Si-H bond activation of alkynylsilanes by group 4 metallocene complexes

Article abstract of DOI:10.1021/ja910527w

The reactivity of variously substituted alkynylsilanes toward selected group 4 metallocene complexes was Investigated. Reactions of the alkynylsilanes R¹C₂SIR²₂H (R¹ = SIMe₃, R² = Me, 1; R¹ = SIMe₃, R² = Ph, 2; R¹ = SIMe₂H, R² = Me, 5) with Cp₂TIMe₂ (Cp = η;5-cyclopentadlenyl) resulted, upon methyl group transfer to the sllyl group, In the previously described titanocene alkyne complexes Cp₂TI(R¹C₂SIR ²₂R³) (R¹ = Me₃SI, R ² = R³ = Me, 3; R₁ = HMe₂SI, R ² = R³ = Me, 6) or the unreported complex 4 (R¹ = Me₃SI, R² = Ph, R³ = Me). The Cp ₂TICI₂/n-BuLI system yielded alkyne complexes 6 and 7 (R¹ = HMe₂SI, R² = Me, R³ = H); no alkyl group transfer was detected. On the other hand, reactions utilizing the Cp₂ZrCI₂/n-BuLI system afforded Inseparable mixtures; however, complexes of the type Cp₂Zr[R¹C ₂SIMe₂(O-Bu)] (R¹ = Me₃SI, 8; R ¹ = HMe₂SI, 9) were detected. Cp₂Hf(A-Bu) ₂ reacted with the alkynylsllanes in a diverse way, depending on the substltuents of the alkyne substrate. The reaction with an excess of alkyne 1 (R¹ = Me₃SI, R² = Me) afforded only an intractable mixture, which contained Me₃SIC₂SIMe ₂(Ai-Bu) (10). Hafnacyclopentadlenes 13-15 as precedented product types were obtained when alkyne 12 (R¹ = Ph, R² = Me) was used. In sharp contrast, the symmetrically substituted alkynes 5 (R¹ = HMe₂SI, R² = Me) and H₂PhSIC ₂SIPhH₂ (18) yielded the hitherto unknown Sl-contalnlng metallacycles 16 and 19. A reaction mechanism leading to these products was proposed and subsequently supported by DFT calculations. in addition, the reduction of Cp₂HfCI₂ with magnesium in THF In the presence of alkynylsilanes was shown to be an alternative route to compounds 14-16 and 19. Presumably due to steric reasons, alkyne 1 could not form any of the product types described above. Nevertheless, It was utilized for the preparation of the PMe₃-stablllzed hafnocene alkyne complex 11.

Full text of DOI:10.1021/ja910527w

Published on Web 03/04/2010
Si-H Bond Activation of Alkynylsilanes by Group 4
Metallocene Complexes
Martin Lamacˇ,*,† Anke Spannenberg, Wolfgang Baumann, Haijun Jiao,
Christine Fischer, Sven Hansen, Perdita Arndt, and Uwe Rosenthal*
Leibniz-Institut fu¨r Katalyse e.V. an der UniVersita¨t Rostock, Albert-Einstein-Strasse 29 a,
18059 Rostock, Germany
Received December 14, 2009; E-mail: uwe.rosenthal@catalysis.de; martin.lamac@jh-inst.cas.cz
Abstract: The reactivity of variously substituted alkynylsilanes toward selected group 4 metallocene
complexes was investigated. Reactions of the alkynylsilanes R1C2SiR2 H (R1 ) SiMe3, R2 ) Me, 1; R1 )
2
SiMe3, R2 ) Ph, 2; R1 ) SiMe2H, R2 ) Me, 5) with Cp2TiMe2 (Cp ) η5-cyclopentadienyl) resulted, upon
methyl group transfer to the silyl group, in the previously described titanocene alkyne complexes
Cp2Ti(R1C2SiR2 R3) (R1 ) Me3Si, R2 ) R3 ) Me, 3; R1 ) HMe2Si, R2 ) R3 ) Me, 6) or the unreported
2
complex 4 (R1 ) Me3Si, R2 ) Ph, R3 ) Me). The Cp2TiCl2/n-BuLi system yielded alkyne complexes 6 and
7 (R1 ) HMe2Si, R2 ) Me, R3 ) H); no alkyl group transfer was detected. On the other hand, reactions
utilizing the Cp2ZrCl2/n-BuLi system afforded inseparable mixtures; however, complexes of the type
Cp2Zr[R1C2SiMe2(n-Bu)] (R1 ) Me3Si, 8; R1 ) HMe2Si, 9) were detected. Cp2Hf(n-Bu)2 reacted with the
alkynylsilanes in a diverse way, depending on the substituents of the alkyne substrate. The reaction with
an excess of alkyne 1 (R1 ) Me3Si, R2 ) Me) afforded only an intractable mixture, which contained
Me3SiC2SiMe2(n-Bu) (10). Hafnacyclopentadienes 13-15 as precedented product types were obtained
when alkyne 12 (R1 ) Ph, R2 ) Me) was used. In sharp contrast, the symmetrically substituted alkynes 5
(R1 ) HMe2Si, R2 ) Me) and H2PhSiC2SiPhH2 (18) yielded the hitherto unknown Si-containing metallacycles
16 and 19. A reaction mechanism leading to these products was proposed and subsequently supported by
DFT calculations. In addition, the reduction of Cp2HfCl2 with magnesium in THF in the presence of
alkynylsilanes was shown to be an alternative route to compounds 14-16 and 19. Presumably due to
steric reasons, alkyne 1 could not form any of the product types described above. Nevertheless, it was
utilized for the preparation of the PMe3-stabilized hafnocene alkyne complex 11.
1. Introduction
of the “free” titanocene fragment (Cp2Ti) suitable for further
reactions. Also, Zr3 and Hf4 constitute analogous alkyne
complexes (stabilized by additional THF, pyridine, or PMe3
ligands in the case of metallocenes bearing unsubstituted
cyclopentadienyl ligands). Nevertheless, some differences in the
reactivity of particular metals are evident. Notably, upon the
reduction of Cp*2HfCl2 (Cp* ) η5-pentamethylcyclopentadi-
enyl) with lithium in the presence of Me3SiC2SiMe3, an
activation of both C-Si and C-H bonds was observed, and
unexpected products such as metallacycle I (Chart 1) were
isolated in addition to the simple alkyne complex.5 These
uncommon bond activations have a potential for further
development of novel synthetic tools or can help to improve
the understanding of particular catalytic processes.
The chemistry of group 4 metallocene alkyne complexes is
well established, especially for Ti and Zr;1 the bis(trimethyl-
silyl)ethyne complex of titanocene, Cp2Ti(η2-Me3SiC2SiMe3),2,1b
is the prototype in particular, which serves as a stable source
† On leave from the J. Heyrovsky´ Institute of Physical Chemistry of
Academy of Sciences of the Czech Republic, Dolejsˇkova 3, 182 23 Prague
8, Czech Republic.
(1) Review: (a) Rosenthal, U.; Burlakov, V. V. In Titanium and Zirconium
in Organic Synthesis; Marek, I., Ed.; Wiley-VCH: Weinheim,
Germany, 2002; pp 355-389. (b) Rosenthal, U.; Burlakov, V. V.;
Arndt, P.; Baumann, W.; Spannenberg, A. Organometallics 2003, 22,
884.
(2) (a) Burlakov, V. V.; Rosenthal, U.; Petrovskii, P. V.; Shur, V. B.;
Vol’pin, M. E. Organomet. Chem. USSR 1988, 1, 526. (b) Burlakov,
V. V.; Polyakov, A. V.; Yanovsky, A. I.; Struchkov, Y. T.; Shur, V. B.;
Vol’pin, M. E.; Rosenthal, U.; Go¨rls, H. J. Organomet. Chem. 1994,
476, 197. (c) Varga, V.; Mach, K.; Pola´sˇek, M.; Sedmera, P.; Hiller,
J.; Thewalt, U.; Troyanov, S. I. J. Organomet. Chem. 1996, 506, 241.
(3) (a) Rosenthal, U.; Ohff, A.; Michalik, M.; Go¨rls, H.; Burlakov, V. V.;
Shur, V. B. Angew. Chem., Int. Ed. Engl. 1993, 32, 1193. (b)
Rosenthal, U.; Ohff, A.; Baumann, W.; Tillack, A.; Go¨rls, H.;
Burlakov, V. V.; Shur, V. B. Z. Anorg. Allg. Chem. 1995, 621, 77.
(4) (a) Beweries, T.; Burlakov, V. V.; Bach, M. A.; Arndt, P.; Baumann,
W.; Spannenberg, A.; Rosenthal, U. Organometallics 2007, 26, 247.
(b) Burlakov, V. V.; Beweries, T.; Bogdanov, V. S.; Arndt, P.;
Baumann, W.; Petrovskii, P. V.; Spannenberg, A.; Lyssenko, K. A.;
Shur, V. B.; Rosenthal, U. Organometallics 2009, 28, 2864.
(5) Beweries, T.; Burlakov, V. V.; Bach, M. A.; Peitz, S.; Arndt, P.;
Baumann, W.; Spannenberg, A.; Rosenthal, U.; Pathak, B.; Jemmis,
E. D. Angew. Chem., Int. Ed. 2007, 46, 6907.
(6) (a) Aitken, C. T.; Harrod, J. F.; Samuel, E. J. Organomet. Chem. 1985,
279, C11. (b) Corey, J. Y.; Zhu, X.-H. J. Organomet. Chem. 1992,
439, 1. (c) Shaltout, R. M.; Corey, J. Y. Tetrahedron 1995, 51, 4309.
(d) Corey, J. Y.; Zhu, X.-H.; Bedard, T. C.; Lange, L. D. Organo-
metallics 1991, 10, 924. (e) Tilley, T. D. Acc. Chem. Res. 1993, 26,
22. (f) Gauvin, F.; Harrod, J. F.; Woo, H.-G. AdV. Organomet. Chem.
1998, 42, 363.
9
10.1021/ja910527w  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 4369–4380 4369

