Environ. Sci. Technol. 2006, 40, 2177-2183
Levels of the Flame Retardants Hexabromocyclododecane and Tetrabromobisphenol A in the Blubber of Harbor Porpoises (Phocoena phocoena) Stranded or Bycaught in the U.K., with Evidence for an Increase in HBCD Concentrations in Recent Years ROBIN J. LAW,* PHILIPPE BERSUDER, COLIN R. ALLCHIN, AND JON BARRY The Centre for Environment, Fisheries and Aquaculture Science, CEFAS Burnham Laboratory, Remembrance Avenue, Burnham on Crouch, Essex CM0 8HA, U.K.
Within the U.K. Marine Mammals Stranding Program, analysis of brominated flame retardants began in 1999. Initially, the focus of attention was the brominated diphenyl ethers. Since the withdrawal of the pentamix and octamix polybrominated diphenyl ether (PBDE) formulations from the EU market prior to August 2004, two other highvolume products, hexabromocyclododecane (HBCD) and tetrabromobisphenol A (TBBP-A), have been included. We report the concentrations of these compounds in the blubber of harbor porpoises (Phocoena phocoena) stranded or dying due to physical trauma in the U.K. during the period 1994-2003. Analysis was undertaken using LC/MS method on a diastereoisomer basis. Eighty-five samples were analyzed for HBCD, and 68 of these for TBBP-A. R-HBCD dominated over the other isomers and was detected in all samples analyzed at concentrations ranging from 10 to 19 200 µg kg-1 wet weight. The maximum concentration was about double that reported in earlier U.K. studies. TBBP-A was detected in only 18 samples and at much lower concentrations, from 6 to 35 µg kg-1 wet weight. Investigation of possible time trends indicated a sharp increase in HBCD concentrations from about 2001 onward, which was not confounded by age (length), sex, nutritional status, or location. This may be a result of changing patterns of use of HBCD following limitations on the production and use of two PBDE formulations within the EU and will feed ongoing risk assessment activities.
Introduction Brominated flame retardants (BFRs) are a diverse group of compounds which are added to a wide range of products to prevent them from catching fire. Since 1981, when the first report identifying polybrominated diphenyl ethers (PBDEs) in fish from the River Viskan (1) demonstrated their presence remote from manufacturing sites, some of these compounds have become the focus of environmental studies. The reasons * Corresponding author email:
[email protected]. 10.1021/es052416o CCC: $33.50 Published on Web 03/07/2006
Published 2006 by the Am. Chem. Soc.
for concern regarding certain BFRs (notably the PBDEs, tetrabromobisphenol A (TBBP-A), and hexabromocyclododecane (HBCD) to date) are their persistent, bioaccumulative, and toxic properties and their scale of use. The scope of monitoring and investigative studies has expanded as regulatory action on some products within the EU (the pentamix and octamix PBDE formulations) has resulted in a change in the balance of products in use. Establishment of time trends in the environment for the most commonly used of the brominated flame retardants (PBDEs, TBBP-A, and HBCD) is therefore of great importance. Study of the brominated diphenyl ethers (BDEs) was implemented within the U.K. Marine Mammals Strandings Program in 1999 following an initial U.K. survey which demonstrated high concentrations of BDEs in samples of both sediments and fish (2), and the data have been reported in a number of publications (3-6). Since the restrictions on the pentamix and octamix PBDE formulations in products within the EU market in 2004, attention has shifted to the two other highvolume flame retardants, HBCD and TBBP-A (7). HBCD is an additive flame retardant used primarily in expanded and extruded polystyrene for thermal insulation in buildings, and TBBP-A is a reactive flame retardant used primarily in electronic circuit boards. World production of HBCD and TBBP-A in 2001 totalled 16 700 t and 119 700 t, respectively. In this paper we report the concentrations of these two compounds in the blubber of harbor porpoises stranded or bycaught in the U.K. between 1994 and 2003. Animals categorized as bycaught include those which died as a result of other physical trauma (generally as a result of an attack by bottlenose dolphins (Tursiops truncatus)) as well as those entrapped in fishing gear.
