Well-defined supported mononuclear tungsten oxo species as olefin metathesis pre-catalysts

Article abstract of DOI:10.1021/cs501294j

[WOCl₄] was grafted on silica dehydroxylated at 200 °C, and the structure of the surface species was elucidated by a combination of spectroscopic and theoretical methods, demonstrating the formation of [(≡SiO)₂WOCl₂] (1a)

Full text of DOI:10.1021/cs501294j

Research Article
pubs.acs.org/acscatalysis
Well-Defined Supported Mononuclear Tungsten Oxo Species as
Olefin Metathesis Pre-Catalysts
†
†
‡
†
†
Yassine Bouhoute, Anthony Garron, Denys Grekov, Nicolas Merle, Kai C. Szeto,
†
⊥
,⊥
‡
is M. Gauvin,*,‡
́
Aimery De Mallmann, Iker Del Rosal, Laurent Maron,* Guillaume Girard, Reg
,‡
,†
Laurent Delevoye,* and Mostafa Taoufik*
†
Laboratoire de Chimie, Catalyse, Polymer
918, F-69616 Villeurbanne Cedex, France
̀ ́ ́
es et Procedes, UMR 5265 CNRS, UCBL, ESCPE Lyon, 43 Boulevard du 11 Novembre
1
‡
Unite
́
de Catalyse et de Chimie du Solide, CNRS UMR 8181, Universite
Laboratoire de Physico-Chimie des Nano-Objets, CNRS UMR 5215, Universite
F-31077 Toulouse, France
S Supporting Information
́
de Lille, F-59655 Villeneuve d’Ascq, France
⊥
́
de Toulouse, INSA, UPS, 135 Avenue de Rangueil,
*
ABSTRACT: [WOCl ] was grafted on silica dehydroxylated at 200 °C, and
4
the structure of the surface species was elucidated by a combination of
spectroscopic and theoretical methods, demonstrating the formation of [(
SiO) WOCl ] (1a) as the major species accompanied by minor monopodal
2
2
17
species [(SiO)WOCl ] (1b). Most noteworthy, EXAFS and O NMR
3
combined to DFT calculations helped elucidate the structure of the surface
species. Alkylation was performed using SnMe , affording methyl species that
4
were also precisely characterized. The alkylated species achieved excellent
performances in isobutene metathesis to 2,3-dimethylbutene.
KEYWORDS: metathesis, surface chemistry, tungsten, tin, alkylation
INTRODUCTION
species through a more practical way than grafting of sensitive
and expensive organometallics. An early example of such
■
1
Olefin metathesis is a reaction of major industrial importance.
The most widely used type of catalyst is WO deposited on an
reactivity was provided by Mol, who used SnMe as an activator
4
3
11
of [WOCl ] to trigger metathesis activity. On the basis of
4
inorganic support such as silica. In this case, it has been
proposed that the active sites are bipodal high valent carbene
these premises, we set out to study the grafting of [WOCl4]
under controlled conditions. We selected silica dehydroxylated
at 200 °C as the support, as it mostly bears vicinal silanols that
2
,3
4
species. Using surface organometallic chemistry approach,
we developed the first example of well-defined (monopodal)
oxo alkyl species that performed efficiently in olefin metathesis,
as a step further toward the development of model species for
12
have been reported to favor the formation of bipodal species.
The alkylation of this grafted complex is then studied, targeting
[
(SiO) WO(CH ) ], closely related to the active species
5
2
3 2
the WO /SiO active sites. Several years later, this approach
2
3
2
proposed in the case of the WO /SiO industrial catalyst. The
6
3 2
was also explored by others, generating organometallic surface
species singly bound to the support through a siloxide moiety.
However, concerns arose, regarding, for example, the lack of
selectivity of the grafting and/or arduous synthesis of molecular
organometallic precursors.
modification of coordination sphere of W is followed step by
step through chemical and spectroscopic characterizations
17
13
(
mass balance analysis, O and C NMR with labeled
compounds, IR, EXAFS) supported by DFT calculations.
We reasoned that cleavage of the metal−chlorine bond can
be a more convenient entry into the anchoring of mononuclear
centers onto an inorganic support. This was shown for
RESULTS AND DISCUSSION
■
[
Grafting of [WOCl
was carried out by a mechanical mixing
] on SiO2−200. The grafting of
4
WOCl ] on SiO
[
[
WOR Cl], where the W−Cl was selectively cleaved to afford
4
2−200
3
7
at 80 °C. After 3 h, volatile compounds were condensed into an
(SiO)WOR ] species (with R = CH tBu). Along the same
3
2
line, previous examples for the grafting of transition metal
chlorides afforded well-defined surface species, as in the case of
Received: August 29, 2014
Revised: October 17, 2014
8
9
10
[
VOCl ], [CrO Cl ] or [TiCl ]. The resulting chloride
3 2 2 4
species could then be alkylated and thus afford tungsten alkyl
©
XXXX American Chemical Society
4232
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
infrared cell (of known volume) in order to quantify HCl
evolved during grafting by infrared spectroscopy and to
establish the mass balance of the reaction. Then, the powder
was washed three times with toluene in order to remove the
consistent with the bond distances found in [OWCl (OAr) ], a
2
2
7
type of molecular complexes where the WO (1.638−1.715
Å), W−O (1.831−1.978 Å), and W−Cl (2.309−2.383 Å) bond
1
3−20
lengths are in the same range.
Similar W−Cl bond
(2.28
Å for W−Cl; 1.8 Å for WO) or for molecular complexes
is bonded to azoxybenzene (2.279 to 2.308 Å
4
23
excess of [WOCl ] and finally rinsed with dry pentane to
distances have also been observed for crystalline WOCl
4
21
4
remove any physisorbed toluene, affording WOCl -SiO
as
4
2−200
22
a pale yellow material. The reaction of [WOCl ] proceeds with
where WOCl
for W−Cl; 1.651 to 1.687 Å for WO) or to a ketone,
CF H Br (2.259 to 2.308 Å for W−Cl;
6 5
4
consumption of almost all free surface silanols of SiO2−200, as
shown from the irreversible disappearance of the silanol
C(O)CHCHNHC
3
−1
1
.664 Å for WO). Similar parameters were obtained when
stretching band at 3747 cm . (Figure 1).
2
fitting the k .χ(k) spectrum. The results thus agree with a major
bis-siloxy structure, [(SiO) WOCl ] 1a (70−80%), along
2
2
with the monosiloxy species, 1b (20−30%). This result is
supported by NMR of the sample [WOCl ] on SiO
4
2−200
selectively enriched on the oxo moiety by 1 O NMR (see
7
below).
To gain further insights on the above-described experimental
data, DFT calculations were carried out to confirm and refine
the understanding of the grafting reaction of [WOCl ] onto
4
silica dehydroxylated at 200 °C (SiO2−200). In the continuation
7
,24−28
of previous theoretical studies,
we considered as surface
models three polyoligosilsesquioxane derivatives: monosilanol
c, and bis-silanols b and ac, with differing connectivity between
the two SiOH groups (Figure 3). These models are suitable for
isolated (model c) and for vicinal silanol groups (models b and
ac). Thus, c accounts for highly dehydroxylated SiO2−700
surface, whereas c, b, and ac represent the various
configurations for the SiO2−200 surface.
