of the ligands. In some cases, the chemoselectivity could
almost be completely shifted in favor of the reduction product
5b (Table 1, entries 6, 9, 10, and 11). Promising results were
observed in the case of ligands JOSIPHOS L2, WALPHOS
L4 ,and L5. Good enantioselectivities were reached in the
case of the syn isomer (Table 1, entries 2, 4, and 5). We
subsequently tested TANIAPHOS ligands L7-L9, which
contain a 1,5-diphosphane unit and hence are capable of
forming an eight-membered chelate ring with the metal. To
our delight, L7 provided a remarkable improvement in the
chemoselectivity in favor of 4b (99:1), with a diastereose-
lectivity (dr 77:23) (Table 1, entries 7 and 8) similar to that
obtained with the L2 or L4 ligands, in favor of the syn adduct
(Table 1, entries 2 and 4). However, the enantiodifferentiation
was drastically enhanced for both isomers of 4b. Under these
conditions, the reaction with L7 furnished adduct 4b with
95% ee and 74% ee for the syn and anti isomers, respectively
(Table 1, entry 7). The catalyst loading can be decreased to
0.1 mol % without any variation in chemo-, diastereo-, and
enantioselectivities (Table 1, entries 1 and 7). Interestingly,
TANIAPHOS L8 and L9 showed poor results in the
reductive aldol reaction compared to L7 (Table 1, entries 9
and 10).
Next, we studied the scope of the copper-catalyzed
asymmetric reductive aldol reaction with respect to the
aldehyde substrates and the TANIAPHOS chiral ligand L7,
under optimal conditions. A variety of aliphatic, aromatic,
or heteroaromatic aldehydes were tested at -78 °C in THF.
However, the low solubility of aromatic substrates at low
temperature in THF or toluene forced us to select a
compromise in which in toluene, at -50 °C, provided the
best results.11 Remarkably, the selectivity of the domino
process did not change when THF was replaced by toluene,
and we observed that all substrates participate successfully
in the reaction (conversion >99%). The chemoselectivity
remains excellent (generally >95:5) with good enantiose-
lectivities but moderate diastereoselectivities (Table 2). The
isolated yields for the corresponding adducts after chromato-
graphic purification were all in the range of 74-99%. For
acyclic aliphatic aldehydes, good chemoselectivities and
moderate diastereoselectivities were observed (entries 1 and
2) and some enantioselectivity was detected in the case of
isobutyraldehyde (eesyn ) 73%) (entry 1).
Table 2. Asymmetric Copper-Catalyzed Reductive Aldol
Reaction between 2 and Various Aldehydes 1 in the Presence of
(R,S)-L7 under the Optimal Conditionsa
conversion
(%)b
eesyn (eeanti
)
entry
R
4:5
syn:anti
(%)c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
i-Pr
t-Bu
Cy
99
99
99
99
99
99
99
99
99
99
99
99
99
94
100:0
77:23
100:0
99:1
100:0
95:5
97:3
74:26
95:5
95:5
95:5
95:5
99:1
64:36
76:24
57:43
77:23
88:12
41:58
44:56
47:53
44:56
60:40
41:59
67:33
51:49
67:33
73 (26)
0 (0)
86 (70)
96 (74)
97 (30)
nd (72)
86 (76)
84 (65)
85 (69)
68 (72)
58 (78)
83 (nd)
86 (76)
34 (44)
Cyd
Cye
C6H5
p-FC6H4
p-CF3C6H4
p-ClC6H4
p-MeOC6H4
o-MeOC6H4
2-thienyld-f
3-thienyl
2-pyridyld
82:18
a All reactions were carried out in toluene (0.25 M) at -50 °C under an
oxygen-free argon atmosphere containing 1 (1.0 equiv), 2 (1.2 equiv), 3
(1 mol %), L7 (1 mol %), and PhSiH3 (1.4 equiv) unless otherwise stated.
b Determined by chiral GC analysis CHIRALSIL-DEX CB (25 m, 0.25 mm,
25 µm). c Configuration determined by comparison with known products.
d At -78 °C. e Ph2SiH2 (1.4 equiv) instead of PhSiH3. f In THF.
good to excellent but the diastereoselectivity remained
moderate, favoring either the syn or the anti isomers (Table
2, entries 6 to 14). In all cases, good to excellent enantiomeric
excesses were obtained (up to 86% for the syn isomer) at
-50 °C. As a general trend, the introduction of a halogen
substituent at the para position (entries 7 to 9) did not change
the selectivity, whereas the replacement of an electron-
withdrawing group, at the para position of benzaldehyde,
by an electron-donating group increased the diastereoselec-
tivity in favor of the syn isomer. Unfortunately, the enan-
tiomeric excess on the syn isomer was slightly decreased
(Table 2, entry 10).
Nevertheless, the domino process was more efficient when
aromatic and heteroaromatic aldehydes were employed. For
the range of substrates studied, the chemoselectivity was
Heteroaromatic aldehydes, such as 2- and 3-thiophene-
substituted aldehydes, also took part efficiently in the domino
sequence to give the syn-4 adducts with rather good
enantioselectivities (Table 2, entries 12 and 13).
(11) General procedure for catalytic reductive aldol reaction: A 10 mL
flame-dried round-bottomed flask, equipped with a magnetic stirrer, was
charged with CuF(PPh3)3‚2MeOH (9.0 mg, 0.01 mmol), ligand (0.01 mmol),
and toluene (4.8 mL). The catalyst solution was stirred for 30 min at room
temperature and phenylsilane (180 µL, 1.40 mmol) was added at the same
temperature. After the solution was cooled at -50 °C, methyl acrylate
(110 µL, 1.20 mmol) and the corresponding aldehyde (1.00 mmol) were
simultaneously added to the solution. The mixture was stirred for 1 h at
-50 °C under argon. Conversion, dr and ee were followed by gas
chromatography (aliquots were hydrolyzed by 1 mL of aqueous NH4F
solution and filtered through a plug of silica). The reaction mixture was
quenched by adding aqueous NH4F solution (5 mL). The aqueous layer
was extracted by diethyl ether (3 × 5 mL). Then, the combined organic
layers were washed with brine (20 mL), dried over anhydrous MgSO4,
filtered, and concentrated under reduced pressure. The crude product was
purified by flash chromatography to yield the corresponding adduct.
We have also investigated the dependence of the structure
of the silane on the domino process. Various silanes were
tested, such as (Me3SiO)2MeSiH, Me2EtOSiH, (Me2SiH)2O,
or PMHS, in the reductive asymmetric aldol reaction process,
with carboxylaldehyde as the electrophile. Unfortunately,
only Ph2SiH2 gave excellent results as the diastereomeric
ratio and the enantiomeric excesses were both improved at
-50 °C in toluene (syn:anti 88:12, eesyn ) 97%) (Table 2,
entries 3 and 4 versus 5). Alas, Ph2SiH2 did not lead to
significantly improved selectivities with the other aldehydes
used.
Org. Lett., Vol. 8, No. 26, 2006
5945