Experimental Investigation and Statistical Analysis of Recycled Asphalt Pavement Mixtures Incorporating Nanomaterials

Abstract

Reclaimed Asphalt Pavement (RAP) materials are used as substitutes for new materials in asphalt pavement construction, leveraging the engineering and commercial benefits of the aged binders and aggregate matrixes in RAP. These asphalt mixtures impart significant variations in volumetric properties and asphalt mixture characteristics. The current study investigates the Marshall properties, moisture susceptibility, and rutting behavior of 24 recycled asphalt mixtures developed with nanosilica and nanoclay. RAP material percent, nanomaterial content, binder grade, and extra binder were considered the factors influencing asphalt mixture performance. The above factors were analyzed using the Response Surface Methodology (RSM) to predict the Marshall and volumetric properties. Also, this investigation covers the moisture susceptibility and rut characteristics of recycled nanomaterial-modified Hot Mix Asphalt (HMA) and Warm Mix Asphalt (WMA) mixes developed with Viscosity Grade 30 (VG-30) and Polymer-Modified Bitumen-40 (PMB-40). The chemical additive Zycotherm was used to develop WMA mixes. The test results indicate that adding RAP material at higher percentages and modifying the binder with nanomaterials affected moisture susceptibility with reduced moisture damage. Recycled nanosilica-modified HMA mixes developed with PMB-40 at higher RAP percentages reported higher tensile strength ratio (TSR) values in contrast with VG-30 mixes, indicating their greater susceptibility toward moisture-induced damage. The rutting potential of all of the recycled asphalt mixture combinations was enhanced by densely packed aggregate structures optimized with nanomaterials, total binder content, and RAP materials developed using the Marshall method. Overall, the nanosilica-modified recycled asphalt mixes developed with PMB40 at higher RAP percentages showed better performance in terms of strength and durability.

1. Introduction

Bitumen, a vital material in road construction, is predominantly obtained through the fractional distillation of crude oil or as a residue from the refining process of crude oil [1,2]. At room temperature, bitumen typically ranges from solid to semi-solid phases [3,4]. The performance of bitumen significantly influences pavement durability, with many studies emphasizing the impact of its rheological properties on asphalt mixtures [5,6]. Temperature susceptibility is one notable limitation of bitumen in flexible pavements. High temperatures cause bitumen to become highly flowable, making it prone to rutting, while at low temperatures, it becomes brittle, leading to cracking [7,8,9].

To address these limitations, conventional bitumen is often mixed with various nanomaterials to boost its physical, rheological, stability, and durability properties in asphalt mixtures [10]. The application of nanotechnology in asphalt mixtures is promising because of the unique morphological characteristics of nanomaterials, such as high surface area, surface free energy, and superior dispersion ability in asphalt materials [11]. The addition of nanomaterials enhances bitumen components at the nano-level, rejuvenating saturates, naphthenic aromatics, polar aromatics, and asphaltenes, which consist of higher molecular weight phenols and heterocyclic compounds affected by various polymers through chemical chains [12,13].

Among the nanomaterials, layered silicate nano clay has emerged as a novel material with the potential to improve asphalt binders’ performance at the macroscale [14]. Studies have shown that adding nano clay improves the rheological responsiveness of asphalt binders, enhances high-temperature performance, and mitigates cracking at intermediate temperatures [15,16,17]. However, some studies have also reported that nano clay can shorten the fatigue durability of asphalt binders [18,19].

The incorporation of nanoparticles into asphalt mixtures has been extensively researched, demonstrating improvements in rheological, engineering, and physical characteristics [20,21,22,23]. These enhancements are attributed to the improved adhesion and cohesion between the asphalt binder and nanoparticles, which prevent crack formation and extend fatigue life while reducing the likelihood of rutting failure [24,25,26,27].

Despite extensive research on nanomaterials, there is a gap in studies examining the combination of nanomaterials with Reclaimed Asphalt Pavement (RAP) in Hot Mix Asphalt (HMA) or Warm Mix Asphalt (WMA). RAP usage ranges from 20 to 30%, with previous studies suggesting that recycled asphalt mixture performance is equivalent to conventional mixtures when RAP content is below 15%. Efforts to produce asphalt mixtures with 100% RAP have been made, but maintaining long-term pavement performance is feasible with up to 30% RAP in the surface layer [28,29,30].