A R T I C L E S
Lamacˇ et al.
Chart 1
σ-bond metathesis with other silanes (primary or secondary),
which is an alternative route to other silyl complexes.13
Metallocene silyl complexes were also isolated as intermediates
in the catalytic dehydrocoupling of silanes or hydrosilylation
reactions. In both processes, Cp2MMe2 (M ) Ti, Zr) were used
as precatalysts, which reacted with primary or secondary silanes.
The structural motifs, which were observed, comprise Ti(III)
dimers bridged by the hydride anion or the hydrosilyl moiety,18
doubly hydride bridged silyl Zr(IV) dimers,19 or paramagnetic
Ti(III) monomers stabilized by a pyridine or tertiary phosphine
ligand.11,20 Such reactions are almost fully restricted to primary
and secondary silanes, while tertiary derivatives are often
claimed to be inert. However, the sole example of a tertiary
silyl Ti(III) complex (II; Chart 1) was prepared in a similar
manner from Cp2TiMe2 and 2 equiv of 2-(dimethylsilyl)pyridine
by Harrod and co-workers.21
Several complexes of titanium and zirconium with alkynes
bearing dimethylsilyl groups (III; Chart 1)22 were previously
prepared either by the alkyne-exchange reaction in the com-
plexes Cp2M(η2-Me3SiC2SiMe3) (M ) Ti, Zr) with the corre-
sponding alkynylsilane or by the reduction of Cp2MCl2 with
magnesium in THF in the presence of the alkynylsilane. It was
shown that an agostic two-electron three-center Si-H-M
interaction involving the hydrosilyl group is present, while the
coordinated alkyne fragment adopts the trans configuration.
Complex III (M ) Ti, R ) SiMe2H; Chart 1) has been also
described by theoretical calculations to have a d2 system with
comparable back-donating strengths of the alkyne and the Si-H
bond.23 The competition between the alkyne and the Si-H
moiety for coordination to the metal center is manifested both
structurally and electronically. Such complexes resemble an
intramolecularly stabilized intermediate of the alkyne hydrosi-
lylation reaction and therefore serve as valuable model com-
pounds for such processes. Analogous complexes have not yet
been reported for hafnium.
Catalyzed reactions involving organic hydrosilanes and group
4 metallocene complexes such as the dehydrogenative coupling
to oligo- and polysilanes,6 or hydrosilylation of unsaturated
organic substrates (alkenes,7 alkynes,8 imines,9 ketones,10 py-
ridines11) are well documented. Metallocene silyl hydride
complexes were postulated as intermediates in some of these
processes. Owing to the rather high reactivity of the M-Si bond,
which can undergo an insertion to unsaturated bonds,12 or
σ-bond metathesis with other Si-H or even C-H bonds,13,6e
relatively few metallocene silyl complexes have been isolated.14
Complexes containing the Cp2M-SiR3 fragment were prepared,
e.g., by the reaction of Cp2MCl2 (M ) Ti, Zr, Hf) with
Al(SiMe3)3 ·OEt2 or LiSiR3,12a,14 to provide chloro silyl or
15
disilyl substituted metallocenes. Another approach utilized the
reaction of Ph2RSiH (R ) H, Ph) with alkene complexes of
Cp2M (M ) Zr, Hf) or with Cp2Ti(PMe3)2 to yield the
metallocene silyl hydride complexes Cp2M(H)SiPh2R (stabilized
by the coordination of an additional PMe3 ligand).7b,16,17 As
already mentioned, metallocene silyl complexes may undergo
In this study, we describe the reactivity of selected group 4
metallocene complexes with various alkynylsilanes, stressing
the differences arising from the different nature of particular
metals and metallocene sources and from steric demands of the
alkyne substrates. For the novel hafnium complexes, we present
detailed density functional investigations concerning the reaction
mechanism of their formation.
(7) (a) Harrod, J. F.; Yun, S. S. Organometallics 1987, 6, 1381. (b)
Takahashi, T.; Hasegawa, M.; Suzuki, N.; Saburi, M.; Rousset, C. J.;
Fanwick, P. E.; Negishi, E. J. Am. Chem. Soc. 1991, 113, 8564. (c)
Corey, J. Y.; Zhu, X.-H. Organometallics 1992, 11, 672. (d) Kesti,
M. R.; Waymouth, R. M. Organometallics 1992, 11, 1095.
(8) Takahashi, T.; Bao, F.; Gao, G.; Ogasawara, M. Org. Lett. 2003, 5,
3479.
(9) (a) Verdaguer, X.; Lange, U. E. W.; Reding, M. T.; Buchwald, S. L.
J. Am. Chem. Soc. 1996, 118, 6784. (b) Tillack, A.; Lefeber, C.;
Peulecke, N.; Thomas, D.; Rosenthal, U. Tetrahedron Lett. 1997, 38,
1533.
2. Results and Discussion
(10) Halterman, R. L.; Ramsey, T. M.; Chen, Z. J. Org. Chem. 1994, 59,
2642.
To explore the interaction of group 4 metallocene alkyl
complexes with alkynyl hydrosilanes, we performed a series of
reactions using Cp2TiMe2, Cp2ZrMe2, Cp2Hf(n-Bu)2, or in situ
generated dialkyl complexes from the Cp2TiCl2/n-BuLi or
(11) (a) Hao, L.; Harrod, J. F.; Lebuis, A.-M.; Mu, Y.; Shu, R.; Samuel,
E.; Woo, H.-G. Angew. Chem., Int. Ed. 1998, 37, 3126. (b) Harrod,
J. F.; Shu, R.; Woo, H.-G.; Samuel, E. Can. J. Chem. 2001, 79, 1075.
(12) (a) Campion, B. K.; Falk, J.; Tilley, T. D. J. Am. Chem. Soc. 1987,
109, 2049. (b) Elsner, F. H.; Tilley, T. D.; Rheingold, A. L.; Geib,
S. J. J. Organomet. Chem. 1988, 358, 169.
(13) (a) Woo, H.-G.; Tilley, T. D. J. Am. Chem. Soc. 1989, 111, 3757. (b)
Woo, H.-G.; Tilley, T. D. J. Am. Chem. Soc. 1989, 111, 8043. (c)
Woo, H.-G.; Heyn, R. H.; Tilley, T. D. J. Am. Chem. Soc. 1992, 114,
5698. (d) Woo, H.-G.; Walzer, J. F.; Tilley, T. D. J. Am. Chem. Soc.
1992, 114, 7047. (e) Casty, G. L.; Lugmair, C. G.; Radu, N. S.; Tilley,
T. D.; Walzer, J. F.; Zargarian, D. Organometallics 1997, 16, 8. (f)
Sadow, A. D.; Tilley, T. D. J. Am. Chem. Soc. 2002, 124, 6814. (g)
Sadow, A. D.; Tilley, T. D. J. Am. Chem. Soc. 2003, 125, 9462.
(14) Nevertheless, compounds with group 4 metal-silicon bonds were
reported as early as the 1960s: Cardin, D. J.; Keppie, S. A.; Kingston,
B. M.; Lappert, M. F. Chem. Commun. 1967, 1035. For other
examples, see refs 11–21.
(17) Spaltenstein, E.; Palma, P.; Kreutzer, K. A.; Willoughby, C. A.; Davis,
W. M.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116, 10308.
(18) Aitken, C. T.; Harrod, J. T.; Samuel, E. J. Am. Chem. Soc. 1986, 108,
4059.
(19) (a) Aitken, C. T.; Harrod, J. T.; Samuel, E. Can. J. Chem. 1986, 64,
1677. (b) Mu, Y.; Aitken, C.; Cote, B.; Harrod, J. F.; Samuel, E. Can.
J. Chem. 1991, 69, 264.
(20) Samuel, E.; Mu, Y.; Harrod, J. F.; Dromzee, Y.; Jeannin, Y. J. Am.
Chem. Soc. 1990, 112, 3435.
(21) Hao, L.; Woo, H.-G.; Lebuis, A.-M.; Samuel, E.; Harrod, J. F. Chem.
Commun. 1998, 2013.
(22) (a) Ohff, A.; Kosse, P.; Baumann, W.; Tillack, A.; Kempe, R.; Go¨rls,
H.; Burlakov, V. V.; Rosenthal, U. J. Am. Chem. Soc. 1995, 117,
10399. (b) Peulecke, N.; Ohff, A.; Kosse, P.; Tillack, A.; Spannenberg,
A.; Kempe, R.; Baumann, W.; Burlakov, V. V.; Rosenthal, U. Chem.
Eur. J. 1998, 4, 1852.
(15) (a) Ro¨sch, L.; Altnau, G.; Erb, W.; Pickardt, J.; Bruncks, N. J.
Organomet. Chem. 1980, 197, 51. (b) Tilley, T. D. Organometallics
1985, 4, 1452.
(16) Kreutzer, K. A.; Fischer, R. A.; Davis, W. M.; Spaltenstein, E.;
Buchwald, S. L. Organometallics 1991, 10, 4031.
(23) Fan, M.-F.; Lin, Z. Organometallics 1997, 16, 494.
9
4370 J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010

