Reduction of α,β-unsaturated ketones with diphenylsilanes bearing several substituents on their phenyl moiety catalyzed by rhodium-phosphine complexes

Article abstract of DOI:10.1246/cl.2007.366

1,4-Addition product was afforded exclusively by rhodiumphosphine complex-catalyzed hydrosilylation by using 2-cyclohexen-1-one and a dihydrosilane bearing an ether substituent in spite of the fact that dihydrosilanes were believed to give 1,2-addition product selectively. Copyright

Full text of DOI:10.1246/cl.2007.366

366
Chemistry Letters Vol.36, No.3 (2007)
Reduction of ꢀ,ꢁ-Unsaturated Ketones with Diphenylsilanes Bearing Several Substituents
on Their Phenyl Moiety Catalyzed by Rhodium–Phosphine Complexes
Daisuke Imao, Miyuki Hayama, Kohta Ishikawa, Tetsuo Ohta,Ã and Yoshihiko Ito
Department of Molecular Science and Technology, Faculty of Engineering, Doshisha University,
Kyotanabe, Kyoto 610-0394
(Received October 25, 2006; CL-061258; E-mail: tota@mail.doshisha.ac.jp)
1,4-Addition product was afforded exclusively by rhodium–
Table 1. Regioselectivity of Rh-catlayzed hydrosilylation of 2-
cyclohexen-1-one with a variety of dihydrosilanesa
phosphine complex-catalyzed hydrosilylation by using 2-cyclo-
hexen-1-one and a dihydrosilane bearing an ether substituent
in spite of the fact that dihydrosilanes were believed to give
1,2-addition product selectively.
Si
Si
O
O
O
R
RhCl(PPh3)3
THF, rt
PhH2Si
1a-f
+
+
1,2-reduction 1,4-reduction
It is known that Rh-catalyzed hydrosilylation of ꢀ,ꢁ-unsat-
urated ketones brings about 1,2-reduction by using dihydro-
silanes, forming allylic alcohols. In contrast, it is also known that
the same reaction brings about 1,4-reduction by using mono-
hydrosilanes,1 forming silyl enol ethers, which can be used as
powerful synthons in Mukaiyama aldol reaction.2 Although
these mechanisms are yet unknown, it is believed that the num-
ber of hydrogen on silicon atom dominates these regioselectivi-
ties.1 In spite of this belief, there were several examples which
disobeyed this rule. For example, Ojima and Kogure reported
that low concentration of dihydrosilane caused increase of 1,4-
addition product,1b so we assumed there must be another hidden
regioselecting factor especially related to some property of
silane itself. In 1995, Zheng and Chan proposed a unique
mechanism in which ketone coordinates to silane rather than
rhodium in silyl–RhIIIH complex to form alkoxysilyl–RhIIIH
species, though, they could find no clear evidence for this
mechanism.1c If their mechanism is true, the silane could be
considered as a Lewis acid in the catalyses.
At the same time, there have been a lot of reports dedicated
to understanding the nature of silicon hypervalency by introduc-
ing many kinds of functional groups to substituents of silicon
which can form intramolecular dative bonds.3 This means these
reports took Lewis acidity of silanes for granted. To our knowl-
edge, there has been no report of catalysis using hydrosilanes
with hypervalency for Rh-catalyzed reduction of ꢀ,ꢁ-unsaturat-
ed ketones. Here, we report catalysis using a dihydrodiphenylsi-
lane with a dative ether group at an appropriate position inverted
its regioselectivity completely.
Encouraged by a report in which a methoxymethyl ether
group has worked as a dative ligand to silane,4 we first envisaged
that the application of dihydrosilane 1b would have some
influence on the selectivity in catalysis. 2-Cyclohexen-1-one
and RhCl(PPh3)3 were used as a substrate and a catalyst. Results
are summarized in Table 1. The catalysis using dihydrosilane 1b
inverted its regioselectivity completely to afford 1,4-addition
product exclusively (Entry 2). To elucidate the role of an oxygen
atom of an ether group more clearly, several alkylated dihydro-
silanes were also tested. Entries 3–5 indicates that steric effect of
alkyl groups had small influence on regioselectivity and gave
regioisomer mixtures. Interestingly, MesPhSiH2 1f, with which
Fu and Tao has obtained higher enantioselectivity than with 1a
in asymmetric hydrosilylation of ketones using planer-chiral
Entry
Silane
R
H
Yield/%b
1,2-/1,4-b
1
2
3
4
5
6
1a
1b
1c
1d
1e
1f
100
95
100
76
71
37
100/0
0/100
90/10
83/17
79/21
27/73
2-CH2OCH3
2-Me
2-Et
2-n-Pr
2,4,6-triMe
aSubstrate (1 mmol), silane (1.5 equiv.), and catalyst
(0.5 mol %) were used for 3 h. bDetermined by 1H NMR
internal standard method.
P,N ligand, showed considerably high 1,4-selectivity.5
Sevearl phosphine ligands were tested in Table 2. Among
Entries 1–3 ortho-substituted monodentate ligand was found
to cause 1,2-fashion reduction. Catalyses using dppf, dppe, and
(S)-binap afforded 1,2-/1,4-fashion reduction mixtures. 1,2-
Fashion reduction product was exclusively afforded by catalyses
using dppp or xantphos. From these results (Tables 1 and 2)
an ether group on 1b was found to be effective for reducing
the substrate in 1,4-fashion; though, congestion around metal
center reverses the regioselectivity to 1,2-fashion.
Next, we applied the other ꢀ,ꢁ-unsaturated ketones to this
catalysis using 1b. During the investigation under several reac-
Table 2. Effect of ligands on regioselectivitya
Si
Si
O
O
O
[RhCl(cod)]2
O
Phosphine ligand
+
+
PhH2Si
1b
THF, rt
1,2-reduction 1,4-reduction
Entry
Ligand (mol %)
Yield/%b
1,2-/1,4-b
1
2
3
4
5
6
7
8
PPh3 (3)
(o-anisyl)PPh2 (3)
(p-anisyl)3P (3)
dppf (1)
39
20
53
65
61
75
65
50
0/100
100/0
0/100
38/62
74/26
87/13
100/0
100/0
(S)-binap (1)
dppe (1)
dppp (1)
xantphos (1)
aSubstrate (1 mol %) and [RhCl(cod)]2 (0.5 mol %) was used
(24 h). bDetermined by 1H NMR internal standard method.
Copyright Ó 2007 The Chemical Society of Japan

