WMe6 tamed by silica: Si-O-WMe5 as an efficient, well-defined species for alkane metathesis, leading to the observation of a supported W-methyl/methylidyne species

Article abstract of DOI:10.1021/ja410747g

The synthesis and full characterization of a well-defined silica-supported ≡Si-O-W(Me)₅ species is reported. Under an inert atmosphere, it is a stable material at moderate temperature, whereas the homoleptic parent complex decomposes above -20 °C, demonstrating the stabilizing effect of immobilization of the molecular complex. Above 70 °C the grafted complex converts into the two methylidyne surface complexes [(≡SiO-)W(≡CH) Me₂] and [(≡SiO-)₂W(≡CH)Me]. All of these silica-supported complexes are active precursors for propane metathesis reactions.

Full text of DOI:10.1021/ja410747g

Article
pubs.acs.org/JACS
WMe Tamed by Silica: Si−O−WMe as an Efficient, Well-Defined
6
5
Species for Alkane Metathesis, Leading to the Observation of a
Supported W−Methyl/Methylidyne Species
†
,†
†
‡
Manoja K. Samantaray, Emmanuel Callens,* Edy Abou-Hamad, Aaron J. Rossini,
‡
†
‡
,†
Cory M. Widdifield, Raju Dey, Lyndon Emsley, and Jean-Marie Basset*
†KAUST Catalysis Center (KCC), King Abdullah University of Science & Technology, 23955-6900 Thuwal, Saudi Arabia
‡Institut de Sciences Analytiques (CNRS, ENS-Lyon, UCB Lyon 1), Universite
rue de la Doua, 69100 Villeurbanne, France
́ ̀ ̀
de Lyon, Centre de RMN a Tres Hauts Champs, 5
*
S Supporting Information
ABSTRACT: The synthesis and full characterization of a
well-defined silica-supported Si−O−W(Me)5 species is
reported. Under an inert atmosphere, it is a stable material
at moderate temperature, whereas the homoleptic parent
complex decomposes above −20 °C, demonstrating the
stabilizing effect of immobilization of the molecular complex.
Above 70 °C the grafted complex converts into the two
methylidyne surface complexes [(SiO−)W(CH)Me ]
2
and [(SiO−) W(CH)Me]. All of these silica-supported
2
complexes are active precursors for propane metathesis
reactions.
INTRODUCTION
It is well-known that W/Ta alkylidene complexes discovered
■
12
13
by Wilkinson and later Schrock are key active catalysts in
olefin metathesis, which is one of the various steps occurring in
single-site alkane metathesis. Thus, the preparation of such
species as single sites on surfaces together with alkyl/hydride
moieties is of high interest for alkane metathesis. However, in
the past, several approaches to synthesize surface methylidene
Transformation of linear alkanes into their lower and higher
homologues via alkane metathesis is an important process in
the petrochemical industry. To date, two main families of
catalytic systems exist for alkane metathesis: (i) a dual-catalyst
system which relies on a dehydrogenation/hydrogenation
catalyst combined with an olefin metathesis catalyst and
ii) a “multifunctional” single-site catalyst supported on various
oxides which is able to achieve these three reactions. Since
the first disclosed silica-supported tantalum hydride, we have
reported various single-site supported catalysts for alkane
metathesis employing Ta and W polyhydrides directly linked to
silica, silica−alumina, and alumina.
1
2
,3
1
4,15
(
species have been used with little success.
Here we envisage
4
,5
that a polymethyl tungsten complex possessing no β-H would
be a suitable alternative candidate to the neopentyl ligand to
generate an in situ surface W−methylidene species in its highest
oxidation state.
5
4
,6,7
These catalysts have
Given that the most active supported catalysts for single-site
been successfully synthesized and characterized at the
molecular and atomic level. Most of them were found to
transform light alkanes into their lower and higher
0
(VI)
6
alkane metathesis are d W
complexes, we propose to
immobilize a well-defined homoleptic hexamethyltungsten
complex to assess if its transformation into a W−methylidene
would affect the catalytic performance of this alkane metathesis
process.
6
,8
homologues. In these instances, the first step of C−H
bond activation occurred on the metal hydride, and the
resulting alkyl species were assumed to undergo either a
process of α- or β-H elimination to give the corresponding
carbene or olefin, both of which are key intermediates for the
Here we use WMe , (1), initially discovered by Wilkinson, as
6
12
the precursor in this strategy. We report the preparation and
characterization, at the molecular level, of a well-defined
supported Si−O−W(Me)5 (2) (Scheme 1), its activity
toward alkane metathesis, and the isolation of a novel silica-
supported W−methyl/methylidyne species.
9
olefin metathesis process. Although the most active catalysts
are generated from surface metal hydrides, supported catalysts
which contain a neopentyl/neopentylidene moiety were also
8
found to be slightly active in alkane metathesis. It was
therefore assumed that an alkyl/hydride functional group is
needed to provide an alkylidene to convert alkene inter-
mediates via a metallacyclobutane.
Received: October 23, 2013
Published: December 26, 2013
1
0,11
©
2013 American Chemical Society
1054
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
Scheme 1. Synthesis of Supported Si−O−W(Me) (2) by
5
Reaction of 1 with Partially Dehydroxylated Silica
RESULTS AND DISCUSSION
■
Preparation and Characterization of Si−O−W(Me)
5
on SiO
by Whan,
. Grafting of 1 on silica has already been reported
2
‑700
6,17
1
though the system, in 1972, was poorly
Figure 1. (A) One-dimensional (1D) 1H MAS solid-state NMR
spectrum of 2 acquired at 600 MHz (14.1 T) with a 22 kHz MAS
frequency, a repetition delay of 5 s, and 8 scans. (B) Two-dimensional
characterized by today’s standards. In the following we re-
examine this step using the appropriate analytical tools of
modern surface organometallic chemistry (e.g., solid-state
1
1
(
2D) H− H double-quantum (DQ)/single-quantum (SQ) and (C)
18
NMR, IR, and elemental analysis).
