Investigation of the Wet and Thermal Conditions Effect on the Micro-Scale Characteristics of Interfacial Transition Zone of Porous Asphalt Mixture

Abstract

Complex environmental factors can significantly influence the micro-properties of porous asphalt (PA) mixture. Therefore, the effects of short-term and long-term thermal aging and moisture immersion on the micromechanics properties, micro-morphology, and chemical element distributions of the interfacial transition zone (ITZ) of PA mixture were studied to reveal the mechanisms by which environmental degradation impact performance by means of nanoindentation (NI), backscattering scanning electron microscope (BSEM), and energy-dispersive X-ray spectroscopy (EDS) analysis techniques. The results show that the ITZ is not the softest part of the PA mixture, and the order of modulus is aggregate > ITZ > asphalt mastic. ITZ thickness is about 10–20 μm. Thermal aging has negligible effect on the width of ITZ. With increasing aging degree, the asphalt mastic and ITZ modulus increase, while water damage reduces the modulus of ITZ and slightly increases the width of ITZ. Thermal aging has little effect on the morphology of ITZ, while water damage will lead to microcracks and micropores in ITZ. Compared with thermal aging, water damage has a greater effect on the morphology of ITZ and leads to softening of the interfacial transition zone and asphalt mastic phase. The width of ITZ obtained by EDS line scanning is basically consistent with that of nanoindentation. Overall, external environmental factors have a more significant effect on the mechanical properties than the micro-morphology of ITZ. The outcomes of this research provide a better understanding of the impact of the service environment on the microscopic characteristics of PA mixture.

1. Introduction

Porous asphalt mixture (PA) has the advantages of drainage, noise reduction and water mist reduction due to its high void ratio (18%–22%) [1,2]. However, compared with dense-graded mixture, PA mixture with a high void ratio is more easily damaged by media such as sunlight, air and water in the environment [3,4,5,6], having a negative impact on the asphalt–aggregate microscopic interface, thus leading to a decline in the macro performance of the mixture [7]. It is necessary to analyze the effects of the external environment on the performance of PA mixture at the micro level, so as to gain insight with the aim of improving its structural durability.

In the process of the mixing and compaction of PA mixture, a series of physical, chemical and mechanical actions between asphalt and filler occur around the surface of the aggregate, thus forming a narrow area around the aggregate, which is called the interfacial transition zone (ITZ) [8]. The microstructure affects the macroscopic characteristics of PA [9,10,11], and ITZ is regarded as a weak area because of its relatively low strength [12]. The ITZ between the aggregate and the asphalt mastic has an important effect on the macroscopic mechanical behavior of the PA mixture, because the generation and development of damage in the PA mixture mainly occur in these low-strength zones [13,14,15]. In addition, studies have shown that during concrete freezing and thawing, ITZ exhibits higher deformation than asphalt mastic and aggregates due to its higher porosity and weaker strength [16]. Determining the properties of the interfacial transition zone (ITZ) between asphalt mastic and aggregate is helpful for understanding some of the mechanisms of asphalt pavement fracture behavior [17]. It is difficult to study ITZ using traditional experimental methods due to the scale of ITZ being at the micro level. In recent years, with the development of nanoindentation technology, it has become possible to use nanoindentation (NI) to directly measure the width and nano-mechanical properties of ITZ [18,19]. In addition, backscatter scanning electron microscopy (BSEM) is an effective tool for visualizing ITZ [20], and is generally used in conjunction with energy-dispersive X-ray spectroscopy (EDS) to qualitatively reveal the chemical characteristics of ITZ [21,22]. According to NI analysis, Huang [23] found that the thickness and elastic modulus of ITZ varied with different components in the recycled concrete aggregate, and ITZ adjacent to the porous components (cement mastic and brick) had greater thickness and lower elastic modulus, and verified the accuracy of the thickness measured by nanoindentation by BSEM and EDS. Zhu [13] also used NI to study the interfacial transition zone between different aggregates and asphalt mastic. It was found that the width of ITZ was 5–20 μm, and the accuracy of NI test results was verified by EDS.

To date, many studies have successfully been performed characterizing the interfacial transition zone of asphalt concrete mixture using different microscopic testing methods, but few studies have addressed the effect of wet and thermal aging on the interfacial transition zone of PA mixture. Therefore, determining the properties of the interfacial transition zone of PA mixture and the influence of external environmental factors is helpful for revealing the influencing mechanism of PA mixture durability [24].

