- Email: [email protected]
Passive air sampling of halogenated polycyclic aromatic hydrocarbons in the largest industrial city in Korea: Spatial distributions and source identification
Journal of Hazardous Materials 382 (2020) 121238
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Passive air sampling of halogenated polycyclic aromatic hydrocarbons in the largest industrial city in Korea: Spatial distributions and source identification
T
Quang Tran Vuonga, Seong-Joon Kima, Tuyet Nam Thi Nguyena, Phan Quang Thanga, ⁎ Sang-Jin Leea, Takeshi Ohurab, Sung-Deuk Choia, a b
School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, 44919, Republic of Korea Faculty of Agriculture, Meijo University, Nagoya, 468-8502, Japan
G R A P H I C A L A B S T R A C T
A R T I C LE I N FO
A B S T R A C T
Editor: D. Aga
Some halogenated polycyclic aromatic hydrocarbons (Halo-PAHs) are known to be more toxic than their corresponding parent PAHs, but studies on Halo-PAHs have been somewhat limited. In this study, passive air samplers were used to monitor Halo-PAH and PAH contamination at 20 sampling sites in Ulsan, one of the largest industrial cities in South Korea. The mean concentrations of Σ24 ClPAHs, Σ11 BrPAHs, and Σ13 PAHs were 207 pg/m3, 84 pg/m3, and 26 ng/m3, respectively. Industrial areas displayed higher concentrations of both HaloPAHs and PAHs than urban and rural areas. Strong correlations between energetically unfavorable Halo-PAHs and their corresponding parent PAHs suggest that the main formation mechanism of Halo-PAHs is not direct halogenation of PAHs. Low molecular weight Halo-PAHs with one halogen atom and their parent PAHs were dominant. The profiles of ClPAHs and BrPAHs in petrochemical, automobile, shipbuilding, and non-ferrous industrial complexes were distinguished. The toxicity equivalency quantities (TEQs) of ClPAHs, BrPAHs, and PAHs at the industrial sites also showed the highest values of 4.2, 0.5, and 18.3 pg-TEQ/m3, respectively, reflecting the high toxicity of Halo-PAHs. To the best of our knowledge, this is the first study reporting atmospheric levels of both ClPAHs and BrPAHs using passive air samplers.
Keywords: PUF-PAS Halo-PAHs ClPAHs BrPAHs Ulsan
⁎
Corresponding author. E-mail address: [email protected] (S.-D. Choi).
https://doi.org/10.1016/j.jhazmat.2019.121238 Received 23 February 2019; Received in revised form 25 August 2019; Accepted 14 September 2019 Available online 17 September 2019 0304-3894/ © 2019 Elsevier B.V. All rights reserved.
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
1. Introduction
South Korea, with a population of more than 1.1 million inhabitants. The Ulsan and Mipo national industrial complex (46 km2) comprises automobile, shipbuilding, petrochemical, and petroleum refining industries, and the Onsan national industrial complex (20 km2) includes petroleum refining and non-ferrous metal industries. Industrial activities from the two national industrial complexes have a significant influence on PAH contamination in Ulsan (Choi et al., 2012; Vu et al., 2010; Lee and Lee, 2004; Dong and Lee, 2009). Halo-PAHs are also expected to be produced from the industrial activities in Ulsan, though no extensive studies on halo-PAHs have been performed in Ulsan yet. In our previous study (Choi et al., 2012), PUF-PASs were deployed in Ulsan, and the spatial distribution and source-receptor relationships of PAHs were investigated. Based on the results of those research efforts, we decided to further monitor Halo-PAHs as well as their corresponding parent PAHs using PUF-PASs. Their levels, profiles, spatial distributions, and potential sources were investigated.
Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous environmental organic pollutants, which have carcinogenic and/or mutagenic properties (Abdel-Shafy and Mansour, 2016; Armstrong and Gibbs, 2009). They are mainly discharged into the atmosphere from biomass burning, vehicular exhausts, and industrial activities (Choi et al., 2012, 2007; Tang et al., 2005). The atmospheric behavior of PAHs is significantly affected by meteorological conditions (wind speed and direction, temperature, and precipitation) (Choi et al., 2012, 2007), photolysis, and reactive species (ozone and hydroxyl radicals) (Reisen and Arey, 2005; Tsapakis and Stephanou, 2005). PAH derivatives and their atmospheric behavior have recently become a more prominent environmental concern. Nitrated PAHs (NPAHs) are emitted directly from combustion processes and produced by secondary reactions of hydroxyl and NO3 radicals with gaseous PAHs (Reisen and Arey, 2005; Tsapakis and Stephanou, 2005). Oxygenated PAHs (OPAHs) are produced by combustion processes and heterogeneous reactions between ozone and particulate PAHs (Tsapakis and Stephanou, 2005). In addition, halogenated PAHs (Halo-PAHs), such as chlorinated PAHs (ClPAHs) and brominated PAHs (BrPAHs) have recently received an increasing amount of attention (Ohura et al., 2018). Incomplete combustion of organic materials with halogen sources such as hydrochloric acid can produce Halo-PAHs (Fujima et al., 2006). Furthermore, they are formed by direct chlorination of the parent PAHs during waste incineration and combustion of polyvinylchloride (PVC) (Fujima et al., 2006; Horii et al., 2008; Miyake et al., 2017). These substances can have adverse effects on human health (Ohura et al., 2007; Yoshino and Urano, 1997). The detection of Halo-PAHs in tap water, road tunnels, and soil has been reported for decades (Nilsson and Oestman, 1993; Ishaq et al., 2003). Nevertheless, research on atmospheric Halo-PAHs has only recently been published (Ohura et al., 2018, 2016; Kakimoto et al., 2014; Ma et al., 2013; Jin et al., 2017a). In these previous studies, only high-volume air samplers (HiVols) were used. Therefore, high-resolution spatial distributions of Halo-PAHs could not be well examined. Polyurethane foam-passive air samplers (PUF-PASs) have been used worldwide because they can be deployed at any site without electricity, where HiVols are impractical (Shoeib and Harner, 2002; Harner et al., 2004). They have been used in many studies on PAHs and persistent organic pollutants (POPs) that contain chlorines or bromines. For instance, the spatial distribution of PAHs in Ulsan, an industrial city, indicated that industrial complexes were the primary source of PAHs (Choi et al., 2012). The vertical distributions of polychlorinated biphenyls (PCBs), PAHs, and organochlorine pesticides (OCPs) were investigated in an urban city, and it was found that the levels of PAHs and PCBs sharply decreased with increasing elevation above the ground (Farrar et al., 2005). On a larger scale, the spatial distributions of PCBs, OCPs, polychlorinated naphthalene (PCNs), PAHs, lindane, hexachlorobenzene (HCB), and polybrominated biphenyl ethers (PBDEs) were studied using PUF-PASs in many European countries to illustrate the feasibility of mapping local, regional, and global sources of POPs (Jaward et al., 2004; Farrar et al., 2006). In addition, worldwide air monitoring of POPs revealed that PCB levels in Africa were relatively high due to the import of electronic waste from developed countries, and the usage of dichlorodiphenyl trichloroethane (DDT) to control malaria has caused high concentrations of DDT in Africa and the Pacific Islands (Bogdal et al., 2013; Harner et al., 2006). Regarding the derivatives of PAHs, PUF-PASs were applied to investigate the spatial distributions of NPAHs and OPAHs in four major urban cities in Nepal, which revealed that the secondary sources associated with photochemical reactions and burning of solid fuels and crop residue are the main sources of NPAHs and OPAHs, respectively (Yadav et al., 2018). However, the assessment of Halo-PAHs using PASs has yet to be addressed. Ulsan Metropolitan City is one of the largest industrial cities in
2. Materials and methods 2.1. Passive air sampling A total of 40 PUF-PASs were deployed in duplicate at 20 sites in Ulsan. The sampling sites were categorized into three classes: industrial (I1, I2, I3, and I4), rural (R1, R2, R3, R4, R5, and R6), and urban (U1, U2, U3, U4, U5, U6, U7, U8, U9, and U10) sites (Fig. 1). The sampling period was from March 8, 2013 to May 31, 2013 (84 days). Prior to deployment, PUF disks were sequentially sonicated with acetone and nhexane for 30 min, respectively. The deployed PUF disks were retrieved and stored in polyethylene zipper bags at −9 °C. 2.2. Instrumental analysis and QA/QC A full description of the instrumental procedure is provided in Text S1 in the Supplementary Material. In summary, retrieved PUF disks were spiked with surrogate standards prior to Soxhlet extraction with nhexane/acetone. The extracts were cleaned up with n-hexane/dichloromethane through silica gel columns. An internal standard was added to the vials prior to instrumental analysis. The target compounds were 24 ClPAHs, 11 BrPAHs, and 13 parent PAHs. Their abbreviations are listed in Table S1 in the Supplementary Material. The target compounds were analyzed using gas chromatography (GC)-electron impact mass spectrometry (MS) in the selected ion monitoring mode (Agilent 7890 GC-5975C MS, USA). The average recoveries were 52%, 98%, and 93% for phenanthrene-d10, chrysene-d12, and perylene-d12, respectively. Method detection limits (MDLs) for each Halo-PAH and PAH species are summarized in Table S2. 2.3. Calculation of air concentrations Ambient air concentrations (CAir) of Halo-PAHs and PAHs were calculated from their accumulated amounts on PUF disks (CPAS) and the passive sampling rate (Rs). In this study, the compound-specific sampling rates in seven sampling zones were estimated using a prediction model. A full description of the calculation method is provided in Text S2. 2.4. Meteorological conditions Meteorological data from seven automatic weather stations (AWSs) in Ulsan (Fig. 1) were obtained from the Korea Meteorological Administration (http://sts.kma.go.kr/). CALPro Plus 7 (TRC Environmental Corp., USA), an MS window version of CALPUFF, was applied to estimate the average wind-field during the sampling period (Fig. S1 in the Supplementary Material). For estimating the sampling rates of each Halo-PAH and PAH species, the 20 sampling sites were matched with 7 AWSs based on this wind pattern. A full description of the 2
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
Fig. 1. Locations of 20 sampling sites, 7 automatic weather stations, and 4 major industrial areas in Ulsan, South Korea.
Fig. 2. Mean concentrations of ClPAHs, BrPAHs (the left axis), and PAHs (the right axis) at 20 sites in Ulsan, Korea. The error bars represent standard deviations. Only concentrations of the detected compounds are plotted.
3. Results and discussion
meteorological conditions is provided in Text S3.
3.1. Levels of Halo-PAHs and PAHs 2.5. Toxicity evaluation The average sampling rates for all the target compounds ranged from 1.85 ± 0.13 m3/day in zone C to 3.69 ± 0.2 m3/day in zone G (Fig. S2). This result is in good agreement with those in previous studies (Chaemfa et al., 2008; He and Balasubramanian, 2010; Pozo et al., 2009; Herkert et al., 2018). The low sampling rate (∼2 m3/day) in zones A and C can be explained by low wind speeds due to the high forest coverage, whereas other zones on the coastline with higher wind speeds showed elevated sampling rates (∼3.5 m3/day) (Figs. S1 and S2). This trend was also observed in the former study (Herkert et al., 2018). In the previous studies, PUF-PASs have been reported to effectively collect gaseous compounds (Shoeib and Harner, 2002; Pozo et al., 2004). Particulate PAHs can also be collected on PUF-PASs to some degree (Harner et al., 2013; Markovic et al., 2015). However, the accumulation mechanism of the particulate pollutants on PUF disks has not been clearly understood, leading to the high uncertainty of the
Toxicity equivalency quantities (TEQs) for ClPAHs, BrPAHs, and PAHs were calculated based on the relative equivalency potencies (REPs). A full description of the calculation method is provided in Text S4.
