J. Am. Chem. Soc. http://doi.org/6j7 (2015)

Nickel catalysis has gone from strength to strength in recent years. The success of nickel is often ascribed to its ability to adopt numerous oxidation states, with all four from 0 to +3 being readily accessible. Typically in cross-coupling reactions, a nickel centre will go through either a Ni(II)/Ni(0) or a Ni(III)/Ni(I) cycle, though recent work has shown that it is possible to go through several oxidation states in one reaction, and this allows some interesting reactivity to be observed. Now, Sarah Tasker and Timothy Jamison at MIT have made use of nickel's interesting redox behaviour, combining nickel-catalysis and photoredox-catalysis to synthesize indolines.

Unlike for indoles — their unsaturated counterpart — one-step annulation reactions to yield indolines are rare and often have limited scope. Based on the groups' previous work in nickel catalysed cross-couplings, they considered combining a Heck coupling with carbon–nitrogen bond formation. It was postulated that a substrate containing an appropriately placed nitrogen atom could intercept one of the intermediates of the Heck reaction, forming a new nitrogen–carbon bond and creating the indoline core. Unfortunately very little product was observed — indeed it appeared that no catalytic turnover was being achieved owing to the difficultly of Csp3−N reductive elimination from a Ni(II) intermediate. Tasker and Jamison then turned to photoredox catalysis in order to convert the stable Ni(II) intermediate into a more labile Ni(III) species via a one-electron oxidation. The use of a Ru(bpy)3(PF6)2 photoredox catalyst and visible-light irradiation gave yields above the catalyst loading and further optimization gave yields of up to 90% of the heterocyclic product.

With a functioning catalytic system in hand, the scope of the coupling was explored. The reaction was shown to be relatively insensitive to electronic effects, with both electron-donating and electron-withdrawing groups tolerated on the arene. It was somewhat more limited by sterics, and arenes with substituents in the 3- or 6-positions gave reduced yields. The alkene coupling partner could be varied widely with generally good yields overall.