Science of the Total Environment 481 (2014) 47–54

Contents lists available at ScienceDirect

Science of the Total Environment

j ourna l homepage: www.e lsev ie r .com/ locate /sc i totenv

Occurrence of brominated flame retardants (BFRs), organochlorinepesticides (OCPs), and polychlorinated biphenyls (PCBs) in agriculturalsoils in a BFR-manufacturing region of North China

Zhi-Cheng Zhu a,b, She-Jun Chen a,⁎, Jing Zheng c, Mi Tian a,b, An-Hong Feng a,b, Xiao-Jun Luo a, Bi-Xian Mai a

a State Key Laboratory of Organic Geochemistry, Guangzhou Institute of Geochemistry, Chinese Academy of Sciences, Guangzhou 510640, Chinab University of Chinese Academy of Sciences, Beijing 100049, Chinac Center for Environmental Health Research, South China Institute of Environmental Sciences, Ministry of Environmental Protection, Guangzhou 510655, China

H I G H L I G H T S

• BFR concentrations in the soil were dominated by deca-BDEs, TBBPA, and DBDPE.• The soil BFR profiles generally resembled the BFR production and use in China.• BDE209 is persistent and has not undergone significant degradation in the soil.• Manufacturing plants are a very significant source of BFRs into the environment.

⁎ Corresponding author. Tel.: +86 20 85290146; fax: +E-mail address: chenshejun@gig.ac.cn (S.-J. Chen).

http://dx.doi.org/10.1016/j.scitotenv.2014.02.0230048-9697/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 2 October 2013Received in revised form 6 February 2014Accepted 7 February 2014Available online 25 February 2014

Keywords:BFRsOCPsPCBsSoilSpatial distributionMass inventory

We investigated the occurrence of brominated flame retardants (BFRs), organochlorine pesticides (OCPs),and polychlorinated biphenyls (PCBs) in the surface soils from the largest BFR-manufacturing and vegetablefarming center (Shouguang) of North China. The total concentrations of BFRs ranged from 39.9 to 8145 ng/gdry weight with a mean of 1947 ng/g. The BFRs were dominated by decabromodiphenylethane (deca-BDEs)and tetrabromobisphenol A (TBBPA), with means of 1127 and 672 ng/g, respectively, followed bydecabromodiphenyl ethane (DBDPE) (111 ng/g) and hexabromocyclododecanes (HBCD) (37.5 ng/g). Thisprofile was generally consistent with the BFR production and use in China, except for TBBPA. Although thelower brominated BDEs (tri- through hepta-BDEs) in the soil may originate from technical deca-BDE mixturesas trace impurities and/or from the degradation of deca-BDEs, deca-BDE was shown to be persistent in the soil.The concentrations of OCPs (44 ng/g) were significantly lower than those of BFRs and displayed a spatial distri-bution opposite to that of BFRs, which was concentrated in the industrial zone. PCBs (with the lowest levels)showed a relatively uniform spatial distribution because of regional diffusive sources. The mass inventories forthe entire land soil (20-cm) were estimated to be 1042, 26, and 3.7 t for BFRs, OCPs, and PCBs, respectively.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Organochlorine pesticides (OCPs), polychlorinated biphenyls(PCBs), and brominated flame retardants (BFRs) represent three well-known classes of man-made organic chemicals and ubiquitous organiccontaminants found in the environment and in living organisms (Huet al., 2008). Many of these halogenated hydrocarbons have beenincluded in the Stockholm Convention on Persistent Organic Pollutants(POPs) because of their persistence, bioaccumulation, long-range trans-port properties, and adverse effects on environmental and humanhealth.

86 20 85290706.

In recent decades, BFRs, which are widely used to manufacture tex-tiles, furniture, electronics, and automotive interiors, have attractedconsiderable research interest. China is a large BFR manufacturer andconsumer because of the rapid economic growth that occurred in recentdecades (Zhou, 2006). The main BFRs currently used in China includetetrabromobisphenol A (TBBPA), polybrominated diphenyl ethers(PBDEs), decabromodiphenyl ethane (DBDPE), and hexabromocyclo-dodecanes (HBCD). The estimated domestic production volumes oftechnical TBBPA, deca-BDE, DBDPE, and HBCD in 2006 were 38,000,20,000, 12,000, 4500, and 4000 t, respectively (Xiao, 2006). 1,2-bis(2,4,6-tribromophenoxy)ethane (BTBPE) is produced and used inChina (Cai, 2008), but information on its production and consumptionis not available. Recent investigations from China have raised concernsabout the extensive presence of BFRs in the environment and in

Fig. 1.Mapof Shouguang in Shandong Province, northernChina, and locations of samplingsites. The red dashed-line circle represents the main industrial zone, the red solid-linecircle represents the downtown area, with the remaining land comprised mostly ofvillages and farmlands.

48 Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

human tissue (Ma et al., 2012; Shi et al., 2013; Zhu et al., 2009). How-ever, information on BFR contamination from the BFR manufacturingareas in China is limited.

Although the production of certain OCPs has been banned fordecades in most countries, these compounds frequently remain indetectable amounts in the environment because of their significanthistorical use and global movement (Szlinder-Richert et al., 2008). Asa large agricultural country, China was a major producer and consumerof OCPs. For example, the total production amountswere approximately4.5 × 106 t for HCHs and 0.44 × 106 t for DDTs between the 1950s and1980s (Zhang et al., 2009). China banned the agricultural use of certainOCPs starting in the early 1980s, but certain small-scale applications arestill permitted (Wang et al., 2007).

