Mix Design Optimization and Performance Evaluation of Ultra-Thin Wearing Courses Incorporating Ceramic Grains as Aggregate

Abstract

The impact of ice and snow in seasonally frozen regions has led to a significant decline in the flatness and skid resistance of highway pavements, creating severe traffic safety hazards. With economic development driving the transition from road construction to maintenance, this study proposes enhancing Ultra-Thin Wearing Course (UTWC) maintenance materials with anti-icing performance and snow-melting properties. The study first employed the Marshall mix design method to develop gradations for two common types of UTWC asphalt mixtures: the dense-graded GT-8 and the open-graded NovaChip^® Type-B. Using the volume substitution method, aggregates were replaced with equivalent volumes of ceramic grains. The optimal asphalt–aggregate ratios for the mixtures with varying ceramic grain contents were determined, and the influence of ceramic grains content on the asphalt–aggregate ratio was analyzed. The results indicate that the optimal asphalt–aggregate ratio increases with higher ceramic grains content. Subsequently, the high-temperature performance, low-temperature performance, and water stability of UTWC with varying ceramic grain contents were evaluated. Overall, NovaChip^® gradation mixtures demonstrated superior road performance compared to GT-8 gradation mixtures. Moreover, an increase in ceramic grains content enhanced the high-temperature performance of UTWC but moderately reduced its low-temperature performance and water stability. Finally, the effects of different ceramic grain contents and snowmelt agent types on the anti-icing and snowmelt properties of UTWC were examined. The results revealed that higher ceramic grains content improved snowmelt effectiveness. Considering the road performance of the specimens, a ceramic grains content of 40% was recommended. Furthermore, calcium chloride (CaCl₂) exhibited superior anti-icing performance compared to other snowmelt agents.

1. Introduction

With rapid economic development and increased traffic, many road infrastructures, especially in harsh climates, have reached their capacity limits [1]. This has spurred the development of innovative maintenance technologies, such as Ultra-Thin Wearing Course (UTWC). First introduced in France in the 1980s, UTWC aimed to address road distress caused by heavy traffic and weather damage. The initial application, Béton Bitumineux Très Mince (BBTM), was a 2–3 cm layer designed to restore surface smoothness, improve skid resistance, and extend pavement life. France later improved this technology by reducing sand content and incorporating modified asphalts, enhancing the mixture’s structural properties and pavement durability [2]. The concept was then extended to other countries, such as Spain, the United States, the United Kingdom, and Japan, all of which began developing their own variations of UTWC to meet specific regional road performance needs. For instance, the United States developed the Open-Graded Friction Course (OGFC) as part of a broader initiative to improve road friction and reduce noise, particularly in warmer climates [3]. By the early 1970s, the United States improved conventional wearing course technologies, successfully developing new thin-layer technologies such as SUP-5 and SMA-5 [4,5]. In the 1990s, the United States introduced NovaChip^®, an open-graded Ultra-Thin Wearing Course known for its excellent drainage properties and resistance to rutting, which was applied to experimental roads such as Talladega and Tallapoosa. These pavements maintained their excellent surface quality even after 10 years of use, leading to the widespread adoption of NovaChip^® [6,7,8]. Similarly, the United Kingdom developed UTWC with a void ratio of around 20%, which significantly reduced road noise by 3–4 dB, equivalent to a 60% reduction in noise levels [9]. In light of these challenges, incorporating additional functionalities into UTWC is an effective approach. Not only can it enhance the road’s performance and durability, but it can also endow the pavement with additional capabilities, such as snow-melting and anti-icing properties, further improving the overall functionality and adaptability of roads.

Salt-based materials play a crucial role in salt-storing pavement de-icing and snow-melting technologies [10]. These materials are classified into aggregate-type, filler-type, and coating-type. Aggregate and filler types involve encapsulating or loading salt-based materials into particles or powders, which replace aggregates or mineral fillers in asphalt mixtures. Coating-type materials incorporate de-icing agents into asphalt emulsions, which are applied to road surfaces for short-term snow-melting [11,12]. The carriers for these materials are typically lipid-resin, inorganic cementitious, or porous. Research has shown that lipid-resin carriers like linseed oil and stearates encapsulate de-icing agents effectively, though they wear over time [12,13,14]. Various approaches, including the use of inorganic cementitious carriers, such as ground bauxite with NaCl, polyurethane foam with CaCl₂, and powder sintering techniques with salt and mineral powders, have been shown to effectively enhance anti-icing and snow-melting performance on road surfaces, demonstrating their potential to achieve significant snow-melting effects [15,16,17]. For porous carriers, the earliest research originated with Siegmund W [18], who employed porous carriers to absorb anti-icing and snow-melting materials when constructing railway pathways. Research on porous carriers has demonstrated effective methods for enhancing snow-melting performance, such as using zeolites and thermoplastic materials combined with alkaline chlorides or alkali metal chlorides. These approaches highlight the feasibility of incorporating snowmelt agents into road materials for effective ice suppression and snow melting [19,20].

