The mechanism of methane elimination in B(C6F5)3-initiated monocyclopentadienyl-ketimide titanium and related olefin polymerization catalysts

Article abstract of DOI:10.1021/ja994378c

A new class of monocyclopentadienyl titanium olefin polymerization catalysts and their activation with B(C₆F₅)₃ is reported herein. Dichlorides Cp[(τ)Bu(R)C=N]TiCl₂ {Cp = C₅H₅, R = (τ)Bu (1a); Cp = C₅Me₅, R = (τ)Bu (2a); Cp = C₅Me₄SiMe₃, R = (τ)Bu (3a); Cp = C₅Me₅, R = CH₂SiMe₃ (4a); Cp = C₅Me₅, R = Me (5a)} were prepared in 50-92% yield from CpTiCl₃ and (τ)Bu(R)C=NLi. Analogous dimethyl compounds 1b-5b were prepare via methylation of dichlorides a using MeMgBr in 89-92% yield. Dimethyl compound 6b (L = C₅Me₅, R = CH(SiMe₃)₂) was prepared directly from Cp*TiMe₃ and (τ)Bu[(Me₃Si)₂CH]C=NH in 40% yield. Dynamic ¹H NMR studies showed that the ketimide ligands in compounds b rotate rapidly about Ti-N on the NMR time scale, with a ΔG(+) of 9.6(6) kcal mol^-1 or less. The mixed alkyl compound Cp*-[(τ)Bu(R)C=N]Ti(CH₃)CH₂SiMe₃ {R = (τ)Bu (7)} was prepared via alkylation of the corresponding methyl chloride derivative with BrMgCH₂SiMe₃. When treated with B(C₆F₅)₃, compounds 1b-6b are rapidly converted into the ion pairs {Cp[(τ)Bu(R)C=N]TiCH₃}⁺[H₃C(B(C₆F₅)₃]^-, 1c-6c; mixed alkyl compound 7 yields the ion pair [Cp*((τ)Bu₂C=N)TiCH₂SiMe₃]⁺[H₃C(B(C₆F₅)₃]^-, 7c, exclusively. Multilinear NMR experiments show that ion pairing is tight in these compounds and that ketimide ligand rotation is occurring with a slightly higher barrier in comparison to the neutral derivatives b. Ion pairs 1c-5c undergo a decomposition processing involving loss of methane and producing the neutral compounds Cp[(τ)Bu(R)C=N]Ti(C₆F₅)[CH₂B(C₆F₅)₂], 1d-5d. The X-ray crystal structure of 1d has been determined. Active chronic compounds are not regenerated from neutral compounds d in the presence of B(C₆F₅)₃ and thus this reaction is a potential deactivation pathway for these particular ion pairs. Detailed kinetic studies on the decomposition of 2c show the reaction to be first order in [2c] with activation parameters of ΔH(+) = 20.6(8) kcal mol^-1 and ΔS(+) = -8.5(10) eu, corresponding to ΔG(+)₂₉₈ of 23.1(8) kcal mol^-1. A substantial kinetic isotope effect of k(H)/k(D) = 9.1(6) was measured using d₆-2c. Further mechanistic experiments, including crossover and examination of alkane elimination from mixed alkyl ion pair 7c, point to a σ-bond metathesis mechanism for the production of compounds d. The implications of our results for other, related catalyst systems are discussed.

Full text of DOI:10.1021/ja994378c

J. Am. Chem. Soc. 2000, 122, 5499-5509
5499
The Mechanism of Methane Elimination in B(C6F5)3-Initiated
Monocyclopentadienyl-Ketimide Titanium and Related Olefin
Polymerization Catalysts
Suobo Zhang,† Warren E. Piers,*,† Xiaoliang Gao,‡ and Masood Parvez†
Contribution from the Department of Chemistry, UniVersity of Calgary, 2500 UniVersity DriVe N.W.,
Calgary, Alberta T2N 1N4, Canada, and NoVa Chemicals Research and Technology Centre,
2928 16th Street N.E., Calgary, Alberta T2E 7K7, Canada
ReceiVed December 15, 1999
Abstract: A new class of monocyclopentadienyl titanium olefin polymerization catalysts and their activation
with B(C6F5)3 is reported herein. Dichlorides Cp[tBu(R)CdN]TiCl2 {Cp ) C5H5, R ) tBu (1a); Cp ) C5Me5,
t
t
R ) Bu (2a); Cp ) C5Me4SiMe3, R ) Bu (3a); Cp ) C5Me5, R ) CH2SiMe3 (4a); Cp ) C5Me5, R ) Me
t
(5a)} were prepared in 50-92% yield from CpTiCl3 and Bu(R)CdNLi. Analogous dimethyl compounds
1b-5b were prepare via methylation of dichlorides a using MeMgBr in 89-92% yield. Dimethyl compound
t
6b (L ) C5Me5, R ) CH(SiMe3)2) was prepared directly from Cp*TiMe3 and Bu[(Me3Si)2CH]CdNH in
1
40% yield. Dynamic H NMR studies showed that the ketimide ligands in compounds b rotate rapidly about
Ti-N on the NMR time scale, with a ∆G‡ of 9.6(6) kcal mol-1 or less. The mixed alkyl compound Cp*-
[tBu(R)CdN]Ti(CH3)CH2SiMe3 {R ) tBu (7)} was prepared via alkylation of the corresponding methyl chloride
derivative with BrMgCH2SiMe3. When treated with B(C6F5)3, compounds 1b-6b are rapidly converted into
the ion pairs {Cp[tBu(R)CdN]TiCH3}+[H3C(B(C6F5)3]-, 1c-6c; mixed alkyl compound 7 yields the ion pair
[Cp*(tBu2CdN)TiCH2SiMe3]+[H3C(B(C6F5)3]-, 7c, exclusively. Multinuclear NMR experiments show that
ion pairing is tight in these compounds and that ketimide ligand rotation is occurring with a slightly higher
barrier in comparison to the neutral derivatives b. Ion pairs 1c-5c undergo a decomposition process involving
loss of methane and producing the neutral compounds Cp[tBu(R)CdN]Ti(C6F5)[CH2B(C6F5)2], 1d-5d. The
X-ray crystal structure of 1d has been determined. Active cationic compounds are not regenerated from neutral
compounds d in the presence of B(C6F5)3 and thus this reaction is a potential deactivation pathway for these
particular ion pairs. Detailed kinetic studies on the decomposition of 2c show the reaction to be first order in
[2c] with activation parameters of ∆H‡ ) 20.6(8) kcal mol-1 and ∆S‡ ) -8.5(10) eu, corresponding to ∆G‡
298
of 23.1(8) kcal mol-1. A substantial kinetic isotope effect of kH/kD ) 9.1(6) was measured using d6-2c. Further
mechanistic experiments, including crossover and examination of alkane elimination from mixed alkyl ion
pair 7c, point to a σ-bond metathesis mechanism for the production of compounds d. The implications of our
results for other, related catalyst systems are discussed.
Introduction
catalysts3 and later employed by Dow4 and Exxon5 to support
more commercially viable titanium derivatives (I, Chart 1).
Continuing in this vein, McConville replaced both Cp ligands
with amido donors, held in a chelating array and substituted
with bulky aryl substitutents (II);6 the latter feature is a key
element in the observed behavior of compounds II as “living”
polymerization catalysts.7,8 More recently, other examples of
Bis-cyclopentadienyl (bis-Cp) complexes of the early transi-
tion metals are highly effective homogeneous olefin polymer-
ization catalysts and the chemistry involved in their operation
is relatively well understood.1 Athough much creativity has been
demonstrated in imbueing this family of catalysts with amazing
versatility, extensive patent coverage has spurred the develop-
ment of catalysts that do not contain the bis-Cp ligand
framework.2 One strategy has been to replace one or both of
the Cp ligands in metallocenes with other donors. A notable
example of this approach is the Cp-amido “constrained geom-
etry” ligand first introduced by Bercaw for scandium-based
(2) (a) Britovsek, G. P.; Gibson, V. C.; Wass, D. F. Angew. Chem., Int.
Ed. 1999, 38, 429. (b) Liang, L.-C.; Schrock, R. R.; Davis, W. M.;
McConville, D. H. J. Am. Chem. Soc. 1999, 121, 5797. (c) Baumann, R.;
Stumpf, R.; Davis, W. M.; Liang, L.-C.; Schrock, R. R. J. Am. Chem. Soc.
1999, 121, 7822.
(3) (a) Piers, W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett
1990, 1, 74. (b) Shapiro, P. J.; Cotter, W. D.; Schaefer, W. P.; Labinger, J.
A.; Bercaw, J. E. J. Am. Chem. Soc. 1994, 116, 4623.
(4) Stevens, J. C.; Timmers, F. J.; Wilson, D. R.; Schmidt, G. F.; Nickias,
P. N.; Rosen, R. K.; Knight, G. W.; Lai, S. European Patent Appl. EP 416
815-A2, 1991 (Dow Chemical Co.).
(5) Canich, J. M.; Hlatky, G. G.; Turner, H. W. PCT Appl. WO 92-
00333, 1992 (Exxon Chemical Co.).
(6) Scollard, J. D.; McConville, D. H.; Vittal, J. J. Macromolecules 1996,
29, 5241.
(7) (a) Scollard, J. D.; McConville, D. H. J. Am. Chem. Soc. 1996, 118,
10008. (b) Scollard, J. D.; McConville, D. H.; Vittal, J. J.; Payne, N. C. J.
Mol. Catal. A: Chem. 1998, 128, 201.
† University of Calgary.
‡ Nova Chemicals Research and Technology Centre.
(1) (a) Mohring, P. C.; Coville, N. J. J. Organomet. Chem. 1994, 479,
1. (b) Brintzinger, H. H.; Fischer, D.; Mulhaupt, R.; Rieger, B.; Waymouth,
R. M. Angew. Chem., Int. Ed. Engl. 1995, 34, 1143. (c) Bochmann, M. J.
Chem. Soc., Dalton Trans. 1996, 255. (d) Marks, T. J. Acc. Chem. Res.
1992, 25, 57. (e) Jordan, R. F. AdV. Organomet. Chem. 1991, 32, 325. (f)
Jordan, R. F.; Bradley, P. K.; LaPointe, R. E.; Taylor, D. F. New J. Chem.
1990, 14, 505. (g) Ziegler Catalysts; Fink, G., Mu¨lhaupt, R., Brintzinger,
H. H., Eds.; Springer: Berlin, 1995. (h) Resconi, L.; Cavallo, L.; Fait, A.;
Piemontesi, F. Chem. ReV. 2000, 100, 1253.
10.1021/ja994378c CCC: $19.00 © 2000 American Chemical Society
Published on Web 05/27/2000

5500 J. Am. Chem. Soc., Vol. 122, No. 23, 2000
Zhang et al.
