Article pubs.acs.org/est
Polychlorinated Biphenyls (PCBs) in Human Hair and Serum from E‑Waste Recycling Workers in Southern China: Concentrations, Chiral Signatures, Correlations, and Source Identification Jing Zheng,† Le-Huan Yu,†,‡ She-Jun Chen,*,‡ Guo-Cheng Hu,† Ke-Hui Chen,‡,§ Xiao Yan,∥ Xiao-Jun Luo,‡ Sukun Zhang,† Yun-Jiang Yu,*,† Zhong-Yi Yang,∥ and Bi-Xian Mai‡
Downloaded via EASTERN KENTUCKY UNIV on January 10, 2019 at 13:06:29 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
†
Center for Environmental Health Research, South China Institute of Environmental Sciences, Ministry of Environmental Protection, Guangzhou 510655, China ‡ State Key Laboratory of Organic Geochemistry and Guangdong Key Laboratory of Environmental Protection and Resources Utilization, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, China § University of Chinese Academy of Sciences, Beijing 100049, China ∥ State Key Laboratory for Biocontrol, School of Life Sciences, Sun Yat-Sen University, Guangzhou 510275, China S Supporting Information *
ABSTRACT: Hair is increasingly used as a biomarker for human exposure to persistent organic pollutants (POPs). However, the internal and external sources of hair POPs remain a controversial issue. This study analyzed polychlorinated biphenyls (PCBs) in human hair and serum from electronic waste recycling workers. The median concentrations were 894 ng/g and 2868 ng/g lipid in hair and serum, respectively. The PCB concentrations in male and female serum were similar, while concentrations in male hair were significantly lower than in female hair. Significant correlations between the hair and serum PCB levels and congener profiles suggest that air is the predominant PCB source in hair and that hair and blood PCB levels are largely dependent on recent accumulation. The PCB95, 132, and 183 chiral signatures in serum were significantly nonracemic, with mean enantiomer fractions (EFs) of 0.440−0.693. Nevertheless, the hair EFs were essentially racemic (mean EFs = 0.495−0.503). Source apportionment using the Chemical Mass Balance model also indicated primary external PCB sources in human hair from the study area. Air, blood, and indoor dust are responsible for, on average, 64.2%, 27.2%, and 8.79% of the hair PCBs, respectively. This study evidenced that hair is a reliable matrix for monitoring human POP exposure.
■
INTRODUCTION
p,p′-DDE correlation, moderate PCB congener correlations, and no correlations for other organochlorines between hair and blood samples. They suggest that organochlorines may originate from different sources.12 These inconsistent observations highlight the importance of identifying internal and external sources of contaminants and the correlations between hair and endogenous tissues. Polychlorinated biphenyls (PCBs) are a well-known class of man-made organic chemicals, which were formerly used in a variety of industrial and commercial applications. They have received significant interest due to their persistent, bioaccumulative, and toxic properties.13 In contrast to the decline of PCB levels in developed countries after their restrictions in the 1970s, high levels have been reported at electrical and electronic waste (e-waste) recycling sites in many developing countries.14 Continuous long-term exposure monitoring efforts are necessary to protect populations living near e-waste sites,
Although blood, breast milk, and adipose have been extensively used for monitoring human exposure to a variety of environmental contaminants, there are increasing studies using hair as an additional biomarker for exposure to persistent organic pollutants (POPs).1−5 Compared to other matrices, hair is easily and noninvasively collected, low-cost, and easily transported and stored for analysis. In addition, hair analysis can reveal both short and long-term exposure.6,7 Each hair follicle root is surrounded by numerous capillary blood vessels that supply substances to the hair.8 Some studies have suggested that hair serves as an indicator of POP levels of endogenous tissues. 7 Poon et al. reported significant correlations (r = 0.345−0.566) between human hair and serum polybrominated diphenyl ether (PBDE) concentrations.9 Hair is enriched with lipids (3.5−4%) and has a tendency to absorb lipophilic organic contaminants from external origins (e.g., from ambient air or dust). Neuber et al. (1999) used preschool children’s hair as a passive sampler to monitor indoor pesticide contamination.10 In addition, positive relationships between PBDE concentrations in indoor dust and human hair have been recently observed.11 Altshul et al. observed a strong © 2016 American Chemical Society
Received: Revised: Accepted: Published: 1579
October 9, 2015 January 5, 2016 January 12, 2016 January 12, 2016 DOI: 10.1021/acs.est.5b04955 Environ. Sci. Technol. 2016, 50, 1579−1586
a
Not detectable.
