Organic Letters
Letter
a
MBH reaction. To date, the backbones used in LBBA
phosphine organocatalysts have mostly been symmetric
scaffolds, whereas nonsymmetric scaffolds containing a
heterocyclic ring have rarely been investigated. As a kind of
prevalent heterocycle, indole has been used as the catalyst4
alone or as a key skeleton in some heterobiaryl systems5−7 and
has showed structural and synthetic advantages. Achiral indole-
aryl-derived phosphine has been utilized as a unique ligand for
Pd-catalyzed cross-coupling reactions5 or as organocatalyst in a
[4 + 1] cyclization reaction.6 Meanwhile, the axially chiral
indole-containing heterobiaryl backbones have been applied as
chiral ligands7 in asymmetric synthesis. However, the non-
symmetric chiral aryl-C2-indole skeleton has never been
installed into phosphine ligands or organocatalysts, probably
due to the lack of an atroposelective procedure for
constructing such architectures. Recently, several procedures
for the enantioselective preparation of chiral indole biaryl
systems have been established,8 and a series of axially chiral
indole derivatives have been prepared. In particular, some
practical procedures for producing chiral aryl-C2-indole
systems have been well established,8f,j,n so that the evaluation
of such scaffolds in LBBA phosphine organocatalysis can be
realized. Considering our continuing interests in organo-
catalysis,9 we intended to develop a novel LBBA phosphine
catalytic system based on axial chiral aryl-C2-indole skeletons
and then to investigate their applications in two types of formal
[4 + 2] cycloaddition reactions (Scheme 1b).
Spirooxindole architecture was first selected as the target
because such a scaffold widely exists in a number of natural
products and biologically active molecules.10 In 2014, Shi
reported a phosphine-catalyzed formal [4 + 2] tandem
cyclization of activated dienes with isatylidenemalononitriles.11
We were encouraged to perform the reaction under the
catalysis of our newly designed and synthesized chiral
phosphines containing an axially chiral naphthyl-C2-indole
scaffold. We began our study with isatylidenemalononitrile 1c
and activated diene 2a as the substrate in the presence of
phosphine catalyst A in toluene. The reaction achieved −79%
ee in <10% yield (Table 1, entry 1). This preliminary result
encouraged us to evaluate catalysts with different substituents
in indole part. Disappointingly, catalysts B, C, and D did not
significantly improve the yield and stereocontrol of the
reaction (Table 1, entries 2−4). Next, the catalyst with a
phosphine part in the indole moiety (catalyst E) was tested
under the same reaction conditions. Interestingly, when the
phosphine functionality shifted to the indole part, the
enantiomeric isomer of product 3c was revised to its
counterpart, and even the absolute configuration of the catalyst
was retained. Moreover, the reaction yield was significantly
increased, albeit the ee value of the product was decreased
(Table 1, entry 5). This result revealed that attaching a
phosphine part into the indole moiety could dramatically
increase the reaction output. The low enantioselectivity was
probably caused by the lack of a hydrogen-donor site between
the catalyst and the substrate. On the basis of this
phenomenon, catalysts F, G, and H were designed to bear a
free hydroxyl group in the naphthalene ring. To our delight,
the reaction proceeded smoothly in the presence of all three
catalysts, and the stereoselectivities were obviously increased.
Among them, catalyst F was found to give better stereo-
selectivity in terms of enantio- and diastereoselectivities (98%
ee, >20:1 dr). The further evaluation of the solvents (Table 1,
entries 9−13), temperature (Table 1, entries 14−18), and
Table 1. Optimization of Reaction Conditions
b
c
d
entry
cat.
solvent
T (°C) yield (%)
ee (%)
dr
1
2
3
4
5
6
7
8
A
B
C
D
E
F
G
H
F
F
F
F
F
F
F
F
F
F
F
F
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
PhCF3
CH2Cl2
CHCl3
THF
CH3CN
toluene
toluene
toluene
toluene
toluene
toluene
toluene
25
25
25
25
25
25
25
25
25
25
25
25
25
35
50
10
0
<10
<10
−79
−85
<10
94
95
89
90
93
85
88
88
96
95
95
94
94
85
93
65
−49
5
98
97
98
98
98
98
97
93
98
97
99
>99
>99
>99
>99
>20:1
>20:1
18:1
18:1
>20:1
15:1
9
10
11
12
13
14
15
16
18:1
15:1
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
>20:1
e
17
18
e
−10
0
0
e f
,
19
e g
,
20
a
Reaction conditions: 1c (0.10 mmol), 2a (0.25 mmol), and catalyst
(10 mol %) in the solvent (2.0 mL) for 12 h under N2, unless
otherwise specified. Isolated yield. Determined by HPLC analysis
b
c
d
using a chiral stationary phase. dr = diastereoselectivity ratio,
e
f
g
determined by 1H NMR. 24 h. 5 mol % catalyst. 2.5 mol % catalyst.
catalyst loading amount (Table 1, entries 19 and 20) revealed
that the best reaction performance could be achieved under the
reaction conditions as follows: 1c (0.10 mmol), 2a (0.25
mmol), and 5 mol % catalyst F in toluene (2.0 mL) at 0 °C for
24 h under N2.
With the optimized reaction conditions in hand (Table 1,
entry 19), the substrate scope of the established enantiose-
lective formal [4 + 2] cascade cyclization reaction was
investigated (Scheme 2). First, different substituent groups
on the nitrogen atom of isatylidenemalononitriles were tested,
and methyl, methyloxymethyl (−MOM), allyl, benzyl, and
phenyl were demonstrated to be compatible with the reaction
conditions and gave products 3a−e in good yields with high
stereoselectivities (up to >99% ee, >20:1 dr). Next, different
electronic properties and positions of the substituent groups
attached to the phenyl ring of isatylidenemalononitriles did not
significantly affect the stereocontrol of the reactions, and
B
Org. Lett. XXXX, XXX, XXX−XXX