Designing the "Search pathway" in the development of a new class of highly efficient stereoselective hydrosilylation catalysts

Article abstract of DOI:10.1002/chem.200500132

A highly efficient class of N-heterocyclic carbene-based catalysts for the asymmetric hydrosilylation of prochiral ketones was discussed. The potential pitfalls associated with the nonlinearity in the behavior of an enantioselective catalyst in response to the variations of key reaction parameters were also described. The modular design of a new catalyst is based essentially on a single assembly step of its structural and functional subunits. It was observed that the direct coupling of oxazolines and N-heterocyclic carbenes leads to chelating C,N ancillary ligands for asymmetric catalysis that combines both an 'anchor' unit and a stereodirecting element.

Full text of DOI:10.1002/chem.200500132

Designing the “Search Pathway” in the Development of a New Class of
Highly Efficient Stereoselective Hydrosilylation Catalysts
Vincent Cꢀsar,[a] Stꢀphane Bellemin-Laponnaz,*[a] Hubert Wadepohl,[b] and
Lutz H. Gade*[a, b]
Dedicated to Professor Brian F. G. Johnson
Abstract: The direct coupling of oxazo-
lines and N-heterocyclic carbenes leads
to chelating C,N ancillary ligands for
asymmetric catalysis that combine both
an “anchor” unit and a stereodirecting
element. Reacting various N-substitut-
ed imidazoles with 2-bromo-4(S)-tert-
butyl- and 2-bromo-4(S)-isopropyloxa-
zoline gave the imidazolium precursors
of the stereodirecting ancillary ligands.
A library of ten different ligand precur-
sors was obtained by using this simple
procedure (65–97% yield). These pro-
tioligands were metalated in a subse-
quent step by reaction with [{Rh(m-
OtBu)(nbd)}2] (nbd=norbornadiene),
generated in situ from KOtBu and
[{RhCl(nbd)}2] giving the correspond-
ing N-heterocyclic carbene complexes
[RhBr(nbd)(oxazolinyl-carbene)] 4a–j
in good yields. X-ray diffraction studies
of two of the rhodium complexes, 4d
and 4j, established a distorted square-
pyramidal coordination geometry with
the bromo ligand occupying the apical
position. The rhodium–carbene bond
length was found to be 2.070(4) ꢀ (4d)
and 2.012(3) ꢀ (4j). Complexes 4a–j
were treated with AgBF4 in dichloro-
methane, giving the active cationic
square-planar catalysts for the hydrosi-
lylation of ketones. As a reference re-
action for the catalyst optimisation, the
hydrosilylation of acetophenone with
diphenylsilane was studied and the
system optimised with respect to the
ing the catalyst (4d) and substrate con-
centrations at À558C established a rate
law for the initial conversion which is
first-order in both substrates as well as
the
catalyst
(Vi =k[4][PhCOMe]-
[Ph2SiH2]). The catalytic system de-
rived from complex 4d was found to
afford high yields and good enantiose-
lectivities in the reduction of various
aryl alkyl ketones (acetophenone: 92%
isolated yield and 90% ee, 2-naphtyl
methyl ketone: 99% yield, 91% ee).
The selectivity for the reduction of pro-
chiral dialkyl ketones is comparable or
even superior to the best previously re-
ported for prochiral nonaromatic ke-
tones; whereas cyclopropyl methyl
ketone is hydrosilylated with an enan-
tioselectivity of 81% ee, the increase of
the steric demand of one of the alkyl
groups leads to improved eeꢁs, reaching
95% ee in the case of tert-butyl methyl
ketone. Linear chain n-alkyl methyl ke-
tones, which are particularly challeng-
ing substrates, are reduced in good
asymmetric induction, such as 2-octa-
none (79% ee) and even 2-butanone
(65% ee).
À
counterion (BF4 ), solvent (THF) and
the silane reducing agent (diphenylsi-
lane). The reaction product (1-phenyle-
thanol) was obtained with the highest
enantiomeric excess (ee) by carrying
out the reaction at À608C, whilst the
enantioselectivity drops upon going
both to lower and higher temperatures.
The observation that the temperature
dependence of the ee values goes
through a maximum indicated a change
in the rate-determining step as the tem-
perature is varied. The determination
of the initial reaction rate in the hydro-
silylation of acetophenone upon vary-
Keywords: carbene ligands · hydro-
silylation · N ligands · oxazolines ·
rhodium
[b] Prof. H. Wadepohl, Prof. L. H. Gade
[a] Dr. V. Cꢂsar, Dr. S. Bellemin-Laponnaz, Prof. L. H. Gade
Laboratoire de Chimie Organomꢂtallique
et de Catalyse (UMR 7513)
Anorganisch-Chemisches Institut, Universitꢃt Heidelberg
Im Neuenheimer Feld 270, 69120 Heidelberg (Germany)
Fax : (+49)6221-545-609
Institut Le Bel, Universitꢂ Louis Pasteur
4 rue Blaise Pascal, 67070 Strasbourg (France)
Fax : (+33)390-241-531
E-mail: lutz.gade@uni-hd.de
E-mail: bellemin@chimie.u-strasbg.fr
gade@chimie.u-strasbg.fr
2862
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200500132
Chem. Eur. J. 2005, 11, 2862 – 2873

FULL PAPER
Introduction
imidazoles (N), which display nucleophilic reactivity, with 2-
bromooxazolines (E) to give the imidazolium precursors
(N–E) of the stereodirecting ancillary ligands. These were
metalated in a subsequent step by reaction with alkoxyrho-
dium complexes as first described by Herrmann et al.[4a]
The development of catalysts for new chemical transforma-
tions or catalytic systems with improved performance for
known reactions is one of the key challenges of current
chemical research. In practical terms, the quest for success-
ful strategies to solve a given problem occupies centre stage.
In a recent perceptive article on the limits of rationality in
catalyst development, Hoveyda et al. have pointed out that
efficient routes to novel catalysts may be based on the
design of a search pathway rather than the preconception of
an individual catalyst structure derived from a complete
mechanistic analysis of a catalytic process.[1] Such search
strategies are conveniently based on highly modular catalyst
systems, their basic structural motifs being derived from the
available mechanistic insight into the catalytic process com-
bined with the known ligand properties of its components.
In an optimised “search algorithm”, the ancillary (stereo-
directing) ligands are assembled in a single step from readily
accessible building blocks. The ligand assembly is followed
by a metallation step giving the (pre)catalyst. The coordina-
tion of the ancillary ligand to the metal should be kinetically
inert in order to give a relatively well-defined active system.
Its structure should therefore combine both an “anchor”
unit and—in the case of asymmetric catalysis—a stereodir-
ecting element.
Preparation of the components for the ligand-coupling step
and assembly of the stereodirecting ancillary ligands: As in-
dicated, 2-bromooxazolines serve as the electrophiles in the
preparation of the imidazolium salts from the respective
imidazole precursors. Following the procedure reported by
Meyers and Novachek, the reaction of the lithiated oxazo-
line (tBuLi, THF, À788C) with 1,2-dibromotetrafluoro-
ethane gave the corresponding 2-bromo-4(S)-isopropyl oxa-
zoline (1a) and 2-bromo-4(S)-tert-butyl oxazoline (1b).[10,11]
Both reagents were isolated by direct bulb-to-bulb distilla-
tion from the reaction mixtures giving solutions in THF
(usually 75–90% w/w).
N-heterocyclic carbenes[2] are excellent “anchor” units for
late transition metals, which form strong metal–carbon
bonds and have thus been widely used in homogeneous cat-
alysis.[2–4] They are straightforward to synthesise, often
chemically more stable than the extensively employed phos-
phines and may be readily combined with other ligating
units, in particular, by the appropriate functionalisation of
the N-atoms in their heterocyclic structures.[4] We recently
reported the direct coupling of oxazolines and N-heterocy-
clic carbenes,[5] which are modelled on previously studied
heterodonor–phosphine stereodirecting ligands for d8-MI/d6-
MIII catalysis.[6] Their highly modular assembly and metala-
tion is designed according to the search pathway delineated
above and was thought to provide an efficient strategy to
develop novel chiral catalysts.[7] In this paper we provide a
full account of the “search pathway” leading to a new,
highly efficient class of N-heterocyclic carbene-based cata-
lysts for the asymmetric hydrosilylation of prochiral ke-
tones.[8,9] We also present the potential pitfalls associated
with the nonlinearity in the behaviour of an enantioselective
catalyst in response to the variation of key reaction parame-
ters. These observations are due to the complexity of the re-
action mechanism.
The imidazoles used for this study are depicted above, all
of which were prepared by the established synthetic proce-
dures.[12–14] The imidazolium salts, which served as “protio”
ligand precursors, were prepared by direct coupling of the 2-
bromooxazolines and the imidazoles in THF at room tem-
perature. The imidazolium salts precipitated as white pow-
ders, which are indefinitely stable under an atmosphere of
nitrogen and were analytically and spectroscopically pure. A
library of ten different ligand precursors was obtained by
using this simple procedure (65–97% yield; Table 1).
Results and Discussion
Library design: As mentioned above, the ideal strategy for
the modular design of a new catalyst is based on essentially
a single assembly step of its structural and functional sub-
units. To this end we directly coupled various N-substituted
Preparation and structure of the rhodium complexes: The
reaction of 3a–i with [{Rh(m-OtBu)(nbd)}2] (nbd=norborna-
diene), generated in situ from KOtBu and [{RhCl(nbd)}2],
Chem. Eur. J. 2005, 11, 2862 – 2873
www.chemeurj.org
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2863