Figure 1. DRIFT spectra of SiO2−200 (a); WOCl -SiO
WOMe-SiO2−200 (c).
(b);
2−200
4
The grafting reaction was thus studied using these models.
Full detailed studies available in the Supporting Information
The amount of W and Cl, as quantified by elemental analysis
of WOCl -SiO are 7.26 and 3.13 wt %, respectively, which
4
2−200
describe the reaction of [WOCl ] with the three silica models.
4
accounts for 2.22 Cl/W. On the other hand, 0.396 mmol of W/
g of silica is grafted, although the concentration of silanols is
In the case of the vicinal silanols models b and ac, this includes
a first study of a monografting step, that yields a [(
1
3
about 0.86 mmol OH/g on SiO2−200, which amounts to a W/
OH ratio of 0.46. Quantification of gas evolved during the
grafting (by integrating the ν(H−Cl) and by comparing to a
calibration curve) showed that 1.7 HCl/W are released. The
results obtained with SiO2−200 are in agreement with the
protonolysis of about two W−Cl moieties by silanols and
formation of a mixture of bisiloxy, [(SiO) WOCl ] major
SiO)WOCl ] species in proximity to a free silanol. In
3
agreement with experimental results, silanolysis of the W−Cl
bond is calculated to proceed efficiently by coordination of the
surface silanol onto the tungsten center followed by HCl
elimination. To summarize, the nature of the silanol group has
a great influence on the geometry and stability of the grafted
complex. As expected, the type of silanol at the silica surface
2
2
species 1a (ca. 80%) and monosiloxy [(SiO)WOCl ] minor
3
determines the grafting mode of [WOCl ]. Indeed, the
4
species 1b (ca. 20%) (Scheme 1).
presence of isolated silanol groups leads to the formation of
monografted complexes, whereas the presence of vicinal silanol
groups induces the formation of bisgrafted species (in this case
the monografted species evolves into the bisgrafted one by
reaction with the neighboring silanol). In the case of vicinal
silanols, contrary to what was observed for group 4 metals
Scheme 1. Grafting of [WOCl ] on SiO
4
2‑200
28
neopentyl (Np) derivatives, the reaction does not stop at the
monosiloxy stage and proceeds to formation of bigrafted
species. Furthermore, the flexibility of the vicinal anchoring site
is also a key factor that dictates the structure of the resulting
isomers.
Characterization of WOCl -SiO2−200 was carried out further
4
by a XAS study, which consisted in recording and analyzing the
Figures 4 and 5 represent the optimized structures for mono-
and bisgrafted species, respectively. Calculated bond lengths are
in the same range as those observed with EXAFS (see Table
S1). Their relative energy indicates a significant preference for
the square pyramidal coordination sphere, with the oxo moiety
in the apical position, for both 1a and 1b species, as in
W L -edge extended X-ray absorption fine structure (EXAFS)
III
spectra of the supported tungsten complexes resulting from the
grafting of [WOCl ] on SiO2−200.
4
For WOCl -SiO2−200, the parameters obtained from the fit of
4
3
the k .χ(k) EXAFS signal are consistent with the following
coordination sphere around W (Figure 2 and Table 1): one
oxygen atom at 1.68(1) Å, assigned to an oxo ligand, a bit less
than two oxygen atoms (1.7 ± 0.3 O) at 1.87(1) Å,
corresponding to σ-bonded siloxy ligands and a bit more
than two chloride centers (2.3 ± 0.3 Cl) at 2.30(1) Å. This is
[WOCl ]. Two types of isomers for 1a are accessible, with
4
siloxy ligands in cis (1aac‑1, Figure 5a, and 1ab‑1, Figure 5d) or
trans (1aac‑2, Figure 5b, and 1ab‑2 Figure 5e) configuration, with
a higher stability for the cis configuration (in the chosen
nomenclature, for instance 1aac‑1 refers to a calculated structure
4
233
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
3
Figure 2. Tungsten L -edge k -weighted EXAFS (left) and corresponding Fourier transform (right, modulus and imaginary part) with comparison
III
to simulated curves for WOCl -SiO2−200. Solid lines: experimental; dashed lines: spherical wave theory.
4
a
Table 1. EXAFS Parameters Obtained for WOCl -SiO2‑200.
4
2
2
type of neighbor
number of neighbors
distance (Å)
σ (Å )
W=O
1.0
1.68(1)
1.87(1)
2.30(1)
0.0012(6)
0.0014(8)
0.0036(6)
W-OSi
1.7(3)
b
W-Cl
2.3(3)
The errors generated by the EXAFS fitting program “RoundMidnight”
a
−1
are indicated in parentheses. Δk: [2.1−18.0 Å ] - ΔR: [0.7−2.4 Å];
2
S
= 0.94; ΔE = 9.5 ± 2.0 eV (the same for all shells); Fit residue: ρ
0
0
2
b
=
4.1%; Quality factor: (Δχ) /ν = 3.85, with ν = 11/19. Shell
constrained to the parameter above.
Figure 4. Optimized structures of complexes resulting from the
of species 1a built with an ac-type silica model). The isomers
grafting of [WOCl ] on an isolated silanol group (c model). ΔG:
4
featuring a trigonal bipyramidal coordination type (1aac‑3
Figure 5c, and 1ab‑3, Figure 5f) are significantly less stable.
,
−1
relative Gibbs free energy of the different isomers (kcal mol ). Atom
color code: blue: W; red: O; green: Cl; pale brown: Si; white: H.
Having established preferred structural types as potential
surface species from theoretical calculations, a further step
would be to correlate such results to experimental evidence and
reach a higher degree of understanding, beyond that accessible
through the above-described techniques (IR and EXAFS,
mostly).
[WCl ] by 1 equiv of 17O-labeled Me SiO*SiMe (obtained
30
6
3
3
1
7
through hydrolysis of Me SiCl with 90%- O-enriched water).
3
Enrichment of the molecular precursor was confirmed by 7O
MAS NMR, as the spectrum comprises a single pattern that
indicates the presence of a single site with isotropic chemical
shift (CS) of 671 ppm (Figure S4). The sideband manifold
results chiefly from chemical shift anisotropy (CSA). Here the
presence of satellite transitions allows for fine adjustment of
further parameters, most precisely, of the quadrupolar coupling
constant (C ) and the asymmetry parameter (η ). Table 2
1
17
We have previously demonstrated in related systems that O
NMR can afford most valuable structural information on oxo
surface species, thanks inter alia to diagnostic quadrupolar and
7,29
chemical shift anisotropic features.
These informative NMR
parameters, when combined with DFT calculations, can afford
access to structural data. We therefore proceeded to
Q
Q
1
7
preparation of O-labeled WOCl -SiO
(WO*Cl4-
gathers the extracted NMR parameters, and Figure S4 in
Supporting Information illustrates the finer influence of
quadrupolar coupling on the outermost side bands (See
4
2−200
1
7
17
SiO2−200) for O NMR studies. O-enriched [WOCl4]
designated as [WO*Cl ]) was prepared by treatment of
(
4
Figure 3. Optimized structures of (a) [WOCl ] and of (b−d) the silanol models. Atom color code: blue: W; red: O; green: Cl; pale brown: Si;
4
white: H.