2.6. Statistical analysis Normality tests, Pearson and Spearman correlation analyses, and rank sum tests were performed using SPSS statistics 22.0 (IBM, USA) and SigmaPlot 12.0 (Systat Software Inc., USA) to support data interpretation. The detailed information about the statistical analysis is provided in Text S5.
3
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
with Ant and Phe imply that ClPAHs can be closely related to their corresponding parent PAHs. However, the most energetically favorable ClPAHs to be formed from the corresponding parent PAHs (e.g., 2ClAnt) were not generated or generated with lower concentrations, whereas the most unfavorable ClPAHs (e.g., 9-ClAnt, 9,10-Cl2Ant, and 9-ClPhe) were found in high concentrations. A previous study reported that ClPAHs were mostly formed in a cooling zone of secondary copper smelters, not by the direct chlorination of PAHs during smelting processes (Jin et al., 2017b). The patterns of Halo-PAHs and PAHs at the emission source can be different from those at the receptor sites due to air dispersion and G/P partitioning behaviors. However, the industrial sites and several urban and rural sites in the present study are close to or inside of the industrial complexes; thus, the patterns of Halo-PAHs and PAHs at the source and receptor sites in Ulsan might be similar to each other. Consequently, it is unlikely that the major ClPAHs were produced by the direct chlorination of parent PAHs. However, it is hard to generalize the formation mechanism of ClPAHs because of not enough data. Due to the limited number of BrPAHs analyzed in this study, the eight detected BrPAHs could not be considered for Gibbs free energies. Among the target BrPAHs, only 2-BrFlu, 7-BrBaA, and 6-BrBaP were significantly correlated with their parent compounds, Flu (r = 0.67, p < 0.01), BaA (r = 0.49, p < 0.05), and BaP (r = 0.51, p < 0.05) (Table S8). In the previous study, BrPAHs were reported to be generated not from the direct bromination of the parent PAHs in iron ore sintering and secondary non-ferrous metal smelting (Xu et al., 2018). The main formation process of BrPAHs might be the secondary reactions between parent PAHs and other precursors in the atmosphere (Ni and Zeng, 2012). A similar formation mechanism has also been suggested for other environmental compartments, such as river sediment (Sun et al., 2011) and soil (Ni and Zeng, 2012). Overall, the exact formation mechanism of Halo-PAHs still remains unclear, and further investigation is required. The total mean concentration of Halo-PAHs in this study was compared with other monitoring data from different cities in China and Japan (Text S6, Table S9). Note that HiVols were used for all the previous studies, whereas PUF-PASs (for mostly gaseous compounds) were used in this study. In addition, characteristics of sampling sites, sampling periods, and target compounds were different from each other. Therefore, this comparison was used for a rough evaluation of the pollution level of Halo-PAHs. Overall, the levels of both ClPAHs and BrPAHs in Ulsan were found to be much higher than those in other cities, but the trend for parent PAHs was reversed. The reason for this discrepancy is unclear, but it seems to be related to heavy industrial activities in Ulsan. Therefore, further studies and long-term monitoring of Halo-PAHs are required.
particulate concentration calculation (Gouin et al., 2010). Therefore, all the following results and discussions in this study will focus mainly on the gaseous phase. Out of the 35 target Halo-PAHs, 16 ClPAHs and 8 BrPAHs were detected (Table S2, Fig. 2). The most abundant compounds were 9ClPhe and 2-BrFlu, which contributed 45% and 40% to the total mean ClPAH and BrPAH concentrations, respectively. The next major compounds were 9-ClAnt (19%) and Cl4Pyr (9%) for ClPAHs and 9-BrAnt (20%) and 9-BrPhe (16%) for BrPAHs. The low molecular weight compounds (LMWs) with one halogen atom were more dominant than the high molecular weight compounds (HMWs) with two or more halogen atoms. Similar trends of gaseous ClPAHs in the ambient air of Shizuoka, Japan have also been reported. LMW-PAHs tend to exist mostly in the gaseous phase, while HMW-PAHs are dominant in the particulate phase (Ohura et al., 2013, 2008). Thus, this observation of gaseous ClPAHs can be explained by the similar gas/particle (G/P) partitioning behavior between Halo-PAHs and their parent PAHs. However, the influence of the number of halogen atoms on the G/P partitioning of Halo-PAHs needs to be further studied (Ohura et al., 2008). For parent PAHs, all the target compounds except for DbahA were detected. Like its derivatives, Phe was dominant, accounting for 40% of the total PAHs, followed by Flt (17%) and Pyr (14%). These results are consistent with those reported in a previous study in Ulsan (Choi et al., 2012). The stability of PAHs in physical and chemical reactions greatly depends on their structural arrangements (Schwarzenbach et al., 2005). For example, linear or 2-ring PAHs can be rapidly photodegraded, whereas angular PAHs are resistant to sunlight (Abdel-Shafy and Mansour, 2016). Furthermore, gaseous PAHs are degraded easily and quickly by photolysis (Finlayson-Pitts and Pitts, 1997). Therefore, the relatively low concentrations of Flu (3.3 ng/m3) and Ant (0.5 ng/m3) compared with another parent PAHs can be explained by their high rates of degradation associated with the aforementioned processes. Moreover, their high reactivity could lead to large formations of their derivatives: 9-ClAnt (39 pg/m3), 2-BrFlu (34 pg/m3), and 9-BrAnt (17 pg/m3) (Fig. 2). Previous studies have shown that direct chlorination of parent PAHs might be a major formation pathway of their daughter Halo-PAH products. For example, it has been reported that monochlorinated derivatives of Flu and Ant were predominantly generated from combustion of polyvinylchloride at lower temperatures compared to di- or tri-homologue (Wang et al., 2003). A study on ClPAH formation in municipal waste incinerators (MSWIs) suggested that these compounds are mainly formed by the chlorination of parent PAHs with Cl2 in the exhaust gas rather than in the incineration furnaces (Yoshino and Urano, 1997). According to the relative Gibbs free energies and the ClPAH formation from waste incineration, monochlorinated PAHs, 2-ClAnt and 3ClPhe, are the most favorable derivatives generated from direct chlorination of the corresponding parent PAHs (Wang et al., 2003; Jin et al., 2017b). However, 2-ClAnt was not detected in any samples in the present study. Instead, 9-ClAnt (19%) was one of the most abundant congeners in ClPAH profiles (Fig. 2), and its direct formation from Ant requires the highest energy level (Jin et al., 2017b). Moreover, 9-ClAnt displayed a strong positive correlation with its corresponding parent PAH (r = 0.67, p < 0.01) (Table S7). In addition, 9,10-Cl2Ant, which requires significantly more energy for the direct chlorination of Ant than the others in di-homologue (Jin et al., 2017b), was also detected (Fig. 2). 9-ClPhe, which requires the highest energy for its direct formation from Phe (Jin et al., 2017b), was the most dominant species in the profile of ClPAHs, and it was significantly correlated with Phe (r = 0.74, p < 0.01), whereas the other Phe derivatives displayed no correlation with Phe (Table S7). For other ClPAHs, even though the mono-derivatives were well correlated with their corresponding parent PAHs (Table S7), their contribution to the total mean concentration of ClPAHs (12%) was much lower than those of 9-ClAnt and 9-ClPhe (19% and 45%, respectively). The strong correlations of 9-ClAnt and 9-ClPhe
3.2. Spatial distributions of Halo-PAHs and PAHs This study is the first one that reports levels and patterns of HaloPAHs in Ulsan, and there is no information or data about their sources in the city. Therefore, the source-receptor relationships for Halo-PAHs were evaluated based on their spatial distributions. The industrial area, as expected, generally displayed higher concentrations of both HaloPAHs and PAHs than the urban and rural areas (Fig. 3). According to the correlation analysis, the concentration of Σ24 ClPAHs and Σ11 BrPAHs at the 20 sampling sites were significantly correlated (r = 0.60, p < 0.01), implying that these two types of Halo-PAHs have similar emission sources and/or formation mechanisms. A stronger correlation was found between Σ24 ClPAHs and Σ13 PAHs (r = 0.71, p < 0.01) (Table S7) than between Σ11 BrPAHs and Σ13 PAHs (r = 0.37, p > 0.05) (Table S8), which suggests that the levels of ClPAHs could be influenced more by their parent compounds. This finding is in line with the relationship between Halo-PAHs and PAHs in a previous study (Ohura et al., 2009). For ClPAHs, the industrial sites showed higher levels 4
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
Fig. 3. Spatial distributions of Σ24 ClPAHs, Σ11 BrPAHs, and Σ13 PAHs in Ulsan, Korea (March 8–May 31, 2013). Three contour maps for each PAH group were drawn using ArcGIS 10.5 (ESRI Inc., USA). Inverse distance weighting (IDW) was employed as an interpolation method.
(609 ± 440 pg/m3) than rural (308 ± 348 pg/m3) and urban (147 ± 92 pg/m3) sites. No statistical difference was found between the urban and rural sites (Mann-Whitney rank sum test, p > 0.05) due to the similar concentrations of ClPAHs at the urban and rural sites, which were located far from the industrial complexes. By contrast, the industrial sites showed significant statistical differences when compared to the urban and rural sites (Mann-Whitney rank sum test, p < 0.01). For BrPAHs, the order of the mean concentrations was different from those of ClPAHs and PAHs: the industrial (204 ± 86 pg/ m3) > urban (119 ± 171 pg/m3) > rural (85 ± 43 pg/m3) sites. There was no statistical difference among them (Mann-Whitney rank sum test, p > 0.05). For PAHs, the mean concentration at the industrial sites (40 ± 11 ng/m3) was higher than those detected at the rural (31 ± 24 ng/m3) and urban (18 ± 11 ng/m3) sites. However, there was no statistical difference among them (Mann-Whitney rank sum test, p > 0.05) because some urban and rural sites near the industrial complexes also displayed relatively high concentrations. A similar spatial distribution for PAHs and statistical results were also reported in our previous study (Choi et al., 2012). The same orders (I > R > U) of PAHs and ClPAHs can also be expected by a strong correlation between them (r = 0.71, p < 0.01) (Table S7), whereas
BrPAHs had a weak correlation of 0.37 (p > 0.05) with PAHs (Table S8) and exhibited a different order of mean concentrations (I > U > R). Therefore, there might be some common sources of ClPAHs and parent PAHs, which are not the potential sources of BrPAHs. These different spatial distributions (or correlation results) warrant further investigation to better understand their source-receptor relationships. Among the rural sites, R1 and R2 showed significantly high concentrations of both ClPAHs and PAHs because they are close to industrial complexes (Fig. 3). Site R1 is located between the automobile industrial complex and the shipbuilding and heavy industrial complex, and sites R2 and I3 are located on the edge of the petrochemical industrial complex. These three sites could be further influenced by south-southwesterly winds blown from the non-ferrous industrial complex (Fig. S1), which is supported by the high concentrations of ClPAHs and PAHs detected at site I4. High concentrations of BrPAHs were observed at site I1 in the automobile industrial complex and I2 and I3 in the petrochemical industrial complex. The two urban sites, U1 and U2, which are relatively close to the industrial complexes, displayed the highest concentrations of BrPAHs, with site U2 potentially being significantly affected by westerly and southwesterly winds transporting BrPAHs from the automobile and petrochemical industrial 5
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
Fig. 4. Levels and fractions of Halo-PAHs and PAHs classified by the number of rings and halogen atoms at the individual sampling sites.
