt in RAP reduces the ductility and impairs the low-temperature anti-cracking performance, which is not conducive to the use of recycled asphalt mixtures in the monsoon freezing zone, and an appropriate increase in the asphalt dosage can make up for the loss of low-temperature performance after the addition of RAP. For this reason, the effect on low-temperature cracking resistance was studied after freeze-thaw cycles by increasing the best asphalt-aggregate ratio by 0.3% and 0.6%, respectively. The destructive strain and fracture energy of recycled asphalt mixtures with different asphalt-aggregate ratios are shown in Figure 4.

As can be seen from Figure 4, the increase in damage strain and fracture energy after increasing the asphalt-aggregate ratio is very small; even if the asphalt-aggregate ratio increased by 0.6%, the damage strain and fracture energy only increased by 2.3% and 1.8%, respectively. Increasing the asphalt-aggregate ratio of the method by increasing the low-temperature cracking resistance of recycled asphalt mixtures seems to be ineffective. Obviously, it is not possible to continue to increase the asphalt-aggregate ratio to improve its low-temperature performance, which will inevitably increase the production cost, contrary to the original intention of adding RAP to reduce construction costs, and at the same time, ensuring high-temperature performance is difficult to meet the requirements.

Figure 5 shows the destructive strain and fracture energy of recycled asphalt mixtures with different asphalt-aggregate ratios after freeze-thaw cycles.

As can be seen from Figure 5, after being subjected to a freeze-thaw cycle, the effect of asphalt-aggregate ratio on the destructive strain is still small, but it has a greater effect on the magnitude of the change in fracture energy; when the asphalt-aggregate ratio increases from 4.0% to 4.6%, the fracture energy increases by 17.1%, which indicates that increasing the asphalt-aggregate ratio can obviously improve the low-temperature crack resistance of recycled asphalt mixtures after a freeze-thaw cycle, and it also illustrates that the evaluation indexes of fracture energy are more sensitive to the asphalt-aggregate ratio after the freeze-thaw cycle. The ratio is sensitive.

Comparing the test results in Figure 4 and Figure 5, after being subjected to freeze-thaw cycles, the destructive strains of recycled asphalt mixtures with different asphalt-aggregate ratios were reduced by 0.3%, 0.2%, and 2%, respectively, and the fracture energies were reduced by 21.8%, 17.3% and 10.0%, respectively. That is to say, the breaking strain has a low sensitivity to freeze-thaw cycle damage, and the fracture energy has a high sensitivity to freeze-thaw cycle damage. It can also be seen that increasing the asphalt-aggregate ratio improves the freeze-thaw damage resistance of recycled asphalt mixtures, which may be because the increase in new asphalt increases the asphalt film thickness, which also decreases the void ratio of the mixture and reduces the water that soaks into the interior of the mixture, and the freeze-thaw damage is smaller. In cold areas, the road surface requires higher cracking resistance to resist damage caused by low-temperature and freezing cycles. The results act as a reminder that engineers need to carefully choose the RAP doping method and asphalt-aggregate ratio to avoid the decline in cracking resistance of the mixture due to the inappropriate RAP content or ingredients.

3.3. Asphalt Types

Rubber powder is made of waste rubber particles. Research shows that rubber-powder-modified asphalt can significantly improve the low-temperature performance of the mixture; the rubber powder used in recycled asphalt mixture can make up for the loss of low-temperature performance of aging asphalt [25,26,27]. Using the external mixing method, add 20% of 20-mesh rubber-powder-modified recycled asphalt mixture, through the Marshall method to determine the rubber-modified recycled asphalt mixture of the optimal asphalt-aggregate ratio of 4.9%, in the optimal asphalt-aggregate ratio, to compare the low-temperature performance of matrix asphalt and rubber-modified asphalt recycled mixtures, as shown in Figure 6.

As can be seen from Figure 6, after adding rubber powder, the destructive strain and fracture energy of recycled asphalt mixtures are greatly improved. They increased by 29.4% and 68.1%, respectively, and the fracture energy is even more than that of the asphalt mixtures without the addition of RAP, which indicates that the rubber powder can be a very good way to improve the low-temperature cracking performance of recycled asphalt mixtures mixes.