PCBs were used in a variety of industrial and commercial applica-tions, such as in transformers, capacitors, and plasticizers; however,the production and usage were discontinued in the 1970s. Recentresearch indicated that buildings, paints, and old appliances found inurban areas are still significant sources of PCBs in the environment inEurope and North America (Diamond et al., 2010; Jamshidi et al.,2007). In certain developing countries, the primitive dismantling ofelectronic and electrical waste (e-waste) has been demonstrated to bea substantial emission source of PCBs (Wang et al., 2011; Wong et al.,2007). Consequently, elevated levels of PCBs in the environment andin human tissue have been reported in parts of Africa and Asia in recentyears (Breivik et al., 2011).

Shouguang town is located in the eastern coastal area of ShandongProvince, which is an important industrial region in North China.Shouguang became the largest BFR-manufacturing center in Chinastarting in the early 2000s because of the region's abundant source ofbromine. However, little information on BFR contamination in thisregion is currently known. Shouguang has also been a famous vegetableproducing center for China (‘Chinese hometown of vegetables’) sincethe mid-1980s. OCPs were very likely used in large quantities in thisregion in the past, despite the consumption data not being available.Soils are a significant sink for POPs and provide a source of POPs trans-ferred in the food web. In the present study, concentrations of BFRs,OCPs, and PCBs, which are currently or have historically been used,were measured in the farmland soils in this BFR-manufacturing andvegetable center in China to examine the contamination status ofthese chemicals in the region. The compositions and spatial distribu-tions were investigated to understand their environmental processesand sources. An estimate of the burden of these POPs residing in thesurface soil was made to evaluate the environmental impact from themanufacturing or application activities in this region.

2. Materials and methods

A detailed description of the materials and instrumental analysis isprovided in the Supplementary material.

2.1. Sample collection

Shouguang has an area of 2180 km2, and its main industries includemarine chemical manufacturing, papermaking, and machine building.There are more than 15 BFR-manufacturing plants in Shouguang,which are mainly concentrated in the industrial zone in the northernarea (Fig. 1). The eastern coastal area of Shandong Province, whereShouguang is located, is also an important industrial region that is char-acterized by chemical manufacturing, household appliance manufac-turing, and machine building industries. As a famous vegetableproducing center of China, Shouguang has 940 km2 of farmland (includ-ing 560 km2 of vegetable land) and an annual vegetable productionvolume of 4 × 106 t in 2012. The land use in the Shouguang is primarilyfor agricultural, residential/commercial, and industrial purposes.

Samples were collected from the agricultural soils, which are pri-marily sand–loam fluvo-aquic soil with a small portion of saline-alkali

soil (near the shore) (Tang et al., 2011). Specifically, thirty-eight soilsamples covering the entire territory of Shouguang were obtained inMay 2008 (Fig. 1). The top 5-cm layer of soil was collected using apre-cleaned stainless steel shovel. Each sample was a combination offour or five subsamples (at a distance of 3–4m) from an area of approx-imately 100 m2. Soil samples were freeze-dried, ground, and homoge-nized by sieving through a stainless steel 80-mesh sieve and stored at−20 °C until extraction.

2.2. Extraction and cleanup

Approximately 20 g samples were Soxhlet extracted with anacetone:hexane (1:1, v:v) mixture for 48 h. Activated copper wasadded to remove elemental sulfur before extraction. The extractwas concentrated to 1–2 mL using a rotary evaporator and solventexchanged to hexane and was then divided into two subsamples. Onesubsample was spiked with 13C-α-, β-, and γ-HBCD for analysis ofTBBPA and HBCD, and the other was spiked with BDE77, BDE181,BDE205, and 13C-BDE209 for analysis of the other BFRs and withPCB30, PCB65, and PCB204 for analysis of the OCPs and PCBs. Bothsubsamples were purified through a silica column packed with neutralsilica (8 cm, 3% deactivated), 44% sulfuric acid silica (8 cm), and anhy-drous sodium sulfate (1 cm) from the bottom to top. The effluentswere concentrated to 200 μL under a gentle nitrogen stream. A knownamount of internal standards (d18-α-, β-, and γ-HBCD, 4-F-BDE67, 3-F-BDE153, BDE118, and BDE128, and PCB24, 82, and 198) was addedto the final extracts prior to instrumental analysis.

49Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

2.3. Instrumental analysis

Tri- through hepta-BDEs (BDE28, 47, 66, 85, 99, 100, 138, 153, 154,and 183), pentabromotoluene (PBT), hexabromobenzene (HBB), penta-bromoethylbenzene (PBEB), and polybrominated biphenyls 153(PBB153) were analyzed by an Agilent 6980N gas chromatographcoupled with an Agilent 5975Bmass spectrometer (GC–MS) using elec-tron capture negative ionization (ECNI) in the selected ion monitoringmode (SIM). A DB-XLB (30 m × 0.25 mm i.d., 0.25 μm film thickness;J&W Scientific, CA) capillary column was used for the separation. Forocta- through deca-BDEs (BDE196, 197, 201, 202, 203, 206, 207, 208,and 209), DBDPE, BTBPT, and PBB209, the analysis was performedwith a Shimadzu 2010 GC–ECNI-MS system equipped with a DB-5HT(15 m × 0.25 mm i.d., 0.10 μm film thickness; J&W Scientific, CA)column. The method detection limits (MDL), defined as the meanblank mass plus three standard deviations, were between 0.002 and0.2 ng/g based on the average soil mass.

