A lumping model for composition- and temperature-dependent partitioning of secondary organic aerosols

An improved lumping model is described for converting multiple semivolatile organic aerosol components into one or more lumped groups that can represent the organic aerosol mixture in atmospheric models. Lumping equations to calculate temperature- and composition-dependent absorption partitioning parameters are presented, together with lumping criteria for dividing components into groups based on component vapor pressure, chemical structure, and activity coefficient in water. Lumping performance of the model is evaluated using a base case mixture of organic aerosol components, random mixtures selected from a set of identified semivolatile organic compounds, and products from the alpha-pinene/ozone reaction. Modeling results show that aerosol composition has a strong influence on SOA partitioning and lumping accuracy. Comparisons of different lumping criteria suggest that lumping components by a combination of vapor pressure and water interactions or by vapor pressure alone will produce optimum results. (C) 2004 Elsevier Ltd. All rights reserved.