last pattern for the radical‐involved C—N bond formation is the cross‐coupling of C‐ and N‐radicals, which is relatively undeveloped compared with the aforementioned two pathways, presumably because of the challenge to provide an appropriate environment for the simultaneous generation of both C‐ and N‐radicals. Moreover, the generated C‐ and N‐radical in the same system are supposed to obey to the principle of persistent radical effect (PRE) to furnish the desired C–N cross‐coupling product rather than the homocoupling ones.
3.4.1 Aryl C(sp2)—N Bond Formation via Radical Cross‐coupling
Among the reported C—N bond construction methods directly from aromatic C(sp2)—H bonds and N—H bonds via radical cross‐coupling pathway, both of the coupling partners are required to be redox active to produce their corresponding radical intermediates upon oxidation. Therefore, in most cases, electron‐rich arenes such as phenols and anilines are chosen as the aryl radical precursors. As for the amine partner, electron‐rich diarylamines and azoles are commonly used. Sulfonamides after deprotonation can also turn into the corresponding radicals upon oxidation, as previously described. For these amine sources, the stabilizing effect for the N‐radicals also appears to be crucial for the success of the radical cross‐coupling.
Scheme 3.39 Mechanistic proposal for the light‐driven direct intramolecular C–N cross‐coupling.
3.4.1.1 Aryl C(sp2)—N Bond Formation Using Diarylamines
In 2016, Xia and coworkers developed a cross‐dehydrogenative C–N coupling between phenols 230 including trimethylsilyl (TMS)‐protected ones and phenothiazines 231 to afford a variety of ortho‐aminated phenols 232 employing K2S2O8 as the external oxidant under mild visible light photoredox conditions (Scheme 3.40) [53]. This protocol requires no additional photocatalyst, which is probably because of the inherent properties of phenothiazines (231) that can absorb visible light to enable the following energy transfer process. Notably, instead of inhibiting the amination or lower the reaction efficiency, additional radical scavenger TEMPO dramatically improves the yield of desired product 232a. Moreover, the reaction system of 3,4‐dimethoxyphenol (230b) and 2‐methoxylphenothiazine (231b), which only yields 232c as a homocoupling product of 231b under the original conditions, does furnish the desired C–N cross‐coupling product 232b with extra 2.0 equiv of TEMPO added. On the basis of some relevant literature, a hypothesis that TEMPO can covalently and reversibly prolong the lifetime of the transient N‐radicals is reasonably proposed by the authors. Mechanistically, under the irradiation of visible light, oxidant K2S2O8 is capable of oxidizing both phenol and phenothiazine to provide C‐radical 235 and N‐radical 236, respectively. The following radical–radical cross‐coupling process generates intermediate 237, which further tautomerizes into the final amination product 232.
Scheme 3.40 Visible‐light‐promoted CDC amination of phenols and phenothiazines. (a) Reactions with additional 2.0 equiv of TEMPO. (b) The mechanistic proposal.
Source: Modified from Zhao et al. [53].
However, the amine scope of this catalyst‐free amination is limited to S‐ or O‐containing cyclic diarylamines, which does not include acyclic ones because they are incapable of absorbing visible light and possess higher oxidation potentials. To accommodate these substrates, Xia's group subsequently developed another photocatalytic strategy to enable the CDC amination of phenols 238 with acyclic diarylamines 239 employing the organic salt 2,4,6‐triphenylpyrylium (TPT) as a photocatalyst (Scheme 3.41) [54]. This protocol is proven compatible with a wide range of electron‐rich acyclic diarylamines and phenols including the natural product (+)‐δ‐tocopherol (240a–240c). Notably, phenothiazine or phenoxazine substrates, which require prolonged time to complete the reaction using the previous method, convert much faster under these conditions. As demonstrated by a series of mechanistic studies including EPR spectroscopy and fluorescence quenching experiments, diaryl amine 239 is able to reductively quench the photoexcited catalyst TPT*, generating the corresponding radical cation 241 along with reduced TPT·−. A radical chain propagation process is indicated by the determined quantum yield (Φ = 19), which is proposed to take place between phenol 238 and N‐centered radical cation 241 to provide phenol radical cation 242 regenerated amine 239. An alternative pathway to reach 242 is through the SET oxidation of 238 by persulfate. Finally, radicals 244 and 243 formed from deprotonation couple with each other to deliver the cross‐coupling product 240.
Scheme 3.41 Visible‐light‐mediated CDC amination of phenols and acyclic diarylamines.
Source: Modified from Zhao et al. [54].
Scheme 3.42 Electro‐oxidative para‐selective C–H/N–H cross‐coupling with hydrogen evolution.
Source: Modified from Liu et al. [55].
An electrochemical protocol via a similar mechanism was described by Lei's group in early 2019, through which various triarylamines can be synthesized by coupling electron‐rich arenes and diarylamine derivatives (Scheme 3.42) [55]. More specifically, a series of para‐selective amination products 248 are furnished from the reactions between aniline derivatives 246 and diarylamines 247. Both coupling partners are found to be redox active during the mechanistic investigation, as exemplified by a plausible mechanism related to 248a. Therefore, 246a and 247a are simultaneously oxidized at the anode to afford radical cation 249a and radical 251a after deprotonation. The subsequent radical/radical cross‐coupling produces cationic species 252a, which turns into the final product 248a after further deprotonation. Shortly after the publication of this work, Song and Li and coworkers also disclosed their independent research, which is very close to Lei's work [56].
3.4.1.2 Aryl C(sp2)—N Bond Formation Using Azoles
In
