COMMUNICATION
Table 1. Optimization studies for the enantioselective nitrocyclopropana-
tion of oxindoles.[a]
Table 2. Synthesis of spiro nitrocyclopropyl oxindoles with two adjacent
tertiary stereogenic centers.[a]
Entry
4
R
R1
T [8C]/
t [h]
Yield
[%][b]
d.r.[c]
19:1
9:1
>19:1
3:1
4:1
7.7:1.6:1
3:1
ee
[%][d]
Entry Base
Solvent t [h] Conversion [%][b] d.r.[b] ee [%][c]
1
2
a
b
b
c
c
d
e
f
H
H
H
H
H
H
H
5-Br
5-Br
5-Cl
6-Cl
5-Me
H
COOEt
COPh
COPh
n-butyl
n-butyl
Ph
2-MePh
COOEt
COOEt
Ph
0/48
0/72
77
52
82
60
75
83
80
76
74
84
65
60
99
98
90
98
93
92
90
90
87
93
90
89
91
97
1
2
3
4
NaHCO3 Toluene 65
NaHCO3 CH2Cl2 24
NaHCO3 CH3CN 24
63
57
40
67
76
47
48
70
100
18:1 98
9.8:1 96
9.4:1 82
9.4:1 97
16:1 98
16:1 98
5.3:1 93
16:1 98
19:1 98
3[e]
4
RT/72
RT/72
RT/72
RT/72
RT/48
RT/24
0/72
RT/48
RT/48
RT/72
RT/72
NaHCO3 THF
NaHCO3 MTBE
NaHCO3 MTBE
24
24
24
24
24
48
5[f]
6[e]
7[e]
8
5
7[d]
8[d]
9[d]
Et3N
Na2CO3
MTBE
MTBE
MTBE
9:1
9[e]
10[e]
11[e]
12[g]
13[h]
f
>19:1
8.6:1.3:1
12:2:1
7:1
10[d,e] Na2CO3
g
h
i
Ph
Ph
COOEt
[a] Unless otherwise specified, the reactions were performed at room
temperature on a 0.1 mmol scale with 1c (10 mol%), 3 (1.5 equiv), base
a
11:1
1
(1 equiv), and 2a (0.2m) [b] Determined by H NMR spectroscopic analy-
sis of the crude mixture. [c] Determined by HPLC analysis on a chiral
stationary phase. [d] Reaction performed with 5 mol% of 1c. [e] Re-
action performed at 08C.
[a] Unless otherwise specified, the reactions were performed on a
0.2 mmol scale with 1c (5 mol%), 3 (1.5 equiv), Na2CO3 (1 equiv), and
1
2a (0.2m). [b] Sum of diastereoisomers. [c] Determined by H NMR spec-
troscopic analysis of the crude mixture. [d] Determined by HPLC analy-
sis on a chiral stationary phase. [e] Reaction performed with 5 mol% of
1d. [f] Reaction performed with 10 mol% of 1d. [g] 99% ee after a single
crystallization. [h] Reaction performed with 2 mol% of 1d.
reaction solvent dramatically affected the conversion, d.r.,
and ee values (Table 1, entries 2–4), with the exception of
methyl tert-butyl ether (MTBE), which gave similar results
to toluene, but with a faster rate of reaction (Table 1,
entry 5). When an organic base stronger than NaHCO3 was
used, the reaction did not proceed efficiently and 4a was
isolated with a lower ee value, probably because of a com-
petitive background reaction (Table 1, entry 8).[9a] Converse-
ly, the addition of one equivalent of Na2CO3 provided the
same level of enantiocontrol with an increase in conversion
and diastereoselectivity (Table 1, entries 9 and 10). The reac-
tion proceeded smoothly with 1c (5 mol%) at 08C for 48 h
to give 4a in 77% yield, with an ee value of 98% and a d.r.
of 19:1.
The optimized reaction conditions were then used to in-
vestigate the synthesis of different spiro nitrocyclopropane
oxindoles and to explore the accessibility of both enantio-
mers (Table 2). Indeed, the use of catalyst 1d (the pseudo-
enantiomeric form of catalyst 1c), allowed access to the op-
posite enantiomers of the spirooxindoles, with an improve-
ment of both the diastereo- and enantiocontrol (Table 2, en-
tries 3, 5 and 9). Activated double bonds with ester and
ketone substituents gave the corresponding cyclopropane as
We believe that the high diastereo- and enantioselectivity
observed in the course of the reaction, is mainly due to the
ability of the bifunctional catalyst to activate the Michael
donor and the electrophilic oxindole simultaneously. This
hypothesis led us to propose a reaction mechanism in which
À
a hydrogen-bond interaction between the N H bonds of the
thiourea moiety and the imidic carbonyl groups[11a–b] of the
oxindole, plays a crucial role in the stereochemical outcome
of the reaction.[16] By placing the oxindole in the correct po-
sition for the oncoming nucleophile on the Si face of the
double bond, the reaction follows the alkylation pathway
shown in Scheme 2.
a
single diastereoisomer with high enantioselectivity
(Table 2, entries 1, 3 and 9). Similarly, the reaction proceed-
ed smoothly when alkyl and aryl substituents were used
(Table 2, entries 4–7 and 10–12). The presence of electron-
withdrawing or donating substituents on the aromatic ring
of the oxindole did not interfere with the general trend of
the results (Table 2, entries 8–12). Remarkably, the use of
2 mol% of 1d gave 4a in quantitative yield, with higher
enantiocontrol, albeit with a lower d.r. value (Table 2,
entry 13).
Scheme 2. Proposed activation mode of the bifunctional catalyst 1c.
The interesting results obtained so far prompted us to
face the synthetic problem associated with the construction
of two adjacent quaternary stereocenters. For this purpose
we tested the reaction of 1-bromo-1-nitroethane (5) with
Chem. Eur. J. 2011, 17, 2842 – 2845
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2843