Enantioselective addition of diethylzinc to aldehydes catalyzed by chiral amino alcohols. Substituent effect and nonlinear effect

Article abstract of DOI:10.1016/S0040-4020(01)00423-9

A new series of chiral β-amino alcohols derived from (S)-leucine, valine, and phenylalanine have been synthesized and evaluated as chiral catalysts for the enantioselective addition of diethylzinc to aldehydes. The β-amino alcohol (1c) possessing a isobutyl substituent and two phenethyl substituents on the carbinol carbon atom was found to be an efficient and optimal ligand to catalyze the diethylzinc addition with high enantioselectivity (up to 97% ee) and good yield. Furthermore, a strong (+)-nonlinear effect (asymmetric amplification) was observed in the enantioselective catalysis, providing high level of enantioselection (93% ee) by use of ligand 1c in 20% ee.

Full text of DOI:10.1016/S0040-4020(01)00423-9

TETRAHEDRON
Pergamon
Tetrahedron 57 '2001) 4825±4829
Enantioselective addition of diethylzinc to aldehydes catalyzed by
chiral amino alcohols. Substituent effect and nonlinear effect
Takahiko Ohga,a Satoshi Umedab and Yasuhiro Kawanamia,p
aDepartment of Biochemistry and Food Science, Faculty of Agriculture, Kagawa University, Miki-cho, Kagawa 761-0795, Japan
bDepartment of Chemistry, Faculty of Education, Kagawa University, Takamatsu, Kagawa 760-8522, Japan
Received 7 March 2001;accepted 12 April 2001
AbstractÐA new series of chiral b-amino alcohols derived from 'S)-leucine, valine, and phenylalanine have been synthesized and
evaluated as chiral catalysts for the enantioselective addition of diethylzinc to aldehydes. The b-amino alcohol '1c) possessing a isobutyl
substituent and two phenethyl substituents on the carbinol carbon atom was found to be an ef®cient and optimal ligand to catalyze the
diethylzinc addition with high enantioselectivity 'up to 97% ee) and good yield. Furthermore, a strong '1)-nonlinear effect 'asymmetric
ampli®cation) was observed in the enantioselective catalysis, providing high level of enantioselection '93% ee) by use of ligand 1c in
20% ee. q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
nonlinear effect in the catalytic asymmetric induction
'Scheme 1).
In order to synthesize optically active secondary alcohols,
the enantioselective additions of dialkylzincs to aldehydes
in the presence of catalytic amounts of chiral ligand have
been extensively investigated and chiral b-amino alcohols
have been shown to be ef®cient chiral ligands.1 Among
them 3-exo-'dimethylamino)isoborneol2 and 'S)-prolinol
derivatives3 have been reported to catalyze the diethylzinc
addition to aromatic aldehydes with excellent enantio-
selectivity. For the addition to aliphatic aldehydes only
proceeds with moderate enantioselectivity. However,
N,N-di-n-butylnorephedrine,4 2-piperidino-1,1,2-triphenyl-
ethanol,5 and N-phenyl¯uorenyl b-amino alcohols6 have
been reported to enhance the enantioselectivity for the
addition to aliphatic aldehydes. Recently, we also reported
that b-amino alcohols derived from 'S)-leucine and bearing
two ¯exible and larger phenethyl groups, compared to a
phenyl and n-butyl group, afforded excellent enantio-
selection7 and revealed that the substituent 'R2) on the car-
binol carbon atom of b-amino alcohols might play an
important role in controlling the enantioselection via a
®ve-membered ethylzinc aminoalkoxide complex as
shown in Fig. 1. Therefore, we anticipated that further
systematic ligand modi®cations could lead to realizing the
stereoselection and developing a rational design of chiral
ligands. In this paper, we report the full detail of the
in¯uence of the substituent of a new series b-amino alcohols
derived from 'S)-leucine, valine, and phenylalanine and the
2. Results and discussion
b-Amino alcohols 1a±d, 2a±c, and 3a±c were readily
Figure 1.
Keywords: catalysis;addition reaction;asymmetric induction;amino alco-
hols;nonlinear effect.
p
Corresponding author. Tel.: 181-87-832-1601;fax: 181-87-832-1417;
e-mail: kawanami@ag.kagawa-u.ac.jp
Scheme 1.
