Table 1. Effect of Additives, Chiral Ligands, Solvent, and Temperature on Zr(OtBu)4-2L* Catalyzed Aldol-Type Condensation of
Benzaldehyde 1 (a, R ) Ph) with Ethyl Diazoacetate
entry
additive (mol %)
liganda
solvent
temp (°C)
time (h)
yield (%)b
ee (%)c
1
2
3
4
5
6
7
8
CH3CN (20)
(S)-4
(S)-4
(S)-5
(S)-6
(S)-7
(S)-4
(S)-4
(S)-5
THF
THF
THF
THF
THF
DME
DME
DME
25
25
25
25
25
25
-23
-35
24
24
24
24
24
24
48
72
35
40
45
23
tr
58
62
65
53
65
58
0
CH3CN(20), H2O(20)
CH3CN(20), H2O(20)
CH3CN(20), H2O(20)
CH3CN(20), H2O(20)
H2O(20)
70
83
87
H2O(20)
H2O(20)
a 20 mol % of the catalyst Zr(OtBu)4-ligand (1:2.2) was applied in all cases. b Refer to the isolated yield after separation by silica gel. c ee’s were
determined by chiral HPLC; Chiracel OJ; hexane/2-propanol ) 96:4.
could catalyze the condensation, we used benzaldehyde as
the substrate and (S)-BINOL (S)-4 as the chiral ligand. After
examination of various metals,4 we were pleased to find that
Kobayashi’s Zr(IV) catalyst,5 prepared from Zr(OtBu)4 and
2.2 equiv of (S)-4,6 is effective for the condensation. With
this catalytic system and THF as solvent and CH3CN as
additive,7 the condensation gave (S)-â-hydroxyl R-diazo
carbonyl compound 3 (a, R ) Ph)8 in 35% yield and 53%
ee (Table 1, entry 1).
experiment is to test the effect of H2O additive in our
reaction. The enantioselectivity could be indeed increased
slightly by adding 20 mol % of H2O, but the yield is still
low (Table 1, entry 2).
The chiral ligands of (S)-BINOL derivatives were then
examined. The hydrogenated (S)-BINOL derivatives (S)-6
and (S)-7 were not efficient in promoting the reaction, while
(S)-BINOL and its (S)-6,6′-dibromo derivative (S)-5 gave
similar results (Table 1, entries 3-5). The catalytic system
with (S)-5 as the chiral ligand was slightly more reactive.
Next, the effect of solvent was studied. Nonpolar solvents,
such as Et2O, CH2Cl2 or PhCH3, were not suitable for this
reaction. Polar solvent CH3O(CH2)2OCH3 (DME), in com-
bination with 1 equiv of H2O as additive, was found to be
superior over THF to give aldol product in 58% yield and
70% ee (Table 1, entry 6). In this solvent, 20 mol % of H2O
is critical in activate the catalyst; however, an excess amount
of H2O will have the opposite effect to deactivate the catalyst.
The temperature of the reaction was found to significantly
affect the enantioselectivity (Table 1, entries 7 and 8). The
reaction at lower temperature could improve the enantio-
selectivity, but it took a longer time for the reaction to
complete. It appeared that at low temperature (S)-6,6′-
dibromo BINOL is the suitable ligand since it could make a
more reactive catalytic system. In any case, at temperatures
below -35 °C the reaction was found to be too slow to be
practically useful.
Since the initial experiments suggested that Zr(OtBu)4-
(S)-BINOL could give higher ee value, this catalytic system
was chosen, and we proceeded to optimize other conditions.
Because Kobayashi et al. reported that Zr(IV) catalysts could
be activated by adding small amount of H2O, the first
This optimized condition was then applied to other
aldehyde substrates (Table 2).9 For some aldehydes (entries
3, 4, 7, 8), we found that the reaction was very slow at -35
°C. We suspected that the product of the condensation might
compete with the substrates to bind to the transition metal
of the catalyst, thus deactivating the catalytic system. To
overcome this possible difficulty, we further added MgBr2
Soc., 2002, 124, 6798. (g) Evans, D.; Tedrow, J. S.; Shaw, J. T.; Downey,
C. W. J. Am. Chem. Soc. 2002, 392. (h) Kumagai, N.; Matsunaga, S.;
Kinoshita, T.; Harada, S.; Okada, S.; Sakamoto, S.; Yamaguchi, K.;
Shibasaki, M. J. Am. Chem. Soc. 2003, 125, 2169.
(3) Jiang, N.; Wang, J. Tetrahedron Lett. 2002, 43, 1285.
(4) We have examined B(III), Ti(IV), and Al(III). The catalyst prepared
from Ti(OiPr)4 and 1 equiv of (S)-4 gave the condensation product in 53%
yield and 15% ee.
(5) Ishitani, H.; Ueno, M.; Kobayashi, S. J. Am. Chem. Soc. 2000, 122,
8180.
(6) The Zr(IV) catalyst prepared with 1 equiv of (S)-4 gave lower
enantioselectivity.
(7) Various additives were examined, including N-methylimidazole. CH3-
CN gave the best results in terms of ee value.
(8) The absolute configuration was determined by converting 3a into a
known compound of ethyl 3-hydroxy-3-phenylpropionate and comparing
the sign of optical rotation. See Supporting Information.
(9) Typical Procedure. Chiral ligand (0.056 mmol) was dissolved in
0.5 mL of anhydrous DME, and then Zr(OtBu)4 (97%, 10 mg, 0.025 mmol)
was added to the solution under N2 at room temperature. After the mixture
stirred for 1 h, N2CHCO2Et (43 mg, 0.375 mmol) was added into the
solution, and then water (0.45 uL, 0.025 mmol) was added. After the solution
was stirred for another 3 h, it was then cooled by a dry ice/CCl4 bath (-23
°C) or dry ice/ClCH2CH2Cl bath (-35 °C). Aldehyde (0.125 mmol) was
added under N2. The solution was stirred for 3 days, and the solvent and
excess N2CHCO2Et were removed with rotovap. The crude residue was
purified with silica gel column (petroleum ether/acetone ) 8:1).
1528
Org. Lett., Vol. 5, No. 9, 2003