Rhodium(I) complexes of new ferrocenyl benzimidazol-2-ylidene ligands:The importance of the chelating effect for ketone hydrosilylation catalysis

Article abstract of DOI:10.1002/ejic.200801163

N-[(1-Phosphanylferrocen-1-yl)methyl] -N-[(2,4,6-trimethyl-phenyl)methyl] -5,6-di-X-benzimidazolium tetrafluoroboratesalts (X = H, 5a and Me, 5b), precursors of new phosphane-benzimidazol-2-ylidene bifunctional ligands, and relatedferrocenyl 5,6-di-X-benz

Full text of DOI:10.1002/ejic.200801163

FULL PAPER
DOI: 10.1002/ejic.200801163
Rhodium(I) Complexes of New Ferrocenyl Benzimidazol-2-ylidene Ligands:
The Importance of the Chelating Effect for Ketone Hydrosilylation Catalysis
Süleyman Gülcemal,[a] Agnès Labande,*[b,c] Jean-Claude Daran,[b,c] Bekir Çetinkaya,[a] and
Rinaldo Poli[b,c,d]
Keywords: Carbenes / P ligands / Ferrocenes / Rhodium / Hydrosilylation
N-[(1-Phosphanylferrocen-1Ј-yl)methyl]-NЈ-[(2,4,6-trimethyl-
phenyl)methyl]-5,6-di-X-benzimidazolium tetrafluoroborate
salts (X = H, 5a and Me, 5b), precursors of new phosphane-
benzimidazol-2-ylidene bifunctional ligands, and related
ferrocenyl 5,6-di-X-benzimidazol-2-ylidene iodide salts (X =
H, 6a and Me, 6b), precursors of monodentate benzimidazol-
2-ylidene ligands, have been prepared for the first time. Cat-
ionic rhodium(I) complexes 7a and 7b and neutral rhodium(I)
complexes 8a and 8b have been obtained in good yields and
have been fully characterised. Cationic rhodium(I) com-
parison with complexes 7, but not isolated. All complexes
showed good activities for the catalytic hydrosilylation of
acetophenone derivatives. The activities are much greater
than for related imidazol-2-ylidene systems, and the cationic
complexes are more active than the neutral complexes; the
highest activity is observed for the more soluble complex 7b.
The use of bidentate ligands proved to be essential for ob-
taining good selectivities of the desired alcohol.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
plexes 10a and 10b were prepared from 8a and 8b, for com- Germany, 2009)
We have tried to improve the catalytic activity of rhodi-
Introduction
um(I) complexes bearing ferrocenylphosphanyl–NHC li-
gands. Indeed, ferrocenyl-substituted NHCs are a class of
ligands that has been little developed and has a great poten-
tial in both asymmetric and non-asymmetric catalysis.[2,14]
We thus prepared new ferrocenylphosphane ligand precur-
sors bearing a benzimidazolium moiety and the corre-
sponding rhodium(I) complexes. Although benzimidazol-
Recent developments in the field of functionalised N-het-
erocyclic carbene (NHC) ligands have proved that they have
great potential in transition-metal catalysis.[1–3] Indeed, the
presence of an NHC ligand and a second structurally dif-
ferent donating group on the metal can radically alter the
catalytic properties. On the other hand, the chelating nature
of these ligands results in the production of highly stable
complexes. In our search for more efficient catalysts, we
have shown that very small structural variations on the bi-
functional ligands dramatically change the reactivity of
rhodium(I) catalysts in the hydrosilylation of ketones.[4,5]
Although NHC ligands bearing a benzimidazole back-
bone offer a vast potential in catalytic applications, their
coordination chemistry has remained relatively unex-
plored.[6–12] The σ-donating properties of benzimidazol-2-
ylidene ligands are between those of unsaturated imidazol-
2-ylidenes and of saturated imidazolin-2-ylidenes. Thus, al-
though they have the topology of unsaturated NHCs, they
show a reactivity typical of saturated carbenes.[11,13]
ium salts with
a ferrocenyl substituent are already
known,[15] we report here the first example of a ferrocenyl-
phosphane bearing a benzimidazolium unit. In order to
evaluate the contribution of the phosphanyl group to catal-
ysis, rhodium(I) complexes bearing monodentate ferrocenyl
NHC ligands were also prepared.
Results and Discussion
Synthesis and Characterisation of Ligand Precursors 5 and 6
The benzimidazolium salts 4a and 4b were obtained in
very good yields from ferrocenyl alcohol 1 and the substi-
tuted benzimidazoles 2a and 2b[7,16] (Figure 1, Scheme 1),
[a] Ege University, Department of Chemistry,
Bornova, 35100 Izmir, Turkey
[b] CNRS, LCC (Laboratoire de Chimie de Coordination),
205, route de Narbonne, 31077 Toulouse, France
Fax: +33-5-61553003
E-mail: agnes.labande@lcc-toulouse.fr
[c] Université de Toulouse, UPS, INPT,
31077 Toulouse, France
[d] Institut Universitaire de France,
103, bd Saint-Michel; 75005 Paris, France
Figure 1. Substituted benzimidazoles.
1806
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1806–1815

Importance of the Chelating Effect for Ketone Hydrosilylation Catalysis
benzimidazolium group. Comparatively, the signal of the
same proton in compounds 6a and 6b is shifted downfield
(10.60 and 10.21 ppm, respectively) because of the stronger
C–H···X– bond polarisation when X = I.[3] The crystal
structures of compounds 5a and 6b have been solved by X-
ray diffraction analysis (Figure 2). The C–C bond of the
imidazolium motif is significantly longer in the case of
benzimidazolium salts 5a and 6b [1.395(3) Å and
1.399(6) Å, respectively] than that of the 1,1Ј-ferrocenyl-
phosphane imidazolium salt previously described by us
[1.330(6) Å].[4] This denotes the delocalisation of the C–C
bond π-electron density over the condensed benzene ring
(the C–C bond lengths within the ring are in the range
1.372–1.402 Å for 5a and in the range 1.354–1.424 Å for
6b), which could be responsible for the different reactivity
of the resulting carbenes. The two structures exhibit hydro-
gen bonds, as is typically observed for imidazolium salts,
–
between a fluorine atom of the BF4 anion and the C62
proton (F3···H62–C62, 2.312 Å) for 5a and between the io-
dide anion and the C1 proton (I1···H1–C1, 2.962 Å) for 6b.
The H bond is significantly shorter in the case of the benz-
Scheme 1. Synthesis of 1,1Ј-disubstituted ferrocenyl benzimidazol-
ium salts.
respectively, by following the procedure developed pre-
viously in our group for related imidazolium salts.[4]
The phosphane group was deprotected by desulfurisation
with Raney nickel to afford the ligand precursors 5a and
5b. Two related systems, either with or without methyl
groups (b and a, respectively) at positions 5 and 6 of the
benzimidazole moiety, were developed in order to check
whether subtle variations could induce electronic changes
on the metal centre or modify the physical properties of the
metal complexes.[12]
In parallel, compounds 6a and 6b were prepared by reac-
tion of N-(ferrocenylmethyl) trimethylammonium iodide (3)
with substituted benzimidazoles in refluxing acetonitrile
(Scheme 2), by following the literature procedure.[17] These
systems will allow a comparison of the performance of the
benzimidazol-2-ylidene/phopshane ligand systems either in
the presence or absence of a chelating effect to be made.
Figure 2. ORTEP views of 5a (top) and 6b (bottom). Ellipsoids are
shown at the 30% probability level. All hydrogen atoms except H2
(top) and H1 (bottom) are omitted for clarity. Selected bond
lengths [Å] and angles [°], 5a: C1–P1 1.811(2), C62–N1 1.325(3),
C62–N2 1.331(3), C63–C68 1.395(3), N1–C62–N2 110.5(2); 6b:
C1–N11 1.321(6), C1–N12 1.321(6), C112–C113 1.399(6), N11–
Scheme 2. Synthesis of ferrocenyl benzimidazolium salts.
All compounds were fully characterised by standard ana-
1
lytical methods. The H NMR spectra of compounds 5a
and 5b show signals at 9.10 and 8.86 ppm, respectively,
which are characteristic of the acidic C2 proton of the C1–N12 110.4(6).
Eur. J. Inorg. Chem. 2009, 1806–1815
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1807

S. Gülcemal, A. Labande, J.-C. Daran, B. Çetinkaya, R. Poli
FULL PAPER
imidazolium salt 5a, which possesses a weakly coordinating
anion, certainly because the F atom is much smaller than
the I atom, although the NMR study (vide supra) indicates
a stronger H bond for the iodide salt.