A modified synthetic protocol was employed for the
1
1
H− H triple-quantum (TQ)/SQ NMR spectra of 2 (both acquired
with 32 scans per t increment, 5 s repetition delay, 32 individual t
increments). (D) C CP/MAS NMR spectrum of 2 acquired at 9.4 T
1
1
1
9
synthesis of 1. With freshly sublimed WCl in CH Cl as
13
6
2
2
1
the starting material, 3 equiv of Me Zn yielded the desired
(v ( H) = 400 MHz) with a 10 kHz MAS frequency, 1000 scans, a 4 s
2
0
complex 1 (12% yield). Solution NMR spectroscopy experi-
repetition delay, and a 2 ms contact time. An exponential line
broadening of 80 Hz was applied prior to Fourier transformation. (E)
1
13
1
13
ments ( H, C, and H− C HSQC) on the product in
1
13
2
9
D H− C CP/MAS dipolar HETCOR spectrum of 2 (acquired at
CD Cl are consistent with the formation of 1 and also agree
2
2
.4 T with an 8.5 kHz MAS frequency, 2000 scans per t increment, a
with previously reported spectroscopic data (see the Support-
1
19
4 s repetition delay, 64 individual t increments, and a 0.2 ms contact
1
ing Information). Next, the grafting of 1 was realized by
stirring a mixture of an excess of 1 and silica which had been
partially dehydroxylated at 700 °C (i.e., SiO2‑700, which contains
13
time). For all spectra depicted here, 2 was 95% C labeled.
0.3 ± 0.1 mmol of silanol groups/g) at −50 °C under an inert
indicate the presence of a signal at or near 0 ppm (in both
spectra), which would indicate methyl transfer to an adjacent
silicon atom of silica and hence the formation of the bipodal
atmosphere of argon. After several washing cycles with pentane
and drying under high vacuum, the resulting yellow powder 2
contains 3.5−3.9 wt % tungsten and 1.1−1.3 wt % carbon as
determined by elemental analysis (C/W ratio 5.0 ± 0.1, in
comparison to the expected value of 5).
species [(Si−O)
W(Me)
][Si-Me] (see Figure S6 in the
2
4
Supporting Information for an analogous example).
The anchoring of 1 on partially dehydroxylated silica would
substitute a strong σ-donor methyl group with a weaker σ-
donor siloxy group. Consequently, we should expect different
chemical shifts for W−methyl groups (trans or cis to the W−
An IR spectrum of 2 showed decreased intensity of the bands
−
1
at 3742 cm , which are associated with isolated and geminal
silanols. For 2, two new groups of bands in the 3014−2878 and
410 cm− regions were observed. These are assigned to
1
1
13
1
O−Si fragment). From the H and C NMR spectra
recorded at 298 K, only one signal was observed. We have
previously reported that grafted molecular species are highly
dynamic, and therefore it is possible that dynamic motion
results in the observation of a set of averaged chemical
ν(CH) and δ(CH) vibrations of the methyl ligands bonded to
tungsten (see the Supporting Information). Hydrogenolysis of
2
at 150 °C produced 5 equiv of CH per W atom. Mass
4
balancing and gas quantification are consistent with 2 being
2
2,23
1
13
assigned to Si−O−W(Me) .
shifts.
Thus, we recorded the H and C solid-state NMR
5
Further spectroscopic analyses of 2 were also conducted with
spectra of 2 at reduced temperatures.
1
Figure 2A,B shows the variable-temperature 13C CP/MAS
solid-state NMR. The H magic-angle spinning (MAS) solid-
1
state NMR spectrum of 2 displays one signal at 2.0 ppm
and H spin echo MAS solid-state NMR spectra of 2, which
13
(
Figure 1A) which autocorrelates in double-quantum (DQ)
had been isotopically enriched to 95% in C. At 301 K, a single
1
3
1
and triple-quantum (TQ) NMR experiments under 22 kHz
resonance is observed at 82.2 and 2.0 ppm in the C and H
NMR spectra, respectively. However, as the temperature of the
20
MAS as shown in Figure 1B,C, respectively. This strong
autocorrelation peak is attributed to the methyl groups (2.0
ppm chemical shift in the single-quantum frequency; 4.0 and
13
sample was reduced, the C resonance broadened until it splits
into two separate resonances centered at 94.2 and 74.1 ppm
with a 1.0:1.8 intensity ratio (Figure 2A). The variation in the
6
.0 ppm in indirect dimensions of the DQ and TQ spectra,
1
3
1
respectively). The C CP/MAS NMR spectrum shows a single
peak at 82 ppm (Figure 1D). This carbon resonance correlates
with the protons at a chemical shift of 2.0 ppm, as indicated in
H MAS NMR spectra when the temperature was decreased
was similar, and two distinct resonances centered at 2.6 and 1.7
1
13
ppm are observed at 196 K. Room-temperature H and
C
1
13
the 2D H− C HETCOR NMR spectrum recorded with a
chemical shifts are observed at the weighted average of the
respective low-temperature chemical shifts. Note that the peak
1
13
contact time of 0.2 ms (Figure 1E). The H and C chemical
shifts are similar to those observed in the solution NMR spectra
of molecular 1. Note that grafting of 1 on oxide supports could
result in the formation of monopodal or bipodal grafted species
due to strained silica ring defects produced after thermal
1
appearing at low temperature at −0.5 ppm in the H NMR
spectra presumably corresponds to residual methane on the
surface.
This clearly indicates that the observation of a single
resonance at room temperature results from a fast chemical
exchange (on the NMR time scale) between the two
2
1
1
13
dehydroxylation. H and C solid-state NMR spectroscopy
of a 13C-enriched sample of 2 (95% C labeled) did not
13
1
055
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
Figure 3. DFT optimized structural model for (A) 2 and (B) selected
local environment for the computational model of the bipodal
W(Me)4 species in 3. Tungsten−carbon and tungsten−oxygen
internuclear distances are indicated in angstroms. Calculated isotropic
1
13
H and C chemical shifts are provided in Table 1. Complete
molecular coordinates are specified in Tables S2 and S3 (Supporting
Information). For clarity, all hydrogen atoms have been omitted.
these distinct bonding environments, the H and 13C chemical
shifts for methyl groups 1, 2, 4 and 3, 5 are calculated to differ
by 0.82 and 18.4 ppm, respectively (Table 1), which is in very
1
Figure 2. Variable-temperature (A) 13C CP/MAS and (B) H spin
echo MAS solid-state NMR spectra of 2 with 95% 13C labeling of the
methyl groups at 16.4 T. Sample temperatures, isotropic chemical
shifts, and intensity scaling factors (which account for differences in
1
Table 1. DFT Calculated H and 13C Nuclear Magnetic
1
1
3
the number of scans) are indicated. A fit of the 196 K C CP/MAS
NMR spectrum is shown (red trace), and the integrated intensities of
Shieldings and Chemical Shifts for Selected Sites in Models
1
13
the two resonances are indicated. Two-dimensional H− C dipolar
HETCOR NMR spectra of 2 acquired at (C) 301 K with a contact
time of 0.75 ms and (D) 196 K with a contact time of 0.25 ms. Both
of 2 and the Supported W(Me) Species in 3
4
1
3
13
1
1
σ ( C)
δ ( C)
σ ( H)
δ ( H)
iso
iso
iso
iso
a
b
a
methyl group
(ppm)
(ppm)
(ppm)
(ppm)
HETCOR spectra were acquired with 16 scans per t increment, 80
1
individual t increments, and t incremented by 64 μs for each t point.
