Investigation of the Formation Mechanism and Environmental Risk of Tire—Pavement Wearing Waste (TPWW)

: Tire—pavement interaction behaviours result in large amounts of wearing waste matter, which attaches to the surface of the pavement and is directly exposed to the surrounding environment. This kind of matter imposes a great challenge to the environment of the road area. The current study is devoted to carrying out a comprehensive investigation of the formation mechanism of tire— pavement wearing waste (TPWW), as well as the resulting environmental risks. A self-developed piece of accelerated polishing equipment, the Harbin advanced polishing machine (HAPM), was employed to simulate the wearing process between vehicle tires and pavement surfaces, and the TPWW was collected to conduct morphological, physical, and chemical characterisations. The results from this study show that the production rate of TPWW decreases with the increase in polishing duration, and the coarse particles (diameters greater than 0.425 mm) account for most of the TPWW obtained. The ﬁne fraction (diameter smaller than 0.425 mm) of the TPWW comprises variously sized and irregularly shaped rubber particles from the tire, as well as uniformly sized and angular ﬁne aggregates. The environmental analysis results show that volatile alkanes (C9–C16) are the major organic contaminants in TPWW. The Open-Graded Friction Course (OGFC) asphalt mixture containing crumb rubber as a modiﬁer showed the highest risk of heavy metal pollution, and special concern must be given to tire materials for the purpose of improving the environmental conditions of road areas. The use of polyurethane as a binder material in the production of pavement mixtures has an environmental beneﬁt in terms of pollution from both organic contaminants and heavy metals.


Introduction
In the service life of pavement, due to the wearing interactions between vehicle tires and pavement surfaces, the wearing of vehicles' mechanical parts, the deposition of vehicles' gas emissions and atmospheric dust, and other reasons, pollutant particles accumulate on pavement surfaces. Among all of these potential sources, the wearing waste particles from the interaction of vehicle tires and pavement surfaces are the main contributors to the quantity of pollutants. TPWW is mechanically generated by the rolling shear of a tire against a pavement surface, and it comprises tire wear, road wear, and brake wear [1,2]. The pavement surface runoff carries a great number of these pollutants and flows through the surrounding soil and water bodies, which causes the accumulation, transmission, and migration of the pollutants in the road area. Unlike the point source pollution from industrial wastewater and urban sewage, which can be uniformly collected and centrally treated, road area pollution is a scattered, sparse plane network system, which greatly increases the difficulty of controlling the pollution.
The formation mechanism and environmental risks of TPWW are matters of general concern. Day et al. first introduced the concept of "urban street dust" to investigate the Three kinds of binders were used in this study, including virgin asphalt, crumb rubber modified asphalt, and polyurethane. The asphalt with a penetration grade of 60/80 (Pen 70), from Sinopec Zhenhai Refining and Chemical Co., Ltd. (Ningbo, China) was used as the virgin asphalt. The crumb rubber modified asphalt was produced from the same asphalt. The polyurethane with an isocyanates-polyhydric alcohol ratio of 800:100 was obtained from BASF SE (Shanghai, China) [22][23][24]. The limestone and andesite were used as filler and aggregate, respectively.
Two types of pavement mixtures, AC-13 and OGFC-13, were selected as the research objects. The AC-13 and OGFC-13 are two commonly used asphalt mixtures in the surface course of asphalt pavement. The AC-13 asphalt mixture contains dense-graded aggregates, and it is of the void content of 3~6%. The OGFC-13 asphalt mixture comprises open-graded aggregates, and the void content of OGFC-13 is generally greater than 18%. Due to the differences in the microstructure of AC-13 and OGFC-13, the technical requirements on the binder materials of the two mixtures are significantly different. The virgin bitumen with the penetration grade of 70 is the most used binder material in AC-13, whereas the modified bitumen binders with higher viscosity are generally used in OGFC-13 to improve the ravelling resistance. The polyurethane has been recently introduced into the construction of permeable pavement and was therefore only used to fabricate the OGFC-13 mixture. The aggregate grading curve of AC-13 and OGFC-13 are shown in Figure 1. The asphalt mixtures were made into plates with a length, width, and height of 300 mm, 300 mm, and 50 mm, respectively, the same as the rutting test samples. The fabricated test specimens are shown in Figure 2. Table 1 presents the types and compositions of all specimens.

