Bulky aryloxide ligand stabilizes a heterogeneous metathesis catalyst

Article abstract of DOI:10.1002/anie.201408880

The reaction of [W(=O)(=CHCMe₂Ph)(dAdPO)₂], containing bulky 2,6-diadamantyl aryloxide ligands, with partially dehydroxylated silica selectively yields a well-defined silica-supported alkylidene complex, [(≡SiO)W(=O)(= CHCMe₂Ph)(dAdPO)]. This fully characterized material is a very active and stable alkene metathesis catalyst, thus allowing loadings as low as 50 ppm in the metathesis of internal alkenes. [(≡SiO)W(=O)(=CHCMe₂Ph)(dAdPO)] also efficiently catalyzes the homocoupling of terminal alkenes, with turnover numbers exceeding 75 000 when ethylene is constantly removed to avoid the formation of the less reactive square-based pyramidal metallacycle resting state.

Full text of DOI:10.1002/anie.201408880

Angewandte
Chemie
DOI: 10.1002/anie.201408880
Supported Catalysts
Bulky Aryloxide Ligand Stabilizes a Heterogeneous Metathesis
Catalyst**
Matthew P. Conley, William P. Forrest, Victor Mougel, Christophe Copꢀret,* and
Richard R. Schrock*
Abstract: The reaction of [W(=O)(=CHCMe Ph)(dAdPO) ],
2
2
containing bulky 2,6-diadamantyl aryloxide ligands, with
partially dehydroxylated silica selectively yields a well-defined
silica-supported alkylidene complex, [(ꢀSiO)W(=O)(=
CHCMe Ph)(dAdPO)]. This fully characterized material is
2
a very active and stable alkene metathesis catalyst, thus
allowing loadings as low as 50 ppm in the metathesis of
internal alkenes. [(ꢀSiO)W(=O)(=CHCMe Ph)(dAdPO)]
2
also efficiently catalyzes the homocoupling of terminal alkenes,
with turnover numbers exceeding 75000 when ethylene is
constantly removed to avoid the formation of the less reactive
square-based pyramidal metallacycle resting state.
Scheme 1. Proposed active sites in WO /SiO (a) and the correspond-
ing well-defined surface (b) and molecular species (c).
3
2
T
he synthesis and design of metal alkylidene complexes over
From a molecular perspective, preparation of W-oxo
alkylidenes was also a challenge. Osborn and co-workers
described a [W(=O)(CH tBu) X] (X = Cl, Br, OR) which
the past 40 years led to highly active alkene metathesis
catalysts which established metathesis as a proven method in
2
3
[
1]
organic and polymer syntheses. In contrast, the classical
heterogeneous metathesis catalysts, oxide-supported MOx
lacks alkylidenes and does not have metathesis activity in the
[
6]
absence of cocatalysts. Supporting [W(=O)(CH tBu) X] or
2
3
(
MOx = metal oxide; M = Mo, W and Re), are still restricted
[W(=O)(CH tBu) ] on silica gives metathesis-active materi-
2 4
to a narrow substrate class, suffer from low activity and/or
stability, and require high operating temperatures (typically
als, though only at elevated temperatures, and a supported
alkylidene was not observed. The first four-coordinate
[
2]
[7]
above 1508C for Mo and 4008C for W). One exception is
W-oxo alkylidene complexes only recently became available
through the introduction of 2,6-disubstituted aryloxide
ligands (Scheme 1c), and show catalytic activity in solution
[3]
Re O /Al O . These catalysts are generally prepared by the
2
7
2
3
dispersion of metal salts onto an oxide support followed by
high-temperature calcination. The catalytically active alkyli-
dene species is generated in situ under reaction conditions at
the high operating temperatures required for Mo and W. The
working active sites have never been observed for these
systems, but are commonly assumed to be an oxo alkylidene
[8]
at 228C.
