Evidence for metal-surface interactions and their role in stabilizing Well-defined immobilized Ru-NHC alkene metathesis catalysts

Article abstract of DOI:10.1021/ja311722k

Secondary interactions are demonstrated to direct the stability of well-defined Ru-NHC-based heterogeneous alkene metathesis catalysts. By providing key stabilization of the active sites, higher catalytic performance is achieved. Specifically, they can be described as interactions between the metal center (active site) and the surface functionality of the support, and they have been detected by surface-enhanced ¹H-²⁹Si NMR spectroscopy of the ligand and ³¹P solid-state NMR of the catalyst precursor. They are present only when the metal center is attached to the surface via a flexible linker (a propyl group), which allows the active site to either react with the substrate or relax, reversibly, to the surface, thus providing stability. In contrast, the use of a rigid linker (here mesitylphenyl) leads to a well-defined active site far away from the surface, stabilized only by a phosphine ligand which under reaction conditions leaves probably irreversibly, leading to faster decomposition and deactivation of the catalysts.

Full text of DOI:10.1021/ja311722k

Article
pubs.acs.org/JACS
Evidence for Metal−Surface Interactions and Their Role in Stabilizing
Well-Defined Immobilized Ru−NHC Alkene Metathesis Catalysts
Manoja K. Samantaray,† Johan Alauzun,‡ David Gajan,§ Santosh Kavitake,† Ahmad Mehdi,‡
Laurent Veyre,† Moreno Lelli,∥ Anne Lesage,∥ Lyndon Emsley,∥ Christophe Coperet,§,*
́
and Chloe Thieuleux*,†
́
†C2P2, UMR 5265, Universite
́
de Lyon, Institut de Chimie de Lyon, UMR 5265 CNRS-Universite
Equipe COMS, ESCPE Lyon, 69616 Villeurbanne, France
‡Institut Charles Gerhardt UMR 5352, Chimie Molec
ulaire et Organisation du Solide, Universite
Cedex 5, France
́
Lyon 1-ESCPE Lyon, LC2P2,
́
́
Montpellier 2, 34095 Montpellier
§Department of Chemistry, ETH Zurich, CH-8093 Zurich, Switzerland
̈
̈
∥Centre de RMN a
̀
Tres
̀
́
Hauts Champs, Universite de Lyon (CNRS/ENS Lyon/UCB Lyon 1), 69100 Villeurbanne, France
S
* Supporting Information
ABSTRACT: Secondary interactions are demonstrated to direct the
stability of well-defined Ru−NHC-based heterogeneous alkene metathesis
catalysts. By providing key stabilization of the active sites, higher catalytic
performance is achieved. Specifically, they can be described as interactions
between the metal center (active site) and the surface functionality of the
support, and they have been detected by surface-enhanced 1H−29Si NMR
spectroscopy of the ligand and 31P solid-state NMR of the catalyst
precursor. They are present only when the metal center is attached to the
surface via a flexible linker (a propyl group), which allows the active site to
either react with the substrate or relax, reversibly, to the surface, thus
providing stability. In contrast, the use of a rigid linker (here mesitylphenyl) leads to a well-defined active site far away from the
surface, stabilized only by a phosphine ligand which under reaction conditions leaves probably irreversibly, leading to faster
decomposition and deactivation of the catalysts.
Here, we show that supported Ru−NHC alkene metathesis
catalysts (Scheme 1) are significantly more stable and display
higher catalytic performances when they are bound to the
surface with a flexible tether. Detailed solid-state NMR
investigations on both the precursors and on the catalysts
suggest that this results from the interaction between the active
metal center and surface functionalities.
Ru−NHC catalysts were prepared via sol−gel strategies
based on supramolecular interactions between the silica
precursor, the functional silane precursor, and the structure-
directing agent to secure the regular localization of the organic
functionalities along the pore channels of the mesostructured
silica matrix,4,5b,7 which were then transformed into supported
NHC−metal complexes. Two classes of catalysts were
investigated having either (i) a flexible alkyl tether (Mat-
RuF) to promote interactions between the metal center (the
active site) and the silica surface (support), here a propyl, or
(ii) a rigid aromatic tether (Mat-RuR) to prevent such
interactions, here a mesitylphenyl (Scheme 1). The ligand−
surface interactions were characterized at an atomic level on the
precursors by dynamic nuclear polarization surface-enhanced
INTRODUCTION
■
Ruthenium-catalyzed alkene metathesis has a profound impact
on a wide range of applications, from performance polymers to
complex natural products and promising pharmaceutical leads.1
Substantial research efforts to improve the performance of
ruthenium alkene metathesis catalysts have been undertaken in
recent years.2 In particular, a considerable amount of work has
focused on obtaining corresponding heterogeneous catalysts
with the aim to improve the efficiency of processes, including
the simplified removal of Ru contaminants.3
In this context, we have recently developed tailored
functional mesostructured materials containing regularly
distributed Ru−N-heterocyclic carbene (NHC) moieties.4
The controlled incorporation of the precursor ligand in the
silica matrix5 and the subsequent surface grafting of Ru
complexes6 yielded well-defined, active, and reusable catalytic
sites. Although we have shown that these Ru−NHC complexes
were highly efficient in the self-metathesis of ethyl oleate, with
performances close to those obtained for equivalent homoge-
neous analogues,4 we had little understanding on the nature of
the active sites and in particular what the role of the surface
functionalities was in affecting the activity and the stability of
the electron deficient Ru catalytic sites.
