Laboratory and Field Testbed Evaluation of the Performance of Recycled Asphalt Mixture Using High-Penetration Asphalt

Abstract

The application of recycled asphalt pavement (RAP) on a large scale is highly promoted to meet the current needs of carbon neutrality and sustainable development purposes. However, a majority of RAP mixture productions are currently relying on the restoring effects provided by the rejuvenators. Therefore, the study focuses on the feasibility of using high penetration asphalt binder (HPAB) in RAP mixture as a replacement for conventional rejuvenators. In this study, a recycled asphalt pavement mixture containing HPAB (RAP-H) was developed to resolve the cracking issue of RAP pavement in winter seasons owing to the rigid behavior of aged binders. To verify the applicability of the RAP-H mixture, the results of the quality standard test and mechanical performance test were compared with the reference RAP mixture having rejuvenator (RAP-R). Through the fatigue cracking test, by using Overlay Tester (OT) device, it was found that all specimens did not reach 93% load reduction after 1000 OT cycles, indicating a satisfied stress-bearing capacity. Additionally, the highest dynamic modulus of 27,275 MPa was found in the modified HPAB mixture, and this result is 4.4% higher than that of the reference mix. In the full-scale testbed, the long-term field applicability of the proposed approach was verified through field test construction. The measurement in practice reveals that the elastic modulus of RAP-H back-calculated from the FWD (Falling Weight Deflectometer) test is increased by more than 50% compared to RAP-R, which resulted in excellent performance characteristics of the HPAB pavement layer. In addition to the efficiency in the surface layer, an improved elastic modulus of the sub-base and subgrade layers in the HPAB section was found to be at 28.6% and 19.5%, respectively, compared to the RAP-R mix. In general, the performance of RAP-H satisfied all of the domestic and international quality and performance standards. The field test results confirmed the possibility of field application by showing performance higher than conventional recycled asphalt pavement.

1. Introduction

Following the Paris Agreement to address pollution and global warming, various efforts have been made to solve environmental issues that include reducing greenhouse gas emissions and using recycled resources. Eco-friendly construction technologies that aim to reduce greenhouse gas emissions, save resources, and promote recycling are receiving much attention in the construction sector [1]. The government also supports various policies to protect the environment; one measure is the recycling of waste asphalt concrete. This high-value-added resource uses the waste generated in the process of repairing old asphalt pavement roads [2,3,4]. Additionally, the development of sustainable asphalt mixtures containing recycled waste material such as fly-ash-based geopolymers and plastics also contributes greatly to green construction purposes [5,6]. The research confirms that the usage of properly treated by-product materials can improve the quality of asphalt mixture in both short-term and long-term performances. In South Korea, the Ministry of Land, Infrastructure, and Transport (MOLIT) and the Ministry of Environment prepared an institutional framework for expanding the production and use of recycled asphalt concrete through revisions to the laws and statutes. Since 2016, road works over a certain scale have been required to use recycled asphalt mixtures that incorporate waste asphalt concrete for more than 40% of the total asphalt mixture requirements [7,8,9,10]. The growing awareness has been influenced in accordance with the national policy to increase the recycling rate of recycled asphalt concrete aggregates to 50% by 2020 [11,12,13]. This results in extensive technology development to promote the usage of RAP on a large scale in various departments of transportation [14,15].

Reclaimed aggregates produced by crushing old asphalt concrete pavement have different aging levels and physical properties, depending on the quality of the asphalt mixtures used, the service period, environmental conditions, and traffic loads [16,17,18]. In particular, the absolute viscosity [5,19] that indicates the aging of old asphalt binders in the reclaimed aggregates in South Korea ranges from 9000 to 265,000 Poise, showing a large deviation [20]. To produce a recycled asphalt mixture with satisfactory quality, an appropriate mixing design should be presented in consideration of the absolute viscosity of the asphalt binder [13].

In general, there are two main techniques used to recycle aged asphalt binders: rejuvenators and soft asphalt binders. The first method involves the use of rejuvenators, commonly called recycling agents [4]. This additive joins the mixing process and restores the virgin microstructure of an aged asphalt binder [11]. When applied, the reclaimed aggregates containing old asphalt binders create a new morphology making it pliable for pavement [21,22,23]. In South Korea, it is specified to use rejuvenators when reclaimed aggregates used in an asphalt mix design exceed 25% [19]. Since the common practice incorporates about 30% reclaimed aggregates, the use of rejuvenators has become orthodox [19,24,25].

