7120
M. Chouhan et al. / Tetrahedron Letters 54 (2013) 7119–7123
Table 2
Effect of solvent and temperaturea
Entry
Solvent
Temp (°C)
Time (h)
Yieldb (%)
drc (trans:cis)
1
2
3
4
5
6
7
8
THF
25
25
25
25
25
20
10
0
36
36
36
36
16
18
20
30
45
40
20
78
95
95
94
50
90:10
90:10
90:10
90:10
90:10
95:05
98:02
98:02
Toluene
EtOH
MeCN
DCM
DCM
DCM
DCM
Scheme 2. Organocatalytic synthesis of 3-spiro-epoxyoxindole.
Epoxidation of (E)-ethyl 2-(1-methyl-2-oxoindolin-3-yli-
dene)acetate (2a) was attempted with natural and synthetic chiral
a
(E)-Ethyl 2-(1-methyl-2-oxoindolin-3-ylidene)acetate (1.0 equiv), UHP (2.0
equiv), Quinine (0.1 equiv).
organocatalysts such as
L
-proline (catalyst A), N-allyl-
L
-prolinol8a
b
(catalyst B), (S)-3-amino-4-methyl-N-phenylpentanamide8b (cata-
lyst C) and quinine (catalyst D) in the presence of different perox-
ides as oxidant (Scheme 2). From the results obtained (Table 1), it
could be inferred that catalysts A, B and C were inefficient and
afforded only a trace amount of the desired product. On the other
hand, reaction with 0.3 equiv of the catalyst D, quinine and
3.0 equiv of aq H2O2 in DCM as solvent at 25 °C afforded the de-
sired product (3a) with poor yield (20%) and a moderate diastereo-
meric ratio of 80:20 as determined from the 1H NMR spectra of the
reaction mixture (Table 1, entry 4). Examination of the catalytic
activity of quinine at lower loading demonstrated that only
0.1 equiv of quinine is sufficient for the reaction (Table 1, entry
9). Interestingly, in a separate study, while screening the different
peroxides as oxidant with varying molar ratios, we observed that
2.0 equiv of urea–hydrogen peroxide facilitates the complete con-
version of the starting material (Table 1, entries 7 and 8). In a con-
trol experiment performed with (E)-ethyl 2-(1-methyl-2-
oxoindolin-3-ylidene)acetate and UHP in DCM without the addi-
tion of quinine, the reaction did not afford the product while the
starting material remained as such even after 30 h (Table 1, entry
10). This clearly reveals the role of quinine not only in imparting
stereoselectivity but also in catalyzing the reaction. Screening var-
ious solvents for the reaction indicated DCM to be the most suit-
able (Table 2). To evaluate the effect of temperature on the
stereoselectivity of the reaction, a set of reactions were performed
at low temperatures of 20 °C, 10 °C and 0 °C (Table 2, entries 6–8).
The best result was obtained at 10 °C where the reaction afforded
the spiro-epoxyoxindole with 94% yield and 98:02 trans diastere-
oselectivity. Surprisingly at 0 °C the reaction was sluggish and
afforded only 50% of the product after 30 h. The trans selectivity
of the reaction was confirmed by NOESY experiment of the major
diastereomer (3b). No nuclear Overhauser effect was observed be-
tween the proton (He) attached to the oxirane ring and the aro-
matic proton (Ar–Ha) which clearly illustrates that the phenyl
ring and He proton are trans to each other (Fig. 2).
Isolated yield.
Diastereoselectivity ratios are based on 1H NMR spectra of the crude products.
c
From the above observations we inferred that reacting 1.0 equiv
of (E)-ethyl 2-(1-methyl-2-oxoindolin-3-ylidene)acetate with
2.0 equiv of UHP and 0.1 equiv of quinine at 10 °C in DCM would
be the ideal condition for the reaction. The scope of the optimized
condition was evaluated on various
a-ylideneoxindoles bearing
electron-withdrawing and donating groups which gave the ben-
zyl/ethyl 2-oxospiro(indolin-3,20-oxirane)-30-carboxylate deriva-
tives in excellent yields and high trans diastereoselectivity
(Table 3). It was also observed that reaction did not discriminate
between N-methyl and the bulky N-benzyl substituted a-ylidene-
oxindoles, affording good yields and diastereoselectivities in all
the cases. Similar results were obtained with ethyl and benzyl ester
derivatives of a-ylideneoxindoles as well.
We envisaged a plausible reaction mechanism for the formation
of trans selective spiro-epoxyoxindoles via transition states TS-A
and TS-B (Fig. 3). In TS-A, quinine activates the nucleophile UHP
and the electrophile
a-ylideneoxindole by hydrogen bonding,
thereby adopting a dual activation role. The bicyclic ring nitrogen
of quinine accepts hydrogen from UHP while the hydroxyl group
activates the carbonyl oxygen of the amide, thus making the b-car-
bon of the double bond more electrophilic in nature. This is fol-
lowed by the attack of the peroxide nucleophile at b-carbon, and
a subsequent rotation about the sigma bond between the
a and b
carbons leads to the stable conformation of TS-B. The enolate then
attacks the peroxy linkage resulting in the formation of trans epox-
ide. The hydroxide ion thus released will gain the proton from qua-
ternary nitrogen to form a molecule of water and regenerates the
quinine for next catalytic cycle.
Based on our previous work7c on water mediated nucleophilic
reactions of epoxides, we investigated the aminolysis of the resul-
tant trans epoxides. As expected the reaction led to ring opening
from the less hindered b-carbon of the epoxide in aqueous
Table 1
Effect of catalysts and oxidantsa
Entry
Catalyst (mol%)
Oxidant (equiv)
Time (h)
Yieldb (%)
drc (trans:cis)
1
2
3
4
5
6
7
8
9
A (30)
B (30)
C (30)
D (30)
D (30)
D (30)
D (30)
D (20)
D (10)
—
aq H2O2 (3.0)
aq H2O2 (3.0)
aq H2O2 (3.0)
aq H2O2 (3.0)
aq TBHP (3.0)
UHP (3.0)
UHP (2.0)
UHP (2.0)
UHP (2.0)
UHP (2.0)
36
36
36
36
36
12
12
14
16
30
Trace
Trace
Trace
20
18
96
95
95
95
Trace
nd
nd
nd
80:20
85:15
90:10
90:10
90:10
90:10
—
10
a
b
c
(E)-Ethyl 2-(1-methyl-2-oxoindolin-3-ylidene)acetate (1.0 equiv).
Isolated yield.
Diastereoselectivity ratios are based on 1H NMR spectra of the crude products.
Chouhan, Mangilal