Received: November 30, 2012
Published: January 30, 2013
© 2013 American Chemical Society
3193
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199

Journal of the American Chemical Society
Article
Scheme 1. (a) Synthesis of NHC−Ru Alkene Metathesis
Catalysts from the Corresponding Imidazolium Derivatives
Scheme 2. Synthesis of Mat-RuR
a
and Structures for (b) Mat-RuF, (c) Mat-RuR, and (d,e) the
b
Corresponding Homogeneous Catalysts
from a molecular precursor having an imidazole terminal group,
which was further transformed into an imidazolium by post-
reaction with benzyl chloride yielding Mat-ImR (Scheme 2; see
Supporting Information for details).
a
R = Ms or CH2Ph, tether = Pr (F) or PhMs (R), OR′ = OiPr or O of
b
Detailed characterization by solid-state NMR spectroscopy of
these materials was then performed, in particular to understand
the ligand orientation with respect to the surface. Two-
dimensional 1H−29Si (Figure 1) were recorded at natural
isotopic abundance by using DNP SENS, a recently introduced
technique that has been shown to provide surface signal
enhancements of up to a factor ∼100 with respect to
conventional NMR instrumentation.8 In DNP SENS, a
paramagnetic polarizing agent is introduced by incipient
wetness impregnation of the material with a radical containing
solution in a suitable solvent. Microwave irradiation of the EPR
transition at low temperature (∼100 K) yields 1H DNP
enhancements of the frozen solid. Cross-polarization can then
the silica surface, and L = PCy3 or silica surface. The silica surface
being amorphous, the structure of the surface species is to be taken as
a representative example based on available data (see text).
NMR spectroscopy (DNP SENS)8 and were shown to
influence the structure of the resulting Ru−NHC surface
species, supported by 31P NMR results, as well as their catalytic
performances.
RESULTS AND DISCUSSIONS
■
The imidazolium materials were prepared using a sol−gel
process in the presence of a structure-directing agent (here
P123, a block copolymer). The targeted material was obtained
by co-hydrolysis and co-condensation of an organotrialkoxy-
silane precursor with tetraethoxysilane as a diluent. A dilution
of 1/30 has been chosen to ensure that the distance between
organic functionality is ca. 1 nm. For the flexible tether, the
material was prepared from iodopropyltriethoxysilane. The
corresponding iodopropyl material was postreacted with
mesitylimidazole to give the imidazolium-containing material
(Mat-ImF) as previously reported.4 All these materials are
highly mesoporous with a narrow pore size distribution
centered at ca. 7 nm according to N2 adsorption/desorption
measurement (calculated by the BJH method using the
adsorption branch of the N2 adsorption isotherm). Trans-
mission electron microscopy and small-angle X-ray diffraction
are consistent with the formation of a 2D hexagonal
arrangement of the porous network (see Supporting
Information). The material with the rigid tether (Mat-ImR),
having similar textural characteristics (pore size, ca. 580 m2/g
and pore volume, ca. 0.8 cm3/g) to Mat-ImF, was prepared
1
be used to transfer the enhanced H magnetization selectively
to the heteronuclei present on the material surface.
NMR samples of Mat-ImR or Mat-ImF were prepared by
impregnation with a 12 mM aqueous solution of TOTAPOL.9
Cross-polarization magic angle spinning (CP-MAS) NMR
spectra were acquired at 9.4 T (400 MHz 1H Larmor
frequency), using a commercial Bruker 400 MHz/263 GHz
gyrotron DNP system (see the Supporting Information for
details). The resulting two-dimensional 1H−29Si correlation
NMR spectra of Mat-ImR and Mat-ImF are shown in Figure 1.
The short mixing time (400 μs) used in the CP step ensures
correlations are observed only between spins in close spatial
proximity to one another. As expected, in both materials we see
correlations between the silicon T-sites and the nearest protons
(aromatic protons for Mat-ImR and aliphatic protons for Mat-
ImF; the weaker intensity of the T3cross-peak in Mat-ImR
compared to the T3 cross-peak in Mat-ImF is accounted for by
1
the longer distance between H and T3-Si spins in Mat-ImR).
In water the SiOH protons are expected to correlate with the Q
3194
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199

Journal of the American Chemical Society
Article
ethane is used as the impregnating solvent for DNP-SENS
(Figure S14).
Mat-ImR and Mat-ImF were converted into the target Ru−
NHC-containing catalytic materials in a two-step sequence: (i)
passivation of the surface silanols by trimethylsilyl groups and
(ii) grafting of [(Cl)2Ru(CHPh)(PCy3)2] using K{N-
(SiMe3)2} (2 equiv/imidazolium) as a base (Scheme 1a). The
Ru loading obtained from elemental analysis (ca. 1.4%wt) shows
the presence of ca. 0.3 Ru per organic functionality for both
systems as typically observed when grafting metal complexes on
imidazolium-containing materials is performed with K{N-
(SiMe3)2}.4,7 A detailed 1H, 13C and 29Si NMR characterization
could not be performed on the catalysts, since so far the DNP
approach provided no signal enhancement in these materials
(probably because of the interaction/reaction of the polarizing
radical and the Ru center). Key structural features were
nevertheless determined from 31P MAS solid-state NMR
spectroscopy. For Mat-RuF only one signal at 47 ppm is
observed, characteristic of P(V) products (Figure 2a,
Figure 1. Contour plots of two-dimensional 9.4 T 1H−29Si HETCOR
DNP SENS spectra for the materials Mat-ImR (a) and Mat-ImF (b).
