5
160
J. Ju et al. / Tetrahedron Letters 54 (2013) 5159–5161
Table 1
Table 2
Substrate scope
a
a
Optimization of reaction conditions
Entry
Ketimine 1
Product 2
Cl
N
cat. Pd/base
toluene, 8h
Cl
+
CO
N
N
N
N
1
1b
1c
2b, 72%
2c, 72%
2d, 56%
Cl
Cl
Cl
1a
2a
O
O
Entry
Catalyst
Base
CO(atm)
Yieldb (%)
N
2
3
1
2
3
4
5
6
7
8
9
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
PdCl
2
2
2
2
2
2
2
2
2
2
(PPh
(PCy
(dppe)(3)
(PCy
(PCy
(PCy
(PCy
(PCy
(PCy
(PCy
3
)
)
2
(3)
(3)
CS
CS
CS
2
CO
2
CO
2
CO
3
3
3
10
10
10
10
10
10
10
5
Trace
81
Trace
72
54
<10
50
Cl
F
Cl
3
2
O
)
3 2
)
3 2
)
3 2
)
3 2
)
3 2
)
3 2
)
3 2
(3)
(3)
(3)
(3)
(3)
(3)
(3)
(1)
(0.5)
2 3
K CO
Na
Bu
3
PO
N
4
N
1d
N
3
F
Cl
Pyrrolidine
O
CS
2
CS
2
CS
2
CS
2
CS
2
CO
3
CO
3
CO
3
CO
3
CO
3
89
1
10
5
Trace
28
1
1
1
0c
1
N
N
4
2e, 31%
PdCl
PdCl
2
(PCy
(PCy
3 2
)
95(84)d
41
1e
Cl
2
2
)
3 2
10
Cl
Cl
O
a
Unless otherwise noted, reactions were carried out using 1.0 mmol 1a, catalyst,
and 1.1 equiv of base in 1.0 mL of toluene in a 25 mL autoclave with CO at 130 °C for
N
5
6
7
8
N
N
2f, 81%
2g, 75%
2h, 83%
2i, 61%
8
h.
1
f
b
Yields of 2a are based on GC by using n-C22
Reactions were carried out at 115 °C.
Isolated yields.
H46 as internal standard.
Cl
Cl
O
c
d
Ph
N
1g
Ph
N-methoxybenzamides and active alkenes. Also, the use of stoichi-
ometric benzoquinone as oxidant and acetic acid as solvent makes
the reaction still not practical in the view of green chemistry. In
O
Ph
N
2
N Ph
continuation of our interest in the C–Cl bond activation and cyclo-
1h
1
1
Cl
carbonylation, we became interested in the development of an
alternative synthetic strategy toward 3-methyleneisoindolin-1-
O
OMe
Cl
1
2
one derivatives via palladium-catalyzed cyclocarbonylation of
ortho-chloro ketimines.
N
OMe
Cl
N
N
1
i
As a model reaction, ortho-chloro ketimine (1a) was employed
as the substrate, while a variety of catalysts and bases were
employed to optimize the reaction conditions (Table 1). Gas chro-
matography (GC) was used to give the yield of each entry by using
O
Cl
Cl
N
9
2j, 50%
1
j
n-C22
To start our study, three Pd(II) catalysts with different ligands
were used with Cs CO as base to explore an efficient catalyst sys-
tem. A common catalyst PdCl (PPh showed rather low activity
entry 1), while a remarkable yield (81%) of the desired product
2a) was obtained by simply replacing the PPh ligand with PCy
entry 2). Catalyst with a bidentate ligand 1,2-bis(diphenylphos-
H46 as internal standard.
O
a
Reactions were carried out using 1.0 mmol of 1, 1.0 mol % of PdCl
.1 equiv of Cs CO in 1.0 mL of toluene in a 25 mL autoclave with 5.0 atm of CO at
2
(PCy
3 2
) , and
2
3
1
1
2 3
2
3 2
)
30 °C for 8 h.
(
(
(
3
3
With the optimized reaction condition in hand, the generality
for the synthesis of isoindolin-1-ones was next examined
Table 2). Dichloro ketimine 1b and 1c were used to afford
phino)ethane (dppe) also showed low activity (entry 3). These re-
sults revealed that the Pd(II) catalyst with trialkylphosphine ligand
1
3
(
PCy
comparable with our previous studies.
By using PdCl (PCy as catalyst, we next surveyed several
bases (entries 4–7). Another carbonate K CO was also suitable
for the reaction besides Cs CO , but gave a lower yield (72%).
Na PO displayed much lower efficiency (54%), and organic bases
such as Bu N and pyrrolidine showed low activity. By using the
PdCl (PCy /Cs CO catalytic system, we further conducted our
3
was an efficient catalyst for the C–Cl bond activation, which is
chloro-substituted products 2b and 2c in 72% yield. Notably, chloro
groups not in the ortho position were inert in the system. This
selectivity in the C–Cl bond cleavage indicates that the C–Cl bond
activation is facilitated by the coordination to Pd using the N atom
in the ketimine (vide infra). Fluoro-substituted ketimine 1d gave
the fluoro-having products 2d in 56%, indicating that the C–F bond
remained intact under the reaction conditions. By using 1e as the
substrate, two isomers are expected to be formed, but only one
of them was obtained in 31% isolated yield, and unfortunately
the stereochemistry of the C@C double bond could not be con-
firmed by 2D NOESY experiment. Ketimines with bulkier N-substi-
tutes or N-aryl substitutes were also employed and the reaction
proceeded smoothly to afford the corresponding 3-methyleneiso-
indolin-1-ones 2f–j in moderate to good yields.
2
3 2
)
2
3
2
3
3
4
3
2
3
)
2
2
3
optimization by varying other reaction conditions. To our delight,
a higher yield (89%) was obtained by reducing the CO pressure to
5
atm (entry 8). However, only trace amount of product was
detected when the reaction was conducted under 1 atm CO (entry
). Decreasing the temperature from 130 °C to 115 °C made the
9
reaction sluggish and gave much lower yield (28%, entry 10).
Interestingly, a much satisfactory yield (95%) was obtained by
reducing the catalyst loading from 3 mol % to 1.0 mol % (entry
A proposed mechanism was shown in Scheme 1, ortho-chloro
ketimine makes an oxidative addition to the Pd catalyst to afford
a five-membered palladacyclic intermediate. Subsequently, CO
insertion occurs and a six-membered palladacyclic intermediate
1
1), may be due to the suppression of the side reactions. However,
further reducing the catalyst loading to 0.5 mol % gave disappoint-
ing yield (41%, entry 12).