The second method is to use soft asphalt binders. These are defined following MOLIT guidelines as binders having a lower performance grade (PG) compared to the recovered recycled asphalt binder. This method is typically used when producing recycled asphalt mixtures having a reclaimed aggregate content of 15% or less [26]. The concept of recycling old asphalt using soft asphalt binders lies in synthesizing both low- and high-penetration binders, creating the desired workability and functionality.

Imad et al. showed that soft asphalt binders can reduce the generation and spread of cracks by effectively absorbing the crack-inducing energy generated in the recycled asphalt pavement due to loading [27]. Al-Qadi reported that the resistance to low-temperature cracking decreases as the ratio of reclaimed aggregate used in recycled asphalt concrete increases, and that this issue can be resolved by using soft asphalt binders with high penetration [27].

Especially, in practical construction, the prolonged low temperature may critically impose overwhelming brittleness issues on the RAP pavement having aged binder, leading to seriously premature cracking at an earlier age [12]. The very high stiffness of aged asphalt binder may be prone to frozen conditions which result in internal pressure under traffic loads, and thereby damaging the entire RAP pavement surface layer [2]. Therefore, the main aim of this research is to develop a new solution to resolve the rigid performance and restore the elastic behavior of RAP pavement. This research also attempts to promote the application of a large amount of RAP for sustainable development purposes [28].

Zaumanis and Mallick argued that the use of soft asphalt binders could resolve the brittleness and early cracking problems of recycled asphalt concrete because they have a lower dynamic modulus than asphalt concrete mixtures using ordinary asphalt binders [29]. Liphardt showed that soft binders need to be used in addition to rejuvenators to recover old binders of reclaimed aggregate. This study also found that the use of soft asphalt and rejuvenators could improve the production and construction workability of recycled asphalt concrete [30]. Yang argued that the use of high-strength recycled asphalt materials could improve the plastic deformation resistance significantly but tends to reduce the crack resistance slightly [2]. The use of soft asphalt binders to solve this problem showed an improved crack resistance. This study also evaluated the plastic deformation resistance according to the use of soft asphalt binders, and the test results satisfied the quality standards [31].

As described above, most of the studies on recycled asphalt mixtures using soft asphalt binders are focused on the use of additives to recover and restore mechanical and chemical performance. Although the utilization of RAP and rejuvenators in asphalt concrete combinations has been extensively investigated, there is very little research on the fabrication of HPAB in RAP mixture. Additionally, there is limited research on the effects of HPAB on the resistivity of RAP mixture to critical pavement damages such as thermal cracking and permanent deformation. Therefore, this study developed a recycled asphalt mixture (RAP-H) using high-penetration asphalt binders to increase the use of reclaimed aggregates in practical construction by conducting experimental research in both laboratory and field testbed. It is expected that the successful application of this approach can broaden the large volume utilization of RAP in practice, since the usage of RAP may currently rely on the incorporation of rejuvenators. Further, the high penetration asphalt binder can represent a vital selection for road and pavement experts, since the production of this binder can be conveniently performed in the asphalt plant.

Subsequently, the applicability of RAP-H was reviewed by comparing and analyzing the laboratory test, performance test, pavement design, and field test results with RAP-R using conventional rejuvenators. Considering the full-scale construction, the falling weight deflectometer test and the modulus back calculation process were applied to verify the actual performance of the reclaimed asphalt pavement having high penetration asphalt.

2. Materials and Methods

2.1. Materials

2.1.1. High-Penetration Asphalt Binder

Soft asphalt binders possessing a penetration value of 120 or greater are classified as high-penetration asphalt binders (HPAB). The asphalt binders are manufactured utilizing the equivalent method as traditional asphalt binders following the refining of petroleum, and they are designed to achieve an absolute viscosity of 650–850 Poise. In contrast to traditional binders that require the incorporation of rejuvenators, HPAB can be employed without any extra additives. In general, the HPAB and reference binders were fabricated from the equivalent PG64-22 type. Through material synthesizing from its great penetration, HPAB can restore the performance of aged asphalt binders when combined with RAP aggregate.

The additional asphaltene components should be uniformly dispersed for the asphalt to have a robust structure in order to recycle aged asphalt binders by restoring their functionality. An asphalt binder classified as HPAB contains aromatic-rich oils derived from petroleum that can distribute asphaltene constituents and has an affinity for old asphalt binder [30]. The HPAB utilized in this research has a viscosity value of 700 Poise and a penetrating value of 145. Additionally, the virgin asphalt binder employed as a comparison group, AP-5, had an absolute viscosity of 2286 Poise and a penetration of 71 (1/10 mm at 25 °C).

Table 1. Physical Properties of HPAB and AP-5.