Chart 1
Scheme 1
Cp(L)TiX2 compounds [L ) OR9 (e.g., III), alkyl,10 NR2,11
NPR3,12 and SR13] have been reported along with their behavior
in the presence of common olefin polymerization activators. 14
Generally, in conjunction with cocatalysts such as methyl-
aluminoxane (MAO) or [Ph3C]+[B(C6F5)4]-,15 the aforemen-
tioned compounds provide catalysts with high activity, stability,
and comonomer incorporation rates, making them viable
alternatives to metallocene-based catalyst systems. On the other
hand, although relatively stable ion pairs are observable in
solution when dimethyl derivatives of these catalysts are treated
with B(C6F5)3,16 they are subject to a deactivation pathway
involving methane loss and conversion of the cationic titanium
complex to inactive neutral compounds (eq 1). Interestingly, a
complexes with ketimide ligands are known,19 only one example
of a monocyclopentadienyl ketimide complex of titanium has
been reported,20 and it has not to our knowledge been examined
within the context of olefin polymerization. Like the examples
I-III discussed above, bis-alkyl derivatives of this family of
compounds constitute an extremely promising group of catalysts
for olefin polymerization under industrially relevant conditions.21
However, when ion pairs are generated with B(C6F5)3, methane
elimination via the process shown in eq 1 is operative. Given
the apparent generality of this ion pair decomposition, we have
used the Cp-ketimides described herein to perform a detailed
mechanistic study on this deactivation process.
similar phenomenon is observed in the constrained geometry
systems I.17 The stoichiometry of this process is relatively well
documented, particularly for II and the aryloxide system Cp-
(OAr)TiMe2 (III), where the LnTi(C6F5)[CH2B(C6F5)2] products
have been crystallographically characterized.9b,18 However, the
intimate mechanism of this deactivation process is poorly
understood.
Results and Discussion
Synthesis. A variety of titanium dichlorides supported by one
Cp donor and one ketimide ligand may be prepared from mono-
Cp trichlorides and the lithium salt of the ketimide (isolated22
t
or generated in situ via reaction of RLi with BuCN) using a
Herein we introduce a new family of Cp(L)TiX2 catalysts,
where L is NdCRR′, a ketimide ligand. Although titanocene
procedure analogous to that originally reported by Leigh.20 The
range of compounds prepared is shown in Scheme 1 and eq 2.
(8) Deng, L.; Ziegler, T.; Woo, T. K.; Margl, P.; Fan, L. Organometallics
1998, 17, 3240.
(9) (a) Vilardo, J. S.; Thorn, M. G.; Fanwick, P. E.; Rothwell, I. P. Chem.
Commun. 1998, 2425. (b) Thorn, M. G.; Vilardo, J. S.; Fanwick, P. E.;
Rothwell, I. P. Chem. Commun. 1998, 2427. (c) Sarsfield, M. J.; Ewart, S.
W.; Tremblay, T. L.; Roszak, A. W.; Baird, M. C. J. Chem. Soc., Dalton
Trans. 1997, 3097. (d) Ewart, S. W.; Sarsfield, M. J.; Jeremic, D.; Tremblay,
T. L.; Williams, E. F.; Baird, M. C. Organometallics 1998, 17, 1502. (e)
For a related chelating system see: Chen, Y.-X.; Fu, P.-F.; Stern, C. L.;
Marks, T. J. Organometallics 1997, 16, 5958.
(10) (a) Mena, M.; Royo, P.; Serrano, R.; Pellinghelli, A.; Tiripicchio,
A. Organometallics 1989, 8, 476. (b) Ewart, S. W.; Baird, M. C. Top. Catal.
1999, 7, 1.
(11) (a) Bai, Y.; Roesky, H. W.; Noltmeyer, M. Z. Anorg. Allg. Chem.
1991, 31, 3887. (b) Schiffino, R. S.; Crowther, D. J. U.S. patent No.
5,625,016, 1997 (Exxon Chemical).
(12) Stephan, D. W.; Stewart, J. C.; Guerin, F.; Spence, R. E. v. H.; Xu,
W.; Harrison, D. G. Organometallics 1999, 18, 1116.
(13) Klapo¨tke, T.; Laskowski, R.; Ko¨pf, H. Z. Naturforsch., Teil B 1987,
42, 777.
(14) Related examples: (a) Richter, J.; Edelmann, F. T.; Noltmeyer, M.;
Schmidt, H.-G.; Shmulinson, M.; Eisen, M. S. J. Mol. Catal. A: Chem.
1998, 130, 149. (b) Yoon, S. C.; Bae, B.-J.; Suh, I.-H.; Park, J. T.
Organometallics 1999, 18, 2049. (c) Amor, F.; Butt, A.; du Plooy, K. E.;
Spaniol, T. P.; Okuda, J. Organometallics 1998, 17, 5836.
(15) (a) Bochmann, M.; Lancaster, S. J. J. Organomet. Chem. 1992, 434,
C1. (b) Chien, J. C. W.; Tsai, W. M.; Rausch, M. D. J. Am. Chem. Soc.
1991, 113, 8570.
(16) (a) Massey, A. G.; Park, A. J. J. Organomet. Chem. 1966, 5, 218.
(b) Piers, W. E.; Chivers, T. Chem. Soc. ReV. 1997, 345.
(17) Allusions to the thermal instability of ion pairs generated from
(CGC)TiMe2 and B(C6F5)3 can be found in the literature, but no details on
the process have been mentioned to our knowledge. We found that this ion
pair indeed decomposes as indicated in eq 1, producing (CGC)Ti(C6F5)-
[CH2B(C6F5)2] (see Experimental Section).
Conversion of the dichloride series (compounds a) to the
analogous dimethyl derivatives (series b) is accomplished via
treatment with 2 equiv of MeMgBr; organolithium reagents lead
to lower yields and less clean reactions due to competitive
reduction at titanium. A further member of the dimethyl series
of compounds, containing the bulky bis-trimethylsilylmethyl-
substituted ketimide (6b), is best generated using a methane
elimination protocol, from Cp*TiMe3 (Cp* ) C5Me5) and the
free ketimine (eq 2).
We required access to mixed alkyl complexes of the Cp-
ketimide ligand set for some of the mechanistic experiments
described below and found that such compounds, exemplified
(19) Bochmann, M.; Wilson, L. M.; Hursthouse, M. B.; Motevalli, M.
Organometallics 1988, 7, 1148.
(20) Latham, I. A.; Leigh, G. J.; Huttner, G.; Jibril, I. J. Chem. Soc.,
Dalton Trans. 1986, 377.
(21) McMeeking, J.; Gao, X.; Spence, R. E. v. H.; Brown, S. J.; Jeremic,
D. PCT WO 99/14250, 1999 (Nova Chemicals Ltd.).
(22) Armstrong, D. R.; Barr, D.; Snaith, R.; Mulvey, R. E.; Wade, K.;
Reed, D. J. Chem. Soc., Dalton Trans. 1987, 1071.
(18) Scollard, J. D.; McConville, D. H.; Rettig, S. J. Organometallics
1997, 16, 1810.

Monocyclopentadienyl Ti Olefin Polymerization Catalysts
J. Am. Chem. Soc., Vol. 122, No. 23, 2000 5501
Scheme 2
equivalent at all accessible temperatures, suggesting a perpen-
dicular orientation of the ketimide ligand (with respect to the
Cp-Ti vector) in the ground state as observed in the solid state
for Cp(nButBuCdN)TiCl2. With this orientation, dimethyl
compounds 4b, and 5b, containing unsymmetrical ketimide
ligands, should have diastereotopic methyl groups. However,
the methyl groups appear as a single resonance down to -80
°C in the 1H NMR spectrum. The most reasonable explanation
for this behavior is rapid rotation of the ketimide ligand about
the Ti-N bond, which exchanges the titanium methyl environ-
ments. For the more sterically bulky 6b, this process is
observable spectroscopically. The signal for the Ti-Me groups
undergoes coalescence behavior and two signals at 0.71 and
0.49 ppm emerge when the sample is cooled to -90 °C.
Analysis of the spectra yields a ∆G‡ of 9.6(6) kcal mol-1 for
ketimide rotation in this compound.26 As the size of R decreases,
this free energy barrier also is lowered and the low-temperature
limits are not accessible for 4b or 5b. The low barrier to rotation
of the ketimide ligand may be due to participation of the CdN
π orbitals in the process, such that the π component to the Ti-N
bond is not lost in the parallel orientation of the ligand,27 the
other limiting configuration during rotation.
by Cp*(tBu2CdN)Ti(CH2SiMe3)CH3, 7, can be prepared cleanly
using the method shown in Scheme 2. Typically, mixed alkyl
titanocenes are prepared from Cp2Ti(R)Cl type precursors, which
themselves are prepared in rather low-yielding monoalkylation
reactions starting from Cp2TiCl2;23 alternatively, oxidative
alkylation of Cp2TiCl derivatives with CdR2 has been used to
access these precursors.24 We discovered that B(C6F5)3 catalyzes
the disproportionation of a given a/b pair of compounds (in
this case, 2a/b), providing an extremely convenient route to the
required methyl chloride derivative, which can be subsequently
alkylated with a second hydrocarbyl group. The disproportion-
ation presumably occurs via B(C6F5)3 abstraction of a methyl
group from 2b, followed by chloride transfer from 2a to the
titanium methyl cation; collapse of the putative [Cp*(tBu2CdN)-
TiCl]+[MeB(C6F5)3]- ion pair furnishes the other equivalent of
Cp*(tBu2CdN)TiMeCl and releases B(C6F5)3. Alkylation of
Cp*(tBu2CdN)TiMeCl is accomplished in the usual way using
a Grignard reagent. This procedure also works for C5H5
compounds 1. The process in Scheme 2 represents a very clean,
high-yielding and possibly general method for preparing LnTi-
(R)Cl compounds, although we have not explored its utility in
other systems (i.e., differing Ln or R) in detail.
Reactions with B(C6F5)3. When activated with B(C6F5)3,
dimethyl compounds 1b-6b are highly active ethylene poly-
merization catalysts at room temperature.28 We thus examined
in detail the reactions of these dimethyl compounds with
B(C6F5)3 to probe the stability of the resulting ion pairs in
solution. Rapid formation of ion pairs 1c-6c is observed upon
admixture of the dimethyl precursor with 1 equiv of B(C6F5)3
in toluene (eq 3). Compounds c are relatively stable at room
The X-ray crystal structure of the related compound Cp-
(nButBuCdN)TiCl2 was reported by Leigh et al. in 1986,20 and
showed the complex to be monomeric with a near-linear Ti-
N-C linkage of 171.3(4)°. In the solid state, the plane of the
temperature under these conditions, slowly decomposing over
the course of ≈24 h as described below. However, they are
persistent enough to fully characterize spectroscopically, and
2c is also isolable as an analytically pure solid.
ketimide ligand is approximately perpendicular to the Cpcentroid
-
Spectroscopic data for the ion pairs 1c-6c suggest that ion-
ion contact is relatively tight in these systems. The positions of
Ti vector. The authors noted that the relatively short Ti-N bond
length of 1.872(4) Å was consistent with some double bond
character in this linkage. For comparison, the Ti-N bond length
in constrained geometry complex I is 1.909 Å,25 indicating that
the π bond between the ketimide ligand and titanium is
somewhat stronger than that in I. On the other hand, in the
unconstrained analogue Cp[N(SiMe3)2]TiCl2, which is perhaps
more closely related to the ketimide compounds, the Ti-N
distance is 1.879(3) Å.11a
1
the H NMR signals for the B-CH3 in 1c-6c are upfield of
the range of ≈1.2-1.4 ppm generally noted for the unassociated
[MeB(C6F5)3]- anion in aromatic solvents.29 Furthermore, the
position of this resonance spans a relatively large range (0.21
to 0.81 ppm) indicating a high degree of sensitivity to the
variations in ancillary ligation. These spectroscopic features are
typical of contact ion pairing in which a Ti-(µ-CH3)-B contact
is present.12,30
All of the neutral titanium ketimide compounds described
above were characterized by NMR spectroscopy and elemental
analysis. The 1H NMR spectroscopic data for the ketimide
complexes 1a-5a and 1b-6b suggest that similar structural
features to those found for Cp(nButBuCdN)TiCl2 obtain in
solution but that rotation about the Ti-N bond of the ketimide
ligand is rapid on the NMR time scale. For Cs symmetric
compounds 1-3, the tBu groups of the ketimide ligand remain
The lack of certain types of temperature dependence in the
1H NMR spectra of ion pairs c, and particularly 3c, indicates
that the associated structure for these ion pairs is rather static
in toluene solution in comparison to other systems. Marks and
(26) Sandstrom, J. Dynamic NMR Spectroscopy; Academic Press: New
York, 1982.
