Rh(I)-catalysed asymmetric hydrosilylation using new oxazoline ligands with potential charge-transfer properties

Article abstract of DOI:10.1002/(SICI)1099-0682(199806)1998:6<783::AID-EJIC783>3.0.CO;2-M

Novel N′-methylated 2-(oxazolin-2-yl)-4,4′-bipyridinium salts, bearing a chiral oxazoline moiety, were tested in the Rh(I)-catalysed enantioselective hydrosilylation. After coordination to rhodium these electron-attracting ligands are supposed to exhibit charge-transfer effects with electron-donating substrates. Therefore, a new catalytic hydrosilylation reaction with 2,5-dimethoxyacetophenone as an electron-rich substrate was developed. The results were compared with those of the non-methylated 2-(oxazolin-2-yl)-4,4′-bipyridine and related ligands. In addition, the new ligands and Rh(I)-complexes were tested in the hydrosilylation of acetophenone.

Full text of DOI:10.1002/(SICI)1099-0682(199806)1998:6<783::AID-EJIC783>3.0.CO;2-M

FULL PAPER
Asymmetric Catalysis, 118[᭛]
Rh(I)-Catalysed Asymmetric Hydrosilylation using New Oxazoline Ligands
with Potential Charge-transfer PropertiesȠ
Henri Brunner* and Reinhard Störiko
Institut für Anorganische Chemie der Universität Regensburg,
D-93040 Regensburg, Germany
Fax: (internat.) ϩ49(0)941/9434439
E-mail: henri.brunner@chemie.uni-regensburg.de
Received September 30, 1997
Keywords: Enantioselective hydrosilylation / Oxazoline ligands / Charge-transfer / 4,4-Bipyridine derivatives /
2,5-Dimethoxyacetophenone
Novel NЈ-methylated 2-(oxazolin-2-yl)-4,4Ј-bipyridinium
salts, bearing a chiral oxazoline moiety, were tested in the
reaction with 2,5-dimethoxyacetophenone as an electron-
rich substrate was developed. The results were compared
Rh(I)-catalysed enantioselective hydrosilylation. After coor- with those of the non-methylated 2-(oxazolin-2-yl)-4,4Ј-bipy-
dination to rhodium these electron-attracting ligands are ridine and related ligands. In addition, the new ligands and
supposed to exhibit charge-transfer effects with electron-do- Rh(I)-complexes were tested in the hydrosilylation of aceto-
nating substrates. Therefore, a new catalytic hydrosilylation
phenone.
Introduction
tried to find evidence for charge-transfer effects in the
Rh(I)-catalysed hydrosilylation using the electron-rich 2,5-
dimethoxyacetophenone as the substrate. As acceptor with
electron-accepting properties we resorted to the Rh(I)-coor-
dinated 2-(oxazolin-2-yl)-4,4Ј-bipyridinium salts 5, de-
scribed in the preceding publication. The adduct catalyst/
substrate should resemble the charge-transfer complex pa-
raquat/hydroquinone (both formulas are shown in the pre-
ceding paper, see Introduction). In the ligands of type 5
chirality is easily introduced into the oxazoline ring. Li-
gands 5 contain the same 2-(pyridin-2-yl)oxazoline chelat-
ing unit as the ligands of type 6 (Scheme 2), which were
In the Rh(I)-catalysed enantioselective hydrosilylation of
prochiral ketones, the substrate (e.g. acetophenone 1a) re-
acts with diphenylsilane to yield the chiral silylalkyl ether
2a (Scheme 1)[1][2][3]. After hydrolysis the chiral 1-phenyl-
ethanol 4a is obtained. Asymmetric induction is brought
about by in situ Rh(I)-catalysts consisting of [Rh(cod)Cl]2
and preferably nitrogen ligands, e.g. chiral oxazo-
lines[4][5][6][7]
.
Scheme 1
Scheme 2
In the preceding paper, we proposed charge-transfer in-
teractions for a possible pre-orientation of the substrate to
be hydrosilylated with respect to the catalytic centre. We
[᭛]
Part 117: H. Brunner, R. Störiko, F. Rominger, Eur. J. Inorg.
Chem. 1998, 771Ϫ781, preceding paper.
