Characteristics of Potentially Toxic Elements, Risk Assessments, and Isotopic Compositions (Cu-Zn-Pb) in the PM 10 Fraction of Road Dust in Busan, South Korea

: The pollution status of ten potentially toxic elements (PTEs), isotopic compositions (Cu, Zn, Pb), and the potential ecological risk posed by them were investigated in the PM 10 fraction of road dust in Busan Metropolitan city, South Korea. Enrichment factors revealed extremely to strongly polluted levels of Sb, Cd, Zn, Pb, and Cu in the PM 10 fraction of road dust, with Sb levels being the highest. Statistical analyses showed that the major cause for contamination with PTEs was non-exhaust trafﬁc emissions such as tire and brake wear. Cu and Zn isotopic compositions of road dust were related to trafﬁc-related emission sources such as brake and tires. Pb isotopic compositions were close to that of road paint, indicating that Pb was a different source from Cu and Zn in this study. No signiﬁcant health risk was posed by the PTEs. Taking into account the total length of road in Busan, a high quantity of PTEs in road dust (PM 10 ) can have serious deleterious effects on the atmospheric environment and ecosystems. The results of metal concentrations and isotopic compositions in road dust will help identify and manage atmospheric ﬁne particle and coastal metal contamination derived from ﬁne road dust.


Introduction
The acceleration of urbanization and the growth of population in cities has increased the magnitude of the negative environmental impact caused by road dust [1]. The total length of roads in Korea in 2019 was 110,714 km, which is an increase of 5401 km over the previous five years [2]. Vehicles are the main contributors of various contaminants that constitute road dust. The number of vehicles in Korea increased from 20.1 million in 2014 to 23.7 million in 2019. Therefore, environmental pollution caused by road dust in urban environments has attracted considerable attention.
Potentially toxic elements (PTEs) are pollutants that can accumulate in the environment because of traffic as well as industrial and mining activities [3][4][5]. The presence of PTEs in road dust, soils, and atmospheric particulate matter have been reported in many countries [6][7][8][9][10][11][12][13]. The concentrations of PTEs in road dust are used as indicators of environmental pollution and for the identification of pollution sources such as traffic and industrial activities.
The fine particles in road dust have a much higher concentration of PTEs compared to coarse particles [14][15][16]. The fine particles (<10 µm) in road dust are easily resuspended into the atmosphere by wind and vehicle movement, and thus, they can be transported over a long distance [16][17][18]. PTEs in fine road dust can pose a greater health risk to the human body via ingestion, inhalation, and dermal contact pathways [19][20][21][22][23][24][25][26].
Fine road dust is an important source of PTE contamination in urban atmosphere and soils. Fugitive road dust particles caused by vehicle movement on paved roads are emitted from the road to the surrounding atmosphere because of turbulence and contribute significantly to the concentration of PM 10 in the atmospheric environment [27][28][29]. The contribution of fugitive dust to total particulate matter is gaining importance because of the increase in the number of vehicles and the detrimental health impacts of high concentrations of heavy metals contained in them [30][31][32]. Several studies have reported that fugitive road dust accounts for 14% to 55% of PM 10 concentrations in India [33], Brazil [34], and European urban areas [35]. In Korea, PM 10 contributes more than 60% of the fine dust that is re-suspended from the paved road [36].
The purpose of this study is to (1) characterize the PTEs in fine particulate matter (PM 10 ) of road dust in Busan Metropolitan in South Korea; (2) assesses the potential ecological risk and non-carcinogenic health risk posed by them; (3) present Cu, Zn, and Pb isotope compositions as a preliminary study for identifying potential pollution sources.

Study Area
The Busan Metropolitan City is the largest port city in South Korea. Busan is located in the southeastern part of South Korea and has a land area of 506 km 2 and an impermeable area of 234.31 km 2 . Busan has a population of 3.4 million, and the population density of the impermeable area is 14,570 persons/km 2 . The average temperature, precipitation, humidity, and wind speed over the past 30 years are 14.7 • C, 1519 mm, 64.7%, and 13.3 km/h, respectively. The road dust in Busan is contaminated with Sb, Cu, Cd, Zn, and Pb caused by transportation activities, and these toxic elements present on the road surfaces are transported to coastal areas through runoff during precipitation [48].