By immobilizing molecular metathesis catalysts onto
[
9]
partially dehydroxylated silica supports
we recently
reported the formation of the supported W-oxo alkylidene
[(ꢀSiO)W(=O)(=CHtBu)(OHMT)] (Scheme 1b; R = mesi-
[4]
species, as shown in Scheme 1a for the WO /SiO catalysts.
tyl, OHMT= 2,6-dimesitylphenoxide), which has higher
3
2
Though the active sites are unknown, the WO /SiO catalyst
activity than the molecular catalyst under the same reaction
3
2
À1 [10]
emerged as an effective alkene metathesis catalyst for the
conditions (TOF = 300 versus 3 min ).
Although
[5]
production of propene by the ethenolysis of 2-butenes.
[(ꢀSiO)W(=O)(=CHtBu)(OHMT)] had catalytic activity
higher than that of other well-defined silica-supported
[
11]
[
+]
[+]
imido complexes,
this catalyst was contaminated with
[
*] Dr. M. P. Conley, Dr. V. Mougel, Prof. Dr. C. Copꢀret
Department of Chemistry and Applied Biosciences
ETH-Zꢁrich
approximately 20% of the unreactive [(ꢀSiO)W(=O)-
(CH tBu)(OHMT)] species, which forms during grafting by
2
Vladimir Prelog Weg 2, CH-8093, Zꢁrich (Switzerland)
E-mail: ccoperet@inorg.chem.ethz.ch
protonation of the alkylidene. In addition, this material also
showed signs of deactivation at low catalyst loading. Herein
[
+]
Dr. W. P. Forrest, Prof. Dr. R. R. Schrock
Department of Chemistry 6-331
we report that incorporation of the very bulky bis(2,6-
[12]
diadamantyl-4-methylphenoxide) (dAdPO)
ligand gives
Massachusetts Institute of Technology
the selective formation of [(ꢀSiO)W(=O)(=CHCMe R)-
2
77 Massachusetts Ave., Cambridge, Massachusetts, 02139 (USA)
(dAdPO)], a highly active and stable catalyst which operates
E-mail: rrs@mit.edu
+
at loadings as low as 50 ppm for internal alkene metathesis
and gives exceptionally high turnover in the homocoupling of
terminal alkene (> 75000).
[
] These authors contributed equally to this work.
[
**] R.R.S. acknowledges the NSF (CHE-1111133) for support of this
research. V.M. was supported by an ETH fellowship (cofunded by
ETH Zurich-Marie Curie Action for People, FEL-08 12-2)
Reacting [(dAdPO) W(=O)(=CHCMe Ph)] (1a) with
2
2
[
13]
silica partially dehydroxylated at 7008C ([SiO2-700]) results
in 0.071 mmol of dAdPOH released per gram of silica during
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201408880.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
À1
13
grafting and a solid with 1.37 wt% W (0.074 mmolWg ),
thus indicating that roughly 27% of the surface silanols
reacted to generate the corresponding monografted species.
The tungsten loading is slightly lower than we obtained for
signals at d = 1.9 and appears at 1.1 ppm. The C CPMAS
spectrum of [(ꢀSiO)W(=O)(=C*HtBu)(dAdPO)] contains
signals for the dAdPO ligand at similar chemical shifts as
for [(ꢀSiO)W(=O)(=CHCMe Ph)(dAdPO)], and also con-
2
[
(ꢀSiO)W(=O)(=CHtBu)(OHMT)], probably because of the
tains the characteristic alkylidene signal at d = 260 ppm. In
larger dAdPO ligand, which is consistent with the much larger
buried volume of the dAdPO ligand (36.8%) compared to
HMTO (24.3%; see Table S1 in the Supporting Informa-
addition, small amounts of [(ꢀSiO)W(=O)(C*H tBu)-
2
(dAdPO) ] are formed as evidenced by the presence of
2
a signal at d = 68 ppm, and presumably results from addition
[
14]
1
tion).