Eur. J. Inorg. Chem. 1998, 783Ϫ788
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ1948/98/0606Ϫ0783 $ 17.50ϩ.50/0
783

H. Brunner, R. Störiko
FULL PAPER
among the first successful chiral nitrogen ligands in enanti- Table 1. Hydrosilylation of 2,5-dimethoxyacetophenone with 5aϪe,
7aϪe, 8e, 9c and 9e; reaction time: 90 h (no solvent) or 18 h (in
oselective catalysis[4][8]
To establish charge-transfer effects in the Rh(I)-catalysed
enantioselective hydrosilylation, the catalytic results of the
.
CCl4); catalyst: 0.23 mol% [Rh(cod)Cl]2 (0.46 mol% Rh) and 2.3
mol% ligand; substrate/Rh: 214:1
ligands of type 5 were compared with those of similar but ligand no of runs silylenol
conver-
ether [%] sion [%]
chemical ee (config.) [%]
yield [%]
less electron-accepting ligands. Obvious differences in the
optical yield might originate from a charge-transfer me-
diated pre-orientation.
7a
2
26
0
95[a]
77
71
77
68
77
60
75
70
88
76
47
76
96
68
55
87
5.8
38.5
27.0
3.2
(S)
(S)
(S)
(R)
(R)
(R)
(R)
(R)
(S)
(S)
(R)
(R)
(R)
(R)
(R)
2 (ϩ CCl4)
5a
7b
5b
7c
5c
9c
7d
5d
7e
2
22
23
34
26
19
11
25
50
21
0
86[a]
99[a]
90[a]
2
2
28.1
7.1
2
100
Hydrosilylation of 2,5-Dimethoxyacetophenone
2
86
44.4
46.3
1.0
10.5
6.7
2 (18 h)
97[a]
2,5-Dimethoxyacetophenone (1b) has already been tested
in the asymmetric hydrosilylation[9] and in two recent pa-
pers it has been reduced with chiral auxiliaries[10][11]. A
standard procedure for the hydrosilylation of 2,5-dimeth-
2
100
2
93[a]
96[a]
96[a]
91[a]
78
4
2 (ϩ CCl4)
52.1
29.0
9.4
5e
4
2
26
30
12
oxyacetophenone with diphenylsilane was set up[12], similar 8e
9e
2 (18 h)
99[a]
44.8
to the acetophenone/diphenylsilane system. Since the ra-
cemic alcohol 4b was not well characterised in the litera-
[a]
No more diphenylsilane detectable by 1H NMR spectroscopy
ture[13], it was synthesised by reduction of 1b with KBH4 in
(80 MHz, CDCl3).
methanol and spectroscopically characterised. The catalytic
1
intermediates 2b and 3b were analysed in situ by H NMR
Figure 1. Comparison of the enantioselectivities of the hydrosilyla-
tion of 2,5-dimethoxyacetophenone with the ligands 7a؊e and
5a؊e
spectroscopy. The silylenol ether 3b originates from the enol
tautomer of 1b and on hydrolysis reverts to the ketone 1b
(Scheme 1). Hence, it reduces the chemical yield of the chi-
ral alcohol 4b.
For quantification of the catalytic results three param-
1
eters were calculated from the H NMR spectra (80 MHz,
CDCl3), which were recorded for each catalysis (reproduc-
ibility about ±5%): (i) the amount of silylenol ether 3 in
the hydrosilylation product: 3/(2ϩ3); (ii) the total degree
of hydrosilylation (conversion): (2ϩ3)/(1ϩ2ϩ3) and (iii) the
chemical yield of the silylalkyl ether 2, which on hydrolysis
is converted to the chiral product 4: (2)/(1ϩ2؉3). The en-
antiomeric excess of the alcohol 4 was determined by GC
with a Chirasil-DEX CB column (reproducibility ±0.5%).
The values in the tables were averaged over usually two in-
dependent runs, differing in most cases less than 5% (con-
version and chemical yield) and 1% (ee), respectively.
