by Z Guo · 2009 · Cited by 37 — with exposure to perfluorooctanoic acid and salts, U.S. EPA, Office of Pollution. Prevention and Toxics, epa/oppt/pfoa/pubs/pfoarisk.pdf

44 KB – 51 Pages

PAGE – 1 ============
EPA/600/R-09/033 March 2009 Perfluorocarboxylic Acid Content in 116 Articles of Commerce Zhishi Guo, Xiaoyu Liu, and Kenneth A. KrebsU.S. Environmental Protection Agency, Office of Research and Development National Risk Management Research Laboratory, Research Triangle Park, NC 27711 and Nancy F. RoacheARCADIS, 4915 Prospectus Dr., Suite F, Durham, NC 27713 National Risk Management Research Laboratory Office of Research and Development U.S. Environmental Protection Agency Research Triangle Park, NC 27711

PAGE – 3 ============
Contents Notice. ii Contents ..iii Abstract iv Acronyms and Abbreviations v Acknowledgments .. vi 1. Introduction 1 2. Conclusions 3 3. Recommendations .. 4 4. Materials and methods . 5 4.1 Sample collection 5 4.2 Verifying the presence of fluorine . 5 4.3 Sample preparation, handling, and storage 6 4.4 Sample extraction and analysis 6 4.5 Quality assurance and quality control .. 8 5. Results.. 9 5.1 Statistics of AOC samples . 9 5.2 Data quality of the analytical results . 9 5.3 Extractable PFCA content in AOC samples .. 10 5.4 Statistics by sample category . 31 6. Discussion 38 6.1 Source strengths 38 6.2 Comparison with literature values .. 40 6.3 Relative abundance of PFCAs .. 40 6.4 Domestic versus imported articles .. 43 6.5 Market trends . 43 7. References 45 Appendix 48 iii

PAGE – 4 ============
Abstract Several recent studies have found elevated levels of perfluorocarboxylic acids (PFCAs) in house dust, suggesting st rongly the presence of indoor sources of these compounds. The main goal of this study was to identify and rank potentially important indoor sources by determining the PFCA content in articles of commerce (AOCs). We analyzed 116 AOC samples purchased from retail outlets in the United States between March 2007 and May 2008 by using a newly developed extraction/analytical method. For these 116 samples, the content of perfluorooctanoic acid (PFOA-C8) ranged from non-detectable to 6750 ng/g, whereas the content of total PFCAs (the sum of C5 to C12 acids) ranged from non-detectable to 47100 ng/g. Given the quantities of articles found in typical homes, it is clear that professional ca rpet-care liquids, pre-treated carpeting, treated floor waxes and sealants, and treated home textile products and upholstery are likely the most important PFCA sources in non-occupational indoor environments. The perfluorochemical-c ontaining AOC market has been in a transition period. Limited data suggest th at the PFCA content in AOCs has shown a downward trend overall. However, definitive confirmation of such a trend will require long-term monitoring. More studies are needed to better understand PFCA transport, exposure routes, and ways to reduce exposures in indoor environments. iv

PAGE – 5 ============
Acronyms and Abbreviations AOC articles of commerce CAS# chemical abstract service registration number HPLC high-performance liquid chromatography IAP internal audit program LC/MS/MS liquid chromatography / tandem mass spectrometry PFC perfluorochemical PFCA perfluorocarboxylic acid PFOA perfluoroocanoic acid PTFE polytetrafluoroethylene RCS recovery check standard RSD relative standard deviation TPFCA total perfluorocarbonyl acids (the sum of C5 to C12 PFCAs) WD wavelength dispersive XRF X-ray fluorescence v

PAGE – 6 ============
Acknowledgments We thank Andrew Lindstrom, Mark Strynar, Shoji Nakayama, and Ed Heithmar of the U.S. EPA and Timothy Begley of the U.S. FDA for technical consultation and assistance; Robert Wright of the U.S. EPA for QA support; Shirley Wasson (retired) and Dean Smith of the U.S. EPA and David Nats chke of Arcadis for operating the X-ray fluorescence spectrometer; and Ivan Dolgov of the U.S. EPA for laboratory support. vi

