Ethylene polymerization with half-sandwich allyl imido complexes of tantalum

Article abstract of DOI:10.1021/om9703494

The imido-based half-sandwich complex Cp * Ta(=N(2,6-diisopropylphenyl))(η₁-C₃H ₅)(η³-C₃H₅)(3) polymerizes ethylene in the presence of [(C₆H₅)₃C]⁺

Full text of DOI:10.1021/om9703494

2500
Organometallics 1997, 16, 2500-2502
Eth ylen e P olym er iza tion w ith Ha lf-Sa n d w ich Allyl Im id o
Com p lexes of Ta n ta lu m
David M. Antonelli,† Ann Leins, and J effrey M. Stryker*
Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G2G2
Received April 24, 1997X
Summary: The imido-based half-sandwich complex
Cp*Ta(dN(2,6-diisopropylphenyl))(η1-C3H5)(η3-C3H5) (3)
polymerizes ethylene in the presence of [(C6H5)3C]+-
[(C6F5)4B]- or B(C6F5)3, while chloride and alkyl deriva-
tives are inactive in the presence of MAO or alkyl ab-
straction reagents. In contrast, however, Cp*Ta(dNSi-
(tert-butyl)3)Cl2 (5) polymerizes ethylene in the presence
of MAO while Cp*Ta(dNSi(tert-butyl)3)(η1-C3H5)(η3-
C3H5) (6) and Cp*Ta(dNSi(tert-butyl)3)Me2 (4) are es-
sentially inactive in the presence of alkyl abstractors.
as a stable ancillary ligand at early transition metal
centers, it might be expected to function as such in
Ziegler-Natta systems.
Previous attempts to develop a niobium or tantalum
half-sandwich imido catalyst for polymerization of ole-
fins have been largely unsuccessful.12,13 Complexes of
the type [CpTa(dNR)Cl2] and Cp*M(dNR)Cl2 (Cp )
C5H5; Cp* ) C5Me5; R ) tert-butyl, 2,6-diisopropyl; M
) Nb, Ta) are not olefin polymerization catalysts in the
presence of excess methylalumoxane (MAO).14 Fur-
thermore, it has not been possible to generate related
half-sandwich alkyl cations12b from dialkyl precursors
using a wide variety of alkyl abstraction reagents.15 The
reason for the inaccessibility of these species is not
understood, but may be related to the propensity for
imido transmetalation16 or bridging to form inactive
dimeric complexes.12b
To extend our previous investigations of group 4
metallocene bis(allyl) rearrangements17 and the reac-
tions of cationic permethylmetallocene allyl complexes,18
the chemistry of the isoelectronic group 5 half-sandwich
imido template was targeted for development. For
zirconium, the crystalline, thermally, stable 16-electron
complex [Cp*2Zr(C3H5)]+[BPh4]- 1 was shown to be a
single-component catalyst for the polymerization of
ethylene.18a,19 The permethylzirconocene methyl and
Since the first synthesis in the early 1980s,1 there
have been numerous suggestions that half-sandwich
imido complexes of the group 5 metals would display
chemistry similar to that observed for the group 4 bent
metallocenes.2,3 This has been attributed to the isolobal
relationship between the two compound classes,4 and
indeed, several observations of structural and reactivity
similarities have been reported for half-sandwich imido
complexes of tantalum and niobium.4,5 Despite two
recent reports of olefin polymerization activity for imido
complexes of vanadium, however, no reports of active
catalysts derived from the heavier group 5 elements
have been published.6 Because of its small size and
relative inertness, the imido ligand is an attractive
alternative to many other dianionic cyclopentadienyl
ligand equivalents, such as dicarbollide,7 borollide,8
trimethylenemethane,9 and butadiene,10 which have
been used for Ziegler-Natta olefin polymerization.
Since the imido ligand has seen extensive use in ring-
opening metathesis polymerization (ROMP) chemistry11
† Current address: Department of Chemistry, University of Sussex,
Brighton, BN1 9RH, England.
X Abstract published in Advance ACS Abstracts, May 15, 1997.
(1) Mayer, J . M.; Curtis, C. J .; Bercaw, J . E. J . Am. Chem. Soc. 1983,
105, 2651.
(2) Gibson, V. C. J . Chem. Soc., Dalton Trans. 1994, 11, 1607.
(3) Antonelli, D. M.; Mountford, P.; Green, M. L. H. J . Organomet.
Chem. 1992, 438, C4.
hydride cations are intractable solids or oils.15b Polym-
erization catalysis is presumably initiated by isomer-
(4) (a) Gibson, V. C.; Williams, D. N.; Clegg, W.; Hockless, D. C. R.