Figure 7 shows the destructive strain and fracture energy of matrix asphalt and rubber-modified asphalt recycled mixtures after freeze-thaw cycles.

From Figure 7, it can be seen that after a freeze-thaw cycle, the destructive strain and fracture energy of rubber-modified asphalt recycled mixtures are still larger than that of matrix asphalt recycled mixtures, indicating that the low-temperature cracking resistance of rubber-modified recycled asphalt mixtures is stronger after a freeze-thaw cycle. Comparing Figure 6 and Figure 7, it is found that after a freeze-thaw cycle, the damage strain and fracture energy of rubber-modified asphalt recycled mixtures decreased by 0.7% and 15.2%, respectively; the reduction in damage strain is comparable to that of base asphalt recycled mixtures, and the reduction in fracture energy is smaller than that of base asphalt recycled mixtures (comparable to that of rubber-modified asphalt recycled mixtures), which indicates that rubber-modified asphalt recycled mixtures are more resistant to the damage caused by the freeze-thaw cycle. The incorporation of rubber powder into asphalt mixtures significantly improves their resistance to freeze-thaw cycles, and the link to practical applications is reflected in a number of ways, such as by improving road durability, promoting the use of sustainable materials, optimizing construction costs, providing design guidance, and facilitating the validation and dissemination of new technologies. The results of the study provide an important reference for practical road construction and maintenance.

3.4. Passing Rate with 4.75 mm

Improving the low-temperature cracking resistance of recycled asphalt mixtures by adjusting the gradation of the mixture is the most desirable way, which will not increase the production process or the production cost. Guo et al. [12] found that increasing the passage rate of 4.75 mm sieve holes in the gradation not only reduces the variability of high-temperature performance of recycled asphalt mixtures, but also improves their low-temperature cracking resistance. In this paper, three gradings with different 4.75 mm sieve passages were designed. The maximum nominal particle size of all three gradings was 19 mm, and the gradation curves were close to the upper, middle, and lower limits of the specification, respectively, to test their low-temperature cracking and splitting resistance properties. Figure 8 shows the destructive strain and fracture energy of the grades with different 4.75 mm sieve passage rates.

As can be seen from Figure 8, with the increase of 4.75 mm sieve passage rate, the destructive strain and fracture energy of recycled asphalt mixture are increased; when the 4.75 mm sieve passage rate reaches 55%, the destructive strain and fracture energy are improved by 63.5% and 23.6%, respectively, and the low-temperature cracking resistance performance is improved obviously. On the one hand, as the 4.75 mm sieve passing rate increases, the finer the gradation, the more dense the asphalt mixture; under the same load, the force area is larger, the actual stress is smaller, and it can withstand greater nominal stress and strain. On the other hand, the finer the gradation, the more aggregate particles per unit volume, because all aggregate particles have asphalt wrapped around them, so the more asphalt in the unit volume of the finely graded mixture, the greater the deformation.

Figure 9 shows the destructive strain and fracture energy of recycled asphalt mixtures with different gradation after being subjected to freeze-thaw cycles.

As can be seen from Figure 9, after freeze-thaw cycle damage, the low-temperature cracking resistance of the grades with high passing rate of 4.75 mm sieve holes is still better, and the destructive strain and fracture energy are still improved by 33.8% and 25.6%. Comparing the destructive strain and fracture energy in Figure 8 and Figure 9, it is found that after freeze-thaw cycle, the destructive strain of AC20G27 and AC20G42 increased, which may be caused by the test error; however, the change rule of fracture energy of the three graded mixes is in line with the expectation, and it decreased by 21.2%, 21.8% and 19.8%, respectively, and the decrease in fracture energy of AC20G55 is slightly lower than that of the other two mixes, which indicates that the fracture energy of AC20G55 is lower than that of the other two graded mixes. The fracture energy reduction in AC20G55 is slightly lower than that of the other two mixes, which indicates that the mix with the high 4.75 mm sieve passing rate has stronger resistance to freeze-thaw damage. In practice, these findings can be used directly to guide the proportioning design of recycled asphalt mixtures to ensure that the mixtures provide adequate low-temperature cracking resistance and durability under different service conditions.