Si-H Bond Activation of Alkynylsilanes
A R T I C L E S
Scheme 1. Reactions of Cp2TiMe2 with Alkynylsilanes
Figure 1. Molecular structure of compound 4. Hydrogen atoms are omitted
for clarity; thermal ellipsoids correspond to the 30% probability level.
Selected bond lengths (Å) and angles (deg): Ti-C1 ) 2.1077(14), Ti-C2
) 2.1155(13), C1-C2 ) 1.290(2), C1-Si1 ) 1.8440(15), C2-Si2 )
1.8366(14); C1-Ti-C2 ) 35.56(5), Si1-C1-C2 ) 145.07(12), C1-C2-Si2
) 139.65(12).
Cp2ZrCl2/n-BuLi system. We initially anticipated a reaction
between Si-H and the alkyl substituent on the metal which,
upon release of the corresponding alkane, would result in the
formation of a M-Si bond. Such a mechanism leading to
metallocene silyl complexes was precedented, as mentioned
above.18–20 In the case of alkynes substituted symmetrically with
two Si-H-containing groups, such reactions could yield inter-
esting metallacycles with silicon atoms and the Ct C triple bond
incorporated in the five-membered ring.24 On the other hand,
observations on the reactivity of Cp2TiMe2 with various alkynes
under thermolytic conditions by Petasis25 and Doxsee26 could
also suggest different reaction pathways for this particular
substrate. As a representative example, Cp2TiMe2 with an excess
of Me3SiC2SiMe3 gives cleanly, upon heating in benzene to
80 °C in the dark, the corresponding titanacyclobutene complex
(IV; Chart 1). The reaction proceeds presumably via the reactive
methylidenetitanocene intermediate.
Scheme 2. Suggested Reaction Mechanism for the Formation of
Products 3, 4, and 6
we suggest a stepwise process (Scheme 2), which includes the
σ-bond metathesis of Si-H and Ti-C bonds to give the silyl
complex. Subsequently, a reductive elimination takes place,27
followed by the coordination of the free alkylated alkynylsilane.
However, we do not have enough experimental evidence for
the formation of any intermediates. It is noteworthy that, when
an excess of the alkynes was used, the reactions yielded only
intractable mixtures containing probably also paramagnetic
species that we could not identify.
Reactions similar to those previously described using the
zirconium analogue Cp2ZrMe2 did not proceed at all. Only after
prolonged reaction time at the reflux temperature, the product
of partial hydrolysis (by traces of moisture), [Cp2Zr(Me)]2O,28
was detected in the reaction mixture by NMR, together with
the unchanged starting complex. Due to such a significantly
lower reactivity of Cp2ZrMe2, the corresponding hafnium
derivative was not tested in this reaction.
2.2. Reactions of Cp2MCl2/n-BuLi Systems (M ) Ti, Zr).
Additionally, metallocene complexes bearing n-butyl groups
were probed as substrates. In the case of Ti and Zr, such
Cp2M(n-Bu)2 species are stable only at low temperatures. They
are generated in situ from Cp2MCl2 and 2 equiv of n-BuLi at
-78 °C. Complicated decomposition pathways of the zirconium
derivative (also known as the Negishi reagent) with increasing
temperature were described in detail by Harrod et al.29 In a
2.1. Reactions of Cp2MMe2 (M ) Ti, Zr). Under our
experimental conditions, no products such as silyl complexes
or titanacyclobutenes were obtained or observed during the
reaction. Instead, methyl group transfer to the Si atom occurred
and the corresponding metallocene alkyne complex was formed.
Thus, the reaction of Cp2TiMe2 with 1 equiv of Me3SiC2SiMe2H
(1; Scheme 1) in n-hexane at 60 °C afforded the known complex
Cp2Ti(η2-Me3SiC2SiMe3) (3). In a similar fashion, using
Me3SiC2SiPh2H (2), the previously unknown alkyne complex
4 was isolated. This compound could be further recrystallized
from n-hexane to yield air- and moisture-sensitive amber crystals
(63%), suitable also for an X-ray structural analysis. The
molecular structure of 4 (Figure 1) reveals an expected mode
of interaction between the bent-metallocene core and the
substituted alkyne ligand. The observed structural parameters
are comparable to those found in compound 3,1b suggesting a
titanacyclopropene structure. Also, NMR and IR spectroscopic
data of 4 are in accordance with the formulated structure and
resemble those of 3.2
When a stoichiometric amount of the symmetrical alkyne
HMe2SiC2SiMe2H (5) was used, a mixture of compounds 3 and
6 in a ca. 1:1 molar ratio was obtained. Compound 6 with one
SiMe2H group participating in the agostic interaction with the
metal center was also previously described.22 Concerning the
mechanism of formation of the previously described products,
(24) An alternative approach leading to such compounds will be described
elsewhere: Lamacˇ, M.; Spannenberg, A.; Jiao, H.; Hansen, S.;
Baumann, W.; Arndt, P.; Rosenthal, U. Angew. Chem., Int. Ed., in
press.
(25) Petasis, N. A.; Fu, D.-K. Organometallics 1993, 12, 3776.
(26) Doxsee, K. M.; Juliette, J. J. J.; Mouser, J. K. M.; Zientara, K.
Organometallics 1993, 12, 4682.
(27) The reductive elimination of silyl groups from titanocene was recently
reported: Zirngast, M.; Flo¨rke, U.; Baumgartner, J.; Marschner, C.
Chem. Commun. 2009, 5538.
(28) The product was identified by NMR spectroscopy. For references, see:
Marsella, J. A.; Huffman, J. C.; Folting, K.; Caulton, K. G. Inorg.
Chim. Acta 1985, 96, 161.
(29) Dioumaev, V. K.; Harrod, J. F. Organometallics 1997, 16, 1452.
9
J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010 4371

