ARTICLE pubs.acs.org/est
Do Temporal and Geographical Patterns of HBCD and PBDE Flame Retardants in U.S. Fish Reflect Evolving Industrial Usage? Da Chen, Mark J. La Guardia, Drew R. Luellen, Ellen Harvey, T. Matteson Mainor, and Robert C. Hale* Department of Environmental and Aquatic Animal Health, Virginia Institute of Marine Science, The College of William and Mary, Gloucester Point, Virginia 23062, United States
bS Supporting Information ABSTRACT: Polybrominated diphenyl ethers (PBDEs) and hexabromocyclododecane (HBCD) are common flame retardants in polymers and textiles. Recognition of the persistent, bioaccumulative, and toxic properties of PBDEs has prompted reductions in their use. In contrast, HBCD has received less scrutiny. The U.S has historically been a dominant BFR consumer. However, the few publications on HBCD in wildlife here suggest modest levels compared to Asian and European studies. In contrast, the HBCD concentrations we detected in U.S. fish are among the highest reported in the world. The temporal trends observed suggest that HBCD use may have risen, and that of Penta-BDE declined, following the 2004 termination of its U.S. manufacture. For example, Hyco River carp collected in 1999 2002 exhibited a mean ∑HBCD (sum of α-, β- and γ-HBCD) concentration of only 13 ng/g (lipid weight basis), but was 4640 ng/g in fish collected in 2006 2007. In contrast, the mean ∑PBDE level in these same fish decreased from 40,700 ng/g in 1999 2002 to 9140 ng/g in 2006 2007. Concentrations of HBCD and PBDEs in several Hyco River fish species exceeded those from rivers less influenced by manufacturing outfalls. Results support the contention that textile-related production, relative to its BFR market share, may release disproportionately large amounts of HBCD to the environment.
’ INTRODUCTION Brominated flame retardants (BFRs) are commonly added to polymers and textiles to reduce their flammability. Of these BFRs, polybrominated diphenyl ethers (PBDEs) have generated the greatest concern to date, due to the persistence, bioaccumulation and toxic (PBT) potentials of some congeners. Studies reveal that PBDEs are now globally dispersed.1 3 Accordingly, the penta- and octa-PBDE mixtures were recently included in the Stockholm Convention on Persistent Organic Pollutants.4 North American usage of PBDEs has historically been the highest in the world due to the demand for polymer products and stringent flame retardancy requirements here.5 For example, in 2001 North America consumed 95%, 44%, and 40% of the total global penta-, deca- and octa-BDE demands, respectively.5 Concentrations of PBDEs in humans and wildlife from North America generally emulate this high usage.1,3 While similar concerns have been raised, the environmental distribution of hexabromocyclododecane (HBCD) has been less extensively studied in North America. North American market statistics for HBCD were last publicly released in 2001 and at that time were about 30% of European demand.5 This lack of transparency regarding recent production has hindered efforts to predict trends in human and wildlife exposure. Some applications of HBCD, r 2011 American Chemical Society
octa- and deca-BDE overlap. Hence, U.S. HBCD usage may have increased following the 2004 cessation in octa-BDE production and pressure to reduce deca-BDE usage. Most HBCD is used in polystyrene thermal insulation boards for the construction industry.6 Greater emphasis on energy efficiency may spur HBCD market demand. Application of HBCD on textiles in the U.S. was reported to constitute less than 1% of total consumption; hence the 2010 EPA Hexabromocyclododecane Action Plan suggested releases from this source are likely modest.6 In contrast, a European HBCD review indicated that textile operations and associated wastewater releases may be important sources to the environment.7 While the apparel segment of the U.S. textile industry has declined dramatically due to foreign competition, BFRs are predominantly used to protect fabrics covering furniture, in curtains and vehicle interiors. Additional releases may arise during the manufacture of high impact polystyrene used in electronics and appliances. Received: April 27, 2011 Accepted: August 24, 2011 Revised: August 19, 2011 Published: August 24, 2011 8254
dx.doi.org/10.1021/es201444w | Environ. Sci. Technol. 2011, 45, 8254–8261
Environmental Science & Technology
ARTICLE
Figure 1. Fish were collected from sites in the Hyco (H1 H3), Dan (D1 D3), and Roanoke River (R1 R7), located within the U.S. mid-Atlantic region. A textile-manufacturing center was located approximately 10 km upstream from H1. Substantial amounts of HBCD and PBDEs entered surface waters via a municipal wastewater treatment plant. The towns of Roanoke, Danville, South Boston (VA) and Roxboro (NC) are also indicated. In 2000, human populations residing in these communities were approximately 95 000, 48 000, 7000, and 8500, respectively.
