1. Introduction
Red mud is a reddish brown colored alkaline solid waste from the alumina refining of bauxite ore [
1]. Bayer red mud is the industrial waste slag produced by the Bayer process in the alumina plant. A ton of alumina products generates about 0.8–1.5 tons of red mud according to the grade of bauxite ores and operating conditions [
2]. China is the largest producer of alumina and red mud in the world [
3]. About 30 million tons of red mud are generated yearly in China, and nearly 4 billion tons of red mud has now accumulated [
4].
The treatment and utilization of red mud has been a huge challenge for the governments and alumina industry around the world, especially in China. In recent years, investigators around the world have researched the treatment and utilization of red mud. These investigations mostly focused on utilizing the red mud to produce building materials (such as bricks, novel inorganic polymer paving blocks, cement, concrete, glass, and ceramics) [
5,
6,
7,
8,
9,
10,
11,
12], recover valuable elements [
13,
14], purify gas [
15], treat water [
16], and improve soil [
17]. The building materials industry has been very interested in the treatment and utilization of red mud, because it can be cost effective. However, the high alkalinity (pH 10–12.5) of red mud results in difficulties when used as raw materials for buildings. Fortunately, asphalt in the asphalt mixture is slightly acidic and the alkalinity of the red mud is beneficial to improving the adhesion between the asphalt and the aggregate, according to chemisorption theory.
Mineral fillers are an important part of asphalt mixtures. It has a great influence on the cohesive performance of asphalt binders [
18,
19]. Traditionally, naturally alkaline limestone powder (LSP) is used as the most common mineral filler in asphalt mixtures. However, limestone is a natural resource and is being exhausted with the increased development of the cement and construction industry in China [
20]. The production of LSP causes a large amount of dust which will pollute the environment. Therefore, many investigators have been studying waste materials to replace the LSP for use as asphalt binders and in asphalt mixtures. Researchers have successfully found several waste materials (such as steel slag, recycled red brick powder, rice husk ash, coal waste powder, fly ash, and waste lime) which can replace LSP as fillers in asphalt mixtures [
21,
22,
23,
24,
25,
26]. The utilization of these waste materials as fillers in asphalt mixtures enlarges their application fields and can make them valuable commodities. Even with these new materials, the requirements needed by the rapid development of the construction and traffic infrastructure in China cannot be met. These materials are not evenly distributed throughout China and their high processing costs also limit their use. Also, only some of these new materials are well suited for use in asphalt mixtures. Many of the materials are not alkaline which adversely affects the adhesive properties between the asphalt and the aggregate. Therefore, additional materials are urgently needed to supplement them.
Red mud, as a waste alkaline material, not only occupies a large amount of land resources, but also results in serious environmental pollution [
27]. Red mud has been widely used as a building material in the field of construction engineering. However, the application of red mud in the field of road engineering has only been reported in the literature a few times [
28,
29], especially in asphalt binder and asphalt mixtures. This paper mainly investigates the performances of asphalt binders with red mud powder (RMP) obtained by special treatment, to replace LSP either partially or completely as a filler.
In order to study the performance of asphalt binder adding RMP, various tests were performed, including penetration, softening point, rotational viscosity (RV), dynamic shear rheometry (DSR), and bending beam rheometry (BBR). Comparisons with asphalt binders prepared using the conventional LSP are also made. RV, DSR, and BBR are the tests used to determine if asphalt binders can be used in Superpave applications. Superpave is one of the asphalt outcomes of the Strategic Highway Research Program (SHRP) of the United States between 1987 and 1992. Superpave presented new tests and standards for asphalt binders, which provides a basis for the of study asphalt binders in this paper.
2. Materials and Methods
2.1. Materials
In this study, the AH-70 base asphalt was supplied by the Koch Asphalt Company (HongKong, China). It was used in the preparation of asphalt binder. As per JTG E20-2011 [
30] of China, the penetration (0.1 mm at 25 °C, 100 g, and 5 s), the softening point and the ductility (at 15 °C and 5 cm/min) of the base asphalt is 69 (0.1 mm), 46.7 °C, 163 cm, respectively.
Natural LSP, which will be used as a reference, was obtained from Baofeng County, which is located in the Henan Province of Central China. The Bayer red mud used in this paper came from the Jiaozuo Zhongzhou alumina refinery in the Henan province of Central China. It is the largest aluminum plant in Henan province and produces about 2 million tons of red mud annually. At this moment, 19 million tons of red mud have already been stockpiled over an area which covers more than 1300 acres, which is depicted in
Figure 1;
Figure 2. The red mud was dried at 105 °C for 5 h, and then crushed by a jaw crusher and ground by ball mill for 30 min to obtain the RMP, as shown in
Figure 3. The basic performances of LSP and Bayer RMP are given in
Table 1. The chemical compositions determined by X-ray fluorescence spectroscopy (XRF) (Panalytical Axios, Almelo, Holland) can be found in
Table 2.
Table 2 shows that the major oxide component of LSP is CaO. CaO is an alkaline compound which affects the bond between the weakly acidic asphalt and the aggregate. However, the concentration of CaO of Bayer RMP is less than in LSP. Therefore, the combination of Bayer RMP and LSP in asphalt binders can help improve the adhesive properties between the asphalt and the aggregate compared to RMP by itself. This may be because the larger amounts of CaO in the LSP can make up for the CaO deficiency in the Bayer RMP.
