Environmental Pollution 254 (2019) 112955

Contents lists available at ScienceDirect

Environmental Pollution

journal homepage: www.elsevier.com/locate/envpol

Sediment records of polybrominated diphenyl ethers (PBDEs) in Huaihe River, : Implications for historical production and household usage of PBDE-containing products*

* Chunnian Da a, c, Ruwei Wang b, , Jingsong Ye a, Shichao Yang d a Department of Biology and Environment Engineering, University, Hefei, 230022, China b Guangdong Key Laboratory of Environmental Pollution and Health, School of Environment, Jinan University, Guangzhou, 510632, China c Key Laboratory for Ecological Environment in Coastal Areas (SOA), Dalian, 116023, China d Top Way Testing Service Co., LTD, , Anhui 300071, China article info abstract

Article history: In recent decades, rapid development of industrialization and urbanization caused adverse impact on the Received 17 March 2019 aqueous ecology and environment of the Huaihe River basin, China. In this work, three 210Pb-dated Received in revised form sediment cores extracted from the middle reach of Huaihe River in Anhui Province, China were analyzed 19 July 2019 to elucidate the temporal trends and sources of polybrominated diphenyl ethers (PBDEs). Source di- Accepted 23 July 2019 agnostics indicated that commercial Deca-BDE, Penta-BDE and Octa-BDE products and debromination of Available online 25 July 2019 higher brominated BDE compounds were likely the PBDE sources in the Huaihe River. The prevalence of BDE-47 in the sediment cores was attributed to the extensive use of commercial Bromkal 70-5DE and Keywords: Polybrominated diphenyl ethers Bromkal DE-71 in the region. BDE-28 was another congener that was prevalent in all sediment samples, Sedimentation suggesting that reductive debromination occurred in the sediments. Dramatic increase of PBDE con- Reductive debromination centrations in both three cores since the post-1980s could be attributed to the rapid expansion of pro- Source analysis duction of electronic and telecommunication equipment and household usage in China. PBDE temporal Huaihe River trends in core S1 located at rural area mainly reflected the regional and national inputs deriving from long distance atmospheric transport, and the positive correlations between PBDE concentration in core S1 and gross domestic product (GDP) and household appliances production volumes (HPVs) were observed. PBDE inputs at site S3 mainly include the transport of contaminated water and re-suspended fine sediment particles from the upstream site S2, which was located in the industrial area and adjacent to e-waste recycling area. The government efforts to protect the environment and improve the e-waste management resulted in the progressive decrease trends in PBDE concentrations in cores S2 and S3. © 2019 Elsevier Ltd. All rights reserved.