Model System 2
1
1
1
1
During t , e-DUMBO-1 homonuclear H decoupling was applied and
proton chemical shifts were corrected by dividing the apparent H
chemical shift dimension by a scaling factor of 0.57. The true sample
1
120.68
121.89
99.64
117.95
103.41
120.17
101.52
−
−
−
−
−
28.46
28.98
27.58
28.66
27.41
28.70
27.50
−
−
−
−
−
1
22
1
2
24
3
7
9
temperatures were calibrated by separately measuring the 5Br
4
2
isotropic chemical shifts and longitudinal relaxation times of KBr.
5
−
−
av 1, 2, 4
av 3, 5
Δδ(3,5−
60.0
78.4
18.4
2.18
3.00
0.82
nonequivalent types of methyl groups which possess a 2:3
1
13
stoichiometry. A low-temperature H− C dipolar HETCOR
1
,2,4)
NMR spectrum clearly shows that the less intense high-
Model System W(Me) in 3
1
13
4
frequency H and C resonances correlate (2.6 ppm, 94.2
ppm) while the more intense low-frequency resonances
correlate (1.7 ppm, 74.1 ppm) (Figure 2D).
1
2
3
4
108.19
−
−
86.8
27.70
−
−
3.23
105.62
92.98
27.55
27.17
28.47
27.91
In order to better understand the local structure of the
supported species in 2 and to explain the observed variable-
temperature NMR data, we performed DFT calculations. Prior
106.64
−
73.2
−
2.72
av 1, 2, 4
106.82
a
The 13C and H chemical shifts were obtained by calibrating shielding
1
to modeling 2, we validated the computational methods by5
values with calculations for other molecular tungsten species for which
chemical shifts have been previously reported (see the Supporting
Information for full details, including the resulting calibration curves).
calculating the structure of molecular 1, which exhibits sd
hybridization at the metal center, as demonstrated with valence
2
6,27
bond theory by Landis et al.
and molecular orbital and ab-
b
The isotropic shielding values of the three protons of each methyl
28
initio calculations by Eisenstein et al. Additionally, we
performed geometry optimizations and NMR chemical shift
calculations for several W molecular systems presented earlier
by Schrock and co-workers for which high-quality experimental
group have been averaged.
good agreement with the observed shift differences of 0.9 ppm
(2.6−1.7 ppm (Figure 2B)) and 20.1 ppm (94.2−74.1 ppm
(Figure 2A)) (see the Supporting Information for further
computational details). The geometry optimized structural
model for the supported species in 2 and the calculated
chemical shift differences are consistent with the low-temper-
ature NMR spectra and indicate that two resonances with a 2:3
intensity ratio between high- and low-frequency shifts are
2
9,30
solution NMR data were available.
Our present DFT
calculation predicts that WMe possesses a trigonal-prismatic
6
geometry (i.e., C3v symmetry), which is in reasonable
agreement with the W−C bond lengths and C−W−C bond
31
angles of the X-ray structure.
The DFT model of 2 also possesses a similar pseudo-
trigonal-prismatic arrangement of the ligands (Figure 3A). Two
of the methyl groups (labeled 3 and 5) are directed away from
the silica surface, while the other three methyl groups (labeled
1
13
expected in both the H and C NMR spectra. Therefore, the
1
13
single resonance observed in the room-temperature H and C
NMR spectra are indeed very likely the result of fast exchange
1, 2, and 4) are closer to the silica surface. As a consequence of
1
056
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
of the inequivalent methyl groups by rapid rotation (or
hopping) about the Si−O−W and different O−W−C bond
axes.
stoichiometry of the methyl groups in the two species, it is
possible to derive an overall ratio of around 2.7 pentamethyl
species for each tetramethyl species. Therefore, variable-
temperature C CP/MAS NMR in combination with quantum
chemical calculations allowed for the quantification of the
surface podality in the systems considered here.
13
Preparation and Characterization of Si−O−W(Me)
5
and (SiO−) W(Me) on SiO
. In addition, we examined
2‑200
2
4
the grafting of WMe on silica which had been partially
6
Evaluation of the Apparent Catalytic Activity of 2 and
for Propane Metathesis. After the successful synthesis and
dehydroxylated at 200 °C (SiO2‑200). Immobilizing an organo-
metallic species on less dehydroxylated silica leads frequently to
3
characterization of complex 2, we investigated its efficiency as a
catalyst precursor for alkane metathesis reactions. Up to now,
two supported catalyst systems were found to be able to
convert alkanes into higher and lower homologues: (i)
Scheme 2. Monopodal and Bipodal Variants of Supported
WMe6
supported metal hydrides MH (M = Ta, W; x = 1−3) and
x
(
(
ii) supported M(neopentyl) (alkylidene/alkylidyne) species
x
4
M = Ta, W, Mo; x = 1−3).
3
Although no catalysts containing only sp alkyl ligands have
been previously disclosed, we hypothesized that complex 2
would be an excellent candidate for the alkane metathesis
reaction. The intuitively easier loss of methane vs neopentane,
via the σ-bond metathesis step, potentially offers a significant
advantage when using catalyst 2 relative to a neopentyl-
containing catalyst.
2
1,32
a mixture of monopodal and bipodal species (Scheme 2).
1
3
C CP/MAS NMR spectra of 1 supported on silica treated at
00 °C (species 3) and 700 °C (species 2) both display similar
In our previous work, the propane metathesis reaction was
the standard catalytic reaction, and thus to compare the
catalytic activity of 2 with earlier results, the catalytic reaction
was conducted under the same reaction conditions (a batch
2
chemical shifts of the methyl groups attached to the W metal at
room temperature. This suggests that we could not distinguish
the monopodal species from the bipodal species of 3 at room
temperature using H/ C solid-state NMR (see Figure S6 in
the Supporting Information). Variable-temperature NMR was
once again used to identify the structural components of 3.
At room temperature, a single broad resonance is observed at
0.4 ppm in the C CP/MAS spectrum of the product
resulting from the grafting of WMe on SiO (3). Upon
cooling, the single resonance of 3 observed at 298 K splits into
two distinct resonances with approximately a 2:3 intensity ratio.
However, at 196 K, there is clearly an additional weak
10
reactor, 1 atm of propane, and over a 5 day period at 150 °C).
1
13
The experimental results confirm our hypothesis of increased
catalytic activity for 2 relative to the prior species. Indeed,
propane was successfully catalyzed when introducing 2 into the
reaction (127 TONs) and appears to compare favorably with
the previously reported inactive catalyst Si−O−W(CtBu)-
1
3
8
(CH
O−WH
tBu)
or the relatively much less reactive complex Si−
6
2‑200
2
2
34
(8 TONs). As anticipated, when 3 was used in the
x
reaction vessel, the propane metathesis reaction was less
efficient (47 TONs), in support of the notion that the higher
functional number of methyl groups on the silica surface
provides better activity (see Table 2).