The blue bands show the expected correlations among the surface Tn
sites and the ortho protons (a) and the CH2−Si protons (b) of the
ligands. In panel b, the red bands show the expected positions for
correlations between the surface Qn silicon atoms and the methyl and
aromatic protons of the ligand. These correlations indicate that the
flexible linker in Mat-ImF is folded to allow interaction with the
surface. To the left of each spectrum the traces perpendicular to the
silicon axes, taken at the silicon frequency indicated by the violet
dashed lines, are shown. Similar conditions were used to acquire both
spectra (full experimental details are given in the Supporting
Information). The DNP enhancements were εH = 30 and 26 for the
spectra acquired on Mat-ImR (a) and Mat-ImF (b), respectively. A
minute amount of sodium 2,2-dimethyl-2-silapentane-5-sulfonate
(DSS), added to the water solution prior to impregnation of the
materials, was used as internal standard to calibrate 1H and 29Si
chemical shifts. A green box in panel a reports the Tn sites with a 2-fold
increased intensity.
sites at around 4 ppm and to give a broad peak superimposed
with the other proton resonances of Mat-ImR and Mat-ImF.
Also as expected, in the material with the rigid tether (Mat-
ImR) we observe correlations between surface Qn sites and the
aromatic protons of the tether. The key finding in these spectra
is that clear correlations are observed between Qn sites and the
aromatic and methyl resonances from the imidazolium and
Figure 2. 31P CPMAS solid-state NMR spectra of (a) Mat-RuF (3584
scans, recycle delay = 2 s and contact time = 2 ms) and of (b) Mat-
RuR (3000 scans, recycle delay = 2 s and contact time = 2 ms); the
peaks at 50 and 37 ppm are in ca. 2:1 ratio (under CP-MAS
conditions, intensities are not quantitative, and ratio only indicative).
All * indicate spinning side bands (MAS frequency = 10 kHz).
1
mesityl fragments in the H−29Si correlation spectrum of Mat-
ImF (Figure 1b). This spectrum shows that the flexible tether
allows the phenyl ring and the attached methyl groups to fold
back onto the surface. Conversely, the absence of correlations
between the methyl resonance of the mesityl ligand and the
surface resonances in Mat-ImR indicates that such ligand−
surface interactions are nonexistent in this material and that this
groups lies away from the surface. The observation for the Mat-
ImF is counterintuitive since, in water, the positively charged
imidazolium ring should prompt the tether to orient away from
the surface, and suggests that this interaction is unexpectedly
strong. The interactions are also observed when tetrachloro-
phosphorus-31 NMR), that result from the reaction of free
PCy3 with the silica surface.11 Note the absence of a
phosphorus signal associated with coordination of PCy3 to
Ru, expected at 36 ppm. This result indicates that the two PCy3
ligands, previously present in [(Cl)2Ru(CHPh)(PCy3)2] are
liberated during grafting and undergo reaction with the surface;
it also implies that the NHC−Ru center attached to the surface
with a flexible tether has no phosphine bound and interacts
with the oxygen of the surface of the material, presumably an
OSiMe3 group or a siloxane bridge,12 to stabilize the otherwise
3195
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199

Journal of the American Chemical Society
Article
unstable 14- electron species (Scheme 1b). In contrast, the
phosphorus-31 CP MAS solid-state NMR spectrum of Mat-
RuR contains two signals of similar intensity for both the P(V)
product and PCy3 coordinated to Ru (Figure 2b). With the
rigid tether, only one free PCy3 ligand is generated, the other
remains coordinated to the Ru center since the rigid tether
prevents stabilization by the surface functionalities (Scheme
1c). Although it was not possible to obtain a direct evidence for
the structure of the alkylidene species via SENS NMR, as was
described for the precursor, the observed absence of
interactions of the imidazolium moiety with the silica surface
in Mat-ImR combined with the presence of PCy3 coordinated
to Ru in Mat-RuR strongly suggests the formation of a NHC−
Ru alkylidene complex coordinated to PCy3 in analogy to the
homogeneous systems (so-called Grubbs-II catalysts, vide
infra). Conversely, the observed evidence of interactions of
the imidazolium moiety with the silica surface in Mat-ImF
combined with the absence of Ru coordinated PCy3 in Mat-
RuF is consistent with the formation of a NHC−Ru alkylidene
species that is most likely stabilized by surface oxygen atoms,
and which could be viewed as an equivalent of the so-called
Grubbs−Hoveyda-II catalysts.2 Note that the corresponding
molecular complexes, RuF and RuR, prepared by the same
approach (reaction of the imidazolium with K{N(SiMe3)2} and
then [(Cl)2Ru(CHPh)(PCy3)2]), have both one NHC and
one PCy3 ligand (Scheme 1d,e), further indicating the
specificity of the surface in Mat-RuF having a flexible tether.
We then investigated the catalytic performance of both
systems with the rigid and flexible tethers, to ask whether the
presence or the absence of metal-surface interactions affects the
catalytic performance of supported catalysts. We examined
three standard reactions (Table 1) that are sensitive to the
structure of the catalysts:
Table 1. Comparison of Catalytic Performance in Well-
Defined Ru−NHC Alkene Metathesis Catalysts
a
Self-metathesis of neat ethyl oleate (EO) was carried out at 40 °C
b
with EO/Ru = 10 000. Ratio of cis to trans products extrapolated to
c
d
0% conversion. No conversion past that time. Initial rates are taken
at the most active stage of the catalyst (after the initiation period).
e
(1) Self-metathesis of ethyl oleate, a pure functionalized Z-
dissymmetric alkene, for which the stereochemical
outcome of the reaction (E/Z ratio) is a fingerprint of
the active sites (each catalyst displays a characteristic E/Z
ratio).4,13
(2) Tandem ring-opening−ring-closing metathesis (RO-
RCM) of cyclooctene. This reaction is very sensitive to
the NHC ligand architecture, providing either dimer/
trimer or polymers.10
The tandem ring-opening−ring-closing metathesis of cyclooctene
(cC8) was carried out at 25 °C in toluene (20 mM cC8) with cC8/Ru
= 10 000. Selectivity in dimers/trimers. Ring-closing metathesis of
f
g
diethyl diallyl malonate (DEDAM) was carried out at 50 °C in toluene
with DEDAM/Ru = 1000; no rates were measure in that case. As
measured by NMR spectroscopy.
h
RuF both display similar very good performance (ca. 1360 vs
1090 h−1 initial rates and 100% conversion), RuR clearly
outperformed its heterogeneous variant (Mat-RuR) in terms of
initial rates (510 h−1 vs 260 h−1) and stability (100% conversion
vs 25% conversion; Table 1). Third, in the RCM of diallyl
diethyl malonate, the catalytic performances were similar for
the RuR and RuF family; the only difference resided in that
homogeneous catalysts were slightly more stable (ca. 90%
conversion) than their heterogeneous analogues (ca. 50%
conversion).