The main objective of this study is to investigate the effect of the wet and thermal conditions on the micro-scale characteristics of ITZ of porous asphalt mixture. To achieve this objective, PA samples were prepared and subjected to different environmental factors, and NI, BSEM and EDS were used to study the micro-mechanics, micro-morphology and chemical properties of the interfacial transition zone between asphalt mastic and aggregate in PA mixture.

2. Materials and Test Methods

2.1. Material Properties and Gradation Design

The raw materials of the PA mixture include SBS-modified asphalt, basalt aggregate, and limestone mineral powder, and their technical properties are able to meet the requirements of Technical Specification for Construction of Highway Asphalt Pavement (JTG F40-2004). The detailed technical properties of asphalt are shown in

Table 1. Basic technical properties of SBS-modified asphalt.

Technical Properties	Detection Result
Penetration (25 °C, 100 g, 5 s)/0.1 mm	55.5
Ductility (5 °C, 5 cm/min)/cm	37.5
Softening point/°C	78.5
Dynamic viscosity (60 °C, Pa·s)	4841

. During the mixing process of the PA mixture, high-viscosity modifier is added, with a mixing amount of 8% of asphalt mass [25], in order to improve the durability of PAC. The gradation composition of the mixture is shown in

Table 2. Mixture gradation.

Mesh Size (mm)	16	13.2	9.5	4.75	2.36	1.18	0.6	0.3	0.15	0.075
Pass rate (%)	100	92.2	60.7	20.5	14.1	11.3	7.6	6	5.3	4.6

, and the optimal asphalt–stone ratio is 4.6%.

2.2. Nanoindentation Test

2.2.1. Sample Preparation

Firstly, the method of accelerated aging under high temperature with forced ventilation found in the Standard Test Methods of Bitumen and Bituminous Mixtures for Highway Engineering (JTG E20-2011) was used to prepare the original, short-term aging and long-term aging gyratory compaction samples of PAC-13. The preparation process of that nanoindentation sample is shown in Figure 1, and the specific preparation method is as follows:

(1) Cutting. Cut a 10-mm-thick slice in the vertical middle of the mixture, and then cut a 10 mm × 10 mm × 10 mm cube of the mixture from the middle of the slice as shown in Figure 1a. To reduce the effect of cutting on the internal structure, the sample was cured in low temperature condition at −20 °C for 24 h before cutting.

(2) Curing of epoxy resin. The nanoindentation specimen requires high flatness, so the surface of the specimen needs to be polished. In order to facilitate polishing and prevent the specimen from tilting during the test, the mixture cube block needs to be embedded in the epoxy resin. First, put the mixture test block into a silica gel mold with a diameter of 20 mm, then mix the epoxy resin and curing agent in a ratio of 3:1, and pour them into the silica gel mold after the bubbles basically disappear as shown in Figure 1b.

(3) Grind and polish. After the resin is cured, use a metallographic grinder to polish the surface of the test sample as shown in Figure 1c, remove the epoxy resin wrapped on the surface, and expose the test surface. The test sample should be polished step by step with 180-, 400-, 800-, 1200-, 1600- and 2000-mesh abrasive paper for 5 min for each grade of abrasive paper. After polishing, spray diamond spray polishing agent on the surface of the test sample, and then further polish on polishing cloth for 20 min. Finally, the sample was cleaned in an ultrasonic cleaner for 3 min to remove dust, diamond particles, and the like. Put it into a sample storage box and air dry it for later use as shown in Figure 1d.

(4) Cure preservation in water bath. In order to study the influence of water on the interfacial transition zone between asphalt and aggregate, the samples of unaged PA mixture prepared according to the above steps need to be cured in water bath, and the curing condition is to soak in water at 60 °C for 1 day. After air drying, it can be used for testing.

After the above preparation, four kinds of sample were obtained under the action of external environment, including: as-received, short-term-aged, long-term-aged and high-temperature water-bath-treated specimens.

2.2.2. Test Method

In this paper, a micro nanoindentation instrument produced by Nano Force Company of America (Nanomechanics, Inc., Oak Ridge, TN, USA.) is used to conduct the nanoindentation test. As shown in Figure 2, a Berkovich indenter, which is more suitable for asphalt materials, is used as the indenter. Firstly, the area of interest (AOI) of asphalt and aggregate is found by transmission electron microscope, as shown in Figure 3, and 10 × 4 grid points (column interval 25 μm, row interval 10 μm) are arranged on AOI as nanoindentation test points. The influences of residual stress and deformation on the next measuring point are eliminated. In accordance with the relevant research results [22], the nanoindentation parameters in this paper were a loading rate of 0.01 mN/s, a maximum load of 0.3 mN, a full load time of 200 s, and the same load control mode was selected for asphalt, asphalt–aggregate interfacial transition zone and aggregate.