Synthesis and Characterisation of Rhodium(I) Complexes 7
and 8
Addition of compounds 5a or 5b to [Rh(cod)(OtBu)]2 –
prepared in situ from [Rh(cod)Cl]2 and tBuOK[18] – in thf
led to rhodium(I) complexes 7a or 7b. [Rh(OMe)(cod)]2 re-
acted with benzimidazolium salts 6a or 6b in dichlorometh-
ane to give complexes 8a or 8b, respectively (Scheme 3). The
reaction with ligand precursors 5a and 5b led directly to
cationic complexes, whereas the neutral form was obtained
in the case of precursors 6a and 6b with coordination of
the iodido ligand.
Figure 3. ORTEP views of 7a (top) and 7b (bottom). Ellipsoids are
shown at the 50% (7a) or 30% (7b) probability level. All hydrogen
atoms are omitted for clarity.
Scheme 3. Synthesis of rhodium complexes.
The 13C NMR signals of the carbene C atoms of com-
plexes 7 appear at δ = 192.5 ppm (7a) and 189.9 ppm (7b)
and lie within the expected range for cationic NHC–RhI
complexes with
a
benzimidazol-2-ylidene moiety.[10,19]
These signals are found at slightly lower fields for com-
plexes 8a (δ =196.4 ppm) and 8b (δ =194.5 ppm).[8,18]
Complexes 7 and 8 gave X-ray quality crystals by slow
diffusion of diethyl ether into a dichloromethane solution
(Figures 3 and 4, respectively). Selected bond lengths and
angles are listed in Table 1 and are compared to those of a
previously described complex 9a (Figure 5).[4] The square-
planar coordination environment in complexes 7 and 8 was
confirmed by X-ray diffraction studies. The cis-NHC–Rh–
Figure 4. ORTEP views of 8a (top) and 8b (bottom). Ellipsoids are
shown at the 50% (8a) or 30% (8b) probability level. All hydrogen
P angles in complexes 7a and 7b [92.06(18)° and 93.78(12)°, atoms are omitted for clarity.
1808
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1806–1815

Importance of the Chelating Effect for Ketone Hydrosilylation Catalysis
respectively] and the cis-NHC–Rh–I angles in complexes 8a [2.064(2) Å].[20] However, this difference is not observed
and 8b [89.57(6)° and 86.1(3)°, respectively] are close to the with neutral complexes 8a and 8b: the Rh1–C(NHC) bond
expected value of 90°. The structures of complexes 7a and lengths are in the typical range observed for NHC–Rh com-
7b reveal slightly shorter Rh1–C(NHC) bond lengths than plexes with coordination of iodido ligands.[8,9,21] There is
those in the imidazol-2-ylidene/phosphane complex 9a as no evidence for a structural effect of the methyl groups in
well as those in a related complex described by Field et al. positions 5 and 6 of the benzene ring (complexes 7b and
8b), as we cannot observe any significant variation in the
Rh1–C(NHC) bond lengths related to this parameter. The
Table 1. Selected bond lengths [Å] and bond angles [°] for com-
plexes 7, 8 and 9a.
mesityl group is found in a roughly orthogonal position to
the benzimidazole backbone, and dihedral angles range
from 86 to 90°, to limit steric interactions.
7a
7b
9a[a]
8a
8b
Bond lengths
Rh1–C(NHC)
Rh1–P1
Rh1–I1
2.018(7)
2.3138(18)
–
2.031(5) 2.047(3) 2.012(2) 1.993(10)
2.3233(12) 2.3345(8) –
2.6848(3) 2.6784(11)
–
Catalytic Hydrosilylation of Methyl Aryl Ketones
–
–
The catalytic activity of NHC complexes 7 and 8 was
evaluated in the hydrosilylation of carbonyl com-
pounds.[22,23] The reactions were typically carried out with
2 mol-% catalyst in thf at room temperature (Scheme 4,
Table 2). The conversion of acetophenone was followed by
TLC, and the crude mixture was analysed by 1H NMR
spectroscopy.
Rh1–C(cod)
2.180(7)
2.181(7)
2.209(7)
2.248(7)
1.355(8)
1.362(8)
2.209(5) 2.215(3) 2.143(2) 2.100(10)
2.209(5) 2.231(3) 2.109(2) 2.091(11)
2.239(4) 2.220(3) 2.234(2) 2.187(10)
2.229(4) 2.217(3) 2.204(2) 2.226(10)
1.367(5) 1.359(4) 1.366(3) 1.364(11)
1.373(6) 1.357(4) 1.360(3) 1.382(11)
N1–C(NHC)
N2–C(NHC)
Bond angles
N1–C(NHC)–N2 105.5(6)
C(NHC)–Rh1–P1 92.06(18)
105.2(4) 104.5(2) 106.02(18) 104.7(8)
93.78(12) 90.83(8)
–
–
C(NHC)–Rh1–I1
–
–
–
89.57(6) 86.1(3)
[a] Data from ref.[4]; chemically equivalent parameters are given on
the same row for all compounds.
Scheme 4. Hydrosilylation of acetophenone and its derivatives with
RhI complexes.
In our first attempts, we were pleased to note that all
complexes were active (Entries 1–2 and 6–7), although a
reaction time of 20 h was needed to obtain reasonable
yields. The yields of the alcohol were much higher in the
case of bifunctional, cationic complexes 7a and 7b. As ob-
served in some of our previous work, the concentration of
acetophenone in the medium can dramatically alter the re-
action rate.[5] We thus carried out the next catalytic tests
with a more concentrated medium (2  instead of 1 ). The
Figure 5. Previously described RhI complexes with an imidazol-2-
ylidene moiety.[4]
Table 2. Hydrosilylation of acetophenone (R = H).[a]
Entry Catalyst
(mol-%)
Additive
Concentration
[]
t
[h]
Conversion[b]
[%]
Ratio silyl ether/silyl enol ether[b] Yield (1-phenylethanol)[c]
[%]
1
2
3
4
5
6
7
8
7a (2)
7b (2)
7a (2)
7b (2)
7b (1)
8a (2)
8b (2)
8a (2)
8b (2)
10a (2)
10b (2)
10a (2)
10b (2)
9a (2)
9b (2)
–
–
–
–
–
–
–
–
1
1
2
2
2
1
1
2
2
2
2
2
2
2
2
20
20
20
2
n.c.
n.c.
80
n.c.
n.c.
96:4
98:2
96:4
n.c.
n.c.
n.c.
73
85
78[d]
100
100
n.c.
n.c.
n.c.
n.c.
100
100
100
100
63
98
2
95
20
20
8
8
2
2
20
2
5 d
5 d
43[d]
23
49[d]
9
–
–
–
n.c.
32
10
11
12
13
14[f]
15[f]
n.c.; 50:50; 59:41[e]
62; 46; 60[e]
83; 66; 58[e]
76[d]
n.c.; 69:31; 58:42[e]
PPh3 (1 equiv.)
PPh3 (1 equiv.)
–
–
78:22
87:13
Ͼ99:1
99:1
84
63[d]
90
87[d]
1
[a] Ketone (1 equiv.), diphenylsilane (1.1 equiv.), room temperature, thf, then hydrolysis with MeOH/HCl. [b] Determined by H NMR
1
signal integration of the crude reaction mixture before hydrolysis, n.c. = not calculated. [c] Determined by H NMR signal integration
of the crude reaction mixture after hydrolysis. [d] Catalyst only partially soluble in thf. [e] Results of three runs. [f] Data from ref.[4]
Eur. J. Inorg. Chem. 2009, 1806–1815
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1809

S. Gülcemal, A. Labande, J.-C. Daran, B. Çetinkaya, R. Poli
FULL PAPER
results were very encouraging, since we obtained a nearly
quantitative yield of the alcohol in 2 h with complex 7b
(Entry 4). A reaction time of 20 h was still needed with
complex 7a to reach 80% conversion of the ketone (Entry
3), probably because of the poor solubility of the catalyst
in thf. In comparison, our previously described complexes
9a and 9b were poorly active,[4] even at this concentration,
as the reactions were not complete after 5 d at room tem-
perature (Entries 14 and 15).
However, by increasing the concentration, the yield did
not improve in the case of neutral complexes 8a and 8b
(Entries 8 and 9). In order to have a better comparison with
cationic complexes 7a and 7b, 1 equiv. AgBF4 was added to
compounds 8a and 8b to give cationic complexes 10a and
10b, respectively (Scheme 5), in which the RhI centre is
probably coordinated by one thf molecule in order to
achieve the preferred square-planar configuration. These
complexes were not isolated but used directly in the cata-
lytic reactions.
