The osmium-silicon triple bond: Synthesis, characterization, and reactivity of an osmium silylyne complex

Article abstract of DOI:10.1021/ja406799y

The first silylyne complex of a metal beyond group 6, [Cp*( ⁱPr₃P)(H)Os≡Si(Trip)][HB(C₆F ₅)₃], was prepared by a new synthetic route involving hydride abstraction from silicon. NMR and DFT computation

Full text of DOI:10.1021/ja406799y

Communication
pubs.acs.org/JACS
The Osmium−Silicon Triple Bond: Synthesis, Characterization, and
Reactivity of an Osmium Silylyne Complex
Paul G. Hayes,†,§ Zhenggang Xu,‡ Chad Beddie,‡,⊥ Jason M. Keith,‡,∥ Michael B. Hall,*,‡
and T. Don Tilley*,†
†Department of Chemistry, University of California, Berkeley, Berkeley, California 94720-1460, United States
‡Department of Chemistry, Texas A&M University, College Station, Texas 77843, United States
S
* Supporting Information
Cp(CO)3MoSi(2,6-Trip2-C6H3) from a base-stabilized halosi-
lylene adduct.7
ABSTRACT: The first silylyne complex of a metal
beyond group 6, [Cp*(iPr3P)(H)OsSi(Trip)][HB-
Despite recent progress in the isolation and characterization,
including recent computational studies8 on metal−ylyne
(C6F5)3], was prepared by a new synthetic route involving
hydride abstraction from silicon. NMR and DFT
computations support the presence of a silylyne ligand,
and NBO and ETS-NOCV analysis revealed the nature of
this Os−Si interaction as a triple bond consisting of a
covalent σ bond and two strong π back-donations. The
discovery of this complex allowed observations of the first
cycloadditions involving a silylyne complex, and terminal
alkynes are shown to react via C−H bond additions across
the OsSi bond.
complexes of the heavier group 14 elements, synthetic control
of structure and bonding in such compounds remains quite
challenging. This is emphasized by the facts that known
complexes of this type generally require an exceptionally bulky
substituent and all of them feature a group 6 metal (Cr, Mo, or
W). In this contribution, the first ylyne complex of a nongroup
6 metal, and a rare example of a silylyne complex, is reported.
This complex, [Cp*(iPr3P)(H)OsSi(Trip)][HB(C6F5)3],
was obtained using the readily available primary silane
(Trip)SiH3 as the silicon-based starting material. Significantly,
this complex represents the first LnMER species to exhibit
alkyne cyclization reactivity, akin to that observed for carbyne
species.1
Initial efforts to generate a group 8 silylyne complex focused
on a strategy related to that used to obtain [Cp*(dmpe)(H)-
MoSiMes]+,6 namely abstraction of an anionic substituent
from a neutral silylene ligand. Thus, reaction of a
bromobenzene solution of Cp*(iPr3P)(H)OsSiH(Trip)9
(1) with the Lewis acid B(C6F5)3 immediately resulted in a
color change from bright orange to ruby red, and formation of
[Cp*(iPr3P)(H)OsSi(Trip)][HB(C6F5)3] (2), as indicated
by multinuclear (1H, 11B, 13C, 19F, 29Si, 31P) NMR spectroscopy
(Scheme 1).