2.2.3. Calculation Principle

A typical load–displacement curve obtained from a nanoindentation experiment is shown in Figure 4, which consists of three stages: a loading phase, a full load phase and an unloading phase. The Young’s modulus can be calculated from the unloading curve:

$$E_r=\frac{1}{\beta }\cdot \frac{S}{2}\cdot \frac{\sqrt{\pi }}{\sqrt{A_c}}$$

(1)

In the formula, S is the contact stiffness, obtained from the initial slope of the unloading curve, $\beta$ is the geometric correction factor, where $\beta =1.034$ for the Berkovich indenter used in this study, $A_c$ is the contact area at peak maximum load, which is related to the contact depth $h_c$ at the half angle of the Berkovich indenter as shown in Equation (2), where $\theta =65.27°$. $P_{max}$ is the maximum load, $h_{max}$ is the maximum indentation depth, $\epsilon$ is a geometric constant, $\epsilon =0.75$.

$$A_c=3\sqrt{3}{h}_c^{2}ta{n}^2\theta$$

(2)

$$h_c=h_{max}-\epsilon \frac{P_{max}}{S}$$

(3)

$E_r$ is the composite modulus of the indenter and the sample. For the independent analysis of the test material, the composite modulus is converted to the elastic modulus. The elastic modulus of the sample can be obtained from Equation (5).

$$\frac{1}{E_r}=\frac{1-v_s^2}{E_s}+\frac{1-v_i^2}{E_i}$$

(4)

$$E_s=\left(1-v_s^2\right)\cdot {\left[\frac{1}{E_r}-\frac{1-v_i^2}{E_i}\right]}^{-1}$$

(5)

In the formula, ${\nu }_s$ represents the Poisson’s ratio of the sample, ${\nu }_i$ represents the Poisson’s ratio of the indenter, the elastic modulus of the indenter used in this study is taken as $E_i=1140$ Gpa, ${\nu }_i=0.07$, the Poisson’s ratio ${\nu }_s$ of the sample has little influence on the calculation result of the elastic modulus, and ${\nu }_s=0.3$ [13] is adopted in this study.

2.3. BSEM and EDS Tests

The morphology of the interfacial transition zone was obtained by scanning electron microscopy (SEM, Hitachi Limited, Tokyo, Japan) in backscattered electron (BSE) mode. The sample preparation method is similar to that of the NI sample. After the epoxy resin is cast, it is cut into cylinders with a height of 5 mm. Since the BSEM test does not require surface flatness, no subsequent grinding is required. The pixel brightness of the BSE image corresponds to the atomic number of the measured phase below the sample surface. Therefore, the voids are the darkest, followed by asphalt and aggregate, so that the BSE image has a clear distinction between asphalt and aggregate. It should be noted that, due to the non-conductivity or poor conductivity of the sample, in order to avoid the impact of the “charging effect”, the sample needs to be sprayed with gold before testing.

The elemental composition and distribution of the sample can be characterized using energy-dispersive X-ray spectroscopy (EDS) detection (Oxford Instruments Nanotechnology Tools Limited, Oxford, UK). In this study, EDS test was carried out with BSE images to obtain the basic mapping of the interfacial transition zone. There are three kinds of EDS test methods: point scanning, line scanning and surface scanning. In this study, surface scanning was used to obtain the distribution of elements in the interfacial transition zone, and then line scanning was used to study the thickness of the interfacial transition zone between asphalt and aggregate with the change in element concentration.

3. Test Results and Analysis of ITZ

3.1. Micromechanical Properties

Taking the original sample as an example, the test section is near the ITZ, as shown in Figure 5. The elastic modulus of each effective indentation was calculated, and a two-dimensional modulus contour map of the test area was drawn, as shown in Figure 6. The asphalt mastic phase is a blue area with the lowest modulus; the red area is the aggregate phase with the highest modulus; the green area in the middle is the asphalt mastic aggregate interfacial transition zone (ITZ), whose modulus is between that of the asphalt mastic and the aggregate. Compared with asphalt mastic, the mechanical properties of ITZ in PA mixture are closer to those of aggregate [26,27]. This may be due to the combination of asphalt and filler (the size is within the scope of nanometers or no more than a few microns) to fill the micro surface pores of the aggregate and penetrate into the aggregate, forming a mechanical interlock. In addition, due to the chemical bonding between asphalt and aggregate and electrostatic force, a harder and stronger asphalt stone composite structure, namely the interfacial transition zone (ITZ), is formed around the aggregate surface than asphalt mastic. It can be seen from the two-dimensional equivalent modulus diagram that the width of the ITZ is roughly between 10–20 μm.

