The Effect of Refined Separation on the Properties of Reclaimed Asphalt Pavement Materials

Abstract

Refined separation not only controls the variability of reclaimed asphalt pavement (RAP), but also improves the mixing ratio of RAP and the quality of recycled asphalt mixtures. This study examines RAP treated with various refined separation frequency parameters, analyzes the variation rules and the variability of RAP aggregate gradation, asphalt content, asphalt properties, and aggregate properties, and calculates the maximum mixing percentage of coarse RAP material by using the gradation variability control method and the asphalt content variability control method. The results show that the variability of gradation and asphalt content of coarsely separated RAP is considerable, and a refined separation process significantly reduces the variability of gradation and asphalt content of RAP; the agglomeration of RAP decreases with an increase in the refined separation frequency; and the RAP agglomeration of three kinds of RAPs (E1, E2, and E3) under a refined separation frequency of 55 Hz reduces by 6.40%, 4.30%, and 4.30%, respectively, as compared with that of coarsely separated RAPs. The asphalt content of the refined separation RAP gradually decreases with an increase in frequency, and the asphalt content of E1 and E2 (55 Hz) was only 0.95% and 1.10%, respectively. The maximum percentage of RAP in recycled asphalt mixtures was calculated using the gradation variability control method and the asphalt content variability control method, respectively. The maximum proportions of RAP were 45% and 33% for A1 (0 Hz), respectively, and the maximum proportions of RAP for E1 (55 Hz) were all 100%. The results of the two methods show that the process of refined separation can increase the maximum proportion of blended RAP materials. They also demonstrate that the refined separation process can increase the maximum blending ratio of coarse RAP materials, thereby improving the quality of the RAP, increasing the proportion of RAP blending, and ensuring the quality of the recycled asphalt mixture. In conclusion, the refined separation process holds promise for maximizing the potential value of RAP and optimizing its recycling, environmental, and economic benefits.

1. Introduction

Reclaimed asphalt pavement (RAP) recycling has become an important development direction for highway transportation [1,2,3]. However, due to the large number of “false-particles” in RAP, there is a high degree of variability in its gradation and asphalt content. The variability of the properties of RAP limits how much of it can be added to recycled materials and poses a challenge for its full utilization [4,5,6,7]. In order to ensure the quality of recycled asphalt mixtures, the general plant-mixed thermal regeneration requires a ratio of RAP addition of 15% to 30% [8,9,10]. For this reason, researchers have proposed asphalt and mineral separation techniques, including physical refined separation methods [11,12,13], chemical solvent separation methods [13,14], and microbial separation methods [15]. Refined separation methods have been emphasized by the industry due to their advantages of high separation efficiency, low environmental impact, and low separation cost [13,16,17,18,19,20].

The variability of RAP materials seriously affects the stability of the performance of hot recycled asphalt mixtures [21], which is one of the main reasons for limiting the maximum addition ratio of RAP. Montañez [22] found that there is significant variability in the mechanical response and performance of the evaluated RAPs, suggesting the necessity of taking into account the variability of RAP when fabricating asphalt mixtures with RAP obtained from different sources. Liu et al. [23] proposed the 12.5 mm particle size as the key sieve size for coarse and fine particles of RAP materials, which can significantly reduce the old asphalt content of RAP materials and the variability of aggregate gradation. Guo [24] found that by increasing the percentage of material passing through a 4.75 mm sieve in the gradation of hot recycled asphalt mixtures, the effect of internal friction angle on the shear strength of the mixture can be reduced, thus effectively improving the negative impact of the variability of RAP gradation on the performance of hot recycled asphalt mixtures. Some researchers focus on binder activation within RAP [25,26,27,28]. Preheating temperature and RAP source were generally considered in previous studies, and both of them were found to have a significant impact on the activation of RAP binders [25,29,30]. Mechanical separation only changes the asphalt content of RAP without changing the binder properties of RAP. Therefore, the effect of mechanical separation on the binder activation of RAP materials can be disregarded in related studies.

Numerous studies have shown that RAP pretreatment techniques can effectively reduce its variability [31,32,33], while refined separation techniques can further improve its utilization in recycled asphalt mixtures [34,35]. Zou et al. [36] used the refined separation process to deal with RAP, and it was found that the asphalt content of the coarse RAP after refined separation was less than 1%. Through refined separation, a large number of false particles are eliminated, and the gradation stability of the recycled asphalt mixture is Improved. Qiu et al. [37] utilized the refined separation process on RAP obtained from a highway expansion project. The results showed that the performances of a recycled asphalt mixture with 80% RAP and a mixture of new asphalt are comparable. Wang et al. [38] also found that the performance of recycled asphalt mixtures is basically the same as new asphalt mixtures. Road performance research based on recycled asphalt mixtures shows that the refined separation process is conducive to enhancing the proportion of RAP blending, improving the quality and stability of recycled asphalt mixtures.