3.5. Freeze-Thaw Cycle Impact

From Figure 2, when RAP is added to asphalt mixtures, it does not meet the requirements of low-temperature cracking resistance in the freezing zone, so it is necessary to improve the low-temperature cracking resistance of recycled asphalt mixtures to improve their applicability in the freezing zone. The fracture energy before and after freeze-thaw cycles under different conditions is shown as

Table 5. The fracture energy before and after freeze-thaw cycles under different conditions.

Groups	RAM	RAP Content (%)	Asphalt-Aggregate Ratio (%)	Asphalt	Fracture Energy (J/m2)
					Before the Freeze-Thaw Cycle	After the Freeze-Thaw Cycle
Different RAP content	AC20	0	3.92	Base asphalt	235.07	178.98
	AC20R10	10	4.04	Base asphalt	227.67	177.00
	AC20R20	20	4.09	Base asphalt	225.43	176.00
	AC20R30	30	4.1	Base asphalt	215.26	156.99
	AC20R40	40	4.3	Base asphalt	213.44	144.99
Different asphalt-aggregate ratio	AC20B40	20	4.09	Base asphalt	225.00	176.00
	AC20B43	20	4.39	Base asphalt	226.00	187.00
	AC20B46	20	4.69	Base asphalt	229.00	206.00
Different asphalt type	AC20B40	20	4.09	Base asphalt	225.43	176.00
	AC20T49	20	4.9	Rubber-modified asphalt	378.85	321.00
Different gradation	AC20G27	20	4.08	Base asphalt	203.00	160.00
	AC20G42	20	4.09	Base asphalt	225.00	176.00
	AC20G55	20	4.15	Base asphalt	251.00	201.00

.

Here, the significance of four factors, RAP dosage, asphalt-aggregate ratio, asphalt type and gradation, on the low-temperature cracking resistance is analyzed by ANOVA, and then the results of the significance analysis are used as the basis for making suggestions on the improvement of the low-temperature cracking performance of reclaimed asphalt mixtures in the seasonal freezing zone. From Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, Figure 8 and Figure 9, the destructive strain is less sensitive to the factors such as RAP dosage, asphalt-aggregate ratio, freeze-thaw cycle, etc., and the fracture energy is more sensitive to these factors, which can effectively differentiate the effects of different factor levels, so this paper takes the fracture energy as the dependent variable of analysis of variance (ANOVA).

ANOVA is a statistical method used to compare whether there is a significant difference in the means of three or more sample groups. It is mainly used to test whether the difference in means between different treatment conditions or groups is greater than the within-group variation, so as to determine whether the experimental treatment has a significant effect on the results. Therefore, it is possible to compare the effect of each influencing factor by testing before and after freeze-thaw cycles. This guides the optimization of experimental conditions or material formulations. The saliency results of four factors (RAP dosage, asphalt-aggregate ratio, asphalt type and gradation) are shown in

Table 6. ANOVA of fracture energy.

Elements	Before the Freeze-Thaw Cycle		After the Freeze-Thaw Cycle
	Saliency Analysis	p-Value	Saliency Analysis	p-Value
RAP content	Saliency	0.0431	Saliency	0.0297
Asphalt-aggregate ratio	Saliency	0.0339	Saliency	0.0241
Asphalt types	Saliency	0.0257	Extremely saliency	0.0026
Gradation	Saliency	0.0415	Saliency	0.0255

. The confidence interval is less than 0.05.