Properties		Characteristic
		HPAB	AP-5
Unaged	Viscosity (60 °C, Poise)	700	2286
	Penetration (25 °C, 0.1 mm)	147	71
	Softening Point (°C)	42.6	46.1
RTFO	G*/sinδ at 58 °C (kPa)	1.148	1.348
PAV	Creep stiffness at −12 °C (MPa)	229	144.07
Creep rate m-value at −12 °C		0.321	0.413
PG		58–28	64–22

lists the physical characteristics and performance grade (PG) test results for both binders, while

Table 2. Specifications of HPAB and AP-5.

	Saturate	Aromatic	Resin	Asphaltene
HPAB	4.98	53.52	21.62	19.88
AP-5	4.26	43.76	22.53	29.45

present the specifications of both binders. Figure 1 and Figure 2 present the testing apparatus used. Moreover, the testing flowchart of the research is presented in Figure 3.

2.1.2. Reclaimed Aggregates

Recycled or reclaimed aggregates have a major impact on the quality and performance of a recycled asphalt concrete mixture. The quality of the reclaimed aggregate is important for high-penetration asphalt binders (HPAB) because the ratio of reclaimed aggregate depends on the content of the old asphalt binder and the degree of aging [19,20,27,32].

This study performed a reclaimed aggregate quality test as a preliminary experiment to examine the feasibility of HPAB. The level of aging, virgin characteristic, manufacturing circumstances, susceptibility to the climate, and many other factors all play a significant impact on the performance of recycled aggregate [33]. This research obtained 30 specimens from 30 randomly chosen regional suppliers to focus on the fundamental quality of recovered aggregate in the area. To assess the aging level, 30 samples underwent quality control examinations following the MOLIT-specified RAP aggregate quality criteria [26]. According to the test results, the mean absolute viscosity of asphalt binder was 33,600 Poise, and the mean content of aged asphalt binder was 4.24%.

Table 3. Physical properties of reclaimed aggregate.

Item	Old Asphalt Content (%)	Absolute Viscosity (Poise)
Reclaimed aggregate	4.24	34,754

shows the asphalt binder content of the reclaimed aggregate used in the study and the absolute viscosity after extraction.

2.1.3. Rejuvenator

In this research, South Korean asphalt manufacturers produced the regenerative asphalt binder AP-5, which is used to develop a reference mixture for comparison purposes with the proposed HPAB. The integrated regenerative ingredient is a bio-oil-based substance produced from renewable sources that allow the aging binder’s characteristics to be regenerated. The organic refined rejuvenator has a general viscosity of 80 cSt at 60 °C and specific gravity of 0.965. The unique bio-restorative ingredient employed in the research has an amphipathic chemical nature that enhances the solvent capacity of the maltene component even at extremely small dosages while also facilitating in dispersing of the highly polar portions [19,21,29]. Therefore, the main benefits are expected to be observed in the areas of low temperature and susceptibility to cracking by restoring the flexibility of the material.

2.2. Mixture Composition

2.2.1. Asphalt Binder Blending

Following the MOLIT (2017) [26], this research employed HPAB to conduct the mixing formulation of RAP mixes. This technique is founded on absolute viscosity [18]. The designed absolute viscosity was determined at 1900 ± 400 Poise, which is the popular value of binders provided by Korean companies. The extraction of HPAB and old asphalt binders from recycled aggregates is illustrated in Figure 4 and

Table 4. Viscosity test of HPAB mixtures.

Reclaimed Aggregate Old Asphalt Content (%)	Recycled Aggregate Asphalt Viscosity (Poise)	High-Penetration Asphalt Binder Viscosity (Poise)	Asphalt Binder Ratio (%)		Target Viscosity (Poise)
			Old	HPAB
4.24	34,754	700	26	74	2156

. By mixing the HPAB and the aged binder following the specified blending proportion, the desired viscosity was obtained [7,24,34].

2.2.2. Combined Aggregate Grading of the Target Mixtures

In terms of the particle size of the reclaimed aggregates, this study set the application ratio of the mixture up to 35% considering the old asphalt binder content of reclaimed aggregate. The combined particle size of the recycled asphalt mixture is shown in Figure 5 and

Table 5. Particle size distribution of aggregate in recycled asphalt mixture.

	Percent Passing (%)
Sieve size (mm)	20	13.2	9.5	4.75	2.36	0.60	0.30	0.15	0.08
Combined gradation	100	99.7	92.1	84.3	61.3	32.8	24.3	12.1	6.36

[35].