(27) A similar low barrier to rotation in compounds Cp(NHtBu)TiCl2
has been rationalized in this fashion: Giolando, G. M.; Kirschbaum, K.;
Graves, L. J.; Bolle, U. Inorg. Chem. 1992, 31, 3887.
(28) Gao, X.; Wang, Q. Unpublished results.
(23) Waters, J. A.; Mortimer, G. A. J. Organomet. Chem. 1970, 22, 417.
(24) Luinstra, G. A.; Teuben, J. H. J. Organomet. Chem. 1991, 420,
337.
(25) Stevens, J. C. In Catalyst Design for Tailor Made Polyolefins,
Studies in Surface Science and Catalysis; Soga, K., Terano, M., Eds;
Elsevier: Amsterdam, 1994; Vol. 89, p 277.
(29) (a) Horton, A. D. Organometallics 1996, 15, 2675. (b) Beck, S.;
Prosenc, M. H.; Brintzinger, H. H. J. Mol. Catal. A: Chem. 1998, 128, 41.
(c) Beswick, C. L.; Marks, T. J. Organometallics 1999, 18, 2410.
(30) Yang, X.; Stern, C. L.; Marks, T. J. J. Am. Chem. Soc. 1994, 116,
10015.

5502 J. Am. Chem. Soc., Vol. 122, No. 23, 2000
Zhang et al.
Scheme 3
Me and B-Me groups also remain distinct at all temperatures
examined, indicating that the borane dissociation/reabstraction
process is also quite slow in these systems. The data indicate
that both exchange processes are slow in the Cp-ketimide ion
pairs relative to metallocene-based ion pairs and must occur
with a higher ∆G‡ than that of the decomposition process
discussed below. The only changes in these spectra are
associated with the resonance for the ketimide tert-butyl group,
which coalesce and reappear as two resonances in the low-
temperature limit where ketimide rotation is slow. This process
is also observable in the ion pair 2c, and free energies of
activation of 11.6(4) and 11.8(9) kcal mol-1 are measured for
2c and 3c, respectively. Although ion pair dissociation/
t
reorganization would also exchange the diastereotopic Bu
groups, since the Cp-methyl groups in 3c remain diastereotopic
t
during Bu coalescence, this latter phenomenon must be a
manifestation of arrested ketimide ligand rotation. The ∆G‡
values for this process in the ion pairs are slightly higher than
that observed in the neutral dimethyl compound 6b (vide supra),
consistent with a higher degree of π-bonding between N and
the more electrophilic titanium centers in the cationic com-
pounds.32
As has been mentioned, ion pairs 1c-5c decompose at room
temperature in toluene solution over the course of several hours
to eliminate methane and produce the neutral complexes Cp-
(tBuRCdN)Ti(C6F5)[CH2B(C6F5)2], 1d-5d (eq 4).9b,18 Ion pair
6c is also susceptible to loss of methane, but the product in this
case is formulated as the metalated species 8 (eq 5); this ion
Figure 1. 1H NMR spectra (400 MHz) of 3c taken at various
temperatures (d8-toluene).
co-workers have shown that ion pairs generated from metal-
locene or constrained geometry catalyst precursors and perfluo-
roaryl borane activators exhibit complex behavior associated
with two distinct dynamic processes, namely ion pair dissocia-
tion/reorganization and borane dissociation/reabstraction. These
two processes as they would be occurring in 3c are illustrated
in Scheme 3. The two dynamic processes can be studied
independently by dynamic NMR techniques, since the former
process exchanges diastereotopic Cp substitutents (A/A′ and
B/B′), while the latter also permutes the two methyl groups (CH3
and C*H3).31
pair is relatively stable toward further reaction. Compounds 1d
and 2d were fully characterized, including an X-ray crystal-
lographic analysis of 1d; 3d-5d were characterized in situ via
multinuclear spectroscopy. Because of the unsymmetrical liga-
tion at titanium, groups with the same connectivity in these
molecules are diastereotopic, although again ketimide ligand
1
Figure 1 shows a series of H NMR spectra for ion pair 3c
t
at various temperatures. As can be seen, separate resonances
for the four Cp methyl groups are observed at all temperatures.
There is some temperature dependence on the chemical shifts
of these resonances, but no hint of the onset of exchange
between the pairs of diastereotopic Cp methyl groups is observed
as the temperature of the samples is raised, although it must be
noted that the decomposition accelerates significantly at higher
temperatures (>60 °C). Furthermore, the signals due to the Ti-
rotation is rapid on the NMR time scale and exchanges Bu
groups in 1d-3d. The methylene protons of the CH2B(C6F5)2
ligand, however, appear as separate resonances; because of their
low intensity and broadness due to coupling with boron, HMQC
experiments33 were required to locate their resonances.
The molecular structure of 1d is shown in Figure 2, along
with selected metrical data; full details are available as Sup-
(32) These observations were made in toluene; in the more polar solvents
C6D5Br and CD2Cl2, the Cp-methyl peaks are close to the fast exchange
limit at room temperature, indicating that, as expected, ion pair reorganiza-
tion is faster in these more polar solvents. Quantitative evaluation of the
temperature-dependent spectra was precluded by the onset of decomposition
at the higher temperatures necessary to reach the fast exchange limit.
(33) Bax, A.; Subramanian, S. J. Magn. Reson.1986, 67, 565.
(31) (a) Deck, P. A.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 6128.
(b) Siedle, A. R.; Newmark, R. A. J. Organomet. Chem. 1995, 497, 119.
(c) Chen, Y. X.; Metz, M. V.; Li, L.; Stern, C. L.; Marks, T. J. J. Am.
Chem. Soc. 1998, 120, 6287. (d) Deck, P. A.; Beswick, C. L.; Marks, T. J.
J. Am. Chem. Soc. 1998, 120, 1772. (e) Luo, L.; Marks, T. J. Top. Catal.
1999, 7, 97.

Monocyclopentadienyl Ti Olefin Polymerization Catalysts
J. Am. Chem. Soc., Vol. 122, No. 23, 2000 5503
and less crowded ligand environment allows for side-on
approach of this ligand. In the present titanium compounds, the
boron ylide ligand is sterically prevented from engaging in such
π-interaction with the metal.
Mechanistic Studies. Compounds d are unreactive toward
B(C6F5)3 and are thus not reactivated in the presence of excess
B(C6F5)3, so the reaction of eq 4 represents a fatal deactivation
pathway for these catalysts. This seems to be a relatively general
fate for catalysts of this type when activated by B(C6F5)3. As
mentioned, both Rothwell9b and McConville18 type catalysts
exhibit this chemistry and the ion pair formed from the
constrained geometry catalyst (C5Me4SiMe2NtBu)TiMe2 and
B(C6F5)3 also decomposes in this manner in the absence of olefin
when heated to 60 °C for 1 h.
Figure 2. Molecular structure of 1d. Selected bond distances (Å):
Ti(1)-N(1), 1.830(4); N(1)-C(7), 1.264(5); C(7)-C(8), 1.556(6);
C(7)-C(12), 1.543(6); Ti(1)-C(16), 2.263(2); Ti(1)-C(6), 2.182(5);
C(6)-B(1), 1.479(6); C(22)-B(1), 1.610(4); C(28)-B(1), 1.6625;
Cpcentrod-Ti(1), 2.033(5). Selected bond angles (deg): N(1)-Ti(1)-
C(6), 100.02(17); N(1)-Ti(1)-C(16), 105.19(13); C(6)-Ti(1)-C(16),
Of the possible mechanisms for this process which may be
envisioned, bimolecular pathways identified for metallocene
systems36 may be excluded on the basis of the deuterium
labeling crossover experiment shown in eq 6. When a 1:1
97.63(14); N(1)-Ti(1)-Cpcentroid, 110.2(18); C(6)-Ti(1)-Cpcentroid
,
122.7(18); C(16)-Ti(1)-Cpcentroid, 118.2(17); Ti(1)-N(1)-C(7), 173.5(3);
N(1)-C(7)-C(8), 116.2(4); N(1)-C(7)-C(12), 118.7(4); C(8)-C(7)-
C(12), 125.0(4); Ti(1)-C(6)-B(1), 110.8(3); C(6)-B(1)-C(22),
118.5(2); C(6)-B(1)-C(28), 121.7(2); C(22)-B(1)-C(28), 119.82(18).
Table 1. Metrical Parameters for 1d and Related Compoundsa
compound
parameter
IIb
IIIc
1d
Ti-C6F5
Ti-CH2
H2C-B
2.191(4)
2.111(4)
1.503(6)
121.5(2)
125.1(3)
360.0
2.176(2)
2.115(2)
2.263(2)
2.182(5)
1.479(6)
97.63(14)
110.8(3)
360.0
mixture of 2b and d6-2b (selectively labeled in the methyl
groups) is treated with a stoichiometric quantity of B(C6F5)3,
only ion pairs 2c and d6-2c are produced; no scrambling of the
methyl groups to produce isotopomers of d6-2c is observed. This
indicates that, unlike dichloride 2a, the dimethyl analogue is
not basic enough to displace [MeB(C6F5)3] from 2c (cf. Scheme
2) and is consistent with the observed lack of ion pair
reorganization in toluene solution for these systems. When this
mixture of 2c/d6-2c is allowed to decompose, CH4 and CD4
are the only methane isotopomers detected by 1H and 2H NMR
spectroscopy, strongly implying an intramolecular pathway for
methane elimination.