1. Introduction commercial PBDEs were produced globally: Penta-BDE (over 70% of BDE-47 and BDE-99), Octa-BDE (over 40% of BDE-183), and Deca- Polybrominated diphenyl ethers (PBDEs) have been widely used BDE (over 98% of BDE-209) (Alaee et al., 2003). In USA and Can- as brominated flame retardants (BFRs) in various commodities (e.g., ada, it was estimated that about 46000, 25000 and 380000 tons of electrical and electronic equipment, textiles, plastics, building commercial penta-BDE, octa-BDE and deca-BDE have been and will materials and furniture upholstery) for more than half a century be consumed during 1970 and 2020 (Abbasi et al., 2015). Moreover, (Alaee et al., 2003; Abbasi et al., 2015; Tiwari et al., 2018). They have Asian countries, especially China, are purported to recycle and become environmental contaminants of great concern due to their dispose of obsolete electronic goods (so-called “e-waste”) imported global distribution, potential bioaccumulation, persistence and from developed countries (Breivik et al., 2016). Penta- and Octa- detrimental health effects (UNEP, 2006, 2017). Three major forms of BDE mixtures have been banned in Europe and the United States since 2004 (EPR, 2002) and globally under the Stockholm Convention (UNEP, 2006), but continue to be released from legacy * This paper has been recommended for acceptance by Charles Wong. e-waste products (Abbasi et al., 2015). Deca-BDE commercial mix- * Corresponding author. tures are still in use as BFRs, although these should have been E-mail address: [email protected] (R. Wang). https://doi.org/10.1016/j.envpol.2019.07.123 0269-7491/© 2019 Elsevier Ltd. All rights reserved. 2 C. Da et al. / Environmental Pollution 254 (2019) 112955 phased out under the Stockholm Convention (UNEP, 2017). BDE- Laboratories (CIL), respectively. Anhydrous sodium sulfate (Sino- 209, the major component of Deca-BDE, may degrade into less pharm Chemical Reagent Co., China) was activated at 500 C for 4 h. brominated congeners which are more toxic and persistent than Neutral silica gel (Riedel de Haen, Germany) and alumina (Darm- BDE-209 (UNEP, 2006). stadt, Germany) were activated at 180 and 250 C for 4 h respec- The composition of PBDEs in environmental contamination tively, and then deactivated with 3% Milli-Q water. depends on the source and contamination history. As PBDEs are generally hydrophobic and persistent in the environment, lacus- 2.2. Field sampling trine sediment, particularly anoxic and light-protected sediment, is a major sink of PBDEs (Alaee et al., 2003; D'Silva et al., 2004). Three sediment cores were collected along the Huaihe River in Consequently, historical variations in PBDE concentrations and July 2017, using a stainless-steel gravity corer. Cores S1 characteristics in undisturbed and well-dated sediment cores can (length ¼ 38 cm), S2 (length ¼ 39 cm), and S3 (length ¼ 36 cm) provide reliable information to assess the ecotoxicological risk were sampled from the river sections passing through City resulting from anthropogenic toxic and hazardous substances and (water depth ¼ 5.7 m), City (water depth ¼ 5.5 m) and improve the regulatory policy (Song et al., 2005). The environ- Wuhe County (water depth ¼ 4.9 m), respectively (Fig. 1). Core S1 mental occurrence of PBDEs in China has received great concern in was located at the upstream in the rural area and far away from recent decades. Previous studies have examined the occurrence of industrial pollution. In contrast, Core S2 was located in the indus- PBDEs in various aquatic systems in China, including coastal areas, trial district of Bengbu city and adjacent to electronic waste recy- such as the East China Sea (Li et al., 2012) and South China Sea (Zhu cling site with intensive e-waste recycling activities for more than et al., 2018), estuaries, such as the Yellow River Delta (Zhao et al., 15 years. Core S3 was located at the downstream in a county nearby 2012; Da et al., 2019), Yangtze River Delta (Chen et al., 2006), and Bengbu. The cores were sectioned into 1.0 cm increments and Pearl River Delta (Zou et al., 2007; Li et al., 2007), tributaries of large immediately delivered to the laboratory. Each segment was wrap- rivers, such as the Huangpu River (Wang et al., 2015a,b) and inland ped in aluminum foil, sealed, and stored at 20 C until analysis. lacustrine deposits, such as at Taihu Lake and Lake (He et al., 2013; Da et al., 2019). However, information about the ef- 2.3. Sediment core dating fect of socioeconomic activities on the historical variations of PBDEs on a regional- and local-scale is still very sparse. 0 0 0 0 The sediment chronology was determined from the specific Huaihe River (30 55 e36 36 N, 111 55 e121 25 E) is the fifth activities of 210Pb, calibrated by those of 137Cs. Sediment samples longest river in China with a total length of approximately 1000 km. were freeze-dried, ground, homogenized, and then placed in sealed It flows through five provinces from west to east including Henan, containers for 20 d to allow equilibration of 226Ra and its daughter Hubei, Anhui, , and Shandong Province, and Anhui is located isotope 214Pb. The activities of 210Pb, 226Ra and 137Cs were measured at the middle reach with a total length of 430 km. As a result of the with -spectrometry at 46.5 keV, 352 keV and 661 keV, respec- rapid growth of economic and industrial development, Huaihe g tively, using a high-purity germanium detector (Canberra GL River was suffered with severe pollution of various environmental 2820R, Canberra Industries, Meriden, CT, USA). The activity of compartments in recent years (Wang et al., 2017a,b; Luo et al., excess 210Pb (210Pb ) was calculated by subtracting 226Ra activities 2018; Bai et al., 2019). In addition, the major industries of textile, ex from total 210Pb activities (Sanchez-Cabeza and Ruiz-Fernandez, household appliances, and fertilizer distributed in this region could 2012). The age of the sediment core samples was calculated using account for large emission of PBDEs (Da et al., 2019). Our previous a continuous rate of supply (CRS) dating model (Appleby, 2002): study (Da et al., 2017) reported that increased industrial and do-     mestic wastewater discharges and agriculture garbage along the 1 Að0Þ Huaihe River significantly contributed to the elevated levels of t i ¼ ln l AðiÞ organochlorine pesticides in the sediment cores, from 0.01 ng g 1 in the bottom cores to 7.2 ng g 1 in the top cores. Building upon this AðiÞ finding, the present study further examined the correlation be- rðiÞ¼l tween the sedimentary histories of PBDEs and anthropogenic ac- Ci tivities, achieving by analyzing three 210Pb-dated sediment cores from three geographically distant sites within the middle reaches of where t(i) is the time elapsed for the sediment formation at depth 210 1 the Huaihe River representing the urban and rural area. The ob- (i), l is the decay constant of Pb (0.031 yr ), A(0) is the total 210 jectives were to establish correlations between the sedimentary inventory of the Pbex in one core, A(i) is the accumulated deposit 210 histories of PBDEs and anthropogenic activities in the region, as of Pbex in the sediment below depth (i), Ci is the content of 210 210 well as to diagnose potential input sources of PBDEs. Pbex in sediment at depth (i). The Pbex dating results and its uncertainty associated with the CRS dating model are shown in ± 2. Materials and methods Fig. S1. Counting statistical errors were better than 3%. The sedi- mentary ages were estimated as 1956e2015 spanning ± e ± 2.1. Materials 59 3.42 years at S1, 1958 2017 spanning 59 3.40 years at S2 and 1960e2016 spanning 56 ± 3.39 years at S3. The estimated average A mixture standard solution containing 39 PBDE congeners (i.e., sedimentary rates for S1, S2 and S3 were 0.64, 0.65 and 0.63 cm 1 BDE-1, BDE-2, BDE-3, BDE-10, BDE-7, BDE-11, BDE-8, BDE-13, BDE- yr , respectively. 12, BDE-15, BDE-30, BDE-32, BDE-17, BDE-25, BDE-28, BDE-33, BDE-35, BDE-37, BDE-75, BDE-49, BDE-71, BDE-47, BDE-66, BDE-77, 2.4. Sample preparation and extraction BDE-100, BDE-119, BDE-99, BDE-118, BDE-116, BDE-85, BDE-126, BDE-155, BDE-154, BDE-153, BDE-138, BDE-166, BDE-183, BDE- In total 113 sediment samples of the three cores were analyzed 181, BDE-190) and an individual BDE-209 were purchased from for PBDEs. Extraction and purification of PBDEs in sediment sam- AccuStandard (New Haven, CT, USA). The internal standard (PCB- ples followed the same procedures as described in our previous 13 204) and surrogate standards ( C12-labeled BDE-138 and BDE-153) studies (Da et al., 2017, 2019; Yuan et al., 2016). Prior to extraction, were obtained from AccuStandard and Cambridge Isotope surrogate standards 13C labeled BDE-138 and BDE-153 were added. C. Da et al. / Environmental Pollution 254 (2019) 112955 3

Fig. 1. Sampling locations of the sediment cores in Huaihe River, China.

