Synthesis, structure, and α-elimination chemistry of hafnocene triarylstannyl complexes

Article abstract of DOI:10.1021/ja0529586

New hafnocene triarylstannyl complexes were prepared and were shown to undergo clean thermal decompositions via α-aryl-elimination to produce the corresponding stannylene and a hafnocene aryl complex. The rate of the decomposition is highly dependent on the nature of the ancillary ligand, with the stabilities of the CpCp*Hf(SnPh₃)X compounds following the order X = NMe₂ > Np (α-agostic) > OMe > Cl > Me. Mechanistic information suggests that α-aryl-elimination may be viewed as a concerted process involving nucleophilic attack of the migrating aryl group onto the electrophilic metal center.

Full text of DOI:10.1021/ja0529586

Published on Web 10/04/2005
Synthesis, Structure, and r-Elimination Chemistry of
Hafnocene Triarylstannyl Complexes
Nathan R. Neale and T. Don Tilley*
Contribution from the Department of Chemistry, UniVersity of California, Berkeley,
Berkeley, California 94720-1460, and the Center for New Directions in Organic Synthesis,
Berkeley, California 94720-1460
Received May 6, 2005; E-mail: tdtilley@berkeley.edu
Abstract: New hafnocene triarylstannyl complexes were prepared and were shown to undergo clean thermal
decompositions via R-aryl-elimination to produce the corresponding stannylene and a hafnocene aryl
complex. The rate of the decomposition is highly dependent on the nature of the ancillary ligand, with the
stabilities of the CpCp*Hf(SnPh3)X compounds following the order X ) NMe2 > Np (R-agostic) > OMe >
Cl > Me. Mechanistic information suggests that R-aryl-elimination may be viewed as a concerted process
involving nucleophilic attack of the migrating aryl group onto the electrophilic metal center.
molecular-weight polystannanes.5 These polymerizations pre-
sumably involve intermediate zirconium stannyl derivatives, but
Introduction
Investigations of d0 metal silyl compounds have revealed a
number of new reactivity modes,1 including insertions of
unsaturated substrates into M-Si bonds2 and σ-bond metathesis
processes that can result in metal-mediated polymerizations of
silanes3 and the productive functionalization of hydrocarbons.4
The reactive (and presumably weak) nature of d0 M-Si bonds,
and established trends in bond energies for the group 14
elements,5 suggested that complexes containing d0 M-E (E )
Ge, Sn, Pb) bonds should also display a rich reaction chemistry.
In fact, we have found that zirconocene derivatives catalyze
the dehydropolymerization of stannanes to the first high
attempts to observe these intermediates in the dehydropolym-
erization reactions have thus far been unsuccessful.5b Although
complexes with d0 M-Sn bonds have been known for a long
time, surprisingly few examples have been reported over the
years.6 In addition, few reactivity studies on such complexes
have been carried out.
A recent report from our laboratories described the isolation
of a stable hydrostannyl d0 complex, CpCp*Hf(SnHMes2)Cl
(Cp* ) η5-C5Me5; Mes ) 2,4,6-trimethylphenyl).7 This complex
was found to react via a process that has not been observed for
analogous silyl complexes: facile R-H-elimination to give
CpCp*Hf(H)Cl and the free stannylene Mes2Sn (eq 1). This
(1) (a) Tilley, T. D. In The Chemistry of Organic Silicon Compounds; Patai,
S., Rappoport, Z., Eds.; Wiley: New York, 1989; Chapter 24. (b) Tilley,
T. D. In The Silicon-Heteroatom Bond; Patai, S., Rappoport, Z., Eds.;
Wiley: New York, 1989; Chapter 10. (c) Eisen, M. S. In The Chemistry
of Organic Silicon Compounds; Patai, S., Apeloig, Y., Eds.; Wiley: New
York, 1998; Vol 2, Part 3, Chapter 35. (d) Corey, J. Y.; Braddock-Wilking,
J. Chem. ReV. 1999, 99, 175-292.
(2) (a) Campion, B. K.; Falk, J.; Tilley, T. D. J. Am. Chem. Soc. 1987, 109,
2049-2056. (b) Arnold, J.; Tilley, T. D. J. Am. Chem. Soc. 1987, 109,
3318-3322. (c) Elsner, F. H.; Tilley, T. D.; Rheingold, A. L.; Geib, S. J.
J. Organomet. Chem. 1988, 358, 169-183. (d) Arnold, J.; Tilley, T. D.;
Rheingold, A. L.; Geib, S. J.; Arif, A. M. J. Am. Chem. Soc. 1989, 111,
149-164. (e) Arnold, J.; Engeler, M. P.; Elsner, F. H.; Heyn, R. H.; Tilley,
T. D. Organometallics 1989, 8, 2284-2286. (f) Roddick, D. M.; Heyn, R.
H.; Tilley, T. D. Organometallics 1989, 8, 324-330. (g) Campion, B. K.;
Heyn, R. H.; Tilley, T. D. J. Am. Chem. Soc. 1990, 112, 2011-2013. (h)
Woo, H.-G.; Tilley, T. D. J. Organomet. Chem. 1990, 393, C6-C9. (i)
Campion, B. K.; Heyn, R. H., Tilley, T. D. Inorg. Chem. 1990, 29, 4355-
4356 and references in the above.
(3) (a) Tilley, T. D. Comments Inorg. Chem. 1990, 10, 37-51. (b) Harrod, J.
F.; Mu, Y.; Samuel, E. Polyhedron 1991, 10, 1239-1245. (c) Corey, J. In
AdVances in Silicon Chemistry; Larson, G., Ed.; JAI Press: Greenwich,
CT, 1991; Vol. 1, p 327-387. (d) Laine, R. M. In Aspects of Homogeneous
Catalysis; Ugo, R., Ed.; Kluwer Academic Publishers: Amsterdam, 1990;
p 37-63.
(4) (a) Sadow, T. D.; Tilley, T. D. Organometallics 2001, 20, 4457-4459. (b)
Sadow, A. D.; Tilley, T. D. J. Am. Chem. Soc. 2002, 124, 6814-6815. (c)
Sadow, A. D.; Tilley, T. D. Angew. Chem. Int. Ed. 2003, 42, 803-805.
(d) Sadow, A. D.; Tilley, T. D. J. Am. Chem. Soc. 2003, 125, 7971-7977.
(e) Sadow, A. D.; Tilley, T. D. J. Am. Chem. Soc. 2003, 125, 9462-9475.
(f) Sadow, A. D.; Tilley, T. D. Organometallics 2003, 22, 3577-3585.
(5) (a) Imori, T.; Tilley, T. D. Chem. Commun. 1993, 21, 1607-1609. (b)
Imori, T.; Lu, V.; Cai, H.; Tilley, T. D. J. Am. Chem. Soc. 1995, 117,
9931-9940. (c) Lu, V.; Tilley, T. D. Macromolecules 1996, 29, 5763-
5764.
R-elimination process appears to be involved in the hafnium-
catalyzed dehydrocoupling of Mes2SnH2 to Mes2HSnSnHMes2,
which occurs via insertion of the stannylene into the Sn-H bond
of Mes2SnH2. Related observations were made in an earlier
attempt to prepare CpCp*Zr(SnPh3)Cl, via the reaction of
CpCp*ZrCl2 with LiSnPh3, which instead produced the phenyl
(6) (a) Coutts, R. S. P.; Wailes, P. C. J. Chem. Soc., Chem. Commun. 1968,
260-261. (b) Creemers, H. M. J. C.; Verbeek, F.; Noltes, J. G. J.
Organomet. Chem. 1968, 15, 125-130. (c) Kingston, B. M.; Lappert, M.
F. J. Chem. Soc., Dalton Trans. 1972, 69. (d) Schumann, H.; Cygon, M. J.
Organomet. Chem. 1978, 144, C43-C45. (e) Razuvaev, F. A.; Kalinina,
G. S.; Fedorova, E. A. J. Organomet. Chem. 1980, 190, 157-165. (f)
Lappert, M. F.; Power, P. P. J. Chem. Soc., Dalton Trans. 1985, 51-57.
(g) Porchia, M.; Casellato, U.; Ossala, F.; Rossetto, G.; Zanella, P.; Graziani,
R. J. Chem. Soc., Chem. Commun. 1986, 1034-1035. (h) Zheng, W.;
Stephan, D. W. Inorg. Chem. 1988, 27, 2386-2388. (i) Woo, H.-G.;
Freeman, W. P.; Tilley, T. D. Organometallics 1992, 11, 2198-2205. (j)
Lutz, M.; Findeis, B.; Haukka, M.; Pakkanen, T. A.; Gade, L. H.
Organometallics 2001, 20, 2505-2509.
(7) Neale, N. R.; Tilley, T. D. J. Am. Chem. Soc. 2002, 124, 3802-3803.
9
10.1021/ja0529586 CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 14745-14755
14745

A R T I C L E S
Neale and Tilley
24 h, 215 °C, 72%; for ClSn(p-(OMe)C6H4)3, 4 h, 275 °C, 70%. These
impure compounds were used without further purification in the
procedures below. For the NMR tube kinetic measurements and all
reactions involving a hafnium hydride, glassware was silylated with
Me3SiCl/chloroform solution (1:9, v/v), washed three times with
acetone, and then rinsed with distilled water and ethanol before oven-
drying.
complex CpCp*Zr(Ph)Cl. In addition, isolated CpCp*Zr(SnPh3)-
Cl (prepared by the σ-bond metathesis reaction of Ph3SnH and
CpCp*Zr[Si(SiMe3)3]Cl) was found to decompose slowly over
days at 80 °C to a number of products, including CpCp*Zr-
(Ph)Cl.6i
The R-elimination process described above is quite unusual,
as degradation of a d0 M-ERnR′ complex (E ) main group
element) to M-R′ and ERn has been reported in only a few
cases. For example, Erker has described the decomposition of
Cp2Zr[C(OMe)Ph2]Cl at room temperature to give Cp2Zr(OMe)-
Cl and tetraphenylethylene (presumably formed by dimerization
of the initial product, diphenylcarbene).8 Related reactions
involve decarbonylations of group 4 metallocene acyl com-
plexes, for which there are many examples.9 Given the pos-
sibility that R-eliminations in d0 stannyl complexes might be
of more general utility in early transition metal chemistry, we
have investigated this reaction type in more detail. Here we
describe the syntheses of a number of new hafnium triarylstannyl
derivatives and mechanistic aspects of R-aryl-elimination pro-
cesses in these complexes.