arbyne (or alkylidyne) complexes with formal metal−
Ccarbon triple bonds have been the focus of considerable
fundamental research on structure, bonding, and reactivity,
especially given the utility of such species in catalysis.1 These
investigations have heightened curiosity in related transition
metal-group 14 congeners, which are expected to exhibit new
and interesting reactivity patterns. The well-known instability of
π bonds involving these heavier elements2 makes such metal−
ylyne complexes of the type LnMER (E = Si, Ge, Sn, Pb)
even more intriguing as synthetic targets. Power and co-
workers reported the first successes along these lines, with
preparation of the terminal germylyne complexes Cp-
(CO)nMGe[2,6-Ar2-C6H3] (n = 2, 3; M = Cr, Mo, W; Ar
= Trip, 2,4,6-iPr3C6H2; Mes, 2,4,6-Me3C6H2), by direct reaction
of anionic [Cp(CO)3M]− complexes with stable ClGe[2,6-Ar2-
C6H3] species.3 Filippou et al. also utilized halogermylene
species to prepare the germylyne complexes X(P−P)2M
Ge(η1-Cp*) (X = Cl, Br, I; M = Mo, W; P−P =
Ph2PCH2CH2PPh2, Et2PCH2CH2PEt2), and X(PMe3)4Mo
Ge(2,6-C6H3-Trip2) via oxidative additions of a Ge−X bond.4
Similar approaches have been used to generate analogous
stannylyne complexes.5 Consistent with the lower stability and
availability of low-valent silicon starting materials, the analogous
silylyne complexes have proven more elusive. The first silylyne
complex, [Cp*(dmpe)(H)MoSiMes][B(C6F5)4], was ob-
tained in this laboratory by abstraction of chloride from the
corresponding silylene complex Cp*(dmpe)(H)MoSi(Cl)-
Mes. This complex features a linear Mo−Si−C arrangement
and a weak Si···H interaction (2JSiH 15 Hz).6 More recently,
Filippou and co-workers have reported the neutral silylyne
Scheme 1. Synthesis and Reactivity of Osmium Silylyne 2
Received: July 4, 2013
Published: August 1, 2013
© 2013 American Chemical Society
11780
dx.doi.org/10.1021/ja406799y | J. Am. Chem. Soc. 2013, 135, 11780−11783

Journal of the American Chemical Society
Communication
Complex 2 was isolated as a thermally sensitive (t1/2 ≈ 30
min in d5-bromobenzene at ambient temperature) solid in 67%
yield. It exhibits a 29Si NMR resonance at δ 321 that is shifted
downfield by 92 ppm relative to the resonance for silylene 1.
This value agrees well with that obtained computationally for
the optimized structure of complex 2 (δ 309). The conversion
of 1 to 2 is also accompanied by the disappearance of the Os−
1
H and Si−H H NMR resonances for 1, and appearance of a
2
characteristic, hydridic doublet at δ −14.5 (1H, JHP = 29 Hz)
and a broad signal at δ 4.39 (1H) attributed to Os−H and B−H
signals, respectively, for silylyne 2. Attempts to measure the
2JSiH value using variable temperature one- and two-dimensional
2
NMR experiments were unsuccessful. Furthermore, JSiH is
computed to be only 7.4 Hz, confirming an extremely weak
OsH···Si interaction. Spectroscopic features of the HB(C6F5)3
Figure 1. DFT optimized structure of 2. All the hydrogen atoms are
hidden except the hydride on the Os atom.
−
anion, specifically the upfield 11B NMR resonance at δ −25 in
combination with a Δδm,p value of 2.6 ppm for the 19F NMR
shifts, indicate that the anion is weakly coordinating.10 In
addition, low temperature NMR studies failed to reveal
evidence for Si−H or Si−F contacts. Thus, extensive solution
NMR data characterize 2 as a terminal silylyne complex as
shown in Scheme 1, and this is supported by DFT
calculations.11
be expected. Furthermore, the computed Os−Si bond length in
2 is 2.176 Å, which is extremely short and contracted by 0.079
Å with respect to 1 (computed as 2.255 Å). This predicted
decrease in bond length is slightly larger than the observed
difference between the Mo−Si bond lengths in Cp*(dmpe)-
(H)MoSi(Cl)Mes (2.288(2) Å) and [Cp*(dmpe)(H)Mo
SiMes][B(C6F5)4] (2.219(2) Å).6 Structures of several known
compounds that share the same Cp*(iPr3P)Os+ fragment
(Figure S1) were optimized to compare Os−Si bond lengths,
and these results also suggest a bond order higher than two for
the Os−Si bond in compound 2 (Table S1).
In an effort to more firmly establish the bonding mode for
the hydride ligand in complex 2, calculations at the DFT
TPSS12 level were undertaken. The two extreme structures 2
and 2′ with the hydrogen located at the osmium and silicon
atoms, respectively, and the transition state having a bridging
hydrogen atom (TS) were examined (Scheme 2). The results
In order to understand the nature of this Os−Si bond,
natural bond orbital (NBO) analysis13 was performed. The
three strongest interactions between Os and Si consist of one σ
bond and two π bonds (Figure 2 and Table S2). The σ bond
has 43% osmium and 57% silicon character, so it is slightly
polarized toward the silicon atom. The osmium component
contains 43% s and 57% d orbital character, and the silicon
component is composed by 55% s and 45% p orbital character,
providing additional evidence for sp hybridization on silicon.
