The radical stabilization energy of a substituted carbon-centered free radical depends on both the functionality of the substituent and the ordinality of the radical.

Chemical intuition suggests that the stabilization of a carbon-centered free radical by a substituent X would be the greatest for a prim and least for a more stable tert radical because of "saturation". However, analysis of a comprehensive recent set of bond dissociation energies computed by Coote and co-workers (Phys. Chem. Chem. Phys. 2010, 12, 9597) and transformed into radical stabilization energies (RSE) suggests that this supposition is often violated. The RSE for a given X depends not only on the nature of X but also on the ordinality (i.e., prim, sec, or tert) of the radical onto which it is substituted. For substituents that stabilize by electron delocalization but also contain electron-withdrawing centers, such as the carbonyl function, the stabilization of XCMe(2)(•) compared with HCMe(2)• is greater than that for XCH(2)• compared with HCH(2)•. However, for substituents that stabilize by lone-pair electron donation, such as N or O centers, the order is strongly reversed. This contrast can be qualitatively rationalized by considering charge-separated VB contributors to the radical structure (R(2)C(+)-X(-•) and R(2)C(-)-X(+•)) and the contrasting effects of methyl substituents on them. This conclusion is not dependent on the particular definition used for RSE.