Angewandte
Communications
Chemie
Table 1: Optimization of reaction conditions.[a]
by using ent-L-RaPr2 (entry 13). In
all cases, the d.r. value of 3a was
over 19:1.
With the optimized reaction
conditions established, various rac-
emic 2H-azirines were tested
(Table 2). Satisfyingly, 2H-azirines
containing either electron-with-
drawing or electron-donating sub-
stituents on the 4-position of the
phenyl ring (R1) were smoothly
converted into the corresponding
products 3b–h in good yields and
Entry
Ligand
Solvent
1a
3a
s[d]
excellent
(entries 2–8).
enantioselectivities
Meanwhile, the
Yield [%][b]
ee [%][c]
<5
7
17
15
6
77
<5
98
90
95
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
7
8
9
L-RaPh
L-RaMe2
L-RaEt2
L-RaPr2
L-PiPr2
L-PrPr2
L-RaPr2
L-RaPr2
L-RaPr2
L-RaPr2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
CH2Cl2
Et2O
THF
67
83
77
82
90
29
95
15
32
46
31
16
20
12
5
62
trace
83
67
49
45
37
65
96
70
33
73
20
40
93
–
2
6
unreacted 1b–h were recovered in
yields of 43–52% with 94–99% ee.
Although substituents on the 3/2-
position of the phenyl group 1i/1j
and disubstituted 1k exhibited
lower reactivities (entries 9–11),
the results were still satisfying
after prolonging the reaction time
and increasing the catalyst loading
to 10 mol%. The compounds 1l and
1m, having a fused ring and naph-
thyl substitutent, respectively, also
underwent the KR process with the
corresponding s factors of 354 and
100 (entries 12 and 13). Although
alkyl groups on the 3-position of the
57
6
4
–
5
7
toluene
10
toluene/Et2O
(v/v, 4:1)
toluene/Et2O
(v/v, 4:1)
toluene/Et2O
(v/v, 4:1)
toluene/Et2O
(v/v, 4:1)
103
11[e]
12[f]
13[g]
L-RaPr2
49
49
50
98
98
49
50
50
94
94
149
149
149
L-RaPr2
ent-L-RaPr2
À98
À94
[a] Unless otherwise noted, all reactions were performed with Sc(OTf)3/ligand (1:1, 5 mol%), 1a
(0.10 mmol), and 2c (0.10 mmol) in solvent (0.5 mL) at 358C for 6 h. [b] Yield of isolated product.
[c] Determined by chiral-phase HPLC analysis. [d] Selectivity factors s, calculated according to the
following equations: s=ln[(1ÀC)(1—ee1)]/ ln[(1ÀC)(1+ee1)], C=(ee1)/(ee1 +ee3). [e] The reaction was
carried out with 2c (0.10 mmol), metal/ligand (1:1.2, 5 mol%), 1a (0.10 mmol), toluene (0.4 mL), Et2O
(0.1 mL) at 358C for 5.5 hours. [f] The reaction was performed with 0.5 mol% catalyst for 23 hours.
2H-azirine
were
compatible
(entries 14 and 15), the substrate
1n, having a methyl group gave low
s factor. In comparison, 1o, having
a more bulky benzyl substituent
resulted in an improved s factor of
41. Next, R2 was varied (entries 16–
20), and the reaction was unbiased
1
[g] The reaction was run for 5.5 hours. The d.r. value of 3a was over 19:1 as detected by H NMR
spectroscopy. THF=tetrahydrofuran.
affected the enantioselectivities of recovered 1a and product
3aa. Improving the steric hindrance from a 2,6-dimethyl to
2,6-diisopropyl group resulted in an increase of the s factors
from 2 to 57 (entries 2–4). Subsequently, l-pipecolic-acid-
derive L-PiPr2 and l-proline-derived L-PrPr2 were investi-
gated (entries 5 and 6), but no better result was obtained. To
further improve the reaction results, various solvents were
screened. When Et2O was used (entry 7), the recovered 1a
was obtained in 95% yield with less than 5% ee. Fortunately,
the reaction rate was improved in both THF and toluene in
spite of the decrease in the s factors (entries 8 and 9).
Additionally, when the mixed solvent of Et2O and toluene
(v/v, 1:4) was added, both the s value and reactivity were
greatly improved (entry 10). Finally, by adjusting the ratio of
metal to ligand to 1:1.2, and the s factors increased up to 149.
The unreacted 1a was recovered in 49% yield with 98% ee,
and 3a was formed in 49% yield with 94% ee (entry 11).
Gratifyingly, the catalyst loading could be reduced to
0.5 mol% without affecting the results (entry 12). As
expected, ent-1a and ent-3a were formed enantioselectively
toward electronic properties, as excellent enantioselectivities
were obtained for both the desired products 3p–s (92–95%
ee) and the recovered 1p–s (93–96% ee). The reaction of rac-
1t, having an alkyl group in the 2-position of the 2H-azirine,
gave the recovered 1t in 78% ee with 20% yield. The absolute
configuration of the product 3q was determined to be (2S, 3S)
by X-ray crystallography analysis.[11]
Next, the synthetic potential of the recovered 2H-azirines
was investigated. In the presence of NaBH4, 1q was converted
to into the unprotected aziridine 1qa in 90% yield with
excellent diastereo- and enantioselectivity (Scheme 2a).
Additionally, when 1a was treated with NaOMe, the nucle-
ophilic imine addition delivered the intermediate A, which
was sequentially protected by BzCl and hydrolyzed with HCl
(6n, aq.) to give the a-amino ketone 4 (Scheme 2b).
Based on the X-ray structure of the N,N’-dioxide/ScIII
complex[9a] and the absolute configuration of the product
3q, a transition state for the KR process was proposed
(Figure 1). In TS-1, the oxindole preferentially attacked (S)-
1 from the Re face of the 2H-azirine via imine formation
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2016, 55, 1 – 5
These are not the final page numbers!