Unexpectedly, the anion exchanged 8e (hexafluorophos-
phate instead of iodine in 5e, Scheme 2) led to a much lower
To compare the NЈ-methylated ligands 5a؊e with the enantioselectivity than 5e (Table 1), a result which does not
non-methylated ligands 7a؊e (Scheme 2) each pair was fit into the picture of charge-transfer interactions between
doubly tested in parallel runs. No CCl4 (see later) was ad- coordinated ligand and substrate.
ded due to the limited solubility of the ionic compounds
5a؊e in chlorinated solvents. For reaction conditions and (Scheme 2) are sterically almost identical to 7c and 7e,
results see Table 1. respectively. As they are definitely less electron-attracting
The 2-(4-phenylpyridin-2-yl)oxazoline ligands 9c and 9e
All the NЈ-methylated ligands 5a-e induced much higher than 5c and 5e, no charge-transfer effects were expected in
optical inductions than their non-methylated counterparts the hydrosilylation of 2,5-dimethoxyacetophenone. Surpris-
7a؊e (Figure 1). The pair 5c/7c exhibited the largest differ- ingly, both ligands yielded relatively high enantioselectivi-
ence with an average ∆ee of about 40%. The degree of ties (9c: 46.3% ee, 9e: 44.8% ee, Table 1), which are slightly
hydrosilylation was generally high (>79%), in some cases better than the NЈ-methylated ligands 5c and 5e and differ
even quantitative. The amount of silylenol ether scattered markedly from their sterically equivalent 4,4Ј-bipyridine
around 14Ϫ58%.
counterparts 7c and 7e. An explanation of these findings
The non-methylated bipyridines 7a and 7e performed might be the free NЈ-position in the ligands 7. It is known
much better if CCl4 was added. No silylenol ether was that for the Rh(I)-catalysed hydrosilylation with nitrogen
formed and the optical yield increased remarkably (Table ligands, a ligand excess is beneficial. This points to equilib-
1). This significant increase in asymmetric induction on ria involving catalytically active species containing at least
conducting the hydrosilylation in CCl4 is known as the CCl4 two ligand molecules, one of which could be even monod-
effect[4][14][15] and will be referred to later.
entate[15]. Only the non-methylated ligands 7 can bind via
784
Eur. J. Inorg. Chem. 1998, 783Ϫ788

Asymmetric Catalysis, 118
FULL PAPER
Figure 2. UV-Spectra in methanol (ca. 10Ϫ5 M)
their free pyridineЈ nitrogen and this might result in the low
optical yields. Due to the methylation of this nitrogen in 5
and due to the phenyl substituent in 9 the formation of such
species is impossible for these ligands.
Also, the bisoxazolines 10 and 11 (Scheme 3) were tested
due to their potential charge-transfer properties if coordi-
nated to two Rh-centres. A Rh/ligand ratio of 2:1 was used
and CCl4 was added (no precipitation as with ligands 5a؊e
was observed). However, no asymmetric induction was ob-
served with 10 and 11[12]
.
Scheme 3
ence on the catalytic results compared to other effects re-
sulting from the nature of the ligands.
Hydrosilylation of Acetophenone
The new ligands were tested in the hydrosilylation of aceto-
phenone with diphenylsilane in CCl4. The results of the ox-
azolines 7a-f and the bisoxazoline 10 are shown in Table 2.
In general good chemical (up to 95%) and optical yields
(up to 74% ee) were obtained. In most cases no silylenol
ether was formed.
Table 2. Hydrosilylation of acetophenone with 7a-f and 10; reaction
time: 18 h; catalyst: 0.24 mol% [Rh(cod)Cl]2 (0.48 mol% Rh) and
2.35 mol% ligand; substrate/Rh: 210:1
ligand no of runs silylenol conver-
chemical ee
(solvent)
ether
[%]
sion
[%]
yield
[%]
(config.)
[%]
So far the hydrosilylation results did not provide defini-
tive proof for charge-transfer interactions between coordi-
nated ligand and substrate. Hence, we investigated the ab-
sorption behaviour of some presumed catalytic intermedi-
ates looking for charge-transfer transitions[16]. The UV
spectra of equally concentrated methanol solutions of 2,5-
dimethoxyacetophenone, [Rh(cod)5e]Cl (prepared in situ),
and [Rh(cod)5e]Cl ϩ 2,5-dimethoxyacetophenone (1:1)
7a
2 (CCl4)
2 (CCl4)
2 (CCl4)
2 (CCl4)
2 (CCl4)
2 (CCl4)
2 (none)
1 (C6F6)
1 (C6HF5)
1 (Cl2-
0
0
92[a]
91[a]
94[a]
78[a]
85[a]
94[a]
98[a]
99[a]
96[a]
96[a]
92
91
94
70
42
94
65
53
85
49
59.8 (S)
66.5 (R)
73.9 (R)
62.9 (S)
29.8 (S)
67.9 (R)
9.6 (R)
7b
7c
0
7d
7
50
0
34
46
11
49
10[b]
7e
7e[c]
7e
4.3 (R)
7e
37.1 (R)
racemate
were recorded. 2,5-Dimethoxyacetophenone showed an ab- 7e
sorption maximum at 330 nm in accord with the litera-
CϭCCl2)
2 (CCl3Br)
2 (CCl4)
7e
7f
Ϫ
14
0
0
70
Ϫ
ture[17]. [Rh(cod)5e]Cl exhibited a maximum at 287 nm and
a shoulder at ca. 380 nm. The spectrum of the 1:1 catalyst/
substrate mixture was only a superposition of the com-
poundsЈ single spectra. No charge-transfer transitions were
found up to 900 nm (Figure 2).
