Structural and mechanistic investigation of a cationic hydrogen-substituted ruthenium silylene catalyst for alkene hydrosilation

Article abstract of DOI:10.1039/c3sc51585k

The cationic ruthenium silylene complex [Cp*(ⁱPr ₃P)Ru(H)₂(SiHMes)][CB₁₁H₆Br ₆], a catalyst for olefin hydrosilations with primary silanes, was isolated and characterized by X-ray crystallography. Relatively strong interactions between the silylene Si atom and Ru-H hydride ligands appear to reflect a highly electrophilic silicon center. The mechanism of olefin hydrosilation was examined by kinetics measurements and other experiments to provide the first experimentally determined mechanism for the catalytic cycle. This mechanism involves a fast, initial addition of the Si-H bond of the silylene complex to the olefin. Subsequent elimination of the product silane produces an unsaturated intermediate, which can be reversibly trapped by olefin or intercepted by the silane substrate. The latter reaction pathway involves activation of the reactant silane by Si-H oxidative addition and α-hydrogen migration to regenerate the key silylene intermediate.

Full text of DOI:10.1039/c3sc51585k

Chemical Science
View Article Online
View Journal | View Issue
EDGE ARTICLE
Structural and mechanistic investigation of a cationic
hydrogen-substituted ruthenium silylene catalyst for
alkene hydrosilation†
Cite this: Chem. Sci., 2013, 4, 3882
*
Meg E. Fasulo, Mark C. Lipke and T. Don Tilley
The cationic ruthenium silylene complex [Cp*(iPr3P)Ru(H)2(SiHMes)][CB11H6Br6], a catalyst for olefin
hydrosilations with primary silanes, was isolated and characterized by X-ray crystallography. Relatively
strong interactions between the silylene Si atom and Ru–H hydride ligands appear to reflect a highly
electrophilic silicon center. The mechanism of olefin hydrosilation was examined by kinetics
measurements and other experiments to provide the first experimentally determined mechanism for the
catalytic cycle. This mechanism involves a fast, initial addition of the Si–H bond of the silylene complex
to the olefin. Subsequent elimination of the product silane produces an unsaturated intermediate,
which can be reversibly trapped by olefin or intercepted by the silane substrate. The latter reaction
pathway involves activation of the reactant silane by Si–H oxidative addition and a-hydrogen migration
to regenerate the key silylene intermediate.
Received 5th June 2013
Accepted 29th July 2013
DOI: 10.1039/c3sc51585k
www.rsc.org/chemicalscience
RuH2(SiHPh$Et2O)][B(C6F5)4] (1), implicated a new hydro-
silation mechanism involving direct addition of the silylene
Introduction
Si–H bond to an olen (Scheme 1).6 The silylene ligand appears
to be regenerated in the catalytic cycle by oxidative addition of
an Si–H bond, followed by 1,2-migration of a second hydrogen
from silicon to ruthenium. Additionally, the silylene function-
ality is necessary for catalysis, since the related silyl complexes
Hydrosilation, the addition of an Si–H bond across an unsatu-
rated bond (C]C, C]O, C]N, etc.), is an extremely important
chemical transformation in the laboratory and in large-scale
industrial applications.1 Commercial hydrosilations utilize
expensive platinum-based catalysts, which are highly effective
but somewhat limited with respect to selectivities and substrate
scope.2 The performance of these catalysts is generally under-
stood within the context of the Chalk–Harrod or modied
Chalk–Harrod mechanisms, which account for the observed
activity as well as the side reactions which lower selectivity.3
Because of the limitations associated with this catalysis, and
especially due to the rising cost of platinum, there is consider-
able interest in the discovery of new types of hydrosilation
catalysts. In this context, the discovery of active hydrosilation
catalysts based on more cost-effective transition metals is an
important research objective.4 Progress toward this goal should
be greatly facilitated by identication of new mechanisms for
catalytic hydrosilation.
Several recent investigations point to the potential utility of
transition metal silylene complexes as reactive intermediates in
the catalytic hydrosilation of olens.5–9 The initial discovery of
hydrosilation catalyzed by a silylene complex, [Cp*(iPr3P)-
Department of Chemistry, University of California-Berkeley, Berkeley, CA, 94720, USA.