Balancing RAP content and its effect on asphalt blend performance is crucial. The efficiency of combining old and virgin asphalt significantly influences the performance of recycled asphalt mixtures [31,32]. Molecular diffusion of old asphalt from RAP particles into new asphalt enhances recycled asphalt mixtures’ performance [33,34]. However, quality control issues and material heterogeneity pose challenges to RAP utilization [35,36].

In summary, extensive research has focused on applying nanomaterials in asphalt modification, particularly nano-silica and nano-clay with virgin and modified binders. However, more research is needed on combining nanomaterials with RAP material in HMA or WMA. Thus, it is essential to investigate these materials’ mechanical, engineering, and deformation performance to design sustainable asphalt mixtures. This investigation employs a Response Surface Methodology (RSM) approach to assess the variability in mechanical and volumetric characteristics of recycled asphalt mixtures with nanomaterials. By analyzing the interactions between the relative proportions of component elements (RAP, nanomaterial, and additional binder), the investigation aims to provide a complete empathetic factor influencing the performance of recycled asphalt mixtures. Statistical analysis ranks the variables affecting these attributes, followed by a detailed explanation of the materials, mixture, specimen preparation, testing procedures, and analyses of findings and critical conclusions.

Response Surface Method (RSM)

RSM employs polynomial functions to relate multiple independent variables to a dependent variable or outcome. This set of numerical and statistical techniques is particularly useful in optimizing system performance through experimental design and analysis [37,38]. A standard second-order response surface model used in the current research activity is shown in Equation (1).

$$y={\beta }_0+{\sum }_{j=1}^k{\beta }_j{X}_j+{\sum }_{j=1}^k{\beta }_{jj}{\left(X_j\right)}^2+{\sum }_{i>j}^k{\beta }_{ij}\cdot X_i{X}_j$$

(1)

where

y is the outcome or response;

β₀, β_j, β_jj, and β_ij are regression coefficients for intercept, linear, quadratic, and interaction terms, respectively;

X_i and X_j are the independent variables.

The Design of Experiments (DoE) method is a crucial tool used in RSM to select combinations of the independent variables. These variables serve as input parameters for the RSM. Before conducting any physical experiments, the range and values of the input parameters are determined to minimize experimental work [39]. The effectiveness of the concluded RSM largely rests on the sampling strategy, which involves selecting specific sites for testing.

Central Composite Design (CCD) is extensively used among various sampling techniques owing to its several advantages [40]. One significant benefit of CCD is its annexation of axial points, which allows for estimating the RSM curvature. Additionally, fewer critical points are needed when experimental errors are predicted, enhancing the accuracy of the developed models.

RSMs have been applied in various contexts within the field of asphalt materials. For instance, they have been used to optimize the incorporation of RAP in WMA and HMA [41] and to evaluate the impact of gradation and lime on the strip-off potential of asphaltic mixtures [42]. These applications underscore the versatility and effectiveness of RSM in optimizing and improving the performance of asphalt mixtures through systematic experimentation and analysis.

2. Materials and Methodology

2.1. Materials

2.1.1. Asphalt Binder

The Viscosity Grade (VG-30) asphalt binder was obtained from Hindustan Colas Limited located in Mumbai, India. In India, VG-30 binder is typically utilized for flexible pavement construction. The physical characteristics of VG-30 are presented in

Table 1. Physical properties of the binder.

Property Tested	Results Obtained		IS 73-2013 Requirements for VG-30 Bitumen
	VG-30	RAP
Penetration at 25 °C	62	47	45 Minimum
Softening point	58	60.2	47 Minimum
Flash Point, °C	310	--	>220
Ductility at 25 °C	75	67	75 Minimum
Specific gravity at 27 °C	1.01	1.07	0.99 Minimum
Kinematic viscosity at 135 °C	452	551	350 Minimum

confirming IS:73-2013 [43].

2.1.2. Warm Mix Asphalt Additive

Zycotherm, a silane-based technology, was used to produce WMA combinations. Compared to other amine-based additives, Zycotherm appears to be more efficient for the water-repellent molecular hydrophobic zone. The silane functional groups in Zycotherm form covalent bonds with hydroxyl groups present on the surface of aggregates and in the asphalt binder. Specifically, the silane groups can bond with silanol groups (Si-OH) on aggregate surfaces (the inorganic phase) and with polar functional groups in the asphalt binder (the organic phase) [24,44]. During hydrolysis in the presence of water, the organic portion of the silane molecule (referring to the hydrocarbon chains attached to the silicon atom) condenses and converts into hydrophobic siloxanes, making the bonding highly resilient to moisture conditions. After hydrolysis, the inorganic part of the silane molecule, specifically the silicon-oxygen (Si-O) portion forms hydrogen bonds with hydroxylated agents on the surfacee of stones. A covalent bond creates a cross-linked siloxane (Si-O-Si) film structure at high temperatures.