NMR spectra were recorded in benzene-d6 solutions (unless other-
wise noted) at 300 or 500 MHz (1H) with Bruker AMX-300 and DRX-
500 spectrometers, at 125.77 MHz (13C{1H}) or at 186.50 MHz
(
119Sn{1H}) with a DRX-500 spectrometer, or at 376.45 MHz (19F-
{1H}) with an AMX-400 spectrometer at ambient temperature and were
referenced to the residual solvent peak. Many of the 1H NMR
resonances for the aryl groups of the triarylstannyl species appear as
complex multiplets due to coupling to both 117Sn and 119Sn nuclei.
Before NMR characterization of compounds 8, 9, and 10, the solvent
of crystallization was removed by crushing the crystals in a mortar
and pestle and placing the resulting powder under vacuum for 4 h.
Elemental and mass spectral analyses were carried out by the Mi-
croanalytics Laboratory at the University of California, Berkeley. IR
samples of solid materials were prepared as Nujol mulls between two
KBr plates. All IR absorptions are reported in cm-1 and were recorded
with a Mattson Infinity 60 MI FTIR spectrometer.
Experimental Section
[Me2C(C5H4)2]Hf(NMe2)2 (1). Toluene (80 mL) was added to a 500-
mL round-bottom Schlenk flask containing Hf(NMe2)4 (4.00 g, 11.3
mmol). A solution of Me2C(C5H5)2 (0.49 g, 2.8 mmol, deoxygenated
by sparging with N2) in toluene (20 mL) was added, and a condenser
fitted with a flow-control adapter was attached. The solution was heated
at 110 °C for 18 h, cooled to room temperature, and filtered via cannula.
The resulting yellow solution was concentrated to ca. 30 mL, pentane
(20 mL) was added, and the biphasic mixture was cooled to -80 °C.
A yellow solid was isolated by cannula filtration, and this impure
product was recrystallized from toluene (15 mL) at -80 °C to give the
product as a bright yellow crystalline solid in 52% yield (2.57 g, 5.88
mmol). Note: upon warming of the yellow solid to room temperature
after cannula filtration at -80 °C, it dissolved in the residual toluene.
However, after removing the toluene under vacuum, a yellow crystalline
solid was obtained. 1H NMR: δ 1.45 (s, 6 H, [Me2C(C5H4)2]), 2.87 (s,
12 H, NMe2), 5.35 (t, 4 H, [Me2C(C5H4)2]), 6.16 (t, 4 H, [Me2C-
(C5H4)2]). 13C{1H} NMR: δ 25.0, 36.9, 48.6, 102.1, 111.7, 128.0. Anal.
Calcd for C17H26N2Hf: C, 46.74; H, 6.00; N, 6.41. Found: C, 46.75;
H, 6.06; N, 6.11.
All manipulations were carried out under a nitrogen atmosphere using
standard Schlenk techniques or a nitrogen-filled glovebox. Dry, oxygen-
free solvents were employed throughout. Pentane, diethyl ether, and
tetrahydrofuran were distilled from sodium/benzophenone, benzene was
distilled from potassium, toluene was distilled from sodium, and
benzene-d6 and toluene-d8 were dried over NaK alloy and Na,
respectively, and then vacuum transferred and stored over 4-Å molecular
sieves. Chloroform was purified by shaking with several small portions
of concentrated H2SO4, washing with several portions of water, and
then drying over CaCl2 before distillation. Trimethylsilyl chloride
(Gelest or Aldrich) was distilled prior to use. The compounds
Hf(NMe2)4,10 Me2C(C5H5)2,11 Ph3SnH,12 PhLi,13 LiNMe2 (from Li0 and
excess HNMe2; isolated by evaporating excess HNMe2),14 CpCp*Hf-
(Me)OTf (6),4a NaOMe,15 (THF)3.5LiSnPh3,16,17 CpCp*Hf(H)Cl,18 and
ClSn(p-FC6H4)319 were prepared according to literature procedures. The
compounds CpCp*Hf(H)OMe and CpCp*Hf(H)NMe2 were prepared
by the reaction of 1 equiv of NaOMe or LiNMe2, respectively, with
CpCp*Hf(H)Cl in THF at room temperature for 12 h. These species
could not be isolated in pure form and were used as prepared. Triaryltin
chlorides ClSn(p-(CF3)C6H4)3 and ClSn(p-(OMe)C6H4)3 were prepared
using the Kocheshkov disproportionation reaction.20 Heating times,
temperatures, and purities were as follows: for ClSn(p-(CF3)C6H4)3,
[Me2C(C5H4)2]Hf(SnPh3)NMe2 (2). A solution of HSnPh3 (0.804
g, 2.29 mmol) in benzene (20 mL) was added to a yellow solution of
[Me2C(C5H4)2]Hf(NMe2)2 (1.00 g, 2.29 mmol) in benzene (30 mL).
The mixture was stirred in the dark at room temperature for 5 min, at
which point benzene was removed in vacuo. The resulting red-orange
foam was extracted with Et2O (2 × 40 mL), and the red-orange ether
solution was filtered via cannula, concentrated to ca. 30 mL, and cooled
to -30 °C. The product was isolated as yellow crystals in 68% yield
(8) Erker, G.; Rosenfeldt, F. Tetrahedron Lett. 1981, 22, 1379-1382.
(9) (a) Fachinetti, G.; Fochi, G.; Floriani, C. J. Chem. Soc. Dalton Trans. 1977,
1946-1950. (b) Tatsumi, K.; Nakamura, A.; Hofmann, P.; Stauffert, P.;
Hoffmann, R. J. Am. Chem. Soc. 1985, 107, 4440-4451. (c) Moloy, K.
G.; Fagan, P. J.; Manriquez, J. M.; Marks, T. J. J. Am. Chem. Soc. 1986,
108, 56-67. (d) AdVanced Inorganic Chemistry, 5th ed.; Cotton, F. A.,
Wilkinson, G.; Wiley: New York, 1988; Chapters 27 and 28. (e) Cavell,
K. J. Coord. Chem. ReV. 1996, 155, 209-243.
1
(1.15 g, 1.55 mmol). H NMR: δ 0.86 (s, 3 H, [Me2C(C5H4)2]), 1.15
(s, 3 H, [Me2C(C5H4)2]), 2.79 (s, 6 H, NMe2), 5.06 (m, 2 H, [Me2C-
(C5H4)2]), 5.38 (m, 2 H, [Me2C(C5H4)2]), 5.50 (m, 2 H, [Me2C(C5H4)2]),
5.68 (m, 2 H, [Me2C(C5H4)2]), 7.20-7.26 (m, 3 H, p-C6H5), 7.32-
7.37 (m, 6 H, m-C6H5), 7.82-7.94 (m, 6 H, o-C6H5). 13C{1H} NMR:
δ 23.0, 24.4, 36.6, 50.7, 99.9, 102.3, 109.3, 111.4, 118.5, 127.7, 128.8,
138.5, 151.7. 119Sn{1H} NMR: δ 67.05. Anal. Calcd for C33H35-
NSnHf: C, 53.36; H, 4.75; N, 1.89. Found: C, 53.59; H, 4.70; N,
1.67.
(10) Diamond, G. M.; Jordan, R. F.; Petersen, J. L. Organometallics 1996, 15,
4030-4037.
(11) Nifant’ev, I. E.; Yarnykh, V. L.; Borzov, M. V.; Mazurchik, B. A.;
Mstyslavsky, V. I.; Roznyatovsky, V. A.; Ustynyuk, Y. A. Organometallics
1991, 10, 3739-3745.
(12) van der Kerk, G. J. M.; Noltes, J. G.; Luijten, J. G. A. J. Appl. Chem.
1957, 7, 366.
(13) Screttas, C. G.; Micha-Screttas, M. Organometallics 1984, 3, 904-907.
(14) Sard, H.; Duffley, R. P.; Razadan, R. K. Synth. Comm. 1983, 13, 813.
(15) Organic Syntheses; Tishler, M., Ed.; John Wiley and Sons: New York:
1959; Vol. 39, p 51.
[Me2C(C5H4)2]Hf(SnPh3)Cl. Trimethylsilyl chloride (0.080 g, 0.74
mmol, 26 equiv) was added to a solution of 2 (0.021 g, 0.028 mmol)
in benzene-d6. Immediately a color change was observed from yellow
to yellow-orange, and after 20 min the product was observed in 74%
(16) Lu, V. Ph.D. Thesis, University of California, Berkeley, 1999.
(17) Gutekunst, G.; Brook, A. G. J. Organomet. Chem. 1982, 225, 1-3.
(18) Woo, H.-G.; Walzer, J. F.; Tilley, T. D. J. Am. Chem. Soc. 1992, 114,
7047-7055.
1
1
yield by H NMR spectroscopy. H NMR: δ 0.63 (s, 3 H, [Me2C-
(C5H4)2]), 0.98 (s, 3 H, [Me2C(C5H4)2]), 4.60 (m, 2 H, [Me2C(C5H4)2]),
5.54 (m, 2 H, [Me2C(C5H4)2]), 5.69 (m, 2 H, [Me2C(C5H4)2]), 7.23 (m,
(19) Schenk, W. A.; Khadra, A.; Burschka, C. J. Organomet. Chem. 1994, 468,
75-86.
(20) Kosheshkov, K. A. Chem. Ber. 1926, 996.
9
14746 J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005

Hafnocene Triarylstannyl Complexes
A R T I C L E S
2 H, [Me2C(C5H4)2]), 7.17-7.20 (m, 3 H, p-C6H5), 7.29-7.33 (m, 6
H, m-C6H5), 7.86-7.95 (m, 6 H, o-C6H5). 119Sn{1H} NMR: δ 97.0.
Alternative Synthesis of [Me2C(C5H4)2]HfCl2 (3).22 Neat ClSiMe3
(2.05 mL, 16.2 mmol, 7 equiv) was added dropwise to a yellow solution
of 1 (1.00 g, 2.29 mmol) in benzene (20 mL). After stirring for 12 h,
an opaque light yellow mixture was obtained. Solvent was removed in
vacuo, leaving a beige residue. Chloroform (100 mL) was added, and
the mixture was filtered via cannula. The resulting pale yellow solution
was concentrated to ca. 10 mL, causing precipitation of a beige powder.
The remaining solution was filtered off, leaving the product as a beige
via cannula, and the resulting deep yellow filtrate was concentrated to
ca. 15 mL and then cooled to -30 °C. The product was isolated by
1
filtration as yellow crystals in 58% yield (0.224 g, 0.303 mmol). H
NMR: δ 1.76 (s, 15 H, C5Me5), 2.84 (s, 6 H, NMe2), 5.72 (s, 3JSnH
)
6.7 Hz, 5 H, C5H5), 7.23 (tt, 3JHH ) 7.3 Hz, 4JHH ) 1.5 Hz, 3 H, p-C6H5),
7.32 (m, 6 H, o-C6H5), 7.82 (m, 6 H, o-C6H5). 13C{1H} NMR: δ 13.01,
49.09, 109.5, 118.3, 127.5, 139.0, 152.5. 119Sn{1H} NMR: δ 74.3. Anal.