PCBs and OCPs (α-,β-,γ-, and δ-HCH, p, p′- and o, p′-DDD, p, p′- ando, p′-DDE, p, p′- and p, p′-DDT, p, p′-DDM and DDMU, heptachlor, hep-tachlor epoxide, aldrin, dieldrin, endrin, endosulfans I and II, endosulfansulfate, endrin aldehyde, endrin ketone, and methoxychlor) wereanalyzed by an Agilent 7890A/5975C GC–MS in electron ionization(EI) mode using a DB-5 ms capillary column (60 m × 0.25 mm i.d.,0.25 μm film thickness; J&W Scientific, CA) for separation. The MDLranged from 0.01 to 0.03 ng/g for PCBs and from 0.003 to 0.026 ng/gfor OCPs.

TBBPA and HBCD isomers (α-, β-, and γ-) were analyzed by anAgilent 1200 series liquid chromatography (LC) system coupled withan Agilent 6410 electrospray triple quadrupole mass spectrometer(MS) in electrospray ionization (ESI) negative ion mode. An XDB-C18(4.6 × 50 mm, 1.8 μm; Agilent) column was used for separation. TheMDLs were 0.002 ng/g for TBBPA and 0.003–0.007 ng/g for HBCDisomers.

2.4. Quality control

A procedural blank was run with each batch of samples. Only traceamounts of BDE47, 206, 207, 208, and 209 were detected in theprocedural blanks, and they were subtracted from the amountsthe sample extracts. The surrogate recoveries were 115 ± 10.6%for BDE77, 101 ± 7.6% for BDE181, 98.9 ± 21.4% for BDE205, 117 ±25.0% for 13C-BDE209, 101 ± 20.4% for PCB30, 112 ± 20.6% for PCB65,125 ± 27.8% for PCB204, 103 ± 12.8% for 13C-α-HBCD, 102 ± 13.2%

Fig. 2. The patterns of relative contributions of BFRs, OCPs, and PCBs in the surface soils in Sh

for 13C-β-HBCD, and 100 ± 13.0% for 13C-γ-HBCD in all the samples.The recoveries of target compounds were 74.7%–122% (standarddeviations b 11.9%) in the spiked blanks and 78.8%–142% (b18.0%) inmatrix spiked samples. The relative standard deviations were within1.71%–17.8% for all compounds in the triplicate samples. Reportedconcentrations were not surrogate-recovery corrected.

2.5. Data analysis

Data analysis, including the Pearson product moment correlation,t-test, and one-way ANOVA, was performed on Sigmaplot 12.0. Concen-tration data that did not follow a normal distribution were log-normalized. A confidence level of 95% was used for the statistical testsand correlation analysis.

3. Results and discussion

3.1. Concentrations and compositions of BFRs

The total concentrations of BFRs in the soil ranged from 39.9 to8145 ng/g, with a geometric mean (GM) of 668 ng/g on a basis of dryweight from the present study (Table S1). The concentrations of totalBFRs and most of the individual compounds followed a log normaldistribution. The total organic carbon (TOC) contents of the soils variedbetween 0.19% and 1.58%, with a mean of 0.65%.

PBDEshad a concentration range of 18.8–5179ng/g (GM=275 ng/g)and were the primary class of BFRs, accounting for more than 50% ofthe BFRs in over 60% of the samples (Fig. 2). PBDEs were dominatedby BDE209 (the predominant component of the technical deca-BDEmixture), with an average percentage contribution of 94%. In contrast,PBDEs from the technical penta- and octa-BDE mixtures had muchlower concentrations in the soils and followed the decreasing orderof BDE183 N BDE153 N BDE47 N BDE99. This finding was consistentwith the deca-BDE mixture having been the predominant technicalPBDE product used and manufactured in China (Xiao, 2006). In fact,the manufacture of technical penta and octa-BDE mixtures in Chinahas not been reported. The composition patterns of tri- throughhepta-BDEs (with elevated contributions from BDE28, 47, and 153),which differed significantly from those in the technical mixtures(Fig. 3), suggest that the primary origin of these congeners in the soilmay not be from technical penta- and octa-BDE mixtures. Instead,these PBDEs were largely derived from technical deca-BDE mixturesas trace impurities and/or from the degradation of highly brominated

ouguang. The black bar (Others) refers to the sum of BTBPE, PBT, PBEB, HBBs, and PBBs.

Fig. 3. The congener profiles of PBDEs derived from technical penta-, octa-, and deca-BDEmixtures in the soils and corresponding technical mixtures. Deca-BDE1 is a productmanufactured in China.

50 Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

BDEs. Thisfindingwas supported by the positive significant correlationsbetween the concentrations of these PBDEs and the deca-BDEs foundin the soil (p b 0.001, Table S3). Moreover, photolytic degradation bysunlight and microbial degradation of highly brominated PBDEs underanaerobic conditions has been observed (He et al., 2006; Hua et al.,2003). It has been suggested that lower brominated congeners havegreater tendency for atmospheric transport than highly brominatedones (Wania and Dugani, 2003). These factors would also lead tochanges in the composition of PBDEs in the soil. It is worthwhile tonote that despite the PBDE degradation potential in the soil, the muchlower concentrations of tri- through hepta-BDEs and the stable deca-BDE compositions in the soil (Fig. 3) demonstrated that BDE209 ispersistent and has not been subject to significant degradation.