Since the 1960s, anti-icing and snow-melting pavements have been proposed as a novel approach to addressing road icing and snow accumulation issues. European countries were the first to initiate research on anti-icing and snow-melting pavements during the 1960s [21]. This technology primarily involves adding de-icing agents, with calcium chloride (CaCl₂) as the main component at a concentration of 90%, into asphalt mixtures. During rainfall or snowfall, water penetrates the mixture, and the de-icing materials diffuse from the higher-concentration interior of the mixture to the lower-concentration surface through osmotic pressure, capillary action, and other mechanisms, achieving anti-icing and snow-melting effects. In 1974, Austria implemented the first anti-icing and snow-melting pavement on the Europe Bridge in Brenner, which demonstrated excellent performance until 1994 [22]. Since then, other countries such as Japan, Germany, and Russia have also developed and implemented similar technologies, using various de-icing agents to achieve effective snow-melting results [23,24].

Ceramic aggregates are well known for their porous structure and are a low-carbon, recycled material capable of effectively adsorbing deicers. Additionally, the implementation of anti-icing and snowmelt road surfaces can significantly enhance driving safety, reduce the costs of ice and snow removal, minimize damage to the pavement caused by ice and snow, and ultimately improve the durability of road materials while potentially lowering maintenance costs—especially in areas where ice and snow accumulation is severe.

This study aims to investigate the incorporation of anti-icing and snowmelt properties into Ultra-Thin Wearing Course (UTWC) asphalt mixtures by adding ceramic aggregates. The goal is to enhance anti-icing performance and improve winter road safety. However, the structural differences between ceramic aggregates and traditional aggregates in the asphalt mixture may influence the final pavement performance. This includes effects on high-temperature stability, low-temperature crack resistance, and overall water stability. To address this gap, this study adopts the Marshall mix design method to explore the optimal asphalt content for mixtures with varying ceramic aggregate contents. This study also investigates the effect of ceramic aggregate content on the asphalt–aggregate ratio. Furthermore, the influence of different ceramic aggregate contents on the high-temperature performance, low-temperature performance, and water stability of UTWC is examined. Using image analysis and chloride ion release rate methods, the impact of different ceramic aggregate contents and deicer types on the anti-icing and snowmelt performance of UTWC is assessed. Based on the findings, this study recommends the optimal ceramic aggregate content to achieve the best snowmelt performance while maintaining the overall performance and service life of the road surface. This provides theoretical support for the widespread application of asphalt mixtures with integrated snowmelt and anti-icing properties.

2. Materials and Methods

2.1. Materials

2.1.1. Aggregates

The performance tests for the coarse and fine aggregates used were conducted in accordance with the Test Methods of Aggregate for Highway Engineering (JTG E42) [25]. The test results are presented in

Table 1. Performance test results and technical specifications for coarse aggregates.

Coarse Aggregate	Aperture Size (mm)		Technical Requirements	Test Methods
	2.36	4.75
Apparent relative density	2.844	2.763	≥2.6	T 0304
Water absorption (%)	1.9	1.2	≤2.0	T 0307
Crushing value (%)	12.9		≤26	T 0316
Abrasion value (%)	12.7		≤28	T 0317

and

Table 2. Performance test results and technical specifications for fine aggregates.

Fine Aggregate	Aperture Size (mm)					Technical Requirements	Test Methods
	0.075	0.15	0.3	0.6	1.18
Apparent relative density	2.762	2.784	2.745	2.795	2.780	≥2.5	T 0304
Angularity (s)	36					≥30	T 0345
Sand equivalent (%)	81					≥60	T 0334

.