Calcd for C38H48.5NO0.75SnHf: C, 55.09; H, 5.90; N, 1.69. Found: C,
54.70; H, 5.58; N, 1.66.
CpCp*Hf(SnPh3)Me·0.5(C5H12) (9). A solution of (THF)3.5LiSnPh3
(0.197 g, 0.324 mmol) in THF (10 mL) was added to a solution of 6
(0.173 g, 0.319 mmol) in THF (10 mL) at -78 °C. The solution was
allowed to come to room temperature with stirring, and stirring was
continued in the dark for 12 h. Solvent was removed in vacuo, leaving
a yellow residue. The residue was extracted with pentane (2 × 25 mL),
the combined extracts were filtered via cannula, and the resulting yellow
solution was concentrated to ca. 25 mL and then cooled to -30 °C.
The product was isolated as yellow crystals in 67% yield (0.168 g,
0.215 mmol). 1H NMR: δ -0.46 (s, Hf-Me, 3 H), 1.73 (s, 15 H,
1
powder in 69% yield (0.667 mg, 1.59 mmol). H NMR: δ 1.10 (s, 6
H, [Me2C(C5H4)2]), 5.10 (t, 3JHH ) 2.6 Hz, 4 H, [Me2C(C5H4)2]), 6.30
3
(t, JHH ) 2.6 Hz, 4 H, [Me2C(C5H4)2]).
Observation of [Me2C(C5H4)2]HfPh2. A solution of 3 (0.0026 g,
0.0062 mmol) in benzene-d6 (0.5 mL) was added to solid (THF)3.5LiSnPh3
(0.0064 g, 0.012 mmol), causing an immediate color change to bright
yellow. A 1H NMR experiment after 15 min suggested that the diphenyl
1
complex had formed. H NMR: δ 1.23 (s, 6 H, [Me2C(C5H4)2]), 5.26
3
3
(t, JHH ) 2.4 Hz, 4 H, [Me2C(C5H4)2]), 6.20 (t, JHH ) 2.4 Hz, 4 H,
[Me2C(C5H4)2]), phenyl resonances obscured by (Ph2Sn)n.
3
3
C5Me5), 5.78 (s, JSnH ) 6.6 Hz, 5 H, C5H5), 7.19 (tt, JHH ) 7.5 Hz,
4JHH ) 1.5 Hz, 3 H, p-C6H5), 7.29 (m, 6 H, m-C6H5), 7.76 (m, 6 H,
o-C6H5). 13C{1H} NMR: δ 12.77, 58.67, 111.4, 119.1, 127.6, 128.8,
138.7, 153.0. 119Sn{1H} NMR: δ 111.5. Anal. Calcd for C36.5H44-
SnHf: C, 56.21; H, 5.69. Found: C, 56.00; H, 5.92.
CpCp*Hf(SnPh3)Cl (4).6i A solution of (THF)3.5LiSnPh3 (0.743 g,
1.22 mmol) in THF (20 mL) was added dropwise to a solution of
CpCp*HfCl2 (0.515 g, 1.15 mmol) in THF (20 mL) at -78 °C. The
solution was allowed to warm slowly to room temperature with stirring
for 12 h in the dark. Solvent was removed from the deep yellow
solution, leaving a yellow foam. The foam was extracted with Et2O (2
× 20 mL), filtered via cannula, concentrated to ca. 15 mL, and cooled
to -80 °C. The product was isolated as yellow crystals in 75% yield
(0.655 g, 0.857 mmol). 119Sn{1H} NMR: δ 112.8.
CpCp*Hf(SnPh3)OMe·0.5(C7H8) (10). A solution of (THF)3.5LiSnPh3
(0.420 g, 0.689 mmol) in THF (10 mL) was added to a solution of 7
(0.299 g, 0.671 mmol) in THF (15 mL) at 0 °C, and the mixture was
stirred in the dark at room temperature for 12 h. Solvent was removed
from the yellow solution in vacuo, and the resulting light yellow foam
was extracted with toluene (20 mL). The extract was filtered via
cannula, and the resulting solution was concentrated to ca. 5 mL and
cooled to -30 °C. The product was isolated as a near colorless (very
pale-yellow) crystalline solid in 72% yield (0.367 g, 0.467 mmol). 1H
CpCp*Hf(NMe2)Cl (5). A solution of LiNMe2 (0.115 g, 2.25 mmol)
in THF (15 mL) was added dropwise to a solution of CpCp*HfCl2
(1.01 g, 2.25 mmol) in THF (25 mL) at -78 °C. The resulting yellow
solution was warmed slowly to room temperature with stirring for 12
h. This reaction was not clean, and a mixture of CpCp*HfCl2, CpCp*Hf-
(NMe2)Cl, and CpCp*Hf(NMe2)2 in a ratio of 0.34:1:0.28 was obtained.
More lithium dimethylamide (0.027 g, 0.53 mmol, 0.24 equiv) in THF
(20 mL) at 0 °C was added to this solution, and the mixture was stirred
for an additional 1 h. Solvent was removed from the yellow solution,
and the resulting yellow residue was extracted with Et2O (2 × 25 mL),
filtered via cannula, concentrated to ca. 30 mL, and cooled to -80 °C.
After 3 days at this temperature the product was isolated by filtration
3
NMR: δ 1.65 (s, 15 H, C5Me5), 5.93 (s, JSnH ) 5 Hz, 5 H, C5H5),
3
4
7.23 (tt, JHH ) 9 Hz, JHH ) 2.1 Hz, 3 H, p-C6H5), 7.34 (m, 6 H,
m-C6H5), 7.90 (m, 6 H, o-C6H5). 13C{1H} NMR: δ 12.49, 62.16, 109.6,
118.1, 127.6, 128.7, 138.8, 152.3. 119Sn{1H} NMR: δ 64.3. Anal. Calcd
for C37.5H42OSnHf: C, 55.89; H, 5.25. Found: C, 56.08; H, 5.42.
CpCp*Hf(SnPh3)Ph (11). A solution of 4 (0.012 g, 0.016 mmol)
and PhLi (0.0014 g, 0.017 mmol) was prepared in benzene-d6 (ca. 0.5
mL) and THF (2 drops). The cloudy red-orange mixture was im-
mediately placed in an NMR tube wrapped with Al foil for protection
from light. The reaction had gone to 81% completion after ca. 10 min,
and after 2 d the product was observed in 91% yield. 1H NMR: δ 1.65
1
as yellow crystals in 53% yield (0.546 g, 1.19 mmol). H NMR: δ
1.84 (s, 15 H, C5Me5), 2.86 (s, 6 H, NMe2), 5.84 (s, 5 H, C5H5). 13C-
{1H} NMR: δ 12.10 (s, C5Me5), 49.33 (s, NMe2), 112.4 (s, C5H5),
119.7 (s, C5Me5). 13C{1H} NMR: δ 12.10, 49.33, 112.4, 119.7. Anal.
Calcd for C17H26NClHf: C, 44.54; H, 5.72; N, 3.06. Found: C, 44.37;
H, 5.79; N, 2.72.
3
(s, 15 H, C5Me5), 5.94 (s, JSnH ) 6.2 Hz, 5 H, C5H5), 6.93 (m, 2 H,
o-C6H5), 7.06 (tt, 3JHH ) 7.3 Hz, 4JHH ) 1.2 Hz, 3 H, p-C6H5), 7.17 (tt,
3JHH ) 11 Hz, 4JHH ) 2 Hz, 3 H, Sn(p-C6H5)3), 7.24 (t, 3JHH ) 13 Hz,
3
6 H, Sn(m-C6H5)3), 7.30 (t, JHH ) 7.4 Hz, 2 H, m-C6H5), 7.74 (m, 6
CpCp*Hf(OMe)Cl (7). A Schlenk tube was charged with CpCp*-
HfCl2 (0.609 g, 1.35 mmol) and NaOMe (0.073 g, 1.35 mmol). THF
(25 mL) was added, and the suspension was stirred at room temperature
for 12 h. Solvent was removed from the cloudy mixture, and the beige
residue was extracted with pentane (2 × 15 mL). The combined extracts
were concentrated to ca. 25 mL and then cooled to -30 °C. The product
was isolated as a light beige crystalline solid in 77% yield (0.460 g,
0.103 mmol). 1H NMR: δ 1.84 (s, 15 H, C5Me5), 3.86 (s, 3 H, OMe),
5.90 (s, 5 H, C5H5). 13C{1H} NMR: δ 11.88, 61.78, 112.5, 119.8. Anal.
Calcd for C16H23OClHf: C, 43.16; H, 5.21. Found: C, 43.05; H, 5.39.
H, Sn(o-C6H5)3). The o-C6H5 resonance could not be definitively
identified. 119Sn{1H} NMR: δ 106.6.
CpCp*Hf(Ph)Cl (12). A thick-walled Teflon-sealable flask was
charged with 4 (0.253 g, 0.331 mmol), and benzene (10 mL) was added.
The flask was wrapped in aluminum foil for protection from light, and
the reaction mixture was then heated to 100 °C for 3 days. Solvent
was removed in vacuo, and the yellow residue was extracted with
pentane (2 × 25 mL). The combined extracts were concentrated to ca.
35 mL and cooled to -80 °C. The product was isolated as light-yellow
1
crystals in 69% yield (0.112 g, 0.228 mmol). H NMR: δ 1.66 (s, 15
CpCp*Hf(SnPh3)NMe2·0.75(C4H10O) (8). A solution of (THF)3.5
-
H, C5Me5), 5.82 (s, 5 H, C5H5), 7.08 (m, 1 H, p-C6H5), 7.26 (m, 2 H,
m-C6H5), 7.25-7.45 (br s, 2 H, o-C6H5). Upon cooling to -70 °C, the
o-C6H5 resonances sharpened in both the 1H and 13C NMR spectra. 1H
NMR (C7D8, -70 °C): δ 1.63 (s, 15 H, C5Me5), 5.74 (s, 5 H, C5H5),
6.88 (d, 3JHH ) 6.5 Hz, 1 H, o-C6H5), p-C6H5 coincident with the aryl
C7D7H resonances, 7.30 (t, 3JHH ) 7.5 Hz, 1 H, m-C6H5), 7.35 (t, 3JHH
LiSnPh3 (0.291 g, 0.477 mmol) in THF (20 mL) was added to a solution
of 5 (0.202 g, 0.477 mmol) in THF (15 mL). The solution was stirred
in the dark at room temperature for 12 h. Solvent was removed from
the yellow-green solution, leaving a yellow foam. The foam was
extracted with Et2O (2 × 15 mL), the combined extracts were filtered
3
) 7.0 Hz, 1 H, m-C6H5), 7.85 (d, JHH ) 7.0 Hz, 1 H, o-C6H5). 13C-
(21) Holt, M. S.; Wilson, W. L.; Nelson, J. H. Chem. ReV. 1989, 89, 11-49.