4
234
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
Figure 6. 17O MAS NMR spectrum of WO*Cl -SiO
at 18.8 T at
4
2−700
a spinning speed of 20 kHz, number of scans: 16384 (a) together with
best fit simulation (b). Asterisks designate the rotor’s ZrO signal.
2
Figure 5. Optimized structures of complexes resulting from the
bigrafting reaction of [WOCl ] on vicinal silanol groups (b and ac
4
ppm (cf. Table 2). Recording the spectrum at lower magnetic
field (9.4 T) confirmed the small value of the quadrupolar
coupling constant, as no broadening was detected (Figure S5).
models). ΔG: relative Gibbs free energy of the different isomers (kcal
−1
mol ). Atom color code: blue: W; red: O; green: Cl; pale brown: Si;
white: H.
From best-fit simulations, an upper C value of 2.0 MHz was
determined.
From comparison with the DFT model species 1bc‑1 and
Q
Table 2. Experimental and DFT-Calculated 17O NMR
Parameters for Mono- and Bigrafted Surface Species
1
b , experimental parameters (CS and C , Table 2) are well
c‑2 Q
CS
ppm)
CQ
(MHz)
Δ
CSA
in line with the former, namely, a square pyramidal
coordination sphere, rather than trigonal bipyramidal, which
allows us to conclude on the coordination geometry of the [(
(
ηQ
(ppm)
ηCSA
[
WOCl4]
1.0
exptl:
671
708
0.7
−330
−479
0.6
SiO)WO*Cl ] 1b surface species. One must also consider that
3
DFT calcd:
0.3
0.01
0.00
the chemical shift anisotropy is not a discriminating parameter
in this case.
bisgrafted species
a
exptl:
1a
810
<2.5
1.2
2.6
5.5
1.4
2.6
4.2
nd
−650
−496
−584
−719
−628
567
0.7
17
The O MAS NMR spectrum of WO*Cl -SiO
4
2−200
DFT calcd: 1aac‑1
806
625
916
798
860
821
0.09
0.28
0.27
0.26
0.15
0.27
0.07
0.74
0.21
0.10
0.90
0.20
recorded at 21.15 T (Figure 7a) results from overlapping of
several components. Prior to further analysis, we have checked
the origin of the signal broadening, which may be due to either
1
1
aac‑2
aac‑3
1
1
1
ab‑1
ab‑2
ab‑3
29
17
large C or to CS distribution. Recording of the O NMR
Q
−591
monografted species
a
exptl:
1b
829
2.0
1.6
5.9
nd
−350
−626
−743
1.0
DFT calcd:
1bc‑1
830
923
0.70
0.02
0.25
0.02
1bc‑2
a
Not determined.
Experimental Section for definitions). DFT calculations of
WOCl4] 17O NMR parameters have not been carried out on
[
an isolated molecule, but they have been determined by taking
into account the specific arrangement found in the crystal
structure (See Supporting Information). As seen in Table 2, a
good agreement is obtained for CS. Both C and CSA values
Q
are in reasonable agreement with experimental features,
considering the accuracy of their determination.
This compound was used for preparation of WO*Cl4-
SiO2−200 by reaction of [WO*Cl ] with SiO2−200. Furthermore,
4
inspired by the conclusions drawn from EXAFS, that indicate
presence of mono- and bisgrafted tungsten centers, we also
prepared material WO*Cl -SiO
from the reaction of
WO*Cl ] with SiO2−700. As this support features only
4
2−700
[
Figure 7. 17O MAS NMR spectrum of WO*Cl -SiO
a spinning speed of 20 kHz, number of scans: 114 688 (a) together
with best fit simulation (b) and individual lineshapes for 1a (black), 1b
(dark gray) and impurity (light gray) (c). Asterisks designate the
4
at 21.15 T at
2−200
4
noninteracting silanols, this will only comprise the monografted
1
7
species [(SiO)WO*Cl ] (1b*). The O MAS NMR
3
spectrum of WO*Cl -SiO
is displayed in Figure 6a. It
features a pattern characteristic of a single site, at CS of 829
4
2−700
rotor’s ZrO signal.
2
4
235
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
spectrum at lower field (9.4 T) affords a signal with comparable
width (as expressed in ppm, see Supporting Information, Figure
S6). This indicates that the width of the signal is mainly due to
chemical shift distribution. The resulting best-fit simulation is
presented in Figure 7b (see Table 2 for corresponding NMR
parameters), and can be decomposed into three components,
based on prior knowledge. As in our previous work, a
Gaussian/Lorentzian line broadening of 1500 Hz was used to
reaction of solid WOCl -SiO
with a vapor pressure of
2−200
4
SnMe * was performed at 80 °C during 12 h, generating
4
material WOMe*-SiO2−200 as a brown material. All volatiles
were transferred to an NMR tube fitted with a Young valve and
13
containing C D . The resulting C NMR spectrum displayed
6
6
signals accounting for SnMe (signal centered at −9.43 ppm
4
1
117/119
13
with a J(
Sn− C) of 320.4/335.1 Hz), as well as for the
expected Me SnCl side-product (−1.74 ppm with a
3
7
1
117/119
31,32
account for NMR parameters distribution. One component at
J(
Sn-CH) of 360.1/377.4 Hz).
No signal to
CS of 829 ppm with large CSA is identified as 1b from
comparison with data described above (Figure 6). It
corresponds to a minor contribution of monografted species
Me SnCl was observed. The DRIFT spectrum of WOMe*-
2
2
SiO2−200 (Figure 1c) displays new peaks compared to WOCl -
4
−1
SiO2−200. Bands at 1455 and 1490 cm , concomitant with the
−
1
[
(SiO)WOCl ] 1b in WO*Cl -SiO2−200, as already
bands in the 2800 to 3000 cm region, are characteristic of
δ(C−H) and ν(C−H) respectively, accounting for methyl
moieties. Moreover, the small amount of remaining isolated
silanol groups is fully consumed after the alkylation, which can
originate from reaction of SnMe to afford [(SiO)SnMe ]
3
4
evidenced by EXAFS. A sharp signal at 480 ppm, featuring
no CSA is also present, deduced from recording the spectrum
at different MAS spinning speeds. The nature of the species
accounting for less than 10% of the total signal is still unknown.
Finally, the best fit simulation also includes a third component
due to bigrafted species [(SiO) WOCl ] 1a*, present in
4
3
33
1
species (vide infra). The H MAS NMR spectrum of
WOMe*-SiO2−200 (Figure 8a) shows two broad signals around
2
2
dominant proportion. This site features a CS of 810 ppm, and it
is characterized by large CSA (−650 ppm) and low C
Q
(
estimated to be less than 2.5 MHz), as determined by best
fit simulation. In the case of supported species 1a, that features
large CSA and low C , and where satellite transitions cannot be
Q
observed, the sideband manifold appears as dominantly affected
by 17O CSA. As no discontinuity appears in each individual
band of the spectrum, only an upper value of quadrupolar
coupling constant can be proposed from best-fit simulation
(Table 2).
