It appeared that the 1,2-aminoalcohol moiety of the cincho-
nine 1 plays a crucial role in the enantiocontrol of this reac-
tion and acts as a bifunctional organocatalyst. Surprisingly,
even the use of commercially available ephedrine (4) result-
ed in the formation of bicycle 9a with moderate enantiose-
lectivity (Table 1, entry 4). These results again emphasize
the importance of the bifunctionality of the catalyst used in
the asymmetric version of this new domino Michael–Henry
reaction.
cally favored diastereomer is produced (Table 2, entry 5).
From these experiments it also becomes apparent that the
reaction time has an impact on the diastereomeric ratio; the
longer the reaction time the lower the diastereomeric ratio
(Table 1, entry 6 and Table 2, entry 1). This indicates a base-
catalyzed isomerization.
Table 2. Influence of catalyst loading, 1,2-dione amount and temperature
on the enantioselective domino Michael–Henry reaction.
Table 1. Evaluation of different bifunctional chiral organocatalysts.
Entry[a] mol% cat. T [8C] 7 [equiv] Yield [%][b] d.r.[c] ee [%][d]
1[e]
2
3
4
5
10
1
1
0.5
1
RT
RT
RT
RT
50
1.2
1.2
1.0
1.0
1.0
94
84
58
44
63
1.7:1 93, 94
1:3.3 92, 94
1:5.5 93, 98
1:7.7 91, 98
1.2:1 91, 98
Entry[a]
Catalyst
t [h]
Yield [%][b]
d.r.[c]
ee [%][d]
1
2
3
4
5
6
1
2
3
4
5
6
3
1.5
24
6
<1
<1
85
86
80
82
92
91
3:1
9:1
8.3:1
5:1
2:1
3:1
45, 24
rac/rac
rac/rac
À36, À30
À83, À89
87, 92
[a] Reaction conditions: 0.2m toluene solution, 1,2-cyclohexadione 7, 8a
(1.0 equiv), catalyst 6. All reactions were stopped after 24 h.[b] Yield of
isolated product after column chromatography. [c] Determined by
1H NMR spectroscopy. [d] Enantiomeric excess was determined by
HPLC.[e] This reaction was completed in less than 15 min.
[a] Reaction conditions: 0.2m toluene solution, 1,2-cyclohexadione
7
(2.0 equiv), 8a (1.0 equiv), 10 mol% catalyst at RT. [b] Yield of isolated
product after column chromatography. [c] Determined by 1H NMR spec-
troscopy. [d] Enantiomeric excess was determined by HPLC.
Having established the optimal reaction conditions, we
decided to explore the scope of this newly developed asym-
metric domino Michael–Henry reaction yielding bicycles
with four stereocenters in a one-pot reaction sequence
(Table 3). In general, various b-nitroalkenes with different
substitution patterns, having both electron-donating
(Table 3, entry 2–5) and electron-withdrawing groups
(Table 3, entry 6–9), can be used in this transformation to
obtain the corresponding products in good yields and with
excellent enantioselectivities throughout. Additionally, for
the first time it was possible to use b-disubstituted nitro-
With this knowledge in hand, we substituted the alcohol
group of cinchonine 1 by a thiourea moiety to obtain cata-
lyst 5[8] with enhanced hydrogen-bond-donor capacity.[9,10]
Application of this catalyst 5 resulted in the opposite enan-
tiomer with better enantioselectivities (Table 1, entry 5). In
addition to the impact the thiourea residue has on enantio-
selectivity, the results show that the absolute configuration
of the cinchonine derivative in the 9-position determines the
stereochemical outcome of the reaction (Figure 1 and
Table 1). Underlining this concept, the cinchonidine-based
thiourea catalyst 6, the pseudoenantiomer of 5, affords the
opposite enantiomer with excellent enantioselectivities
(Table 1, entry 6).
For further optimization, the substrate concentration, as
well as catalyst loading, was varied (Table 2). These experi-
ments revealed that the best results with regard to reactivity
and selectivity were obtained with 1.0 mol% catalyst
(Table 2, entry 2). In varying substrate concentration, the se-
lectivity can be improved by decreasing the amount of 7 to
1.2 equivalents (Table 2, entry 1). The lower substrate con-
centration of the nucleophile 7 causes statistically less
Brønsted base activation and thus, an increase of bifunction-
al substrate activation is observed, which is fundamental for
obtaining high enantioselectivities. Further reduction of the
catalyst loading to 0.5 mol% also resulted in excellent enan-
tioselectivities; however, lower yields were observed and the
diastereomeric ratio was reversed. The latter observation in-
dicates a base-catalyzed epimerization at the a-nitro posi-
tion of the product 9a. This is also observed if the reaction
is performed at 508C where again more the thermodynami-
AHCTUNGTREGsNNUN tyrenes, which lead to the bicycles 9j and 9k with two
neighboring quaternary stereocenters (Table 3, entry 10 and
11). Interestingly, the products 9j and 9k could be isolated
almost as a single diastereomer and with excellent enantio-
selectivities.
The molecular structure of 9k was determined by X-ray
crystal-structure analysis and revealed that the kinetically
favored product is the major diastereoisomer in which the
aryl moiety and the nitro group have a syn-configuration
(Figure 2).
Mechanistically we propose that the Michael–Henry
domino reaction is catalyzed by a bifunctional thiourea cin-
chonidine involving a hydrogen-bond and base-catalyzed ac-
tivation of the nitrostyrene and 1,2-dione, respectively. With
regard to the base-induced epimerization of the products
our experimental results indicate that two possible pathways
exist, a slow and a fast one (Scheme 1).
The fast pathway is a deprotonation–reprotonation of the
a-nitrocarbon atom due to the lower basicity as compared
to the alcohol, yielding the thermodynamically favored dia-
stereoisomer.
4174
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 4173 – 4176