Environ. Sci. Technol. 2010, 44, 749–754
Field Evaluation of Polyurethane Foam Passive Air Samplers to Assess Airborne PAHs in Occupational Environments P E R N I L L A B O H L I N , * ,† K E V I N C . J O N E S , ‡ AND BO STRANDBERG† Department of Occupational and Environmental Medicine, The Sahlgrenska Academy at University of Gothenburg, SE-40530 Gothenburg, Sweden, and Centre for Chemicals Management, Lancaster Environment Centre, Lancaster University, Lancaster, LA1 4YQ, U.K.
Received July 31, 2009. Revised manuscript received November 11, 2009. Accepted November 19, 2009.
There is a need for simple air sampling techniques to enable routine monitoring of the occupational exposure to polycyclic aromatic hydrocarbons (PAHs) in compliance with occupational exposure limits. Other gas-phase contaminants can be monitored in workplaces using passive samplers but this is currently not the case for PAHs. Here, polyurethane foam (PUF) disk passive air samplers (PAS), routinely used for outdoor air monitoring of PAHs and POPs, were assessed for their suitability in an indoor occupational environment against: ability to accumulate detectable levels within 1-2 weeks; quantitative sampling of benzo(a)pyrene (BaP), precision, uptake kinetics, influence of shelter design, and performance of 16 deuterated PAHs as depuration compounds (DCs). Sampling rates (Rvalues) for PAHs in PUF-PAS, estimated by comparison to lowvolume active samplers, and the loss of DCs, varied for individual PAHs (1-10 m3 day-1) but were found to be in the same order of magnitude for both gas-phase and particleassociated PAHs including BaP. Only one PAH (Acy) fulfilled the DC criteria of >40% loss during the 2 week exposure. These results suggest that PUF-PAS are potentially useful tools for PAHs in occupational environments in screening workplaces and identifying sources/hotspots - although unlikely to replace active sampling.
Introduction Monitoring of polycyclic aromatic hydrocarbons (PAHs) in occupational environments is important given the potential adverse health effects associated with high exposure among workers. Workplaces where high levels of PAHs and high exposure may occur include iron and coke plants, alloy industries, aluminum plants, and traffic-associated occupations (1). The US Environmental Protection Agency (US EPA) has identified a group of 16 PAHs as priority pollutants (Table 1) including some classified as carcinogens or possible carcinogens to humans by the International Agency for Research on Cancer (IARC). To ensure a healthy and safe working environment, national or regional (EU) legislation * Author for correspondence: Pernilla Bohlin, E-mail: pernilla.bohlin@ amm.gu.se Tel: (+46)-31-7863618, Fax: (+46)-31-409728. † The Sahlgrenska Academy at University of Gothenburg. ‡ Lancaster University. 10.1021/es902318g
2010 American Chemical Society
Published on Web 12/15/2009
have set air quality standards and occupational exposure limits (OEL) to pollutants. The OELs for PAHs are based on time-weighted average (TWA) air concentrations of benzo(a)pyrene (BaP) - a known carcinogen - for typical occupational exposure periods. Employers and/or regulatory agencies are therefore required to conduct routine monitoring of PAHs in the workplace in compliance with the TWA limits and may subsequently need to control the workers’ exposure. This is currently limited by the expense and complications associated with commonly used sampling techniques, that is active samplers. This is the most accurate technique since the volume of air can be controlled but there are several drawbacks which strongly limit their application for routine monitoring in workplaces. These include the high cost of sampler purchase and maintenance, the need for qualified personnel, noise nuisance and disturbance for workers, potential damage/limitations to sampler performance in harsh occupational conditions, and possible sampling artifacts (evaporation from particles on the filter, and indirect photochemical degradation). There is therefore a high demand and incentive for a new sampling technique that can overtake these obstacles and simplify/enable routine monitoring, determine workers’ exposure or screen PAHs quickly, cheaply, and easily at many locations through a workplace. Passive air sampling (PAS) techniques are shown to be good substitutes to active samplers for monitoring of PAHs in ambient air and to monitor gas-phase compounds; volatile organic compounds (VOCs), O3, NOx, SOx, and so forth, in occupational settings (2, 3). A recent study used a PAS for PAHs in an occupational environment with the purpose of assessing air genotoxicity (4). However, despite the obvious need, PAS techniques for monitoring the exposure to PAHs in occupational settings have not been studied. This is presumably because the sampling