“
shoulder” visible on the low-frequency peak (Figure S7,
Supporting Information). This feature becomes even more
pronounced at 150 K. This third resonance is expected to
correspond to a bis-grafted tetramethyl species, since it was not
observed in any of the NMR spectra of 2. DFT calculations
Table 2. Propane Metathesis Reaction: Activity (TON) and
Alkane Product Selectivities of W Catalyst Precursors 2 and
3 at 150 °C
using a model structure of the bipodal species (SiO−) W-
2
(
Me) were performed to confirm the assignment of the NMR
4
b
product selectivity (%)
spectrum (Figure 3B).
TON
The DFT model of the bipodal supported species in 3
indicates that three of the carbon nuclei possess magnetic
shielding values between 105.6 and 108.2 ppm and are likely
shifted to higher frequency relative to the analogous group of
three methyl carbons in 2 (average shifts are 73.2 ppm vs 60.0
ppm). The remaining fourth carbon nucleus is predicted to
possess a chemical shift that is 13.6 ppm higher than that for
catalyst
precursor
(conversion
a
c
d
(%))
methane ethane butanes pentanes

SiO2‑700
−
127 (12)
2
54
33/4
6/1
W(Me)5

SiO2‑200
−
47 (5)
7
56
22/2.5
9/2
W(Me)
x
a
TON is expressed in (mol of propane transformed)/(mol of W).
b
33
the other three methyl groups (86.8 ppm). The additional
The selectivities are defined as the amount of product over the total
experimentally observed shift at 75.5 ppm in the 150 K 13
C
c
amount of products. Ratio of linear and branched alkanes. C
/i-C
4
.
4
d
C /i-C .
NMR spectrum is therefore assigned to three of the carbons of
the bipodal species in 3 (Figure S7, Supporting Information).
We then assume that the fourth carbon of the bipodal species in
overlaps with the high-frequency shifts of the monopodal
species in 3 (which are centered at 91.6 ppm). From a fit of the
5
5
The alkane product distributions when using these two
different supported species in the reaction vessel are very
similar: the major alkane products are ethane and butanes, and
the minor products are methane and pentanes. These products
are produced since a [2 + 2] cycloaddition of propene with W−
alkylidenes would yield two different W−metallacyclobutanes
as intermediates. The steric interactions between positions
[1,2] and [1,3] of the substituents on the W−metal-
3
13
1
50 K C CP/MAS NMR spectrum of 3 acquired at 9.4 T, the
ratio of the pentamethyl to tetramethyl carbon nuclei was
found to be 0.77:0.23 (after subtracting 0.056 from the high-
frequency integral to account for overlap with the fourth carbon
of the monopodal species). After accounting for the
1
057
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
35
lacyclobutanes direct the alkene selectivity, which upon
hydrogenolysis yields the observed alkanes (see Scheme S1 in
35
the Supporting Information). The formation of branched
alkanes results from the competitive σ bond activation of CH2
versus CH groups of the propane, which is well-documented
3
4
in the literature.
NMR Studies of the Thermal Transformation of 2. The
above observations all suggest that the reaction proceeds
through a W−methylidene intermediate. In order to induce the
formation of this species, and in the hope of isolating the
methylidene, we therefore studied the thermal stability of 2 in
the absence of substrate directly by solid-state NMR.
Heating a supported sample of 2 which was enriched in 13C
(
>95%) from 298 to 345 K leads to the observation of several
new NMR signals. After maintaining the temperature at 345 K
for 12 h, most of the Si−O−W(Me) had converted. The
5
spectra of the converted material suggest that the products are
1
Figure 5. (A) 1D H spin−echo MAS solid-state NMR spectrum of
(SiO) W(CH)Me ] after maintaining the temperature of 95%
the W−methyl/methylidyne species 5 and 6 in Scheme 3.
[
x
y
1
3
C -labeled 2 at 345 K for 12 h (acquired on a 600 MHz NMR
Scheme 3. Formation of the W−Methylidyne Species upon
Heating 2
spectrometer at a 20 kHz MAS spinning frequency, number of scans 8,
repetition delay 5 s). (B) 2D 1H− H DQ and (C) H− H TQ
acquired on a 600 MHz NMR spectrometer at 22 kHz MAS spinning
frequency with a back-to-back recoupling sequence, number of scans
28, repetition delay 5 s, number of t increments 128, with the
1
1
1
(
1
1
1
3
increment set equal to one rotor period of 45.45 μs) (D) C CP/
MAS NMR spectrum (10 kHz MAS at the same field as above,
number of scans 20000, repetition delay 4 s, contact time 2 ms, line
broadening 80 Hz). (E) 2D CP/MAS HETCOR NMR spectrum
acquired with short contact times of 0.2 ms (8.5 kHz MAS, number of
1
scans per increment 4000, repetition delay 4 s, number of t
increments 32, line broadening 80 Hz).
1
In the converted material the H NMR spectrum (Figure 4)
exhibits four major new signals at 1.1, 1.4, 4.1, and 7.6 ppm.
(
peak at −12 ppm) (Figure S8, Supporting Information). The
signal at 2.0 ppm likely corresponds to unreacted silanol. The
13
C CP/MAS NMR spectrum (Figure 5D) displays three
signals at 40, 44, and 48 ppm, and at high frequency a signal at
1
13
2
98 ppm is observed. Additionally, the 2D H− C HETCOR
NMR spectrum (Figure 5E) with a short contact time (0.2 ms)
shows a correlation between the methyl protons (1.4 and 1.1
ppm) and these two carbon atoms (44 and 40 ppm),
respectively, and a correlation between the methyl protons
centered at 4.1 ppm with the carbon at 48 ppm allows the
assignment of the carbon−proton pairs to the individual methyl
groups. Furthermore, the strong correlation between the
carbon and proton signals at 298 and 7.6 ppm, respectively,
strongly supports the assignment of a methylidyne moiety
1
13
(
WCH) (Figure 5E). Chemical shift values for H and C in
1
Figure 4. H spin echo MAS solid-state NMR spectra of the thermal
this range are also reasonably consistent with DFT calculations
for a model silica surface-supported system that contains a W
CH functional group (see Figure S9 and Tables S5 and S6 in
the Supporting Information).
transformation of 2 (acquired on a 600 MHz NMR spectrometer with
20 kHz MAS frequency, number of scans 8, repetition delay 5 s). The
true sample temperatures were calibrated by separately measuring the
7
9
Br isotropic chemical shifts and longitudinal relaxation times of
Furthermore, a correlation in the DQ/SQ NMR correlation
25
KBr.