A R T I C L E S
Lamacˇ et al.
Scheme 4. Reactions of “Cp2Hf” Sources with Alkynylsilane 1
Scheme 3. Reactions of the Cp2TiCl2/n-BuLi System with
Alkynylsilanes
simplified conception, the zirconocene 1-butene complex is
formed, which can easily generate the “free” zirconocene
fragment in reactions with various substrates. The titanium
analogue behaves similarly but is even less thermally stable,
which corresponds with our observations.
The Cp2TiCl2/n-BuLi system in the presence of stoichiometric
amounts of alkynes 1 and 5 produced the known complexes 6
and 7, respectively, upon warming the mixtures of reagents to
room temperature (Scheme 3). The “Cp2Ti” species was
probably generated already at low temperature and immediately
stabilized by coordination of the alkynes. On the other hand,
the Cp2ZrCl2/n-BuLi system facilitated the n-butyl group transfer
to alkynylsilanes 1 and 5. Unfortunately, the reactions using in
situ generated dialkyl complexes (in particular the zirconocene
system) produced various amounts of unidentified byproducts,
arguably as a result of multiple decomposition pathways of the
metallocene precursor. The isolation of analytically pure samples
of the described products was impossible because these com-
pounds (as well as the impurities) were oily solids that were
extremely soluble, even in cold hydrocarbon solvents. In general,
alkyne complexes of the type Cp2Zr[R1C2SiMe2(n-Bu)] (R1 )
Me3Si, 8; R1 ) HMe2Si, 9), in which the n-butyl group was
present at one Si atom, were major components of the reaction
mixtures, as concluded from NMR spectra and the HR ESI-
MS analysis (see the Supporting Information for details). Also,
free alkynes 10 and 17 (see below), which resulted from the
n-butyl group transfer from Cp2Zr(n-Bu)2 to one or both silicon
atoms of the substrate 1 or 5, were later identified in the mixtures
by comparison of their spectral data. A similar formal transfer
of the n-butyl group from the in situ generated Cp2Zr(n-Bu)2 to
Ph2SiH2 was previously reported.7b Formal hydrosilylation of
1-butene in the coordination sphere of Zr and subsequent
coordination of the alkyne to the metallocene fragment or a
consecutive process beginning with the Zr-C and Si-H σ-bond
metathesis13 involving the n-butyl substituent are both possible
explanations of the reaction mechanism.
2.3. Reactions of “Cp2Hf” Sources. In contrast to the Ti and
Zr n-butyl complexes, the hafnium analogue Cp2Hf(n-Bu)2 can
be prepared as a defined stable solid material.4b Owing to its
enhanced stability, higher temperatures are necessary to initiate
the reaction. In the case of this metallocene source, structures
of the reaction products turned out to be dependent on the
substitution of the alkynylsilane used, particularly on steric
properties of the substituents present on silicon. To demonstrate
these differences, we utilized an even broader set of alkynyl-
silanes. In contrast to the previous experiments, using an excess
(2.5-3 mol equiv) of the corresponding alkynylsilane proved
to be essential to obtain well-defined products, as is shown
below.
respect to 1 was used, and compound 10 was isolated for
characterization purposes from the mixture upon hydrolysis of
hafnium byproduct and subsequent chromatography on alumina.
The reduction of Cp2HfCl2 with magnesium in the presence
of 1 (Scheme 4) led also to an intractable mixture, as was the
case when Me3SiC2SiMe3 was employed.4a However, the
presence of the alkyne complex Cp2Hf(HMe2SiC2SiMe3) was
detected in the mixture by EI-MS.30 Unfortunately, we were
not able to isolate this compound without further stabilizing
ligands. When the reduction was performed in the presence of
PMe3 together with 1, the phosphine-stabilized alkyne complex
11 was formed. A highly air- and moisture-sensitive orange
crystalline material was isolated and characterized. Notably, a
mixture of two isomers (molar ratio ca. 12:1) with respect to
the mutual orientation of the alkyne moiety and the phosphine
ligand is observed in solution by NMR spectroscopy. However,
NMR as well as IR spectra show expected features for the major
isomer, such as high-field-shifted doublets of the alkyne carbon
atoms in 13C spectra (179.7 and 216.4 ppm, respectively), a
singlet at -7.4 ppm in the 31P spectrum, and bands attributable
to the Si-H stretch (2086 cm-1) and the coordinated triple bond
stretch (1562 cm-1) in the IR spectrum. The solid-state structure
of complex 11 (the preferentially crystallized major isomer) was
determined by X-ray diffraction (Figure 2). It reveals a structure
analogous to the known complex of Me3SiC2SiMe3.4a The less
sterically demanding dimethylsilyl group is expectedly oriented
toward the PMe3 ligand, while no further interaction of the Si-H
bond with the coordinatively saturated metallocene core could
be found.
An experiment performed under conditions analogous to those
previously described starting with Cp2Hf(n-Bu)2 and alkyne
PhC2SiMe2H (12) yielded a mixture of hafnacyclopentadienes
13-15 in the molar ratio ca. 5:2:1 (Scheme 5). By recrystal-
lization, we were only able to obtain a mixture of compounds
13 and 14. Hafnacyclopentadienes of this kind are known
product types of reactions between hafnocene alkyl complexes
or in situ formed “free” hafnocene and certain alkynes.31,4a
Analogous structures were also found for Ti and Zr with alkyne
12 as the substrate.22b,32 Notably, the dominant product of the
reaction using Cp2Hf(n-Bu)2 contains the n-butyl group attached
to one of the Si atoms as a consequence of the transfer from
the hafnocene precursor. NMR spectra are consistent with the
(30) See the Supporting Information for details.
Reaction of Cp2Hf(n-Bu)2 with either 1 or 3 molar equiv of
1 in toluene at 110 °C gave only mixtures of unidentified
decomposition products of the hafnocene precursor (Scheme
4). The sole identified compound was the yet unreported product
of the n-butyl group transfer to the silyl group, alkyne 10. In
an additional experiment, a slight excess of Cp2Hf(n-Bu)2 with
(31) (a) Atwood, J. L.; Hunter, W. E.; Rausch, M. D. J. Am. Chem. Soc.
1976, 98, 2454. (b) Sabada, M. B.; Farona, M. F.; Zarate, E. A.;
Youngs, W. J. J. Organomet. Chem. 1988, 338, 347. (c) Yousaf, S. M.;
Farona, M. F.; Shively, R. J.; Youngs, W. J. J. Organomet. Chem.
1989, 363, 281.
(32) Kempe, R.; Spannenberg, A.; Kosse, P.; Peulecke, N.; Rosenthal, U.
Z. Kristallogr.-New Cryst. Struct. 1998, 213, 789.
9
4372 J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010

Si-H Bond Activation of Alkynylsilanes
A R T I C L E S
removed under vacuum. The starting compound seems to be
consumed within ca. 2.5 h, while the formation of product begins
as soon as the reaction temperature is reached and then continues
slowly for approximately 3 h. In this manner, product 16 was
obtained in good yield (88%) as an orange solid, only slightly
contaminated with 17, and was further recrystallized from
n-hexane at -78 °C for analytical purposes.
In yet another experiment, Cp2HfCl2 was reduced by the
stoichiometric amount of magnesium in THF in the presence
of 3 equiv of alkyne 5 (Scheme 6). Again, compound 16 was
isolated in sufficient purity after 5 days at 60 °C, although in
slightly lower isolated yield.
To probe the behavior of expectedly more reactive secondary
silyl groups, SiPhH2, in comparison to the SiMe2H substituents
used throughout this work, we made use of another symmetrical
alkynylsilane, H2PhSiC2SiPhH2 (18). Similarly to the previous
reactions, 18 reacted with Cp2Hf(n-Bu)2 to give the metallacycle
19 in 84% yield (Scheme 6) as a racemic mixture with the chiral
center present at the silicon atom that is incorporated into the
five-membered ring. Compound 19 was also prepared via the
reduction of Cp2HfCl2 with magnesium in THF (Scheme 6, 64%
yield).
Figure 2. Molecular structure of compound 11. Atoms with higher
occupancy are shown for the disordered SiMe3 group. Hydrogen atoms are
omitted for clarity, except for H1. Thermal ellipsoids correspond to the
30% probability level. Selected bond lengths (Å) and angles (deg): Hf-C1
) 2.254(6), Hf-C2 ) 2.196(7), C1-C2 ) 1.296(9), C1-Si1 ) 1.852(7),
C2-Si2 ) 1.863(8); C1-Hf-C2 ) 33.8(2), C1-C2-Si2 ) 141.4(6),
C2-C1-Si1 ) 130.0(6).
The spectroscopic features of compounds 16 and 19 are
comparable, indicating a similar structural framework. In the
1H NMR spectra of 16, three doublets of SiMe2H Me-groups
and one singlet of the quarternary silyl group can be identified
in the expected range. Three resonances of the silyl-bound
protons appear also with δ values typical for free dimethylsilyl
groups having no interaction with the metal center. The same
situation in the case of compound 19 gives rise to a complex
set of signals in this region due to mutually interacting
nonequivalent Si-H protons. A doublet due to the proton of
the exocyclic double bond can be found at 7.31 ppm for 16
and at 7.55 ppm for 19, respectively. The intrinsic chirality of
the latter complex is demonstrated also by the presence of two
signals for cyclopentadienyl groups. Carbon resonances of the
hafnasilacyclopentene ring appear as three distinct high-field-
shifted signals: for 16, 179.7 (C2; see Figure 3 for numbering),
244.6 (C3), and 263.4 (C1) ppm, respectively; for 19, 171.9
(C2; see Figure 4 for numbering), 242.4 (C3), and 263.0 (C1)
ppm, respectively. The signal at 159.8 (16) or 157.4 ppm (19)
belongs to the exocyclic sp2 carbon C4. All four silicon
resonances for both compounds appear in the 29Si spectra, with
unexceptional values ranging from -23.3 to -52.7 ppm for 16
or from -47.9 to -65.7 ppm for 19. For novel compounds,
including 16 and 19, we managed to determine homo- and
heteronuclear coupling constants involving silicon. Such data
are rarely reported and are thus presented in the Supporting
Information (Table S5). The IR spectrum of 16 shows three
bands at 2137, 2115, and 2080 cm-1, respectively, attributable
to the Si-H bond vibrations, which are comparable to those of
the free alkyne 5.22b Analogously, the IR spectrum of 19 displays
bands in the region of free Si-H bond vibrations (2145-2065
cm-1). The observed frequencies are only slightly shifted to
lower energies compared to those of the free alkyne 18.33
The solid-state structures of 16 (Figure 3) and rac-19 (Figure
4) were corroborated by single-crystal X-ray diffraction studies.
The hafnasilacyclopentene ring resembles the structural motif
of compound I (Chart 1), although in the case of the described
compounds, additional substituents are present, and the mech-
formulated structures, and furthermore, molecular ions for
compounds 13 and 14 were observed in EI mass spectra and
the composition of [M + H]+ species was corroborated with
the aid of the HR ESI-MS analysis.
An experiment utilizing the reduction of Cp2HfCl2 with Mg
in THF at 60 °C was subsequently performed (Scheme 5). In
the presence of alkyne 12, metallacyclic products 14 and 15
were obtained in the molar ratio ca. 1:4.5. When the solvent
was changed to toluene and the mixture was heated to 110 °C,
isomerization of 15 to 14 was observed, which led to the 14:15
molar ratio of ca. 15:1 after 48 h. A similar behavior was
observed also for Ti and Zr,22b,32 whereas the initially formed
unsymmetrically substituted kinetic product (in our case rep-
resented by 15) isomerizes within longer reaction times to the
thermodynamically stable symmetrical product, analogous to
14. The symmetrically substituted product with SiMe2H groups
in ꢀ positions of the hafnacyclopentadiene ring was not formed,
presumably due to steric reasons, which is in line with the
previous observations involving other group 4 metals. Com-
pound 14 was isolated in a sufficiently pure state to perform
elemental analysis, and both 14 and 15 gave consistent NMR
spectra and were successfully analyzed using MS techniques.
In contrast to previous cases, the use of alkyne 5 resulted in
a markedly different behavior. Cp2Hf(n-Bu)2 reacted with 3
equiv of 5 in toluene at 110 °C to produce the unusual
metallacycle 16 (Scheme 6). When only 1 equiv of the alkyne
was used, 16 was also the sole identified Hf-containing product.
A portion of the alkyne was, however, transformed into the
doubly n-butylated derivative 17, which was found in the
resulting mixture by analogy to the reaction utilizing 1. Similarly
as for 10, an excess of Cp2Hf(n-Bu)2 was used to prepare the
previously unreported compound 17 for characterization pur-
poses. The concurrent process leading to this byproduct can be
effectively suppressed by using an excess of 5 (3 equiv).
According to the in situ NMR observation of this reaction (1H
NMR performed in toluene-d8 at 100 °C in 5 min intervals),
the alkynylsilane still acts as an acceptor of n-Bu groups from
the hafnocene substrate, but under such conditions, the monoalky-
lated derivative is preferentially formed, which can be later
(33) Ru¨dinger, C.; Bissinger, P.; Beruda, H.; Schmidbaur, H. Organome-
tallics 1992, 11, 2867.
9
J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010 4373