The elastic modulus is used to determine the thickness and mechanical properties of the interfacial transition zone. In the change trend chart of the elastic modulus, the “sudden drop” zone in the middle is generally identified as the interfacial transition zone, from which the thickness of the interfacial transition zone can be determined. Analyzing 40 data points consisting of 10 rows and four columns, the average value of each row of data was counted, and the curve graph of modulus was drawn, as shown in Figure 7. The width of ITZ before and after the aging of the PA mixture is similar, at about 18 μm. In addition, the width of ITZ increased slightly to about 21 μm after water immersion, which may be caused by the loosening of ITZ caused by water damage. The average value of modulus of each phase was calculated, as shown in

Table 3. Average modulus of each phase under different working conditions.

Project	Original			Short-Term Aging			Long-Term Aging			60 °C Water Bath 1 d
	AM	ITZ	BA	AM	ITZ	BA	AM	ITZ	BA	AM	ITZ	BA
Modulus (GPa)	4.72	15.81	40.00	5.82	17.46	39.81	8.21	20.36	40.12	3.28	12.21	39.54

Note: AM stands for asphalt mastic; BA stands for basalt.

. With increasing aging degree, the modulus of asphalt mastic phase and ITZ increased, but this had little effect on the modulus of aggregate phase, which was due to the increase in asphalt and hardening of asphalt caused by aging. This shows that aging has little effect on ITZ width, mainly with respect to asphalt mastic phase and ITZ mechanical properties. Contrary to the aging effect, the modulus of asphalt mastic phase and ITZ decreased after the action of water. It can be seen that the damage of water will lead to the softening of the interfacial transition zone and asphalt mastic.

3.2. Micro Morphology Characteristics

In this study, the BSEM test is used to calibrate and photograph the interfacial transition zone of samples under four working conditions (original sample, short-term aging, long-term aging and immersion). As shown in Figure 8, the interface between the interfacial transition zone and the aggregate is obvious, but there is no obvious interface between the interfacial transition zone and the asphalt mastic. There is no obvious difference in the interfacial transition zone between the original sample and the short-term aging sample. The interfacial transition zone and asphalt of asphalt mastic phase of long-term aging samples have wrinkle phenomenon and rough surface. This phenomenon is caused by the long-term aging of asphalt, the degradation of butadiene, and the destruction of long-chain polymer to form a shorter SBS polymer structure, which makes it impossible to fully surround filler particles, and the volatilization of light components reduces the lubricity of asphalt, which leads to a little microcracks can be seen in the ITZ. For the sample immersed in water bath at 60 °C for 1 h, the morphology of the sample is obviously different from that of the original sample, which is mainly reflected in the appearance of microspores in the asphalt mastic phase and the interfacial transition zone (ITZ), as shown by the yellow area, and the appearance of micro cracks in the ITZ. The asphalt mastic phase is separated from the aggregate phase, as shown by the blue area. In addition, it is observed that the asphalt in the ITZ decreases and becomes thinner, and more filler surfaces are exposed. The main reason for the above phenomenon is that the erosion of water forms a water film on the surface of the aggregate, resulting in an asphalt–water–aggregate system. The accumulation of water leads to the enhancement of the isolation effect of the water molecular film on the adhesion of the asphalt–aggregate [28], which reduces the adhesion properties of the asphalt to the aggregate, resulting in the stripping of the asphalt mastic phase and the aggregate phase. It can be seen that, compared with aging, water damage has a greater impact on the morphology of the interfacial transition zone.