Sampling and Measurement of PTEs
Road dust samples were collected from 25 sites in Busan during May 2014. The sampling locations and the study area are shown in Figure 1. Road dust was obtained using a cordless vacuum cleaner (DC35, Dyson Limited, Malmesbury, UK) covering an area of 0.25 m 2 (0.5 m × 0.5 m) from the curb, and each composite sample comprised of 4-5 subsamples to ensure representativeness. The vacuum cleaner was disassembled and cleaned prior to sampling at each site to prevent cross contamination. The samples were stored in a zipper bag, dried in an oven at 40 • C, and weighed. Road dust containing particles of less than 10 µm (PM 10 ) size was separated using a nylon sieve and a vibratory sieve shaker (Analysette 3 PRO, Fritsch GmbH, Germany) in the laboratory.
The analysis of PTEs in the PM 10 fraction of road dust was performed through total digestion using a high-purity (Suprapur ® grade) acid mixture containing HNO 3 , HF, and HClO 4 . About 0.1 g of the sample was decomposed in a Teflon digestion vessel using the acid mixture on a hot plate at 180 • C, evaporated, and then redissolved using 2% HNO3. Al, Cr, Co, Ni, Cu, Zn, As, Cd, Sb, and Pb were measured using inductively coupled plasma mass spectrometry (ICP-MS, iCAP-Q, Thermo Fisher Scientific, Braunschweig, Germany). Hg was measured using a Hg analyzer (Hydra-C, Teledyne Leeman Labs, Hudson, NH, USA). Duplicate measurements were performed to ensure the reliability of the analysis. Data accuracy was ensured using certified reference materials MESS-4 and PACS-3, and the recoveries ranged from 95.7% to 102.4%.
Data accuracy was ensured using certified reference materials MESS-4 and PACS-3, and the recoveries ranged from 95.7% to 102.4%.

Pollution and Ecological Risk Assessments
The enrichment factor (EF) was used to assess the contamination levels of individual metals using the following equation [49]: The EF values were classified into five classes [50], and the classification is provided in Table S1.
Single factor ecological risk degree (E ) and potential ecological risk (PER) were used to assess the ecological risk of the ten PTEs on their toxicity response using the following equations [51]: where Ci and Bi are the measured and geological background concentrations by Rudnick and Gao [52], respectively, T is the toxic response factor (Hg = 40, Cd = 30, As = 10, Sb = 7, Co = Cu = Ni = Pb = 5, Cr = 2, and Zn = 1) [51,53,54]. E and PER values were classified into five and four classes, respectively (Table S1).

Health Risk Assessment
Health risk assessment was carried out considering the exposure of adults and children, through the pathways of ingestion, inhalation, and dermal contact, to the ten PTEs present in the PM10 fraction of road dust by calculating the non-carcinogenic hazards based on the guidelines of [55,56].
The average daily dose (ADD) for the exposure assessment of PTEs via ingestion (ADDing), inhalation (ADDinh), and dermal contact (ADDderm) were calculated using the follow equations:

Pollution and Ecological Risk Assessments
The enrichment factor (EF) was used to assess the contamination levels of individual metals using the following equation [49]: The EF values were classified into five classes [50], and the classification is provided in Table S1.
Single factor ecological risk degree (E i r ) and potential ecological risk (PER) were used to assess the ecological risk of the ten PTEs on their toxicity response using the following equations [51]: where C i and B i are the measured and geological background concentrations by Rudnick and Gao [52], respectively, T i r is the toxic response factor (Hg = 40, Cd = 30, As = 10, Sb = 7, Co = Cu = Ni = Pb = 5, Cr = 2, and Zn = 1) [51,53,54]. E i r and PER values were classified into five and four classes, respectively (Table S1).