The infrared spectrum of this material contains
of the silanol onto the smaller neopentylidene ligand. The H–
À1
13
[16]
a band at 3624 cm , thus indicating that some surface silanols
C HETCOR spectrum
(dAdPO)] is shown in Figure 1 and contains a crosspeak at
of [(ꢀSiO)W(=O)(=C*HtBu)-
[
11b]
are interacting with the aromatic ligand,
and that there is
also a significant fraction of unreacted isolated silanols,
thereby suggesting that the rather bulky [(dAdPO) W(=O)(=
2
[15]
CHCMe Ph)] complex cannot access some surface silanols.
2
Carbon elemental analysis shows the presence of 3.45 wt%,
which corresponds to 39 Æ 1 C/W, which is in very close
agreement with the expected value (37 C/W) for the selective
formation of [(ꢀSiO)W(=O)(=CHCMe Ph)(dAdPO)] (2a) as
2
shown in Scheme 2. These data are in contrast with those of
the the corresponding HMTO system.
1
13
Figure 1. Two-dimensional H– C HETCOR spectrum of 2b recorded
with a CP contact time of 600 ms. Other experimental details are given
1
3
in the Supporting Information. The one-dimensional C CPMAS spec-
trum is shown above the two-dimensional plots together with the
assignment of the resonances.
Scheme 2. Selective formation of 2a by grafting 1a on silica partially
dehydroxylated at 7008C.
1
The H magic angle spinning (MAS) NMR spectrum of
d = 9.3 ppm which correlates with the alkylidene signal. These
results are in excellent agreement with the solution data of
1b*.
[
(ꢀSiO)W(=O)(=CHCMe Ph)(dAdPO)] contains signals at
2
d = 1.9 ppm, which are assigned to the adamantyl fragments
and the methyl groups of the neophylidene, and at d =
We evaluated the catalytic activity of [(ꢀSiO)W(=O)(=
7
.1 ppm, which represents the phenyl group of the dAdPO
CHMe Ph)(dAdPO)] (2a) in cis-4-nonene metathesis at 308C
2
and neophylidene ligands. The spectrum also contains a broad
resonance centered at d = 9.1 ppm, which is assigned to the
alkylidene proton and is in good agreement with the NMR
(Table 1): 2a converts 5000 equivalents of cis-4-nonene into
Table 1: Catalytic data for [(ꢀSiO)W(=O)(=CHCMe Ph)(dAdPO)].
1
3
2
spectrum of 1a in solution. The C cross polarization magic
angle spinning (CPMAS) NMR spectrum of [(ꢀSiO)W(=
[
b]
Substrate
mol% TOF
Time to equilibrium conversion
O)(=CHCMe Ph)(dAdPO)] contains signals at d = 19, 30, 36,
2
cis-4-nonene
cis-4-nonene
ethyl oleate
0.02
0.005
0.05
0.1
325 (22%)
356 (5%)
133 (15%)
5 (2%)
10 min
5 h
ca. 6 h
4
2, 126, and 140 ppm from the dAdPO ligand and neo-
phylidene groups. The alkylidene signal was not observed,
and is typical of supported alkylidenes at natural abundance.
[
a]
1-nonene
42% conversion after 24 h
1
3
We prepared the corresponding C-labeled neopentylidene
[a] Run in a flow reactor. [b] Turnover frequency (TOF) at 3 min, given in
À1
equivalent [W(=O)(=C*HtBu)(dAdPO) ] (1b*), with 50%
min with the corresponding conversions given in brackets. [c] Pro-
2
1
3
ductive turnover numbers.
C enrichment at the alkylidene carbon atom, and grafted
this compound onto [SiO2-700]. Similar to the grafting of 1,
about 24% of the surface silanols react under grafting
conditions as determined from mass balance analysis. The
80:20 E/Z mixtures of 4-octenes and 5-decenes in 10 minutes.