These values were compared with those of the DFT-
optimized structures (Table 2). The combination of CS, CQ,
and CSA set of parameters allows for selection of 1aac‑1 and
1
1
ab‑1 as the closest candidates. Interestingly, they are also the
most stable configurations in the series (Figure 5). The NMR
lineshapes for all the calculated species (thus including 1aac‑2
aac‑3, 1a , and 1a ) are presented in Figure S7. Their
Figure 8. (a) H MAS NMR spectrum of WOMe*-SiO
(500
2−200
1
3
MHz, 8 scans, relaxation delay: 2 s, spinning speed: 10 kHz) (b)
C
,
CP MAS NMR spectrum of WOMe*-SiO2−200 (125.7 MHz, 30 000
scans, relaxation delay: 2 s, CP contact time: 2 ms, spinning speed: 10
kHz) with 100% 13C isotopic enrichment of the methyl carbon.
1
b‑2
b‑3
comparison further demonstrates the impact of the anisotropy
onto the resulting signal and confirms our selection of models
1
aac‑1 and 1a . Thus, we can conclude that surface species
1.2 and 0.1 ppm, accounting for the methyl moieties bound on
tungsten and tin, respectively. Similarly, the 13C CP MAS
spectrum features a sharp and intense signal at 46 ppm (Figure
8b), assigned to the characteristic W−Me fragment of [(
SiO) W(O)Me ], as observed in the closely related [{(t-
b‑1
[
(SiO) WOCl ] 1a preferentially adopts a square pyramidal
2
2
configuration (as in the case of 1b), with cis siloxy moieties.
The two DFT-calculated models (1aac‑1 and 1a ) illustrate the
intrinsic heterogeneity of the SiO2−200 surface, which comprises
unevenly spaced vicinal silanols: This translates into different
b‑1
2
2
3
4
Bu) SiO} W(O)Me ] complex (CS
: 47.0 ppm).
W‑Me
3
2
2
1
7
SiO-W-OSi angles, but the WO O NMR parameters (CS,
Moreover, a shoulder at 53 ppm is also seen, which may
arise from alkylation of monopodal species 1b. In addition,
peaks at −7 and −2 ppm are assigned to SnMe fragments on
C ) are not significantly affected by this geometrical variation.
Q
1
7
This emphasizes the interest of probing the O NMR
parameters of the oxo group to unveil the nature of the first
coordination sphere. From the best-fit simulation of the
33
silica, as previously documented. Formation of surface tin
methyl species is probably due to reaction of SnMe4 or
spectrum of WO*Cl -SiO2−200, one can also estimate a relative
SnMe Cl (formed during the alkylation) with the remaining
4
3
proportion of 35/65 for mono versus bigrafted species.
However, experimental uncertainties do not allow for non-
ambiguous quantification, when relating this distribution to that
obtained from elemental and mass-balance analyses, and from
EXAFS.
free silanol groups on the surface, as also observed by
DRIFT.
Further characterization of WOMe-SiO2−200 was carried out
3
3,35,36
by EXAFS. Clearly, when compared to WOCl -SiO2−200, the
4
W−Cl scattering pathway decreases after alkylation (Figure 9).
1
3
3
Reactivity of WOCl -SiO
with ( C-Labeled)
The parameters extracted from the fit of the k .χ(k) EXAFS
4
2−200
SnMe . As mentioned in the introduction, previous elements
spectrum (Table 3) are consistent with one oxo ligand at
1.70(1) Å, ca. two oxygen atoms at 1.88(1) Å and three carbon
atoms at 2.12(1) Å, respectively, assigned to siloxide and
methyl ligands, consistently with the bond distances found in
[(Silox) (n-Bu) WO] (1.705(2) Å for WO; 1.876 to 1.882 Å
4
indicate that use of tetra-alkyl tin derivatives is an efficient entry
into the generation of metathesis active species from tungsten
chloride precursors. We therefore attempted to generate surface
alkyl derivatives using readily available tetramethyl tin as an
alkylating agent. In order to ease spectroscopic investigations,
we prepared 13C-labeled SnMe , designated as SnMe *. The
2
2
3
4
for W−O; 2.098 to 2.154 Å for W−C) or in (O-
37
ArCH N(Me)CH Ar−O)(Me) WO (1.694−1.706 Å for
4
4
2
2
2
4
236
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
3
Figure 9. Tungsten L -edge k -weighted EXAFS (left), for WOMe-SiO
and corresponding Fourier transform (right; modulus and imaginary
III
2−200
part). Solid lines: experimental; dashed lines: spherical wave theory.
a
39
Table 3. EXAFS Parameters Obtained for WOMe-SiO2‑200
methyl ligand to generate a methylidene as active center.
Wolczanski described thermally stable complexes
2
2
type of neighbor
number of neighbors
distance (Å)
σ (Å )
34
[
(R SiO) WO R ] that comprise trans siloxide fragments,
3 2 2 2
W=O
1.0
1.70(1)
1.88(1)
2.12(2)
2.36(2)
2.91(3)
0.0017(8)
0.0014(5)
0.006(4)
0.002(2)
0.002(2)
in a distorted trigonal pyramid geometry. In comparison, the
peculiarity of the grafted system is the surface-enforced cis
configuration of the siloxy groups, as demonstrated above by
W-OSi
1.8(2)
b
W-CH3
1.9(4)
W-Cl
0.3(2)
1.0
17
O NMR combined DFT. This may have an impact on the
W--O(Si)2
formation of the carbene function by the α-H-transfer.
However, further mechanistic studies would be necessary to
elucidate this point.
The catalytic activity of WOMe-SiO2−200 was assessed in the
self-metathesis of isobutene into ethylene and 2,3-dimethylbu-
tenes (2,3-DMBs) (Scheme 2). The latter compounds are
a
The errors generated by the EXAFS fitting program “RoundMid-
−1
night” are indicated in parentheses. Δk: [2.2 − 18.0 Å ] - ΔR: [0.7 −
2
.9 Å] ([0.7 − 2.4 Å], when considering only the first coordination
2
sphere without Cl); S0 = 0.94; ΔE = 8.7 ± 2.0 eV (the same for all
0
2
shells); fit residue: ρ = 10.5%; quality factor: (Δχ) /ν = 3.81, with ν =
2
1
1/24 ([(Δχ) /ν] = 4.99 with ν = 11/19, considering only three
1
backscatterers, oxo, −O, and −C, in the first coordination sphere of
b
W). Shell constrained to the parameters above and below.
Scheme 2. Isobutene Self-Metathesis
WO; 1.887 to 1.919 Å for W−O; 2.170 to 2.191 Å for W−
C). The W−C distance obtained by EXAFS seems a bit short
compared to the last example; however, W(O)−Me distances
valuable additives for gasoline after a simple hydrogenation to
2,3-dimethylbutane with an impressive research octane number
(RON) of 103.5 and low RVP (Reid vapor pressure). 2,3-
38
as short as 2.052 Å have already been reported. Similar
2
40
parameters were obtained when fitting the k .χ(k) spectrum.
