Radical and polar reactivity of radical ions in solution. A new look at an old problem.

It is proposed that both polar and radical reactions should be considered when discussing radical ion reactivity. The fact that only the polar reactivity has dominated previous discussions is pointed out. It is argued that if the configuration mixing model is applicable for the estimation of radical cation-nucleophile (or cation-nucleophile) combination reaction barriers, the theory should also explain the reactivity of the proton toward anionic nucleophiles. The model severely overestimates reaction barriers for the latter reactions. The discussion is centered around the following hypothesis: In the absence of severe steric effects, the reactivity of radical ions, especially evident in the reactions with radicals (for example very few radical cations react with dioxygen at measurable rates), is dominated by the degree of coupling between charge and radical centers. Evidence from the literature supporting the hypothesis is presented for a number of different classes of radical ion reactions. Further work to test the validity of the hypothesis is proposed for many of the reaction types. Triarylmethyl radicals and carbenium ions are suggested as models to which the radical and polar reactivities of 9,10-diarylanthracene radical cations may be compared. It is concluded that the radical cation/carbenium ion comparison (for the reaction with acetate ion) would show similar reactivities while the radical cation/free radical comparison (for the reaction with dioxygen) would fail, since no reaction at all would be observed with the radical cation while the free radical reacts rapidly.