À
apparent that the flexibility of the C C bond associated
with the amino and hydroxyl groups in the ligand had
in terms of yield (68%) and ee value (88%) was demon-
strated (Table 3, entry 5). Catalyst loadings also had a signifi-
cant effect on the enantioselectivities; a 2.5 mol% loading
of catalyst gave the highest ee value (96%, Table 4, entry 2).
a strong influence on the coordination with CuACTHNUGTRNEUNG(OAc)2·H2O,
and consequently on the enantioselectivity of the reaction.
Ligands 1a–c were clearly superior to 2a–c in terms of ee
values, among which 1a distinguished itself as the best
ligand, although the product was obtained with only a mod-
erate ee value.
Table 4. The effect of the catalyst loading on the Henry reaction.[a]
In the subsequent studies, the reaction parameters, includ-
ing metal salts, solvents and catalyst loadings, were opti-
mized. From the data listed in Table 2, we noted that when
Entry
Cat. loading [%]
Yield [%][b]
ee [%][c]
1
2
3
4
5
6
1
2.5
5
10
15
20
24
42
57
68
64
66
95
96
94
88
90
90
Table 2. The effect of a metal salt on the Henry reaction.[a]
[a] All reactions were performed on a 1 mmol scale. [b] Yield of isolated
product. [c] The ee value was determined by chiral HPLC analysis.
Entry
Metal salt
Cu(OAc)2·H2O
CuCl2·H2O
CuSO4·5H2O
Yield
[%][b]
ee
[%][c]
1
2
3
4
5
6
7
8
9
G
63
25
25
22
20
96
77
7
67
37
15
11
0.4
4.3
1.7
2.7
–
With the optimized reaction conditions in hand, we began
to explore the generality of the method and thus examined
a variety of aromatic aldehydes (Table 5). To our delight, all
of the substrates used in this study (regardless of whether
the aromatic ring contained electron-withdrawing or elec-
tron-donating groups at the ortho, meta or para positions),
gave the corresponding R-enriched products in moderate to
good yield of isolated products (51–95%) with excellent
enantioselectivities in most cases (ꢀ90% ee). Further ex-
periments revealed that aromatic aldehydes functionalized
with electron-donating groups (Table 5, entries 28–38) fur-
nished the corresponding products with higher enantioselec-
tivities (ꢀ97% ee) than those with electron-withdrawing
groups. The only exception was 1-naphthaldehyde (91% ee,
Table 5, entry 31), probably due to its larger steric demand.
It is also worth noting that those substrates functionalized at
the 2,6-positions, for example, 2-chloro-6-fluorobenzalde-
Cu
Cu(OTf)2
Zn(OTf)2
La(OTf)3
In(OTf)3
Bi(OTf)3
(NO3)2·H2O
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
N
ND
[a] All reactions were performed on a 1 mmol scale. [b] Yield of isolated
product. [c] The ee value was determined by chiral HPLC analysis.
other copper salts were used instead of copper acetate, the
yields and ee values decreased dramatically (Table 2, en-
tries 2–5). The most likely reason for this is that the moder-
ately charged acetate anion facilitates the deprotonation of
nitromethane as a prelude to the aldol-addition process.[10]
Other metal triflates were also tested; however, the reac-
tions proceeded with very poor enantioselectivities (Table 2,
entries 6–9). Further studies showed that the reaction was
also highly sensitive to the nature of solvent employed
(Table 3); diethyl ether was found to be the superior solvent
hyde
(Table 5,
entry 15),
2,6-difluorobenzaldehyde
(entry 16), 2,4,6-trifluorobenzaldehyde (entry 17), 2,6-di-
chlorobenzaldehyde (entry 22), 2,4,6-trimethylbenzaldehyde
(entry 37), and 2,6-dimethoxybenzaldehyde (entry 38), gave
the corresponding products with enantioselectivities higher
than 97% ee.
Table 3. The effect of the solvent on the Henry reaction.[a]
Although there is a lack of structural information with re-
gards to the catalytically active species, we have proposed
a simplified distorted tetragonal-pyrimidal copper species
based on the known model.[10] Figure 1 shows the optimized
Entry
Solvent
Yield
[%][b]
ee
[%][c]
À
structures of the transition states and products for the C C
bond-forming step; A1 leads to the formation of the R-con-
figured alcohol A2, whereas B1 results in the formation of
the S-configured product B2, which have energy barriers of
20.06 and 23.02 kcalmolÀ1, respectively. This proposal is in
good agreement with the experimental observation that the
reaction could be conducted under ambient conditions, and
it also indicates that the pathway to the R-configured prod-
uct is kinetically more favorable. Detailed inspection re-
vealed that A1, in comparison with B1, was stabilized by the
1
2
3
4
5
6
7
8
MeOH
iPrOH
CH2Cl2
THF
50
54
19
44
68
48
13
40
62
81
73
88
88
77
75
82
Et2O
hexane
toulene
CH3CN
[a] All reactions were performed on a 1 mmol scale. [b] Yield of isolated
product. [c] The ee value was determined by chiral HPLC analysis.
&
2
&
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!