Environ. Sci. Technol. 2010, 44, 1987–1991
Levels and Potential Sources of Decabromodiphenyl Ethane (DBDPE) and Decabromodiphenyl Ether (DecaBDE) in Lake and Marine Sediments in Sweden NIKLAS RICKLUND,* AMELIE KIERKEGAARD, AND MICHAEL S. MCLACHLAN Department of Applied Environmental Science (ITM), Stockholm University, SE-10691 Stockholm, Sweden
Received December 07, 2009. Revised manuscript received January 27, 2010. Accepted January 28, 2010.
Decabromodiphenyl ethane (DBDPE) is a brominated flame retardant (BFR) used as a replacement for the structurally similar decabromodiphenyl ether (decaBDE), which is a regulated environmental contaminant of concern. DBDPE has been found in indoor dust, sewage sludge, sediment, and biota, but little is known about its occurrence and distribution in the environment. In this paper, sediment was analyzed from 11 isolated Swedish lakes and along a transect running from central Stockholm through the Stockholm archipelago to the Baltic Sea. DBDPE was present in all samples. In lake sediment, the levels ranged from 0.23 to 11 ng/g d.wt. and were very similar to the levels of decaBDE (0.48-11 ng/g d.wt.). Since the lakes have no known point sources of BFRs, their presence in the sediments provides evidence for long-range atmospheric transport and deposition. In the marine sediment, the DBDPE and decaBDE levels decreased by a factor of 20-50 over 40 km from the inner harbor to the outer archipelago. There the DBDPE and decaBDE levels were similar to the levels in nearby isolated lakes. The results indicate that contamination of the Swedish environment with DBDPE has already approached that of decaBDE, and that this contamination is primarily occurring via the atmosphere.
Introduction Decabromodiphenyl ethane (DBDPE) is a brominated flame retardant (BFR) that has been produced and used for more than 20 years (1). It is a commercially important alternative to decabromodiphenyl ether (decaBDE); both are used as additive flame retardants in different plastic and textile applications. DecaBDE has been shown to be ubiquitously present in the environment (2-5). The usage of decaBDE within the EU was recently restricted in 2008 through a decision by the European Court of Justice (6), which annulled the exemption of this BFR from chemicals that are prohibited in electrical and electronic equipment (which represents ∼80% of total decaBDE usage) because they are likely to pose risks to health or the environment (7). This has strengthened the position of DBDPE as a replacement chemical. DBDPE is structurally similar to decaBDE, which * *Corresponding author phone: +4686747565; fax: +4686747637; e-mail:
[email protected]. 10.1021/es903701q
2010 American Chemical Society
Published on Web 02/10/2010
suggests that it may behave similarly in the environment (the molecular structures are shown in Supporting Information (SI) Figure S1) In view of this and the increasing usage, it would seem prudent to scrutinize the environmental behavior of DBDPE. Measurements of DBDPE in the environment are, however, relatively few. After the discovery of DBDPE in sediment, sewage sludge, and indoor air was reported by Kierkegaard et al. in 2004 (8), DBDPE was also found in sludge from Spain (9) and Canada (10). More recently, DBDPE has been reported in dust (11, 12) and in biotic samples: a benthic food chain from North America (13), Panda bears from China (14), and birds from China (15, 16) and North America (17). A recent screening survey of DBDPE in international sludge samples confirmed that it is a widespread contaminant (18). DBDPE was found in samples from each of the 12 countries included in the survey, and it was concluded that DBDPE may be a worldwide concern. The release of DBDPE from the technosphere was explored in a recent mass balance study of DBDPE in a wastewater treatment plant (WWTP) in Stockholm, Sweden (19). It was shown that DBDPE is being released with the WWTP effluent to Stockholm harbor. Given the hydrophobicity of DBDPE, it would be expected to be sequestered into the sediment, which could then be transported outward through the Stockholm archipelago, leading to contamination of the Baltic Sea. In other studies, long-range transport (20) and atmospheric deposition (21, 22) to remote lake sediment has been shown to occur for decaBDE. Since DBDPE has been found in air (23) and tree bark (24), long-range atmospheric transport and deposition of DBDPE may also be an important source of environmental contamination. In this study we explored these two vectors of DBDPE into the aquatic environment. The influence of emissions to water from an urban center was studied by analyzing DBDPE in marine sediment collected along a transect from Stockholm Harbour to the outer ranges of the Stockholm Archipelago. The influence of long-range atmospheric transport and deposition was explored using sediment collected from isolated Swedish lakes.
