Atmospheric Fate of Acyloxyl Radicals: Decomposition versus Isomerization

Abstract Acyloxyl radicals (RCO2) are a common type of alkoxyl radical formed in the gas-phase oxidation of organic compounds in the atmosphere. Previous theoretical studies have predicted that the decomposition of RCO2 to an alkyl radical R and CO2 is a very fast reaction, suggesting it to be the exclusive fate for RCO2. In contrast, previous experimental work invokes intramolecular H-shift isomerization reactions of RCO2 in order to explain the formation of observed products. To resolve this apparent discrepancy, we investigate the electronic structure and energetics associated with the RCO2 decomposition reaction. Generally, reactions are studied on the electronic ground state of the reactant. However, using CH3CO2 as a model system, we show that it is not possible to accurately describe the CO2 loss reaction with only a single electronic state. Instead, we find that it is necessary to involve the three lowest-lying electronic states to describe the reactants and transition states accurately. Our state-averaged multireference calculations for the CO2 loss from CH3CO2 find a barrier of ∼7 kcal mol–1, corresponding to a rate coefficient of ∼108 s–1 at 298 K. Calculations for larger RCO2 test systems, where R is CH3CH2, (CH3)2CH, or (CH3)3C, show barriers that are 2–3 kcal mol–1 lower, corresponding to 1–2 orders-of-magnitude faster decomposition reactions. Thus, CO2 loss will likely dominate the fate of RCO2 but exceptionally fast H-shift reactions can be competitive. To support our theoretical calculations, we carry out OH oxidation experiments using glutaraldehyde as a model system and conclusively demonstrate that CO2 loss occurs for a variety of functionalized RCO2 with an empirically constrained lower-limit reaction rate coefficient of ∼107 s–1.