Communications
Table 1: Optimization studies.
phenyl-substituted tin derivative gave the lowest yield,
whereas the trimethyl-substituted systems provided the high-
est yield. Along this line, allyldialkylphenyltin derivatives
provided intermediate yields (1d and 1e; Table 1, entries 16
and 17). Switching from methyl to butyl substituents did not
influence reactivity to a large extent (compare Table 1, entries
6 and 14, and 16 and 17).
Under optimized conditions we reacted various isomeri-
cally pure 2-alkenyltributylstannanes with 2-nitrosopyridine
using catalyst loadings of 10 and 5 mol% (Table 2). Z-2-
Entry
Ligand
Cat.
[mol%]
Prod.
Yield
[%][a]
ee
[%][b]
1
(S)-Binap
10
10
10
10
10
10
5
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1c
1c
1d
1e
31
35
43
52
77
77
73
58
–
66[c]
77
2
(R)-Segphos
3
(R)-Solphos
61
4
5
6
7
8
9
10
11[e]
12
13
14
15
16
17
(S)-Difluorphos
(R,P)-Walphos-CH3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
–
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
(R,P)-Walphos-CF3
86[c]
74
96
96
96
2
10[d]
–
–
–
–
Table 2: Variation of the allyltin compound.
–
10
10
5
10
5
<5
47
45
76
71
63
62
96 (99)[f]
96 (99)[f]
98
98
97
96
10
10
Entry R1
R2
Prod. Yield
[%][a,b]
d.r.
ee
[%][c]
[a] Yield of isolated product (reactions conducted at 0.14 mmol scale).
[b] Determined by HPLC on a chiral stationary phase (see the Supporting
Information). [c] R enantiomer formed. [d] With ligand in the absence of
[Cu(MeCN)4]PF6. [e] In THF as a solvent. [f] After crystallization.
1
2
3
4
5
6
7
8
Me
Et
n-Pr
n-Hex
CH2CH2CH(CH3)2
iPr
tBu
CH2OCH2Ph
OMe
H
H
H
Me
H
H
H
H
H
H
H
H
H
2a
2b
2c
2d
2e
2 f
2g
2h
2i
>99 (97) >99:1 >99
83 (81) >99:1
80 (78) >99:1
99
98
84 (82) >99:1 >99
81 (80) >99:1 >99
81 (80) >99:1 >99
71 (63) >99:1 >99
further increased to 96% ee without affecting the yield by
switching to the CF3 congener (Table 1, entry 6). Reducing
the catalyst loading to 5 and 2 mol% led to slightly lower
yields while the excellent selectivity was maintained (Table 1,
entries 7 and 8, respectively). The ligand alone did not
catalyze the process (Table 1, entry 9) and a background
reaction did not occur (Table 1, entry 10). In THF only a trace
amount (< 5%) of the product was formed (Table 1,
entry 11). With triphenylallyltin, isoxazolidine 1b was
formed in significantly lower yield but also excellent selec-
tivity (Table 1, entries 12 and 13). Enantiomerically pure
material was isolated after a single crystallization. Based on
the X-ray structure of 1b the absolute configuration (S en-
antiomer) was unambiguously determined (Figure 1).[10]
Absolute configuration of other products was assigned in
analogy. The highest selectivity (98% ee) was achieved with
allyltrimethyltin (Table 1, entries 14 and 15).
93 (93) >99:1
94 (93) >99:1
98
9
96
10
11
12
13
14
Me 2j
Et 2k
Ph 2l
72 (72)
2:1
98[d]
64
45 (44) >99:1
29 (26) >99:1
45
Me 2m
63 (61)
–
96
CH2CH2CH(CH3)2 Me 2n
46 (44) >99:1 >99
[a] Yield of isolated product (reactions conducted at 0.14 mmol scale).
[b] Yield with 10 mol% cat. loading and (in brackets) with 5 mol% cat.
loading. [c] Determined by HPLC on a chiral stationary phase (see the
Supporting Information). [d] Minor isomer 2a was formed with
>99% ee.
Butenyltributyltin was the most reactive substrate in these
studies and the formal [3+2] cycloaddition delivered adduct
2a in a quantitative yield with perfect selectivity (Table 2,
entry 1). At 5 mol% catalyst loading the yield was still very
high (97%). For this stannane we conducted the reaction
using 0.5 mol% of the catalyst (run on a 1 mmol scale) and
obtained 2a as a single isomer in 96% yield. The cis relative
configuration and the regiochemistry were unambiguously
assigned by 1H NMR spectroscopy (see the Supporting
Information). Increasing the size of the R1 substituent in the
cis-2-alkenyltributylstannanes led to slightly lower but still
very good yields, perfect diastereoselectivties, and excellent
enantioselectivities (Table 2, entries 2–8). Also stannylated
enol ethers proved to be good substrates, as shown by the
preparation of 2i (Table 2, entry 9).
It is evident from these results that the size of the “non-
allylic” substituents at Sn strongly influences reactivity: the
To investigate the stereospecificity of the annulation we
tested E-2-butenyltributyltin and found that reaction was
sluggish compared to the transformation with the cis isomer.
Surprisingly, reaction was not stereospecific and 2j was
isolated along with the minor isomer 2a (ratio 2:1). Enantio-
Figure 1. X-ray crystal structure of 1b.
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 11257 –11260