Each sample of 10 g was freeze-dried and Soxhlet extracted for 48 h the peak areas. with a mixture of acetone and n-hexane (v:v ¼ 1:1). Copper strips activated with dilute HCl were added to remove elemental sulfur 2.6. Quality assurance/quality control and statistics analysis from the extract. The extract was concentrated to ca. 1 mL with a rotary evaporator under reduced pressure. The concentrated To avoid any organic contamination, all glassware was pre- fi extract was puri ed/fractionated with sulfuric acid silica and cleaned with distilled water, acetone and MeOH, and then was alumina (44% sulfuric acid, w/w) combined column chromatog- baked at 450 C for 5 h before use. One procedural blank was pro- raphy. The PBDE fraction was eluted by 35 mL of n-hexane, followed cessed in the same manner in every batch to check for interference ¼ by 70 mL of n-hexane: dichloromethane mixture (v:v 1:1). The or contamination, and no target substance was detected in the second eluate was reduced to ca. 5 mL by rotary evaporation and blank sample. One replicate sample were processed for every five then to 1 mL under a gentle stream of high purity N2 and spiked field samples to test the reliability of the method. PBDE concen- with the internal standard before instrumental analysis. trations were calculated by five-point standard calibration curves, and correlation coefficient (R2) of 0.999 or above were obtained. 1 2.5. Instrumental analysis The mixed standards with 40 PBDEs (10 ng g for each congener) were spiked into blank samples, and their mean recoveries are Quantitative analyses of PBDEs other than BDE-209 (Br -Br - shown in Supporting Information Table S1. The surrogate recovery 3 9 ± ± ± BDEs) were performed using an Agilent 7890 Series gas chro- rates ranged from 87.1 4.3% to 106 6.3%, and 80.5 6.1% to ± 13 13 matograph interfaced with an Agilent 5975C mass spectrometer in 94.3 5.2% for C12-BDE-138 and C12-BDE-153, respectively. The selected ion mode, equipped with a 30 m 0.25 mm i.d. DB-5MS reported concentrations were not corrected according to the sur- column with a film thickness of 0.25 mm. Ultra-high purity helium rogate recovery rates. Replicate sample analyses varied within < was used as the carrier gas. The column oven temperature was set acceptable limits (SRD 10%). The limits of detection (LOD) 1 (defined as a signal-to-noise ratio of 3), ranging from 2.0 to at 60 C and maintained for 2 min, then ramped at 10 C min to 1 200 C for 2 min, increased at 20 C min 1 to 300 C, and main- 18.5 pg g , was determined according to U.S. EPA method (U.S. EPA, tained for 10 min. The temperatures of the ion source and interface 2016)(Table S1). line were set at 150 C and 280 C, respectively. BDE-209 was Statistical analysis of the data was carried out using the SPSS analyzed using Thermo Trace Ultra gas chromatograph coupled to a 16.0 (Chicago, IL, USA). The principal component analysis was Thermo DSQ II mass spectrometer. The gas chromatograph was fit conducted with VARIMAX rotation and a factor adjustment of 2. with a DB-5HT capillary column (15 m 0.25 mm i.d., 0.25 mm film The T-test was used to investigate if there are any potential dif- thickness). The mass spectrometer was operated in selective ion ferences in levels of PBDEs among three sampling sites and the fi ¼ monitoring mode and the ion source was employed in positive ion different samples. The level of signi cance was set at p 0.05. electron impact mode at 70 eV. The oven temperature was set at 120 C and maintained for 2 min, ramped at 20 C min 1 to 300 C, 3. Results and discussion and maintained for 12 min. The ion source and interface tempera- tures were both set to 300 C. Quantitative ions were as follows: m/ 3.1. Occurrence of sediment PBDEs and ecological implications z 248, 250 for BDE-1, -2, and -3; m/z 328, 168, 326 for BDE-10, -7, -11, 8, 13, 12, 15; m/z 246, 248, 406, 408 for BDE-30, The detection rate of individual PBDEs ranged from 48.2 to 98.1% -32, 17, 25, 28, 33, 35, 37; m/z 326, 486 for BDE-75, in S1, 56.8e96.7% in S2 and 59.4e96.3% in S3, reflecting the -49, 71, 47, 66, 77; m/z 404, 406, 564, 566 for BDE-100, extensive use of PBDEs historically in the study region. Overall, 10 -119, 99, 118, 116, 85, 126; m/z 484, 644 for BDE-155, out of 40 target PBDE congeners were detected in all samples. In- -154, 153, 138, 166; m/z 562, 564, 722, 724 for BDE-183, dividual PBDE concentrations in the three sediment cores showed a -181, 190; m/z 799 for BDE-209; m/z 430, 428 for PCB-204; m/z similar sequence, i.e., BDE-209 > BDE-47 > BDE-183 > BDE- 13 13 for C12-labeled BDE-190; m/z 302.0, 371.9, 373.9 for C12-BDE- 154 > BDE-100 > BDE-99 > BDE-28 > BDE-85 > BDE-37 > BDE-153. 13 1 138; m/z 495.8, 655.7 for C12-BDE-153. The internal standard PCB- Moreover, S10PBDE concentrations ranged from 0.27 to 4.9 ng g 204 (100 mLof10mgL 1) were added to each sample to normalize dry weight (mean: 3.4 ng g 1) for sediment core S1, 0.69e6.0 ng g 1 4 C. Da et al. / Environmental Pollution 254 (2019) 112955

Table 1 Concentration of PBDEs (ng g 1 dw) in the sediment cores.

PBDE congeners Concentrations of PBDE Congeners (ng g 1) Mean

Site S1 Site S2 Site S3

Mean Range Detection Mean Range Detection Mean Range Detection rate (%) rate (%) rate (%)

