Environ Monit Assess (2011) 174:259–270 DOI 10.1007/s10661-010-1455-y
Occurrence, distribution, and source of polybrominated diphenyl ethers in soil and leaves from Shenzhen Special Economic Zone, China Pei-Heng Qin · Hong-Gang Ni · Yang-Sheng Liu · Ye-Hong Shi · Hui Zeng
Received: 31 October 2009 / Accepted: 6 April 2010 / Published online: 2 May 2010 © Springer Science+Business Media B.V. 2010
Abstract Polybrominated diphenyl ethers (PBDEs) were measured in soil and three plant species samples taken at different land use areas in Shenzhen China. The concentrations of 7 BDEs (BDE-28, BDE-47, BDE-99, BDE100, BDE-153, BDE-154, and BDE-183) and BDE-209 in the surface soils ranged from 0.23 to 271 and 8.9 to 5,956 ng/g dry weight (dw), respectively. These figures are comparable to that in the soils of electronic waste dismantling sites. BDE-209 was the predominant congener (contributes 85–99% of 8 PBDEs (7 PBDEs plus BDE-209)) in soils. The regression slopes of total organic carbon and individual BDE congeners were rather gentle, indicating that factors other than soil organic matter regulated the soil concentrations. Proximity to sources of deposition processes might be the major factors. In the plant leaves, 7 BDEs and BDE-209 concentrations ranged from 1.29 to 5.91 and 5.49 to 28.2 ng/g dw, respectively. BDE-209 is also
P.-H. Qin · H.-G. Ni · Y.-S. Liu · Y.-H. Shi · H. Zeng (B) The Key Laboratory for Environmental and Urban Sciences, Shenzhen Graduate School, Peking University, Shenzhen, 518055, China e-mail:
[email protected] H. Zeng College of Urban and Environmental Sciences, Peking University, Beijing, 100871, China
the dominant component, but the contribution was much lower compared with that in soils. Bauhinia purpurea Linn. and Michelia alba DC. show some similarities on the uptake of PBDEs, while Ficus microcarpa var. pusillifolia is different from them. The correlations between plant leaf concentrations and predicted gaseous concentrations were moderate, indicating that gaseous concentration did not influence the leaf concentration significantly. Keywords PBDEs · Soil · Plant leaves · Land use · Source
Introduction Polybrominated diphenyl ethers (PBDEs), which were widely used in different kinds of consumer products, have attracted more attention in recent years due to their increased levels in both environment and human tissues (Alaee et al. 2003; Hites 2004). Additionally, because of their bioaccumulation and potential toxicity, the European Union and several US states have banned the use of the penta-BDE and octa-BDE in commercial products (Hassanin et al. 2004), while deca-BDE was partially banned in some regions or countries (Heart 2008). In China, the annual consumption of PBDEs has increased at an estimated rate of 8%, and deca-BDE is the dominant product of the
260
PBDEs technical mixture (Zou et al. 2007). This may further deteriorate PBDE contaminations in China. Soils are a major reservoir and sink for the existing persistent organic pollutants (POPs) (Meijer et al. 2002). Therefore, investigating the occurrence of PBDEs in soil can help us better understand the environmental fate and behavior of PBDEs. A large number of environmental samples from hotspot areas have been analyzed for PBDEs (Matscheko et al. 2002; Leung et al. 2007; Jin et al. 2008a; Yang et al. 2008), but studies on PBDEs in typical urbanized areas are seldom conducted, especially on the correlation between soil PBDEs pollution and urbanization (Harrad and Hunter 2006; Li et al. 2008). The vegetation compartment likely plays an intermediary role between the air and the soil, enhancing air–surface exchange (Gouin and Harner 2003). Herbage (Hassanin et al. 2005) and tree bark (Zhu and Hites 2006) have been used as environmental passive samplers to investigate the long-distance atmospheric transport of PBDEs. Some research showed that the contamination of PBDEs in the leaf samples had good correlation with the soils around them (Yang et al. 2009), while the plant uptake of POPs from soils has been shown in experiments to be negligible (Barber et al. 2004). The relationships between soil and plant samples were complicated and need further studies. But there are seldom reports on PBDE concentrations in leaves, especially in urbanized areas. To better understand soil PBDEs pollution and their transfer among soil, plants, and air in urbanized areas, investigation of PBDEs in soils and plants are very necessary. The Shenzhen Special Economic Zone (SEZ) is one of the original SEZs with an area of 395.81 km2 . In the recent three decades, the SEZ has experienced a fast urbanization and industrialization process. The immediate result was a rapid growth of labor-intensive processing industries. The main industrial products include personal computer, television, hard disk machine, liquid crystal display components, monitor, telephone, habiliment, and so on. All these products have the feasibility to use PBDEs as burning inhibitor in order to meet the flame-resistant standard both at home and abroad (Mai et al. 2005; Heart 2008).