Health Risk Assessment
Health risk assessment was carried out considering the exposure of adults and children, through the pathways of ingestion, inhalation, and dermal contact, to the ten PTEs present in the PM 10 fraction of road dust by calculating the non-carcinogenic hazards based on the guidelines of [55,56].
The average daily dose (ADD) for the exposure assessment of PTEs via ingestion (ADD ing ), inhalation (ADD inh ), and dermal contact (ADD derm ) were calculated using the follow equations: where C i represents the concentration of PTEs (mg/kg) in the PM 10 fraction of road dust. All the other parameters present in these equations and their corresponding values for adults and children are provided in Table S2 [57]. The hazard quotient (HQ) was calculated using the rate of the corresponding reference dose (RfD) for each exposure pathway i to determine the non-carcinogenic risks posed by the PTEs in road dust. The RfD values reported by Ferreira-Baptista and De Miguel [58] and Ma et al." [59] were used in this study.
If HQ > 1, a potential adverse health effect may occur, whereas if HQ < 1, an adverse health effect will not occur.
The hazard index (HI) for each PTE was also calculated using the following equation: If the HI value exceeds the threshold value of 1, non-carcinogenic effects are likely to occur.

Measurement of Cu, Zn, and Pb Isotopes
Isotopic compositions of Cu, Zn, and Pb isotopes were measured using multi-collector ICP-MS (MC-ICP-MS) at the Korea Institute of Ocean Science and Technology (KIOST). Analytical methods and instrumental settings followed [48]. The Cu and Zn isotopic data for each sample represent the average of 3-4 standard sample brackets. Samples of 100-200 ppb were introduced into the Teflon nebulizer through a 100 µL min −1 . Cu and Zn isotopic compositions are expressed in delta notation from isotopic standards (ERM-AE647, IRMM-3702). The concentrations of PTEs in the PM 10 fraction of road dust are summarized in Table 1. The spatial distribution of these elements is illustrated in Figure S1. The average concentrations of PTEs in the PM 10 fraction of the road dust are as follows: Zn (3007 mg/kg), Cu (513 mg/kg), Pb (480 mg/kg), Cr (300 mg/kg), Ni (75.6 mg/kg), Sb (61.6 mg/kg), Co (19.8 mg/kg), As (17.0 mg/kg), Cd (5.6 mg/kg), and Hg (0.6 mg/kg) (Figure 2a).
High concentrations of Cu, Zn, Cd, Pb, and Hg were observed in sampling sites R11-R16 that experience heavy traffic because of the presence of large departmental stores, city halls, and fish markets. Many studies have reported that pollution by these elements is mainly caused by traffic activities [60][61][62][63]. Previous studies have reported that heavy traffic intensity contributes to high levels of heavy metals in the fine size fraction of road dust [64][65][66][67]. In this study, the concentrations of PTEs were relatively low in sampling sites that experience low traffic volume and areas with smaller road widths.

