Catalytic functionalization of hydrocarbons by σ-bond-metathesis chemistry: Dehydrosilylation of methane with a scandium catalyst

Article abstract of DOI:10.1002/anie.200390213

C-H activation with scandium: Reaction of [Cp₂*ScCH₃] (1, Cp* = C₅Me₅) with MesSiH₃ (Mes=2,4,6-C₆H₂Me₃) yielded a new scandium silyl complex [Cp₂*ScSiH₂Mes] (2) by an unusual σ-bond metathesis pathway. Complex 2 activates the C-H bonds of benzene and methane. In the presence of excess Ph₂SiH₂, the catalytic dehydrosilation of saturated and unsaturated hydrocarbons, including methane, was accomplished, with [Cp₂*ScH] (see scheme; R=CH₃, C₃H₅, CH=CMe₂) which can be formed from 1 or 2. This appears to be the first example of catalytic methane functionalization by σbond metathesis.

Full text of DOI:10.1002/anie.200390213

Angewandte
Chemie
Catalytic Methane Activation
nistic investigations indicate that this reactivity is possibly the
result of the ability of silicon, unlike carbon, to adopt the
b position of a four-centered transition state.[10]
Catalytic Functionalization of Hydrocarbons by s-
Bond-Metathesis Chemistry: Dehydrosilylation of
Methane with a Scandium Catalyst**
ꢀ
The recent discovery of arene C H activation by a
cationic, hafnium silyl complex suggests that hydrocarbon
conversions might be based on catalysts that possess highly
reactive metal–silicon bonds [Eq. (2)].[11] The possibility that
scandocene–silicon bonds might be highly reactive in this
sense was suggested by the similarity between the electron
Aaron D. Sadow and T. Don Tilley*
The selective, catalytic functionalization of saturated hydro-
carbons represents one of the most important challenges in
chemical research.[1] While some progress has been made,[2–5]
there are very few processes which allow conversion of the
cheapest and most abundant hydrocarbon, methane. A few
homogeneous catalytic conversions of methane have been
developed, but none of these are efficient enough to be
employed in routine chemical syntheses. One type of
homogeneous catalytic system, originally described by Shilov
and co-workers,[1a] features a platinum-group-metal complex
as the catalyst and converts methane into simple derivatives
of the type MeX (X = Cl, OH, OSO3H, O2CCF3) under acidic
conditions.[1,3] Metal oxo complexes, such as [NBu4]VO3 with
pyrazine-2-carboxylic acid and methane monooxygenase,
have also been found to catalyze selective oxidations of
methane with O2 and/or peroxides in protic solvents to yield
methanol or its derivatives.[4] Although several stoichiometric
reactions of transition-metal complexes with methane are
known,[1,5] this reactivity has not yet provided useful catalytic
processes.
count and ionic radius of the metal center in the complexes
[Cp02HfSiR3]+ and [Cp02ScSiR3],[12] and by the known ability of
*
[Cp2 Sc] derivatives to activate hydrocarbons by s-bond
metathesis.[6b]
Methods for the generation of d0-metal–silicon bonds
have involved s-bond metathesis reactions of hydrosilanes
with hydride or silyl derivatives.[10] However, d0-metal alkyl
derivatives generally react with hydrosilanes by alkyl transfer
to silicon and formation of a metal hydride.[8,13] It was
*
surprising, then, to discover that [Cp2 ScMe] (1) reacted with
1.2 equivalents of MesSiH3 (Mes = 2,4,6-C6H2Me3) with elim-
ination of methane to produce a bright-yellow solution of
*
[Cp2 ScSiH2Mes] [2; 3 h, room temperature, [D12]cyclohex-
*
ane; Eq. (3)]. The competitive formation of [Cp2 ScH] (3) and
New approaches to the development of catalysts for
methane conversion might involve s-bond metathesis steps.