Existing research focuses more on the performance of recycled asphalt mixtures with refined separation RAP, but less on the evaluation of the quality of refined separation RAP. The quality of RAP after refined separation is the key process of recycled asphalt mixtures, which directly affects the final road performance and the durability of recycled asphalt mixtures. Therefore, in order to ensure the quality of recycled asphalt mixtures, it is very necessary to carry out a systematic evaluation study of RAP produced by on-site refined separation equipment.

This paper focuses on the effect of refined separation on the properties of reclaimed asphalt pavement materials. The variability of RAP aggregate gradation, aggregate properties, asphalt content, and asphalt properties after refined separation were analyzed. Meanwhile, the effects of different refined separation frequencies on the properties of RAP materials were quantitatively evaluated. The method of calculating the maximum proportion of coarse RAP that can be added to a recycled asphalt mixture is provided in this study, and the maximum proportion of RAP that can be added after refined separation was calculated by the method of controlling the variability of gradation and the method of controlling the variability of asphalt content, respectively. Its findings will significantly contribute to maximizing the utilization of RAP materials and optimizing recycled asphalt mixture design.

2. Refined Separation Principle

The refined separation equipment comprises several sub-systems designed for RAP refined separation functionality. It includes the feeding system, screening system, centrifugal impact crusher, roller crusher, dust removal system, discharging system, etc. The refined separation equipment and principle are shown in Figure 1. The core of the equipment is the centrifugal impact crusher (Figure 1); RAP enters the centrifugal impact crusher, it obtains kinetic energy from the high-speed rotor, and it is thrown out. The thrown RAP collides and rubs violently against the crusher’s baffle and falling RAPs. The RAP “false-particles” along the asphalt interface between the aggregates are crushed, so that the coarse RAP (≥5 mm) and the fine RAP (0–5 mm) are broken. The coarse RAP (≥5 mm) is separated from the fine RAP (0–5 mm). The surface of coarse RAP contains little asphalt due to collision and friction, which can realize the separation of asphalt and coarse RAP aggregate. The effect of refined separation of RAP is related to the rotational speed of the crusher, the nature of the RAP and its temperature, and the position of the crusher plate, of which the rotational speed of the crusher is the key influence factor. This test is based on the engineering site; it is difficult to change the position of the crusher plate and the nature and temperature of RAP. Therefore, the rotor speed was adjusted by changing the crusher frequency, which can change the refined separation effect for RAP. The ambient temperature during this test ranges from 23 to 28 °C, and the RAP used is the modified asphalt mixture from the middle surface layer on the highway.

3. Materials and Methods

3.1. RAP Materials

A 120 t/h processing capacity refined separation equipment was used to finely separate RAP into 3 grades (0–5 mm, 5–10 mm, 10–20 mm) at different frequencies (0 Hz (original material), 30 Hz, 40 Hz, 50 Hz, 55 Hz). The selected RAP sample is asphalt pavement surface material, which is milled on-site in Beijing, and the old asphalt in the RAP material is a mixture of modified bitumen and plain bitumen. The RAP sample numbers used for the test are shown in

Table 1. Abbreviations of RAP samples with different refined separation frequency.

Size (mm)	Refined Separation Frequency (Hz)
	0	30	40	50	55
10–20	A1	B1	C1	D1	E1
5–10	A2	B2	C2	D2	E2
0–5	A3	B3	C3	D3	E3

. The materials were taken from four positions of the same stockpile, which are indicated by ① ② ③ ④, respectively. The RAP (10–20 mm) samples under different refined separation frequencies are given in this paper, as shown in Figure 2.

3.2. Test Methods

In this paper, after refined separation, the aggregate and old asphalt in RAP of each grain size were separated by extraction and sieving, and the aggregate gradation and asphalt content were tested and the variability was calculated after RAP extraction. The properties of the aggregate (angularity, elongated and flat particle content, crushing value) and the properties of the old asphalt (penetration, softening point, and ductility) were tested. Furthermore, based on the evaluation of the test results, the maximum addition ratio of the coarse RAP material was calculated by using the variability control method of the gradation and the variability control method of the asphalt content, respectively.