(22) Shaltout, R. M.; Corey, J. C. Tetrahedron 1995, 51, 4309-4320.
{1H} NMR (C7D8, -70 °C): δ 12.28, 113.7, 120.4, 125.2, 127.8, 139.7,
9
J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005 14747

A R T I C L E S
Neale and Tilley
139.9, 196.1. Anal. Calcd for C21H25HfCl: C, 51.33; H, 5.13. Found:
C, 51.27; H, 5.37.
7.01 (m, 6 H, m-FC6H4), 7.53-7.67 (m, 6 H, o-FC6H4). 13C{1H}
NMR: δ 13.0, 112.1, 116.0, 121.2, 139.9, 146.2, 163.9. 19F{1H}
NMR: δ -114.3. 119Sn{1H} NMR: δ 116.5. Anal. Calcd for C32H32F3-
ClSnHf: C, 48.44; H, 3.94. Found: C, 48.53; H, 3.80.
CpCp*Hf(H)Ph (13). CpCp*Hf(H)Cl (0.200 g, 0.482 mmol) and
PhLi (0.041 g, 0.49 mmol) were combined in a Teflon-sealable flask.
THF (10 mL) was vacuum transferred onto the solids, and the mixture
was slowly warmed to room temperature. After ca. 10 min at ambient
temperature, solvent was removed in vacuo, leaving a beige oil. The
oil was extracted with toluene (10 mL) and filtered through a bed of
Celite (1 cm). The resulting solution was concentrated to ca. 2 mL and
cooled to -30 °C. The product was isolated as a beige solid in 75%
yield (0.164 g, 0.359 mmol). 1H NMR: δ 1.78 (s, 15 H, C5Me5), 5.79
(s, 5 H, C5H5), 6.89 (br d, 3JHH ) 6.5 Hz, 2 H, m-C6H5), 7.29 (tt, 3JHH
CpCp*Hf{Sn[p-(CF3)C6H4]3}Cl (20). HSn[(CF3)C6H4]3 (0.263 g,
0.474 mmol) and CpCp*Hf(H)Cl (0.200 g, 0.482 mmol) were added
to a Teflon-sealable Schlenk flask, and THF (10 mL) was vacuum
transferred onto the solids. The mixture was stirred and allowed to
come to room temperature over 1 h in the dark. Solvent was removed
from the resulting orange solution in vacuo, leaving an orange foam.
This foam was extracted with pentane (15 mL) and filtered via cannula.
The resulting orange pentane solution was concentrated to ca. 12 mL
and cooled to -30 °C. Impure material was isolated via cannula
filtration as an orange, crystalline solid. Recrystallization from an Et2O
solution (3 mL) layered with pentane (3 mL) at -80 °C gave
analytically pure product as a yellow powder in 26% yield (0.120 g,
0.124 mmol). 1H NMR: δ 1.67 (s, 15 H, C5Me5), 5.63 (s, 3JSnH ) 7.9
Hz, 5 H, C5H5), 7.49 (m, 6 H, m-CF3C6H4), 7.62-7.69 (m, 6 H, o-(CF3)-
C6H4). 13C{1H} NMR: δ 13.0, 112.1, 121.2, 125.3, 125.7, 130.4, 138.5,
156.2. 19F{1H} NMR: δ -62.20. 119Sn{1H} NMR: δ 106.4. Anal.
Calcd for C36H32F9ClSnHf: C, 44.66; H, 3.33. Found: C, 44.60; H,
3.29.
4
3
) 7.0 Hz, JHH ) 1.3 Hz, 1 H, p-C6H5), 7.32 (t, JHH ) 7.5 Hz, 2 H,
o-C6H5), 13.02 (s, 1 H, Hf-H). 13C{1H} NMR: δ 12.48, 110.1, 118.1,
125.8, 127.5, 201.2. Anal. Calcd for C21H46Hf: C, 55.20; H, 5.74.
Found: C, 54.94; H, 5.82.
CpCp*Hf(SnPh3)Np (14). A solution of NpLi (0.027 g, 0.35 mmol)
in toluene (15 mL) was added dropwise to a solution of 4 (0.266 g,
0.348 mmol) in THF (10 mL) at -78 °C. This solution was allowed to
gradually warm to room temperature over 2 h, at which point solvent
was removed in vacuo. The resulting yellow-brown oil was extracted
with Et2O (1 × 10 mL, 1 × 5 mL), and the combined extracts were
concentrated to ca. 5 mL. Pentane (2 mL) was added, and the resulting
solution was cooled to -30 °C. A deep orange solution was filtered
away from the sticky yellow residue that had precipitated. After
filtration, the orange filtrate precipitated yellow crystals at room
temperature, which contained the product in ca. 70% purity. These
yellow crystals were redissolved in Et2O (10 mL), and the resulting
yellow solution was concentrated to ca. 2 mL and cooled to -30 °C.
The product (80% purity) was isolated by cannula filtration as yellow
crystals in 14% yield, including the impurities (0.038 g, 0.048 mmol).
Additional crystallizations did not further purify the product. 1H
Kinetic Study of the r-Elimination of Metal Stannyl Complexes.
Samples for kinetic studies with 4 were prepared as follows: 4 (ca.
0.015 µmol) and ferrocene (ca. 0.010 µmol) were weighed into a 2.00
( 0.01 mL volumetric flask, and toluene-d8 was added to give a total
volume of 2.00 mL. Two portions (ca. 0.6 mL each) of this solution
were transferred to separate J. Young NMR tubes, which were wrapped
in aluminum foil to protect the solution from light. The tubes were
lowered into a temperature-controlled oil bath (temperature variation
e1.0 °C). Data points were collected by removing the tubes from the
oil and immediately cooling to room temperature. The tubes were
typically at room temperature for about 15 min before being returned
to the oil bath. The interim time at ambient temperature was not
included in the data analysis. Samples for other kinetic studies of 2, 4,
8, 9, 10, 18, 19, and 20 were prepared by combining the Hf-SnAr3
compound (ca. 0.015 µmol) and ferrocene (ca. 0.010 µmol) in a 1.00
( 0.01 mL volumetric flask, and toluene-d8 was added to give a total
volume of 1.00 mL.
2
2
NMR: δ -3.81 (d, 1 H, JHH ) 9.9 Hz, J117/119SnH ) 39.8, 48.3 Hz,
R-agostic, CH2CMe3), 1.08 (s, CH2CMe3), 1.76 (s, 15 H, C5Me5), 2.60
2
2
3
(d, 1 H, JHH ) 9.9 Hz, JSnH ) 39.8 Hz, CH2CMe3), 5.84 (s, JSnH
)
6.5 Hz, 5 H, C5H5), 7.17-7.21 (m, 3 H, p-C6H5), 7.28-7.31 (m, 6 H,
m-C6H5), 7.81-7.88 (m, 6 H, o-C6H5). 13C{1H} NMR: δ 12.24, 35.80,
40.95, 108.5, 118.1, 124.8, 127.7, 128.7, 138.7, 152.5. 119Sn{1H}
NMR: δ 91.3.
1
Data points were gathered by H NMR spectroscopy, and the rate
CpCp*Hf{Sn[p-(OMe)C6H4]3}Cl (18). HSn[p-(OMe)C6H4]3 (0.266
g, 0.603 mmol) and CpCp*Hf(H)Cl (0.250 g, 0.602 mmol) were added
to a Teflon-sealable Schlenk flask, and THF (10 mL) was vacuum
transferred onto the solids. The mixture was stirred and warmed to
room temperature over 10 min in the dark. Solvent was removed from
the resulting orange solution in vacuo, leaving an orange-yellow foam.
This foam was extracted with toluene (40 mL), and the extract was
filtered via cannula. The resulting solution was concentrated to ca. 15
mL, layered with Et2O (10 mL), and cooled to -30 °C. The product
was isolated via cannula filtration as a yellow powder in 46% yield
of disappearance of hafnium stannyl species was monitored by
integrating the C5Me5 or C5H5 peak relative to that of Cp2Fe. Rate
constants were calculated from first-order plots using data from the
first three to five half-lives. Finally, during some reactions other
decomposition species besides the R-elimination product were observed
as impurities. In the case where impurities were observed, the rate of
decomposition of 4 slowed slightly (by a factor of 0.8) relative to the
case where clean decomposition products were generated. However,
by silylating all glassware, 99% conversion of 4 to the R-elimination
product was observed.
1
(0.237 g, 0.277 mmol). H NMR: δ 1.83 (s, 15 H, C5Me5), 3.38 (s, 9
3
H, (OMe)C6H4), 5.87 (s, JSnH ) 7.0 Hz, 5 H, C5H5), 7.01 (m, 6 H,
Results
m-(OMe)C6H4), 7.81-7.88 (m, 6 H, o-(OMe)C6H4). 13C{1H} NMR:
δ 13.1, 54.9, 112.1, 115.0, 120.9, 139.7, 142.3, 160.2. 119Sn{1H}
NMR: δ 120.9. Anal. Calcd for C36H41O3ClSnHf: C, 50.61; H, 4.84.
Found: C, 50.94; H, 4.81.
As described above, earlier investigations established the
operation of a facile R-H-elimination process for a hafnium
hydrostannyl complex, and additional observations suggested
that aryl groups in zirconium stannyl complexes might also
undergo R-elimination. To investigate this possibility in detail,
we targeted the synthesis of pure triarylstannyl complexes of
hafnium. It was thought that such hafnium complexes might
be more stable than the analogous zirconium derivatives and
therefore be more readily isolated in pure form. Stable triaryl-
stannyl complexes of this type could serve as a starting point
in the search for well-behaved R-aryl-elimination reactions. It
seemed that such systems should also allow mechanistic studies
on this novel transformation. For the latter purpose, we desired
CpCp*Hf[Sn(p-FC6H4)3]Cl (19). HSn(p-FC6H4)3 (0.175 g, 0.432
mmol) and CpCp*Hf(H)Cl (0.179 g, 0.431 mmol) were added to a
Teflon-sealable Schlenk flask, and THF (10 mL) was vacuum trans-
ferred onto the solids. The mixture was stirred and allowed to come to
room temperature in the dark for 12 h. Solvent was removed from the
resulting yellow solution in vacuo, leaving a yellow foam. This foam
was extracted with pentane (40 mL), and the extract was filtered via
cannula. The resulting yellow solution was concentrated to ca. 10 mL
and cooled to -30 °C. The product was isolated via cannula filtration
1
as a yellow, crystalline solid in 59% yield (0.208 g, 0.254 mmol). H
3
NMR: δ 1.72 (s, 15 H, C5Me5), 5.72 (s, JSnH ) 7.6 Hz, 5 H, C5H5),
9
14748 J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005

Hafnocene Triarylstannyl Complexes
A R T I C L E S
a series of hafnium stannyl complexes for which the steric and
electronic properties at both Hf and Sn were varied. This was
accomplished by the synthesis of hafnocene stannyl derivatives
with various cyclopentadienyl-based ancillary ligands (for
varying steric factors) and with different aryl groups in the
stannyl ligand (for varying electronic factors).
after 10 min, and the amount of this compound increased
marginally over time (20% after 20 min, and 23% after 7 h).