Furthermore, the fit was improved by considering a further
layer of Cl atom backscatterers at 2.37(2) Å (only ca. 0.3 such
neighbors compared to 2.3 in the starting material), suggesting
that the Cl/Me exchange may not be complete upon reaction
Dimethylbutenes are generally prepared by dimerization of
propylene on nickel catalysts such as in the IFP Difasol process,
which consist of a nickel salt in the presence of bulky basic
41
phosphine dissolved in ionic liquid. However, these processes
are dependent on propylene, which has recently met increasing
demand as starting material, resulting in a notable increase of its
with SnMe , and a layer of oxygen backscatterer at 2.91(4) Å
4
which would be due to oxygen of surface siloxanes of silica. The
results obtained are thus in good agreement with a Cl/Me
42
price. Consequently, the development of a new approach to
obtain 2,3-DMBs directly from alternative feeds, for example,
readily available isobutene, is highly desirable.
exchange after the reaction of WOCl -SiO
with SnMe4
4
2−200
leading mainly to [(SiO) WOMe ], with most probably [(
2
2
SiO)WOMe Cl] as minor species (ca. 20−30%). This result is
Isobutene self-metathesis is rather challenging due to the less
thermodynamically favored intermediate, as it bears gem-methyl
substituents in the [1,2]-positions, generating a sterically
encumbered intermediate. Productive metathesis of isobutene
is thus disfavored for steric reasons which accounts for the
relatively moderate activity observed in the conversion of
2
1
3
in accordance with C CP NMR features that indicate the
presence of minor monopodal W-Me containing species.
Catalytic Studies in Isobutene Metathesis. Exposure of
1
00 equiv of propene to WOMe-SiO2−200 at 150 °C in a batch
reactor for 2 h led to the formation of equilibrated mixture of
metathesis products: propene, ethene, and 2-butene (trans/cis
ratio of 1.9, Figure S8), along with 0.9 equiv of methane per
tungsten, in agreement with the formation of the expected
43
isobutene in comparison to linear alkenes. When isobutene
self-metathesis is performed by using WOMe-SiO2−200 in a
dynamic flow reactor (W = 7.10 wt %, T = 150 °C, P = 4 bar,
total flow rate 10 mL·min− (50% isobutene in Ar)), the
reaction proceeds with an initial maximal conversion of 9%
(thermodynamic equilibrium at 12.5%) before reaching a
pseudoplateau of 7% (Figure 10), affording a cumulated turn
over number of 680 after 70 h. The product selectivities at the
steady-state are branched hexenes (50%), ethene (49%) and
1
carbenic catalytic active species [(SiO)WO(CH )]. As
2
reported in the literature in the case of monopodal [(
SiO)(WO)Np ], attempts to observe alkylidene species by
3
heating (80 °C) prior to substrate addition were not
5
successful. The carbene ligand necessary for catalysis to
proceed is probably generated in situ by α-H-transfer from a
4
237
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
stability and selectivity toward ethylene and 2,3-dimethyl-2-
butene. The reaction proceeds most probably by generation of
bipodal tungsten oxo carbene species that resembles the
proposed model for the industrial metathesis catalyst WO3/
SiO . Further studies will be devoted to the selective
2
preparation and the characterization of such species and to
the assessment of its catalytic performances.
EXPERIMENTAL SECTION
■
All experiments were carried out by using standard Schlenk and
glovebox techniques. Solvents were purified and dried
according to standard procedures. C D (SDS) was distilled
6
6
over Na/benzophenone and stored over 3 Å molecular sieves.
1
7
O-Enriched [WOCl ] was prepared by using the reported
4
1
7
30
13
procedure and 90% O-labeled water. 100% C-labeled
Figure 10. Conversion of isobutene catalyzed by WOMe-SiO2−200
(
SnMe * was prepared from literature procedure, by alkylation
4
13
46
7.10 wt % W): (a) conversion of isobutene (■) and T.O.N (□).
of SnCl using CH MgCl in n-butylether.
4 3
Propene (Air Liquide, 99.5%) was dried and deoxygenated
before use by passing it through freshly regenerated molecular
diisobutenes (1%). The main hexenes component is 2,3-
dimethyl-2-butene (DMB-2) with high relative selectivity
sieves (3 Å) and R3−15 catalysts (BASF). Isobutene (Scott,
̅
(98%) obtained by self-metathesis of isobutene (Figure 11).
99%) was used as received. SiO
and SiO
were
2
‑(200)
2‑(700)
2
prepared from Aerosil silica (Degussa, specific area of 200 m
−
1
g ), which was partly dehydroxylated at 200 and 700 °C,
respectively, under high vacuum (10− Torr) for 15 h to give a
5
2
−1
white solid having a specific surface area of 190 m g and
containing 2.3 and 0.7 OH nm− respectively. Gas-phase
analyses were performed on a Hewlett-Packard 5890 series II
gas chromatograph equipped with a flame ionization detector
and an Al O /KCl on fused silica column (50 m × 0.32 mm).
2
2
3
Elemental analyses were performed at the Pascher Mikroana-
lytisches Labor at Remagen-Bandorf (Germany). IR spectra
were recorded on a Nicolet 6700 FT-IR spectrometer by using
a DRIFT cell equipped with CaF window. The samples were
2
prepared under argon within a glovebox. Typically, 64 scans
−1
were accumulated for each spectrum (resolution 4 cm ).
NMR Characterization. Solution NMR spectra were
recorded on an Avance-300 Bruker spectrometer. All chemical
1
13
shifts were measured relative to residual H or C resonances
1
in the deuterated solvent: C D , δ 7.15 ppm for H, 128 ppm
Figure 11. Relative selectivities in branched hexenes.
6
6
13
1
13
for C. H and C solid-state NMR spectra were recorded on
a Bruker Avance-500 spectrometer with a conventional double-
resonance 4 mm CP MAS probe at the Laboratoire de Chimie
Very low selectivity toward DMB-1 (2%) results from DMB-2
isomerization. The present catalyst provides a second example
for tungsten-based supported catalyst (apart from W−H/
Al O ) active in the highly challenging isobutene self-
́
Organometallique de Surface. The samples were introduced
under argon in a zirconia rotor (4 mm), which was then tightly
closed. In all experiments, the sample was spun at 10 kHz
unless otherwise specified. Chemical shifts were given with
2
3
4
4,45
metathesis.
1
13
respect to TMS as external reference for H and C NMR. The
O solid-state NMR spectra were acquired on a Bruker Avance
17
CONCLUSION
The grafting of [WOCl ] on SiO
■
1
17
17
under controlled
2−200
III 800 ( H, 800.13 MHz; O, 108.47 MHz) and 900 ( O,
122.11 MHz) spectrometers. The 17O MAS NMR spectra at
18.8 and 21.15 T were acquired using a single pulse sequence
(pulse excitation of 2 μs at an RF field strength of 60 kHz) at
spinning frequencies ranging from 17.6 to 20 kHz (3.2 mm
rotor diameter) to avoid overlapping of spinning sidebands
with CS resonances and to determine isotropic chemical shifts.
No proton decoupling was applied. The recycle delay was 1 s.
4
conditions was studied using experimental and theoretical
approach. Combination of spectroscopic evidence backed up by
DFT calculations demonstrated that the grafting results in
bipodal tungsten oxo bischloride surface species [(
SiO) WOCl ] as the major component, along with minor
2
2
monopodal tungsten oxo trischloride centers [(SiO)-
17
WOCl ]. O NMR demonstrated that both these centers
3
adopt a square pyramidal configuration, with the oxo ligand in
Preparation and Characterization of WOCl -SiO2−200.