With the default NBO parameters, both π bonds are
characterized as donor−acceptor interactions (Table S2), i.e.
back-donations from doubly occupied Os d orbitals to empty Si
p orbitals. The polarization of this bond gives rise to a highly
electrophilic silicon atom with an NPA (natural population
analysis) charge of 1.14. In order to obtain a more quantitative
understanding about the energetics of this bonding interaction,
a combination of the extended transition state (ETS)14 energy
decomposition scheme and the natural orbital for chemical
valence (NOCV)15 density decomposition approach was
employed to study the interactions between the two closed
shell fragments: Cp*(iPr3P)(H)Os and [Si(Trip)]+.16 In the
ETS decomposition scheme the two attractive interactions, the
covalent interaction (also known as orbital interaction, −168.5
kcal/mol) and the electrostatic interaction (−166.1 kcal/mol),
have nearly equal contributions; these attractive interactions are
offset by a large Pauli repulsion (206.5 kcal/mol) (Table S3).
The NOCV analysis suggests strong π contributions to the
overall orbital interactions. Such strong π interactions are
consistent with the short Os−Si bond. The high reactivity of
this bond may reflect ease of polarization of the Os−Si π bonds
from nearly covalent to donor−acceptor interactions. Detailed
computational studies on the reactivity of 2 are underway.
The mechanism by which 2 is generated may involve direct
abstraction of the silylene hydrogen by B(C6F5)3, or abstraction
of the osmium hydride ligand followed by α-hydrogen
Scheme 2. Structural Models for 2
Table 1. Selected Bond Lengths (Å), Bond Angle (deg), and
Solvated Free Energies (kcal mol−1)
ΔG
Os−Si
Os−H
Si−H
Os−Si−C
2
0.0
20.3
14.8
2.176
2.254
2.303
1.635
2.487
3.298
2.760
1.550
1.499
168.0
165.0
135.4
TS
2′
in Table 1 show that 2′ is 14.8 kcal mol−1 higher in energy
(ΔG) than 2 and that it can easily transform to 2 through the
bridging hydride transition state (TS), with a free-energy
barrier of only 5.5 kcal mol−1. Furthermore, the computed 29Si
NMR resonance of 2′ (δ 282) does not match the experimental
value. Thus, 2 is the exclusive product formed in the hydride
abstraction reaction.
The optimized structure of 2 has some noteworthy
characteristics (Figure 1). The Os−Si−C bond angle is
168.0°, a value which suggests that the Si atom has nearly sp
hybridization. This angle is much larger than the computed
value of 135.4° in 2′ (Table 1), where sp2 hybridization would
11781
dx.doi.org/10.1021/ja406799y | J. Am. Chem. Soc. 2013, 135, 11780−11783

Journal of the American Chemical Society
Communication
Figure 2. Three bonding interactions in compound 2 described by natural bond orbital analysis. 1π and 2π are back-donations from filled osmium d
orbitals to empty silicon p orbitals.
migration from silicon to osmium. It is not currently possible to
distinguish between these mechanisms, but a possibly relevant
observation is that, under comparable experimental conditions,
there is no appreciable reaction between B(C6F5)3 and
Cp*(iPr3P)(H)OsSiH(dmp) (dmp = 2,6-Mes2-C6H3).9
Given the much greater steric hindrance associated with the
silicon center of the latter silylene complex (vs that for 1), this
result suggests that Si−H abstraction may be operative in the
formation of 2.
alkylidyne and phosphinidene complexes). As with complex 3,
−
11B and 19F NMR spectra indicate that the HB(C6F5)3 anion
of 4 remains intact and weakly coordinating.