2.1.2. Mineral Powder

For the filler material, limestone was chosen [26]. The technical indexes of the filler were tested according to the Test Methods of Aggregate for Highway Engineering (JTG E42) standard method. The results in

Table 3. Performance test results and technical specifications of mineral powder.

Technical Specifications of Mineral Powder		Test Results	Technical Requirements	Test Methods
Apparent relative density		2.798	≥2.5	T 0352
Particle size range	<0.6 mm	100	100	T 0351
	<0.15 mm	92.7	90~100
	<0.075 mm	80.3	75~100
Hydrophilicity coefficient		0.7	<1	T 0353
plasticity index		1.9	<4	T 0354
Appearance		No agglomeration or lump formation	No agglomeration or lump formation	-

follow the technical requirements.

2.1.3. High-Viscosity, High-Elasticity Asphalt

In order to ensure the stability of the overall pavement performance, the asphalt used in the composition of the Ultra-Thin Wearing Course (UTWC) material is required to have high specifications, typically involving the use of high-viscosity modified asphalt [27,28]. The design thickness of an Ultra-Thin Wearing Course is generally between 15 and 25 mm. Ordinary asphalt is often insufficient to provide the necessary constraint for the skeleton structure, resulting in inadequate pavement strength. Under heavy traffic loading, the pavement may suffer from severe rutting, which negatively impacts the road’s usability. To meet the functional requirements of the Ultra-Thin Wearing Course, this study selects a high-viscosity, low-penetration high-elasticity modified asphalt. This type of asphalt can effectively constrain the skeleton structure of the coarse aggregate. The performance test results of the high-viscosity and high-elasticity asphalt are shown in

Table 4. Performance test results and technical specifications of high-viscosity, high-elasticity asphalt.

Test Items	Unit	Technical Requirements	Test Results	Test Methods
Penetration degree (25 °C, 5 s, 100 g)	0.1 mm	30~60	37.0	T 0604
Softening point	°C	≥90	95.8	T 0606
Ductility (5 °C)	cm	≥20	22.0	T 0605
Flash point	°C	≥230	337	T 0611
Elastic recovery (25 °C)	%	≥95	103	T 0662
Dynamic viscosity (60 °C)	Pa·s	>400,000	486,870	T 0620

.

2.1.4. Ceramic Grains

Ceramic grains, primarily made from shale, clay, coal gangue, and other raw materials, are a type of lightweight aggregate produced by crushing, screening, and high-temperature calcination to achieve expansion. These grains are characterized by their porous structure, low density, high porosity, and high softening coefficient. Depending on the raw materials used, ceramic grains can be categorized into different types such as biological sludge ceramic grains, shale ceramic grains, clay ceramic grains, coal gangue ceramic grains, and waste ceramic grains. Among them, clay ceramic grains and fly ash ceramic grains are typically used as lightweight aggregate concrete for thermal insulation or structural purposes. Shale ceramic grains are mainly used as filtration materials or construction materials. Waste ceramic grains and biological sludge ceramic grains replace fly ash in the production of ceramic grains, thus conserving fly ash and contributing to environmental protection. Based on the shape of ceramic grains, they can be classified into gravel-type ceramic grains, spherical ceramic grains, and cylindrical ceramic grains. Cylindrical and spherical ceramic grains have a more regular shape, which results in poorer adhesion properties and are typically used as landscaping materials. Therefore, this study selects gravel-type shale ceramic grains, as shown in Figure 1. This type of ceramic grain can replace part of the coarse aggregate to form a mineral skeleton structure with a certain gradation. When mixed with a certain amount of asphalt, it forms an asphalt concrete with a certain strength.

The density grade of the ceramic grains is 500 level. In accordance with the Test Methods of Aggregate for Highway Engineering (JTG E42) [25], the apparent density, coefficient of variation, and crushing value were tested. The basic performance indicators of the ceramic grains are shown in

Table 5. Basic performance indicators of ceramic grains.

Grain Size (mm)	Apparent Density (g/cm3)	Water Absorption (%)	Crushing Value (%)
5–10	1.189	9.3	67.1

.