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020'01)00423-9

4826
T. Ohga et al. / Tetrahedron 5712001) 4825±4829
Scheme 2. 'a) R2MgBr, THF '28±47%);'b) I'CH 2)5I, K2CO3, CH3CN,
re¯ux '49±74%).
prepared via a two-step sequence from commercially avail-
able 'S)-leucine ethyl ester hydrochloride, 'S)-valine methyl
ester hydrochloride, and 'S)-phenylalanine methyl ester
hydrochloride, respectively, according to the previously
described procedure.7 Addition of different carbon chain
of Grignard reagents to the amino acid esters followed by
alkylation with 1,5-diiodopentane and potassium carbonate
afforded the corresponding b-amino alcohols without a loss
of enantiopurity as shown in Scheme 2.
Figure 2. Substituent effect 'R2).
the ligand 2a±c derived from 'S)-valine 'entries 9±11) and
the ligand 3a±c derived from 'S)-phenylalanine 'entries
13±15) showed the similar substitution effect 'Fig. 3).
Thus, the chiral ligand 1c with two ¯exible phenethyl sub-
stituent was found to be the most ef®cient and optimal
ligand, giving 'R)-1-phenyl-1-propanol with 97% ee and
'R)-3-nonanol with 88% ee. These results suggest that the
in¯uence of amino acid residue 'R1) is negligible and the
¯exibility of the substituent 'R2) at the carbinol carbon atom
could play a critical role in the enantioselection of the
addition reaction.
First, we examined the enantioselective addition of diethyl-
zinc to benzaldehyde and n-hexylaldehyde to heptanal, which
were used as a model reaction, in the presence of 10 mol% of
the chiral amino alcohol ligands 1a±d, 2a±c, and 3a±c in
hexane at 08C. These results are summarized in Table 1.
Comparing the amino alcohol ligands derived from 'S)-
leucine, ligand 1c with a phenethyl substituent 'R2) was
the optimal ligand to provide higher enantioselectivity
than that of ligands 1a, 1b, and 1d with a phenyl, benzyl,
and longer 3-phenylpropyl group, respectively 'entries
1±4). The similar tendencies were observed in the addition
to n-hexylaldehyde and ligand 1c showed the highest enan-
tioselectivity 'entries 5±8) as shown in Fig. 2. Furthermore,
Since Kagan et al. reported nonlinear effect in asymmetric
synthesis,8 it has been established that nonlinear effect
provides useful information on the reaction mechanism
such as the structure and aggregation of active species
involved in asymmetric reaction, especially in enantioselec-
tive catalysis.9 A strong '1)-nonlinear effect 'asymmetric
ampli®cation) as observed in the addition to aldehydes cata-
lyzed by DAIB10 and 1-piperidino-3,3-dimethyl-2-butanol
'PDB)11 is particularly valuable for generating products
Table 1. Enantioselective addition of diethylzinc to aldehydes using
various ligands 1a±d, 2a±c, and 3a±c
Entry
Ligand
RCHO
Ph
Yield '%)
Ee '%)a
Con®g.b
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1a
1b
1c
1d
2a
2b
2c
2c
3a
3b
3c
3c
88
68
90
56
76
66
81
58
63
59
86
68
64
77
69
64
85
84
97
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
R
92
n-C6H13
78c
84c
88c
76c
82
9
Ph
10
11
12
13
14
15
16
82
97
n-C6H13
Ph
86c
82
67
96
n-C6H13
84c
a
Determined by HPLC analysis using DAICEL Chiralcel OB 'hexane/
i-PrOH 98:2).
The absolute con®gurations of the resulting alcohols were determined to
be 'R) by comparison of the speci®c rotations 'Ref. 3).
Determined by HPLC analysis using DAICEL Chiralcel OD 'hexane/
i-PrOH 800:1) after benzoylation.
b
c
Figure 3. Substituent effect 'R1).

T. Ohga et al. / Tetrahedron 5712001) 4825±4829
4827
mixing with 'R)-1c as shown in Fig. 4, for example, pro-
viding high level of enantioselection '93% ee) by use of
ligand 1c in 20% ee. This phenomenon could be attributed
to the stability of heterochiral 'R)´'S)-dimer as in the
mechanism proposed by Noyori et al.12 Also discussion of
the enantioselections are based on the carbinol stereocenter
of DAIB and PDB, but ligand 1c has only one chiral center
at amino acid residue as shown in Fig. 5. These facts, there-
fore, suggest that the steric interaction between the isobutyl
group on the chiral center and one of two phenethyl groups
should render the achiral carbinol atom of ligand 1c confor-
mationally chiral center.