Table 2. Physical properties of Zycotherm.

Properties	Test Results
Specific gravity	0.97
Appearance	Liquid
Color	Pale-yellow
Flash point	90 °C
Freezing point	5 °C

illustrates the properties of Zycotherm.

2.1.3. Natural Aggregate and RAP Material

The RAP material was gathered from Yelahanka, Bengaluru–Hyderabad connecting route, which runs through NH-44. The top 40 mm of the pavement was removed using the controlled milling technique after six years of service since this road needed to be resurfaced due to potholes and other distress. The asphalt content, aggregate gradation, and aggregate quality were measured to characterize the RAP material. The asphalt content in RAP material was 3.47%, determined through the centrifuge extraction technique; the aggregates from RAP were collected separately. RAP aggregates were subjected to tests such as crushing, impact, abrasion, specific gravity, water absorption, and gradation analysis in accordance with MoRTH, 2013 [45]. The water absorption of natural aggregates was at a rate of about 1.2%, which is less than the threshold value, indicating that an anti-stripping agent was not required to develop the asphalt mixture.

Table 3. Physical properties of natural and RAP aggregates.

Property	Natural Aggregate	RAP Aggregate
Specific gravity	2.65	2.63
Los Angeles abrasion (%)	20	23
Impact value (%)	18	22
Crushing value (%)	20	23
Combined F&E (%)	21	26
Water absorption (%)	1.2	1.8

presents the physical properties of the aggregates tested in the study presented herein. Similarly, Figure 1 presents the gradation of natural RAP aggregate material.

2.1.4. Nanosilica (NS)

Nanosilica was procured from local vendors in Yelahanka, Bangalore.

Table 4. Physical properties of nanosilica.

Description	Purity	Surface Area (m2/g)	Appearance (Color and Form)	Melting Point	Density (g/cm3)
Remark	>82%	325	White and crystal	1700	2.1

and

Table 5. Chemical composition of nanosilica.

Constituent	SiO2	Al2O3	Fe2O3	MgO	CaO	TiO2	K2O
%	99.86	0.02	0.05	0.04	0.03	-	-

represent its physical and chemical properties, respectively. The increased contact surface between asphalt and nanomaterial is prompted by the layered and even surface shown in Scanning Electron Microscope (SEM) micrographs, as illustrated in Figure 2a. Figure 2b illustrates the Energy-Dispersive X-ray (EDX) results and Figure 2c shows the X-ray Diffraction (XRD) analysis results which were used to study elemental analysis results.

Asphalt mixture with nanomaterial has better resistance to moisture sensitivity compared to control mixes. The precipitates are primarily composed of calcium hydroxide and calcium silicate hydrogen (C-S-H). The Bitumen bond strength test suggests that the additional nanomaterials can strengthen the asphalt mastic’s binding and increase its resistance to moisture damage. Furthermore, asphalt’s adhesive property is strengthened when its size is reduced from the micron to the nanoscale; as a result, it can significantly increase resistance to moisture damage. The utilization of nanosilica in any mix exhibits good dispersion because of its high surface area, pozzolanic activity, and adsorption capacity.

2.1.5. Nanoclay (NC)

Nanoclay was procured from local vendors in Yelahanka, Bangalore.

Table 6. Physical properties of the nanoclay.

Description	Montmorillonite (%)	Colour	pH	Liquid Limit (%)	Plastic Limit (%)	Shrinkage Limit (%)
Remark	>80	Grey	9.5	280	45	29

and

Table 7. Chemical composition of the nanoclay.

Constituent	Al2O3	SiO2	Fe2O3	CaO	MgO	TiO2	K2O
%	21.1	69.9	2.58	0.59	4.23	0.49	0.34

list the physical and chemical characteristics. Figure 3a shows the morphological features of clay particles examined using Scanning Electron Microscopy (SEM) that reveal the amorphous phase with a greater surface area prompts improved adhesion properties for asphalt material and nanoclay. Figure 3b illustrates the XRD elemental analysis and Figure 3c depicts the EDX analysis, which exhibits the bond behavior between asphalt aggregate matrixes enhanced by traces of elements.