Pollution Assessment of PTEs
The average EF value of Sb was the highest among the PTEs (average 183 mg/kg, minimum: 64.8 mg/kg, and maximum: 292 mg/kg). The average of EF values occurs in the following order: Sb > Cd > Zn > Pb > Cu > As > Cr > Ni > Hg > Co (Figure 2b). The Sb and Cd in the PM10 fraction of the road dust had EF values exceeding 32, indicating the extremely polluted level at all the sampling sites. The average EF value for Zn (53.4) indicates an extremely polluted level. The average EF of Cu was 21.7, which corresponds to the strongly polluted level with three sampling sites corresponding to the extremely polluted level. The reasons for the high levels of contamination with Sb, Cu, Zn, and Cd are wearing of brakes and tires [70,76,77].
Substantial quantities of Sb compounds are used in the production of polyethylene terephthalate as catalysts and brake pads as lubricants [78,79]. Of the 54 μg of Sb emitted per braking event per car, 32 μg is apportioned in the PM10 fraction and 22 μg is apportioned in the PM2.5 fraction [76,80,81]. Bukowiecki et al. [82] reported that the Sb emission factors of Sb were 11 μg km −1 vehicle −1 for light-duty vehicles and 86 μg km −1 vehicle −1 for heavy-duty vehicles. Garg et al. [83] concluded that about 30% of brake wear was released into the airborne particulate matter and the emission rate ranged from 3.3 to 8.8 mg/km. Extreme enrichment of Sb in road dust, stream sediments, and particulate matter in runoff has been observed in Korea [48,84,85]. The average values of EF for As show that the PM10   9 Logiewa et al., 2020 [16] (road dust at Krakow); 10 Vlasov et al., 2021 [21] (road dust).
The highest concentrations of Cr, Ni, and Co were observed in sampling sites R21 and R23 that are located in an industrial area containing metal-and steel-related facilities (processing and cutting). Cr and Ni are the most common metals used for alloying because of their high hardness and corrosion resistance, and they are widely used in steel production. Daigo et al. [71] reported that 85% and 61% of the global production of Cr and Ni, respectively, are used in stainless steel production. Stainless steel contains more than 10% of Cr and Ni. Various alloying metals are extensively used in food, beverage, and medical industries and in the manufacturing of equipment, machinery, and vehicle. Co is released from stainless steel and Ni-based alloy powders and Cr-alloy powders [72,73]. The highest concentrations of Sb and Pb were found in R22, which has a tire replacement shop. Particles released from the tire and brake pad replacement process spread to the surrounding, resulting in a high concentration of these elements in road dust.
The average concentrations of Co, Ni, Cu, Zn, Cd, Sb, Pb, and Hg in the PM 10 fraction of road dust were 1.1 (Ni)-4.1 (Hg) times higher than those of the bulk road dust samples [48]. The concentrations of PTEs in this study were lower than that of the soil pollution standard for roads and factory sites in Korea [74]; however, they were higher than the fine particulate fraction of road dust in other cities [14,16,21,57,61,67,69,70,75].

Pollution Assessment of PTEs
The average EF value of Sb was the highest among the PTEs (average 183 mg/kg, minimum: 64.8 mg/kg, and maximum: 292 mg/kg). The average of EF values occurs in the following order: Sb > Cd > Zn > Pb > Cu > As > Cr > Ni > Hg > Co (Figure 2b). The Sb and Cd in the PM 10 fraction of the road dust had EF values exceeding 32, indicating the extremely polluted level at all the sampling sites. The average EF value for Zn (53.4) indicates an extremely polluted level. The average EF of Cu was 21.7, which corresponds to the strongly polluted level with three sampling sites corresponding to the extremely polluted level. The reasons for the high levels of contamination with Sb, Cu, Zn, and Cd are wearing of brakes and tires [70,76,77].
Substantial quantities of Sb compounds are used in the production of polyethylene terephthalate as catalysts and brake pads as lubricants [78,79]. Of the 54 µg of Sb emitted per braking event per car, 32 µg is apportioned in the PM 10 fraction and 22 µg is apportioned in the PM 2.5 fraction [76,80,81]. Bukowiecki et al. [82] reported that the Sb emission factors of Sb were 11 µg km −1 vehicle −1 for light-duty vehicles and 86 µg km −1 vehicle −1 for heavy-duty vehicles. Garg et al. [83] concluded that about 30% of brake wear was released into the airborne particulate matter and the emission rate ranged from 3.3 to 8.8 mg/km. Extreme enrichment of Sb in road dust, stream sediments, and particulate matter in runoff has been observed in Korea [48,84,85]. The average values of EF for As show that the PM 10 fraction of road dust was moderately polluted, and Cr, Ni, Hg, and Co were slightly polluted to unpolluted levels.

Ecological Risk Assessments
A comparison of the single factor ecological risk degree (E i r ) for the measured PTEs is presented in Figure 2c. The average E i r values of Cd (1872), Sb (1078), and Hg (489) exceeded 320, which indicates that these elements in the PM 10 fraction of road dust correspond to the extreme ecological risk level because of their high concentrations. The average E i r values of the PTEs in the descending order were as follows: Cd > Sb > Hg > Pb > Cu > Zn > As > Ni > Cr > Co. The average E i r values of Pb and Cu indicate considerable potential risk. The average E i r value of Zn was 44.9, indicating moderate potential risk. The average E i r value of As, Ni, Cr, and Co were less than 40, indicating low potential risk. The average PER value (3772), which represents the comprehensive ecological risk of the ten PTEs, significantly exceeded 600, indicating high ecological risk for all the sampling sites in the PM 10 fraction of road dust.