L. H. Gade, S. Bellemin-Laponnaz et al.
Table 1. Synthesis of the imidazolium salts 3a–j and the rhodium com-
plexes 4a–j.
3,4
R’
R
3,4
R’
R
a
b
c
d
e
Ph
o-tol
mes
mes
tBu
tBu
iPr
tBu
tBu
f
2-tBuC6H4
CH2Ph
CHPh2
fluorenyl
CH(napht)2
tBu
tBu
tBu
tBu
tBu
g
h
i
2,6-(iPr)2Ph
j
yielded the corresponding N-heterocyclic carbene complexes
4a–j in good yields (above 80%), except for 4i in which the
concurrent deprotonation of the acidic proton of the fluo-
renyl group lead to partial decomposition (Table 1).[15] All
complexes were purified by recrystallisation from dichloro-
methane/pentane, characterised by elemental analysis and
NMR spectroscopy, and were found to be stable in air. Par-
ticularly diagnostic for the formation of the carbene com-
plexes was the observation of the 13C NMR resonances of
the carbene carbon nuclei,[4,5] at d=184–191 ppm with cou-
pling constants J(Rh,C) of approximately 55 Hz. Further-
more, a significant oxazoline n(C=N) vibrational band shift
to lower wave numbers (of about 20 cmÀ1) indicated the co-
ordination of the oxazolinyl unit to the metal centre.
Two of rhodium complexes, compounds 4d and 4j, were
characterised by X-ray diffraction, and their molecular
structures are depicted in Figure 1 along with the principal
bond lengths and angles. Both complexes possess distorted
square-pyramidal coordination geometries with the bromo
ligand occupying the apical position. Given this coordination
geometry and the arrangement of the ligands, the observa-
tion of two diastereomers was in principle possible. Howev-
er, both in solution and in the solid state, only one diaste-
reoisomer was observed. In the molecular structures of 4d
and 4j the bromo ligand was found to adopt the same orien-
tation relative to the tert-butyl substituent of the ancillary
ligand, probably due to steric repulsion between the latter
and the norbornadiene ligand. As observed for related com-
pounds, the mesityl ring in compound 4d is oriented almost
orthogonally to the imidazolyl ring (dihedral angle: C10-N3-
C11-C12 81.68).[5] The rhodium–carbene bond length was
found to be 2.070(4) ꢀ and the trans influence of the N-het-
Figure 1. Top: Molecular structure of complex 4d. Selected bond lengths
À
À
À
(ꢀ) and angles (8): Rh Br, 2.6775(5); Rh N1, 2.285(3); Rh C10
À
À
2.001(4); Rh C20/C21, 2.231(4)/2.231(4); Rh C23/C24, 2.084(4),
2.070(4); N1-Rh-Br, 91.72(9); N1-Rh-C10, 77.0(1); N1-Rh-C20, 113.6(2);
N1-Rh-C21, 95.4(2); N1-Rh-C23, 128.3(2); N1-Rh-C24, 168.3(2); C10-
Rh-Br, 96.5(1); C20/C24-Rh-Br, 84.8(1)/99.9(1); C21/C23-Rh-Br,
116.4(1)/139.9(1); C20-Rh-C24, 66.4(2); C21-Rh-C23, 65.1(2). Bottom:
Molecular structure of complex 4j. Selected bond lengths (ꢀ) and angles
À
À
À
À
(8): Rh Br, 2.6914(4); Rh N1, 2.265(3); Rh C10, 2.012(3); Rh C32/C33,
À
À
2.097(3)/2.069(3); Rh C35/C36, 2.218(3)/2.217(3); C32 C33, 1.435(5);
À
C35 C36, 1.384(5); N1-Rh-Br, 89.97(7); N1-Rh-C10, 77.4(1); N1-Rh-
C32, 126.2(1); N1-Rh-C33, 166.2(1); N1-Rh-C35, 112.1(1); N1-Rh-C36,
92.5(1); C10-Rh-Br, 90.76(8); C32/C36-Rh-Br, 143.8(1)/119.99(9); C33/
C35-Rh-Br, 103.6(1)/88.58(9); C32-Rh-C36, 65.1(1); C33-Rh-C35,
66.6(1).
À
In this case, the C H bond of the di(1-naphthyl)methyl
group points towards the metal centre and the naphthyl
groups arrange themselves in a way which minimises the in-
teractions with the norbornadiene. The C-Rh-N bite angles
in the two molecular structures found for the complexes 4d
and 4j are 77.0(1) and 77.4(1)8, respectively. This reflects
the rigidity of this family of oxazolinyl–carbene ligands,
which impose a defined geometry upon coordination to a
metal centre.
À
erocyclic carbene is manifested in the difference of Rh C
bond lengths to the C=C units in the norbornadiene (see se-
lected bond data in legend of Figure 1).
In the 1H NMR spectra of complexes 4a–j, recorded at
258C, only two signals for the four norbornadiene olefin
protons are observed, indicating fast chemical exchange at
that temperature. In the light of our previous studies with
Basically the same structural features were observed for
compound 4j (rhodium–carbene bond length: 2.012(3) ꢀ).
2864
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2862 – 2873

Hydrosilylation Catalysts
FULL PAPER
achiral oxazolinylcarbene rhodium complexes, the dynamic
processes observed in solution are due to a sequence of
Berry pseudorotations combined with a fast bromide disso-
ciation and association mechanism.[5b] The latter equilibrates
the two diastereoisomers (derived from the arrangement of
the ligands around the rhodium centre) very rapidly in so-
1
lution and thus leads to average H and 13C NMR spectra of
the two forms.
À
While rotation around the CAryl NImidazoyl bond in imidazo-
lium salts 3a–f is nonhindered and rapid on the NMR time-
scale at room temperature, complexation to the “RhBr-
(nbd)” fragment blocks this rotation for 4c–e at 258C, lead-
ing to the observation of, for example, two sets of signals for
the ortho-methyl (4c and 4d) and isopropyl groups (4e;
Figure 2).
Complexes 4b and 4 f possess ortho-monosubstituted aryl
substituents, which may adopt two inequivalent orientations,
thus allowing the existence of two atropisomeric conformers
in solution. The two conformers of 4 f (with a bulky tert-
butyl group as substituent) are present in a ratio of 77:23 at
258C. This ratio is obtained under kinetic control during the
formation of the complexes and then frozen, since the acti-
vation barrier for the interconversion of the conformers is
too high to allow their equilibration even at higher tempera-
tures (below the decomposition point of the complex in so-
lution). In contrast, the two atropisomeric forms of 4b inter-
convert rapidly on the NMR timescale at 258C, leading to a
single average signal at d=2.17 ppm of the ortho-methyl
group in the two exchanging species (Figure 3). At À58C
this resonance coalesces and two signals at d=2.38 and
2.03 ppm are observed at À518C. In the slow exchange
regime below this temperature the ratio of the two atrop-
isomers corresponds to an apparent equilibrium constant of
1.4.
Figure 3. Variable temperature 1H NMR spectrum of complex 4b in the
region 2.5–1.8 ppm (recorded in CD2Cl2, 300 MHz).
none with diphenylsilane at 258C. Complexes 4a–j were
used in the presence of AgBF4 in dichloromethane, giving
the active cationic square-planar catalysts. With a catalyst
loading of 1.0 mol% at room temperature, satisfactory iso-
lated yields of the secondary alcohol were obtained in all
cases (from 70% for the catalytic conversion with complex
4i to 93% with complex 4j). Figure 4 summarises the enan-
tioselectivities that were obtained for all complexes under
these standardised reaction conditions.
Not unexpectedly, the level of enantioselectivity observed
is extremely dependent upon the substituents and there are
several notable aspects. Whereas a phenyl substituent on the
imidazole ring of the catalyst derived from 4a gives racemic
1-phenylethanol, the use of the bulkier mesityl substituent
(in 4d) results in an enantiomeric excess of 65%. The use of
an aryl group as imidazolidine substituent gives the products
with absolute S configuration (given the S configuration at
Hydrosilylation of ketones
Catalyst optimisation: As a reference reaction for the cata-
lyst optimisation we chose the hydrosilylation of acetophe-
À
Figure 2. Schematic representation of the influence of the aryl group on the rotation NIm CAr in complexes 4a–f.
Chem. Eur. J. 2005, 11, 2862 – 2873
www.chemeurj.org
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2865

L. H. Gade, S. Bellemin-Laponnaz et al.
Figure 4. Enantioselectivity data for the hydrosilylation of acetophenone with diphenylsilane with a catalyst loading of 1.0 mol% in the presence of
AgBF4 (experimental conditions: CH2Cl2, room temperature, 5 h then hydrolysis with K2CO3 in methanol).
À
the substituent position in the oxazoline units). However, re-
placement of the aryl by an arylmethyl substituent at the
NHC-unit, as in catalysts 4g–i, the R enantiomer was ob-
tained. The only exception to this observation is the deriva-
tive with the sterically very demanding di(1-naphthyl)methyl
group (4j), yielding again the S-configured product. Finally,
the results obtained with the two oxazolines employed in
the ligand assembly established a higher selectivity for the
derivatives with the tert-butyl groups relative to those with
isopropyl substituents in the 4-position of the oxazoline. As
a result of this first screening, we selected precatalyst 4d for
further studies of the catalyst performance.
OTsÀ >BPh4À >PF6À >OTfÀ >B(C6F5)4
,
whereas CH2Cl2
was found to give the best results among various polar sol-
vents that were tested. In several previous studies on asym-
metric hydrosilylation catalysis it was found that high enan-
tioselectivity can be afforded by using diarylsilanes that con-
tain sterically demanding aryl groups.[8c–e,17] In the case at
hand, the enantioselectivity was found to be dependent on
the choice of the silane (Table 2). On the one hand, methyl-
phenylsilane provided a nearly racemic product (3% ee;
entry 1), whilst the use of diarylsilanes with greater steric
demand than diphenylsilane equally led to a marked de-
crease in enantioselectivity (entries 4–6). Furthermore, the
activity of the catalyst was significantly lower with these
sterically crowded silanes.
The effect of counterion and the choice of solvent on the
catalyst selectivity was investigated for the hydrosilylation
of acetophenone by using the cationic catalyst derived from
4d with a slight excess of various silver salts.[16] A slight de-
Temperature dependence of the catalyst selectivity and its im-
plications for the mechanism of stereoselection: Using the
À
crease in enantioselectivity is observed in the series BF4
>
2866
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2862 – 2873