Scheme 6. Side products generated during hydrosilylation reactions.
In order to see whether the phosphane was necessary to
maintain
a good selectivity, we added 1 equiv. tri-
phenylphosphane to complexes 10a and 10b prior to the
reaction (Entries 12 and 13). The reaction was complete
after 2 h with complex 10b, but needed 20 h with complex
10a, as the addition of PPh3 decreases the solubility of the
complex in thf. Analysis of the reaction mixture before and
after hydrolysis showed an increase in selectivity in both
cases, which confirms our hypothesis. However, the selectiv-
ity was lower than that obtained with complexes 7a and 7b,
which shows clearly that the chelating effect is beneficial for
reaction selectivity.
Encouraged by the results obtained with complex 7b, we
decided to lower the catalyst loading to 1 mol-% (Entry 5)
and were pleased to observe that the activity was preserved,
with a total conversion and 95% yield in only 2 h. Catalytic
tests with the acetophenone derivatives allowed us to deter-
mine the efficiency of complexes 7a and 7b (2 mol-%) with
a variation in the electronic or steric properties (Scheme 4,
Table 3). Whereas marked differences were observed with
complex 7a, catalyst 7b shows a very high activity irrespec-
tive of the substituents on acetophenone. Indeed, even the
presence of an ortho substituent did not lower the yield (En-
try 6).
Scheme 5. Generation of cationic RhI complexes bearing mono-
dentate NHC ligands.
Both complexes were completely soluble in thf, in con-
trast to their neutral precursors. It has been reported in
some other cases that the use of AgX (X = OSO2CF3, BF4,
PF6) was required in order to improve the catalytic ac-
tivity,[23–25] and this proved to be the case also with our
systems. Indeed, the conversion of acetophenone was com-
Table 3. Hydrosilylation of acetophenone derivatives.[a]
Entry
R
Catalyst
t [h]
Yield [%][b]
1
2
3
4
5
6
4-OMe
4-OMe
4-F
4-F
2-Me
2-Me
7a
7b
7a
7b
7a
7b
20
2
20
2
20
2
87
97
71
98
52
92
1
plete after only 2 h, as shown by the H NMR spectra of
the reaction mixture before hydrolysis of the silyl ether (En-
tries 10 and 11). However, the NMR yields calculated after
hydrolysis were rather low and hardly reproducible, which
shows a lack of selectivity of these systems for the forma-
1
[a] Ketone (1 equiv.), diphenylsilane (1.1 equiv.), room temperature,
2 mol-% catalyst, thf, then hydrolysis with MeOH/HCl. [b] Deter-
tion of the expected alcohol. The H NMR spectra of the
reaction mixtures before hydrolysis suggest the formation
of a silyl enol ether as by-product (Scheme 6), which would
then be reconverted to acetophenone by the hydrolysis step.
Indeed, the 1H NMR spectra before hydrolysis show two
1
mined by H NMR signal integration of the crude products after
hydrolysis.
The higher catalytic activity observed at higher ketone
doublets at δ = 4.65 and 5.05 ppm (=CH2, J = 2.4 Hz), concentrations agrees with previous reports of a saturation
which is characteristic of the silyl enol ether (the doublet at effect: the reaction was found to be first order with respect
δ = 5.05 ppm, however, is often hidden by the Si–H singlet to the ketone at low concentrations, but eventually became
of Ph2SiH2 at δ = 5.06 ppm). These signals are absent from independent of the ketone concentration at high concentra-
the 1H NMR spectra of the mixtures after hydrolysis, tions.[27–29] The mechanism of this catalytic process has
whereas two new signals, typical of acetophenone, appear been addressed by many authors. Although a few variations
at 2.6 ppm (CH3, s) and 8.0 ppm (Ar-H, d, J = 7.5 Hz). have been discussed, the most commonly accepted mecha-
This process is frequently observed in ketone hydro- nism remains that originally proposed,[27,30–32] which is
silylation.[26]
based on a rate-determining Si–H oxidative addition pro-
1810
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1806–1815

Importance of the Chelating Effect for Ketone Hydrosilylation Catalysis
Scheme 7. Proposed mechanism for the formation of saturated and unsaturated hydrosilylation products.
cess to yield a RhIII silyl hydride intermediate. This step posedly, the absence of the phosphane ligand renders the β-
would be followed by ketone coordination, insertion into H elimination pathway more easily accessible through the
the metal–silyl bond, and final reductive elimination. Inter- availability of an additional open coordination site.
estingly, the observation by Kolb and Hetflejs of a satura-
tion effect (while working with a cationic system) was inter-
preted as indicative of reversible complexation of the ketone Conclusions
prior to the rate-determining silane addition,[28] but Giering
The benzimidazol-2-ylidene unit gives highly active
later discarded this variant when he could properly model,
hydrosilylation rhodium catalysts that prove to be superior
on the basis of the other mechanism, the kinetic results that
to imidazol-2-ylidene. Moreover, bidentate NHC ligands
he obtained with a neutral system, which also included a
bearing a phosphanyl group ensure a much better selectivity
saturation effect in the ketone.[27] However, the data fittings
in the desired product than monodentate NHC ligands. The
were described as unsatisfactory unless additional processes
introduction of methyl substituents on the benzene ring of
of catalyst decomposition were included at various places
the benzimidazol-2-ylidene ligand enhances the solubility of
in the catalytic cycle. We believe that the interpretation of-
the rhodium complex in the reaction medium. This work
fered by Kolb and Hetflejs is reasonable and probably ap-
was carried out with achiral ligands, and we are now plan-
plies to our system, for the following reasons. The cod li-
ning to prepare planar chiral 1,2-disubstituted ferrocenyl
gand is probably lost rapidly from the coordination sphere
ligands, using the methodology developed here, for applica-
by the hydrosilylation or hydrogenation processes, as also
tions to asymmetric hydrosilylation.
observed or proposed in other cases.[27,32–34] The resulting
[(NHCFcPPh2)Rh]+ system is then likely to be coordinated
by two solvent molecules, but the lability of these should
Experimental Section
easily allow ketone coordination, at a rate faster than that
of the rate-determining silane oxidative addition step. Thus,
the rate-determining silane oxidative addition may preferen-
tially occur on the ketone-substituted complex, as shown in
Scheme 7, when the ketone concentration is high, whereas
the order of events may be inverted (pathway indicated with
dashed arrows in Scheme 7) at low ketone concentrations.
Finally, the formation of the unsaturated silyl ether can
be easily rationalised on the basis of a competition between
the reductive elimination of the hydrosilylation product and
a β-H elimination step from a common intermediate
(Scheme 7). Although the mechanism of ketone hydro-
silylation is not as well established as that of olefin hydro-
silylation,[35] there is good evidence supporting a ketone in-
sertion into the Rh–Si bond followed by C–H reductive eli-
mination, rather than the alternative insertion into the Rh–
All reactions were carried out under a dry argon atmosphere using
Schlenk glassware and vacuum line techniques. Solvents for synthe-
ses were dried and degassed by standard methods before use. Spec-
tra were recorded on Bruker ARX250, AV300 or DPX300 spec-
trometers. All spectra were recorded in CDCl3, unless otherwise
stated. Mass spectra were obtained from acetonitrile solutions on
a TSQ7000 instrument from ThermoElectron. (1Ј-Diphenylthio-
phosphanylferrocen-1-yl)methanol (1) was prepared according to
our previously published procedure.[4,37] [Rh(OMe)(cod)]2,[38] 2a,
2b[7]and 3[39] were prepared according to literature procedures.
Benzimidazolium Salt 4a: HBF4 (35 µL, 54 wt.-% in Et2O) was
added quickly to a solution of ferrocenyl alcohol 1 (100 mg,
0.23 mmol) in degassed dichloromethane (5 mL). N-(2,4,6-Trimeth-
ylbenzyl) benzimidazole (2a, 86 mg, 0.35 mmol) was then added
immediately. The mixture was washed with 2  aq. HCl, water, sat-
urated aq. NaHCO3 and water again. The organic phase was dried
H bond followed by Si–C reductive elimination.[24,36] Sup- (MgSO4), filtered, and concentrated in vacuo. The residue was
Eur. J. Inorg. Chem. 2009, 1806–1815
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1811

S. Gülcemal, A. Labande, J.-C. Daran, B. Çetinkaya, R. Poli
FULL PAPER
purified by column chromatography on silica gel (eluent: CH2Cl2/
acetone, 9:1) to give a yellow–orange solid (121 mg, 70% yield).
