Comparison of filter materials used for sampling of mutagens and polycyclic aromatic hydrocarbons in ambient airborne particles

Abstract The filter material used in high-volume sampling of ambient airborne particles was investigated for its influence on mutagenicity and PAH content of the particles. The particles were collected simultaneously with glass-fibre filters (G filters) and Teflon filters (T filters) or Teflon-coated filters (Tc filters). Sampling was carried out 17 times at different locations in The Netherlands. The use of G filters lead to slightly higher mutagenicity in almost all experiments. In addition every sample collected with this filter material contained substantially higher concentrations of the lower molecular weight PAH which are partly present in the vapour phase. In contrast some experiments revealed slightly lower concentrations of non-volatile PAH which are sensitive to chemical conversion under atmospheric conditions. A multivariate analysis of the concentrations did hardly show any influence of the filter material. The main factor affecting the PAH profile was volatility; factor analysis revealed chemical reactivity as a second factor. The differences between the concentrations of the volatile PAH can be explained by the lower pressure drop across the G filters, resulting in less artifactual evaporation. In addition, adsorption of these compounds from the gas phase on this filter material was found. The latter phenomenon may also explain the differences in mutagenicity, as it has been shown that PAH adsorbed on glass-fibre material react with gaseous air components to form mutagens. The lower concentrations of non-volatile PAH on the G filters points directly to conversion which may also have contributed to the differences in mutagenicity. It is not possible to decide whether the chemical conversion on the T and the Tc filters (demonstrated with factor analysis) has an artifactual origin or takes place in the ambient air before the air reaches the filter. It is concluded, that the differences in mutagenicity between the filter materials are too small to support the use of T or Tc filters instead of G filters in mutagenicity studies. The adsorption of volatile PAH on filter material makes it uncertain to what extent high-volume sampling by filtration leads to over- or underestimation of the volatile PAH present in the particulate phase of the air.