Environ. Sci. Technol. 2010, 44, 3059–3065
Modification and Calibration of a Passive Air Sampler for Monitoring Vapor and Particulate Phase Brominated Flame Retardants in Indoor Air: Application to Car Interiors M O H A M E D A B O U - E L W A F A A B D A L L A H * ,†,‡ AND STUART HARRAD† Division of Environmental Health and Risk Management, School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham, B15 2TT, United Kingdom, and Department of Analytical Chemistry, Faculty of Pharmacy, Assiut University, 71526 Assiut, Egypt
Received January 14, 2010. Revised manuscript received March 01, 2010. Accepted March 08, 2010.
A passive air sampler was modified to monitor both vapor and particulate phase brominated flame retardants (BFRs) in indoor air using polyurethane foam disks and glass fiber filters (GFF). Significant correlation (p < 0.01) was observed between passive (ng day-1) and active sampler (ng m-3) derived BFR concentrations in an office microenvironment (r ) 0.94 and 0.89 for vapor and particulate phase BFRs, respectively). A calibration experiment was performed where concentrations of target BFRs were obtained for an office using a low volume active sampler operated over a 50 day period alongside passive samplers. The passive uptake rates of each studied BFR ranged between (0.558-1.509 ng day-1) and (0.448-0.579 ng day-1) for vapor and particulate phases, respectively. The passive entrapment of particles by the GFF was investigated using environmental scanning electron microscopy which revealed gravitational deposition of particles as the main mechanism involved. The developed sampler was applied to monitor BFR concentrations in 21 cars. Average concentrations of ΣHBCDs, TBBP-A, and Σtetra-deca BDEs were 400, 3, and 2200 pg m-3 in cabins and 400, 1, and 1600 pg m-3 in trunks. No significant differences (p < 0.05) were observed between levels of ΣHBCDs and Σtrito hexa- BDEs in cabins and trunks. However, TBBP-A, BDE-209, and ΣPBDEs concentrations were significantly higher in vehicle cabins.
Introduction Brominated flame retardants (BFRs) constitute a diverse group of compounds used to prevent or minimize fire hazards. The most widely used BFRs are tetrabromobisphenol A (TBBP-A) with a global demand of 170 000 t in 2004 (1), alongside decabromodiphenyl ether, hexabromocyclododecanes (HBCD), and pentabromodiphenyl ether, for which the worldwide market demands in 2001 were respectively 56 100 t, 16 700 t, and 7 500 t (1). The potential adverse effects * Corresponding author phone: +44 121 414 5431; fax: +44 121 414 3078; e-mail:
[email protected]. † University of Birmingham. ‡ Assiut University. 10.1021/es100146r
2010 American Chemical Society
Published on Web 03/15/2010
on human health as a result of exposure to such chemicals include endocrine disruption, neurotoxicity, liver microsomal enzyme induction, immunotoxicity, and carcinogenicity (2-4). Therefore, there is an increasing interest in monitoring the levels of BFRs in indoor air and dust. However, less information is available on the concentrations of BFRs in indoor air than in dust (5-7). Notwithstanding this, we have reported previously on concentrations of airborne (vapor phase only) trihexa PBDEs in 25 cars using PUF-disk passive samplers, and revealed concentrations to be substantial in some vehicles (9). However, for practical reasons we deployed the samplers in the trunks only, and the extent to which this approach can represent accurately the exposure of vehicle occupants is unknown due to the possibility of intravehicle variations in BFR concentrations. Specifically, it is possible that concentrations in the passenger cabin to which vehicle occupants are exposed, may differ significantly from those in the trunk. In previous studies, our research group reported on the calibration and application of polyurethane foam (PUF) based passive air samplers for monitoring tri- to hexa- polybrominated diphenyl ethers (PBDEs) (8, 9), as well as R-, β-, and γ-HBCDs (10, 11) in indoor air from different microenvironment categories. However, PUF disks are known to sample effectively only the vapor phase (12) which renders them inappropriate for monitoring airborne levels of contaminants that exist primarily in the particulate phase such as TBBP-A (11) and higher (hepta- to deca-) PBDEs (13). Thus, active air sampling incorporating a particulate filter as well as a vapor phase sorbent is the current method of choice for measuring airborne concentrations of TBBP-A and higher PBDEs. Recently, a glass fiber filter (GFF) coupled with a PUF disk has been used successfully for collecting both vapor and particulate phase polycyclic aromatic hydrocarbons (PAHs) in outdoor air using passive sampling (14). Owing to the advantages conferred by passive sampling over active air sampling including ease of use, inobtrusiveness during deployment, low cost, no electricity requirements, as well as the provision of time weighted average concentrations; we investigate here the feasibility of employing passive air sampling techniques for monitoring concentrations of BFRs in indoor air in both the vapor and particulate phases. Therefore, the aims of the current study are (a) to adapt an existing passive air sampler (PAS) configuration to collect both vapor and particulate phase BFRs in indoor air; (b) to calibrate this PAS against a low-volume active air sampler to determine passive air sampling rates for target BFRs; (c) to investigate the mechanism of incorporation of particulates on the GFF inside PAS using environmental scanning electron microscopy (ESEM); (d) to compare the concentrations of target BFRs derived from the PAS to those obtained from active air sampling devices in different indoor microenvironments; and (e) to use the PAS to monitor concentrations of target BFRs in the trunks and cabins of UK cars, and assess whether there exists any significant intravehicle differences in concentrations.