C-Ti-C
B-C-Ti
98.73(8)
∑
C-B-C
a Bond distances in Å; bond angles in deg. b Reference 15. c Ref-
erence 9b.
porting Information. The ketimide ligand is structurally similar
to that found in Cp(tBunBuCdN)TiCl2,20 with a near linear
C(7)-N(1)-Ti(1) angle of 173.5(5)° and a Ti(1)-N(1) distance
of 1.830(4) Å. For comparison, key data for 1d and crystallo-
graphically characterized examples of this type of ion pair
decomposition product for catalyst precursors II and III (Chart
1) are given in Table 1. The titanium center is tetrahedrally
coordinated in this chiral “piano stool” molecule and the metrical
parameters associated with the geometry about titanium are
similar to those found in Rothwell’s related compound (Table
1). The -CH2B(C6F5)2 ligand is essentially a σ-alkyl donor to
titanium in all of these complexes as indicated by the normal
Ti-CH2 distances of 2.182(5) (1d), 2.111(4) (II), and 2.115(2)
Å (III) (cf. the value of 2.170(2) Å in Cp2TiMe234), values of
Ti-C-B which are near the tetrahedral ideal, and trigonal planar
boron centers. We have noted recently, however, that this ligand
may be viewed as being a boron ylide (i.e. [CH2dB(C6F5)2]-),
which can also interact with metals in a side-on binding mode
reminiscent of alkene bonding. This type of bonding is in fact
observed in the tantalum complex IV,35 where the larger metal
The unimolecularity of methane elimination is also indicated
by kinetic studies carried out by monitoring the reaction using
1H NMR spectroscopy in d8-toluene (ꢀ ) 2.37). No intermedi-
ates are observed in the conversion of 2c to 2d and the reaction
is cleanly first order in the concentration of the ion pair over
several half-lives with a rate constant of 4.5(3) × 10-5 s-1 at
room temperature. A typical first-order plot is given in Figure
3, along with the analogous data collected for the conversion
of d6-2c to d2-2d. A substantial kinetic isotope effect of 9.1(6)
is observed in this system. The decomposition of 2c was also
followed at several different temperatures, and from the resulting
Eyring plot (Figure 4), activation parameters of ∆H‡ ) 20.6(8)
kcal mol-1 and ∆S‡ ) -8.5(10) eu are obtained. The rate of
the reaction is accelerated somewhat when carried out in a more
(34) Thewalt, U.; Worthle, T. J. Organomet. Chem. 1994, 464, C17.
(35) Cook, K. S.; Piers, W. E.; Rettig, S. J. Organometallics 1999, 18,
1575.
(36) Bochmann, M.; Cuenca, T.; Hardy, D. T. J. Organomet. Chem. 1994,
484, C10.

5504 J. Am. Chem. Soc., Vol. 122, No. 23, 2000
Zhang et al.
tion of an ion pair derived from a mixed alkyl precursor, since
both ion pair reorganization and borane dissociation/reabstrac-
tion processes are slow in these systems. It was for this purpose
that mixed alkyl derivative 7 was prepared; as shown at the top
of Scheme 4, reaction of 7 with B(C6F5)3 results exclusively in
abstraction of the less sterically hindered methyl group29c,38 to
give ion pair 7c.39 The spectroscopic data for 7c suggest that it
also has a static contact ion pair structure.
This selectivity is crucial for distinguishing between the
mechanistic possibilities put forward in Scheme 4. The first
(path a) involves elimination of alkane from the neutral dialkyl
compound and necessitates borane dissociation prior to loss of
RH. This is a possibility since the diamido complex [(C6H11)2N]2-
TiMe2 has been observed to decompose with loss of methane
to form the dimeric µ-methylidene derivative {[(C6H11)2N]2Ti-
(µ-CH2)}2.40 For ion pair 7c, this path would be expected to
yield a mixture of alkanes and alkylidene products; electrophilic
attack by B(C6F5)3 on the alkylidene ligands would lead to
products 7d and 9.41 The second possible pathway (path b) is
reminiscent of the alkane elimination process thought to be
operative in the formation of Tebbe’s reagent.42 This path would
require ion pair dissociation to attain the concerted six-
membered transition state via which RH loss and -C6F5 transfer
to titanium occurs in a concerted fashion. Methane and titanium
complex 9 would be the only products expected from this
pathway.
Figure 3. Typical first-order plots for the thermal decomposition of
ion pairs c, using 2c and d6-2c as examples.
Since we have already established that both borane dissocia-
tion and ion pair reorganization are slow in these systems
(relative to the decomposition process), a priori neither of these
pathways seem likely. A third option, path c, involves a σ-bond
metathesis reaction43 between Ti-CH3 and H-CH2B(C6F5)3.
From ion pair 7c, this path would yield only SiMe4 and 2d as
products. As Figure 5 shows, this is the product mixture which
results when 7c is allowed to decompose at room temperature.
While the rate of decomposition for this ion pair is qualitatively
much slower than that observed for the related ion pair 2c, the
product mixture clearly consists of 2d and SiMe4; within the
detection limits of 1H NMR spectroscopy, no methane is
produced, nor are any signals consistent with the generation of
9 observed. This experiment provides strong support for the
σ-bond metathetical mechanistic option of path c. While σ-bond
metathesis reactions involving the Zr-C bonds of zirconium-
based cations are relatively common,1e,44 to our knowledge, such
reactions involving the Ti-C bonds for cationic titanium alkyls
and C-H bonds are less well defined.
Figure 4. Eyring plot for the thermal decomposition of ion pair 2c at
various temperatures.
Table 2. First-Order Rate Constants for the Decomposition of Ion
Pairs 1-5c
ion pair
T (K)
solvent
10-4k (s-1
)
1c
2c
3c
4c
5c
d6-2c
2c
2c
2c
2c
2c
2c
2c
298
298
298
298
298
298
302
310
317
325
333
340
298
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C7D8
C6D5Br
1.49(5)
0.45(5)
0.89(5)
0.25(5)
0.18(5)
0.04(5)
0.99(5)
2.66(5)
3.91(5)
11.05(5)
23.87(5)
51.06(5)
1.16(5)
In this picture of the reaction, σ-bond metathesis occurs from
a tight contact ion pair prior to full dissociation of the
[MeB(C6F5)3]- anion from the titanium cation via a typical
(38) Temme, B.; Erker, G. J. Organomet. Chem. 1995, 488, 177.
(39) Use of the less hindered C5H5 analogue of 7, i.e., C5H5(L)Ti(CH3)CH2-
SiMe3, resulted in competitive trimethylsilylmethyl abstraction (≈30%).
(40) Scoles, L.; Minhas, R.; Duchateau, R.; Jubb, J.; Gambarotta, S.
Organometallics 1994, 13, 4978.
(41) While the order in [Ti] is unknown for the loss of methane from
[(C6H11)2N]2TiMe2, the conclusions here would be the same for a bimo-
lecular elimination of RH.
(42) (a) Tebbe, F. N.; Parshall, G. W.; Reddy, G. S. J. Am. Chem. Soc.
1978, 100, 3611. (b) Tebbe, F. N.; Harlow, R. L. J. Am. Chem. Soc. 1980,
102, 6149. (c) Ott, K. C.; deBoer, E. J. M.; Grubbs, R. H. Organometallics
1984, 3, 223.
(43) Thompson, M. E.; Baxter, S. M.; Bulls, A. R.; Burger, B. J.; Nolan,
M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw, J. E. J. Am. Chem.
Soc. 1987, 109, 203 and references therein.
(44) (a) Jordan, R. F.; Taylor, D. F.; Baenziger, N. C. Organometallics
1990, 9, 1546. (b) Jordan, R. F.; LaPointe, R. E.; Bradley, P. K.; Baenziger,
N. C. Organometallics 1989, 8, 2892. (c) Siedle, A. R.; Lamanna, W. M.;
Newmark, R. A.; Schroepfer, J. N. J. Mol. Catal. A: Chem. 1998, 128,
257. (d) Eshuis, J. J. W.; Tan, Y. Y.; Meetsma, A.; Teuben, J. H.; Renkema,
J.; Evens, G. G. Organometallics 1992, 11, 362.
polar solvent (kobs ) 11.1(5) × 10-5 s-1 in C6D5Br, ꢀ ) 5.40).37
For comparison, first-order rate constants for the decomposition
of compounds 1c and 3c-5c were also obtained; all of the rate
constants measured are collected in Table 2.
Three reasonable unimolecular pathways for methane elimi-
nation are shown in Scheme 4. In principle, these may be
distinguished by analysis of the products formed via decomposi-
(37) Attempts to measure the rate of this reaction in CD2Cl2 (ꢀ ) 9.08)
were hampered by side reactions which we believe involved chloride
abstraction from the solvent.

Monocyclopentadienyl Ti Olefin Polymerization Catalysts
J. Am. Chem. Soc., Vol. 122, No. 23, 2000 5505
Scheme 4
than two neutral reactants coming together to engage in
metathesis. The dissociative character of both path a and path
b would be expected to exhibit a positive activation entropy.
The large kH/kD of 9.1(6) at room temperature is less smoothly
rationalized, given that a quite different value of 2.8(2) (80 °C)
was found in the σ-bond methathesis reaction of (d15-Cp*)2-
ScCH3 with C6X6 (X ) H, D).43 It should be noted, however,
that in contact ion pairs involving [MeB(C6F5)3]-, the anions
typically interact with the cation through two or more of the
C-H bonds of the abstracted methyl group.12,30 Furthermore,
an inversion of stereochemistry during the abstraction process
at the methyl carbon is indicative of substantial hybridization
changes at this carbon atom in the abstraction/reattachment
process.45 It is therefore not inconceivable that the observed
kH/kD is a composite of a moderate normal primary kinetic
isotope effect associated with C-H(D) bond breakage in the
σ-bond metathesis reaction and some relatively large secondary
kinetic isotope effects associated with the two C-H(D) bonds
not directly involved in the elimination of CH(D)4. Indeed,
secondary effects associated with the C-H(D) bonds of the
nonabstracted methyl group may also be contributing to the
overall effect.
Figure 5. 1H NMR spectra (400 MHz) of the decomposition of ion
pair 7c (d8-toluene). The peak marked with an asterisk is not due to
CH4.
4-centered transition state. The activation parameters obtained
for the decomposition of 2c are consistent with this notion. The
activation entropy of -8.5(10) eu is substantially less negative
than those observed for other σ-bond metathesis reactions (≈
-34 eu43), but this is consistent with the reaction developing
from the contact ion pair, an inherently more organized structure
(45) Spence R. E. v H.; Piers, W. E.; Sun, Y.; Parvez, M.; MacGillivray,
L. R.; Zaworotko, M. J. Organometallics 1998, 17, 2459.

5506 J. Am. Chem. Soc., Vol. 122, No. 23, 2000
Zhang et al.
Although we were unable to obtain the activation barrier for
ion pair dissociation for any of these compounds, clearly it is
higher than 23.1(8) kcal mol-1, the value for ∆G‡ at 298 K for
the methane elimination process in 2c. This is substantially
higher than ∆G‡ for ion pair dissociation of 14.2(10) kcal mol-1
reported for the constrained geometry system I in d8-toluene31d
and that of 12.4(5) kcal mol-1 (-25 °C) reported for alkoxide
derivative III in the same solvent. This perhaps indicates that
the ketimide ligand is a comparatively poorer donor versus the
amide or alkoxide moieties, although the diverse steric properties
of these three systems undoubtedly play a role in this effect.