Although the actual active species are unclear, the addition
reaction catalyzed by chiral amino alcohol ligand 1c might
proceed through the similar transition state model as
proposed by Noyori et al. 'Fig. 6).12 This model is partially
supported by MM2 minimized conformation of ligand 1c
itself, which clearly shows that one of the phenethyl groups
locates in a pseudo-axial position as shown in Fig. 7. There-
fore, the aldehyde might be attacked on its re-face at the
upper side of the dinuclear Zn complex to produce the corre-
sponding 'R)-secondary alcohol. It is noteworthy that
conformationally ¯exible moiety could serve to relay and
amplify the stereochemical information of the relatively
remote stereogenic center via the chiral relay system, ensur-
ing ef®cient stereocontrol of approaching aldehydes and
diethylzinc.
Figure 4. Nonlinear effect.
Figure 5.
with high ee from ligands with low ee. Therefore, we
investigated the nonlinear effect in this enantioselective
catalysis next. The similar asymmetric ampli®cation,
'1)-nonlinear effect, was observed in the addition using
enatiomerically impure ligands 1c that was prepared via
3. Conclusion
Systematic study of a new series of chiral b-amino alcohols
derived from 'S)-leucine, valine, and phenylalanine showed
that ligand 1c possessing two phenethyl substituents was an
ef®cient and optimal ligand to catalyze the enantioselective
addition of diethylzinc to aromatic and aliphatic aldehydes
with high enantioselectivity. Furthermore, '1)-nonlinear
effect, asymmetric ampli®cation, was observed in the
addition catalyzed by ligand 1c. These results suggest that
the achiral carbinol atom with two phenethyl substituents
should be conformationally restricted by the proximity of
the chiral amino acid residue and this chiral relay system
could be a factor in the ef®ciency of this ligand. Thus, this
ligand design bearing chiral relay system would lead to
developing further ef®cient chiral ligands.
Figure 6. Transition state model with chiral relay system.
Figure 7. MM2 minimized conformation of ligand 1c.

4828
T. Ohga et al. / Tetrahedron 5712001) 4825±4829
4. Experimental
0.96 'd, 3H, Jˆ6.4 Hz), 1.11±2.03 'm, 9H), 2.14±2.98
'm, 9H), 5.15 's, 1H), 6.90±7.67 'm, 10H); 13C NMR
'CDCl3) d 138.3, 138.1, 130.9, 130.6, 128.2, 127.6, 127.3,
125.7, 74.7, 65.9, 43.7, 41.9, 36.0, 26.3, 24.6, 20.8. EI
HRMS C25H35NO 'M1) 365.2721. Found 365.2725.
4.1. Data for compounds
IR spectra were determined using a Shimadzu IR-435 spec-
trophotometer. 1H NMR and 13C NMR spectra were
recorded at 90 and 23 MHz using a JEOL JNM-EX90
spectrometer, respectively. All NMR spectra were taken in
CDCl3 solution with TMS as the internal standard. Mass
spectra were recorded on a JEOL JMS-SX102A mass
spectrometer. Optical rotations were determined on a
Yanagimoto OR-50 polarimeter. The HPLC analysis was
carried out using a DAICEL Chiralcel OB or OD column
'0.46£25 cm2) with a Shimadzu LC-6A. THF was distilled
from sodium benzophenone ketyl. TLC was carried out on
Merck glass plates precoated with silica gel 60F-254
'0.25 mm) and column chromatography was performed
using Merck 23±400 mesh silica gel. Diethylzinc was
purchased from the Aldrich Chemical Co. 'S)-Leucine
ethyl ester hydrochloride, 'S)-valine methyl ester hydro-
chloride, 'S)-phenylalanine methyl ester hydrochloride,
and the other reagents were obtained from Tokyo Kasei
Kogyo Co. or Wako Pure Chemical Ind. Chiral b-amino
alcohols were prepared according to the previously
described procedure for 1a and 1c.7
4.1.3. (S)-3-Methyl-1,1-diphenyl-2-piperidino-1-butanol
25
2a. Yield 34 and 59%: [a]D ˆ186.9 'c 3.71, CHCl3);IR
'neat) 3350, 3050, 3020, 2910, 2840, 1730, 1600, 1490,
1
1445, 995, 745, 695, 615 cm21; H NMR 'CDCl3) d 0.60
'd, 3H, Jˆ6.6 Hz), 1.10 'd, 3H, Jˆ6.2 Hz), 1.15±1.85
'm,7H), 1.90±2.78 'm,4H), 3.00 'd, 1H, Jˆ10.8 Hz), 6.23
's, 1H), 7.12±7.94 'm, 10H); 13C NMR 'CDCl3) d 146.2,
142.9, 128.0, 127.6, 127.5, 126.9, 126.8, 81.8, 78.0, 52.4,
29.2, 27.2, 24.6, 23.5, 23.4. EI HRMS C22H29NO 'M1)