The differences in catalytic performance between the two
heterogeneous catalysts are fully consistent with the fact that
metal-surface interactions have a dramatic effect on the
reactivity and stabilization of reactive intermediates,15 hence
the slightly lower activity of the heterogeneous catalysts and the
large difference in productivity between Mat-RuF and Mat-
RuR. Considering that for all metathesis catalysts a 14-electron
active species must be generated through loss of a coordinating
ligand (PCy3 for Mat-RuR and for the homogeneous catalysts
vs the surface oxygen functionalities for Mat-RuF), the
aforementioned observation can be interpreted as follows
(Scheme 3). For Mat-RuR this step is likely irreversible, in
contrast to the homogeneous catalysts, since free PCy3 reacts
(3) Ring-closing metathesis (RCM) of diallyl diethyl
malonate, a classical alkene metathesis substrate used
to evaluate catalyst activity and stability.14
In the metathesis of ethyl oleate, it is noteworthy that Mat-
RuF rapidly converts ethyl oleate to the expected thermody-
namic mixtures (ca. 50% conv.) within 5 h. Under identical
conditions Mat-RuR deactivates rapidly, leading to only 17%
conversion; allowing longer reaction times does not improve
conversion. This behavior is in sharp contrast with their
homogeneous equivalents as both RuF and RuR convert ethyl
oleate to the thermodynamic equilibrium mixture in less than 5
h. Additionally note that all of the catalysts display similar Z/E
ratios (ca. 2) at extrapolated 0% conversion, which indicates
that homogeneous and heterogeneous catalysts may have
related catalytically active site structures.10
In the tandem RO-RCM of cyclooctene, we observed that
Mat-RuF again displayed much greater catalytic performance
than Mat-RuR, especially in terms of stability since the latter
could only reach 25% conversion (vs 95% conversion for the
former). Although Mat-RuF and its homogeneous analogue
3196
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199

Journal of the American Chemical Society
Article
center in Pascher, Germany. Analysis by gas phase chromatography
was performed on a Hewlett-Packard 5890 series II GC apparatus
equipped with a FID detector and a Fame column (50 m × 0.25 mm)
to monitor the metathesis of ethyl oleate and on an Agilent
Technologies 7890A GC apparatus equipped with a FID detector
and a HP5 column (30 m × 0.32 mm) for RCM of DEDAM and RO-
RCM of cyclooctene. Split ratio of GC was set to 50:1. Sample
injection was 1 μL with a 10-μL syringe. Liquid-state NMR spectra
were recorded using a Bruker AC 300 and solid-state NMR spectra
were recorded under MAS on Bruker Avance 300 and 500 MHz
spectrometers with a conventional double resonance 4 mm CP-MAS
probe. The MAS frequency was set to 10 kHz for all of the
experiments reported here. The samples were introduced in a 4 mm
zirconia rotor in the glovebox and tightly closed.
Scheme 3. Formation of the Active Species and Successive
Reactions
Surface-Enhanced NMR by Dynamic Nuclear Polarization
(SENS). All SENS experiments were performed using a solid-state
DNP-NMR spectrometer designed by Bruker-Biospin with specifica-
tions given by Rosay et al.16 This system consists of a wide-bore 9.4 T
magnet (ωH/(2π) = 400 MHz, ωSi/(2π) = 79.4 MHz) with a Bruker
Avance III spectrometer console, and is equipped with a triple
resonance 3.2 mm low-temperature CPMAS probe. DNP is achieved
by irradiating the sample with microwaves at a frequency of 263 GHz.
The microwaves are generated by a gyrotron and delivered to the
sample by a corrugated waveguide with ∼5 W of power reaching the
sample. Sapphire rotors (endowed with ZrO2 caps) were used for
optimal microwave penetration. The MAS spinning frequencies were
regulated to 8 kHz 2 Hz, the sample temperatures were ∼100 K. 1H
and 29Si Chemical shifts are referenced to DSS at 0 ppm. The pulse
sequence for 2D 1H−29Si HETCOR DNP SENS correlation spectra is
shown in Figure S1. The polarization delay (for DNP) between scans
was 1.5 s in all experiments. The 1H π/2 pulse length was 2.8 μs (ν1 =
89 kHz). A linear amplitude ramp (from 90% to 100% of the nominal
with surface functionalities. After initiation, the propagating
alkylidene species in Mat-RuR is therefore a free unstable 14-
electron reactive species. In the self-metathesis of ethyl oleate,
this highly reactive species will either react or decompose very
rapidly. In contrast, for Mat-RuF, this species can either react
with the alkene moiety of ethyl oleate or coordinate back to the
oxygen functionalities of the surface, providing a stable resting
state for the catalytic active sites. On the other hand, in the
RCM of diethyl diallyl malonate, the alkylidene intermediates
bear a γ-alkenyl fragment, which can stabilize a 14-electron
active species. This would explain why there is hardly any
difference between the two heterogeneous catalysts in this
specific reaction; both being stabilized by the pendant alkenyl
group. However, in tandem RO-RCM, backbiting of the remote
pendant alkenyl functionality is probably not as efficient, and
surface interactions again become important for the stabiliza-
tion of unstable 14-electron Ru intermediates, hence the
difference in stability of Mat-RuF and Mat-RuR.