Materials and Methods Target BFRs. The following BFRs were investigated in this study: TBBP-A, R-HBCD, β-HBCD, γ-HBCD, BDE-47, BDE85, BDE-99, BDE-100, BDE-153, BDE-154, BDE-183, BDE196, BDE-197, BDE-203, BDE-206, BDE-207, BDE-208, and BDE-209. Passive Air Sampler Configuration. Figure 1 shows the passive sampler developed. It consists of two sampling media: VOL. 44, NO. 8, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3059
FIGURE 1. Schematic diagram of our passive air sampler configuration. a PUF disk (140 mm diameter, 12 mm thickness, 360.6 cm2 surface area, 0.02 g cm-3 density, PACS, Leicester, UK) and a glass fiber filter (GFF, 12.5 cm diameter, 1 µm pore size, Whatman, UK) used for sampling vapor and particulate phase BFRs, respectively. The sampling media are fully sheltered between two different size stainless steel housings (18 cm, one liter bottom housing and 23 cm, two liters top housing). To minimize gravitational deposition of particles, the PUF disk was mounted at the top of the shelter with only the downward face exposed to air. The inner sampler surface was lined fully by aluminum foil (prerinsed with CH2Cl2) in an attempt to minimize the particle-scavenging potential of the sampler housing reported previously (8). Such scavenging was found to limit the capacity of a similar passive air sampler design for sampling higher molecular weight compounds especially at lower concentrations (8). The GFF was suspended in the middle of the sampler housing supported by a stainless steel thin wire mesh disk (3 mm aperture size, 12.5 cm diameter) mounted on the central screw to collect particulates. PUF disks were pre-extracted with CH2Cl2 for 8 h in a Soxhlet apparatus while GFFs were preconditioned by heating at 450 °C for 5 h prior to field deployment. Prior to sampling, both sampling media were treated with 10 ng of both d18-R-HBCD and 13C-BDE-100 as sampling evaluation standards (SESs) designed to provide a quantitative measure of sampling efficiency and analyte breakthrough during sampling. Shelters were cleaned carefully and solvent rinsed to remove potential contamination prior to each deployment. Assessment of the Passive Air Sampler Performance. Five passive samplers were deployed for 45 days at a height of 180 cm (with a minimum distance of 50 cm between samplers) in an office microenvironment in January 2008. A low volume active sampler (details can be found in the Supporting Information (SI)) was used to collect three samples (each comprising 45 m3) from the same office at the start, middle, and end of the passive sampling period. Passive Air Sampler Calibration. To obtain quantitative data on airborne contaminant concentrations from passive samplers, one needs to know both the mass of contaminant sequestered by the sampler, and the volume of air sampled over its deployment. To achieve this, a calibration exercise was conducted involving simultaneous monitoring of target BFRs in indoor air using the passive sampler over different exposure times, and calibration against time weighted average concentrations of the same compounds derived via simultaneous low volume active air sampling in the same microenvironment. Passive samplers (n ) 8) were deployed over 50 days in March 2008 in an office microenvironment at a height of 150 cm with a minimum distance between samplers of 50 cm. PUF disks and GFFs were harvested at 10 day intervals over the 50 days of the experiment. To ensure that detectable concentrations were provided by the passive samplers at the 10 and 20 day sampling intervals, three and two samplers were harvested and combined for analysis at 3060
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 8, 2010
