À
Table 1. Optimization of the intramolecular C H amination of N-phenyl-
pyridin-2-amine (3a).
(Table 1, entries 12 and 13) promoted the reaction to a sig-
nificantly lesser extent. Other copper salts (Table 1, en-
tries 14 and 15) failed to effect the conversion.
Subsequently, the influence of the structure of the acid
additive was investigated (Table 1).[17] Other carboxylic acids
such as butyric, pivalic, and benzoic acid gave essentially the
same conversion of 3a to 4a after a reaction time of 18 h
(Table 1, entries 16–18). Stronger acids such as trifluoroace-
tic acid can also be used but only in a catalytic amount
(Table 1, entries 19 and 20). Also non-carboxylic acid con-
taining acids are allowed (Table 1, entries 21 and 22), but
again when the acid is strong the use of a stoichiometric
amount is detrimental (Table 1, entry 23). As the conversion
data in Table 1 do not take into account side reactions of
the substrate 3a, we looked at isolated yields in the pres-
ence of acetic acid and several substituted benzoic acids pos-
sessing different pKa values.
Interestingly, a significant difference in the isolated yields
and amount of recovered substrate was found for the stud-
ied acids (acetic acid: 60% 4a, 10% 3a; 4-methoxybenzoic
acid (pKa: 4.47): 52% 4a, 30% 3a; benzoic acid (pKa: 4.17):
75% 4a, 16% 3a; 3,4,5-trifluorobenzoic acid (TFBA) (pKa:
3.46): 92% 4a, no 3a).[18] These results indicate that the se-
lectivity of the reaction is dependent on the type of acid ad-
ditive used. In the case of acetic acid and 4-methoxybenzoic
acid, the missing mass balance is 22% and 10%, respective-
ly, in comparison with the experiment involving 3,4,5-tri-
fluorobenzoic acid (Table 2, entry 2). 3,4,5-Trifluorobenzoic
acid is clearly a superior additive, additionally providing a
faster reaction and full conversion of 3a. In the case of less
acidic additives (acetic, 4-methoxybenzoic, and benzoic
acid), starting material was recovered. A similar selectivity
trend was observed for N-(4-fluorophenyl)pyridin-2-amine
(3b) with the yield of the desired compound increased with
decreasing pKa value (4-methoxybenzoic acid: 38% 4b;
benzoic acid: 68% 4b; 3,4,5-trifluorobenzoic acid: 75%
4b). In this case, however, no starting material remained.
Ultra performance liquid chromatography (UPLC) analysis
of the crude reaction mixtures showed various amounts of
substrate dimers depending on the acid used. We therefore
currently believe that the pKa related selectivity of the acid
can be rationalized by the control of competing intra- and
Entry Solvent
Catalyst
Additive mol%
T
Conv.
[8C] [%][a,b]
1
2
3
4
5
6
7
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO/1%
H2O v/v
Cu
Cu(OAc)2·H2O AcOH
Cu(OAc)2·H2O AcOH
Cu(OAc)2·H2O
Cu(OAc)2·H2O AcOH
Cu(OAc)2 AcOH
Cu(OAc)2·H2O AcOH
(OAc)2·H2O
–
–
15
100
–
100 43
100 47
100 53
120 82
120 92
120 65
120 90
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
T
–
T
15
15
15
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
8
DMSO/10% Cu
(OAc)2·H2O AcOH
15
120 19
120 17
H2O v/v
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
DMSO
9
Cu
Cu
Cu(O2CCF3)2
Cu(OTf)2
A
100
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
A
120
0
N
AcOH
AcOH
AcOH
AcOH
AcOH
15
15
15
15
15
120 84
120 28
120 29
120
120
120 91
120 92
120 90
120 95
120 17
120 92
120 89
AHCTUNGTRENNUNG
CuSO4
CuCl2
Cu(OH)2
5
0
Cu
Cu(OAc)2·H2O PivOH
Cu(OAc)2·H2O BzOH
Cu(OAc)2·H2O TFA
Cu(OAc)2·H2O TFA
Cu(OAc)2·H2O NH4Cl
Cu(OAc)2·H2O HCl
Cu(OAc)2·H2O HCl
Cu(OAc)2·H2O TFBA
A
T
15
15
15
100
15
15
AHCTUNGTRENNUNG
N
G
AHCTUNGTRENNUNG
N
E
100
15
120
2
AHCTUNGTRENNUNG
120 15[c,d,e]
[a] Reactions were performed under an atmosphere of O2 (ca. 1 atm) at a
concentration of 0.5m in standard (undried) DMSO, unless otherwise in-
dicated with 15 mol% CuII salt. [b] Conversion of 3a to 4a determined
by HPLC after 18 h. [c] Under an Ar atmosphere, after 24 h. [d] Con-
firmed by 1H NMR spectroscopy of the crude reaction mixture. [e] With
AcOH (15 mol%) in either DMSO or DMF under an Ar atmosphere,
similar results were obtained.
served as with one equivalent (Table 1, entry 2). At higher
temperature (1208C) a better conversion was achieved
(Table 1, entries 4 and 5). Because the addition of 15 mol%
acid was optimal, we decided to perform further optimiza-
tions with a catalytic amount of acid. Interestingly, anhy-
drous CuII
ACHTUNGTRENNUNG(OAc)2 proved less effective than the monohy-
À
drate (Table 1, entry 6), and the reaction proved not to be
sensitive to the presence of 1% v/v water (Table 1, entry 7),
but a higher amount was detrimental (Table 1, entry 8). The
use of dimethoxyethane, dioxane, acetonitrile, 1-propanol,
or toluene as solvent failed to effect the cyclization. Only
solvents in which a S=O or C=O moiety was present al-
lowed the reaction (DMSO, sulfolane, NMP, DMA, and
DMF).[15] Among these DMSO and DMF gave the highest
conversions, and the former was arbitrarily chosen for fur-
ther studies.[15] Interestingly, the use of base instead of acid
(acetate and carbonate) had an inhibitory effect on the con-
version (Table 1, entries 9 and 10), and bidentate amine li-
gands[16] completely shut the reaction down. CuII catalysts
with carboxylate counterions performed well (Table 1, en-
tries 5 and 11), whilst those with a sulfonate or sulfate group
intermolecular C N bond formation. This will be studied in
future research.
The results obtained for substrates 3a and 3b clearly indi-
cated that the use of a catalytic amount of 3,4,5-trifluoro-
benzoic acid is superior and therefore the scope was as-
sessed for these optimized reaction conditions [CuII-
.
(OAc)2 H2O
(15 mol%),
3,4,5-trifluorobenzoic
acid
(15 mol%), DMSO, 1208C] for a set of substituted N-phe-
nylpyridin-2-amines (Table 2). Both electron-donating
(OMe, entry 8) and electron-withdrawing groups (Cl,
entry 11; COOEt, entry 12; CF3, entry 16) in the para posi-
tion of the benzene ring were tolerated, although the ethox-
ycarbonyl and trifluoromethyl group required a 20% load-
ing of catalyst (Table 2, entries 13 and 17). With 15 mol%
CuII
ACHTNUTRGNE(UNG OAc)2 at least 15% of substrate 3 could be recovered.
6316
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 6315 – 6320