3.3. Chemical Element Distribution

Due to the possible mismatch between nanomechanical properties and inhomogeneities, the determination of ITZ thickness based on nanoindentation test results alone may not be sufficient. To obtain more accurate thicknesses, ITZ in PA mixtures was identified using elemental analysis. Figure 9 is an element distribution diagram of the interfacial transition region under different working conditions. The brightness of the color represents the content of the element in the element mapping image. The brighter the color, the higher the content. It can be seen from Figure 9 that the asphalt mastic phase has the phenomenon of high calcium and low silicon, which is mainly due to the fact that the filler used in this study is limestone powder (with high calcium carbonate content), so the phenomenon of high calcium and low silicon occurs. The aggregate is basalt (with high silica content), so the aggregate phase is characterized by high silica and low calcium. Aging has no obvious effect on this phenomenon, but the effect of high temperature water bath can increase the content of calcium element in asphalt mastic phase, which also indicates that the effect of water will damage asphalt, reduce the adhesion between asphalt and filler, and expose the surface of filler, thus increasing the content of calcium element in mapping image, which is also consistent with the phenomenon observed in BSE image. After long-term aging, the brightness of O element in asphalt mastic increases slightly and the brightness of C element decreases, which indicates that the content of O element increases and the content of C element decreases after long-term aging, which is caused by asphalt aging, while short-term aging has little effect.

However, the mapping image can only demonstrate the element distribution, and cannot effectively identify the thickness of the interfacial transition zone. In this study, the length of the interfacial transition zone is 70 μm, in order to determine the thickness of the interfacial transition zone according to the changes in the number of elements.

Figure 10 shows the change in the quantity of main elements (C, Ca, Si and O) along the line of samples under different working conditions. Since each sample is made of the same material and the aggregate is basalt (with high silicon dioxide content), the Si element in the aggregate phase fluctuates at a high level, gradually decreasing when it reaches the interfacial transition zone, and then fluctuates at a low level in the asphalt mastic phase, and the O element shows the same trend. The trend for Ca and C is the opposite. Because the filler used is limestone (with high calcium carbonate content), the amount of Ca element fluctuates at a low level in the aggregate phase gradually increases in the interfacial transition zone, and then fluctuates at a higher level in the asphalt mastic phase. It can be seen from Figure 10 that the number of elements in the ITZ is between that of aggregate and asphalt mastic. According to the change in element content, the width of the ITZ is about 20 μm for the original sample, short-term aging sample and long-term aging sample, and about 23 μm for the sample immersed in water at 60 °C for 1 d. The above results are basically consistent with those obtained by nanoindentation. Therefore, the external environmental factors have little influence on the ITZ width, mainly on its mechanical properties.

4. Conclusions and Findings

The microscopic mechanical properties, microscopic morphology characteristics and chemical element distribution of the interfacial transition zone of PA mixture were studied on the basis of nanoindentation, BSEM and EDS tests, and the influence of wet and thermal environments on the microscopic properties of the interfacial transition zone was analyzed. The following conclusions and findings can be drawn:

(1) Nanoindentation grid arrangement is able to effectively reflect the three-phase performance distribution law of asphalt mastic-ITZ-aggregate. The interfacial transition zone is not the softest part of the PA mixture, and the modulus can be ordered as aggregate > ITZ > asphalt mastic. According to the equivalent modulus diagram, it is found that the thickness of ITZ is roughly between 10 and 20 μm, which is consistent with previous research conclusions.

(2) Aging has no effect on the width of ITZ, mainly affecting its mechanical properties. With increasing degree of aging, the modulus of asphalt mastic and ITZ increased. Water damage slightly increased the width of ITZ, which may be due to the loosening of ITZ caused by water, and the decreasing modulus of asphalt mastic and ITZ caused by water.

(3) There was no obvious difference between the BSE images of the interfacial transition zone of the original sample and the short-term aging pattern. The interfacial transition zone of the long-term aging sample and the asphalt of the asphalt mastic phase have wrinkles and rough surfaces; micropores and microcracks appear in the asphalt mastic phase and the interfacial transition zone of the water-immersed sample, and the erosion of water reduces the adhesion performance between asphalt and aggregate, resulting in the peeling of the asphalt mastic phase and aggregate phase. In addition, the high-temperature water bath caused asphalt loss of ITZ. Compared with aging, water damage has a greater impact on the morphology of the interfacial transition zone.

(4) The amount of elements in the interfacial transition zone is between the aggregate and the asphalt mastic, and the width of the interfacial transition zone can be well characterized by line scanning. The width of the interface transition zone of samples for the original and aged samples was about 20 μm, while that for the water immersed samples was about 23 μm, whereby the results obtained from EDS analysis were almost the same as the results obtained by nanoindentation.

Notably, the thermal and moisture characteristics have obvious influence with modulus and morphology of ITZ, which have been assumed to be the reason for the stripping of PA. Therefore, for future research, an investigation of the relationship between the difference of ITZ and the micro properties, especially the durability of PA, is required.