Interestingly, terminal alkynes exhibit a different mode of
reactivity toward 2, involving C−H addition across the OsSi
triple bond. Thus, addition of RCCH to 2 afforded the new
silylene complexes [Cp*(iPr3P)(H)(RCC)OsSiH(Trip)]-
t
i
1
[HB(C6F5)3] (5, R = Bu; 6, R = Pr3Si), identified by H, 11B,
19F, and 31P NMR spectroscopy (Scheme 1). Silylene
complexes 5 and 6 exhibit characteristic downfield Si−H
resonances in the 1H NMR spectra, at δ 9.5 and 11.6,
respectively. These unusual C−H activations might be
envisioned as proceeding by two possible mechanisms (Scheme
3): (1) hydrogen migration from osmium to silicon followed by
Since the thermal sensitivity of 2 precluded growth of X-ray
quality crystals, reactivity studies were undertaken to confirm
its identity. Treatment of a bromobenzene solution of silylyne 2
with diphenylacetylene afforded the unprecedented cyclo-
addition product, [Cp*(iPr3P)(H)OsSi(Trip)(PhC
CPh)][HB(C6F5)3] (3), as an analytically pure yellow powder
in 88% yield (Scheme 1). Complex 3 exhibits an upfield-shifted
29Si NMR silylene resonance at δ 110 (calculated: δ 102; see
Supporting Information (SI)) and diagnostic quaternary signals
at 164.3 (2JCP = 13 Hz) and 197.5 (3JCP = 3.1 Hz) in the
13C{1H} NMR spectrum assigned as Cα and Cβ, respectively.
Scheme 3. Mechanism of Formation of Complex 6
1
Resonances in the H, 11B, and 19F NMR spectra attributed to
−
the HB(C6F5)3 anion remained essentially unchanged from
those of complex 2. Unlike silylyne 2, metallosilacyclobutadiene
3 is stable indefinitely under an inert atmosphere at low
temperature. The cycloaddition does not appear to be
reversible, as evidenced by the lack of reaction between 3
and either tBuCCH or iPr3SiCCH (90 min in d5-bromobenzene
solution). Despite well documented examples of similar
cycloadditions for carbyne complexes,1 complex 3 represents
the first example of addition of an alkyne across a ME
functionality. Furthermore, direct cycloadditions of this type
have rarely been observed for silylene complexes, but are of
great interest in the context of metathesis reactivity.17
A related cycloaddition was observed upon reaction of 2 with
18
t
1 equiv of the phosphaalkyne BuCP in bromobenzene
solution, to cleanly produce the orange complex [Cp*(iPr3P)-
(H)OsSi(Trip)(PCtBu)][HB(C6F5)3] (4), as determined
by multinuclear NMR spectroscopy. Upon washing with
pentane, complex 4 was isolated as an analytically pure, orange
powder in 77% yield. Diagnostic resonances in the 29Si, 31P, and
13C NMR spectra at δ 82, 516, and 278 for the silicon,
phosphorus, and carbon ring atoms, respectively, suggest that 4
is best described by a resonance structure containing osmium
phosphinidene19 (OsP) and silene20 (SiC) moieties
(Scheme 1). Notably, this interpretation agrees well with
computations that predict 29Si, 31P, and 13C NMR chemical
shifts for complex 4 (96, 527, and 298 ppm, respectively). This
reactivity suggests that silylyne complexes similar to 2 might
provide useful routes to other triply bonded species (e.g.,
C−H oxidative addition at Os (Path A) or (2) direct C−H
addition across the OsSi bond (Path B). To distinguish
between these pathways, a deuterium-labeling experiment was
performed. Treatment of a d5-bromobenzene solution of
[Cp*(iPr3P)(D)OsSiTrip] [DB(C6F5)3] (2-d2) with 1
equiv of iPr3SiCCH afforded [Cp*(iPr3P)(D)(iPr3C
C)OsSiH(Trip)][DB(C6F5)3] (6-d2) as the exclusive
product, thereby confirming that path B is the operative
mechanism. No OsD/SiH or SiH/BD exchange was observed
after 6 h at ambient temperature.