2.2. Experimental Methods

2.2.1. Mixture Design

According to the functional requirements of road surfaces, the wearing course is classified into dense-graded and porous-graded types. This study selects two gradation types for the Ultra-Thin Wearing Course to investigate their anti-icing and snowmelt performance: GT-8 dense-graded asphalt mixture and NovaChip^® Type-B porous-graded asphalt mixture. The Marshall mix design method was employed in accordance with the provisions of the Technical Specification for Construction of Highway Asphalt Pavements (JTG F40) [26]. Four Marshall samples with heights of 63.5 ± 1.3 mm were prepared for each group for testing. The optimum asphalt content of two kinds of Ultra-Thin Wearing Course were determined and verified based on the volumetric indexes and Marshall stability.

2.2.2. Rutting Test

Numerous studies have shown that dynamic stability is commonly used to evaluate the high-temperature deformation resistance of asphalt mixtures, thereby reflecting their rutting resistance under high-temperature conditions. In this study, dynamic stability was adopted as the evaluation index. Rutting specimens with dimensions of 300 mm × 300 mm × 50 mm were prepared based on different ceramic grain contents and their corresponding optimal oil–stone ratios. Prior to testing, all specimens were placed in a thermostatic oven at 60 °C and conditioned for more than 5 h. According to Standard Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering (JTG E20-2011) [29], during the test, the testing wheel moved back and forth across the center of the rutting plate at a speed of 42 passes per minute. At t1 = 45 min and t2 = 60 min, displacement sensors automatically recorded the deformation values, d1 and d2, respectively. The dynamic stability was then calculated based on the deformation values.

2.2.3. Bending Test

The small beam bending test is used to evaluate the cracking resistance of specimens under low-temperature conditions. The evaluation indicators include flexural tensile strength (RB), maximum bending strain (εB), and flexural stiffness modulus (SB). In this test, rutting plate specimens with ceramic grain contents of 0%, 20%, 40%, and 60% were first prepared. These were then cut into small beam specimens with dimensions of 250 mm × 30 mm × 35 mm. The small beam specimens were conditioned at a temperature of −10 ± 0.5 °C−10 ± 0.5 °C for more than 4 h prior to testing. The small beam specimens were then subjected to a loading test using a UTM universal testing machine at a loading rate of 50 mm/min until failure occurred. This test was conducted to evaluate the low-temperature tensile performance of the asphalt mixtures. The load-deflection curves for the span were ultimately obtained. The test followed the procedures outlined in the Standard Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering (JTG E20-2011) [29]. During the test, the upper loading plate was positioned at the center and tightly contacted the sample to prevent any impact load caused by gaps between the plate and the specimen. The specimen was symmetrically placed on the supports, with its vertical axis aligned with the loading direction. The low-temperature small beam bending test process for asphalt mixtures with different ceramic grain contents is shown in Figure 2.

2.2.4. Immersing Marshall Test

When external water enters the voids of the asphalt pavement, under the influence of vehicle loads and temperature–moisture-induced expansion and contraction, dynamic water pressure or vacuum suction occurs within the asphalt mixture. This leads to water entering the gaps between the asphalt and aggregates, causing the pavement to become loose, develop raveling, potholes, and other distress. This type of damage is referred to as water damage in asphalt.

The water stability of ceramic grain-modified asphalt mixtures was investigated using the immersion Marshall test. According to the Standard Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering (JTG E20-2011) [29], asphalt mixtures with different ceramic grain contents were compacted with 75 blows on both sides to form 8 Marshall specimens. The specimens were divided into two groups. One group was tested following the conventional Marshall stability test method, while the other group was immersed in a 60 °C thermostatic water bath for 48 h before performing the Marshall stability test.

2.2.5. Quantitative Analysis of Pores and Cracks

As a functional wearing course, the Ultra-Thin Wearing Course (UTWC) must not only meet the road performance requirements of the asphalt mixture but also possess excellent anti-icing performance and snowmelt properties. In this study, the Pores and Cracks Analysis System (PCAS), developed by Nanjing University, was used to quantitatively analyze the road’s ice and snow melting images. PCAS has a range of powerful functions, including image acquisition, calibration, enhancement, and processing. It can convert various particle and pore images into binary format, automatically remove noise, and output their geometric and statistical parameters, displaying vector graphics. Then, using the color range and color tolerance features in the image processing software, pixel blocks were selected to determine the proportion of black and white pixels, which allowed the assessment of residual snow on the road surface. The snowfall during the test was 8.6 mm. The effects of snowmelt properties and anti-icing performance of the road before and after PCAS treatment are shown in Figure 3 and Figure 4.