Indeed, the first reports of stoichiometric, homogeneous
*
methane activation, by [Cp2 MMe] (M = Sc (3), Y, Lu; Cp* =
h5-C5Me5) complexes, suggest the reaction proceeds via such
transition states [Eq. (1)].[6] Despite this early breakthrough
involving the degenerate exchange of methyl groups, produc-
tive reactions of methane by s-bond metathesis have not been
reported. A possible limitation to the development of such
processes (e.g., carbon–carbon coupling) is the apparent
restriction that carbon cannot adopt the position b to the
metal center in a four-centered transition state.[7] In contrast,
a few catalytic processes involving d0 (and fnd0) metal catalysts
and silane substrates have been discovered (e.g., olefin
hydrosilation and silane dehydropolymerization).[8,9] Mecha-
MesMeSiH2 (ca. 5%) was also observed, but addition of a
slight excess of MesSiH3 converted the by-product 3 into 2 (an
optimized preparation of 2 is given in the Experimental
Section).
Complex 2 reacted in [D6]benzene at 658C to form the
*
scandium phenyl complex [Cp2 ScC6D5] ([D5]4; ca. 70% after
1
60 min at 90% conversion, by H NMR spectroscopy). The
primary by-product in this reaction is the hydride [D1]3 (ca.
12% after 60 mins at 658C), which may form by reaction of a
*
[Cp2 ScR] species with hydrogen or MesSiH3, or by the
thermal decomposition of 2 (t1/2 = 8.5h at
508C in [D12]cyclohexane; 95% yield of
2
3). A H NMR spectrum of the reaction
mixture (in [D6]benzene) indicates that
MesSiD3 is the major silane product, but
also reveals the presence of a trace
amount (< 5%) of the dehydrocoupling
[*] Prof. T. D. Tilley, A. D. Sadow
Department of Chemistry
product MesD2SiSiD2Mes; there is no evidence for the
formation of [([Dn]Cp*)2ScC6D5]. The conversion of 2 into
the phenyl complex 4 in benzene is approximately two-times
faster than it is in [D6]benzene, which indicates that CH bond
activation is involved in the rate-determining step. Plots of [2]
versus time reveal that the reaction rate increases as the
thermolysis proceeds, which suggests that a product or
intermediate promotes the reaction. A reasonable mechanism
University of California, Berkeley
Berkeley, CA 94720-1460 (USA)
Fax: (+1)510-642-8940
E-mail: tilley@cchem.berkeley.edu
[**] Acknowledgment is made to the National Science Foundation for
their generous support of this work, and to Dr. Richard Andersen,
Dr. John Bercaw, and Dr. Mark Thompson for helpful discussions.
Angew. Chem. Int. Ed. 2003, 42, No. 7
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4207-0803 $ 20.00+.50/0
803

Communications
for this reaction, involving hydrogen as a catalyst, is analogous
catalysis were complicated by the presence of several species
in the reaction mixture. The solution remained homogeneous
over the course of the reaction and precipitation of insoluble
species, such as [{(h1:h5-C5Me4CH2)(h5-Cp*)Sc}2],[6b] did not
occur. As 3 was depleted over the course of the reaction
(because of slow decomposition), the rate of formation of
Ph2MeSiH decreased, which suggests that 3 is involved in the
catalytic cycle.
*
to that proposed for the reaction of [Cp2 SmCH(SiMe3)2] with
H2Si(SiMe3)2.[14]
Reaction of 2 with methane (14 equiv) in [D12]cyclohex-
ane occurred slowly at room temperature over four days to
give MesSiH3 and MesMeSiH2 (85% and 15%, respectively,
by GC-MS), and 3 (42%) as the major scandium-containing
product [Eq. (4)]. At intermediate stages of the reaction the
methyl complex 1 was observed in trace quantities. As
Mechanistic investigations of the catalytic functionaliza-
tion of methane with Ph2SiH2 have thus
ꢀ
far focused separately on the Si C bond
ꢀ
formation and C H activation steps. The
ꢀ
mechanism of Si C bond formation
could proceed via a four-centered transition state in which a
described above, 1 reacted with MesSiH3 to give 2 and CH4
as the primary kinetic products. Taken together, these results
indicate the existence of the coupled equilibria of
Equation (4), for which the thermodynamic products are
MesMeSiH2 and the scandium hydride 3. However, just as in
the metalation of benzene discussed above, it is not clear
ꢀ
Sc Me derivative reacts with Ph2SiH2 to yield 3 and
Ph2MeSiH (methyl transfer), or by reaction of a scandium
silyl species with methane (silyl transfer). The methylation of
Ph2SiH2 by 1 to give Ph2MeSiH and 3 in [D6]benzene or
[D12]cyclohexane (t1/2 ꢁ 45min at 25 8C) was readily observed;
note the sharp contrast between this reaction and that of 1
with MesSiH3, which produced methane and 2 (see above;
[Eq. (2)]).[15] In contrast, the reaction of 2 with a large excess
of methane was significantly slower and less efficient than the
reaction of 1 with Ph2SiH2. These observations lead us to
ꢀ
whether the C H activation step involves 2 or 3.