BDE-28 (Tri-BDEs) 0.02 0.002 48.2 0.19 0.003 65.3 0.23 0.001 59.4 0.15 e0.25 e0.45 e0.48 BDE-37 (Tri-BDEs) 0.01 0.01e0.21 59.2 0.06 0.001 63.1 BDL BDL BDL 0.04 e0.12 BDE-47 (Tetra-BDEs) 0.16 0.05e0.37 94.2 1.6 0.03e2.0 89.9 1.5 0.001e3.1 85.5 1.1 BDE-85 (Penta-BDEs) 0.01 0.001 59.3 0.31 0.001e1.0 56.8 0.10 0.001e1.0 61.2 0.14 e0.71 BDE-99 (Penta-BDEs) 0.12 0.07e0.30 91.4 0.23 0.001 91.9 0.18 0.002 90.1 0.18 e0.57 e0.43 BDE-100 (Penta- 0.24 0.01e0.33 91.8 0.41 0.002e1.0 93.5 0.34 0.003 89.5 0.33 BDEs) e0.78 BDE-153 (Hexa-BDEs) 0.02 0.01e0.19 61.3 BDL BDL BDL BDL BDL BDL 0.02 BDE-154 (Hexa-BDEs) 0.13 0.01e0.36 60.2 0.63 0.002e1.0 67.4 0.54 0.02e1.0 78.1 0.43 BDE-183 (Hepta- 0.28 0.03e0.42 97.5 1.2 0.001e1.8 95.4 0.78 0.03e2.1 91.3 0.76 BDEs)

ƩBr3-Br9-BDEs 1.0 0.27e2.6 73.7 4.7 0.69e6.0 85.7 3.6 0.04e6.1 79.1 3.1 BDE-209 (Deca-BDE) 2.4 1.1e4.9 98.1 3.4 1.0e4.9 96.7 4.2 0.02e5.6 96.3 3.3

Ʃ10PBDEs 3.4 0.27e4.9 82.9 8.0 0.69e6.0 91.2 7.8 0.02e6.1 83.3 6.4 BDL: below detection limit.

(mean: 8.0 ng g 1) for sediment core S2 and 0.02e6.1 ng g 1 the three sediment cores are illustrated in Fig. 2. The predomi- (mean: 7.8 ng g 1) for sediment core S3 (Table 1 and Table S2). The nation of BDE-209, which contributes 50e76%, 32e49% and variation in concentrations of S10PBDE among the three different 19e63% from 1956 to 2015 in the core S1, from 1958 to 2017 in the sites was primarily attributed to their geographic positions and core S2, from 1960 to 2016 in the core S3, respectively, suggested surrounding anthropogenic activities: Site S2 was adjacent to a the extensive usage of Deca-BDE mixtures as a brominated flame recycling center for electronic products, which was responsible for retardant in China. A similar result was also reported in the sedi- the highest concentration; in addition, PBDE concentrations were ments from the Yellow River Estuary (Yuan et al., 2016), Southern more likely to accumulate in the downstream site (S2 and S3) Yellow Sea (Wang et al., 2017a,b), East China Sea (Li et al., 2016a,b), compared to the upstream site (S1). Bohai Sea (Pan et al., 2011), and Deep Bay in South China (Qiu et al., PBDEs could bioaccumulate in food chains (El-Nahhal, 2018). 2010). Among the PBDE congeners, BDE-47 and BDE-183 were the Therefore, the ecological impacts of PBDE contaminants on aquatic most abundant compounds, accounting for up to 17% and 12% of organic organisms should be investigated. The environmental safety standards for PBDEs in sediments have not been established Table 2a in China, so an environmental quality guideline (EQG) established PBDE concentrations (ng g 1) in core S1 and EQGs. by Canada (Environment Canada, 2010) was used to evaluate the environmental ecological risk of PBDEs to subaqueous organisms in Year Depth (cm) Tri-BDEs Tetra-BDEs Penta-BDEs Hexa-BDEs Deca-BDEs the Huaihe River. The EQGs for PBDEs were 44, 39, 0.4, 440, and 2015 1 0.02 0.03 0.05 0.05 4.3 19 ng g 1 dry weight for the Tri-BDEs, Tetra-BDEs, Penta-BDEs, 2014 2 0.01 0.03 0.06 0.04 4.6 Hexa-BDEs, and Deca-BDEs, respectively (Environment Canada, 2013 3 0.02 0.02 0.05 0.05 4.4 2013 4 0.04 0.04 0.17 0.96 4.2 2010). As seen in Table 1, most of the PBDE levels were less than 2012 5 0.04 0.07 0.18 0.12 4.2 1 the EQGs, except Hexa-BDEs with an average value 0.43 ng g for 2011 6 0.02 0.08 0.11 0.17 4.9 BDE-154, implying that there was a negligible hazard to the or- 2010 7 0.01 0.11 0.22 0.17 4.1 ganisms in the study areas. 2009 8 0.001 0.07 0.15 0.17 4.1 Moreover, the toxicological data of PBDEs and the estimation of 2008 9 0.002 0.04 0.15 0.15 4.3 2007 10 0.002 0.03 0.11 0.11 4.4 hazard quotients (HQ) were used to quantify the ecological impacts 2006 11 0.004 0.03 0.01 0.11 4.5 of PBDEs on sediment-dwelling organisms in the watersheds 2005 12 0.002 0.04 0.008 0.09 4.1 (Environment Canada, 2013). The HQ is defined as the ratio of PBDE 2005 13 0.004 0.02 0.005 0.08 3.9 concentrations in sediments to the critical effect concentration 2004 14 0.002 0.02 0.002 0.003 3.8 2002 15 0.001 0.01 0.002 0.001 3.7 below which no adverse effect is expected (PNECs), which is 1999 16 0.001 0.04 0.005 0.09 3.1 1 determined by Environment Canada (2013) to be 0.03 mg kg and 1999 17 0.001 0.03 0.15 0.1 2.8 1 73 mg kg for Penta-BDE and Deca-BDE, respectively. HQ < 0.1 1998 18 0.004 0.02 0.12 0.08 0.48 indicates no hazard; 0.1 HQ < 1 low hazard; 1 HQ < 10 moder- 1997 19 0.004 0.01 0.01 0.04 2.8 ate hazard; HQ 10 high hazard. The HQ values for the two com- 1995 20 0.006 0.02 0.02 0.05 2.8 1993 21 0.006 0.01 0.01 0.02 2.3 mercial PBDEs were determined to be all less than 0.1 in the 1992 22 0.005 0.01 0.01 0.04 1.8 sediments from the three sites (Table 2a,b,c), demonstrating that 1991 23 0.004 0.007 0.01 0.04 3.8 there was no hazard to the organisms in the study areas. 1990 24 0.004 0.02 0.01 0.05 2.1 1988 25 0.003 0.06 0.01 0.09 0.12 1985 26 0.002 0.06 0.1 0.13 1.8 3.2. Compositions and sources of PBDEs 1983 27 0.001 0.004 0.09 0.11 0.29 1981 28 0 0.004 0.1 0.1 0.09 EQGs 44 39 0.4 440 19 The relative compositions of PBDEs in the different periods for C. Da et al. / Environmental Pollution 254 (2019) 112955 5

