Catalytic antimony-antimony bond formation through stibinidene elimination from zirconocene and hafnocene complexes

Article abstract of DOI:10.1002/anie.200600318

(Figure Presented) Bringing together: Zirconium and hafnium d⁰ metallocene complexes catalyze the dehydrocoupling of the stibines MesSbH ₂ (Mes = mesityl) and dmpSbH₂ (dmp = 2,6-dimesitylphenyl) to Sb₄Mes₄ and (dmp)Sb=Sb(dmp), respectively, with liberation of H₂. Kinetic analysis shows that thermal decomposition of hafnium stibide complexes proceeds through α-hydrogen migration and stibinidene elimination.

Full text of DOI:10.1002/anie.200600318

Communications
Dehydrocoupling
DOI: 10.1002/anie.200600318
Catalytic Antimony–Antimony Bond Formation
through Stibinidene Elimination from
Zirconocene and Hafnocene Complexes**
Rory Waterman and T. Don Tilley*
Metal-catalyzed dehydrocoupling reactions have emerged as
an important synthetic method for the generation of element–
element bonds and the production of new types of inorganic
oligomers and polymers.[1] Current information indicates that
these coupling reactions occur by a number of different
mechanisms. For silane couplings catalyzed by d0-metal
complexes, s-bond metathesis reactions appear to be impor-
tant.[2] However, related catalysts mediate similar Sn Sn
À
bond formations and dehydropolymerizations by a mecha-
nism that involves a-hydrogen migration and stannylene
elimination, followed by rapid insertion of stannylene units
[3]
À
À
into Sn H or M Sn bonds (Scheme 1). The latter mecha-
Scheme 1. Possible mechanism for stannane dehydrocoupling.
nism represents an unusual pathway in transition-metal
chemistry and is of interest in determining how generally
useful this process might be. This elimination mechanism is
expected to be most favored for the heavier main-group
elements, which readily form stable low-valent species. These
elements should also possess weak element–hydrogen bonds
which could additionally promote a-hydrogen migration. To
explore the possible utility of this unusual reaction type, we
investigated the dehydrocoupling of stibines. Herein, we
À
provide two examples of metal-catalyzed Sb Sb bond for-
mation and evidence that these reactions occur through a-
hydrogen migration with stibinidene (DSbR) elimination from
d0-metal–SbHR derivatives. For Group 15 elements, catalytic
dehydrocouplings have been reported for primary phos-
phines[4] and Lewis acid/phosphine adducts.[5]
[*] Dr. R. Waterman, Prof. Dr. T. D. Tilley
Department of Chemistry
University of California, Berkeley
Berkeley, CA 94720-1460 (USA)
Fax: (+1)510-642-8940
E-mail: tdtilley@berkeley.edu
[**] This work was supported by the US National Science Foundation
(Grant no. 0132099). The authors acknowledge the Miller Institute
for Basic Research in Science for a Research Fellowship (R.W.) and
Professorship (T.D.T.).
Supportinginformation for this article is available on the WWW
under http://www.angewandte.org or from the author.
2926
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2926 –2929

