**1. Introduction**

Asphalt pavement is a kind of continuous pavement without joints, which is the main form of highway pavement and has many advantages, such as smoothness, low noise, durability, anti-skid properties and easy maintenance [1]. The mixing temperature during the production process of hot-mixed asphalt mixtures (HMAs) is around 150–180 ◦C [2,3], and these need to be kept at A high temperature during the production, transportation and paving stages [4]. However, heating the asphalt mixture to high temperature requires a lot of fuel consumption [5], and emits greenhouse gases and toxic gases [6]. Besides this, the high construction temperature of HMA limits its application in cold areas [7]. The concepts of saving energy and consumption reduction have promoted the green production of asphalt materials, which has made researchers pay attention to the warm mix asphalt (WMA) [3,8].

WMA technology originated in Europe [9], and has been widely used all over the world [10]. The purpose of WMA technology is to reduce the viscosity of asphalt binder at high temperature [11,12] by various methods such as the use of organic viscosity reduction

**Citation:** Li, Y.; Feng, J.; Chen, A.; Wu, F.; Wu, S.; Liu, Q.; Gong, R. Effects of Low-Temperature Construction Additives (LCAs) on the Performance of Asphalt Mixtures. *Materials* **2022**, *15*, 677. https:// doi.org/10.3390/ma15020677

Academic Editor: Krzysztof Schabowicz

Received: 11 December 2021 Accepted: 11 January 2022 Published: 17 January 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

additives [13], the foaming viscosity reduction method [14] or using surface-active viscosity reduction additives [15] without reducing the construction quality and mixture performance of the resulting asphalt mixtures, so as to reduce the mixing temperature of these mixtures [8,16]. Compared with HMA, WMA can reduce fuel consumption by more than 40% and reduce the asphalt aging effects caused by high temperatures [17]. Liseane [18] studied the emission and energy consumption of different kinds of asphalt mixtures, and found that the formation temperature of WMA was 20–50 ◦C lower than that of HMA, and the greenhouse gas emissions were decreased by more than 40%. Maríadel [19] found that the reduction of the mixing temperature can effectively reduce the harmful gases such as CO2 and SO2 produced during the asphalt mixture production process. Mosa [20] used alum instead of traditional WMA additives. The results showed that alum could reduce the mixing temperature and compaction temperature by 25.5 ◦C and 20 ◦C, respectively. Gao [21] indicated that a small amount of WMA additive can reduce the viscosity of asphalt by more than 80%. Huang [22] and others studied the fatigue and mechanical properties of WMA by the discrete element model and found that the fatigue and mechanical properties of warm-mixed rubber asphalt mixture are better than those of ordinary asphalt mixtures. Although WMA is widely used and is considered a mature technology, studies have pointed out that its performance is actually rather poor. For example, Kim [23] studied the properties of WMA and HMA with the same performance graded (PG) asphalt binder and aggregate gradation, and the results showed that WMA had poorer rutting resistance than HMA. Tan [24] used the Marshall design method to study the physical, mechanical properties and mixture performance of WMA. The results showed that the optimum asphalt content of WMA is 0.1–0.2% higher than that of HMA, the air voids and flow value of WMA were slightly higher than those of HMA, and that the immersion residue Marshall strength of WMA is increased, but the freeze-thaw indirect tensile strength decreases.

Low-temperature construction additives (LCAs) are a new environmentally friendly asphalt modifier made from waste plastics or waste rubber [25]. The mixing temperature of LCA-modified asphalt mixtures (LCA-AMs) is even lower than that of WMA, but higher than the mixing temperature of cold mix asphalt mixture (CMA) [26]. Like WMA, LCAs can solve the technical problems of serious pollution and excessive energy consumption of HMA. LCA-AMs can provide a better viscosity reduction effect, and the preparation temperature of asphalt mixtures can be decreased by more than 40 ◦C, therefore the mixtures have better workability at lower temperatures (90–110 ◦C), and energy consumption and harmful gas emissions are reduced. They can be directly used for repairing and maintaining the surface layer of new pavement or old pavement [27]. LCA- modified asphalt binders have good storage stability, and it can be used at any time. Due to the small size of the LCA molecules, they may help improve the low-temperature performance and fatigue performance of mixtures [28]. Although LCA-AMs have many advantages such as greater workability, the initial mechanical properties of LCA-AMs are considered insufficient because LCAs reduce the viscosity of asphalt binders [29,30].

In this paper, LCA-AMs were designed and prepared. Firstly, the mechanical properties and preparation parameters of asphalt mixtures with and without LCAs were studied by the Marshall method. The influence of mixing methods and dosage of LCAs on the strength of LCA-AMs were investigated, so as to determine the best LCA-AM preparation parameters. Then a low-dose of epoxy resin (ER) (Hunan Baxiongdi New Material CO., LTD., Changsha, China) was applied to improve the initial mechanical properties of the LCA-AMs. Finally, mixture performance tests were conducted to study the rutting resistance, low-temperature crack resistance, water sensitivity and fatigue resistance of the thus prepared LCA-AMs.