species at R5 (63%) and R6 (65%), other compounds were accounted for quite equally (6%), and 1-ClPyr contributed very small proportions at R5 and R6 (3% and 0%, respectively) (Fig. S3a and b). Moreover, the profile of ClPAHs at site I4 with 9-ClPhe (27%), 9-ClAnt (12%), 3-ClFlt (4%), and 1-ClPyr (26%) (Fig. S3b) was similar to the profile of gaseous ClPAHs found in secondary copper smelters (Jin et al., 2017b). Hence, the four ClPAHs seem to be indicative of non-ferrous industrial operations. Regarding BrPAH profiles, the group of Br-3 rings accounted for 96% of Σ11 BrPAHs at site I1 (Fig. 4d). Particularly, the fraction of 2BrFlu (75%) at site I1 was higher than those at the other three industrial sites (Fig. S3d). This pattern might affect the high contribution of 2BrFlu (70%) at site U2, which is likely due to the influence of the northwesterly winds. Thus, 2-BrFlu could be a typical marker for the automobile industry. The group of Br-3 rings accounted for the majority at virtually every site; however, the Br2-4 ring group was the most abundant at site I2 (68%) and comprised 39% of the total BrPAH concentration at site U6 (Fig. 4c and d), while the concentration of Br24 rings (4,7-Br2BaA and 7,12-Br2BaA) was the second highest at site U2 (62 pg/m3, 11%) (Fig. S3c and d). These three sites are close to each other, and southwesterly and easterly winds might increase the influence of the petrochemical industries on the BrPAH profiles at sites U6 and U2. Given the absence of Br2-3 rings at all the sampling sites, the profiles of BrPAHs in Ulsan were quite different from the gaseous profiles in Beijing (Jin et al., 2017a). This difference might be partly explained by different sampling conditions and environmental factors, but the reasons remain unclear due to limited information. Although it was reported that BrPAHs might not share the same potential sources of ClPAHs in soil (Ni and Zeng, 2012), the information on the main sources and the formation mechanism of BrPAHs is still limited. For the profiles of PAHs, 3 and 4 rings accounted for the majority of
complexes. In general, the sampling sites close to the industrial complexes showed relatively high concentrations of PAHs. Furthermore, urban site U3, located on the western side of a medium-industrial belt (i.e., Jungsan and Maegok general industrial complexes, mainly producing auto parts for vehicles), showed relatively high concentrations of PAHs because of heavy-duty vehicle traffic in addition to the nearby industrial activities. 3.3. Profiles of Halo-PAHs and PAHs Levels and fractions for each group categorized by the number of halogen atoms and rings at each sampling site are presented in Fig. 4. For ClPAHs, there was a noticeable difference among the four industrial sites. The average fraction of the first group (Cl-3 rings) at sites I1 and I2 was 68%, which is twice that found at sites I3 and I4 (30%) (Fig. 4a and b). Similar profiles were observed at sites U2 and R2, which are located in the vicinity of I1 and I2 under the influence of northwesterly winds during the sampling period. Particularly, two compounds (9ClPhe and 9-ClAnt) accounted for 67% and 69% of the Σ24 ClPAHs at sites I1 and I2, respectively (Fig. S3a and b). This pattern was consistent with the findings in a multi-industrial area in Tokyo Bay, Japan (Ohura et al., 2018). Therefore, those species could be considered as markers associated with the automobile and shipbuilding industries. Cl-4 rings accounted for 42% and 30% of the Σ24 ClPAHs at sites I3 and I4, respectively (Fig. 4a and b). More specifically, site I4 had the highest fractions of 3-ClFlt (4%) and 1-ClPyr (26%) among the sites, whereas site I3 had the highest proportion of 6-ClChr (30%) (Fig. S3a and b). The profiles of ClPAHs at sites U1 (4% of 3-ClFlt) and R1 (5% of 1ClPyr) might be also influenced by the high fractions of those two compounds at site I4 combined with the influence of the south-southwesterly winds (Fig. S1). Even though 9-ClPhe was the most abundant 6
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
(18.29 pg-TEQ/m3) was higher than those at the urban (12.98 pg-TEQ/ m3) and rural (14.94 pg-TEQ/m3) sites (Fig. 5). There was no statistical difference (Mann-Whitney rank sum test, p > 0.05) among them because several urban and rural sites, located in the vicinity of the industrial complexes, also showed relatively high TEQ values. The total TEQ value in Ulsan (15 pg-TEQ/m3) was comparable with that in Shizuoka, Japan (Table S10) (Ohura et al., 2009). Meanwhile, the total PAH TEQs in Ulsan was three times those in Busan, South Korea and Tokyo, Japan, and one-third of those in Nagoya, Japan and Beijing, China (Ohura et al., 2018, 2016; Kakimoto et al., 2014; Jin et al., 2017a). According to the REPBaP values for individual compounds (Table S11), virtually all Halo-PAH species were more toxic than their corresponding parent PAHs. Particularly, 6-ClChr is 233 times more toxic than its parent compound. Furthermore, for 3- and 4-ring ClPAHs, the toxicity significantly increases with the number of halogen substitutions. The REPBaP values of 3,9-Cl2Phe and 3,8-Cl2Flt are about 80- and 1,900-fold higher than those of Phe and Flt, respectively. The reason for this toxicological trend of Halo-PAHs is still unclear; however, AhR activity, CYP1A1 mRNA expression in MCF-7 cells, and Ames mutagenicity on bacteria strains (TA98 and TA100) for Halo-PAHs and parent PAHs were confirmed (Ohura et al., 2007; Yoshino and Urano, 1997). Hence, the toxicological mechanism of Halo-PAHs based on the number and position of the substituted halogen atoms deserves further attention.
Fig. 5. Spatial distributions of the total toxicity equivalency quantities (TEQs) of Halo-PAHs and PAHs at each site.
PAH species at every sampling site (about 86%) (Fig. 4e and f). In particular, Phe (40%), Flt (17%), Pyr (14%), and Flu (13%) showed higher mean fractions (Fig. S3e and f), which is consistent with the previously reported profiles for gaseous PAHs in Ulsan (Choi et al., 2012; Nguyen et al., 2018). In these former studies, contamination characterization and source identification of parent PAHs were accomplished using statistical tools (e.g., principal component analysis and positive matrix factorization) and diagnostic ratios. Therefore, additional interpretation and further discussions have not been included in this study.
4. Conclusion The air pollution by ClPAHs, BrPAHs, and PAHs in the multi-industrial city of Ulsan was investigated using the passive sampling technique. Overall, Ulsan, especially the industrial complexes, showed higher concentrations of both ClPAHs and BrPAHs than other cities. Major wind patterns played a vital role in the dispersion of Halo-PAHs from the industrial zones to the surrounding urban and rural areas. According to the spatial distributions and correlation analysis, the total concentrations of Halo-PAHs and parent PAHs were significantly correlated, suggesting they have similar potential sources. However, the most energetically unfavorable Halo-PAHs exhibited strong correlations with their corresponding parent compounds, indicating that the direct halogenation of PAHs is not a major formation pathway of individual Halo-PAHs. In this regard, more in-depth investigations on the formation mechanisms and identification of major anthropogenic sources are necessary. Despite the comparable TEQs of PAHs with those of other cities, Ulsan showed significantly high TEQs of both ClPAHs and BrPAHs because of the great contributions of the most toxic species, possibly linked to industrial activities. On the basis of this first passive sampling study on Halo-PAHs, we are conducting seasonal monitoring using PUF-PASs and HiVols for more understanding of the contamination characteristics of Halo-PAHs.
3.4. Toxicity evaluation of Halo-PAHs and PAHs Total TEQs of 11 ClPAHs, 8 BrPAHs, and 11 PAHs at the individual sampling site are presented in Fig. 5. The average TEQ of ClPAHs at the industrial sites (4.21 pg-TEQ/m3) was significantly higher than those at the urban (0.72 pg-TEQ/m3, Mann-Whitney rank sum test, p < 0.01) and rural sites (0.42 pg-TEQ/m3, Mann-Whitney rank sum test, p < 0.01). The highest ClPAH TEQ at site I3 (13 pg-TEQ/m3) was due to the most abundant proportion (30%) of the second most toxic ClPAH (6-ClChr, REPBaP = 2.1). Sites I4, R1, and R2 also showed higher concentrations of ClPAHs; however, 3,8-Cl2Flt (REPBaP = 5.7) and 6-ClChr only accounted for small fractions (< 2%) in their profiles. For BrPAHs, the industrial sites (0.53 pg-TEQ/m3) showed a higher average TEQ than the urban (0.16 pg-TEQ/m3) and rural sites (0.10 pg-TEQ/m3); however, there was no statistical difference among them (MannWhitney rank sum test, p > 0.05). Sites I2 (2 pg-TEQ/m3) and U2 (1 pg-TEQ/m3) showed the highest BrPAH TEQs due to the major contributions from 4,7-Br2BaA, the second most toxic BrPAHs (REPBaP = 0.77), in their profiles (I2: 38% and U2: 10%). In general, the total TEQs of both ClPAHs and BrPAHs in this study were found to be substantially higher than those in other Northeast Asian cities (Table S10). For example, the total TEQ of ClPAHs in Ulsan was 31 times higher than that in Tokyo, Japan (Ohura et al., 2018), 14 times higher than the neighboring Korean city of Busan (Kakimoto et al., 2014), and 9 times higher than that in Beijing, China (Jin et al., 2017a). For BrPAHs, the total TEQ in Ulsan was about fourfold higher than that in Shizuoka, Japan (Ohura et al., 2009), and 121-fold higher than that in Beijing, China (Jin et al., 2017a). Despite limited reports, the comparison data suggest that Ulsan is more contaminated by Halo-PAHs than other cities, and further studies on source identification are essential. Regarding parent PAHs, the average TEQ at the industrial sites
Acknowledgements The research was supported by the 2019 Research Fund (1.190011.01) of UNIST and the National Research Foundation of Korea (NRF) (NRF-2017R1A2B4003229 and 2017M3D8A1092015). We are thankful for the discussions with Dr. Tom Harner for estimating KOA values of Halo-PAHs and calculation of ambient air concentrations. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jhazmat.2019.121238. References Abdel-Shafy, H.I., Mansour, M.S.M., 2016. A review on polycyclic aromatic
7
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