As can be seen from Table 6, four conditions have significant effects on the fracture energy before and after freeze-thaw cycles. For a specific project, generally the RAP dosage and asphalt type have been determined in advance, and the design of recycled asphalt mixtures can only be improved by adjusting the asphalt-aggregate ratio and gradation to improve the low-temperature cracking resistance. However, the change in fracture energy values under different asphalt-aggregate ratios is relatively small. Additionally, a high asphalt-aggregate ratio increases asphalt usage, which raises the cost of asphalt mixtures and defeats the original purpose of cost-saving. Therefore, the perspective of gradation optimization is the most desirable method, which does not increase the cost of material use or the production process, and can significantly save the cost of low-temperature cracking resistance improvement. Although increasing the 4.75 mm sieve passing rate may reduce the high-temperature performance of the mixture, due to the fact that aging asphalt in the RAP has a better cohesion, it can make up for part of the loss of high temperature performance due to finer gradation. The 4.75 mm sieve passing rate of 55% of the recycled mixture for high temperature rutting test found that its dynamic stability still meets the specification requirements. Only three grades of recycled asphalt mixtures have been selected to improve the low-temperature performance of recycled asphalt mixtures cracking performance. Further research should focus on grading further, e.g., the balanced low-temperature cracking performance of recycled asphalt mixtures, and the high-temperature performance of the relationship between the 4.75 mm sieve maximum throughput rate and the grading curve.

Although four conditions have significant effects on the fracture energy of recycled asphalt mixtures not subjected to freeze-thaw cycling, the changes in the fracture energy at different factor levels under the same influencing factor are relatively small. The differences in the fracture energy at different factor levels after freeze-thaw cycling increase significantly. Therefore, in order to better simulate the low-temperature environment and mixture performance in the monsoon freezing zone, it is recommended to use the fracture energy after freeze-thaw cycles as an evaluation index of low-temperature cracking resistance of asphalt mixtures in the monsoon freezing zone.

4. Conclusions

The use of RAP for highway construction in seasonal freezing areas has become a trend of development; however, because the environmental conditions of the seasonal freezing area are more demanding on the low-temperature performance of recycled asphalt mixtures, it is necessary to further study the low-temperature performance law of recycled asphalt mixtures based on the environmental conditions of the seasonal freezing area, and to analyze the feasibility of the program to improve the low-temperature performance of recycled asphalt mixtures in the seasonal freezing area. To this end, this paper investigates the effects of different RAP dosages, asphalt-aggregate ratios, asphalt types and gradations on the low-temperature performance of recycled asphalt mixtures under conventional conditions and freeze-thaw cycles, and obtains the following three conclusions:

(1)

Under conventional conditions and freeze-thaw cycles, the destructive strain and fracture energy of recycled asphalt mixtures decreased with the increase in RAP dosage, asphalt-aggregate ratio, and 4.75 mm sieve passing rate. This can guide the proportioning design of recycled asphalt mixtures and optimize the construction and material selection of recycled asphalt pavements in engineering.

(2)

The addition of rubber powder significantly increased the damage strain and fracture energy after freeze-thaw cycles. Rubber-modified asphalt recycled mixtures are more resistant to damage caused by freeze-thaw cycles. In practical applications, the use of rubber-modified asphalt recycled mixtures can extend the service life of roads and reduce the cost of frequent maintenance and repair. This is important for road construction and maintenance in cold regions.

(3)

In engineering, low-temperature cracking resistance can be improved by adjusting the asphalt-aggregate ratio and gradation in the mixture design. Increasing the sieve passage rate of 4.75 mm can simultaneously optimize cracking resistance and reduce costs.

(4)

The low-temperature cracking resistance test was mostly conducted using beam specimens without freeze-thaw cycles, which do not reflect the significant differences in low-temperature cracking resistance among various recycled mixtures. However, the fracture energy after freeze-thaw cycles exhibits significant changes, suggesting it could serve as a new evaluation index for recycled mixes in seasonal freezing areas.

However, only three gradation curves were investigated to improve low-temperature crack resistance. There is a need to further determine the optimum 4.75 mm sieve passages and grading curves. The low-temperature crack resistance, high-temperature performance, and long-term durability of recycled asphalt mixtures can be further discussed and evaluated in a balanced manner with the help of finite element analysis in subsequent studies.