Optically active lithium-alkoxide catalyzed asymmetric reduction of imines with trimethoxyhydrosilane

Article abstract of DOI:10.1055/s-2003-37527

Optically active lithium alkoxide catalyzed asymmetric reduction of imines with trimethoxhydroysilane in moderate enantioselectivity (up to 72% ee).

Full text of DOI:10.1055/s-2003-37527

LETTER
561
Optically Active Lithium-Alkoxide Catalyzed Asymmetric Reduction of Imi-
nes with Trimethoxyhydrosilane
A
symmetric Reduc
i
tion of
s
Imines
w
a
ith Trimeth
s
oxyhydro
h
silane i Nishikori, Ritsuko Yoshihara, Akira Hosomi*
Department of Chemistry, Graduate School of Pure and Applied Science, University of Tsukuba, CREST JST (Japan Science and
Technology), Tsukuba, Ibaraki 305-8571, Japan
Fax +81(298)536503; E-mail: hosomi@chem.tsukuba.ac.jp
Received 19 January 2003
was shown (entry 7). In contrast, enantioselectivity was
Abstract: Optically active lithium alkoxide catalyzed asymmetric
enhanced up to 57% ee by using (R)-1,1′-bi-2-naphthol 4
reduction of imines with trimethoxhydroysilane in moderate enan-
tioselectivity (up to 72% ee).
(entry 8). Then we carried out modification of 4 and ex-
amined the reaction. However, enantioselectivities were
decreased to only low level (entry 9, 10). We also exam-
Key words: asymmetric reduction, lithium alkoxide, imine, solvent
effect
ined the reaction with monolithio salt of 4. Contrary to the
reduction of ketones,5b lower enantioselectivity of 36% ee
than that of dilithiosalt of 4 was shown (entry 12).
Hydrosilanes have come into wide use in transition metal-
catalyzed asymmetric hydrosilylation of carbon-carbon or
carbon-heteroatom double bonds.1 In 1986 we first found
Table 1 Asymmetric Reduction of Imine 1a with Various
Catalystsa
that in situ generated lithium alkoxide catalyzed reduction
of carbonyl compounds with trialkoxysilanes.2,3 We also
found the reaction proceeded in diastereoselective and
enantioselective manner4 (Scheme 1). For example, re-
duction of acetophenone with dilithiosalt of (S)-prolinol
(0.4 mol%) showed moderate enantioselectivity (52% ee)
at room temperature. After that, some groups reported im-
provement of the system. Kagan et al. demonstrated that
mono lithiosalt of (R)-1,1′-bi-2-naphthol (5 mol%) cata-
lyzed the reduction of acetophenone at 0 °C with 61% ee.5
Ts
Ts
Catalyst, BuLi
HSi(OMe)3
N
HN
Ph
Ph
1a
2a
Entry Catalyst
Yield
(%)b
Ee
Config.d
(%)c
13
13
4
1
2
(S)-Isoleucinol
(S)-Valinol
56
28
45
48
60
37
68
53
37
27
49
70
S
S
O
OH
*
R*OH, BuLi (1 or 2 equiv.)
HSi(OMe)3
3
(S)-Phenylalaninol
S
Ph
Ph
4
(S)-Prolinol
1
S
R*OH : (S)-Prolinol (0.4 mol %)
: (R)-1,1'-Bi-2-naphthol (5 mol%)
52 % ee
61 % ee
5
(R,R)-2,3-Butanediol
28
32
2
R
R
R
R
R
R
–
6
(R,R)-Hydrobenzoin
Scheme 1 Enantioselective reduction of acetophenone
7
3
We also found that lithium methoxide catalyzed reduction
of imines with trimethoxyhydrosilane.6 Thus, we expect-
ed that the method is able to apply for enantioselective re-
duction of imines.7 In this paper, we describe the
enantioselective reduction of imines with trimethoxyhy-
drosilane by using mono or dilithiosalts of optically active
alcohols as catalyst.
8
4
57
21
2
9
5
10
11e
12e
6
(R)-Phenylethylalcohol
0
4
36
R
At first, we examined the reaction of 1a with catalysts us-
ing various dilithio salts in tetrahydrofuran (Table 1).
Amino alcohols showed only poor enantioselectivity (en-
try 1–4). (R,R)-2,3-Butanediol and (R,R)-hydrobenzoin
showed modest enantioselectivity of 28% ee and 32% ee,
a All reactions were carried out with 20 mol% of catalyst, 40 mol% of
BuLi and 2 equiv of (MeO)3SiH at r.t. for 48 h in THF.
b Isolated yields.
c Determined by HPLC analysis using optically active column
(DAICEL CHIRALPAK AD-H, hexane–2-propanol, 9:1).
respectively (entry 5,6). Using tartrate-derived chiral 1,4- d Configuration was determined by the comparison of the elution
order with the authentic sample.
diol 3 (TADDOL, Figure 1), very poor enantioselectivity
e The reaction was carried out in the presence of 20 mol% of BuLi.
Synlett 2003, No. 4, Print: 12 03 2003.
Art Id.1437-2096,E;2003,0,04,0561,0563,ftx,en;U13703ST.pdf.
© Georg Thieme Verlag Stuttgart · New York
ISSN 0936-5214