of air monitoring by passive air sampling in five continents. Trends Analyt. Chem. 46, 150–161. Harner, T., Pozo, K., Gouin, T., Macdonald, A.-M., Hung, H., Cainey, J., Peters, A., 2006. Global pilot study for persistent organic pollutants (POPs) using PUF disk passive air samplers. Environ. Pollut. 144, 445–452. Yadav, I.C., Devi, N.L., Singh, V.K., Li, J., Zhang, G., 2018. Concentrations, sources and health risk of nitrated- and oxygenated-polycyclic aromatic hydrocarbon in urban indoor air and dust from four cities of Nepal. Sci. Total Environ. 643, 1013–1023. Vu, V.-T., Lee, B.-K., Ny, M.T., Kim, J.-T., 2010. A Study on Characteristics of Organic Carbon and Polycyclic Aromatic Hydrocarbons (PAHs) in PM10 at the Residential and Industrial Areas in Ulsan of Korea. International Forum on Strategic Technology, pp. 263–266. Lee, B.-K., Lee, C.-B., 2004. Development of an improved dry and wet deposition collector and the atmospheric deposition of PAHs onto Ulsan Bay, Korea. Atmos. Environ. 38, 863–871. Dong, T.T.T., Lee, B.-K., 2009. Characteristics, toxicity, and source apportionment of polycylic aromatic hydrocarbons (PAHs) in road dust of Ulsan, Korea. Chemosphere 74, 1245–1253. Chaemfa, C., Barber, J.L., Gocht, T., Harner, T., Holoubek, I., Klanova, J., Jones, K.C., 2008. Field calibration of polyurethane foam (PUF) disk passive air samplers for PCBs and OC pesticides. Environ. Pollut. 156, 1290–1297. He, J., Balasubramanian, R., 2010. A comparative evaluation of passive and active samplers for measurements of gaseous semi-volatile organic compounds in the tropical atmosphere. Atmos. Environ. 44, 884–891. Pozo, K., Harner, T., Lee, S.C., Wania, F., Muir, D.C.G., Jones, K.C., 2009. Seasonally resolved concentrations of persistent organic pollutants in the global atmosphere from the first year of the GAPS study. Environ. Sci. Technol. 43, 796–803. Herkert, N.J., Spak, Scott N., Smith, A., Schuster, J.K., Harner, T., Martinez, A., Hornbuckle, K.C., 2018. Calibration and evaluation of PUF-PAS sampling rates across the Global Atmospheric Passive Sampling (GAPS) network. Environ. Sci.: Processes Impacts. 20, 210–219. Pozo, K., Harner, T., Shoeib, M., Urrutia, R., Barra, R., Parra, O., Focardi, S., 2004. Passive-sampler derived air concentrations of persistent organic pollutants on a north-south transect in Chile. Environ. Sci. Technol. 38, 6529–6537. Harner, T., Su, K., Genualdi, S., Karpowicz, J., Ahrens, L., Mihele, C., Schuster, J., Charland, J.-P., Narayan, J., 2013. Calibration and application of PUF disk passive air samplers for tracking polycyclic aromatic compounds (PACs). Atmos. Environ. 75, 123–128. Markovic, M.Z., Prokop, S., Staebler, R.M., Liggio, J., Harner, T., 2015. Evaluation of the particle infiltration efficiency of three passive samplers and the PS-1 active air sampler. Atmos. Environ. 112, 289–293. Gouin, T., Wilkinson, D., Hummel, S., Meyer, B., Culley, A., 2010. Polycyclic aromatic hydrocarbons in air and snow from Fairbanks, Alaska. Atmos. Pollut. Res. 1, 9–15. Ohura, T., Horii, Y., Kojima, M., Kamiya, Y., 2013. Diurnal variability of chlorinated polycyclic aromatic hydrocarbons in urban air, Japan. Atmos. Environ. 81, 84–91. Ohura, T., Fujima, S., Amagai, T., Shinomiya, M., 2008. Chlorinated polycyclic aromatic hydrocarbons in the atmosphere: seasonal levels, gas-particle partitioning, and origin. Environ. Sci. Technol. 42, 3296–3302. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2005. Environmental Organic Chemistry. John Wiley & Sons, Inc., Hoboken, New Jersey, USA. Finlayson-Pitts, B.J., Pitts Jr., J.N., 1997. Tropospheric air pollution: ozone, airborne toxics, polycyclic aromatic hydrocarbons, and particles. Science 276, 1045–1052. Wang, D., Xu, X., Chu, S., Zhang, D., 2003. Analysis and structure prediction of chlorinated polycyclic aromatic hydrocarbons released from combustion of polyvinylchloride. Chemosphere 53, 495–503. Jin, R., Liu, G., Zheng, M., Jiang, X., Zhao, Y., Yang, L., Wu, X., Xu, Y., 2017b. Secondary copper smelters as sources of chlorinated and brominated polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 51, 7945–7953. Xu, Y., Yang, L., Zheng, M., Jin, R., Wu, X., Li, C., Liu, G., 2018. Chlorinated and brominated polycyclic aromatic hydrocarbons from metallurgical plants. Environ. Sci. Technol. 52, 7334–7342. Ni, H.-G., Zeng, E.Y., 2012. Environmental and human exposure to soil chlorinated and brominated polycyclic aromatic hydrocarbons in an urbanized region. Environ. Toxicol. Chem. 31, 1494–1500. Sun, J.-L., Ni, H.-G., Zeng, H., 2011. Occurrence of chlorinated and brominated polycyclic aromatic hydrocarbons in surface sediments in Shenzhen, South China and its relationship to urbanization. J. Environ. Monit. 13, 2775–2781. Ohura, T., Sawada, K., Amagai, T., Shinomiya, M., 2009. Discovery of novel halogenated polycyclic aromatic hydrocarbons in urban particulate matters: occurrence, photostability, and AhR activity. Environ. Sci. Technol. 43, 2269–2275. Nguyen, T.N.T., Jung, K.-S., Son, J.M., Kwon, H.-O., Choi, S.-D., 2018. Seasonal variation, phase distribution, and source identification of atmospheric polycyclic aromatic hydrocarbons at a semi-rural site in Ulsan, South Korea. Environ. Pollut. 236, 529–539.
hydrocarbons: source, environmental impact, effect on human health and remediation. Egypt. J. Pet. 25, 107–123. Armstrong, B.G., Gibbs, G., 2009. Exposure–response relationship between lung cancer and polycyclic aromatic hydrocarbons (PAHs). Occup. Environ. Med. 66, 740–746. Choi, S.-D., Baek, S.-Y., Chang, Y.-S., 2007. Influence of a large steel complex on the spatial distribution of volatile polycyclic aromatic hydrocarbons (PAHs) determined by passive air sampling using membrane-enclosed copolymer (MECOP). Atmos. Environ. 41, 6255–6264. Choi, S.-D., Kwon, H.-O., Lee, Y.-S., Park, E.-J., Oh, J.-Y., 2012. Improving the spatial resolution of atmospheric polycyclic aromatic hydrocarbons using passive air samplers in a multi-industrial city. J. Hazard. Mater. 241–242, 252–258. Tang, N., Hattori, T., Taga, R., Igarashi, K., Yang, X., Tamura, K., Kakimoto, H., Mishukov, V.F., Toriba, A., Kizu, R., Hayakawa, K., 2005. Polycyclic aromatic hydrocarbons and nitropolycyclic aromatic hydrocarbons in urban air particulates and their relationship to emission sources in the Pan-Japan Sea countries. Atmos. Environ. 39, 5817–5826. Reisen, F., Arey, J., 2005. Atmospheric reactions influence seasonal PAH and Nitro-PAH concentrations in the Los Angeles basin. Environ. Sci. Technol. 39, 64–73. Tsapakis, M., Stephanou, E.G., 2005. Polycyclic aromatic hydrocarbons in the atmosphere of the Eastern Mediterranean. Environ. Sci. Technol. 39, 6584–6590. Ohura, T., Horii, Y., Yamashita, N., 2018. Spatial distribution and exposure risks of ambient chlorinated polycyclic aromatic hydrocarbons in Tokyo Bay area and network approach to source impacts. Environ. Pollut. 232, 367–374. Fujima, S., Ohura, T., Amagai, T., 2006. Simultaneous determination of gaseous and particulate chlorinated polycyclic aromatic hydrocarbons in emissions from the scorching of polyvinylidene chloride film. Chemosphere 65, 1983–1989. Horii, Y., Ok, G., Ohura, T., Kannan, K., 2008. Occurrence and profiles of chlorinated and brominated polycyclic aromatic hydrocarbons in waste incinerators. Environ. Sci. Technol. 42, 1904–1909. Miyake, Y., Tokumura, M., Wang, Q., Amagai, T., Horii, Y., Kannan, K., 2017. Mechanism of formation of chlorinated pyrene during combustion of polyvinyl chloride. Environ. Sci. Technol. 51, 14100–14106. Ohura, T., Morita, M., Makino, M., Amagai, T., Shimoi, K., 2007. Aryl hydrocarbon receptor-mediated effects of chlorinated polycyclic aromatic hydrocarbons. Chem. Res. Toxicol. 20, 1237–1241. Yoshino, H., Urano, K., 1997. Formation of chlorinated PAHs in exhaust gas from municipal waste incinerators, and their mutagenic activities. Toxicol. Environ. Chem. 63, 233–246. Nilsson, U.L., Oestman, C.E., 1993. Chlorinated polycyclic aromatic hydrocarbons: method of analysis and their occurrence in urban air. Environ. Sci. Technol. 27, 1826–1831. Ishaq, R., Näf, C., Zebühr, Y., Broman, D., Järnberg, U., 2003. PCBs, PCNs, PCDD/Fs, PAHs and Cl-PAHs in air and water particulate samples-patterns and variations. Chemosphere 50, 1131–1150. Ohura, T., Kamiya, Y., Ikemori, F., 2016. Local and seasonal variations in concentrations of chlorinated polycyclic aromatic hydrocarbons associated with particles in a Japanese megacity. J. Hazard. Mater. 312, 254–261. Kakimoto, K., Nagayoshi, H., Konishi, Y., Kajimura, K., Ohura, T., Hayakawa, K., Toriba, A., 2014. Atmospheric chlorinated polycyclic aromatic hydrocarbons in East Asia. Chemosphere 111, 40–46. Ma, J., Chen, Z., Wu, M., Feng, J., Horii, Y., Ohura, T., Kannan, K., 2013. Airborne PM2.5/ PM10-associated chlorinated polycyclic aromatic hydrocarbons and their parent compounds in a suburban area in Shanghai, China. Environ. Sci. Technol. 47, 7615–7623. Jin, R., Liu, G., Jiang, X., Liang, Y., Fiedler, H., Yang, L., Zhu, Q., Xu, Y., Gao, L., Su, G., Xiao, K., Zheng, M., 2017a. Profiles, sources and potential exposures of parent, chlorinated and brominated polycyclic aromatic hydrocarbons in haze associated atmosphere. Sci. Total Environ. 593, 390–398. Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air samplers for persistent organic pollutants. Environ. Sci. Technol. 36, 4142–4151. Harner, T., Shoeib, M., Diamond, M., Stern, G., Rosenberg, B., 2004. Using passive air samplers to assess urban-rural trends for persistent organic pollutants. 1. Polychlorinated biphenyls and organochlorine pesticides. Environ. Sci. Technol. 38, 4474–4483. Farrar, N.J., Harner, T., Shoeib, M., Sweetman, A., Jones, K.C., 2005. Field deployment of thin film passive air samplers for persistent organic pollutants: a study in the urban atmospheric boundary layer. Environ. Sci. Technol. 39, 42–48. Jaward, F.M., Farrar, N.J., Harner, T., Sweetman, A.J., Jones, K.C., 2004. Passive air sampling of PCBs, PBDEs, and organochlorine pesticides across Europe. Environ. Sci. Technol. 38, 34–41. Farrar, N.J., Prevedouros, K., Harner, T., Sweetman, A.J., Jones, K.C., 2006. Continental scale passive air sampling of persistent organic pollutants using rapidly equilibrating thin films (POGs). Environ. Pollut. 144, 423–433. Bogdal, C., Scheringer, M., Abad, E., Abalos, M., van Bavel, B., Hagberg, J., Fiedler, H., 2013. Worldwide distribution of persistent organic pollutants in air, including results
8
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Passive air sampling of halogenated polycyclic aromatic hydrocarbons in the largest industrial city in Korea: Spatial distributions and source identification
T
Quang Tran Vuonga, Seong-Joon Kima, Tuyet Nam Thi Nguyena, Phan Quang Thanga, ⁎ Sang-Jin Leea, Takeshi Ohurab, Sung-Deuk Choia, a b
School of Urban and Environmental Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan, 44919, Republic of Korea Faculty of Agriculture, Meijo University, Nagoya, 468-8502, Japan
G R A P H I C A L A B S T R A C T
A R T I C LE I N FO
A B S T R A C T
Editor: D. Aga
Some halogenated polycyclic aromatic hydrocarbons (Halo-PAHs) are known to be more toxic than their corresponding parent PAHs, but studies on Halo-PAHs have been somewhat limited. In this study, passive air samplers were used to monitor Halo-PAH and PAH contamination at 20 sampling sites in Ulsan, one of the largest industrial cities in South Korea. The mean concentrations of Σ24 ClPAHs, Σ11 BrPAHs, and Σ13 PAHs were 207 pg/m3, 84 pg/m3, and 26 ng/m3, respectively. Industrial areas displayed higher concentrations of both HaloPAHs and PAHs than urban and rural areas. Strong correlations between energetically unfavorable Halo-PAHs and their corresponding parent PAHs suggest that the main formation mechanism of Halo-PAHs is not direct halogenation of PAHs. Low molecular weight Halo-PAHs with one halogen atom and their parent PAHs were dominant. The profiles of ClPAHs and BrPAHs in petrochemical, automobile, shipbuilding, and non-ferrous industrial complexes were distinguished. The toxicity equivalency quantities (TEQs) of ClPAHs, BrPAHs, and PAHs at the industrial sites also showed the highest values of 4.2, 0.5, and 18.3 pg-TEQ/m3, respectively, reflecting the high toxicity of Halo-PAHs. To the best of our knowledge, this is the first study reporting atmospheric levels of both ClPAHs and BrPAHs using passive air samplers.