Benzimidazolium Salt 5b: Raney nickel (2.4 g, suspension in water)
was washed with degassed methanol (3ϫ10 mL), degassed diethyl
ether (3ϫ10 mL) and, finally, degassed acetonitrile (3ϫ10 mL). A
C40H38BF4FeN2PS (752.1): calcd. C 63.82, H 5.05, N 3.72; found
1
C 63.60, H 5.32, N 3.64. H NMR (300 MHz, CDCl3, 25 °C): δ = solution of 4b (450 mg, 0.58 mmol) in acetonitrile (10 mL) was
9.30 (s, 1 H, NCHN+), 7.98 (d, JH,H = 8.4 Hz, 1 H, benzimidazole),
7.29 (d, JH,H = 8.4 Hz, 1 H, benzimidazole), 7.43–7.67 (m, 12 H,
benzimidazole + ArH), 6.98 (s, 2 H, Mes), 5.61 (s, 2 H, CH2-Mes),
added; the mixture was stirred at room temperature for 24 h, fil-
tered through a short path of Celite® and rinsed with acetonitrile.
The solution was concentrated and the orange solid was dried in
5.50 (s, 2 H, CH2-Fc), 4.75, 4.71, 4.48, 4.08 (t, JH,H = 1.8 Hz, 8 H, vacuo (370 mg, 85% yield). C42H42BF4FeN2P (748.1): calcd. C
Cp), 2.33 (s, 3 H, p-CH3 Mes), 2.26 (s, 6 H, o-CH3 Mes) ppm. 67.37, H 5.61, N 3.74; found C 67.23, H 5.62, N 3.88. 1H NMR
13C{1H} NMR (75.5 MHz, CDCl3, 25 °C): δ = 141.32 (NCN+), (300 MHz, CDCl3, 25 °C): δ = 8.86 (s, 1 H, NCHN+), 7.59 (s, 1
140.20, 137.93, 134.55, 133.39, 131.69, 131.55, 131.49, 131.41,
128.49, 128.33, 127.26, 124.32, 113.89, 113.41 (Ar-C), 130.31
(Mes), 80.73, 74.31, 74.15, 73.47, 73.33, 72.06, 70.91 (Cp), 47.01
(CH2-Mes), 46.98 (CH2-Fc), 21.09 (p-CH3 Mes), 19.84 (o-CH3
H), 7.41–7.32 (m, 8 H), 7.31 (s, 1 H), 7.26 (s, 1 H), 7.11 (s, 1 H,
benzimidazole + Ar-H), 6.98 (s, 2 H, Mes), 5.51, (s, 2 H, CH2-
Mes), 5.21, 5.07 (s, 2 H, CH2-Fc), 4.58, (t, JH,H = 1.8 Hz, 2 H,
Cp), 4.41 (s, 1 H, Cp), 4.33 (t, JH,H = 1.8 Hz, 2 H, Cp), 4.21 (t,
Mes) ppm. 31P{1H} NMR (121.5 MHz, CDCl3, 25 °C): δ = JH,H = 1.8 Hz, 1 H, Cp), 4.13 (d, JH,H = 1.8 Hz, 2 H, Cp), 2.45,
41.6 ppm.
2.41 (s, 6 H, o-CH3 Mes), 2.35, 2.33 (s, 3 H, p-CH3 Mes), 2.25
(d, JH,H = 4.2 Hz, 6 H, CH3 benzimidazole) ppm. 13C{1H} NMR
(75.5 MHz, CDCl3, 25 °C): δ = 139.66 (NCN+), 138.87, 137.97,
137.33, 137.25, 133.53, 128.73, 128.31, 113.16, 112.97 (Ar-C),
130.19 (Mes), 79.82, 74.10, 72.45, 70.60, 70.37, 65.85 (Cp), 47.03
(CH2-Mes), 46.40 (CH2-Fc), 21.06, 20.76, 20.73 (CH3 Mes), 19.75
(CH3 benzimidazole) ppm. 31P{1H} NMR (121.5 MHz, CDCl3,
25 °C): δ = –17.1 ppm.
Benzimidazolium Salt 4b: HBF4 (35 µL, 54 wt.-% in Et2O) was
added quickly to a solution of ferrocenyl alcohol 1 (100 mg,
0.23 mmol) in degassed dichloromethane (5 mL). N-(2,4,6-Trimeth-
ylbenzyl)-5,6-dimethyl benzimidazole (2b, 96 mg, 0.35 mmol) was
then added immediately. The mixture was washed with 2  aq.
HCl, water, saturated aq. NaHCO3 and water again. The organic
phase was dried (MgSO4), filtered, and concentrated in vacuo. The
residue was purified by column chromatography on silica gel (elu-
ent: CH2Cl2/acetone, 9:1) to give a yellow–orange solid (108 mg,
60% yield). C42H42BF4FeN2PS (780.1): calcd. C 64.61, H 5.38, N
Benzimidazolium Salt 6a: (Ferrocenylmethyl)trimethylammonium
iodide (3, 465 mg, 1.2 mmol) was slowly added to a solution of
N-(2,4,6-trimethylbenzyl) benzimidazole (2a, 325 mg, 1.3 mmol) in
acetonitrile (10 mL). The mixture was heated to reflux for 48 h,
1
3.59; found C 64.67, H 5.82, N 3.85. H NMR (300 MHz, CDCl3,
25 °C): δ = 9.01 (s, 1 H, NCHN+), 7.78–7.70 (m, 3 H), 7.58 (s, 1 after which the solvent was removed in vacuo. The residue was
H), 7.51–7.45 (m, 5 H), 7.31 (s, 1 H), 7.26 (s, 1 H), 7.10 (s, 1 H,
benzimidazole + Ar-H), 6.99 (s, 2 H, Mes), 5.51, 5.39 (s, 2 H, CH2-
Mes), 5.32, 5.21 (s, 2 H, CH2-Fc), 4.74, (t, JH,H = 1.8 Hz, 1 H,
purified by column chromatography on silica gel (eluent: CH2Cl2/
acetone, 8:2) to give an orange solid (485 mg, 70% yield).
C28H29FeIN2 (578.8): calcd. C 58.05, H 5.01, N 4.84; found C
Cp), 4.66 (t, JH,H = 1.8 Hz, 1 H, Cp), 4.55 (d, JH,H = 1.8 Hz, 2 H, 58.10, H 5.08, N 4.37. 1H NMR (300 MHz, CDCl3, 25 °C): δ =
Cp), 4.48–4.47 (m, 1 H, Cp), 4.30 (s, 1 H, Cp), 4.07 (t, JH,H
1.8 Hz, 1 H, Cp), 4.00 (s, 1 H, Cp), 2.49, 2.44, 2.41 (s, 6 H, o-CH3
Mes), 2.33 (t, JH,H =.3 Hz, 3 H, p-CH3 Mes), 2.26 (d, JH,H
1.2 Hz, 6 H, CH3 benzimidazole) ppm. 13C{1H} NMR (75.5 MHz,
=
10.60 (s, 1 H, NCHN+), 7.77, 7.30 (dd, JH,H = 8.4 Hz, 2 H, benz-
imidazole), 7.57, 7.48 (dt, JH,H = 6.2 Hz, 2 H, benzimidazole), 6.96
(s, 2 H, Mes), 5.72 (s, 2 H, CH2-Mes), 5.32 (s, 2 H, CH2-Fc), 4.55,
4.31, 4.23 (t, JH,H = 1.8 Hz, 9 H, Cp), 2.32 (s, 9 H, CH3Mes) ppm.
=
CDCl3, 25 °C): δ = 139.71 (NCN+), 137.92, 137.87, 137.54, 137.42, 13C{1H} NMR (75.5 MHz, CDCl3, 25 °C): δ = 140.96 (NCN+),
133.46, 131.69, 131.54, 131.40, 131.36, 128.48, 128.36, 128.32,
140.03, 137.98, 131.35 (d, JC,H = 4.5 Hz), 127.22 (d, JC,H
=
128.20, 124.49, 120.31, 113.51, 112.89, 109.77 (Ar-C), 130.22
10.6 Hz), 124.65, 113.67 (d, JC,H = 12.8 Hz, Ar-C), 130.32 (Mes),
(Mes), 81.01, 74.28, 74.11, 73.46, 73.33, 72.18, 72.05, 71.93, 70.84, 79.20, 69.77, 69.64, 69.55, 69.48, 69.35, 69.30 (Cp), 48.09 (CH2-
69.29, 69.22, 60.10 (Cp), 46.66, 46.49 (CH2-Mes), 43.03 (CH2-Fc),
21.08, 21.04 (o-CH3 Mes), 20.78, 20.65, 20.29 (p-CH3 Mes), 19.78,
19.56 (CH3 benzimidazole) ppm. 31P{1H} NMR (121.5 MHz,
CDCl3, 25 °C): δ = 41.5 ppm.