Nanoclay0Modified Bitumen (NCMB) and Nanosilica-Modified Bitumen (NSMB) samples were prepared by adding the optimized 4% percentage of nanomaterials to heated bitumen [46]. To ensure the uniform mix of nanomaterial into asphalt, a high shear mixer with a speed of 4000 rpm was utilized without lowering the asphalt’s temperature. Warm Mix Asphalt (WMA) and Polymer-Modified Bitumen (PMB) of grade 40 were the other two different types of binders used for the study. Warm Mix Asphalt (WMA) was prepared by adding the chemical additive Zycotherm at 0.1% by weight of bitumen.

Table 8. Physical properties of modified binders.

Properties	NCMB	NSMB	WMA	PMB
Penetration at 25 °C	51	59	67	46
Softening point, °C	58	57	54	55
Ductility at 25 °C	81	85	70	72
Kinematic Viscosity at 135 °C	775	763	471	540

represents the physical characteristics of the various binders used in the work.

2.2. Methods

To assess the uncertainty and sensitivity of the recycled asphalt blends, a number of experimental investigations and theoretical analyses were conducted. The steps involved the determination of input parameters affecting the mechanical and volumetric properties using the DoE approach to generate a test matrix with variation in input parameters. Then, we carried out the preparation of the asphalt mixtures and tested their mechanical and volumetric properties to develop Response Surface Models (RSMs) and use the statistical analysis to determine which parameters affect the mechanical and volumetric properties of the recycled asphalt mixtures developed in the study presented herein. Figure 4 displays the flowchart related to the current study.

Four input characteristics that have a substantial impact on the RAP quality were found based on the literature. These are the following percentages: percentage of total binder (%TB), percentage of nanomaterial, percentage of RAP (%RAP), and percentage of extra binder. If all other factors are the same, a rise in binder viscosity makes specimen compaction more challenging. Comparably, higher RAP content causes higher binder viscosity, which in turn causes more compaction effort. Blend viscosity decreases with the increase in extra binder content if the other parameters stay unchanged [33]. Compaction effort was significantly impacted by this. On the other hand, changes in volumetric attributes were evident, along with an increase in total binder content.

By applying the DoE technique, the crucial combinations of all of these factors were found, as depicted in Figure 5 and Figure 6. The Marshall properties of recycled asphalt blends are predicted using Equations (2)–(5) to deviate significantly from design guidelines when any of the parameters approach severe values. Lower and higher bounds for each of the following individual input parameters were set as percentage of RAP (0–45%), percentage of nanomaterial (2–4%), percentage of total binder (4–6%), and percentage of extra binder (1–2%). The limits for binder percentages were selected considering the dense graded mix. The statistical interaction between RAP and NS, RAP and Bitumen, NS and Bitumen are presented in Figure 5. Similarly, RAP and NC, RAP and Bitumen, and NC and Bitumen are presented in Figure 6. The statistical interaction plot between the nanomaterials and bitumen provides the optimum mixes and the variation for usage in recycled asphalt pavement. Similarly, the obtained predicted equations for NS and NC indicate validation of experimental results to determine the optimum usage in nanomaterial-modified recycled mixes.

Using the CCD technique, different combinations of these four input factors were generated in the second stage. Within the top and lower bounds, five distinct levels were found. Consequently, CCD was used to identify 17 pairings. These preset component proportions were used to generate recycled asphalt mixtures in the third phase. The ratio of each component (RAP, nanomaterial, and additional binder) to the total mixture was calculated using the numerical values of the input parameters at each unique combination. For each combination, six replicates of recycled asphalt mixtures were prepared, resulting in a total of 102 individual specimens (17 combinations × 6 replicates). This replication ensures robustness and statistical validity in the results. The volumetric and Marshall parameters were then computed for every combination by applying the Marshall method of mix design.