At lower catalyst loading (0.005 mol%, 50 ppm), the sub-
1
H MAS NMR spectrum of [(ꢀSiO)W(=O)(=C*HtBu)-
(
dAdPO)] (2b*) is similar to that of [(ꢀSiO)W(=O)(= strate is fully converted in 5 hours, and the initial turnover
À1
CHCMe Ph)(dAdPO)], though in this case the tert-butyl
frequency (TOF) is about 350 min . Under these reaction
2
neopentylidene resonance is separated from the adamantyl
conditions the molecular 1a shows less than 1% conversion of
2
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
cis-4-nonene after 24 hours, probably because of the presence
a frit was loaded with the supported catalyst and connected to
[12]
of two bulky dAdPO ligands. Replacing one of them by the
locally small surface siloxy ligand (with 20.6% buried volume
a flask containing 1-hexene. The alkene was refluxed at 658C
under a slow flow of argon and distilled over the catalyst bed
which allowed the liquid to pass through the catalyst bed and
the frit, and ethylene to be vented. Under these reaction
conditions, we loaded this reactor with 1-hexene several times
over 70 hours and obtained 76300 turnovers. This finding
illustrates the stability of the catalyst under conditions that
allow for the removal of ethylene, which is known to
[
11c]
using 111 b-cristobalite as a model)
dramatically increases
the reactivity of the metal alkylidene, and the catalyst stability
[
17]
is increased through surface-site isolation. We also found
that ethyl oleate, a substrate incompatible with classical
heterogeneous catalysts is fully converted to equilibrium in
À1
less than 6 hours with an initial TOF of 133 min with
[19]
0
.05 mol% catalyst. The initial TOF values obtained with 2a
deactivate some metathesis catalysts.
are slightly higher than observed for the corresponding
HMTO analogue, probably because of the cleaner grafting
of 1a on the silica surface. In addition, 2a is less prone to
deactivation, as illustrated by the much shorter equilibration
time measured for cis-4-nonene metathesis (10 min versus
Overall, we described the synthesis and the character-
ization of a well-defined silica-supported W-oxo-based meta-
thesis catalysts incorporating the bulkier dAdPO ligand. This
ligand improves the stability of the catalyst, thus allowing
a decrease in catalyst loading, down to 50 ppm. This hetero-
geneous catalyst is also compatible with alkenes bearing an
ester functionality, that is, ethyl oleate, in contrast to the
classical tungsten catalysts. The high stability of this system
allows the efficient homocoupling of terminal alkenes which
are a challenging substrate class because it yields ethylene as
a co-product, which tends to deactivate—reversibly or not—
metathesis catalysts. In this case, ethylene forms a very stable
unsubstituted metallacyclobutane which is not prone to
cycloreversion, hence lower metathesis rates and reduced
catalyst performance can be overcome by continuous ethyl-
ene venting under flow conditions, thus allowing very high
turnover numbers (> 75000) to be reached. This study shows
that combining a very bulky molecular ligand (here AdPO)
and a small surface siloxy group provides an optimal ligand
set to achieve both high reactivity and stability. We are
exploring this concept to design more efficient heterogeneous
catalysts.
6
0 min for [(ꢀSiO)W(=O)(=CHtBu)(OHMT)]).
In sharp contrast, 1-nonene gave sluggish results with only
4
5
2% conversion after 24 hours and a low initial TOF of
À1
min
even with significantly higher catalyst loading
(
0.1 mol%). We reasoned that this stark difference in activity
between internal and terminal alkenes was due to ethylene
build-up in the reaction mixture, thus leading to stable
[12]
metallacyclobutanes and/or irreversible deactivation.
To
test this hypothesis, we generated the stable metallacycle
[
2
(ꢀSiO)W(=O)(*CH *CH *CH )(dAdPO)] (3) by reacting
2
2
2
1
3
a with excess bis( C-labeled) ethylene (10 equiv/W) and it
1
3
resulted in the release 0.9 equivalents of mono- C-labelled 3-
methyl-3-phenyl-1-butene per tungsten (Scheme 3). The
Received: September 8, 2014
Published online: && &&, &&&&
Keywords: alkenes · metathesis · metallacycles ·
.