Polyhedron 1986, 8, 1819. (b) Siemling, U.; Gibson, V. C. J . Organomet.
Chem. 1992, 426, C5. (c) Poole, A. D.; Gibson, V. C.; Clegg, W. J . Chem.
Soc., Chem. Commun. 1992, 237. (d) Williams, D. N.; Mitchell, J . P.;
Poole, A. D.; Siemling, U.; Clegg, W.; Hockless, D. C. R., O’Neil, P. A.;
Gibson, V. C. J . Chem. Soc., Dalton Trans. 1992, 9, 739.
(5) (a) Poole, A. D.; Williams, D. N.; Kenwright, A. M.; Gibson, V.
C.; Clegg, W.; Hockless, D. C. R.; O’Neil, P. A. Organometallics 1993,
12, 2549. Gibson, V. C.; Poole, A. D. J . Chem. Soc., Chem. Commun.
1995, 2261.
(10) Mashima, K.; Fujikawa, S.; Nakamura, A. J . Am. Chem. Soc.
1993, 115, 10990. Mashima, K.; Fujikawa, S.; Tanaka, Y.; Urata, H.;
Oshiki, T.; Tanaka, E.; Nakamura, A. Organometallics 1995, 14, 2633.
(11) (a) Schrock, R. R. Acc. Chem. Res. 1990, 23, 158. (b) Torecki,
R.; Schrock, R. R. J . Am. Chem. Soc. 1990, 112, 2448. (c) Parshall, G.
W.; Nugent, W. A. Chemtech 1988, 18, 184.
(6) (a) CpV(dNtol)Cl2 (tol ) toluyl) in conjunction with MAO or
diethylaluminum chloride, see: Coles, M. P.; Gibson, V. C. Polym. Bull.
1994, 33, 529. An analogous niobium complex showed but marginal
activity. (b) Hydrotris(pyrazolyl)borato imido complexes of vanadium,
see: Scheuer, S.; Fischer, J .; Kress, J . Organometallics 1995, 14, 2627.
(c) Bis(imido) complexes of chromium are also active, see: Poole, A.
D.; Gibson, V. C. J . Chem. Soc., Chem. Commun. 1995, 2261.
(7) (a) Crowther, D. J .; Baenziger, N. C.; J ordan, R. F. J . Am. Chem.
Soc. 1991, 113, 1455. (b) Bazan, G. C.; Schaefer, W. P.; Bercaw, J . E.
Organometallics 1993, 12, 2126.
(12) (a) Antonelli, D. M.; Mountford, P.; Green, M. L. H. J . Orga-
nomet. Chem. 1992, 438, 13. (b) Antonelli, D. M.; Gomes, P. T.;
Mountford, P.; Green, M. L. H. J . Chem. Soc. Dalton, in press.
(13) Gibson, V. C. Personal communication to D.A., 1992.
(14) Coles, P. M.; Dalby, C. I.; Gibson, V. C.; Clegg, W.; Elsegood,
M. R. J . J . Chem. Soc., Chem. Commun. 1995, 1709.
(15) (a) J ordan, R. F. Adv. Organomet. Chem. 1991, 32, 325. (b)
Hlatky, G. G.; Turner, H. W.; Eckman, R. R. J . Am. Chem. Soc. 1989,
111, 13, 763. (c) Turner, H. W. Eur. Patent Application 277004, 1988.
(d) Chien, J . C. W.; Tsai, W.-M.; Rausch, M. D. J . Am. Chem. Soc. 1991
113, 8570. (e) Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc.
1991, 113, 3623. Yang, X.; Stern, C. L.; Marks, T. J . J . Am. Chem.
Soc. 1994, 116, 10015.
(8) (a) Herberich, G. E. J . Organomet. Chem. 1987, 319, 9. (b) Quan,
R. W.; Bazan, G. C.; Kiely, A. F.; Schaefer, W. P.; Bercaw, J . E. J . Am.
Chem. Soc. 1994, 116, 4489.
(9) Bazan, G. C.; Rodriguez, G.; Cleary, B. P. J . Am. Chem. Soc.
1994, 116, 4489.
(16) Mitchell, J . P. Ph.D. Thesis, University of Durham, Durham,
U.K., 1992.
