Simulations of SOA Oxidative Potentials Using Explicitly Predicted Products Formed from Atmospheric Processes of Various Hydrocarbons

Abstract

Reactive oxygen species (ROS) in secondary organic aerosol (SOA) cause oxidative stress in airway cells and harm respiratory health. The UNIfied Partitioning Aerosol Reaction (UNIPAR) model predicting SOA formation by multiphase reactions of hydrocarbons was applied to estimate the oxidative potential (OP) of SOA generated from the photooxidation of phenol, naphthalene, toluene, and α-pinene. In the UNIPAR-OP model, SOA chemical compositions, predicted with lumping species using explicitly predicted products, were enabled to estimate ROS, including hydroperoxides (ROOH), conjugated carbonyls (CJCB), peroxyacyl nitrates (PAN), and naphthoquinones (QUINONE) under varying environments. The performance of UNIPAR-OP was evaluated by comparing simulations with data from chemical assays (dithiothreitol assay for OP and 4-nitrophenylboronic acid assay for ROOH). Both predictions and observations revealed that SOA OP is more sensitive to hydrocarbon type than NO_x levels showing the highest OP with naphthalene SOA and the lowest with α-pinene. The dominance of subgroup ROS in OP varies by SOA, with CJCB significant in phenol and toluene SOA, QUINONE in naphthalene SOA, and ROOH and PAN in α-pinene SOA. UNIPAR-OP can predict regional differences in OP under varying SOA compositions and NO_x levels, with quantification of subgroup contributions and identification of SOA components that are associated with OP.