Angewandte
Chemie
À
The stibines MesSbH2 (1; Mes = mesityl) and dmpSbH2
(2; dmp = 2,6-dimesitylphenyl) were readily prepared by
reduction of the dichloro precursors using LiAlH4 to afford
analytically pure colorless crystals in 53 and 81% yield,
respectively [Eq. (1)].[6] As with (2,6-Trip2C6H3)SbH2 (Trip =
2,4,6-iPr3C6H2),[6c] stibine 2 (m.p. 174–1768C) shows consid-
from the Sb H resonance of 1 (d = 3.12 ppm). Installation of
the stibide ligand presumably occurs through s-bond meta-
thesis to form H2. The s-bond metathesis reaction that forms
stibide complex 5 is fast (t1/2 < 2 min) relative to the reaction
of [Sn(H)2Mes2] with 4 to form [CpCp*HfCl{Sn(H)Mes2}] (t1/2
À
ꢀ 75 min). The facile activation of an Sb H bond is consistent
with the known, small bond-dissociation energy for SbH3
(61 kcalmolÀ1).[8] The isotope effect associated with the
À
Sb H bond activation (kH/kD = 1.2(2)) was determined by a
competition experiment that involved treatment of 4 with an
excess of 1 and [D2]1 (1:1) in C6D6. This isotope effect is small
relative to that expected for a s-bond metathesis mecha-
nism;[2] however, pre-equilibrium coordination of the stibine
to the Hf center could explain the observed isotope effect.[3a]
Compound 5 exhibits limited thermal stability in solution
and could not be isolated as a pure solid. Thermal decom-
position of 5 led to quantitative formation of tetrastibene 3
erably greater thermal stability than 1, which decomposes in
the solid state over a period of hours at ambient temperature
to insoluble products. The deuterated stibine MesSbD2
([D2]1) was similarly prepared by reduction of MesSbCl2
using LiAlD4 in 43% yield of the isolated product. The IR
1
and hafnium hydride 4 [Eq. (3)], as determined by H and
13C NMR spectroscopy. Reaction of a solution of 4 in C6D6
with [D2]1 resulted in evolution of HD and quantitative
formation of [CpCp*HfCl{Sb(D)Mes}] ([D1]5), as deter-
mined by NMR spectroscopy. The thermal decomposition of
[D1]5 provided 3 and hafnium deuteride [CpCp*Hf(D)Cl] in
nearly quantitative yields.
spectrum of [D2]1 exhibits a nSb stretch (1341 cmÀ1) that is
À
D
shifted with respect to the nSb H stretch of 1 (1873 cmÀ1) by the
À
expected amount.
Zirconocene complexes [Cp2Zr(H)Cl] and [Cp2ZrMe2]
(Cp = cyclopentadienyl) react with 20equivalents of stibine 1
in C6D6 to liberate H2 and produce the known tetrastibene
Sb4Mes4 (3) in high yield [ > 95%; Eq. (2)].[6b] It is noteworthy
The sequence of stoichiometric reactions in Equation (3)
appears to involve elementary steps that account for the
À
catalytic dehydrocoupling of MesSbH2 by 4. Thus, the Sb Sb
bond-forming process was investigated in more detail. The
decomposition of 5 follows first-order kinetics (k = 1.02(6) ”
10À4 sÀ1) over approximately five half-lives, as evaluated by
1H NMR spectroscopy. Decomposition of [D1]5 proceeds at a
rate of k = 2.53(5) ” 10À5 sÀ1, which gives a kinetic isotope
effect (kH/kD) of 4.0for this reaction. This relatively large
À
isotope effect is consistent with significant Sb H bond
cleavage in the transition state. Additional evidence for an
a-hydrogen migration/stibinidene elimination pathway comes
from an Eyring analysis, which provided the activation
parameters DH° = 27.3(1) kcalmolÀ1 and DS° = À20.0(1) eu
(for T= 2.2–46.98C). These values are similar to those
obtained for the a-stannylene elimination reaction of
[CpCp*HfCl(SnPh3)], which results in the formation of
[CpCp*HfClPh] and diphenylstannylene,[3c] and are consis-
tent with an ordered transition state [Eq. (4)].
that these reactions occur in the dark, but ambient lighting
accelerates the dehydrocoupling. Qualitatively, [Cp2Zr(H)Cl]
(< 0.5 h) is a much faster catalyst than [Cp2ZrMe2] ( ꢀ 16 h),
likely a result of a slower initial reaction of 1 with dimethyl
zirconocene.[2] Intermediate species were not observed by
1H NMR spectroscopy during the course of these reactions.
Stiochiometric dehydrocoupling of phenylstibine (PhSbH2)
has been reported by reaction with alkyllithium reagents,
sodium metal, or hydrogen traps (styrene or phenylacety-
lene).[7]
The hafnium hydride [CpCp*Hf(H)Cl] (4; Cp* = pen-
tamethylcyclopentadienyl) is a much slower catalyst than
[Cp2Zr(H)Cl] for the dehydrocoupling of Equation (2)
( ꢀ 8 h), and addition of one equivalent of 1 to a solution of
4 in C6D6 resulted in rapid evolution of H2 and near-
quantitative formation of [CpCp*HfCl{Sb(H)Mes}] (5), as
Presumably, the reaction shown in Equation (4) involves
the elimination of mesitylstibinidene (DSbMes), which con-
denses to form the observed tetrastibene 4. An attempt to
trap free DSbMes was made by addition of
1
identified by H and 13C NMR spectroscopy [Eq. (3)]. The
À
Sb H proton of 5 resonates at d = 2.46 ppm, an upfield shift
2,3-dimethyl-1,3-butadiene
(10–
60equiv), an effective trap for free
stibinidene,[9] to solutions of 5 in C6D6.
However, the presence of the added
Angew. Chem. Int. Ed. 2006, 45, 2926 –2929
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2927