PCB28 PCB52 PCB66 PCB74 PCB95 PCB99 PCB101 PCB105 PCB118 PCB128 PCB138 PCB153 PCB164 PCB170/190 PCB177 PCB180 PCB187 PCB209 Total
129 49.9 42.9 27.4 50.1 31.4 73.2 23.0 48.4 8.40 20.4 28.1 8.48 3.09 1.88 5.87 3.18 0.27 611
median
147 65.5 46.8 29.5 64.1 31.9 83.4 38.2 75.9 11.6 27.5 33.6 10.2 4.15 2.30 6.23 3.67 0.27 682
mean
male hair range 32.4−300 8.83−168 12.0−107 7.58−64.6 13.1−160 7.42−111 17.0−274 6.14−191 12.8−349 1.92−58.9 5.59−125 8.65−146 1.91−43.2 0.64−13.1 0.47−6.68 1.88−14.3 1.02−8.54 n.d.-1.46 161−2020
346 192 117 70.8 189 73.8 239 85.4 174 32.3 77.7 98.7 31.0 12.8 6.42 15.8 8.99 1.55 1781
median 281 162 103 61.8 167 73.7 220 100 206 36.7 82.5 101 32.2 13.6 6.01 17.1 9.14 2.02 1675
mean
female hair range 72.1−479 36.4−288 17.9−190 11.8−114 29.5−294 11.5−184 30.8−481 13.4−315 26.1−582 4.05−110 12.0−228 14.9−255 5.06−78.5 3.09−33.6 1.25−11.3 4.86−35.9 2.65−17.1 n.d.-6.28 314−3514
176 n.d. 160 286 141 223 35.9 125 384 57.6 301 341 90.2 78.9 15.7 105 51.8 34.1 2851
median
median 537 n.d. 273 317 97.0 159 40.1 206 561 58.5 276 289 47.3 49.1 13.7 71.0 42.6 38.6 2888
range n.d.a−724 n.d.−406 n.d.−799 23.6−299 n.d.−904 n.d.−261 n.d.−928 29.3−2353 n.d.−163 n.d.−957 n.d.−997 n.d.−283 12.6−210 n.d.−75.5 n.d.−376 n.d.−164 n.d.−81.8 315−8976
240 162 330 137 271 64.2 209 579 59.4 356 404 116 90.8 21.6 133 54.4 33.2 3263
mean
male serum
Table 1. PCB Concentrations in Hair (ng/g) and Serum (ng/g lipid) from E-Waste Recycling Workers in Southern China
262 367 127 272 90.4 279 797 71.0 378 428 95.1 92.3 22.3 122 47.6 35.8 3980
494
mean
female serum range
n.d.−506 n.d.−983 36.7−299 n.d.−943 n.d.−331 n.d.−1241 49.5−2850 n.d.−250 n.d.−1216 n.d.−1153 n.d.−345 18.7−238 n.d.−66.2 n.d.−336 n.d.−120 n.d.−71.21 256−11200
n.d.−1445
Environmental Science & Technology Article
1580
DOI: 10.1021/acs.est.5b04955 Environ. Sci. Technol. 2016, 50, 1579−1586
Article
Environmental Science & Technology
surface, which was also recommended by a recent study.23 They were then freeze-dried, cut into small pieces (2−3 mm), and homogenized. Hair was incubated overnight (12 h) in hydrochloric acid (4 M) and a hexane/dichloromethane mixture (4:1, v/v). PCBs in the incubation medium were extracted via liquid−liquid extraction with a hexane/dichloromethane mixture (4:1, v/v), which was repeated three times. The combined extracts were purified on a silica/alumina column and condensed, after which internal standards (PCB 24, 82, and 198) were added. PCBs were analyzed by an Agilent 6890 gas chromatograph equipped with a 5975B mass spectrometer in electron impact ionization (EI) mode (GC-EI-MS). A DB-5 ms capillary column (60 m × 0.25 mm i.d., 0.25 μm film thickness) was used to separate the PCB congeners. Enantiomer analysis was conducted using the Agilent GC-EI-MS. The atropisomeric PCB 132 and 183 enantiomers were separated on a BGB-172 (30 m × 0.25 mm i.d., 0.18 μm film thickness) capillary column, while PCB 95 was separated on a Chirasil-Dex (25 m × 0.25 mm i.d., 0.25 μm film thickness) column. Further instrumental analysis information is provided in the Supporting Information (SI). Quality Control. The PCB 65 and 204 surrogate standard recoveries ranged from 65 to 98% and from 72 to 118% in the serum samples, and 60−121% and 58−115% in the hair samples, respectively. The final results were not recovery corrected. Serum (Milli-Q water) and hair sample (hydrochloric acid and a hexane/dichloromethane mixture) procedural blanks were run with each sample batch. Only trace amounts (
