Organic Letters
Letter
Scheme 1. Catalytic Approach in Acceptorless
Dehydrogenation of Alcohols
2b−d in very good yield. p-Methoxybenzyl alcohol and 3,5-
dimethoxybenzyl alcohol provided the products 2e and 2f in
95 and 86% yield, respectively. Notably, p-methylthiobenzyl
alcohol was tolerated by cobalt catalysis, which resulted in p-
methylthiobenzoic acid 2g in 78% yield. The electron-
withdrawing group on arylmethanols diminished their
reactivity toward the oxidation reaction, which resulted in
comparatively low yields of the corresponding carboxylic acids.
Thus m-chloro, p-bromo, and p-fluorobenzyl alcohols were
subjected to the cobalt-catalyzed oxidation reaction and the
products 2h−j, obtained in 61−82% yield. Aryl halides are
known to undergo a reduction reaction to the corresponding
arenes when subjected to transition-metal catalysis. However,
no such arene formation was observed under this catalytic
condition (entries 8−10, Table 2). Similarly, the reaction of p-
trifluoromethyl- and p-nitrobenzyl alcohol provided the
carboxylic acids 2k and 2l in 71 and 55% yields, respectively.
Despite the low yield observed for p-nitrobenzoic acid (2l), it
is gratifying to note that the nitro functionality is tolerated;
nitro compounds are incompatible in iridium-catalyzed
oxidation reactions and are under-investigated in this trans-
formation. Biaryl methanols are also tolerated. When 1-
naphthalenemethanol was subjected to the reaction, 1-
naphthalene carboxylic acid (2m) was isolated in 67% yield.
Heteroarene functionalities are susceptible to a hydrogenation
reaction under the acceptorless oxidation, where the catalyst
can utilize the hydrogen liberated from alcohol oxidation.
Remarkably, heteroarylmethanol compounds are well tolerated
in the cobalt-catalyzed oxidation reaction, although the
observed reactivity is low in comparison with that of other
arylmethanols. When 2-pyridinemethanol and 4-pyridineme-
thanol were subjected to oxidation, the corresponding
carboxylate salts 2n and 2o were isolated in 59 and 67%
yield, respectively. Furfuryl alcohol and p-aminobenzyl alcohol
provided the corresponding carboxylic acids 2p and 2q in poor
yield. The cobalt catalysis is very well compatible with diols.
Phthalic acid (2r) and terephthalic acid (2s) were obtained
from the corresponding diols in good yield. However, 2,6-
pyridinedimethanol provided the dicarboxylate salt 2t in only
mmol) and KOH (1.2 mmol) with catalyst 1 at 140 °C for 24
h provided the benzoic acid in 49% yield upon acidic work-up
(
entry 1, Table 1). Furthermore, an increase in the base load
Table 1. Optimization for Catalytic Dehydrogenation of
a
Alcohol Catalyzed by 1
b
entry
cat (mol %)
base (equiv)
time (h)
yield (%)
1
2
3
4
5
6
1 (1)
1 (1)
1 (2)
1 (2)
1.2
1.5
1.2
1.5
1.5
1.5
24
24
16
16
24
24
49
81
73
93
3
c
CoBr (2)
7
2
a
Reaction conditions: Alcohol (1 mmol), catalyst, base, and toluene
4
9% yield.
b
(
2 mL) were taken in a sealed tube and heated to 140 °C. Isolated
c
Furthermore, the scope of the reaction was explored with a
yield of product after acidic work-up with 1 M HCl. Control reaction
variety of aliphatic primary alcohols in the catalytic synthesis of
carboxylic acid. A variety of aliphatic primary alcohols provided
moderate to good yields (Table 3). When ethanol and 1-
propanol were subjected to catalysis, potassium acetate and
potassium propanoate salts 3a and 3b were obtained in 67 and
performed without catalyst.
from 1.2 to 1.5 equiv enhanced the yield of benzoic acid to
8
1% (entry 2, Table 1). Performing a similar catalytic reaction
using 2 mol % of catalyst 1 and 1.2 equiv of KOH resulted in a
diminished yield (73%, entry 3, Table 1). Thus an experiment
using 2 mol % of catalyst 1 and 1.5 equiv of KOH was
performed, which provided benzoic acid in 93% yield (entry 4,
Table 1). Without a catalyst, the reaction of alcohols with a
base alone was performed, which failed to provide carboxylic
acid in an appreciable amount (entry 5, Table 1), and this
control experiment confirmed the necessity for a catalyst in this
73% yield, respectively. Long-chain linear alcohols such as 1-
hexanol and 1-heptanol underwent facile oxidation to provide
the carboxylic acids 3c,d in good yield. Similarly, cyclo-
butylmethanol, cyclohexylmethanol, and bicyclo[2.2.1]-
heptylmethanol were oxidized to the corresponding carboxylic
acids 3e−g in good yield. Phenethyl alcohol and 3-phenyl-
propanol were transformed to phenylacetic acid (3h) and 3-
phenylpropanoic acid (3i) in 75 and 81% yield, respectively.
Iridium catalysts were not compatible with alkene com-
transformation. Furthermore, the use of simple CoBr (2 mol
2
%
) as a catalyst provided benzoic acid in 7% yield, reiterating
10
the importance of the designed cobalt pincer catalyst 1.
With the optimized reaction condition in hand, an
assortment of alcohols were subjected to the oxidation to
develop the scope of the reaction as well as to identify the
limitation of this catalytic protocol (Table 2). Arylmethanols
with electron-donating groups such as p-methyl, p-isopropyl,
and p-tert-butyl provided the corresponding carboxylic acids
pounds. Although ruthenium catalysts were tolerated, the
5
,6,8
olefin functionality underwent in situ hydrogenation.
On
the contrary, when cinnamyl alcohols were reacted with cobalt
catalyst 1 and a base, cinnamic acid (3j) was isolated in 78%
yield. Such selective oxidation of alcohol functionality by
catalyst 1, even in the presence of a competing olefin motif, is
remarkable. 1,5-Pentane diol and hexamethylene diol were also
B
Org. Lett. XXXX, XXX, XXX−XXX