Reversible insertion of carbenes into ruthenium-silicon bonds

Article abstract of DOI:10.1021/ja410822p

The five-coordinate carbene complexes [Ru{κP,P,Si-Si(Me)(C ₆H₄-2-PiPr₂)₂}Cl(=CHR)] (2, R = Ph; 3, R = SiMe₃), analogues of the Grubbs catalyst, were prepared from the dimer [Ru(μ-Cl){κP,P,Si-Si(Me)(C₆H₄-2- PiPr₂)₂}]₂ (1) and the corresponding diazoalkane N₂CHR. The particular structural features that result from the presence of a strongly trans directing silyl group at the pincer ligand of these complexes are discussed on the basis of NMR information and the crystal structure of the vinylidene analogue [Ru{κP,P,Si-Si(Me)(C ₆H₄-2-PiPr₂)₂}Cl(=C=CHPh)] (4), which was also obtained from 1 and phenylacetylene. The reactions of 3 with reagents such as P(OMe)₃, CO, NCMe, and K(acac) illustrate that the first response of these carbene complexes to an increase of the coordination number around ruthenium is the insertion of the carbene ligand into the Ru-Si bond. These reactions also indicate that the insertion process is reversible and allows typical transformations of carbene ligands such as C-H functionalizations via carbene insertion (in the acac ligand) or the formation of ketene from CO. In addition, the reactions of 3 with terminal alkynes such as phenylacetylene or 3,3-dimethyl-1-butyne show that the inserted carbenes can also undergo reactions typical of metal-bound alkyls such as alkyne insertion and C-H reductive elimination.

Full text of DOI:10.1021/ja410822p

Article
pubs.acs.org/JACS
Reversible Insertion of Carbenes into Ruthenium−Silicon Bonds
́
María J. Bernal, Olga Torres, Marta Martın,* and Eduardo Sola*
́
́
́ ́
Instituto de Sıntesis Quımica y Catalisis Homogenea (ISQCH), CSIC-Universidad de Zaragoza, E-50009 Zaragoza, Spain
S
* Supporting Information
ABSTRACT: The five-coordinate carbene complexes [Ru{κP,P,Si-Si(Me)(C6H4-
2-PiPr2)2}Cl(CHR)] (2, R = Ph; 3, R = SiMe3), analogues of the Grubbs
catalyst, were prepared from the dimer [Ru(μ-Cl){κP,P,Si-Si(Me)(C6H4-2-
PiPr2)2}]2 (1) and the corresponding diazoalkane N2CHR. The particular
structural features that result from the presence of a strongly trans directing silyl
group at the pincer ligand of these complexes are discussed on the basis of NMR
information and the crystal structure of the vinylidene analogue [Ru{κP,P,Si-
Si(Me)(C6H4-2-PiPr2)2}Cl(CCHPh)] (4), which was also obtained from 1
and phenylacetylene. The reactions of 3 with reagents such as P(OMe)3, CO, NCMe, and K(acac) illustrate that the first
response of these carbene complexes to an increase of the coordination number around ruthenium is the insertion of the carbene
ligand into the Ru−Si bond. These reactions also indicate that the insertion process is reversible and allows typical
transformations of carbene ligands such as C−H functionalizations via carbene insertion (in the acac ligand) or the formation of
ketene from CO. In addition, the reactions of 3 with terminal alkynes such as phenylacetylene or 3,3-dimethyl-1-butyne show
that the inserted carbenes can also undergo reactions typical of metal-bound alkyls such as alkyne insertion and C−H reductive
elimination.
dominated by the title reaction and do not work as olefin
metathesis catalysts.
INTRODUCTION
■
As our mechanistic understanding of coordination and
organometallic chemistry increases, we are able to recognize
more reactions in which the ligands play a role beyond the
mere stereoelectronic influence. Thus, it is increasingly
common to find in the chemical literature labels such as
“noninnocent”,1 “cooperating”,2 or bifunctional,3 to highlight
certain properties of some ligands that render them particularly
suitable or even essential for transformations occurring at the
coordination sphere of metal complexes. Sometimes, the origin
of such ligand contributions is an electronic reorganization: a
valence tautomerism or resonance.1,4 In other cases, it is a
reversible bond cleavage or formation that happens in the
presence of reagents.2,3,5,6 Several biochemically relevant
systems exploit examples of the first type,7 while the latter is
at the heart of highly efficient catalytic cycles8 and valuable
stoichiometric reaction sequences.9 This work features a
recourse of the latter type for pincer ligands that contain a
silyl donor group (κP,P,Si), which consists of a reversible
insertion of carbenes into the metal−silicon bond.
Our interest in these types of pincers was originally due to
their capability of controlling the coordination geometry of
unsaturated metal complexes via the large trans influence of the
silyl group.10,11 We intended to exploit this property to shape
analogues of the Grubbs catalyst with the vacant coordination
site cis to the carbene ligand.12 In our hypothesis, such five-
coordinate ruthenium complexes could work as olefin meta-
thesis catalysts without the need for preactivating via ligand
dissociation,13 thus potentially leading to more robust,
compatible, and reusable catalysts. The following pages show
that even though the target nonisostructural analogues of the
Grubbs catalyst are readily accessible, their behavior is
RESULTS AND DISCUSSION
■
Grubbs Catalyst Analogues Based on a κP,P,Si Pincer.