Health Risk Assessments
The values of the HQ and HI are provided in Table 2. For adults, the average values of HQ ing , HQ inh , and HQ derm were <1, indicating no non-carcinogenic risks posed by PTEs in the PM 10 fraction of road dust (Figure 3). The HQ ing values of Cr, Cd, and Sb were higher than those of HQ inh and HQ derm . The average HI values for adults in decreasing order were as follows: Sb > Cr > Pb > As > Cd > Cu > Zn > Ni > Co > Hg (Figure 3a). The HI for adults was <1, indicating no non-carcinogenic health risk by PTEs in the PM 10 fraction of road dust. decreased in the following order: ingestion > dermal contact > inhalation in the PM10 fraction of road dust ( Table 2). This suggests that ingestion is the major exposure pathway for most toxic elements except for Cr, Cd, and Sb for adults and children. Dermal contact was the main exposure pathway for Cr, Cd, and Sb. The HI value for children was higher than that for adults. The quantity of the PM10 fraction in road dust per unit area of road surface ranged from 0.01 to 6.9 g/m 2 with an average of 2.4 g/m 2 . The average quantities of PTEs in the PM10 fraction of road dust were 0.71 mg/m 2 for Cr, 0.05 mg/m 2 for Co, 0.19 mg/m 2 for Ni, 1.26 mg/m 2 for Cu, 7.75 mg/m 2 for Zn, 0.04 mg/m 2 for As, 0.01 mg/m 2 for Cd, 0.15 mg/m 2 for Sb, 1.29 mg/m 2 for Pb, and 0.001 mg/m 2 for Hg, respectively. Considering the total length of paved roads in Busan at 3313 km, significant quantities of PTEs have accumulated on the road surfaces in its urban area.

Statistical Analysis
To understand the sources of the PTEs, principal component analysis (PCA) and Pearson's correlation analysis were conducted using SPSS (version 18.0). The Kaiser-Meyer-Olkin (KMO) test value was 0.625, and the Bartlett test result was p < 0.001, indicating that the data set was suitable for PCA analysis. Two principal components explaining 63.0% of the total variance were extracted (Figure 4). PC1 explained 40.8% of the total variance, and it had a positive loading on Cu, Zn, Cd, Sb, and Pb with a strong positive correlation between these elements ( Table 3). The PCA result and the significant degree of pollution of these elements in road dust are attributed to traffic sources. PC2 explained 22.2% of the total variance, and it had a positive loading on Cr, Co, and Ni with positive correlations between them. The pollution levels of these elements were not significant, The average value of HQ for children in terms of non-carcinogenic risk through the three pathways was less than 1. The HI values for children were higher than those for adults. However, HQ ing values of the sampling sites in industrial areas for Cr and Pb were >1, indicating that potential adverse health effects can occur. HQ inh and HQ derm values of all the sampling sites and PTEs were <1. The HI values for children were in the decreasing order as follows: Sb > Pb > Cr > As > Cu > Zn > Cd > Ni > Co > Hg (Figure 3b). The HI values for children in industrial areas (R21, R23, and R25) for Cr and near the tire replacement facility (R22) for Sb and Pb exceeded 1, indicating a potential non-carcinogenic effect of the PM 10 fraction of road dust. The total HQ values of PTEs for both adults and children decreased in the following order: ingestion > dermal contact > inhalation in the PM 10 fraction of road dust (Table 2). This suggests that ingestion is the major exposure pathway for most toxic elements except for Cr, Cd, and Sb for adults and children. Dermal contact was the main exposure pathway for Cr, Cd, and Sb. The HI value for children was higher than that for adults.
The quantity of the PM 10 fraction in road dust per unit area of road surface ranged from 0.01 to 6.9 g/m 2 with an average of 2.4 g/m 2 . The average quantities of PTEs in the PM 10 fraction of road dust were 0.71 mg/m 2 for Cr, 0.05 mg/m 2 for Co, 0.19 mg/m 2 for Ni, 1.26 mg/m 2 for Cu, 7.75 mg/m 2 for Zn, 0.04 mg/m 2 for As, 0.01 mg/m 2 for Cd, 0.15 mg/m 2 for Sb, 1.29 mg/m 2 for Pb, and 0.001 mg/m 2 for Hg, respectively. Considering the total length of paved roads in Busan at 3313 km, significant quantities of PTEs have accumulated on the road surfaces in its urban area.