Wearing Test
The wearing tests were conducted by the Harbin Advanced Polishing Machine (HAPM), as shown in Figure 3. The HAPM contains six units: an acceleration system, loading system, workbench system, dispenser system, protection system and operation control system. The acceleration system was used to set the loading rate and provide power for the tests. The radial tire (185/60 R 14, produced by Giti Company) was used for the HAPM test in the current study. The section width, aspect ratio and hub diameter of the tire are 185 mm, 60, and 14 inches, respectively. The wearing tests were conducted by the Harbin Advanced Polishing Machine (HAPM), as shown in Figure 3. The HAPM contains six units: an acceleration system, loading system, workbench system, dispenser system, protection system and operation control system. The acceleration system was used to set the loading rate and provide power for the tests. The radial tire (185/60 R 14, produced by Giti Company) was used for the HAPM test in the current study. The section width, aspect ratio and hub diameter of the tire are 185 mm, 60, and 14 inches, respectively. During the test, the specimen was fixed in the slot of the workbench, and the surface of the specimen was flush with the workbench. The silica sand was selected as the polishing medium, since previous research has demonstrated that the silica sand can best simulate the effect of pavement dust. A total of 50 g of silica sand was uniformly distributed on the specimen surface through the dispenser system. The water was continuously sprayed at the rate of 60 mL/min through the sprinkler in the dispenser system during the loading process to cool the wearing system and reduce the wearing noise. The contact pressure between the tire and sample surface was set as 1500 N, and the rotation rate of the tire was set as 100 rad/min. The major parameters for the wearing test are given in Table 2. The determination of the parameter values was based on our previous research in combination with a trial-and-error test [25][26][27]. Each wear test cycle was 30 min, and 5 cycles were carried out for every specimen. The wearing products were collected for subsequent analysis after every cycle test. During the test, the specimen was fixed in the slot of the workbench, and the surface of the specimen was flush with the workbench. The silica sand was selected as the polishing medium, since previous research has demonstrated that the silica sand can best simulate the effect of pavement dust. A total of 50 g of silica sand was uniformly distributed on the specimen surface through the dispenser system. The water was continuously sprayed at the rate of 60 mL/min through the sprinkler in the dispenser system during the loading process to cool the wearing system and reduce the wearing noise. The contact pressure between the tire and sample surface was set as 1500 N, and the rotation rate of the tire was set as 100 rad/min. The major parameters for the wearing test are given in Table 2. The determination of the parameter values was based on our previous research in combination with a trial-and-error test [25][26][27]. Each wear test cycle was 30 min, and 5 cycles were carried out for every specimen. The wearing products were collected for subsequent analysis after every cycle test.

Physical Analysis of TPWW
For purpose of analysing the physical characteristics of the TPWW, the vacuum freezedrying method was employed for the solid-liquid separation of the original wearing products. After that, the volumes of the TPWW matters were measured, and the particle size distributions of the TPWW were determined through the sieve test. The morphological properties of the fine TPWW particles (size smaller than 425 µm) were also characterized with the help of the Scanning Electron Microscope (SEM) test.