2.2. Preparation of the Asphalt Binder
In this study, five types of filler were used to prepare the asphalt binder. These fillers were composed of 0%, 25%, 50%, 75%, and 100% RMP instead of LSP respectively. Filler combinations are given in
Table 3. For these fillers, seven different filler-to-asphalt (F/A) ratios were studied: 0.3, 0.6, 0.9, 1.2, 1.5, 1.8 and 2.1, respectively.
During the preparation of the asphalt binder, the base asphalt was heated in the mixing container inside the oven at 140 °C for 60 min until melt. According to the chosen F/A, a certain quantity of the filler heated at 140 °C was mixed with the base asphalt and stirred for 20 min at a speed of 800 rpm, until an even mixture was obtained. The asphalt binders were kept at room temperature for the subsequent performance tests.
2.3. Methods
In this work, the main test methods of the asphalt binder included penetration, softening point, rotational viscosity, dynamic shear rheometry (DSR) and bending beam rheometry (BBR). Softening point, rotational viscosity (RV), and penetration are the basic performance parameters of road asphalt, which have been applied to evaluating asphalt binder [
31,
32]. Penetration (0.1 mm at 100 g and 5 s) at 25 °C, softening point (ring and ball), and rotational viscosity at 135 °C of the asphalt binder were tested according to JTG E20-2011 [
30] of China. Rotational viscosity was measured using a rotational viscometer (NDJ-1C, CHANGJI, Shanghai, China). The asphalt sample was poured into a chamber holder and then inserted into the RV chamber to achieve the desired temperature of 135 °C. Viscosity of the asphalt binder was determined at 135 °C. A cylindrical spindle was submerged in the chamber and was rotated at a speed of 20 rpm.
In this study, dynamic shear rheometry (DSR) (DHR-I, TA, New Castle, DE, USA) and bending beam rheometry (BBR) (ATS, PA, USA) were applied to characterize the rheological performances of all asphalt binders. The dynamic shear rheometry (DSR) test was performed at 60 °C at a fixed frequency of 10 rad/s. Parallel plates with a diameter of 25 mm were used to prepare binder samples with the thickness of 2 mm. The asphalt binders were heated until they became sufficiently fluid enough to be poured into silicone molds to obtain the DSR test samples. The test was carried out with 25 mm diameter, 1 mm gap geometry at 60 °C. Through the DSR test, the complex shear modulus (G*), phase angle (δ), and rutting factor (G*/sin δ) of the asphalt binder were analyzed in detail. The bending beam rheometry (BBR) test was performed to determine the creep response of the asphalt binder at −12 °C and a loading time of 240 s. The size of the specimen was 127 × 6.35 × 12.7 mm. During the test, a beam of the specimen was submerged in a constant temperature bath and kept for 60 min. A constant load of about 100 g was applied after preloading on to the rectangular beam. The beam was supported by stainless steel half rounds on both ends. By measuring the deflection of the center of specimen continuously, the creep stiffness (S) and the rate of change of creep stiffness (m-value) of the asphalt binders were obtained.
4. Conclusions
This study evaluated the performance of asphalt binders using RMP as a filler, instead of the commonly used LSP. Based on the results of the studies performed, the following main conclusions can be given:
- (1)
In the Bayer red mud, Fe2O3, Al2O3, and SiO2 were the dominant components accounting for nearly 67% of the total weight. Na2O was detected in the Bayer red mud samples and contributes a strong alkaline effect, with a concentration of 3.21%. The higher Na2O concentration indicates why the red mud powders have higher PH values in comparison with more traditional natural mineral fillers. Moreover, this alkalinity may be beneficial in improving the adhesion between the asphalt and aggregates, due to the weak acidity of the asphalt.
- (2)
As the F/A ratio increases, the softening point increases gradually, and the penetration decreases linearly. The effects of RMP on the softening point and the penetration is more prominent than LSP.
- (3)
Exponential functions provide the best descriptions of the relationship between the rotational viscosity and the F/A ratio. The higher the amount of RMP in the binder, the more noticeably the rotational viscosity changes, and the smaller the F/A ratio which corresponds to the Superpave requirement of 3 Pa·s. For RMP100 asphalt binder, the F/A ratio that meets this requirement is 1.2.
- (4)
G* and G*/sin(δ) have an exponential relationship with the F/A ratio. RMP75 asphalt binder has the highest proportion of elastic component, because δ is the smallest. Moreover, the deformation resistance and the rutting resistance of the RMP75 asphalt binder at high temperature are the strongest in this work. LSP (RMP0) asphalt binder is the opposite, with the weakest deformation and rutting resistance.
- (5)
As the F/A ratio increases, the low temperature properties (S and m-values) of the asphalt binders decrease gradually. With increased substitution of RMP for LSP, the low temperature properties decrease more significantly. In order to meet the Superpave requirements, S values need to be less than 300 MPa, and the F/A ratio for the binders in this work must be less than 0.9. In addition, with the exception of the F/A ratios of 1.8 and 2.1 for the RMP100 asphalt binder, all m-values meet the Superpave requirement of not less than 0.3.
With the rapid development of road construction in China, the demand for asphalt fillers is very large. However, since the crushing and ball-milling processes lead to large amounts of environmental pollution, many traditional filler plants have been closed. This has led to a sharp increase in the price of traditional fillers. Considering the drying and transportation costs, the application cost of red mud as filler in asphalt mixture should be lower than traditional fillers. Moreover, the use of red mud can reduce land occupation and environmental pollution. In short, considering the high temperature deformation resistance, low temperature crack resistance, easy construction, and economic factors, it is suggested that the best ratio of RMP, instead of traditional LSP, is 75%. At this point, the optimal F/A of asphalt binder is 1.0.