#### **2. Materials and Experimental Methods**

*2.1. Materials*

2.1.1. Asphalt Binder

The original asphalt binder was 70# base asphalt binder (Hubei Guochuang Hi-tech Materials CO., LTD., Wuhan, China). The technical performance of this asphalt binder is shown in Table 1.


**Table 1.** Technical information of the 70# asphalt binder [27].

### 2.1.2. LCAs

LCAs are new environmentally friendly additives for asphalt mixtures, made from waste plastics and waste rubber. Among them, benzyl ethylene block copolymer accounts for about 80%, while ER and other additives account for about 20%. Figure 1a shows a typical LCA. After mixing with the asphalt binder, molecules containing amide groups are formed by ionization and recombination, and can form a dense film on the surface of asphalt molecules. LCAs can reduce the surface free energy and surface tension of asphalt molecules, resulting in a decrease in viscosity. At the same time, the lipophilic groups such as higher aliphatic chains, phenolic groups and benzene ring groups in ER can be adsorbed by asphalt molecules, which improves the stability of their dispersion system [31–33]. The mechanism of action of LCAs is shown in Figure 1b.

**Figure 1.** LCAs and their modification process in asphalt binder. (**a**) LCAs; (**b**) LCAs' mechanism of action.

#### 2.1.3. LCAs Modified Asphalt Binder

The LCA-modified asphalt binder was made by the melt blending method. First the 70# asphalt binder was heated to 110 ◦C. Then, 9% LCA was added to the 70# asphalt binder and shear mixed for 30 min. The optimum 9% dosage of LCA was determined in our previous research [27]. The technical information of the LCA-modified asphalt binder is shown in Table 2.


**Table 2.** Technical information of the LCA-modified asphalt binder.

#### 2.1.4. Aggregates and Filler

The aggregates were divided into the coarse aggregate (≥2.36 mm) and fine aggregate (0–2.36 mm). The coarse aggregates were basalt, the fine aggregates were limestone, and the filler was limestone powder. The technical properties of the aggregates and fillers are listed in Tables 3–5.

**Table 3.** Technical properties of coarse aggregate.


**Table 4.** Technical properties of fine aggregate.


**Table 5.** Technical properties of mineral powder.


*2.2. Design of LCA-Modified Asphalt Mixtures (LAC-AMs)*

2.2.1. Mix Design of LCA-AMs

A skeleton dense-graded mixture was used in this research, where the nominal maximum aggregate size of the mixture was 13 mm (LCAs-13). The aggregates were divided into four grades: 10–15 mm, 5–10 mm, 3–5 mm and 0–3 mm. The composite aggregate gradation range of LCAs is shown in Figure 2, and it has some common areas with the Chinese dense-graded mixture (AC-13). Both AC-13 and LCAs-13 are commonly used mixtures for the surface layer of asphalt pavement. The target gradation of LCAs-13 is

shown in Figure 2, and the mixing ratio of aggregates is shown in Table 6. The asphalt content was 4.8% by weight of aggregate.

**Figure 2.** Composite aggregate gradation of asphalt mixture.

**Table 6.** Mixing ratio of aggregates.


2.2.2. Manufacturing Temperature of LCA-AMs

The heating temperature of conventional HMA is clearly specified in JTG F40-2004. The heating and mixing temperatures for LCA-AMs are shown in Table 7. As it can be seen from Table 7, compared with the conventional HMA, the heating temperature of asphalt binder is reduced by 40 ◦C, the heating temperature of aggregate is reduced by 50 ◦C, and the mixing temperature of LCA-AMs is reduced by 30 ◦C.

**Table 7.** Construction temperature of conventional HMA and LCAs-AM.


2.2.3. Preparation Method of LAC-AMs

The aggregates, filler and LCA-modified asphalt binder were mixed at 120 ◦C for 180 s. The prepared LAC-AM was cured at 110 ◦C for 4 h. The high temperature curing could accelerate the volatility of volatile components in the LCA, increase the viscosity of LCAmodified asphalt binder, and promote the development of strength in the LAC-AM. The cured LAC-AM was mixed again at 110 ◦C for 120 s to prepare specimens, which were then stored at room temperature (25 ◦C) for 72 h before further tests. The preparation method is shown in Figure 3.

**Figure 3.** Preparation method of LAC-AMs.

#### *2.3. Experimental Methods*

#### 2.3.1. Volumetric Properties and Mechanical Performance Tests

The specimens were compacted using the Marshall method at 110 ◦C with both sides of specimens compacted 100 times, the specimens were then cured for 72 h at room temperature (25 ◦C). The volumetric properties and mechanical properties of specimens, including the volume of air voids (VV), the voids in mineral aggregate (VMA), the voids filled with asphalt (VFA), the Marshall stability (MS) and flow value (FL), were tested based on JTG E20-2011.

#### 2.3.2. ER Dosage and Mixing Methods

Epoxy resin (ER) at different dosages (0.3%, 0.6%, 1.0%, 2.0% and 4.0%) and different mixing methods were applied to study the differences between adding ER to LCA-AMs before and after curing, as shown in Figure 4. Method 1 involved adding ER before curing, while in Method 2 ER was added after curing for 4 h at 110 ◦C. The MS of specimens was tested after curing 72 h at room temperature (25 ◦C).