Heavy Metals Analysis
In a spectroscopic elemental analysis sample preparation, acid digestion is an important step of the entire analytical procedure. In this research, HNO 3 , HClO 4 and HCl were used for acid digestion. Samples were accurately weighed (0.1 g each) and placed in a 150 mL conical flask. Then, 2.5 mL of HNO 3 and 7.5 mL of HCl were mixed and poured into a conical flask slowly. Then, the mixture was boiled gently for about 30 min. After cooling, 3 mL of HClO 4 was added, and the mixture was gently boiled until dense white Sustainability 2021, 13, 8172 6 of 14 fumes appeared. Later, the mixture was allowed to cool, and 5 mL of deionized water was added, followed by further boiling until the fumes were totally released. Afterwards, 5 mL of 10% HCl was added, and the mixture was gently boiled for about 10 min. During the digestion procedures, the inner walls of the flasks were washed with 2 mL of 2% HNO 3 to prevent the loss of the samples, and at the last part of the digestion processes, the samples were filtered with filter paper. Then, a sufficient amount of HNO 3 was added to make the final volume up to 50 mL.
The content of heavy metal elements in raw pavement materials and wear products was determined by inductively coupled plasma atomic emission spectrometry (ICP-AES). The samples were aspirated in the atomizer by the nebulizer to measure the analyte concentration. The plasma flow rate, auxiliary gas flow rate, atomizer flow rate and injection flow rate were set as 15 L/min, 0.2 L/min, 0.8 L/min, 1.5 mL/min, respectively. The standard soil sample was used as the control sample. In this research, Cd, Pb, Cr, Cu, and Zn were target elements to be measured. The concentration (C) of the target element in the sample was calculated by Equation (1) as follows: where C is the concentration of the target element in the sample, mg/kg; C i is the measured concentration of the target element in the sample, µg/mL; V is the metered volume of the sample, mL; m is the mass of the sample, g.

Organic Pollutants Analysis
The gas chromatography-mass spectrometry (GC-MS) method was used to determine the composition of organic compounds in wear products. An ultrasonic extraction pretreatment was needed for the collection of the TPWW products before the GC-MS test. About 2~3 g of the sample was mixed with anhydrous sodium sulphate to remove all moisture. Then, 40 mL of n-hexane was added, and the mixture was treated by ultrasonic extraction for 30 min. Later, the sample was filtered with a filter paper with 0.45 µm particle retention. The filtrate was used for measuring the chromatogram and target ion abundance of the sample in the GC-MS test. A schematic flowchart of the experimental test in this study is shown in Figure 4.

Formulation of TPWW
The measured volumes of the TPWW after every 30 min of polishing duration are shown in Figure 5. It can be easily found that the amount of TPWW generally decreases as the polishing procedure continues. The exception is that of group I at the end of the 120

Formulation of TPWW
The measured volumes of the TPWW after every 30 min of polishing duration are shown in Figure 5. It can be easily found that the amount of TPWW generally decreases as the polishing procedure continues. The exception is that of group I at the end of the 120 min polishing duration. The slight increase in the TPWW at the end of the 120 min polishing duration was found to be resulted from the stripping of aggregates by observation. The results from Figure 5b indicate that the polishing resistance of the polyurethane (PU) mixtures is better than that of the bituminous mixtures, and the polishing resistance of the AC-13 asphalt mixture is better than the OGFC-13 asphalt mixture.

Particle size distribution of TPWW
The wearing products directly collected from the HAPM test contain a certain amount of particles whose sizes are greater than 1.7 mm. However, this fraction of TPWW was eliminated during the following analysis due to the following reasons: (i) this fraction

Particle Size Distribution of TPWW
The wearing products directly collected from the HAPM test contain a certain amount of particles whose sizes are greater than 1.7 mm. However, this fraction of TPWW was eliminated during the following analysis due to the following reasons: (i) this fraction of particle is the mineral aggregates directly stripped from the test sample surface, and they are very different from the wearing products in terms of formation mechanism; (ii) these particles are of higher volume and mass values than that of the finer particles, and they will hold a dominant position in both volume and mass of the total wearing products if they are taken into consideration. However, these particles hardly contribute to the organic contaminant and heavy metal elements of the road area environment. The remaining fraction (size smaller than 1.7 mm) constitutes both aggregate and rubber wearing particles. They were divided into five groups which were of the particle size range of 0~0.105 mm, 0.105~0.25 mm, 0.25~0.425 mm, 0.425~0.85 mm, and 0.85~1.7 mm, respectively. The identified particle size distributions of different groups of TPWW are shown in Figure 6. It is found from Figure 6 that the coarse particles (0.425~1.7 mm) dominate the TPWW regardless of the asphalt mixture type, which accounts for nearly 90% of the total collections. Among the three sample groups, group I is of the least content of coarse particles which can be attributed to the denser microstructure of the AC-13 asphalt mixture. wearing particles. They were divided into five groups which were of the particle size range of 0~0.105 mm, 0.105~0.25 mm, 0.25~0.425 mm, 0.425~0.85 mm, and 0.85~1.7 mm, respectively. The identified particle size distributions of different groups of TPWW are shown in Figure 6. It is found from Figure 6 that the coarse particles (0.425~1.7 mm) dominate the TPWW regardless of the asphalt mixture type, which accounts for nearly 90% of the total collections. Among the three sample groups, group I is of the least content of coarse particles which can be attributed to the denser microstructure of the AC-13 asphalt mixture.