The dimer [Ru(μ-Cl){κP,P,Si-Si(Me)(C6H4-2-PiPr2)2}]2 (1)
turned out to be a suitable precursor for the preparation of
carbene complexes. It was obtained from the reaction of [Ru(μ-
Cl)2(cod)]n with the silane HSi(Me)(C6H4-2-PiPr2)2 in the
presence of triethylamine. The structure of 1 determined by X-
ray crystallography, virtually C2 symmetric (Figure 1), is similar
to that reported by Tobish, Turculet, et al. for the analogous
complex with cyclohexyl instead of isopropyl groups.14 Also
similarly to this precedent, the NMR spectra of 1 in solution at
room temperature indicate an average higher symmetry, by
showing for example a single broad resonance in the 31P{1H}
spectrum corresponding to the four phosphorus atoms.
Compound 1 was found to react with diazoalkanes such as
phenyldiazomethane and trimethylsilyldiazomethane, releasing
N2 and forming the carbene complexes [Ru{κP,P,Si-Si(Me)-
(C6H4-2-PiPr2)2}Cl(CHR)] (2, R = Ph; 3, R = SiMe3),
respectively (Scheme 1). The commercial ethyl diazoacetate,
however, did not react cleanly under the same conditions, thus
showing that this synthetic method is not applicable to any
diazoalkane reagent. Actually, only the formation of 3 seems to
be clean and selective, as the reaction leading to compound 2
always gave rise to variable amounts of still unidentified
byproducts.
Received: October 22, 2013
Published: December 1, 2013
© 2013 American Chemical Society
19008
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
phenylacetylene at room temperature (Scheme 2). The
synthesis required several equivalents of the alkyne reagent,
Scheme 2
as 4 was formed together with the product of phenylacetylene
dimerization (Z)-1,4-diphenyl-1-buten-3-yne. Similar selective
dimerization processes have already been observed for the
parent vinylidene complexes [RuCl2(CCHPh)(PR3)2] (R
= iPr, Cy) as catalysts.15
Figure 1. Crystal structure of complex 1 at the 50% probability level.
The hydrogen atoms are omitted for clarity. For bond distances and
angles see the Supporting Information.
Scheme 1
Figure 2 details the angles in the equatorial plane of the five-
coordinate complex 4 to stress the differences with those found
in the aforementioned vinylidene analogues of the Grubbs
catalyst [RuCl2(CCHPh)(PR3)2]. In the latter, the vacant
coordination site clearly sits trans to the strongly σ-donor
vinylidene ligand, with angles between the chlorides of 161° (R
= iPr),16 and 158° (R = Cy).16,17 In contrast, the silyl function
of the κP,P,Si pincer of 4 seems capable of opening up the trans
angle to 144° even in the presence of the vinylidene, whose
trans angle closes to 131°. In favor of attributing these
differences mainly to a competition among the trans influences
of the equatorial ligands, the related derivative [RuCl(NO)(
CCHPh)(PPh3)2] also shows the widest angle (139°) trans
to the trans-directing bent nitrosyl ligand,18 while that trans to
vinylidene is just 113°.19
Complexes 2 and 3 show NMR spectra consistent with
equivalent and mutually trans PiPr2 groups, indicating a change
of the coordination mode of the κP,P,Si ligand from fac (in 1)
to mer. Accordingly, the carbene ligand CHs give rise to
1
characteristic triplet signals in both the H and 13C{1H} NMR
Among the various possible isomers for the products of the
above reactions, those represented in Schemes 1 and 2 are the
major isomers (2 and 3) or the only species detectable by
NMR (4). In this isomer, the carbene or vinylidene ligand
occupies the (less hindered) coordination position syn to the
methyl at silicon, while the substituent at the carbene or
vinylidene avoids the vicinity of this methyl. This is inferred, for
spectra: at δ 12.51 (JHP = 6.2) and 264.69 (JCP = 3.1),
respectively, for 2 and at δ 12.44 (JHP = 1.8) and 274.95 (JCP
4.9) for 3.
=
Although, so far, we have not been able to grow good crystals
for any of these carbene complexes, their likely structural
features can be discussed on the basis of the X-ray diffraction
structure of the analogous vinylidene compound [Ru{κP,P,Si-
Si(Me)(C6H4-2-PiPr2)2}Cl(CCHPh)] (4; Figure 2). This
derivative was also prepared from precursor 1 by reaction with
1
all complexes, from the observation of NOE between the H
NMR signals corresponding to the SiMe and the carbene or
vinylidene hydrogen and is confirmed by the crystal structure of
4. Such a product selectivity suggests that the κP,P,Si pincer
provides by itself a congested coordination environment.
Furthermore, other NMR parameters suggest that the pincer
can adapt its coordination angles to accommodate bulky
equatorial ligands. This is illustrated in Figure 3 by means of
two of the 13C{1H} NMR signals that are sensitive to the
magnitude of the JPP′ coupling constant, because they
correspond to carbons of the pincer backbone for which the
two chemically equivalent 31P nuclei are magnetically
inequivalent. The signals of complex 4, whose crystal structure
indicates a P−Ru−P angle of 165°, are virtual triplets
characteristic of strongly coupled phosphorus. This is also the
case for the phenyl-substituted carbene 2 but not for the
trimethylsilyl-substituted derivative 3, whose signals show
features halfway between the virtual triplet and those typical
of weakly coupled phosphorus. The latter are exemplified in the
figure by the signals of complex 1, the structure of which shows
P−Ru−P angles of 95 and 105° and whose 303 K NMR
spectrum indicates an averaged JPP′ coupling constant of about
Figure 2. (left) Crystal structure of 4 (50% probability level) and
(right) structural details of the equatorial plane of the complex. The
hydrogen atoms are omitted for clarity. For other bond distances and
angles see the Supporting Information.