PAGE – 8 ============
hypothesized that they can contribute to human exposure to PFCAs either directly (e.g., by dermal contact and hand-to-mouth transfer) or indirectly (e.g., inhalation of suspended particles from treated carpet and other interior surfaces). There have been several studies of the PFCA content in AOCs, but most of them report a single compound ΠPFOA. In 2005, Washburn and his colleagues reported the PFOA content in 14 article groups based on theoretical calculations and analytical measurements. Of these groups, pre-treated carpeting and carpeting treated with carpet- care solution had the highest PFOA loadings: 0.2 to 0.6 and 0.2 to 2 mg of PFOA per kg of article, respectively [7]. Studies by other researcherstick cookware, food contact paper, thread sealant tape, and dental floss [17-20] . Data for other PFCAs in AOCs are rather scarce. One study by Sinclair, et al. reported the C5 to C12 PFCA content in three brands of popcorn packaging paper [20] . The main goal of this study was to identify the major PFCA sources in no occupational, indoor environments by determin ing the content of these chemicals in a variety of AOCs and rank them in terms of source strengths. Policy-makers and manufacturers can use the data for risk mana gement purposes. The re sults also provide a snapshot of the current uses of PFCAs on the market and may serve as baseline data for future, long-term monitoring. 2

PAGE – 9 ============
2. Conclusions We analyzed 116 AOC samples purchased fr om retail outlets in the United States between March 2008 and May 2008 to determine the extractable cont ent of C5 to C12 PFCAs using a newly developed extraction/analytical method. To the authors™ best knowledge, this is the first time that the C5 to C12 PFCA contents in a wide variety of AOCs are been reported. The PFCA contents in these samples cover a broad range, from nondetectable to as high as 6750 ng/g for PFOA and from non-detectable to 47100 ng/g for total PFCAs (i.e., the sum of C5 to C12) . In typical American homes with carpeted floors, pre-treated carpet and commercial carpet-care liquids are likely the most significant PFCA sources among the 13 articl e categories studied. For homes without carpeting, floor waxes and stone/tile/wood sealants that contain fluorotelomers products are important sources of PFCAs. Other pot entially important indoor sources include treated home textile, upholstery and apparel, and household carpet/fabric care liquids and foams. The data presented in this report may help explain why PFCAs are frequently detected in house dust. While the exact mechanisms by which PFCAs are transferred from sources to dust are not well characteri zed, existing data strongly suggest that AOCs may contribute to indoor human exposures to PFCAs either directly (dermal contact and hand-to-mouth activities) or indi rectly (inhala tion of dust). 3

PAGE – 10 ============
3. Recommendations Further research is needed in the following areas to better understand the effect of PFCA-containing AOCs on human exposure: (1) PFCA transfer from sources to indoor air and surfaces; (2) the re lationship between AOCs and inhalation exposure; (3) the significance of dermal exposur e; (4) risk management measures for reducing PFCA levels in polluted homes; and (5) monitoring of the market transition on a global scale over an extended period of time. 4

PAGE – 11 ============
4. Materials and Methods 4.1 Sample Collection AOC samples were purchased from retail outlets in the United States. Before collecting samples, a survey was conducted to determine the availability of AOCs that contained or were treated with fluorinated chemicals. Sample candidates were identified based on one of the following claims by the ve ndors: (a) the article contains fluorinated chemicals identifiable by th eir trade names (e.g., Scotchga rd, GoreTex, and Teflon); (b) the article contains fluorinated chemicals identifiable by the chemical names (e.g., polytetrafluoroethylene or PTFE); or (c) the article was identified as having certain properties that are common for articles treated with fluorinated chemicals (e.g., stain resistant, water repellent, and anti-grease). Sample candidates were purchased from local stores in the Raleigh and Durham areas of North Carolina, in Atlanta and New York City, and from on-line stores. While it was not the goal of this study to obtain statistically representative samples for the entire U.S. market, efforts were made to maximize the representativeness of the samples by considering the following factors whenever applicable: article category, trade name of the fluoropolymer or fluorotelomer product, brand name of the article, price range (high, medium, and low), store type (chain stores, high-end stores, low-end stores, and specialty stores), and country of origin. 4.2 Verifying the Presence of Fluorine Sample articles obtained from the market were first screened for the presence of fluorine to exclude those with false claims and those that achieved certain surface properties (e.g., anti-grease) without using fluorinated chemicals. Sample articles containing less than 0.01% fluorine by weight were discarded. Typically, 0.05 to 0.5% of the fluorochemical by weight of the articl e is used to ensure durable repellency [15] . For articles made of polytetrafluoroethylene (PTF E), such as thread sealant tape and some dental floss, the fluorine content can be greater than 70%. Thus, the 0.01% cut-off provided an adequate safe margin to ensure that all sample articles treated with fluorinated chemicals were included for fu rther analysis. The fluorine content was determined by wavelength dispersive (W D) X-ray fluorescence (XRF) spectrometry. Instrument and operating paramete rs are provided in Table 4-1. 5

44 KB – 51 Pages