Morphology of TPWW
The morphology of the fine particles (smaller than 0.425 mm) was characterized through SEM, and the results are shown in Figures 7 and 8. The selection of the fine particles was based on the fact that the TPWW of a diameter smaller than 500 µm can be well dissolved in water media, as has been demonstrated in previous research [28]. The mesoscale morphologies (see Figure 7) of the TPWW indicate that the finalized materials were composed of varying-sized and irregular-shaped rubber particles from the tire, as well as uniform-sized and angular fine aggregates. The microscale morphologies (found in Figure 8) of the fine aggregates and rubber particles exhibit significantly different features. The rubber particles were obviously porous, with a great number of fine fillers adsorbed on the surface. The aggregation phenomenon of the fine fillers can also be observed in certain regions. On the other hand, the aggregate particles had relatively smooth surfaces and profound angularity.

Morphology of TPWW
The morphology of the fine particles (smaller than 0.425 mm) was characterized through SEM, and the results are shown in Figures 7 and 8. The selection of the fine particles was based on the fact that the TPWW of a diameter smaller than 500 µm can be well dissolved in water media, as has been demonstrated in previous research [28]. The mesoscale morphologies (see Figure 7) of the TPWW indicate that the finalized materials were composed of varying-sized and irregular-shaped rubber particles from the tire, as well as uniform-sized and angular fine aggregates. The microscale morphologies (found in Figure 8) of the fine aggregates and rubber particles exhibit significantly different features. The rubber particles were obviously porous, with a great number of fine fillers adsorbed on the surface. The aggregation phenomenon of the fine fillers can also be observed in certain regions. On the other hand, the aggregate particles had relatively smooth surfaces and profound angularity. mesoscale morphologies (see Figure 7) of the TPWW indicate that the finalized mater were composed of varying-sized and irregular-shaped rubber particles from the tire well as uniform-sized and angular fine aggregates. The microscale morphologies (fou in Figure 8) of the fine aggregates and rubber particles exhibit significantly differ features. The rubber particles were obviously porous, with a great number of fine fill adsorbed on the surface. The aggregation phenomenon of the fine fillers can also observed in certain regions. On the other hand, the aggregate particles had relativ smooth surfaces and profound angularity.

Organic Contaminant Content of TPWW
The measured results of the gas chromatography test for the three groups of samp are presented in Figure 9, Figure 10, Figure 11, respectively. The mass content of ea component was determined by the gas chromatography area, and the calculation resu are shown in Figure 12. It is seen that the alkanes account for nearly 80% of the to TPWW matters, followed by the haloalkanes (8~10%). The benzene series with strong toxicity shows a relatively low content of no more than 5%. The contents of alkenes for t three groups are generally no more than 1%. The remaining matter includes some este ethers, alcohols, etc. The group II sample is found to contain more benzene ser substances compared to the group I sample. The increased benzene series substances a introduced by the use of crumb rubber modifiers. The substitution of polyurethane as t binding material leads to a decrease in the content of alkanes as well as an increase benzene series substances. The group III sample shows the highest content of remaini substances (7.54%). This is resulted from the increased esters and aldehydes (includi methyl stearate, petroselinic acid methyl ester, benzaldehyde, etc.) which are the b products of polyurethane.