2.2.6. Determination of Chloride Ion Content

To further investigate the release pattern of effective components in the snowmelt agent, this study employed a chloride ion content rapid determination instrument to measure the chloride ion leaching concentration of the Ultra-Thin Wearing Course. This approach helps better analyze the release rate and pattern of the snowmelt agent. The chloride ion content rapid determination instrument primarily uses the conductivity analysis method, which is a commonly used technique for determining the concentration of strong electrolyte solutions. The principle is that strong electrolytes dissociate into cations and anions when dissolved in water, and under the influence of an electric field, they move in opposite directions, creating an electric current. The concentration of the solution is determined by measuring the conductivity, making this method quick, efficient, and easy to use. Specifically, the ion concentration is inferred by analyzing the linear relationship between the calibration solutions of three different NaCl concentrations (0.005 mol/L, 0.05 mol/L, and 0.5 mol/L) and the corresponding potential differences, thus calculating the molar concentration of chloride ions. First, rutting plate specimens with thicknesses of 20 mm were molded and submerged in 1500 mL of distilled water. Every 30 min, three 100 mL samples were taken. The chloride ion contents in these samples were measured using the chloride ion content rapid determination instrument, as shown in Figure 5, by determining the conductivity of the samples. Ultimately, the release pattern of the effective components in the snowmelt agent was obtained.

3. Results

3.1. Mixture Design of Ultra-Thin Wearing Course

3.1.1. Aggregate Gradations

(1)

Dense-Graded Mixture

Chen Fuda [30] adopted the Coarse Aggregate Voids Filling (CAVF) method to design a highly durable ultra-thin asphalt wearing course (GT-8). GT-8 is primarily based on the theory of coarse aggregate void filling. By adjusting the passing rate of the critical sieve size of 2.36 mm, this method reduces the amount of fine aggregate, increases the proportion of coarse aggregate, and simultaneously enhances the asphalt content. This approach improves the interlocking capacity and compactness of the aggregate skeleton, ultimately forming a special skeleton-dense continuous gradation.

The mix design for this gradation consists of basalt, limestone, and mineral powder. By adjusting the blending ratios of aggregates in different size ranges, the gradation design curve is obtained, as shown in Figure 6.

(2)

Open-Graded Mixture

NovaChip^® is categorized into three types: A, B, and C. The nominal maximum aggregate size (NMAS) for Type A is 9.5 mm, for Type B is 13.2 mm, and for Type C is 16 mm. Considering that the Ultra-Thin Wearing Course designed in this study has a thickness of 20 mm, Type A gradation is deemed most suitable for subsequent mix design.

The materials used in the mix design for this gradation include basalt, limestone, and mineral powder. By adjusting the blending ratios of aggregates in different size ranges, the gradation design curve was obtained, as shown in Figure 7.

3.1.2. Determination of Asphalt Content

Replacing a portion of the 4.75–9.5 mm aggregates in the Ultra-Thin Wearing Course with ceramic grains of the same size can impact the performance of the Ultra-Thin Wearing Course. Insufficient incorporation of ceramic grains fails to achieve the desired anti-icing and snowmelt properties, while excessive incorporation may compromise the road performance of the wearing course. Therefore, the dosage of ceramic grains is a critical factor in this study. Liu Dongxu [31], considering both the mechanical properties and skid resistance of ceramic grains, suggested that the dosage should not exceed 60%. The specific amount should be determined based on the road’s performance requirements to ensure the service life and safety of the ceramic grain-enhanced pavement.

In this study, ceramic grains were used to replace 0%, 20%, 40%, and 60% of the 4.75–9.5 mm aggregates in the mixture by an equal volume substitution method. Standard cylindrical specimens were then prepared using the Marshall method, and the performance of the Marshall specimens for both gradations was tested. The properties evaluated included bulk density, stability, flow value, void content, and effective asphalt saturation. Based on these properties, the optimal oil–stone ratio for each gradation with varying ceramic grain content was determined according to the Technical Specification for Construction of Highway Asphalt Pavements (JTG F40) [26], and the results are shown in

Table 6. The optimal oil–stone ratio of the ceramic grain asphalt mixture for the GT-8 gradation.

Ceramic grainscontent(%)	0	20	40	60
Optimal oil–stone ratio(%)	6.7	7.1	7.3	7.8

and

Table 7. The optimal oil–stone ratio of the ceramic grain asphalt mixture for the NovaChip® gradation.

