1502
J. Ma et al. / Tetrahedron Letters 56 (2015) 1501–1504
Table 1
Table 2
Optimization of the reaction conditionsa
Substrate scope of allylation of 3-substituted oxindolesa
R1
R1
N
Bn
Bn
NH
conditions
BF3 Et2O (0.4 equiv)
O
+
O
+
2
O
O
R
R
Cl3C
O
N
N
N
R2
R2
1a-l
3a-l
1a
2
3a
Entry Substrate
Bn
1
Product
Bn
3
Yieldb (%)
Entry
2 (equiv)
Acid (equiv)
Solvent
Yieldb (%)
1
2
3
4
5
6
7
8
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
1.2
2.0
3.0
2.0
CF3CO2H (0.4)
CF3SO3H (0.4)
BF3ꢀEt2O (0.4)
AlCl3(0.4)
DCM
DCM
DCM
DCM
DCM
THF
3
7
1
2
3
4
5
6
1a
3a 81c
3b 78
3c 75
3d 74
3e 76
3f 93
O
O
30
—
Trace
—
N
N
Bn
Bn
Cu(CF3SO2)2 (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.8)
BF3ꢀEt2O (0.2)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
BF3ꢀEt2O (0.4)
F
F
1b
1c
1d
1e
1f
O
O
Toluene
ACN
4
N
N
Trace
31
58
61
57
44
63
75
80
9c
10
11
12
13
14d
15d
16d
DCM
DCM
DCM
DCM
DCM
DCM
DCM
DCM
Bn
Bn
Br
Br
O
O
N
N
2.0
Bn
Bn
1.0 * 2
1.0 * 3
0.5 * 6
O
O
N
N
a
Reactions were carried out with 1a (0.20 mmol), catalyst and 2 in solvent
(1.0 mL) at room temperature for 36 h.
Bn
Bn
O
O
b
Isolated yield.
Reaction time is 115 h.
Compound 2 was added portionwise over a 5 h interval.
c
O
O
N
N
d
Ph
Ph
O
O
examined (Table 1). As expected, the desired 3-allyl-3-benzyl-1-
methylindolin-2-one (3a) was obtained. Treatment of 1a with 2
in the presence of 0.4 equiv of trifluoromethanesulfonic acid in
CH2Cl2 at room temperature gave 3a in 7% yield (Table 1, entry
2). A variety of Lewis acids were evaluated in the model reaction
(Table 1, entries 1–5), and BF3ꢀOEt2 was found to give the desired
3a in 30% yield (Table 1, entry 3). Other Brønsted acids (H3PO4,
H2SO4) and Lewis acids (SnCl4, FeCl3, ZnCl2) all resulted in only a
trace amount of product. Other solvents such as toluene, THF,
and acetonitrile diminished the reaction clearly (Table 1, entries
6–8). Notably, the starting material 1a was always recovered dur-
ing the process. So the reaction time was prolonged, however, sim-
ilar results were obtained (Table 1, entry 9). To our delight, the
yields were clearly increased when 2.0 equiv of 2 was employed
(Table 1, entry 10). The reason may be ascribed to the correspond-
ing allylic carbocation intermediate being unstable. Indeed, using a
large excess of 2 resulted in a higher yield (Table 1, entries 10, 11
vs entry 3). In contrast, no improvement was achieved by changing
the catalyst amount (Table 1, entry 12 vs entry 10). Further optimi-
zation of the reaction conditions revealed that addition of 3.0 equiv
of 2 by 6 times dramatically enhanced the yield to 80% (Table 1,
entry 16).
N
N
Ph
Ph
O
O
7
1g
3g 89
N
N
Bn
Bn
Ph
Ph
F
F
8
1h
1i
3h 90
3i 87
3j 85
3k 86
O
O
N
N
Ph
Ph
Br
Br
9
O
O
N
N
Ph
Ph
10
11
1j
O
O
N
N
Ph
Ph
O
O
1k
O
O
N
N
Bn
Bn
Having identified the optimized conditions, we next examined
this new method in a range of substrates with different substitu-
tion patterns. The results are summarized in Table 2. Overall, a
variety of 3-allyl-30-substituted oxindole derivatives were success-
fully prepared, and various substituents on the aromatic ring were
found to be tolerable in this process. Substrates with electron-defi-
cient or electron-rich substitutions at the aromatic ring offered
good yields (Table 2, entries 2–5), indicating that the electron
effect plays little role in this reaction. Reaction of 1,3-dimethylin-
dolin-2-one with 2 under the optimal conditions gave the desired
product in only 24% yield. It seemed that the reaction started from
an acid catalyzed enolization of oxindole at 3-position, thus a 3-
aryl substituted group will be beneficial to such a ketone-enol
O
O
12
1l
3l 73
N
Bn
N
Bn
a
Reactions were carried out with 1a–l (0.20 mmol), BF3ꢀEt2O (0.08 mmol,
0.4 equiv) and 2 (0.60 mmol, 3.0 equiv) in DCM (1.0 mL) at room temperature for
36 h.
b
Isolated yield.
2.0 mmol scale.
c
Generally, the desired 3-allylation oxindole products were
obtained with higher yields than those of 3-benzyl substituted
oxindole substrates (Table 2, entries 6–11 vs entries 1–5). Similar
a previous observation, electron-deficient or electron-rich
substituted 3-phenyl oxindole substrates all gave the desired ally-
to
equilibrium through
p–p conjugation. Then 3-phenyl substituted
oxindole substrates were prepared and subjected to this reaction.