Nonetheless, the susceptibility of ion pairs [Cp(L)TiMe]+[MeB-
(C6F5)3]- toward this σ-bond metathetical deactivation process
is likely related to the donor abilities of L. Ligands which are
able to delocalize positive charge from titanium should ef-
fectively lower the electrophilicity of the cationic titanium center
and make it a poorer σ-bond metathesis partner. This also in
all likelihood lowers the free energy barrier to anion dissociation
relative to σ-bond metathesis by decreasing the electrostatic
attraction between the ions. However, to the extent each family
of catalysts undergoes methane loss with essentially equal
facility, these effects are less relevant under above ambient
conditions.
Although the range of rate constants is quite small, compari-
sons within the various ketimide-supported ion pairs reported
here may be rationalized on this basis as well. For example,
the rate of methane evolution in 4c is roughly half that observed
for 2c. The only difference between these compounds is that
for 4c, R ) CH2SiMe2, while in 2c, R ) tBu. Presumably, the
trimethylsilylmethyl substituent is able to stabilize positive
charge on the ketimide carbon via the â-silicon effect,46 lowering
the titanium center’s electrophilicity and raising the barrier to
σ-bond metathesis. Indeed, the apparently slower rate of alkane
elimination from 7c may also have its root in the â-silicon effect,
since the -CH2SiMe3 group might be expected to partially
stabilize the positive charge on titanium47 through hypercon-
jugation.48 As Table 2 also indicates, the decomposition of C5H5-
substituted 1c is approximately 3 times faster than C5Me5-ligated
2c. Although steric effects cannot be discounted here, we believe
the difference in the stabilities of these two species is primarily
due to the more electron donating Cp* ligand decreasing the
electrophilicity of the titanium center.
Conclusions. A detailed understanding of ion-ion interac-
tions in olefin polymerization catalysts is crucial for the design
of new and superior catalysts for this important process. In this
paper, we have shown that the commonly observed methane
elimination process for ion pairs of general formula [Cp(L)TiC-
H3]+[H3CB(C6F5)3]- takes place via a σ-bond metathetical
elimination of methane from a contact ion pair. In the absence
of monomer, this process should be more facile than dissociation
of the contact ion pair, and once it has taken place, the catalyst
is resistant to reactivation. This process is probably related to
the observed evolution of CH4 in MAO-activated metallocene
systems,49 which has been proposed to occur via σ-bond
metathesis between the active cationic center and the C-H
bonds of aluminum (or aluminate)-bound methyl groups.1b
Unlike the deactivation products 1d-5d, reactivation is possible
in the MAO-activated systems, possibly mediated by the
“AlMe3” inevitably present in the MAO.50
For catalysts based on the Cp(L)Ti molecular fragment and
activated by B(C6F5)3, the properties of L are likely important
for influencing the facility of alkane elimination. A ligand L
which will attenuate the electrophilicity of the titanium center
by delocalization of some of the cationic charge renders the
catalyst less susceptible to this deactivation pathway. The
prevalence of this termination step in B(C6F5)3-activated
catalysts is also related, presumably, to the electron-rich nature
of the C-H bonds in the methyl borate anion. It is notable that
each of the catalysts discussed perform at a significantly higher
level when other types of activators are employed.
Experimental Section
General Details. Unless otherwise noted all manipulations were
carried out under argon using an Innovative Technology System One
drybox and/or standard Schlenk techniques on double manifold vacuum
lines.51 Toluene, hexanes, and THF were dried and deoxygenated using
the Grubbs solvent purification system52 and were stored in evacuated
glass vessels over titanocene53 or sodium benzophenone. Deuterated
NMR solvents d6-benzene (C6D6) and d8-toluene (C7D8) were dried
and distilled from sodium/benzophenone ketyl, and d2-dichloromethane
(CD2Cl2) and d5-bromobenzene (C6D5Br) were dried and distilled from
calcium hydride. NMR spectra were recorded on Bruker AC 200, AM
400, or AMX2 300 MHz spectrometers at room temperature in C7D8
unless otherwise specified. Proton and carbon spectra were referenced
to solvent signals, boron spectra to external BF3‚Et2O at 0.0 ppm, and
fluorine spectra to CFCl3 at 0.0 ppm. NMR data are given in ppm; 13
C
resonances for the C6F5 groups were not obtained. Elemental analyses
were performed in the microanalytical laboratory of the Department
of Chemistry at the University of Calgary. Trimethylacetonitrile, 1,2,3,4-
tetramethylcyclopentadiene, (C5Me5)TiCl3, and (C5H5)TiCl3 (Aldrich
Chemicals) were used as received. C5Me4SiMe3,54 (C5Me5)TiMe3,55 and
tBu[CH(SiMe3)2]CdNSiMe356 were synthesized using literature meth-
t
t
ods. Ketimide ligands Bu(R)CdNLi (R ) Bu, CH2SiMe3, Me) were
t
generated in situ by treating BuCN with RLi in THF.
Synthesis of Cp(tBu2CdN)TiCl2, 1a. A solution of Bu2CdNLi
t
(0.67 g, 4.55 mmol) in toluene (10 mL) was added slowly to CpTiCl3
(1.0 g, 4.55 mmol) in toluene (15 mL) at -78 °C. A yellow to orange
color change was observed. The reaction mixture was warmed to room
temperature and stirred for 12 h. The mixture was filtered, the filtrate
was concentrated to ≈5 mL, and hexane (≈25 mL) was added. The
product crystallized at -30 °C as purple crystals. Yield: 90% (1.3 g).
Anal. Calcd for C14H23Cl2NTi: C, 51.88; H, 7.15; N, 4.32. Found: C,
52.20; H, 7.25; N, 4.28. 1H NMR: δ 6.12 (s, 5H, C5H5), 1.04 (s, 18H,
C(CH3)3). 13C NMR: δ 204.06 (CdN), 117.08 (C5H5), 46.71 (CCH3),
30.16 (C(CH3)3).
t
Synthesis of Cp*(tBu2CdN)TiCl2, 2a. A solution of Bu2CdNLi
(0.5 g, 3.45 mmol) in toluene (10 mL) was added slowly to Cp*TiCl3
(1.0 g, 3.45 mmol) in toluene (15 mL) at -78 °C. The reaction mixture
was warmed to room temperature and stirred for 12 h. The red solution
was filtered and concentrated to 5 mL and hexane (25 mL) was added.
The product crystallized at -30 °C as red orange crystals. Yield: 92%
(1.2 g). Anal. Calcd for C19H33Cl2NTi: C, 57.88; H, 8.43; N, 3.55.
1
Found: C, 57.77; H, 8.70; N, 3.61. H NMR: δ 1.98 (s, 15H, C5-
(CH3)5), 1.15 (s, 18H, C(CH3)3). 13C NMR: δ 202.35 (CdN), 128.61
(C5Me5), 46.87 (CCH3), 30.68 (CCH3), 13.28 (C5(CH3)5).
(50) Reddy, S. S.; Shasidhar, G.; Sivaram, S. Macromolecules 1993, 26,
1180.
(51) Burger, B. J.; Bercaw, J. E. Experimental Organometallic Chemistry;
Wayda, A. L., Darensbourg, M. Y., Eds.; ACS Symp. Ser. 357; American
Chemical Society: Washington, DC, 1987.
(52) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.
(53) Marvich, R. H.; Brintzinger, H. H. J. Am. Chem. Soc. 1971, 93,
2046.
(46) Lambert, J. B. Tetrahedron 1990, 46, 2677.
(47) Eisch, J. J.; Piotrowski, A. M.; Brownstein, S. K.; Gabe, E. J.; Lee,
F. L. J. Am. Chem. Soc. 1985, 107, 7219.
(48) Koga, N.; Morokuma, K. J. Am. Chem. Soc. 1988, 110, 108.
(49) (a) Kaminsky, W.; Steiger, R. Polyhedron 1988, 7, 2375. (b)
Kaminsky, W.; Bark, A.; Steiger, R. J. Mol Catal. 1992, 74, 109.
(54) Courtot, P.; Pichon, R.; Salaun, J. Y.; Toupet, L. Can. J. Chem.
1991, 69, 661.
(55) Mena, M.; Royo, P.; Serrano, R.; Pellinghell, M. A.; Tiripicchio,
A. Organometallics 1989, 8, 476.
(56) Hitchcock, P. B.; Lappert, M. F.; Layh, M. J. Organomet. Chem.
1997, 529, 243.

Monocyclopentadienyl Ti Olefin Polymerization Catalysts
J. Am. Chem. Soc., Vol. 122, No. 23, 2000 5507
Synthesis of (C5Me4SiMe3)TiCl3. BuLi (10 mL of a 1.6 M solution
in hexanes) was slowly added to a stirred solution of C5Me4HSiMe3
(3.15 g, 16 mmol) in hexane (60 mL) at -78 °C; the mixture was
stirred and warmed to room temperature over 8 h. The hexanes were
pumped off under vacuum. The residue was suspended in THF (50
mL) and Me3SiCl (1.75 g, 16 mmol) was added. The resulting solution
was stirred for 6 h. The reaction mixture was poured into 100 mL of
cold, distilled water, and the organosilicon product was extracted into
hexane. The extract was concentrated to give 1,1-bis-trimethylsilyl-
2,3,4,5-tetramethylcyclopentadiene as a yellow oil, (purity, >90% by
13C NMR: δ 194.51 (CdN), 129.03, 128.92, 125.86 (C5(CH3)4), 52.51
(Ti-CH3, JC-H ) 118.9 Hz), 47.21 (C(CH3)3), 31.41(C(CH3)3), 15.41
(C5(CH3)4), 12.54 (C5(CH3)4), 0.56 (SiCH3)3).
Synthesis of Cp*[tBu(Me3SiCH2)CdN]TiMe2, 4b. A procedure
analogous to that employed for preparing 2b was used starting from
4a (1.0 g, 2.35 mmol). Yield: 90% (0.81 g). Anal. Calcd for C21H41-
NSiTi: C, 65.76; H, 10.77; N, 3.65. Found: C, 65.74; H, 10.29; N,
3.34. 1H NMR (C6D6): δ 2.08 (s, 2H, CH2Si), 1.93 (s, 15H, C5(CH3)3),
1.09 (s, 9H, C(CH3)3), 0.51 (s, 6H, Ti-CH3), 0.15 (s, 9H, Si(CH3)3).
13C NMR: δ 192.29 (CdN), 121.16 (C5(CH3)5), 51.02 (Ti-CH3, JC-H
) 118.9 Hz), 45.36 (C(CH3)3), 31.34 (CH2Si), 28.79 (C(CH3)3), 12.37
(C5(CH3)5).
1
GC/MS, 3.8 g, 85%). H NMR (CDCl3): δ 1.97(s, 6H, C5CH3), 1.82
(s, 6H, C5CH3), 0.01 (s, 18H, SiCH3). C5Me4(SiMe3)2 (1.75 g, 6.6 mmol)
was slowly added to titanium tetrachloride (1.25 g, 6.6 mmol) dissolved
in 20 mL of toluene. The reaction mixture was stirred for a further 12
h and the solvent then evaporated under vacuum to leave red precipitate
of (C5Me4SiMe3)TiCl3. Yield: 96% (2.2 g). Anal. Calcd for C12H25-
Cl3SiTi: C, 40.99; H, 7.17. Found: C, 40.52; H, 6.86. 1H NMR
(CDCl3): δ 2.51, 2.31 (s, 12H, C5CH3), 0.41 (s, 9H, SiCH3).
