Tri-decabrominated diphenyl ethers and hexabromocyclododecane in indoor air and dust from Stockholm microenvironments 2: Indoor sources and human exposure

Abstract

Data on polybrominated diphenyl ether (PBDE) and hexabromocyclododecane (HBCD) concentrations from Stockholm, Sweden, indoor microenvironments were combined with information from detailed questionnaires regarding the sampling location characteristics, including furnishing and equipment present. These were used to elucidate relationships between possible flame-retarded sources and the contaminant concentrations found in air and dust. Median concentration ranges of ΣPenta-, ΣOcta-, ΣDecaBDE and HBCD from all microenvironments were 19–570, 1.7–280, 29–3200 and < 1.6–2 pg/m³ in air and 22–240, 6.1–80, 330–1400 and 45–340 ng/g in dust, respectively. Significant correlations were found between concentrations of some PBDEs and HBCD in air and/or dust and the presence of electronic/electrical devices, foam furniture, PUF mattresses and synthetic bed pillows in, as well as floor area and construction year of the microenvironment. Car interiors were a source to indoor air in dealership halls. Using median and maximum concentrations of ΣPenta-, ΣOcta-, ΣDecaBDE and HBCD in air and dust, adult and toddler (12–24 months) intakes from inhalation and dust ingestion were estimated. Toddlers had higher estimated intakes of ΣPenta-, ΣDecaBDE and HBCD (7.8, 43, 7.6 ng/d, respectively) from dust ingestion than adults (5.8, 38, 6.0 ng/d, respectively). Air inhalation in offices was also an important exposure pathway for ΣPenta-, ΣOcta- and ΣDecaBDE in adults. For ΣPentaBDE and HBCD, air inhalation and dust ingestion play minor roles when compared to previously published Swedish dietary intakes (median exposures). However, in worst case scenarios using maximum concentrations, dust ingestion may represent 77 and 95% of toddler intake for ΣPentaBDE and HBCD, respectively.

Introduction

The polybrominated diphenyl ether (PBDE) technical products, here designated as Penta mix, Octa mix and Deca mix, as well as hexabromocyclododecane (HBCD) are additive brominated flame retardants (BFRs). They are or have been used in textile coatings, furniture foams, and plastics in TVs, computers, various electric and electronic equipment. HBCD has been used in expanded and extruded polystyrene, commonly used as insulation in buildings. These products are found in homes, office buildings, public buildings such as day care centers and schools as well as in cars, subways, trains, airplanes, and other means of transport. Thus, emissions from flame-retarded items present in indoor environments are now considered probable sources of human exposure (Harrad et al., 2010). However, attempts to correlate concentrations of BFRs in indoor air and dust with room or home characteristics, building materials and/or counts of potentially flame-retarded items have met with mixed success (Allen et al., 2007, Cunha et al., 2010, Harrad et al., 2004, Harrad et al., 2009, Hazrati and Harrad, 2006, Stapleton et al., 2005, Toms et al., 2009, Wilford et al., 2004, Wilford et al., 2005, Wu et al., 2007, Zhang et al., 2011). In part, this may have been due to the sample sizes used, only studying one type of microenvironment (usually homes), the number of flame-retarded products included (typically PCs, TVs, foam furniture), which in turn leads to low variability in the number and types of flame-retarded products within and between microenvironments. Other factors may also affect concentrations of BFRs in indoor air and dust, including room temperature, the ventilation rate and whether or not windows to outdoors are open or closed. Misclassification of flame-retarded items may also be a problem as many items included in counts may not contain BFRs, as has been shown in one study using X-ray fluorescence (XRF) to measure bromine in items (Allen et al., 2008a). In that study, no significant correlations were seen between concentrations of PBDEs in dust and counts of TVs and foam furniture, but when the items were measured for bromine content with XRF and counts only included items containing bromine, significant correlations were seen (Allen et al., 2008a). In another study, significant correlations were found between concentrations of Penta mix-related congeners in air and counts of electrical equipment and polyurethane foam (PUF) chairs when a larger variety of microenvironments were included (Harrad et al., 2004).

Concentrations of BFRs in air and dust have also been used to estimate adult and toddler exposure from air inhalation and dust ingestion in various countries from Europe and North America (Abdallah et al., 2008a, Abdallah et al., 2008b, Allen et al., 2007, D'Hollander et al., 2010, Fromme et al., 2009, Harrad et al., 2006, Harrad et al., 2008a, Roosens et al., 2009a, Roosens et al., 2009b, Roosens et al., 2010, Wilford et al., 2004, Wilford et al., 2005). Only a few such studies have included both PBDEs and HBCD and no previous estimates for Sweden exist.

In the first part of this study (Thuresson et al., in press), we determined concentrations of tri-decabrominated diphenyl ethers and HBCD in indoor air and dust samples from houses, apartments, day care centers, offices, cars and car dealership halls. These represent a wide range of microenvironments with more heterogeneous characteristics and contents of flame-retarded items. In the current study, the analytical results for homes, day care centers and offices were used together with detailed information from questionnaires of microenvironment characteristics and contents of potentially flame-retarded items to explore associations between BFR concentrations and the presence of possible indoor sources. The results from air and dust analyses were also used to estimate adult and toddler exposure in Sweden to PBDEs and HBCD from the different indoor environments including cars. The estimated intakes from inhalation and dust ingestion were also compared to estimated dietary intakes of Penta mix-related congeners and HBCD from previously published Swedish studies and BDE-209 from a recent Belgian study (Roosens et al., 2010).