Simulating and explaining passive air sampling rates for semi-volatile compounds on polyurethane foam (PUF) disks

Polyurethane foam (PUF) passive air samplers are widely deployed as an inexpensive and practical way to sample semi-volatile pollutants. However, concentration estimates from passive sampling to date have relied on estimating constant empirical mass transfer rates, which add unquantified uncertainties to concentrations and any spatial and temporal information they contain. Here we present a method for modeling hourly sampling rates, mass transfer, and concentrations from hourly meteorology using first principle chemistry, physics, and fluid dynamics. This approach provides a way to quantify and explain the observed effects of meteorology on spatial and seasonal variability in congener-specific sampling rates and analyte concentrations; to assess PUF saturation; and to recover average concentration at a reference temperature. Modeled sampling rates are evaluated for gas-phase polychlorinated biphenyls (PCBs) at an urban network of seven flying saucer samplers in Chicago, Illinois during 2008 using local meteorological observations and those simulated by the Weather Research and Forecasting (WRF) model at urban to regional scales. Modeled sampling rates are compared with results from depuration compounds, and temporal trends in analyte concentrations evaluated through comparison with active samplers. The model simulates average sampling rates within 17.2 (±16.4) % of those determined from depuration compounds. Results highlight that sampling rates are highly variable at hourly and daily scales, sensitive to spatial and temporal resolution in meteorology, and strongly congener-dependent, with predictable relationships between congeners. Both modeled and depuration-based sampling rates for low molecular weight congeners differ from high molecular weight congers by >16% consistently throughout the year, indicating the need to use congener-specific sampling rates when determining concentrations from passive samplers. We quantify the importance of each simulated process to sampling rates and mass transfer, and assess uncertainty contributed by each process, including advection, molecular diffusion, volatilization, and turbulent and laminar flow regimes within the sampler housing. Through this method, hourly congener-specific sampling rates, analyte concentrations, and PUF uptake can be simulated anywhere, at any time, from observed or modeled hourly meteorology.