1
RF field strength) was used for the H channel, with a 400 μs CP
contact time τCP and a nominal RF field amplitude (ν1) of 64 kHz for
1H and 60 kHz for 29Si. SPINAL-6417 proton decoupling was applied
during the acquisition of the 29Si signal with ν1 = 89 kHz. The
eDUMBO-122 scheme18 was used for 1H homonuclear decoupling
during the indirect evolution time with ν1 = 89 kHz (and a basic
eDUMBO cycle of 34 μs). A scaling factor of 0.56 was applied to
correct the 1H chemical shift scale.18 Quadrature detection was
achieved using the States-TPPI19 scheme by incrementing the phase of
CONCLUSION
■
In conclusion, our study shows that the performance of the
catalytic centers in immobilized Ru−NHC systems is greatly
influenced by the nature of the tether, a flexible tether
providing higher stability. Experimental evidence suggests that
this results from surface interactions, which enable the
stabilization of reactive intermediates (as resting states), a
phenomenon typically encountered in bio- or homogeneous
catalysts where residues and ligands play an essential role. This
opens new perspectives in the design of heterogeneous catalysts
where interactions with the surface, or even tailored secondary
surface functional groups, can be implemented to favor or
disfavor a specific reaction pathway.
1
the H spin-lock pulse of the CP step. A total of 128 t1 increments of
64 μs each were recorded, the overall acquisition times in t1 and t2
were 4.3 and 11.4 ms, respectively. The spectra in Figure 1A,B (main
text) were recorded with 16 and 128 scans per increment, respectively.
The 2D spectra were processed with a 2048 × 1024 complex points,
and 80 Hz of exponential line broadening was applied in both
dimensions prior to Fourier transformation. Proton DNP enhance-
ment factors (eH) were determined by scaling the intensities of the
1
direct excitation H spectra obtained under the same experimental
conditions with or without MW irradiation.
Preparation of Homogeneous Catalysts.
Synthesis of RuF. RuF was prepared according to the reported
procedure.4
EXPERIMENTAL SECTION
■
Synthesis of RuR. To the solution of 4-{4′-(tri(isopropoxy)silyl)-
phenyl}-2,6-dimethylphenyl imidazole (0.63 g, 1.41 mmol) in toluene
(10 mL) was added benzyl chloride (0.18 g, 1.41 mmol). After heating
the reaction mixture for 24 h under reflux, the solid was filtered,
washed with pentane (2 × 20 mL), and dried under vacuum to yield 1-
{4′-[4″-(tri(isopropoxy)silylphenyl]-2′,6′-dimethylphenyl}-3-benzyl-
General Procedure. All organometallic syntheses, passivation
steps and grafting experiments for the introduction of the metal
complex were carried out under argon using standard Schlenk
techniques, using dry and degassed solvents. Toluene was freshly
distilled over NaK under Ar in the presence of benzophenone ketyl.
Dichloromethane was freshly distilled over P2O5. Tetraethoxysilane
(TEOS) was distilled over Mg under Ar. Triethylamine, Pluronic
P123, tetramethylsilyl bromide (TMSBr), potassium hexamethyldisil-
azide K{N(SiMe3)2} (0.5 M in toluene), and Grubbs I catalyst
[Cl2(PCy3)2Ru(CHPh)] were bought from Sigma-Aldrich, and
chlorobenzyltriethoxysilane and 3-chloropropyltriethoxysilane were
from ABCR. Cyclooctene was purchased from Aldrich, distilled over
Na prior to use. Diethyl diallyl malonate (DEDAM) was purchased
from Sigma-Aldrich and purified over alumina prior to use. Ethyl
oleate was purchased from Nu-Chek Prep and purified over alumina
prior to use. Elemental analyses were performed at the microanalysis
1
imidazolium chloride as a white solid (ImR). H NMR (CD2Cl2, 300
MHz): δ (ppm) = 11.25 (s, 1H). 7.77 (d, 3JHH = 8 Hz, 2H), 7.62−7.60
(m, 5H), 7.51−7.44 (m, 7H), 7.43 (s, 1H, NCHCHN), 7.21 (s, 1H,
3
NCHCHN), 5.95 (s, 2H, CH2), 4.30 (sept, 3H, JHH = 6 Hz
3
CH(CH3)2), 2.21 (s, 6H, o-CH3), 1.23 (d, JHH = 6 Hz, 18H,
(−CH(CH3)2)3. 13C NMR (CD2Cl2, 75 MHz): δ (ppm) = 143.7,
140.7, 139.3, 135.4, 135.2, 133.8, 129.4, 129.3, 129.0, 127.9, 126.4,
123.1, 122.1, 65.5 (-CH(CH3)2), 25.3 (−CH(CH3)2), 17.8 (Ar−CH3).
HRMS (ESI+): m/z 543.3026 [M−Cl]+, i.e., calculated 543.3037. To a
mixture of ImR (0.35 g, 0.60 mmol) in 5 mL of toluene, was added
3197
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199

Journal of the American Chemical Society
Article
1.45 mL (0.72 mmol) of a 0.5 M toluene solution of K{N(SiMe3)2}.