Hydrosilylation Catalysts
FULL PAPER
Table 2. Hydrosilylation of acetophenone with catalyst 4d as a function
of the silane.
inverse absolute temperature, is displayed alongside
(Figure 5, bottom).
If the selectivity determining step had remained the same
over the whole temperature range, one would have expected
a linear relationship of ln(S/R) versus TÀ1 and the observa-
tion that the temperature dependence of the ee values goes
through a maximum indicates a change in the selectivity de-
termining step as the temperature is varied. Such behaviour,
occasionally referred to as an isoinversion relationship, has
been observed inter alia in the asymmetric dihydroxylation
of olefins,[19] the diastereoselective reduction of 2-tert-butyl-
cyclohexanone[20] and the photochemically induced Paterno–
Bꢅchi reaction of chiral phenyl glycoxalates with cyclic ole-
fins.[21] Recently, Scharf and co-workers observed such a
nonlinear temperature dependence for the enantioselective
hydrosilylation of ketones using chiral cyclic monophos-
phonite rhodium catalysts.[22]
In a detailed analysis of this phenomenon, Hale and Ridd
concluded that the observation of a maximum in a ln(S/R)
versus TÀ1 plot may be taken as evidence for two selectivity
determining stages in a reaction.[23] In our case, the low-tem-
perature slope may be identified with the enthalpic and en-
tropic discrimination involved in that regime (DDH°
ꢀ15 kJmolÀ1, DDS° ꢀ96 JmolÀ1 KÀ1), while the region on
the side of the maximum towards higher temperature may
be viewed as representing a (rather broad) transitional
regime with enthalpic and entropic characteristics of a
second rate-determining step as the limit. As for hydrosilyla-
tion catalysis, the observations are consistent with the mech-
anistic scheme put forward by Ojima[24] in which the reversi-
ble enantioface differentiating coordination of the ketone to
the hydridorhodium complex and the subsequent irreversi-
Entry
Silane
ee [%]
1
2
3
4
5
6
MePhSiH2
Ph2SiH2
(p-tol)PhSiH2
(o-tol)PhSiH2
(a-naphtyl)PhSiH2
(o-tol)2SiH2
3
65
58
À10
À7
À12
optimised catalyst and the reaction conditions identified in
the systematic study described above, the temperature de-
pendence of the catalyst performance was determined in the
range of À78 to +258C. Interestingly, we found that the re-
action product (1-phenylethanol) is obtained with the high-
est ee by carrying out the reaction at À608C, whilst the
enantioselectivity drops upon going both to lower and
higher temperatures as is shown in Figure 5 (top). As alter-
native way to represent this result, an Eyring diagram,[18]
plotting ln(S/R) (S/R=product enantiomer ratio) versus the
À
ble insertion of the carbonyl into the Si Rh bond
(Scheme 1) provide two stages at which the stereochemical
outcome of the hydrosilylation can be determined.
Scheme 1. Schematic presentation of the two stages (A and B) that deter-
mine the stereochemistry of the product.
Figure 5. Top: Temperature dependence of the asymmetric hydrosilyla-
tion of acetophenone in the presence of system 4d/AgBF4 as catalyst
(1.0 mol% in CH2Cl2) (ee vs temperature); Bottom: Eyring diagram
(ln(S/R) vs TÀ1).
The determination of the initial reaction rate in the hy-
drosilylation of acetophenone upon varying the catalyst
(4d) and substrate concentrations at À558C established a
Chem. Eur. J. 2005, 11, 2862 – 2873
www.chemeurj.org
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2867

L. H. Gade, S. Bellemin-Laponnaz et al.
Table 3. Catalytic asymmetric hydrosilylation of aryl alkyl ketones with
catalyst 4d.
rate law for the initial conversion that is first-order in both
substrates as well as the catalyst (Vi =k[4d][PhCOMe]-
[Ph2SiH2]).[25]
The first-order dependence on the catalyst concentration
was only observed at catalyst loadings below 1.0ꢆ
10À2 molLÀ1. At higher catalyst loadings, a marked nonlinear
behaviour of the initial rate was observed which was accom-
panied by a similar concentration profile for the enantio-
meric excess (Figure 6). We attribute this behaviour to cata-
lyst aggregation at higher concentrations, accompanied by a
change of colour of the reaction solutions from yellow to
deep orange upon increasing the concentration of the rhodi-
um catalyst.
Entry
1
Ketone
ee [%][a]
Yield [%]
92
90
2
3
91
75
99
90
4
5
86
70
82
94
6
7
8
9
88
91
78
85
92
90
88
84
Figure 6. Kinetics of the hydrosilylation of acetophenone at varying con-
~
centrations of catalyst ( ) among with the enantiomeric excess of the
*
product ( ).
10
11
81
89
80
93
The “proof of the pudding…”—hydrosilylation of ketones
with catalyst 4d: The key to the high enantioselectivity ob-
served in the hydrosilylation of ketones with the cationic
active catalyst derived from complex 4d is its remarkably
high catalytic activity, which permits its application at the
optimised temperature of À608C with acceptable reaction
times and excellent yields. Under these conditions the enan-
tioselectivity is generally good for aryl/alkyl ketones and
truly remarkable in the case of the reduction of prochiral di-
alkyl ketones (vide infra).
between 78 and 91% were found (entries 6–9), while cyclic
aromatic ketones were reduced with equally good enantiose-
lectivity (entries 10, 11).
Hydrosilylation of aromatic ketones: Under the optimised
conditions described above, the cationic catalyst derived
from complex 4d was found to afford high yields and good
enantioselectivities in the reduction of various aryl alkyl ke-
tones (Table 3). Acetophenone is reduced in 92% isolated
yield and 90% ee (entry 1), while 2-naphthyl methyl ketone
is isolated in 99% yield with an ee of 91% (entry 2). We
found that ortho-substitution of the aryl substituent resulted
in lower enantioselectivities (entries 3, 4) and a similar trend
is observed for substrates with bulkier alkyl groups at the
CO unit (entry 5). Various derivatives with electronically
different substituents have been investigated for which eeꢁs
Hydrosilylation of dialkyl ketones: For the hydrosilylation of
dialkyl ketones it has previously proved to be more difficult
to obtain high enantioselectivity.[8] Table 4 summarises the
results obtained with various nonaromatic ketones in the
presence of catalytic system derived from 4d and the opti-
mised experimental conditions. Whereas cyclopropyl methyl
ketone is hydrosilylated with an enantioselectivity of
81% ee (entry 1),[26] the increase of the steric demand of
one of the alkyl groups leads to improved eeꢁs, reaching
95% ee in the case of tert-butyl methyl ketone (entries 1–4).
Linear chain n-alkyl methyl ketones, which are particularly
challenging substrates, are reduced in good asymmetric in-
2868
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2862 – 2873

Hydrosilylation Catalysts
FULL PAPER
Table 4. Catalytic asymmetric hydrosilylation of dialkyl ketones with cat-
alyst 4d.
Experimental Section
All manipulations were performed under an inert atmosphere of dry ni-
trogen by using standard Schlenk techniques or by working in a glove
box. THF and diethyl ether were distilled from sodium/benzophenone.
Pentane was distilled from a sodium/potassium alloy, dichloromethane
was dried over CaH2 and subsequently distilled. H and 13C NMR spectra
1
Entry
1
Ketone
ee [%]
Yield [%]
63[a]
were recorded on a Bruker Avance 300 NMR spectrometer at 300 MHz
and 75 MHz and were referenced using the residual proton solvent peak
(1H) or carbon resonance (13C). Infrared spectra were obtained on a FT-
IR Perkin Elmer 1600 spectrometer and the mass spectra were recorded
by the “service de spectromꢂtrie de masse de lꢁUniversitꢂ Louis Pasteur”
on an Autospec HF mass spectrometer. The elemental analyses were per-
formed by the analytical services of the Strasbourg and Heidelberg
Chemistry Departments. [{RhCl(nbd)}2] was prepared according to a lit-
erature procedure.[27] (1-Naphth)PhSiH2 and (o-Tol)PhSiH2 were pre-
pared by reaction of (1-Naphth)MgBr or (o-Tol)MgBr with PhSiCl3 in
THF and further reduction with LiAlH4 in the same flask.[28] (o-Tol)2SiH2
was prepared by using HSiCl3 as the starting chlorosilane. Potassium tert-
butoxide was sublimed prior to use. RhCl3·3H2O was provided by BASF
AG (Ludwigshafen) and tert-leucine by Degussa AG. All other reagents
were commercially available and used as received.
81
2
3
88
89
53[a]
98
4
5
95
77
70[a]
97
Preparation of the ligands
(4S)-2-Bromo-4-isopropyloxazoline (1a): tBuLi (21.7 mL, 1.7m in pen-
tane, 36.8 mmol) was added dropwise to a solution of (4S)-4-isopropylox-
azoline (3.79 g, 33.5 mmol) in anhydrous THF (100 mL) at À788C over
5 min. The resulting yellow solution was then stirred for an additionnal
15 min prior to the addition of 1,2-dibromo-1,1,2,2-tetrafluoroethane
(4.84 mL, 36.8 mmol). The solution was then allowed to warm to ambient
temperature overnight, and was concentrated to about 15 mL. The
brownish mixture was purified by a short bulb-to-bulb distillation to yield
a colorless solution of the expected bromooxazoline in THF (concentra-
6
7
8
79
74
65
95
88[b]
n.d.[b]
1
tion of about 77% w/w) (4.6 g, 55%). H NMR (CDCl3): d=4.45 (dd, J=
[a] Moderate yield due to the volatility of the product. [b] Reaction car-
8.3, 9.7 Hz, 1H; CH2), 4.15 (pseudo-t, J=8.2 Hz, 1H; CH2), 3.96 (m, 1H;
CHoxa), 1.80 (m, 1H; CH(CH3)), 0.99 (d, J=6.7 Hz, 3H; CH(CH3)2),
0.91 ppm (d, J=6.7 Hz, 3H; CH(CH3)2); 13C {1H} NMR (CDCl3): d=
141.7 (NCO), 73.2 (CH2oxa), 72.9 (CHoxa), 32.5 (CH(CH3)2), 18.4,
18.1 ppm (CH(CH3)2).
ried out at À408C.
duction (entries 5–8), such as 2-octanone (95% isolated
yield, 79% ee, entry 6) and even for the extreme case of 2-
butanone (65% ee, entry 8).[27] The results displayed in
Table 4 are comparable or even superior to the best previ-
ously reported for prochiral nonaromatic ketones.[8d,28]
(4S)-2-Bromo-4-tert-butyloxazoline (1b): tBuLi (10.8 mL, 1.7m in pen-
tane, 18.3 mmol) was added dropwise to a solution of (4S)-4-tert-butylox-
azoline (2.11 g, 16.6 mmol) in anhydrous THF (50 mL) at À788C over
5 min. The resulting yellow solution was then stirred for an additionnal
15 min prior to the addition of 1,2-dibromo-1,1,2,2-tetrafluoroethane
(2.4 mL, 18.3 mmol). The solution was then allowed to warm to ambient
temperature over night, and was concentrated to about 5 mL. The light
brownish mixture was purified by a short bulb-to-bulb distillation to yield
a colorless solution of the expected bromooxazoline in THF (concentra-
tion of about 88% w/w) (2.25 g, 66%). 1H NMR (CDCl3): d=4.38 (dd,
J=8.4, 10.0 Hz, 1H; CH2), 4.24 (dd, J=8.0, 8.4 Hz, 1H; CH2), 3.92 (dd,
J=8.0, 10.0 Hz, 1H; CHoxa), 0.92 ppm (s, 9H; C(CH3)); 13C {1H} NMR
(CDCl3): d=141.4 (NCO), 76.3 (CH), 71.6 (CH2), 35.6 (C(CH3)3),
25.5 ppm (C(CH3)3).
Conclusion
In this work we have devised a successful search strategy for
a new class of rhodium-based highly stereoselective hydrosi-
lylation catalysts. By choice of a modular catalyst design the
rapid identification of a system with excellent catalytic activ-
ity and enantioselectivity was achieved. The process of opti-
misation revealed in particular a complex dependence of the
catalyst performance on the temperature and the catalyst
loading. This is, inter alia, indicative of mechanistic com-
plexity of the system which defies a preconceived “design”
approach in its true sense. In view of the simplicity of the
ligand and catalyst assembly, the combination of similar
building blocks for the development of catalysts for other
organic transformations appears to be promising. Such work
is currently under way.
Imidazolium salts—general procedure: The imidazole (1–5 mmol) was
added to a concentrated solution of the bromooxazoline (1.15 equiv) in
THF (c>2m) and the solution was stirred during 2–5 days at ambient
temperature. During the course of this period, the mixture became solid
and had to be crushed with a spatula. The colourless solid was washed
several times with Et2O (2 mLmmolÀ1) and dried in vacuo.
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-phenylimidazolium bromide
(3a): Yield 70%; 1H NMR (CDCl3): d=10.47 (pseudo-t, 4J=1.6 Hz, 1H;
NCHN), 8.38 (dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 8.04 (m, 2H;
CHPh),7.97 (dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 7.50 (m, 3H;
CHPh), 4.74 (dd, 2J=8.8 Hz, 3J=9.8 Hz, 1H; CH2oxa), 4.54 (pseudo-t, J=
8.6 Hz, 1H; CH2oxa), 4.12 (dd, 3J=8.5 Hz, 3J=9.8 Hz, 1H; CHoxa),
0.9 ppm (s, 9H; C(CH3)3); 13C {1H} NMR (CDCl3): d=149.3 (NCO),
135.0, 133.9 (N2C, CPh), 131.0 (CHPh), 130.5 (CHPh), 123.6 (CHim), 122.9
(CHPh), 120.8 (CHim), 75.1 (CHoxa), 73.1 (CH2oxa), 33.9 (C(CH3)3),
Chem. Eur. J. 2005, 11, 2862 – 2873
www.chemeurj.org
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2869