These results suggest that the reaction of 2 with Me3SiCl does
not go to completion, and perhaps for this reason [Me2C(C5H4)2]-
Hf(SnPh3)Cl could not be isolated from preparatory-scale
reactions. This difficulty may result from the volatility of Me3-
SiCl, which is removed by evaporation during workup, resulting
in a shift of the equilibrium back to 2. Given the problems in
converting the stannyl amide 2 to additional stannyl derivatives,
we explored other synthetic routes to hafnocene stannyl
complexes.
Synthesis of Hafnium Stannyl Complexes by Salt Elimina-
tion. Another method previously reported for the synthesis of
hafnium stannyl compounds is based on salt elimination between
a metal halide and a stannyllithium reagent.6a,c-d,i For application
of this method, a hafnocene dichloride was desired. The addition
of an excess (7 equiv) of Me3SiCl to 1 provided the hafnium
dichloride starting material 3 in good yield (69%). The reaction
of this complex with LiSnPh3 (1 equiv) in benzene-d6 did not
produce the expected compound [Me2C(C5H4)2]Hf(SnPh3)Cl,
but instead gave a mixture of [Me2C(C5H4)2]Hf(Ph)Cl (47%),
[Me2C(C5H4)2]HfPh2 (22%), and unreacted 3 (25%). The
monophenyl species is characterized by ligand resonances in
the 1H NMR spectrum that reflect the unsymmetrical nature of
this compound [δ 1.15 and 1.17 (Me); 4.97, 5.41, 6.16, and
6.21 (C5H4)], while those for the diphenyl derivative are
consistent with C2 symmetry [δ 1.23 (Me); 5.26 and 6.20
(C5H4)]. Both of these compounds were independently generated
in reactions of 3 and PhLi (benzene-d6/THF solution) and
identified by 1H NMR spectroscopy. The 1:2 reaction of 3 with
LiSnPh3 (benzene-d6, room temperature, 15 min) also resulted
only in phenylation of the hafnium center, to give the diphenyl
complex [Me2C(C5H4)2]HfPh2 in 95% yield by 1H NMR
spectroscopy (eq 3). These results suggested that, like CpCp*Zr-
Synthesis of Hafnium Stannyl Complexes by Amine
Elimination. An established synthetic method for the prepara-
tion of metal stannyl derivatives involves the elimination of an
amine upon reaction of a metal amido complex with a
hydrostannane.21 For example, tetrakis(triphenylstannyl)titanium
was synthesized from Ph3SnH and Ti(NMe2)4.6b For application
of this method to the preparation of a hafnocene stannyl
complex, the bis(amido) complex [Me2C(C5H4)2]Hf(NMe2)2 (1)
was prepared in 52% yield by refluxing a toluene solution of
Me2C(C5H5)222 and Hf(NMe2)4 for 18 h. This methylene-bridged
dicyclopentadienyl ligand, which provides a very open metal-
locene framework,16 was expected to present minimal steric
resistance to rearrangements involving the metal center.
The reaction of 1 with Ph3SnH in benzene at room temper-
ature produced a red-orange solution from which [Me2C(C5H4)2]-
Hf(SnPh3)NMe2 (2) was isolated as air-sensitive, yellow crystals
in 68% yield (eq 2). The 1H NMR spectrum of 2 contains four
resonances (δ 5.06, 5.38, 5.50, and 5.68), corresponding to the
cyclopentadienyl ligand protons, and two peaks for the diaste-
reotopic methyl groups of the ligand backbone (δ 0.86 and 1.15).
This complex is further characterized by a 119Sn NMR shift of
δ 67.1, which is in a downfield region similar to that observed
for another d0 hafnium triarylstannyl derivative (Cp*2Hf(SnPh3)-
Cl; δ 114.7).16,22 As was previously noted for Zr and Hf stannyl
compounds,6i 2 is light-sensitive and decomposes completely
under ambient room lighting after 4 days (room temperature,
benzene-d6 solution), to HNMe2 (1 equiv), Ph6Sn2 (0.3 equiv),
and a complex mixture of metal species. Due to the light
sensitivity of such hafnium stannyls, the complexes described
here were manipulated under minimal ambient lighting condi-
tions.
For access to a series of related stannyl derivatives for which
the adjacent ancillary ligand is varied, it was of interest to
substitute the dimethylamide group in 2. Attempts to replace
the amide with halide substituents using HNMe2·HCl, MeI, or
acetyl chloride in benzene-d6 gave complex mixtures of
products. However, when trimethylsilyl chloride (26 equiv) was
added to 2 at room temperature, a 74% conversion to [Me2C-
(C5H4)2]Hf(SnPh3)Cl was observed after 20 min (benzene-d6
solution, by 1H NMR spectroscopy). After 8 h, this new stannyl
complex had decomposed to a number of species, including
[Me2C(C5H4)2]HfCl2 (3). The addition of 1 equiv of Me3SiCl
to 2 provided [Me2C(C5H4)2]Hf(SnPh3)Cl in a low yield (13%)
(SnPh3)Cl,6i the desired triphenylstannyl complexes are unstable
under the reaction conditions and decompose via an R-elimina-
tion process.23
Since this methylene-bridged ligand system appeared to give
hafnium stannyl compounds that were unstable toward R-elim-
ination, we turned to the CpCp*MLn mixed-ring system, which
had previously been useful in the synthesis of metal stannyl
complexes.6i,7 Fortunately, this ligand set has allowed the
synthesis and isolation of a series of stannyl complexes for use
in studies on R-aryl-elimination. Bright yellow crystals of
CpCp*Hf(SnPh3)Cl (4) were obtained as previously reported
from LiSnPh3 and CpCp*HfCl2,6i and this compound was
isolated in an improved yield of 75% (see Experimental Section).
Other hafnocene starting materials, CpCp*Hf(NMe2)Cl (5),
(23) Note that redistribution processes, which may be catalyzed by the LiCl
byproduct, may also occur, and in general, this could give more than one
product. See: Alcock, N. W.; Clase, H. J.; Duncalf, D. J.; Hart, S. L.;
McCamley, A.; McCormack, P. J.; Taylor, P. C. J. Organomet. Chem. 2000,
605, 45-54. However, the R-elimination process observed in eq 3 should
be unaffected by any such Ph/Cl scrambling, as both chloride ligands are
substituted.
9
J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005 14749

A R T I C L E S
Neale and Tilley
CpCp*Hf(Me)OTf (6), and CpCp*Hf(OMe)Cl (7), were pre-
pared from salt metathesis reactions between CpCp*HfCl2 and
LiNMe2 (5) or NaOMe (7) or by a published procedure (6).4a
Reactions of compounds 5-7 with LiSnPh3 resulted in isolation
of the stannyl complexes CpCp*Hf(SnPh3)NMe2 (8), CpCp*Hf-
(SnPh3)Me (9), and CpCp*Hf(SnPh3)OMe (10) in good yields
(58-72%) (eq 4). All three of these stannyl derivatives
is quite rapid. Compound 13 was independently isolated from
the reaction of CpCp*Hf(H)Cl with PhLi in THF, and the
1
terminal hydride ligand is characterized by a H NMR shift at
δ 13.02 and an infrared stretch at 1600 cm-1
.
To investigate the effect of a sterically demanding ligand on
R-elimination chemistry, an attempt was made to prepare the
neopentyl derivative CpCp*Hf(SnPh3)Np (Np ) neopentyl).
Unexpectedly, the reaction of 4 with NpLi (Np ) neopentyl)
gave a mixture consistent with both LiCl and LiSnPh3 elimina-
tion. After slowly warming a toluene solution from -78 °C to
room temperature over 2 h, CpCp*Hf(SnPh3)Np (14, 54%), the
triphenylstannyl phenyl derivative 11 (16%), an unidentified
product (16%), and unreacted 4 (14%) were present. Repeated
crystallizations from Et2O allowed the isolation of a small
amount of 14 [80% pure; impurities were 4 (13%) and 11 (6%)]
as yellow crystals in 14% yield. Compound 14 appears to exhibit
crystallize with solvent [8·0.75(C4H10O), 9·0.5(C5H12), 10·0.5-
(C7H8)], and the solvent-free complexes may be obtained by
crushing the crystals to a fine powder and then applying a
vacuum. These compounds are air-sensitive and (like 2) light-
sensitive in solution. Finally, similar to other hafnocene stannyl
compounds,6i,7 4, 8, 9, and 10 exhibit a yellow color that results
from weak, broad LMCT transitions centered at ca. 350 nm.
For example, the UV-visible spectra for the two dimethylamide
derivatives contain absorptions at 358 nm (2, ꢀ ) 4700 dm3
mol-1 cm-1) and 340 nm (8, ꢀ ) 5000 dm3 mol-1 cm-1).
Another hafnium stannyl complex was generated by the
reaction of 1 equiv of PhLi with 4 in benzene-d6/THF. This
reaction mixture produced a deep red solution containing one
major species in 91% yield, which appears to be CpCp*Hf-
(SnPh3)Ph (11) based on 1H and 119Sn NMR spectroscopy. The
downfield 119Sn NMR shift observed for 11 (δ 106.6) is
diagnostic for hafnocene triarylstannyl compounds, which are
characterized by resonances near δ 100 (cf. δ 112.7 for 4, δ
74.3 for 8, δ 111.5 for 9, and δ 64.3 for 10). Unfortunately,
attempts to isolate the pure compound from preparative scale
reactions were not successful. The same species was formed
from the reaction of CpCp*Hf(Ph)Cl (12) and LiSnPh3 (1 equiv),
but again the product could not be isolated cleanly via
crystallization from a variety of solvents (toluene, toluene/
pentane, toluene/Et2O, in varying ratios). Interestingly, 11 was
also produced by the addition of 2 equiv of LiSnPh3 to
CpCp*HfCl2 (benzene-d6 solution), presumably via the inter-
mediate CpCp*Hf(SnPh3)2, which may undergo rapid R-elim-
ination of Ph2Sn. These results clearly indicate that the CpCp*
ligand set provides greater stability to hafnium triphenylstannyl
derivatives than does the less sterically demanding Me2C(C5H4)2
ligand.