Communications
À
diene had no effect on the rates of decomposition or
formation of Sb4Mes4 as the exclusive Sb-containing product.
C(11) bond of 6 is roughly coplanar with the Hf Cp centroid
vector—an arrangement likely to reduce steric interactions.
However, such a geometry also allows for donation from the
lone pair on the antimony center to a vacant orbital on Hf.[13]
Current evidence prohibits definitive assignment of such an
interaction.
The inability to trap free stibinidene is consistent with a rapid
0
À
Sb Sb bond-forming event. In d -metal-catalyzed dehydro-
coupling of secondary stannanes, stannylene fragments are
difficult to trap and appear to rapidly insert into metal–
hydride and metal–tin bonds.[3]
Thermal decomposition of 6 proceeded smoothly in C6D6
=
The treatment of solutions of 4 in hexane with dmpSbH2
resulted in the evolution of H2 and the formation of a red
solution, from which analytically pure red crystals of
[CpCp*HfCl{Sb(H)dmp}] (6) were obtained in 89%
yield [Eq. (5)]. Complex 6 was characterized by NMR
and IR spectroscopic analysis and a single-crystal X-ray
diffraction study. Diagnostic spectroscopic features for 6
include a 1H NMR resonance at d = 2.49 ppm for the
stibide ligand proton (cf. d = 2.46 ppm for 5) and a IR
to give 4 and the reported distibene (dmp)Sb Sb(dmp) [7;
Eq. (6)].[6b] Furthermore, 4 catalyzed the dehydrocoupling of
dmpSbH2 in C6D6 solution to form H2 and 7 in > 95% yield, as
1
determined by H NMR spectroscopy. A preliminary kinetic
investigation indicates that thermal decomposition proceeds
with a first-order dependence on the concentration of 6 (k =
1.29(5) ” 10À5 sÀ1), which is consistent with an a-hydrogen
migration/stibinidene extrusion mechanism.
In summary, Group 4 metal complexes have been shown
to catalytically dehydrocouple stibines. These dehydrocou-
pling reactions with the hafnium catalyst [CpCp*Hf(H)Cl]
show first-order dependence on the hafnium stibide species, a
large primary kinetic isotope effect, and activation parame-
ters that are consistent with a mechanism that involves a-
hydrogen migration and elimination of a stibinidene fragment
(a-stibinidene elimination). Efforts to extend this chemistry,
which involves elimination of low-valent fragments, to
catalytic reactions of additional stibines, other elements, and
new processes are ongoing.
stretch at nSbH = 1830cm À1, a value 41 cmÀ1 lower in energy
than that for dmpSbH2. The latter observation is consistent
with a metal-bound stibide ligand.[10]
The solid-state structure of 6 is shown in Figure 1.[11] The
À
Hf Sb bond length is 3.0035(8) Š; a search of the Cambridge
Structural Database provided no examples of Group 4 metals
with stibine or stibide ligands. The structurally characterized
complex most closely related to 6 is [Cp2Nb(H)2(SbPh2)],
À
which features a Nb Sb bond length of 2.8929(4) Š and a
pyramidal Sb center (angles at Sb: ꢀ = 306.38).[12] The Sb
À
Received: January 25, 2006
Published online: March 23, 2006
Keywords: antimony · dehydrocoupling· elimination · hafnium ·
.
stibinidenes
[1] a) F. Gauvin, J. F. Harrod, H. G. Woo, Adv. Organomet. Chem.
1998, 42, 363 – 405; b) J. Y. Corey, Adv. Organomet. Chem. 2004,
51, 1 – 52; c) T. D. Tilley, Comments Inorg. Chem. 1990, 10, 37 –
46.
[2] T. D. Tilley, Acc. Chem. Res. 1993, 26, 22 – 29.
[3] a) N. R. Neale, T. D. Tilley, J. Am. Chem. Soc. 2002, 124, 3802 –
3803; b) N. R. Neale, T. D. Tilley, Tetrahedron 2004, 60, 7247 –
7260; c) N. R. Neale, T. D. Tilley, J. Am. Chem. Soc. 2005, 127,
14745 – 14755.
[4] a) N. Etkin, M. C. Fermin, D. W. Stephan, J. Am. Chem. Soc.
1997, 119, 2954 – 2955; b) M. C. Fermin, D. W. Stephan, J. Am.
Chem. Soc. 1995, 117, 12645 – 12646; c) V. P. W. Böhm, M.
Brookhart, Angew. Chem. 2001, 113, 4832 – 4834; Angew. Chem.
Int. Ed. 2001, 40, 4694 – 4696.
Figure 1. Perspective view of one enantiomer of
[CpCp*HfCl{Sb(H)dmp}] (6), with thermal ellipsoids drawn at the
35% probability level. Hydrogen atoms have been omitted for clarity.
[5] C. A. Jaska, I. Manners in Inorganic Chemistry in Focus II (Eds:
G. Meyer, D. Naumann, L. Wesemann), Wiley-VCH, Weinheim,
2005, pp. 53 – 64.
À
À
Select bond lengths [Š] and angles [8]: Hf Sb 3.0035(8), Hf Cl
[6] a) Primary stibine dmpSbH2 is novel; synthesis of MesSbH2 has
been reported recently: H. J. Breunig, M. E. Ghesner, E. Lork,
À
2.407(2), Sb C(11) 2.180(8); Cl-Hf-Sb 96.62(6), C(11)-Sb-Hf 112.2(2).
2928 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2926 –2929