E-mail: tdtilley@berkeley.edu
† Electronic supplementary information (ESI) available: Synthetic details,
characterization data, crystallographic data, and computational details. CCDC
942935. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c3sc51585k
Scheme 1
3882 | Chem. Sci., 2013, 4, 3882–3887
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
Cp*(iPr3P)RuH2(SiHPhOTf) and Cp*(iPr3P)Ru(H)Cl(SiH2Ph) are analogous complex [Cp*(iPr3P)RuH2(]SiHMes)][B(C6F5)4]
2
not active catalysts. This catalysis features a strict selectivity for (29Si ¼ 228.9 ppm, JSiH ¼ 58.2 Hz).11 Interestingly, the borate
primary silanes (to exclusively give secondary silane products), and carborane salts display different thermal stabilities – while
and no reactivity is observed with secondary and tertiary silanes. [Cp*(iPr3P)RuH2(]SiHMes)][B(C6F5)4] can be stored for 1 to 2
Additionally, these olen hydrosilations are highly selective in weeks at ꢀ35 ꢁC before signicant decomposition, 2 can be
ꢁ
exclusively producing anti-Markovnikov products. Typical stored for 2 months at ꢀ35 C without decomposition and is
byproducts of hydrosilation catalysis with platinum, such as stable in C6D5Br solution at room temperature for ca. one week.
vinyl silanes, isomerized olen, and silane redistribution
Single crystals suitable for X-ray diffraction were grown by
products, are not observed with 1 as catalyst. Studies with vapor diffusion of pentane into a solution of 2 in C6H5F over
analogous osmium silylene complexes indicate that the cationic 48 h at ꢀ35 ꢁC (Fig. 1). No interactions are observed between the
nature of the catalytic species is required for insertion of the carborane anion and the silicon center. The Ru–Si bond
olen into the Si–H bond.7
distance of 2.246(1) A is slightly longer than analogous
˚
Given the potential utility of new catalysis based on reactions distances in the neutral ruthenium silylene complexes
13
˚
A)
of a silylene ligand, it is important to obtain detailed mecha- Cp*(CO)(H)Ru]SiH[C(SiMe3)3]
(2.220(2)
and
i
14
˚
nistic information for hydrosilations catalyzed by 1 and related Cp*( Pr2MeP)(H)Ru]SiH(Trip) (2.205(1) A) but is similar to
complexes. Current understanding of the mechanism is based that found in the cationic complex [Cp*(Me3P)2Ru]SiMe2]-
only on studies of stoichiometric transformations, and by DFT [B(C6F5)4] (2.238(3) A). This observed bond lengthening may
15
˚
investigations on simplied model systems.8 Signicantly, be attributed to a greater contribution from a silicenium reso-
kinetic studies of the catalytic process have not been reported, nance form, with a concomitant reduction in the Ru–Si bond
and structural information on key catalytic species of the type order.7 The ruthenium hydride ligands were located in the
[Cp*(iPr3P)RuH2(]SiHR)]+ has not been available. In addition, Fourier difference map; however, the position of H(2) was not
kinetic studies on catalysis by 1 has been complicated by lack of stable under renement and was xed. The Ru–H(1) distance of
˚
a suitable solvent that is unreactive and nonvolatile over an 1.73(6) A is longer than analogous Ru–H distances observed in
i
appropriate temperature range. Here, we report the synthesis Cp*( Pr3P)Ru(H)2(SiHMesCl)16 (Ru–H(1) ¼ 1.50(9) A; Ru–H(2) ¼
˚
and structural characterization of [Cp*(iPr3P)RuH2(]SiHMes)]- 1.59(7) A) and Cp*( Pr3P)Ru(H)2(SiPh2OTf)11 (Ru–H(1) ¼ 1.41(4)
i
˚
˚
˚
˚
[CB11H6Br6] (2), a hydrosilation catalyst and model for the key A; Ru–H(2) ¼ 1.51(5) A), and the Si/H(1) distance of 1.74(6) A is
silylene intermediate. Most signicantly, kinetic studies with remarkably short, indicating a strong Ru–H/Si interaction in 2.
this catalyst provide the rst detailed description of the mech- For comparison, the neutral complex Cp*(iPr2MeP)(H)Ru]
anism of this silylene-mediated hydrosilation, and provide a SiH(Trip) displays a RuH/Si distance of 2.21(4) A.14 The silicon
˚
rm basis for future catalyst development.
atom is coplanar with P, Ru, and the Cp* centroid (sum of
angles ¼ 360.0(5)ꢁ) suggesting that the hydrides may be equiv-
alent and related by a molecular mirror plane.