1
an unusual H NMR spectrum due to an R-agostic interaction
between one of the neopentyl methylene C-H bonds and Hf.
The two diastereotopic methylene protons are observed at δ
-3.80 and 2.59 (2JHH ) 9.9 Hz), and the C-H coupling
constants obtained from a proton-coupled heteronuclear multiple
quantum coherence (HMQC) experiment are 87.2 Hz (δ -3.80)
and 109 Hz (δ 2.59). The C-H coupling constant of 114 Hz
for the methyl group of CpCp*Hf(SnPh3)Me (9) lends further
support to the characterization of 14 as possessing an R-agostic
neopentyl group.24 Finally, the 13C NMR resonance for the
methylene group in 14 is observed at δ 124.8, far downfield
from the resonance observed for the hafnium methyl group in
9 (δ 58.67). It is of note that the infrared spectrum of this
compound revealed no peaks in a region of lower frequency,
as might be expected for compounds with an R-agostic
interaction,23 but the C-H coupling constants strongly suggest
the presence of a secondary interaction between the CH2 group
and Hf.
Synthesis of Hafnium Stannyl Complexes by σ-Bond
Metathesis. It has previously been observed that σ-bond
metathesis can be used to prepare metal stannyl complexes, such
as CpCp*Zr(SnPh3)Cl, that could not be obtained through salt
metathesis.6i In such procedures, a hydrostannane (R3SnH or
R2SnH2) reacts with a metal hydride or silyl complex (M-H
or M-SiR3) to eliminate H2 or HSiR3 with formation of a
M-Sn bond. This method can be synthetically more convenient
than that involving salt metathesis, since hydrostannanes are in
general more readily available than stannyllithium reagents. In
the following syntheses, purification of the stannyl complexes
was simplified by using hydrogen rather than silane elimination,
as hydrogen is more readily removed from the product.
To probe electronic factors in the migration of an aryl group
from Sn to Hf, a series of para-substitued triarylstannyl species
An attempt to obtain a stannyl hydride complex was based
on reaction of the hydride CpCp*Hf(H)Cl18 with LiSnPh3. As
shown in eq 5, this reaction exclusively produced the phenyl
1
was synthesized. By H NMR spectroscopy, it was observed
that reactions of CpCp*Hf(H)Cl with various stannanes Ar3-
SnH [Ar ) p-(OMe)C6H4 (15), p-FC6H4 (16), and p-(CF3)C6H4
(17)] produced good yields (68-84%) of the stannyl complexes
CpCp*Hf[Sn(p-(OMe)C6H4)3]Cl (18), CpCp*Hf[Sn(p-FC6H4)3]-
Cl (19), and CpCp*Hf[Sn(p-(CF3)C6H4)3]Cl (20) (eq 6). The
lower isolated yields (26-59%) for 18-20 appear to result from
competing decomposition of the hafnium hydride starting
hydride complex CpCp*Hf(H)Ph (13), even at -40 °C (by 1H
NMR spectroscopy, in toluene-d8 solution). This result suggests
that the R-elimination of a phenyl group in CpCp*Hf(SnPh3)H
(24) (a) Brookhart, M.; Green, M. L. H.; Pardy, R. B. A. J. Chem. Soc. Chem.
Commun. 1983, 691-693. (b) Brookhart, M.; Green, M. L. H. J.
Organomet. Chem. 1983, 250, 395-408.
9
14750 J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005

Hafnocene Triarylstannyl Complexes
A R T I C L E S
material during synthetic manipulations. The new stannyl
complexes are yellow in the solid state and in solution and give
rise to 119Sn NMR resonances of δ 120.9 (18), 116.5 (19), and
106.4 (20). The pattern of increasingly upfield 119Sn NMR shifts
is consistent with a decrease in electron density at Sn in going
from 18 to 19 to 20.25
Figure 1. ORTEP diagram of CpCp*Hf(SnPh3)NMe2•0.75(C4H10O) [8·
0.75(C4H10O)]. Only one of the two molecules in the unit cell is shown.
The hydrogen atoms and the C4H10O molecules were removed for clarity.
Atoms are shown as 50% probablility ellipsoids.
Crystallographic Studies. Given the paucity of structural
data for d0 metal stannyl complexes,6g,h,j,7 it was of interest to
investigate several of the complexes described above by X-ray
crystallography. The compounds [Me2C(C5H4)2]Hf(SnPh3)NMe2
(2) and CpCp*Hf(SnPh3)X [X ) Cl (4), X ) NMe2 (8), X )
Me (9), and X ) OMe (10)] all form single crystals that
provided suitable diffraction patterns for crystallographic char-
acterization (the ORTEP diagram for CpCp*Hf(SnPh3)NMe2·
0.75(C4H10) is shown in Figure 1). Surprisingly, the Hf-Sn bond
distances in these complexes do not vary considerably (2.94-
2.97 Å, Table 1). The dimethylamide derivatives 2 and 8 were
found to possess planar nitrogen atoms (sum of angles about
nitrogen 180°), and the dihedral C(Me)-N-Hf-Sn angle for
both species is approximately 60°. Dihedral angles of ca. 60°
are typical for amide ligands in metallocene complexes exhibit-
ing steric strain.26 The corresponding C(Me)-O-Hf-Sn angle
in the methoxy-substituted species 10 is slightly greater (81.9°),
presumably due to reduced steric repulsion between the Cp*
and OMe ligands.
Table 1. Selected Crystallographic Data
compound
2
4
8
9
10
Hf-Sn bond
distance (Å)
Sn-Hf-X
angle (deg)
2.9428(7) 2.9650(4) (a) 2.9694(8) 2.9740(5) 2.9556(5)
(b) 2.9658(8)
93.3(2)
87.65(4) (a) 90.6(2)
92.7(2)
X ) Me X ) OMe
90.6(2)
X ) NMe2 X ) Cl
(b) 89.7(2)
X ) NMe2
(a) 117.3
average
116.8
116.8
115.8
114.3
Hf-Sn-C(Ph)
angle (deg)
(b) 117.7
condensation was observed upon the elimination of Mes2Sn from
CpCp*Hf(SnHMes2)Cl, and this stannylene was trapped by 2,3-
dimethylbutadiene29 to give 1,1′-dimesityl-3,4-dimethylstanna-
cyclopent-3-ene.7 However, in the presence of 9 equiv of 2,3-
Mechanistic Studies on r-Aryl-Elimination Reactions. In
an initial examination of R-aryl-elimination in these hafnium
stannyl compounds, a toluene-d8 solution of the triphenylstannyl
chloride derivative 4 was heated to 100 °C. After 4 days,
complete conversion of 4 to the phenyl chloride 12 (99% yield)
was observed by 1H NMR spectroscopy (eq 7). By 119Sn NMR
spectroscopy, the cyclic polystannanes (Ph2Sn)5 and (Ph2Sn)6
were observed as the only tin-containing products (δ -205.9
and -217.2, respectively).5b,c,27,28 Presumably, these products
result from elimination of the stannylene Ph2Sn, which then
condenses to form the cyclic species. Similar stannylene
dimethylbutadiene (9 equiv), the reaction of eq 7 did not produce
the anticipated stannacyclopentene. The lack of efficient trapping
in this case may be due to the very rapid oligomerization of
Ph2Sn (k ) 108 M-1 s-1) to (Ph2Sn)n.21 A small amount of
yellow precipitate was also observed during the course of the
reaction, which was insoluble in benzene or toluene and was
only sparingly soluble in THF (despite coloring THF solutions
pale yellow, no peaks were observed from a GPC trace of this
(25) (a) Brown, H. C.; Okamoto, Y. J. Am. Chem. Soc. 1958, 80, 4979-4987.
(b) Swain, C. G.; Lupton, Jr., E. C. J. Am. Chem. Soc. 1968, 80, 4328-
4337. (c) Johnson, C. D. The Hammett Equation; Cambridge University
Press: London, 1973.
(26) Hillhouse, G. L.; Bulls, A. R.; Santarsiero, B. D.; Bercaw, J. E. Organo-
metallics 1988, 7, 1309-1312.
(27) Jousseaume, B.; Noiret, N.; Pereyre, M.; Saux, A.; France`s, J.-M.
Organometallics 1994, 13, 1034-1038.
(28) The 119Sn NMR shift of δ -208 previously reported for (Ph2Sn)6 was
observed in chloroform-d1. Dra¨ger, B.; Mathiasch, B.; Ross, L.; Ross, M.
Z. Anorg. Allg. Chem. 1983, 506, 99-109.
(29) Neumann, W. P. Chem. ReV. 1991, 91, 311-334.
9
J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005 14751

A R T I C L E S
Neale and Tilley
8 did not decompose at an appreciable rate until 115 °C.
Complex 9 decomposes 64 times faster than 4 (at 70 °C), which
undergoes R-elimination 230 times faster than 8 (at 115 °C).
On the basis of these differences, it can be determined that the
dimethylamide species 8 is more stable than the methyl
derivative 9 by at least a factor of 280. Thus, the π-donating
NMe2 ligand greatly stabilizes hafnocene stannyl complexes
toward R-phenyl-elimination, relative to a σ-only ligand such
as methyl.
Interestingly, there is a significant difference in the stabilities
of the two amide derivatives, as evidenced by the rate constants
of k ) 9.4 × 10-5 s-1 (2) and k ) 4.3 × 10-7 s-1 (8) for
disappearance of the starting compound at 115 °C. After heating
1
2 for 1 week at this temperature, the species identified by H
NMR spectroscopy were unreacted 2 (4%), [Me2C(C5H4)2]Hf-
(Ph)NMe2 (71%), and the bis(amido) complex 1 (13%). The
decomposition of 8 at 115 °C was monitored for 58 days, and
after this time the R-elimination product CpCp*Hf(Ph)NMe2
(66%), unreacted 8 (11%), and the bis(amido) product CpCp*Hf-
(NMe2)2 (2%) were present. These rate data provide additional
evidence that the methylene-bridged ligand system affords
stannyl complexes that are more susceptible to R-elimination
than analogous complexes of the CpCp* ligand set.