Ceramic grainscontent (%)	0	20	40	60
Optimal oil–stone ratio(%)	4.9	5.4	5.5	6.0

.

Based on the table, it can be concluded that as the ceramic grain content increases, the optimal oil–stone ratio also increases. This is because the porous structure of ceramic grains absorbs part of the asphalt, and increasing the ceramic grain content leads to a higher asphalt requirement. The optimal oil–stone ratio for the ceramic grain asphalt mixture of the GT-8 gradation is greater than that of the NovaChip^® gradation. This is because the GT-8 gradation, with its denser packing, requires more asphalt to fill the voids created by the ceramic grains. In contrast, the NovaChip^® gradation, being more open-graded, has larger voids that require less asphalt to achieve the desired performance. The reason is that, in dense gradations, the aggregate particles are more compactly packed, resulting in a larger total surface area. Asphalt must coat this increased surface area to ensure proper bonding, leading to higher asphalt demand.

3.2. Evaluation of Road Performance

According to the Technical Specification for Construction of Highway Asphalt Pavements (JTG F40), the road performance of the Ultra-Thin Wearing Course for both NovaChip^® and GT-8 gradations was evaluated to investigate the impact of ceramic grains with snowmelt agent adsorption on the high-temperature performance, low-temperature performance, and water stability of the Ultra-Thin Wearing Course.

3.2.1. Rutting Test

The rutting test uses dynamic stability as the evaluation index to assess the high-temperature stability of asphalt mixtures, reflecting their ability to resist rutting deformation under high-temperature conditions. The rutting test results for NovaChip^® and GT-8 gradations with different ceramic grain contents are shown in Figure 8.

According to Figure 8, it can be observed that due to the use of high-viscosity, high-elasticity asphalt, both types of pavements exhibit excellent high-temperature properties, with dynamic stability approximately reaching 8000 passes/mm or more. For the GT-8 gradation, when the ceramic grain content does not exceed 40%, the dynamic stability increases with the ceramic grain content. A similar trend is observed for the NovaChip^® gradation asphalt mixture. However, when the ceramic grain content reaches 60%, the dynamic stability significantly decreases. The surface texture of ceramic grains is more intricate, forming a better skeleton between aggregates. Additionally, by replacing coarse aggregates with equal volumes of ceramic grains, some asphalt is absorbed by the ceramic grains, reducing the asphalt content. This leads to poorer high-temperature flow properties of the asphalt, enhancing the high-temperature performance of the asphalt mixture with ceramic grains. However, because ceramic grains are porous aggregates, when their content is too high, the crushability of the porous composite aggregate is severely affected, reducing the overall rutting resistance of the mixture. As a result, the specimen experiences crushing under the test wheel, which increases the rutting depth and decreases dynamic stability.

Additionally, when comparing the two gradations with the same ceramic grain content, the NovaChip^® gradation asphalt mixture contains approximately 37% more coarse aggregates and 38% less fine aggregates. This composition results in a more robust aggregate skeleton structure, providing stronger deformation resistance. Consequently, the dynamic stability of the NovaChip^® gradation is generally higher than that of the GT-8 gradation, indicating superior high-temperature performance.

3.2.2. Bending Test

(1)

Failure strain and bending stiffness modulus

The failure strain and bending stiffness modulus can directly reflect the low-temperature performance of asphalt mixtures. When the specimen fails, a larger failure strain at the bottom of the tested beam and a smaller bending stiffness modulus indicate better low-temperature cracking resistance of the mixture. The results of the bending test for asphalt mixtures with different ceramic grain contents are shown in Figure 9.

The primary function of the Ultra-Thin Wearing Course asphalt mixture designed in this study is to provide anti-icing and snowmelt properties, making it particularly suitable for severe winter regions. According to the Technical Specification for Construction of Highway Asphalt Pavements (JTG F40), the requirement for the maximum flexural strain of modified asphalt mixtures in such regions is no less than 3000 με. As shown in Figure 9, all specimens meet this requirement except for the GT-8 gradation with 60% ceramic grain content. With increasing ceramic grain content, the bending stiffness modulus decreases, the maximum flexural strain first increases and then decreases, and the flexural-tensile strength first decreases and then increases. The specimen with 20% ceramic grain content achieves the highest maximum flexural strain and the lowest bending stiffness modulus.