After stirring the reaction mixture for 20 min, [Cl2(PCy3)2Ru(
CHPh)] (0.44 g, 0.54 mmol) in 10 mL of toluene was added. The
reaction mixture was stirred for 3 h and filtered over Celite. After
removal of the solvent under vacuum the residual solid was crystallized
in toluene and pentane yielding 0.16 g of RuR (42%). 1H NMR
(CD2Cl2, 300 MHz): δ (ppm) = 19.24 (s, 1H), 7.93 (br, 1H), 7.76−
7.49 (m, 5H), 7.39 (t, 1H, J = 7.5 Hz), 7.25−6.98 (m, 5H), 6.88 (br,
2H), 5.92 (s, 2H), 4.32 (m, 1H, -CH(CH3)2) 2.30 (m, 4H), 2.10 (s,
6H), 1.64−1.53 (m, 18H), 1.24 (d, 18H, J = 6 Hz) 1.22−1.07 (m,
14H). 31P NMR (CD2Cl2, 125 MHz): δ (ppm) = 36.4 (s, 1P). 13C
material was dried under high vacuum (10−5 mbar) at room
temperature for 22 h to yield 0.35 g of a light greenish Mat-RuR.
1H solid-state NMR (300 MHz): δ (ppm) = 7.3 (aromatic H), 0.0
(Me3Si). 13C CP-MAS solid-state NMR (75 MHz): δ (ppm) = 134−
128 (aromatic C), 54 (NCH2Ph), 26 (PCy3), 17 (Me2C6H2), 0
(Me3Si). 29Si CP-MAS solid-state NMR: δ (ppm) = +11.9 (OSiMe3),
−82.3 (T3), −101.5 (Q3), −109.6 (Q4). 31P solid-state NMR (75
MHz): δ (ppm) = 50 (P(V)), 37 (Ru-PCy3). Elemental analysis: N =
1.13%wt; Ru = 1.40%wt, P = 0.40%wt. N/Ru ratio obtained: 0.17
(expected 0.5).
Representative Procedure for Metathesis Reactions.
Self-Metathesis of Ethyl Oleate. General Procedure. All metathesis
experiments were carried out under an inert atmosphere of argon. In a
typical run, a 5 mL Schlenk flask was loaded with ethyl oleate and Ru-
catalysts [Mat-RuR, Mat-RuF, Ru-R, and Ru-F] in a 10000:1 ratio,
and the reaction mixture was heated at 40 °C. After certain interval of
time an aliquot of the reaction mixture was drawn, quenched with
ethyl acetate, and analyzed by GC.
Ring-Opening−Ring-Closing Metathesis (RO-RCM) of cis-Cyclo-
octene. The reaction was carried out as described above using a ∼20
mM solution of cyclooctene in toluene with 0.01% of Ru catalysts, a
reaction temperature of 25 °C, and eicosane as internal standard.
Ring-Closing Metathesis (RCM) of DEDAM. The reaction was
carried out as described above using a ∼20 mM solution of DEDAM
with 0.1% of Ru catalysts, a reaction temperature of 50 °C, and
eicosane as internal standard.
NMR (CD2Cl2, 75 MHz): δ (ppm) = 294.4 (HCRu), 187.7 (CNHC
-
Ru, JP−C = 75 Hz), 151.8, 141.9, 141.3, 138.5, 137.3, 135.7, 135.4,
132.2, 130.3, 129.8, 129.3, 129.0, 128.6, 126.7, 124.3, 121.5, 65.9
(-CH(CH3)2), 55.5 (-NCH2Ph), 31.8 (d, JP−C = 16.5 Hz, ipso-CCy),
29.9 (meta-CCy), 28.1 (d, JP−C = 9.8 Hz ortho-CCy), 26.9 (para-CCy),
25.7 (−CH(CH3)2), 18.7 (Ar−CH3). HRMS (ESI+): m/z 1049.4478
[M−Cl]+, i.e., calculated 1049.4493.
Preparation of Heterogeneous Catalysts.
Synthesis of Mat-RuF. Mat-RuF was synthesized according to
literature procedure.4
Synthesis of Mat-RuR. A mixture of 2.70 g of P123 dissolved in an
aqueous HCl solution (109 mL, pH 1.5) was added to a mixture of
TEOS (6.07 g, 29.2 mmol) and 4-{4′-(tri(isopropoxy)silyl)phenyl}-
2,6-dimethylphenylimidazole (0.44 g, 0.97 mmol) at room temper-
ature. The reaction mixture was stirred for 90 min giving rise to a
microemulsion (transparent mixture). To the reaction mixture heated
at 45 °C was added small amount of NaF (15 mg) under stirring
(mixture composition: 0.04 F-/1 TEOS/0.033 of 4-{4′-(tri-
(isopropoxy)silyl)phenyl}-2,6-dimethylphenylimidazole/0.016 P123/
0.12 HCl/220 H2O). The mixture was left at 45 °C under stirring
for 72 h. The resulting solid was filtered and washed with acetone. The
surfactant was removed by an extraction with ethanol using a Soxhlet
during 24 h. After filtration and drying at 135 °C under vacuum 2.95 g
of Mat-R was obtained. Elemental analysis: N = 1.02%wt, Si = 40.0%wt.
Si/N ratio obtained: 19 (expected 20). To Mat-R (0.8 g) was added
toluene (15 mL) and benzyl chloride (0.62 mL, 15 equiv). After
heating the reaction mixture at 135 °C for 72h, the solid was recovered
by filtration and washed successively with toluene (3 × 100 mL),
acetone (3 × 100 mL) and diethyl ether (3 × 50 mL). The solid was
then dried for 14 h under high vacuum (10−5 mbar) at 135 °C,
affording 0.75 g of nonhydrolyzed material. To this material (0.7 g)
was added 5 mL of a 2 M aqueous HCl. After heating the reaction
mixture at 45 °C for 2 h under stirring, the solid was filtered and
washed successively with water (3 × 100 mL), acetone (3 × 100 mL)
and diethyl ether (3 × 50 mL). The solid was then dried for 5−6 h
under high vacuum (10−5 mbar) at 135 °C, to afford 0.6 g of
hydrolyzed material. To this material (0.55 g) was added a mixture of
pyridine (4.1 mL), water (4.1 mL), and 2 M aqueous HCl (0.7 mL).5a
After heating the reaction mixture at 70 °C for 22 h, the solid was
filtered and washed successively with water (3 × 100 mL), acetone (3
× 100 mL), and diethyl ether (3 × 100 mL). The solid was then dried
for 14 h under high vacuum (10−5 mbar) at 135 °C, affording 0.5 g of
Mat-ImR. 1H solid-state NMR (500 MHz): 7.6 ppm (aromatic H), 2.1
(Me2C6H2). 13C CP-MAS solid-state NMR (125 MHz): 134−127
ppm (Car) 54 ppm (NCH2Ph), 15 ppm (Me2C6H2). To the
suspension of Mat-ImR (0.4 g) in toluene (28.0 mL) was added
triethylamine (5.1 mL) and TMSBr (2.5 mL) at room temperature.