Predicted stability equation for nanosilica:

Stability = 12.5 + 0.190 RAP − 0.75 NS + 1.46 BITUMEN − 0.004064 RAP × RAP + 0.0933 NS × NS − 1.57 BITUMEN × BITUMEN + 0.0063 RAP × NS − 0.0093 RAP × BITUMEN + 0.161 NS × BITUMEN

(2)

Predicted flow equation for nanosilica:

Flow = 9.7 − 0.461 RAP − 3.5 NS + 2.0 BITUMEN + 0.00145 RAP × RAP + 0.073 NS × NS − 0.75 BITUMEN × BITUMEN + 0.0200 RAP × NS + 0.0867 RAP × BITUMEN + 0.625 NS × BITUMEN

(3)

Predicted stability equation for nanoclay:

Stability = −35.5 + 1.163 RAP − 1.26 NC + 17.55 BITUMEN − 0.00794 RAP × RAP − 0.037 NC × NC − 1.572 BITUMEN × BITUMEN + 0.0131 RAP × NC − 0.1691 RAP × BITUMEN + 0.337 NC × BITUMEN

(4)

Predicted flow equation for nanoclay:

Flow = −19.9 − 0.074 RAP − 0.19 NC + 10.18 BITUMEN + 0.00080 RAP × RAP + 0.0891 NC × NC − 1.190 BITUMEN × BITUMEN − 0.0021 RAP × NC + 0.0250 RAP × BITUMEN − 0.037 NC × BITUMEN

(5)

2.3. Experimental Plan

A maximum aggregate size of 19 mm was used for Bituminous Concrete (BC) Grade 1, in accordance with MoRTH (2013) specifications [45]. Nanosilica and nano clay-modified binders were incorporated into graded aggregates to produce HMA and WMA mixes, with an optimum binder content of 4%, as established in the previous literature [45]. The same procedures were followed for a comparative study using high-grade polymer-modified (PMB-40) binders. A total of 24 mixes were evaluated: 12 prepared with viscosity-grade bitumen and the remaining 12 with PMB-40-grade bitumen, varying the RAP percentages at 0, 15, 30, and 45%.

2.4. Marshall Stability and Flow Value

The Marshall Method of mix design was used to develop recycled mixes with a minimum bitumen proportion of 5.2% as required by ASTM D 155. The Optimum Bitumen Content (OBC) for the binder component (RAP bitumen + Fresh Bitumen) was determined for 4% air voids. Compacted asphalt mixtures are displayed in Figure 7.

After meeting the Marshall properties listed in MoRTH (2013) [45], the nanomaterial content of recycled mixtures was chosen in the following phase. Additionally, based on the higher stability and lower flow value (2–4 mm) of the desired mixes prepared with the optimum RAP content, the optimal nanomaterial content was determined, satisfying the requirements of volumetric properties. This study investigates the performance characteristics of 4% nanomaterial-modified asphalt mixtures with varying Reclaimed Asphalt Pavement (RAP) contents of 0, 15, 30, and 45%.

2.4.1. Indirect Tensile Strength (ITS)

The ITS test was performed on all of the mixes with RAP content of 0, 15, 30, and 45% according to ASTM D6931 [47]. Similarly, the 5 to 45 °C temperature range was used to determine the recycled mixes’ static tensile load. The temperature-dependent variation in ITS simulates the thermal characteristics of asphalt mixtures. Equation (6) was used to calculate the recycled mixes’ ITS (kPa) value.

ITS = 2000 × P/(π D T)

(6)

where P = ultimate load (N), T = Thickness of sample (mm), and D = Diameter of Sample (mm).

2.4.2. Tensile Strength Ratio (TSR)

The moisture content had a significant effect on the asphalt mixes’ longevity. This moisture constrains the bitumen’s and aggregate matrix’s bonding properties, allowing windows to be created for raveling and stripping. The moisture resistance of the conditioned sample—which was kept in a water bath for 24 h at 60 °C—was evaluated using retained Marshall stability (RMS) in comparison with unconditioned specimens. TSR was calculated as the ratio of the ITS value of the conditioned sample at 25 °C to the ITS value of the unconditioned sample at 25 °C. According to MoRTH (2013) [45], the minimum TSR value is 80%. Equation (7) is used to determine the TSR value. A higher TSR indicates that the mixture will function well and have strong defense against moisture-related harm. Every blend should satisfy the 80% minimum requirement for the TSR estimate.