2.2.3. Computing the Mixing Proportion of Recycled Asphalt

The link between the particle size of aggregate and the asphalt concentration of the RAP mix was determined following Equation (1). Based on the calculation, 4.6% was found to be the acceptable asphalt proportion in the research [36].

$$P_b=0.035a+0.045b+Xc+F$$

(1)

where:

$P_b$ (%): Estimated binder volume;

$a$ (%): Ratio of aggregate that was contained in 2.5 mm;

$b$ (%): Ratio of aggregate that passes through 2.5 mm and was contained in 0.08 mm;

$c$ (%): Ratio of aggregate that passes through 0.08 mm;

$X$: determined according to the value of $c$, $X$ = 0.2 when $c$ $\le$5%, $X$ = 0.18 when 6 $\le$ $c$ $\le$ 10%, and $X$ = 0.15 when 11% $\le$ $c$ $\le$ 15%;

$F$: is calculated by the absorption ratio of aggregate, 0.7%.

$$P_{nb}=\frac{\left({100}^2-r{P}_{sb}\right)P_b}{100\left(100-P_{sb}\right)}-\frac{(100-r)P_{sb}}{100-P_{sb}}$$

(2)

where:

$P_{nb}$ (%): New asphalt content added in RAP mix;

$P_b$ (%): Estimated asphalt content in RAP mix;

$P_{sb}$ (%): Asphalt content in the recycled aggregate;

$r$ (%): Ratio of recycled aggregate in 100 of total aggregates.

Table 6. Mix Design of reclaimed asphalt mixture.

Mix Type	Aggregate (%)		Asphalt Binder (%)		Filler	Total (%)
	Recycled	New	Old + HPAB	Old + AP-5
RAP-H	29.5	62.5	4.6	-	3.4	100
RAP-R			-	4.6

shows the final mixture design of recycled asphalt binder calculated using Equation (2) considering the estimated asphalt content derived using Equation (1), the binder content of the reclaimed aggregate, and the ratio between recycled and virgin aggregates [36].

The equivalent mix design was applied to the mixture using rejuvenators to compare the effectiveness, and the amount of rejuvenator mixed in was 7% of the amount of AP-5.

2.2.4. Asphalt Mixture Laboratory Test

For the Marshall stability test commonly used in mix design, the compaction direction when fabricating the specimens and the loading direction of the test are perpendicular to each other (see Figure 6). Moreover, the properties of the mixtures were evaluated by performing indirect tensile strength and tensile strength ratio tests.

Table 7. Standard reference values for recycled asphalt mixtures.

Test Item	Test Method (Korea Standard, KS)	Standard Value (Korea Standard, KS)
Stability (N)	KS F 2337 [37]	Over 5000
Flow (1/100 cm)	KS F 2337 [37]	20~40
Air void (%)	KS F 2364 [38]	3~6
Voids Filled with Asphalt (%)	-	65~80
Voids in Mineral Aggregate (%)	-	Over 14.0
Tensile Strength Ratio	KS F 2398 [39]	Over 0.75
Indirect Tensile Strength (N/mm2)	KS F 2382 [40]	Over 0.80
Toughness (N.mm)	KS F 2382 [40]	Over 8000
Dynamic stability	KS F 2374 [41]	Over 750
After viscosity (Poise)	KS F 2381 [42], KS M2247 [34]	Under 5000

shows the test methods and standard test range values for recycled asphalt mixtures in South Korea.

2.3. Performance Test

2.3.1. Hamburg Wheel Tracking Test

Rutting and moisture degradation were evaluated using the HWT test. This experiment determines the susceptibility of the mixture to an early failure caused by inadequate aggregate–binder bonding, inadequate aggregate packing, excessive moisture, and inappropriate binder rigidity. The AASHTO T 324 [43] guideline outlines the methods for performing the HWT test. According to the specification, a wheel load having a weight of 705 ± 4.5 N must be moved throughout the top of samples (slab/cubical) that are soaked in a steam bath that is kept at a consistent heat of 50 °C. The apparatus has the capacity to examine two samples at once. The wheels fluctuate at 52 ± 2 passes per min and comprise a size of 203 mm and a width of 47 mm. The wet HWT assessment in this research is carried out under identical circumstances as those described in AASHTO T 324.

2.3.2. Crack Resistance Test

The Overlay test (OT) is a simple performance test used to characterize the crack reflection potential of asphalt mixes in the laboratory. The OT evaluation was conducted at 25 °C using specimens following TxDOT Tex-248-F [44]. The number of cycles to failure was obtained to determine the cracking resistance potential. The amount of OT loops needed to generate a 93% reduction in the original load as determined within the initial cycle was used to determine the number of iterations until breakdown. The experiment was automatically terminated if a 93% drop in the starting load was not achieved in the defined highest quantity of repetitions. The test was set to run for a total of 1000 loading cycles until being terminated.