Table 2b (8e13%) and BDE-28 (0.1e0.2%) (Laguardia et al., 2006) and were 1 PBDE concentrations (ng g ) in core S2 and EQGs. used as flame retardants until 2004 (US. EPA, 2017). The relatively Year Depth (cm) Tri-BDEs Tetra-BDEs Penta-BDEs Hexa-BDEs Deca-BDEs high abundance of BDE-183 implies the wide usage of commercial

2017 1 0.08 0.16 0.19 0.03 3.6 Octa-BDE. It was surprising that BDE-28 was also widely detected 2016 2 0.07 0.16 0.12 0.04 3.1 in the cores although it is not present in Penta-, Octa- or Deca-BDEs. 2015 3 0.09 0.18 0.2 0.06 4.3 Comparable results were also observed in the East China Sea (Wang 2014 4 0.01 0.28 0.2 0.1 4.6 et al., 2016), southern Yellow Sea (Wang et al., 2017a,b), and Bohai 2013 5 0.1 0.28 0.23 0.12 4.4 Sea (Pan et al., 2011). It has been reported that BDE-28 can be 2013 6 0.1 0.27 0.34 0.58 5.2 2012 7 0.13 0.39 0.37 0.58 5.2 formed through debromination of BDE-209 via photodecomposi- 2011 8 0.1 0.39 0.39 0.58 8.9 tion or, microbially mediated transformation in the environment 2010 9 0.07 0.47 0.4 0.95 7.1 (Bezares-Cruz et al., 2004). Therefore, the prevalence of BDE-28 in 2009 10 0.14 0.51 0.4 0.11 7.1 the sediment cores suggested that reductive debromination prob- 2008 11 0.1 0.46 0.3 0.21 6.3 2007 12 0.08 0.29 0.29 0.23 5.4 ably took place in the sediments. Deca-BDE can also debrominate in 2006 13 0.02 0.25 0.18 0.13 5.5 sediments to Hepta-, Octa-, and Nona-BDEs (Orihel et al., 2016). 2005 14 0.03 0.23 0.17 0.1 5.1 The principal component analysis was performed on the 2005 15 0.02 0.28 0.14 0.09 3.9 measured PBDEs to identify possible sources. As seen in Fig. 3, the 2004 16 0.01 0.25 0.12 0.06 3.8 first two major components distributed as PCF-1 and PCF-2 were 2002 17 0.01 0.1 0.13 0.05 3.7 1999 18 0.01 0.29 0.16 0.33 3.1 1998 19 0.03 0.19 0.15 0.23 2.5 1997 20 0.04 0.3 0.11 0.06 2.8 1995 21 0.06 0.28 0.11 0.08 2.8 1993 22 0.06 0.19 0.01 0.04 2.3 1992 23 0.04 0.15 0.008 0.04 2.8 1991 24 0.04 0.1 0.008 0.04 3.8 1990 25 0.02 0.45 0.006 0.04 2.1 1988 26 0.007 0.29 0.1 0.06 2.1 1985 27 0.005 0.2 0.11 0.19 2.8 1983 28 0.005 0.3 0.005 0.06 2.3 1981 29 0.006 0.3 0.006 0.05 2.1 EQGs 44 39 0.4 440 19

Table 2c PBDE concentrations (ng g 1) in core S3 and EQGs.

Year Depth (cm) Tri-BDEs Tetra-BDEs Penta-BDEs Hexa-BDEs Deca-BDEs

2016 1 0.01 0.03 0.05 0.02 4.0 2015 2 0.01 0.02 0.07 0.04 4.0 2014 3 0.01 0.01 0.1 0.04 4.4 2013 4 0.07 0.03 0.1 0.31 4.6 2012 5 0.1 0.51 0.09 0.5 5.1 2011 6 0.19 0.67 0.2 0.14 6.9 2010 7 0.21 0.91 0.99 0.17 6.1 2009 8 0.22 1.0 0.55 0.15 6.1 2008 9 0.11 0.61 0.41 0.1 6.3 2007 10 0.03 0.08 0.1 0.01 6.5 2006 11 0.02 0.01 0.005 0.03 7.5 2005 12 0.01 0.07 0.005 0.04 6.1 2005 13 0.004 0.04 0.005 0.008 6.9 2004 14 0.003 0.09 0.001 0.13 6.8 2002 15 0.004 0.03 0.002 0.41 5.7 1999 16 0.04 0.44 0.001 0.35 5.1 1998 17 0.07 0.21 0.001 0.001 4.5 1997 18 0.03 0.02 0.01 0.01 4.8 1995 19 0.02 0.01 0.01 0.02 4.8 1993 20 0.02 0.83 0.01 0.04 4.3 1992 21 0.12 0.31 0.02 0.004 3.8 1991 22 0.02 0.02 0.01 0.001 4.8 1990 23 0.02 0.01 0.06 0.35 4.1 1988 24 0.02 0.01 0.001 0.002 4.1 1985 25 0.001 0.01 0.002 0.001 3.8 1983 26 0.002 0.001 0.001 0.001 3.0 1981 27 0.001 0.002 0.001 0.001 3.0 EQGs 44 39 0.4 440 19

S10PBDEs, respectively. It suggests that Octa-BDE and Penta-BDE were also the significant sources of PBDEs in the Huaihe River. The prevalence of congener BDE-47 in the three sediment cores might relate to the extensive use of commercial less-brominated mixtures of Bromkal 70-5DE and Bromkal DE-71, which were e e composed of BDE-99 (45 49%), BDE-47 (38 43%), BDE-100 Fig. 2. Compositional pattern of PBDEs in sediment core. 6 C. Da et al. / Environmental Pollution 254 (2019) 112955

