Liang Zeng et al.
FULL PAPERS
Results and Discussion
Bases also influenced the progress of the coupling re-
action, and 3 equivalents of K3PO4 provided the high-
First, 2-bromobenzoic acid and methylamine hydro- est yield (compare entries 3, 13 and 14). Changing the
chloride were chosen as the model substrates to opti- amount of base was also attempted, and the reaction
mize the catalytic conditions, including copper cata- afforded a slightly lower yield when 2.5 equivalents of
lysts, ligands, bases and solvents in the N-arylation at K3PO4 were used (entry 15). After the optimization
room temperature as shown in Table 1. Several process of catalysts, ligands, solvents and bases, the
copper salts, CuCl, CuBr, CuI, CuSO4, CuCl2 and following coupling reactions were carried out under
CuACHTUNGTRENNUNG(OAc)2 (entries 1–6), were tested using DMF as our standard conditions: 10 mol% CuI as the catalyst,
the solvent, 20 mol% rac-BINOL as the ligand, 20 mol% rac-BINOL as the ligand, DMF as the sol-
K3PO4 (relative to 2-bromobenzoic acid) as the base, vent and 2 or 3 equivalents of K3PO4 (2 equivalents of
and the results showed that CuI was the most effec- base for free amines, 3 equivalents of base for amine
tive catalyst (entry 3). Other ligands, such as l-proline hydrochlorides) as the base (relative to 2-halobenzoic
and N,N’-dimethylethylenediamine, were tested (en- acids) at room temperature under a nitrogen atmos-
tries 7 and 8), and rac-BINOL gave the highest yield phere.
(96%, entry 3). The reaction provided a 58% yield in
The optimized amination procedure was then ap-
the absence of ligand (entry 9). We also investigated plied to a variety of 2-halobenzoic acids and aliphatic
the effect of solvents (compare entries 3, 10–12), amines to evaluate the synthetic potential of this
DMSO was slightly inferior to DMF (entry 10), but method as shown in Table 2. For 2-bromo- and 2-io-
the others were bad solvents (entries 11 and 12). dobenzoic acid derivatives, the coupling reactions pro-
vided good to excellent yields (entries 1–22), but 2-
chlorobenzoic acid was a poorer substrate (entries 23
Table 1. Copper-catalyzed coupling of 2-bromobenzoic acid
with methylamine hydrochloride: optimization of the cata-
lytic conditions.[a]
and 24). Importantly, the copper-catalyzed amination
proceeded with regioselectivity, and the reactions
only occurred on the carbon-halogen bonds adjacent
to the carboxyl group at room temperature, which
was probably because of the accelerating effect of the
adjacent carboxylate group (see reaction mechanism
below). For example, couplings of 2-bromo-4-halo-
benzoic acids (1d and 1e) with aliphatic amines only
yielded N-alkyl-4-haloanthranilic acids (entries 16–
20), however, N-alkyl-2-bromoanthranilic acids were
not observed. Various functionalities were tolerated
including alkenyl (entries 5, 12, 15, 17, 19, 22 and 24),
hydroxy (entry 6), and ester groups (entry 10) in the
aliphatic amines and nitro (entries 11–15), carbon-hal-
ogen bonds (except ortho carbon-halogen bonds of
carboxylates) (entries 16–20) in the 2-halobenzoic
acid derivatives. We attempted to increase the
amount of K3PO4 for the slow reactions (such as en-
tries 8, 9 and 17–20 in Table 2), but the couplings did
not give higher yields. Electronic effects in the 2-halo-
benzoic acids including electron-rich and electron-de-
ficient groups did not show any evident difference.
The procedure is insensitive to moisture. The results
above show that our amination procedure provides a
convenient access to a wide range of N-alkylanthranil-
ic acids from readily available, unprotected 2-halo-
benzoic acids and aliphatic amine derivatives.
Entry Cat.
Ligand Solvent
Base
Yield
[%][b]
1
2
3
4
5
6
7
8
CuCl
CuBr
CuI
CuSO4
CuCl2
A
A
A
A
A
A
B
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMF
DMSO
toluene
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
trace
20
96
0
trace
0
25
67
58
70
0
CuACHTUNGTRENNUNG(OAc)2
CuI
CuI
CuI
CuI
CuI
CuI
CuI
CuI
CuI
C
–
9
10
11
12
13
14
15
A
A
A
A
A
A
1,4-dioxane K3PO4
0
DMF
DMF
DMF
Cs2CO3 35
K2CO3 trace
K3PO4
We also investigated the mechanism of formation
of N-alkylanthranilic acids. The reaction of 2-(2-bro-
mophenyl)acetic acid with methylamine hydrochlo-
ride provided product 3u in only 34% yield under the
catalysis conditions as shown in Scheme 1, Eq. (1),
and the structure of 3u was identified by NMR and
IR spectroscopy (see Supporting Information). The
results showed the importance of the position of the
91[c]
[a]
Reaction conditions: 2-bromobenzoic acid (1.0 mmol),
methylamine hydrochloride (1.5 mmol), catalyst
(0.1 mmol), ligand (0.2 mmol), base (3 mmol), solvent
(3 mL) at room temperature (~258C) under a nitrogen
atmosphere.
Isolated yield.
2.5 equiv. of K3PO4 were used as the base.
[b]
[c]
1672
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 1671 – 1676