Mes), 47.12 (CH2-Fc), 21.13 (p-CH3 Mes), 20.44 (d, JC,H = 4.5 Hz,
o-CH3 Mes) ppm.
Benzimidazolium Salt 6b: (Ferrocenylmethyl)trimethylammonium
iodide (3, 465 mg, 1.2 mmol) was slowly added to a solution of
N-(2,4,6-trimethylbenzyl)-5,6-dimethyl benzimidazole (2b, 361 mg,
1.3 mmol) in acetonitrile (10 mL). The mixture was heated to reflux
for 48 h, after which the solvent was removed in vacuo. The residue
was purified by column chromatography on silica gel (eluent:
CH2Cl2/acetone, 8:2) to give an orange solid (590 mg, 81% yield).
C30H33FeIN2 (606.8): calcd. C 59.33, H 5.44, N 4.61; found C
61.01, H 6.08, N 4.80. 1H NMR (300 MHz, CDCl3, 25 °C): δ =
10.21 (s, 1 H, NCHN+), 7.11, 7.27, 7.32, 7.50, 7.59 (s, 2 H, benz-
Benzimidazolium Salt 5a: Raney nickel (2.2 g, suspension in water)
was washed with degassed methanol (3ϫ10 mL), degassed diethyl
ether (3ϫ10 mL) and, finally, degassed acetonitrile (3ϫ10 mL). A
solution of 4a (500 mg, 0.67 mmol) in acetonitrile (10 mL) was
added; the mixture was stirred at room temperature for 24 h, fil-
tered through a short path of Celite® and rinsed with acetonitrile.
The solution was concentrated, and the orange solid was dried in
vacuo (420 mg, 87% yield). C40H38BF4FeN2P (720.1): calcd. C
66.66, H 5.28, N 3.89; found C 65.01, H 5.99, N 3.92. 1H NMR imidazole), 6.98 (s, 2 H, Mes), 5.64 (d, JH,H = 8.7 Hz, 2 H, CH2-
(300 MHz, CDCl3, 25 °C): δ = 9.10 (s, 1 H, NCHN+), 7.70–7.32 Mes), 5.22 (s, 2 H, CH2-Fc), 4.52, (t, JH,H = 2.1 Hz, 2 H, Cp), 4.30,
(m, 12 H, benzimidazole + Ar-H), 6.96 (s, 2 H, Mes), 5.60 (s, 2 H,
CH2-Mes), 5.15 (s, 2 H, CH2-Fc), 4.58, 4.36, 4.22, 4.13 (t, JH,H
1.8 Hz, 8 H, Cp), 2.32 (s, 3 H, p-CH3 Mes), 2.25 (s, 6 H, o-CH3
Mes) ppm. 13C{1H} NMR (75.5 MHz, CDCl3, 25 °C): δ = 141.20
(NCN+), 140.14, 138.00, 133.64, 131.52, 131.33, 129.62, 128.89,
128.30, 127.22, 127.04, 124.49, 113.60, 113.53 (Ar-C), 130.26
(s, 5 H, Cp), 4.22 (t, JH,H = 2.1 Hz, 2 H, Cp), 2.43 (d, JH,H =
=
10.2 Hz, 6 H, o-CH3-Mes), 2.34 (d, JH,H = 8.1 Hz, 6 H, CH3 benz-
imidazole), 2.27 (s, 3 H, p-CH3-Mes) ppm. 13C{1H} NMR
(75.5 MHz, CDCl3, 25 °C): δ = 140.95 (NCN+), 140.00, 139.86,
138.72, 137.99, 137.90, 137.31, 137.26, 132.00, 130.29, 130.12,
129.97, 129.62, 127.37, 124.82, 120.35, 113.18, 113.04, 109.78, (Ar-
(Mes), 79.54, 72.56, 70.73, 70.45 (Cp), 47.31 (CH2-Mes), 46.91 C), 131.15 (Mes), 79.53 (Cp), 69.32 (m, Cp), 47.77, 46.59 (CH2-
(CH2-Fc), 21.07 (p-CH3 Mes), 19.85 (o-CH3 Mes) ppm. 31P{1H}
NMR (121.5 MHz, CDCl3, 25 °C): δ = –17.2 ppm.
Mes), 43.05 (CH2-Fc), 19.58 (p-CH3 Mes), 20.31, 20.42 (CH3 benz-
imidazole), 20.75, 21.92 (o-CH3 Mes) ppm.
1812
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1806–1815

Importance of the Chelating Effect for Ketone Hydrosilylation Catalysis
Rhodium Complex 7a: tBuOK (23.3 mg, 0.21 mmol) and [Rh(cod)-
Cl]2 (47.6 mg, 0.097 mmol) were placed in a Schlenk tube; the sol-
ids were degassed, and thf (10 mL) was added. The solution was
stirred for 30 min at room temperature and then slowly added to
5a (139 mg, 0.194 mmol), which was precooled to –78 °C. The mix-
ture was stirred at –78 °C for 15 min and at room temperature for
2 h. The solvent was evaporated, and the residue taken up into
degassed CH2Cl2 (10 mL). AgBF4 (56.7 mg, 0.291 mmol) was
added, and the mixture was stirred for 30 min at room temperature
and filtered through Celite®. The solvent was evaporated in vacuo,
and the residue was filtered through silica gel (eluent: CH2Cl2/ace-
tone, 95:5) to give 7a as an orange solid (154 mg, 86% yield).
column chromatography on silica gel (eluent: CH2Cl2) to give a
yellow–orange solid (135 mg, 86% yield). C36H40FeIN2Rh (785.7):
1
calcd. C 54.98, H 5.09, N 3.56; found C 52.85, H 4.71, N 3.32. H
NMR (250 MHz, CDCl3, 25 °C): δ = 7.21 (d, JH,H = 15 Hz, 1 H,
benzimidazole), 6.94 (t, JH,H = 7.5 Hz, 1 H, benzimidazole), 6.78
(t, JH,H = 7.5 Hz, 1 H, benzimidazole), 6.32 (benzimidazole), 6.95
(s, 2 H, Mes), 6.10 (t, JH,H = 12.5 Hz, 2 H, CH2-Mes), 5.83 (q,
JH,H = 15 Hz, 2 H, CH2-Fc), 4.83 (d, JH,H = 1.8 Hz, 2 H, Cp), 4.48
(d, JH,H = 1.8 Hz, 2 H, Cp), 4.35 (s, 2 H, Cp), 4.19 (t, JH,H
=
1.8 Hz, 2 H, Cp), 5.52, 5.39 (s, 2 H, cod-CH), 3.66 (s, 2 H, cod-
CH), 2.49 (s, 4 H, cod-CH2), 1.94 (s, 4 H, cod-CH2), 2.37 (s, 3
H, p-CH3 Mes), 2.32 (s, 6 H, o-CH3 Mes) ppm. 13C{1H} NMR
C48H49BF4FeN2PRh (929.6): calcd. C 61.96, H 5.27, N 3.01; found (62.9 MHz, CDCl3, 25 °C): δ = 196.38 (d, JRh,C = 50.6 Hz,
1
C 61.55, H 5.60, N 2.98. H NMR (300 MHz, CDCl3, 25 °C): δ = Ccarbene), 138.82, 138.53, 135.86, 134.69, 128.20, 122.05, 121.47,
7.51–7.30 (m, 10 H, Ar-H), 7.62 (d, JH,H = 8.4 Hz, 1 H, benzimid-
azole), 7.46 (d, JH,H = 15.3 Hz, 1 H, benzimidazole), 6.26 (d, JH,H
= 15.3 Hz, 1 H, benzimidazole), 5.77 (d, JH,H = 8.4 Hz, 1 H, benz-
imidazole), 6.93 (s, 2 H, Mes), 5.47 (d, JH,H = 15 Hz, 2 H, CH2-
Mes), 4.97 (d, JH,H = 15 Hz, 2 H, CH2-Fc), 4.49, 4.24, 4.11 (s, 8
H, Cp), 5.18 (s, 2 H, cod-CH), 4.40 (s, 2 H, cod-CH), 2.53 (s, 4 H,
cod-CH2), 2.31 (s, 4 H, cod-CH2), 2.37 (s, 6 H, o-CH3 Mes), 1.82
(s, 3 H, p-CH3 Mes) ppm. 13C{1H} NMR (75.5 MHz, CDCl3,
25 °C): δ = 192.52 (d, JRh,C = 50.8 Hz, Ccarbene), 139.78, 138.55,
134.11, 131.96, 131.65, 130.92, 129.96, 128.71, 126.10, 123.65,
111.38, 110.93 (Ar-C), 129.82 (Mes), 71.26, 69.98, 69.31, 68.44
(Cp), 99.13, 96.50, 73.46 (cod-CH), 30.90, 30.42 (cod-CH2), 50.80
(CH2-Fc), 47.12 (CH2-Mes), 21.12 (o-CH3 Mes), 19.89 (p-CH3
Mes) ppm. 31P{1H} NMR (121.5 MHz, CDCl3, 25 °C): δ = 18.4
(d, JP,Rh = 154.3 Hz) ppm.