The strong distortion of 2 away from an idealized four-legged
piano stool structure, with the hydride ligands signicantly dis-
placed toward silicon, is reminiscent of similar distortions found
Results and discussion
Synthesis and structural characterization of cationic silylene
complex [Cp*(iPr3P)RuH2(]SiHMes)]+ (2)
A primary objective of this investigation was to obtain detailed in piano-stool silyl hydride complexes of ruthenium and is
structural data for a [Cp*(iPr3P)RuH2(]SiHR)]+ silylene complex, usually attributed to donation of hydride electron density to the
since such species are thought to engage in the key Si–C bond- electrophilic silicon center.17 For Cp*(iPr3P)Ru(H)2(SiHMesCl),
forming event in hydrosilation catalysis. Interestingly, DFT
calculations on the simplied model systems [Cp(H3P)RuH2(]
SiHR)]+ (R]H,8a Ph8b) predict a distorted four-legged piano stool
structure, with the presence of two strong ruthenium hydride–
silicon interactions. Numerous initial efforts to obtain X-ray
quality crystals for structural studies focused on cationic silylene
complexes containing the [B(C6F5)4]ꢀ anion, and were not
successful. Thus, attention turned to use of other weakly coor-
dinating anions, and in this regard the carborane anion
[CB11H6Br6]ꢀ seemed particularly promising given its proven
utility in providing crystalline samples of iridium silylene
complexes and other cationic silicon species.10
Reaction of Cp*(iPr3P)RuH2(SiHMesOTf)11 with [Et3Si]-
[CB11H6Br6]12 in C6H5F followed by precipitation with pentane
provided [Cp*(iPr3P)RuH2(]SiHMes)][CB11H6Br6] (2) as an
Fig. 1 Molecular structure of 2 displaying thermal ellipsoids at the 50% prob-
orange solid. The 29Si NMR spectrum reveals a resonance at
ability level. Selected H-atoms, carborane anion, and fluorobenzene molecule
228.7 ppm, and in the 1H NMR spectrum the ruthenium hydride
˚
have been omitted for clarity. Selected bond lengths (A): Ru(1)–Si(1)
¼
¼
2
ligands at ꢀ11.35 ppm display a JSiH coupling constant of
2.2461(13), Ru(1)–P(1)
¼
2.3662(12), Ru(1)–H(1)
¼
1.73(6), Ru(1)–H(2)
62.3 Hz. These values are similar to those observed for the 1.5981(4), Si(1)–H(3) ¼ 1.32(6).
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 3882–3887 | 3883

View Article Online
Chemical Science
Edge Article
this distortion from a regular, piano-stool geometry results in P– C6D5Br solution containing 1 mol% loading of 2 and an internal
Ru–H angles (78, 79ꢁ) that are substantially wider than the Si–Ru– standard. The reaction temperature was then raised to 80 ꢁC, and
H angles (60, 64ꢁ). The stronger distortion for 2, reected in a reactions were allowed to proceed for 1 h. As observed for both
greater difference between the P–Ru–H (80, 85ꢁ) and Si–Ru–H (50, [Cp*(iPr3P)RuH2(SiHPh$Et2O)][B(C6F5)4]6 and [(PNP)(H)Ir]Si(H)-
54ꢁ) angles, appears to result from the more strongly electrophilic Mes][B(C6F5)4],22 the catalysis is regioselective and exclusively
nature of the three-coordinate, sp2-hybridized silicon center in produces the anti-Markovnikov, secondary silane product
this type of cationic structure.18 In addition, the small hydrogen (H2SiRR0), which is inactive to further hydrosilation.
substituent at the silicon center of 2 allows the complex to readily
Unlike 1, compound 2 catalyzes the redistribution of PhSiH3
adopt a conformation that maximizes interactions of the under these reaction conditions, and this results in lower yields
hydrides with the formally empty p orbital at silicon, with the for the main hydrosilation product. For example, the catalytic
silylene plane bisecting the H–Ru–H angle.19 Thus, the silicon– reaction mixture of PhSiH3 and 1-hexene aer 1 h at 80 ꢁC in
hydride interactions appear to be a consequence of the unusually C6D5Br contains PhHexSiH2 (60%), PhSiH3 (8%), HexSiH3 (19%),
electrophilic nature of the silicon center and 2, and in catalysis and Ph2SiH2 (10%). For comparison, the B(C6F5)4ꢀ analogue of 2,
must be readily displaced by nucleophilic attack of an incoming [Cp*(iPr3P)RuH2(]SiHMes)][B(C6F5)4],11 produces only the
olen substrate.