TSR (%) = (ITS of conditioned sample/ITS of unconditioned sample) × 100

(7)

2.4.3. Wheel Rut Test (WRT)

WRT simulates the field rutting depth, and it is used to assess the rutting potential of recycled mixes. The study investigates asphalt mixtures with varying RAP content like 0%, 15%, 30%, and 45%, following the guidelines outlined in JTG E20-2011 [48]. The recycled mixtures were conditioned for six hours at 60 °C before this test. As seen in Figure 8, the test setup comprises a wheel load of 150 N, a contact pressure of 700 kPa, and a loading rate of 42 passes per minute. A 300 × 300 × 50 mm asphalt slab was created using a wheel rut shaper intended for 7% air gaps. The asphalt slab was tested in a damp state, with 10,000 passes or the failure criterion of rut depth of 6 mm applied, whichever happens first. Additionally, during the last fifteen minutes of the hour-long test conducted to establish the effect of RAP and nanomaterial on various BC-1 mix characteristics, the number of wheel turns required to cause 1 mm of rutting was used to determine the dynamic stability of the mixtures. Rutting, measured as rut depths d₁ (mm) at 45 min and d₂ (mm) at 60 min, at a test temperature of 50 °C (122 °F) to simulate high-temperature conditions, was used to determine the dynamic stability of the mixtures. Also, the deformation t1 and t2 were measured at 45 min and 60 min respectively. DS is calculated using Equation (8).

Dynamic stability, (DS) = [15N/(d2−d1)] × C1 C2 (passes/min)

(8)

where N = no of passes per minute (42 passes/min), d1 = rut depth at 60 min, and d2 = rut depth at 45 min. C1 and C2 are machine coefficients: C1 = 1; C2 = 2.

3. Results and Discussion

3.1. Marshall Properties

The output and input relationship between different factors, such as the total binder content, nanomaterial, and RAP, was determined using an RSM. The quantitative comparison was then carried out by plotting the predicted values against the experimental values shown in Figure 9 and Figure 10. Most of the values were positioned along the trend line. The coefficient of determination (R²) values associated with Marshall stability and Marshall flow were 0.93 and 0.92, respectively. The accuracy of the model was checked using RMS error. Due to the complexity of the interactions between the different factors, the second-phase response surface effectively captured the patterns of volumetric and Marshall properties. However, it did not thoroughly analyze the isolated effects of each variable. To mitigate these effects, contour plots were developed with two variables held constant. Figure 11 and Figure 12 present contour and surface plots.

From Figure 11 and Figure 12, the nanomaterial and percentage of RAP content were kept constant at their mean values, and the percentage of total binder and binder grade were varied. RAP typically contains higher binder content than original mixes, with increased coarse aggregate angularity and fine particle percentage [49,50]. The porous nature of RAP particles and their aged binder contribute to increased porosity in RAP-inclusive mixtures. This porosity affects mechanical properties, with recycled mixes showing lower modulus, higher wheel tracking rates, and reduced fatigue life compared to virgin mixes. Increased RAP content leads to poor compaction of asphalt mixtures, resulting in increased air voids, whereas in the case of Voids in Mineral Aggregate (VMA), a decrease in value was observed with an increased percentage of total binder and RAP content [51,52].

Due to poor interlock and high porosity, RAP material offers high resistance to compaction. Therefore, at a similar binder content, an increase in RAP material results in more voids. Even with a higher demand for binders, the minimal contribution of RAP material to the overall binder content can lead to a degree of inefficiency in the mixture. As a result, the VMA increased when the total binder content increased. On the other hand, when the RAP content increased, the VMA decreased.

An increase in Marshall stability (MS) values was observed up to a replacement of 30% RAP from the natural aggregate. For a certain RAP content, the total binder content has a considerable effect on Marshall stability due to the conglomeration effect of RAP [37]. Also, the high stiffness of RAP material is reported to have high Marshall stability and decreased flow value [38]. This can be ascribed to poor aggregate interlock, the viscosity of the binder, and the binder content requirement of recycled asphalt mixtures [33].

3.2. Indirect Tensile Strength (ITS)

The observed ITS values of conventional HMA and WMA mixes (at 0% RAP) are 1461 kPa and 1193 kPa, respectively. This indicates that the addition of Zycotherm, even though it results in a reduction in compaction temperature from 160 to 130 °C, also results in a reduction in the tensile strength values of asphalt mixes. When RAP was incorporated into HMA and WMA asphalt mixes at 15, 30, and 45%, an increase in tensile strength value was observed up to 30% RAP replacement when compared to conventional mixes with 0% RAP. At 45% RAP replacement, the tensile strength value was observed to be slightly more than the conventional mixes but lower than 30% RAP replacement. The reason for this response could be that, at higher RAP content, the ITS value is dominated by the RAP proportion, which is stiffer and more robust than a fresh binder. The addition of RAP material into WMA mixes reported similar performance in the case of ITS value [39].

