handled under inert atmosphere; others (e.g., boronic esters)
are costly and require multiple-step synthesis. A desirable
alternative is to directly couple CO2 with a C-H bond of
the organic substrates,1,4,20 with the best example being the
synthesis of salicyclic acid by direct carboxylation of
phenol.1,4 Here we report the direct carboxylation of aromatic
heterocycles following a similar strategy.
experiments showed that no reaction occurred in other
solvents (entries 7 and 8, Table 1). Without CO2, the reaction
did not proceed (entry 9, Table 1), suggesting that the COO
moiety in the product does not originate from Cs2CO3. The
pure product, 2-benzothiazolecarboxylic acid, can be isolated
as a white solid in 98% yield from a preparative reaction.
The compound decarboxylates slowly in solution (20%
decomposition in 5 h). The acid can be converted to a stable
ester by reacting with an alkyl halide.37
We chose benzothiazole as the initial substrate following
our work on direct alkylation of similar heterocycles.21
Furthermore, direct C-H functionalization of these hetero-
cycles with carbon (sp2 and sp) nucleophiles is now ubiq-
uitous.22-32 The C(2)-H is slightly acidic (pKa ) 27 in
DMSO)33 and might form carbon anion in the presence of a
base such as tert-butoxide, phosphate, or carbonate. We
hypothesized that coupling of this carbon anion with CO2 is
possible under appropriate reaction conditions and with an
active catalyst. As Cu is active in many C-H functional-
ization reactions of heterocycles22,23,25,29-31,34 and in car-
boxylation of boronic esters,11,15 we initially tried CuI as a
catalyst. Indeed, using LiOBut as the base, carboxylation of
benzothiazole proceeded well at 125 °C in DMF and gave
full conversion (entry 1, Table 1). 2-Benzothiazolecarboxylic
The optimized conditions can be applied for the
carboxylation of other heterocycles (Table 2). Because of
the limited stability of the heterocyclic carboxylic acids,
the products of carboxylation, the carboxylates, were
directly converted to stable esters in a one-pot procedure.37
Substituted benzothiazoles can be carboxylated, including
those containing sensitive nitrile and keto groups (entries
1-3, Table 2). Benzoxazoles are suitable substrates, and
aryl ester and Cl groups are tolerated (entries 4-7, Table
2). Naphtho[1,2-d]oxazole can be coupled (entry 8, Table
2), 5-phenyloxazoles are carboxylated at the C(2) position
(entries 9-11, Table 2), and 2-aryl-1,3,4-oxadiazoles can
be successfully carboxylated as well (entries 12-16, Table
2). The reactions are compatible with aryl-Cl, Br, CF3,
and OMe groups. Unfortunately, imidazole and thiophene
derivatives cannot be coupled, probably because of the
low acidity of their C(2)-H bonds. 2-Aryl-1,3,4-thiadia-
zoles gave ring-opened products under the carboxylation
conditions.
Table 1. Optimization of Conditions for Direct Carboxylation of
Benzothiazolea
(20) Olah, G. A.; Torok, A.; Joschek, J. P.; Bucsi, I.; Esteves, P. M.;
Rasul, G.; Prakash, G. K. S. J. Am. Chem. Soc. 2002, 124, 11379–11391.
(21) Vechorkin, O.; Proust, V.; Hu, X. L. Angew. Chem., Int. Ed. 2010,
49, 3061–3064.
entry
conditions
conversion (%)
1
2
3
4
5
6
7
8
9
DMF, 125 °C, 5 mol % CuI, LiOBut
DMF, 125 °C, LiOBut
DMF, 125 °C, NaOMe or NaOH or KOH
DMF, 125 °C, K2CO3
DMF, 125 °C, K3PO4
DMF, 125 °C, Cs2CO3
dioxane or toluene, 125 °C, Cs2CO3
THF or CH3CN, 90 °C, Cs2CO3
DMF, 125 °C, Cs2CO3, no CO2
100
100
0
10
(22) Daugulis, O.; Do, H. Q.; Shabashov, D. Acc. Chem. Res. 2009, 42,
1074–1086.
(23) Do, H. O.; Khan, R. M. K.; Daugulis, O. J. Am. Chem. Soc. 2008,
130, 15185–15192.
(24) Hachiya, H.; Hirano, K.; Satoh, T.; Miura, M. Angew. Chem., Int.
Ed. 2010, 49, 2202–2205.
20
100 (95b)
(25) Kitahara, M.; Hirano, K.; Tsurugi, H.; Satoh, T.; Miura, M.
Chem.sEur. J. 2010, 16, 1772–1775.
0
0
0
(26) Canivet, J.; Yamaguchi, J.; Ban, I.; Itami, K. Org. Lett. 2009, 11,
1733–1736.
(27) Turner, G. L.; Morris, J. A.; Greaney, M. F. Angew. Chem., Int.
Ed. 2007, 46, 7996–8000.
a Reaction scale: benzothiazole (1 mmol), base (1.2 mmol), and solvent
(2 mL). b Isolated yield.
(28) Ackermann, L.; Althammer, A.; Fenner, S. Angew. Chem., Int. Ed.
2009, 48, 201–204.
(29) Besselievre, F.; Piguel, S. Angew. Chem., Int. Ed. 2009, 48, 9553–
9556.
(30) Zhao, D. B.; Wang, W. H.; Yang, F.; Lan, J. B.; Yang, L.; Gao,
G.; You, J. S. Angew. Chem., Int. Ed. 2009, 48, 3296–3300.
(31) Huang, J. K.; Chan, J.; Chen, Y.; Borths, C. J.; Baucom, K. D.;
Larsen, R. D.; Faul, M. M. J. Am. Chem. Soc. 2010, 132, 3674–3675.
(32) Nakao, Y.; Kashihara, N.; Kanyiva, K. S.; Hiyama, T. Angew.
Chem., Int. Ed. 2010, 49, 4451–4454.
acid was identified as the only detectable product by NMR.
To our surprise, control experiment showed that the same
reaction occurred even without CuI under otherwise identical
conditions (entry 2, Table 1). Thus, the carboxylation does
not require a catalyst. Consistent results were obtained with
a 400 mBar over pressure of CO2. Both 97% (Sigma-Aldrich)
and 99.9% LiOBut (Alfa Aesar) worked. Other bases were
then tested. NaOMe, NaOH, and KOH were ineffective
(entries 3, Table 1). The use of K2CO3 and K3PO4 resulted
in low conversions (10-20%, entries 4 and 5, Table 1).
Cs2CO3 (purity from 98-99.995%) could be used and gave
reproducible results (entry 6).35 Since Cs2CO3 is less basic
than LiOBut, it is used for further reactions.36 Further
(33) Bordwell pKa table, available at http://www.chem.wisc.edu/areas/
reich/pkatable/index.htm.
(34) Kawano, T.; Hirano, K.; Satoh, T.; Miura, M. J. Am. Chem. Soc.
2010, 132, 6900–6901.
(35) Six of the seven commercial Cs2CO3 tested mediated the carboxy-
lation and gave reproducible results. The six sources are Aldrich (98% and
99.995%), Acros (99.5%), Alfa Aesar (99%), VWR (extra pure), and Chem-
Impex (99.9%). Only the Cs2CO3 from Alfa Aesar (99.994%) did not work.
Control experiment showed that it is likely due to the fact that this Cs2CO3
contains ca. 1 equiv of water, which inhibits the reaction. See Supporting
Information for details.
(36) Flessner, T.; Doye, S. J. Prakt. Chem. 1999, 341, 186–190.
(37) See Supporting Information.
3568
Org. Lett., Vol. 12, No. 15, 2010