81[a]
31.8 (R)
[a]
No more diphenylsilane detectable by 1H NMR spectroscopy
[b]
[c]
(80 MHz, CDCl3). Ϫ
90 h.
1.16 mol% ligand. Ϫ
Reaction time:
By substituting the electron-rich 2,5-dimethoxyaceto-
Figure 3 compares the enantioselectivities obtained with
phenone for the much less electron-donating acetophenone the ligands 7a؊f and the reported values (comparable con-
(to be described next), charge-transfer interactions should ditions) of the 2-(pyridin-2-yl)oxazolines 6a؊f (Scheme
be strongly reduced. However, 5e used as a cocatalyst in 1)[4]
.
the hydrosilylation of acetophenone gave a higher optical
The 2-(4,4Ј-bipyridin-2-yl)oxazolines 7a؊c and 7e in-
induction than the non-methylated 7e (17.8% compared to duced a slightly higher optical induction compared to the
9.6% ee, similar amounts of silylenol ether and chemical corresponding 2-(pyridin-2-yl)oxazolines 6. Only 7d and 7f
yields under identical conditions)[12]
.
performed worse than their counterparts 6d and 6f. The
All these experiments showed that the desired charge- results with the tert-butyl substituted 7f were surprising, be-
transfer interactions, if existent at all, had only little influ- cause the enantiomeric excess dropped appreciably with re-
Eur. J. Inorg. Chem. 1998, 783Ϫ788
785

H. Brunner, R. Störiko
FULL PAPER
Figure 3. Comparison of the enantioselectivities of the hydrosilyla-
Using acetophenone as the prochiral substrate and 7e as
the cocatalyst the optical induction in CCl4 increased by
58.3% compared to the solvent-free reaction (Table 2).
Therefore, we studied the hydrosilylation of acetophenone
using the ligand 7e in other polyhalide solvents (Table 2).
None of these solvents increased the enantioselectivity com-
pared to CCl4. Replacing hexafluorobenzene by penta-
fluorobenzene the content of silylenol ether decreased from
46 to 11%, while the optical yield increased from 4.3 to
37.1%. In perchloroethylene only racemic alcohol was pro-
duced, whereas CCl3Br totally inhibited the hydrosilylation.
This inhibition had been reported for similar reactions[14]
and explained by an oxidative addition of CCl3Br to Rh(I)
yielding a catalytically inactive Rh(III)-species.
tion of acetophenone with the ligands 7a؊f and 6a؊f (values of
6a؊f taken from the literature[4]
)
The activity of in situ catalysts consisting of [Rh(cod)Cl]2
and the 2-(4-phenylpyridin-2-yl)oxazolines 9c and 9e de-
pended on the stage, when the additive CCl4 came into play
(Table 3). If the in situ catalyst (always easily recognised by
its red colour) was prepared in the presence of CCl4, no
hydrosilylation activity was observed. If the in situ catalyst
was prepared in acetophenone and CCl4 was added later
together with the diphenylsilane, smooth reaction took
place as usual.
gard to 6f, which still is the best 2-(pyridin-2-yl)oxazoline
for asymmetric hydrosilylation[4]
.
With the bisoxazoline 10 less than half the optical yield
of its mono-oxazoline equivalent 7d was obtained (Table 2),
while the amount of silylenol ether was more than seven-
fold. The pyrazine ligand 11 did not yield any optically ac-
tive product at all (three double experiments with different
concentrations of 11)[12]
.