Although TBBPA accounted for nearly half of theproduction volumesof BFRs in China circa 2006, the concentrations of TBBPA were lowerthan those of PBDEs in most soils in the studied area and varied signifi-cantly from 1.64 to 7758 ng/g with a GM of 107 ng/g. It is not clearwhether this resulted from differences in the production quantities ofdeca-BDE and TBBPA manufactured in this region (though the quanti-ties are not known). However, a likely reason for this finding resultedfrom differences in the physicochemical properties between TBBPAand other BFR compounds. TBBPA has a much higher water solubilitythat increases exponentially with increasing pH (0.171–4.16 mg/L)(Kuramochi et al., 2008) compared to other BFR compounds (e.g., 0.02–0.03 mg/L for BDE209) (EHC, 2009). Shouguang is a coastal regionthat has a highly saline-alkali soil (Liu, 2012). Consequently, there is agreater tendency for TBBPA to migrate down to deeper soils alongwith water than there is with other BFRs, thus reducing its concentra-tion in the surface soil (Arnon et al., 2006). Alternatively, TBBPA maybe covalently bound to soil humic substances through the hydroxylgroups hampering their extraction to organic solvents (Luo et al.,2010), although a significant correlation was not observed betweenthe TBBPA concentrations and TOC contents (Table S3).

As a replacement for deca-BDE products, DBDPE showed concentra-tions quite lower than PBDEs in the soil, ranging from 12.0 to 344 ng/gwith a GM of 66.4 ng/g. One explanation was its short manufacturehistory in China (from 2003), with another being that a considerablenumber of DBDPE manufacturing plants are located in East China. Theconcentrations of HBCD were also lower, with a GM of 5.91 ng/g(between 0.30 and 280 ng/g). In most soil samples, γ-HBCD was themost abundant isomer. HBCD in soils at the point-source sites (withconcentrations of 30.4–280 ng/g, which were much higher than othersites, Fig. 4) was expected to originate from direct emission by themanufacturing activities. However, HBCD diastereoisomer composi-tions in the soil (on average, 12.0% α-HBCD, 18.5% β-HBCD, and 67.5%γ-HBCD) were different fromwhat was reported in the Chinese techni-cal HBCD product (12.0% α-HBCD, 11.0% β-HBCD, and 77.0% γ-HBCD)(Yu et al., 2008), with enrichment of only β-HBCD, which has rarelybeen reported in the literature (Covaci et al., 2006). This could be attrib-uted to the HBCD product manufactured in this region having differentdiastereoisomer compositions. Another possible explanation is theapplication of industrial sewage sludges to the soils because elevatedβ-HBCD contributions in sewage sludges were observed (Morris et al.,2004). There was also a significant difference in the stereoisomerpatterns in soils between the point-source and non-point-source sites(23.8% α-HBCD, 19.7% β-HBCD, and 56.4% γ-HBCD) (p b 0.028) withan enrichment ofα-HBCD and depletion ofγ-HBCD (Fig. 4). Enrichmentof theα-HBCDdiastereoisomer in the environment, especially in the air,relative to the technical formulation has frequently been found (Covaciet al., 2006). It is unknownwhether thiswas because of the difference indiastereoisomer-specific vapor pressures of HBCD, which to our knowl-edge have not been reported. Another explanation was a shift fromγ-HBCD to α-HBCD induced by photolytic isomerization and/or prefer-ential degradation of γ-HBCD, which was previously observed (Daviset al., 2005; Harrad et al., 2009). Other BFRs, including BTBPE, PBT,PBEB, HBBs, and PBBs, showed the lowest concentrations and lowdetection frequencies in the soil, indicating that these BFRs were notmanufactured in this region. Little information was found on themanu-facture of these BFRs in China.

The concentrations of PBDEs in the present study were markedlyhigher than the PBDE levels in the Chinese farmland soil (except forsoil from Shanghai) but comparable to levels in certain urban soils andsoils from e-waste dumping sites in China (Fig. S1). The soil PBDEconcentrations in the present study were also high compared to therecently reported levels in other locations in the world. Currently,limited data are available for other BFRs in soils. The concentrations ofTBBPA were at levels in the range of 0.30–32.2 ng/g in soils from Spain(Sánchez-Brunete et al., 2009) and ranged from nondetectable (nd) to5.6 ng/g in farmland soil and 26–104 ng/g in soils from an e-waste site

Fig. 4. The HBCD isomer patterns in soils from point sources (S7, 8, 9, 12, 34, 36, 37, and 38) and non-point sources in Shouguang.

51Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

in Beijing, China (Xu et al., 2012). HBCD mean concentrations of 0.22–9.99 ng/g in industrial soils from South China (Gao et al., 2011) and amean concentration of 0.02 ng/g in background soils in East Chinahave been reported (Meng et al., 2011). The HBCD concentrations insoils at open waste dumping sites in Asian developing countries (withmeans of 0.04–0.54 ng/g) (Eguchi et al., 2012) were also low. Highlevels have been found in soils from point-source sites in Europe(110–23,200 ng/g) (Petersen et al., 2004). The DBDPE concentrationsin farmland soils from South China (nd–35.8 ng/g) (Shi et al., 2009)and in various soils from Indonesia (nd–7.6 ng/g) (Ilyas et al., 2011)were both lower than those in the present study (average 111 ng/g).

3.2. Concentrations and compositions of OCPs and PCBs

The concentrations of OCPs in the soils ranged from 7.17 to 171 ng/g(GM = 32.3 ng/g) and their levels in most samples were lower thanthose of PBDEs and TBBPA, comparable to those of DBDPE, and obviouslyhigher than the levels of other BFRs (Table S2). TheOCPswere dominatedby HCHs (55 ± 22%) and DDTs (37 ± 21%), which were detected in allthe samples, with GMs of 16.5 and 10.2 ng/g, respectively. In four of the38 samples, the concentrations exceeded China's Soil EnvironmentQuality Standard for both HCHs and DDTs (50 ng/g) (GAQS, 1995).Dieldrin and endosulfan sulfate were detected in less than 10% of thesamples, indicating a small use of these pesticides in the region. Theaverage level of DDT residues in the present study was close to themedian of the mean DDT concentrations in soils in China and otherlocations in the world that were recently reported. However, the HCHlevel in the present study was higher than most HCH values in thesesoils (Fig. S2). This reflected the extensive use of technical HCH orlindane in this region.