4
apical position. Further treatment with SnMe results in Cl/Me
A mixture of finely ground [WOCl ] (160 mg, 0.47 mmol) and
4
4
exchange within the surface species, as shown mainly by EXAFS
SiO2−200 (1 g) was stirred at 80 °C (3h) under dynamic vacuum
while all volatile compounds were condensed into a cold trap.
Toluene was then added and the solid was washed 5 times. The
resulting powder was dried under vacuum (10− Torr). Analysis
1
13
and H and C MAS NMR. This material is active in alkene
metathesis, achieving efficient conversion of the challenging
isobutene substrate. Importantly, this catalyst shows high
5
4
238
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
2
by infrared spectroscopy of the condensed volatiles indicated
the formation of 674 μmol of HCl during the grafting for ca.
C = e qQ /h = eV Q /h, |V | ≥ |V | ≥ |V |
Q
33
33
22
11
(2)
(3)
395 μmol of grafted tungsten (1.7 HCl/W). Elemental analysis:
η = (V − V )/V , (1 ≥ η ≥ 0)
Q Q
11
22
33
W 7.26 wt %; Cl 3.13 wt %(2.2 Cl/W).
Preparation and Characterization of WO*Cl -SiO2−700.
4
ΔCSA = δ − (δ + δ )/2 = 3δ/2, |δ − δ |
1
7
zz
xx
yy
zz
CS
A mixture of finely ground 90% O-labeled [WO*Cl ] (120
4
mg, 0.35 mmol) and SiO2−700 (1 g) were stirred at 80 °C (3h)
under dynamic vacuum, while all volatile compounds were
condensed into a cold trap. Toluene was then added, and the
solid was washed 5 times. The resulting powder was dried
under vacuum (10 Torr). Analysis by infrared spectroscopy
of the condensed volatiles indicated the formation of 227 μmol
of HCl during the grafting for ca. 245 μmol of grafted tungsten
≥ |δ − δ | ≥ |δ − δ |
xx
CS
yy
CS
(4)
(5)
with the reduced anisotropy, δ = δ33 − δCS
ηCSA = (δ − δ )/(δ − δ ), (1 ≥ η ≥ 0)
−5
yy
xx
zz
CS
CSA
DFT Methodological Details. All DFT calculations were
48
performed with Gaussian 03. Calculations were carried out at
the DFT level of theory using the hybrid functional
(
(
0.9 HCl/W). Elemental analysis: W 4.5 wt %; Cl 2.71 wt %
3.1 Cl/W).
49−54
B3PW91.
Geometry optimizations were achieved without
Alkylation of WOCl -SiO
with SnMe . One gram of
any symmetry restriction. Calculations of vibrational frequen-
cies were systematically done in order to characterize the nature
of stationary points. Gibbs free energies were obtained at P = 1
atm and T = 298.15 K within the harmonic approximation for
frequencies. Stuttgart effective core potentials and their
4
2−200
4
material WOCl -SiO
was exposed to SnMe4 vapor
4
2−200
pressure at 80 °C in a batch reactor. After 12 h, all volatile
compounds were condensed into a NMR tube (fitted with a
Young valve) containing C D . The resulting powder was dried
6
6
−5
1
55
under vacuum (10 Torr). H MAS NMR (300 MHz) δ 0.1,
.2 ppm. 13C CP MAS NMR (75.4 MHz) δ 46, −2, −7 ppm.
Elemental analysis: W 7.10 wt %; C 1.20 wt %; Sn 0.7 wt % (ca.
associated basis set were used for silicon and tungsten. The
1
basis sets were augmented by a set of polarization functions (ζd
= 0.284 for Si and ζf = 0.823 for W). Hydrogen, chlorine, and
oxygen atoms were treated with 6-31G(d,p) double-ζ basis
2
.6 C/Wand 0.15 Sn/W).
Propene Metathesis. Twenty-five milligrams of WOMe-
56,57
17
sets.
The optimized structures were used for O NMR
calculations. These calculations were also performed using a
SiO2−200 was transferred to a 500 mL Schlenk flask in the
glovebox. About 100 equiv (with respect to tungsten) of
purified propene was then introduced into the flask and then
heated to 150 °C from room temperature (heating rate: 4 °C/
min). The gas-phase composition in the flask was determined
by GC (HP 5890, 30 m × 0.32 mm KCl/Al O column, FID).
Isobutene Metathesis. The evaluation of the activity of
the catalyst WOMe-SiO2−200 in isobutene metathesis has been
performed in a continuous flow reactor (Microactivity-
Reference, PID Eng&Tech). The catalytic test were performed
higher Dunning’s correlation consistent basis set cc-PVTZ for
5
8,59
the oxygen atoms.
In all cases, among the various theories
available to compute chemical shielding tensors, the Gauge
Including Atomic Orbital (GIAO) method has been adopted
6
0−65
for the numerous advantages it presents.
Typically, in
2
3
1
7
order to compare our calculations with experimental values, O
chemical shielding has been converted to chemical shift using
the usual equation: δ = σiso ref − σiso sample, where σ
is the
iso
iso ref
isotropic 17O chemical shielding of the liquid water. In the
7
−1
−1
continuation of our previous studies, an internal reference is
at 150 °C, with a flow rate of 10 mL·min (5 mL·min
−1
used for the calibration of the σ
The O quadrupolar coupling constant C and the asymmetry
value: σiso ref = 292.2 ppm.
iso ref
isobutene + 5 mL·min argon), pressure = 4 bar, and 0.25 g of
17
Q
catalyst). Isobutene was introduced by a high pressure syringe
PMHP50−500, Top Industrie). The products were analyzed
by online GC (Agilent 7890A) equipped with KCl/Al O
parameter η , which describes the interaction between nuclear
Q
(
quadrupolar moment of the oxygen nuclei with the electric field
gradient (EFG) arisen at these sites, are calculated from the
EFG tensor eigenvalues V , V , and V .
2
3
column and FID. The conversion and selectivity were
11
22
33
calculated from carbon numbers.
SIMPSON Simulation of 17O MAS NMR Spectra. All 17O
EXAFS. X-ray absorption spectra were acquired at the ESRF,
using the Swiss-Norwegian beamline BM01B (proposal no. 01-
01-905), at room temperature at the tungsten LIII edge, with a
double crystal Si(111) monochromator detuned 70% to reduce
the higher harmonics of the beam. The spectra were recorded
in the transmission mode between 9.98 and 11.45 keV. The
supported W sample was packaged within an argon-filled
glovebox in a double airtight sample holder equipped with
kapton windows. This type of cell has already been used and
proved to be very efficient for air-sensitive compounds such as
MAS NMR numerical spectra were calculated using gcompute
47
method implemented in SIMPSON software package. This
package described each line shape with nine NMR interaction
parameters, which included the isotropic chemical shift (δCS)
defined in eq 1, the quadrupolar coupling constant (C ), and
quadrupolar asymmetry parameter (η ) defined in eq 2 and eq
Q
Q
3
(
, the CSA (ΔCSA) and chemical shift asymmetry parameter
ηCSA) (in the Haeberlen convention) defined in eq 4 and eq 5,
and the Euler angles (α, β, γ). Here the Euler angles were kept
to zero due to the impossibility to assess them with the
available experimental data sets. Each numerical simulation was
performed with the zcw4180 crystallite file and 30 gamma
angles, with the calculated FID inclusive of all quadrupolar
satellites. A line broadening of 600, 800, and 1500 Hz was
[
(SiO)HfNp ] species supported onto Aerosil silica(800)
3
where the Hf−C contribution could be clearly distinguished
66
from Hf−O by EXAFS. The spectra analyzed were the results
of four such acquisitions, and no evolution could be observed
between the first and last acquisition. The data analyses were
performed by standard procedures using in particular the
67
applied for simulated spectra of [WO*Cl ], WO*Cl -SiO
4
4
2−700
program “Athena” and the EXAFS fitting program “Round-
Midnight”, from the “MAX” package, using spherical waves.