A R T I C L E S
Lamacˇ et al.
Scheme 5. Reactions of “Cp2Hf” Sources with Alkynylsilane 12
Scheme 6. Reactions of “Cp2Hf” Sources with Alkynylsilanes 5 and 18
anism leading to this structure is undoubtedly different. In the
16, 1.346(5) Å for 19), which adopts the E configuration with
the SiMe2H or SiPhH2 group pointing out from the metallocene
core. Other bonds fall into the range typical for single bonds of
the corresponding type. For example, distances between hafnium
and the ring carbons C1 and C3 are 2.251(2) and 2.225(3) Å
case of compound 19, the racemic mixture of enantiomers gave
j
rise to a structure conformable with the centrosymmetric P1
space group. The C1-C2 bond in both complexes clearly has
a double-bond character, which is demonstrated by the lengths
of 1.362(3) Å (for 16) and 1.360(4) Å (for 19), respectively.
The same is true of the exocyclic C3-C4 bond (1.351(4) Å for
Figure 3. Molecular structure of compound 16. Only one position of the
disordered cyclopentadienyl rings is shown. Hydrogen atoms are omitted
for clarity, except for Si-H protons (H1, H2, H4B), and H4A. Thermal
ellipsoids correspond to the 30% probability level. Selected bond lengths
(Å) and angles (deg): Hf-C1 ) 2.251(2), Hf-C3 ) 2.225(3), C1-C2 )
1.362(3), C2-Si3 ) 1.892(3), C3-Si3 ) 1.880(3), C3-C4 ) 1.351(4);
C1-Hf-C3 ) 91.08(10), Hf-C1-C2 ) 115.1(2), C1-C2-Si3 )
120.0(2), C2-Si3-C3 ) 107.37(12), Hf-C3-Si3 ) 106.29(14), Hf-C3-C4
) 131.1(2).
Figure 4. Molecular structure of rac-19 (the S enantiomer is shown). Only
one position of the disordered Cp ring C34-C38 is shown. Hydrogen atoms
are omitted for clarity, except for Si-H protons (H1A, H1B, H2A, H2B,
H3, H4B, H4C) and H4A. Thermal ellipsoids correspond to the 30%
probability level. Selected bond lengths (Å) and angles (deg): Hf-C1 )
2.253(3), Hf-C3 ) 2.240(3), C1-C2 ) 1.360(4), C2-Si3 ) 1.886(3),
C3-Si3 ) 1.878(3), C3-C4 ) 1.346(5); C1-Hf-C3 ) 88.90(11),
Hf-C1-C2 ) 116.9(2), C1-C2-Si3 ) 118.7(2), C2-Si3-C3 )
107.04(13), Hf-C3-Si3 ) 106.8(2), Hf-C3-C4 ) 130.2(2).
9
4374 J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010

Si-H Bond Activation of Alkynylsilanes
A R T I C L E S
Scheme 7. Proposed Mechanism of Formation of Compounds 16
and 19
172.82(12)°. The five-membered ring Hf-C1-C2-Si3-C3 in
19 is slightly deformed (e.g., the torsion angles C1-Hf-C3-Si3
and Hf-C1-C2-Si3 are 11.0(1) and 8.7(3)°, respectively).
Pendant phenyl substituents of SiPhH2 groups are twisted from
the plane of the central five-membered ring in order to meet
steric demands. Phenyl rings C17-C22 and C23-C28 are
involved in a weak intramolecular π-π interaction (distance
of ring centroids 3.93 Å, dihedral angle between ring planes
14.51°, slip angles34 31.68, 21.61°), while phenyl groups
C5-C10 of two adjacent molecules in the crystal structure
interact similarly, being in a virtually coplanar orientation
(distance of ring centroids 3.98 Å, slip angle 30.22°).
3. Mechanistic Considerations
Concerning the mechanism of formation of metallacycles 16
and 19, we suggest that both Cp2Hf(n-Bu)2 (at 110 °C) and the
Cp2HfCl2/Mg system generate the “free” hafnocene that is
immediately stabilized by the coordination of the corresponding
alkyne. The resulting metallocene alkyne complex could be
similar to those reported22 for Ti and Zr with the trans
configuration of the coordinated alkyne and an agostic Si-H-M
interaction (A in Scheme 7). Cp2Hf(n-Bu)2 is expected to
eliminate n-butane under reaction conditions upon the formation
ofanalkenecomplex.Thiscan,indeed,undergothealkene-alkyne
exchange, affording intermediate A. At this point, also the
concurrent processsa formal hydrosilylation of 1-butenescould
take place, leading to the observed side product 17.35
for 16 and 2.253(3) and 2.240(3) Å for 19, respectively, which
is comparable to distances found in I.5 The five-membered ring
Hf-C1-C2-Si3-C3 in 16 is nearly planar (mean deviation
of 0.02 Å), and also the C4 sp2 carbon atom and silicon atoms
Si1, Si2, and Si4 do not markedly deviate from this plane. Only
the SiMe2H group on C2 is slightly bent out from the plane of
the metallacycle; the torsion angle Hf-C1-C2-Si2 is -
Subsequent activation of the Si-H bond in A is a prerequisite
for the reorganization leading to the main product. In this
Scheme 8. BP86 Computed Bond Distances (Å), Gibbs Free Energies (∆G, kcal/mol), and Enthalpies (∆H, kcal/mol, in Parentheses) for the
Isomerization Reaction of the Suggested Intermediate A to Ca
a See Scheme 7. Cyclopentadienyl rings on Hf were omitted for simplicity.
9
J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010 4375

A R T I C L E S
Lamacˇ et al.
Scheme 9. BP86 Computed Bond Distances (Å), Gibbs Free Energies (∆G, kcal/mol), and Enthalpies (∆H, kcal/mol, in Parentheses) for the
Insertion Reaction of the Second Molecule of the Alkyne Substrate to Intermediate Ca
a See Scheme 7. Cyclopentadienyl rings on Hf were omitted for simplicity.
manner, a silyl hydride complex (B) should be formed, which
could transform into a hafnasilacyclopropane structure (C).
We also cannot exclude the direct formation of intermediate B
by the oxidative addition of alkynylsilane to the “free” metal-
locene species. A structural motif similar to that of C was
observed in the structure of the ruthenium complex reported
by Jones36 or in the species proposed by Mori and co-workers
as an intermediate during reactions of Cp2ZrCl2 with
Me2PhSiLi.37 Under these conditions, the presence of a
zirconium-silene complex (or silazirconacyclopropane) was
anticipated and such ideas were supported by further reactions
with alkynes that yielded silazirconacyclopentenes via insertion
of the alkyne into the Zr-Si bond. However, these compounds
were observed only spectroscopically and the mechanism was
deduced from products of the subsequent hydrolysis, which
yielded the correspondingly substituted vinylsilanes. Obviously,
a similar process of insertion could lead in our case to the
formation of the isolated products 16 and 19. The in situ NMR
monitoring (toluene-d8, 100 °C, 5 min interval) of the reaction
leading to 16 revealed the formation of one major intermediate
with a resonance for Cp protons at 4.87 ppm. According to Mori
et al., the zirconocene silene complex exhibited a signal for Cp
protons at lower chemical shifts compared to the zirconacyclo-
pentyne product.37a On the basis of this analogy, silene complex
C could be present in our system. The concentration of the
observed species culminated after ca. 40 min, when it reached
approximately 30% of the sum of molar concentrations of the
reactant and the product. Then the signal gradually dropped
away as the product was formed. In addition, a small signal at
-3.94 ppm (approximately 5% intensity of the latter signal)
was present, which can be explained by the silyl group proton
in an agostic interaction with the metal possibly belonging to
(34) Angles between centroid-centroid vector and normal to the respective
plane.
(35) Metathesis of σ-bonds involving the starting Cp2Hf(n-Bu)2 is another
possible pathway. Such transfer of a methyl group from Hf complexes
to PhSiH3 has also been reported: Sadow, A. D.; Tilley, T. D.
Organometallics 2003, 22, 3577.
(36) Yin, J.; Klosin, J.; Abboud, K. A.; Jones, W. M. J. Am. Chem. Soc.
1995, 117, 3298.
(37) (a) Mori, M.; Kuroda, S.; Dekura, F. J. Am. Chem. Soc. 1999, 121,
5591. (b) Kuroda, S.; Dekura, F.; Sato, Y.; Mori, M. J. Am. Chem.
Soc. 2001, 123, 4139. (c) Mori, M. Top. Organomet. Chem. 2005,
10, 41.
9
4376 J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010

