Research Polybrominated Diphenyl Ethers in House Dust and Clothes Dryer Lint H E A T H E R M . S T A P L E T O N , * ,† NATHAN G. DODDER,† JOHN H. OFFENBERG,‡ MICHELE M. SCHANTZ,† AND STEPHEN A. WISE† National Institute of Standards and Technology, 100 Bureau Drive, Mailstop 8392, Gaithersburg, Maryland 20899, and Environmental Protection Agency, Human Exposure and Atmospheric Sciences Division, Research Triangle Park, North Carolina 27711
Few studies have measured the flame retardants polybrominated diphenyl ethers (PBDEs) in the indoor environment. Here, we report measurements of PBDEs in house dust samples collected from the Washington, D.C. metropolitan area in the United States. Dust samples were analyzed for 22 individual PBDE congeners and our results found PBDEs present in every sample. Concentrations of total PBDEs ranged from 780 ng/g dry mass to 30 100 ng/g dry mass. The dominant congeners observed in the dust samples were congeners associated with the pentaBDE and decaBDE commercial mixtures. Ancillary data were collected on the homes and examined for any correlations with total PBDE concentrations. No correlations were observed with year of house construction, type of flooring (i.e., hardwood vs carpet) or the number of television sets or personal computers in the home. However, a significant inverse correlation (p < 0.05) was observed between the area of the home and the contribution of BDE 209 to the total PBDE concentration in dust. Using estimates of inadvertent dust ingestion (0.02-0.2 g/day) by young children (ages 1-4), we estimate ingestion of total PBDEs to range from 120 to 6000 ng/day. Clothes dryer lint was also sampled and analyzed for PBDEs from five of the homes and were present in all five samples ranging from 480 to 3080 ng/g dry mass. This study demonstrates that PBDEs are prevalent at relatively high concentrations within homes where people, and particularly young children, may be susceptible to exposure.
Introduction Polybrominated diphenyl ethers (PBDEs) are brominated flame retardant chemicals that are applied to many common products found within homes such as furniture, carpeting, mattresses, televisions, coffee makers and hair dryers (1). Because of fire safety standards, many of these products are required to contain flame retardant chemicals which significantly delay the onset and spread of fire. The use of flame retardants in these products is estimated to have helped save * Corresponding author phone: (301)975-8578; fax: (301)977-0685; e-mail:
[email protected]. † National Institute of Standards and Technology. ‡ Environmental Protection Agency. 10.1021/es0486824 Not subject to U.S. Copyright. Publ. 2005 Am. Chem. Soc. Published on Web 12/29/2004
millions of dollars in property damage and to have saved many lives (2). However, the heavy use of these chemicals, and the manner in which they are applied, has caused PBDEs to leach from the treated materials and accumulate in animals and humans. Reports have now demonstrated that concentrations of PBDEs have been increasing rapidly over the past 25 years in many environmental matrixes (3, 4) and in human serum and breast milk (5-7). PBDE concentrations have been measured in human serum, adipose tissue, and breast milk and all have shown that PBDE levels are about 17 times higher in North American individuals than those from individuals in Europe (8). The European Union has phased out the use of two of the three commercial mixtures of PBDEs (pentaBDE and octaBDE) this year and the state of California will follow in 2008. Because of this phase out, North America now uses 98% of the world market demand for pentaBDE, the commercial mixture which contains the congeners that are commonly observed in human tissues. Great Lakes Chemical Co., the sole producer of pentaBDE in the United States, has chosen to voluntarily phase out the production of pentaBDE by the end of 2004 (9); however, products that contain pentaBDE (i.e., furniture and carpets) will most likely remain in most homes for years. The third commercial