Chemistry Letters Vol.36, No.3 (2007)
367
O
O
OH
L
L
L
Cl
RhCl(PPh3)3
Rh
O
2 M HCl
0.3 mol %
OSiHPh2
Si
O
Ph
1b 1.3 equiv.
L
H
F
99%
80:20
I
2.5 h
L
L
O
O
Rh Cl
1,4- /1,2- reduction
A
H
21:79
3% recovered
O
Ph
H Si
Ph2SiO
L
O
O
E
PhR1SiH2
33:67
H
L
Rh H
Rh
L
L
O
0:100
11% recovered
1,2-fashion
R1 = Ph
1,4-fashion
Cl
5:95
Cl
14% recovered
R1 =
O
H
H
Ph
Ph Si
L
Scheme 1. Regioselectivities of several ꢀ,ꢁ-unsaturated ke-
tones using 1b with no solvent.
O
Ph
H Si
Ph
Si
H
R1
O
Rh H
Cl
BMOM
BPh
H
L
L
L
O
L
Rh
Cl
Rh H
Cl
NTs
L
O
NHTs
OSiHAr2
80%
D
O
1.2 equiv.
0.8 equiv.
1b
G
O
H2O 1 equiv.
NHiPr2 5%
DMF 0.5 mL
2.5 h
cat.
C
2.5 h
90% from silyl enol ether
1 equiv.
anti:syn = 94:6
Scheme 3. Plausible Rh-catalyzed hydrosilylation mechanisms
with 2-cyclohexen-1-one using diphenylsilane and 1b.
Scheme 2. One-pot synthesis procedure for a ꢁ-amino ketone.
and 1b presumably because the Rh has no vacant orbital due
to the extremely wide bite angle of bidentate xantphos.8 Sub-
strates with steric hidrance around their ꢁ-carbon would prob-
ably disfavored the ꢂ-coordination interaction with the complex,
causing 1,2-reduction preferentially.
In summary, we found that 1,4-addition product was
exclusively afforded by rhodium-catalyzed hydrosilylation of
2-cyclohexen-1-one by using dihydrosilane bearing an ether
substituent in spite of the fact that dihydrosilanes were believed
to give 1,2-addition product selectively.
tion conditions, we noticed that catalyses using some ketones
produced 1,2-/1,4-reduction mixtures by using H2SiPh2. Reac-
tion conditions reported by Ojima and Kogure1b where no sol-
vent was used is considered to be the most effective conditions
for dihydrosilanes to demonstrate high 1,2-reduction selectivity.
The results with several ꢀ,ꢁ-unsaturated ketones by using no
solvent are summarized in Scheme 1. The result with 2-cyclo-
hexen-1-one showed worse selectivity; though, it still remained
high 1,4-selectivity. Catalyses with the other ketones under the
same conditions resulted in dramatic decrease the ratios of their
1,4-selectivity due probably to the steric hindrance around their
ꢁ-carbon.
As a supporting evidence for the 1,4-reduction, we carried
out one-pot Mannich-type reaction depicted in Scheme 2.
We anticipated that the silyl enol ether catalytically generated
in situ with 1b should remain a hydride ligand on a silicon
atom. These kind of silyl enol ethers have been known to exhibit
high reactivity in base-catalyzed Mannich-type reaction in the
presence of water and diisopropylamine as a base.6