S0276-7333(97)00349-X CCC: $14.00 © 1997 American Chemical Society

Communications
Organometallics, Vol. 16, No. 12, 1997 2501
ization of the fluxional allyl ligand in complex 1 from
η3- to η1-coordination to generate an active 14-electron
complex. Nucleophilic addition to this complex occurs
at the central carbon of the allyl ligand, supporting a
bonding picture in which positive charge from the highly
electrophilic metal center is delocalized onto the central
carbon.18a These results suggested that cationic allyl
complexes of the group 5 metals might be more stable
than the corresponding alkyl derivatives, providing
sufficient electronic stabilization to be isolated or, at a
minimum, to be prepared in situ and used as polymer-
ization catalysts.
with acetone, gives polyethylene, pure by infrared
spectroscopy and elemental analysis.22 An active cata-
lyst is also obtained from complex 3 upon treatment
with B(C6F5)3 (1 equiv) in fluorobenzene under eth-
ylene. No significant polymerization is observed using
this template in toluene.
In contrast to the permethylzirconocene system, all
attempts to isolate a stable cationic allyl complex from
the reaction between complex 3 and [Ph3C]+[B(C6F5)4]-
lead only to the formation of intractable oils, as do other
standard approaches to cation generation. The reaction
between 3 and [Cp2Fe]+[B(C6F5)4]-,23 for example, gives
a paramagnetic green oil with no catalytic activity.
Protonolysis using anilinium salts or treatment with
[(C6F5)3B] leads to the isolation of intractable mixtures,
which also show no catalytic activity.
15e
4d
Treatment of Cp*Ta(dN(2,6-diisopropylphenyl))Cl2
(2) with allylmagnesium bromide (2 equiv) leads to the
formation of the fluxional bis(allyl) complex 3.20,21
Complex 3 reacts with [Ph3C]+[B(C6F5)4]-15d in fluo-
robenzene under 80 psi of ethylene to yield an orange
solution. Within a few minutes formation of polyeth-
To improve our understanding of the chemistry of this
template, a second half-sandwich imido system was
constructed. The bulky (tri-tert-butylsilyl)imido ligand
sterically shields low-coordinate metal centers, prevent-
ing dimerization by blocking the γ-sphere of the metal
while maintaining a reasonably unencumbered region
in the immediate vicinity of the metal center.24 On this
basis, we anticipated that complexes of the type [Cp*M-
(dNSi(tert-butyl)3)R]+ (M ) Nb, Ta; R ) allyl, alkyl, H)
would be more stable toward deactivation than those
arylimido and alkylimido systems studied previously.
Analogous to a previously developed procedure,1 yellow
crystalline [Cp*Ta(dNSi(tert-butylsilyl)3)Me2] (4) was
prepared in 87% yield by heating Cp*TaMe3Cl25 and
LiNHSi(tert-butyl3)24 in toluene overnight at 100 °C,
followed by removal of the solvent in vacuo, tritura-
tion with pentane and crystallization from cold pen-
ylene is evident and after approximately 0.5 h stirring
completely stops. Workup using HCl/MeOH, followed
by filtration and washing of the resultant white solid
tane.20,21 Heating complex
4
with
2 equiv of
(17) (a) Tjaden, E. B.; Stryker, J . M. J . Am. Chem. Soc. 1993, 115,
2083. (b) Casty, G. C.; Stryker, J . M. J . Am. Chem. Soc. 1995, 117,
7814.
[iPr2EtNH]+Cl- in THF at 70 °C gave [Cp*Ta(dNSi-
(tert-butyl)3)Cl2] (5) in 84% yield after crystallization
from pentane/toluene.20,21
(18) (a) Tjaden, E. B.; Casty, G. L.; Stryker, J . M. J . Am. Chem.
Soc. 1993, 115, 9814. See also: Tjaden, E. B. Ph.D. Dissertation,
Indiana University, Bloomington, IN, 1993. (b) Antonelli, D. M.;
Tjaden, E. B.; Stryker, J . M. Organometallics 1994, 13, 763.
(19) Erker has recently demonstrated that zirconocene butadiene
complexes upon treatment with strong Lewis acids polymerize ethyl-
ene, a reaction proposed to proceed via intermediate formation of a
cationic, boron-substituted η3-allylic intermediate, see: Temme, B.;
Erker, G.; Karl, J .; Heinrich, L.; Frolich, R.; Kotila, S. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1755.
(20) Experimental procedures and compound characterization data
are included as Supporting Information.