It is well known, that Rh(I)-catalysed hydrosilylations
with nitrogen ligands tend to give higher chemical and, in
Since the Rh-complex 12 (Scheme 3), containing 9c as
particular, optical yields in the presence of CCl4 (ЉCCl4 ef- the ligand, had been synthesised (preceding paper), it was
fectЉ)[4][14][15]. So far the highest increase in the asymmetric tested in three different catalytic runs (Table 3). 12 proved
induction by conducting the catalysis in CCl4 has been to be a poor catalyst with respect to the chemical and op-
18.3% ee compared to toluene solution or the solvent-free tical yield. However, by adding a fourfold amount of ligand
procedure[15]. As already mentioned, the bipyridine ligands establishing the Rh/ligand ratio of 1:5, a catalytic system
7a and 7e improved the optical yields in the hydrosilylation resulted, which formed almost no silylenol ether and which
of 2,5-dimethoxyacetophenone in CCl4 by striking 32.7 and gave 74.8% ee, exceeding the asymmetric induction of the
45.4% ee, respectively (Table 1).
in situ system (70.5% ee) to some extent.
Table 3. Hydrosilylation of acetophenone with the ligands 9c, 9e and the complexes 12؊14; reaction time: 18 h (if not otherwise stated);
0.48 mol% Rh; substrate/Rh: 210:1
catalytic system
no of runs
(solvent)
silylenol ether
[%]
conversion
[%]
chemical yield
[%]
ee (config.)
[%]
[Rh(cod)Cl]2 ϩ
2.35 mol% 9e
[Rh(cod)Cl]2 ϩ
2.35 mol% 9e
[Rh(cod)Cl]2 ϩ
2.35 mol% 9e
[Rh(cod)Cl]2 ϩ
2.35 mol% 9c
[Rh(cod)Cl]2 ϩ
2.35 mol% 9c
[Rh(cod)Cl]2 ϩ
2.35 mol% 9c
12[c]
1 (none)
13
Ϫ
0
99[a]
0, 0[b]
93[a]
100
86
54.9
Ϫ
(R)
3 (CCl4 from
the beginning)
1 (CCl4 added
with H2SiPh2)
0, 0[b]
93
64.2
45.0
Ϫ
(R)
(R)
2 (none)
22
Ϫ
5
78
(CCl4 from1
the beginning)
1 (CCl4 added
with H2SiPh2)
1 (CCl4 from
the beginning)
1 (CCl4 added
with H2SiPh2)
2 (CCl4 added
with H2SiPh2)
1 (CCl4)[c]
0
0
95[a]
96[a]
92[a]
90
70.5
2.9
(R)
(R)
(R)
62
76
36
12[c]
22
2.6
12 ϩ
1.9 mol% 9c[d]
2
17
29
25
0
84
83
68
61
57
85
74.8
33.6
34.6
14.4
66.2
(R)
(R)
(R)
(R)
(R)
13
81[a]
86[a]
74[a]
85
1 (CCl4)[d]
14
2 (CCl4)[d]
14 ϩ 1.9 mol% 7e
2 (CCl4)[c]
[b]
[c]
[a] No more diphenylsilane detectable by 1H NMR spectroscopy (80 MHz, CDCl3). Ϫ
After 90 h. Ϫ
Reaction time: 66 h. Ϫ
[d] Reaction time: 42 h.
786
Eur. J. Inorg. Chem. 1998, 783Ϫ788

Asymmetric Catalysis, 118
FULL PAPER
(l ϭ 25 m, &osol; ϭ 0.25 mm), integrator: Varian 4290, retention
times (160°C): 8.6Ϫ8.8 min [(R)-1-(2,5-dimethoxyphenyl)ethanol]
and 9.2Ϫ9.4 min [(S)-1-(2,5-dimethoxyphenyl)ethanol].
All these results show that the inhibition of the catalyst
by CCl4 is only possible during its formation. Once the
catalyst is formed, CCl4 cannot inhibit its activity. Instead,
it leads to a lower amount of silylenol ether and a higher
enantioselectivity compared to the hydrosilylation without
CCl4 (Table 3). The ЉinertnessЉ of the prepared Rh(I)-com-
plex 12 towards CCl4 was also proven by recording 1H
NMR spectra of 12 in CDCl3 before and after adding 25
vol% of CCl4. Even 24 h after the addition no changes of
the original spectrum were evident.