323.2251. Found 323.2257.
4.1.4. (S)-2-Benzyl-4-methyl-1-phenyl-3-piperidino-2-
25
pentanol 2b. Yield 32 and 74%: [a]D ˆ147.6 'c 3.47,
CHCl3);IR 'neat) 3395, 3080, 3060, 3020, 2920, 2840,
1
1735, 1595, 1490, 1450, 1235, 1085, 745, 695 cm21; H
NMR 'CDCl3) d 0.55±1.74 'm, 7H), 1.00 'd, 3H, Jˆ
6.0 Hz), 1.26 'd, 3H, Jˆ5.4 Hz), 1.83±3.08 'm, 9H), 6.30
'S, 1H), 6.97±7.78 'm, 10H); 13C NMR 'CDCl3) d 139.1,
138.7, 131.0, 127.6, 127.4, 125.8, 125.5, 73.7, 73.5, 51.9,
44.3, 44.2, 28.3, 27.1, 24.6, 23.6, 23.1. EI HRMS C24H33NO
'M1) 351.2564. Found 35.2569.
4.1.1. (S)-8-Methyl-1-phenyl-4-(3-phenylpropyl)-5-piperi-
dino-4-octanol 1d. 'S)-Leucine ethyl ester hydrochloride
'587 mg, 3.0 mmol) was suspended in Et2O '15 ml) and
neutralized with 10% aqueous NaOH solution. The aqueous
solution was separated and extracted with Et2O '3£5 ml)
and the ethereal solution was dried over MgSO4. The solu-
tion of the resulting amino acid ester in THF '1.5 ml) was
slowly added to Ph'CH2)3MgBr that was prepared from Mg
'292 mg, 12 mmol) and Ph'CH2)3Br '1.8 ml, 12 mmol) in
THF '7.2 ml) and the mixture was stirred for 18 h at room
temperature. An aqueous saturated NH4Cl solution was
added to the reaction mixture and the THF solution was
decanted. The aqueous solution was twice washed with
ethyl acetate. After evaporation of the combined organic
solution, the residue was ¯ash chromatographed 'ethyl
acetate/MeOH, 6:1) to give the amino alcohol '352 mg,
33%). To the amino alcohol in acetonitrile '10 ml), 1,5-
diiodopentane '178 ml, 1.2 mmol) and K2CO3 '331 mg,
2.4 mmol) were added and the mixture was re¯uxed for
19 h. After ®ltration of the reaction mixture and evapo-
ration, the residue was dissolved with CH2Cl2 and dried
over MgSO4. The residue was ¯ash chromatographed
'hexane/ethyl acetate, 10:1) to give a colorless oil 1d
4.1.5. (S)-5-Methyl-1-phenyl-3-(2-phenylethyl)-4-pipero-
25
dino-3-hexanol 2c. Yield 47 and 56%: [a]D ˆ114.9 'c
4.37, CHCl3);IR 'neat) 3400, 3080, 3050, 3020, 2930,
2850, 1735, 1600, 1490, 1445, 1385, 1085, 735, 695,
1
630 cm21; H NMR 'CDCl3) d 1.08 'd, 6H, Jˆ6.2 Hz),
1.40±3.16 'm, 20H), 5.67 's, 1H), 6.98±7.55 'm, 10H);
13C NMR 'CDCl3) d 143.7, 143.2, 128.4, 128.3, 128.2,
125.5, 125.4, 75.2, 72.7, 52.9, 40.1, 39.8, 30.5, 30.0, 28.3,
27.3, 24.7, 23.9, 22.6. EI HRMS C26H37NO 'M1) 379.2877.