L. H. Gade, S. Bellemin-Laponnaz et al.
24.9 ppm (C(CH3)3); MS (FAB): m/z (%): 270 (100) [M+ÀBr]; FT-IR
(KBr): n˜ =1694 cmÀ1 (s, C=N); elemental analysis calcd (%) for
C16H20BrN3O (350.26): C 54.87, H 5.76, N 12.00; found: C 54.67, H 5.64,
N 11.93.
(C(CH3)3oxa), 32.0 (C(CH3)3ortho), 25.8 ppm (C(CH3)3oxa); MS (ESI): m/z
(%): 326.22 (100) [M+ÀBr], 327.22 (22) [M+ÀBr+H]; FT-IR (KBr): n˜ =
1695.1 cmÀ1 (s, C=N).
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-benzylimidazolium bromide
(3g): Yield: 91%; 1H NMR (CDCl3): d=10.91 (brs, 1H; NCHN), 7.78
(brs, 1H; CH4/5-Im), 7.71–7.66 (m, 3H; CH4/5-Im, 2CHPh), 7.38–7.35 (m,
3H; CHPh), 6.05 (s, 2H; CH2Ph), 4.67 (dd, 2J=8.9 Hz, 3J=9.8 Hz, 1H;
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-(2-tolyl)imidazolium
bro-
mide (3b): Yield 84%; 1H NMR (CDCl3): d=10.03 (pseudo-t, 4J=
3
4
1.6 Hz, 1H; NCHN), 8.06 (dd, J=2.1 Hz, J=1.6 Hz, 1H; CH4/5-Im), 7.94
(dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 7.82 (m, 1H; CHtol), 7.44–7.26
(m, 3H; CHtol), 4.73 (dd, 2J=8.7 Hz, 3J=9.9 Hz, 1H; CH2oxa), 4.54
(pseudo-t, J=8.6 Hz, 1H; CH2oxa), 4.12 (dd, 3J=8.5 Hz, 3J=9.8 Hz, 1H;
CHoxa), 2.31 (s, 3H; CH3tol), 0.96 ppm (s, 9H; C(CH3)3); 13C {1H} NMR
(CDCl3): d=149.3 (NCO), 136.7 (N2C), 133.3 (Ctol), 133.2(Ctol), 132.0
(CHtol), 131.6 (CHtol), 127.9 (CHtol), 127.7 (CHtol), 125.8 (CHim), 120.6
(CHim), 75.1 (CHoxa), 73.1 (CH2oxa), 33.9 (C(CH3)3), 25.8 (C(CH3)3),
18.01 ppm (CH3tol); MS (ESI): m/z (%): 284.17 (100) [M+ÀBr], 285.17
(13) [M+ÀBr+H]; FT-IR (KBr): n˜ =1696.4 cmÀ1 (s, C=N).
3
3
CH2oxa), 4.47 (pseudo-t, J=8.7 Hz, 1H; CH2oxa), 4.08 (dd, J=8.6 Hz, J=
9.8 Hz, 1H; CHoxa), 0.90 ppm (s, 9H; C(CH3)3); 13C {1H} NMR (CDCl3):
d=149.2 (NCO), 136.7 (N2C), 132.7 (CPh), 129.7 (CHPh), 129.7 (CHPh),
129.5 (CHPh), 123.6 (CHim), 119.5 (CHim), 75.0 (CHoxa), 73.0 (CH2oxa),
54.0 (CH2Ph), 33.8 (C(CH3)3), 25.6 ppm (C(CH3)3); MS (ESI): m/z (%):
284.16 (100) [M+ÀBr], 285.17 (16) [M+ÀBr+H]; FT-IR (KBr): n˜ =
1699 cmÀ1 (s, C=N).
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-(diphenylmethyl)imidazo-
lium bromide (3h): Yield: 86%; 1H NMR (CDCl3): d=10.73 (brs, 1H;
NCHN), 8.43 (s, 1H; CHPh2), 7.80 (t, J=1.9 Hz, 1H; CH4/5-Im), 7.43–7.37
(m, 10H; CHPh), 7.35 (pseudo-t, J=1.75 Hz, 1H; CH4/5-Im), 4.70 (dd, 2J=
9.0 Hz, 3J=9.8 Hz, 1H; CH2oxa), 4.52 (pseudo-t, J=8.8 Hz, 1H; CH2oxa),
4.11 (dd, 3J=8.7 Hz, 3J=9.8 Hz, 1H; CHoxa), 0.94 ppm (s, 9H; C(CH3)3);
13C {1H} NMR (CDCl3): d=149.2 (NCO), 137.5 (N2C), 135.9 (CPh), 129.5,
129.4, 128.6 (CHPh), 122.7 (CHim), 119.3 (CHim), 75.1 (CHoxa), 73.1
(CH2oxa), 67.0 (CHPh2), 33.9 (C(CH3)3), 25.7 ppm (C(CH3)3); MS (ESI):
m/z (%): 360.22 (100) [M+ÀBr], 361.22 (85) [M+ÀBr+H]; FT-IR (KBr):
n˜ =1698 cmÀ1 (s, C=N); elemental analysis calcd (%) for C23H26BrN3O
(440.38): C 62.73, H 5.95, N 9.54; found: C 62.52, H 5.93, N 9.48.
1-[(S)-4-Isopropyl-4,5-dihydrooxazol-2-yl]-3-mesitylimidazolium bromide
1
4
(3c): Yield: 80%; H NMR (CDCl3): d=10.38 (pseudo-t, J=1.6 Hz, 1H;
NCHN), 8.17 (dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 7.77 (dd, 3J=
2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 6.95 (s, 2H; CHmes), 4.82 (dd, 2J=
8.5 Hz, 3J=9.4 Hz, 1H; CH2oxa), 4.47 (pseudo-t, 2J=8.5 Hz, 3J=8.5 Hz,
1H; CH2oxa), 4.12 (m, 1H; CHoxa), 2.28 (s, 3H; CH3para), 2.10 (s, 6H;
CH3ortho), 1.85 (m, 1H; CH(CH3)2), 1.00 (d, J=6.7 Hz, 3H; CH(CH3)2),
O.91 ppm (d, J=6.7 Hz, 3H; CH(CH3)2); 13C {1H} NMR (CDCl3): d=
149.4 (NCO), 141.7 (Cmes), 137.3 (N2C), 134.0 (Cmes), 130.1 (Cmes), 130.0
(CHmes), 126.0 (CHim), 121.2 (CHim), 75.0 (CH2oxa), 71.9 (CHoxa), 32.7
(CH(CH3)2), 21.2 (CH3para), 18.8 (CH(CH3)2), 18.5 (CH(CH3)2), 18.0 ppm
(CH3ortho); MS (ESI): m/z (%): 298 (84) [M+ÀBr], 187 (100) [M+
ÀBrÀC6H8NO]; FT-IR (KBr): n˜ =1696 cmÀ1 (s, C=N); elemental analysis
calcd (%) for C18H24BrN3O (378.31): C 57.15, H 6.39, N 11.10; found: C
58.41, H 6.43, N 11.10.
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-fluorenylimidazolium
bro-
mide (3i): Yield: 92%; 1H NMR (CDCl3): d=11.86 (brs, 1H; NCHN),
7.92 (brs, 1H; CH4/5-Im), 7.76–7.67 (m, 5H; CH4/5-Im, 4CHfluo), 7.49–7.33
(m, 4H; CHfluo), 4.72 (dd, J=9.1, 9.7 Hz, 1H; CH2oxa), 4.55 (pseudo-t, J=
8.8 Hz, 1H; CH2oxa), 4.12 (dd, 3J=8.7 Hz, 3J=9.8 Hz, 1H; CHoxa),
0.94 ppm (s, 9H; C(CH3)3); 13C {1H} NMR (CDCl3): d=149.3 (NCO),
141.0, 139.3, 138.4 (N2C, Cfluo, Cfluo), 130.7, 128.8, 126.2 (CHfluo), 120.9
(CHim), 120.7 (CHfluo), 119.7 (CHim), 75.0 (CHoxa), 73.1 (CH2oxa), 63.6
(CH-fluo), 33.9 (C(CH3)3), 25.7 ppm (C(CH3)3); MS (ESI): m/z (%):
358.19 (100) [M+ÀBr], 359.19 (18) [M+ÀBr+H]; FT-IR (KBr): n˜ =
1699 cmÀ1 (s, C=N); elemental analysis calcd (%) for C23H24BrN3O
(438.37): C 63.02, H 5.52, N 9.58; found: C 63.50, H 5.80, N 9.30.
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-mesitylimidazolium bromide
(3d): Yield: 71% (reaction time: 5d); 1H NMR (CDCl3): d=10.43
4
3
4
(pseudo-t, J=1.6 Hz, 1H; NCHN), 8.18 (dd, J=2.1 Hz, J=1.6 Hz, 1H;
CH4/5-Im), 7.80 (dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 6.96 (s, 2H;
CHmes), 4.76 (dd, 2J=8.8 Hz, 3J=9.9 Hz, 1H; CH2oxa), 4.55 (pseudo-t, J=
8.6 Hz, 1H; CH2oxa), 4.14 (dd, 3J=8.8 Hz, 3J=9.9 Hz, 1H; CHoxa), 2.28
(s, 3H; CH3para), 2.10 (s, 6H; CH3ortho), 0.93 ppm (s, 9H; C(CH3)3); 13C
{1H} NMR (CDCl3): d=149.4 (NCO), 141.8 (Cmes), 137.4 (N2C), 134.0
(Cmes), 130.2 (Cmes), 130.0 (CHmes), 126.1 (CHim), 121.2 (CHim), 75.2
(CHoxa), 73.1 (CH2oxa), 33.9 (C(CH3)3), 25.8 (C(CH3)3), 21.1 (CH3para),
18.0 ppm (CH3ortho); MS (ESI): m/z: 312.2038 [M+ÀBr]; FT-IR (KBr):
n˜ =1699 cmÀ1 (s, C=N); elemental analysis calcd (%) for C19H26BrN3O
(392.34): C 58.17, H 6.68, N 10.71, found: C 57.50, H 6.66, N 10.42.
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-[di(a-naphtyl)methyl]imida-
zolium bromide (3j): Yield: 97%; 1H NMR (CDCl3): d=11.16 (s, 1H;
NCHN), 9.65 (s, 1H; CHNp2), 8.49–8.39 (m, 2H; CHNp), 7.93–7.87 (m,
4H; CHNp), 7.77 (brs, 1H; CH4/5-Im), 7.57–7.50 (m, 4H; CHNp), 7.36–7.26
2
(m, 3H; CH4/5-Im, 2CHNp), 6.92–6.89 (m, 2H; CHNp), 4.67 (dd, J=9.0 Hz,
3J=9.8 Hz, 1H; CH2oxa), 4.50 (pseudo-t, J=8.8 Hz, 1H; CH2oxa), 4.10
(dd, 3J=8.7 Hz, 3J=9.8 Hz, 1H; CHoxa), 0.93 ppm (s, 9H; C(CH3)3); 13C
{1H} NMR (CDCl3): d=149.2 (NCO), 138.0 (N2C), 134.1, 132.5, 132.4
(CNp), 130.7, 128.7, 128.3, 125.9, 124.8, 124.3, 124.2 (CHNp), 123.6 (CHim),
118.9 (CHim), 75.1 (CHoxa), 73.1 (CH2oxa), 62.5 (CHNp2), 33.8 (C(CH3)3),
25.7 ppm (C(CH3)3); MS (ESI): m/z (%): 267.10 (37) [Np2CH]+, 460.22
(100) [M+ÀBr].
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl)-3-(2,6-diisopropylphenyl)imi-
dazolium bromide (3e): Yield: 96% (reaction time 4d); 1H NMR
(CDCl3): d=10.56 (brs, 1H; NCHN), 8.35 (dd, J=2.1 Hz, 4J=1.6 Hz,
3
4
3
1H; CH4/5-Im), 7.68 (dd, J=2.1 Hz, J=1.6 Hz, 1H; CH4/5-Im), 7.56 (t, J=
8.0 Hz, 1H; CHAr), 7.34 (d, 3J=8.0 Hz, 2H; CHAr), 4.86 (pseudo-t, J=
9.5 Hz, 1H; CH2oxa), 4.65 (pseudo-t, J=8.6 Hz, 1H; CH2oxa), 4.21
(pseudo-t, J=9.5 Hz, 1H; CHoxa), 2.43 (m, 2H; CH(CH3)2), 1.29 (d, 3J=
Preparation of the complexes—general procedure: [{RhCl(nbd)}2]
(0.25 mmol) and 2.2 equivalents of potassium tert-butoxide were placed
in a Schlenk tube. THF was added by syring so that the solution was
0.02m with respect to the Rh precursor. After stirring for 30 minutes at
ambient temperature, the mixture was slowly added to a suspension of
2.0 equivalents of the imidazolium salt in THF (0.02m with respect to the
imidazolium salt) at À788C. The mixture was allowed to warm to ambi-
ent temperature overnight and was centrifuged. The supernatant solution
was separated and the volatiles were removed in vacuo. The crude solid
was washed two or three times with pentane (15 mLmmolÀ1) and dried.
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-phenylimidazol-2-ylidene}rhodium(i) (4a): Yield: 87%; 1H NMR
(CDCl3): d=8.03 (m, 2H; CHPh), 7.56–7.46 (m, 3H; CHPh), 7.33 (d, 3J=
2.1 Hz, 1H; CH4/5-Im), 6.97 (d, 3J=2.1 Hz, 1H; CH4/5-Im), 4.78 (dd, 2J=
8.7 Hz, 3J=9.7 Hz, 1H; CH2oxa), 4.65 (dd, 2J=8.6 Hz, 3J=7.3 Hz, 1H;
CH2oxa), 4.25 (dd, 3J=7.3 Hz, 3J=9.8 Hz, 1H; CHoxa), 3.58 (brs, 2H;
CH2/3/5/6-nbd), 3.49 (brs, 2H; CH2/3/5/6-nbd), 3.42 (brs, 2H; CH1/4-nbd), 1.02 (s,
9H; C(CH3)3), 1.00 ppm (s, 2H; CH2nbd); 13C {1H} NMR (CDCl3): d=
3
3
6.8 Hz, 6H; CH(CH3)2), 1.18 (d, J=6.8 Hz, 3H; CH(CH3)2), 1.17 (d, J=
6.8 Hz, 3H; CH(CH3)2), 0.99 ppm (s, 9H; C(CH3)3); 13C {1H} NMR
(CDCl3): d=149.3 (NCO), 141.0, 137.7, 132.4 (CAr, N2C), 126.5 (CHAr),
124.9 (CHim), 121.2 (CHim), 75.2 (CHoxa), 73.4 (CH2oxa), 33.8 (C(CH3)3),
28.9 (CH(CH3)2), 25.9 (C(CH3)3), 24.5 (CH(CH3)2), 24.2(CH(CH3)2); MS
(ESI): m/z (%): 354.25 (100) [M+ÀBr], 355.25 (22) [M+ÀBr+H]; FT-IR
(KBr): n˜ =1692.7 cmÀ1 (s, C=N).
1-[(S)-4-tert-Butyl-4,5-dihydrooxazol-2-yl]-3-(2-tert-butylphenyl)imidazo-
lium bromide (3 f): Yield: 70%; 1H NMR (CDCl3): d=9.84 (brs, 1H;
NCHN), 8.10 (dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 7.92 (m, 1H;
CHAr), 7.71 (dd, 3J=2.1 Hz, 4J=1.6 Hz, 1H; CH4/5-Im), 7.61 (m, 1H;
CHAr), 7.54 (m, 1H; CHAr), 7.39 (m, 1H; CHAr), 4.77 (pseudo-t, J=
9.2 Hz, 1H; CH2oxa), 4.59 (pseudo-t, J=8.6 Hz, 1H; CH2oxa), 4.18 (dd,
3
3J=8.7 Hz, J=9.8 Hz, 1H; CHoxa), 1.24 (s, 9H; C(CH3)3ortho),0.96 ppm (s,
9H; C(CH3)3oxa); 13C {1H} NMR (CDCl3): d=149.2 (NCO), 145.3
(CAr),137.9 (N2C), 132.2 (CAr), 132.0, 130.6, 128.7, 128.1 (CHAr), 127.2
(CHim), 120.0 (CHim), 75.2 (CHoxa), 73.1 (CH2oxa), 35.9 (C(CH3)3ortho), 33.9
2870
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2862 – 2873