Statistical Analysis
To understand the sources of the PTEs, principal component analysis (PCA) and Pearson's correlation analysis were conducted using SPSS (version 18.0). The Kaiser-Meyer-Olkin (KMO) test value was 0.625, and the Bartlett test result was p < 0.001, indicating that the data set was suitable for PCA analysis. Two principal components explaining 63.0% of the total variance were extracted (Figure 4). PC1 explained 40.8% of the total variance, and it had a positive loading on Cu, Zn, Cd, Sb, and Pb with a strong positive correlation between these elements ( Table 3). The PCA result and the significant degree of pollution of these elements in road dust are attributed to traffic sources. PC2 explained 22.2% of the total variance, and it had a positive loading on Cr, Co, and Ni with positive correlations between them. The pollution levels of these elements were not significant, and these are attributed to natural origins except for sampling sites in industrial areas.

Elemental Ratios
Some researchers have found that brake and tire wear show a specific pattern of some heavy metals such as Fe, Ti, Cu, Zn, Zr, Mo, Sn, Sb, Cd, and Pb [86][87][88][89]. In many studies, elemental concentrations and ratios between elements such as Zn/Cu and Cu/Sb were used as tracers to evaluate the relative contributions and contamination sources of PTEs in road dust [90][91][92]. Since brake pads and linings contain Sb2O3 and Sb2S3, a high concentration of Sb is present in road dust and atmospheric particulate matter [93]. The EF values show that Sb contamination was more than that of other elements. PM10 emission factors from brake and tire wear were reported at 2.9-7.5 mg and 3.0-13.0 mg km −1 vehicle −1 , respectively [83,[94][95][96][97][98]. The total number of vehicles in Busan is 1.4 million, and the average daily driving distance is 39.6 km vehicle −1 . Therefore, a substantial quantity of PTEs generated by these traffic activities would have accumulated on the road surface.
Sanders et al. [99] reported that 50-70% of the total wear lining material was released in the form of airborne particles. Grigoratos and Martini [100] reported that brake wear can contribute up to 55% of the mass of the total non-exhaust traffic-related PM10 emissions and up to 21% of the mass of the total transportation-related PM10 emissions in urban environments. Tires lose approximately 1.0-1.5 kg of weight over their lifetime, of which the PM10 percentage is less than 10% [101,102]. In this study, the average Zn/Cu and Cu/Sb ratios were 6.1 ± 1.3 and 8.9 ± 3.4, respectively. This indicates that the dominant pollution source of the PM10 fraction of the road dust was non-exhaust traffic activities such as wear