19009
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
Scheme 3
Figure 3. Examples of 13C{1H} NMR signals corresponding to A
nuclei in AXX′ spin systems (X = 31P) for complexes 1, 3, and 4.
20 Hz. Even though the signal to noise ratio of the 13C{1H}
NMR spectrum of 3 does not allow observation of those side
lines of the AXX′ spin system that make possible the evaluation
of the JPP′ constant, the shape of these signals can be considered
diagnostic of a distortion of the mer pincer toward a fac
coordination.
Carbene Insertion. The possible catalytic activity of the
unsaturated carbene complexes 2 and 3 in olefin metathesis was
investigated through reactions and protocols previously
proposed in the literature as a standard system of character-
ization.20 None of these experiments gave positive results. The
only noticeable observation was the slow isomerization of
substrates such as 1,5-cyclooctadiene, diethyldiallyl malonate,
allylbenzene, and cis-1,4-diacetoxy-2-butene, which was not
further investigated. The in situ NMR examination of reactions
with excess ethylene also did not produce noticeable results.
However, in contrast to the inertness suggested by these initial
experiments, many reagents other than olefins were observed to
react with these five-coordinate complexes. The three selected
reactions of complex 3 depicted in Scheme 3 support the
description and discussion of the elementary process that was
repeatedly observed upon ligand or reagent coordination: the
insertion of the carbene ligand into the Ru−Si bond.
Figure 4. 31P{1H} NMR spectra of the reaction of 3 with 1 equiv of
P(OMe)3 in acetone-d6 at various temperatures.
After addition of 1 equiv of trimethyl phosphite, the solutions
1
of 3 in acetone-d6 showed broadened room-temperature H
and 31P{1H} NMR spectra (Figure 4). The spectra at low
temperature revealed the formation of two new products which,
according to their sets of JPP coupling constants (major
product, 416.9, 78.3, and 17.3 Hz; minor product, 236.4, 42.3,
and 26.3 Hz), show mer arrangements of three inequivalent
phosphorus atoms. In view of the typical values for 31P
chemical shifts, the major product should have two mutually cis
PiPr2 groups, while in the minor product these groups seem to
be trans. The major product could be identified from its NMR
signals as complex 5 (Scheme 3), although due to the weakness
significant change in the NMR spectra upon reaction is the
pronounced shift of the 29Si{1H} signal corresponding to the
silicon of the pincer (Figure 5), which moves from δ 50.72 (t,
JSiP = 11.7 Hz) to δ −9.76 (dd, JSiP = 21.1 and 6.8 Hz). A
common explanation for these two NMR features would imply
an insertion of the carbene ligand into the Ru−Si bond. Such a
process generates a chiral center and therefore inequivalent
pincer phosphorus atoms, whatever the ligand arrangement.
Taking this into account and in view of the 29Si{1H} NMR
signals observed for the minor product of this reaction (Figure
5), this second complex should also contain a κC,P,P pincer
generated via carbene insertion, but coordinated in a mer
fashion. Furthermore, the VT evolution of the NMR spectra
(31P in Figure 4) suggests that the spectral broadening can be
attributed to the exchange between both isomers of 5 in
equilibrium.
1
of the H NMR NOE effects at the low temperatures required
to observe narrow spectra, some details of its stereochemistry
remain unknown. Notably, its 13C{1H} NMR spectrum at 193
K does not show any signal attributable to a carbene ligand.
Instead, an apparent quartet at very high field (δ −42.60, CH)
with JCP coupling constants of about 8 Hz can be attributed to a
new Ru−alkyl moiety cis to three phosphorus atoms. A second
19010
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
reaction of 7 with trimethylsilyldiazomethane was investigated
because it could enable a catalytic synthesis of the ketene.22
However, the only new products detected by NMR in this
reaction were those of carbene dimerization.
The observed evolution of 6 strongly suggests that the
process of carbene insertion into the Ru−Si bond is reversible.
A much clearer evidence of that was obtained from the last
reaction of Scheme 3, in which the acetonitrile provoked not
only the carbene insertion but also the displacement of the
chloride outside the coordination sphere of the metal. The
single product of this reaction, the cationic complex 8, showed
broadened NMR spectra above 240 K due to facile acetonitrile
1
dissociation. At low temperature, the H, 13C{1H}, and 29Si-
{1H} (Figure 5) NMR spectra were fully consistent with the
proposed structure. It is worth noting that when the solutions
of 8 were dried in vacuo and the residue was redissolved, the
NMR spectra indicated the quantitative regeneration of the
carbene precursor 3. In fact, this represented a difficulty in
isolating 8 as a pure solid, which was overcome by removing
the chloride from the solution with silver triflate, thus isolating
8 as its triflate salt.