Si-H Bond Activation of Alkynylsilanes
A R T I C L E S
Table 1. Crystallographic Data for 4, 11, 16, and 19
4
11
16
rac-19
formula
Mr
C28H32Si2Ti
472.62
C20H35HfPSi2
541.12
C22H38HfSi4
593.37
C38H38HfSi4
785.53
cryst syst
space group
T (K)
triclinic
P1
200(2)
monoclinic
P21/c
200(2)
monoclinic
C2/c
200(2)
triclinic
P1
200(2)
j
j
a (Å)
b (Å)
c (Å)
8.5394(4)
10.3251(5)
16.5699(8)
94.324(4)
101.046(3)
114.318(3)
1287.09(11)
2
16.8109(5)
10.2196(3)
14.9047(5)
90
109.929(2)
90
2407.30(13)
4
1.493
32.0481(10)
9.1762(2)
20.1421(7)
90
113.035(2)
90
5451.1(3)
8
1.446
8.1828(2)
10.4511(3)
20.5913(6)
88.056(2)
84.173(2)
80.283(2)
1726.47(9)
2
R (deg)
ꢀ (deg)
γ (deg)
V (Å3)
Z
Dcalcd (g cm-3
)
1.219
0.44
1.511
3.19
µ(Mo KR) (mm-1
cryst size (mm)
color, habit
)
4.50
4.01
0.37 × 0.34 × 0.12
yellow, prism
0.809-0.922
24 788/6939
4795
0.20 × 0.15 × 0.08
orange, prism
0.431-0.735
31 646/4476
3158
0.47 × 0.20 × 0.17
yellow, prism
0.316-0.576
40 215/5776
4264
0.42 × 0.23 × 0.03
yellow, plate
0.520-0.926
26 806/7322
6175
transmissn coeff
no. of total/unique rflns
no. of obsd rflns (I > 2σ(I))
Rint (%)
2.75
7.65
2.83
3.42
no. of params
284
184
207
366
R1 (I > 2σ(I)) (%)
wR2 (all data) (%)
GOF on F2
3.22
7.57
0.86
0.35, -0.25
3.63
6.97
0.97
1.58, -0.87
1.99
4.33
0.89
0.70, -0.66
2.46
5.43
0.91
0.98, -0.61
∆F (e Å-3
)
the hafnocene alkyne complex Cp2Hf(HMe2SiC2SiMe2H). We
performed additional NMR measurements at room temperature
of the mixture that was heated to 110 °C for 40 min, and
we have observed an analogous composition of the mixture,
while the Cp signal appeared as a broad signal at 4.86 ppm and
the Si-H signal moved slightly to -4.09 ppm. Unfortunately,
in the 29Si spectra, we were not able to find corresponding
signals, possibly due to certain broadening, which did not allow
us to unambiguously prove the presence of the suggested
intermediates.
the final product (represented by D1 in Scheme 9). The origin
of the presented energetic values is A1 and H3SiC2SiH3 as the
substrate.
A1 is more stable than the corresponding hafnocene alkyne
complex without the agostic interaction (having a cis configu-
ration of the coordinated alkyne), which is in agreement with
the behavior of the corresponding Ti and Zr complexes from
the review of both experiments22 and computations (see ref 23
and the Supporting Information). However, we could not find
the corresponding intermediate with the broken H-Si bond, and
optimizations for this species led back to A1, which can be
explained by the charge separation between the negatively
charged hydride and the positively charged SiH2 group. In
addition, attempts to locate the corresponding species with only
H-Si bond coordination leading to an oxidative addition
resulted in the structure B1, which is higher in energy than A1
by 10.93 kcal/mol. Such hafnocene silyl hydride complexes have
been described, and the observed structural parameters for
Cp*CpHf(H)[SiH(SiMe3)2] (Hf-Si ) 2.744(1) Å),13e the solid-
state structure of which was determined, correspond with our
model. However, we also could not find the corresponding
transition state from A1 to B1, and this process can be
considered as the only energy-raising process. The next step
from B1 to B2 can be considered as the coordination of the
silapropargyl moiety of B1, while B2 is higher in energy than
A1 by 5.69 kcal/mol. It is interesting to note that B2 has hydride
and silaallenyl coordination and can be considered as a rotational
isomer of A1 by rotating the silaallene group by approximately
180°.
While compounds 16 and 19 clearly represented the preferred
structure type for alkynes 5 and 18, respectively, the unsym-
metrical alkyne 1 did not give such products. Presumably, the
steric bulk of one SiMe3 group precludes the formation of the
corresponding metallacycle, as is possible in the presence of
two less sterically demanding SiMe2H or SiPhH2 groups. We
can speculate that the insertion step to the Hf-Si bond is
impossible for this substrate, while the intermediate alkyne
complex is not sufficiently stable to be isolated.
4. Computational Analysis
Along with the experimental investigations we carried out
density functional theory calculations for understanding the
reaction mechanism proposed in Scheme 7. Instead of the
substituted alkynylsilane substrates, we used the parent
H3SiC2SiH3 as a model substrate. This model is appropriate,
since both alkynylsilanes 5 and 18 formed the same structure
pattern in products. In our discussion we have used the lettering
system A-D, conformable with the previous usage (Scheme
7), for the stationary points on the potential energy surface
(Schemes 8 and 9).
As proposed in Scheme 7, the reaction has two basic steps
toward the formation of products: (1) the formation of the alkyne
complex A along with its isomerization to the corresponding
silene complex C (cf. Scheme 8, structures A1-C3), and (2)
the selective insertion of the second alkyne molecule to afford
The next step is a hydride transfer from B2 through the
transition state B3-TS to C1; the computed barrier of 2.47 kcal/
mol for this process is rather low. An agostic C-H-Hf
interaction is present in C1. The subsequent step is the breaking
of this agostic interaction and the bending out of the vinyl group
(through the transition state C2-TS), resulting in the formation
of a silene complex (C3). As shown in Scheme 8, the
transformation from A1 to C3 raises the energy by only 5.58
9
J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010 4377