The observed behavior of methane as a methylating
reagent suggested that catalytic methane functionalization
might be possible with the appropriate organosilane [Eq. (2)],
where RH = CH4). A screening of several silanes with the
catalyst 3 revealed that Ph2SiH2 provided the best results in
terms of the minimization of competitive side reactions (e.g.,
dehydrocoupling and redistribution). A mixture of 3 and
10 equivalents of Ph2SiH2 in [D12]cyclohexane reacted under
approximately 7 atm of CH4 in a Young's tube at 808C to
yield Ph2MeSiH (by GC-MS and NMR spectroscopy) in
substoichiometric quantities (ca. 0.4 equiv after 1 week). The
reaction rate is dependent on methane concentration; heating
a cyclohexane solution of Ph2SiH2 and 1 to 808C under
150 atm of methane produced five equivalents of Ph2MeSiH
after 1 week, one equivalent of which was derived directly
from 1 [Eq. (5)]. Increasing the amount of added Ph2SiH2 to
ꢀ
favor methyl transfer from Sc to Si as the Si C bond-forming
step in the catalysis outlined in Equation (5).
Attempts to directly detect the activation of methane by 3
have been unsuccessful. Heating [D12]cyclohexane solutions
of 3 under 7–150 atm of CH4 to 808C for 4 days, followed by
release of the pressure, did not produce observable quantities
of 1 (by 1H NMR spectroscopy). Note, however, that a
reaction between [Cp2 ScD] ([D1]3) and CH4 is implied by the
observed incorporation of deuterium into methane in the
presence of excess D2 or [D6]benzene.[6b] The possible
*
*
participation of [Cp2 ScSiHPh2] (5) in methane activation is
suggested by the observed reactions of isoelectronic
[Cp2Hf(SiHMes2)]+ (Cp = h-C5H5) with the C H bonds of
ꢀ
both benzene and toluene.[11] However, 5 could not be
detected under catalytic conditions, and attempts to isolate
it have failed.
The dehydrogenative silylation of other hydrocarbons can
*
also be mediated by [Cp2 Sc] derivatives. For example, the
*
20 equivalents did not substantially affect the rate of reaction,
as approximately five equivalents of Ph2MeSiH were detected
after a week in both cases. Apparently, the rate-limiting step
vinyl complex [Cp2 ScCHCMe2] slowly catalyzed the coupling
of Ph2SiH2 (8 equiv) with isobutylene (18 equiv) at 508C to
produce the vinyl silane Ph2(Me2CCH)SiH (ca. 2 turnovers
after 20 days in [D12]cyclohexane).[6b] With PhSiH3 and
isobutylene, a mixture of hydrosilation (Me2HCCH2SiH2Ph)
and dehydrosilation (Me2CCHSiH2Ph) products (3:2 ratio;
3 turnovers, 2 days in [D12]cyclohexane), as well as minor
amounts of dehydrocoupling products, were observed. Com-
pound 3 also catalyzed the dehydrocoupling of cyclopropane
and Ph2SiH2 in [D12]cyclohexane at 808C (2.5turnovers,
20 days). Interestingly, benzene and pentane were not suit-
able substrates for this dehydrosilation process.
Herein we described a new approach for the selective,
catalytic conversion of methane. The catalytic cycles reported
(for hydrocarbon dehydrosilations) are believed to involve
only s-bond-metathesis steps. Such mechanisms are therefore
quite analogous to those previously proposed for d0-metal-
ꢀ
in the catalytic cycle is C H bond activation. Though the
methane conversion is slow, the reaction is reasonably
selective with 75% of the consumed Ph2SiH2 being converted
into Ph2MeSiH. It seems likely that some of the Ph2SiH2
ꢀ
reacts by competitive Si Ph hydrogenolysis and silane
dehydropolymerization, but no other products were detected
by GC after the catalyst had been removed by an aqueous
workup. Increasing the temperature to 1008C decreased the
amount of Ph2MeSiH produced (< 1 turnover, by NMR
spectroscopy), apparently as a result of rapid decomposition
of the catalyst 3 at this temperature.