The methoxy derivative 10 exhibited only a 2-fold increase
in stability (k (100 °C) ) 8.8 × 10-6 s-1) relative to that for
the chloride 4 and decomposed to give CpCp*Hf(Ph)OMe in
96% yield after 3.5 days. Further, comparisons of rates lead to
the conclusion that 10 is 23 times less stable than 8. This
relatively rapid rate of decomposition is somewhat surprising,
since the σ+π-donating abilities of OMe and NMe2 are expected
to be comparable and should lead to similarly stable com-
pounds.30 However, in this case, the lower steric bulk of OMe
may be an important factor in determining the higher rate of
R-elimination for 10.
Finally, decomposition of the neopentyl complex CpCp*Hf-
(SnPh3)Np (14) at 100 °C occurred with a rate constant of k )
4 × 10-6 s-1 (over three half-lives), providing evidence that
this stannyl complex is 160 times more stable than the methyl
derivative 9 (determined from similar relationships as above).
Although there is an added electronic effect due to the R-agostic
interaction observed between this ligand and Hf, it is likely that
the enhanced steric bulk of the neopentyl ligand is the primary
factor in slowing the rate of decomposition relative to the methyl
derivative 9 (vide infra). In summary, the order of stability
toward R-phenyl-elimination provided by the ancillary ligands
is NMe2 > Np > OMe > Cl > Me.
Figure 2. Eyring plot for the rate of disappearance of [4] at different
temperatures.
solution). This insoluble precipitate could therefore be due to
higher molecular weight (Ph2Sn)n, as it has previously been
noted that Ph2Sn oligomers (Mw/Mn ) 2200/900) are only
slightly soluble in THF.5b Further support for this conclusion
comes from an elemental analysis of this precipitate, which gave
a carbon/hydrogen ratio consistent with (Ph2Sn)n. The reaction
of eq 7 appears to be irreversible, as the product mixture did
1
not contain observable quantities of 4 (by H NMR spectros-
copy) after 1 week at room temperature.
The disappearance of 4, monitored by NMR spectroscopy,
obeys first-order kinetics. An Eyring plot (Figure 2) of rate data
for the temperature range 70-115 °C provided the activation
parameters ∆Hq ) 24 (1) kcal/mol and ∆Sq ) -15 (1) eu. These
numbers are consistent with an ordered transition state, as might
be expected for a unimolecular R-elimination in which a phenyl
group migrates from Sn to Hf. In the polar solvent o-
dichlorobenzene (ꢀ ) 10.12; versus ꢀ ) 2.379 for toluene), the
rate of R-elimination at 100 °C was not significantly affected
(k ) 2.5 × 10-5 s-1; versus k ) 2.3 × 10-5 s-1 in toluene).
Thus, the reaction appears to involve a relatively nonpolar
transition state.
Despite the fact that the final product mixtures for some of
the decompositions discussed herein suggest a nonselective
process (observed yields of hafnium aryl products 66-99%),
their clean, first-order kinetics (with respect to decay of the
initial stannyl complex) are consistent with a single, initial step
in the decomposition process. We propose that this step
corresponds to an R-aryl-elimination. The unidentified side
products likely result from secondary reactions involving the
Hf-Ar product at these high temperatures (e.g. redistribution
at hafnium, reactions with the solvent, etc.).
To investigate the influence of electronic inductive effects
on the rate of R-elimination, kinetic studies of the decomposi-
tions of CpCp*Hf(SnPh3)Cl (4), CpCp*Hf[Sn(p-(OMe)C6H4)3]-
Cl (18), CpCp*Hf[Sn(p-FC6H4)3]Cl (19), and CpCp*Hf[Sn(p-
(CF3)C6H4)3]Cl (20) were undertaken. The fastest rate of
R-elimination was observed for the p-methoxy-substituted
derivative 18 [k (100 °C) ) 1.8 × 10-4 s-1; 95% yield of
CpCp*Hf(p-(OMe)C6H4)Cl], while the p-trifluoromethyl species
20 was the most stable complex in this series [k (100 °C) )
3.1 × 10-6 s-1; 85% yield of CpCp*Hf(p-(CF3)C6H4)Cl]. The
The influence of ancillary ligands on the rate of R-elimination
was investigated with the compounds [Me2C(C5H4)2]Hf(SnPh3)-
NMe2 (2) and CpCp*Hf(SnPh3)X [X ) NMe2 (8), Me (9), and
OMe (10)]. The methyl derivative 9 was found to be the least
stable, decomposing very rapidly at 70 °C (k ) 9.6 × 10-5
s-1; 5.6 × 10-6 s-1 at 45 °C), to CpCp*Hf(Ph)Me in 82% yield.
In contrast to the relatively rapid decomposition of the methyl
derivative 9, the dimethylamide-substituted compounds 2 and
p-fluorophenyl compound 19 [k (100 °C) ) 2.9 × 10-5 s-1
;
98% yield of CpCp*Hf(p- FC6H4)Cl] was found to undergo
R-elimination at the same rate as 4. The Hammett plot
(30) Caulton, K. G. New J. Chem. 1994, 18, 25-41.
9
14752 J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005

Hafnocene Triarylstannyl Complexes
A R T I C L E S
Given the planar geometry for the nitrogen atoms in both
dimethylamide derivatives 2 and 8, we can assume the existence
of a π-bond between the hafnium and nitrogen atoms. As was
previously noted, π-donation of this type is typical for early
metal amide complexes.26 Despite the interaction between the
nitrogen lone pair and hafnium, rotation about the Hf-N bond
1
is quite facile (by H NMR spectroscopy; the amide methyl
groups are equivalent in toluene-d8 solution down to -80 °C).
Also, π-bonding may be assumed for the Hf-O interaction in
10, and the larger C(Me)-O-Hf-Sn dihedral angle of 81.9°
(cf. ca. 60° for the C(Me)-N-Hf-Sn dihedral angles in 2 and
8) is presumably made possible by the less bulky methoxy
ancillary ligand (vide supra). Since 10 undergoes R-phenyl-
elimination significantly faster than 2 and 8, this result suggests
that steric properties of the ancillary ligands are quite important
in determining the rate of decomposition.
The results presented here provide further evidence that
hafnocene stannyl complexes have a marked tendency to
decompose via R-elimination. Interestingly, this type of de-
composition reaction appears to be quite rare for d0 complexes,
although few analogous d0 metal-main group systems have been
investigated. More generally, migratory R-eliminations are well-
known in transition metal chemistry, but such reactions are
typically associated with higher dn configurations. Such re-
arrangements usually result in conversion of a complex of the
type LnM-ERnR′ to LnM(dERn)R′. Thus, in this case the
“eliminated” fragment remains bonded to the metal center. For
this type of R-elimination reaction, at least two electrons are
required for the reacting metal center (dn configurations with n
g 2). For example, Green’s complex [Cp2W(C2H4)(CH3)]+[PF6]-
loses ethylene and undergoes R-H-elimination to produce the
methylidene hydride [Cp2W(dCH2)(H)]+[PF6]-.33 Also, Schrock
found that the reduction of Cp*Ta(CH2CMe3)Cl3 with 2 equiv
of Na/Hg amalgam in the presence of PMe3 gives the R-migra-
tion product Cp*Ta(dCHCMe3)(H)(PMe3)Cl.34 Decarbonyla-
tion9b-e and desulfination35 reactions from transition metal acyl
and sulfinato complexes, respectively, represent related pro-
cesses.
Figure 3. Hammett correlation for R-aryl-elimination.
constructed from these data (Figure 3) represents a linear
correlation with a negative slope (F ) -2.13). The magnitude
of this F value indicates that there is a significant electronic
effect on the rate of decomposition, and the negative sign implies
that electron donor groups promote the R-elimination process.
Discussion
A variety of methods have been reported for the synthesis of
d0 stannyl complexes.21,31 In the investigations described here,
methods based on amine elimination, salt elimination, and
σ-bond metathesis were used to prepare a number of new
hafnium stannyl compounds. The salt elimination route, em-
ploying a stannyllithium derivative, is somewhat limited in its
utility. For example, the syntheses of CpCp*Zr(SnPh3)Cl6i and
CpCp*Hf(SnMe3)Cl32 were not successful using salt elimination
but could be achieved via σ-bond metathesis routes. As
previously noted (vide supra), hydrostannanes are more practical
stannylating reagents, as they are more convenient than stan-
nyllithium reagents to prepare and store. In general, the most
useful synthetic route to hafnocene stannyl species appears to
involve the σ-bond metathesis reaction of a metal hydride with
a hydrostannane.
The Hf-Sn bond distances for [Me2C(C5H4)2]Hf(SnPh3)-
NMe2 (2), CpCp*Hf(SnPh3)Cl (4), CpCp*Hf(SnPh3)NMe2 (8),
CpCp*Hf(SnPh3)Me (9), and CpCp*Hf(SnPh3)OMe (10) do not
vary significantly. Further, these Hf-Sn bond distances are
slightly shorter than the analogous values reported for the
sterically encumbered stannyl complexes CpCp*Hf(SnHMes2)-
Cl [3.0073(6) Å]7 and {MeSi[SiMe2N(4-CH3C6H4)]3}SnHfCp2-
Cl [3.0231(2) Å].6j Thus, these results suggest that the Hf-Sn
bond in compounds 2, 4, 8, 9, and 10 is not under significant
steric pressure. Finally, a greater Hf-Sn bond length does not
necessarily lead to an increased rate of R-elimination, as might
have been expected if this factor was significant in destabilizing
the complex.
Reactions that are more relevant to those described here
involve elimination of an ERn species from a LnM-ERnR′
complex, to form LnM-R′. In cases where LnM-R′ is the first
observed product, the dn electron configuration is typically d0,
which often leads to a weak Ln(R′)M-ERn interaction and rapid
loss of ERn. This is seen, for example, in decarbonylations of
d0 acyl derivatives.8,9a Other examples of this type of reactivity
involve higher dn configurations and elimination of a relatively
stable ERn species. For example, CpM(CO)3PbMe3 (M ) Cr,
Mo, W; d4) and CpFe(CO)2PbMe3 (d6) decompose thermally
or photolytically to the corresponding methyl species, with
elimination of Me2Pb.36
R-Elimination processes related to those described here have
also been described for main group compounds in which the
reacting main group center is closed shell. Such R-elimination
(33) (a) Cooper, J. N.; Green, M. L. H. J. Chem. Soc. Chem. Comm. 1974,
208-209 and 761-762 (b) Cooper, J. N.; Green, M. L. H. J. Chem. Soc.
Dalton Trans. 1979, 1121-1127.
(34) Schrock, R. R. Acc. Chem. Res. 1979, 12, 98-104.
(35) (a) Kubota, M.; Blake, D. M. J. Am. Chem. Soc. 1971, 93, 1368-1373.