The RAP Is subjected to different collision Intensities at different refined separation frequencies, which have different effects on the morphology of the aggregates. Therefore, aggregate angularity was tested with a coarse aggregate analyzer. The RAP with different refined separation frequency was extracted and sieved, and the elongated and flat content of the coarse aggregate was tested according to the JTG E42-2005 (T0312) [39], and the crushing value was tested according to the JTG E42-2005 (T0316) [39]. Asphalt penetration, softening point and ductility indexes were tested according to the JTG E20-2011 [40].

The refined separation evaluation indexes selected for this study are summarized in Figure 3, and the calculation of the agglomeration degree is described in Section 3.2 of this paper.

4. The Effect Evaluation of Refined Separation on RAP

4.1. Gradation Variability of RAP

Controlling the variability of RAP gradation is an important objective of the refined separation process. Because the test needs to analyze the old asphalt, each gradation of RAP aggregate is obtained by the centrifugal separation method. The specific method is as follows. Firstly, the RAP samples were soaked with trichloroethylene, and then the mixture and solvent were poured into the centrifugal extractor to separate the asphalt from the minerals. After refined separation, the results of extraction and sieving of RAP aggregate gradation are shown in Figure 4, and the coefficients of variation (CVs) of the gradation of the aggregates are shown in Figure 5.

As seen from Figure 4, by comparing the gradation curves, there is a notable difference in the sieve passing percentage between the sieve gradation curves of A1 and A2, while the consistency of the sieve gradation curves of E1 and E2 is better. From Figure 5, for the three sizes of RAP, the coefficients of variation of A1, A2, and A3 (0 Hz) are larger, and the coefficients of variation tend to increase while the particle size decreases. In addition, the coefficient of variation of 4.75 mm in A1 (0 Hz) is 13.2%, while the coefficient of variation of 4.75 mm in E1 (55 Hz) is 7.2%, and the coefficients of variation for RAPs treated with refined separation are significantly lower compared with those of the unfine separated RAPs (0 Hz). For the 5–10 mm RAP, the difference in E2 (after refined separation) was significantly smaller than that of A2. For the 0–5 mm RAP, the difference in E3 was slightly smaller than A3. Compared with the coarse RAP (>5 mm), the effect of refined separation on the RAP of 0–5 mm particle size was significantly smaller. By analyzing the results of the three sizes of RAP, the refined separation process reduces the variability of aggregate gradation of the RAP and makes RAP gradation have better stability.

4.2. Agglomeration Degree of RAP

The agglomeration degree characteristic of RAP, also known as a false particle, is an important reason for the large variability of RAP gradation. In order to evaluate the agglomeration of RAP, this paper establishes the agglomeration degree evaluation index of RAP based on the difference of the sieve residue before and after the extraction of old material, and the calculation model is shown as follows:

$$I_c=\frac{{\displaystyle \sum _{i=1}^n\left|P_i^b-P_i^a\right|}}{n}$$

(1)

where I_c is RAP agglomeration degree (%); P_i^b is the passing percentage of each sieve size before RAP is extracted; P_i^a is the passing percentage of each sieve size after RAP is extracted.

The results of agglomeration degree (I_c) value of RAP materials with different refined separation frequencies are presented in Figure 6.

As shown in Figure 6, the RAP agglomeration gradually decreased with the increase in the frequency of refined separation, and the trends of the three particle sizes of RAP agglomeration were basically the same. For A1, A2, and A3, the I_c values of the agglomeration indexes were 8.02%, 5.42%, and 5.64%, respectively. And I_c values for E1, E2, and E3 are 1.62%, 1.12%, and 1.34%, respectively. Compared with I_c of A1, A2, and A3, the I_c values of E1, E2, and E3 decreased by 6.40%, 4.30%, and 4.30%, respectively. The agglomeration of RAP after refined separation decreased significantly, i.e., the content of false particles was reduced, which made RAP closer to the real mineral material. This is mainly due to the fact that the RAP entering the centrifugal impact crusher obtains kinetic energy, and the kinetic energy increases with the increase in frequency. It causes the RAP clusters to break after collision and friction. And after the friction and impact effect, the old asphalt binder is stripped from the surface of the coarse RAP particles. This effectively reduces the content of false particles in the RAP and improves the quality and stability of the RAP.

4.3. Aggregate Properties of RAP

The angularity, elongated and flat particle content, and crushing values of the aggregates after RAP extraction at different refined separation frequencies are presented in

Table 2. Aggregate properties of RAP after extraction.