The isomer composition of HCHs consists of 31.5 ± 11.6% β-HCH,30.2 ± 15.2% δ-HCH, 27.6 ± 6.37% γ-HCH, and 10.6 ± 4.89% α-HCH.This pattern differed substantially from those for the technical HCHsand lindane formulations, which were dominated by α-HCH (60%–70%) and γ-HCH (N99%), respectively. It has been found that the HCHisomers can be degraded under both aerobic and anaerobic conditionsin soils (Wu et al., 1997). The result is in agreement with the removalrates of HCH isomers in agriculture soils (α-HCH N γ-HCH N δ-HCH ≫β-HCH) observed by Chessells et al. (1988). This suggests that historiccontamination was largely responsible for the HCH residues in the

present soils. The ratios of α-HCH/γ-HCH (4–7 for technical HCH andnearly zero for lindane) are frequently used to differentiate the possibleorigins of HCHs from technical HCH or lindane (Zhang et al., 2004). Theratios ofα-HCH/γ-HCH ranged from 0.11 to 0.78 in the soils in the pres-ent study, which were lower than most of the ratios recently found inChinese soils (Fig. S3), implying that lindane was continuously used inthis region and resulted in a continuous input of γ-HCH into the soil.

The technical DDT is composed of 77% p,p′-DDT, 15% o,p′-DDT, andtrace amounts of DDE and DDD (EHC, 1979). For the soils in this study,p,p′-DDT (44.4 ± 20.3%) and p,p′-DDE (33.2 ± 15.3%) constituted themajority of the DDTs, followed by o,p′-DDT (11.1 ± 13.9%) and o,p′-DDE (6.99 ± 9.54%). DDE and DDD are the major metabolites andbreakdown products of DDT under aerobic and anaerobic conditions,respectively, in the environment. Thus, the ratio of p,p′-DDT/(p,p′-DDE + p,p′-DDD) is typically used as an indicator of the resident timeand degree of degradation of p,p′-DDT in the environment. The ratios(0–4.86) in the present study were greater than 1 in 70% of the samples(Fig. S4), suggesting a fresh input of DDT into the soils. In addition, theratios of o,p′-DDT/p,p′-DDT in the soils ranging from 0 to 1.5 (with amean of 0.28) were much closer to that of the technical DDT (0.2)than dicofol (~7) (Guo et al., 2009), indicating that application of dicofolmade a small contribution to the DDT contamination in the soils in thisregion compared to other input sources. A further degradation productof DDD, p,p′-dichlorodiphenylmethane (DDM), was detected inmost ofthe samples (68%) with a concentration of up to 2.47 ng/g.

The concentrations of PCBs in the soils ranged from1.46 to 19.2 ng/g,with an average of 5.89 ng/g (Table S2). The levels of legacy PCBs weresignificantly lower than those for BFRs and OCPs in the soils. This wasattributed to the small production and consumption amount in China(approximately 10,000 t between 1965 and 1974) (Xing et al., 2005).The PCB levels were within the concentration range in soils recentlyreported from East, Northeast, and North China (average 1.07–35.5 ng/g) (Teng et al., 2008; Wang et al., 2008, 2010; Wu et al., 2011;Zhang et al., 2007) and comparable to the PCB levels in the global back-ground soil (average 5.41 ng/g) (Meijer et al., 2003). Nevertheless, thevalues in the present study were higher than the PCB concentrations(0.14–1.84 ng/g, average 0.52 ng/g) in soils across China (includingurban, rural, and background soils) (Ren et al., 2007). The PCB homologcomposition was similar to that in the Chinese technical PCB productbut different from that in soils across China (Fig. S5) in which tri-,

52 Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

tetra-, and di-PCBs constituted the majority of PCBs (Ren et al., 2007).These findings indicated that the soils in this region were subject tohigher PCB pollution than most Chinese soils because of the intensiveindustrial activities in and around Shouguang region.

3.3. Spatial distributions

Plots in Fig. 5 display the spatial distributions of the target contami-nants (categorized by PBDEs, TBBPA, DBDPE, HBCD, OCPs, and PCBs) insoils in the studied area, which were predicted by the ordinary Kriginginterpolationmethod. High contamination of TBBPA in the soil was con-centrated in the east of the economic development zone. Comparedwith TBBPA, soils in the entire economic development zonewere highlycontaminated with PBDEs, indicating that deca-BDE mixture wasmanufactured extensively in the economic development zone. The spa-tial distribution of DBDPE contamination had a significant resemblanceto that of PBDEs, suggesting that the manufacture of the technicaldeca-BDE mixture was most likely replaced by DBDPE in the BFR-manufacturing plants in Shouguang. The distribution of HBCD followedthat of TBBPA, displaying a high contamination east of the economicdevelopment zone.