Two other Rh(I)-complexes 13 and 14 (Scheme 3) were
tested in the hydrosilylation of acetophenone (Table 3). Al-
though they only differ in the nature of their anion, the
BF4-complex 13 gave a much higher optical yield than the
PF6-complex 14. The enantioselectivities of these complexes
(Rh/ligand ratio: 1:1) are quite low compared to the in situ
catalysts (Rh/ligand ratio: 1:5). In line with this, addition
of a fourfold excess of 7e to 14 resulted in similar chemical
and optical yields as the in situ catalyst of [Rh(cod)Cl]2 and
7e (Tables 3 and 2).
1-(2,5-Dimethoxyphenyl)ethyl Diphenylsilyl Ether (2b). Ϫ 1H
3
NMR (250 MHz, CDCl3): δ ϭ 1.45 (d, J ϭ 6.2 Hz, 3 H, CH3),
3
3.65 (s, 3 H, OCH3), 3.73 (s, 3 H, OCH3Ј), 5.36 (q, J ϭ 6.2 Hz, 1
H, CHCH3), 5.43 (s, 1 H, SiH), 6.71 / 6.72 / 7.19 (3 m, 3 H, H-3,
H-4, H-6), 7.29Ϫ7.45 (m, 6 H, Ph-H), 7.56Ϫ7.65 (m, 4 H, Ph-H).
1-(2,5-Dimethoxyphenyl)ethenyl Diphenylsilyl Ether (3b)[18]. Ϫ
1H NMR (250 MHz, CDCl3): δ ϭ 3.69 (s, 3 H, OCH3), 3.72 (s, 3
H, OCH3Ј), 4.77 (d, 2 J ϭ 1.2 Hz, 1 H, CϭCHHЈ), 5.10 (d, 2 J ϭ
1.2 Hz, 1 H, CϭCHHЈ), 5.64 (s, 1 H, SiH), 6.79Ϫ6.81 (m, 2 H,
Ar-H), 7.15 (m, 1 H, Ar-H), 7.27Ϫ7.49 and 7.52Ϫ7.71 (m, 10 H,
Ph-H).
rac-1-(2,5-Dimethoxyphenyl)ethanol (4b): 2,5-Dimethoxyaceto-
phenone (1.5 ml, 9.5 mmol) and KBH4 (800 mg, 14.8 mmol) in dry
methanol (10 ml) were stirred for 15 h. After acidifying with 2 
HCl (10 ml) the solution was extracted with ether (3ϫ 10 ml). The
combined organic layers were washed with water (10 ml) and brine
(10 ml), and dried over MgSO4. Evaporation of the solvent yielded
1.73 g (100%) of the oily alcohol 4b. Ϫ 1H NMR (250 MHz,
CDCl3): δ ϭ 1.48 (d, 3J ϭ 6.4 Hz, 3 H, CH3), 2.76 (d, 3J ϭ 4.8
Hz, 1 H, OH), 3.76 (s, 3 H, OCH3), 3.81 (s, 3 H, OCH3Ј), 5.05 (dq,
3J ϭ 6.4 Hz, 3J ϭ 4.8 Hz, 1 H, CH3CHOH), 6.74 (dd, 3J ϭ 8.7
We thank the Fonds der Chemischen Industrie for providing a
´
Kekule scholarship for R.S. and for financial support.
4
3
5
Hz, J ϭ 2.8 Hz, 1 H, H-4), 6.79 (dd, J ϭ 8.7 Hz, J ϭ 0.8 Hz, 1
H, H-3), 6.94 (dd, 4J ϭ 2.8 Hz, 5J ϭ 0.8 Hz, 1 H, H-6). Ϫ 13C
NMR (62.9 MHz, CDCl3): δ ϭ 23.1 (CH3), 55.8 and 55.9 (OCH3),
Experimental Section
Chromatography: Merck silica gel 60 (63Ϫ200 mesh). Ϫ Elemen-
tal analysis: Microanalytical Laboratory, University of Regensburg. 66.4 (CHOH), 111.6 / 112.4 / 112.5 (C-3, C-4, C-6), 134.9 (C-1),
Ϫ H / 13C NMR: Bruker AW-80 (80 MHz, T ϭ 31°C) and AC-
250 (250 / 62.9 MHz, T ϭ 24°C), TMS as internal standard. Ϫ 167 (95) [Mϩ Ϫ CH3], 152 (24), 139 (88), 137 (32), 124 (33), 43
MS: Finnigan MAT 311 A (EI, 70 eV). Ϫ IR: Beckman IR 4240
(29). Ϫ IR (film): ν (cmϪ1) ϭ 3400 s (OϪH), 2960/2945 s (CϪH),
150.7 (C-2), 154.0 (C-5). Ϫ MS (EI): m/z (%) ϭ 182 (100) [Mϩ],
1
˜
(film between NaCl plates), only characteristic bands are listed. Ϫ 2830 s (OCϪH), 1275/1210/1070/1045/1020 s (CϪO). Ϫ C10H14O3
UV/Vis: Kontron Instruments Spectrophotometer UVIKON 922. (182.22): calcd. C 65.91, H 7.74; found C 65.64, H 7.81.