The program FEFF8 was used to calculate theoretical files for
phases and amplitudes based on model clusters of atoms. The
value of the scale factor, S = 0.94, was determined from the k
68
and WO*Cl -SiO2−200, respectively, with Gaussian/Lorentzian
4
ratio of 0.6.
69
δCS = (δ + δ + δ )/3
(1)
2
2
11
22
33
0
4
239
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
3
and k χ(k) spectra of a reference compound, a sample of
(9) Demmelmaier, C. A.; White, R. E.; van Bokhoven, J. A.; Scott, S.
L. J. Phys. Chem. C 2008, 112, 6439−6449.
[
WO(CH tBu) Cl] complex diluted in BN and carefully mixed
2 3
(10) Popoff, N.; Espinas, J.; Goure, E.; Boyron, O.; Le Roux, E.;
and pressed as a pellet (one oxo at 1.70(1) Å, three carbon
atoms at 2.10(1) Å and one chlorine at 2.43(1) Å in the firs7t
coordination sphere, with three carbon atoms at 3.28(3) Å).
The refinements were performed by fitting the structural
parameters N , R , σ and the energy shift, ΔE (the same for all
Basset, J. M.; Gauvin, R. M.; De Mallmann, A.; Taoufik, M. Macromol.
Rapid Commun. 2011, 32, 1921−1924.
(11) van Roosmalen, A. J.; Polder, K.; Mol, J. C. J. Mol. Catal. 1980,
8
, 185−190.
12) Tosin, G.; Delgado, M.; Baudouin, A.; Santini, C. C.; Bayard, F.;
Basset, J. M. Organometallics 2010, 29, 1312−1322.
13) Gomez, F. J.; Manak, M. S.; Abboud, K. A.; Wagener, K. B. J.
i
i
i
0
(
shells). The fit residue, ρ (%), was calculated by the following
formula:
(
3
3
2
Mol. Catal. A: Chem. 2000, 160, 145−156.
∑k [k χ (k) − k χ (k)]
exp
cal
2
(14) Lehtonen, A.; Sillanpaa, R. Polyhedron 2002, 21, 349−352.
(15) Lehtonen, A.; Sillanpaa, R. Inorg. Chem. 2004, 43, 6501−6506.
ρ =
*100
3
∑k [k χ (k)]
exp
(16) Millar, A. J.; White, J. M.; Doonan, C. J.; Young, C. G. Inorg.
Chem. 2000, 39, 5151−5155.
As recommended by the Standards and Criteria Committee
2
(17) Nayab, S.; Park, W.; Woo, H. Y.; Sung, I. K.; Hwang, W. S.; Lee,
H. Polyhedron 2012, 42, 102−109.
of the International XAFS Society, the quality factor, (Δχ) /ν,
where ν is the number of degrees of freedom in the signal, was
calculated, and its minimization considered in order to control
the number of variable parameters in the fits.
(18) Nugent, W. A.; Feldman, J.; Calabrese, J. C. J. Am. Chem. Soc.
1
(
2
(
995, 117, 8992−8998.
19) Redshaw, C.; Elsegood, M. R. J. Eur. J. Inorg. Chem. 2003,
071−2074.
20) Redshaw, C.; Humphrey, S. M. Polyhedron 2006, 25, 1946−
1954.
ASSOCIATED CONTENT
Supporting Information
■
*
S
(
66.
21) Hess, H.; Hartung, H. Z. Anorg. Allg. Chem. 1966, 344, 157−
Additional spectroscopic data, DFT calculations on the
mechanistic aspects of the grafting reaction, and coordinates
of DFT-optimized structures. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
(
22) Bassi, I. W.; Scordamaglia, R. J. Organomet. Chem. 1975, 99,
27−134.
23) Sergienko, V. S.; Abramenko, V. L.; Ilyukhin, A. B. Zh. Neorg.
1
(
Khim. 1997, 42, 945−951.
AUTHOR INFORMATION
Corresponding Authors
E-mail: mostafa.taoufik@univ-lyon1.fr.
E-mail: laurent.delevoye@ensc-lille.fr.
E-mail: regis.gauvin@ensc-lille.fr.
E-mail: laurent.maron@irsamc.ups-tlse.fr.
Notes
The authors declare no competing financial interest.
■
*
*
*
*
(24) Del Rosal, I.; Gerber, I. C.; Poteau, R.; Maron, L. J. Phys. Chem.
A 2010, 114, 6322−6330.
(25) Del Rosal, I.; Poteau, R.; Maron, L. Dalton Trans. 2011, 40,
11211−11227.
(26) Del Rosal, I.; Poteau, R.; Maron, L. Dalton Trans. 2011, 40,
1
(
1228−11240.
27) Del Rosal, I.; Tschan, M. J. L.; Gauvin, R. M.; Maron, L.;
Thomas, C. M. Polym. Chem. 2012, 3, 1730−1739.
28) Popoff, N.; Espinas, J.; Pelletier, J.; Macqueron, B.; Szeto, K. C.;
(
Boyron, O.; Boisson, C.; Del Rosal, I.; Maron, L.; De Mallmann, A.;
ACKNOWLEDGMENTS
We thank the CNRS, the French Ministry of Research and
Higher Education, and the Agence Nationale de la Recherche
■
Gauvin, R. M.; Taoufik, M. Chem. Eur. J. 2013, 19, 964−973.
(29) Merle, N.; Trebosc, J.; Baudouin, A.; Del Rosal, I.; Maron, L.;
Szeto, K.; Genelot, M.; Mortreux, A.; Taoufik, M.; Delevoye, L.;
Gauvin, R. M. J. Am. Chem. Soc. 2012, 134, 9263−9275.
(
ANR-12-BS07-0021-01, OXOCAT) for their generous sup-
(30) Gibson, V. C.; Kee, T. P.; Shaw, A. Polyhedron 1988, 7, 579−
port. Financial support from the TGIR RMN THC Fr3050 for
conducting the research is gratefully acknowledged. ESRF is
acknowledged for providing beamtime, as well as Prof. E. Le
Roux (University of Bergen, Norway) and Hermann Emerich
5
(
5
(
4
(
80.
31) Verdonck, L.; van der Kelen, G. P. J. Organomet. Chem. 1966, 5,
32−536.
32) Verdonck, L.; van der Kelen, G. P. J. Organomet. Chem. 1972,
0, 143−150.
33) Legagneux, N.; de Mallmann, A.; Grinenval, E.; Basset, J. M.;
(
ESRF) for their kind support during the XAS acquisition.