**Figure 4.** Different ER mixing methods.

#### 2.3.3. Rutting Resistance Test

The slabs used for the rutting resistance tests were prepared using the wheel tracking method (JTG E20 T0703). The rutting resistance tests were conducted according to the criteria of JTG E20 T0910. The rutting resistance was tested by the wheel track test machine (Beijing Aerospace Keyu Test Instrument, Beijing, China) at a wheel rolling load of 0.7 MPa and a rate of 42 times/min. The test temperature was 60 ◦C. After curing at 25 ◦C for 72 h, specimens (300 mm × 300 mm × 50 mm) were put in the wheel track machine for 5 h at 60 ◦C. The dynamic stability (DS) of LCAs-AM was then calculated according to Equation (1).

$$\text{DS} = \frac{(t\_2 - t\_1) \times N}{(d\_2 - d\_1)} \tag{1}$$

where DS is the dynamic stability of asphalt mixture (times/mm), *t*<sup>1</sup> and *t*<sup>2</sup> is the 45 min and 60 min, respectively, *d*<sup>1</sup> and *d*<sup>2</sup> are the rutting depths of asphalt mixtures at *t*<sup>1</sup> and *t*<sup>2</sup> respectively (mm), *N* is the speed of the testing wheel (42 times/min).

#### 2.3.4. Low-Temperature Resistance Test

The low-temperature crack resistance was tested by semi-circular bending (SCB) tests at the loading rate of 1.27 mm/min using a Universal Testing Machine (UTM-100); the distance between two supports was set to 75 mm. The specimen (ϕ 101.6 mm × 63.5 mm specimen) was cut into two semicircles, and a notch (5 mm length × 3 mm width) was cut along the height direction from the midpoint of the semicircle, as shown in Figure 5. The specimens were put into the UTM at −10 ◦C and 0 ◦C for 4 h before testing.

**Figure 5.** Load-displacement curve and test setup of SCB tests.

#### 2.3.5. Water Sensitivity Test

The immersion Marshall test and freeze–thaw split test were used to study the water sensitivity. The specimens for both tests (ϕ 101.6 mm × 63.5 mm) were tested at loading rates of 50 mm/min. Both sides of the specimens were compacted 100 times and 75 times, for the immersion Marshall test and freeze–thaw split test, respectively. The immersion Marshall test was conducted according to the criteria of JTG E20 T0709. The samples of unconditional group and conditional group were immersed in the 60 ◦C water bath for 30 min and 48 h respectively. The freeze–thaw split test was conducted according to the criteris of JTG E20 T0729. The samples of unconditional group were immersed in the 25 ◦C water bath for 2 h; for the conditional group, the Marshall specimens were first vacuum-filled with water, and the vacuum degree was 97.3–98.7 kPa, then frozen at −18 ◦C for 16 h. after that, put the specimens in 60 ◦C water bath for 24 h, and finally put in a 25 ◦C water bath for 2 h.

#### 2.3.6. Fatigue Resistance Test

The fatigue resistance was tested by the recycle semi-circular bending (R-SCB) test under different stress ratios of 0.3, 0.4, 0.5, 0.6 and 0.7 at the loading frequency of 2 Hz using UTM-100. The specimens were conditioned at 25 ◦C for 4 h, and the distance between supports was set to 75 mm.

#### **3. Results and Discussions**

#### *3.1. Volumetric Properties of LCAs-AM*

The volumetric properties and mechanical performance of LCAs-AM, including the volume of air voids (VV), the voids in mineral aggregate (VMA), the voids filled with asphalt (VFA), the Marshall stability (MS) and flow value (FL), are shown in Table 8. The comparison of volumetric properties and mechanical performance between 70# asphalt mixture (70#-AM) and LCA-AM is shown in Table 9. From Table 8, all of the volumetric properties meet the design requirements. From Table 9, compared to the 70#-AM, the VFA and MS of LCA-AM decreased by 1.5% and 28.3%, respectively, while the VV, VMA and FL of the LCA-AM increased by 7.5%, 4.0% and 25.7%, respectively. This indicates that adding LCAs decreases the viscosity of the asphalt mixture, resulting in the decrease of its MS.

**Table 8.** Volumetric properties and mechanical performance of LCAs-AM.


**Table 9.** Comparison of volumetric properties and mechanical performance between 70#-AM and LCAs-AM.


The LCAs in asphalt binder produce amide-containing molecules by ionised recombination, forming a dense film on the surface of the molecules. With the increase of temperature, the surface energy of the molecules decreases and subsequently their surface tension decreases, leading to a decrease in viscosity. As the viscosity decreases, the cohesion of the mixture is reduced, leading to a decrease in MS and an increase in FL. Although the average MS of 7.56 kN is slightly below the specification requirement of 8 kN, the LCA-AM strength will improve as the volatile solvents in the LCA gradually evaporate. In addition, the LCA contains around 20% ER, which can be used to improve the initial mechanical performance of LCA-AMs by adding a small amount of epoxy additive.