Synthesis of Cp*[tBu(Me)CdN]TiMe2, 5b. A procedure analogous
to that employed for preparing 2b was used starting from 5a (1.0 g,
2.84 mmol). Yield: 92% (0.81 g). Anal. Calcd for C18H33NTi: C, 69.44;
H, 10.68; N, 4.49. Found: C, 69.01; H, 10.48; N, 4.36. 1H NMR
(C6D6): δ 1.90 (s, 15H, C5(CH3)3), 1.85 (s, 3H, CH3), 1.07 (s, 9H,
C(CH3)3), 0.53 (s, 6H, Ti-CH3). 13C NMR: δ 187.63 (CdN), 120.87
(C5(CH3)3), 49.13 (Ti-CH3, JC-H ) 118.9), 43.99 (C(CH3)3), 28.33
(C(CH3)3), 20.97 (CH3), 12.01 (C5(CH3)5).
Synthesis of C5Me4SiMe3(tBu2CdN)TiCl2, 3a. A procedure analo-
gous to that used in the preparation of 2a was employed using tBu2Cd
NLi (0.42 g, 2.89 mmol) and C5Me4SiMe3TiCl3 (1.0 g, 2.89 mmol) to
give 3a as orange crystals. Yield: 90% (1.17 g). Anal. Calcd for C14H23-
Cl2NTi: C, 51.88; H, 7.15; N, 4.32. Found: C, 52.20; H, 7.25; N,
4.28. 1H NMR (C6D6): δ 2.29 (s, 6H, C5(CH3)4), 1.87 (s, 6H, C5(CH3)4),
1.19 (s, 18H, C(CH3)3), 0.59 (s, 9H, Si(CH3)3). 13C NMR: δ 202.07
(CdN), 130.62, 129.73, 129.08 (C5(CH3)4), 46.93 (C(CH3)3), 30.62
(C(CH3)3), 16.55, 12.89 (C5(CH3)4)), 1.76 (SiCH3).
t
Synthesis of Bu[(SiMe3)2CH]CdNH. HCl (4.4 mL, 4.4 mmol, 1
M in Et2O) was added via syringe to a solution of But[CH(SiMe3)2]Cd
NSiMe3 (1.32 g, 4.4 mmol) in Et2O (20 mL) at -78 °C. The solution
was allowed to warm to room temperature over 30 min; the solvent
was removed under reduced pressure to give a lime green oil in
1
quantitative yield. H NMR: δ 1.66 (s, 1 H, NH), 1.10 (s, 1H, CH),
0.96 (s, 9H, C(CH3)3), 0.14 (s, 18H, Si(CH3)3).
Synthesis of Cp*[tBu(Me3SiCH2)CdN]TiCl2, 4a. A procedure
analogous to that used for the synthesis of 2a was employed from (tBu)-
(CH2SiMe3)CdNLi (0.51 g, 3.46 mmol) and Cp*TiCl3 (1.0 g, 3.46
mmol). Compound 4a was isolated as red crystals. Yield: 50% (0.74
g). Anal. Calcd for C18H35NCl2SiTi: C, 52.43; H, 8.56; N, 3.40.
Synthesis of Cp*{tBu[(SiMe3)2CH]CdN}TiMe2, 6b. tBu[(SiMe3)2-
CH]CdNH (0.67 g, 2.71 mmol) in toluene (10 mL) was added to a
solution of Cp*TiMe3 (0.62 g, 2.71 mmol) in toluene (10 mL). The
mixture was stirred for 48 h at room temperature. The solvent was
pumped off and the residue recrystallized from pentane to give yellow
crystals of 6b. Yield: 40% (0.5 g). Anal. Calcd for C24H49NSi2Ti: C,
63.26; H, 10.84; N, 3.07. Found: C, 62.84; H, 10.35; N, 3.47. 1H
NMR: δ 2.53 (s, 1H, CH), 1.94 (s, 15H, C5(CH3)3), 1.05 (s, 9H,
C(CH3)3), 0.54 (s, 6H, Ti-CH3), 0.27 (s, 18H, Si(CH3)3). 13C NMR:
δ 195.46 (CdN), 109.41 (C5(CH3)5), 52.36 (Ti-CH3, JC-H ) 119.6
Hz), 46.57 (C(CH3)3), 34.33 (CH), 29.72 (C(CH3)3),12.41 (C5(CH3)5),
2.15 (Si(CH3)3).
1
Found: C, 51.91; H, 8.42; N, 3.73. H NMR: δ 1.99 (s, 15H, C5-
(CH3)5), 1.89 (s, 2H, CH2SiMe3), 1.03 (s, 9H, C(CH3)3), 0.16 (s, 9H,
Si(CH3)3). 13C NMR: δ 198.51 (CdN), 129.09 (C5(CH3)5), 45.43
(C(CH3)3), 31.09 (CH2), 28.69 (C(CH3)3), 13.19 (C5(CH3)5, 0.86 (Si-
(CH3)3).
Synthesis of Cp*[tBu(Me)CdN]TiCl2, 5a. A procedure analogous
t
to that used for the synthesis of 2a was employed from Bu(Me)Cd
Synthesis of Cp*(tBu2CdN)TiMeCl. Dimethyl compound 2b (200
mg, 0.57 mmol), dichloride 2a (200 mg, 0.51 mmol), and B(C6F5)3
(23 mg, 0.045 mmol) were charged into a 50 mL reaction flask in the
glovebox. On the vacuum line, toluene (20 mL) was then vacuum
transferred into the flask at -78 °C. The mixture was slowly warmed
to room temperature and stirred for 1 h and the solvent removed in
vacuo. The residue was extracted with hexanes (25 mL) and the slurry
was filtered. The filtrate was concentrated to 10 mL. The product
crystallized at -30 °C and isolated as red orange crystals. Yield: 90%
(0.35 g). Anal. Calcd for C20H36ClNTi: C, 64.26; H, 9.71; N, 3.75.
Found: C, 64.17; H, 10.38; N, 3.85. 1H NMR (C6D6): δ 1.91 (s, 15H,
C5(CH3)5), 1.18 (s, 18H, C(CH3)3), 0.90 (s, 3H, Ti-CH3). 13C NMR:
δ 199.90 (CdN), 124.16 (C5(CH3)5), 55.41 (Ti-CH3, JC-H ) 122.9
Hz), 46.68 (C(CH3)3), 31.17 (C(CH3)3), 12.83 (C5(CH3)5).
Synthesis of Cp*(tBu2CdN)Ti(CH2SiMe3)Me, 7. Me3SiCH2MgBr
(0.6 mL of a 1 M solution in Et2O) was added to a toluene solution
(15 mL) of Cp*(tBu2CdN)TiMeCl (0.2 g, 0.53 mmol) at -78 °C. After
the addition was complete, the solution was allowed to warm to room
temperature at which time the solvent was removed in vacuo. The
residue was extracted with hexanes (30 mL) and the slurry filtered.
The filtrate was pumped to dryness to give the pure product as an orange
oil. Yield: 98% (0.22 g) Anal. Calcd for C24H47NSiTi: C, 67.73; H,
11.13; N, 3.29. Found: C, 67.01; H, 10.84; N, 3.47. 1H NMR (C6D6):
δ 1.89 (s, 15H, C5(CH3)5), 1.23 (s, 18H, C(CH3)3), 0.54 (s, 3H, Ti-
CH3), 0.28 (s, 9H, Si(CH3)3), 0.76 (d, 1H, CH2, 2JHH ) 11.6 Hz), 0.09
(d, 1H, CH2). 13C NMR (C6D6): δ 196.36 (CdN), 120.98 (C5(CH3)5),
65.87 (TiCH2, JC-H ) 109.7 Hz), 54.59 (Ti-CH3, JC-H ) 120.1 Hz),
46.32 (C(CH3)3), 30.84 (C(CH3)3), 12.47 (C5(CH3)5), 2.84 (Si(CH3)3).
In Situ Generation of [Cp(tBu2CdN)TiMe]+[MeB(C6F5)3]-, 1c.
1b (7.7 mg, 0.027 mmol) and B(C6F5)3 (14 mg, 0.027 mmol) were
dissolved in C7D8 in a J-Young NMR tube. Reaction was immediate
and the sample was assayed by NMR spectroscopy. 1H NMR: δ 5.67
NLi (0.36 g, 3.46 mmol) and Cp*TiCl3 (1.0 g, 3.46 mmol). Compound
5a was isolated as orange crystals. Yield: 75% (0.91 g). Anal. Calcd
for C15H27Cl2NTi: C, 52.96; H, 8.00; N, 4.12. Found: C, 52.62; H,
7.68; N, 4.56. 1H NMR: δ 1.95 (s, 15H, C5(CH3)5), 1.65 (s, 3H, CH3),
0.91 (s, 9H, C(CH3)3. 13C NMR: δ 192.52 (CdN), 130.34 (C5(CH3)5),
44.21 (C(CH3)3), 27.88 (C(CH3)3), 22.42 (CH3), 13.22 (C5(CH3)5).
Synthesis of Cp*(tBu2CdN)TiMe2, 2b. MeMgBr (1.7 mL of a 3M
in Et2O) was added to a toluene solution (30 mL) of dichloride 2a (1.0
g, 2.52 mmol) at -78 °C. After the addition was complete, the solution
was warmed to room temperature; after 30 min the solvent was removed
in vacuo. The residue was extracted with hexanes (30 mL) and the
slurry was filtered. The filtrate was pumped to dryness to give the pure
product as an orange solid. Yield: 89% (0.89 g). Anal. Calcd for C21H39-
NTi: C, 71.36; H, 11.35; N, 3.96. Found: C, 71.44; H, 11.35; N, 3.98.
1H NMR (C6D6): δ 1.89 (s, 15H, C5(CH3)5), 1.28 (s, 18H, C(CH3)3),
0.53 (s, 6H, Ti-CH3). 13C NMR: δ 197.04 (CdN), 112.90 (C5(CH3)5),
51.54 (Ti-CH3, JC-H 118.9 Hz), 46.43 (C(CH3)3), 31.17 (C(CH3)3),
12.36 (C5(CH3)5). d6-2b was prepared in an analogous fashion with
CD3MgBr.
Synthesis of Cp(tBu2CdN)TiMe2, 1b. A procedure analogous to
that employed for preparing 2b was used starting from 1a (1.0 g, 3.08
mmol). Yield: 90% (0.79 g). Anal. Calcd for C16H29NTi: C, 67.83;
H, 10.31; N, 4.70. Found: C, 67.63; H, 10.29; N, 4.94. 1H NMR
(C6D6): δ 6.04 (s, 5H, C5H5), 1.16 (s, 18H, C(CH3)3), 0.73 (s, 6H,
Ti-CH3). 13C NMR: δ 197.21 (CdN), 112.66 (C5H5), 51.54 (Ti-
CH3, JC-H ) 120.6 Hz), 46.31 (C(CH3)3), 31.01 (C(CH3)3).