In summary, a new route to a transition metal silylyne
complex, involving hydride abstraction from silicon, has been
established. NMR and DFT computations helped to validate
the proposed structure of this complex. Unlike previous
11782
dx.doi.org/10.1021/ja406799y | J. Am. Chem. Soc. 2013, 135, 11780−11783

Journal of the American Chemical Society
Communication
Mork, B. V.; Simons, R. S.; Power, P. P. J. Am. Chem. Soc. 2000, 122,
650.
examples of complexes with triple bonds between a transition
metal and a heavy group 14 element, the new silylyne complex
does not feature a group 6 transition metal center. NBO and
ETS-NOCV analysis revealed the nature of this Os−Si bond as
a triple bond consisting of a covalent σ bond and two strong π
back-donations. Significantly, the discovery of complex 2 has
allowed the exploration of new reactivity for a metal−silicon
triple bond. In particular, the observed additions of multiple
bonds, and C−H bonds, across the OsSi bond would seem
to portend a rich and varied chemical reactivity for LnMER
species and the possibility of new chemical structures and
previously unobserved transformations of fundamental signifi-
cance. This work is also an excellent demonstration of close
interplays between experiment and theory in studying novel
compounds.
(4) (a) Filippou, A. C.; Philippopoulos, A. I.; Portius, P.; Neumann,
D. U. Angew. Chem., Int. Ed. 2000, 39, 2778. (b) Filippou, A. C.;
Portius, P.; Philippopoulos, A. I. Organometallics 2002, 21, 653.
(c) Filippou, A. C.; Schnakenburg, G.; Philippopoulos, A. I.;
Weidemann, N. Angew. Chem., Int. Ed. 2005, 44, 5979. (d) Filippou,
A. C.; Weidemann, N.; Philippopoulos, A. I.; Schnakenburg, G. Angew.
Chem., Int. Ed. 2006, 45, 5987.
(5) (a) Filippou, A. C.; Philippopoulos, A. I.; Schnakenburg, G.
Organometallics 2003, 22, 3339. (b) Filippou, A. C.; Portius, P.;
Philippopoulos, A. I.; Rohde, H. Angew. Chem., Int. Ed. 2003, 42, 445.
(6) Mork, B. V.; Tilley, T. D. Angew. Chem., Int. Ed. 2003, 42, 357.
(7) Filippou, A. C.; Chernov, O.; Stumpf, K. W.; Schnakenburg, G.
Angew. Chem., Int. Ed. 2010, 49, 3296.
(8) (a) Pandey, K. K.; Lein, M.; Frenking, G. J. Am. Chem. Soc. 2003,
125, 1660. (b) Filippou, A. C.; Rohde, H.; Schnakenburg, G. Angew.
́
Chem., Int. Ed. 2004, 43, 2243. (c) Pandey, K. K.; Lledos, A. Inorg.
ASSOCIATED CONTENT
Chem. 2009, 48, 2748. (d) Pandey, K. K.; Patidar, P.; Power, P. P.
Inorg. Chem. 2011, 50, 7080.
■
S
* Supporting Information
(9) Hayes, P. G.; Beddie, C.; Hall, M. B.; Waterman, R.; Tilley, T. D.
J. Am. Chem. Soc. 2006, 128, 428.
(10) Horton, A. D. Organometallics 1996, 15, 2675.
Detailed experimental procedure, NMR spectra, computational
details, and coordinates of important molecules. This material is
available free of charge via the Internet at http://pubs.acs.org.
(11) Frisch, M. J., et al. Gaussian 09, revision B.01; Gaussian, Inc.:
Wallingford, CT, 2009. Please refer to the SI for the full citation.
(12) Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E. Phys.
Rev. Lett. 2003, 91, 146401.
(13) NBO 5.9 was used in this work. Glendening, E. D.; Badenhoop,
J. K.; Reed, A. E.; Carpenter, J. E.; Bohmann, J. A.; Morales, C. M.;
Weinhold, F. NBO 5.9; Theoretical Chemistry Institute, University of
Wisconsin: Madison, WI, 2011; http://www.chem.wisc.edu/∼nbo5.
(14) Ziegler, T.; Rauk, A. Theor. Chim. Acta 1977, 46, 1.
(15) (a) Mitoraj, M. P.; Michalak, A.; Ziegler, T. J. Chem. Theory
Comput. 2009, 5, 962. (b) Mitoraj, M.; Michalak, A. J. Mol. Model.
2007, 13, 347. (c) Michalak, A.; Mitoraj, M.; Ziegler, T. J. Phys. Chem.