Ϫ All liquids were distilled and stored under argon. Ϫ η4-1,5-
Cyclooctadiene ϭ cod.
Asymmetric Hydrosilylation of Acetophenone: 10 mg (0.02 mmol)
of [Rh(cod)Cl]2 (0.04 mmol Rh) and ligand (0.2 mmol, if not other-
wise stated) were dissolved in acetophenone (1.0 ml, 8.5 mmol)
under argon. Usually, 2.0 ml of CCl4 were added and the solution
was stirred at r. t. for 30 min. After cooling to 0°C for at least 30
min diphenylsilane (1.6 ml, 8.6 mmol) was added and stirring in the
ice bath, which was warming-up, continued for the period quoted.
Asymmetric Hydrosilylation of 2,5-Dimethoxyacetophenone: 8 mg
(0.016 mmol) of [Rh(cod)Cl]2 (0.032 mmol Rh) and ligand (0.16
mmol, if not otherwise stated) were dissolved in 2,5-dimethoxyace-
tophenone (1.1 ml, 7.0 mmol) under argon. Eventually, 1.7 ml of
CCl4 were added and the solution was stirred at r. t. for 30 min.
After cooling to 0°C for at least 30 min diphenylsilane (1.3 ml, 7.0
mmol) was added and stirring in the ice bath, which was warming-
up, continued for the period quoted.
To determine the amount of silylenol ether, the degree of hydrosi-
lylation and the chemical yield, a sample was taken and a 1H NMR
(CDCl3, 80 MHz) recorded. The following integrals were used: δ ϭ
5.70 ppm (s, SiH, silylenol ether IE), δ ϭ 5.40 (s, SiH, silylalkyl
ether IA), and δ ϭ 2.50 (s, CH3, acetophenone IAP). Calculations:
To determine the amount of silylenol ether, the degree of hydrosi-
lylation and the chemical yield, a sample was taken and a 1H NMR
spectrum (CDCl3, 80 MHz) recorded. The following integrals were
used: δ ϭ 5.43 (s, SiH) and 5.36 (q, CH) together (silylalkyl ether
IA), δ ϭ 5.10 and 4.77 (2 d, CH2) together (silylenol ether IE), and
δ ϭ 2.60 (s, CH3, 2,5-dimethoxyacetophenone IDMAP). Calcu-
lations: silylenol ether [%] ϭ [IE /(IA ϩ IE)]·100, conversion [%] ϭ
[(3 IE ϩ 3 IA)/(3 IA ϩ 3 IE ϩ 2 IDMAP)]·100, and chemical yield
[%] ϭ [3IA/(3IA ϩ 3IE ϩ 2IDMAP) ]·100.
silylenol ether [%] ϭ [IE /(IA ϩ IE)]·100, conversion [%] ϭ [(3 IE
ϩ
3 IA)/(3 IA ϩ 3 IE ϩ IAP)]·100, and chemical yield [%] ϭ [3 IA/(3
IA ϩ 3 IE ϩ IAP)]·100.
