5
534
M. Wang et al. / Tetrahedron Letters 56 (2015) 5533–5536
CO
2
H
previous work
O
OH
O
P
O
N
O
OH
N
F
O
N
O
O
HO
F
N
H
Bn
N
N
∗
O
P
O
N
NBn
CH
3
N
(1)
(2)
∗
N
HO
H
BnN
C-N formation
O
O
3
1
2
AS701666
L-902688
present work
R2
CH
3
O
R2
O
O
XnM
TsO
N
N
O
R1
OCH
O
TsOH
MXn
2
R
1
NBn
BnN
O
N
H
3
CO
H CO
C-C formation
O
OCH3
Bn
1
A
3
3
H CO
OH
O
3
3
4
Scheme 2. Proposed protocol for the synthesis of functionalized pyrrolidin-2-one
Isonuevamine
Aknadilactam
derivatives.
Figure 1. Pharmaceutical and bioactive 5-alkenyl-2-pyrrolidinone derivatives.
Table 1
a
indoles with N-benzyl-
(
a
,b-unsaturated-
Scheme 2, Eq. 1). In connection with our previous work, we
envisioned that a cooperative catalysis consisting of Lewis acid
and Brønsted acid may promote the reaction of ,b-unsaturated-
-lactams and olefins. As shown in Scheme 2, N-acyliminium ion
c
-lactam as electrophiles
Optimization of reaction conditions
1
1
8
O
Ph
Ph
Lewis acid, BrØnsted acid
NBn
+
Ph
a
o
O
Ph
Solvent, T C, 6 h
N
Bn
c
1
2a
3a
intermediates would be generated in the presence of Brønsted acid.
Then, a Prins-type reaction of A and olefin 2 might occur and
enable the generation of 5-alkenyl-2-pyrrolidinones in the cat-
alytic system of Lewis acids and Brønsted acids.
Entry
Lewis acid
Fe(OTs)
Y(OTf)
Brønsted acid
Solvent
T (°C)
Yieldb (%)
1
2
3
4
5
6
7
8
9
3
TsOH
TsOH
TsOH
TsOH
TsOH
PivOH
TTCEc
TTCE
TTCE
TTCE
TTCE
TTCE
TTCE
TTCE
50
50
50
50
50
50
50
50
50
50
50
50
50
60
40
50
50
50
17
59
46
57
42
13
55
43
20
Trace
18
28
Trace
59
3
ÁxH
2
O
O
Eu(OTf)
Zn(OTf)
3
ÁxH
2
Results and discussion
2
Yb(OTf)
3
ÁxH
2
O
O
O
O
c
Y(OTf)
Y(OTf)
Y(OTf)
3
ÁxH
3
ÁxH
3
ÁxH
2
2
2
To validate our hypothesis, we began our study on synthesis of
c
HOTf
5
-alkenyl-2-pyrrolidinones by examining the reaction between
N-benzyl- ,b-unsaturated- -lactam 1 and 1,1-diphenylethylene
a in the presence of a catalytic amount of Lewis acid and Brønsted
acid. To our delight, the employment of Fe(OTs) as the Lewis acid
NH
2
SO
3
H
a
c
Y(OTf) ÁxH O
TsOH
TsOH
TsOH
TsOH
TsOH
TsOH
TsOH
TsOH
None
None
CH Cl2
3
3
3
3
3
3
3
2
2
2
2
2
2
2
2
2
10
Y(OTf)
Y(OTf)
Y(OTf)
Y(OTf)
Y(OTf)
Y(OTf)
None
ÁxH
ÁxH
ÁxH
ÁxH
ÁxH
ÁxH
O
O
O
O
O
O
CCl
4
c
11
12
13
14
DCE
CH NO
3
3
2
and TsOH as the Brønsted acid resulted in the formation of the
desired product 3a in 17% yield at 50 °C in TTCE. Encouraged by
this result, we investigated the reaction conditions in detail, and
the results are summarized in Table 1. The effect of Lewis acids
was first evaluated. The reaction efficiency was dramatically
improved and product 3a was obtained in 59% yield (Table1, entry
Toluene
TTCE
TTCE
TTCE
TTCE
15
42
1
1
1
6
7
8
Trace
Trace
NR
Y(OTf)
None
3
ÁxH
2
O
TTCE
a
Reaction conditions:
Brønsted acid (15 mol %) in solvent (1 mL) under N
1 (0.2 mmol), 2a (0.4 mmol), Lewis acid (5 mol %),
2
) when a catalytic amount of Y(OTf)
catalytic system. An extensive screening of Lewis acids demon-
strated that other Lewis acids such as Eu(OTf) O, Zn(OTf)
ÁxH
and Yb(OTf) O could also catalyze this transformation, but
ÁxH
none of these catalysts could improve the yield (Table 1, entries
–5). Next, Brønsted acids were examined and the results revealed
3
ÁxH
2
O was introduced to the
2
at T °C for 6 h.
b
Isolated yield.
3
2
2
,
c
PivOH = 2,2-Dimethylpropanoate, HOTf = Trifluoromethanesulfonic acid, TTCE =
1,1,2,2-Tetrachloroethane, DCE = 1,2-Dichloroethane.
3
2
3
proceed more efficiently in polar solvents than in non-polar ones,
and TTCE was found to be the best choice. In addition, with either
an increase or decrease in the temperature, inferior results were
observed. Furthermore, the control reactions demonstrated that
the product 3a was rarely formed in the absence of either Lewis
acid or Brønsted acid. These results declared that Lewis acid and
Brønsted acid were essential for promoting the reactivity of the
transformation.
With the optimized reaction conditions in hand, we subse-
quently evaluated the scope of olefins. Initially, a series of 1,1-disub-
stitutedethylenes were subjected to the present reaction, and the
corresponding products 3a–3e were obtained in moderate to good
yields (Table 2, entries 1–5). The reactions were significantly
affected by electronic properties of the alkenes. In general, the alke-
nes with electron-donating group exhibited higher reactivities to
give the corresponding products. For example, 1-methoxy-4-(1-
phenylvinyl)benzene 2c, bearing electron-donating group in the
phenyl ring, gave the desired products 3c in 84% yield (Table 2, entry
that TsOH was the best choice, while other Brønsted acids gave
lower yields (Table 1, entries 6–11). To achieve better results, the
effect of solvent was also studied. The reaction was found to
O
R2
R1
180 C
o
OR
R
1
Path A
Path B
O
R2
NHR3
N
R
3
R2
R2
X
R1
1
) PBr
3
R1
O
N
R
2) t-BuOK
O
N
R
3
3
O
R2
R1
+
1) NaBH4
2) HCl
NR3
MgCl
R1
O
Path C
R2
N
R
O
3
O
R2
R2
NR3
+
R
1
This work
O
R1
N
R
3
). 1-Methoxy-4-(prop-1-en-2-yl)benzene 2e showed highest
3
activity to give the corresponding product 3e in 85% yield (Table 2,
Scheme 1. Synthetic methods of 5-alkenyl-2-pyrrolidinones.
entry 5). For the 1,1-disubstituted alkenes substituted with