It has recently been demonstrated that the O<sub>2</sub>SO<sub>3</sub><sup>−</sup> ion forms in the atmosphere as a natural consequence of ionizing radiation. Here, we present a density functional theory-based study of the reactions of O<sub>2</sub>SO<sub>3</sub><sup>−</sup> with O<sub>3</sub>. The most important reactions are (a) oxidation to O<sub>2</sub>SO<sub>3</sub><sup>−</sup> and (b) cluster decomposition into SO<sub>3</sub>, O<sub>2</sub> and O<sub>3</sub><sup>−</sup>. The former reaction is highly exothermic, and the nascent O<sub>2</sub>SO<sub>3</sub><sup>−</sup> will rapidly decompose into SO<sub>4</sub><sup>−</sup> and O<sub>2</sub>. If the origin of O<sub>2</sub>SO<sub>3</sub><sup>−</sup> is SO<sub>2</sub> oxidation by O<sub>3</sub><sup>−</sup>, the latter reaction closes a catalytic cycle wherein SO<sub>2</sub> is oxidized to SO<sub>3</sub>. The relative rate between the two major sinks for O<sub>2</sub>SO<sub>3</sub><sup>−</sup> is assessed, thereby providing a measure of the maximum turnover number of ion-catalysed SO<sub>2</sub> oxidation, i.e. how many SO<sub>2</sub> can be oxidized per free electron. The rate ratio between reactions (a) and (b) is significantly altered by the presence or absence of a single water molecule, but reaction (b) is in general much more probable. Although we are unable to assess the overall importance of this cycle in the real atmosphere due to the unknown influence of CO<sub>2</sub> and NO<sub>x</sub>, we roughly estimate that ion-induced catalysis may contribute with several percent of H<sub>2</sub>SO<sub>4</sub> levels in typical CO<sub>2</sub>-free and low NO<sub>x</sub> reaction chambers, e.g. the CLOUD chamber at CERN.