hydrosilation product PhHexSiH2 (98%) under the same reaction
The structure and bonding in 2 were examined by DFT conditions. The redistribution chemistry associated with 2 was
calculations to provide additional information on the nature of found to be dependent on the nature of the organosilane
the Ru–H–Si interactions. The solid-state structure of 2 was used substrate, as the alkylsilane substrate CySiH3 results in no
as a starting point for a DFT geometry optimization calculation redistribution products under the same conditions with 2 as
that indicated the presence of two equivalent, strong Si–H catalyst. In the absence of olen, 2 reacted with 10 equiv. of
interactions (Si–H distances 1.683, 1.676 A for 2-DFT).20 These Si– PhSiH3 at 80 ꢁC in C6D5Br over 1 h to give Ph3SiH (16%), Ph2SiH2
˚
H distances are quite short, and suggest that 2 could be described (27%), PhSiH3 (27%), and SiH4 (30%). Under analogous condi-
as having a structure featuring an h3-H2SiHMes ligand.18 This tions, no redistribution is observed for [Cp*(iPr3P)RuH2-
bonding description is supported by NBO and NMLO calcula- (]SiHMes)][B(C6F5)4] with 10 equiv. of PhSiH3. These interesting
tions on 2-DFT, which are consistent with the presence of two Si– results imply that the anion plays a crucial role in dening the
H bonds that donate strongly to the ruthenium center. Addi- catalytic chemistry. A detailed explanation for this difference is
tionally, these calculations do not indicate the presence of a currently lacking, but it would seem to reect the nature of
strong, direct s bond between the ruthenium and silicon. Thus, 2 cation–anion interactions in solution. In this context, note that
appears to feature considerable h3-H2SiHMes character and is redistribution at silicon has been shown to occur via bimolecular
therefore related to [PhBP3]RuH(h3-H2SiRR0) complexes that also reactions of neutral silyl and cationic silylene Cp*Ru complexes
exhibit substantial electrophilicity at silicon.18 However, the NMR in solution.23 Interestingly, a ten-fold decrease in the concentra-
data for 2 (29Si d 228.7 ppm, JSiH ¼ 62.3 Hz) indicate that this tion of 2 gave much lower conversion of 10 equiv. of PhSiH3 at 80
complex features a higher degree of silylene character and weaker ꢁC in C6D5Br over 1 h to the redistribution products Ph2SiH2 (5%)
Si–H interactions (for [PhBP3]RuH(h3-H2SiRR0); 29Si d z 150 and SiH4 (3%) (92% of the PhSiH3 remained).
ppm, 1JSiH z 100 Hz). These comparisons suggest that there is a
strong relationship between ruthenium(IV) silylene dihydride and
ruthenium(II) h3-silane structures, and perhaps a continuum
represented by the contributing resonance structures of Scheme
2. Complete cleavage of two Si–H bonds in the coordinated silane
results in the dihydride structure on the le, whereas incomplete
Si–H activation leads to the h3-silane complex on the right. This is
consistent with observed similarities between silylene dihydride
and h3-silane complexes,18 and is related to the continuum that
has been established for h2-HSiR3 and silyl hydride complexes.21
Mechanistic studies on hydrosilation with 2
Mechanistic studies were undertaken using CySiH3 and
1-octene as substrates, to avoid possible complications from
silane redistribution. Kinetic measurements were monitored by
loss of silane at 80 ꢁC in bromobenzene-d5, and the appearance
Table 1 Catalytic hydrosilation by 2a
Silane
Alkene
Yield
Catalytic hydrosilation with 2
PhSiH3
1-Octene
1-Hexene
Cyclopentene
Cyclohexene
t-Butylethylene
1-Octene
73%
60%
67%
79%
79%
>98%
97%
>98%
94%
>98%
Compound 2 was evaluated as a catalyst for the hydrosilation of a
variety of olens, with the primary silanes PhSiH3 and CySiH3
(Table 1). In a typical hydrosilation experiment, equimolar
amounts of the primary silane and the olen were added to a
CySiH3
1-Hexene
Cyclopentene
Cyclohexene
t-Butylethylene
a
Reactions were conducted with 1 mol% of 2 in 0.5 mL of C6D5Br at 80
ꢁC for 1 h. Yields were determined by integration of 1H NMR resonances
with respect to an internal standard (C6Me6).