Fig. 3. Principal component analysis for PBDEs in the sediment core samples.

extracted, accounted for 99.9% of the data variation. PCF-1 with the debromination of higher brominated compounds (Lv et al., explained 55.3% of total variance, which showed high loadings on 2015). a group of PBDEs, including BDE-47, BDE-85, BDE-99, BDE-209, BDE-100, BDE-28, BDE-37, BDE-154 and BDE-183, among which BDE-47, BDE-154 and BDE-183 are representative of industrial 3.3. Temporal trends in PBDEs Penta-BDE and Octa-BDE products, and they showed analogous environmental behavior and source. Yuan et al. (2016) have also Core S1 was collected from the rural area and far away from the reported that there was a significant correlation among BDE-37, industrial center, thus the historical accumulation is unlikely BDE-154, and BDE-183 in the surface sediment from the Yellow affected by local PBDE input. PBDE concentrations in core S1 River Estuary in China. BDE-37 could be derived from the debro- increased substantially since the 1980s, coinciding with the mination of higher brominated BDE mixtures (Fan et al., 2014). increasing production, usage and disposal of commercial PBDE Therefore, PCF-1 suggested pollution by industrial Penta-BDE and products over this period. The Chinese trade in household appli- Octa-BDE products, as well as the debromination of higher ances and computers has increased greatly since the early 1980s, brominated BDE compounds. PCF-2 accounted for 44.7% of the total which promoted China as a global processing factory for electronic variability, which showed high loadings on BDE-153 associated and telecommunication equipment (household appliance, com- puters and phones) from the late 2010s (Jiang et al., 2011). In

Fig. 4. Time trends of PBDE concentrations in the sediment cores. C. Da et al. / Environmental Pollution 254 (2019) 112955 7 addition, PBDE concentrations showed no obvious decline toward commercial penta- and octa-BDEs have been banned in 2007 in the sediment surface dated to 2015 (Fig. 4), implying the contin- China to comply with Stockholm Convention (Zhang et al., 2009). uous releasing of PBDEs resulting from the usage and circulation of In contrast to the upstream rural site S1, core S2 and S3 were legacy PBDE products, although the production and new usage of collected from the highly populated areas which were susceptible