The reaction mixture was stirred overnight. The solid was further
filtered and washed successively with toluene (3 × 20 mL) and
dichloromethane (3 × 20 mL). The solid was dried under high
vacuum (10−5 mbar) for 16 h at 135 °C, affording 0.4 g of passivated
Mat-ImR. To the suspension of passivated Mat-ImR (0.38 g, 1.0
equiv) in toluene (2.0 mL) was slowly added a 0.5 M toluene solution
of K{N(SiMe3)2} (0.4 mL, 1.2 equiv) at room temperature. After
stirring for 30 min, [Cl2(PCy3)2Ru(CHPh)] (0.16 g, 1.2 equiv) in
toluene (4.0 mL) was slowly added at room temperature to the
reaction mixture. After stirring the reaction mixture for 20 h, the solid
was filtered and washed successively with toluene (2 × 20 mL) and
dichloromethane (3 × 20 mL) till the filtrate was colorless. The
ASSOCIATED CONTENT
* Supporting Information
■
S
Synthesis scheme for molecular catalyst Ru-R, pulse sequences
for DNP, additional characterization spectra of products, and
catalytic test. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
ccoperet@ethz.ch; thieuleux@cpe.fr
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported in part by the “Agence Nationale de
la Recherche” (ANR-08-BLAN-0151). Financial support is
acknowledged from EQUIPEX Contract No. ANR-10-EQPX-
47-01 and ETH Zurich. The authors wish to thank Drs. Werner
̈
Maas, Shane Pawsey, and Melanie Rosay and Bruker Biospin
Corporation (Billerica, U.S.A.) for support with the DNP-
SSNMR spectrometer.
REFERENCES
■
(1) (a) Grubbs, R. H. Angew. Chem., Int. Ed. 2006, 45, 3760−3765.
(b) Vougioukalakis, G. C.; Grubbs, R. H. Chem. Rev. 2010, 110, 1746−
1787.
(2) (a) Herrmann, W. A. Angew. Chem. Int. Ed. 2002, 41, 1290−1309.
(b) Deshmukh, P. H.; Blechert, S. Dalton Trans. 2007, 2479−2491.
(c) Monfette, S.; Fogg, D. E. Chem. Rev. 2009, 109, 3783−3816.
(d) Samojlowicz, C.; Bienek, M.; Grela, K. Chem. Rev. 2009, 109,
3708−3748. (e) Hoveyda, A. H.; Malcolmson, S. J.; Meek, S. J.;
Zhugralin, A. R. Angew. Chem., Int. Ed. 2010, 49, 34−44. (f) Nolan, S.
P.; Clavier, H. Chem. Soc. Rev. 2010, 39, 3305−3316.
(3) For reviews, see: (a) Buchmeiser, M. R. New J. Chem. 2004, 28,
́
549−557. (b) Coperet, C.; Basset, J. M. Adv. Synth. Catal. 2007, 349,
78−92. (c) Clavier, H.; Grela, K.; Kirschning, A.; Mauduit, M.; Nolan,
S. P. Angew. Chem., Int. Ed. 2007, 46, 6786−6801. (d) Buchmeiser, M.
H. Chem. Rev. 2009, 109, 303−321. For seminal and selected reports
in the field: (e) Mayr, M.; Mayr, B.; Buchmeiser, M. R. Angew. Chem.,
3198
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199

Journal of the American Chemical Society
Article
Int. Ed. 2001, 40, 3839−3842. (f) Pruehs, S.; Lehmann, C. W.;
Fuerstner, A. Organometallics 2004, 23, 280−287. (g) Halbach, T. S.;
Mix, S.; Fischer, D.; Maechling, S.; Krause, J. O.; Sievers, C.; Blechert,
S.; Nuyken, O.; Buchmeiser, M. R. J. Org. Chem. 2005, 70, 4687−4694.
(h) Allen, D. P.; vanWingerden, M. M.; Grubbs, R. H Org. Lett. 2009,
11, 1261−1264.
(18) Fung, B. M.; Khitrin, A. K.; Ermolaev, K. J. Magn. Reson. 2000,
142, 97−101.
(19) Marion, D.; Ikura, M.; Tschudin, R.; Bax, A. J. Magn. Reson.
1989, 85, 393−399.
(4) Karame, I.; Boualleg, M.; Camus, J. M.; Maishal, T. K.; Alauzun,
J.; Basset, J. M.; Coper
Reye,
C.; Veyre, L.; Thieuleux, C. Chem.Eur. J. 2009, 15, 11820−
11823.
́
et, C.; Corriu, R. J. P.; Jeanneau, E.; Mehdi, A.;
́
́
(5) (a) Mehdi, A.; Reye, C.; Corriu, R. J. P. Chem. Soc. Rev. 2011, 40,
563−574. (b) Hoffmann, F.; Cornelius, M.; Morell, J.; Froba, M.
Angew. Chem., Int. Ed. 2006, 45, 3216−3251.