A R T I C L E S
Lamacˇ et al.
kcal/mol and the single step with the highest energy is about
11 kcal/mol. All this indicates a slightly endogenic process with
moderate energy barriers.
steric bulk of the SiMe3 group prevented the formation of
metallacyclic products from two alkyne molecules. The hafnocene
alkyne complex of 1 was isolable only after utilizing the
additional stabilizing PMe3 ligand. On the other hand, sym-
metrical alkynes 5 and 18 provided novel five-membered Hf-
and Si-containing metallacycles with an exocyclic double bond
(16 and 19, respectively). DFT calculations confirmed the
plausibility of the proposed intermediates and some of the
elementary steps in the suggested mechanism of their formation.
The latter type of reactivity revealed, in comparison to Ti and
Zr, a significant difference in the hafnium chemistry arising from
the possible exceptional rearrangement after the Si-H bond
activation in alkynylsilanes.
Postulating the formation of silene complex C3, we were
furthermore interested in the selective insertion of a second
alkyne molecule leading to the corresponding products. As
shown in C3, there are two possibilities for the alkyne insertion:
one into the Hf-Si bond and the other into the Hf-C bond. It
should be noted that the insertion process into both Hf-Si and
Hf-C bonds is a single-step process leading directly to the
product and does not form any intermediates as a stepwise
process. As shown in Scheme 9, the insertion into the Hf-Si
bond via C4-TS-1 has an activation free energy lower than that
into the Hf-C bond via C4-TS-2 (27.79 vs 31.25 kcal/mol),
while the corresponding insertion product D2 is more stable
than D1 (-13.96 vs -12.72 kcal/mol). The difference (3.46
kcal/mol) in activation free energy confirms the selective
formation of D1, as found experimentally (Scheme 7), and the
process is kinetically controlled. The enhanced thermodynamic
stability of D2 over D1 is due to the formation of an additional
agostic interaction. Calculated structural parameters for the
product D1 are in good agreement with the experimentally
obtained values for both 16 (Figure 3) and 19 (Figure 4).
6. Experimental Section
6.1. General Considerations. All manipulations were carried
out under an atmosphere of argon using standard Schlenk tech-
niques. Prior to use, solvents (including deuterated) were predried,
freshly distilled from sodium tetraethylaluminate, and stored under
argon. Cp2MCl2 (M ) Ti, Zr, Hf) and Cp2TiMe2 (5% solution in
toluene) were purchased from MCAT (Metallocene Catalysts &
Life Science Technologies, Konstanz, Germany). Alkyne 12 was
purchased from ABCR. Cp2ZrMe2,40 Cp2Hf(n-Bu)2,4b and alkynes
1,41a 2,41a 5,41b and 1833 were synthesized according to published
procedures. The following instruments were used. NMR: Bruker
AV 400 and AV 300 spectrometers were used; chemical shifts (1H,
13C, 29Si) are given relative to SiMe4 and are referenced to residual
solvent signals (benzene-d6, δH 7.16, δC 128.0; toluene-d8, δH 2.03,
δC 20.4; THF-d8, δH 1.73, δC 25.2). Where necessary, signals were
assigned with the aid of 2D NMR experiments: COSY, HMQC,
HMBC, NOESY. MS: MAT 95-XP (EI- and CI-MS) and Agilent
Technologies 6210 (TOF LC-MS) and 6890N+5973N (GC-MS).
IR: Bruker Alpha-P. Elemental analyses: Leco TruSpec Micro
CHNS. Melting points: Bu¨chi 535 apparatus, capillaries sealed
under Ar.
6.2. X-ray Structure Determination. Crystals suitable for
measurement were obtained from n-hexane at -40 °C (4), by slow
evaporation of an n-hexane solution (11 and 19) or benzene-d6
solution (16) under a stream of argon. Diffraction data were
collected on a STOE IPDS II diffractometer using graphite-
monochromated Mo KR radiation. The structures were solved by
direct methods (SHELXS-9742) and refined by full-matrix least-
squares techniques on F2 (SHELXL-9742). For all compounds a
numerical absorption correction was applied. For compound 11,
both Cp rings in the structure were treated as rigid groups and the
disordered SiMe3 group was modeled over two positions. The
structures of compounds 16 and 19 contained two disordered
cyclopentadienyl ligands. For 16, each of them was successfully
modeled over two positions. In the case of 19, one was treated as
a rigid group (C29-C33), while the other was modeled over two
positions (C34-C38). All fully occupied non-hydrogen atoms were
refined anisotropically. Hydrogen atoms (except H1 of compound
11, H1, H2, H4A, and H4B of 16, and H1A, H1B, H2A, H2B, H3,
H4B, and H4C of 19, which were located in the difference map)
are placed in idealized positions and refined using a riding model.
The DIAMOND43 program was used for graphical representations.
Crystallographic data for compounds 4, 11, 16, and 19 are given
in Table 1.
5. Conclusions
We have demonstrated a diversity in the behavior of
alkynylsilanes in reactions with metallocene alkyl complexes
as well as with in situ generated “free” metallocenes. Depending
on the metal, substituents of the metallocene precursor, and
substituents on the alkyne, respectively, different product types
were formed. Processes during these reactions include the Si-H
bond activation of the silane that subsequently facilitates the
creation of new Si-C bonds. On the other hand, interaction of
the alkyne with the metallocene core is an important phenom-
enon. In reactions with alkynylsilanes, Ti and Zr obviously
preferred the alkyne coordination. Metallocene alkyne com-
plexes were thus obtained, some of which were synthesized
previously in a different manner. Nevertheless, the typical
feature of the reactions utilizing Cp2TiMe2 was methyl group
transfer to the Si atom, which unequivocally demonstrates the
previous Si-H bond activation.38 When the Cp2MCl2/n-BuLi
(M ) Ti, Zr) systems were used, an analogous transfer of n-butyl
groups was detected only in the case of Zr. This process,
however, can also be viewed as a hydrosilylation reaction, since
the presence of metallocene alkene complexes can be anticipated
under such reaction conditions.39 It should be noted that the
Cp2ZrCl2/n-BuLi system yielded inseparable mixtures of prod-
ucts and therefore turned out to be an inconvenient “Cp2Zr”
source in this particular case in comparison to the Cp2ZrCl2/
Mg system.22
In contrast to Ti and Zr, Hf preferred the formation of
metallacycles in most cases. However, steric features of the
substituents on the alkyne substrate played a key role and, in
some cases, n-butyl group transfer to Si was also observed. The
less sterically demanding alkyne 12 facilitated the coupling of
two substrate molecules to give exclusively hafnacyclopenta-
dienes as known types of products. In the case of alkyne 1, the
(40) Wailes, P. C.; Weigold, H.; Bell, A. P. J. Organomet. Chem. 1972,
34, 155.
(41) (a) Ruffolo, R.; Decken, A.; Girard, L.; Gupta, H. K.; Brook, M. A.;
McGlinchey, M. J. Organometallics 1994, 13, 4328. (b) Wrackmeyer,
B.; Kehr, G.; Su¨ss, J.; Molla, E. J. Organomet. Chem. 1998, 562,
207.
(42) Sheldrick, G. M. Acta Crystallogr. 2008, A64, 112.
(43) Brandenburg, K. DIAMOND, Version 3.1e; Crystal Impact GbR,
Bonn, Germany, 2007.
(38) A similar methyl transfer process from Cp2TiMe2 to silanes was briefly
mentioned in the studies of Harrod and co-workers.11b,18
(39) We are aware of the fact that such an explanation is speculative.
Notably, the decomposition of Cp2Zr(n-Bu)2 was shown upon detailed
analysis to be a complicated set of reactions rather than simple
elimination of butane and the formation of a butene complex.29
9
4378 J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010