2.3.3. Dynamic Modulus Test

A crucial factor in quality analyses and road depth planning is the dynamic modulus. Based on AASHTO T342-11, the experiment was examined at standard temps of 10, 5, 20, 40, and 54 °C and frequencies of 25, 10, 5, 1, 0.5, and 0.1 Hz [45]. The results were fitted utilizing the sigmoid function to produce the master curves of the dynamic modulus, as shown below:

$$\log \left|E^{*}\right|=\delta +\frac{\alpha }{1+e^{\beta -\gamma log\left(f_r\right)}}$$

(3)

where $\left|E^{*}\right|$ the dynamic modulus (MPa), $\delta$ = the minimum value of $\left|E^{*}\right|$, $\delta$+ $\alpha$ = the maximum value of $\left|E^{*}\right|$, $\beta$ and $\gamma$ = the regression parameters describing the shape between asymptotes and inflection points, and $f_r$ = the reduced frequency (Hz).

$$f_r=\tau +a_T$$

(4)

where $\tau$ is the frequency at the specified temp (Hz), $a_T$ is the shift factor that transforms the frequency at the temp measured to the frequency at the reference temperature.

$$\mathrm{og}\left|a_T\right|=\delta +\frac{C_1\left(T-T_d\right)}{C_2+\left(T-T_{ref}\right)}$$

(5)

where:

• $C_1$ and $C_2$ are the regression parameters,
• $T$ is the measured temp.
• $T_d$ is the defined temp.
• $T_{ref}$ is the reference temp. (20 °C).

3. Field Test

3.1. Field Test Outline

For this study, field testing was conducted to evaluate the applicability of the developed mix design materials. The field test construction was carried out on a four-lane road in the metropolitan city of Daegu, South Korea. Figure 7 shows the test section schematic top view image and the cross-sectional diagram of each pavement layer. The field test sections include two main pavements: RAP pavement having HPAB, and AP-5 with rejuvenator. Each section was constructed at 300 m long and 3.5 m wide. The thickness of the surface layer, base layer, subbase layer, and subgrade is 5 cm, 10 cm, 20 cm, and 100 cm, respectively. The field test was conducted in the early wintertime in South Korea (Late November). Figure 8 shows the actual pavement construction for the field test.

3.2. Falling Weight Deflectometer Test

As shown in Figure 9, this study used the FWD test to evaluate the pavement structure after field test construction. The analysis was performed to compare the structural characteristics of each section. This is considered the non-destructive testing device used to evaluate the physical properties of pavement structures by subjecting the surface to an impact load and obtaining the surface deflection bowl measured by a series of sensors installed on the pavement surface. The standard load is 4082 kgf, and the diameter of the load plate is 30 cm. When a dynamic load is applied, the geophones measure the deflection at each location [46,47,48]. The deflection bowl is calculated from the measured FWD deflections, Dr, at geophones positioned at different offsets, r in mm, from the center of loading. Deflections resulting from dropped weights such as 4082 kgf are measured at these offsets [49]. These discrete measuring points are illustrated in Figure 10. The discrete measured points on the deflection bowl enable the calculation of deflection parameters describing various zones or areas of the whole deflection bowl [50]. In this study, the FWD test was performed at 20 m length intervals for each type of recycled asphalt mixture lane. The ambient temperature was about 9 °C, and the surface temperature of the asphalt concrete layer was about 10 °C when the field test was performed in Daegu, Republic of Korea.

FWD Analysis

In this study, the pavement structure was analyzed using the most general equations for pavement structure evaluation [50,51]. In addition, it was evaluated by dividing the structural condition grade according to

Table 8. Parameters for the functional level of the pavement.

	Curvature Index (μm)
Structural Condition	Maximum	RoC	SCI	BDI	BCI
Sound	Lower than 500	Lower than 100	Lower than 200	Lower than 100	Lower than 50
Warning	500$~$750	50$~$100	200$~$400	100$~$200	50$~$100
Severe	Higher than 750	Higher than 50	Higher than 400	Higher than 200	Higher than 100

[52]. The most frequently used FWD data to analyze the quality of the pavement are the deflection in the center of the load plate ($D_0$) and the AREA, which is an analysis of the deflection basin characteristics. Equation (6) shows how to calculate the AREA:

$$\mathrm{AREA}=6\left(1+2\frac{D_{300}}{D_0}+\frac{D_{600}}{D_0}+\frac{D_{900}}{D_0}\right)$$

(6)

where $D_{300}$: Deflection at a distance of 300 mm from the middle of the plate (μm), $D_{600}$: Deflection at a distance of 600 mm from the middle of the plate (μm), and $D_{900}$: Deflection at a distance of 900 mm from the middle of the plate (μm).