(2)

Bending failure energy density

Zhou et al. [32] combined low-temperature deformation capacity and bearing capacity, representing them by the bending failure energy density, as shown in Formula 1. Using bending failure energy density provides a better evaluation of the low-temperature properties of asphalt mixtures. The greater the energy consumed during failure, the better the low-temperature crack resistance. The calculation results of the bending failure energy density are shown in Figure 10.

The equation is as follows:

$${\omega }_f={\displaystyle \frac{d\omega }{d\upsilon }}={\displaystyle {\int }_0^{{\epsilon }_c}\sigma (\epsilon )d\epsilon }$$

(1)

where ω_f is the strain energy density function; σ(ε) is the function of stress with respect to strain; and ε_c is the strain corresponding to maximum stress.

As shown in Figure 10, the bending failure energy density of pavements with both gradations decreases with the increase in ceramic grain content. Meanwhile, the difference in bending failure energy density between the two gradations is minimal. The bending failure energy density of asphalt mixtures with NovaChip^® gradation is higher than that of mixtures with GT-8 gradation, but the difference does not exceed 9%. Pavements with higher ceramic grain content require more asphalt. High-viscosity, high-elasticity asphalt mixtures exhibit more pronounced brittle characteristics at low temperatures, making them more prone to fracture. Therefore, as the ceramic grain content increases, the bending failure energy density of the asphalt mixture decreases, resulting in poorer low-temperature properties. A comprehensive analysis of the two low-temperature performance evaluation methods indicates that, compared to a ceramic grain content of 60%, the low-temperature property differences at lower ceramic grain contents are insignificant.

3.2.3. Immersing Marshall Test

Based on the immersion residual stability test results for different ceramic grain contents, the residual stability variation curve is plotted, as shown in Figure 11.

According to the Technical Specification for Construction of Highway Asphalt Pavements (JTG F40) [26], the residual stability in the Immersing Marshall Test must not be less than 80%. As shown in Figure 10, the results for both gradations meet this requirement after adding ceramic grains. Additionally, with increasing ceramic grain content, the residual stability of the specimens shows a slight decreasing trend, with a reduction of approximately 1%. This is attributed to the increase in the void ratio of the asphalt mixture caused by the addition of ceramic grains, leading to reduced stability after immersion. The residual stability of the NovaChip^® gradation asphalt mixture is consistently higher than that of the GT-8 gradation. This can be explained by the critical void ratio for water damage in asphalt mixtures, which is 5% to 14%. When the void ratio is below 5%, the pavement is dense enough to prevent water intrusion, and when it exceeds 14%, water drains quickly from the structure. Since the void ratio of NovaChip^® exceeds 15%, water does not remain in the pavement structure, significantly reducing water-induced damage.

3.2.4. Snowmelt Property of Different Ceramic Grain Contents

The Ultra-Thin Wearing Course, as a functional wearing layer, is required to meet the performance demands of asphalt mixtures for road applications and exhibit excellent anti-icing performance and snowmelt properties. The snow melting effects of four types of specimens with different snowmelt agent contents, such as 0, 20%, 40%, and 60%, are shown in Figure 12. The PACS software was used to quantitatively calculate the snow accumulation on specimens with varying ceramic grain contents. The calculation results are shown in Figure 13.

As shown in Figure 12, with continuous snowfall, a thin layer of snow appeared on the road surface. The specimen without ceramic grains content showed accumulation of snow on its surface. After calculation, the snow accumulation on specimens with different ceramic grain contents accounted for 54.35%, 48.45%, 32.54%, and 22.60% of the total road surface area, respectively. It can be concluded that as the ceramic grains content increases, the amount of ice and snow remaining on the road decreases, resulting in better snowmelt properties.

3.2.5. The Impact of Different Ceramic Grain Contents on Chloride Ion Release

Asphalt mixture specimens containing snowmelt agents release Cl⁻ ions when soaked in water. Therefore, a chloride ion content rapid determination instrument was used to detect the release pattern of the snowmelt agent in the Ultra-Thin Wearing Course, providing a quantitative evaluation of its anti-icing and snowmelt properties. The variation in chloride ion concentration in the solution with soaking time under different ceramic grain contents is shown in Figure 14.

