Computational
chemistry at the G3(MP2)-RAD//M06-2X/6-31+G(d,p)//SMD
level of theory was used to study the oxidation of a test set of methyl
adducts of nitroxide radicals and methyl adducts of Blatter’s
radical, a Kuhn verdazyl and two oxo-verdazyls. The barriers and the
reaction energies of the SN2 reactions of the oxidized
species with pyridine were also studied with a view to identify species
with both low oxidation potentials and low SN2 barriers,
so as to broaden the functional group tolerance of in situ electrochemical
methylation compared with TEMPO–Me (1-methoxy-2,2,6,6-tetramethylpiperidine).
Within the alkoxyamines, the oxidation potentials covered a range
of 0.5 V, with trends explicable in terms of electrostatics, ring
strain, and charge transfer. The oxidation potentials of oxo-verdazyl
adducts, verdazyl adducts, and particularly the methyl adducts of
Blatter’s radical were considerably low due to the ability
of their extensive π-systems to stabilize a positive charge.
As expected, the SN2 reaction energies of the oxidized
substrate became less favorable as the oxidation potential decreases.
Unfortunately, this also meant that the barriers increased due to
the excellent Evans–Polanyi correlation (R
2 = 0.92). Nonetheless, 7-methoxy-7-azadispiro[5.1.5.836]hexadecane, N,N-di-tert-butyl-O-methylhydroxylamine,
and particularly 1-methoxy-2,2,5,5-tetramethylpyrrolidine were identified
as suitable candidates for broadening the scope of in situ electrochemical
methylation while maintaining comparable kinetics to known reagents.