spectrum between the SiCH group at −0.5 ppm and the
3
methyl groups at 4.1 ppm supports transfer of a methyl group
The signals at 1.1, 1.4, and 4.1 ppm autocorrelate in 2D DQ
to the silica and suggests the formation of bipodal species 6
1
1
13
1
and TQ H− H homonuclear dipolar correlation spectra and
are assigned to different methyl groups (Figure 5B,C). The
proton resonance at 7.6 ppm displays no autocorrelation in the
DQ and TQ spectra (Figure 5B,C). The broad signal at −0.5
ppm is assigned to methane and methyl groups transferred to
silica (i.e., SiMe), which is supported by an autocorrelation
( C, 48 ppm; H, 4.1 ppm) (Scheme 3). Since no correlation
with the other two methyl groups is observed, these two
13
1
inequivalent methyl groups ( C, 44 and 40 ppm; H, 1.4 and
1.1 ppm) can be assigned to the monopodal species 5. The
methyl groups of both species 5 and 6 correlate with the
methylidyne moiety as observed in both DQ and TQ NMR
experiments (Figure 5B,C).
2
9
in DQ and TQ (Figure 5B,C) and also by Si CP/MAS NMR
1
058
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
Backing for the Existence of Transient Methylidene
Intermediates by Formation of W−Methyl/Methylidyne
Species. Together, these studies suggest that 2 evolves upon
thermal treatment into a mixture of unprecedented mono- and
bipodal W−methyl/methylidyne species. This plausibly sup-
ports the formation of the transient W−methylidene
intermediate 4 (Scheme 3). We have identified that the grafted
deactivated with KCl was used with helium as the carrier gas at 32.1
kPa. Each analysis was carried out under the same conditions: a flow
rate of 1.5 mL/min and an isotherm at 80 °C.
Liquid-State Nuclear Magnetic Resonance Spectroscopy. All
liquid-state NMR spectra were recorded on Bruker Avance 600 MHz
spectrometers. All chemical shifts were measured relative to the
1
13
residual H or C resonance in the deuteurated solvent: CD Cl , 5.32
2
2
1
13
ppm for H, 53.5 ppm for C.
Solid-State Nuclear Magnetic Resonance Spectroscopy.
WMe species can evolve into a W−methylidyne-containing
6
1
13
species, which would not be otherwise observable in a
One-dimensional H MAS and C CP/MAS solid state NMR spectra
were recorded on Bruker AVANCE III spectrometers operating at 400,
00, or 700 MHz resonance frequencies for H. Si NMR solid-state
36
comparable homogeneous system. These supported W−
methylidyne species 5 and 6 were also used as precursors for
propane metathesis and produced ethane and butane with
traces of methane and pentanes with a TON of 50 after 120 h
at 150 °C. We find that they are less active than the
pentamethyl compound 2. We very tentatively hypothesize that
this is again due to the presence of fewer methyl groups. If we
assume that the first step in the process was σ bond activation,
it would then be easier for species 2 to achieve this, in
comparison to species 5 or 6.
1
29
5
NMR was recorded using a 400 MHz Bruker AVANCE III
spectrometer. Experiments at 400 MHz employed a conventional
double-resonance 4 mm CP/MAS probe, while experiments at 700
MHz utilized a 3.2 mm HCN triple-resonance probe. For the 150 K
experiment on 3 at 400 MHz, a Bruker low-temperature 3.2 mm
double-resonance probe was employed. In all cases the samples were
packed into rotors under an inert atmosphere inside gloveboxes. Dry
nitrogen gas was utilized for sample spinning to prevent degradation of
the samples. NMR chemical shifts are reported with respect to the
external references TMS and adamantane. For 13C and Si CP/MAS
NMR experiments, the following sequence was used: 90° pulse on the
proton (pulse length 2.4 s), then a cross-polarization step with a
29
CONCLUSION
We described the grafting of WMe on variously dehydroxy-
lated silica (at 200 and 700 °C) surfaces using surface
organometallic strategies and tools. Notably, solid-state NMR
combined with computational modeling offers significant
support for the structure of a well-defined supported W
■
13
6
contact time of typically 2 ms, and finally acquisition of the C and
29
Si NMR signal under high-power proton decoupling. The delay
between the scans was set to 5 s to allow the complete relaxation of
1
the H nuclei, and the number of scans ranged between 3000 and 5000
for 13C and 30000 and 50000 for Si and was 32 for H. An
exponential apodization function corresponding to a line broadening
of 80 Hz was applied prior to Fourier transformation.
29
1
species, Si−O−WMe , a surface species that is much more
5
stable than the homoleptic parent complex in solution or in the
solid state. Furthermore, the grafting of this unique WMe6
homoleptic species allowed us to observe by solid state NMR
the temperature dependence of the methyl ligand fluxionality at
room temperature. Solid-state NMR has been used to
qualitatively determine the podality (i.e., monopodal vs
bipodal) of the grafted complex on silica. Then, thermal
1
13
The 2D H− C heteronuclear correlation (HETCOR) solid state
NMR spectroscopy experiments were conducted on a Bruker
AVANCE III spectrometer using a 3.2 mm MAS probe. The
experiments were performed according to the following scheme: 90°
13
13
proton pulse, t evolution period, CP to C, and detection of the
C
1
magnetization under TPPM decoupling. For the cross-polarization
step, a ramped radio frequency (RF) field centered at 75 kHz was
applied to the protons, while the 13C channel RF field was matched to
studies on Si−O−WMe (2) led us to isolate a novel
5
supported W−methylidyne/methyl complex, which was also
confirmed by experimental and theoretical studies. All of these
complexes are more active than the previously reported silica-
supported W complexes in alkane metathesis, with a TON of
obtain an optimal signal. A total of 32 t increments with 2000 scans
1
each were collected. The sample spinning frequency was 8.5 kHz.
Using a short contact time (0.5 ms) for the CP step, the polarization
transfer in the dipolar correlation experiment was verified to be
selective for the first coordination sphere about the tungsten: that is, to
1
27 at 150 °C for Si−O−WMe5.
1
13
lead to correlations only between pairs of attached H− C spins (C−
H directly bonded).