As expected, a ꢁ-amino ketone was obtained in 72% (90%
from a silyl enol ether) with high anti-selectivity in one-pot syn-
thetic procedure. In contrast, less than 5% of the same product
was obtained by using HSiMe2Ph in the same procedure. This
result would clearly support that a silyl enol ether bearing a
hydride ligand on its silicon is involved in this reaction.
From these results, we postulated mechanisms of rhodium-
catalyzed hydrosilylation of 2-cyclohexen-1-one as depicted in
Scheme 3. In 1,2-fashion, our results could be explained by
the idea of Zheng and Chan in which a carbonyl oxygen atom
would coordinate to Si rather than Rh, depicted as D.
When 1b is used, its ethereal oxygen atom would coordinate
at its silicon Lewis acid orbital intramolecularly, drawn as G.
Therefore, the rhodium metal would predominantly behave as
a Lewis acid to give 1,4-addition product probably through
ꢂ-coordination by the C=C double bond of the substrate.7 1,2-
Fashion reduction was exhibited by a Rh–xantphos catalyst
This work was partially supported by a grant to RCAST at
Doshisha University from the Ministry of Education, Culture,
Sports, Science and Techonlogy, Japan.
This paper is dedicated to the heartfelt memory of the late
Professor Yoshihiko Ito of Doshisha University.
References and Notes
1
a) I. Ojima, Z. Li, J. Zhu, in The Chemistry of Organic Silicon
Compounds, ed. by Z. Rappoport, Y. Apeloig, Wiley, Chichester,
U.K., 1998, Vol. 2, Chap. 29. b) I. Ojima, T. Kogure, Organo-
metallics 1982, 1, 1390. c) G. Z. Zheng, T. H. Chan, Organometal-
lics 1995, 14, 70.
2
a) M. Sawamura, Y. Ito, in Catalytic Asymmetric Synthesis, 2nd
ed., ed. by I. Ojima, Wiley, New York, 2000, Chap. 8B1,
pp. 493–512. b) E. M. Carreira, in Catalytic Asymmetric Synthesis,
2nd ed., ed. by I. Ojima, Wiley, New York, 2000, Chap. 8B2,
pp. 513–541.
3
4
D. Kost, I. Kalikhuman, in The Chemistry of Organic Silicon
Compounds, ed. by Z. Rappoport, Y. Apeloig, Wiley, Chichester,
U.K., 1998, Vol. 2, Chap. 23.
A. Mix, U. H. Berlekamp, H.-G. Stammler, B. Neumann, P. Jutzi,
J. Organomet. Chem. 1996, 521, 177.
5
6
B. Tao, G. C. Fu, Angew. Chem., Int. Ed. 2002, 41, 3892.
K. Miura, K. Tamaki, T. Nakagawa, A. Hosomi, Angew. Chem.,
Int. Ed. 2000, 39, 1958.
7
8
T. Hayashi, K. Yamasaki, Chem. Rev. 2003, 103, 2829.
P. C. J. Kamer, P. W. M. N. van Leeuwen, J. N. H. Reek, Acc.
Chem. Res. 2001, 34, 895.
9
Supporting Information is available electronically on the CSJ-
Journal Web site, http://www.csj.jp/journals/chem-lett/index.html.
Published on the web (Advance View) February 3, 2007; doi:10.1246/cl.2007.366