562
H. Nishikori et al.
LETTER
tion of 2 equivalents of N,N,N′,N′-tetramethylethylenedi-
amine gave the product of opposite absolute configuration
with moderate enantioselectivity of 59% ee (entry 11).
We carried out further examination of other solvents, but
more effective one was not found (entry, 8, 14, 15). We
paid attention to the two solvent systems, which were tet-
rahydrofuran and diethyl ether–TMEDA, and examined
temperature effect for each ones. In case of tetrahydrofu-
ran, the reaction at lower temperature (0 °C) improved
enantioselectivity up to 65% ee (entry 6). But the reaction
at further lower temperature –20 °C), enantioselectivity
was decreased (entry 7). The diethyl ether–TMEDA sys-
tem at lower temperature, on the other hand, only modest
enantioselectivities were shown (entry 12, 13). From
these noticeable solvent and TMEDA effects, it is sug-
gested that lithium cation deeply concern in the reaction,
although the detailed reaction mechanism is not clear at
present.8
R
Ph Ph
O
O
OH
OH
Ph
4 : R = H
5 : R = Me
6 : R = SiPh3
OH
OH
R
Ph
Ph
3 (TADDOL)
R
Figure 1
Next, we carried out optimization of the reaction condi-
tion by using dilithiosalt of 4 (base, solvent, additive and
temperature, Table 2). Using lithium bistrimethylsilyla-
mide (LiHMDS) in place of butyllithium, or adding
N,N,N′,N′-tetramethylethylenediamine (TMEDA) in tet-
rahydrofuran, little influence was observed on enantiose-
lectivity (entry 2, 5). We also carried out examination of
counter cation effect. Using NaHMDS or KHMDS as
base, only trace amounts of amine were observed (entry 3,
4). Contrary to tetrahydrofuran, noticeable solvent effect,
reverse of the absolute configuration of the product, ap-
peared by using diethyl ether that is less polar solvent (en-
try 9). Lithium bistrimethylsilylamide enhanced the
enantioselectivity to 31% ee (entry 10). Moreover, addi-
Reduction of some other imines was also examined under
optimized conditions, and the reduction of imine 1b
showed moderate enantioselectivity of 72% ee
(Scheme 2).
Table 2 Asymmetric Reduction of Imine 1a:a Effect of Base, Solvent, Additive, and Temperature
Entry
1
Base
Solvent
THF
Additive
Temp (°C)
r.t.
Yield (%)b
Eec (%)
57
54
–
Config.d
BuLi
–
53
38
trace
trace
52
43
47
44
30
65
63
75
67
28
4
R
R
–
2
LiHMDSe
NaHMDSf
KHMDSg
BuLi
THF
–
r.t.
3
THF
–
r.t.
4
THF
–
r.t.
–
–
5
THF
TMEDAh
r.t.
51
65
55
25
12
31
59
19
1
R
R
R
R
S
6
BuLi
THF
–
0
7
BuLi
THF
–
–20
r.t.
8
BuLi
DME
Et2O
–
9
BuLi
–
r.t.
10
11
12
13
14
15
LiHMDSe
BuLi
Et2O
–
r.t.
S
Et2O
TMEDAh
r.t.
R
R
R
S
BuLi
Et2O
TMEDAh
0
BuLi
Et2O
TMEDAh
–20
r.t.
BuLi
i-Pr2O
Toluene
–
–
6
BuLi
r.t.
11
S
a All reactions were carried out with 20 mol% of 4, 40 mol% of base and 2 equiv of (MeO)3SiH for 48 h.
b Isolated yields.
c Determined by HPLC analysis using optically active column (DAICEL CHIRALPAK AD-H, hexane–2-propanol, 9:1).
d Configuration was determined by the comparison of the elution order with the authentic sample.
e LiHMDS = Lithium bistrimethylsilylamide.
f NaHMDS = Sodium bistrimethylsilylamide.
g KHMDS = Potasium bistrimethylsilylamide.
h Reaction was carried out in the presence of 2 equiv of TMEDA.
Synlett 2003, No. 4, 561–563 ISSN 0936-5214 © Thieme Stuttgart · New York