EXPERIMENTAL SECTION
General Procedure. All experiments were carried out by using
standard Schlenk and glovebox techniques under an inert nitrogen
atmosphere. The syntheses and the treatments of the surface species
were carried out using high-vacuum lines (<10−5 mbar) and glovebox
1
1
■
H− H Multiple-Quantum Spectroscopy. Two-dimensional
double-quantum (DQ) and triple-quantum (TQ) experiments were
recorded on a Bruker AVANCE III spectrometer operating at 600
MHz with a conventional double-resonance 3.2 mm CP/MAS probe,
according to the following general scheme: excitation of DQ
techniques. Pentane was distilled from a Na/K alloy under N and
coherences, t
recorded in a rotor synchronized fashion in t
increment equal to one rotor period (45.45 μs). One cycle of the
1
evolution, z filter, and detection. The spectra were
by setting the t
1
2
dichloromethane from CaH . Both solvents were degassed through
1
2
freeze−pump−thaw cycles. SiO2 and SiO2‑200 were prepared from
‑700
2
Aerosil silica from Degussa (specific area of 200 m /g), which were
standard back-to-back (BABA) recoupling sequences was used for the
3
7
partly dehydroxylated at either 700 or 200 °C under high vacuum
excitation and reconversion period. Quadrature detection in w was
1
−
5
(
<10 mbar) for 24 h to give a white solid having a specific surface
achieved using the States-TPPI method. An MAS frequency of 22 kHz
2
area of 190 m /g and containing respectively 0.5−0.7 and 2.4−2.6
OH/nm . Hydrogen and propane were dried and deoxygenated before
was used. The 90° proton pulse length was 2.5 μs, while a recycle delay
2
of 5 s was used. A total of 128 t increments with 32 scans per each
1
use by passage through a mixture of freshly regenerated molecular
increment were recorded. The DQ frequency in the w dimension
1
sieves (3 Å) and R3
̅
corresponds to the sum of two single-quantum (SQ) frequencies of
a Nicolet 6700 FT-IR spectrometer by using a DRIFT cell equipped
with CaF2 windows. The IR samples were prepared under argon
within a glovebox. Typically, 64 scans were accumulated for each
the two coupled protons and correlates in the w dimension with the
2
3
8
two corresponding proton resonances. The TQ frequency in the w1
dimension corresponds to the sum of the three SQ frequencies of the
spectrum (resolution 4 cm ). Elemental analyses were performed at
Mikroanalytisches Labor Pascher (Germany). Gas-phase analysis of
alkanes was performed using an Agilent 6850 gas chromatography
column with a split injector coupled with an FID (flame ionization
detector). An HP-PLOT Al O KCl 30 m × 0.53 mm, 20.00 mm
three coupled protons and correlates in the w dimension with the
2
three individual proton resonances. Conversely, groups of less than
three equivalent spins will not give rise to diagonal signals in this
spectrum.
Preparation of Hexamethyltungsten, WMe (1). The molec-
6
capillary column coated with a stationary phase of aluminum oxide
ular precursor WMe (1) was prepared from WCl and (CH ) Zn,
6
6
3 2
1
059
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
1
2,19,39
22
following the literature procedure.
To a mixture of WCl (1.80
previously optimized structure of Emsley and co-workers. The
6
g, 4.5 mmol) in dichloromethane (25 mL) was added (CH ) Zn (13.6
surface species was exchanged from that of the prior study with the
species of interest here, and the surface was truncated at the second
coordination sphere for the magnetic shielding tensor computation.
Where applicable, oxygen atoms from the silica surface were
terminated with hydrogen atoms. For all monopodal surfaces, the
silicon and oxygen atoms of the surface were fully frozen, while the
surface species of interest, as well as the H atoms terminating the silica
surface, were fully optimized at the level of theory specified earlier.
Optimized geometries and energies, as well as a more complete
disclosure of the computed magnetic shielding values, are provided in
the Supporting Information.
3
2
mmol, 1.0 M in heptane) at −80 °C, and after addition, the reaction
mixture was warmed to −35 °C and stirred at this temperature for
another 30 min. After successive filtrations with pentane and removal
of the solvent, the red solid 1 was obtained (0.16 g, 12%). Caution!
This 12e compound is highly unstable and is prone to violent
4
0
1
decomposition. H NMR (CD Cl , 600 MHz): δ (ppm) 1.65 (s,
2
2
1
3
1
8H, WCH ). C NMR (CD Cl , 150 MHz): δ (ppm) 82 (s, 6C,
3
2
2
1
83
13
W−
J
= 47 Hz, WCH ). HSQC confirms the correlation between the
C
3
1
13
H and C NMR signals.
The 13C-enriched W(CH ) was synthesized as described below:
3
6
1
3
13
13
C-enriched ( CH ) Zn was prepared from a suspension of CH Li
For models of bipodal species on SiO2‑200, the silica surface was
3
2
3
5
0
taken from Sautet and co-workers. The surface corresponds to a slice
along the [001] plane of β-cristobalite and was found to be a stable
phase of silica under the approximate conditions of the material under
study (i.e., 200 °C and near-ambient pressures). Taking the original
structure proposed by the authors (denoted as 001-4 in the original
manuscript), a small cluster was selected and terminated with
hydrogen atoms at all but two central positions. For these two
positions, the surface species was placed. All framework atoms, except
for the directly bound oxygen atoms, were held fixed, while the
molecule on the surface was subjected to a geometry optimization.
In comparison to the monopodal species, many computational
parameters remained identical (i.e., basis set, exchange-correlation
functional, ZORA), but they were supplemented by a dispersion term
(given by Grimme’s three-parameter description) due to the large
size of the cluster. The additional dispersion correction leads to a
slightly more favorable interaction between the molecule and the
surface. As before, optimized geometries and energies, as well as a
more complete disclosure of the computed magnetic shielding values,
are provided in the Supporting Information.
and ZnCl (2/1), with subsequent synthetic steps being analogous to
2
41
those provided above.
Preparation of WMe on SiO
(2). A solution of 1 in pentane
2‑700
6
(
150 mg, 1.2 equiv with respect to the amount of surface-accessible
silanols) was reacted with 1.8 g of Aerosil SiO2‑700 at −50 °C for 1 h,
was warmed to −30 °C, and was stirred for an additional 2 h. At the
end of the reaction, the resulting yellow solid was washed with pentane
−5
(
3 × 20 mL) and dried under dynamic vacuum (<10 Torr, 1 h). IR
1 1
−
data (cm ): 3742, 3014, 2981, 2946, 2878, 1410. H solid-state NMR
1
3
(
(
1
400 MHz): δ (ppm) 2.0 (W−CH ). C CP/MAS solid-state NMR
3
100 MHz): δ (ppm) 82.0 (W−CH ). Anal. Found: W, 3.5−3.9; C,
3
.1−1.3. The C/W ratio obtained was 5.0 ± 0.1 (5 was expected).
Preparation of 3. The same procedure above was carried out with
5
1
1
Aerosil SiO2‑200 dehydroxylated at 200 °C. H solid-state NMR (400
13
MHz): δ (ppm) 1.95 (W−CH ). C CP/MAS solid-state NMR (100
3
MHz): δ (ppm) 80.4 (W−CH ). Anal. Found: W, 3.49; C, 1.04. The
3
C/W ratio obtained was 4.6 ± 0.1 (4.7 was expected).