Environ Monit Assess (2011) 174:259–270
This has created severe environmental pollution of PBDEs. However, there is no study on PBDEs in soils and plants from Shenzhen by now. In this context, the objective of this study was to determine the concentrations of eight BDE congeners, including BDE-28, BDE-47, BDE-99, BDE-100, BDE-153, BDE-154, BDE-183, and BDE-209 in soil and plant leaves from the SEZ, China. In addition, the potential sources, correlation between soil PBDEs and total organic carbon (TOC), and correlation between PBDEs in air and leaf were examined to elucidate the transport process of targets among soil, plants, and air in Shenzhen.
Materials and methods Sampling Six land use classes including industrial area, residential area, commercial area, warehouse area, urban park, and country park were chosen as representative sampling sites. Industrial area mainly consisted of manufacturing factories. Urban park consisted of any location with plantation in urban area for leisure. Rural hilly areas and some orchard farms were designated as country park. Large land area used for cargo storage was defined as warehouse area. Considering the influence of municipal administration construction on surface soil, we made sure the soil of each sample site was unchanged for at least 10 years. Twenty-eight surface soil samples (0–5 cm) were collected from different land use types in May 2008 (Fig. 1). Soil samples were collected using a stainless steel spatula cleaned by clean paper before each sampling. Each soil sample was a composite of five subsamples collected from a square block (10 × 10 m). Apart from that, totally 18 plant leaf samples were collected from six sampling sites in different land use areas including three urban afforestation tree species at each sampling site: Bauhinia purpurea Linn., Michelia alba DC., and Ficus microcarpa var. pusillifolia. All samples were wrapped in aluminum foil, sealed in polyethylene bags, and then transported to the laboratory immediately. The leaves were rinsed with distilled water carefully to remove the dust on the surface. After being freeze-dried, the soils
Environ Monit Assess (2011) 174:259–270
261
Fig. 1 Schematic map of the general study area and major sampling cores symbolized by dots; f lags present plant collection sites
were ground and sieved trough a 0.15-mm sieve and the leaves were crushed into powder. Then they were stored frozen at −4◦ C until analysis. Standard materials and reagents The individual standards were obtained from the Cambridge Isotope Laboratory (USA), including eight native PBDEs (BDE-28, BDE-47, BDE-99, BDE-100, BDE-153, BDE-154, BDE183, and BDE-209), eight 13 C-labeled PBDEs (BDE-28, BDE-47, BDE-99, BDE-100, BDE153, BDE-154, BDE-183, and BDE-209) as surrogate standard, and 13 C-BDE-139 as internal standard. All solvents used (hexane, acetone, and dichloromethane) were of pesticide grade (Promochem, Germany). In the present study, 7 PBDEs was defined as the sum of BDE-28, BDE-47, BDE-99, BDE-100, BDE-153, BDE154, and BDE-183. 7 PBDEs plus BDE-209 were designated as 8 PBDEs. PBDEs analysis Then soil samples (1 g) and plant samples (5 g) were extracted for 60 min with dichloromethane (150 ml) by ASE300 (Dionex, America). Before extraction, the samples were spiked with 25-μl (40 pg/μl) surrogate standards. The extracts were concentrated to 5 ml by a rotary evaporator and 5 ml sulfuric acid was added (98% H2 SO4 ) to remove lipid and wax. Then the extract was further cleaned with three multilayer columns. First, the