Index	Size (mm)	Refined Separation Frequency (Hz)
		0	30	40	50	55
Angularity	5–10	17.90	17.53	17.29	17.18	17.16
	10–20	17.63	17.25	17.19	16.89	16.82
Elongated and flat particle content (%)	5–10	6.39	5.29	4.79	3.99	3.19
	10–20	6.68	5.39	4.80	4.20	3.20
Crushed value (%)	10–20	6.19	5.60	5.30	5.10	5.00

.

As can be seen from Table 2, the coarse aggregate angularity gradually decreases with the increase in refined separation frequency. This is mainly due to the fact that the collision and friction of RAP in the crusher become stronger with the increase in frequency, and the shape of aggregate tends to be “spherical”, resulting in the decrease in aggregate angularity. Aggregate angularity has a great influence on the performance of asphalt mixtures, especially on the high-temperature performance of asphalt mixtures. Therefore, it is crucial not to increase the frequency of refined separation too much to improve the separation effect, as it may significantly impact the angularity of the coarse aggregate, thereby affecting the high-temperature performance of recycled asphalt mixtures.

From Table 2, the elongated and flat particle content of aggregate after refined separation is lower than that of the original aggregate, and the elongated and flat particle content of the aggregates decreases gradually with the increase in the refined separation frequency. When the frequency of refined separation is 55 Hz, the elongated and flat particle contents of 5–10 mm and 10–20 mm aggregates are 3.19% and 3.20%, respectively, which is 3.20% and 3.48% lower than that of non-refined separated aggregates (0 Hz). The elongated and flat particles are easily broken and crushed during the refined separation process, which results in a decrease in the elongated and flat content of the aggregate.

From the crushing value of aggregates in Table 2, it can be seen that the crushing value of aggregates after refined separation is lower than that of aggregates without fine separation, and with the increase in the frequency of fine separation, the crushing value of aggregates gradually decreases. This is due to the fact that the refined separation process reduces the aggregate elongated and flat particle content, and the aggregate surface angle is also reduced, which in turn reduces the aggregate crushed, and with the increase in the frequency of fine separation, the aggregate tends to be more “rounded”, and the aggregate is crushed less.

4.4. Asphalt Content Variability of RAP

The asphalt contents of RAP with different refined separation frequencies are shown in Figure 7.

As shown in Figure 7, for the 0–5 mm RAP, the asphalt content of A3 (0 Hz) has a small change from the asphalt content of the refined separated RAP. The asphalt content of the 5–10 mm and 10–20 mm RAP decreases with the increase in the frequency. The asphalt content of A1 and A2 (0 Hz) is 3.28% and 3.68%, and that of E1 and E2 (55 Hz) is 0.95% and 1.10%, respectively. Compared with the same specifications, the asphalt content of RAP after refined separation (55 Hz) decreased by 2.33% and 2.58%, respectively. In coarser RAP after the refined separation process, the surface old asphalt binder is stripped after the repeated collision friction effect, resulting in a significant decrease in the RAP asphalt content.

The asphalt content and its coefficient of variation after extraction of unfine separation RAPs (A1, A2, and A3) and 55 Hz refined separation RAPs (E1, E2, and E3) are shown in Figure 7.

As can be seen in Figure 7, the coefficients of variation for asphalt content of the 55 Hz refined separation RAPs (E1, E2 and E3) are significantly lower than those of the unfine separation RAPs (A1, A2, and A3). The coefficients of variation are 10.7%, 11.8%, and 8.4% for A1, A2, and A3, respectively, and 6.1%, 7.4%, and 2.6% for the fine separated E1, E2, and E3, respectively. Compared with A1, A2, and A3 (0 Hz), the coefficients of variation of asphalt content of E1, E2, and E3 (55 Hz) are reduced by 4.6%, 4.4%, and 5.8%, respectively. The refined separation process significantly reduces the variability of asphalt content in RAP.

4.5. Asphalt Properties of RAP

The asphalt properties after extraction of RAP are shown in Figure 8.

As can be seen from Figure 8, the average values of penetration, softening point, and ductility of asphalt obtained by extraction before and after RAP refined separation change less. Moreover, with the increase in separation frequency, the standard error of the three performance indicators of asphalt gradually decreases. This is due to the RAP in the centrifugal impact crusher collision process undergoing further mixing, so that the asphalt performance variability is reduced. However, during the refined separation process, the RAPs collide and rub against each other, and this process is a physical process, which has less influence on the asphalt performance index.