To improve the performance of RAP-incorporated HMA and WMA mixes, the recycled asphalt mixtures were modified with optimized nanomaterials such as nano silica and nanoclay at 4%. Nanomaterial-modified recycled asphalt mixtures also resulted in a similar trend in ITS value when compared with unmodified recycled asphalt mixes, i.e., 30% RAP replacement showed an increase in tensile strength. Among the two nanomaterials compared, the nanosilica-modified HMA and WMA mixes resulted in a higher ITS value than the nanoclay-modified HMA and WMA mixes.

In the case of nanosilica-modified recycled asphalt mixes, the highest ITS value observed for HMA and WMA mixes was 1616 kPa and 1454 kPa, respectively, achieved at 30% RAP replacement. Conversely, the lowest ITS values were recorded at 45% RAP replacement. To enhance the performance of HMA and WMA mixes incorporating 45% RAP, we conducted a study using nanomaterial-modified Polymer-Modified Bitumen (PMB 40) mixes. The ITS values of nanomaterial-modified PMB recycled asphalt mixes exhibited a similar trend with an increase in RAP content. Specifically, nanomaterial-modified PMB mixes at 45% RAP content showed higher tensile strength values compared to conventional mixes at 0% RAP and unmodified mixes at 45% RAP. Figure 13 and Figure 14 depict the variations in ITS values observed for nanosilica- and nanoclay-modified recycled asphalt mixes with varying RAP content.

Overall, improvement in ITS value can be expected at 30% RAP content compared to control mixes. It can also be observed that the highest ITS value when compared to the three RAP variations was for all mixes with 30% RAP. Also, improved performance can be observed for nanosilica-modified PMB recycled mixes with 45% RAP.

3.3. Tensile Strength Ratio (TSR)

The TSR value of HMA and WMA mixes with RAP material is depicted in Figure 15 and Figure 16. The HMA mix with 0% RAP was found to have a TSR value of 0.82, exceeding the minimum suggested value of 0.80 as per MoRTH (2013) [45], and hence passed the threshold value. Meanwhile, the WMA mix with 0% RAP observed a TSR value of 0.78, suggesting that Zycotherm-based WMA could reduce the mix’s ability to withstand moisture, but when added with RAP, it behaves differently.

In the case of nanomaterial-modified recycled mixes up to 30% RAP replacement, an increase in TSR value was observed, indicating improved resistance of the mixes to moisture susceptibility. Among the nanomaterials considered, nanosilica-modified recycled mixes showed better performance when compared to nanoclay-modified recycled mixes, indicating the highest resistance to moisture damage [53]. The better performance of nanosilica-modified mixes can be attributed to the enhanced bonding behavior of nanosilica due to increased contact surface area.

The highest values observed are 0.9 and 0.88 for nanosilica-modified recycled HMA and WMA mixes, respectively, at 30% RAP replacement, and the lowest value was observed for recycled mixes with 45% RAP replacement. This phenomenon was observed due to changes in the volumetric properties of asphalt mixtures with 45% RAP content. A similar phenomenon was observed in the research carried out by researcher Singh et al. [39].

In the case of a higher RAP percentage of 45%, the nanomaterial-modified PMB recycled asphalt mixtures showed comparatively higher TSR values when compared to conventional mixes. The nanosilica-modified PMB recycled HMA mixes showed the highest TSR value of 0.85, which is observed to be higher when compared to conventional mixes. In the case of WMA mixes, the highest TSR value was observed at 0.89 for the nanosilica-modified PMB recycled mixes at 30% RAP replacement. It was also observed that all mixes passed the minimum required value of 0.80 except for the control WMA mix and nanoclay-modified recycled WMA mix at 45% RAP replacement.

Therefore, from the perspective of moisture damage resistance potential, adding RAP up to 30% may be considered appropriate with HMA and WMA mixtures for both binder types.