Although insertion reactions of metal-bound carbenes are
relatively common,23,24 the reactions of Scheme 3 represent, to
the best of our knowledge, the first examples of such an
insertion into a metal−silicon bond. Furthermore, reversible
migratory insertions of carbenes such as these are rare.25,26
Perhaps the closest precedent for this reaction can be found in
the treatment of the iridium(I) complex [Ir{κN,P,P-N(C6H3-4-
Me-2-PiPr2)2}(CHOtBu)] with excess CO, which causes the
migratory insertion of the alkoxycarbene into the Ir−N bond of
the amidophosphine pincer.26 Such a process was also
proposed to be reversible on the basis of the product thermal
evolution, which resembles that of 6. On the other hand, the
involvement of the silicon atom of κP,P,Si pincers in processes
of bond cleavage and formation has been frequently
observed,27−29 a fact suggesting that such pincers are less
inert than their κC,P,P or κN,P,P counterparts.30 In particular,
oxo groups and hydroxo ligands have been reported to insert
respectively into Ru−Si29 and Ir−Si28 bonds of complexes with
κP,P,Si pincers. Moreover, the reversible formation of a Si−C
bond by reductive elimination might be involved in the unusual
hydrocarboxylation of allenes with CO2 catalyzed by a
palladium κP,P,Si complex.31
Figure 5. 29Si{1H} NMR spectra of 3 and the products of its reactions
with P(OMe)3 (5), excess CO (6) and excess NCMe (8). JSiP coupling
constants are shown in blue. Asterisks denote signals and constants
corresponding to the minor products.
The second example in Scheme 3, the reaction of 3 with CO,
was monitored by NMR in CD2Cl2 solution. After bubbling of
CO and subsequent removal of the excess CO with bubbling
argon, a mixture of two new complexes was again observed.
The NMR spectra of both products display the aforementioned
features diagnostic of the carbene insertion: inequivalent
phosphorus atoms, negative chemical shifts for the 29Si{1H}
NMR signals of the pincer, and high-field CH multiplets in the
13C{1H} NMR spectrum (apparent triplets in this case). The
major product (ca. 90%), complex 6 in Scheme 3, shows two
CO 13C{1H} NMR signals indicative of ligands trans to
phosphorus: doublets of doublets at δ 199.02 and 198.09 with
JCP coupling constants of 77.2, 17.2 Hz and 74.7, 20.2 Hz,
respectively. Also in view of its CO 13C{1H} NMR signals, the
minor product of this reaction should be an isomer of 6 with
the chloride trans to phosphorus, although the spectra do not
allow us to distinguish between the two possibilities.
X-ray diffraction confirmation of the carbene migratory
insertion and further evidence of its reversibility were obtained
from the reaction sequence of Scheme 4: initiated by a
Complex 6 was found to slowly transform (days at room
temperature) into the five-coordinate complex [Ru{κP,P,Si-
Si(Me)(C6H4-2-PiPr2)2}Cl (CO)] (7) and trimethylsilylketene.
The latter was identified in the reaction mixture by means of its
reported 1H, 13C{1H}, and 29Si{1H} NMR signals.21 This
reactive compound was formed just as a transient species,
disappearing to give products that could not be identified.
Complex 7 was formed as a single isomer displaying a syn
relative orientation of the CO ligand and the methyl at silicon.
This compound could also be prepared by alternative methods
such as the reaction of precursor [RuHCl(CO)(PPh3)3] with
the silane HSi(Me)(C6H4-2-PiPr2)2 or, more simply, by
addition of a drop of ethanol during the synthesis of 1. The
crystal structure of 7 can be found in the Supporting
Information. Its most noticeable feature is an angle trans to
silicon of 178°, which supports the structural arguments based
on trans influences given in the previous section. The possible
Scheme 4
substitution of the chloride ligand by acetylacetonate (acac). As
in the previous examples, this reaction was expected to increase
the coordination number of the complex, thus triggering
carbene migration.
The structure of the expected reaction product, complex 9, is
shown in Figure 6. It can be described as square pyramidal, with
the vacant site trans to one of the phosphorus atoms of the new
19011
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
of the 13C{1H} NMR signals of the pincer backbone that are
sensitive to the magnitude of the JPP coupling constant (see
Figure 3 and accompanying discussion), which are also similar
to those of 1, thereby indicating a fac coordination mode of the
pincer.
Together, Schemes 1 and 4 compose a reaction sequence for
acac functionalization via insertion of the carbene into a C−H
bond.32 These types of functionalizations are known to be
catalyzed by a variety of transition-metal complexes, especially
of rhodium and coinage metals.33 Ruthenium carbenes,
however, are thought to be too unreactive for such
functionalizations, as only a few exceptional examples of
intramolecular reactions have been reported.34 In the present
case, carbene insertion into the Ru−Si bond should certainly
contribute to accommodate the anionic substrate in the
coordination sphere of ruthenium and might be behind the
success of the process.
Further Transformations of the Inserted Carbene.
None of the aforementioned complexes featuring inserted
carbenes was found to be stable in solution for long periods,
although, with the exception of 6 and 9, their evolutions turned
out to be too unselective and could not be understood. Yet, the
acac functionalization leading to 10 (Scheme 4) and the release
of ketene from 6 (Scheme 3) demonstrate that the “masked
carbenes” of these complexes remain useful for new bond
formations via migratory deinsertion from the silyl-substituted
metal alkyls. Nevertheless, carbene deinsertion is not the only
option for these alkyls, as inferred from the two reactions
between 3 and terminal alkynes shown in Scheme 5.
Figure 6. Crystal structure of complex 9 at the 50% probability level.
