Communications
J . Org. Chem., Vol. 63, No. 2, 1998 223
ciently promoted by N-hydroxyphthalimide (NHPI) com-
bined with a small amount of transition metal salts to
produce the corresponding oxygenated products. We have
now developed a new type of catalytic free-radical carboxyl-
ation of polycyclic alkanes such as adamantanes with CO/
air using N-hydroxyphthalimide (NHPI), which serves as an
efficient radical catalyst. In this paper, we report the
catalytic free-radical carboxylation of alkanes by NHPI
without any photoactivation under mild conditions, i.e.,
lower CO pressure (up to 15 atm) and temperature (below
present carboxylation of 1 was efficiently catalyzed by NHPI
alone under mild conditions.
The carboxylation of 2 in the presence of NHPI (30 mol
%) gave dicarboxylic acid 3 in 57% yield at 74% conversion
(eq 2, Supporting Information). The formation of 3 is
interesting from a synthetic point of view, because dicar-
boxylic acid 3 is difficult to obtain from the conventional
carboxylation of 1 with CO.
Similarly, 1,3-dimethyladamantane was also carboxylated
under these conditions to give the corresponding adamantyl
monocarboxylic acid (55%) and dicarboxylic acid (7%) in 77%
conversion (eq 3, Supporting Information). The carboxyla-
tion of endo-tricyclo[5.2.1.02 ]decane (9) by NHPI (30 mol
%) combined with a slight amount of Co(acac)2 (0.5 mol %)
gave monocarboxylic acid 10 as the major product (55%) (eq
4, Supporting Information). The same reaction of 9 in the
absence of Co(acac)2 resulted in the lowering of the yield of
10.
1
00 °C).
A typical reaction was carried out as follows. To a solution
,6
of NHPI (0.1 mmol) in a mixed solvent of acetic acid (3 mL)
and 1,2-dichloroethane (3 mL) in a glass-lining autoclave
was added adamantane (1) (1 mmol), and then 1 atm of air
and 15 atm of CO was charged. The reaction was carried
out at 95 °C for 4 h. Table 1 shows the representative
results for the carboxylation of 1 under various reaction
conditions.
A plausible mechanism of the present NHPI-catalyzed
radical carboxylation of 1 with CO/air system is outlined in
Scheme 1. The first step is the generation of phthalimide-
N-oxyl (PINO) from NHPI with O2.1 The PINO abstracts
a tertiary hydrogen from 1 to form an adamantyl radical
4d
(
A), which is the key intermediate in this reaction. It is
1
e
probable that the radical A is readily trapped with CO,
giving adamantaneformyl radical (B), which on subsequent
reaction with O2, produces adamantanecarboperoxy radical
(C) and eventually carboxylic acid 2.
To clarify the participation of the acyladamantyl cation
The carboxylation of 1 with CO/air in the presence of
NHPI gave 1-adamantanecarboxylic acid (2), 1,3-adaman-
tanedicarboxylic acid (3), 2-adamantanecarboxylic acid (4),
and several oxygenated products such as 1-adamantanol (5)
and 2-adamantanone (6) as well as 1-acetoxyadamantane
in the formation of carboxylic acid 2, the carboxylation of 1
in the presence of H218O was examined under the same
conditions as run 1 in Table 1. The mass spectrum of the
1
8
resulting 2 showed that the ratio of 2/ O labeled-2 is 94/6.
This suggests that most of the carboxyl oxygen contained
(7). Although the selectivity of these products depended
1
8
in 2 comes from CO and air, but not from H2 O. If the
NHPI-catalyzed carboxylation of 1 passes through the
acyladamantyl cation as a transient intermediate, a consid-
erable amount of 2 in which 18O was incorporated in the
molecule should be formed.
markedly on the reaction conditions, 1-adamantanecarb-
aldehyde (8) was not formed in all reactions examined.