Synthesis of 5 and 6. In a glass reactor, 1.25 g of 2 was added and
heated at 100 °C (ramped at 60 °C/h) for 12 h to produce a dark gray
powder which was a mixture of the monopodal and bipodal species 5
−1
1
and 6. IR data (cm ): 3741, 2967, 2929, 2899. H solid-state NMR
ASSOCIATED CONTENT
■
(
400 MHz): δ (ppm) −0.5 (s, Si−CH ), 1.1 (s, W−CH ), 1.4 (s, W−
3
3
1
3
*
S
Supporting Information
CH ), 2.0 (s, SiOH), 4.1 (s, W−CH ), 7.6 (s, WCH). C CP/MAS
3
3
solid-state NMR (100 MHz): δ (ppm) 40 (s, W−CH ), 44 (s, W−
Figures and tables giving solution NMR spectra of 1, additional
3
29
1
CH ), 48 (s, W−CH ), 298 (s, WCH). Si CP/MAS solid-state
solid-state NMR spectra of 2 and 3, variable-temperature H,
3
3
13
29
NMR (80 MHz): δ (ppm) −12.3 (SiCH ), −100 (Q3), −109.7
3
C, and Si solid-state NMR spectra of 3, and information on
(
Q4). Anal. Found: W, 3.18; C, 0.6. The C/W ratio obtained was 2.9
optimized geometries and full disclosure of the calculated
magnetic shielding values for the computational models used.
This material is available free of charge via the Internet at
http://pubs.acs.org.
±
0.1 (3 was expected).
Procedure for the Quantification of Methane Released
during Hydrogenolysis. A sample of 2 (0.020 mmol/W, 100 mg)
and dry H (786 hPa) was added in a batch reactor of known volume
2
(
480 mL). The reaction mixture was heated to 130 °C for 10 h. Next,
an aliquot of the gas phase was released and analyzed by GC. Gas-
AUTHOR INFORMATION
■
phase analysis gave 0.098 mmol of CH , corresponding to a C/W ratio
of 4.9 ± 0.1 (5 was expected).
4
Corresponding Author
Typical Procedure for Propane Metathesis Reactions. A
mixture of a potential catalytic material (0.013 mmol/W) and dry
propane (980−1013 hPa) were heated to 150 °C in a batch reactor of
known volume (480 mL) over a 5 day period. At the end of the run, an
aliquot was drawn and analyzed by GC. The selectivities are defined as
the amount of a particular product molecule over the total amount of
products.
jeanmarie.basset@kaust.edu.sa (J.-M.B.); emmanuel.callens@
kaust.edu.sa (E.C.).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
Computational Details. All quantum chemical computations
The authors acknowledge the KAUST Nuclear Magnetic
Resonance Core Lab and technical assistance of Dr. Kazuo
Yamauchi. The authors also thank Dr. Nassima Riache for her
were performed using the Amsterdam Density Functional (ADF)
4
2,43
software, version 2010.02,
which is produced by Scientific
Computing & Modeling (SCM). For both the geometry optimization
1
13
and the magnetic shielding calculations ( H and C), the zeroth-order
̂
fruitful discussions. Mr. Benoit Vincenti is thanked for inert-
4
4−46
regular approximation (ZORA)
was used to include relativistic
atmosphere sample preparation prior to solid-state NMR
experiments upon 2. A.J.R. acknowledges support from an
EU Marie-Curie IIF fellowship (PIIF-GA-2010-274574).
C.M.W. acknowledges the Natural Sciences and Engineering
Council of Canada (NSERC) for a postdoctoral fellowship.
Computational resources for DFT calculations were provided
by the Pole Scientifique de Modelisation Numerique (PSMN).
̂ ́ ́
This work was supported by funds from King Abdullah
University of Science and Technology.
4
7
effects and included both scalar and spin−orbit contributions. All
calculations used an all-electron triple-ζ basis set which included
polarization functions (i.e., TZ2P) and the generalized gradient
approximation (GGA) exchange-correlation functional developed by
48,49
Perdew, Burke, and Ernzerhof (i.e., PBE).
Magnetic shielding
calculations were carried out using the “NMR” module which comes
bundled with the ADF software.
For models of monopodal species on SiO2‑700, the silica surface
selected for NMR property computations was taken from the
1
060
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061

Journal of the American Chemical Society
Article
much better platform on which to compare and contrast the two
structures. For example, the range of the carbon chemical shifts for the
bipodal species is reduced relative to the monopodal species, and in
addition the bipodal species would possess carbon chemical shifts
which are generally larger than those for the monopodal species.
REFERENCES
■
(
1) Basset, J. M.; Coper
́
et, C.; Soulivong, D.; Taoufik, M.; Thivolle-
Cazat, J. Angew. Chem., Int. Ed. 2006, 45, 6082−6085.
(
(
2) Burnett, R. L.; Hughes, T. R. J. Catal. 1973, 31, 55−64.
3) Haibach, M. C.; Kundu, S.; Brookhart, M.; Goldman, A. S. Acc.
(
́
34) Le Roux, E.; Taoufik, M.; Baudouin, A.; Coperet, C.; Thivolle-
Chem. Res. 2012, 45, 947−958.
4) Basset, J. M.; Coperet, C.; Soulivong, D.; Taoufik, M.; Cazat, J. T.
Acc. Chem. Res. 2010, 43, 323−334.
5) Vidal, V.; Theolier, A.; Thivolle-Cazat, J.; Basset, J. M. Science
997, 276, 99−102.
6) Le Roux, E.; Taoufik, M.; Coper
Thivolle-Cazat, J.; Basset, J. M.; Maunders, B. M.; Sunley, G. J. Angew.
Chem., Int. Ed. 2005, 44, 6755−6758.
7) Taoufik, M.; Le Roux, E.; Thivolle-Cazat, J.; Coperet, C.; Basset,
́
Cazat, J.; Basset, J. M.; Maunders, B. M.; Sunley, G. J. Adv. Synth.
Catal. 2007, 349, 231−237.
(
́
(35) (a) Le Roux, E.; Chabanas, M.; Baudouin, A.; de Mallmann, A.;
(
1
(
Coper
́
et, C.; Quadrelli, E. A.; Thivolle-Cazat, J.; Basset, J. M.; Lukens,
W.; Lesage, A.; Emsley, L.; Sunley, G. J. J. Am. Chem. Soc. 2004, 126,
́
et, C.; de Mallmann, A.;
1
1
(
3391−13399. (b) Leconte, M.; Basset, J. M. J. Am. Chem. Soc. 1979,
01, 7296−7302.
