Organic Letters
Letter
a
(Scheme 1A). In contrast, three-component intermolecular
cyanoalkylcarbofunctionalization of alkenes, which would
efficiently forge two C−C bonds and rapidly achieve high
levels of molecular complexity, remains challenging and more
underdeveloped. One prominent challenge is an unwanted
competitive two-component reaction. Recently, elegant works
from the group of Xiao and Chen revealed the viability of 1,2-
cyanoalkylaryllation11 and 1,2-cyanoalkylalkynylation12 of
alkenes through a photoinduced, copper-catalyzed radical
relay (Scheme 1B). Nevertheless, vicinal cyanoalkylacylation
of alkenes, to the best of our knowledge, is still unexplored to
date. Given the ubiquity and paramount importance of
carbonyl group in almost every facet of chemical science,13
as well as the prevalence of cyanoalkyl moieties in many
natural products and pharmaceuticals,14 an efficient and
modular method for 1,2-cyanoalkylacylation of alkenes would
be highly desirable.
Enlightened by recent striking radical transformations
mediated by N-heterocyclic carbene (NHC),15−17 we envi-
sioned that the merging of iminyl radical chemistry with NHC
organocatalysis might provide a new opportunity for
accomplishment of vicinal cyanoalkylacylation of alkenes with
widely available aldehydes. In this regard, we became intrigued
by the reducing enolate form of Breslow intermediate, which
can form a persistent radical intermediate through single-
electron oxidation as reported in the pioneering studies of
Fukuzumi group (Scheme 1C).18 Building on this knowledge,
recently, the Ohmiya group used redox-active carboxylic esters
as SET oxidants to fulfill the 1,2-alkylacylation of alkenes via
NHC-catalyzed radical relay.19 Prompted by this elegant study,
fluoroalkyl reagents20 and Katritzky pyridinium salts21 were
subsequently identified as viable SET oxidants by other groups.
Despite these venerable advances, in contrast to NHC-enabled
electron-pair-transfer reactions,22 the realm of NHC catalyzed
radical transformations is still in its infancy,15b especially with
respect to vicinal alkene dicarbofunctionalization. Herein, we
report an unprecedented protocol for the intermolecular
radical 1,2-cyanoalkylacylation of alkenes with oxime esters
and aldehydes through NHC organocatalysis in a completely
regioselective fashion (Scheme 1D). This modular method
grants a rapid, flexible access to densely functionalized versatile
ketonitrile architectures under mild, transition-metal-free, and
redox-neutral conditions.
Our investigations commenced by reacting O-(tert-butox-
ycarbonyl) oxime 1a with styrene 2a and 4-chlorobenzalde-
hyde 3a in the presence of the NHC catalyst. An abbreviated
outline of optimization studies is provided in Table 1. In its
optimal manifestation, a 79% yield of the three-component
coupling product, ζ-keto nitrile 4aaa, was obtained from 1a
(0.15 mmol), 2a (0.2 mmol), and 3a (0.1 mmol) using 1.5
equiv of Cs2CO3 in concert with the N-2,6-diisopropylphenyl-
substituted cycloheptane-fused thiazolium precatalyst C1 (10
mol %) in DCM at 60 °C (entry 1). The acyl and γ-cyanoalkyl
fragments were incorporated into the α and β positions of
styrene, respectively, in an exclusively regioselective fashion. A
small quantity of the byproduct 5aa was observed due to the
competitive two-component cross-coupling of 1a and 3a.
The screening of NHC catalysts unveiled that both the
backbone and the N-substituent of the NHC precursors were
essential for the reaction efficiency (Table 1, entries 2−4). The
reaction saw an obvious decrease in yield when conducted
using NHC precursors possessing a cyclohexane or dimethyl
backbone or a smaller N-mesityl group. Switching the R group
Table 1. Optimization of the Reaction Conditions
yield
(4aaa, %)
yield
b
b
entry deviation from standard conditions
(5aa, %)
1
2
3
4
5
none
79
56
55
45
64
14
10
11
15
18
C2 instead of C1
C3 instead of C1
C4 instead of C1
1a′ (R = 4-CF3C6H4CO) instead of
1a
6
7
1a′′ (R = C6F5CO) instead of 1a
using Li2CO3, Na2CO3, K2CO3, or
K3PO4
40
<15
15
<5
8
9
DBU instead of Cs2CO3
NEt3 instead of Cs2CO3
NMM instead of Cs2CO3
using pyridine or 2,6-lutidine
1.0 equiv of Cs2CO3
ratio of 1a:2a:3a = 2:2:1
ratio of 1a:2a:3a = 1.2:2:1
ratio of 1a:2a:3a = 1.5:1.5:1
no NHC catalyst
10
32
10
0
48
56
62
42
0
0
9
0
10
11
12
13
14
15
16
17
0
10
18
14
18
0
no base
0
0
a
Reaction conditions: 1a (0.15 mmol), 2a (0.2 mmol), 3a (0.1
mmol), C1 (10 mol %), and Cs2CO3 (0.15 mmol) in DCM (1.0 mL)
at 60 °C for 12 h under Ar. Yield determined by H NMR analysis
using 1,3,5-trimethoxybenzene as an internal standard. Boc = tert-
butyloxycarbonyl, DBU = 1,8-diazabicyclo[5.4.0]-7-undecene. NMM
= N-methylmorpholine.
b
1
of O-acyl oxime from Boc to 4-(trifluoromethyl)benzoyl or
perfluorobenzoyl rendered an erosion in the reactivity (entries
5 and 6). The specific identity of the base proved critical
(entries 7−11), as did the choice of the solvent system (see the
equivalent of Cs2CO3 employed led to lower yields (entry 12).
Further refinement of the molar ratio of 1a and 2a to 3a did
not boost the product yield (entries 13−15). As speculated,
control studies confirmed that the NHC catalyst and base were
indispensable to this vicinal alkene dicarbofunctionalization
reaction (entries 16 and 17).
With an optimized set of conditions established, the scope of
this NHC organocatalytic three-component cyanoalkylacyla-
tion of alkenes was then explored (Scheme 2). The initial focus
was on evaluating alkene diversity. Aside from 2a, vinyl arenes
bearing either electron-withdrawing or -donating substituents
on the phenyl ring could all engage in this dicarbofunction-
alization reaction and furnish the desired ζ-keto nitriles in
moderate to good yields (4aba−4aga). Chloro-substituents at
the meta- and ortho-position of styrene were well tolerated
under the standard conditions (4afa and 4aga). Naphthyl-,
thiophene-, and pyridine-substituted olefins also proved to be
competent coupling partners (2h−2k) for the facile assembly
B
Org. Lett. XXXX, XXX, XXX−XXX