When the reaction was carried out using molecular oxygen
in place of air, oxidative products were formed as major
products (run 2). Needless to say, the reaction did not take
place without air (run 3). Among the solvents examined, a
mixed solvent consisting of acetic acid and 1,2-dichloro-
ethane (1/5 v/v %) gave the best yield of 2 (56%) at 75%
conversion (run 4). The present reaction proceeded even at
In addition, the reaction of 1-bromoadamantane (12) with
CO/air by the Ryu system using AIBN/Bu3SnH,13 which can
be excluded the formation of the acyladamantyl cation as
the intermediate, gave also 2 in 54% yield (eq 5). These
6
0 °C, although a prolonged reaction time was necessary to
obtain satisfactory yield of 2 (run 5). The carboxylation of
using N-hydroxysuccinimide, an analogue of NHPI, under
these conditions gave 2 in slightly lower selectivity (40%)
1
(run 6).
The reaction did not take place in the presence of
observations suggest that the NHPI-catalyzed carboxylation
of 1 under the influence of CO/air proceeds through the
acyladamantyl radical rather than the acyladamantyl cation.
Although a part of the C seems to decompose to adamantyl
radical A, carbon dioxide, and dioxygen, the resulting A can
also react with CO to regenerate the B.
The fact that no aldehyde 8 was formed in the present
reaction may indicate that the 8 formed is rapidly converted
into carboxylic acid 2 under these conditions. In fact, the
independent reaction of 8 under the influence of NHPI and
CO/air gave 2 in 57% yield along with adamantane 1 (12%),
hydroquinone (0.1 mol %) (run 10). This shows that the
NHPI-catalyzed carboxylation involves the radical process
in the reaction step. In the present carboxylation of 1, the
relative reactivity of the tertiary hydrogen to the secondary
hydrogen estimated from run 1 was ca. 12.15 This value was
almost the same as that by the NHPI-catalyzed aerobic
oxidation of 1,1 which may reflect a reaction path in the
present carboxylation analogous to that of the aerobic
oxidation reported previously.
4c
In contrast to the NHPI-catalyzed aerobic oxidation of
alkanes where the reaction is markedly enhanced by the
addition of a transition metal salt such as Co(acac)2, the
1
6
-adamantanol 5 (5%), and 1-acetoxyadamantane 7 (1%) (eq
). The formation of 1, 5 and 7 from 8 suggests that the
reaction between 1 and 8 is reversible path.
(14) In a previous paper, we demonstrated that PINO is easily generated
by exposing NHPI to molecular oxygen. (a) Ishii, Y.; Nakayama, K.; Takeno,
M.; Sakaguchi, S.; Iwahama, T.; Nishiyama, Y. J . Org. Chem. 1995, 60,
Ack n ow led gm en t. This work was partly supported by
Grant-in-Aid for Scientific Research (No.09238255) on
Priority Areas (No. 283, “Innovative Synthetic Reactions”
from Monbusho.
3
934. (b) Iwahama, T.; Sakaguchi, S.; Nishiyama, Y.; Ishii, Y. Tetrahedron
Lett. 1995, 36, 6923. (c) Ishii, Y.; Kato, S.; Iwahama, T.; Sakaguchi, S.
Tetrahedron Lett. 1996, 37, 4993. (d) Ishii, Y.; Iwahama, T.; Sakaguchi, S.;
Nakayama, K.; Nishiyama, Y. J . Org. Chem. 1996, 61, 4520. (e) Ishii, Y. J .
Mol. Catal. A: Chem. 1997, 123. (f) Yoshino, Y.; Hayashi, Y.; Iwahama, T.;
Sakaguchi, S.; Ishii, Y. J . Org. Chem. 1997, 62, 6810.
Su p p or tin g In for m a tion Ava ila ble: Equations 2-4 and 6
characterization data for all compounds (16 pages).
3
°
2°
(
15) The ratio of C -H/C -H was estimated from total yield of 2, 3, 5,
and 7/total yield of 4 and 6.
J O9718556