Organic Contaminant Content of TPWW
The measured results of the gas chromatography test for the three groups of samples are presented in FIgures 9-11, respectively. The mass content of each component was determined by the gas chromatography area, and the calculation results are shown in Figure 12. It is seen that the alkanes account for nearly 80% of the total TPWW matters, followed by the haloalkanes (8~10%). The benzene series with stronger toxicity shows a relatively low content of no more than 5%. The contents of alkenes for the three groups are generally no more than 1%. The remaining matter includes some esters, ethers, alcohols, etc. The group II sample is found to contain more benzene series substances compared to the group I sample. The increased benzene series substances are introduced by the use of crumb rubber modifiers. The substitution of polyurethane as the binding material leads to a decrease in the content of alkanes as well as an increase in benzene series substances. The group III sample shows the highest content of remaining substances (7.54%). This is resulted from the increased esters and aldehydes (including methyl stearate, petroselinic acid methyl ester, benzaldehyde, etc.) which are the by-products of polyurethane.

Organic Contaminant Content of TPWW
The measured results of the gas chromatography test for the three groups of samples are presented in Figure 9, Figure 10, Figure 11, respectively. The mass content of each component was determined by the gas chromatography area, and the calculation results are shown in Figure 12. It is seen that the alkanes account for nearly 80% of the total TPWW matters, followed by the haloalkanes (8~10%). The benzene series with stronger toxicity shows a relatively low content of no more than 5%. The contents of alkenes for the three groups are generally no more than 1%. The remaining matter includes some esters, ethers, alcohols, etc. The group II sample is found to contain more benzene series substances compared to the group I sample. The increased benzene series substances are introduced by the use of crumb rubber modifiers. The substitution of polyurethane as the binding material leads to a decrease in the content of alkanes as well as an increase in benzene series substances. The group III sample shows the highest content of remaining substances (7.54%). This is resulted from the increased esters and aldehydes (including methyl stearate, petroselinic acid methyl ester, benzaldehyde, etc.) which are the byproducts of polyurethane.    The separated components in the gas chromatography analysis were further characterized by the mass spectrometry method. By comparing the obtained results to the standard mass spectra information in the NIST.17 database, the major compositions of the TPWW matters were identified as shown in Table 3. It shows that the alkanes are of ~75% of the total detectable substance types, followed by the benzene series with an approximate content of 15%. The haloalkanes, alkenes, and others (esters, ethers, alcohols, etc.) make up ~10% of the content. Among the alkanes, the volatile C9-C16 alkanes account for nearly half of the detected substances. The boiling point of alkane increases with the carbon number and carbon chain, and the alkanes with a carbon number of 9 generally have a boiling point of nearly 150 °C. This is reasonable considering that the mixing temperature of asphalt mixture is normally in the range of 140~150 °C. The alkanes whose carbon numbers are smaller than 9 can be readily volatilized during the mixing procedure, and therefore, they hardly exist in the TPWW. The benzene series substances mainly include methylbenzene, naphthalene, anthracene, phenol, and their derivatives.   The separated components in the gas chromatography analysis were furth characterized by the mass spectrometry method. By comparing the obtained results to th standard mass spectra information in the NIST.17 database, the major compositions of th TPWW matters were identified as shown in Table 3. It shows that the alkanes are of ~75 of the total detectable substance types, followed by the benzene series with a approximate content of 15%. The haloalkanes, alkenes, and others (esters, ethers, alcoho etc.) make up ~10% of the content. Among the alkanes, the volatile C9-C16 alkan account for nearly half of the detected substances. The boiling point of alkane increas with the carbon number and carbon chain, and the alkanes with a carbon number of generally have a boiling point of nearly 150 °C. This is reasonable considering that th mixing temperature of asphalt mixture is normally in the range of 140~150 °C. The alkan whose carbon numbers are smaller than 9 can be readily volatilized during the mixin procedure, and therefore, they hardly exist in the TPWW. The benzene series substanc mainly include methylbenzene, naphthalene, anthracene, phenol, and their derivatives  The separated components in the gas chromatography analysis were further characterized by the mass spectrometry method. By comparing the obtained results to the standard mass spectra information in the NIST.17 database, the major compositions of the TPWW matters were identified as shown in Table 3. It shows that the alkanes are of~75% of the total detectable substance types, followed by the benzene series with an approximate content of 15%. The haloalkanes, alkenes, and others (esters, ethers, alcohols, etc.) make up 10% of the content. Among the alkanes, the volatile C9-C16 alkanes account for nearly half of the detected substances. The boiling point of alkane increases with the carbon number and carbon chain, and the alkanes with a carbon number of 9 generally have a boiling point of nearly 150 • C. This is reasonable considering that the mixing temperature of asphalt mixture is normally in the range of 140~150 • C. The alkanes whose carbon numbers are smaller than 9 can be readily volatilized during the mixing procedure, and therefore, they hardly exist in the TPWW. The benzene series substances mainly include methylbenzene, naphthalene, anthracene, phenol, and their derivatives. To summarize, the organic contaminant characteristic of TPWW is closely related to the binder materials, the mixing conditions, the chemical composition of the tire, and the pavement surface properties. From the aspect of pavement materials, the virgin asphalt resulted TPWW shows low toxicity, and the addition of asphalt modifiers can cause increased toxicity. A higher mixing temperature can increase the loss of the volatile fractions in the asphalt binder, which in turn decreases the volatility of the TPWW. Comparing to the asphaltic mixtures, the use of polyurethane shows additional environmental benefits [29][30][31].

