Angewandte
Communications
Chemie
the asymmetric umpolung addition of trifluoromethyl keti-
mines to electrophiles has been documented so far.[6] Very
recently, Deng and co-workers successfully applied this
umpolung strategy to the synthesis of optically active
trifluoromethyl amines, with a chiral tertiary stereocenter,
from readily available trifluoromethyl ketimines (1) and enals
by the employing a chiral phase-transfer catalyst (PTC;
Scheme 1b). Inspired by Dengꢀs work and our continual
interest in asymmetric phosphine catalysis,[7,8] we envisaged
that the Morita–Baylis–Hillman (MBH) carbonates 2,
another very readily available and widely used compounds
in organocatalytic reactions,[9] would generate the chiral
quaternary phosphonium salts 3 in the presence of chiral
phosphine catalyst (Scheme 1c). Subsequent SN2’ reaction
with the in situ generated 2-azaallyl anions 4 would deliver
the corresponding enantioenriched trifloromethyl amines 5
with a chiral tertiary stereocenter. Herein, we report a novel
phosphine-catalyzed and highly enantioselective umpolung
addition of trifluoromethyl ketimines to MBH carbonates
under mild reaction conditions, and it provides facile access to
optically active trifluoromethyl amines with a chiral tertiary
stereocenter. Moreover, synthetically valuable a-methylene
g-lactams could be achieved through further transformations
of the products by the use of a simple protocol.
Figure 2. The screened chiral phosphine catalysts.
catalysts were examined at room temperature. Our recently
developed chiral phosphine catalysts Xiao-Phos (S,RS)-X1
and (S,RS)-X10 (Figure 2), as well as Wei-Phos (S,RS)-W1
gave the adduct 5aa in moderate enantioselectivities (Table 1,
entries 1–3). The substituent effect at the ortho-position of the
phenyl ring, such as in (S,RS)-P1–P8, was then investigated.
(S,RS)-P5, with the rigid 9-anthracenyl substituent gives better
results (entry 8). To further enhance the capability of hydro-
gen-bonding or proton donation, two new catalysts, (S)-P9
with 3,5-bistrifluoromethylphenyl thiourea and (S)-P10 with
3,5-bistrifluoromethyl benzoyl, were prepared from P5 in two
steps (entries 12 and 13). To our delight, 78% yield and 81%
ee could be obtained for the product when using (S)-P10 as
the catalyst. Further studies showed that the ee values and
yields could be further improved with increasing the size of
the ester group of the MBH carbonate from methyl to ethyl,
isopropyl, and tert-butyl (entries 14–17), and the adduct 5ae
was obtained in 85% yield with 98% ee from 2e. The
evaluation of the solvents showed that toluene was the best
reaction medium in terms of reactivity and enantioselectivity
(entries 17–19). And an 82% yield with 99% ee could be
furnished when running the reaction of 1a with 2e at À108C
instead of room temperature. Remarkably, the catalyst
loading could be further reduced to 2.5 mol%, thus delivering
52% yield and 94% ee.
With the optimized reaction conditions in hand, the scope
with respect to the aryl trifluoromethyl ketimines 1 was
investigated by reaction with 2e (Scheme 2). To our delight,
the aryl trifluoromethyl ketimines bearing diverse functional
groups, such as the halogens (F, Cl, Br), an electron-with-
drawing group (CF3), and electron-donating groups (tBu, Me,
OMe) at the para-position of the phenyl ring generally react
well, thus delivering the corresponding products 5de–he in
good yields with 93–99% ee values under the optimal reaction
conditions. Further substrate scope investigation demon-
strated that the meta-substituent of the aryl trifluoromethyl
ketimines is also compatible and delivers the desired products
5ie–je in good yields and high enantioselectivities. High
ee values and good yields could be also obtained when phenyl
was replaced by other aryl groups such as naphthyl (5ke) and
thienyl (5le). For example, the thienyl-derived 1l delivered
the corresponding amine 5le in 70% yield with 96% ee.
Finally, it is noteworthy that this reaction is amenable to gram
scale without obvious loss of the efficiency and enantioselec-
tivity, even with a lower catalyst loading. 5ae could be
With this hypothesis in mind, we chose the phenyl
trifluoromethyl ketamine 1a with the MBH carbonate 2a as
model substrates (Table 1). Awide variety of chiral phosphine
Table 1: Reaction optimization.[a]
Entry
2, R=
Cat.
Solvent
Yield [%][b] (ee [%])[c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21[d]
22[e]
22[f]
2a, Me
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2a
2b, Et
2c, Bn
2d, iPr
2e, tBu
2e
(S,RS)-X1
(S,RS)-X10
(S,RS)-W1
(S,RS)-P1
(S,RS)-P2
(S,RS)-P3
(S,RS)-P4
(S,RS)-P5
(S,RS)-P6
(S,RS)-P7
(S,RS)-P8
(S)-P9
(S)-P10
(S)-P10
(S)-P10
(S)-P10
(S)-P10
(S)-P10
(S)-P10
(S)-P10
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
PhMe
THF
5aa, 61 (62)
5aa, 67 (63)
5aa, 64 (42)
5aa, 67 (60)
5aa, 62 (57)
5aa, 65 (62)
5aa, 68 (58)
5aa, 72 (75)
5aa, 67 (55)
5aa, 66 (66)
5aa, 71 (60)
5aa, 70 (42)
5aa, 78 (81)
5ab, 74 (92)
5ac, 77 (80)
5ad, 75 (96)
5ae, 85 (98)
5ae, 78 (95)
5ae, 84 (95)
5ae, 79 (94)
5ae, 82 (99)
5ae, 73 (97)
5ae, 52 (94)
2e
2e
2e
2e
DCM
Et2O
PhMe
PhMe
PhMe
(S)-P10
(S)-P10
(S)-P10
2e
[a] Unless otherwise specified, 1a (0.2 mmol), 2 (0.3 mmol), catalyst
(0.02 mmol), solvent (2 mL), RT. [b] Yield of isolated product. [c] Deter-
mined by HPLC analysis using a chiral stationary phase. [d] 08C.
[e] 5.0 mol% catalyst, 24 h. [f] 2.5 mol% catalyst, 48 h.
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 6
These are not the final page numbers!