The integrity of the surface layer is evaluated by the surface curvature index ($SCI$), while this property of the base layer was termed the base curvature index ($BCI$), and that of the sub-based is assessed by the base damage index ($BDI$). The displacement increases as each index’s value increases.

Equations (7)–(9) show how to calculate each index [50,51]:

$$SCI=D_0-D_{300}$$

(7)

$$BDI=D_{300}-D_{600}$$

(8)

$$BCI=D_{600}+D_{900}$$

(9)

Reflecting the structure behavior of the pavement layers, the radius of curvature ($RoC$) is determined using the formula in Equation (10) below:

$$RoC=\frac{L^2}{2D_0\left(1-\frac{D_{200}}{D_0}\right)}$$

(10)

where, $L$ is designed at 200 (mm), and $D_{200}$ represents the deformation at 200 mm from the load plate center (μm).

Moreover, an inverse analysis was performed using Modulus 6.0. The inverse analysis was performed by using the thickness of each layer and the assumed elastic modulus range of each layer to calculate the deflection based on the multilayer elastic theory. The calculations were repeated by adjusting the elastic modulus for each layer until the calculated deflection matched the actual measured results. From this process, the elastic modulus of each layer can be accurately estimated.

4. Laboratory Test Result

4.1. General Analysis

Based on the results of testing for each mix type, all of the mixtures satisfied the quality standards in South Korea.

Table 9. The property test result of RAP-H and RAP-R.

Test Item	Characteristic		Standard Value [34,37,38,39,40,41,42]
	RAP-H	RAP-R
Stability (N)	10,580	14,044	Over 7500
Flow (1/100 cm)	36	35	20~40
Air void (%)	3.8	4.1	3~6
Voids filled with asphalt (%)	76.1	77	65~80
Voids in mineral aggregate (%)	14.7	17.7	Over 14.0
Tensile strength ratio	0.84	0.78	Over 0.75
Indirect tensile strength (N/mm2)	1.52	1.4	Over 0.80
Toughness (N.mm)	16,315	14,790	Over 8000
Dynamic stability (cycle/mm)	1983	2.831	Over 750
Final viscosity (Poise)	4350	4122	Under 5000

shows the property test results of each asphalt mixture. The laboratory tests for RAP-H and RAP-R showed that all of the mixtures meet the quality of South Korean standards, including Marshall stability, flow, porosity, saturation, VMA, and VFA. Further, both the indirect tensile strength and toughness test results attained the required standard value. It is defined that HPAB contributes to the dispersion of the asphaltene component of the aged binder from the reclaimed aggregate. Asphalt binders can be represented by a maltene component that is soluble in n-hexane and an asphaltene component that is not dissolved. However, in the aged asphalt binder, the maltene component is reduced and the asphaltene component is increased due to the phenomena such as oxidation and polymerization. In order to restore the performance of the aged asphalt binder and reuse it, the increased asphaltene components must be dispersed again so that the asphalt mixture can have a stable structure [27,29,30]. In the case of HPAB, compared to AP-5 with rejuvenator it is considered to have played a sufficient role in restoring the performance of the aged asphalt binder of the reclaimed aggregate, as it is composed of relatively more maltene components and less asphaltene than AP-5.

4.2. Performance Test Result

4.2.1. HWT Test

Figure 11 shows the HWT test results for the RAP-H and RAP-R mixtures. In general, the State Transportation Authority of the United States stipulates the use of an asphalt mixture that loads 20,000 cycles. Moreover, stripping should occur after at least 10,000 cycles and does not exceed 20 mm of settling at the end of the cycles [25,31]. In both asphalt mixture types, it was found that the settlement did not exceed 20 mm after 20,000 cycles. The final settlement of RAP-H and RAP-R was 16.2 mm and 11.5 mm, respectively.

As shown in Table 9, both RAP-H and RAP-R satisfied the Korean Standard, but the Marshall stability of the RAP-R mixture obtained about 32% higher compared to the RAP-H mix. Based on the data from the Marshall stability and HWT test, it seems that the RAP-R mix has a slightly higher resistance to rutting and settlement compared to the RAP-H mixture. From the test results it was determined that the moisture resistance was relatively low, and this should be improved by using anti-stripping agents.