these times, respectively. PUF disks and GFFs were analyzed separately. The analyte masses present in the combined samples for the 10 and 20 day sampling intervals were normalized to a single sampler equivalent mass for calibration purposes. Concurrent with the deployment of the passive air samplers, three low-volume active samplers were operated simultaneously, each providing a single active air sample (details in the SI) covering the full 50 day duration of the calibration. Comparison of BFR Concentrations Derived via Passive Sampling with Those Obtained via Active Air Sampling. Between June and October 2008, two passive air samplers were deployed at each of five different indoor microenvironments (two homes and three offices) for 45 days. In each microenvironment, two low volume active air samples were collected (sample volume 45 m3) at the start and end of the passive sampling period. In addition, three high volume active air samples (sample volume 45 m3, details in the SI) were collected in the studied office microenvironments on the 10th, 20th, and 30th day of sampling. For reasons of excessive noise, high volume air samples were not collected in homes. For the purposes of comparison, concentrations of target BFRs collected by each air sampling technique were normalized subsequently to one sample from each studied microenvironment. ESEM for Investigating the Mechanism of Particulate Matter Sampling by the GFF Incorporated in the PAS. Following deployment, small circular pieces of GFFs (2 mm diameter) were adhered to aluminum stubs using doublesided carbon sticky tabs (Agar Scientific) prior to coating with evaporated gold in an Emscope SC500 evaporation unit. Microscopic examination was conducted at high vacuum in an FEI XL-30 FEG ESEM. Chemical analysis for the determination of bromine content was carried out by energy dispersive X-ray energy dispersive spectrometry (EDS) microanalysis using an Oxford Instruments INCA instrument. Air Sampling in Cars. Passive air samplers supported by secure stainless steel cradles were deployed for 45 days to provide a time-integrated sample in 21 private cars in Birmingham, UK (SI Table SI-1). Sampling took place between October 2008 and April 2009. Two samplers were deployed in 18 cars; one in the trunk and the other in the cabin. For operational reasons, it was possible only to deploy samplers in the trunk in one car, and in the cabin only in two vehicles. Passive air samplers were assembled on-site and at the end of deployment period; PUF disks and GFFs were stored separately at 4 °C until extraction. Analytical Protocols. PUF disks and GFFs were analyzed separately. To minimize the particle scavenging potential of the wire mesh filter support (14), the wire mesh was extracted with the GFF. Briefly, each sampling medium was spiked with internal standard mixture (10 ng of each of 13C-labeled R-HBCD, β-HBCD, γ-HBCD, TBBP-A, BDE-47, BDE-99, BDE153, BDE-128, and BDE-209 in hexane) prior to Soxhlet extraction with CH2Cl2 for 8 h. The crude extracts were concentrated to 0.5 mL using a Zymark Turbovap II then purified by loading onto SPE cartridges filled with 8 g of precleaned acidified silica (44% concentrated sulfuric acid, w/w). The analytes were eluted with 25 mL of hexane:CH2Cl2 (1:1, v/v). The eluate was evaporated to dryness under a gentle stream of N2, then reconstituted in 100 µL of methanol containing d18-γ-HBCD (25 pg µL-1) as recovery determination (or syringe) standard, used to determine the recoveries of internal standards for QA/QC purposes. Sample analysis was carried out using an LC-MS/MS system composed of a dual pump Shimadzu LC-20AB Prominence liquid chromatograph equipped with SIL-20A autosampler, a DGU-20A3 vacuum degasser coupled to a Sciex API 2000 triple quadrupole mass spectrometer. Full details of the analytical methodology used for determination
TABLE 1. Comparison of Average Concentrations of Target BFRs in Gaseous and Particulate Phases of Indoor Air Derived from Passive Samplers (ng Day-1; n = 5) and Active Sampler (ng m-3; n = 3). Standard Deviations Are Given in Parentheses active sampler (pg m-3) R-HBCD β-HBCD γ-HBCD TBBP-A BDE-47 BDE-85 BDE-99 BDE-100 BDE-153 BDE-154 BDE-183 BDE-196 BDE-197 BDE-203 BDE-206 BDE-207 BDE-208 BDE-209
passive sampler (pg day-1)
gas (G)
particle (P)
Σ G+P
gas
particle
Σ G+P
28.4 (4.1) 7.5 (1.7) 54.2 (9.3)