mixture, decaBDE, is still used without regulation. It is the most heavily used with an estimated world market demand of 56 000 tons as reported in 2001 (10). DecaBDE is usually incorporated into high-impact polystyrene that is commonly found in the casings of TV sets and computers. Studies have suggested that dietary exposure is the most likely route by which people accumulate PBDEs (11). PBDEs are similar in structure to polychlorinated biphenyls (PCBs), which accumulate through dietary ingestion (12). However, in contrast to PBDEs, PCBs were used primarily in electrical transformers and capacitors that were external to the home environment. Leaching of PCBs from transformers in landfills, industrial regions, and at point sources all contributed to high levels of PCBs in the environment and their biomagnification in aquatic food chains. PBDEs, in contrast, are applied to products found in almost every home in percentages as great as 30% by mass of the product to which they are applied (1). Studies comparing indoor versus outdoor air have shown that concentrations are greater in homes (1.643 times greater) than they are in the outdoor environment (13). The large concentration of PBDEs in residential use products suggests that the home environment may also be a significant source of human exposure to PBDEs. In the present study, we collected dust from 16 homes in the Washington, D.C. metropolitan area and one home in Charleston, SC, to analyze for PBDEs. Dust has been used as an indictor of indoor exposure to pollutants such as lead and pesticides (14, 15). Young children are particularly vulnerable to contaminants found in dust as they are often in close contact with floors and dusty surfaces and have a greater propensity to put their hands and objects in their mouths. Past studies have set a precedent for contaminant exposure via dust in young children, as house dust has been positively correlated with children’s blood lead levels (1517). PBDE congeners found in the pentaBDE commercial mixture cause neurobehavioral deficits and disrupt thyroid hormone homeostasis in rodent and human in vitro studies (18-21). Fewer studies have been conducted examining the toxicity of decaBDE; however, some studies using mice and VOL. 39, NO. 4, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
925
TABLE 1. Ancillary Information Collected from Each House or Apartment Sampled no. of sample
city
typea
area (m2)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
California, MD Gaithersburg, MD Lusby, MD Gaithersburg, MD Lusby, MD Lusby, MD Silver Spring, MD Charleston, SC Silver Spring, MD Arlington, VA Silver Spring, MD Lusby, MD Oakton, VA Alexandria, VA Germantown, MD Gaithersburg, MD Washington, D.C.
H H H H H H H H T T B B A A A A A
213.7 204.4 185.8 185.8 167.2 167.2 157.9 116.1 223 139.4 65 55.7 101.3 83.6 80.8 62.2 55.7
year of construction
floor typeb
foam pieces
TVs
computers
hours computer used/week
1987 1974 1992 1967 1986 2002 1949 1994 1960 1983 1923 1987 1984 1989 1987 2000 1960
H H C H C C H C H H C C C C C C C
9 5 3 2 3 3 13 2 3 3 4 2 2 3 2 1 1
2 2 2 4 1 2 2 1 1 2 1 1 2 1 1 1 1
1 0 1 3 3 1 2 1 1 1 1 0 2 1 1 1 1
168 0 15 50 168 2 6 168 15 7 14 0 80 168 10 7 4
a Key to type of home: H (single-family detached home), T (townhouse), B (basement apartment), A (one- or two-bedroom apartment). type: C (carpeted), H (hardwood floors with area rugs).
rats suggest that chronic exposure may lead to neoplastic nodules in the liver and increases in the incidence of hepatocellular adenomas and carcinomas (22, 23). Very little is known about the direct effects of chronic PBDE exposure on people and particularly on babies and young children who are in sensitive developmental stages. Our objectives in this study were to measure PBDEs in house dust and determine the potential for exposure of PBDEs to young children via house dust. In addition, clothes dryer lint was analyzed for PBDEs as an alternate matrix for assessing the levels of PBDEs within the home and to determine if PBDEs may be adsorbing to clothes.