Fig. 5. Relationships of PBDEs in the sediment cores with the GDP and household appliance in the Huaihe River. 8 C. Da et al. / Environmental Pollution 254 (2019) 112955 to PBDE inputs from local anthropogenic activities. The depth the main source contributions in the Huaihe River. The prevalence profiles for both Br3-Br9-BDEs and BDE-209 showed similar pat- of BDE-28 in the sediment cores suggested that reductive debro- terns between sites S2 and S3, suggesting the same local PBDE mination probably took place in the sediments. The sources anal- pollution source in this region. In addition, PBDE inputs at site S3 ysis indicated that debromination of higher brominated BDE mainly include the transport of contaminated water and re- compounds, commercial Deca-BDE, Penta-BDE and Octa-BDE suspended fine sediment particles from the upstream site S2, products were likely source materials in the Huaihe River. which was located in the industrial area and adjacent to e-waste recycling area. PBDE concentrations in core S2 and S3 generally Acknowledgements increased from 1960s to late 1980s, followed by substantial growth to a maximum around 2008, and then declined steadily thereafter This work is sponsored by National Natural Science Foundation (Fig. 4). The earlier detection of PBDEs for both Br3-Br9-BDEs and of China (41773099), Guangdong Provincial Key Laboratory of Soil fi BDE-209 in sediment cores before their rst emergence was also and Groundwater Pollution Control (No. 2017B030301012), Anhui documented in other studies, such as the Southern Yellow Sea in Provincial Natural Science Foundation (1608085MD78), Key Labo- China around 1910s (Wang et al., 2017a,b), East China Sea around ratory for Ecological Environment in Coastal Areas (201802), Key 1930s (Wang et al., 2016) and Deep Bay in South China around projects of Anhui province university outstanding youth talent 1940s (Qiu et al., 2010). It was explained by the down-core diffusion support program (gxyqZD2016274, gxgnfx2018032). We would like of PBDEs resulting from resuspension and bioturbation (Yang et al., to thank Dr. Derek Muir for giving many constructive suggestions. 2016). Moreover, possible sediment mixing during the period of Special thanks are given to anonymous reviewers for their useful e fi 1960s 1980s was also supported by the examination of the pro les suggestions and comments. of 210Pb concentration versus cumulative dry mass of the sediment (Fig. S2). On the other hand, Chinese household consumption Appendix A. Supplementary data expenditure increased about 18-fold from 1990 to 2013 (Shang et al., 2016). Therefore, dramatic increase of PBDE concentrations Supplementary data to this article can be found online at in core S2 and S3 after 1980s could be attributed to the rapid https://doi.org/10.1016/j.envpol.2019.07.123. production expansion of electronic and telecommunication equipment and household usage in China. References 3.4. Comparison with records of anthropogenic production Abbasi, G., Buser, A.M., Soehl, A., et al., 2015. Stocks and flows of PBDEs in products from use to waste in the U.S. and Canada from 1970 to 2020. Environ. Sci. To further investigate the influence of economic growth and Technol. 49, 1521e1528. industrialization on the historical deposition of PBDEs, the histor- Alaee, M.P., Arias, P., Bergman, Å., 2003. An overview of commercially used brominated flame retardants, their applications, their use patterns in different ical PBDE concentrations were decoupled from the annual gross countries/regions and possible modes of release. Environ. Int. 29, 683e689. domestic product (GDP) of Anhui Province and the household ap- Appleby, P.G., 2002. Chronostratigraphic Techniques in Recent Sediments. Tracking pliances production volumes (HPVs) of China. From 1971 to 2015, Environmental Change Using Lake Sediments. Springer, Dordrecht, pp. 171e203. Anhui's GDP grew ca. 28-fold, from 0.08 to 2.2 trillion yuan. Bai, Y., Ruan, X.H., Xie, X.C., et al., 2019. Antibiotic resistome profile based on Meanwhile, Chinese HPVs increased ca. 1390-fold from 0.01 to 13.9 metagenomics in raw surface drinking water source and the influence of billion over the same time period (Fig. 5)(http://www.stats.gov.cn). environmental factor: a case study in Huaihe River Basin, China. Environ. Pollut. 248, 438e447. The positive correlations between PBDE concentration in core S1 Bezares-Cruz, J., Jafvert, C.T., Hua, I., 2004. Solar photodecomposition of deca- 2 2 and GDP (SBr3-Br9-BDEs: R ¼ 0.7; BDE-209: R ¼ 0.46) and HPVs bromodiphenyl ether: products and quantum yield. Environ. Sci. Technol. 38, 2 2 e (SBr3-Br9-BDEs: R ¼ 0.68; BDE-209: R ¼ 0.47) were observed 4149 4156. (Fig. S3). This relationship is supported by a recent study conducted Breivik, K., Armitage, J.M., Wania, F., et al., 2016. Tracking the global distribution of persistent organic pollutants accounting for E-waste exports to developing at the Southern Yellow Sea in China, East China Sea, and Pearl River regions. Environ. Sci. Technol. 50, 789e805. Estuary in South China (Wang et al., 2017a,b). In addition, this re- Chen, S.J., Gao, X.J., Mai, B.X., et al., 2006. Polybrominated diphenyl ethers in surface sediments of the Yangtze River Delta: levels, distribution and potential hy- sults suggested that PBDE temporal trends in core S1 located at fl e fl drodynamic in uence. Environ. Pollut. 144, 951 957. rural area mainly re ect the regional and national inputs deriving Da, C.N., Wu, K., Jin, J., et al., 2017. Levels and sources of organochlorine pesticides in from long distance atmospheric transport. In comparison, weak surface sediment from Anhui reach of Huaihe River, China. Bull. Environ. correlations between PBDE concentrations and Anhui's GDP and Contam. Toxicol. 98, 784e790. Da, C.N., Wu, K., Ye, J., et al., 2019. Temporal trends of polybrominated diphenyl Chinese HPVs were observed for both core S2 and core S3 (Fig. S3). ethers in the sediment cores from different areas in China. Ecotox. Environ. Safe. This is reasonable because sites S2 and S3 located at highly popu- 171, 222e230. lated areas mainly received PBDE inputs from local anthropogenic D'Silva, K., Fernandes, A., Rose, M., 2004. Brominated organic micropollutants e igniting the flame retardant issue. Crit. Rev. Environ. Sci. Technol. 34, 141e207. activities (e.g., e-waste recycling). PBDE concentrations in cores S2 El-Nahhal, Y., 2018. Toxicity of some aquatic pollutants to fish. Environ. Monit. and S3 decreased gradually in recent years thanks to government Assess. 190, 1e15. efforts to protect the environment and improve the e-waste man- Environment Canada, 2010. Risk Management Strategy for Polybrominated Diphenyl Ethers (PBDEs) 2010. Chemicals Sectors Directorate. Environmental agement (Lu et al., 2015), although the GDP and HPV have shown a Stewardship Branch, Ottawa ON, p. 30. steady increase over this period. Environment Canada, 2013. Federal Environmental Quality Guidelines for Poly- brominated Diphenyl Ethers (PBDEs). February 2013. e 4. Conclusions EPR European Parliament Report, 2002. A5-0437/2002. EUROPEAN Union Press Release, 30 January 2001. Fan, Y., Lan, J., Zhao, Z., et al., 2014. Sedimentary records of hydroxylated and PBDE concentrations in the dated sediment cores collected from methoxylated polybrominated diphenyl ethers in the southern Yellow sea. Mar. e Huaihe River were mostly affected by the historical usage of PBDEs Pollut. Bull. 84, 366 372. He, W., Qin, N., Kong, X., et al., 2013. Polybrominated diphenyl ethers (PBDEs) in the and environmental policies in China. The covariation between surface sediments and suspended particulate matter (SPM) from Lake Chaohu, PBDE concentrations and intensity of human activities since the a large shallow Chinese lake. Sci. Total Environ. 463, 1163e1173. 1970s implied that PBDE releases were mainly affected by rapid Jiang, J.J., Lee, C.L., Fang, M.D., et al., 2011. Polybrominated diphenyl ethers and polychlorinated biphenyls in sediments of Southwest Taiwan: regional char- economic growth and industrialization. The compositional patterns acteristics and potential sources. Mar. Pollut. Bull. 62, 815e823. of PBDEs showed that Deca-BDE, Octa-BDE, and Penta-BDE were Laguardia, M.J., Hale, R.C., Harvey, E., 2006. Detailed polybrominated diphenyl ether C. Da et al. / Environmental Pollution 254 (2019) 112955 9

