Organic Letters
Letter
render a chiral ion pair comprising the enolate and the azolium
ion. The activated chiral enolate is then expected to react with
a suitable azodicarboxylate to enable synthesis of a wide range
of open-chain 1,3-dicarbonyl compounds containing an N-
substituted quaternary stereocenter with excellent enantiose-
lectivity and yield.
We began our studies using β-ketoamide 1a as a model
substrate and di-tert-butyl azodicarboxylate (2a) as the
electrophilic nitrogen source and widely employed chiral
azolium salt 3a as the precatalyst under various reaction
conditions (Table 1). When the reaction was performed using
lectivity of product 4d was observed with amide 1d. Following
this observation, other ketoamides bearing o-ethyl- and o-
isopropyl-substituted aniline units were examined. While the
reaction efficiency could be improved with almost all of the
ketoamides derived from the ortho-substituted aniline unit,
ketoamide 1e found to be the most effective (entry 11).
Importantly, the loading of 3d could be reduced further. An
optimal catalyst loading for the enantioselective amination
reaction was set to 10 mol % (entries 13−15).
Having established the optimized reaction conditions, we
direct our attention toward evaluating the substrate scope of
the reaction (Figure 1). Accordingly, acylamides bearing
alkene and alkyne functional groups were subjected to the
reaction. In all of the cases, the reaction afforded potentially
useful acyclic α-amino ketones 5−8 in excellent results.
Ketoamides having different functionalized alkyl moieties at
the α-position gave products 9−13 in excellent yields and ee’s.
The reaction worked equally well with α-benzyl-substituted
1,3-ketoamides to afford products 14−21 in nearly perfect ee
values and yields. Importantly, products 10 and 19 were
prepared in a >1.3 g quantity with essentially no change in ee
and yield, thus establishing the synthetic potential of the
method.
a
Table 1. Reaction Development
Aromatic ketone-containing substrates were tolerated in this
protocol. However, they require a slightly high catalyst loading
(20 mol %) and a longer reaction time. Substrates containing
phenyl as well as aromatic ring having F, Cl, Br (irrespective of
their position), CN, CF3, Me, and OMe substituents worked
well in this method (22−30). Different heteroaromatic ring-
containing substrates underwent the α-amination reaction to
deliver products 31−33 in good yields and ee’s. The
ketoamide bearing a 2-naphthyl ring was found to be a
challenging substrate (34).
The NHC-catalyzed enantioselective amination could be
applied to acyclic 1,3-amidoesters possessing different alkyl
substituents at the α-position. Like β-ketoamides, 1,3-
amidoesters exhibited very similar reactivity toward the
catalytic amination reaction. For α-methyl-substituted ami-
doesters, the enantioselectivity of the products decreases
gradually with increasing steric bulk on the ester unit (35−37).
Notably, the reaction worked well with sterically unbiased 1,3-
amidoesters (35 and 38). With this protocol, 1,3-amidoesters
bearing synthetically useful functional groups like allyl,
propargyl, Br, and OTBS were aminated smoothly (39−42,
respectively). Furthermore, the enantioselective amination of
α-benzyl-substituted amidoesters worked in equal efficiency in
this catalytic process (43 and 44).
Cyclic ketoamides could also be aminated with this catalytic
amination process in excellent enantioselectivity. Nevertheless,
the chemical reactivity of the process appeared to be strongly
dependent on the ring size of the cyclic ketoamides. The
reaction afforded products 45 and 46 with eight- and seven-
membered cyclic substrates, respectively, in excellent yields
and ee’s. However, a sharp decrease in the chemical yield of
product 47 was realized despite an excellent ee with a six-
membered cyclic ketoamide, and no product formation was
realized when a cyclopentanone-derived substrate was used.
The increasing conformational rigidity in six- or five-
membered cyclic substrates may be the reason for such an
observation.
b
catalyst
time
(h)
temp
yield
(%)
ee
entry
(mol %)
amide
(°C)
(%)
1
2
3
4
5
6
7
8
3a (20)
3a (20)
3a (20)
3b (20)
3c (20)
3d (20)
3e (20)
3d (20)
3d (20)
3d (20)
3d (20)
3d (20)
3d (15)
3d (10)
3d (5)
1a
1a
1a
1a
1a
1a
1a
1b
1c
1d
1e
1f
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
2.0
1.25
1.25
1.5
2.0
2.0
0
−40
−78
−40
−40
−40
−40
−40
−40
−40
−40
−40
−40
−40
−40
83
74
63
76
75
78
36
92
88
90
96
92
97
96
48
29
61
64
60
42
81
40
80
83
90
90
88
90
90
88
9
10
11
12
13
14
15
1e
1e
1e
a
Reaction conditions: 1a (0.05 mmol), 2a (0.055 mmol), LiHMDS
(16−4 mol %), 4 Å molecular sieves (35 mg) in toluene (1.0 mL);
isolated yield. ee determined by HPLC analysis on a chiral stationary
b
phase.
20 mol % 3a, 16 mol % LiHMDS, and 4 Å molecular sieves in
toluene at 0 °C, aminated product 4a was isolated in 83% yield
with 29% ee (entry 1). The enantioselectivity of 4a could be
increased to 61% by carrying out the reaction at −40 °C (entry
2). Further decreasing the reaction temperature did not
improve the reaction outcome (entry 3). To obtain 4a in an
acceptable ee, other 1-amino-2-indanol-derived azolium salts
3b−e were tested (entries 4−7, respectively). Among them,
azolium salt 3d was found to be promising. At this point, we
decided to induce steric bulk on the aniline moiety of the
ketoamide, anticipating further improvement in the ee value of
the desired product. Hence, p-, m-, and o-toluidine-derived β-
ketoamides 1b−d were tested in this reaction (entries 8−10,
respectively). Indeed, a sharp improvement in the enantiose-
To illustrate the synthetic utility of the method, products 10
and 19 were transformed to a variety of valuable
enantioenriched compounds (Scheme 2; see the Supporting
B
Org. Lett. XXXX, XXX, XXX−XXX