The OCP concentrations had less spatial variation in the soils com-pared to the BFRs and showed a spatial distribution that was generallyopposite to that of the BFRs, with relatively high contamination occur-ring in soils outside the economic development zone. In fact, negativecorrelations were found between concentrations of OCPs and BFRs(Table S3). This is not surprising considering that Shouguang is animportant agricultural area and pesticideswere commonly used in agri-culture. Similarly, PCBs also showed a relatively uniform distribution,

Fig. 5. Spatial distributions of PBDEs, TBBPA, DBDPE, HBCD, P

suggesting diffusive sources to the soils. This was supported by thepositive correlations (r= 0.646, p b 0.001) between the PCB concentra-tions and TOC contents in the soils. There was also a positive correlation(r = 0.650, p b 0.001) between OCPs and TOC. A significant reason forthis is the tendency to partition organic matter for hydrophobic organiccompounds after long-term residence in the soil. The result alsosuggests that organic matter plays a vital role in determining the distri-bution PCBs and OCPs in the soils. However, there was a lack of correla-tion betweenmost BFRs and TOC and significantly negative correlationswith TOC were found for DBDPE and HBCD (r = −0.348, p = 0.032;and r = −0.341, p = 0.036, respectively). This result confirmed thatthe contamination of BFRs in the soils was primarily derived frompoint sources (manufacturing activities).

3.4. Mass inventory

The soils in the studied region (Shouguang) were divided into 45compartments, each of which had an area of approximately 44 km2

(Fig. 1). The inventory (I, t) was calculated by the following equation:

I ¼Xn

i¼1CiAdρ; ð1Þ

where n is the land segment number, Ci is the average soil concentrationof the pollutants (ng/g) within segment i, which was estimated by theconcentration (or average concentration) at the contained site(s), A isthe area of each compartment (km2), d is the soil thickness (cm), andρ is the soil bulk density (g/cm3). With an assumed soil density of1.43 g dry soil per cm3 wet soil (He et al., 2010) and a soil thickness of20 cm, the inventories of BFRs, OCPs, and PCBs in the agricultural soil

CBs, and OCPs (ng/g) in the surface soils in Shouguang.

53Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

(1100 km2) were 373, 15.4, and 2.1 t, respectively. Based on thismethod, their mass inventories in the soils for the entire region wereestimated to be 1042, 26, and 3.7 t, respectively (Table 1). This demon-strated that during the past several years, there were approximately635, 344, and 61 t of PBDEs, TBBPA, and DBDPE, respectively, releasedinto local soils from manufacturing of these BFR products. Manufactur-ing plants in this region are a very significant source of BFRs in theenvironment; however, the primary input pathway (e.g., irrigation oratmospheric deposition) to the soil is not clear. The estimated PBDEinventory in the soil in the present study was much larger than whatwas found (~48.5 t) in the Pearl River Delta, South China, an importantcity cluster with a much larger area (Zou et al., 2007). The results alsoindicated that after their ban for nearly 30 years in China, approximately7.4 t of HCH and 7.2 t of DDT residues remain in the surface agriculturalsoil.

4. Conclusion

This work provides information on the contamination of BFRs, OCPs,and PCBs in the surface soils from a BFR-manufacturing region and largevegetable farming center of North China. The levels of BFRs have mark-edly exceeded those of historically used OCPs as a result of rapid growthof the BFR-manufacturing industry in the Shouguang region in recentdecade. PBDEs, TBBPA, and DBDPE were the primary BFRs, which isconsistentwith their production and use in China. However, continuousinputs of lindane and technical DDT into the soil were still observed. Ourresults indicated that approximately 1042 t of BFRs has been releasedinto local soil from manufacturing facilities since the early 2000s andthat approximately 26 t of OCP residues remains in the surface soilafter their 30-year ban in China. Further research is urgently requiredto ascertain the air emission factors and inventories of BFRs resultingfrom their manufacture, emission and transport pathways into theatmosphere and soil, the impact of BFRs on the surroundings, and therisk of human exposure to these pollutants.

Conflict of interest

We state that there are no conflicts of interest (financial, personalor other relationships) with other people or organizations that couldinappropriately influence, or be perceived to influence, their work.

Acknowledgments

This study was financially supported by the National ScienceFoundation of China (Nos. 41273115, 41230639 and 41121063). Thisis contribution No. IS-1828 from GIGCAS.

Table 1The mass inventories of BFRs, OCPs, and PCBs in the surface soils (top 20 cm) inShouguang.

Chemical Mass inventory (t) Mass inventory per unit(g m−2)

Agricultural land Entire land Agricultural land Entire land

PBDEs 222 635 0.202 0.318DBDPE 26.3 61.1 0.024 0.031HBCD 6.23 9.44 0.006 0.005TBBPA 118 334 0.107 0.167BFRs 373 1042 0.340 0.521HCHs 7.42 12.9 0.007 0.006DDTs 7.15 11.8 0.007 0.006OCPs 15.4 26.2 0.014 0.013PCBs 2.12 3.70 0.002 0.003

Appendix A. Supplementary data

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.scitotenv.2014.02.023.

References

Arnon S, Ronen Z, Yakirevich A, Adar E. Evaluation of soil flushing potential forclean-up of desert soil contaminated by industrial wastewater. Chemosphere2006;62:17–25.

Breivik K, Gioia R, Chakraborty P, Zhang G, Jones KC. Are reductions in industrial organiccontaminants emissions in rich countries achieved partly by export of toxic wastes?Environ Sci Technol 2011;45:9154–60.

Cai YY. Current major brominated flame retardants. Handbook of modern technology offlame retardarce; 2008. p. 35–45. [in Chinese].

Chessells MJ, Hawker DW, Connell DW, Papajcsik IA. Factors influencing the distributionof lindane and isomers in soil of an agricultural environment. Chemosphere 1988;17:1741–9.

