Organic Letters
Letter
a
acetylpyridines have been a biocatalyst (transaminase)
approach,16 reduction of an oxime and then diastereomeric
salt resolution,17 and reduction of a chiral sulfonamide.18
Herein, we report a highly efficient asymmetric synthesis of
corresponding chiral amines via DARA of 2-acetyl-6-
substituted pyridines with >99% conversion and >99% ee.
We started our research with 2-acetyl-6-methoxypyridine
(1a) as a model substrate in the presence of a commercially
available BINAP-based ruthenium catalyst, Ru(OAc)2{(S)-
binap}, and various ammonium salts as the nitrogen source in
THF under 0.8 MPa of hydrogen gas pressure (Table 1). This
Table 2. Effect of RuX{(S)-binap} on DARA of 1a
entry
RuX{(S)-binap}
conversion (%)
% ee of 2a
1
2
3
Ru(OAc)2{(S)-binap}
Ru(OCOCF3)2{(S)-binap}
RuCl2{(S)-binap}
98.8
96.7
96.7
97.9
98.0
98.7
a
Reactions were conducted for 17 h at 90 °C in THF using a
RuX{(S)-binap} under 0.8 MPa of hydrogen pressure.
a
Table 1. Effect of Ammonium Salt on DARA of 1a
a
Table 3. Effect of Reaction Solvent on DARA of 1a
b
,c
d
entry
NH4X
conversion (%)
% ee of 2a
f
1
2
3
NH4OAc
47.6
no data
97.9
77.6
HPLC area %
NH4CO2CF3
98.8
90.5
e
entry
solvent
methanol
ethanol
2-propanol
THF
toluene
EtOAc
trifluoroethanol
CH2Cl2
1a
2a
2a′
2.3
2.4
1.3
1.6
10.4
5.0
7.3
NH4SA
a
1
2
3
4
5
6
7
8
38.7
41.6
52.8
3.6
33.7
13.9
46.5
0.3
1.7
4.5
6.0
72.2
44.0
60.9
4.5
Reactions were conducted for 17 h at 90 °C in THF using
b
Ru(OAc)2{(S)-binap}under 0.8 MPa of hydrogen pressure. Deter-
mined by HPLC analysis. Conditions: L-column-2 ODS (5 μm, 150
mm × 4.6 mm), mobile phase A consisting of a 190/5/5 phosphate
buffer/MeCN/MeOH mixture, mobile phase B consisting of a 50/
140/10 phosphate buffer/MeCN/MeOH mixture, phosphate buffer
preparation, 1.3 g of H3PO4 and 9 mL of 1 mol/L aqueous KOH/
1000 mL of water, gradient of 0% to 100% B from 0 to 32.5 min and
100% B from 32.5 to 40 min, flow rate of 1.0 mL/min, oven
temperature of 25 °C, UV detection at 254 nm, tR = 9.1 and 9.5 min
48.3
32.4
a
Reactions were conducted for 17 h at 90 °C in the indicated solvent
c
using Ru(OAc)2{(S)-binap} under 0.8 MPa of hydrogen pressure.
(2a), tR = 24.8 min (1a). Calculation method: 100-HPLC area % of
d
1a. Determined by chiral HPLC analysis. Conditions: CHIRALPAK
(2a′) as an undesired product (entries 5, 6, and 8). The
enantiomeric excess of 2a was not measured when checking
the conversion.
AY-RH column (5 μm, 150 mm × 4.6 mm), 65/35 0.1% AcOH
aqueous solution/MeCN mixture, 1.0 mL/min, UV 254 nm. The
sample was derivatized to the N-benzoylated form with BzCl and
Et3N. tR = 16.2 min (major N-benzoylated enantiomer), tR = 17.7 min
The substrate scope was then evaluated with the optimized
DARA conditions in hand (Scheme 2). A series of 2-acetyl-6-
substituted pyridines were successfully converted to the chiral
corresponding amines with good to excellent enantioselectivity
(>94% ee) regardless of the type of substituent group, such as
aryl, alkyl, and halogen (entries 2a−2f). When 2-acetyl-6-aryl-
substituted pyridines were used, the enantioselectivity tended
to be decreased by an increased level of electron withdrawal
(entries 2f and 2g). Although 2-acetylisoquinoline was
converted to the corresponding amine, the enantioselectivity
was not good (entry 2h). However, DARA of 2-acetylpyridine,
2-acetyl-4-methoxypyridine, and 2-acetyl-5-methoxypyridine
did not proceed (entries 2s, 2v, and 2w) and 2-acetyl-3-
bromopyridine gave poor conversion (2n). These results
revealed that the substituent at position 6 on the pyridine ring
was extremely important for excellent enantioselectivity and
conversion. Among the 2-acetyl-6-substituted pyridines, the
conversion of 2-acetyl-6-methylesterpyridine (2o) was poor
and 2-acetyl-6-pyridone (2x) showed no reaction. Diaryl
ketone (2t) and fused ring ketone (2u) also showed no
reaction. With replacement of the pyridyl group to phenyl or
naphthyl groups, most of the substrate was recovered (2p−2r).
However, when a β-keto ester or β-keto amide was used
instead of the 2-acetylpyridines, high conversion and
enantioselectivity were achieved similar to 2-acetyl-6-substi-
tuted pyridine (2i−2l). Interestingly, 1-(pyridin-2-yl)propan-2-
one, which is extended by one carbon between the carbonyl
e
f
(minor N-benzoylated enantiomer). Ammonium salicylate. Complex
mixture.
nitrogen source screen found that ammonium trifluoroacetate
was superior to ammonium acetate or salicylate, providing (S)-
1-(6-methoxypyridin-2-yl)ethan-1-amine (2a) with the highest
level of enantioselectivity (entries 1−3). We chose to perform
the study under low-hydrogen pressure (0.8 MPa) conditions
that are operationally friendly from an industrial viewpoint.
To identify the effect of the ruthenium catalyst, some
commercially available RuX{(S)-binap} catalysts were inves-
tigated using ammonium trifluoroacetate as the ammonium
source (Table 2). All catalysts gave the same results in terms of
conversion and enantioselectivity. These results unambigu-
ously showed that trifluoroacetate anion from ammonium
trifluoroacetate would readily interconvert with acetate or
chloride anions on the ruthenium center. Moreover, Noyori
and co-workers have prepared Ru(OCOCF3)2{(R)-binap} by
treating Ru(OAc)2{(R)-binap} with 2 equiv of trifluoroacetic
acid in CH2Cl2.19
A variety of solvents were examined, and THF was identified
as optimal, furnishing good conversion and amine selectivity
(Table 3, entry 4). Alcohols such as methanol, ethanol, and 2-
propanol gave some lipophilic substances on HPLC analysis
(entries 1−3 and 7). Hydrophobic solvents such as toluene,
EtOAc, and CH2Cl2 gave relatively more alcohol formation
3365
Org. Lett. 2021, 23, 3364−3367