(21) Preparative details.20 Complex 3: To a slurry of Cp*Ta(dN-
(2,6-diisopropylphenyl))Cl2,4d 2 (1.36 g, 2.42 mmol), in Et2O (50 mL)
at room temperature was added allyl magnesium chloride (2.4 mL,
2.0 M in THF, 4.8 mmol) dropwise by syringe under
a nitrogen
atmosphere. The resulting light yellow suspension was stirred for 2
h, diluted with pentane (50 mL), filtered through Celite, and evapo-
rated in vacuo to dryness. The residue was taken up in warm pentane
(30 mL), filtered through Celite, and cooled to -78 °C overnight to
In contrast to the arylimido system, in the presence
of MAO, complex 5 is an ethylene polymerization
catalyst.26 Qualitatively, it is more active than the
catalyst derived from bis(allyl) complex 3, although the
yield pale yellow crystals. Yield: 0.968 g (73%). Anal. Calcd for C28H42
-
NTa: C, 58.63; H, 7.32. Found: C, 58.11; H, 7.41. 1H NMR (C6D6): δ
1.38 (d, J ) 8.4 Hz, 12H, CHMe2), 1.70 (s, 15H, Me5C5), 3.19 (br s, 8H,
Ht), 3.87 (sept J ) 8.4 Hz, 2H, CHMe2), 6.13 (quint J ) 11.3 Hz, 2H,
Hc), 6.96-7.28 (m, 3H, phenyl). 1H NMR (C7D8, -40 °C): 3.19 (d, J )
11.3 Hz, 8H). Complex 5: A toluene suspension (100 mL) of Cp*TaMe3-
Cl (5.30 g, 13.33 mmol) and LiN(H)Si(tBu)324 (2.95 g, 13.33 mmol) under
nitrogen was stirred at 70 °C for 24 h, until all the lithium salt had
dissolved. The solvent was removed in vacuo, and the orange residue
was washed with pentane (2 × 10 mL), concentrated, and cooled to
-40 °C to yield orange microcrystals of Cp*Ta[dNSi(tBu)3]Me2 (4).
Yield: 4.70 g, (89%). Anal. Calcd for C24H48NSiTa: C, 51.46; H, 8.64;
N, 2.50. Found: C, 51.17; H, 8.46; N, 2.44. A mixture of complex 4
(1.53 g, 2.73 mmol) and [iPr2Et NH]+Cl- (1.13 g, 6.84 mmol) were
heated to reflux in tetrahydrofuran (50 mL) for 24 h. The solvent was
removed under reduced pressure, and the residue was triturated with
toluene and filtered through Celite. Concentration, layering with
pentane, and cooling to -40 °C gave 1.38 g (84%) Cp*Ta[dNSi(tBu)3]-
Cl2 (5) as bright yellow crystals. Anal. Calcd for C22Cl2H42NSiTa: C,
43.96; H, 7.04; N, 2.33. Found: C, 44.59; H, 7.30; N, 2.46. 1H NMR
(22) The catalyst generated from complex 3 and [Ph3C]+[B(C6F5)4]-
in fluorobenzene at room temperature functions modestly, with an
activity of 38 000 g of polyethylene per mol of catalyst per hour. The
polymer obtained was completely soluble in 1,2,4-trichlorobenzene at
140 °C, giving an average molecular weight of 1.29 × 106 and
polydispersity (D) of 17.4 by GPC analysis.20 The broad molecular
weight distribution strongly suggests that this system does not produce
a single-site catalyst.
(23) Campbell, R. E., J r. Eur. Pat. Appl. EP 421659; Chem. Abstr.
1991, 115, 72440j. The blue crystalline reagent is prepared by
precipitation of the ferricinium cation from water on addition of LiB-
(C6F5)4, a procedure completely analogous to the preparation of the
tetraphenylborate salt.
(24) See, for example: (a) Schaller, C. P.; Bonanno, J . B.; Wolczanski,
P. T. J . Am. Chem. Soc. 1994, 116, 4133. (b) Cummins, C. C.;
Wolczanski, P. T. J . Am. Chem. Soc. 1988, 110, 8731.
t
(C6D6): δ 1.95 (s, 15H, Me5C5), 1.34 (s, 27H, Bu).
(25) Schrock, R. R.; Sharp, P. R. J . Am. Chem. Soc. 1978, 100, 2396.

2502 Organometallics, Vol. 16, No. 12, 1997
Communications
very different polymerization procedures preclude any
direct comparison. Why complex 5 is active under these
conditions and the other half-sandwich imido complexes
in this series are not is poorly understood, but it can be
tentatively attributed to the bulkiness of the tert-butyl
groups, which may prevent dimerization of the active
catalyst or inhibit MAO attack on the basic imido
nitrogen center.