NMR spectroscopy · tungsten
Scheme 3. Formation of the parent metallacyclobutane 3 by reaction of
1
3
13
2
a with C-labeled ethylene. *= C.
1
3
[1] a) A. H. Hoveyda, A. R. Zhugralin, Nature 2007, 450, 243 – 251;
b) A. H. Hoveyda, J. Org. Chem. 2014, 79, 4763 – 4792; c) Hand-
book of Olefin Metathesis, Vol. 2 (Ed.: R. H. Grubbs), Wiley-
VCH, Weinheim, 2003.
[2] R. L. Banks, G. C. Bailey, Ind. Eng. Chem. Prod. Res. Dev. 1964,
3, 170 – 173.
resulting solid shows the expected signal in the C CPMAS
1
13
spectrum and the H- C HETCOR for the formation of the
surface square-pyramidyl (SP) metallacycle (3 ). The for-
SP
mation of the SP metallacycle is also consistent with DFT
calculations which indicate this isomer is generally more
stable than the trigonal bipyramidyl (TBP) metallacycle,
[
3] J. C. Mol, J. Mol. Catal. A 2004, 213, 39 – 45.
[
18]
[4] S. Lwin, I. E. Wachs, ACS Catal. 2014, 4, 2505 – 2520.
particularly for tungsten oxo species.
[
[
5] N. Popoff, E. Mazoyer, J. Pelletier, R. M. Gauvin, M. Taoufik,
Chem. Soc. Rev. 2013, 42, 9035 – 9054.
6] a) J. R. M. Kress, M. J. M. Russell, M. G. Wesolek, J. A. Osborn,
J. Chem. Soc. Chem. Commun. 1980, 431 – 432; b) J. Kress, M.
Wesolek, J. P. Leny, J. A. Osborn, J. Chem. Soc. Chem. Commun.
1981, 1039 – 1040.
Reacting 3 with cis-4-nonene (0.02%) under the opti-
mized reaction conditions resulted in full conversion of the
substrate after an induction period and a significantly reduced
À1
À1
initial TOF (17 min versus 325 min for 2a). These results
are consistent with metathesis inhibition (reversible deacti-
vation) by ethylene through the formation of the stable and
less reactive unsubstituted metallacycle intermediate. Since
terminal alkene homocoupling results in significant ethylene
build-up in the reaction mixture we conducted this reaction
with a reactor that allowed removal of ethylene during the
reaction (see the Supporting Information for details). Briefly,
[
7] a) E. Mazoyer, N. Merle, A. d. Mallmann, J.-M. Basset, E.
Berrier, L. Delevoye, J.-F. Paul, C. P. Nicholas, R. M. Gauvin, M.
Taoufik, Chem. Commun. 2010, 46, 8944 – 8946; b) N. Merle, G.
Girard, N. Popoff, A. De Mallmann, Y. Bouhoute, J. Trꢀbosc, E.
Berrier, J.-F. Paul, C. P. Nicholas, I. Del Rosal, L. Maron,
R. g. M. Gauvin, L. Delevoye, M. Taoufik, Inorg. Chem. 2013,
52, 10119 – 10130.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
[
8] a) D. V. Peryshkov, R. R. Schrock, M. K. Takase, P. Mꢁller,
A. H. Hoveyda, J. Am. Chem. Soc. 2011, 133, 20754 – 20757;
b) D. V. Peryshkov, R. R. Schrock, Organometallics 2012, 31,
[14] A. Poater, B. Cosenza, A. Correa, S. Giudice, F. Ragone, V.
Scarano, L. Cavallo, Eur. J. Inorg. Chem. 2009, 1759 – 1766.
