Atmospheric gas-particle partitioning of organic compounds: Comparison of sampling methods

Abstract

A conventional glass-fiber filter, a low-pressure cascade impactor and an electrostatic precipitator, each combined with two gas traps, were used to establish the significance of sampling artifacts occurring during the investigation of atmospheric gas-particle partitioning. The absolute concentrations and the gas-particle partitioning of polychlorinated benzenes, pentachloroanisole, hexachlorocyclohexanes, PCBs, p,p′-DDE, p,p′-DDT and some PAHs were determined. The total amounts of substance collected in the impactor and the filter system were similar, whereas the levels of most compounds in the precipitator system differed due to substance alteration in the sampler. The electrostatic precipitator therefore does not seem to be a reliable device for such measurements. The comparison of gas-particle partitioning showed particle bound fractions of the organochlorine compounds to be systematically higher in the impactor-adsorbent combination than in the filter-adsorbent sampler. This was attributed to the particle aggregation in the impactor preventing the chemicals on particle surfaces from an exchange with the gas phase. However, these differences did not exceed a factor of two and were on average in the range of 36%. This indicates that sampling artifacts through the use of glass-fiber filter-adsorbent systems for investigations of atmospheric gas-particle partitioning are less drastic than discussed in the literature.