The hydrogen atoms are omitted for clarity. For bond distances and
angles see the Supporting Information.
fac κC,P,P pincer generated after carbene insertion (P−Ru−P
angle 104°). Although from just a consideration of trans
influences this is not the most likely ligand arrangement, it
seems to be the one sterically favored, as it gives more space to
accommodate the bulky SiMe3 group. In any case, the trans
influences of the alkyl and phosphorus arms of the pincer are
not that different, as inferred from the two Ru−O distances of
the acac ligand: 2.1385(13) Å trans to C and 2.0960(13) Å
trans to P. In line with these considerations, it should be
mentioned that the room-temperature NMR spectra of 9 are
Scheme 5
1
inconsistent with the solid-state structure, since both the H
and 13C{1H} NMR signals corresponding to the acac indicate a
symmetry element that relates the halves of the ligand. The
spectra at low temperature, however, show the decoalescence of
each NMR signal into two signals with the approximate relative
integration 3:1 (in toluene-d8). Then, the structure of 9 shown
in Figure 6 should be in equilibrium with a second coordination
isomer of similar energy, whose fast exchange on the NMR
time scale with 9 averages the acac signals but not those of the
κC,P,P pincer.
Complex 9 was observed to evolve in solution into a new
product, 10. The reaction took several days at room
temperature but just a few hours in toluene at 363 K. The
NMR spectra of 10 indicate a symmetric structure with
equivalent 31P nuclei: a feature by itself demonstrative of the
reversal of the carbene insertion. Consistently, the 29Si{1H}
NMR spectrum of the compound recovers the low-field signal
characteristic of the pincer’s κP,P,Si coordination: in this case a
triplet at δ 69.37 with a JSiP coupling constant of 19.0 Hz. The
NMR spectra also confirm that the SiMe3 group remains at the
complex, although the 13C{1H} NMR spectrum does not show
any CH signal attributable to a Ru-bound alkyl or carbene
ligand. Instead, the disappearance of the acac CH and the
presence of a new methylene group is evident from the 1H and
13C{1H} NMR spectra and their correlation experiments. All of
these NMR features point to the structure of 10 shown in
Scheme 4. This proposal is further supported by the similarity
between the NMR spectra of 10 and those recorded for the
closely related complex [Ru{κP,P,Si-Si(Me)(C6H4-2-PiPr2)2}-
(acac)] (11), which was prepared for this purpose from 1 and
K(acac). The similarities between 10 and 11 include the shape
The reaction of 3 with phenylacetylene at room temperature
was complete in a few minutes, giving complex 12 together
with traces of the alkyne dimerization product (Z)-1,4-
diphenyl-1-buten-3-yne. The 1H, 13C{1H}, and 29Si{1H}
NMR spectra of 12 display many similarities with those of
the carbene insertion products described above, although the
characteristic 13C{1H} NMR high-field signal corresponding to
the Ru-bound carbon is now a CH2 instead of a CH (a doublet
at δ −19.59 with a JCP coupling constant of 6.8 Hz). This signal
correlates in the 1H,13C hsqc NMR spectrum with those of two
inequivalent hydrogens: a doublet at δ −0.02 showing a JHH
coupling constant of 9.0 Hz and a multiplet at δ −0.34 which
additionally shows two JHP coupling constants of 11.5 and 2.7
Hz. In addition to the signals from carbons of the former
κP,P,Si ligand skeleton and a phenyl ring (broad), the 13C{1H}
19012
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
NMR spectrum shows two resonances corresponding to
quaternary carbons at δ 185.13 (dd, JCP = 17.6 and 6.7 Hz)
and 151.46 (dd, both JCP = 1.8 Hz). These latter signals,
together with that at high field, suggest the presence of a η3-allyl
ligand, as shown in the proposed structure of 12 (Scheme 5). In
closely related to 13 has been discussed in depth.40 The bond
distances within the ruthenacyclopropene ring are consistent
with the expected presence of a Ru−C double bond,38 Ru−
C(31) = 1.832(2) Å, Ru−C(30) = 2.183(2) Å, and C(30)−
C(31) = 1.431(3) Å, and so are the 13C{1H} NMR signals
corresponding to the alkenyl carbons, an apparent triplet at δ
330.45 (JCP = 5.9 Hz) and a singlet at δ 53.81.
1
favor of this proposal, the H NMR nOesy spectrum indicates
1
the proximity of the SiMe and SiMe3 groups, while the H,13C
NMR hmbc correlation reveals long-range couplings between
both of these groups and the CH2. Furthermore, other potential
ways of combining a carbene and an alkyne in the coordination
sphere of the metal35 can be discarded, because they would lead
to either highly unsaturated complexes or compounds
containing both intact κP,P,Si pincers and ligands with NMR-
evident carbenoid quaternary carbons. Moreover, the most
plausible pathway toward 12 from 3 and phenylacetylene is
compatible with the literature results and the ease of the
reaction. Assuming that, as shown in the previous pages, the
alkyne coordination to ruthenium causes carbene insertion into
the Ru−Si bond, the formation of 12 would consist of a further
insertion of the alkyne in the Ru−alkyl moiety36 followed by a
1,3-hydrogen shift, transforming the resulting alkenyl ligand
into an allyl.37
The second example of Scheme 5, the reaction between 3
and 3,3-dimethyl-1-butyne to form complex 13, is much more
elaborate than the previous example and, consistently, was
observed to be much slower, taking several hours in
dichloromethane solution at room temperature. Our monitor-
ing of the transformation by NMR indicated that no
intermediate accumulates up to concentrations allowing
identification. The structure of 13 (Figure 7) meets several
The structure of 13 again suggests a reaction pathway
initiated by alkyne coordination and carbene insertion into the
Ru−Si bond, although the presence of the CH2Si(Me)3
substituent at the silicon of the former pincer seems to indicate
that in this example the C−H activation of the alkyne is
preferred over its insertion into the Ru−C bond. The formation
of 13 also entails the very unusual insertion of a putative alkynyl
ligand into a Si−C bond, a process that is unlikely to occur
directly even with the help of an unsaturated metal center.
