Journal of the American Chemical Society
Communication
redox catalysis and Cu catalysis would allow radical
carbocyanation of 1,3-dienes with carboxylic acid derived
NHP esters and TMSCN (Figure 1b). This promising plan
was designed to assemble a crucial allylcopper complex from
1,3-dienes, chiral Cu(II) complexes and photochemically
generated alkyl radicals, finally affording significant chiral
allyl nitrile products. Obviously, performing the reaction under
mild and neutral conditions can avoid the racemization of
sensitive chiral allyl cyanides.
only a trace amount of product (Table 1, entry 4). In addition,
variation in chiral ligands (L2−L4) did not lead to any
improvement on the results (Table 1, entries 5−7), while
increasing the equivalents of substrates 1a and 3 resulted in a
slightly enhanced yield (Table 1, entry 8, 54% yield and 91%
ee). We then identified the optimal photocatalysts (Table 1,
entries 9−11). When perylene (PC4) was used as the
photocatalyst and white LEDs were used as the light source,
similar reaction efficiency and enantiocontrol were afforded
(Table 1, entry 11), and increasing the reactant concentration
obviously improved the reaction efficiency (Table 1, entry 12:
71% yield and 92% ee). Control experiments confirmed that
photocatalysts, Cu catalysts, and light are all indispensable for
this multicomponent reaction.51
With this idea in mind, we first examined the multi-
component carbocyanation reaction between 1,3-dinene 1a,
NHP ester 2a, and TMSCN (3) (Table 1). Delightedly, we
a
Table 1. Condition Optimization
Having established the optimal reaction conditions, we
started to examine the generality of this methodology. As
highlighted in Table 2, a variety of bench-stable NHP esters
derived from readily available cyclic and linear alkylcarboxylic
acids were effective in this reaction, producing the correspond-
ing chiral allyl cyanide products in 66−73% yields and 89−
96% ee (Table 2, 4aa−4af). Many NHP esters with abundant
functional groups, such as alkenyl, alkynyl and chloride groups,
are also amenable to this synergetic catalysis system (4ag−
4ai). Next, we probed the scope of 1,3-diene components. As
illustrated in Table 2, regardless of the electronic properties
and substitution pattern of the benzene ring in 1,3-dienes,
chiral allyl cyanides were afforded in moderate to good yields
with high enantioselectivity (4bd−4hd, 50−67% yields and
92−97% ee). Moreover, thiophenyl- and furanyl-substituted
1,3-dienes 1i and 1j were converted to products 4id and 4jd in
70% yield with 94% ee and 73% yield with 95% ee,
respectively. Pyridinyl-substituted 1,3-diene 1k can also be
successfully converted to the desired product in moderate yield
and enantioselectivity (4kd, 45% yield and 70% ee). In
addition to monosubstituted 1,3-dienes, polysubstituted 1,3-
dienes were also investigated under the standard conditions.
For example, when an additional methyl or bromo substituent
was introduced at the β-position of the phenyl group, the
corresponding chiral allyl cyanides were obtained in moderate
yields with excellent enantioselectivities (4ld, 70% yield and
97% ee; 4md, 42% yield and 97% ee). We also found that the
1,1-disubstituted diene 1n and 1,4-disubstituted internal diene
1o were effective substrates, affording corresponding products
in promising results (4nd: 60% yield and 74% ee; 4od: 36%
yield, 79% ee, 96% ee, and 2.8:1 d.r.).52 However, internal
acyclic diene 1p failed to proceed with the carbocyanation at
the current stage. It is worth noting that nonactivated diene 1q
can be applied to this reaction, affording the desired product
4qd in 33% yield and 7% ee, which indicates that the aryl
group may be important to stabilize the generated allyl radical
and induce the enantioselectivity. Finally, the NHP ester
derived from commercially available drug dehydrocholin could
also be converted to the corresponding allyl cyanide in
moderate yield and good selectivity (4aj, 41% yield, >19:1
d.r.), which also proves the good functional group tolerance of
this strategy.
a
Conditions: 1,3-diene 1a (0.4 mmol), NHP ester 2a (0.2 mmol),
TMSCN 3 (0.3 mmol), Cu source (5 mol %), chiral ligand (6 mol
%), and PC (5 mol %) in anhydrous DCE (2 mL) at 29 °C for 24 h
under the irradiation of 2 × 3 W purple LEDs (light intensity = 37.4
b
mw/cm2). NMR yield determined using 1,3,5-trimethoxybenzene as
c
d
an internal standard. Determined by chiral HPLC analysis. Using
1,3-diene 1a (0.6 mmol) and TMSCN 3 (0.6 mmol). Using 2 × 3 W
white LEDs, 36 h. Using DCE (0.5 mL), isolated yield in
e
f
parentheses. ND: not determined.
found that the E-allyl cyanide product 4aa was formed in 40%
yield and 90% ee after irradiation with purple LEDs for 24 h in
the presence of organic photocatalyst PC1, Cu(MeCN)4BF4,
and chiral bisoxazoline ligand L1 (Table 1, entry 1). Notably,
To highlight the synthetic utility of the methodology, we
performed asymmetric radical carbocyanation on a gram scale.
Chiral product 4ad was facilely produced in 62% yield with
95% ee starting from 1,3-diene 1a, NHP ester 2d, and TMSCN
(Figure 2a). Furthermore, we converted allyl cyanide 4ad into
many other useful chiral compounds by simple operations
(Figure 2b). For example, treatment of 4ad under a H2
1
no Z-allyl cyanide was detected through H NMR analysis of
the reaction mixture. Screening of Cu sources indicated that
Cu(I) catalyst selection was crucial for this transformation. For
example, replacement of Cu(MeCN)4BF4 with CuOAc and
CuCN did not obviously affect the reaction efficiency and
selectivity (Table 1, entries 2 and 3), and Cu(OTf)2 afforded
4169
J. Am. Chem. Soc. 2021, 143, 4168−4173