concentrated extract was purified on a glass column (10 mm i.d.) packed with, from the bottom to top, little absorbent cotton, activated silica (1 g), 2% potassium hydroxide silica (3 g), activated silica (1 g), 44% sulfuric acid silica (4 g), 22% sulfuric acid silica (6 g), 1 g of activated silica, 10% silver nitrate silica (3 g), and anhydrous sodium sulfate (6 g). The fraction was eluted on this glass column with 150 ml of dichloromethane/hexane (1:9 in volume), and the eluant was concentrated to ∼0.5 ml by a rotary evaporator. Then the concentrated eluant was purified again on a glass column (8 mm i.d.) filled from the bottom to top with little absorbent cotton, acid alumina (2 g), and anhydrous sodium sulfate (2 g). The sample was eluted with 100 ml hexane/dichloromethane (1:1 in volume) and then the eluant was concentrated to ∼0.5 ml. Finally, the concentrated eluant was further purified on another glass column (8 mm i.d.) filled from the bottom to top with little absorbent cotton, florisil (2 g), and anhydrous sodium sulfate (2 g) and then the sample was eluted with 120 ml hexane/dichloromethane (98:2 in volume). The eluant was concentrated to ∼0.5 ml by a rotary evaporator and further reduced to 20 μl under a gentle nitrogen stream. An internal standard, 10 μl (100 pg/μl) 13 C-BDE139, was added to the final extract prior to instrumental analysis. The quantification was performed on a highresolution gas chromatograph coupled with a high-resolution mass spectrometer (MSTAION 700-D, JEOL) with an electron impact ion source.
262
Environ Monit Assess (2011) 174:259–270
The electron energy of the mass spectrometric detector was 38 eV, the ion source temperature was 285◦ C, and the mass resolution was above 10,000. A DB-5MS (30 m × 0.25 mm i.d. with 0.25 μm film thickness) capillary column was used to separate seven BDE congeners which were then quantified with an internal calibration procedure. In addition, a DB-5MS (15 m × 0.25 mm i.d. with 0.25 μm film thickness) capillary column was employed to analyze the heavily brominated BDE-209 that was then quantified with an external calibration method. Extract injection was conducted in the splitless mode. The oven and injection temperature were 110◦ C and 280◦ C, respectively. Pure helium was used as the carrier gas, with a column flow of 1.2 ml/min. For 7 PBDEs, column temperature was initiated at 90◦ C (held for 2 min), then ramped to 210◦ C at 25◦ C/min (held for 1 min), 275◦ C at 10◦ C/min (held for 10 min), and 330◦ C at 25◦ C/min (held for 10 min). For BDE-209, the initial temperature was 110◦ C (held for 3 min), then increased to 285◦ C at 15◦ C/min (held for 4 min) and 330◦ C at 10◦ C/min (held for 10 min).
