Evidence for CT Mechanism

Cohen has recently reviewed the large amount of research performed on photoreduction by amines 152 . The overall reaction is different for tertiary amines than for primary and secondary, since radicals with the structure R2C—NH—R are able to reduce ground state ketone in the same way that a-hydroxy radicals do. 

Photoreduction by amines differs from photoreduction by alcohols in two respects quantum yields are always lower than maximal and rate constants for amine quenching of triplet ketones are very large. These two facts led Cohen 153 and Davidson 154 to suggest that amines react with excited carbonyl compounds by electron transfer followed by proton transfer from the amine radical cation. 

It was later pointed out that a charge transfer rather than electron transfer process was more likely 155 . Now most people would call the product formed by interaction of a triplet ketone and an amine an exciplex, although there is no evidence for any low energy phosphorescence or for reversal to monomeric excited ketone. 

The best evidence for a CT process rather than direct hydrogen abstraction involves the values of kT s-butyl- and ferf-butylamine display much the same value 156 triethylamine and ferf-butyldimethylamine are equally reactive and some 50 times more so than primary amines 155 . Thus the rate constant for reaction is independent not only of the type of C—H bond a to the nitrogen but also of the presence or absence of a-hydrogens. Such evidence demands that abstraction of an a-hydrogen not be involved in the rate-determining quenching reaction. Moreover, the relative reactivity of amines (tertiary secondary primary) is proportional to the ease with which they are oxidized. 

Although all amines quench triplet ketones rapidly, the efficiency with which the interaction produces radicals varies considerably. 

---