110.71, 110.60 (Ar-C), 129.51 (Mes), 82.08, 71.15, 69.50, 69.22,
68.91, 67.76 (Cp), 97.74, 97.63, 97.55, 97.44, 72.50, 72.28, 72.11,
71.89 (cod-CH), 32.73, 32.08, 30.16, 29.00 (cod-CH2), 50.45 (CH2-
Mes), 49.92 (CH2-Fc), 21.11 (p-CH3 Mes) 21.07 (o-CH3 Mes) ppm.
Rhodium Complex 8b: A mixture of 6b (121 mg, 0.2 mmol) and
[Rh(OMe)(cod)]2 (48.4 mg, 0.1 mmol) in CH2Cl2 (5 mL) was
stirred at room temperature for 6 h. The volume of the solvent was
reduced to 2 mL, pentane (10 mL) was added, and the yellow pre-
cipitate was recovered by filtration. The residue was purified by
column chromatography on silica gel (eluent: CH2Cl2) to give a
yellow–orange solid (138 mg, 84% yield). C38H44FeIN2Rh (813.7):
1
calcd. C 56.04, H 5.41, N 3.44; found C 56.85, H 5.81, N 3.33. H
NMR (300 MHz, CDCl3, 25 °C): δ = 6.97 (s, 2 H, benzimidazole),
6.95 (s, 2 H, Mes), 6.03 (d, JH,H = 15 Hz, 1 H, CH2-Mes), 6.23 (d,
JH,H = 15 Hz, 1 H, CH2-Mes), 5.78 (q, JH,H = 12 Hz, 2 H, CH2-
Fc), 4.80 (d, JH,H = 1.2 Hz, 2 H, Cp), 4.47 (d, JH,H = 1.2 Hz, 2 H,
Cp), 4.33 (d, JH,H = 1.2 Hz, 2 H, Cp), 4.17 (t, JH,H = 1.2 Hz, 2 H,
Cp), 5.46, 5.35 (s, 2 H, cod-CH), 3.60 (s, 2 H, cod-CH), 2.47 (s, 4
H, cod-CH2), 1.95 (s, 4 H, cod-CH2), 2.37 (s, 3 H, p-CH3 Mes),
2.31 (s, 6 H, o-CH3 Mes), 2.17 (s, 6 H, CH3 benzimidazole) ppm.
Rhodium Complex 7b: tBuOK (19 mg, 0.171 mmol) and [Rh(cod)-
Cl]2 (38.5 mg, 0.078 mmol) were placed in a Schlenk tube; the sol-
ids were degassed, and thf (10 mL) was added. The solution was
stirred for 30 min at room temperature and then slowly added to
5b (117 mg, 0.156 mmol), which was precooled to –78 °C. The mix-
ture was stirred at –78 °C for 15 min and at room temperature for
2 h. The solvent was evaporated, and the residue taken up into
degassed CH2Cl2 (10 mL). AgBF4 (45.6 mg, 0.234 mmol) was
added, and the mixture was stirred for 30 min at room temperature
and filtered through Celite®. The solvent was evaporated in vacuo,
and the residue was filtered through silica gel (eluent: CH2Cl2/ace-
tone, 95:5) to give 7b as an orange solid (126 mg, 84% yield).
C50H53BF4FeN2PRh (957.6): calcd. C 62.66, H 5.54, N 2.92; found
13C{1H} NMR (75.5 MHz, CDCl3, 25 °C): δ = 194.51 (d, JRh,C
=
50.8 Hz, Ccarbene), 138.83, 138.37, 134.51, 133.32, 130.61, 130.26,
129.35, 128.56, 111.15 (Ar-C), 129.47 (Mes), 82.43, 71.07, 69.38,
68.76, 67.69 (Cp), 97.36, 97.27, 97.16, 97.08, 72.35, 72.16, 71.98,
71.80 (cod-CH), 32.70, 32.10, 30.11, 29.03 (cod-CH2), 50.08 (CH2-
Mes), 49.68 (CH2-Fc), 21.06 (p-CH3 Mes) 20.29 (o-CH3 Mes),
20.15 (CH3 benzimidazole) ppm.
Representative Procedure for RhI-Catalysed Hydrosilylation of
Ketones: The RhI complex (10–6 mol, 2 mol-%) was placed in a
flame-dried Schlenk flask under argon. Anhydrous thf (0.25 mL)
and then acetophenone (56 µL, 4.8ϫ10–4 mol) were added by sy-
ringe. Diphenylsilane (100 µL, 5.4ϫ10–4 mol) was added slowly,
the reaction mixture was stirred at room temperature, and the con-
version of acetophenone followed by TLC analysis. After the reac-
tion was complete, the solvent was evaporated in vacuo, and the
residue analysed by 1H NMR (CDCl3) spectroscopy. The NMR
sample was taken up into CH2Cl2, MeOH (1 mL) was added, and
the reaction mixture stirred for 1 h at room temperature. After the
addition of 1  HCl (1 mL), the reaction mixture was stirred for
one additional hour at room temperature. The phases were sepa-
rated, the aqueous phase extracted with CH2Cl2, and the organic
phases were dried (MgSO4), filtered, and concentrated in vacuo.
1
C 61.40, H 5.66, N 2.89. H NMR (300 MHz, CDCl3, 25 °C): δ =
7.51–7.39 (m, 10 H, Ar-H), 7.33 (s, 1 H, benzimidazole), 5.39 (s, 1
H, benzimidazole), 6.94 (s, 2 H, Mes), 5.33 (d, JH,H = 9 Hz, 2 H,
CH2-Mes), 4.87 (d, JH,H = 15 Hz, 2 H, CH2-Fc), 4.48, 4.45, 4.25,
4.15 (s, 8 H, Cp), 5.14 (s, 2 H, cod-CH), 4.38 (s, 2 H, cod-CH),
2.52 (s, 4 H, cod-CH2), 2.41 (s, 4 H, cod-CH2), 2.39 (s, 6 H, o-CH3
Mes), 1.99 (s, 3 H, p-CH3 Mes), 1.81 (s, 6 H, CH3 benzimidazole)
ppm. 13C{1H} NMR (75.5 MHz, CDCl3, 25 °C): δ = 189.84 (d,
JRh,C = 50.7 Hz, Ccarbene), 139.69, 138.68, 134.27, 134.10, 133.87,
132.80, 132.67, 131.64, 130.86, 129.33, 129.19, 128.75, 128.62,
126.40, 111.87, 110.87 (Ar-C), 129.60 (Mes), 71.31, 69.91, 69.24,
68.43 (Cp), 98.89, 96.47, 96.33, 73.58 (cod-CH), 30.93, 30.43, 22.34
(cod-CH2), 50.52 (CH2-Fc), 46.99 (CH2-Mes), 21.06 (o-CH3 Mes),
20.57, 20.26 (p-CH3 Mes), 19.89 (CH3 benzimidazole) ppm.
31P{1H} NMR (121.5 MHz, CDCl3, 25 °C): δ = 18.55 (d, JP,Rh
=
1
The residue was analysed by H NMR spectroscopy.