The relative standard deviations were commercial area >
Table 1 PBDE concentrations in soils (in nanograms per kilogram, dw) Classified soil
Sample 7 PBDEs BDE-209 no. Range Mean Median Range
Industrial area 6 Residential area 6 Commercial area 3 Urban park 4 Country park 5 Warehouse land 4 Total 28
6.7–39.5 18.7 0.6–9.0 2.9 1.4–271.0 91.6 0.7–7.1 3.3 0.2–0.8 0.5 1.2–5.4 3.0 0.2–271.0 14.6
14.8 2.0 2.0 2.8 0.4 2.8 2.1
Mean
8 PBDEs Median Range
232.0–5,956.0 1,306.0 331.0 11.4–181.0 71.7 55.5 27.5–109.0 59.2 41.0 17.3–43.5 35.9 41.4 8.9–19.2 13.4 11.6 39.4–338.0 125.0 60.8 8.9–5,955.0 307.4 48.3
Mean
Median
239.0–5,995.0 1,325.0 346.0 12.0–190.0 74.6 57.5 29.0–381.0 150.8 43.0 17.9–50.7 39.2 44.2 9.2–19.9 13.9 12.0 40.6–344.0 127.9 63.6 9.4–6,035.0 336.6 54.5
Environ Monit Assess (2011) 174:259–270
warehouse area > residential area > urban park > country park. Concentrations in industrial areas are approximately 1–2 orders of magnitude higher than those of other land use areas, suggesting the wide use of deca-BDEs in Shenzhen manufacturing industry. It also suggested that PBDE concentrations were strongly associated with land use type. Congener compositions were dominated by BDE-209 (85–99%, except one outlier sample in which BDE-209 contributed 29% of 8 PBDEs) in soils. This is consistent with other studies in the Pearl River Delta (PRD) which concluded that the large consumption of deca-BDE in electrical and electronic equipment manufacturing is likely responsible for the high BDE-209 concentrations in soils in the industrial site (Chen et al. 2006; Luo et al. 2009). The commercial areas were in the center of SEZ where large numbers of electronic goods were sold lead to the more serious contamination in soils than that in the residential area. Soil contamination in warehouse areas may be due to the emission from varying commercial products containing PBDEs in the course of cargo storage and transportation. While the country parks were mainly in suburbs away from the factories and residential area, so the soil contamination was the least, which was in line with a previous study that reported that soil concentrations clearly decreased with increasing distance from the city center (Harrad and Hunter 2006). The comparisons among different studies were difficult because the PBDE congeners reported were different. We tried to compare the results in our study with those of other studies based on the same congeners (Table 2). It is amazing that the levels of BDE-209 obtained in industrial areas in the present study (232–5,956 ng/g dw; average, 1,306 ng/g dw) were in the same order of magnitude as in the electronic waste (e-waste) sites and PBDE production sites in China (Jin et al. 2008b; Luo et al. 2009). The BDE-209 levels in soils collected from the acid leaching sites in Guiyu (average, 1,270 ng/g dw; Leung et al. 2007), in road soils collected near the dismantling workshops in Qingyuan (69–6,320 ng/g dw; average, 1,539 ng/g dw; Luo et al. 2009), and in soils collected around a PBDE production plant in Laizhou (5,258–7,120 ng/g dw; average, 6,189 ng/g dw; Jin et al. 2008a) were comparable to the concen-
263
trations in industrial area soils collected from the greenbelt or bare land in industrial areas where there were various manufacturing factories. But for 7 PBDEs, the average concentration in soils collected from the acid leaching sites in Guiyu was approximately 25 times higher than that from the industrial area in this study. Noticeably, the levels of 5 BDE (BDE-47, BDE-99, BDE100, BDE-153, and BDE-154) in soils collected from the country park was within the ranges of woodland soil concentrations reported in UK and Norway (5 BDE, including BDE-47, BDE-99, BDE-100, BDE-153, and BDE-154; Hassanin et al. 