́
(6) (a) Coperet, C.; Chabanas, M.; Petroff Saint-Arroman, R.; Basset,
J. M. Angew. Chem., Int. Ed. 2003, 42, 156−181. (b) Thomas, J. M.;
Raja, R.; Lewis, D. W. Angew. Chem., Int. Ed. 2005, 44, 6456−6482.
(c) Tada, M.; Iwasawa, Y. Coord. Chem. Rev. 2007, 251, 2702−2716.
(d) Wegener, S. L.; Marks, T. J.; Stair, P. C. Acc. Chem. Res. 2012, 45,
206−214.
́
(7) Maishal, T. K.; Alauzun, J.; Basset, J.-M.; Coperet, C.; Corriu, R.
J. P.; Jeanneau, E.; Mehdi, A.; Reye, C.; Veyre, L.; Thieuleux, C. Angew.
Chem., Int. Ed. 2008, 47, 8654−8656.
(8) (a) Lesage, A.; Lelli, M.; Gajan, D.; Caporini, M. A.; Vitzthum, V.;
Mieville, P.; Alauzun, J.; Roussey, A.; Thieuleux, C.; Mehdi, A.;
́
Bodenhausen, G.; Coperet, C.; Emsley, L. J. Am. Chem. Soc. 2010, 132,
15459−15461. (b) Lelli, M.; Gajan, D.; Lesage, A.; Caporini, M. A.;
Vitzthum, V.; Mieville, P.; Heroguel, F.; Rascon, F.; Roussey, A.;
́
Thieuleux, C.; Boualleg, M.; Veyre, L.; Bodenhausen, G.; Coperet, C.;
Emsley, L. J. Am. Chem. Soc. 2011, 133, 2104−2107. (c) Rossini, A. J.;
Zagdoun, A.; Lelli, M.; Gajan, D.; Rascon, F.; Rosay, M.; Maas, W. E.;
́
Coperet, C.; Lesage, A.; Emsley, L. Chem. Sci. 2012, 3, 108−115.
(d) Zagdoun, A.; Casano, G.; Ouari, O.; Lapadula, G.; Rossini, A. J.;
Lelli, M.; Baffert, M.; Gajan, D.; Veyre, L.; Maas, W. E.; Rosay, M.;
́
Weber, R. T.; Thieuleux, C.; Coperet, C.; Lesage, A.; Tordo, P.;
Emsley, L. J. Am. Chem. Soc. 2012, 134, 2284−2291. (e) Zagdoun, A.;
Rossini, A. J.; Gajan, D.; Bourdolle, A.; Ouari, O.; Rosay, M.; Maas, W.
́
E.; Tordo, P.; Lelli, M.; Emsley, L.; Lesage, A.; Coperet, C. Chem.
Commun. 2012, 48, 654−656. (f) Rossini, A. J.; Zagdoun, A.; Lelli, M.;
Canivet, J.; Aguado, S.; Ouari, O.; Tordo, P.; Rosay, M.; Maas, W. E.;
Coper
Ed. 2012, 51, 123−127. (g) Vitzthum, V.; Mieville, P.; Carnevale, D.;
Caporini, M. A.; Gajan, D.; Coperet, C.; Lelli, M.; Zagdoun, A.;
́
et, C.; Farrusseng, D.; Emsley, L.; Lesage, A. Angew. Chem., Int.
́
Rossini, A. J.; Lesage, A.; Emsley, L.; Bodenhausen, G. Chem. Commun.
2012, 48, 1988−1990.
(9) Song, C. S.; Hu, K. N.; Joo, C. G.; Swager, T. M.; Griffin, R. G. J.
Am. Chem. Soc. 2006, 128, 11385−11390.
(10) Kavitake, S.; Samantaray, M. K.; Dehn, R.; Deuerlein, S.;
Limbach, M.; Schachner, J. A.; Jeanneau, E.; Coper
Dalton Trans. 2011, 40, 12443−12446.
́
et, C.; Thieuleux, C.
(11) (a) Sommer, J.; Yang, Y.; Rambow, D.; Bluemel, J. Inorg. Chem.
2004, 43, 7561−7563. (b) Hilliard, C. R.; Bhuvanesh, N.; Gladysz, J.
A.; Bluemel, J. Dalton Trans. 2012, 41, 1742−1754.
́
(12) Chabanas, M.; Baudouin, A.; Coperet, C.; Basset, J. M.; Lukens,
W.; Lesage, A.; Hediger, S.; Emsley, L. J. Am. Chem. Soc. 2003, 125,
492−504.
́
(13) Coperet, C. Beilstein J. Org. Chem. 2011, 7, 13−21.
(14) Ritter, T.; Hejl, A.; Wenzel, A. G.; Funk, T. W.; Grubbs, R. H.
Organometallics 2006, 25, 5740−5745.
(15) (a) Blanc, F.; Coperet, C.; Thivolle-Cazat, J.; Basset, J.-M.;
́
Lesage, A.; Emsley, L.; Sinha, A.; Schrock, R. R. Angew. Chem., Int. Ed.
2006, 45, 1216−1220. (b) Coperet, C. Pure Appl. Chem. 2009, 81,
585−596.
́
(16) Rosay, M.; Tometich, L.; Pawsey, S.; Bader, R.; Schauwecker, R.;
Blank, M.; Borchard, P. M.; Cauffman, S. R.; Felch, K. L.; Weber, R.
T.; Temkin, R. J.; Griffin, R. G.; Maas, W. E. Phys. Chem. Chem. Phys.
2010, 12, 5850−5860.
(17) Elena, B.; de Paepe, G.; Emsley, L. Chem. Phys. Lett. 2004, 398,
532−538.
3199
dx.doi.org/10.1021/ja311722k | J. Am. Chem. Soc. 2013, 135, 3193−3199