[15] D. Gajan, C. Copꢀret, New J. Chem. 2011, 35, 2403 – 2408.
[16] F. Blanc, C. Copꢀret, A. Lesage, L. Emsley, Chem. Soc. Rev.
2008, 37, 518 – 526.
[17] F. Blanc, C. Copꢀret, J. Thivolle-Cazat, J.-M. Basset, A. Lesage,
L. Emsley, A. Sinha, R. R. Schrock, Angew. Chem. Int. Ed. 2006,
45, 1216 – 1220; Angew. Chem. 2006, 118, 1238 – 1242.
[18] a) X. Solans-Monfort, E. Clot, C. Copꢀret, O. Eisenstein, J. Am.
Chem. Soc. 2005, 127, 14015 – 14025; b) A. Poater, X. Solans-
Monfort, E. Clot, C. Copꢀret, O. Eisenstein, J. Am. Chem. Soc.
2007, 129, 8207 – 8216; c) X. Solans-Monfort, C. Copꢀret, O.
Eisenstein, J. Am. Chem. Soc. 2010, 132, 7750 – 7757; d) X.
Solans-Monfort, C. Copꢀret, O. Eisenstein, Organometallics
2012, 31, 6812 – 6822.
7278 – 7286.
[
9] a) M. Chabanas, A. Baudouin, C. Copꢀret, J.-M. Basset, J. Am.
Chem. Soc. 2001, 123, 2062 – 2063; b) C. Copꢀret, Dalton Trans.
2007, 5498 – 5504.
[
10] M. P. Conley, V. Mougel, D. V. Peryshkov, W. P. Forrest, D.
Gajan, A. Lesage, L. Emsley, C. Copꢀret, R. R. Schrock, J. Am.
Chem. Soc. 2013, 135, 19068 – 19070.
11] a) F. Blanc, R. Berthoud, A. Salameh, J.-M. Basset, C. Copꢀret,
R. Singh, R. R. Schrock, J. Am. Chem. Soc. 2007, 129, 8434 –
[
8435; b) B. Rhers, A. Salameh, A. Baudouin, E. A. Quadrelli, M.
Taoufik, C. Copꢀret, F. Lefebvre, J.-M. Basset, X. Solans-
Monfort, O. Eisenstein, W. W. Lukens, L. P. H. Lopez, A. Sinha,
R. R. Schrock, Organometallics 2006, 25, 3554 – 3557; c) V.
Mougel, C. Copꢀret, Chem. Sci. 2014, 5, 2475 – 2481.
[19] a) A.-M. Leduc, A. Salameh, D. Soulivong, M. Chabanas, J.-M.
Basset, C. Copꢀret, X. Solans-Monfort, E. Clot, O. Eisenstein,
V. P. W. Bçhm, M. Rçper, J. Am. Chem. Soc. 2008, 130, 6288 –
6297; b) G. C. Vougioukalakis, R. H. Grubbs, Chem. Rev. 2010,
110, 1746 – 1787.
[
[
12] D. V. Peryshkov, W. P. Forrest, R. R. Schrock, S. J. Smith, P.
Mꢁller, Organometallics 2013, 32, 5256 – 5259.
13] F. Rascꢂn, R. Wischert, C. Copꢀret, Chem. Sci. 2011, 2, 1449 –
1456.
4
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
These are not the final page numbers!

Angewandte
Chemie
Communications
Supported Catalysts
Like a rock: Grafting a W-oxo alkylidene,
containing bulky adamantyl-substituted
phenoxy ligands, onto partially dehy-
droxylated silica generates a very active
and stable metathesis catalyst. This large
ligand allows efficient terminal alkene
homocoupling with a turnover number
(TON) exceeding 75000.
M. P. Conley, W. P. Forrest, V. Mougel,
C. Copꢁret,*
R. R. Schrock*
&&&& — &&&&
Bulky Aryloxide Ligand Stabilizes
a Heterogeneous Metathesis Catalyst
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!