For comparison, the bis(allyl) silylimido complex 6
was prepared by treatment of complex 5 with allylmag-
nesium bromide (2 equiv) in diethyl ether.20 The 1H-
NMR spectrum of 6 exhibits two signals in a 1:4 ratio
for the central and terminal protons, respectively, of the
allyl ligands, indicative of rapid equilibration between
η3- and η1-allyl coordination modes.18a An X-ray crystal
F igu r e 1. ORTEP diagram of Cp*Ta(dNSi(tert-butyl)3(η1-
C3H5)(η3-C3H5), 6 (most hydrogen atoms omitted for clar-
ity).20 Selected bond distances (Å): Ta-N, 1.799(5); N-Si,
1.735(5); Ta-C1, 2.307(7); Ta-C2, 2.400(7); Ta-C3, 2.492-
(7); Ta-C4, 2.265(7); C1-C2, 1.406(11); C2-C3, 1.381(11);
C4-C5, 1.470(11); C5-C6, 1.309(11). Selected bond angles
(deg): Ta-N-Si, 175.7(3); C1-C2-C3, 121.6(8); Ta-C1-
C2, 76.3(4); Ta-C3-C2, 70.0(4); C4-C5-C6, 128.2(9); Ta-
structure determination of complex 6 confirms that the
allyl ligands are η1- and η3-coordinated in the solid state
(Figure 1) and shows that the Ta-N-Si bond angle is
a nearly linear 175.7(3)°,20 similar to that observed in
related complexes.27
C4-C5, 119.8(5). Final residuals: R1 ) 0.0308, wR2
)
0.0738.
Surprisingly and in contrast to the closely analogous
complex 3, however, bis(allyl) complex 6 is not an eth-
ylene polymerization catalyst in the presence of [Ph3C]+-
[B(C6F5)4]- or B(C6F5)3. Dimethyl complex 4 is also
inactive in the presence of a variety of alkyl abstraction
reagents previously used to activate dimethylzirconocene
complexes toward olefin polymerization. These results
are, at present, most difficult to rationalize.
In summary, imido-based Ziegler-Natta catalysts
have been prepared through the use of ligands known
to stabilize highly reactive, coordinatively unsaturated,
Lewis acidic metal centers. Although some interesting
but as yet difficult to understand observations have been
made, these results demonstrate that the imido ligand
can provide an alternative to other dianionic cyclopen-
tadienyl analogues in the construction of group 5 olefin
polymerization catalysts. This work also demonstrates
that the allyl ligand is capable of supporting polymer-
ization activity in a catalyst system for which the
corresponding alkyl derivatives fail. Taken together,
these results reveal that very subtle changes in pre-
catalyst structure and activation can lead to dramatic
differences in the resulting catalyst structure and
activity. Further investigation is intended to define the
specific function of the allyl ligand and evaluate its
potential application to other catalyst systems.
Ack n ow led gm en t. We are particularly grateful to
Dr. Peter Hoang and NOVA Chemicals, Ltd., of Calgary,
Alberta, for the GPC analyses of polymer products. We
also thank Dr. Bob McDonald of the University of
Alberta Structure Determination Laboratory for the
X-ray crystallography. Financial support from the
Natural Sciences and Engineering Research Council of
Canada (NSERC) and the University of Alberta Depart-
ment of Chemistry is gratefully acknowledged.
(26) Polymerization procedure: To a Fischer-Porter bottle contain-
ing MAO (200 equiv) in toluene (5 mL) at room temperature under
ethylene pressure (65 psi) was added a solution of compound 5 (50.6
mg, 0.083 mmol) in 5 mL of toluene. Polymerization was stopped after
20 min by addition of 6N HCl-MeOH (1:1, 20 mL). The recovered
polyethylene was washed with MeOH and dried in vacuo to a constant
weight (1.686 g). The activity under these conditions was 61 000 g of
polyethylene per mol of catalyst per h. A significant portion of the
material obtained was insoluble in 1,2,4-trichlorobenzene at 140 °C,
reflecting the production of high molecular weight polymer. The soluble
portion was analyzed by GPC, giving an average molecular weight )
1.09 × 106 and D ) 3.6.
Su p p or tin g In for m a tion Ava ila ble: Text giving the
experimental, spectroscopic, and analytical data for complexes
3-6, the polymerization experiments, and the crystallographic
details for complex 6 and tables of the GPC experimental
conditions, calibration procedure, molecular weight in poly-
styrene equivalents, and crystallographic data, atomic coor-
dinates, bond distances, and bond angles for 6 (11 pages).
Ordering information is given on any current masthead
page.
(27) Nugent, W. A.; Mayer, J . M. Metal-Ligand Multiple Bonds;
Wiley Interscience: New York, 1988.
OM9703494