The composition of the catalytic reaction mixture was
1
probed by H NMR spectroscopy at low methane pressures
(ca. 8 atm), but attempts to study the mechanism of the
804
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4207-0804 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 7

Angewandte
Chemie
[7] W. E. Piers, P. J. Shapiro, E. E. Bunel, J. E. Bercaw, Synlett 1990,
catalyzed dehydropolymerizations of silanes and stan-
nanes.[10,16] Jordan and Taylor have reported the 1,2-addition
74.
[8] a) P.-F. Fu, L. Brard, Y. Li, T. J. Marks, J. Am. Chem. Soc. 1995,
117, 7157, and references therein; b) T. I. Gountchev, T. D.
Tilley, Organometallics 1999, 18, 5661; c) G. A. Molander, M.
Julius, J. Am. Chem. Soc. 1995, 117, 4415; d) A. K. Dash, I.
Gourevich, J. Q. Wang, J. Wang, M. Kapon, M. S. Eisen,
Organometallics 2001, 20, 5084.
[9] a) C. T. Aitken, J. F. Harrod, E. Samuel, J. Am. Chem. Soc. 1986,
108, 4059; b) J. Y. Corey, X.-H. Zhu, T. C. Bedard, L. D. Lange,
Organometallics, 1991, 10, 115; c) T. D. Tilley, Acc. Chem. Res.
1993, 26, 22.
ꢀ
of a C H bond of picoline to an olefin (hydroalkylation),
which is catalyzed by a cationic d0 bis(Cp) zirconium complex
by a mechanism involving s-bond metathesis and olefin
insertion.[17] Dehydrogenative silation of terminal alkynes has
been observed as a competitive process to hydrosilation.[18]
*
Note that the ability of [Cp2 ScH] (3) to catalyze hydrocarbon
dehydrosilation requires that competitive processes, such as
silane polymerization and redistribution, are slow relative to
[10] a) H.-G. Woo, J. F. Walzer, T. D. Tilley, J. Am. Chem. Soc. 1992,
114, 7047; b) G. L. Casty, C. G. Lugmair, N. S. Radu, T. D. Tilley,
J. F. Walzer, D. Zargarian, Organometallics 1997, 16, 8; c) N. S.
Radu, T. D. Tilley, J. Am. Chem. Soc. 1992, 114, 8293.
[11] A. D. Sadow, T. D. Tilley, J. Am. Chem. Soc. 2002, 124, 6814.
[12] M. E. Thompson, J. E. Bercaw, Pure Appl. Chem. 1984, 56, 1.
[13] A. Z. Voskoboynikov, I. N. Parshina, A. K. Shestakova, K. P.
Butin, I. P. Beletskaya, L. G. Kuz'mina, J. A. K. Howard, Orga-
nometallics 1997, 16, 4041.
[14] N. S. Radu, T. D. Tilley, J. Am. Chem. Soc. 1995, 117, 5863.
[15] Reactions of 1 with organosilanes are highly sensitive to the
nature of substituents at silicon; A. D. Sadow, T. D. Tilley,
unpublished results.
ꢀ
C H bond activation. This characteristic of the scandium
system described here is unusual, and further investigations
will address mechanistic issues related to this selectivity.
Experimental Section
All manipulations were performed either on a Schlenk line under an
Ar atmosphere or in a N2-filled drybox (M. Braun). All solvents and
reagents were purified by standard procedures.