2004). One possible explanation is that decaBDE (BDE-209 contributes 97–98%) is the dominant flame retardant used in the study area. In addition, the feasibility of deca-BDE debromination is very small in the soil (Leung et al. 2007). The levels of PBDEs in Shenzhen soils were higher than that of other areas (Zou et al. 2007) in the PRD. This can be ascribed to the widespread use of various consumer products such as rubber, plastics, and electronic appliances containing brominated flame retardants. Compositional patterns and potential sources Figure 2 presents the PBDE compositions (excluding BDE-209) in different land use soils. The relative abundances of BDE-47, BDE-100, and BDE-183 were generally greater than other congeners in soils (except BDE-183 abundance, which was low in commercial area soils), which was different from the findings of previous studies that BDE-47 and BDE-99 were the predominant compounds in soils (Harrad and Hunter 2006; Zou et al. 2007). BDE-47, BDE-99, BDE-100, BDE-153, and BDE-154 were usually found in the technical penta-BDE mixture, with BDE-47 and BDE-99 being the two predominant compounds (Sjodin et al. 1998; Mai et al. 2005). There was a high possibility that BDE-47, BDE-99, and BDE100 in soils came from the use of penta-BDE in the study area (Mai et al. 2005). But it is unclear why the abundance of BDE-100 was higher than that of BDE-99 in the present study. One possible explanation was that planting might influence PBDE concentrations and composition in soils and plant uptake of POPs can also contribute
Near polyurethane foam manufacturing plant (n = 1) Around the PBDEs manufacturing plant (n = 2) Residential area near PBDEs manufacturing plant (n = 3)
Hale et al. (2002) Jin et al. (2008a) Jin et al. (2008b)
153.2 (71.2–235.2)c 44.9 (22.7–65.6)c
This study This study This study This study This study This study
Luo et al. (2009) Zou et al. (2007) Zou et al. (2007) Leung et al. (2007) Leung et al. (2007) Leung et al. (2007) Leung et al. (2007)
Luo et al. (2009)
76.0e 6,188.8 (5,258–7,120) 1,446.9 (1,080–1,814)
1,306.4 (232.1–5,955.8) 71.7 (11.4–181.0) 59.2 (27.5–109.0) 124.8 (39.4–338.4) 35.9 (17.3–43.5) 13.4 (8.9–19.1)
20.1 (5.3–29.4) 13.8 (2.38–66.6) 70.5 (25.7–102) 1,270 48,600 37.3 2.7
0.39 (0.10–0.58)b 0.98c 7.76c 473.2d 3,994d 4.32d 0.67d 18.7 (6.7–39.5)c 2.9 (0.6–9.0)c 91.6 (1.4–271.5)c 3.0 (1.2–5.3)c 3.3 (0.7–7.1)c 0.5 (0.2–0.8)c
59.8 (46.4–75.2)
4.5 (3.3–6.0)b
Industrial area (n = 6) Residential area (n = 6) Commercial area (n = 3) Warehouse area (n = 4) Urban park (n = 4) Country park (n = 5)
32.2 (3.5–178.5)
10.0 (1.7–28.8)b
Luo et al. (2009)
Luo et al. (2009)
1,539.3 (69.1–6,319.6)
1,149.8 (121.7–3,159.0)b
Road soils near dismantling workshop (n = 29) Farmland soils near dismantling workshop (n = 18) Farmland soils near an E&E equipment manufacturing zone (n = 3) Rural farmland soils (n = 3) Nonpoint source soils (n = 33) Point source soils (n = 3) Soils in acid leaching (n = 3) Combusted residues (n = 3) Soils in rice field (n = 3) Soils in reservoir (n = 3)
Ref. Hassanin et al. (2004) Hassanin et al. (2004) Hassanin et al. (2004)
BDE-209
0.44 (0.02–5.00)a 1.80 (0.08–5.6)a 0.71 (0.09–2.6)a
PBDEs
Grassland soils (n = 21) Woodland soils (n = 21) Woodland soils (n = 24)
Location
b PBDEs
includes BDE-47, BDE-99, BDE-100, BDE-153, and BDE-154 includes BDE-2, BDE-17, BDE-28, BDE-47, BDE-49, BDE-66, BDE-75, BDE-99, BDE-100, BDE-138, BDE-153, BDE-154, BDE-155, BDE-183, BDE190, BDE-196, BDE-197, BDE-203, BDE-206, BDE-207, and BDE-208 c PBDEs includes BDE-28, BDE-47, BDE-99, BDE-100, BDE-153, BDE-154, and BDE-183 d PBDEs includes BDE-28/33, BDE-47, BDE-99, BDE-100, BDE-153, BDE-154, and BDE-183 e PBDEs includes BDE-47, BDE-99, and BDE-100
a PBDEs
Laizhou Bay, Shandong, China
