In the late 1990s I was working on a chemistry project involving cascade reactions of C-centred radicals (CCRs). At that time, N-centered radicals (NCRs) were little used compared to their C-centered relatives and I was wondering if things had changed much since then.
Considering that the first example of the Hofmann-Löffler-Freytag reaction was reported in 18831 it seems that the development of the chemistry of NCRs has been fairly slow. More recently, in 2008, Zard2 described NCRs as a forgotten species with significant synthetic potential in a comprehensive review of intramolecular NCR cyclisation. NCRs have suffered from limitations because traditional methods for their generation have relied on the synthesis of N-X precursors and required harsh conditions for bond homolysis (Scheme 1, path A). New developments in metal catalysis (path B) and visible-light photocatalysis (path C) have resulted in greater use of NCR chemistry (for C-N bond formation) and advances in this field were reviewed by Zhang3 last year.
Scheme 13 Strategies in C-N bond formation based on NCR chemistry
Generation of NCRs using visible-light photocatalysis is particularly attractive due to low catalyst loading and mild conditions, with most reactions occurring at room temperature without the need for highly reactive radical initiators. This technology has been widely applied to radical amination chemistry in the last three years. An interesting example of this approach entitled ‘’Visible-Light-Mediated Generation of Nitrogen Centered Radicals: Transition Metal Free Hydroimination and Iminohydroxylation Cyclisation Reactions’’ was published by Leonori.4 One of the transformations described is a 5-exo–trig cyclisation of iminyl radicals to give pyrrolines without the use of a transition metal catalyst (Scheme 2). A diverse range of oximes were converted to the corresponding pyrrolines, in good yields, and bicyclic heterocycles were also prepared using this methodology.
Scheme 24 Visible-light mediated hydroimination
The mechanism of this reaction involves single electron transfer (SET) reduction of the aryl oxime ether (A) by a visible-light exited photocatalyst (*PC), followed by N-O bond fragmentation to give iminyl radical (D) which can then cyclise onto the pendant alkene (Scheme 3). The role of cyclohexadiene (CHD) is to reduce the intermediate radical (E), followed by a second reduction of the photocatalyst (PC), which is then irradiated (PC®PC*) completing the catalytic cycle. Cyclic voltammetry studies suggested that that the key single electron transfer (SET) reduction, by excited state organic dye eosin Y, would only be efficient when using nitro-substituted aryl oxime ethers (1a-1d) with suitable reduction potentials. The 2,4-dinitro-aryl oxime (1a) was found to be the best substrate.
Scheme 34 Proposed photoredox cycle and electrochemical studies. EY = eosin Y, ppy = 2-phenylpyridine
It was found that a different activation mode could be used to generate the iminyl radical without the presence of a photocatalyst (Scheme 4). When a solution of the aryl oxime ether (2a) and triethylamine in acetonitrile was irradiated with visible-light in the presence of CHD the previously observed hydroimination product (3a) was obtained, and also an unexpected iminoalcohol (4a). When CHD was not in the reaction mixture iminoalcohol (4a) was the main product. The mechanism proposed for this process involves formation of an unusual electron donor-acceptor complex (5) which undergoes SET giving the dipolar species (6). This species can then fragment, and undergo 5-exo-trig cyclisation to give radical (7). If the radical is not reduced by CHD it can be oxidised by attacking the nitro group (8) which after N-O bond homolysis, and hydrogen atom abstraction, gives the iminoalcohol (4a). Evidence for the source of the oxygen was provided by generation of 2-NO-4-NO2C6H3OH (10) from the reaction which was in contrast to 2,4-dinitrophenol that was produced from the original hydroimination.
Scheme 44 Initial findings and proposed reaction mechanism for the iminohydroxylation
The chemistry of NCRs has been developing in recent years, but they still remain less well used than CCRs. Significant problems, such a functional group compatibility, need to be solved before they can become routinely used for C-N bond construction alongside traditional ionic chemistry. Visible-light photocatalysis seems very likely to play a role in future developments.
Blog written by Michael Annis
1. J. L. Jeffrey and R. Sarpong, Chem. Sci., 2013, 4, 4092
2. S. Z. Zard, Chem. Soc. Rev., 2008, 37, 1603
3. T. Xiong and Q. Zhang, Chem. Soc. Rev., 2016, 45, 3069
4. J. Davies, S. G. Booth, S. Esaffi, R. A. W. Dryfe and D. Leonori, Angew. Chem. Int. Ed. 2015, 54, 14017