REFERENCES
Lefebvre, F. Inorg. Chem. 2009, 48, 8718−8722.
(34) Rosenfeld, D. C.; Kuiper, D. S.; Lobkovsky, E. B.; Wolczanski, P.
T. Polyhedron 2006, 25, 251−258.
(35) Corma, A.; Navarro, M. T.; Renz, M. J. Catal. 2003, 219, 242−
246.
■
(
1) Mol, J. C. J. Mol. Catal. A: Chem. 2004, 213, 39−45.
2) Grunert, W.; Feldhaus, R.; Anders, K.; Shpiro, E. S.; Minachev, K.
(
M. J. Catal. 1989, 120, 444−456.
3) Popoff, N.; Mazoyer, E.; Pelletier, J.; Gauvin, R. M.; Taoufik, M.
Chem. Soc. Rev. 2013, 42, 9035−9054.
4) Basset, J. M.; Psaro, R.; Roberto, D.; Ugo, R. Modern Surface
Organometallic Chemistry; Wiley-VCH: Weinheim, 2009.
5) Mazoyer, E.; Merle, N.; de Mallmann, A.; Basset, J.-M.; Berrier,
E.; Delevoye, L.; Paul, J.-F.; Nicholas, C. P.; Gauvin, R. M.; Taoufik,
M. Chem. Commun. 2010, 46, 8944−8946.
6) Conley, M. P.; Mougel, V.; Peryshkov, D. V.; Forrest, W. P.;
Gajan, D.; Lesage, A.; Emsley, L.; Coperet, C.; Schrock, R. R. J. Am.
Chem. Soc. 2013, 135, 19068−19070.
7) Merle, N.; Girard, G.; Popoff, N.; De Mallmann, A.; Bouhoute,
Y.; Trebosc, J.; Berrier, E.; Paul, J. F.; Nicholas, C. P.; Del Rosal, I.;
Maron, L.; Gauvin, R. M.; Delevoye, L.; Taoufik, M. Inorg. Chem.
013, 52, 10119−10130.
8) Rice, G. L.; Scott, S. L. Langmuir 1997, 13, 1545−1551.
(
(36) Samuel, P. P.; Shylesh, S.; Singh, A. P. J. Mol. Catal. A: Chem.
(
2007, 266, 11−20.
(37) Lehtonen, A.; Sillanpaa, R. J. Organomet. Chem. 2007, 692,
2361−2364.
(
(38) Stavropoulos, P.; Wilkinson, G.; Motevalli, M.; Hursthouse, M.
B. Polyhedron 1987, 6, 1081−1087.
(
(39) Schrock, R. R. Chem. Rev. 2002, 102, 145−179.
́
́
(40) Hancsok, J.; Magyar, S.; Nguyen, K. V. S.; Keresztury, L.; Valkai,
L. Pet. Coal 2003, 45, 99−104.
(
(41) Olivier-Bourbigou, H.; Magna, L.; Morvan, D. Appl. Catal., A
2010, 373, 1−56.
(42) Plotkin, J. S. Catal. Today 2005, 106, 10−14.
(43) Szeto, K. C.; Mazoyer, E.; Merle, N.; Norsic, S.; Basset, J. M.;
Nichoas, C. P.; Taoufik, M. ACS Catal. 2013, 3, 2162−2168.
2
(
4
240
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241

ACS Catalysis
Research Article
(
44) Garron, A.; Stoffelbach, F.; Merle, N.; Szeto, K. C.; Thivolle-
Cazat, J.; Basset, J. M.; Norsic, S.; Taoufik, M. Catal. Sci. Technol.
012, 2, 2453−2455.
45) Merle, N.; Stoffelbach, F.; Taoufik, M.; Le Roux, E.; Thivolle-
Cazat, J.; Basset, J. M. Chem. Commun. 2009, 2523−2525.
46) Edgell, W. F.; Ward, C. H. J. Am. Chem. Soc. 1954, 76, 1169−
169.
47) Bak, M.; Rasmussen, J. T.; Nielsen, N. C. J. Magn. Reson. 2000,
47, 296−330.
48) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
(69) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D. Phys.
Rev. B 1998, 58, 7565−7576.
2
(
(
1
(
1
(
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A.; Jr., V., T.; Kudin,
K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone,
V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene,
M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.;
Morokuma, K.; Voth, G. A.; Salvador, P.; Dannen-berg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.;
Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.;
Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.;
Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen,
W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian03, revision B.05;
Gaussian, Inc.: Wallingford, CT, 2003.
(
49) Burke, K.; Perdew, J. P.; Wang, Y.. Electronic Density
Functional Theory: Recent Progress and New Directions. InDerivation
of a Generalized Gradient Approximation: the PW91 Density
Functional;Dobson, J. F., Vignale, G., Mas, M. P., Eds.; Plenum:
New York, 1998.
(
50) Perdew, J. P. Electronic Structure of Solids. InUnified Theory of
Exchange and Correlation Beyond the Local Density Approximation;
Ziesche, P., Eschrig, H., Eds.; Akademie: Berlin, 1991.
(
51) Perdew, J. P.; Burke, K.; Wang, Y. Phys. Rev. B 1996, 54, 16533−
6539.
52) Perdew, J. P.; Burke, K.; Wang, Y. Phys. Rev. B 1998, 57, 14999−
4999.
53) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.;
Pederson, M. R.; Singh, D. J.; Fiolhais, C. Phys. Rev. B 1992, 46, 6671−
687.
54) Perdew, J. P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.;
Pederson, M. R.; Singh, D. J.; Fiolhais, C. Phys. Rev. B 1993, 48, 4978−
978.
55) Kuchle, W.; Dolg, M.; Stoll, H.; Preuss, H. Mol. Phys. 1991, 74,
245−1263.
56) Harihara, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213−
22.
57) Hehre, W. J.; Ditchfie, R; Pople, J. A. J. Chem. Phys. 1972, 56,
1
(
1
(
6
(
4
(
1
(
2
(
2
(
(
1
257−2261.
58) Davidson, E. R. Chem. Phys. Lett. 1996, 260, 514−518.
59) Woon, D. E.; Dunning, T. H. J. Chem. Phys. 1993, 98, 1358−
371.
(
(
60) Ditchfield, R. Mol. Phys. 1974, 27, 789−807.
61) Dodds, J. L.; McWeeny, R.; Sadlej, A. J. Mol. Phys. 1980, 41,
1
(
1
419−1430.
62) Junk, P. C.; Steed, J. W. J. Organomet. Chem. 1999, 587, 191−
94.
(
63) London, F. J. Phys. Radium 1937, 8, 397−409.
(
64) McWeeny, R. Phys. Rev. 1962, 126, 1028−1034.
(
65) Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990,
1
(
12, 8251−8260.
66) Tosin, G.; Santini, C. C.; Taoufik, M.; De Mallmann, A.; Basset,
J. M. Organometallics 2006, 25, 3324−3335.
(
(
67) Ravel, B.; Newville, M. J. Synchrotron Rad. 2005, 12, 537−541.
68) Michalowicz, A.; Moscovici, J.; Muller-Bouvet, D.; Provost, K. J.
Phys. Conf. Ser. 2009, 190, 012034.
4
241
dx.doi.org/10.1021/cs501294j | ACS Catal. 2014, 4, 4232−4241