LETTER
Asymmetric Reduction of Imines with Trimethoxyhydrosilane
563
Acknowledgment
Ns
N
Ns
4 (20 mol%), LiHMDS
HN
This work was partly supported by Grant-in-Aid for Scientific
Research from the Japan Society for the Promotion of Science
(JSPS). We thank Dow Corning Toray Silicone Co. Ltd. and Shin-
Etsu Chemical Co. Ltd. for a gift of organosilicon compounds.
HSi(OMe)3, THF, –20 °C
Ph
Ph
*
1b
2b
Ns = p-Nitrobenzenesulfonyl
35%, 72% ee
Scheme 2 Enantioselective reduction of Imine 1b
References
(1) For the review of asymmetric hydrosilylation, see for
example: Nishiyama, H.; Itoh, K. In Catalytic Asymmetric
Synthesis, 2nd ed.; Ojima, I., Ed.; VCH Publisher, Inc.: New
York, 2000, 111.
(2) For reviews of extracoordination at silicon, see: (a) Brook,
M. A. Silicon in Organic, Organometallic, and Polymer
Chemistry; Interscience Publisher, Inc.: New York, 2000,
Chap. 4, 97. (b) Hosomi, A. Reviews on Heteroatom
Chemistry 1992, 7, 214.
(3) Hosomi, A.; Hayashida, H.; Kohra, S.; Tominaga, Y. J.
Chem. Soc., Chem. Commun. 1986, 1411.
(4) (a) Kohra, S.; Hayashida, H.; Tominaga, Y.; Hosomi, A.
Tetrahedron Lett. 1988, 29, 89. (b) Hojo, M.; Fujii, A.;
Murakami, C.; Aihara, H.; Hosomi, A. Tetrahedron Lett.
1995, 36, 571.
(5) (a) Pini, D.; Iuliano, A.; Salvadori, P. Tetrahedron:
Asymmetry 1992, 3, 693. (b) Schiffers, R.; Kagan, H. B.
Synlett 1997, 1175. (c) LaRonde, F. J.; Brook, M. A.
Tetrahedron Lett. 1999, 40, 3507.
(6) Hojo, M.; Murakami, C.; Fujii, A.; Hosomi, A. Tetrahedron
Lett. 1999, 40, 911.
(7) Matsumura et al. reported proline derivatives catalyzed
asymmetric reduction with trichlorosilane and moderate
enantioselectivity of 66% ee was achieved: Iwasaki, F.;
Onomura, O.; Mishima, K.; Kanematsu, T.; Maki, T.;
Matsumura, Y. Tetrahedron Lett. 2001, 42, 2525.
(8) We have already found that diastereoselectivity was
switched by selection of solvent in the reduction of α,β-
epoxyketones. See ref.4b
Typical experimental procedure is exemplified with the
reduction of imine 1a with dilithio salt of 4 as chiral cata-
lyst: To a solution of (R)-1,1′-bi-2-naphthol 4 (11.5 mg,
0.04 mmol) in dry THF (1.0 mL) was added BuLi (1.57 N
in hexane solution, 51.0 µL, 0.08 mmol) under nitrogen,
and stirred for 15 minutes. After stirred, this solution was
cooled to 0 °C. To this solution were added trimethoxyhy-
drosilane (0.40 mmol, 55 µL) and imine 1a (0.20 mmol,
54.7 mg), and the mixture was stirred for 72 hours at 0 °C.
The reaction mixture was quenched with aqueous
NaHCO3 (0.1 N, 1.3 mL), extracted with diethyl ether,
washed with brine, dried over Na2SO4, and concentrated
in vacuo. The residue was purified by column chromatog-
raphy on silica gel to give amine 2a of 65% ee in 43%
yield.
In conclusion, we were able to demonstrate that the lithi-
um alkoxide-catalyzed enantioselective reduction system
could be applied to imines in moderate enantioselectivity,
and absolute configuration of product was switched by
selection of solvents. Further investigation is in progress
in our laboratory.
Synlett 2003, No. 4, 561–563 ISSN 0936-5214 © Thieme Stuttgart · New York