4.2.2. Crack Resistance Test

Figure 12 and

Table 10. Result of crack resistance test.

Mix Type	Number of Cycles	Reduction in Load (%)
RAP-H	1000	76.1
RAP-R	1000	75.6

illustrate the results of the fatigue cracking tests. The reliability of the fatigue cracking test has been confirmed through field application after development and is being applied as a quality standard for asphalt concrete pavement in several states, including the Texas Highway Administration [47,48,49]. In general, both mixtures share the equivalent trend in the load reduction curve. The result of the test on reduction in load for RAP-H and RAP-R mixes was 76.1% and 75.6%, respectively. The RAP-R mix was found to have a small decrease in load after 1000 repeated loading cycles compared to the RAP-H mixture, but the difference was insignificant at 0.5%. Through the fatigue cracking test it was found that all specimens did not reach 93% load reduction in 1000 cycles, satisfying the test criteria. Therefore, both RAP-H and RAP-R mixes are appropriate for application on the surface layer of asphalt concrete pavement.

4.2.3. Dynamic Modulus Test Results

The test results of RAP-H and RAP-R mixtures and the correlation between load frequency and dynamic modulus at each temperature is shown in Figure 13 and Figure 14. Similar to viscosity, load frequency is a variable that has a strong impact on the dynamic modulus. As a result of the experiment it was determined that the dynamic modulus increased according to the increase in frequency (from 0.1 Hz to 25 Hz). Moreover, the dynamic modulus of elasticity measured by frequency was found to be higher in mix RAP-H. In the dynamic modulus relationship according to temperature, it is reasonable to find a decline in values as the temperature increases. By using the load frequency and dynamic modulus graph for each temperature, the master curve is developed as shown in Figure 14. In addition, the highest dynamic modulus of the sample using the HPAB was discovered at the frequency range of 2.5 to 5.0 Hz, the usual low-speed traffic loading range.

4.3. FWD Test Result

4.3.1. Deflection Basin Analysis (D0/AREA)

Figure 15 shows the trend of FWD deflection according to the field test construction sections. As shown in Figure 15a, the deflection data for each section show significant differences at locations close to the load plate, and the gap in the deflection data between two geophones becomes smaller as the gap from the load plate increases, ultimately resulting in convergence. Figure 15b compares the deflection loading plate D0 data for each section. Based on the analysis of the data, the D0 range of the RAP-H sections was 200~350 $\mathsf{\mu }\mathrm{m}$, and the range of the RAP-R sections was about 200~400 $\mathsf{\mu }\mathrm{m}$. Thus, the deflection loading plate D0 of RAP-H was noticeably smaller than RAP-R, demonstrating a strong settlement resistance. These results indicate that the RAP-H sections have better pavement structural characteristics. However, the AREA values in Figure 15c do not show significant differences according to each section. In general, AREA is a deflection basin characteristic that increases as the depth of the pavement layer increases, and likewise as the stiffness of the upper layer increases. Therefore, the overall pavement characteristics (including the subgrade) are considered to be excellent if the D0 is relatively small and the AREA is large. However, since the test results only showed a significant difference in D0 and almost no difference in AREA, there are limitations in terms of distinguishing the superiority of each section’s structural characteristics through this analysis.

4.3.2. Layer Index Result

Figure 16a compares the SCI by using the deflection data. SCI is an index representing the structural condition of the surface layer. Generally, the structural condition rating is classified as a warning range for an SCI value from 200 to 400, and a severe zone for a value higher than 400. Both mixtures were considered to have decent structural conditions of the surface layer, since all measured locations exhibited SCI values lower than 150. Especially, the RAP-H sections have very low SCI, indicating the characteristics of the base layer are relatively superior compared to the conventional RAP-R with a rejuvenator. It may be attributable to the incorporation of HPAB helping restore the performance of the aged binder. Meanwhile, the combination of conventional rejuvenators may partly replenish the maltene component in old asphalt.

The BDI index provides the status of the structural condition of the sub-base layer as shown in Figure 16b. In general, the BDI of the RAP-H sections belongs to the sound category which outperforms the BDI of the RAP-R construction method. In the RAP-R sections, the BDI exceeds 100 at several points, showing that the structural condition of the sub-base layer is classified as a warning zone having a value of 100 ≤ BDI ≤ 200. Additionally, compared to the BDI findings, Figure 16c indicates an equivalent trend in the BCI index of subgrade condition since this value of RAP-R reaches the warning zone at multiple points, demonstrating that the structural condition of the subgrade layer is relatively poor.