Scheme 2
3884 | Chem. Sci., 2013, 4, 3882–3887
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
of product correlates with silane loss. First-order dependence of type of structural change has previously been observed and
the reaction rate on the concentration of 2 was established by attributed to steric inuence on rotation about the metal–
varying the loading of catalyst 2 from 1 to 4 mol% (Fig. S1†). silicon bond, and consequently the magnitude of the non-
Preliminary studies indicated that the reaction is not rst order classical Ru–H–Si interactions. Thus, the displacement of such
in olen and suggested an inverse dependence. To determine interactions in 2 could contribute to the observed inverse KIE
the reaction order in olen more precisely, pseudo-rst order since the transfer of deuteride from a bridging position to a
reaction conditions were established for various concentrations terminal position is more favorable than for hydride.24 The
of olen (3.2–7.5 M) with 0.16 M silane and 1 mol% catalyst 2. observed difference between the KIE for the overall reaction and
For these conditions, plots of [olen] vs. 1/kobs were linear, that for olen insertion indicates that the latter is not the RDS.
indicating an inverse-order dependence on olen (Fig. S2†). Additionally, the catalytic reaction of 1-octene and CySiD3 by 2
1
2
Interestingly, the concentration of olen inuences the order of in C6D5Br resulted in a single isotopomer by H and H NMR
the reaction with respect to CySiH3. At low concentrations of spectroscopy, consistent with a concerted reaction (eqn (1)). The
olen (0.16 M), the reaction rate is independent of the competition experiment of 10 equiv. of CySiH3, 10 equiv. of
concentration of CySiH3, and depends only on the concentra- CySiD3, 100 equiv. of 1-octene, and 1 equiv. of 2 in C6D5Br at
tion of 2. At higher concentrations of olen (3.2–7.5 M), the 80 ꢁC for 1 h gave a mixture of isotopomers, preventing the
reaction order in silane was determined to be rst order based determination of a KIE.
on linear plots of ln[silane] vs. time (Fig. S7†), measured over 4
half-lives. This information suggests that under typical catalytic
conditions (1 : 1 ratio of silane : olen), the rate determining
step does not involve silane or olen, but that at much higher
concentrations of olen, the silane must effectively compete
with the olen for binding to the catalyst. The complex inu-
ence of olen concentration on the reaction rate provides
unanticipated, new information about the mechanism of this
Qualitative observations provide further evidence that the
RDS occurs during the exchange of the secondary silane product
for the primary silane substrate to regenerate the H-substituted
silylene complex. The stoichiometric reaction of 2 with 1 equiv.
hydrosilation.
of 1-hexene in C6D5Br at room temperature proceeds instanta-
Further investigation of the catalytic mechanism involved a
neously with a color change from bright orange to bright yellow
series of isotopic labeling experiments. An experiment to
to give the secondary silylene complex [Cp*(iPr3P)RuH2-
determine the kinetic isotope effect (KIE) for catalysis involved
(]SiMes(Hex))][CB11H6Br6] (3) (eqn (2)). The 29Si NMR spec-
the competition reaction of 1 equiv. of 1-octene with 10 equiv. of
trum reveals a resonance at 264.7 ppm, and in the 1H NMR
CySiH3 and 10 equiv. of CySiD3, in the presence of 1 mol% of 2
spectrum the ruthenium hydride ligands at ꢀ11.62 ppm do not
ꢁ
in C6D5Br at 80 C over 1 h. The KIE was found to be kH/kD
¼
display a 2JSiH coupling constant, indicating little or no Ru–H/
Si interaction.11 Addition of 10 equiv. of CySiH3 to 3 in C6D5Br
resulted in no observable reaction aer 3 h at room tempera-
ture. However, heating this reaction mixture at 80 ꢁC for 5 min
resulted in complete conversion to [Cp*(iPr3P)RuH2(]SiHCy)]-
[CB11H6Br6] (eqn (2)). Efforts to observe an intermediate in the
reaction of 3 with 1–10 equiv. of CySiH3 by NMR spectroscopy
were unsuccessful. In order to probe the reversibility of product
release, a model reaction involving 20 equiv. of MesHexSiH2
and the silylene complex [Cp*(iPr3P)RuH2(]SiHMes)]-
[CB11H6Br6] (2) in C6D5Br solution was monitored by NMR
spectroscopy. Aer 1 h at room temperature, an equilibrium
mixture of 2 and 3 (10 : 1) was established and 0.1 equiv. of free
H3SiMes was observed in solution, indicating that product
formation is somewhat reversible (Keq z 2000).
1.5(1), which suggests that an Si–H bond is broken or formed
during the rate determining step (RDS). Based on the previously
suggested mechanism, there are three steps that involve
cleavage of an Si–H bond: initial Si–H activation of the silane at
the metal center, a-H migration from silicon to ruthenium to
produce the silylene ligand, and insertion of an olen into the
H-substituted silylene ligand. The elimination of the product
silane involves the formation of two Si–H bonds by processes
that are the reverse of the Si–H activation steps.