Synthesis of C5Me4SiMe3(tBu2CdN)TiMe2, 3b. A procedure
analogous to that employed for preparing 2b was used starting from
3a (1.0 g, 2.2 mmol). Yield: 90% (0.82 g). Anal. Calcd for C23H45-
NSiTi: C, 67.12; H, 11.02; N, 3.40. Found: C, 67.27; H, 11.08; N,
3.34. 1H NMR (C6D6): δ 2.15 (s, 6H, C5(CH3)4), 1.69 (s, 6H, C5(CH3)4),
1.22 (s, 18H, C(CH3)3), 0.66 (s, 6H, Ti-CH3), 0.44 (s, 9H, Si(CH3)3).

5508 J. Am. Chem. Soc., Vol. 122, No. 23, 2000
Zhang et al.
(s, 5H, C5H5), 0.97 (s, 3H, Ti-CH3), 0.72 (s, 18H, (C(CH3)3), 0.55
(br. 3H, BCH3). 13C NMR: δ 214.04 (CdN), 116.5 (C5H5), 64.1
(TiCH3, JC-H ) 122.3 Hz), 47.00 (C(CH3)3), 40.08 (CH3B), 29.47
(C(CH3)3). 19F NMR: δ -133.3 (d, 3JF-F ) 28.3 Hz, 6F, o-F), -159.1
129.27 (C5(CH3)5), 94.83 (TiCH2, JC-H ) 123.1 Hz), 47.00 (C(CH3)3),
32.08 (BCH3), 30.54 (C(CH3)3), 11.95 (C5(CH3)5), 2.67 (Si(CH3)3). 19
F
3
3
NMR: δ -131.4 (d, JF-F ) 22.62 Hz, 6F, o-F), -159.8 (t, JF-F
)
3
18.87 Hz, 3F, p-F), -164.5 (t, JF-F ) 18.87 Hz). 11B NMR: -14.0.
Over the course of 48 h, 7c decomposed to neutral complex 2d (see
below for spectroscopic data) and SiMe4.
3
3
(t, JF-F ) 30.0 Hz, 3F, p-F), -164.1 (t, JF-F ) 30.0 Hz, 6F, m-F).
11B NMR: δ -13.4.
Synthesis of Cp*[(tBu2CdN)TiMe]+[MeB(C6F5)3]-, 2c. 2b (0.11
g, 3.1 mmol) and B(C6F5)3 (0.16 g, 3.1 mmol) were charged into a 50
mL reaction flask in the glovebox. On the vacuum line, toluene (15
mL) was condensed into the flask at -78 °C. The mixture was stirred
for 30 min and the solvent removed in vacuo. The residue was triturated
with hexanes to give an orange solid. Yield: 92% (0.25 g). Anal. Calcd
for C38H39BF15NTi: C, 54.12; H, 4.54; N, 1.61. Found: C, 53.69; H,
4.37; N, 1.66. 1H NMR: δ 1.52 (s, 15H, C5(CH3)5), 0.82 (s, 18H,
C(CH3)3), 1.00 (s, 3H, TiCH3), 0.57 (br. 3H, BCH3). 13C NMR: δ
210.79 (CdN), 128.82 (C5(CH3)5), 64.83 (TiCH3, JC-H ) 110.4 Hz),
46.71 (C(CH3)3), 34.29 (CH3B), 29.93 (C(CH3)3), 12.13 (C5(CH3)5. 19F
Synthesis of Cp(tBu2CdN)Ti(C6F5)[CH2B(C6F5)2], 1d. Dimethyl
derivative 1b (0.1 g, 0.35 mmol) and B(C6F5)3 (0.18 g, 0.35 mmol)
were loaded into a 50 mL reaction flask in the glovebox. Toluene (15
mL) was condensed into the vessel at -78 °C, and the solution was
stirred at room temperature for 24 h. The solvent was removed under
vacuum and the residue extracted with hexane and filtered. The solvent
was removed under vacuum and the red solid collected. Yield: 92%
(0.25 g). Anal. Calcd for C33H25BF15NTi: C, 50.08; H, 3.28; N, 1.82.
1
Found: C, 49.58; H, 3.26; N, 1.33. H NMR (C6D6): δ 5.18 (s, 10H,
C5H5), 3.89 (br, 1H, CH2B), 3.55 (br, 1H, CH2B), 1.16 (s, 18H,
C(CH3)3). 13C NMR (C6D6): δ 208.49 (CdN), 116.46 (C5H5), 106.06
(CH2B), 46.38 (C(CH3)3), 30.21 (C(CH3)3). 19F NMR (C6D6): -112.1
(d, 2F, JF-F ) 21.0 Hz, o-F), -127.6, -129.7 (br, 4F, o-F), -152.7
(br, 2F, p-F), -154.6 (m, 4F, p-F), -160.9 (m, 2F, m-F), -161.5 (br,
4F, m-F). 11B NMR (CD2Cl2): δ 51.8.
Synthesis of Cp*(tBu2CdN)Ti(C6F5)[CH2B(C6F5)2], 2d. Dimethyl
derivative 2b (0.1 g, 0.28 mmol) and B(C6F5)3 (0.15 g, 0.28 mmol)
were loaded into a 50 mL reaction flask in the glovebox. Toluene (15
mL) was condensed in at -78 °C, and the solution was stirred at room
temperature for 24 h. The solvent was removed under vacuum to yield
a red solid. Yield: 90% (0.22 g). Anal. Calcd for C38H35BF15NTi: C,
53.73; H, 4.15; N, 1.65. Found: C, 53.03; H, 4.15; N, 1.32. 1H NMR
(C6D6): δ 3.52 (br, 1H, CH2Ti), 2.68 (br, 1H, CH2Ti), 1.64 (s, 15H,
C5(CH3)5), 0.99 (s, 18H, C(CH3)3). 13C NMR (C6D6): δ 212.30 (Cd
N), 130.27 (C5(CH3)5), 102.88 (CH2B), 47.81 (C(CH3)3), 31.08
(C(CH3)3), 13.85 ((C5(CH3)5). 19F NMR (C6D6): -109.4 (d, 2F, o-F,
3JF-F ) 21.0 Hz), -130.1 (br, 4F, o-F), -154.3 (br, 2F, p-F), -154.6
(m, 4F, p-F), -162.2 (m, 2F, m-F), -164.2 (br, 4F, m-F). 11B NMR
(CD2Cl2): δ 50.2.
3
3
NMR: δ -132.8 (d, JF-F ) 20.18 Hz, 6F, o-F), -159.5 (t, JF-F
)
19.14 Hz, 3F, p-F), -164.3 (t, 3JF-F ) 21.21 Hz, 6F, m-F). 11B NMR:
δ -13.5.
In Situ Generation of [C5Me4SiMe3(tBu2CdN)TiMe]+[MeB-
(C6F5)3]-, 3c. 3b (7.7 mg, 0.019 mmol) and B(C6F5)3 (9.5 mg, 0.019
mmol) were dissolved in C7D8 in a J-Young NMR tube. Reaction was
immediate and the sample was assayed by NMR spectroscopy. 1H
NMR: δ 1.80, 1.68, 1.45, 1.36 (s, 12H, C5(CH3)4), 0.84 (s, 18H,
C(CH3)3), 1.19 (s, 3H, TiCH3), 0.65 (br. 3H, BCH3), 0.14 (s, 9H, Si-
(CH3)3). 13C NMR: δ 211.04 (CdN), 136.81, 131.58, 130.45 (C5-
(CH3)4), 66.13 (TiCH3, JC-H ) 123.1 Hz), 47.05 (C(CH3)3), 34.08
(BCH3), 30.06 (C(CH3)3), 12.17, 12.47, 14.68, 15.74 (C5(CH3)4), 0.84
(Si(CH3)3). 19F NMR: δ -132.5 (d, 3JF-F ) 30.0 Hz, 6F, o-F), -159.8
3
3
(t, JF-F ) 30.0 Hz, 3F, p-F), -164.4 (t, JF-F ) 28.3 Hz, 6F, m-F).
11B NMR: δ -14.3.
In Situ Generation of Cp*{[tBu(Me3SiCH2)CdN]TiMe}+[MeB-
(C6F5)3]-, 4c. 4b (7.8 mg, 0.02 mmol) and B(C6F5)3 (10.3 mg,0.02
mmol) were dissolved in C7D8 in a J-Young NMR tube. Reaction was
immediate and the sample was assayed by NMR spectroscopy. 1H
NMR: δ 2.02 (d, 1H, CH2Si), 1.49 (d, 1H, CH2Si), 1.56 (s, 15H, C5-
(CH3)5), 0.89 (s, 3H, TiCH3), 0.78 (br, 3H, BCH3), -0.03 (s, 9H, Si-
(CH3)3). 13C NMR: δ 207.08 (CdN), 65.09 (TiCH3, JC-H ) 123.1
Hz), 44.78 (C(CH3)3), 34.03 (BCH3), 27.82 (C(CH3)3, 0.14 (Si(CH3)3),
11.95 (C5(CH3)5). 19F NMR: δ -132.7 (d, 3JF-F ) 24.31 Hz, 6F, o-F),
In Situ Generation of C5Me4SiMe3(tBu2CdN)Ti(C6F5)[CH2B-
(C6F5)2], 3d. Dimethyl derivative 3b (8.2 mg, 0.026 mmol) and B(C6F5)3
(13.5 mg, 0.026 mmol) were loaded into a J-Young NMR tube and
dissolved in C7D8. Ion pair 3c was produced immediately and
decomposed at room temperature to 3d with elimination of CH4 over
1
the course of 15 h. H NMR: δ 2.12, 1.96, 1.68, 1.53 (s, 12H, C5-
3
3
-159.6 (t, JF-F ) 19.66 Hz, 3F, p-F), -164.2 (t, JF-F ) 19.66 Hz,
(CH3)4), 1.08 (s, 18H, C(CH3)3), 0.22 (s, 9H, Si(CH3)3). 13C NMR: δ
211.91 (CdN), 106.76 (CH2B), 47.11 (C(CH3)3), 30.16 (C(CH3)3),
16.77, 16.17, 12.97 (C5(CH3)4), 1.48 (Si(CH3)3). 19F NMR: -107.2 (d,
2F, o-F, 3JF-F ) 22.1 Hz), -130.1 (br, 4F, o-F), -153.9 (br, 2F, p-F),
6F, m-F). 11B NMR: δ -14.4.
In Situ Generation of Cp*{[tBu(Me)CdN]TiMe}+[MeB(C6F5)3]-,
5c. 5b (8.2 mg, 0.026 mmol) and B(C6F5)3 (13.5 mg, 0.026 mmol)
were dissolved in C7D8 in a J-Young NMR tube. Reaction was
immediate and the sample was assayed by NMR spectroscopy. 1H
NMR: δ 1.57 (s, 3H, TiCH3), 1.51 (s, 15H, C5(CH3)5), 1.45 (s, 3H,
CH3), 0.81 (br, 3H, BCH3), 0.70 (s, 9H, C(CH3)3). 13C NMR: δ 208.54
(CdN), 61.86 (TiCH3, JC-H ) 121.7 Hz), 44.28 (C(CH3)3), 33.17
(BCH3), 26.75 (C(CH3)3), 11.42 (C5(CH3)5). 19F NMR: δ -132.4 (d,
3
-154.4 (t, 4F, p-F, JF-F ) 19.2 Hz), -162.1 (m, 2F, m-F), -163.3
(br, 4F, m-F). 11B NMR: δ 63.6.