Keywords: PUF-PAS Halo-PAHs ClPAHs BrPAHs Ulsan
⁎
Corresponding author. E-mail address: [email protected] (S.-D. Choi).
https://doi.org/10.1016/j.jhazmat.2019.121238 Received 23 February 2019; Received in revised form 25 August 2019; Accepted 14 September 2019 Available online 17 September 2019 0304-3894/ © 2019 Elsevier B.V. All rights reserved.
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
1. Introduction
South Korea, with a population of more than 1.1 million inhabitants. The Ulsan and Mipo national industrial complex (46 km2) comprises automobile, shipbuilding, petrochemical, and petroleum refining industries, and the Onsan national industrial complex (20 km2) includes petroleum refining and non-ferrous metal industries. Industrial activities from the two national industrial complexes have a significant influence on PAH contamination in Ulsan (Choi et al., 2012; Vu et al., 2010; Lee and Lee, 2004; Dong and Lee, 2009). Halo-PAHs are also expected to be produced from the industrial activities in Ulsan, though no extensive studies on halo-PAHs have been performed in Ulsan yet. In our previous study (Choi et al., 2012), PUF-PASs were deployed in Ulsan, and the spatial distribution and source-receptor relationships of PAHs were investigated. Based on the results of those research efforts, we decided to further monitor Halo-PAHs as well as their corresponding parent PAHs using PUF-PASs. Their levels, profiles, spatial distributions, and potential sources were investigated.
Polycyclic aromatic hydrocarbons (PAHs) are ubiquitous environmental organic pollutants, which have carcinogenic and/or mutagenic properties (Abdel-Shafy and Mansour, 2016; Armstrong and Gibbs, 2009). They are mainly discharged into the atmosphere from biomass burning, vehicular exhausts, and industrial activities (Choi et al., 2012, 2007; Tang et al., 2005). The atmospheric behavior of PAHs is significantly affected by meteorological conditions (wind speed and direction, temperature, and precipitation) (Choi et al., 2012, 2007), photolysis, and reactive species (ozone and hydroxyl radicals) (Reisen and Arey, 2005; Tsapakis and Stephanou, 2005). PAH derivatives and their atmospheric behavior have recently become a more prominent environmental concern. Nitrated PAHs (NPAHs) are emitted directly from combustion processes and produced by secondary reactions of hydroxyl and NO3 radicals with gaseous PAHs (Reisen and Arey, 2005; Tsapakis and Stephanou, 2005). Oxygenated PAHs (OPAHs) are produced by combustion processes and heterogeneous reactions between ozone and particulate PAHs (Tsapakis and Stephanou, 2005). In addition, halogenated PAHs (Halo-PAHs), such as chlorinated PAHs (ClPAHs) and brominated PAHs (BrPAHs) have recently received an increasing amount of attention (Ohura et al., 2018). Incomplete combustion of organic materials with halogen sources such as hydrochloric acid can produce Halo-PAHs (Fujima et al., 2006). Furthermore, they are formed by direct chlorination of the parent PAHs during waste incineration and combustion of polyvinylchloride (PVC) (Fujima et al., 2006; Horii et al., 2008; Miyake et al., 2017). These substances can have adverse effects on human health (Ohura et al., 2007; Yoshino and Urano, 1997). The detection of Halo-PAHs in tap water, road tunnels, and soil has been reported for decades (Nilsson and Oestman, 1993; Ishaq et al., 2003). Nevertheless, research on atmospheric Halo-PAHs has only recently been published (Ohura et al., 2018, 2016; Kakimoto et al., 2014; Ma et al., 2013; Jin et al., 2017a). In these previous studies, only high-volume air samplers (HiVols) were used. Therefore, high-resolution spatial distributions of Halo-PAHs could not be well examined. Polyurethane foam-passive air samplers (PUF-PASs) have been used worldwide because they can be deployed at any site without electricity, where HiVols are impractical (Shoeib and Harner, 2002; Harner et al., 2004). They have been used in many studies on PAHs and persistent organic pollutants (POPs) that contain chlorines or bromines. For instance, the spatial distribution of PAHs in Ulsan, an industrial city, indicated that industrial complexes were the primary source of PAHs (Choi et al., 2012). The vertical distributions of polychlorinated biphenyls (PCBs), PAHs, and organochlorine pesticides (OCPs) were investigated in an urban city, and it was found that the levels of PAHs and PCBs sharply decreased with increasing elevation above the ground (Farrar et al., 2005). On a larger scale, the spatial distributions of PCBs, OCPs, polychlorinated naphthalene (PCNs), PAHs, lindane, hexachlorobenzene (HCB), and polybrominated biphenyl ethers (PBDEs) were studied using PUF-PASs in many European countries to illustrate the feasibility of mapping local, regional, and global sources of POPs (Jaward et al., 2004; Farrar et al., 2006). In addition, worldwide air monitoring of POPs revealed that PCB levels in Africa were relatively high due to the import of electronic waste from developed countries, and the usage of dichlorodiphenyl trichloroethane (DDT) to control malaria has caused high concentrations of DDT in Africa and the Pacific Islands (Bogdal et al., 2013; Harner et al., 2006). Regarding the derivatives of PAHs, PUF-PASs were applied to investigate the spatial distributions of NPAHs and OPAHs in four major urban cities in Nepal, which revealed that the secondary sources associated with photochemical reactions and burning of solid fuels and crop residue are the main sources of NPAHs and OPAHs, respectively (Yadav et al., 2018). However, the assessment of Halo-PAHs using PASs has yet to be addressed. Ulsan Metropolitan City is one of the largest industrial cities in
2. Materials and methods 2.1. Passive air sampling A total of 40 PUF-PASs were deployed in duplicate at 20 sites in Ulsan. The sampling sites were categorized into three classes: industrial (I1, I2, I3, and I4), rural (R1, R2, R3, R4, R5, and R6), and urban (U1, U2, U3, U4, U5, U6, U7, U8, U9, and U10) sites (Fig. 1). The sampling period was from March 8, 2013 to May 31, 2013 (84 days). Prior to deployment, PUF disks were sequentially sonicated with acetone and nhexane for 30 min, respectively. The deployed PUF disks were retrieved and stored in polyethylene zipper bags at −9 °C. 2.2. Instrumental analysis and QA/QC A full description of the instrumental procedure is provided in Text S1 in the Supplementary Material. In summary, retrieved PUF disks were spiked with surrogate standards prior to Soxhlet extraction with nhexane/acetone. The extracts were cleaned up with n-hexane/dichloromethane through silica gel columns. An internal standard was added to the vials prior to instrumental analysis. The target compounds were 24 ClPAHs, 11 BrPAHs, and 13 parent PAHs. Their abbreviations are listed in Table S1 in the Supplementary Material. The target compounds were analyzed using gas chromatography (GC)-electron impact mass spectrometry (MS) in the selected ion monitoring mode (Agilent 7890 GC-5975C MS, USA). The average recoveries were 52%, 98%, and 93% for phenanthrene-d10, chrysene-d12, and perylene-d12, respectively. Method detection limits (MDLs) for each Halo-PAH and PAH species are summarized in Table S2. 2.3. Calculation of air concentrations Ambient air concentrations (CAir) of Halo-PAHs and PAHs were calculated from their accumulated amounts on PUF disks (CPAS) and the passive sampling rate (Rs). In this study, the compound-specific sampling rates in seven sampling zones were estimated using a prediction model. A full description of the calculation method is provided in Text S2. 2.4. Meteorological conditions Meteorological data from seven automatic weather stations (AWSs) in Ulsan (Fig. 1) were obtained from the Korea Meteorological Administration (http://sts.kma.go.kr/). CALPro Plus 7 (TRC Environmental Corp., USA), an MS window version of CALPUFF, was applied to estimate the average wind-field during the sampling period (Fig. S1 in the Supplementary Material). For estimating the sampling rates of each Halo-PAH and PAH species, the 20 sampling sites were matched with 7 AWSs based on this wind pattern. A full description of the 2
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
Fig. 1. Locations of 20 sampling sites, 7 automatic weather stations, and 4 major industrial areas in Ulsan, South Korea.
Fig. 2. Mean concentrations of ClPAHs, BrPAHs (the left axis), and PAHs (the right axis) at 20 sites in Ulsan, Korea. The error bars represent standard deviations. Only concentrations of the detected compounds are plotted.
3. Results and discussion
meteorological conditions is provided in Text S3.
3.1. Levels of Halo-PAHs and PAHs 2.5. Toxicity evaluation The average sampling rates for all the target compounds ranged from 1.85 ± 0.13 m3/day in zone C to 3.69 ± 0.2 m3/day in zone G (Fig. S2). This result is in good agreement with those in previous studies (Chaemfa et al., 2008; He and Balasubramanian, 2010; Pozo et al., 2009; Herkert et al., 2018). The low sampling rate (∼2 m3/day) in zones A and C can be explained by low wind speeds due to the high forest coverage, whereas other zones on the coastline with higher wind speeds showed elevated sampling rates (∼3.5 m3/day) (Figs. S1 and S2). This trend was also observed in the former study (Herkert et al., 2018). In the previous studies, PUF-PASs have been reported to effectively collect gaseous compounds (Shoeib and Harner, 2002; Pozo et al., 2004). Particulate PAHs can also be collected on PUF-PASs to some degree (Harner et al., 2013; Markovic et al., 2015). However, the accumulation mechanism of the particulate pollutants on PUF disks has not been clearly understood, leading to the high uncertainty of the
Toxicity equivalency quantities (TEQs) for ClPAHs, BrPAHs, and PAHs were calculated based on the relative equivalency potencies (REPs). A full description of the calculation method is provided in Text S4.
