Angewandte
Chemie
Cross-Coupling
Visible-Light-Catalyzed Direct Benzylic C(sp3)–H Amination Reaction
by Cross-Dehydrogenative Coupling
Ganesh Pandey* and Ramkrishna Laha
Abstract: A conceptually new and synthetically valuable cross-
dehydrogenative benzylic C(sp3)–H amination reaction is
reported by visible-light photoredox catalysis. This protocol
employs DCA (9,10-dicyanoanthracene) as a visible-light-
absorbing photoredox catalyst and an amide as the nitrogen
source without the need of either a transition metal or an
external oxidant.
(CDC) using more environmental friendly metal salts[9,10] or
metal-free conditions,[11] but these strategies require strong
oxidizing agents. Radical-based cross-coupling reactions[12]
are also reported in this area but selectivity remains the
main issue in these approaches. Sarpong and co-workers have
reported an interesting strategy for intramolecular nitrogen
incorporation by oxidation of a benzylic C,N-di-anion using
iodine as an oxidant[13] but the major constraint of this
strategy remains functional group tolerance owing to the use
of an excess of strong base.
It becomes apparent from the above introduction that
during the last decade several fundamental reactions have
been discovered for direct C–H amination but the majority of
them either uses costly and toxic transition-metal catalysts or
requires strong oxidizing agents and the majority of them also
suffers from regio- and chemoselectivity issues. Therefore, it
was felt necessary that there is an urgent requirement of
developing an entirely new chemistry for the C–H amination
reaction which should overcome all the problems associated
with earlier methods and would also address environmental
issues (see Scheme 1). In the context of fulfilling this
À
D
irect amination of a saturated C H bond is emerging as an
[1]
À
important methodology for C N bond formation. The
nitrogen source in the majority of these reactions is a metal-
catalyzed nitrenoid that inserts into the C(sp3)–H bond using
either organic azides,[2] iminoiodanes[3] or haloamine deriva-
tives (ZNNaX)[4] as nitrene precursors. While non-selectivity,
low yield, and employment of a highly electrophilic nitrene
source typically marked the former processes, use of hyper-
valent iodine reagents and the generation of a stoichiometric
amount of environmentally non-benign iodobenzene remains
a major drawback of the later approaches. Although, an
alternative approach of generating nitrene and its insertion
À
into the C H bond by rhodium-catalyzed decomposition of
N-tosyloxycarbamates[5] overcomes some of these problems,
the high costs of the rhodium metal and its non-recyclability
limit its use. Furthermore, hypervalent N-trifluoromethylsul-
fonylimino-l3-bromane[6] has also been used as an active
organo nitrenoid species for the C–H amination reaction but
this strategy produces a stoichiometric amount of trifluoro-
methylbromobenzene as a by-product. A divergent regiose-
lectivity of intramolecular C–H amination has also been
reported by generating nitrenoid species from two differently
engineered variants of P450BM3 where one is favoring
À
amination of benzylic C H bonds and the other is favoring
[7]
À
homo-benzylic C H bonds, but this approach is highly
substrate specific. Apart from using metal/organo nitrenoids,
another interesting methodology in this area is reported by
Scheme 1. Intermolecular cross-dehydrogenative benzylic C(sp3)-H
amination reactions.
directed metal-catalyzed inter- and intramolecular amination
[8]
À
of C H bonds. However, use of a directing group as an
appendage limits its scope. Efforts have also been directed to
replace costly transition metals in C–H amination reactions
by developing catalytic cross-dehydrogenative couplings
challenge, our attention was drawn towards one of our own
recent reports wherein a benzylic C H bond was functional-
À
[14]
À
ized for C O bond formation by photoredox catalysis.
À
However, when the same protocol was attempted for C N
bond formation, the reaction failed due to competitive
electron-transfer processes between amine and alkyl aryl
groups.
Therefore, we envisioned an entirely new strategy to
generate first the benzylic radical 6 by H abstraction from
a captodative aminyl radical 5,[15] which was perceived to be
generated by a one-electron photoredox oxidation of 2a using
a singlet excited state of 9,10-dicyanoanthracene (DCA) as an
electron acceptor[16] followed by one electron oxidation of 6
[*] Prof. G. Pandey, R. Laha
Molecular Synthesis Laboratory
Centre of Bio-Medical Research (CBMR)
Sanjay Gandhi Postgraduate Institute of
Medical Science Campus Raebareli Road
Lucknow-226014 (India)
E-mail: gp.pandey@cbmr.res.in
Supporting information for this article is available on the WWW
Angew. Chem. Int. Ed. 2015, 54, 14875 –14879
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14875