For the olen insertion step, the related complex [Cp*(iPr3P)-
OsH2(]SiH(Trip))][B(C6F5)4]7 was reported to exhibit an inverse
KIE of kH/kD ¼ 0.8(1), and this was attributed to a secondary
effect resulting from rehybridization of the silylene ligand
(Sisp2–H/D / Sisp3–H/D) upon coordination of the olen to
silicon. To determine the KIE associated with the ruthenium
system described here, a competition experiment between 2 and
[Cp*(iPr3P)RuD2(]SiDMes)][CB11H6Br6] (2-d3) was conducted
by combining 1 equiv. of 2 with 1 equiv. of 2-d3 in C6D5Br, fol-
lowed by 0.5 equiv. of 1-octene and the internal standard
Ph2Si(CH3)(CD3). Aer 5 min at room temperature, the distri-
bution of products reected an inverse KIE of kH/kD ¼ 0.77(4)
(by NMR spectroscopy), which may result from the secondary,
rehybridization effect mentioned above. However, note that this
insertion step results in a loss of the intramolecular Ru–H–Si
Activation parameters for the catalytic reaction were deter-
mined by kinetic measurements over the temperature range of
‡
ꢁ
60–90 C. An Eyring plot provides the values DH ¼ 32(2) kcal
moꢁlꢀ1, DS‡ ¼ 19(1) J Kꢀ1 molꢀ1, and DG‡ ¼ 25(2) kcal molꢀ1 (at
80 C; Fig. S3†). If the silane exchange process (CySiH3 activa-
tion and CyRSiH2 elimination) were to have appreciable
2
interactions, as indicated by the lack of an observable JSiH
coupling constant in the new silylene complex (vide infra). This
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 3882–3887 | 3885

View Article Online
Chemical Science
Edge Article
associative character, then a negative entropy of activation complex of [Cp*(iPr3P)Ru]+ stabilized by coordinated solvent.
might be expected. Thus, the large, positive value for DS‡ The substrate/product exchange may be somewhat reversible,
suggests that the rate-determining step involves dissociation of as indicated by the observed reaction of 2 with MesHexSiH2, but
the product silane (CyRSiH2) from the ruthenium center prior to the displacement of the secondary silane product by the
binding of either substrate molecule. This is consistent with the primary silane appears to be strongly favored.
experimentally determined rate law under relatively low
Notably, under high concentrations of olen, D can instead
substrate loadings, which shows that the rate depends only on by trapped by olen to give E. The formation of E represents a
the concentration of the catalyst. Notably, this aspect of the non-productive equilibrium that is reected in the inverse-order
mechanism differs from calculations of Beddie and Hall, on rate dependence on olen. Experiments to observe a complex of
reaction of the model system [Cp(H3P)RuH2(]SiHEt)]+ with type E involved addition of 1–20 equiv. of 1-octene to 2 in
SiH4.8a The latter study concluded that the transition state for C6D5Br, but NMR spectra of the resulting solutions at ꢀ30 to 25
this step involves elimination of EtSiH3 from the model ꢁC did not provide denitive evidence for the formation of an
complex [Cp(H3P)RuH(SiH3)(h2-HSiH2Et)]+, which already olen complex, and were limited by the freezing point of the
features a new substrate molecule added to ruthenium.
solvent (ꢀ30.8 ꢁC). For the product forming pathway, CySiH3
From the kinetic and mechanistic information described must compete with the olen for binding to ruthenium, and
above, the catalytic cycle of Scheme 3 is proposed. The rst step thus the reaction is rst order in CySiH3 when the concentration
involves a rapid, concerted addition of the Si–H bond of the H- of olen is high. Under these conditions, saturation behavior
substituted silylene complex A across the carbon–carbon double was observed for CySiH3, and a maximum rate was observed at
bond of the olen substrate. This step appears to be strongly concentrations of 3.5 M silane and greater (Fig. S8†).