Elemental Ratios
Some researchers have found that brake and tire wear show a specific pattern of some heavy metals such as Fe, Ti, Cu, Zn, Zr, Mo, Sn, Sb, Cd, and Pb [86][87][88][89]. In many studies, elemental concentrations and ratios between elements such as Zn/Cu and Cu/Sb were used as tracers to evaluate the relative contributions and contamination sources of PTEs in road dust [90][91][92]. Since brake pads and linings contain Sb 2 O 3 and Sb 2 S 3 , a high concentration of Sb is present in road dust and atmospheric particulate matter [93]. The EF values show that Sb contamination was more than that of other elements. PM 10 emission factors from brake and tire wear were reported at 2.9-7.5 mg and 3.0-13.0 mg km −1 vehicle −1 , respectively [83,[94][95][96][97][98]. The total number of vehicles in Busan is 1.4 million, and the average daily driving distance is 39.6 km vehicle −1 . Therefore, a substantial quantity of PTEs generated by these traffic activities would have accumulated on the road surface.
Sanders et al. [99] reported that 50-70% of the total wear lining material was released in the form of airborne particles. Grigoratos and Martini [100] reported that brake wear can contribute up to 55% of the mass of the total non-exhaust traffic-related PM 10 emissions and up to 21% of the mass of the total transportation-related PM 10 emissions in urban environments. Tires lose approximately 1.0-1.5 kg of weight over their lifetime, of which the PM 10 percentage is less than 10% [101,102]. In this study, the average Zn/Cu and Cu/Sb ratios were 6.1 ± 1.3 and 8.9 ± 3.4, respectively. This indicates that the dominant pollution source of the PM 10 fraction of the road dust was non-exhaust traffic activities such as wear of brakes and tires.

Cu, Zn, and Pb Isotopic Compositions
Cu isotopic compositions (δ 65 Cu AE647 ) of road dust (<10 µm) in this study ranged from −0.04 to +0.13‰ (mean: +0.05‰ ± 0.05, sd, n = 25). The δ 65 Cu AE647 values of road dust (<10 µm) were closer to those of the brake pads from Korea (unpublished data) than brake pads from the UK [103] (Figure 5a). It also indicates that the isotopic compositions of the brake pads can differ depending on the manufacturers. Zn isotopic compositions (δ 66 Zn IRMM3702 ) of road dust (<10 µm) in this study ranged from −0.17 to −0.03 (mean: −0.11‰ ± 0.03, sd, n = 25). Road dust (<10 µm) was distinguished from road paint (UK; [103]) and brake pads (UK; [103]) in the plot of concentrations and isotopic compositions of Zn. However, road dust (<10 µm) was similar to brake pads (Korea; unpublished data) and tires (UK; [103]) in Figure 5b, suggesting that the Zn sources in road dust were closely related to traffic-related sources. Previous studies reported that brake pads and tires had significantly high Cu and Zn concentrations, respectively, and these were major sources of road dust from traffic-related emissions [88,95]. Our results also showed that Cu and Zn sources of road dust (<10 µm) were associated with traffic-related emissions. The Pb isotopic composition ( 206 Pb/ 207 Pb) of road dust (<10 µm) in this study ranged from 1.1432 to 1.1606 (mean: 1.1514 ± 0.0036, sd, n = 25). Road dust (<10 µm) overlapped with road paint (UK; [103]) and was distinguished from brake pads (UK; [103], Korea; unpublished data) and tires (UK) in Figure 5c. Previous study reported that the mean of 206 Pb/ 207 Pb values in lead-based paint ranged from 1.083 to 1.183 (mean: 1.117 ± 0.030, sd, n = 15; [104]). Anthropogenic Pb sources in urban environments derive from various sources such as lead-based paint, traffic-related exhaust, and industrial activities [105]. Additionally, lead-based paints are common sources of lead-contaminated soil and road dust [106][107][108]. Pb sources of road dust (<10 µm) in this study appear to be different sources from Zn and Cu.