36) Chiu, K. W.; Jones, R. A.; Wilkinson, G.; Galas, A. M. R.;
(
Hursthouse, M. B.; Malik, K. M. A. J. Chem. Soc., Dalton Trans. 1981,
204−1211.
37) Sommer, W.; Gottwald, J.; Demco, D. E.; Spiess, H. W. J. Magn.
Reson. 1995, 113, 131−134.
38) Rataboul, F.; Baudouin, A.; Thieuleux, C.; Veyre, L.; Coper
J. M.; Maunders, B.; Sunley, G. J. Top. Catal. 2006, 40, 65−70.
8) Rascon, F.; Coperet, C. J. Organomet. Chem. 2011, 696, 4121−
131.
9) Chauvin, Y. Angew. Chem., Int. Ed. 2006, 45, 3740−3747.
10) Coperet, C. Chem. Rev. 2010, 110, 656−680.
11) Blanc, F.; Berthoud, R.; Coperet, C.; Lesage, A.; Emsley, L.;
1
(
(
4
(
(
́
(
́
et,
́
C.; Thivolle-Cazat, J.; Basset, J. M.; Lesage, A.; Emsley, L. J. Am. Chem.
Soc. 2004, 126, 12541−12550.
(
́
Singh, R.; Kreickmann, T.; Schrock, R. R. Proc. Natl. Acad. Sci. U.S.A.
008, 105, 12123−12127.
12) Shortland, A. J.; Wilkinson, G. J. Chem. Soc., Chem. Commun.
972, 318.
13) Schrock, R. R. J. Am. Chem. Soc. 1974, 96, 6796−6797.
14) Buffon, R.; Leconte, M.; Choplin, A.; Basset, J. M. J. Chem. Soc.,
(39) Shortland, A. J.; Wilkinson, G. J. Chem. Soc., Dalton Trans. 1973,
2
(
1
(
8
(
(
72−876.
40) Seppelt, K. Science 1996, 272, 182−183.
41) DuMez, D. D.; Mayer, J. M. J. Am. Chem. Soc. 1996, 118,
1
2416−12423.
42) te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Guerra, C. F.;
van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem.
001, 22, 931−967.
43) Guerra, C. F.; Snijders, J. G.; te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391−403.
44) van Lenthe, E.; Ehlers, A.; Baerends, E. J. J. Chem. Phys. 1999,
10, 8943−8953.
45) van Leeuwen, R.; Baerends, E. J. Int. J. Quantum Chem. 1994, 52,
11−730.
46) van Lenthe, E.; Baerends, E. J.; Snijders, J. G. J. Chem. Phys.
993, 99, 4597−4610.
47) van Lenthe, E.; Snijders, J. G.; Baerends, E. J. J. Chem. Phys.
996, 105, 6505−6516.
48) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78,
396.
49) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77,
865−3868.
50) Rozanska, X.; Delbecq, F.; Sautet, P. Phys. Chem. Chem. Phys.
010, 12, 14930−14940.
51) Grimme, S.; Antony, J.; Ehrlich, S.; Krieg, H. J. Chem. Phys.
010, 132, 154104−154123.
(
(
Dalton Trans. 1994, 1723−1729.
15) Le Roux, E.; Taoufik, M.; Chabanas, M.; Alcor, D.; Baudouin,
A.; Coperet, C.; Thivolle-Cazat, J.; Basset, J. M.; Lesage, A.; Hediger,
S.; Emsley, L. Organometallics 2005, 24, 4274−4279.
16) Smith, J.; Mowat, W.; Whan, D. A.; Ebsworth, E. A. V. J. Chem.
Soc., Dalton Trans. 1974, 1742−1746.
17) Mowat, W.; Smith, J.; Whan, D. A. J. Chem. Soc., Chem.
Commun. 1974, 34−35.
18) Coperet, C.; Chabanas, M.; Saint-Arroman, R. P.; Basset, J. M.
Angew. Chem., Int. Edit. 2003, 42, 156−181.
19) Kleinhenz, S.; Pfennig, V.; Seppelt, K. Chem. Eur. J. 1998, 4,
687−1691.
20) Geen, H.; Titman, J. J.; Gottwald, J.; Spiess, H. W. Chem. Phys.
Lett. 1994, 227, 79−86.
21) Fleischman, S. D.; Scott, S. L. J. Am. Chem. Soc. 2011, 133,
847−4855.
22) Blanc, F.; Basset, J. M.; Coper
(
2
(
́
(
(
1
(
7
(
1
(
1
(
1
(
3
(
2
(
2
(
(
́
(
1
(
(
4
(
́
et, C.; Sinha, A.; Tonzetich, Z. J.;
Schrock, R. R.; Solans-Monfort, X.; Clot, E.; Eisenstein, O.; Lesage, A.;
Emsley, L. J. Am. Chem. Soc. 2008, 130, 5886−5900.
23) Chen, Y.; Callens, E.; Abou-Hamad, E.; Merle, N.; White, A. J.
P.; Taoufik, M.; Coper
Int. Ed. 2012, 51, 11886−11889.
24) Elena, B.; de Paepe, G.; Emsley, L. Chem. Phys. Lett. 2004, 398,
(
́
et, C.; Le Roux, E.; Basset, J. M. Angew. Chem.,
(
5
(
(
32−538.
25) Thurber, K. R.; Tycko, R. J. Magn. Reson. 2009, 196, 84−87.
26) Landis, C. R.; Cleveland, T.; Firman, T. K. J. Am. Chem. Soc.
1
995, 117, 1859−1860.
27) Landis, C. R.; Cleveland, T.; Firman, T. K. Science 1996, 272,
82.
28) (a) Demolliens, A.; Jean, Y.; Eisenstein, O. Organometallics
986, 5, 1457−1464. (b) Kang, S. K.; Albright, T. A.; Eisenstein, O.
Inorg. Chem. 1989, 28, 1611−1613.
29) Freudenberger, J. H.; Schrock, R. R. Organometallics 1985, 4,
937−1944.
30) Kreickmann, T.; Arndt, S.; Schrock, R. R.; Muller, P.
(
1
(
1
(
1
(
Organometallics 2007, 26, 5702−5711.
(
(
(
31) Pfennig, V.; Seppelt, K. Science 1996, 271, 626−628.
32) Gajan, D.; Coperet, C. New J. Chem. 2011, 35, 2403−2408.
33) It is important to point out here that it is difficult to compare
́
the calculated magnetic shielding values for the monopodal and
bipodal structural models on an absolute basis, as the surfaces used in
the models are very different. With this being said, the differences (i.e.,
using a relative basis) between the magnetic shielding values offer a
1
061
dx.doi.org/10.1021/ja410747g | J. Am. Chem. Soc. 2014, 136, 1054−1061