promoted by silicenium character in A (see resonance structures
The rate expression for consumption of the reactant silane in
A and A0 in Scheme 3),7 and produces a disubstituted silylene the catalytic cycle in Scheme 3, derived from application of the
complex (B) with displacement of the intramolecular Ru–H/Si steady state approximation to species D, is given in eqn (3) (see
interactions. The lack of Ru–H/Si interactions in B (indicated ESI†). This rate law predicts that saturation should be observed
by the NMR data) is thought to result from the steric inuence with respect to [silane] when k5[silane] is much larger than the
of two non-hydrogen substituents at silicon, which induces other terms in the denominator. However, at higher concen-
rotation about the Ru–Si bond such that donation of electron trations of olen or product, the [silane] term in the numerator
density from the lateral Ru–H bonds into a 3p orbital of the is not canceled by this same term in the denominator, and the
silicon center is less favorable.11,19 Next, complex B undergoes reaction becomes rst order with respect to [silane], and inverse
reversible a-hydride migration from ruthenium to silicon to order in [olen] or [product]. Thus, this rate law is consistent
produce the unsaturated (16 electron) ruthenium silyl complex with the aforementioned experimental observations for this
C. Reductive elimination of the product Si–H bond generates an system, including the variable dependence of rate on [silane]
unsaturated (formally 14 eꢀ) complex D, to which CySiH3 and the inverse dependence on [olen]. Additionally, under
rapidly adds by oxidative addition and 1,2-hydride migration to normal catalytic conditions, a silylene complex consistent with
regenerate the silylene complex A. Complex D, which is impli- structure B was observed by NMR spectroscopy as the resting
cated by the kinetic data, was not observed but it may exist as a state of the catalytic cycle, as expected when [olen] is too small
to inuence the reaction rate (i.e. [E] is small). Under these
conditions, the rate of consumption of [silane] decreases at high
conversion (>85% conversion), which is consistent with an
inverse dependence on [product] at relatively high concentra-
tions of product (Fig. S4†).
À
Á
ꢀk5½silaneꢂ k3K2½Bꢂ þ k ½Eꢂ
d½silaneꢂ
dt
½productꢂ þ k4½alkeneꢂ þꢀk45½silaneꢂ
À
Á
¼
(3)
k
ꢀ3
Conclusions
In conclusion, use of the anion [CB11H6Br6]ꢀ has enabled a
structural analysis of the cationic silylene complex [Cp*(iPr3P)-
RuH2(SiHMes)][CB11H6Br6] (2), which exhibits very short Ru–H/
Si distances. These short contacts appear to reect the presence
of a strongly electrophilic silicon center, which is a key chemical
property associated with hydrosilation activity. In the catalytic
cycle, nucleophilic attack of the alkene onto the silicon center
directly precedes insertion of the alkene into the Si–H bond of a
silylene ligand.6–8 Additionally, complex 2 displays greater
thermal stability relative to that of the [B(C6F5)4]ꢀ analogue 1,
allowing for in-depth kinetic and mechanistic studies. Most
Scheme 3
3886 | Chem. Sci., 2013, 4, 3882–3887
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Edge Article
Chemical Science
¨
importantly, these investigations provide the rst experimental
W. A. Herrmann and F. E. Kuhn, J. Am. Chem. Soc., 2011,
basis for a detailed mechanism of the catalytic cycle associated
133, 1589–1596.
with silylene-mediated alkene hydrosilation. Interestingly, acti- 10 (a) E. Calimano and T. D. Tilley, Organometallics, 2010, 27,
vation of the olen is relatively rapid in this catalysis, and
the rate-determining process is associated with elimination of
the product silane. These studies provide a working model for the
design of less expensive, more efficient, and/or more selective
catalysts that operate via related mechanisms.
1680–1692; (b) C. A. Reed, Acc. Chem. Res., 1998, 31, 133–
139; (c) K.-C. Kim, C. A. Reed, D. W. Elliott, L. J. Mueller,
F. Tham, L. Lin and J. B. Lambert, Science, 2002, 297, 825–
827; (d) Z. Xie, R. Bau, A. Benesi and C. A. Reed,
Organometallics, 1995, 14, 3933–3941.
11 M. E. Fasulo, P. B. Glaser and T. D. Tilley, Organometallics,
2011, 30, 5524–5531.
12 C. A. Reed, Acc. Chem. Res., 2010, 43, 121–128.
13 M. Ochiai, H. Hashimoto and H. Tobita, Angew. Chem., Int.
Ed., 2007, 46, 8192–8194.
14 P. G. Hayes, R. Waterman, P. B. Glaser and T. D. Tilley,
Organometallics, 2009, 28, 5082–5089.
15 S. K. Grumbine, T. D. Tilley, F. P. Arnold and A. L. Rheingold,
J. Am. Chem. Soc., 1994, 116, 5495–5496.
Notes and references
1 (a) D. Troegel and J. Stohrer, Coord. Chem. Rev., 2011, 255,
1440–1459; (b) B. Marciniec, Silicon Chem., 2002, 1, 155; (c)
M. A. Brook, Silicon in Organic, Organometallic, and Polymer
Chemistry, Wiley, New York, 2000; (d) I. Ojima, Z. Li and
J. Zhu, The Chemistry of Organic Silicon Compounds, Wiley,
Avon, 1998, ch. 29.