Heavy Metal Content of TPWW
The measured heavy metal contents of the virgin asphalt binder, modified asphalt binder, tire particles, and the collected TPWW materials (particle size smaller than 0.425 mm) are presented in Table 4. Results from Table 4 show that the virgin asphalt binder contains relatively low heavy metal among the four raw materials. The contents of Pb, Cr, and Zn elements in the tire particles were approximately 2, 4, and 40 times higher than that of the virgin asphalt binder, respectively. The higher contents of Cr and Zn heavy metals in the modified asphalt compared to the virgin asphalt are expected to be sourced from the crumb rubber modifier. It is also found from Table 4 that the collected TPWW from group II samples has the highest contents of all the five detected heavy metals. The higher contents of heavy metals for the group II TPWW matters can be resulted from both the use of crumb rubber modifier and the greater amount of accumulated TPWW. Even though that the asphalt pavement composed of the OGFC-13 mixture has the advantage of higher skid resistance and noise reduction, the associated environmental risks must be considered. In general, the group III samples have the lowest content of the five heavy metal elements, which implies a potential environmental benefit of using polyurethane for the replacement of asphalt binder materials.
The TPWW is the major component of the pavement surface dust and pavement water contaminants. For the purpose of comparison, the current study collected the reported data of the heavy metal element content of the pavement surface dust in several cities (see Table 5). Detailed information regarding the location as well as the heavy metal contents are given in Table 3. It is found that the contents of the Cd element in the TPWW matters are generally higher than that of the measured pavement surface dust, especially for the group II samples with modified asphalt binder. The contents of the Pb element in groups I, II, and III are lower than that of the reported data from different cities. However, the measured values of Cr and Zn content in pavement surface dust are found to be lower than that of the TPWW matters which indicating an enrichment effect of the Cr and Zn elements in the wearing process. On the other hand, the Cu contents in pavement surface dust are slightly higher than that of the lab-obtained samples.

Conclusions
This study contributes to a comprehensive understanding of the TPWW by conducting a variety of polishing tests and materials characterization. The major conclusions from this study include: The production rate of TPWW decreases with the increasing polishing duration, and the coarse fraction (diameter greater than 0.425 mm) accounts for nearly 90% of the total TPWW. Among the fine TPWW matters, the rubber particles are of a porous microstructure with fine fillers adsorbed on the surface, while the aggregate particles had a relatively smooth surface and profound angularity.
Alkanes are the main source of organic contaminants in the TPWW matters, with an average percentage of 80%, and half of the detectable alkanes are volatile C9-C16 substances. The toxic benzene series substances show a relatively low content which is no more than 5%. The tire is found to be closely related to the heavy metal pollution in the road area, and the potential environmental risks arising from the use of crumb rubber modifier and OGFC asphalt mixture must be carefully treated.
The findings from the current study provide insights into the formation mechanism and environmental risk of the tire-pavement TPWW matters. However, the mode of occurrences of the detected heavy metal elements remains unclear, and the bioavailability of heavy metals is not involved. Besides, pollutant transportation is not involved in the current study, and should be further identified in future research.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author. The data are not publicly available due to confidentiality promised to research participants.