Hydrosilylation Catalysts
FULL PAPER
149.3 (NCO), 138.9 (CPh), 129.0, 128.8, 125.8 (CHPh), 122.8 (CHim), 117.7
(CHim), 74.8 (CHoxa), 73.1 (CH2oxa), 60.3 (CH2nbd), 48.8 (CH1/4-nbd), 34.4
(C(CH3)3), 25.8 ppm (C(CH3)3); MS (ESI): m/z: 464.1299 [M+ÀBr]; FT-
IR (KBr): n˜ =1675 cmÀ1 (s, C=N).
298 K) in a ratio of 1:3.1; they are labelled “major” for major isomer and
“minor” for minor isomer. 1H NMR (CDCl3, 298 K): d=8.20–8.18 (m,
1H; CHAr, major), 7.70–7.67 (m, 1H; CHAr, minor), 7.40–7.30 (m, 3H;
CHAr, isomers taken together), 7.29 (d, 3J=2.1 Hz, 1H; CH4/5-Im, minor),
7.20 (d, 3J=2.1 Hz, 1H; CH4/5-Im, major), 6.84 (d, 3J=2.1 Hz, 1H; CH4/5-
Im, minor), 6.78 (d, 3J=2.1 Hz, 1H; CH4/5-Im, major), 4.79 (dd, J=8.8 Hz,
J=9.6 Hz, 1H; CH2oxa, major), 4.73–4.66 (m, 1H (major), 1H (minor);
CH2oxa), 4.59 (dd, J=7.5 Hz, J=8.4 Hz, 1H; CH2oxa, minor), 4.20 (dd,
3J=6.5 Hz, 3J=9.8 Hz, 1H; CHoxa, major), 4.20 (dd, 3J=7.2 Hz, 3J=
9.6 Hz, 1H; CHoxa, minor), 3.70 (brs, 2H; CH2/3/5/6-nbd, major), 3.52 (brs,
2H; CH1/4-nbd, minor), 3.37 (brs, 2H; CH1/4-nbd, major), 3.22 (brs, 2H;
CH2/3/5/6-nbd, major), 1.20 (s, 9H; C(CH3)3Ph, minor), 1.17 (s, 9H; C-
(CH3)3Ph, major), 1.04 (s, 9H; C(CH3)3oxa, minor), 0.93 ppm (brs, 11H;
9C(CH3)3oxa, major; 2CH2nbd, isomers taken together); 13C {1H} NMR
(CDCl3): d=187.2 (d, 1J(103Rh,13C)=55 Hz, N2C), 156.5 (NCO, major),
155.9 (NCO, minor), 145.3, 136.6 (CAr, minor), 144.6, 135.7 (CAr, major),
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-(2-tolyl)imidazol-2-ylidene}rhodium(i) (4b): Yield: 89%; 1H NMR
(CDCl3): d=7.32 (br, 2H; CHPh), 7.24 (br, 2H; CH4/5Im, CHPh), 6.72 (d,
3J=2.1 Hz, 1H; CH4/5-Im), 4.78 (pseudo-t, J=9.2 Hz, 1H; CH2oxa), 4.66
2
3
3
3
(dd, J=8.5 Hz, J=6.6 Hz, 1H; CH2oxa), 4.18 (dd, J=6.5 Hz, J=9.7 Hz,
1H; CHoxa), 3.69 (brs, 2H; CH2/3/5/6-nbd), 3.35 (s, 2H; CH1/4-nbd), 3.21 (brs,
2H; CH2/3/5/6-nbd), 2.18 (br, 3H; CH3tol), 1.00 (s, 2H; CH2nbd), 0.93 ppm (s,
9H; C(CH3)3); 13C {1H} NMR (CDCl3): d=186.8 (d, 1J(103 h13C)=55 Hz,
N2C), 156.2 (NCO), 137.2 (Ctol), 130.5, 129.6 (CHtol), 129.2 (Ctol), 126.8
(CHtol), 123.6 (CHim), 116.1 (CHim), 74.6 (CHoxa), 73.8 (CH2oxa), 59.6
(CH2nbd), 48.4 (CH1/4-nbd), 34.5 (C(CH3)3), 25.7 (C(CH3)3), 18.0 ppm
(CH3tol); MS (ESI): m/z: 478.1348 [M+ÀBr]; FT-IR (KBr): n˜ =1674 cmÀ1
(s, C=N).
133.0, 129.6, 128.6, 127.1 (CHAr, major), 132.5, 129.7, 128.6, 126.7 (CHAr
,
minor), 126.2, 115.3 (CHIm, major), 125.9, 116.7 (CHIm, minor), 74.8
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-isopropyl-4,5-dihydrooxazol-2-
yl]-3-mesitylimidazol-2-ylidene}rhodium(i) (4c): Yield: 67%; 1H NMR
(CDCl3): d=7.27 (d, 3J=2.1 Hz, 1H; CH4/5-Im), 6.88 (s, 2H; CHmes), 6.39
(d, 3J=2.1 Hz, 1H; CH4/5-Im), 4.82 (pseudo-t, J=8.5 Hz, 1H; CH2oxa),
4.60 (pseudo-t, J=8.5 Hz, 1H; CH2oxa), 4.34 (m, 1H; CHoxa), 3.66 (brs,
4H; CH2/3/5/6-nbd), 3.45 (brs, 2H; CH1/4-nbd), 2.29 (s, 3H; CH3para), 2.20 (s,
3H; CH3ortho), 2.17 (s, 3H; CH3ortho), 1.95 (m, 1H; CH(CH3)2), 0.99 (d,
J=6.8 Hz, 3H; CH(CH3)2), 0.96 (s, 2H; CH2nbd), 0.91 ppm (d, J=6.7 Hz,
3H; CH(CH3)2); 13C {1H} NMR (CDCl3): d=184.4 (d, 1J(103Rh,13C)=
55 Hz, N2C), 156.7 (NCO), 139.4, 135.8, 135.4, 133.6 (Cmes), 128.8
(CHmes), 123.5 (CHim), 115.2 (CHim), 74.1 (CH2oxa), 68.9 (CHoxa), 60.0
(CH2nbd), 48.8 (CH1/4-nbd), 30.6 (CH(CH3)2), 21.2 (CH3para), 19.2 (CH-
(CH3)2), 18.7, 18.6 (CH3ortho), 16.4 ppm (CH(CH3)2); MS (ESI): m/z (%):
492.1432 (100) [M+ÀBr]; FT-IR (KBr): 1669.5 cmÀ1 (s, C=N); elemental
analysis calcd (%) for C25H31BrN3ORh (572.34): C 52.46, H 5.46, N 7.34;
found: C 52.18, H 5.47, N 7.27.
(CHoxa, major), 74.0 (CH2oxa, major), 73.9 (CHoxa, minor), 73.0 (CH2oxa
minor), 60.9 (CH2nbd minor), 59.6 (CH2nbd major), 49.3 (CH1/4-nbd
,
,
,
,
minor), 48.3 (CH1/4-nbd, major), 36.2, 34.2 (2 C(CH3)3, minor), 36.0, 34.8
(2C(CH3)3, major), 32.2 (C(CH3)3Ph, major), 32.1 (C(CH3)3Ph, minor),
26.0 (C(CH3)3oxa, minor), 25.7 ppm (C(CH3)3oxa, major); MS (ESI): m/z
(%): 520.1827 (100) [M+ÀBr], 521.1861 (27) [M+ÀBr+H]; FT-IR (KBr):
n˜ =1676 cmÀ1 (s, C=N).
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-benzylimidazol-2-ylidene}rhodium(i) (4g): Yield: 88%; 1H NMR
(CDCl3): d=7.55–7.25 (m, 5H; CHPh), 7.21 (d, J=2.1 Hz, 1H; CH4/5-Im),
6.60 (d, J=2.1 Hz, 1H; CH4/5-Im), 5.58 (s, 2H; CH2Ph), 4.70 (dd, 2J=
8.7 Hz, 3J=9.7 Hz, 1H; CH2oxa), 4.56 (dd, 2J=8.5 Hz, 3J=7.5 Hz, 1H;
3
CH2oxa), 4.19 (dd, 3J=7.5 Hz, J=9.8 Hz, 1H; CHoxa), 3.93 (br, 4H; CH2/
3/5/6-nbd), 3.61 (brs, 2H; CH1/4-nbd), 1.16 (s, 2H; CH2nbd), 1.00 ppm (s, 9H;
C(CH3)3); 13C {1H} NMR (CDCl3): d=191.2 (d, 1J(103Rh,13C)=56 Hz,
N2C), 154.7 (NCO), 135.6 (CPh), 129.0, 128.3, 128.1 (CHPh), 121.3 (CHim),
119.5 (CHim), 74.7 (CHoxa), 72.3 (CH2oxa),), 61.6 (CH2nbd), 54.6 (CH2Ph),
49.9 (CH1/4-nbd), 33.3 (C(CH3)3), 25.9 ppm (C(CH3)3); MS (ESI): m/z:
478.1426 [M+ÀBr]; FT-IR (KBr): n˜ =1676 cmÀ1 (s, C=N).
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-mesitylimidazol-2-ylidene}rhodium(i) (4d): Yield: 89%; crystallisa-
tion from CH2Cl2/pentane gave orange crystals suitable for an X-ray dif-
1
3
fraction studies; H NMR (CDCl3): d=7.29 (d, J=2.1 Hz, 1H; CH4/5-Im),
6.96 (s, 1H; CHmes), 6.88 (s, 1H; CHmes), 6.59 (d, 3J=2.1 Hz, 1H; CH4/5-
Im), 4.79 (dd, 2J=8.7 Hz, 3J=9.7 Hz, 1H; CH2oxa), 4.68 (pseudo-t, J=
6.7 Hz, 1H; CH2oxa), 4.23 (dd, 3J=6.7 Hz, 3J=9.7 Hz, 1H; CHoxa), 3.63
(brs, 2H; CH2/3/5/6-nbd), 3.40 (brs, 2H; CH1/4-nbd), 3.36 (brs, 2H; CH2/3/5/6-
nbd), 2.39 (s, 3H; CH3ortho), 2.32 (s, 3H; CH3ortho), 2.03 (s, 3H; CH3para),
0.98 (s, 9H; C(CH3)3), 0.98 ppm (s, 2H; CH2nbd); 13C {1H} NMR (CDCl3):
d=156.4 (NCO), 139.4, 136.7, 134.8, 134.0 (Cmes), 129.3, 128.4 (CHmes),
123.6 (CHim), 116.5 (CHim), 74.7 (CHoxa), 73.8 (CH2oxa), 59.7 (CH2nbd),
48.5 (CH1/4-nbd), 34.5 (C(CH3)3), 25.8 (C(CH3)3), 21.1 (CH3para), 19.4,
18.0 ppm (CH3ortho); MS (ESI): m/z: 506.1764 [M+ÀBr]; FT-IR (KBr):
n˜ =1674 cmÀ1 (s, C=N); elemental analysis calcd (%) for C26H33BrN3ORh
(586.38): C 53.26, H 5.67, N 7.16; found: C 53.05, H 5.63, N 6.99.
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-(diphenylmethyl)imidazol-2-ylidene}rhodium(i) (4h): Yield: 87%,
1H NMR (CDCl3): d=7.96 (s, 1H; CHPh2), 7.37–7.18 (m, 11H; CH4/5-Im
,
10CHPh), 6.57 (t, 3J=2.1 Hz, 1H; CH4/5-Im), 4.73 (pseudo-t, J=9.2 Hz,
1H; CH2oxa), 4.59 (pseudo-t, J=8.0 Hz, 1H; CH2oxa), 4.19 (dd, 3J=
7.5 Hz, 3J=9.8 Hz, 1H; CHoxa), 3.73 (br, 4H; CH2/3/5/6-nbd), 3.47 (brs, 2H;
CH1/4-nbd), 1.08 (s, 2H; CH2nbd), 1.01 ppm (s, 9H; C(CH3)3); 13C {1H}
NMR (CDCl3): d=192.5 (d, 1J(103Rh,13C)=55 Hz, N2C), 154.8 (NCO),
139.4, 139.0 (CPh), 129.7–127.8 (CHPh), 120.8 (CHim), 118.8 (CHim), 74.8
(CHoxa), 72.3 (CH2oxa), 67.4 (CHPh2), 61.6 (CH2nbd), 49.8 (CH1/4-nbd), 34.3
(C(CH3)3), 25.8 ppm (C(CH3)3); MS (ESI): m/z (%): 462.10 (100) [M+
ÀBrÀnbd], 554.16 (68) [M+ÀBr]; FT-IR (KBr): n˜ =1671 cmÀ1 (s, C=N).
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-fluorenylimidazol-2-ylidene}rhodium(i) (4i): Yield: 58%; 1H NMR
(CDCl3): d=7.93–7.89 (m, 1H; CHfluo), 7.77–7.68 (m, 2H; CHfluo), 7.47–
7.26 (m, 6H; CH-fluo, 5CHfluo), 7.14 (d, J=1.5 Hz, 1H; CH4/5-Im), 6.02
(d, J=1.5 Hz, 1H; CH4/5-Im), 4.75 (pseudo-t, J=8.8 Hz, 1H; CH2oxa), 4.62
(pseudo-t, J=7.6 Hz, 1H; CH2oxa), 4.29 (dd, 3J=7.8 Hz, 3J=9.4 Hz, 1H;
CHoxa), 3.19 (br, 4H; CH2/3/5/6-nbd), 3.74 (brs, 2H; CH1/4-nbd), 1.28 (s, 2H;
CH2nbd), 1.09 ppm (s, 9H; C(CH3)3); 13C {1H} NMR (CDCl3): d=154.6
(NCO), 142.2, 141.1, 140.9, 140.3 (Cfluo), 129.6, 129.5, 128.6, 128.2, 126.7,
125.4, 120.3, 120.0 (CHfluo), 119.6 (CHim), 118.9 (CHim), 75.0 (CHoxa), 72.3
(CH2oxa), 65.1 (CH-fluo), 61.7 (CH2nbd), 50.0 (CH1/4-nbd), 34.4 (C(CH3)3),
25.9 ppm (C(CH3)3); MS (ESI): m/z (%): 165.09 (5) [fluo]+, 358.19 (17)
[L]+, 460.09 (13) [M+ÀBrÀnbd], 554.16 (68) [M+ÀBr]; FT-IR (KBr):
n˜ =1673 cmÀ1 (s, C=N).
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-(2,6-diisopropylphenyl)imidazol-2-ylidene}rhodium(i) (4e): Yield:
80%; 1H NMR (CDCl3): d=7.47–7.42 (m, 1H; CHAr), 7.33 (d, 3J=
2.1 Hz, 1H; CH4/5-Im), 7.27–7.22 (m, 2H; CHAr), 6.67 (d, 3J=2.1 Hz, 1H;
CH4/5-Im), 4.76 (dd, 2J=8.7 Hz, 3J=9.7 Hz, 1H; CH2oxa), 4.64 (pseudo-t,
J=6.7 Hz, 1H; CH2oxa), 4.21 (dd, 3J=6.9 Hz, 3J=9.7 Hz, 1H; CHoxa),
3.53 (brs, 4H; CH2/3/5/6-nbd), 3.36 (brs, 2H; CH1/4-nbd), 2.93 (m, 1H; CH-
(CH3)2), 2.77 (m, 1H; CH(CH3)2), 1.38 (d, 3J=6.8 Hz, 3H; CH(CH3)2),
1.36 (d, 3J=6.8 Hz, 3H; CH(CH3)2), 1.02 (s, 9H; C(CH3)3), 1.02 (s, 2H;
CH2nbd), 0.99 (d, 3J=7.1 Hz, 3H; CH(CH3)2), 0.97 ppm (d, 3J=7.1 Hz,
3H; CH(CH3)2); 13C {1H} NMR (CDCl3): d=187.3 (d, 1J(103Rh,13C)=
55 Hz, N2C), 156.0 (NCO), 146.6, 146.2, 134.2 (CAr), 130.3 (CHAr), 125.1
(CHim), 123.9, 123.7 (CHAr), 116.5 (CHim), 74.3 (CHoxa), 73.3 (CH2oxa),
60.4 (CH2nbd), 48.8 (CH1/4-nbd), 34.3 (C(CH3)3), 28.1 (CH(CH3)2), 26.1
(CH(CH3)2), 25.9 (C(CH3)3), 23.2 ppm (CH(CH3)2); MS (ESI): m/z:
548.2250 [M+ÀBr]; FT-IR (KBr): n˜ =1670.5 cmÀ1 (s, C=N).
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-[di(1-naphtyl)methyl]imidazol-2-ylidene}rhodium(i) (4j): Yield:
92%; 1H NMR (CDCl3): d=8.69 (brs, 1H; CHNp2), 7.86–7.37 (m, 12H;
CHNp), 7.11 (d, J=2.1 Hz, 1H; CH4/5-Im), 6.89–6.86 (m, 2H; CHNp), 6.43
(d, J=2.1 Hz, 1H; CH4/5-Im), 4.74 (pseudo-t, J=9.3 Hz, 1H; CH2oxa), 4.62
Bromo-(h4-2,5-norbornadiene)-{1-[(S)-4-tert-butyl-4,5-dihydrooxazol-2-
yl]-3-(2-tert-butylphenyl)imidazol-2-ylidene}rhodium(i) (4 f): Yield: 81%.
Two isomers are observed in the 1H and 13C NMR spectra (recorded at
Chem. Eur. J. 2005, 11, 2862 – 2873
www.chemeurj.org
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2871