Found 379.2881.
4.1.6. (S)-1,1,3-Triphenyl-2-piperidino-1-propanol 3a.
25
Yield 29 and 69%: [a]D ˆ110.8 'c 7.61, CHCl3);IR
'neat) 3420, 3060, 3020, 2920, 2840, 1725, 1700, 1655,
1
1595, 1490, 1445, 1270, 1150, 1015, 725, 695 cm21; H
NMR 'CDCl3) d 0.98±2.55 'm, 10H), 2.72 'dd, 1H, Jˆ
11.6, 14.7 Hz), 3.15 'dd, 1H, Jˆ2.4, 14.7 Hz), 3.90 'dd,
1H, Jˆ2.5, 11.6 Hz), 6.35 's, 1H), 7.02±7.78 'm, 15H);
13C NMR 'CDCl3) d 145.6, 144.4, 140.2, 128.9, 128.3,
128.0, 127.8, 127.5, 127.2, 127.1, 126.6, 126.1, 77.5, 74.1,
52.9, 34.1, 26.9, 24.2. EI HRMS C26H29NO 'M1) 371.2251.
Found 371.2252.
25
'325 mg, 77%);[ a]D ˆ10.2 'c 4.35, CHCl3);IR 'neat)
3300, 3020, 3000, 2920, 2840, 2800, 1605, 1490, 1450,
1
1160, 1095, 1030, 984, 745, 695 cm21; H NMR 'CDCl3)
4.1.7. (S)-2-Benzyl-1,4-diphenyl-3-piperidino-2-butanol
25
3b. Yield 35 and 70%: mp 101.58C;[ a]D ˆ19.8 'c
d 0.91 'd, 6H, Jˆ6.3 Hz), 1.08±2.13 'm, 17H), 2.13±2.94
'm, 9H), 4.38 's, 1H), 6.95±7.48 'm, 10H); 13C NMR
'CDCl3) d 142.7, 128.4, 128.2, 125.5, 74.1, 67.6, 36.6,
36.2, 35.9, 35.5, 27.2, 26.6, 25.2, 24.7, 24.3, 21.3. EI
HRMS C29H43NO 'M1) 421.3347. Found 421.3349.
12.61, CHCl3);IR 'KBr) 3360, 3120, 3090, 3030, 2940,
2840, 1595, 1490, 1450, 1385, 1085, 1030, 748, 740, 720,
1
695 cm21; H NMR 'CDCl3) d 1.02±1.65 'm, 6H), 2.06±
3.08 'm, 11H), 5.05 's, 1H), 6.91±7.60 'm, 15H); 13C NMR
'CDCl3) d 140.0, 139.2, 139.0, 130.9, 130.6, 129.0, 128.0,
127.7, 126.0, 125.9, 74.9, 70.0, 53.2, 44.1, 42.3, 32.2, 27.0,
24.2. EI HR MS C28H33NO 'M1) 431.2462. Found 431.2467.
4.1.2. (S)-2-Benzyl-5-methyl-1-phenyl-3-piperidino-2-
25
hexanol 1b. Yield 33 and 49%: [a]D ˆ131.3 'c 3.77,
CHCl3);IR 'neat) 3280, 3080, 3020, 2920, 2850, 2820,
1605, 1490, 1450, 1390, 1160, 1090, 1030, 980, 740,
4.1.8. (S)-1,5-Diphenyl-3-(2-phenylethyl)-2-piperidino-3-
25
1
695 cm21; H NMR 'CDCl3) d 0.64 'd, 3H, Jˆ6.4 Hz),
pentanol 3c. Yield 28 and 61%: [a]D ˆ129.7 'c 1.45,

T. Ohga et al. / Tetrahedron 5712001) 4825±4829
4829
CHCl3);IR 'neat) 3400, 3040, 3010, 2920, 2840, 1605,
References
1
1490, 1450, 1240, 1090, 1030, 740, 695 cm21; H NMR
'CDCl3) d 1.13±2.33 'm, 10H), 2.33±3.27 'm, 12H),
6.94±7.49 'm, 15H); 13C NMR 'CDCl3) d 143.4, 142.9,
140.7, 129.1, 129.0, 128.4, 128.3, 126.2, 126.1, 125.7,
125.6, 73.7, 71.4, 53.5, 44.5, 39.8, 38.5, 32.0, 30.0, 27.1,
24.4. EI HRMS C30H37NO 'M1) 427.2877. Found
427.2881.