2.6. Statistical analysis Normality tests, Pearson and Spearman correlation analyses, and rank sum tests were performed using SPSS statistics 22.0 (IBM, USA) and SigmaPlot 12.0 (Systat Software Inc., USA) to support data interpretation. The detailed information about the statistical analysis is provided in Text S5.
3
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
with Ant and Phe imply that ClPAHs can be closely related to their corresponding parent PAHs. However, the most energetically favorable ClPAHs to be formed from the corresponding parent PAHs (e.g., 2ClAnt) were not generated or generated with lower concentrations, whereas the most unfavorable ClPAHs (e.g., 9-ClAnt, 9,10-Cl2Ant, and 9-ClPhe) were found in high concentrations. A previous study reported that ClPAHs were mostly formed in a cooling zone of secondary copper smelters, not by the direct chlorination of PAHs during smelting processes (Jin et al., 2017b). The patterns of Halo-PAHs and PAHs at the emission source can be different from those at the receptor sites due to air dispersion and G/P partitioning behaviors. However, the industrial sites and several urban and rural sites in the present study are close to or inside of the industrial complexes; thus, the patterns of Halo-PAHs and PAHs at the source and receptor sites in Ulsan might be similar to each other. Consequently, it is unlikely that the major ClPAHs were produced by the direct chlorination of parent PAHs. However, it is hard to generalize the formation mechanism of ClPAHs because of not enough data. Due to the limited number of BrPAHs analyzed in this study, the eight detected BrPAHs could not be considered for Gibbs free energies. Among the target BrPAHs, only 2-BrFlu, 7-BrBaA, and 6-BrBaP were significantly correlated with their parent compounds, Flu (r = 0.67, p < 0.01), BaA (r = 0.49, p < 0.05), and BaP (r = 0.51, p < 0.05) (Table S8). In the previous study, BrPAHs were reported to be generated not from the direct bromination of the parent PAHs in iron ore sintering and secondary non-ferrous metal smelting (Xu et al., 2018). The main formation process of BrPAHs might be the secondary reactions between parent PAHs and other precursors in the atmosphere (Ni and Zeng, 2012). A similar formation mechanism has also been suggested for other environmental compartments, such as river sediment (Sun et al., 2011) and soil (Ni and Zeng, 2012). Overall, the exact formation mechanism of Halo-PAHs still remains unclear, and further investigation is required. The total mean concentration of Halo-PAHs in this study was compared with other monitoring data from different cities in China and Japan (Text S6, Table S9). Note that HiVols were used for all the previous studies, whereas PUF-PASs (for mostly gaseous compounds) were used in this study. In addition, characteristics of sampling sites, sampling periods, and target compounds were different from each other. Therefore, this comparison was used for a rough evaluation of the pollution level of Halo-PAHs. Overall, the levels of both ClPAHs and BrPAHs in Ulsan were found to be much higher than those in other cities, but the trend for parent PAHs was reversed. The reason for this discrepancy is unclear, but it seems to be related to heavy industrial activities in Ulsan. Therefore, further studies and long-term monitoring of Halo-PAHs are required.
particulate concentration calculation (Gouin et al., 2010). Therefore, all the following results and discussions in this study will focus mainly on the gaseous phase. Out of the 35 target Halo-PAHs, 16 ClPAHs and 8 BrPAHs were detected (Table S2, Fig. 2). The most abundant compounds were 9ClPhe and 2-BrFlu, which contributed 45% and 40% to the total mean ClPAH and BrPAH concentrations, respectively. The next major compounds were 9-ClAnt (19%) and Cl4Pyr (9%) for ClPAHs and 9-BrAnt (20%) and 9-BrPhe (16%) for BrPAHs. The low molecular weight compounds (LMWs) with one halogen atom were more dominant than the high molecular weight compounds (HMWs) with two or more halogen atoms. Similar trends of gaseous ClPAHs in the ambient air of Shizuoka, Japan have also been reported. LMW-PAHs tend to exist mostly in the gaseous phase, while HMW-PAHs are dominant in the particulate phase (Ohura et al., 2013, 2008). Thus, this observation of gaseous ClPAHs can be explained by the similar gas/particle (G/P) partitioning behavior between Halo-PAHs and their parent PAHs. However, the influence of the number of halogen atoms on the G/P partitioning of Halo-PAHs needs to be further studied (Ohura et al., 2008). For parent PAHs, all the target compounds except for DbahA were detected. Like its derivatives, Phe was dominant, accounting for 40% of the total PAHs, followed by Flt (17%) and Pyr (14%). These results are consistent with those reported in a previous study in Ulsan (Choi et al., 2012). The stability of PAHs in physical and chemical reactions greatly depends on their structural arrangements (Schwarzenbach et al., 2005). For example, linear or 2-ring PAHs can be rapidly photodegraded, whereas angular PAHs are resistant to sunlight (Abdel-Shafy and Mansour, 2016). Furthermore, gaseous PAHs are degraded easily and quickly by photolysis (Finlayson-Pitts and Pitts, 1997). Therefore, the relatively low concentrations of Flu (3.3 ng/m3) and Ant (0.5 ng/m3) compared with another parent PAHs can be explained by their high rates of degradation associated with the aforementioned processes. Moreover, their high reactivity could lead to large formations of their derivatives: 9-ClAnt (39 pg/m3), 2-BrFlu (34 pg/m3), and 9-BrAnt (17 pg/m3) (Fig. 2). Previous studies have shown that direct chlorination of parent PAHs might be a major formation pathway of their daughter Halo-PAH products. For example, it has been reported that monochlorinated derivatives of Flu and Ant were predominantly generated from combustion of polyvinylchloride at lower temperatures compared to di- or tri-homologue (Wang et al., 2003). A study on ClPAH formation in municipal waste incinerators (MSWIs) suggested that these compounds are mainly formed by the chlorination of parent PAHs with Cl2 in the exhaust gas rather than in the incineration furnaces (Yoshino and Urano, 1997). According to the relative Gibbs free energies and the ClPAH formation from waste incineration, monochlorinated PAHs, 2-ClAnt and 3ClPhe, are the most favorable derivatives generated from direct chlorination of the corresponding parent PAHs (Wang et al., 2003; Jin et al., 2017b). However, 2-ClAnt was not detected in any samples in the present study. Instead, 9-ClAnt (19%) was one of the most abundant congeners in ClPAH profiles (Fig. 2), and its direct formation from Ant requires the highest energy level (Jin et al., 2017b). Moreover, 9-ClAnt displayed a strong positive correlation with its corresponding parent PAH (r = 0.67, p < 0.01) (Table S7). In addition, 9,10-Cl2Ant, which requires significantly more energy for the direct chlorination of Ant than the others in di-homologue (Jin et al., 2017b), was also detected (Fig. 2). 9-ClPhe, which requires the highest energy for its direct formation from Phe (Jin et al., 2017b), was the most dominant species in the profile of ClPAHs, and it was significantly correlated with Phe (r = 0.74, p < 0.01), whereas the other Phe derivatives displayed no correlation with Phe (Table S7). For other ClPAHs, even though the mono-derivatives were well correlated with their corresponding parent PAHs (Table S7), their contribution to the total mean concentration of ClPAHs (12%) was much lower than those of 9-ClAnt and 9-ClPhe (19% and 45%, respectively). The strong correlations of 9-ClAnt and 9-ClPhe
3.2. Spatial distributions of Halo-PAHs and PAHs This study is the first one that reports levels and patterns of HaloPAHs in Ulsan, and there is no information or data about their sources in the city. Therefore, the source-receptor relationships for Halo-PAHs were evaluated based on their spatial distributions. The industrial area, as expected, generally displayed higher concentrations of both HaloPAHs and PAHs than the urban and rural areas (Fig. 3). According to the correlation analysis, the concentration of Σ24 ClPAHs and Σ11 BrPAHs at the 20 sampling sites were significantly correlated (r = 0.60, p < 0.01), implying that these two types of Halo-PAHs have similar emission sources and/or formation mechanisms. A stronger correlation was found between Σ24 ClPAHs and Σ13 PAHs (r = 0.71, p < 0.01) (Table S7) than between Σ11 BrPAHs and Σ13 PAHs (r = 0.37, p > 0.05) (Table S8), which suggests that the levels of ClPAHs could be influenced more by their parent compounds. This finding is in line with the relationship between Halo-PAHs and PAHs in a previous study (Ohura et al., 2009). For ClPAHs, the industrial sites showed higher levels 4
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
Fig. 3. Spatial distributions of Σ24 ClPAHs, Σ11 BrPAHs, and Σ13 PAHs in Ulsan, Korea (March 8–May 31, 2013). Three contour maps for each PAH group were drawn using ArcGIS 10.5 (ESRI Inc., USA). Inverse distance weighting (IDW) was employed as an interpolation method.
(609 ± 440 pg/m3) than rural (308 ± 348 pg/m3) and urban (147 ± 92 pg/m3) sites. No statistical difference was found between the urban and rural sites (Mann-Whitney rank sum test, p > 0.05) due to the similar concentrations of ClPAHs at the urban and rural sites, which were located far from the industrial complexes. By contrast, the industrial sites showed significant statistical differences when compared to the urban and rural sites (Mann-Whitney rank sum test, p < 0.01). For BrPAHs, the order of the mean concentrations was different from those of ClPAHs and PAHs: the industrial (204 ± 86 pg/ m3) > urban (119 ± 171 pg/m3) > rural (85 ± 43 pg/m3) sites. There was no statistical difference among them (Mann-Whitney rank sum test, p > 0.05). For PAHs, the mean concentration at the industrial sites (40 ± 11 ng/m3) was higher than those detected at the rural (31 ± 24 ng/m3) and urban (18 ± 11 ng/m3) sites. However, there was no statistical difference among them (Mann-Whitney rank sum test, p > 0.05) because some urban and rural sites near the industrial complexes also displayed relatively high concentrations. A similar spatial distribution for PAHs and statistical results were also reported in our previous study (Choi et al., 2012). The same orders (I > R > U) of PAHs and ClPAHs can also be expected by a strong correlation between them (r = 0.71, p < 0.01) (Table S7), whereas
BrPAHs had a weak correlation of 0.37 (p > 0.05) with PAHs (Table S8) and exhibited a different order of mean concentrations (I > U > R). Therefore, there might be some common sources of ClPAHs and parent PAHs, which are not the potential sources of BrPAHs. These different spatial distributions (or correlation results) warrant further investigation to better understand their source-receptor relationships. Among the rural sites, R1 and R2 showed significantly high concentrations of both ClPAHs and PAHs because they are close to industrial complexes (Fig. 3). Site R1 is located between the automobile industrial complex and the shipbuilding and heavy industrial complex, and sites R2 and I3 are located on the edge of the petrochemical industrial complex. These three sites could be further influenced by south-southwesterly winds blown from the non-ferrous industrial complex (Fig. S1), which is supported by the high concentrations of ClPAHs and PAHs detected at site I4. High concentrations of BrPAHs were observed at site I1 in the automobile industrial complex and I2 and I3 in the petrochemical industrial complex. The two urban sites, U1 and U2, which are relatively close to the industrial complexes, displayed the highest concentrations of BrPAHs, with site U2 potentially being significantly affected by westerly and southwesterly winds transporting BrPAHs from the automobile and petrochemical industrial 5
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
Fig. 4. Levels and fractions of Halo-PAHs and PAHs classified by the number of rings and halogen atoms at the individual sampling sites.