Instead, the process is likely to involve Si−C bond cleavage and
formation via oxidative additions and reductive eliminations,
respectively, as such steps have been found to be facile in
κP,P,Si pincer complexes of Ni, Pd, Rh, and Ir.27,28,31 In any
case, regardless of the details of each structure, the reactions of
Scheme 5 show that the inserted carbenes can indeed
participate in reactions typical of alkyls such as migratory
insertions and reductive eliminations.
CONCLUSIONS
■
The κP,P,Si pincer ligand [Si(Me)(C6H4-2-PiPr2)2] permits the
preparation of five-coordinate ruthenium carbene complexes
isoelectronic with the Grubbs catalyst but not isostructural, as
they can offer a coordination site cis to the carbene ligand. The
reactions with a variety of incoming ligands and other reagents
indicate that the first response of these complexes to an
increase of the coordination number around the metal is the
insertion of the carbene ligand into the Ru−Si bond. Such an
insertion generates an alkyl moiety as well as a new
coordination vacancy that enables further reactions. Since the
insertion is reversible, these reactions can be those typical of
carbene ligands, such as C−H bond functionalizations and
ketene formation from CO. Alternatively, they can be those
characteristic of metal-bound alkyls, such as C−H reductive
eliminations and alkyne insertions. Ultimately, this capability of
the pincer contributes to improve and diversify the reactivity of
carbene ligands and therefore might constitute an exploitable
ligand resource. Our current research efforts aim to recognize
transformations that can benefit from this resource and to
extend it to chemical entities other than carbenes.
ASSOCIATED CONTENT
■
Figure 7. Crystal structure of complex 13 at the 50% probability level.
The hydrogen atoms are omitted for clarity. For bond distances and
angles see the Supporting Information.
S
* Supporting Information
Text, figures, and CIF files giving experimental details for
compound syntheses, characterization data, and full crystallo-
graphic descriptions. This material is available free of charge via
the Internet at http://pubs.acs.org.
notable features. It contains a new pincer ligand with an alkenyl
arm that coordinates in an η2 fashion, forming a 1-
ruthenacyclopropene. This type of alkenyl coordination has
been previously observed in complexes of Mo, W, Re, and Os38
but not of Ru, in spite of the fact that it has been proposed to
rationalize the stereochemical outcome of several trans-
formations catalyzed by complexes of this metal.39,40 One
such transformation is the anti addition of silanes to alkynes, in
which the possible participation of η2-alkenyl intermediates
AUTHOR INFORMATION
■
Corresponding Author
martam@unizar.es; sola@unizar.es
Notes
The authors declare no competing financial interest.
19013
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
́
́
́
Gamasa, M. P.; Rodrıguez-Alvarez, Y.; Garcıa-Granda, S. Organo-
metallics 2002, 21, 4815.
ACKNOWLEDGMENTS
■
This research was supported by the Spanish MINECO (Grants
CTQ2009-08023 and CTQ2012-31774) and CSIC (Grant PIE
201280E109).
(16) Katayama, H.; Ozawa, F. Organometallics 1998, 17, 5190.
(17) Wolf, J.; Stuer, W.; Grunwald, C.; Gevert, O.; Laubender, M.;
Werner, H. Eur. .J. Inorg. Chem. 1998, 1827.
(18) Coe, B. J.; Glenwright, S. J. Coord. Chem. Rev. 2000, 203, 5.
(19) Yamaguchi, M.; Arikawa, Y.; Nishimura, Y.; Umakoshi, K.;
Onishi, M. Chem. Commun. 2009, 2911.
(20) Ritter, T.; Hejl, A.; Wenzel, A. G.; Funk, T. W.; Grubbs, R. H.
Organometallics 2006, 25, 5740.
(21) (a) Allen, A. D.; Egle, I.; Janoschek, R.; Liu, H. W.; Ma, J.;
́
Marra, R.; Tidwell, T. T. Chem. Lett. 1996, 45. (b) Valentí, E.; Pericas,
M. A.; Serratosa, F. J. Org. Chem. 1990, 55, 395. (c) Black, T. H.;
Farrell, J. R.; Probst, D. A.; Zotz, M. C. Synth. Commun. 2002, 32,
2083.
(22) For precedents of catalytic formation of ketenes: (a) Paul, N.
D.; Chirila, A.; Lu, H.; Zhang, X. P.; de Bruin, B. Chem. Eur. J. 2013,
19, 12953. (b) Zhang, Z.; Liu, Y.; Ling, L.; Li, Y.; Dong, Y.; Gong, M.;
Zhao, X.; Zhang, Y.; Wang, J. J. Am. Chem. Soc. 2011, 133, 4330.
(23) For leading references on metal-bound carbene insertions:
Franssen, N. M. G.; Walters, A. J. C.; Reek, J. N. H.; de Bruin, B. Catal.
Sci. Technol. 2011, 1, 153.
