Angewandte
Chemie
tion conditions, and they generally provided the anti adducts
as the major products (Table 1, entries 3–5). The enantiose-
lectivities for the anti adducts was excellent, although in most
cases the corresponding syn adducts had lower ee values
(however, their absolute configuration is not clear). Intra-
molecular hydrogen bonding presumably plays a key role in
this undesired observation, as evident from the fact that
reaction of 3-methylbutanal with 7 in methanol still gave syn-
adduct 13a as the major product (Table 1, entry 6).
Although further experiments are required for a detailed
mechanistic investigation of this unusual stereoselectivity,
tentative proposals are outlined in Scheme 2. These possible
Scheme 1. Synthesis of nitroolefin 7 and its Michael addition with
aldehyde 10a catalyzed by O-TMS protected diphenylprolinols 9.
TMS=trimethylsilyl, DMAP=4-(dimethylamino)pyridine.
attempted to convert the major isomer 11a into oseltamivir
(for the reaction sequence, see Scheme 4), and surprisingly
found that the final product was the enantiomer of oseltami-
vir. This result indicated that 11a is the (2S,3R) isomer, not
the (2R,3S) isomer that was predicted according to the
transition-state model for Michael additions to simple nitro-
olefins.[7]
When 3-methylbutanal was used as a Michael donor,
another interesting result was observed: anti adduct 12a was
determined to be the major product (Table 1, entry 1).
However, the diastereoselectivity was not satisfactory. As
only one recent example of the anti-selective asymmetric
Michael reaction of aldehydes and nitroolefins has been
reported,[11] we decided to try to enhance the anti/syn-
isomeric ratio by changing the reaction conditions. After
several experiments, we were pleased to discover that higher
selectivity could be obtained by adding 4 ꢁ molecular sieves
to the reaction, using acetic acid as the additive, and slightly
reducing the reaction temperature (Table 1, entry 2). Other
aldehydes were then examined under these optimized reac-
Scheme 2. Possible reaction pathway for Michael addition of nitro-
olefin 6 with aldehydes.
pathways are based on the acyclic synclinal transition-state
model for enamine-based Michael additions, as proposed by
Seebach and Golinski.[12] We realized that isomerization of 7
´
did not occur in chloroform, because the intramolecular
hydrogen bond was too strong. As a result, 7 might directly
interact with E enamines to form transition-state A, which in
turn gave the anti-selective adducts. However, there should be
a marked steric repulsion between the R group and the amido
moiety in transition-state A. When R was the more bulky
OCH(CH2CH3)2 group, the steric repulsion was very high,
which predominantly led to reaction of the Z enamine[13] with
7. This model could be used to rationalize the absolute
configuration of syn-adduct 11a, which was different to the
absolute configuration of the products that were generated
from interaction of E nitroolefins with E enamines.
Table 1: Organocatalytic Michael of cis-olefin 7 with aldehydes.[a]
As it is difficult to convert 7 into its trans isomer (8), we
decided to obtain the corresponding trans olefins by removing
the intramolecular hydrogen bond through the introduction
of another N substituent. Accordingly, exposure of amine 6 to
a solution of phthaloyl dichloride and triethylamine in
methylene chloride afforded 14 in 90% yield (for exper-
imental details, see the Supporting Information). Next, we
investigated the Michael reaction of 14 with n-butyraldehyde
under different reaction conditions (Table 2). In the presence
of 5 mol% 9c, the reaction proceeded well in chloroform to
afford the desired syn adduct (15a) with good yield and 99%
ee (Table 2, entry 1). The diastereoselectivity could be
improved by changing the solvent to acetonitrile (Table 2,
entry 2), and further increased by reducing the reaction
temperature (Table 2, entry 3). The highest ratio of syn/
anti isomers (14:1) was observed when the catalyst was
Entry
T
[8C]
t [h] Product
Yield
[%][b]
anti/syn
ee
ratio[c]
[%][d]
1
2
3
4
5
À5 12
12a: R=iPr
90[e]
98
3:1
7:1
9:1
6:1
4:1
94 (74)
98 (84)
98 (96)
93 (41)
94 (55)
À10
À10
À10
À10
9
2.5 12b: R=Et
1.5 12c: R=Bn
95[f]
93
3
12d:
80
R=(CH2)3Cl)
12a: R=iPr
6
25
3
91[g]
1:1.4
68 (94)
[a] Reaction conditions:
(0.04 mmol), HOAc (0.04 mmol), 4 ꢀ M.S. (50 mg), CHCl3 (0.4 mL).
[b] Yield of isolated product. [c] Determined by H NMR spectroscopy.
[d] Determined by HPLC on a chiral stationary phase. The values in
parentheses are for the syn isomer. [e] PhCO2H was used as the additive
and 4 ꢀ M.S. was absent. [f] 5 mol% catalyst was used. [g] Methanol was
used as the solvent and 4 ꢀ M.S. was absent.
7 (0.2 mmol), aldehyde (0.4 mmol), 9a
1
Angew. Chem. Int. Ed. 2010, 49, 4656 –4660
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4657