Experimental Section Sampling, Analysis, and Analytical Quality Control. The porpoises sampled were collected within the marine mammal stranding program funded by the U.K. government. Animals which are found stranded or are bycaught and which are classified as freshly dead or only slightly decomposed are taken for postmortem study in order to establish cause of death. Contaminant analyses are conducted in tissues from selected animals with the aim of exploring possible links between contaminant burden and death due to infectious disease. The detailed protocols applied in the postmortem studies are described elsewhere (8). Tissue samples are stored frozen at -20 °C prior to analysis and are then thawed and homogenized prior to extraction. Thoroughly homogenized samples of blubber were mixed with sodium sulfate and extracted with a 1:1 (v/v) acetone:n-hexane mix in a Soxhlet apparatus for 4 h (9). Sample extract cleanup was performed by gel permeation chromatography (GPC) followed by a sulfuric acid treatment. GPC cleanup was performed on a Agilent Series 1100 HPLC system (Agilent Technologies, Waldbronn, Germany) using two Envirogel columns (Waters Corporation, Milford, MA) (19 × 150 mm and 19 × 300 mm) in series and a mobile phase composed of ethyl acetate: cyclohexane [1:1 v/v]. A suitable volume of Soxhlet extract, determined as a function of the lipid content, was measured, and known amounts of surrogate internal standards (d18R-HBCD, d18-β-HBCD, d18-γ-HBCD, and 13C12 TBBP-A) were added. This was concentrated to 1.5 mL, and 900 µL was injected onto the GPC system, previously calibrated using a mixture of corn oil, bis(2-ethylhexyl)phthalate, methoxyclor, perylene, and sulfur so as to establish the fraction collection times. Fractions were collected from ca. 15 to 22 min, evaporated to dryness, and reconstituted to 1 mL using HPLC VOL. 40, NO. 7, 2006 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
2177
FIGURE 1. LC/MS chromatogram of HBCD in the blubber of a porpoise. The d18- substituted HBCD isomers are internal (surrogate) standards. grade n-hexane prior to sulfuric acid cleanup. Sulfuric acid cleanup of the GPC fraction was performed following EPA Method 3665A (10). The final hexane extract was concentrated to dryness and reconstituted to 120 µL using methanol. The quantitative determination of HBCD and TBBP-A was conducted using LC-MS using a Surveyor HPLC system and a LCQ Advantage ion-trap mass spectrometer (ThermoFinnigan, San Jose, CA), equipped with an electrospray (ESI) interface operated in the negative ionization mode. The HPLC column was a 100 mm × 2.1 mm i.d. (3 µm particle size) Hypersil GOLD C18 column protected by a SecuriGuard cartridge, and the injection volume was 15 µL. LC conditions were selected so as to yield baseline resolution for the three HBCD diastereoisomers, and these are given in Supporting Information Table S1 . The limits of quantification (LOQs) for R-, β-, γ-HBCD and TBBP-A were calculated using the lowest calibration level standard (0.010 ng µL-1) for each sample. An example chromatogram is given in Figure 1. In the absence of reference materials certified for HBCD or TBBP-A, cod muscle spiked with 24.8 µg kg-1 total-HBCD (4.0 µg kg-1 R-HBCD; 3.2 µg kg-1 β-HBCD; 17.6 µg kg-1 γ-HBCD) in methanol was used as a laboratory reference material (LRM). Recoveries were found to be 90% ((10%) R-HBCD, 98% ((7%) β-HBCD, and 71% ((10%) γ-HBCD (n)5). Procedural blanks and the LRM were analyzed within all sample batches so that the routine performance of the method could be monitored. The analysis of four samples with high concentrations (SW2001/85D, SW2002/214, SW2003/260, and SW2003/271) was repeated, yielding results within 3%, 5%, 7%, and 18%, respectively, of the original concentrations. The use of 13C-labeled TBBP-A and deuterated HBCD isomers as surrogate internal standards also balances potential ion suppression caused by matrix effects during LC-MS analyses (11).
Results and Discussion Concentrations of HBCD and TBBP-A. The porpoises analyzed were stranded or bycaught on U.K. coasts between 1994 and 2003. Of the 85 animals studied, 63 were classified as bycaught. Dorsal blubber thicknesses determined during 2178
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 40, NO. 7, 2006
postmortem varied from 5 to 32 mm, and blubber lipid contents varied from 56 to 92%. With the exception of three animals, all had relatively high lipid contents (> 80%). Table 1 lists the sampling dates and locations and gives information on sex, body length, and age (where available). Ages were determined by quantification of growth-layer groups from analyses of decalcified tooth sections (12). Eighty-five animals were analyzed for HBCD on a diastereomer basis, and 68 were also analyzed for TBBP-A. Instrumental problems prevented the quantification of TBBP-A in one sample batch. Concentrations of HBCD and TBBP-A determined in blubber are given in Table 1. R-HBCD dominated as in other biota samples (7) and was detected in all samples analyzed at concentrations ranging from 10 to 19 210 µg kg-1 wet weight (10.8 to 21 300 µg kg-1 on a lipid basis), and the range of concentrations for the sum of the three HBCD isomers was similar. The highest concentration was observed in a male porpoise, SW2003/257C, bycaught off the coast of Yorkshire, NE England, in 2003. This animal was 110 cm long and likely to have been a juvenile. TBBP-A was detected in only 18 samples, and the concentrations were much lower, ranging from not detected to 35 µg kg-1 wet weight (up to 38 µg kg-1 on a lipid basis) (Table 1). Comparison with Other Studies. Morris et al. (13) reported concentrations of HBCD and TBBP-A determined using LC/MS in the blubber of nine harbor porpoises from the North Sea, yielding concentration ranges of