PBDE production sites US
Manufacturing sites Shenzhen SEZ, Guangdong, China
Guiyu, Guangdong, China
PRD, Guangdong, China
Norway E-waste sites Qingyuan, Guangdong, China
Background UK
Region
Table 2 Comparison of PBDE levels in soils between this study and other studies (in nanograms per gram, dw)
264 Environ Monit Assess (2011) 174:259–270
Environ Monit Assess (2011) 174:259–270
265
Fig. 2 Percentage composition of PBDE congeners in the soils of SEZ
to loss of target compounds (like BDE-99) from soils (Mattina et al. 2004; Luo et al. 2009), for most soils we collected were covered with grass or shrub. BDE-183 was the major component in the octa-BDE mixture, and the relatively higher abundances found in the present study indicated that octa-BDE was also used in SEZ, while the quantity was much lower than that of deca-BDE. Correlation between PBDEs and TOC The relationship between the concentrations of individual BDE congeners and surface soil TOC was examined by regression analysis (Fig. 3). Significant correlations were found for BDE-28 (r = 0.42, P < 0.01), BDE-47 (r = 0.54, P < 0.01), BDE-99 (r = 0.55, P < 0.01), BDE-100 (r = 0.50, P < 0.01), BDE-153 (r = 0.63, P < 0.01), BDE154 (r = 0.48, P < 0.01), BDE-183 (r = 0.33, P < 0.05), and BDE-209 (r = 0.37, P < 0.05). The soil organic matter (SOM) fraction is 1.7 times the TOC (Bidleman and Leone 2004). Plots of POPs concentration against %SOM provide useful information on POPs air–soil exchange and their tendency to hop (Hassanin et al. 2004). If regressions give a slope of 0, soil POP concentrations are obviously affected by factors other than SOM, e.g., proximity to sources of deposition processes. In contrast, if a steep slope is obtained, this indicates that the POPs can reenter the atmosphere (preferentially from soils of lower %SOM), redeposit, and tend to be retained in soils of higher %SOM content. Repeated air–soil exchange would result in steeper slopes over time (Gouin et al. 2004). As noted
in Fig. 3, the regression slopes were rather gentle for PBDE congeners, indicating that factors other than %SOM have a major influence on soil concentrations. Proximity to sources of deposition processes might be the major factors. PBDE concentrations in leaves The concentrations of 7 PBDEs and BDE-209 in leaves ranged from 1.30 to 5.90 and 5.5 to 28.2 ng/g dw, respectively (Fig. 4). BDE-209 was also the dominant component in leaves, but the contribution was much lower compared with that in soils due to the lower vapor pressure and low bioaccumulation potential of BDE-209 (Yang et al. 2008). The variation of PBDE levels in leaves from different land use was not significant except that the samples from commercial areas were higher than those from other areas. Shenzhen is located in a subtropical monsoon area with frequent precipitations and wind in spring, so the distribution of PBDE congeners tended to be even in the atmosphere. For PBDE congeners (log KOA > 6 and log KAW > −6), atmospheric deposition can be regarded as the major source of leaves PBDEs (Cousins and Mackay 2001). As for most PBDE congeners, log KOA > 11 (log KOA = 9.46, 10.53, 11.32, 11.18, 11.86, 11.93, and 11.96 for BDE-28, BDE-47, BDE-99, BDE-100, BDE-153, BDE-154, and BDE-183, respectively; Harner and Shoeib 2002), so the atmospheric deposition can be found predominately in the particulate phase (Barber et al. 2004). The even atmosphere levels of PBDEs lead to even leaves PBDE levels via atmospheric particulate deposition. Therefore, all
266
Environ Monit Assess (2011) 174:259–270
Fig. 3 Regression analysis between logarithms of PBDE congeners and log TOC: a BDE-28, b BDE-47, c BDE-99, d BDE-100, e BDE-153, f BDE-154, g BDE-183, h BDE-209
.5
3.0 (a) BDE-28 y=0.3968x-1.3003 r=0.4202 P