(PBDE) congener composition of the widely used penta-, octa-, and deca-PBDE Convention on POPs, p. 46. technical flame-retardant mixtures. Environ. Sci. Technol. 40, 6247e6254. UNEP, 2017. Risk Profile on Decabromodiphenyl Ether (Commercial Mixture, C- Li, A., Tanabe, S., Jiang, G., et al., 2007. Persistent Organic Pollutants in Asia e decaBDE) UNEP/POPS/POPRC.10/10/Add.2. United Nations Environment Pro- Sources, Distributions, Transport and Fate, Developments in Environmental gramme, Stockholm Convention on Persistent Organic Pollutants, Geneva, Science, vol. 7. Elsevier, UK. Switzerland, p. 58. Li, W.L., Ma, W.L., Jia, H.L., et al., 2016a. Polybrominated diphenyl ethers (PBDEs) in US. EPA, 2016. Definition and Procedure for the Determination of the Method surface soils across five Asian countries: levels, spatial distribution, and source Detection Limit. Revision 2 Google Scholar. contribution. Environ. Sci. Technol. 50, 12779e12788. US. EPA, 2017. Polybrominated Diphenyl Ethers (PBDEs) Action Plan. http://www. Li, Y., Lin, T., Chen, Y., et al., 2012. Polybrominated diphenyl ethers (PBDEs) in epa.gov/sites/production/files/2015-09/documents/pbdes_ap_2009_1230_final. sediments of the coastal East China Sea: occurrence, distribution and mass pdfGoogleScholar. (Accessed 16 November 2017). inventory. Environ. Pollut. 171, 155e161. Wang, G.G., Feng, L.J., Qi, J.S., et al., 2017a. Influence of human activities and organic Li, Y., Lin, T., Hu, L., et al., 2016b. Time trends of polybrominated diphenyl ethers in matters on occurrence of polybrominated diphenyl ethers in marine sediment east China seas: response to the booming of PBDE pollution industry in China. core: a case study in the Southern Yellow Sea, China. Chemosphere 189, Environ. Int. 92e93, 507e514. 104e114. Lu, C.Y., Zhang, L., Zhong, Y.G., et al., 2015. An overview of E-wastes management in Wang, J., Liu, G.J., Lu, L.L., et al., 2015a. Geochemical normalization and assessment China. J. Mater. Cycles Waste 17, 1e12. of heavy metals (Cu, Pb, Zn, and Ni) in sediments from the Huaihe River, Anhui, Luo, Z., Zuo, Q., Shao, Q., 2018. A new framework for assessing river ecosystem China. Catena 129, 30e38. health with consideration of human service demand. Sci. Total Environ. 640, Wang, J., Liu, G.J., Liu, H.Q., et al., 2017b. Multivariate statistical evaluation of dis- 442e453. solved trace elements and a water quality assessment in the middle reaches of Lv, J., Zhang, Y., Zhao, X., et al., 2015. Polybrominated diphenyl ethers (PBDEs) and Huaihe River, Anhui, China. Sci. Total Environ. 538, 421e431. polychlorinated biphenyls (PCBs) in sediments of Liaohe River: levels, spatial Wang, P., Shang, H., Li, H., et al., 2016. PBDEs, PCBs and PCDD/FS in the sediments and temporal distribution, possible sources, and inventory. Environ. Sci. Pollut. from seven major river basins in China: occurrence, congener profile and spatial Res. 22, 4256e4264. tendency. Chemosphere 144, 13e20. Orihel, D.M., Bisbicos, T., Darling, C.T.R., et al., 2016. Probing the debromination of Wang, X.T., Chen, L., Wang, X.K., et al., 2015b. Occurrence, profiles, and ecological the flame retardant decabromodiphenyl ether in sediments of a boreal lake. risks of polybrominated diphenyl ethers (PBDEs) in river sediments of Environ. Toxicol. Chem. 35, 573e583. Shanghai, China. Chemosphere 133, 22e30. Pan, X., Tang, J., Li, J., et al., 2011. Polybrominated diphenyl ethers (PBDEs) in the Yang, C., Rose, N.L., Turner, S.D., et al., 2016. Hexabromocyclododecanes, poly- riverine and marine sediments of the Laizhou Bay area, north China. J. Environ. brominated diphenyl ethers, and polychlorinated biphenyls in radiometrically Monit. 13, 886. dated sediment cores from English lakes, ~ 1950epresent. Sci. Total Environ. Qiu, Y.W., Zhang, G., Guo, L.L., et al., 2010. Bioaccumulation and historical deposition 541, 721e728. of polybrominated diphenyl ethers (PBDEs) in Deep Bay, South China. Mar. Yuan, Z., Liu, G., Lam, M.H.W., et al., 2016. Occurrence and levels of polybrominated Environ. Res. 70, 219e226. diphenyl ethers in surface sediments from the Yellow River Estuary, China. Sanchez-Cabeza, J.A., Ruiz-Fernandez, A.C., 2012. 210Pb sediment radiochronology: Environ. Pollut. 212, 147e154. an integrated formulation and classification of dating models. Geochim. Cos- Zhang, X.L., Luo, X.J., Chen, S.J., et al., 2009. Spatial distribution and vertical profile of mochim. Acta. 82, 183e200. polybrominated diphenyl ethers, tetrabromobisphenol A, and deca- Shang, X., Dong, G., Zhang, H., et al., 2016. Polybrominated diphenyl ethers (PBDEs) bromodiphenylethane in river sediment from an industrialized region of South and indicator polychlorinated biphenyls (PCBs) in various marine fish from China. Environ. Pollut. 157, 1917e1923. Zhoushan fishery, China. Food Control 67, 240e246. Zhao, H., Chen, J., Qiao, X., et al., 2012. Polybrominated diphenyl ethers in soils of Song, W.L., Ford, J.C., Li, A., et al., 2005. Polybrominated diphenyl ethers in the the modern Yellow River Delta, China: occurrence, distribution and inventory. sediments of the great lakes. 3. Lakes Ontario and Erie. Environ. Sci. Technol. 39, Chemosphere 88, 791e797. 5600e5605. Zhu, B., Lam, J.C., Lam, P.K., 2018. Halogenated flame retardants (HFRs) in surface Tiwari, M., Sahu, S.K., Bhangare, R.C., et al., 2018. Polybrominated diphenyl ethers sediment from the Pearl River Delta region and Mirs Bay, South China. Mar. (PBDEs) in core sediments from creek ecosystem: occurrence, geochronology, Pollut. Bull. 129, 899e904. and source contribution. Environ. Geochem. Health. 40, 2587e2601. Zou, M.Y., Ran, Y., Gong, J., et al., 2007. Polybrominated diphenyl ethers in watershed UNEP, 2006. Risk Profile for Commercial Pentabromodiphenyl Ether. UNEP/POPS/ soils of the Pearl River Delta, China: occurrence, inventory, and fate. Environ. POPRC.2/17/Add.1. United Nations Environmental Program, Stockholm Sci. Technol. 41, 8262e8267.