5. The Calculation Method of Maximum Addition Proportion of RAP for Recycled Asphalt Mixture

5.1. Maximum Addition Proportion of RAP under the Control of Gradation Variability

The Technical Specification for construction of Highway Asphalt Pavement (JTG F40-2004) specifies that the sieve passing percentage of each mineral aggregate material should meet the corresponding quality requirements [41]. The coefficient of variation of RAP is larger than that of the new mineral material, and when the proportion of RAP added to the recycled mixture is too large, it will result in the recycled asphalt mixture not meeting the quality requirements. Therefore, in order to ensure the quality of the recycled asphalt mixtures, it is necessary to calculate the maximum proportion of RAP based on different coefficients of variation. Assuming the RAP blending ratio X, the gradation variability should meet the requirements of Equation (2) [5]:

VX ≤ V′

(2)

where V is the standard deviation of RAP passing percentage of each sieve size, %; X is the RAP addition ratio in asphalt mixture; V′ is the specification value of hot mix recycled asphalt gradation passing percentage of each sieve size, %.

Due to the new aggregate gradation sieve, throughput variability is not considered, so the recycled asphalt mixture variability requirements should be slightly more stringent than the ordinary hot mix asphalt mixture. It is recommended that the hot mix recycled asphalt mixture variability meet the requirements of

Table 3. Grading quality requirements of hot mixed recycled asphalt mixtures.

Sieve Size (mm)	Number of Samples	Quality Requirements
		Highway
0.075 mm	RAP material sample number n: n ≥ 4	XV0.075 ≤ 1%
≤2.36 mm		XV≤2.36 ≤ 3%
≥4.75 mm		XV≥4.75 ≤ 4%

.

As can be seen from Table 3, when the RAP variability is large, the quality requirements can be satisfied by reducing the RAP addition ratio, and when the RAP variability is small, it is favorable to increase the RAP addition ratio.

Relevant studies [18,42] show that the passing percentage of each sieve size of RAP aggregate material can be characterized by being distributed, and the overall standard deviation can be estimated from the samples according to the mathematical and statistical methods. According to the theory of confidence interval, the overall standard deviation is calculated by Equation (3).

$$S=\frac{\sigma }{\sqrt{n}}{t}_{\alpha /2}$$

(3)

where S is the RAP overall standard deviation, %; σ is the RAP sample standard deviation, %; n is the number of samples; α is the confidence level; for t_α/2, check the t-distribution table.

When the confidence level α = 0.05 and the degree of freedom (n − 1) is 3, check the table to obtain t_α/2 = 3.182. The overall standard deviation of the four specifications of RAP (A1, A2, E1, E2), i.e., V value, can be obtained, which is brought into the formula in Table 3 to obtain the quality requirement under the control of gradation variability, as shown in

Table 4. Quality requirements for aggregate gradation.

Sieve Size (mm)	Quality Requirements
	A1	A2	E1	E2
19	0.0%X ≤ 4%	0.0%X ≤ 4%	-	-
16	4.9%X ≤ 4%	1.1%X ≤ 4%	-	-
13.2	8.8%X ≤ 4%	3.8%X ≤ 4%	-	-
9.5	6.8%X ≤ 4%	2.9%X ≤ 4%	0.0%X ≤ 4%	0.0%X ≤ 4%
4.75	6.1%X ≤ 4%	1.0%X ≤ 4%	6.1%X ≤ 4%	0.5%X ≤ 4%
2.36	5.0%X ≤ 3%	0.7%X ≤ 3%	4.7%X ≤ 3%	0.4%X ≤ 3%
1.18	5.0%X ≤ 3%	0.6%X ≤ 3%	3.5%X ≤ 3%	0.3%X ≤ 3%
0.6	4.3%X ≤ 3%	0.5%X ≤ 3%	3.4%X ≤ 3%	0.2%X ≤ 3%
0.3	4.2%X ≤ 3%	0.5%X ≤ 3%	3.2%X ≤ 3%	0.2%X ≤ 3%
0.15	3.5%X ≤ 3%	0.4%X ≤ 3%	3.4%X ≤ 3%	0.2%X ≤ 3%
0.075	1.6%X ≤ 1%	0.5%X ≤ 1%	1.5%X ≤ 1%	0.2%X ≤ 1%

.