3.4. Wheel Rut Test (WRT)

The rut resistance of the recycled asphalt mixtures was evaluated through a wheel track test as per the E20-2011 specification (JTG E20-2011). For each asphalt mixture, three slabs of dimensions 300 mm × 300 mm × 50 mm were cast at a specified mixing and compaction temperature of 155 ± 5 °C and 145 ± 5 °C, respectively. The samples were subjected to a tire pressure of 0.7 ± 0.05 MPa through a static-type solid rubber tire with a diameter and width of 200 mm and 50 mm, respectively. The speed of the wheel is 42 ± 1 cycles/min for 60 min at a travel distance of 230 ± 10 mm. The relationship between the rut depth and the number of passes was obtained for the asphalt mixtures. The Dynamic Stability (DS) of the mixtures was computed using the rut depth for 45 min (RD₄₅) and 60 min (RD₆₀) as given in Equation (9).

DS = 630/RD₄₅ − RD₆₀

(9)

In Figure 17 and Figure 18, it can be observed that in the case of nanomaterial-modified recycled HMA and WMA mixes, the rut depth value decreased with the increase in RAP content up to 30% RAP replacement. Nanosilica-modified recycled HMA and WMA mixes showed a lower rut depth value, indicating higher resistance of the recycled asphalt mixes to rutting failure. The highest rut depth value was observed for nanoclay-modified recycled WMA mixes with 45% RAP, indicating the least rut resistance compared to all other mixes. When the RAP content was increased to 45%, the rut depth value was observed to be the highest among all of the mixes, stipulating more susceptibility of the mixture to rut failure. Meanwhile, nanomaterial-modified PMB recycled asphalt mixes are reported to have a significant effect on rut behavior, as depicted in Figure 18. Figure 19 and Figure 20 represent the number of cycles required to meet the failure criteria. It was observed that mixes with the highest resistance to rutting failure showed a lower rut depth value with a higher number of cycles for loading failure. It was observed that the nanosilica-modified PMB recycled HMA and WMA mixes at 30% RAP showed the lowest rut depth of 4.1 mm and 4.3 mm, respectively, with the highest number of loading cycles for failure. The dynamic stability of recycled asphalt mixtures also increases with asphalt content initially and then gradually decreases, attaining the maximum value at similar binder content for both types of asphalt mixtures [54]. This is due to excessive and insufficient asphalt being primarily responsible for the rut resistance of recycled asphalt mixtures [38,55]. The internal friction and the cohesiveness of recycled asphalt materials determine their rutting performance [41]. It is always desirable to have optimum asphalt content for adequate stability at high temperatures; nevertheless, the dynamic stability of the asphalt mixture falls short of the standard criterion of DS ≥ 600 cycle min as outlined in JTG F20-2004 [48]. The dynamic stability of VG-30- and PMB-40-grade recycled asphalt mixtures is greatly affected by binder content. Overall, the nanosilica-modified recycled asphalt mixtures of PMB-40 grade exhibited better mechanical properties and performance compared to all other mixes.

4. Conclusions

The objective of this study was to determine the effects of various factors on the Marshall and volumetric characteristics of asphalt mixtures. Through the use of RSM, the researchers were able to identify the different factors that can affect the performance of these mixtures:

• Based on these response surfaces, the results show that the variability in the recycled asphalt mixture was reduced by the binder quantity. Therefore, to avoid uncertainties regarding the strength parameters and volumetric qualities of recycled asphalt mixtures, an appropriate proportion of total binder content (additional bitumen and aged binder from RAP) might be employed.
• It was found that the RAP content had the biggest impact on the stability of the recycled asphalt mixtures. On the contrary, nanomaterials and the total binder content had the most significant effects on the flow values and volume properties.
• The effects of the nanomaterial on the recycled asphalt mixture were evaluated through indirect tensile strength (ITS) and tensile strength ratio tests. The test results indicate that the nanomaterial enhanced the moisture susceptibility and low-temperature cracking resistance of the recycled asphalt mixture.
• The high rut resistance of the recycled asphalt mixture is expected due to its densely packed aggregate structure formed with greater dispersion of nanomaterial in the binder (VG-30 and PMB 40) sufficient to have interparticle contact in the asphalt matrix.
• In recycled asphalt mixtures with nanomaterials, the optimum asphalt content is considered. However, RAP materials and mixture (HMA/WMA) types should be estimated for their threshold value to avoid lubrication between the aggregate matrix and binders prone to severe rut depth.