Angewandte
Chemie
a functionalization with nitroalkenes.[11] It should also be
pointed out that there is only one literature precedent dealing
with an enantioselective intermolecular [2+2] cycloaddition
between enals and electron-rich alkenes which proceeds
under iminium catalysis,[12] and the dienamine route pre-
sented herein is still an unexplored approach[13] which is also
a complementary methodology to reported example, as it
allows the [2+2] cyclocondensation of enals with an electron-
poor alkene. In addition, this new process leads to the
formation of cyclobutanes that are not obviously accessible
using other methods.
at a lower temperature led to 3a with an excellent ee value,
but in much lower yield (entry 6). With the aim of improving
this latter parameter, we next considered the possibility of
a dual activation approach for both the aldehyde and the
nitroalkene by using the combination of the amine catalyst 4a
and the achiral thiourea 5a, which was expected to engage in
selective hydrogen bonding with the nitroalkene.[14] When we
used this catalyst combination, the yield increased to 70%
without affecting the enantioselectivity (entry 7), and an even
better result was obtained when thiourea 5b was employed
(entry 13).
We started our study by surveying the best reaction
conditions for the transformation using the reaction of the
enal 1a with nitroalkene 2a as a model system (Table 1). We
Having established the best protocol for the reaction, we
next surveyed the scope of this transformation with respect to
the a,b-unsaturated aldehydes and nitroolefins. As summar-
ized in Table 2, the reaction behaved well in all cases tested
Table 1: Screening for the best experimental conditions using the
reaction of aldehyde 1a and nitroalkene 2a as a model system.[a]
Table 2: Scope of the reaction.[a]
Entry
3
R1
R2
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
7
8
3a
3b
3c
3d
3e
3 f
3g
3h
3i
Ph
Ph
Ph
Ph
Ph
86
88
77
73
72
91
91
90
67
69
52
38
91
92
92
89
95
94
94
94
92
94
94
85
4-MeC6H4
4-FC6H4
2-thienyl
Ph
4-MeC6H4
4-FC6H4
4-MeOC6H4
Ph
Entry
4
Co-catalyst
Solvent
T [8C]
Yield [%][b]
ee [%][c]
4-MeOC6H4
4-MeOC6H4
4-MeOC6H4
4-MeOC6H4
4-ClC6H4
4-ClC6H4
2-thienyl
Ph
1
2
3
4
5
6
7
8
4a
4b
4a
4a
4a
4a
4a
4a
PhCO2H
PhCO2H
PhCO2H
PhCO2H
PhCO2H
PhCO2H
5a
CHCl3
CHCl3
toluene
THF
4
4
4
4
4
À20
À20
À20
59
0
56
<5
<5
47
74
–
86
n.d.
n.d.
90
91
91
9
10
11
12
3j
3k
3l
4-MeC6H4
Ph
Et
EtOH
toluene
toluene
toluene
70
86
5b
[a] All reactions carried out on 0.35 mmol scale of 1a. [b] Yield of pure
product as a 1:1 mixture of a and b anomers (NMR analysis) after
column chromatography. [c] Determined by HPLC analysis of the
corresponding lactone.
[a] Reaction conditions: 1a (0.35 mmol), 2a (0.52 mmol), catalyst 4
(20%), and co-catalyst (20%) were stirred in the specified solvent and
temperature for 72 h. [b] Yield of pure product as a 1:1 mixture of a and
b anomers after column chromatography. [c] Determined by HPLC
analysis of the corresponding lactone. n.d.=not determined, THF=
tetrahydrofuran.
when g-aryl-substituted a,b-unsaturated aldehydes were
employed, regardless of the electronic nature of the g sub-
stituent. In this sense, both the electron-rich (1b) and
electron-poor (1c) aryl-substituted enals reacted efficiently
with a-hydroxymethylnitrostyrene (2a) to furnish the corre-
sponding bicyclic adducts 3b and 3c, respectively, in good
yields and excellent enantioselectivities (entries 2 and 3). The
g-heteroaryl-substituted enal 1d also performed well in the
reaction with 2a, thus delivering 3d in good yield and
enantioselectivity (entry 4). Regarding the nitrostyrene
reagent we evaluated the use of the 2a and other derivatives
incorporating electron-donating or electron-withdrawing sub-
stituents at the aryl moiety, and, in all cases the reaction also
proceeded with similar levels of chemical efficiency and
stereocontrol (entries 5–11). We also evaluated the use of a b-
alkyl-substituted a,b-unsaturated aldehyde such as 2-hexenal
(1 f), which led to the formation of the desired cyclobutane
adduct with a somewhat lower yield, but still with good levels
of stereocontrol (entry 12). This final experiment also points
toward the ability of the g-aryl substituent on the enal
initially tested the performance of catalyst 4a under standard
reaction conditions, which involve the use of benzoic acid as
Brønsted acid co-catalyst in CHCl3 at 48C (entry 1). Under
these reaction conditions, the adduct 3a was isolated in
moderate yield with a promising 74% ee. Importantly, NMR
analysis of the crude reaction mixture indicated that 3a had
been formed as a 1:1 mixture of a and b anomers. No other
diastereoisomers could be detected, which indicated that the
formal [2+2] cycloaddition process also had taken place with
complete diastereocontrol. Increasing the steric bulk at the
aryl moieties of the diarylprolinol catalyst (entry 2) did not
result in any significant improvement. We next evaluated the
influence of the solvent by adopting 4a as the most efficient
catalyst, and observed that running the reaction in toluene led
to a slight increase in the enantioselectivity (entry 3), whereas
other more polar solvents were incompatible with this
transformation (entries 4 and 5). We also found that working
Angew. Chem. Int. Ed. 2012, 51, 4104 –4107
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4105