In 1980 Steckhan and Schmidt introduced the use of 4,4’,4”-tribromotriphenylamine (1a) and several more highly halogenated derivatives for the electrocatalytic anodic oxidation of organic substrates,1 and a sizeable number of such transformations have been reported.2 The relatively modest oxidation potential of 1 (+0.78 vs Ag/0.1 M AgNO3 reference) imposes restrictions, however, on the range of substrates that can treated in this manner. For this reason, we synthesized a series of triarylamines bearing several electron-withdrawing groups with the expectation that these would be useful for effecting catalytic oxidation of substrates that high oxidation potentials.3 We had found earlier that such substances can be difficult to oxidize anodically.4We have had some success using these more electronegatively substituted triarylamines.5 However, the situation is not as simple as it may appear. For example, the generalization is often made that mediated oxidation can be effected at potentials up to 500-600 mv positive of the oxidation potential of the substrate, but the latter depends, inter alia, upon the rate of further chemical reaction of the oxidized form of the substrate. Other assumptions are that electronegative substituents increase the oxidative power of the triarylamine catalyst, but not any of its other chemical properties and that the amine cation radical is stable on the time scale of the electrolysis. As we will report here, we find that both of these assumptions are often unwarranted for triarylamines. We recently carried out a series of controlled potential electrolyses of a 10:1 mole ratio of trans-stilbene and 4,4’,4”-trimethyl-2,2’,2”-trinitrotriphenylamine (1b) in aqueous acetonitrile. Under these conditions it was found that the initially produced dark blue solution of 1b+ (the cation radical of 1b) quickly turns first violet and then red and much unreacted stilbene remains. The color change demonstrated that 1 + is unstable under these conditions. A subsequent electrolysis of 1 in the absence of stilbene, followed by flash chromatography of the electrolysis mixture afforded a red solid whose formula (from exact mass measurement by mass spectrometry) is C21H16N4O8. We assign structure 2 (a substituted N-phenylphenoxazine) to this substance because a reasonable mechanistic path to it can be written involving three consecutive ECE processes involving nucleophilic attack by water upon 1 +to introduce the oxygen atoms onto the rings.One might conclude from this that the presence of three highly withdrawing nitro groups in 1 + is responsible for its instability by increasing its electrophilicity and that a less highly nitrated derivative would not take this path. This is not so; more recently, we have found that anodic oxidation of the mononitro triphenylamine 1c in aqueous acetonitrile under the same conditions also affords a red substance to which we again assign a phenoxazine structure. Although 1b and 1c both exhibit reversible cyclic voltammograms, this is misleading. In fact, the cation radicals from both 1b and 1c both undergo rather efficient nucleophilic attack by water in experiments of longer time scale. Phenoxazine formation is apparently efficient from 1c even though its oxidation potential is much lower than that of 1b. Decomposition of triphenylamine cation radicals may be a general pattern; the tribromide 1a has also been observed to decompose during preparative scale electrolysis.6 Finally, we note that instability of cation radicals, with its concomitant undesirable impact on electrosynthetic applications, may prove to be problematical in other organic electrocatalytic systems, particularly when they bear one or more strongly electron-withdrawing groups.
Read full abstract