Methyl and silyl mesolytic dissociations in the radical cations and radical anions of but-1-ene, allylsilane, hexa-1,3-diene, and penta-2,4-dienylsilane. CAS-MCSCF and Coupled Cluster theoretical study

Methyl- or silyl- dissociation in the CH2=CHCH2-XH3 (a-XH2•+) and CH2=CHCH=CHCH2-XH3 (p-XH3•+) radical cations (X = C, Si) yields a+ or p+ and XH3. Similarly, the radical anions a-CH3•- and p-CH3•- give the Π-delocalized anion and CH3• preferentially. In contrast, a-SiH3•- and p-SiH33•- prefer to dissociate into the Π-delocalized radical and silide. All reactions are endoergic: by 43 - 50 kcal mol-1 in the radical cations, and easier to some extent in the radical anions, that require 29 - 33 kcal mol-1 (X = C), and 13 - 14 kcal mol-1 (X = Si). The fragmentation energy profiles do not present significant barriers for the backwards process in the case of the radical cations. All radical anions exhibit an energy maximum along the dissociation pathway, but the barrier is lower than the dissociation limit. Fragmentation is "activated" more in the anions than in the cations with respect to homolysis in the corresponding neutrals (that requires 72 - 81 kcal mol-1). Wavefunction analysis indicates that the C-X bond cleavage in the hydrocarbon radical ions, although formally comparable to a homolytic process, is at variance with this model, due to the spin recoupling of one of the two C-X bond electrons with the originally unpaired electron. This is basically true also for the silyl-substituted radical anions, in which the initial more delocalized charge distribution might suggest some heterolytic character of the bond cleavage.