species at R5 (63%) and R6 (65%), other compounds were accounted for quite equally (6%), and 1-ClPyr contributed very small proportions at R5 and R6 (3% and 0%, respectively) (Fig. S3a and b). Moreover, the profile of ClPAHs at site I4 with 9-ClPhe (27%), 9-ClAnt (12%), 3-ClFlt (4%), and 1-ClPyr (26%) (Fig. S3b) was similar to the profile of gaseous ClPAHs found in secondary copper smelters (Jin et al., 2017b). Hence, the four ClPAHs seem to be indicative of non-ferrous industrial operations. Regarding BrPAH profiles, the group of Br-3 rings accounted for 96% of Σ11 BrPAHs at site I1 (Fig. 4d). Particularly, the fraction of 2BrFlu (75%) at site I1 was higher than those at the other three industrial sites (Fig. S3d). This pattern might affect the high contribution of 2BrFlu (70%) at site U2, which is likely due to the influence of the northwesterly winds. Thus, 2-BrFlu could be a typical marker for the automobile industry. The group of Br-3 rings accounted for the majority at virtually every site; however, the Br2-4 ring group was the most abundant at site I2 (68%) and comprised 39% of the total BrPAH concentration at site U6 (Fig. 4c and d), while the concentration of Br24 rings (4,7-Br2BaA and 7,12-Br2BaA) was the second highest at site U2 (62 pg/m3, 11%) (Fig. S3c and d). These three sites are close to each other, and southwesterly and easterly winds might increase the influence of the petrochemical industries on the BrPAH profiles at sites U6 and U2. Given the absence of Br2-3 rings at all the sampling sites, the profiles of BrPAHs in Ulsan were quite different from the gaseous profiles in Beijing (Jin et al., 2017a). This difference might be partly explained by different sampling conditions and environmental factors, but the reasons remain unclear due to limited information. Although it was reported that BrPAHs might not share the same potential sources of ClPAHs in soil (Ni and Zeng, 2012), the information on the main sources and the formation mechanism of BrPAHs is still limited. For the profiles of PAHs, 3 and 4 rings accounted for the majority of
complexes. In general, the sampling sites close to the industrial complexes showed relatively high concentrations of PAHs. Furthermore, urban site U3, located on the western side of a medium-industrial belt (i.e., Jungsan and Maegok general industrial complexes, mainly producing auto parts for vehicles), showed relatively high concentrations of PAHs because of heavy-duty vehicle traffic in addition to the nearby industrial activities. 3.3. Profiles of Halo-PAHs and PAHs Levels and fractions for each group categorized by the number of halogen atoms and rings at each sampling site are presented in Fig. 4. For ClPAHs, there was a noticeable difference among the four industrial sites. The average fraction of the first group (Cl-3 rings) at sites I1 and I2 was 68%, which is twice that found at sites I3 and I4 (30%) (Fig. 4a and b). Similar profiles were observed at sites U2 and R2, which are located in the vicinity of I1 and I2 under the influence of northwesterly winds during the sampling period. Particularly, two compounds (9ClPhe and 9-ClAnt) accounted for 67% and 69% of the Σ24 ClPAHs at sites I1 and I2, respectively (Fig. S3a and b). This pattern was consistent with the findings in a multi-industrial area in Tokyo Bay, Japan (Ohura et al., 2018). Therefore, those species could be considered as markers associated with the automobile and shipbuilding industries. Cl-4 rings accounted for 42% and 30% of the Σ24 ClPAHs at sites I3 and I4, respectively (Fig. 4a and b). More specifically, site I4 had the highest fractions of 3-ClFlt (4%) and 1-ClPyr (26%) among the sites, whereas site I3 had the highest proportion of 6-ClChr (30%) (Fig. S3a and b). The profiles of ClPAHs at sites U1 (4% of 3-ClFlt) and R1 (5% of 1ClPyr) might be also influenced by the high fractions of those two compounds at site I4 combined with the influence of the south-southwesterly winds (Fig. S1). Even though 9-ClPhe was the most abundant 6
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
(18.29 pg-TEQ/m3) was higher than those at the urban (12.98 pg-TEQ/ m3) and rural (14.94 pg-TEQ/m3) sites (Fig. 5). There was no statistical difference (Mann-Whitney rank sum test, p > 0.05) among them because several urban and rural sites, located in the vicinity of the industrial complexes, also showed relatively high TEQ values. The total TEQ value in Ulsan (15 pg-TEQ/m3) was comparable with that in Shizuoka, Japan (Table S10) (Ohura et al., 2009). Meanwhile, the total PAH TEQs in Ulsan was three times those in Busan, South Korea and Tokyo, Japan, and one-third of those in Nagoya, Japan and Beijing, China (Ohura et al., 2018, 2016; Kakimoto et al., 2014; Jin et al., 2017a). According to the REPBaP values for individual compounds (Table S11), virtually all Halo-PAH species were more toxic than their corresponding parent PAHs. Particularly, 6-ClChr is 233 times more toxic than its parent compound. Furthermore, for 3- and 4-ring ClPAHs, the toxicity significantly increases with the number of halogen substitutions. The REPBaP values of 3,9-Cl2Phe and 3,8-Cl2Flt are about 80- and 1,900-fold higher than those of Phe and Flt, respectively. The reason for this toxicological trend of Halo-PAHs is still unclear; however, AhR activity, CYP1A1 mRNA expression in MCF-7 cells, and Ames mutagenicity on bacteria strains (TA98 and TA100) for Halo-PAHs and parent PAHs were confirmed (Ohura et al., 2007; Yoshino and Urano, 1997). Hence, the toxicological mechanism of Halo-PAHs based on the number and position of the substituted halogen atoms deserves further attention.
Fig. 5. Spatial distributions of the total toxicity equivalency quantities (TEQs) of Halo-PAHs and PAHs at each site.
PAH species at every sampling site (about 86%) (Fig. 4e and f). In particular, Phe (40%), Flt (17%), Pyr (14%), and Flu (13%) showed higher mean fractions (Fig. S3e and f), which is consistent with the previously reported profiles for gaseous PAHs in Ulsan (Choi et al., 2012; Nguyen et al., 2018). In these former studies, contamination characterization and source identification of parent PAHs were accomplished using statistical tools (e.g., principal component analysis and positive matrix factorization) and diagnostic ratios. Therefore, additional interpretation and further discussions have not been included in this study.
4. Conclusion The air pollution by ClPAHs, BrPAHs, and PAHs in the multi-industrial city of Ulsan was investigated using the passive sampling technique. Overall, Ulsan, especially the industrial complexes, showed higher concentrations of both ClPAHs and BrPAHs than other cities. Major wind patterns played a vital role in the dispersion of Halo-PAHs from the industrial zones to the surrounding urban and rural areas. According to the spatial distributions and correlation analysis, the total concentrations of Halo-PAHs and parent PAHs were significantly correlated, suggesting they have similar potential sources. However, the most energetically unfavorable Halo-PAHs exhibited strong correlations with their corresponding parent compounds, indicating that the direct halogenation of PAHs is not a major formation pathway of individual Halo-PAHs. In this regard, more in-depth investigations on the formation mechanisms and identification of major anthropogenic sources are necessary. Despite the comparable TEQs of PAHs with those of other cities, Ulsan showed significantly high TEQs of both ClPAHs and BrPAHs because of the great contributions of the most toxic species, possibly linked to industrial activities. On the basis of this first passive sampling study on Halo-PAHs, we are conducting seasonal monitoring using PUF-PASs and HiVols for more understanding of the contamination characteristics of Halo-PAHs.
3.4. Toxicity evaluation of Halo-PAHs and PAHs Total TEQs of 11 ClPAHs, 8 BrPAHs, and 11 PAHs at the individual sampling site are presented in Fig. 5. The average TEQ of ClPAHs at the industrial sites (4.21 pg-TEQ/m3) was significantly higher than those at the urban (0.72 pg-TEQ/m3, Mann-Whitney rank sum test, p < 0.01) and rural sites (0.42 pg-TEQ/m3, Mann-Whitney rank sum test, p < 0.01). The highest ClPAH TEQ at site I3 (13 pg-TEQ/m3) was due to the most abundant proportion (30%) of the second most toxic ClPAH (6-ClChr, REPBaP = 2.1). Sites I4, R1, and R2 also showed higher concentrations of ClPAHs; however, 3,8-Cl2Flt (REPBaP = 5.7) and 6-ClChr only accounted for small fractions (< 2%) in their profiles. For BrPAHs, the industrial sites (0.53 pg-TEQ/m3) showed a higher average TEQ than the urban (0.16 pg-TEQ/m3) and rural sites (0.10 pg-TEQ/m3); however, there was no statistical difference among them (MannWhitney rank sum test, p > 0.05). Sites I2 (2 pg-TEQ/m3) and U2 (1 pg-TEQ/m3) showed the highest BrPAH TEQs due to the major contributions from 4,7-Br2BaA, the second most toxic BrPAHs (REPBaP = 0.77), in their profiles (I2: 38% and U2: 10%). In general, the total TEQs of both ClPAHs and BrPAHs in this study were found to be substantially higher than those in other Northeast Asian cities (Table S10). For example, the total TEQ of ClPAHs in Ulsan was 31 times higher than that in Tokyo, Japan (Ohura et al., 2018), 14 times higher than the neighboring Korean city of Busan (Kakimoto et al., 2014), and 9 times higher than that in Beijing, China (Jin et al., 2017a). For BrPAHs, the total TEQ in Ulsan was about fourfold higher than that in Shizuoka, Japan (Ohura et al., 2009), and 121-fold higher than that in Beijing, China (Jin et al., 2017a). Despite limited reports, the comparison data suggest that Ulsan is more contaminated by Halo-PAHs than other cities, and further studies on source identification are essential. Regarding parent PAHs, the average TEQ at the industrial sites
Acknowledgements The research was supported by the 2019 Research Fund (1.190011.01) of UNIST and the National Research Foundation of Korea (NRF) (NRF-2017R1A2B4003229 and 2017M3D8A1092015). We are thankful for the discussions with Dr. Tom Harner for estimating KOA values of Halo-PAHs and calculation of ambient air concentrations. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jhazmat.2019.121238. References Abdel-Shafy, H.I., Mansour, M.S.M., 2016. A review on polycyclic aromatic
7
Journal of Hazardous Materials 382 (2020) 121238
Q.T. Vuong, et al.