A better understanding of avenues for HBCD release is necessary in order to control its environmental introduction and reduce associated impacts. To date, the few North American HBCD studies available have mainly focused on wildlife populations exposed via diffuse sources.8 15 Accordingly, reported levels in these organisms have been modest, to an extent mollifying concerns about HBCD here. Little research has evaluated contamination of freshwater fish near U.S. industrial sources and wastewater treatment plant (WWTP) discharges. The greatest contamination would be expected first near such facilities, before disseminating toward more remote locales. Our study objectives here were as follows: (1) assess HBCD and PBDE concentrations in U.S. freshwater fish, especially near facilities handling BFRs; (2) investigate differences in fish concentrations between the early and mid to late 2000s in light of evolving U.S. market demand for BFRs; and (3) investigate the distribution of HBCD diastereomers in fish to better understand their tissue-specific uptake and metabolism.
’ EXPERIMENTAL SECTION Samples. Sampling was centered in southeastern Virginia and northeastern North Carolina (Figure 1). This region is not densely populated or industrialized, but historically has been a center for textile production. Fish and sediments from several of these sites were previously observed to exhibit substantial PBDE contamination originating from local plastics and textile operations.16 Effluents from some of these facilities passed through WWTPs before entering surface waters. A total of 189 individual adult fish were collected via electrofishing from sites in the Hyco, Dan and Roanoke Rivers from May to October of 1999 2002 and 183 additional fish in 2006 2007 (Table 1). The five species sampled were common carp (Cyprinus carpio), flathead catfish (Pylodictus olivaris), channel catfish (Ictalurus punctatus), redhorse sucker (Moxostoma sp.), and gizzard shad (Dorosoma cepedianum). Fish weight and total length were determined. Specimens were then wrapped in solvent-rinsed aluminum foil, sealed in plastic bags, and maintained at 20 °C until
analysis. Hyco River surface sediments were collected from site H2 in 2005 and 2007. Analysis. Fish were filleted, as a major goal of the original sampling was to establish fish consumption advisories for humans for regulated pollutants. Both individual fish fillets and single species composites of fillets from multiple individuals were analyzed (for details, see Supporting Information, SI, Table S1). Liver, adipose, and undigested food (mostly small clams, e.g. Corbicula sp.) remaining in the digestive tracts were also removed from the Hyco River carp to evaluate within-fish HBCD diastereomer patterns. Details of the extraction, purification, and instrumental analyses procedures are described elsewhere (ref 17 and in the SI). Briefly, all samples were freeze-dried and surrogate standards added (13C-labeled α-HBCD; Wellington Laboratories, Ontario, Canada and PCB-204; Ultra Scientific, North Kingstown, U.S.). Dried samples were extracted using accelerated solvent extraction with dichloromethane. Extractable lipids were gravimetrically determined. Extracts were purified by size exclusion, followed by silica gel liquid chromatography. Purified extracts were spiked with internal standards (d18-labeled αHBCD; Wellington Laboratories, Ontario, Canada and decachlorodiphenyl ether; Ultra Scientific, North Kingstown, U.S.). The PBDE congeners (BDE-28, 47, 49, 99, 100, 153, 154, 183, and 209) were determined by GC/MS in the electron-capture negative ionization (ECNI) mode (see SI for details). Separation of HBCD diastereomers was achieved on an ultraperformance liquid chromatograph (UPLC). The UPLC was interfaced with a 3200 Q Trap triple quadrupole/linear ion trap mass spectrometer (MS; Applied Biosystems/MDS Sciex; Toronto, Canada). The MS was equipped with a TurboIonSpray electrospray ionization (ESI) probe operated in the multiple reaction monitoring (MRM) mode for quantitative determination (see SI for details). QA/QC and Data Analysis. Laboratory blanks were analyzed coincident with the samples to assess potential introduction of contaminants. These were below quantitation limits. Spiking tests were conducted to evaluate the recovery of PBDE congeners and HBCD diastereomers (see SI). The mean 8255
dx.doi.org/10.1021/es201444w |Environ. Sci. Technol. 2011, 45, 8254–8261
1510 (1510) 4.6 (4.6) 0.5 (0.5)
