C(70) oxides and ozonides and the mechanism of ozonolysis on the fullerene surface. A theoretical study.

A series of ab initio calculations have been carried out to determine why the a,b- and c,c-isomers are the most commonly observed mono-oxides of C(70) in ozonolysis reactions, when existing calculations in the literature report that these structures are not the most stable conformations. We show that the a,b- and c,c-isomers are the two most stable structures on the C(70)O(3) potential energy surface, which suggests that the reaction pathway toward oxide formation must proceed via the corresponding ozonide structure. From our calculations, we offer a mechanism for the thermally induced dissociation of C(70)O(3) that share the first two steps with the general mechanism for ozonolysis of alkenes proposed by Criegee. We suggest further steps that involve C(70)O(3) losing O(2) in its triplet or singlet state, thus leaving C(70)O in its triplet or singlet state, respectively. A pair of products in their singlet states seems to be more likely for the decomposition of a,b-C(70)O(3), which ultimately leads to the closed a,b-C(70)O epoxide structure. For c,c-C(70)O(3), the more thermodynamically favorable route is the triplet channel, resulting in the triplet open c,c-C(70)O oxidoannulene structure, which may subsequently decay to the singlet ground state c,c-C(70)O epoxide form. This finding offers an alternative interpretation of the experimental observations which reported an open d,d-C(70)O oxidoannulene structure as the metastable intermediate.