The Reaction of Ozone with the Hydroxide Ion: Mechanistic Considerations Based on Thermokinetic and Quantum Chemical Calculations and the Role of HO₄⁻ in Superoxide Dismutation

The reaction of OH(-) with O(3) eventually leads to the formation of *OH radicals. In the original mechanistic concept (J. Staehelin, J. Hoigné, Environ. Sci. Technol. 1982, 16, 676-681), it was suggested that the first step occurred by O transfer: OH(-)+O(3)-->HO(2)(-)+O(2) and that *OH was generated in the subsequent reaction(s) of HO(2)(-) with O(3) (the peroxone process). This mechanistic concept has now been revised on the basis of thermokinetic and quantum chemical calculations. A one-step O transfer such as that mentioned above would require the release of O(2) in its excited singlet state ((1)O(2), O(2)((1)Delta(g))); this state lies 95.5 kJ mol(-1) above the triplet ground state ((3)O(2), O(2)((3)Sigma(g)(-))). The low experimental rate constant of 70 M(-1) s(-1) is not incompatible with such a reaction. However, according to our calculations, the reaction of OH(-) with O(3) to form an adduct (OH(-)+O(3)-->HO(4)(-); DeltaG=3.5 kJ mol(-1)) is a much better candidate for the rate-determining step as compared with the significantly more endergonic O transfer (DeltaG=26.7 kJ mol(-1)). Hence, we favor this reaction; all the more so as numerous precedents of similar ozone adduct formation are known in the literature. Three potential decay routes of the adduct HO(4)(-) have been probed: HO(4)(-)-->HO(2)(-)+(1)O(2) is spin allowed, but markedly endergonic (DeltaG=23.2 kJ mol(-1)). HO(4)(-)-->HO(2)(-)+(3)O(2) is spin forbidden (DeltaG=-73.3 kJ mol(-1)). The decay into radicals, HO(4)(-)-->HO(2)*+O(2)(*-), is spin allowed and less endergonic (DeltaG=14.8 kJ mol(-1)) than HO(4)(-)-->HO(2)(-)+(1)O(2). It is thus HO(4)(-)-->HO(2)*+O(2)(*-) by which HO(4)(-) decays. It is noted that a large contribution of the reverse of this reaction, HO(2)*+O(2)(*-)-->HO(4)(-), followed by HO(4)(-)-->HO(2)(-)+(3)O(2), now explains why the measured rate of the bimolecular decay of HO(2)* and O(2)(*-) into HO(2)(-)+O(2) (k=1 x 10(8) M(-1) s(-1)) is below diffusion controlled. Because k for the process HO(4)(-)-->HO(2)*+O(2)(*-) is much larger than k for the reverse of OH(-)+O(3)-->HO(4)(-), the forward reaction OH(-)+O(3)-->HO(4)(-) is practically irreversible.