2: Neat MesSiH3 (0.120 g, 0.798 mmol) was added to solid
*
[Cp2 ScMe] (1; 0.0485g, 0.1468 mmol). The bright-yellow solid which
formed was washed with cold pentane (3 ” 2 mL), yielding 2. The
washings were cooled to ꢀ308C from which additional compound
could be isolated: yield 0.030 g, 44%, m.p. 1758C, elemental analysis:
calcd (%) for C29H43ScSi: C 74.96, H 9.33; found: C 74.98, H 9.31;
1H NMR (500 MHz, [D6]Benzene, 258C, TMS): d = 6.988 (s, 2H,
C6H2Me3), 4.345(s, 2H, SiH), 2.581 (s, 6H, o-C6H2Me3), 2.295(s, 3H,
p-C6H2Me3), 1.814 ppm (s, 30H, C5Me5); 13C{1H} NMR (125MHz):
d = 160.70 (C6H2Me3), 143.98 (C6H2Me3), 141.56 (C6H2Me3), 135.83
(C6H2Me3), 123.00 (C5Me5), 26.45( o-C6H2Me3), 21.71 (p-C6H2Me3),
[16] T. Imori, V. Lu, H. Cai, T. D. Tilley, J. Am. Chem. Soc. 1995, 117,
9931.
[17] R. F. Jordan, D. F. Taylor, J. Am. Chem. Soc. 1989, 111, 778.
[18] A. K. Dash, J. Q. Wang, M. S. Eisen, Organometallics 1999, 18,
4727.
11.86 ppm (C5Me5); 29Si{1H} NMR (99 MHz): d = ꢀ71.0 ppm (1JSiH
=
135Hz); IR (KBr): n˜ = 2014 cmꢀ1 (Si-H).
Received: August 26, 2002 [Z50046]
[1] a) J. A. Labinger, J. E. Bercaw, Nature 2002, 417, 5 07; b) R. H.
Crabtree, J. Chem. Soc. Dalton Trans. 2001, 2437; c) A. Sen, Acc.
Chem. Res. 1998, 31, 550; d) S. S. Stahl, J. A. Labinger, J. E.
Bercaw, Angew. Chem. 1998, 110, 2298; Angew. Chem. Int. Ed.
1998, 37, 2180; e) A. E. Shilov, G. B. Shul'pin, Chem. Rev. 1997,
97, 2879; f) B. A. Arndtsen, R. G. Bergman, T. A. Mobley, T. H.
Peterson, Acc. Chem. Res. 1995, 28, 154.
[2] Some recent representative examples of catalytic hydrocarbon
functionalizations: a) H. Chen, S. Schlecht, T. C. Semple, J. F.
Hartwig, Science 2000, 287, 1995; b) J.-Y. Cho, M. K. Tse, D.
Holmes, R. E. Maleczka, M. R. Smith, Science 2002, 295, 305;
c) C. Jia, D. Piao, J. Oyamada, W. Lu, T. Kitamura, Y. Fujiwara,
Science 2000, 287, 1992; d) F. Liu, E. B. Pak, B. Singh, C. M.
Jensen, A. S. Goldman, J. Am. Chem. Soc. 1999, 121, 4086.
[3] R. A. Periana, D. J. Taube, S. Gamble, H. Taube, T. Satoh, H.
Fujii, Science 1998, 280, 560.
[4] a) G. V. Nizova, G. Suss-Fink, G. B. Shul'pin, Chem. Commun.
1997, 397; b) M. N. Vargaftik, I. P. Stolarov, I. I. Moiseev, J.
Chem. Soc. Chem. Commun. 1990, 1049; c) M. Merkx, D. A.
Kopp, M. H. Sazinsky, J. L. Blazyk, J. Müller, S. J. Lippard,
Angew. Chem. 2001, 113, 2860; Angew. Chem. Int. Ed. 2001, 40,
2782.
[5] R. H. Crabtree, Chem. Rev. 1995, 95, 987, and references therein.
[6] a) P. L. Watson, G. W. Parshall, Acc. Chem. Res. 1985, 18, 5 1;
b) M. E. Thompson, S. M. Baxter, A. R. Bulls, B. J. Burger, M. C.
Nolan, B. D. Santarsiero, W. P. Schaefer, J. E. Bercaw, J. Am.
Chem. Soc. 1987, 109, 203; c) C. M. Fendrick, T. J. Marks, J. Am.
Chem. Soc. 1986, 108, 425.
Angew. Chem. Int. Ed. 2003, 42, No. 7
ꢀ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4207-0805 $ 20.00+.50/0
805