Hydrolysis was performed by adding methanol (10 ml) and a few
crystals of p-TosOH. After stirring at r. t. for 30 min the solvents
were evaporated and the residue was distilled in a kugelrohr appa-
ratus at 100Ϫ120°C / ca. 1 Torr. The enantiomeric excess was deter-
mined by injecting 0.4 µl of a solution of the distillate (3Ϫ4 drops)
in 1 ml of CH2Cl2 (Merck Uvasol ) into a Fisons 8130 gas chroma-
tograph. Column: Chrompack Chirasil-DEX CB (l ϭ 25 m, &
osol; ϭ 0.25 mm), integrator: Varian 4290, retention times (118°C):
7.3Ϫ7.7 min [(R)-1-phenylethanol] and 8.0Ϫ8.3 min [(S)-1-phenyl-
ethanol].
Hydrolysis was performed by adding methanol (8 ml) and a few
crystals of p-TosOH. After stirring at r. t. for 30 min the solvents
were evaporated. Ca. 0.1 ml of the residue were purified by chroma-
tography (4Ϫ5 cm SiO2 in a Pasteur pipette) with CH2Cl2 as eluent
(Rf ϭ 0.12). The first fraction (ca. 3 ml) was discarded, the follow-
ing 10 ml were collected and evaporated. The enantiomeric excess
was determined by injecting 0.4 µl of a solution of the resulting oil
(1Ϫ2 drops) in 1 ml of CH2Cl2 (Merck Uvasol ) into a Fisons
8130 gas chromatograph. Column: Chrompack Chirasil-DEX CB
Ƞ
Dedicated to Professor H. Nöth on the occasion of his 70th
birthday.
Eur. J. Inorg. Chem. 1998, 783Ϫ788
787

H. Brunner, R. Störiko
FULL PAPER
[1]
[9]
H. Brunner, H. Nishiyama, K. Itoh, Asymmetric Hydrosilyl-
H. Brunner, A. Kürzinger, J. Organomet. Chem. 1988, 346,
413Ϫ424.
ation in Catalytic Asymmetric Synthesis (Ed.: I. Ojima), VCH
Publishers, New York 1993, 303Ϫ322.
H. Brunner, Transition Metal Catalyzed Reactions in Houben-
[10]
[11]
P. Ramachandran, P. Veeraraghavan, B. Gong, H. C. Brown,
Tetrahedron Lett. 1994, 35, 2141Ϫ2144.
[2]
Weyl, Vol. E 21d, Stereoselective Synthesis (Eds.: G. Helmchen,
R. W. Hoffmann, J. Mulzer, E. Schaumann), Thieme, Stuttgart,
New York 1995, 4074Ϫ4081.
Y.-J. Cherng, J.-M. Fang, T.-J. Lu, Tetrahedron: Asymmetry
1995, 6, 89Ϫ92.
[12]
[13]
[14]
[15]
R. Störiko, Dissertation, Universität Regensburg 1997.
M. Mayen, E. Marechal, Bull. Soc. Chim. Fr. 1972, 4662Ϫ4681.
H. Brunner, C. Huber, Z. Naturforsch. 1991, 43b, 1145Ϫ1148.
H. Brunner, P. Brandl, Tetrahedron: Asymmetry 1991, 2,
919Ϫ930.
[3]
H. Brunner, W. Zettlmeier, Handbook of Enantioselective Ca-
talysis, Vol. I & II, VCH, Weinheim 1993.
[4]
H. Brunner, U. Obermann, Chem. Ber. 1989, 122, 499Ϫ507.
[5]
G. Balavoine, J. C. Client, I. Lellouche, Tetrahedron Lett. 1989,
[16]
30, 5141Ϫ5144.
See for example: L. J. Andrews, R. M. Keefer, Molecular Com-
plexes in Organic Chemistry, Holden-Day, San Francisco, Lon-
don, Amsterdam 1964.
[6]
H. Nishiyama, H. Sakaguchi, T. Nakamura, M. Horihata, M.
Kondo, K. Itoh, Organometallics 1989, 8, 846Ϫ848.
[7]
[17]
[18]
H. Nishiyama, M. Kondo, T. Nakamura, K. Itoh, Organomet-
R. Matsushima, Y. Hiramatsu, T. Kitanishi, J. Chem. Soc., Per-
kin Trans. 2 1988, 847Ϫ850.
allics 1991, 10, 500Ϫ508.
[8]
H. Brunner, U. Obermann, P. Wimmer, J. Organomet. Chem.
Determined by analysis of a mixture consisting of ca. 25% 3b
and ca. 75% 2b.
1986, 316, C1ϪC3.
[97225]
788
Eur. J. Inorg. Chem. 1998, 783Ϫ788