Finally, the RoC comparison between the RAP-H and RAP-R sections is shown in Figure 16d. At almost all counting zones the radius of curvature of RAP-H sections shows a greater value compared to RAP-R sections. Therefore, the test results verify the effectiveness of reinforcement in structural conditions of the surface by using the newly developed solution.

4.3.3. Back Calculation Result

This study performed inverse analysis using FWD deflection data. Figure 17a compares the results of analyzing the elastic modulus of the asphalt pavement layer. The average elastic modulus of the surface layers in the RAP-H and RAP-R sections was 4560 MPa and 3035 MPa, respectively. This shows that the performance of the asphalt pavement layer in the RAP-H sections was better compared to the section having RAP-R. The higher elastic modulus recorded from RAP-H pavement also confirms the restoration of aged asphalt binder by using HPAB. Within an appropriate mixing ratio it can be concluded that HPAB can disperse well in the old asphalt, restoring its maltene component and significantly contributing to the original viscous-elastic characteristic of the asphalt binder.

The elastic modulus of the sub-base layer is shown in Figure 17b. In general, the elastic modulus of the sub-base layer in the RAP-H and RAP-R sections was ranging from 200 to 400 MPa, except for some data. The elastic modulus of the sub-base and subgrade layer confirms the enhancement efficiency by using HPAB on RAP mixtures. Based on the findings from Figure 17b,c, an improved average elastic modulus of the sub-base and subgrade layers for RAP-H was found to be at 28.6% and 19.5%, respectively, compared to the RAP-R mix. The enhancement effectiveness was minimized in the sub-grade layer, and it can be explained by the stress dissipation ability provided by the asphalt concrete layers.

In terms of summarizing the indices analyzed by using the FWD data and the results of the inverse analysis, the physical characteristic of the asphalt pavement layer (surface layer + base layer) of the RAP-H sections outperform the RAP-R. The properties of the lower layers in the RAP-H sections also show greater strength compared to the section having a rejuvenator. Therefore, the application of HPAB in RAP mixture is very prominent since this proposed solution not only recovers the performance of aged asphalt binder, but also contributes to the strength of the whole pavement structure and provides a simple construction process.

5. Conclusions

This study performed quality tests through laboratory mix design of high-penetration asphalt (HPAB) mixtures having reclaimed aggregates. Then, the field construction of RAP pavements was performed to verify the effectiveness of reinforcement between the newly developed solution with high penetration asphalt (RAP-H) and the conventional asphalt binder having rejuvenator (RAP-R). The main findings are as follows:

• The laboratory test results showed that RAP-H satisfied South Korean quality standards for recycled asphalt pavement in terms of Marshall stability, flow, porosity, saturation, VMA, and VFA. Moreover, HPAB with relatively more maltenes and fewer asphaltenes presented effectiveness in dispersing existing asphaltenes in aged asphalt binder found in RAP, thus achieving a more stable and durable mixture;
• According to the Hamburg wheel tracking test results, the settlement of RAP-H was 16.2 mm after 20,000 loading cycles, which satisfied the settlement standard of the State Transportation Authority of the USA (20 mm or less). Regarding the fatigue test results, RAP-H satisfied the criteria by only showing a 0.5% difference from the conventional RAP-R;
• The results of the dynamic modulus test indicate that the dynamic modulus of RAP-H was higher than that of RAP-R. Especially, the dynamic modulus was highest in the low-speed traffic load range (from 2.5 to 5.0 Hz), showing excellent durability against wheel load. The usage of RAP mixture having high penetration asphalt may be more effective in intersection zone application;
• There were no difficulties during the construction of the surface layer paved by the RAP-H in the pilot test. According to the FWD test results, the application of RAP-H had affected the increment of the elastic modulus on overall asphalt concrete layers, showing an above-equal effect compared to asphalt concrete layers using RAP-R. It may be attributed to the incorporation of HPAB, which helps restore the performance of the aged asphalt binder by balancing its chemical components namely: maltenes and asphaltenes.
• In summary, the performance of RAP-H satisfied all of the domestic and international quality and performance standards. The field test results confirmed the possibility of field application by showing performance equivalent to or higher than conventional recycled asphalt pavement. However, the usage of high-penetration asphalt binders should be verified through additional curing and production conditions. There are considerable regional and national discrepancies in the preservation conditions of high-penetration asphalt (i.e., temperature). Accordingly, evaluating various characteristics of production and storage settings from various areas functioned as a sub-objective of future work. In addition, future studies need to include long-term monitoring and durability tests through various field applications before conducting large-scale construction in practice.