(24) For carbene migratory insertions in related complexes:
(a) Burrell, A. K.; Clark, G. R.; Rickard, C. E. F.; Roper, W. R.;
Wright, A. H. J. Chem. Soc., Dalton Trans. 1991, 609. See also:
(b) Hill, A. F.; Roper, W. R.; Waters, J. M.; Wright, A. H. J. Am. Chem.
Soc. 1983, 105, 5939. (c) Galan, B. R.; Pitak, M.; Gembicky, M.;
Keister, J. B.; Diver, S. T. J. Am. Chem. Soc. 2009, 131, 6822.
(25) Latos-Grazynski, L.; Cheng, R.-J.; La Mar, G. N.; Balch, A. L. J.
Am. Chem. Soc. 1981, 103, 4270.
REFERENCES
■
(1) (a) Hindson, K.; de Bruin, B. Eur. J. Inorg. Chem. 2012, 340.
(b) Kaim, W. Inorg. Chem. 2011, 50, 9752.
(2) Grutzmacher, H. Angew. Chem., Int. Ed. 2008, 47, 1814.
̈
(3) Crabtree, R. H. New J. Chem. 2011, 35, 18.
(4) (a) Luca, O. R.; Crabtree, R. H. Chem. Soc. Rev. 2013, 42, 1440.
(b) Lyaskovskyy, V.; de Bruin, B. ACS Catal. 2012, 2, 270. (c) Dzik,
W. I.; van der Vlugt, J. I.; Reek, J. N. H.; de Bruin, B. Angew. Chem., Int.
Ed. 2011, 50, 3356. (d) Chirik, P. J.; Wieghardt, K. Science 2010, 327,
794.
(5) (a) Annibale, V. T.; Song, D. RSC Adv. 2013, 3, 11432.
(b) Askevold, B.; Roesky, H. W.; Schneider, S. ChemCatChem 2012, 4,
307. (c) Gunanathan, C.; Milstein, D. Acc. Chem. Res. 2011, 44, 588.
(d) Ikariya, T. Bull. Chem. Soc. Jpn. 2011, 84, 1.
(6) For leading references of bifunctional ligands that exploit
supramolecular interactions see ref 3 and: (a) Dydio, P.; Detz, R. J.;
Reek, J. N. H. J. Am. Chem. Soc. 2013, 135, 10817. (b) Fackler, P.;
Huber, S. M.; Bach, T. J. Am. Chem. Soc. 2012, 134, 12869.
(7) (a) Kaim, W.; Schwederski, B. Coord. Chem. Rev. 2010, 254, 1580.
(b) Buckel, W. Angew. Chem., Int. Ed. 2009, 48, 6779. (c) Stubbe, J.;
Nocera, D. G.; Yee, C. S.; Chang, M. C. Y. Chem. Rev. 2003, 103, 2167.
(d) Whittaker, J. W. Chem. Rev. 2003, 103, 2347. (e) Stubbe, J.; van
der Donk, W. A. Chem. Rev. 1998, 98, 705.
(8) (a) Milstein, D. Top. Catal. 2010, 53, 915. (b) Ikariya, T.;
Blacker, A. J. Acc. Chem. Res. 2007, 40, 1300. (c) Conley, B. L.;
Pennington-Boggio, M. K.; Boz, E.; Williams, T. J. Chem. Rev. 2010,
(26) Whited, M. T.; Grubbs, R. H. Organometallics 2008, 27, 5737.
(27) (a) Mitton, S. J.; McDonald, R.; Turculet, L. Angew. Chem., Int.
Ed. 2009, 48, 8568. (b) Mitton, S. J.; McDonald, R.; Turculet, L.
Polyhedron 2013, 52, 750. (c) Kameo, H.; Ishii, S.; Nakazawa, H.
Dalton Trans. 2013, 42, 4663.
110, 2294. (d) Kaß, M.; Friedrich, A.; Drees, M.; Schneider, S. Angew.
̈
Chem., Int. Ed. 2009, 48, 905. (e) Noyori, R.; Yamakawa, M.;
Hashiguchi, S. J. Org. Chem. 2001, 66, 7931. (f) Dang, Y.; Qu, S.;
Wang, Z.-X.; Wang, X. Organometallics 2013, 32, 2804.
́
́
(28) Garcıa-Camprubí, A.; Martın, M.; Sola, E. Inorg. Chem. 2010,
49, 10649.
(9) (a) Kohl, S. W.; Weiner, L.; Schwartsburd, L.; Konstantinovski,
L.; Shimon, L. J. W.; Ben-David, Y.; Iron, M. A.; Milstein, D. Science
2009, 324, 74. (b) Hetterscheid, D. G. H.; van der Vlugt, J. I.; de
Bruin, B.; Reek, J. N. H. Angew. Chem., Int. Ed. 2009, 48, 8178.
(c) Ackermann, J. Chem. Rev. 2011, 111, 1315.
(29) Stobart, S. R.; Zhou, X.; Cea-Olivares, R.; Toscazo, A.
Organometallics 2001, 20, 4766.
(30) For leading references: (a) Montag, M.; Efremenko, I.; Diskin-
Posner, Y.; Ben-David, Y.; Martin, J. M. L.; Milstein, D. Organo-
metallics 2012, 31, 505. See also: (b) Steinke, T.; Shaw, B. K.; Jong,
H.; Patrick, B. O.; Fryzuk, M. D.; Green, J. C. J. Am. Chem. Soc. 2009,
131, 10461.
(31) (a) Takaya, J.; Iwasawa, N. Organometallics 2009, 28, 6636.