159.2 Hz) ppm.
Rhodium Complex 8a: A mixture of 6a (116 mg, 0.2 mmol) and
X-ray Diffraction Studies: A single crystal of each compound was
[Rh(OMe)(cod)]2 (48.4 mg, 0.1 mmol) in CH2Cl2 (5 mL) was mounted under inert perfluoropolyether at the tip of a glass fibre
stirred at room temperature for 6 h. The volume of the solvent was
reduced to 2 mL, pentane (10 mL) was added, and the yellow pre-
cipitate was recovered by filtration. The residue was purified by
and cooled in the cryostream of either an Oxford-Diffraction XCA-
LIBUR CCD diffractometer for 6b, 7b, 7a and 8a, a Stoe IPDS
diffractometer for 8b or a Bruker APEXII CCD diffractometer for
Eur. J. Inorg. Chem. 2009, 1806–1815
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1813

S. Gülcemal, A. Labande, J.-C. Daran, B. Çetinkaya, R. Poli
5a. Data were collected by using the monochromatic Mo-Kα radia-
FULL PAPER
Table 4. Crystal data and structure refinement parameters for the
benzimidazolium salts.
tion (λ = 0.71073 Å). The structures were solved by direct methods
[40]
(SIR97)
and refined by least-squares procedures on F2 with
Compound
5a
6b
SHELXL-97.[41] All H atoms attached to carbon were introduced
in the calculation in idealised positions and treated as riding mod-
els. In compound 6b, some residual electron density was difficult
Empirical formula
Formula weight
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
β [°]
C40H38BF4FeN2P
720.35
180(2)
0.71073
monoclinic
P21/c
16.5808(16)
16.7482(15)
12.9508(11)
90
(C30H33FeIN2)2
1208.67
180(2)
[42]
to model, and, therefore, the SQUEEZE function of PLATON
0.71073
was used to eliminate the contribution of the electron density in
the solvent region from the intensity data, and the solvent-free
model was employed for the final refinement. There are two cavities
of about 173 Å3 per unit cell. PLATON estimated that each cavity
contains 12 electrons, which may correspond to half a molecule of
acetonitrile as suggested by chemical analyses. In compound 7b
and 8a, the SQUEEZE function was also used. In 7b, there are two
cavities of 470 Å3 per unit cell. PLATON estimated that each cavity
contains 190 electrons, which may correspond to roughly a mixture
of four CH2Cl2 and four (C2H5)2O molecules within the cell. In
compound 8a, there are two cavities of 172 Å3 per unit cell, which
may roughly correspond to a mixture of two CH2Cl2 and two
(C2H5)2O molecules within the cell. The drawing of the molecules
was realised with the help of ORTEP32.[43] Crystal data and refine-
ment parameters are shown in Tables 4 and 5. CCDC-710144,
-710145, -710146, -710147, CCDC-711381, and -711382 contain
the supplementary crystallographic data for this paper. These data
can be obtained free of charge from The Cambridge Crystallo-
graphic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.
triclinic
¯
P1
10.0316(8)
13.7879(10)
22.6738(14)
103.671(6)
97.411(6)
102.822(7)
2916.3(4)
2
92.832(5)
90
3592.0(6)
4
γ [°]
Volume [Å3]
Z
Density (calculated) [Mg/m3] 1.332
1.376
1.594
1224
Absorption coefficient [mm–1] 0.516
F(000)
Crystal size [mm]
θ range [°]
Reflections collected
Independent reflections (Rint
Completeness [%]
1496
0.84ϫ0.163ϫ0.068 0.75ϫ0.43ϫ0.26
2.38–27.02
120450
7757 (0.0836)
98.7
multiscan
1.0, 0.404
F2
7757/0/445
1.031
0.0418, 0.1013
0.0686, 0.1154
0.611/–0.467
2.77–27.10
23199
12820 (0.0407)
99.5
multiscan
1.0, 0.635
F2
12820/0/623
1.078
)
Absorption correction
Max., min. transmission
Refinement method
Data/restraints/parameters
Goodness-of-fit on F2
R1, wR2 [IϾ2σ(I)]
R1, wR2 (all data)
Acknowledgments
We thank the Centre National de la Recherche Scientifique
(CNRS) and the Institut Universitaire de France (IUF) for support
of this work.
0.0608, 0.1374
0.0766, 0.1434
1.536/–1.232
Residual density [eÅ–3]
Table 5. Crystal data and structure refinement parameters for the rhodium complexes.
Compound
7a
7b
8a
8b
Empirical formula
Formula weight
Temperature [K]
Wavelength [Å]
Crystal system
Space group
a [Å]
b [Å]
c [Å]
α [°]
C46H45BF4FeN2PRh
891.57
180(2)
0.71073
monoclinic
P21/c
10.4291(4)
27.7380(12)
14.1065(6)
90
C50H51BF4FeN2PRh
956.47
180(2)
0.71073
monoclinic
P21/n
14.6210(15)
13.7691(11)
24.724(2)
90
C36H40FeIN2Rh
786.36
180(2)
0.71073
tetragonal
I4
28.3187(10)
28.3187(10)
8.0375(6)
90
C38H44FeIN2Rh
814.41
296(2)
0.71073
monoclinic
P21/n
12.5094(14)
15.0659(12)
18.3417(19)
90
β [°]
γ [°]
91.019
90
4080.1(3)
4
1.515
91.663(3)
90
4975.3(8)
4
1.277
0.703
90
90
6445.7(6)
8
1.621
98.151(13)
90
3421.9(6)
4
1.581
1.837
Volume [Å3]
Z
Density (calculated) [Mg/m3]
Absorption coefficient [mm–1]
F(000)
0.855
1912
1.947
3152
1968
1640
Crystal size [mm]
θ range [°]
Reflections collected
Independent reflections (Rint
Completeness [%]
0.336ϫ0.125ϫ0.080
2.82–26.37
30948
8322 (0.0756)
99.9
0.293ϫ0.085ϫ0.039
1.64–22.12
79794
6134 (0.1124)
98.8
0.44ϫ0.08ϫ0.071
2.27–33.22
128931
12376 (0.0463)
99.9
0.40ϫ0.40ϫ0.16
1.76–24.20
27214
5458 (0.1564)
99.0
)
Absorption correction
Max., min. transmission
Refinement method
multiscan
1.0, 0.970
F2
semiempirical from equivalents semiempirical from equivalents
multiscan
0.5312, 0.4354
F2
1.0, 0.784
1.0, 0.881
F2
F2
Data/restraints/parameters
Goodness-of-fit on F2
R1, wR2 [IϾ2σ(I)]
8322/51/526
1.243
0.0792, 0.1418
0.1146, 0.1500
1.068/–1.517
6134/0/546
1.076
0.0406, 0.0978
0.0598, 0.1051
0.508/–0.403
12376/1/373
1.035
0.0294, 0.0674
0.0336, 0.0689
0.600/–0.359
5458/0/393
0.790
0.0544, 0.1175
0.1360, 0.1409
1.200/–1.130
R1, wR2 (all data)
Residual density [eÅ–3]
1814
www.eurjic.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2009, 1806–1815

Importance of the Chelating Effect for Ketone Hydrosilylation Catalysis
[19] A. S. Veige, M. S. Jeletic, R. J. Lowry, PCT Int.
WO2008101197, 2008.
[20] L. D. Field, B. A. Messerle, K. Q. Vuong, P. Turner, Organome-
tallics 2005, 24, 4241–4250.
[1]
[2]
a) P. L. Arnold, S. T. Liddle, Chem. Commun. 2006, 3959–3971;
O. Kühl, Chem. Soc. Rev. 2007, 36, 592–607; b) A. T. Norm-
and, K. J. Cavell, Eur. J. Inorg. Chem. 2008, 2781–2800; c) J.
Wolf, A. Labande, J.-C. Daran, R. Poli, J. Organomet. Chem.
2006, 691, 433–443; d) J. Wolf, A. Labande, M. Natella, J.-C.
Daran, R. Poli, J. Mol. Catal. A 2006, 259, 205–212.
a) S. Gischig, A. Togni, Organometallics 2004, 23, 2479–2487;
b) H. Seo, H.-J. Park, B. Y. Kim, J. H. Lee, S. U. Son, Y. K.
Chung, Organometallics 2003, 22, 618–620; c) H. Willms, W.
Frank, C. Ganter, Chem. Eur. J. 2008, 14, 2719–2729.
J.-c. Shi, P. Yang, Q. Tong, L. Jia, Dalton Trans. 2008, 938–945.
A. Labande, J.-C. Daran, E. Manoury, R. Poli, Eur. J. Inorg.
Chem. 2007, 1205–1209.
[21] a) H. Seo, B. Y. Kim, J. H. Lee, H.-J. Park, S. U. Son, Y. K.
Chung, Organometallics 2003, 22, 4783–4791; b) M. Poyatos,
M. Sanaú, E. Peris, Inorg. Chem. 2003, 42, 2572–2576; c) M. V.
Baker, S. K. Brayshaw, B. W. Skelton, A. H. White, Inorg.
Chim. Acta 2004, 357, 2841–2849; d) G. T. S. Andavan, E. B.
Bauer, C. S. Letko, T. K. Hollis, F. S. Tham, J. Organomet.
Chem. 2005, 690, 5938–5947; e) H. Karabiyik, R. Kilinçarslan,
M. Aygün, B. Çetinkaya, O. Büyükgüngür, J. Coord. Chem.
2007, 60, 393–399.