A 2008, 112, 1933.
(16) ETS-NOCV calculations were performed by the ADF2012
program. (a) Fonseca Guerra, C.; Snijders, J. G.; te Velde, G.;
Baerends, E. J. Theor. Chem. Acc. 1998, 99, 391−403. (b) te Velde, G.;
Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra, C.; van Gisbergen,
S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput. Chem. 2001, 22, 931−
967. (c) ADF2012; SCM, Theoretical Chemistry, Vrije Universiteit,
Amsterdam, The Netherlands, http://www.scm.com. A full list of
authors and contributors can be found in the SI.
(17) (a) Mitchell, G. P.; Tilley, T. D. J. Am. Chem. Soc. 1997, 119,
11236. (b) Lee, V. Y.; Aoki, S.; Yokoyama, T.; Horiguchi, S.;
Sekiguchi, A.; Gornitzka, H.; Guo, J.-D.; Nagase, S. J. Am. Chem. Soc.
2013, 135, 2987.
AUTHOR INFORMATION
■
Corresponding Author
tdtilley@berkeley.edu; mbhall@tamu.edu
Present Addresses
§Department of Chemistry and Biochemistry, University of
Lethbridge, Lethbridge, Alberta, T1K 3M4, Canada.
⊥Chemical Strategies Division, Health Canada, 269 Laurier
Ave. W, 5-058, Ottawa, ON K1A 0K9 AL 4905B, Canada.
∥Department of Chemistry, Colgate University, 13 Oak Drive,
Hamilton, NY 13346, USA.
Notes
The authors declare no competing financial interests.
ACKNOWLEDGMENTS
■
This work was supported by the National Science Foundation
under Grant No. CHE-1265674. We thank Professors C. C.
Cummins and J. Figueroa for helpful discussions and the
generous gift of (tBuC(O)P)Nb(N[tBu]Mes)3. A postdoctoral
fellowship was provided by the Natural Sciences and Engineer-
ing Research Council (NSERC) of Canada (P.G.H.). M.B.H.
thanks the NSF (CHE-0910552, MRI-0216275) and The
Welch Foundation (A-0648) for support. The authors would
like to thanks Drs. Qian Peng and Rory Waterman for helpful
discussions.
(18) Figueroa, J. S.; Cummins, C. C. J. Am. Chem. Soc. 2004, 126,
13916.
(19) Rosenberg, L. Coord. Chem. Rev. 2012, 256, 606.
(20) Ottosson, H.; Eklof, A. M. Coord. Chem. Rev. 2008, 252, 1287.
̈
REFERENCES
■
(1) (a) Bunz, U. H. F. Acc. Chem. Res. 2001, 34, 998. (b) Schrock, R.
R. Chem. Rev. 2002, 102, 145. (c) Furstner, A.; Davies, P. W. Chem.
Commun. 2005, 2307. (d) Schrock, R. R.; Czekelius, C. Adv. Synth.
Catal. 2007, 349, 55. (e) Zhang, W.; Moore, J. S. Adv. Synth. Catal.
2007, 349, 93. (f) Wu, X.; Tamm, M. Beilstein J. Org. Chem. 2011, 7,
82. (g) Heppekausen, J.; Stade, R.; Kondoh, A.; Seidel, G.; Goddard,
R.; Furstner, A. Chem.Eur. J. 2012, 18, 10281.
̈
(2) (a) Jutzi, P. Angew. Chem., Int. Ed. Engl. 1975, 14, 232.
(b) Gusel’nikov, L. E.; Nametkin, N. S. Chem. Rev. 1979, 79, 529.
(c) West, R. Angew. Chem., Int. Ed. Engl. 1987, 26, 1201. (d) Power, P.
P. Chem. Rev. 1999, 99, 3463. (e) Waterman, R.; Hayes, P. G.; Tilley,
T. D. Acc. Chem. Res. 2007, 40, 712.
(3) (a) Simons, R. S.; Power, P. P. J. Am. Chem. Soc. 1996, 118,
11966. (b) Pu, L.; Twamley, B.; Haubrich, S. T.; Olmstead, M. M.;
11783
dx.doi.org/10.1021/ja406799y | J. Am. Chem. Soc. 2013, 135, 11780−11783