2
C.-T. Feng et al. / Tetrahedron Letters xxx (2016) xxx–xxx
N
O
Ph
N
O
N
OEt
H
O
4a
1
N
NO2
N
Ph
OEt
7H2O. CeCl3 (10 mol %)
7H2O. CeCl3 (10 mol %)
N
2a
Ph
H
3
5a
our previous work
cascade cyclization
Ph
N
N
H
6
Scheme 2. Cascade cyclization and acylation.
The applications of cerium compounds to organic transforma-
group to 7-position of the indole ring marginally reduced the reac-
tion yield (Table 2, 5t).
tions have been extensively studied.13 They have attracted much
attention in organic synthesis because they are inexpensive, easy
to handle, and highly tolerant to air and moisture. Therefore, the
development of cerium-catalyzed methods would be of a great
value. Herein, we report a cerium-catalyzed C3-selective acylation
of N–H indoles using nitroolefins as acylating reagents. To our
knowledge, this is the first example of using the nitroolefins as
acylating reagents in acylation reactions.
To gain insight into the possible mechanism of the reaction, a
few control experiments were carried out. In the presence of the
radical scavenger 2,2,6,6-tetramethyl-1-piperidinyloxy (TEMPO),
a negative influence on the yield was observed (Scheme 3, Eq. 1).
This implied that some radical intermediates should be involved
in the reaction. Next, instead of nitroolefin, benzaldehyde was used
to react with 2-(pyridin-2-yl)-1H-indole under the optimal reac-
tion conditions. Almost no desired product was obtained in this
case (Scheme 3, Eq. 2). When the reaction of 2-(pyridin-2-yl)-1H-
indole (4a) and trans-(Z)nitrostyrene (2a) was performed in the
cerium catalytic system under argon gas, the desired product
(5a) was obtained in 93% yield (Scheme 3, Eq. 3). This result indi-
cated that the nitro group might act as an internal oxidant. When
2-phenyl-1H-indole (Scheme 3, Eq. 4) or 2-(pyridin-4-yl)-1H-
indole (Scheme 3, Eq. 5) was used to react with nitroolefin (2a),
michael adducts were separated as major product. The above
results suggest that the existence of pyridin-2-yl group has an
unclear intramolecular effect on the product formation process.
To look into the prospect of this new acylation reaction, a model
reaction using easily available 2-(pyridin-2-yl)-1H-indole (4a) and
trans-(Z)nitrostyrene (2a) was investigated in detail by varying the
catalyst in order to develop appropriate conditions (Table 1). We
screened various Lewis acids and found that other Lewis acids,
including copper salt, iron salt, nickel salt, cobalt salt, are either
ineffective or less effective than CeCl3Á7H2O (Table 1, entries 1–
6). The reaction gave a poor yield in the absence of catalyst (Table 1,
entry 7). On the other hand, the organic solvent was also found to
play an important role in the reaction. Among various solvents
screened, ethanol was found to be the most suitable solvent. Other
solvents reduced the yield of this reaction (Table 1, entries 8–13).
The yield of 5a decreased to 72% when lowering the temperature
from 120 °C to 100 °C (Table 1, entry 14). In recent years, it has
been found that the CeCl3Á7H2O–NaI system as an efficient Lewis
acid activator has a wide range of interesting applications in
organic chemistry.13b However, the introduction of 0.1 equiv of
NaI into the reaction system significantly suppressed the reaction
(Table 1, entry 15).
Table 1
Optimization of the reaction conditionsa
O2N
10 mol% cat.
+
O
N
H
With the optimal conditions in hand, the substrate scope and
limitation of the reaction was explored. First of all, the substrate
scope of nitroolefins 2 was examined. As illustrated in Table 2, aro-
matic nitroalkenes with both electron-rich (4-Me, 4-OMe, 4-NMe2)
and electron-deficient (CF3) substituents on the aromatic ring par-
ticipated in this reaction smoothly to afford the expected products
in good to excellent yields. Generally, an electron-donating sub-
stituent on the aromatic ring has a negative effect on the yield.
Unprotected phenol functions were well tolerated under the opti-
mized conditions, when R1 was replaced by a 2-OH group, the reac-
tion gave a lower yield in comparison with that R1 was a 4-OH
group (Table 2, entries 5f and 5n). This implied that steric effect
had an influence on the reaction. The reaction of ring-fused (5o)
nitroalkene also afforded the corresponding product in good yields.
However, less-reactive alkyl substituted nitroalkenes did not give
the target products under the current reaction conditions.
Next, the effect of the substituents attached on the indole ring
was examined. The halogen, including fluoro, chloro, or bromo
group, had little influence on the reaction (Table 2, 5q–5s), thus pro-
viding the possibility to allow additional coupling reactions. The
electron-donating group at 5-position of the indole ring was also
well tolerated, affording the corresponding product in excellent
yield (Table 2, 5p). But the introduction of an electron-donating
N
N
N
H
4a
2a
5a
Entry
Catalyst
Solvent
Yieldb (%)
1
2
3
4
5
6
7
8
CuCl2Á2H2O
FeCl3Á6H2O
NiCl2Á6H2O
CoCl2Á6H2O
CeCl3Á7H2O
TsOHÁH2O
—
CeCl3Á7H2O
CeCl3Á7H2O
CeCl3Á7H2O
CeCl3Á7H2O
CeCl3Á7H2O
CeCl3Á7H2O
CeCl3Á7H2O
CeCl3Á7H2O
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
H2O
DMF
MeCN
THF
CH2Cl2
MeOH
EtOH
EtOH
32
25
38
Traces
90
29
36
Traces
Traces
21
Traces
Traces
85
9
10
11
12
13
14c
15d
72
43
a
Reaction conditions: 4a (0.20 mmol), 2a (0.30 mmol), catalyst (0.02 mmol),
solvent (2 ml), 120 °C, 12 h.
b
Isolated yield.
The reaction was at 100 °C.
0.02 mmol of NaI was used.
c
d