[22] a) A. R. Chianese, R. H. Crabtree, Organometallics 2005, 24,
4432–4436; b) C. Rivera, R. H. Crabtree, J. Mol. Catal. A 2004,
222, 59–73; c) J. W. Faller, P. P. Fontaine, Organometallics 2006,
25, 5887–5893.
[3]
[4]
[5]
[6]
J. Wolf, A. Labande, J.-C. Daran, R. Poli, Eur. J. Inorg. Chem.
2007, 5069–5079.
a) I. Özdemir, Y. Gök, N. Gürbüz, E. Çetinkaya, B. Çetinkaya,
Heteroat. Chem. 2004, 15, 419–423; b) I. Özdemir, Y. Gök, N.
Gürbüz, E. Çetinkaya, B. Çetinkaya, Synth. Commun. 2004,
34, 4135–4144; c) M. Dincer, N. Özdemir, S. Gülcemal, B. Çet-
inkaya, O. Buyukgungor, Acta Crystallogr., Sect. C: Cryst.
Struct. Commun. 2006, 62, m252-m254; d) A. G. Gökçe, S.
Gülcemal, M. Aygün, B. Çetinkaya, O. Büyükgüngör, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun. 2006, 62, m535-
m537; e) S. Demir, I. Özdemir, B. Çetinkaya, Appl. Organomet.
Chem. 2006, 20, 254–259; f) M. Yigit, B. Yigit, I. Özdemir, E.
Çetinkaya, B. Çetinkaya, Appl. Organomet. Chem. 2006, 20,
322–327; g) M. Dincer, N. Özdemir, S. Gülcemal, B. Çetinkaya,
Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 2007, 63,
m228-m230; h) B. Kosar, H. Türkmen, B. Çetinkaya, O.
Büyükgüngür, Acta Crystallogr., Sect. E: Struct. Rep. Online
2007, 63, m619-m620; i) H. Türkmen, S. DenizaltI, I. Özdemir,
E. Çetinkaya, B. Çetinkaya, J. Organomet. Chem. 2008, 693,
425–434; j) S. Yasar, I. Özdemir, B. Çetinkaya, J.-L. Renaud,
C. Bruneau, Eur. J. Org. Chem. 2008, 2142–2149; k) M. S.
Jeletic, I. Ghiviriga, K. A. Abboud, A. S. Veige, Organometal-
lics 2007, 26, 5267–5270.
[23] V. César, S. Bellemin-Laponnaz, H. Wadepohl, L. H. Gade,
Chem. Eur. J. 2005, 11, 2862–2873.
[24] H. Nishiyama, H. Sakaguchi, T. Nakamura, M. Horihata, M.
Kondo, K. Itoh, Organometallics 1989, 8, 846–848.
[25] a) H. Nishiyama, M. Kondo, T. Nakamura, K. Itoh, Organo-
metallics 1991, 10, 500–508; b) H. Nishiyama, S. B. Park, K.
Itoh, Tetrahedron: Asymmetry 1992, 3, 1029–1034; c) H. Nishi-
yama, S. Yamaguchi, M. Kondo, K. Itoh, J. Org. Chem. 1992,
57, 4306–4309; d) D. Cuervo, M. P. Gamasa, J. Gimeno, J. Mol.
Catal. A 2006, 249, 60–64.
[26] a) T. Hayashi, C. Hayashi, Y. Uozumi, Tetrahedron: Asymmetry
1995, 6, 2503–2506; b) H. Brunner, R. Störiko, Eur. J. Inorg.
Chem. 1998, 783–788.
[27] C. Reyes, A. Prock, W. P. Giering, Organometallics 2002, 21,
546–554.
[28] I. Kolb, J. Hetflejs, Collect. Czech. Chem. Commun. 1980, 45,
2808–2816.
[29] T. E. Waldman, G. Schaefer, D. P. Riley, ACS Symp. Ser. 1993,
517, 58–74.
[30] I. Ojima, M. Nihonyanagi, T. Kogure, M. Kumagai, S. Horiu-
chi, K. Nakatsugawa, J. Organomet. Chem. 1975, 94, 449–461.
[31] a) R. Corriu, J. J. E. Moreau, J. Organomet. Chem. 1975, 85,
19–33; b) I. Ojima, T. Kogure, M. Kumagai, S. Horiuchi, T.
Sato, J. Organomet. Chem. 1976, 122, 83–97.
[7]
[8]
[9]
I. Özdemir, N. Sahin, Y. Gök, S. Demir, B. Çetinkaya, J. Mol.
Catal. A 2005, 234, 181–185.
W.-L. Duan, M. Shi, G.-B. Rong, Chem. Commun. 2003, 2916–
2917.
a) W. A. Herrmann, J. Schutz, G. D. Frey, E. Herdtweck, Orga-
nometallics 2006, 25, 2437–2448; b) T. Chen, X.-G. Liu, M.
Shi, Tetrahedron 2007, 63, 4874–4880.
[32] T. Imamoto, T. Itoh, Y. Yamanoi, R. Narui, K. Yoshida, Tetra-
hedron: Asymmetry 2006, 17, 560–565.
[33] M. F. Lappert, R. K. Maskell, J. Organomet. Chem. 1984, 264,
217–228.
[10] R. J. Lowry, M. K. Veige, O. Clément, K. A. Abboud, I. Ghivi-
riga, A. S. Veige, Organometallics 2008, 27, 5184–5195.
[11] O. Buisine, G. Berthon-Gelloz, J.-F. Briere, S. Sterin, G. Mig-
nani, P. Branlard, B. Tinant, J.-P. Declercq, I. E. Marko, Chem.
Commun. 2005, 3856–3858.
[12] C. J. O’Brien, E. A. B. Kantchev, G. A. Chass, N. Hadei, A. C.
Hopkinson, M. G. Organ, D. H. Setiadi, T.-H. Tang, D.-C.
Fang, Tetrahedron 2005, 61, 9723–9735.
[13] F. E. Hahn, L. Wittenbecher, R. Boese, D. Bläser, Chem. Eur.
J. 1999, 5, 1931–1935.
[14] a) C. Bolm, M. Kesselgruber, G. Raabe, Organometallics 2002,
21, 707–710; b) Y. Yuan, G. Raabe, C. Bolm, J. Organomet.
Chem. 2005, 690, 5747–5752.
[15] a) B. Bildstein, M. Malaun, H. Kopacka, K.-H. Ongania, K.
Wurst, J. Organomet. Chem. 1998, 552, 45–61; b) B. Bildstein,
M. Malaun, H. Kopacka, K.-H. Ongania, K. Wurst, J. Or-
ganomet. Chem. 1999, 572, 177–187.
[16] 2b was prepared according to a slightly modified procedure
from ref.[7]: nBuOH was used instead of EtOH.
[17] F. Demirhan, Ö. Yildirim, B. Çetinkaya, Transition Met. Chem.
2003, 28, 558–562.
[34] J. E. Hill, T. A. Nile, J. Organomet. Chem. 1977, 137, 293–300.
[35] A. K. Roy, in Advances in Organometallic Chemistry, vol. 55
(Eds.: R. West, A. F. Hill, M. J. Fink), Academic Press, 2007,
pp. 1–59.
[36] a) I. Ojima, T. Kogure, M. Kumagai, J. Org. Chem. 1977, 42,
1671–1679; b) R. Kuwano, M. Sawamura, J. Shirai, M. Taka-
hashi, Y. Ito, Bull. Chem. Soc. Jpn. 2000, 73, 485–496; c) T.
Imamoto, T. Itoh, Y. Yamanoi, R. Narui, K. Yoshida, Tetrahe-
dron: Asymmetry 2006, 17, 560–565.
[37] a) T.-Y. Dong, L.-L. Lai, J. Organomet. Chem. 1996, 509, 131–
134; b) T.-Y. Dong, P.-H. Ho, C.-K. Chang, J. Chin. Chem.
Soc. 2000, 47, 421–424.
[38] R. Uson, L. A. Oro, J. A. Cabeza, Inorg. Synth. 1995, 23, 126–
130.
[39] D. Lednicer, C. R. Hauser, Org. Synth. 1960, 40, 31.
[40] A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R.
Spagn, J. Appl. Crystallogr. 1999, 32, 115–119.
[41] G. M. Sheldrick, SHELXL97, Göttingen, Germany, 1998.
[42] A. L. Spek, J. Appl. Crystallogr. 2003, 36, 7–13.
[43] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565.
Received: November 30, 2008
[18] C. Köcher, W. A. Herrmann, J. Organomet. Chem. 1997, 532,
261–265.
Published Online: February 19, 2009
Eur. J. Inorg. Chem. 2009, 1806–1815
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1815