As shown in Figure 14, the solution soaked with the specimen without ceramic grains showed almost no chloride ion leaching. Within 180 min, the three groups of specimens with ceramic grain contents of 20%, 40%, and 60% all exhibited an increase in chloride ion concentration over time. Furthermore, as the ceramic grains content increased, the growth in chloride ion concentration became more pronounced. In all experimental groups, it was observed that chloride ions in the Ultra-Thin Wearing Course leached rapidly during the first 30 min, and after 30 min, the leaching rate slowed down.

3.2.6. Anti-Icing Performance of Different Types of Snowmelt Agents

In investigating the anti-icing performance of different types of snowmelt agents, the ceramic grains content was set at 40%. To evaluate the anti-icing performance of the specimens, three indicators were used: freezing rate, bubble rate, and ice formation rate, which represent the worst, intermediate, and best anti-icing performance, respectively. The snowmelt agents used in the experiment included NaCl, MgCl₂, and CaCl₂. Three sets of rutting plate specimens were prepared with these three snowmelt agents. A total of 400 mL of water was spread on the surface of the specimens, which were then frozen at −5 °C and −10 °C for 2 h. The freezing conditions of the surface water of the specimens are shown in Figure 15.

The PACS software was used to quantitatively calculate the freezing, bubble, and ice formation rates of liquid water on the surfaces of different specimens. At −5 °C, the specimen without any snowmelt agent had its surface water completely frozen, with a freezing rate of 23.73%. The specimen with NaCl as the snowmelt agent had partially frozen surface water, with a significant amount of bubbles, resulting in a bubble rate of 15.07%. The specimens with MgCl₂ and CaCl₂ as snowmelt agents showed no freezing on the surface, maintaining the liquid state, with an ice formation rate of 0%. At −10 °C, the specimens without snowmelt agents, and those with NaCl and MgCl₂ as snowmelt agents, all had their surface water completely frozen, with freezing rates of 43.26%, 37.15%, and 28.25%, respectively. The specimen with CaCl₂ showed partial freezing of the surface water and some bubbles, with a bubble rate of 19.34%. Thus, the anti-icing performance of CaCl₂ was the best, while NaCl performed the worst. In colder regions, for better anti-icing performance, it is recommended to replace NaCl with MgCl₂ or CaCl₂.

Chloride-based deicers (such as NaCl with MgCl₂ or CaCl₂) accelerate the melting of ice and snow by lowering their freezing points. However, during application, there are certain environmental risks associated with their use. Therefore, it is recommended that ceramic aggregates soaked in CaCl₂ should be thoroughly washed before use to remove any residues, minimizing adverse effects on the environment and pavement performance. Additionally, it is advised to monitor chloride ion release during the actual construction process to assess the long-term impact on the surrounding environment, enabling preventive measures and timely intervention if necessary.

3.2.7. The Impact of Different Snowmelt Agents on Chloride Ion Release

The effect of different snowmelt agents on chloride ion release is shown in Figure 16.

As shown in Figure 16, the solution soaked with the specimen without ceramic grains showed almost no chloride ion leaching. During the entire test period, the three groups of specimens with ceramic grain contents of 20%, 40%, and 60% all showed an overall increase in chloride ion concentration with the passage of time. In the first 30 min, chloride ions leached rapidly, and after 30 min, the leaching rate gradually slowed down.

4. Conclusions

This study designed two gradations of Ultra-Thin Wearing Course (UTWC), GT-8 and NovaChip^®, to investigate the effects of ceramic grain content on UTWC road performance and snowmelt property. It also explored the influence of different snowmelt agents on the anti-icing performance of UTWC and proposed the optimal ceramic grain content. The main conclusions are as follows:

• This study examined the high-temperature performance, low-temperature performance, and water stability of GT-8 and NovaChip^® gradations of UTWC. The effects of varying ceramic grain contents on the road performance of asphalt mixtures were analyzed. Overall, the road performance of the NovaChip^® gradation was superior to that of the GT-8 gradation.
• As the ceramic grain content with adsorbed snowmelt agents increased, the release of Cl⁻ ions within the compatible time frame was maximized, and the snowmelt property of the UTWC improved.
• Among the tested snowmelt agents, CaCl₂ demonstrated the best anti-icing performance, while NaCl was the least effective. In colder regions, replacing NaCl with MgCl₂ or CaCl₂ is recommended to achieve better anti-icing performance.
• Considering the road performance, snowmelt property, and anti-icing performance of UTWC, a ceramic grain content of approximately 40% is recommended.