Materials and Methods Dust samples were collected from 16 homes in the Washington, D.C. metropolitan area and one home in Charleston, SC (Table 1), between January and March of 2004. Dust was collected from the floor in the main family room of all homes with a small commercial vacuum (Euro-Pro model, 900 W) equipped with a hose. Dust entering the vacuum first passed through a 1-mm wire mesh before being collected on a standard coffee filter (white-bleached filter for 12-cup drip coffee maker) which was inserted between the basket and HEPA filter. Between sample collections, the vacuum was thoroughly cleaned with hot water and a methanol rinse, and the coffee filter was replaced. In each home, dust was collected by vacuuming the rugs or hardwood floors (most with area rugs) until sufficient mass (0.1-0.5 g) was collected on the filter (approximately 15-30 min of vacuuming). The dust samples that were collected were assumed to be dry and no measurement of water content was performed. In place of a field blank, sodium sulfate powder was spread across a tile floor and collected with the vacuum in a manner similar to the dust samples. After vacuuming, the dust or field blank was scraped off the filter into precleaned glass jars using methanol-rinsed spatulas and taken back to the laboratory for extraction. Lint was collected from homes by removing the lint trap from the clothes dryer and collecting all the material off the lint trap, which was then wrapped in clean aluminum foil and placed in a plastic bag until extraction. An occupant from each home was asked to complete a survey regarding some ancillary parameters of the home. Data collected from each home included year of construction, area, the number of foam-containing couches and chairs, 926
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 4, 2005
b
Flooring
the number of TVs and computers, and the number of hours each week the computer was left on. Observations were also made regarding whether the home contained predominantly hardwood or carpeted floors in the main family room. Prior to extraction, between 50 and 100 ng (in hexane) of a 13C-labeled BDE 209 (2,2′,3,3′,4,4′,5,5′,6,6′-decabromodiphenyl ether, BDE 209L) and a 13C-labeled chlorinated diphenyl ether (2,2′,3,4,5-pentachlordiphenyl ether, CDE 86L, both from Cambridge Isotope Laboratories, Andover, MA) were added to each sample as internal standards. Dust samples were extracted using pressurized fluid extraction (Dionex model ASE 200) with dichloromethane. All samples, blanks, and calibration solutions were extracted using the following program parameters: temperature at 100 °C, heat time for 5 min, static time for 5 min, and pressure at 13.8 MPa (2000 psig), for three cycles. The extract was reduced in volume to 0.5 mL using an automated evaporation system and solvent exchanged to hexane. The extract was further cleaned using precleaned silica Sep-Pak cartridges (Waters Co., Milford, MA). Cartridges were first cleaned with 10 mL of hexane and the extract was eluted using 20 mL hexane. After concentration to 0.5 mL, the final extract was measured for PBDEs using an Agilent 6890 series gas chromatograph coupled to an Agilent 5973 mass spectrometer (GC/MS). Quantification of BDE congeners was performed with a GC/MS using negative chemical ionization and was operated in the selected ion monitoring mode. All BDEs were quantified using ions 79 and 81 (bromide ions) with the exception of BDE 209, which was monitored with ions 487 and 409. The 22 individual BDE congeners that were quantified in this study include triBDEs: 17, 28; tetraBDEs: 71, 47, 66; pentaBDEs: 100, 99, 85; hexaBDEs: 154, 153, 138, 156; heptaBDEs: 184, 183, 191, 190; octaBDEs: 197, 196; nonaBDES: 208, 207, 206, and the fully brominated BDE 209. A 15 m × 0.25 mm (i.d.) 5% phenyl methylpolysiloxane capillary column (0.25-µm film thickness; J & W Scientific) was used for the separation of the BDE congeners, and all injections were performed with cool on-column injection. The inlet was programmed to follow the oven temperature program which was 80 °C for 2 min followed by a temperature ramp of 12 °C/min to 140 °C, then another ramp from 140 °C to 280 °C at 5 °C/min (held for an additional 20 min at 280 °C). The ion source was held at a constant temperature of 200 °C, the quadrupole was held at 100 °C, and the temperature of the transfer line was held at 280 °C. The method detection
FIGURE 1. GC/ECNI-MS chromatogram comparison of PBDE standard and house dust collected from sample 13. Ions 79 and 81 were used to monitor tri- through nonaBDEs, while ion 487 and 409 were used to monitor BDE 209. limit (MDL) for the PBDE congeners ranged from 1 ng/g dry mass (for BDE 28) to 6 ng/g dry mass (for BDE 209). Instrumental detection limits (S/N ) 50) for BDE congeners ranged from 0.1 to 0.4 ng. Field blanks contained traces of BDEs 47, 99, and 209 but all values were low enough (