Role of Criegee Intermediates in Secondary Sulfate Aerosol Formation in Nocturnal Power Plant Plumes in the Southeast US

Criegee intermediates (CI) from ozonolysis of biogenic volatile organic compounds (BVOC) have been suggested to be important atmospheric oxidants. However, due to their low atmospheric concentrations, possible high reactivity with water vapor, and unconstrained thermal unimolecular decay rates, their impact on atmospheric oxidation of trace species such as SO2 and NO2 remains uncertain. In this study, we investigate the formation of secondary sulfate aerosols (SSA) in nocturnal power plant plumes in the Southeastern US. These plumes have large mixing ratios of SO2 and NO that make reaction with CI competitive with other pathways, such as thermal unimolecular decay and water vapor reaction. The background into which these plumes are emitted has high levels of BVOC and O-3, whose reaction produces a large source of CI. Observed nighttime power plant plume intercepts had measurable sulfate aerosol, ranging from 0.7-1.2% of the total plume sulfur (SO2 + sulfate) on a molar basis. In the absence of photochemical OH oxidation, these observed sulfate levels can be compared to calculated CI + SO2 production. We present a plume dispersion model that simulates the chemical evolution of these nighttime plumes and compare the results to observed sulfate aerosol. Thermal unimolecular decay of CI is the largest uncertainty. In the absence of thermal unimolecular CI decay, CI reactions with SO2 in the dark account for up to 41% of the total observed sulfate aerosol, with the remainder attributable to reaction of SO2 with secondary OH and direct emission. Conversely, with a thermal unimolecular decay rate for all CI of 200 s(-1), equivalent to the highest measured rate, CI reactions with SO2 accounted for only S.7% of the total SSA. A second uncertainty is the rate coefficients for larger, and as yet unmeasured, CI species. The most important CI in the modeled scenario is the C, compound, CH2OO, which accounts for up to 50% of the CIs produced from isoprene. C-4 CIs may contribute up to 40% of the CIs produced and are expected to have substantially slower thermal unimolecular decay rates and water vapor reaction rate coefficients. Therefore, the model results may be a lower limit to the CI contribution to SSA. Calculated nighttime (10 h) total SO2 oxidation was 1.8%, of which 1.1% was due to CI + SO2, and the remainder to secondary OH + SO2. This compares to daytime (14 h) SO2 oxidation rates of 4% due to photochemical OH + SO2 reaction.