Conclusions
Recently, PTEs in the fine particle fraction of road dust have been attracting significant attention because they can easily resuspend and affect the atmosphere and human health. Ten PTEs in the PM10 fraction of road dust from the port city of Busan were investigated to evaluate the pollution status, identify isotopic compositions (Cu, Zn, Pb), and assess the potential ecological risks posed by them. The average concentrations of the studied PTEs (mg/kg) in decreasing order are as follows: Zn (3007) > Cu (513) > Pb (480) > Cr (300) > Ni (75.6) > Sb (61.6) > Co (19.8) > As (17.0) > Cd (5.6) > Hg (0.6). The enrichment factors showed extremely polluted level in terms of Sb, Cd, Zn, and Pb, and strongly polluted levels for Cu. Based on average EF value, the PTEs can be ranked as follows: Sb > Cd > Zn > Pb > Cu > As > Cr > Ni > Hg > Co. Ecological risk assessment indicated that the The average δ 65 Cu AE647 and δ 66 Zn IRMM3702 values of this study are slightly higher than those of heavily industrialized areas [109]. On the other hand, the average 206 Pb/ 207 Pb ratio of this study had a lower value than industrial areas [109]. In the plots of concentration and isotopic composition for Cu and Pb, the urban area (this study) can be clearly distinguished from the industrial area ( Figure S2) [109]. This industrial area provided a clear signal, as it was heavily contaminated with Cu and Pb compared to Zn. Additionally, it was difficult to distinguish between urban and industrial areas in the plot of concentration and isotopic values for Zn ( Figure S2).
Recent studies reported the contribution of pollution sources using various modeling programs (Iso-source, SIAR, PMF) in atmospheric fine dust [110][111][112]. Endmembers of various pollution sources will enable more accurate source tracking, and evaluation of the contribution rate for specific pollutants using these models can help improve source identification. However, metallic isotope fingerprints of various pollution sources are still insufficient in South Korea. Therefore, it is important to collect endmembers of various pollutants to trace these sources precisely by applying these models in further study. Especially, isotopic fingerprints of products from each country are also needed, since traffic-related sources may differ depending on the manufacturing countries.

Conclusions
Recently, PTEs in the fine particle fraction of road dust have been attracting significant attention because they can easily resuspend and affect the atmosphere and human health. Ten PTEs in the PM 10 fraction of road dust from the port city of Busan were investigated to evaluate the pollution status, identify isotopic compositions (Cu, Zn, Pb), and assess the potential ecological risks posed by them. The average concentrations of the studied PTEs (mg/kg) in decreasing order are as follows: Zn (3007) > Cu (513) > Pb (480) > Cr (300) > Ni (75.6) > Sb (61.6) > Co (19.8) > As (17.0) > Cd (5.6) > Hg (0.6). The enrichment factors showed extremely polluted level in terms of Sb, Cd, Zn, and Pb, and strongly polluted levels for Cu. Based on average EF value, the PTEs can be ranked as follows: Sb > Cd > Zn > Pb > Cu > As > Cr > Ni > Hg > Co. Ecological risk assessment indicated that the PM 10 fraction of road dust poses extreme risk of Cd, Sb, and Hg contamination, considerable risk of Pb and Cu, and moderate risk of Cu. Considering the high total length of road in the study area and the quantity of PM 10 contained in the road dust, some PTEs (Sb, Cd, Zn, Pb, and Cu) can seriously affect the atmospheric environment and human health. The degree of pollution and statistical analyses such as Pearson's correlation and PCA analysis showed that the PTEs in the PM 10 fraction of road dust were mainly due to the abrasion of tires and brake pads. The result of Cu isotopic compositions (δ 65 Cu AE647 ) indicated that Cu source of road dust (<10 µm) was more related to brake pads from Korea than that of the UK. The δ 66 Zn IRMM3702 had similar values to brake pads (Korea) and tires (UK), indicating Zn source associated with traffic-related emission sources. Pb isotopic ratios ( 206 Pb/ 207 Pb) of road dust (<10 µm) were close to road paint (UK), suggesting that the dominant Pb source in this study was different from Cu and Zn. The approach using multi-isotope has been successfully carried out, and it is meaningful to compare the isotopic composition of road dust (<10 µm) with potential pollution sources in this study. Long-term monitoring is required to efficiently reduce/manage fine road dust polluted with high concentrations of PTEs and to understand the impact of PTEs in road dust on the environments and the relationship between road dust and atmospheric particles.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/atmos12091229/s1, Figure S1: Spatial distribution of potentially toxic elements in PM 10 of road dust in this study, Figure S2: Plots of concentrations and isotopic values of Cu, Zn, and Pb (a: this study, b: Jeong et al., 2021), Table S1: Classification of enrichment factor (EF) and potential ecological risk factor (E i r ) for each element and potential ecological risk index (PER), Table S2: Key parameters for the health risk assessment through three pathways used in this study (  Data Availability Statement: All data for this study are available from the corresponding author on request.