1. Oguni, N.;Omai, T. Tetrahedron Lett. 1984, 25, 2823. 'a) For
review see: Noyori, R.;Kitamura, M. Angew. Chem., Int. Ed.
Engl. 1991, 30, 49. 'b) Soai, K.;Niwa, S. Chem. Rev. 1992, 92,
833.
2. Kitamura, M.;Suga, S.;Kawai, K.;Noyori, R. J. Am. Chem.
Soc. 1986, 108, 6071.
3. Soai, K.;Ookawa, A.;Kaba, T.;Ogawa, K. J. Am. Chem. Soc.
1987, 109, 7111.
4.1.9. Typical procedure for enantioselective addition of
diethylzinc to aldehyde: (R)-1-phenyl-1-propanol. To a
solution of 2c '38.0 mg, 0.1 mmol, 10 mol%) in hexane
'2.2 ml) was added benzaldehyde '101 ml, 1.0 mmol)
under an argon atmosphere and the resulting solution was
stirred at room temperature for 20 min. Diethylzinc '1.0 M
solution in hexane, 2.2 ml, 2.2 mmol) was added to the solu-
tion at 08C and the mixture was stirred for 12 h. After being
quenched with aqueous saturated NH4Cl solution, the
mixture was extracted with CH2Cl2, dried 'MgSO4), and
concentrated. The residue was puri®ed by ¯ash column
chromatography 'hexane/ethyl acetate 4:1) to give 'R)-1-
phenyl-1-propanol '117.7 mg, 86%) and the recovered 2c
'27.1 mg, 71%). The ee was determined to be 97% by HPLC
analysis using a DAICEL Chiralcel OB column 'hexane/
i-PrOH 98:2, ¯ow rate: 0.5 ml/min). For 3-nonanol, after
4-nitrobenzoylation, OD column 'hexane/i-PrOH 800:1,
¯ow rate: 0.5 ml/min).
4. Soai, K.;Yokoyama, S.;Hayasaka, T. J. Org. Chem. 1991, 56,
4264.
5. 'a) Sola, L.;Reddy, K. S.;Vidal-Ferran, A.;Moyano, A.;Pericas,
M. A.;Riera, A.;Alvarez-Larena, A.;Piniella, J. J. Org. Chem.
1998, 63, 7078. 'b) Reddy, K. S.;Sola, L.;Moyano, A.;Pericas,
M. A.;Riera, A. J. Org. Chem. 1999, 64, 3969.
6. Paleo, M. R.;Cabeza, I.;Sardina, F. J. J. Org. Chem. 2000, 65,
2108.
7. Kawanami, Y.;Mitsuie, T.;Miki, M.;Sakamoto, T.;Nishitani,
K. Tetrahedron 2000, 56, 175.
8. Puchot, C.;Samuel, O.;Dunach, E.;Zhao, S.;Agami, C.;
Kagan, H. B. J. Am. Chem. Soc. 1986, 108, 2353.
9. Girard, C.;Kagan, H. B. Angew. Chem. Int. Ed. Engl. 1998,
37, 2922.
10. 'a) Kitamura, M.;Okada, S.;Suga, S.;Noyori, R.
J. Am.
Chem. Soc. 1989, 111, 4028. 'b) Kitamura, M.;Yamakawa,
M.;Oka, H.;Suga, S.;Noyori, R. Chem. Eur. J. 1996, 2, 1173.
'c) Kitamura, M.;Suga, S.;Oka, H.;Noyori, R. J. Am. Chem.
Soc. 1998, 120, 9800.
11. Oguni, N.;Matsuda, Y.;Kaneko, T. J. Am. Chem. Soc. 1988,
110, 7877.
Acknowledgements
12. 'a) Yamakawa, M.;Noyori, R. J. Am. Chem. Soc. 1995, 117,
6237. 'b) Kitamura, M.;Oka, H.;Noyori, R. Tetrahedron
1999, 55, 3605.
We are grateful to Professor Tsutomu Katsuki 'Kyushu
University) for helpful discussion.