Covaci A, Gerecke AC, Law RJ, Voorspoels S, Kohler M, Heeb NV, et al. Hexabromocyclo-dodecanes (HBCDs) in the environment and humans: a review. Environ Sci Technol2006;40:3679–88.

Davis JW, Gonsior S, Marty G, Ariano J. The transformation of hexabromocyclododecane inaerobic and anaerobic soils and aquatic sediments. Water Res 2005;39:1075–84.

Diamond ML, Melymuk L, Csiszar SA, Robson M. Estimation of PCB stocks, emissions,and urban fate: will our policies reduce concentrations and exposure? Environ SciTechnol 2010;44:2777–83.

Eguchi A, Isobe T, Ramu K, Tue NM, Sudaryanto A, Devanathan G, et al. Soil contaminationby brominated flame retardants in open waste dumping sites in Asian developingcountries. Chemosphere 2012;90:2365–71.

EHC (Environmental Health Criteria 162). ICPS. Retrieved 2009-07-20. bhttp://www.inchem.org/documents/ehc/ehc/ehc162.htm#SectionNumber:2.1/N.

EHC (Environmental Health Criteria 9). DDT and its derivatives. World Health Organiza-tion; 1979 [bhttp://www.inchem.org/documents/ehc/ehc/ehc009.htmN].

Gao ST, Wang JZ, Yu ZQ, Guo QR, Sheng GY, Fu JM. Hexabromocyclododecanes in surfacesoils from e-waste recycling areas and industrial areas in South China: concentrations,diastereoisomer- and enantiomer-specific profiles, and inventory. Environ SciTechnol 2011;45:2093–9.

General Administration of Quality Supervision (GAQS). Inspection and quarantine:quality standard for soil environment, GB 15618–1995; 1995.

Guo Y, Yu HY, Zeng EY. Occurrence, source diagnosis, and biological effect assessment ofDDT and its metabolites in various environmental compartments of the Pearl RiverDelta, South China: a review. Environ Pollut 2009;157:1753–63.

Harrad S, AbdallahMAE, Covaci A. Causes of variability in concentrations anddiastereomerpatterns of hexabromocyclododecanes in indoor dust. Environ Int 2009;35:573–9.

He J, Robrock KR, Alvarez-Cohen L. Microbial reductive debromination of polybrominateddiphenyl ethers (PBDEs). Environ Sci Technol 2006;40:4429–34.

He Y, Hu KL, Huang YF, Li BG, Chen DL. Analysis of the anisotropic spatial variability andthree-dimensional computer simulation of agricultural soil bulk density in an alluvialplain of north China. Math Comput Model 2010;51:1351–6.

Hu GC, Luo XJ, Dai JY, Zhang XL, Wu H, Zhang CL, et al. Brominated flame retardants,polychlorinated biphenyls, and organochlorine pesticides in captive giant panda(Ailuropoda melanoleuca) and red panda (Ailurus fulgens) from China. Environ SciTechnol 2008;42:4704–9.

Hua I, Kang N, Jafvert CT, Fábrega-Duque JR. Heterogeneous photochemical reactions ofdecabromodiphenyl ether. Environ Toxicol Chem 2003;22:798–804.

Ilyas M, Sudaryanto A, Setiawan IE, Riyadi AS, Isobe T, Ogawa S, et al. Characterization ofpolychlorinated biphenyls and brominated flame retardants in surface soils fromSurabaya, Indonesia. Chemosphere 2011;83:783–91.

Jamshidi A, Hunter S, Hazrati S, Harrad S. Concentrations and chiral signatures ofpolychlorinated biphenyls in outdoor and indoor air and soil in a major UK conurba-tion. Environ Sci Technol 2007;41:2153–8.

Kuramochi H, Kawamoto K, Miyazaki K, Nagahama K, Maeda K, Li XW, et al. Determina-tion of physicochemical properties of tetrabromobisphenol A. Environ Toxicol Chem2008;27:2413–8.

Liu XL. Research of Saline-alkali tolerant plant in Weifang coastal area of Shandong. AdvOrganomet Hortic China 2012:446–9.

Luo XJ, Zhang XL, Chen SJ, Mai BX. Free and bound polybrominated diphenyl ethersand tetrabromobisphenol A in freshwater sediments. Mar Pollut Bull 2010;60:718–24.

Ma J, Qiu XH, Zhang JL, Duan XL, Zhu T. State of polybrominated diphenyl ethers in China:an overview. Chemosphere 2012;88:769–78.

Meijer SN, OckendenWA, Sweetman A, Breivik K, Grimalt JO, Jones KC. Global distributionand budget of PCBs and HCB in background surface soils: implications or sources andenvironmental processes. Environ Sci Technol 2003;37:667–72.

Meng XZ, Duan YP, Yang C, Pan ZY,Wen ZH, Chen L. Occurrence, sources, and inventory ofhexabromocyclododecanes (HBCDs) in soils from Chongming Island, the YangtzeRiver Delta (YRD). Chemosphere 2011;82:725–31.

Morris S, Allchin CR, Zegers BN, Haftka JJ, Boon JP, Belpaire C, et al. Distribution and fate ofHBCD and TBBPA brominated flame retardants in north sea estuaries and aquaticfood webs. Environ Sci Technol 2004;38:5497–504.

Petersen M, Hamm S, Schäfer A, Esser U. Comparative GC/MS and LC/MS detection ofhexabromocyclododecane (HBCD) in soil and water samples. OrganohalogenCompd 2004;66:224–31.

54 Z.-C. Zhu et al. / Science of the Total Environment 481 (2014) 47–54

Ren NQ, Que MX, Li YF, Liu Y, Wan XN, Xu DD, et al. Polychlorinated biphenyls in Chinesesurface soils. Environ Sci Technol 2007;41:3871–6.