of air monitoring by passive air sampling in five continents. Trends Analyt. Chem. 46, 150–161. Harner, T., Pozo, K., Gouin, T., Macdonald, A.-M., Hung, H., Cainey, J., Peters, A., 2006. Global pilot study for persistent organic pollutants (POPs) using PUF disk passive air samplers. Environ. Pollut. 144, 445–452. Yadav, I.C., Devi, N.L., Singh, V.K., Li, J., Zhang, G., 2018. Concentrations, sources and health risk of nitrated- and oxygenated-polycyclic aromatic hydrocarbon in urban indoor air and dust from four cities of Nepal. Sci. Total Environ. 643, 1013–1023. Vu, V.-T., Lee, B.-K., Ny, M.T., Kim, J.-T., 2010. A Study on Characteristics of Organic Carbon and Polycyclic Aromatic Hydrocarbons (PAHs) in PM10 at the Residential and Industrial Areas in Ulsan of Korea. International Forum on Strategic Technology, pp. 263–266. Lee, B.-K., Lee, C.-B., 2004. Development of an improved dry and wet deposition collector and the atmospheric deposition of PAHs onto Ulsan Bay, Korea. Atmos. Environ. 38, 863–871. Dong, T.T.T., Lee, B.-K., 2009. Characteristics, toxicity, and source apportionment of polycylic aromatic hydrocarbons (PAHs) in road dust of Ulsan, Korea. Chemosphere 74, 1245–1253. Chaemfa, C., Barber, J.L., Gocht, T., Harner, T., Holoubek, I., Klanova, J., Jones, K.C., 2008. Field calibration of polyurethane foam (PUF) disk passive air samplers for PCBs and OC pesticides. Environ. Pollut. 156, 1290–1297. He, J., Balasubramanian, R., 2010. A comparative evaluation of passive and active samplers for measurements of gaseous semi-volatile organic compounds in the tropical atmosphere. Atmos. Environ. 44, 884–891. Pozo, K., Harner, T., Lee, S.C., Wania, F., Muir, D.C.G., Jones, K.C., 2009. Seasonally resolved concentrations of persistent organic pollutants in the global atmosphere from the first year of the GAPS study. Environ. Sci. Technol. 43, 796–803. Herkert, N.J., Spak, Scott N., Smith, A., Schuster, J.K., Harner, T., Martinez, A., Hornbuckle, K.C., 2018. Calibration and evaluation of PUF-PAS sampling rates across the Global Atmospheric Passive Sampling (GAPS) network. Environ. Sci.: Processes Impacts. 20, 210–219. Pozo, K., Harner, T., Shoeib, M., Urrutia, R., Barra, R., Parra, O., Focardi, S., 2004. Passive-sampler derived air concentrations of persistent organic pollutants on a north-south transect in Chile. Environ. Sci. Technol. 38, 6529–6537. Harner, T., Su, K., Genualdi, S., Karpowicz, J., Ahrens, L., Mihele, C., Schuster, J., Charland, J.-P., Narayan, J., 2013. Calibration and application of PUF disk passive air samplers for tracking polycyclic aromatic compounds (PACs). Atmos. Environ. 75, 123–128. Markovic, M.Z., Prokop, S., Staebler, R.M., Liggio, J., Harner, T., 2015. Evaluation of the particle infiltration efficiency of three passive samplers and the PS-1 active air sampler. Atmos. Environ. 112, 289–293. Gouin, T., Wilkinson, D., Hummel, S., Meyer, B., Culley, A., 2010. Polycyclic aromatic hydrocarbons in air and snow from Fairbanks, Alaska. Atmos. Pollut. Res. 1, 9–15. Ohura, T., Horii, Y., Kojima, M., Kamiya, Y., 2013. Diurnal variability of chlorinated polycyclic aromatic hydrocarbons in urban air, Japan. Atmos. Environ. 81, 84–91. Ohura, T., Fujima, S., Amagai, T., Shinomiya, M., 2008. Chlorinated polycyclic aromatic hydrocarbons in the atmosphere: seasonal levels, gas-particle partitioning, and origin. Environ. Sci. Technol. 42, 3296–3302. Schwarzenbach, R.P., Gschwend, P.M., Imboden, D.M., 2005. Environmental Organic Chemistry. John Wiley & Sons, Inc., Hoboken, New Jersey, USA. Finlayson-Pitts, B.J., Pitts Jr., J.N., 1997. Tropospheric air pollution: ozone, airborne toxics, polycyclic aromatic hydrocarbons, and particles. Science 276, 1045–1052. Wang, D., Xu, X., Chu, S., Zhang, D., 2003. Analysis and structure prediction of chlorinated polycyclic aromatic hydrocarbons released from combustion of polyvinylchloride. Chemosphere 53, 495–503. Jin, R., Liu, G., Zheng, M., Jiang, X., Zhao, Y., Yang, L., Wu, X., Xu, Y., 2017b. Secondary copper smelters as sources of chlorinated and brominated polycyclic aromatic hydrocarbons. Environ. Sci. Technol. 51, 7945–7953. Xu, Y., Yang, L., Zheng, M., Jin, R., Wu, X., Li, C., Liu, G., 2018. Chlorinated and brominated polycyclic aromatic hydrocarbons from metallurgical plants. Environ. Sci. Technol. 52, 7334–7342. Ni, H.-G., Zeng, E.Y., 2012. Environmental and human exposure to soil chlorinated and brominated polycyclic aromatic hydrocarbons in an urbanized region. Environ. Toxicol. Chem. 31, 1494–1500. Sun, J.-L., Ni, H.-G., Zeng, H., 2011. Occurrence of chlorinated and brominated polycyclic aromatic hydrocarbons in surface sediments in Shenzhen, South China and its relationship to urbanization. J. Environ. Monit. 13, 2775–2781. Ohura, T., Sawada, K., Amagai, T., Shinomiya, M., 2009. Discovery of novel halogenated polycyclic aromatic hydrocarbons in urban particulate matters: occurrence, photostability, and AhR activity. Environ. Sci. Technol. 43, 2269–2275. Nguyen, T.N.T., Jung, K.-S., Son, J.M., Kwon, H.-O., Choi, S.-D., 2018. Seasonal variation, phase distribution, and source identification of atmospheric polycyclic aromatic hydrocarbons at a semi-rural site in Ulsan, South Korea. Environ. Pollut. 236, 529–539.
hydrocarbons: source, environmental impact, effect on human health and remediation. Egypt. J. Pet. 25, 107–123. Armstrong, B.G., Gibbs, G., 2009. Exposure–response relationship between lung cancer and polycyclic aromatic hydrocarbons (PAHs). Occup. Environ. Med. 66, 740–746. Choi, S.-D., Baek, S.-Y., Chang, Y.-S., 2007. Influence of a large steel complex on the spatial distribution of volatile polycyclic aromatic hydrocarbons (PAHs) determined by passive air sampling using membrane-enclosed copolymer (MECOP). Atmos. Environ. 41, 6255–6264. Choi, S.-D., Kwon, H.-O., Lee, Y.-S., Park, E.-J., Oh, J.-Y., 2012. Improving the spatial resolution of atmospheric polycyclic aromatic hydrocarbons using passive air samplers in a multi-industrial city. J. Hazard. Mater. 241–242, 252–258. Tang, N., Hattori, T., Taga, R., Igarashi, K., Yang, X., Tamura, K., Kakimoto, H., Mishukov, V.F., Toriba, A., Kizu, R., Hayakawa, K., 2005. Polycyclic aromatic hydrocarbons and nitropolycyclic aromatic hydrocarbons in urban air particulates and their relationship to emission sources in the Pan-Japan Sea countries. Atmos. Environ. 39, 5817–5826. Reisen, F., Arey, J., 2005. Atmospheric reactions influence seasonal PAH and Nitro-PAH concentrations in the Los Angeles basin. Environ. Sci. Technol. 39, 64–73. Tsapakis, M., Stephanou, E.G., 2005. Polycyclic aromatic hydrocarbons in the atmosphere of the Eastern Mediterranean. Environ. Sci. Technol. 39, 6584–6590. Ohura, T., Horii, Y., Yamashita, N., 2018. Spatial distribution and exposure risks of ambient chlorinated polycyclic aromatic hydrocarbons in Tokyo Bay area and network approach to source impacts. Environ. Pollut. 232, 367–374. Fujima, S., Ohura, T., Amagai, T., 2006. Simultaneous determination of gaseous and particulate chlorinated polycyclic aromatic hydrocarbons in emissions from the scorching of polyvinylidene chloride film. Chemosphere 65, 1983–1989. Horii, Y., Ok, G., Ohura, T., Kannan, K., 2008. Occurrence and profiles of chlorinated and brominated polycyclic aromatic hydrocarbons in waste incinerators. Environ. Sci. Technol. 42, 1904–1909. Miyake, Y., Tokumura, M., Wang, Q., Amagai, T., Horii, Y., Kannan, K., 2017. Mechanism of formation of chlorinated pyrene during combustion of polyvinyl chloride. Environ. Sci. Technol. 51, 14100–14106. Ohura, T., Morita, M., Makino, M., Amagai, T., Shimoi, K., 2007. Aryl hydrocarbon receptor-mediated effects of chlorinated polycyclic aromatic hydrocarbons. Chem. Res. Toxicol. 20, 1237–1241. Yoshino, H., Urano, K., 1997. Formation of chlorinated PAHs in exhaust gas from municipal waste incinerators, and their mutagenic activities. Toxicol. Environ. Chem. 63, 233–246. Nilsson, U.L., Oestman, C.E., 1993. Chlorinated polycyclic aromatic hydrocarbons: method of analysis and their occurrence in urban air. Environ. Sci. Technol. 27, 1826–1831. Ishaq, R., Näf, C., Zebühr, Y., Broman, D., Järnberg, U., 2003. PCBs, PCNs, PCDD/Fs, PAHs and Cl-PAHs in air and water particulate samples-patterns and variations. Chemosphere 50, 1131–1150. Ohura, T., Kamiya, Y., Ikemori, F., 2016. Local and seasonal variations in concentrations of chlorinated polycyclic aromatic hydrocarbons associated with particles in a Japanese megacity. J. Hazard. Mater. 312, 254–261. Kakimoto, K., Nagayoshi, H., Konishi, Y., Kajimura, K., Ohura, T., Hayakawa, K., Toriba, A., 2014. Atmospheric chlorinated polycyclic aromatic hydrocarbons in East Asia. Chemosphere 111, 40–46. Ma, J., Chen, Z., Wu, M., Feng, J., Horii, Y., Ohura, T., Kannan, K., 2013. Airborne PM2.5/ PM10-associated chlorinated polycyclic aromatic hydrocarbons and their parent compounds in a suburban area in Shanghai, China. Environ. Sci. Technol. 47, 7615–7623. Jin, R., Liu, G., Jiang, X., Liang, Y., Fiedler, H., Yang, L., Zhu, Q., Xu, Y., Gao, L., Su, G., Xiao, K., Zheng, M., 2017a. Profiles, sources and potential exposures of parent, chlorinated and brominated polycyclic aromatic hydrocarbons in haze associated atmosphere. Sci. Total Environ. 593, 390–398. Shoeib, M., Harner, T., 2002. Characterization and comparison of three passive air samplers for persistent organic pollutants. Environ. Sci. Technol. 36, 4142–4151. Harner, T., Shoeib, M., Diamond, M., Stern, G., Rosenberg, B., 2004. Using passive air samplers to assess urban-rural trends for persistent organic pollutants. 1. Polychlorinated biphenyls and organochlorine pesticides. Environ. Sci. Technol. 38, 4474–4483. Farrar, N.J., Harner, T., Shoeib, M., Sweetman, A., Jones, K.C., 2005. Field deployment of thin film passive air samplers for persistent organic pollutants: a study in the urban atmospheric boundary layer. Environ. Sci. Technol. 39, 42–48. Jaward, F.M., Farrar, N.J., Harner, T., Sweetman, A.J., Jones, K.C., 2004. Passive air sampling of PCBs, PBDEs, and organochlorine pesticides across Europe. Environ. Sci. Technol. 38, 34–41. Farrar, N.J., Prevedouros, K., Harner, T., Sweetman, A.J., Jones, K.C., 2006. Continental scale passive air sampling of persistent organic pollutants using rapidly equilibrating thin films (POGs). Environ. Pollut. 144, 423–433. Bogdal, C., Scheringer, M., Abad, E., Abalos, M., van Bavel, B., Hagberg, J., Fiedler, H., 2013. Worldwide distribution of persistent organic pollutants in air, including results
8