In Situ Generation of Cp*[tBu(Me3SiCH2)CdN]Ti(C6F5)[CH2B-
(C6F5)2], 4d. Dimethyl derivative 4b (8.2 mg, 0.026 mmol) and B(C6F5)3
(13.5 mg, 0.026 mmol) were loaded into a J-Young NMR tube and
dissolved in C7D8. Ion pair 4c was produced immediately and
decomposed at room temperature to 4d with elimination of CH4 over
3
3JF-F ) 21.73 Hz, 6F, o-F), -159.4 (t, JF-F ) 19.11 Hz, 3F, p-F),
3
-164.2 (t, JF-F ) 19.11 Hz, 6F, m-F). 11B NMR: δ -13.3.
1
In Situ Generation of [Cp*{tBu[(Me3Si)2CH]CdN}TiMe]+[MeB-
(C6F5)3]-, 6c. 6b (7.2 mg, 0.016 mmol) and B(C6F5)3 (8.1 mg, 0.016
mmol) were dissolved in C7D8 in a J-Young NMR tube. Reaction was
immediate and the sample was assayed by NMR spectroscopy. 1H
NMR: δ 2.32 (s, 1H, CHSi), 1.23 (s, 3H, TiCH3), 0.64 (s, 9H, C(CH3)3),
0.21 (br, 3H, BCH3), 0.00 (br, 18H, Si(CH3)3). 13C NMR: δ 211.05
(CdN), 128.61 (C5(CH3)5), 59.50 (TiCH3, JC-H ) 124.44 Hz), 44.43
(C(CH3)3), 37.78 (CH), 29.25 (BCH3), 28.67 (C(CH3)3), 12.57 (C5-
the course of 15 h. H NMR: δ 2.95 (br, 2H, CH2B), 2.26 (d, 1H,
CH2Si), 2.19 (d, 1H, CH2Si), 1.72 (s, 15H, C5(CH3)5), 0.76 (s, 9H,
C(CH3)3), 0.13 (s, 9H, Si(CH3)3). 13C NMR: 205.57 (CdN), 133.39
(C5(CH3)5), 101.61 (CH2B), 44.26 (C(CH3)3), 35.86 (CH2Si), 28.69
(C(CH3)3), 12.78 (C5(CH3)5), 0.53 (Si(CH3)3). 19F NMR: δ -108.7
(br, 2F, o-F), -129.7 (br, 4F, o-F), -152.8 (br, 2F, p-F), -154.2 (t,
3
3
1F, p-F, JF-F ) 19.14 Hz), -161.6 (t, 2F, m-F, JF-F ) 19.66 Hz),
-163.11 (br, 4F, m-F). 11B NMR: δ 62.2.
3
(CH3)5), 1.72 (Si(CH3)3). 19F NMR: δ -132.4 (d, JF-F ) 21.73 Hz,
In Situ Generation of Cp*[tBu(Me)CdN]Ti(C6F5)[CH2B(C6F5)2],
5d. Dimethyl derivative 5b (8 mg, 0.026 mmol) and B(C6F5)3 (13.1
mg, 0.026 mmol) were loaded into a J-Young NMR tube and dissolved
in C7D8. Ion pair 5c was produced immediately and decomposed at
room temperature to 5d with elimination of CH4 over the course of 15
6F, o-F), -159.6 (br, 3F, p-F), -164.2 (br, 6F, m-F). 11B NMR: δ
-13.3.
In Situ Generation of [Cp*(tBu2CdN)Ti(CH2SiMe3)]+[MeB-
(C6F5)3]-, 7c. Mixed dialkyl derivative 7 (7 mg, 0.016 mmol) and
B(C6F5)3 (8.4 mg, 0.016 mmol) were dissolved in C7D8 in a J-Young
NMR tube. Reaction was immediate and the sample was assayed by
1
h. H NMR: δ 3.15 (br, 1H, CH2B), 2.95 (br, 1H, CH2B), 1.81 (s,
15H, C5(CH3)5), 1.54 (s, 3H, CCH3), 0.81 (s, 9H, C(CH3)3). 13C NMR:
208.1 (CdN), 133.3 (C5(CH3)5), 105.2 (CH2B), 46.8 (C(CH3)3), 29.2
(C(CH3)3), 28.7 (CCH3), 12.8 (C5(CH3)5). 19F NMR: δ -113.1 (br,
2F, o-F), -130.5 (br, 4F, o-F), -153.8 (br, 2F, p-F), -154.5 (t, 1F,
1
NMR spectroscopy. H NMR: δ 1.62 (s, 15H, C5(CH3)5), 1.38 (br,
1H, TiCH2Si), 1.22 (br, 1H, TiCH2Si), 0.91 (s, 18H, C(CH3)3), 0.63
(br, 3H, BCH3), 0.01 (s, 9H, Si(CH3)3). 13C NMR: δ 210.86 (CdN),

Monocyclopentadienyl Ti Olefin Polymerization Catalysts
J. Am. Chem. Soc., Vol. 122, No. 23, 2000 5509
p-F), -159.1 (t, 2F, m-F, 3JF-F ) 19.6 Hz), -164.1 (br, 4F, m-F). 11
NMR: δ 56.8.
B
Table 3. Data Collection and Structure Refinement Details for 1d
formula
fw
cryst syst
a, Å
b, Å
c, Å
R, deg
â, deg
γ, deg
V, Å3
space group
Z
C33H25NBF15Ti‚0.46(C7H8/C6H14)
In Situ Generation of Ion Pair 8. Dimethyl derivative 6b (8.2 mg,
0.018 mmol) and B(C6F5)3 (9.2 mg, 0.018 mmol) were loaded into a
J-Young NMR tube and dissolved in C7D8. Ion pair 6c was produced
immediately and decomposed at room temperature to 8 with elimination
of CH4 over the course of 15 h. 1H NMR: δ 2.22 (s, 1H, CH), 1.83 (s,
15H, C5(CH3)5), 0.74 (s, 9H, C(CH3)3), 0.53 (d, 1H, TiCH2), 0.48 (d,
1H, TiCH2), 0.27 (br, 3H, BCH3), 0.04 (s, 3H, SiCH3), -0.11(s, 3H,
SiCH3), -0.03 (s, 9H, Si(CH3)3). 13C NMR: δ 206.37 (CdN), 69.98
(CH2Si), 42.84 (C(CH3)3), 38.42 (CH), 29.80 (BCH3), 28.36 (C(CH3)3),
12.49 (C5(CH3)5), 3.23 (Si(CH3)2), 2.29 (Si(CH3)3). 19F NMR: δ -132.3
861.40
triclinic
13.469(6)
14.866(5)
10.893(4)
109.16(3)
112.08(3)
74.06(3)
1880(1)
P1h
2
3
3
(d, 2F, o-F, JF-F ) 19.6 Hz), -159.7 (t, 1F, p-F, JF-F ) 18.62 Hz),
F(000)
876.00
1.521
0.333
0.0588
0.1622
1.014
-164.40 (m, 2F, m-F, JF-F ) 18.62 Hz). 11B NMR: δ -13.6.
3
d
calc, mg m-3
In Situ Decomposition of [(C5Me4SiMe2NtBu)TiCH3][CH3B-
(C6F5)2]. The constrained geometry ion pair was generated according
to the literature procedure in d8-toleune. The sample was heated at 60
°C until decomposition was complete. The product of this decomposi-
tion, [C5Me4SiMe2NtBu]Ti(C6F5)[CH2B(C6F5)2], was characterized by
1H and 19F NMR spectroscopy. 1H NMR: δ 2.63 (br, 2H, CH2B), 2.06
(s, 3H, C5(CH3)4), 1.85 (s, 3H, C5(CH3)4), 1.48 (s, 3H, C5(CH3)4), 1.44
(s, 3H, C5(CH3)4), 1.14 (s, 9H, C(CH3)3), 0.43 (s, 3H, SiCH3), 0.41 (s,
3H, SiCH3). 19F NMR: δ -106.5 (br, 1F, o-F), -117.6 (br, 1F, o-F),
µ, mm-1
R
Rw
gof
techniques. Phenyl rings were constrained as regular hexagons. The
non-hydrogen atoms were refined anisotropically. Disordered molecules
of solvents C7H8 and C6H14 were located over inversion centers with
electron density scrambled over a large region; 10 sites were allowed
with partial occupancy factors which led to the equivalent of 6 C-atoms,
giving a 0.46 contribution of the C7H8 and C6H14 solvates. Hydrogen
atoms of the complex were included at geometrically idealized positions
and were not refined; H atoms of the solvates were ignored. All
calculations were performed using the TEXSAN57 crystallographic
software package of Molecular Structure Corporation.
3
-130.3 (br, 4F, o-F), -152.6 (t, 1F, p-F, JF-F ) 19.7 Hz), -153.4
(br, 2F, p-F), -160.2 (br, 1F, m-F), -162.0 (br, 1F, m-F), -162.1 (br,
4F, m-F).
Kinetic Studies. A 1 mL volumetric flask was charged with dimethyl
compound 1b-5b (0.05 mmol) and B(C6F5)3 (0.05 mmol). The solution
was brought to volume with d8-toluene. A J-Young NMR tube was
charged with 0.5 mL of this solution. The disappearance of the ion
Acknowledgment. Financial support for this work was
provided by Nova Chemicals Corporation of Calgary, Alberta.
W.E.P. thanks the Alfred P. Sloan Foundation for a Research
Fellowship (1996-00).
pair complexes Cp(L)TiMe+MeB(C6F5)3 was followed by H NMR
spectroscopy for a period of at least 3 half-lives by integration of
suitable baseline separated resonances. Alternatively, the concentrations
of ion pair complexes were determined by integration against a Cp2Fe
standard. A relaxation delay of 5 s (based on T1 measurements on Cp2-
Fe and 2c) was used to ensure that measured integrals accurately
reflected solution concentrations. Measurements used to determine the
isotope effect were perfromed in triplicate.
-
1
Supporting Information Available: Experimental details
and full listings of crystallographic data, atomic parameters,
hydrogen parameters, atomic coordinates, and complete bond
distances, angles, and torsion angles for 1d (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
X-ray Crystallographic Analysis of 1d. Suitable crystals were
coated with Paratone-8277 oil (Exxon) and mounted onto a glass fiber.
Crystal data and refinement details are collected in Table 3. Measure-
ments were made on a Rigaku AFC6S diffractometer using graphite
monochromated Mo KR radiation (λ ) 0.71069 Å) at -103 °C. The
structure was solved by direct methods and expanded using Fourier
JA994378C
(57) Crystal Structure Analysis Package, Molecular Structure Corporation,
1985 and 1992.