In order to make the recycled asphalt mixture meet the quality requirements, there exists a maximum addition ratio of RAP additions, and the maximum additive ratio is calculated according to Equation (4) [5]:

$$X_{\max }=\min \left\{0.04/S_{\ge 4.75},0.03/S_{\le 2.36},0.01/S_{0.075}\right\}$$

(4)

where X_max is the maximum addition proportion of RAP; S_≥4.75 is the overall standard deviation of the passing percentage with sieve size greater than 4.75, %; S_≤2.36 is the overall standard deviation of the passing percentage with sieve size less than 2.36, %; S_0.075 is the overall standard deviation of the passing percentage with sieve size 0.075, %. It should be noted that when recycled asphalt mixtures are designed, the addition ratio of RAP should not exceed X_max. The specific RAP addition ratio should be determined based on the mixture design requirements and the properties of the recycled asphalt mixture.

The maximum addition proportion of RAP was calculated according to Equation (4) and the results are shown in Figure 9.

From Figure 9, it can be seen that the calculated values for the maximum addition proportion without refined separation of A1 and A2 are 45% and 64%, respectively. After refined separation, the maximum addition proportion for both E1 and E2 was 100%. The above results demonstrate that the maximum addition proportion of RAP is significantly increased after refined separation. The calculated value of the maximum addition proportion of RAP can be determined by the control method. In order to ensure the stability of the quality of the recycled pavement, in the actual recycled asphalt mixture design and application, it can effectively control the variability of the recycled asphalt mixture when the addition proportion of RAP is less than the calculated maximum proportion value.

5.2. Maximum Addition Proportion of RAP under the Control of Asphalt Content Variability

In order to investigate the effect of asphalt content variability on the maximum addition ratio of RAP, the asphalt content variability control method was used to calculate the maximum addition ratio of RAP. The Technical Specification JTG F40-2004 requires that the asphalt–aggregate ratio not exceed ±0.3%, and considering the variability of new asphalt and construction processes, the variability of RAP asphalt content is controlled at ±0.2%. Referring to the calculation method of RAP gradation variability control (i.e., Table 4), the maximum addition ratio of RAP under asphalt content variability control was calculated, and the results are shown in Figure 9.

As can be seen from Figure 9, the maximum addition ratios of A1 and A2 (0 Hz) are 33% and 29%, respectively. The maximum addition ratios of E1 and E2 (55 Hz) are all 100%. Compared to the unfine separation RAP, the addition ratio of RAP after refined separation increased by 67% and 71%, respectively. The results obtained from the calculations of the asphalt content variability control method and the gradation variability control method are consistent with the fact that the refined separation process increases the maximum addition percentage of coarse RAP material in the recycled asphalt mixtures. It should be noted that in order to ensure the stability of the quality of the recycled asphalt mixture, the addition proportion of RAP should be less than the minimum of the values calculated by the two control methods, so as to effectively control the variability of recycled asphalt mixtures.

6. Conclusions

This study systematically evaluated and analyzed the effect of refined separation on aggregate gradation, aggregate properties, asphalt content, and asphalt properties of RAP. The calculation method of the maximum addition proportion of RAP material using gradation variability control and asphalt content control is also given, respectively. The following conclusions can be drawn:

(1)

The refined separation process can effectively reduce the variability of RAP and enhance the property stability of RAP.

(2)

The RAP false particles decreased significantly after refined separation and were closer to the natural mineral material. RAP agglomeration degree decreases with increasing frequency parameters of refined separation. Comparing with the I_c values of A1, A2, and A3, those of E1, E2, and E3 decreased by 6.40%, 4.30%, and 4.30%, respectively.

(3)

The asphalt content of the coarse RAP materials after refined separation decreases gradually with increasing frequency parameter. For 0–5 mm grain size RAP, the asphalt content of A3 changed less compared to that of the refined separated RAP.

(4)

Based on the quality control requirements of hot mix asphalt mixtures, the maximum blending percentage of RAP coarse material was calculated using the gradation variability control method and asphalt content variability control method, respectively. The refined separation process can significantly increase the maximum addition ratio of RAP material.

(5)

Through refined separation, it is conducive to improving the addition proportion of RAP in the recycled asphalt mixture, ensuring the quality of the recycled asphalt mixture, and further maximizing the potential value of RAP. This enhances the environmental benefits of recycling RAP and economic benefits.

Future research should focus on the application of fine RAP after refined decomposition to further maximize the utilization of RAP materials. The refined separation method of fine RAP should be improved to realize efficient asphalt–aggregate separation to reduce the variability of RAP materials. It is an important way to realize the high-value recycling of RAP. Currently, the research predominantly centers on the refined separation process of RAP materials, with limited consideration for practical applications. It is imperative to conduct in-depth research on the application of refined separation RAP.