(b) Takaya, J.; Iwasawa, N. J. Am. Chem. Soc. 2008, 130, 15254. See
also: (c) North, M. Angew. Chem., Int. Ed. 2009, 48, 4104.
(32) Doyle, M. P.; Duffy, R.; Ratnikov, M.; Zhou, L. Chem. Rev.
2010, 110, 704.
(10) Sola, E.; García-Camprubí, A.; Andres
J. Am. Chem. Soc. 2010, 132, 9111.
́
, J. L.; Martín, M.; Plou, P.
(11) For recent work in silyl-pincer complexes and leading
references: (a) Kirai, N.; Takaya, J.; Iwasawa, N. J. Am. Chem. Soc.
2013, 135, 2493. (b) Wu, S.; Li, X.; Xiong, Z.; Xu, W.; Lu, Y.; Sun, H.
Organometallics 2013, 32, 3227. (c) Joost, M.; Mallet-Ladeira, S.;
Miqueu, K.; Amgoune, A.; Bourissou, D. Organometallics 2013, 32,
898. (d) Mitton, S. J.; Turculet, L. Chem. Eur. J. 2012, 18, 15258.
(e) Fang, H.; Choe, Y.-K.; Li, Y.; Shimada, S. Chem. Asian J. 2011, 6,
2512. (f) Yang, J.; Del Rosal, I.; Fasulo, M.; Sangtrirutnugul, P.;
Maron, L.; Tilley, T. D. Organometallics 2010, 29, 5544. (g) Zhou, X.;
Stobart, S. R. Organometallics 2001, 20, 1898. See also: (h) Montiel-
́
́
(33) Dıaz-Requejo, M. M.; Perez, P. J. Chem. Rev. 2008, 108, 3379.
(34) Lo, V. K.-Y.; Guo, Z.; Choi, M. K.-W.; Yu, W.-Y.; Huang, J.-S.;
Che, C.-M. J. Am. Chem. Soc. 2012, 134, 7588.
(35) (a) Trnka, T. M.; Day, M. W.; Grubbs, R. H. Organometallics
2001, 20, 3845. (b) O’Connor, J. M.; Baldridge, K. K.; Velez, C. L.;
́
Rheingold, A. L.; Moore, C. E. J. Am. Chem. Soc. 2013, 135, 8826 and
Palma, V.; Munoz-Hernan
́
́
dez, M. A.; Cuevas-Chavez, C. A.; Vendier,
̃
L.; Grellier, M.; Sabo-Etienne, S. Inorg. Chem. 2013, 52, 9798.
references therein. See also: (c) Le Paih, J.; Vovard-Le Bray, C.;
(i) Bruck, A.; Gallego, D.; Wang, W.; Irran, E.; Driess, M.; Hartwig, J.
̈
́
Derien, S.; Dixneuf, P. H. J. Am. Chem. Soc. 2010, 132, 7391.
F. Angew. Chem., Int. Ed. 2012, 51, 11478.
(36) For leading references on alkyne insertions into M−C bonds,
see: Li, L.; Jiao, Y.; Brennessel, W. W.; Jones, W. D. Organometallics
2010, 29, 4593.
(12) (a) Schwab, P.; Grubbs, R. H.; Ziller, J. W. J. Am. Chem. Soc.
1996, 118, 100. (b) Love, J. A.; Sanford, M. S.; Day, M. W.; Grubbs, R.
H. J. Am. Chem. Soc. 2003, 125, 10103.
(37) For an example of 1,3-hydrogen migration, see: (a) MacDougall,
T. J.; Samant, R. G.; Trepanier, S. J.; Ferguson, M. J.; McDonald, R.;
Cowie, M. Organometallics 2012, 31, 1857. For related isomerizations
of hydrocarbyl ligands leading to allyls, see: (b) Esteruelas, M. A.;
(13) For leading references on the mechanism of olefin metathesis:
Ashworth, I. W.; Hillier, I. H.; Nelson, D. J.; Percy, J. M.; Vincent, M.
A. ACS Catal. 2013, 3, 1929.
(14) MacInnis, M. C.; McDonald, R.; Ferguson, M. J.; Tobisch, S.;
Lahoz, F. J.; Onate, E.; Oro, L. A.; Sola, E. J. Am. Chem. Soc. 1996, 118,
89.
(38) Frohnapfel, D. S.; Templeton, J. L. Coord. Chem. Rev. 2000,
206−207, 199.
Turculet, L. J. Am. Chem. Soc. 2011, 133, 13622.
̃
(15) (a) Katayama, H.; Yari, H.; Tanaka, M.; Ozawa, F. Chem.
Commun. 2005, 4336. See also: (b) Lee, J.-H.; Cauton, K. G. J.
Organomet. Chem. 2008, 693, 1664. (c) Bassetti, M.; Marini, S.;
19014
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015

Journal of the American Chemical Society
Article
(39) Imazaki, Y.; Shirakawa, E.; Ueno, R.; Hayashi, T. J. Am. Chem.
Soc. 2012, 134, 14760.
(40) (a) Crabtree, R. H. New J. Chem. 2003, 27, 771. (b) Chung, L.
W.; Wu, Y.-D.; Trost, B. M.; Ball, Z. T. J. Am. Chem. Soc. 2003, 125,
11578.
19015
dx.doi.org/10.1021/ja410822p | J. Am. Chem. Soc. 2013, 135, 19008−19015