The influence of aerosol size and organic carbon content on gas/particle partitioning of polycyclic aromatic hydrocarbons (PAHs)

Concentrations of polycyclic aromatic hydrocarbons (PAHs) were measured on size segregated airborne particulate matter and in the gas phase during 12 h periods in urban Chicago and over southern Lake Michigan during July 1994 and January 1995 as part of the atmospheric exchange over lakes and oceans (AEOLOS) project. In the 19 paired gas phase and size segregated particle samples, partition coefficients of polycyclic aromatic hydrocarbons are well correlated with the sub-cooled liquid vapor pressure, (pl0) within an aerosol size class. However, partition coefficients differ systematically according to particle size and aerosol organic carbon content. Approximately 68% of the observed variability in measured PAH partition coefficients is explained by a three dimensional multiple linear regression that includes vapor pressure of the compound of interest, aerosol particle aerodynamic equivalent diameter, aerosol organic carbon content and interaction terms. However, addition of both particle size and the fraction organic carbon terms, while statistically significant, appear to be of minimal importance in improving our ability to model gas/particle partitioning in the atmosphere. The influence of either particle size or fraction organic carbon is nearly as large as the calculated random error in partition coefficients, and suggests that an important factor in predicting gas/particle partitioning has not yet been incorporated in the current model.