of PAHs is more challenging. They are semivolatile, that is some are almost exclusively in the gas phase, whereas others are almost exclusively associated to particles at ambient temperatures. The sampler’s ability to trap particles is especially important for applications in occupational exposure assessments because the most carcinogenic PAH compound, BaP, is mainly (>90%) associated to particles and OELs are based on the concentration of BaP. The theory of PAS has been developed based on compounds in the gas phase (5, 6), whereas the uptake behavior of particles and their associated chemicals may be more complex due to their different deposition mechanisms (i.e., sedimentation, impaction, interception, diffusion). Moreover, total and individual PAHs are typically at orders of magnitude lower concentrations than VOCs in occupational air (7). This requires a larger volume of air to be retained by the sampler (i.e., increased sampling time and/or sampler size) compared to PAS for VOCs, to give sufficient sensitivity. A possible PAS for PAHs in occupational settings is the polyurethane foam (PUF) disk sampler. This is the most widely used PAS for PAHs and other persistent organic pollutants (POPs) in outdoor and recently indoor air monitoring (8, 9). The PUF-PAS appears to be the only PAS in routine use that potentially traps particles to a similar extent as gas-phase compounds (at least the fine mode 40% loss. It is also preferable to use several DCs to cover a wide range of volatilities and to screen R-values for the full range of target compound classes. Hence, the suitability of 16 US EPA PAHs with different volatility properties as DCs was tested for the first time in an indoor environment exposure. The average losses (CDC/CDC,0), calculated log KOA and estimated log KPUF-A are shown in Table S4 of the Supporting Information. This is the first report of estimated log KPUF-A for PAHs. Unfortunately, the lack of experimental log KPUF-A likely limits the quality of the R-values estimated with method 2. No differences in losses of DCs were found between the different sampling points in the factory. This is in agreement with the results from method 1 and confirms minor influence from wind speed as suspected in indoor environments. Only the most volatile PAHs (Acy, Ace, and Flu) met the original DC criteria and had a loss >20% after the 2 week exposure time, whereas Flu failed the criteria of >40%. Acy was the only compound showing consistent high losses and provides the best option as a DC. Smaller or no losses were found for PAHs with higher KOA. However, the stability of the heavier PAHs provides good options for usage as surrogate DCs experiencing no loss. Exposure times of 1 week or shorter may require even more volatile DCs to assess a loss >40%, for example 2 rings PAHs. The estimated R-values using the three PAHs with a loss >20% ranged between 3-6 m3 day-1 (Figure 2), whereas Acy indicate R-value of 3.4 m3 day-1. This is slightly higher than
the R-value obtained using method 1. The reason for this is unclear but may depend on uncertainty in the estimated values of log KPUF-A. The R-values obtained from a given DC can presumably be applied to a full range of target compounds in the gas phase. In the case of PAHs, some concern arises due to their wide range of volatilities and high level of particle associations. It is therefore reasonable to assume that the average DC-derived R-value can be applied to the other PAHs in the gas phase but not to particle-associated ones. The uptake of particle-associated PAHs (in particular large particles) will depend on processes different from the diffusivity theory for DCs (i.e., dependent on the TSP, aerodynamic of particles etc.). However, the average R-value from the volatile DCs could potentially be applied to PAHs associated to ultrafine to fine particles, which have similar diffusivity as the gasphase PAHs, although this needs to be demonstrated. The results from method 1 show that in the present conditions PUF-PAS trapped the particle-associated PAHs as efficiently as the gas-phase PAHs indicating a large fraction of fine mode particles within the factory and thereby the potential applicability of the DC-derived R-values for the particle phase too under these conditions. Testing the Accuracy of PUF-PAS for Estimating Air Concentrations. To assess the accuracy and precision in deriving air concentrations with PUF-PAS, we applied the obtained R-values from sampling period 1 to the PUF-PAS’ results from sampling period 2. The estimated PUF-PASderived air concentrations were compared to the air concentrations measured with the active samplers showing a factor of 1-2 from the measured ones for 3-ring PAHs, 1.5-5 for 4-5-ring PAHs, 1-3.5 for 6-ring PAHs. This performance is lower than the accuracy criterion from NIOSH of