Sánchez-Brunete C, Miguel E, Tadeo JL. Determination of tetrabromobisphenol-A,tetrachlorobisphenol-A and bisphenol-A in soil by ultrasonic assisted extractionand gas chromatography–mass spectrometry. J Chromatogr A 2009;1216:5497–503.

Shi T, Chen SJ, Luo XJ, Zhang XL, Tang CM, Luo Y, et al. Occurrence of brominated flameretardants other than polybrominated diphenyl ethers in environmental and biotasamples from southern China. Chemosphere 2009;74:910–6.

Shi ZX, Jiao Y, Hu Y, Sun ZW, Zhou XQ, Feng JF, et al. Levels of tetrabromobisphenol A,hexabromocyclododecanes and polybrominated diphenyl ethers in human milkfrom the general population in Beijing, China. Sci Total Environ 2013;452:10–8.

Szlinder-Richert J, Barska I, Mazerski J, Usydus Z. Organochlorine pesticides in fish fromthe southern Baltic Sea: levels, bioaccumulation features and temporal trends duringthe 1995–2006 period. Mar Pollut Bull 2008;56:927–40.

Tang HB, Liao CY, Sun CZ. Spatio-temporal characteristics of soil nutrients in the vegetablegreenhouse in Shouguang of Shan dong Province. Acta Agric Boreali-Occident Sin2011;20:134–7. [in Chinese].

Teng Y, ZhengMK, Luo YM, Gao J, Li ZG,Wu LH. Spatial distribution of soil PCBs congenersin typical area of Yangtze River delta region. Environ Sci 2008;29:3477–82.

Wang F, Jiang X, Bian YR, Yao FX, Gao HJ, Yu GF, et al. Organochlorine pesticides in soilsunder different land usage in the Taihu Lake region, China. J Environ Sci 2007;19:584–90.

WangDG, YangM, Jia HL, Zhou L, Li YF. Levels, distributions and profiles of polychlorinatedbiphenyls in surface soils of Dalian, China. Chemosphere 2008;73:38–42.

Wang H, An Q, Dong YH, Li DC, Velde B. Contamination and congener profiles ofpolychlorinated biphenyls from different agricultural top soils in a county of theTailake Region, China. J Hazard Mater 2010;176:1027–31.

Wang Y, Luo CL, Li J, Yin H, Li XD, Zhang G. Characterization and risk assessment ofpolychlorinated biphenyls in soils and vegetations near an electronic waste recyclingsite, South China. Chemosphere 2011;85:344–50.

Wania F, Dugani CB. Assessing the long-range transport potential of polybrominateddiphenyl ethers: a comparison of four multimedia models. Environ Toxicol Chem2003;22:1252–61.

Wong MH, Wu SC, DengWJ, Yu XZ, Luo Q, Leung AOW, et al. Export of toxic chemicals—areview of the case of uncontrolled electronic-waste recycling. Environ Pollut2007;149:131–40.

WuWZ, Xu Y, Schramm KW, Kettrup A. Study of sorption, biodegradation and isomeriza-tion of HCH in stimulated sediment/water system. Chemosphere 1997;35:1887–94.

Wu S, Xia XH, Yang LY, Liu H. Distribution, source and risk assessment of polychlorinatedbiphenyls (PCBs) in urban soils of Beijing, China. Chemosphere 2011;82:732–8.

Xiao J. A perspective on the development of brominated flame retardants in China; 2006[in Chinese].

Xing Y, Lu YL, Dawson RW, Shi YJ, Zhang H, Wang TY, et al. A spatial temporal assessmentof pollution from PCBs in China. Chemosphere 2005;60:731–9.

Xu T, Wang J, Liu SZ, Lü C, Shelver WL, Li QX, et al. A highly sensitive and selective immu-noassay for the detection of tetrabromobisphenol A in soil and sediment. Anal ChimActa 2012;751:119–27.

Yu ZQ, Chen LG, Mai BX, Wu MH, Sheng GY, Fu JM, et al. Diastereoisomer- andenantiomer-specific profiles of hexabromocyclododecane in the atmosphere of anurban city in South China. Environ Sci Technol 2008;42:3996–4001.

Zhang ZL, Huang J, Yu G, Hong HS. Occurrence of PAHs, PCBs and organochlorine pesti-cides in the Tonghui River of Beijing, China. Environ Pollut 2004;130:249–61.

Zhang JY, Qiu LM, He J, Liao Y, Luo YM. Occurrence and congeners specific ofpolychlorinated biphenyls in agricultural soils from Southern Jiangsu, China. J EnvironSci 2007;19:338–42.

Zhang LF, Dong L, Shi SX, Zhou L, Zhang T, Huang YR. Organochlorine pesticides contam-ination in surface soils from two pesticide factories in Southeast China. Chemosphere2009;77:628–33.

Zhou ZM. Implement of administrative measure on the control of pollution caused byelectronic information products and the exemption of deca-BDE mixture. FlameRetard Mater Technol 2006;4:15–6.

Zhu LY, Ma BL, Li JG, Wu YN, Gong J. Distribution of polybrominated diphenyl ethers inbreast milk from North China: implication of exposure pathways. Chemosphere2009;74:1429–34.

Zou MY, Ran Y, Gong J, Mai BX, Zeng EY. Polybrominated diphenyl ethers in watershedsoils of the Pearl River Delta, China: occurrence, inventory, and fate. Environ SciTechnol 2007;41:8262–7.