L. H. Gade, S. Bellemin-Laponnaz et al.
(pseudo-t, J=7.8 Hz, 1H; CH2oxa), 4.19 (dd, 3J=7.8 Hz, 3J=9.8 Hz, 1H;
CHoxa), 3.5–3.3 (br, 4H; CH2/3/5/6-nbd), 3.00 (brs, 2H; CH1/4-nbd), 1.00 (s,
9H; C(CH3)3), 0.74 ppm (s, 2H; CH2nbd); 13C {1H} NMR (CDCl3): d=
155.3 (NCO), 133.9, 133.8, 131.4, 130.8 (CNp), 129.8, 129.2, 128.5, 128.2,
127.3, 126.8, 126.4, 125.7, 125.3, 125.0, 124.7, 121.4 (CHNp), 127.8 (CHim),
123.6 (CHim), 74.9 (CHoxa), 72.7 (CH2oxa), 61.9 (CHNp2), 60.8 (CH2nbd),
49.0 (CH1/4-nbd), 34.2 (C(CH3)3), 25.6 ppm (C(CH3)3); MS (ESI): m/z (%):
562.13 (11) [M+ÀBrÀnbd], 654.20 (100) [M+ÀBr]; FT-IR (KBr): n˜ =
1671.2 cmÀ1 (s, C=N).
Table 5. X-ray experimental data for compounds 4d and 4j.
4d 4j
C26H33BrN3ORh
formula
C38H37BrN3ORh·
CH2Cl2
819.45
Mr
586.39
crystal system
space group
a [ꢀ]
b [ꢀ]
c [ꢀ]
orthorhombic
P212121
9.7117(1)
13.7354(2)
18.3991(2)
90
monoclinic
P21
12.5400(5)
11.4205(5)
13.4866(6)
116.8570(10)
1723.1(1)
2
General procedure for the hydrosilylation catalysis: In a small Schlenk
tube, an orange solution of complex 4d (17.6 mg, 0.030 mmol) in CH2Cl2
(0.6 mL) was added to AgBF4 (7.0 mg, 0.012 mmol). The resulting mix-
ture was stirred for about 5 min and then filtered through a pad of Celite
and rinsed with an additional of CH2Cl2 (0.3 mL). The red solution was
divided into three equal parts (for three different catalytic runs; 0.3 mL
each) and each vial was topped up to 0.5 mL. After addition of the de-
sired ketone (1.0 mmol) at ambient temperature, the reaction mixture
was cooled to À608C and diphenylsilane (210 mL, 1.1 mmol) was added
dropwise over a period of 2 min. The bright yellow reaction mixture was
stirred at À608C for 10 h. A solution of K2CO3 in methanol (0.1%,
2.0 mL) was then added and the resulting mixture was stirred for at least
4 h at room temperature. After evaporation of the solvents, the product
was purified by column chromatography (2ꢆ10 cm, SiO2, pentane/Et2O:
85/15). Yields refer to isolated yields of compounds estimated to be
b [8]
V [ꢀ3]
2454.33(5)
4
Z
colour
orange
0.25ꢆ0.20ꢆ0.20
1.59
1192
2.347
orange
0.25ꢆ0.18ꢆ0.08
1.579
832
1.847
crystal size [mm]
1calcd [gcmÀ3
F000
]
m [mmÀ1
T [K]
]
173
100
2q range [8]
5–60
4–64
index range
À13ꢁhꢁ13
À19ꢁkꢁ19
À25ꢁlꢁ25
7139
À18ꢁhꢁ16
À14ꢁkꢁ17
0ꢁlꢁ20
10567
reflns measured
observed reflns
1
>95% pure as determined by H NMR spectroscopy. Yields and eeꢁs are
3461
9533
the average of at least two corroborating runs. Absolute configurations
of enantiomerically enriched alcohols were determined by comparing the
sign of their optical activities with those reported in literature.
I>3s(I)
289
I>2s(I)
443
parameters
R (obsvd data)
wR
0.030
0.048
0.038
Kinetic studies: Hydrosilylation studies were carried out at À558C in
CD2Cl2 as solvent. Each experiment was conducted with freshly prepared
rhodium catalyst by reacting complex 4d with AgBF4 in CD2Cl2 followed
by filtration through Celite. The solution was then transferred into a
NMR tube with the desired amount of acetophenone. In all cases, the
sample was allowed to stand at À558C before adding the diphenyl silane.
The progress of the reaction was monitored by measuring the disappear-
ance of the ketone and appearance of the product by 1H NMR spectros-
copy.
wR2 (all data)
Flack parameter
GOOF
0.091
0.001(6)
1.033
0.01(1)
1.020
0.532
largest difference peak [eꢀ3]
1.206
BASF (Ludwigshafen) and Degussa (Hanau) is gratefully acknowledged.
We also thank Dr A. De Cian and N. Gruber for the X-ray diffraction
studies.
Crystal structure determinations: Suitable crystals of the complexes 4d
and 4j were obtained by layering concentrated solutions of the samples
in dichloromethane with pentane and allowing slow diffusion at room
temperature. Intensity data were collected at low temperature on Nonius
Kappa CCD (complex 4d) and Bruker Smart 1000 CCD (complex 4j)
diffractometers. The radiation used was graphite-monochromated MoKa
(l=0.71073 ꢀ). A semi-empirical absorption correction was applied. The
structures were solved by using direct methods and refined by full-matrix
least-squares methods. For complex 4d, hydrogen atom positions were
taken from difference Fourier maps; they were introduced as fixed con-
[1] A. H. Hoveyda, A. W. Hird, M. A. Kaprzynski, Chem. Commun.
2004, 1779.
[2] Reviews: a) W. A. Herrmann, Angew. Chem. 2002, 114, 1342;
Angew . Chem. Int. Ed. 2002, 41, 1290; b) D. Bourissou, O. Guerret,
F. Gabbaꢈ, G. Bertrand, Chem. Rev. 2000, 100, 39; c) W. A. Herr-
mann, C. Kçcher, Angew. Chem. 1997, 109, 2257; Angew. Chem. Int.
Ed. Engl. 1997, 36, 2162.
[3] a) J. A. Chamizo, P. B. Hitchcock, H. A. Jasim, M. F. Lappert, J. Or-
ganomet. Chem. 1993, 451, 89; b) W. A. Herrmann, C. Kçcher, L. J.
Goossen, G. R. J. Artus, Chem. Eur. J. 1996, 2, 1627; c) D. S.
McGuinness, K. J. Cavell, Organometallics 2000, 19, 741; d) C. Yang,
H. M. Lee, S. P. Nolan, Org. Lett. 2001, 3, 1511; e) P. L. Arnold,
A. C. Scarisbrick, A. J. Blake, C. Wilson, Chem. Commun. 2001,
2340; f) A. A. Danopoulos, S. Winston, T. Gelbrich, M. B. Hurst-
house, R. P. Tooze, Chem. Commun. 2002, 482; g) A. A. Danopou-
los, S. Winston, M. B. Hursthouse, J. Chem. Soc. Dalton Trans. 2002,
3090; h) P. L. Arnold, Heteroat. Chem. 2002, 13, 534.
[4] a) W. A. Herrmann, L. J. Goossen, M. Spiegler, Organometallics
1998, 17, 2162; b) J. C. C. Chen, I. J. B. Lin, Organometallics 2000,
19, 5113; c) A. A. D. Tulloch, A. A. Danopoulos, R. P. Tooze, S. M.
Cafferkey, S. Kleinhenz, M. B. Hursthouse, Chem. Commun. 2000,
1247; d) E. Peris, J. A. Loch, J. Mata, R. H. Crabtree, Chem.
Commun. 2001, 201; e) S. Grꢅndemann, A. Kovacevic, M. Albrecht,
J. W. Faller, R. H. Crabtree, Chem. Commun. 2001, 2274; f) S. Grꢅn-
demann, M. Albrecht, J. A. Loch, J. W. Faller, R. H. Crabtree, Orga-
nometallics 2001, 20, 5485; g) A. A. D. Tulloch, A. A. Danopoulos,
G. J. Tizzard, S. J. Coles, M. B. Hursthouse, R. S. Hay-Motherwell,
W. B. Motherwell, Chem. Commun. 2001, 1270; h) S. Grꢅndemann,
À
tributors in the structure factor calculations with fixed coordinates (C H:
0.95 ꢀ) and isotropic temperature factors (B(H)=1.3 Beqv(C) ꢀ2), but
not refined. For complex 4j, hydrogen atoms were input at calculated po-
sitions except for the olefinic hydrogens of the norbornadiene ligand,
À
which were taken from Fourier maps and refined with the C H distance
constrained to 0.95 ꢀ. The calculations were performed by using the pro-
grams OpenMoleN,[31] SHELXS-86[32] and SHELXL-97.[33] Graphical rep-
resentations were drawn with PLATON.[34] Crystal data and experimental
details are given in Table 5.
CCDC 221154 and 262660 contain the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
Acknowledgements
We thank the CNRS (France) and the Institut Universitaire de France
for support of this work and the Ministꢇre de lꢁEducation Nationale de la
Recherche et de la Technologie for a Ph.D. grant to V.C. Support by
2872
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 2862 – 2873