Angewandte
Chemie
Z. Anorg. Allg. Chem. 2005, 631, 851 – 856; b) M. Ates, H. J.
Breunig, A. Soltani-Neshan, M. Z. Tegeler, Z. Naturforsch. B
1986, 41, 321 – 326; c) B. Twamley, C. D. Sofield, M. M. Olm-
stead, P. P. Power, J. Am. Chem. Soc. 1999, 121, 3357 – 3367.
[7] a) K. Issleib, A. Balszuweit, Z. Anorg. Allg. Chem. 1975, 418,
158 – 166; b) K. Issleib, A. Balszuweit, Z. Anorg. Allg. Chem.
1976, 419, 87 – 91.
[8] F. A. Cotton, G. Wilkinson, C. A. Murillo, M. Bochmann,
Advanced Inorganic Chemistry, 6th ed., Wiley, New York,
1999, p. 88.
[9] T. Sasamori, Y. Arai, N. Takeda, R. Okazaki, N. Tokitoh, Chem.
Lett. 2001, 42 – 43.
[10] H.-G. Woo, J. F. Walzer, T. D. Tilley, J. Am. Chem. Soc. 1992, 114,
7047 – 7055.
[11] Crystal structure analysis of 6: C39H45ClHfSb, Mr = 849.44,
¯
triclinic, P1, a = 9.070(2), b = 11.309(2), c = 18.282(3) Š, a =
83.691(4), b = 80.809(3), g = 71.477(3)8, V= 1751.7(5) Š3, Z =
, , F(000) = 838, q = 1.13–
2, 1 = 1.611 gcmÀ3 m = 3.833 mmÀ1
26.398. Final R1 = 5.20%, wR2 = 11.55%, GOF = 0.963 for 379
parameters and 11448 reflections (of which 6913 with I > 2s(I)),
max positive and negative peaks in DF map 3.777 and
À1.298 eŠÀ3. Data were collected at 163(2) K Bruker Platform
goniometer (MoKa radiation, 0.71073 Š) equipped with a CCD
detector (Smart Apex). A Patterson search was used to locate
the Hf and Sb atoms; all non-H atoms were converted and
refined anisotropically. Hydrogen atoms were refined isotropi-
cally, fixed at calculated positions. Despite large residual
electron density proximal to the Sb center, a stable hydrogen
atom could not be refined at that location. CCDC-285852
contains the supplementary crystallographic data for this paper.
These data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[12] G. I. Nikonov, L. G. Kuzmina, J. A. K. Howard, Organometallics
1997, 16, 3723 – 3725.
[13] R. T. Baker, J. F. Whitney, S. S. Wreford, Organometallics 1983,
2, 1049 – 1051.
Angew. Chem. Int. Ed. 2006, 45, 2926 –2929
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 2929