16 B. K. Campion, R. H. Heyn and T. D. Tilley, J. Chem. Soc.,
Chem. Commun., 1992, 1201–1203.
2 (a) J. L. Speier, J. A. Webster and G. H. Barnes, J. Am. Chem.
Soc., 1957, 79, 974–979; (b) B. D. Karstedt, US Patent
3,775,452, 1973; (c) J. Stein, L. N. Lewis, Y. Gao and
R. A. Scott, J. Am. Chem. Soc., 1999, 121, 3693–3703.
3 (a) A. J. Chalk and J. F. Harrod, J. Am. Chem. Soc., 1965, 87,
16–21; (b) F. Seitz and M. S. Wrighton, Angew. Chem., Int.
Ed. Engl., 1988, 27, 289–291; (c) S. B. Duckett and
R. N. Perutz, Organometallics, 1992, 11, 90–98.
4 (a) J. Koller and R. G. Bergman, Organometallics, 2012, 31,
2530–2533; (b) J. Yang and T. D. Tilley, Angew. Chem., Int.
Ed., 2010, 49, 10186–10188; (c) S. C. Bart, E. Lobkovsky and
P. J. Chirik, J. Am. Chem. Soc., 2004, 126, 13794–13807; (d)
M. Brookhart and B. E. Grant, J. Am. Chem. Soc., 1993, 115,
2151–2156.
17 (a) M. A. Rankin, D. F. MacLean, G. Schatte, R. McDonald
and M. Stradiotto, J. Am. Chem. Soc., 2007, 129, 15855–
15864; (b) A. L. Osipov, S. M. Gerdov, L. G. Kuzmina,
J. A. K. Howard and G. I. Nikonov, Organometallics, 2005,
24, 587–602; (c) S. B. Duckett, L. G. Kuzmina and
G. I. Nikonov, Inorg. Chem. Commun., 2000, 3, 126–
128; (d) V. Rodriguez, B. Donnadieu, S. Sabo-Etienne
and B. Chaudret, Organometallics, 1998, 17, 3809–
3814.
18 M. C. Lipke and T. D. Tilley, J. Am. Chem. Soc., 2011, 133,
16374–16377.
19 A. Shinohara, J. McBee and T. D. Tilley, Inorg. Chem., 2009,
48, 8081–8083.
5 R. Waterman, P. G. Hayes and T. D. Tilley, Acc. Chem. Res.,
2007, 40, 712–719.
6 P. B. Glaser and T. D. Tilley, J. Am. Chem. Soc., 2003, 125,
13640–13641.
20 DFT calculations were performed with Gaussian 09 using the
B3PW91 functional and LANL2DZ/6-31G** basis sets for Ru/
main group elements, respectively.
21 (a) U. Schubert, Adv. Organomet. Chem., 1990, 30, 151–187;
(b) W. Scherer, P. Meixner, J. E. Barquera-Lozada, C. Hauf,
7 P. G. Hayes, C. Beddie, M. B. Hall, R. Waterman and
T. D. Tilley, J. Am. Chem. Soc., 2006, 128, 428–429.
8 (a) C. Beddie and M. B. Hall, J. Am. Chem. Soc., 2004, 126,
13564–13565; (b) U. Bohme, J. Organomet. Chem., 2006,
691, 4400–4410.
¨
A. Obenhuber, A. Bruck, D. J. Wolstenholme, K. Ruhland,
D. Leusser and D. Stalke, Angew. Chem., Int. Ed., 2013, 52,
6092–6096.
22 E. Calimano and T. D. Tilley, J. Am. Chem. Soc., 2008, 130,
9226–9227.
23 S. R. Klei, T. D. Tilley and R. G. Bergman, Organometallics,
2002, 21, 3376–3387.
24 R. B. Calvert and J. R. Shapley, J. Am. Chem. Soc., 1978, 100,
7726–7727.
9 Transition metal silylene complexes have also been
proposed as intermediates in the catalytic hydrosilation
of carbonyl compounds. See: (a) N. Schneider, M.
Finger, C. Haferkemper, S. Bellemin-Laponnaz,
P. Hofmann and L. H. Gade, Angew. Chem., Int. Ed.,
2009, 48, 1609–1613; (b) P. Gigler, B. Bechlars,
This journal is ª The Royal Society of Chemistry 2013
Chem. Sci., 2013, 4, 3882–3887 | 3887