Hydrosilylation Catalysts
FULL PAPER
A. Kovacevic, M. Albrecht, J. W. Faller, R. H. Crabtree, J. Am.
Chem. Soc. 2002, 124, 10473; i) A. A. Danopoulos, S. Winston, W. B.
Motherwell, Chem. Commun. 2002, 1376; j) J. A. Loch, M. Albrecht,
E. Peris, J. Mata, J. W. Faller, R. H. Crabtree, Organometallics 2002,
21, 700; k) J. J. Van Veldhuizen, S. B. Garber, J. S. Kingsbury, A. H.
Hoveyda, J. Am. Chem. Soc. 2002, 124, 4954; l) A. A. Danopoulos,
S. Winston, M. B. Hursthouse, J. Chem. Soc. Dalton Trans. 2002,
3090; m) A. A. D. Tulloch, S. Winston, A. A. Danopoulos, G. East-
ham, M. B. Hursthouse, Dalton Trans. 2003, 699; n) M. C. Perry, X.
Cui, M. T. Powell, D.-R. Hou, J. H. Reibenspies, K. Burgess, J. Am.
Chem. Soc. 2003, 125, 113; o) H. Seo, H.-J. Park, B. Y. Kim, J. H.
Lee, S. U. Son, Y. K. Chung, Organometallics 2003, 22, 618; p) L. G.
Bonnet, R. E. Douthwaite, B. M. Kariuki, Organometallics 2003, 22,
4187.
[15] C. Kçcher, W. A. Herrmann, J. Organomet. Chem. 1997, 532, 261.
[16] The formation of the square-planar cationic rhodium species as an
active catalyst appears to be crucial, since the use of catalyst 4d
without a silver salt gave poor enantioselectivity and yield (13% ee;
53% yield).
[17] a) G. Balavoine, J. C. Clinet, I. Lellouche, Tetrahedron Lett. 1989,
30, 5141; b) Y. Yamanoi, T. Imamoto, J. Org. Chem. 1999, 64, 2988;
c) H. Tsuruta, T. Imamoto, Tetrahedron: Asymmetry 1999, 10, 877.
[18] S. Glasstone, K. J. Laidler, H. Eyring, The Theory of Rate Processes,
McGraw-Hill, New York, 1941, Chapte 4.
[19] T. Gçbel, K. B. Sharpless, Angew. Chem. 1993, 105, 1417; Angew.
Chem. Int. Ed. Engl. 1993, 32, 1329.
[20] J. Brunner, N. Hoffmann, H. D. Scharf, Tetrahedron 1994, 50, 6819.
[21] a) H. Buschmann, H.-D. Scharf, N. Hoffmann, M. W. Plath, J. Run-
sink, J. Am. Chem. Soc. 1989, 111, 5367; review: b) H. Buschmann,
H.-D. Scharf, N. Hoffmann, P. Esser, Angew. Chem. 1991, 103, 480;
Angew. Chem. Int. Ed. Engl. 1991, 30, 477.
[5] a) V. Cꢂsar, S. Bellemin-Laponnaz, L. H. Gade, Organometallics
2002, 21, 5204; b) V. Cꢂsar, S. Bellemin-Laponnaz, L. H. Gade, Eur.
J. Inorg. Chem. 2004, 3436.
[6] a) G. Helmchem, A. Pfaltz, Acc. Chem. Res. 2000, 33, 326; b) For a
general review, covering phosphine–oxazoline ligands in catalysis,
see: P. Braunstein, F. Naud, Angew. Chem. 2001, 113, 702; Angew.
Chem. Int. Ed. 2001, 40, 681.
[7] Reviews on N-heterocyclic-carbene–transition-metal complexes in
asymmetric catalysis: a) M. C. Perry, K. Burgess, Tetrahedron:
Asymmetry 2003, 14, 951; b) V. Cꢂsar, S. Bellemin-Laponnaz, L. H.
Gade, Chem. Soc. Rev. 2004, 33, 619.
[8] a) D. Enders, H. Gielen, K. Beuer, Tetrahedron: Asymmetry 1997, 8,
3571; b) W.-L. Duan, M. Shi, G.-B. Rong, Chem. Commun. 2003,
2976; the state of the art in the asymmetric hydrosilylation of ke-
tones with chiral phosphine ligands: c) M. Sawamura, R. Kuwano,
Y. Ito, Angew. Chem. 1994, 106, 92; Angew. Chem. Int. Ed. Engl.
1994, 33, 111; d) B. Tao, G. C. Fu, Angew. Chem. 2002, 114, 4048;
Angew. Chem. Int. Ed. 2002, 41, 3892; e) D. A. Evans, F. E. Michael,
J. S. Tedrow, K. R. Campos, J. Am. Chem. Soc. 2003, 125, 3534.
[9] A preliminary account of this work: V. Cꢂsar, S. Bellemin-Lapon-
naz, L. H. Gade, Angew. Chem. 2004, 116, 1036; Angew. Chem. Int.
Ed. 2004, 43, 1014.
[10] A. I. Meyers, K. A. Novachek, Tetrahedron Lett. 1996, 37, 1747.
[11] We also recently reported a very efficient way to synthesise trisoxa-
zolines by using 2-bromooxazoline, see: a) S. Bellemin-Laponnaz,
L. H. Gade, Chem. Commun. 2002, 1286; b) S. Bellemin-Laponnaz,
L. H. Gade, Angew. Chem. 2002, 114, 3623; Angew. Chem. Int. Ed.
2002, 41, 3473.
[22] a) D. Haag, J. Runsink, H.-D. Scharf, Organometallics 1998, 17, 398;
see also: b) D. Enders, H. Gielen, K. Beuer, Tetrahedron: Asymme-
try 1997, 8, 3571.
[23] a) K. J. Hale, J. H. Ridd, J. Chem. Soc. Chem. Commun. 1995, 357;
b) K. J. Hale, J. H. Ridd, J. Chem. Soc. Perkin Trans. 1 1995, 1601.
[24] I. Ojima, T. Kogure, M. Kumagai, S. Horiuchi, Y. Sato, J. Organo-
met. Chem. 1976, 122, 83.
[25] For a recent kinetic study of hydrosilylation of acetophenone with a
rhodium/phosphine catalyst, see: C. Reyes, A. Prock, W. P. Giering,
Organometallics 2002, 21, 546, and references therein.
[26] The lower yields in the reduction of some of the diakyl ketones are
due to the volatility of the products. The conversions are usually
1
complete as determined by H NMR spectroscopy.
[27] However, we found almost no conversion with this substrate at
À608C. For that reason, the reaction was conducted at À408C. Com-
plete conversion was observed after 10 h by 1H NMR spectroscopy.
[28] a) Q. Jiang, Y. Jiang, D. Xiao, P. Cao, X. Zhang, Angew. Chem.
1998, 110, 1203; Angew. Chem. Int. Ed. 1998, 37, 1100; b) R.
Kowano, M. Sawamura, J. Shirai, M. Takahashi, Y. Ito, Bull. Chem.
Soc. Jpn. 2000, 73, 485.
[29] E. W. Abel, M. A. Bennett, G. Wilkinson, J. Chem. Soc. 1959, 3178.
[30] T. Masuda, J. K. Stille, J. Am. Chem. Soc. 1978, 100, 268.
[31] OpenMoleN, Interactive Intelligent Structure Solution, Nonius
B. V., Delft (The Netherlands), 1997.
[32] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467.
[33] G. M. Sheldrick, SHELXL-97, Universitꢃt Gçttingen (Germany),
1997.
[12] A. J. Arduengo,
III, F. P. Gentry, Jr., P. K. Taverkere, H. E.
Howard, III, US Patent 6 177 575, 2001.
[13] See for the different methods: a) C. van der Stelt, DE Patent No.
2130673, 1974; b) F. G. Bordwell, J. P. Cheng, S. E. Seyedrezai, C. A.
Wilson, J. Am. Chem. Soc. 1988, 110, 8179; c) W. Draber, E. Regel,
DE Patent No. 2009020, 1971.
[34] A. L. Spek, PLATON, Utrecht University (The Netherlands), 1980–
2004; A. L. Spek, Acta Crystallogr. Sect. A 1990, 46, C31.
Received: February 7, 2005
[14] A. L. Johnson (E. I. Dupont), U. S. Patent No. 3637731, 1972.
Published online: March 3, 2005
Chem. Eur. J. 2005, 11, 2862 – 2873
www.chemeurj.org
ꢄ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2873