Si-H Bond Activation of Alkynylsilanes
A R T I C L E S
6.3. Reactions of Cp2TiMe2 with Alkynylsilanes: General
Procedure. A commercial 5% toluene solution (3.0 mL) of
Cp2TiMe2 was evaporated to dryness in the dark, and the solid (145
mg, 0.70 mmol) was dissolved in n-hexane (10 mL). The corre-
sponding alkyne (0.77 mmol) was introduced, and the resulting
orange solution was heated to reflux for 1.5 h, during which time
the color changed to dark amber. After it was cooled to room
temperature, the solution was evaporated under vacuum, the solid
residue extracted with n-hexane (all solids were readily soluble),
the extract filtered, and the solvent removed in vacuo, leaving an
amber solid.
Thus, the reaction of alkyne 1 afforded complex 3 of >90% purity
in virtually quantitative yield, while the reaction of 5 yielded a
mixture of 3 and 6 in a ca. 1:1 ratio, together with some minor
unidentified compounds (up to 10 mol %). The reaction products
were not further purified. NMR data for complexes 32c and 64b
were in agreement with those previously reported. Compound 4
was recrystallized from n-hexane at -40 °C to give dark amber
crystals. Yield: 208 mg (63%). Mp: 118-120 °C dec. Anal. Calcd
for C28H32Si2Ti (472.59 g mol-1): C, 71.16; H, 6.83. Found: C,
71.04; H, 7.04. NMR (400 MHz, 297 K, benzene-d6): 1H, δ -0.36
(s, 9 H, SiMe3), 0.11 (s, 3 H, SiMePh2), 6.35 (s, 10 H, C5H5),
7.12-7.25 (m, 10 H, SiMePh2); 13C{1H}, δ -1.4 (SiMePh2), 0.4
(SiMe3), 117.8 (C5H5), 127.9, 129.2, 134.8 (3 × SiMePh2 CH),
138.4 (SiMePh2 Cipso), 236.0 (CSiPh2Me), 248.2 (CSiMe3);
29Si{1H}, δ -21.9 (SiPh2Me), -14.3 (SiMe3). IR (Nujol): ν˜/cm-1
1648 (m, Ct C). MS (CI, isobutane): m/z 472 [M]+, 295
[Me3SiC2SiMe2C5H6N + H]+.
6.4. Reaction of the Cp2TiCl2/n-BuLi System with Alkynes
1 and 5. A solution of Cp2TiCl2 (100 mg, 0.40 mmol) in THF (10
mL) was cooled in a dry ice-ethanol bath to -78 °C, and n-BuLi
(0.50 mL of 1.6 M solution in hexanes, 0.80 mmol) was slowly
introduced. After the mixture was stirred for 10 min at -78 °C,
the alkyne 1 (70 mg, 0.45 mmol) was added to the deep red mixture,
which was then warmed slowly to room temperature and stirred
for 3 h, while the color changed to dark brown. All volatiles were
evaporated under vacuum, and the residue was extracted with
n-hexane (15 mL). Removal of the solvent afforded a dark amber
residue, which was identified by NMR as complex 6 accompanied
by some minor unidentified byproducts (up to 10 mol %).
3JHH ) 3.6 Hz, 6 H, Me2 at Si4), 0.28 (d, 3JHH ) 3.8 Hz, 6 H, Me2
at Si2), 0.37 (d, 3JHH ) 3.8 Hz, 6 H, Me2 at Si1), 0.47 (s, 6 H, Me2
3
3
at Si3), 3.80 (hept, JHH ) 3.8 Hz, 1 H, Si1-H), 4.37 (hept, JHH
3
) 3.8 Hz, 1 H, Si2-H, and dhept, JHH ) 3.6, and 7.1 Hz, 1 H,
Si4-H), 5.78 (s, 10 H, C5H5), 7.31 (d, 3JHH ) 7.1 Hz, 1 H, C4-H);
13C{1H}, δ -2.9 (Me at Si42), -0.4 (Me2 at Si2), -0.3 (Me2 at
Si1), 2.5 (Me2 at Si3), 111.2 (C5H5), 159.8 (C4), 179.7 (C2), 244.6
(C3), 263.4 (C1); 29Si{1H}, δ -52.7 (Si3), -40.5 (1JSiH ) 171 Hz,
Si1), -30.0 (1JSiH ) 185 Hz, Si4), -23.3 (1JSiH ) 182 Hz, Si2). IR
(Nujol): ν˜/cm-1 2137, 2115, 2080 (3 × m, Si-H). MS (CI,
isobutane): m/z 594 [M]+, 283 [C12H27Si4]+.
6.6. Preparation of Complex rac-19. Method a. In an analogy
to the preparation of 16, alkyne 18 (169 mg, 0.71 mmol) and
Cp2Hf(n-Bu)2 (100 mg, 0.24 mmol) were reacted in toluene (10
mL). Crude compound 19 was isolated as an orange solid
contaminated with 18. The crystalline 19 was subsequently washed
with n-pentane to remove the impurities. Yield: 159 mg (84%).
Method b. According to the same procedure as for 16, Cp2HfCl2
(190 mg, 0.50 mmol), alkyne 18 (358 mg, 1.50 mmol), and
magnesium turnings (13 mg, 0.53 mmol) reacted in THF (10 mL)
at 60 °C for 2 days to give the crude crystalline 19 contaminated
with 18, which was treated in the same manner as above or
recrystallized from n-hexane at -78 °C. Yield: 251 mg (64%). Mp:
130-131 °C. Anal. Calcd for C38H38HfSi4 (785.54 g mol-1): C,
58.10; H, 4.88. Found: C, 58.10; H, 5.16. NMR (400 MHz, 297 K,
benzene-d6; numbering according to atom labeling in the molecular
structure, see Figure 4): 1H, δ 4.66 (dd, 2JHH ) 5.7 Hz, 3JHH ) 3.9
Hz, 1 H, Si4-H), 4.78 (d, 2JHH ) 5.5 Hz, 1 H, Si1-H), 4.85 (dd,
3
2
2JHH ) 5.7 Hz, JHH ) 4.7 Hz, 1 H, Si4-H), 4.95 (d, JHH ) 5.8
2
Hz, 1 H, Si2-H), 4.98 (d, JHH ) 5.5 Hz, 1 H, Si1-H), 5.07 (d,
2JHH ) 5.8 Hz, 1 H, Si2-H), 5.40 (d, 4JHH ) 1.4 Hz, 1 H, Si3-H),
3
5.73, 5.76 (2× s, 10 H, C5H5), 7.55 (ddd, JHH ) 3.9 and 4.7 Hz,
4JHH ) 1.4 Hz, 1 H, C4-H), 7.03-7.88 (m, 20 H, Ph); 13C{1H},
δ 111.6, 112.2 (2 × C5H5), 128.1, 128.3, 128.5, 129.4, 129.6, 129.8,
133.6, 133.7, 135.7, 136.2, 136.6, 136.8 (12 × Ph), 157.4 (C4),
171.9 (C2), 242.4 (C3), 263.0 (C1), signals of four Ph groups in
the range between 128.0 and 137.0 are partially overlapped and
are not unambiguously assigned; 29Si{1H}, δ -65.7 (Si3), -55.9
(Si1), -50.4 (Si4), -47.9 (Si2). IR (Nujol): ν˜/cm-1 2145, 2125 (2
× m, Si-H), 2092 (s, Si-H), 2065 (m, Si-H). MS (CI, isobutane):
m/z 785 [M]+.
In the same manner, the reaction utilizing alkyne 5 (63 mg, 0.44
mmol) yielded compound 7 (purity varied between ca. 80 and 90%).
Compounds 6 and 7 were identified by the comparison of NMR
spectra with previously published data.22b Owing to the high
solubility of impurities as well as 6 and 7, and their reluctance to
crystallize under such conditions, the reaction products could not
be isolated in a pure state.
6.5. Preparation of Complex 16. Method a. Alkyne 5 (252
mg, 1.77 mmol) was added to a solution of Cp2Hf(n-Bu)2 (250
mg, 0.59 mmol) in toluene (15 mL) at room temperature. The
mixture was heated to 110 °C for 7 h. After the mixture was cooled
to room temperature, all volatiles were removed under vacuum to
leave an orange solid, which was identified as compound 16 of ca.
95% purity, contaminated only by traces of 17. Yield: 309 mg
(88%). The product can be further recrystallized from a concentrated
n-hexane solution at -78 °C.
6.7. Computational Details. Our computations have been
carried out at the BP86 level of density functional theory as
implemented in the Gaussian 03 program.44 We have used the
standard 6-311+G(d,p) basis set for H, C, and Si atoms and the
LANL2DZ basis set for Ti, Zr, and Hf.45 All structures have been
optimized in C1 symmetry without any constraints, and all optimized
structures have been characterized either as energy minima without
imaginary frequencies or as transition states with only one imaginary
frequency by frequency calculations at the same level. For
discussion we have used the computed Gibbs free energies (∆G)
at 298 K and the computed enthalpies (∆H, 298 K) that are given
for comparison. For comparison we have also carried out MP2
single-point energy calculations by using the same basis sets and
the BP86 optimized geometries.
Our benchmark calculations on the acetylene complex with and
without H-Si agostic interactions show that BP86 not only
reproduces structural parameters very well but also gives the correct
relative energies; i.e., BP86 favors the structure with an agostic
interaction by ∆G ) -4.01 kcal/mol (∆H ) -4.92 kcal/mol) for
M ) Zr and ∆G ) -3.06 kcal/mol (∆H ) -3.55 kcal/mol) for M
) Hf. However, for M ) Ti, the structure with an agostic interaction
is less stable by ∆G ) 3.33 kcal/mol (∆H ) -0.08 kcal/mol,
indicating the entropy effect). MP2 single-point energy calculations
including the thermal correction to Gibbs free energy give the
correct energetic order; i.e., the structure with an agostic interaction
Method b. Cp2HfCl2 (150 mg, 0.40 mmol), alkyne 5 (145 mg,
1.02 mmol), and magnesium turnings (10 mg, 0.41 mmol) were
suspended in THF (10 mL), and the mixture was heated to 60 °C
and stirred for 5 days at this temperature, during which time the
color changed from pale yellow to dark amber. After the mixture
was cooled to room temperature, all volatiles were evaporated and
the residue was extracted with n-hexane (2 × 10 mL). Removal of
the solvent from extracts afforded 167 mg (70%) of 16 with a purity
similar to that in the previous case. Characterization data for 16
are as follows. Mp: 140-141 °C dec. Anal. Calcd for C22H38HfSi4
(593.37 g mol-1): C, 44.53; H, 6.45. Found: C, 44.58; H, 6.45.
NMR (400 MHz, 297 K, benzene-d6; numbering according to atom
(44) Frisch, M. J. et al.; Gaussian 03, Revision D.01; Gaussian, Inc.:
Wallingford, CT, 2004.
(45) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299.
1
labeling in the molecular structure, see Figure 3): H, δ 0.22 (d,
9
J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010 4379

A R T I C L E S
Lamacˇ et al.
is more stable by ∆G ) 1.24 kcal/mol for Ti, ∆G ) 7.37 kcal/mol
for Zr, and ∆G ) 5.58 kcal/mol for M ) Hf.
Regina Jesse, for assistance. This work was supported by the
Deutsche Forschungsgemeinschaft (Project No. GRK 1213).
In addition, we have also done all calculations by using the well-
accepted B3LYP method with the same basis set. B3LYP can
reproduce the geometric parameters but gives the opposite order
of stability; i.e., B3LYP favors the structure without an agostic
interaction by ∆G ) 7.29 kcal/mol (∆H ) 6.66 kcal/mol) for M
) Ti, ∆G ) 0.73 kcal/mol (∆H ) 0.67 kcal/mol) for M ) Zr, and
∆G ) 2.44 kcal/mol (∆H ) 2.19 kcal/mol) for M ) Hf.
On the basis of these energetic data, the energy difference
between two structural forms should not be too large; we have used
the BP86 method for our calculations, and the MP2 energetic data
are included. All these data are listed in the Supporting Information.
Supporting Information Available: Text and tables giving
synthetic procedures for compounds 10, 11, 13-15, and 17
along with their characterization data, experimental procedures
for reactions of the Cp2ZrCl2/n-BuLi system with alkynes 1 and
5 and of hafnocene precursors with alkyne 1, all computational
details (benchmark calculations, electronic energies, computed
Cartesian coordinates, and selected structural parameters), Si-Si
and Si-C coupling constants for 4, 14, 16, and 19, complete
ref 44, and CIF files giving crystallographic data for 4, 11, 16,
and rac-19. This material is available free of charge via the
Internet at http://pubs.acs.org.
Acknowledgment. This work is dedicated to Prof. Dietmar
Seyferth in appreciation of his advancement of the area of
organometallic chemistry. We thank our technical staff, in particular
JA910527W
9
4380 J. AM. CHEM. SOC. VOL. 132, NO. 12, 2010