Table 1. Ligand evaluation in the Henry reaction.[a]
Table 2. Aldehyde scope of the CuACTHNGUTERNNUG
(OAc)2/4d system.[a]
ee [%][b]
Entry
Ligand
Yield [%]
ee [%][b]
Entry
R
Product
t [h]
Yield [%]
1
2
3
4
5
3
9
85
14
45
99
35
95
65
84
97
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
CH
CH
CH
G
7a
7b
7c
7d
7e
7 f
7g
7h
7i
7j
7k
7l
7m
7n
7o
7p
7q
7r
48
48
48
48
48
48
48
24
30
86
24
96
48
60
60
60
48
60
99
99
99
91
89
98
99
99
74
99
99
99
76
87
99
88
81
95
97 (R)
97 (R)
97 (R)
99 (R)
98 (R)
98 (R)
98 (R)
95 (R)
94 (R)
96 (R)
92 (R)
95 (R)
97 (R)
96 (R)
96 (R)
96 (R)
96 (R)
97 (R)
4a
4b
4c
4d
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
[a] Reactions were carried out with hexanal (0.5 mmol), CH3NO2
(5.0 mmol), Cu(OAc)2·H2O (2.5 mol%), and ligand (2.5 mol%) in etha-
nol (1.0 mL) at room temperature for 48 h. [b] Enantiomeric excesses
were determined by HPLC on a Chiral OD-H column.
ACHTUNGTRENNUNG
PhCH=CH
Ph
4-NO2C6H4
2-MeOC6H4
4-PhC6H4
4-FC6H4
4-ClC6H4
4-BrC6H4
2-naphthyl
2-furyl
loading of CuACHTUNGTRENNUNG(OAc)2·H2O (2.5 mol%) and of the ligand (2.5
mol%) in ethanol. The ligand screening results showed that
the reactivity and enantioselectivity are dependent upon
both the chiral diamine fragment and the substituents of the
side chain which were derived from the a-amino alcohol
(R2). In comparison with 3, 4a provided a better yield and
higher enantioselectivity. Replacing the R2 substituents with
methyl or benzyl (Bn) groups did not improve the enantio-
selectivity. Compared with ligand 4a, with a cyclohexyl frag-
ment, ligand 4d, containing a 1,2-diphenylethylenediamine
backbone, gave a better enantioselectivity (97% ee). More-
over, since all diamine ligands only provide racemic prod-
ucts (see the Supporting Information), this proves the validi-
ty of the bis(sulfonamide)–diamine ligand design. The bis-
(sulfonamide)–diamine ligands have significant advantages
over the traditional chiral diamines and sulfonamides used
for the Henry reaction.
[a] Reactions were carried out with aldehyde (0.5 mmol), CH3NO2
(5.0 mmol), Cu(OAc)2·H2O (2.5 mol%), and ligand 4d (2.5 mol%) in
ethanol (1.0 mL) at room temperature. [b] Enantiomeric excesses were
determined by HPLC and the absolute configurations (R) were estab-
lished by comparison to literature data.
Scheme 3. An application of a nitroaldol adduct.
Under the optimized conditions, the aliphatic aldehyde
substrate scope of the Henry reaction was explored
(Table 2). To our surprise, the reactions proceeded very
well; excellent yields accompanied with excellent enantiose-
lectivities were achieved in most cases. It is notable that nei-
ther the length of the carbon chain, nor its steric bulk has
an obvious effect on the enantioselectivity. Importantly, a-
branched aliphatic aldehydes, such as isobutyraldehyde, piv-
alaldehyde, and cyclohexanecarboxaldehyde, resulted in the
formation of the b-nitroalcohol with excellent enantioselec-
tivities of up to 99% ee (Table 1, entries 4–7). Furthermore,
the optimized catalyst was also used for the Henry reaction
of aromatic aldehydes (Table 2, entries 9–18). In general, all
of the investigated aromatic aldehydes provided the corre-
sponding adducts in excellent enantioselectivities. The aro-
matic substrates containing electron-donating groups tended
to afford better results than those with electron-withdrawing
groups (Table 2, entries 11–16). Even for the bulkier 2-naph-
thaldehyde and 2-furylaldehyde, high enantioselectivities
still remained (Table 2, entries 17–18).
generate the corresponding aziridine 9 without any loss of
stereochemical information. Aziridines constitute the funda-
mental synthetic tool for the generation of some N-contain-
ing natural products and also serve as chiral building blocks,
auxiliaries, and ligands in organic synthesis.[14] They are also
a key intermediate in the synthesis of our bis(sulfonamide)–
diamine ligand.
The bis(sulfonamide)–diamine ligands were also prelimi-
narily examined in Henry reactions in which other nitroal-
kanes were used. We found that the catalyst system of
ligand 4a with CuBr was highly effective for the Henry reac-
tion between 1-nitropropane and cyclohexanecarboxalde-
hyde (Scheme 4). Excellent stereocontrol and the apparently
favorable formation of the syn diastereomer were observed
in this reaction. The ratio of the syn to anti isomers was de-
termined to be 97:3. The syn diastereoisomer was produced
To further demonstrate the applicability of the new bis-
(sulfonamide)–diamine ligand, the transformation of the ni-
troaldol adduct 7d was performed as shown in Scheme 3. In-
itially, 7d was reduced to furnish (R)-(À)-3-methyl-1-amino-
butan-2-ol (8).[13] Then, 8 was cyclized by sulfonylation to
Scheme 4. A Henry reaction of 1-nitropropane.
8260
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 8259 – 8261