(b) Downs, R. L.; Wojcicki, A. Inorg. Chim. Acta 1978, 27, 91-103.
(36) (a) Pannell, K. H. J. Organomet. Chem. 1980, 198, 37-40. (b) Pannell, K.
H.; Kapoor, R. N. J. Organomet. Chem. 1981, 214, 47-52 and 1984, 269,
59-63.
(31) Petz, W. Chem. ReV. 1986, 86, 1019-1047.
(32) Neale, N. R.; Tilley, T. D. Tetrahedron 2004, 60, 7247-7260.
9
J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005 14753

A R T I C L E S
Neale and Tilley
decomposition pathways have been extensively studied for oligo-
and polysilanes.37 These studies suggest that polysilanes R3Si-
(SiR2)nR undergo R-elimination as a major decomposition
pathway, to give R3Si(SiR2)n-1R and R2Si. Such reactions have
been observed in other group 14 compounds, such as PhMe2-
GeSiMe3, which thermally decomposes via competitive Me2Si
and Me2Ge elimination.38
phenonium ion having an sp3-hybridized carbon atom on the
phenyl ring.40 In subsequent mechanistic studies that supported
the existence of this phenonium ion, Brown et al. determined
that for the migration of para-substituted phenyl groups, the
Hammett correlation has a regression constant of F ) -1.46.41
The negative F value observed for the Hammett correlation
described here suggests that these reactions may be described
as nucleophilic migrations of the aryl group. We therefore
propose the concerted transition state A (eq 9), which involves
We have previously shown that CpCp*Hf(SnHMes2)Cl
undergoes decomposition via stannylene elimination,7 and this
now appears to be a general decomposition mode for d0 hafnium
stannyl complexes. For the triarylstannyl compounds described
here, solution-phase thermal decompositions are well-behaved
and amenable to kinetic studies, which have provided insight
into the mechanism of this reaction. The R-elimination decom-
position process was studied in most detail for the parent
compound CpCp*Hf(SnPh3)Cl (4). Kinetic studies of this
transformation revealed a first-order decomposition process,
which is consistent with an R-elimination pathway in which
intramolecular phenyl migration occurs. The activation param-
eters suggest an ordered transition state (∆Sq < 0), which might
be expected for migration of the phenyl group from Sn to Hf.
Also, since a polar solvent has almost no effect on the rate of
reaction, we conclude that significant charge separation is not
involved in the rate-determining step.
A Hammett correlation based on data derived from the para-
substituted triaryl derivatives gave a negative slope (F ) -2.13),
which indicates transition state stabilization (and rate enhance-
ment) by electron donor groups. Related Hammett correlations
have previously been determined for reactions involving the
migration of para-substituted phenyl groups to an electrophilic
center. A classic example of a migration of an aryl group to an
electrophilic carbon center is found in the pinacol rearrangement
(eq 8).39 Cram suggested that the stereospecificity of this process
an interaction of the electrophilic hafnium center with the
nucleophilic migrating group. Related transition states, possess-
ing this three-center, four-electron bonding, have previously
been proposed for R-eliminations in disilanes37d,e,j and distan-
nanes42 (e.g., eq 10). A second, perhaps less significant, factor
that may contribute to this Hammett correlation relates to
stabilization of the stannylene elimination product, as electron-
donating groups are known to stabilize such species.29
Further evidence in support of this mechanism was obtained
from kinetic studies of the thermal decompositions of CpCp*Hf-
(SnAr3)X (X ) Cl, NMe2, Me, OMe, and Np). The observed
influence of the X ligand on the stability of the complex (NMe2
> Np (R-agostic) > OMe > Cl > Me . H, SnR3) appears to
reflect the π-donating ability of this ancillary ligand. A greater
degree of π-donation should result in a less electrophilic hafnium
center, and therefore a slower R-elimination rate. This π-dona-
tion should populate the otherwise vacant metallocene a1
orbital,43 which appears to be required to accommodate the
migrating aryl group. The effect of π-bonding is therefore to
decrease the interaction of this orbital with the Sn-C(Ph) bond.
Presumably, this orbital is also required for formation of the
initial product of elimination, the 18-electron stannylene com-
plex CpCp*Hf(Ar)(X)(SnAr2). This unobserved intermediate (B)
would possess a very weakly bound stannylene ligand (eq 9).
Although formation of such an intermediate is not required for
the mechanism proposed here, similar species have been
observed as products of analogous transformations (e.g., de-
may be explained by a mechanism involving an intermediate
(37) (a) Pitt, C. G. In Homoaromatic Rings, Chains and Macromolecules of
Main-Group Elements; Rheingold, A. L., Ed.; Elsevier Scientific: New
York, 1977; Chapter 8. (b) Braunstein, P.; Morise, X. Chem. ReV. 2000,
100, 3541-3552 and references therein. Specific reviews on polysilanes:
(c) Ziegler, J. M. Mol. Cryst. Liq. Cryst. 1990, 190, 265-282. (d)
Vanderwielen, A. J.; Ring, M. A.; O’Neal, H. E. J. Am. Chem. Soc. 1975,
97, 993-998. (e) Dzarnoski, J.; Rickborn, S. F.; O’Neal, H. E.; Ring, M.
A. Organometallics 1982, 1, 1217-1220. (f) Nares, K. E.; Harris, M. E.;
Ring, M. A.; O’Neal, H. E. Organometallics 1989, 8, 1964-1967. (g)
Walker, K. L.; Jardine, R. E.; Ring, M. A.; O’Neal, H. E. Int. J. Chem.
Kinet. 1998, 30, 69-88. (h) Ring, M. A.; O’Neal, H. E.; Walker, K. L.
Int. J. Chem. Kinet. 1998, 30, 89-97. (i) Trefonas, P., III; West, R.; Miller,
R. D. J. Am. Chem. Soc. 1985, 107, 2737-2742. (j) Karatsu, T.; Miller, R.
D.; Sooriyakumaran, R.; Michl, J. J. Am. Chem. Soc. 1989, 111, 1140-
1141. (k) Walsh, R. Organometallics 1988, 7, 75-77. (l) Bell, T. N.;
Perkins, K. A.; Perkins, P. G. J. Phys. Chem. 1984, 88, 116-118.
(38) Bobbitt, K.; Maloney, V. M.; Gaspar, P. P. Organometallics 1991, 10,
2772-2777.
(40) Cram, D. J. J. Am. Chem. Soc. 1949, 71, 3863-3870.
(41) (a) Brown, H. C.; Kim, C. J.; Lancelot, C. J.; Schleyer, P. v. R. J. Am.
Chem. Soc. 1970, 92, 5244-5245. (b) Brown, H. C.; Kim, C. J. J. Am.
Chem. Soc. 1971, 93, 5765-5773.
(42) (a) Schro¨er, U.; Neumann, W. P. Angew. Chem. Int. Ed. Eng. 1975, 14,
246-247. (b) Scherping, K.-H.; Neumann, W. P. Organometallics 1982,
1, 1017-1020. (c) Fobbe, H.; Neumann, W. P. J. Organomet. Chem. 1986,
303, 87-98.
(39) Bachmann, W. E.; Ferguson, J. W. J. Am. Chem. Soc. 1934, 56, 2081-
2084.
(43) Albright, T. A.; Burdett, J. K.; Whangbo, M.-H. Orbital Interactions in
Chemistry; Wiley: New York, 1985; Section 20.4.
9
14754 J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005

Hafnocene Triarylstannyl Complexes
A R T I C L E S
Conclusions
carbonylation reactions of transition metal acyl complexes, for
which the rates are slowed by π-donating ancillary ligands).44
The nature of the bis(cyclopentadienyl) ligand set also appears
to play a significant role in determining the stability of the
stannyl complex. Decompositions of the dimethylamide com-
pounds 2 and 8 demonstrate that, relative to the methylene-
bridged [Me2C(C5H4)2] ligand, the mixed-ring CpCp* ligand
set stabilizes a hafnium stannyl complex toward R-elimination
by over 2 orders of magnitude. The difference between these
two ligand systems may stem from both electronic and steric
factors. First, the permethylated Cp* ligand is significantly more
electron-donating than a cyclopentadienyl ring in [Me2C-
(C5H4)2], and the ansa structure of the latter ligand leads to less
efficient donation to the metal center.45 In addition, it is likely
that the more bulky ligand system stabilizes the hafnium stannyl
complexes by inhibiting the migration process. This steric effect
appears to be operative in the much slower decomposition of
CpCp*Hf(SnPh3)NMe2 (8) compared to CpCp*Hf(SnPh3)OMe
(10), since NMe2 and OMe are expected to have similar π-donor
abilities,30 and in the drastically reduced decomposition rate for
CpCp*Hf(SnPh3)Np (14) relative to CpCp*Hf(SnPh3)Me (9).
The results reported here describe a previously little known
reaction type for d0 metal complexes. More research with early
transition metal-main group compounds may well show that
this elimination process represents a common transformation
for such systems. Since R-elimination appears to be rather facile
for d0 group 4 stannyl compounds, this is likely an important
process in the dehydropolymerization of secondary hydrostan-
nanes to polystannanes by d0 transition-metal catalysts.32 Finally,
this chemistry may have significant implications for the
development of new catalytic reactions, and these issues will
be addressed in future investigations.
Acknowledgment is made to the National Science Founda-
tion for their generous support of this work.
Supporting Information Available: Detailed 13C NMR
characterization and IR and melting point data for new
compounds; synthesis for stannanes 15, 16, and 17; crystal data
(for 2, 4, 8, 9, 10) and ORTEP diagrams (for 2, 4, 9, 10); and
tables of kinetic data, rate constants for the disappearance of
CpCp*Hf[Sn(p-XC6H4)3]Cl, and rate constants for the disap-
pearance of 2, 4, 8, 9, and 10 at different temperatures are
available. This material is available free of charge via the
Internet at http://pubs.acs.org.
(44) (a) Marsella, J. A.; Moloy, K. G.; Caulton, K. G. J. Organomet. Chem.
1980, 201, 389-398. (b) Marsella, J. A.; Huffman, J. C.; Caulton, K. G.;
Longato, B.; Norton, J. R. J. Am. Chem. Soc. 1982, 104, 6360-6368.
(45) Zachmanoglou, C. E.; Docrat, A.; Bridgewater, B. M.; Parkin, G.; Brandow,
C. G.; Bercaw, J. E.; Jardine, C. N.; Lyall, M.; Green, J. C.; Keister, J. B.
J. Am. Chem. Soc. 2002, 124, 9525-9546.
JA0529586
9
J. AM. CHEM. SOC. VOL. 127, NO. 42, 2005 14755