Materials and Methods Sampling. Lake Sediment. Surface sediment samples from 11 Swedish lakes were collected in the autumn of 2007 (see SI Figure S2 and Table S1 for the locations and coordinates of the sampled lakes). A core sampler was used to sample the top 0-2 cm of sediment (custom-made by the Swedish University of Agricultural Science, Uppsala). From each lake, sediment from three sites was sampled and mixed. The samples were collected by the Swedish Museum of Natural History within the national Swedish freshwater monitoring program for contaminants in biological samples. The lakes were selected to avoid direct anthropogenic sources of contaminants. More information about the monitoring program is available at http://www.nrm.se/download/ 18.4e32c81078a8d9249800013277/Limniska2002.pdf. After collection, the samples were stored in amber glass jars in a freezer (-20 °C). Prior to extraction the samples were lyophilized, ground, and mixed. Marine Sediment. Sediment cores were collected in the spring of 2008 from seven different locations along a transect from central Stockholm through the Stockholm archipelago (see SI Figure S3 and Table S2 for the locations and coordinates of the sampling sites). Two samples were collected from each location. Sampling was performed using a Gemini twin corer. The distinctly laminated sediment cores VOL. 44, NO. 6, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1987
were subsampled, and the slices corresponding to the years 2004-2006 were transferred to glass jars. These were covered with alumina foil, sealed, and stored in a freezer (-20 °C). Prior to extraction the samples were lyophilized, ground, and mixed. Analysis. Determination of the Loss on Ignition (LOI). The loss on ignition LOI was calculated from the loss in mass of predried sediment as a result of heating to 500 °C for 2 h. Extraction and Cleanup. The method developed for analysis of DBDPE in sludge by Ricklund et al. (18) was used for the sediment samples after some modifications. Between 7.4 and 13.4 g of the dried, ground sediment was spiked with 13C-labeled decaBDE (9 ng) and extracted with dichloromethane (DCM) using an accelerated solvent extractor (ASE 300, Dionex). The ASE parameters were set to 100 °C, two 5 min static cycles, 100% flush volume, and 60 s purge time with nitrogen gas. The extract in the ASE collection bottle was gently reduced in volume under nitrogen gas and transferred to a test tube. The DCM was changed to hexane by evaporating the extract under nitrogen gas to near dryness, with subsequent addition of ∼2 mL of hexane. This procedure was repeated twice. After this, about 8 mL of concentrated sulphuric acid was added to the extract to remove the bulk of coextracted organic material and the mixture was inverted for ∼30 s. Following centrifugation for 10 min at 2000 rpm, the hexane phase was transferred to a new test tube. The sulfuric acid phase was re-extracted with a small amount of hexane (e0.5 mL). Sulfur was removed from the hexane fraction by adding 2 mL potassium hydroxide (0.5 M) in ethanol (99%) and heating in a 45 °C water bath for 20 min. The alkaline treatment was interrupted by the addition of 4 mL water and the hexane phase was transferred to new test tubes. The water-ethanol phase was re-extracted with a small amount of hexane. The extracts were then evaporated to approximately 500 µL and fractionated on a column of 0.6 g preconditioned aminopropyl gel with 0.5 g acid silica (40%) added to the top. The analytes were eluted with 8 mL hexane, of which the fraction between 3-8 mL was collected. This fraction was evaporated with nitrogen to 50 µL. The volumetric standard,13C-labeled 2,2,3,4,4,5,5-heptachlorobiphenyl (13C-CB180, 0.63 ng), was added to the lake sediment samples prior to analysis with GC-LRMS. GC-LRMS Analysis. The samples were analyzed with a gas chromatograph (GC, HP 5890 II, Agilent Technol.) equipped with a split/splitless injector. A 12 m, 0.25 mm i.d., 0.1 µm film thickness DB5MS column (J&W Scientific, Agilent Technologies Inc.) was employed. The extracts were injected splitless at a temperature of 275 °C. The GC temperature program was as follows: start at 80 °C and hold for 3 min; ramp 20 °C min-1 to 200 °C; ramp 6 °C min-1 to 315 °C and hold for 10 min. The transfer line temperature and the ion source temperature were kept at 280 and 180 °C, respectively. The helium carrier gas flow was 1.4 mL min-1. The GC was coupled to a Finnigan MAT SSQ 7000 mass spectrometer, which was operated in electron capture negative ionization (ECNI) mode with NH3 as moderating gas. For DBDPE the ions m/z ) 78.9 and m/z ) 80.9 were monitored. For decaBDE and 13C-decaBDE the ions m/z ) 484.6; 486.6 and m/z ) 494.6; 496.6 were used, respectively. For the volumetric standard, 13C-CB180, the ions m/z ) 405.8 and m/z)407.8 were used. Identification and Quantification. The identification of DBDPE and native and labeled decaBDE was based on the retention time of the fragment ions monitored. Quantification was performed using eight point calibration curves of each chemical. 13C-labeled decaBDE was used as a surrogate standard for the quantification of both decaBDE and DBDPE. Quality Assurance. Exposure of samples to UV-light was minimized by covering all lamps in the laboratory with UV1988
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 6, 2010
protective film. All glassware was preheated to 450 °C and rinsed with acetone before use. The ASE-cups were washed and extracted empty before being filled with sample. The laboratory procedures were tested for background levels of DBDPE and decaBDE. Procedural blank samples that included the lyophilization, extraction, and cleanup were analyzed in parallel to the samples. The recovery of the surrogate standard from the lake sediment samples was quantified using the volumetric standard. To obtain information about the local variation of BFR concentrations along the marine sediment transect, sample pairs of the marine sediment were analyzed. Additional tests made with other samples included the efficiency of the extraction method; the relative recovery of DBDPE to the surrogate standard and the precision of the method.
Results and Discussion Quality Control. Blank Samples. For the lake sediment, four out of five procedural blank samples did not contain any detectable levels of DBDPE (