Cationic metallocene polymerization catalysts based on tetrakis(pentafluorophenyl)borate and its derivatives. probing the limits of anion noncoordination via a synthetic, solution dynamic, structural, and catalytic olefin polymerization study

842

Organometallics 1997, 16, 842-857

Ca tion ic Meta llocen e P olym er iza tion Ca ta lysts Ba sed on

Tetr a k is(p en ta flu or op h en yl)bor a te a n d Its Der iva tives.

P r obin g th e Lim its of An ion “Non coor d in a tion ” via a

Syn th etic, Solu tion Dyn a m ic, Str u ctu r a l, a n d Ca ta lytic

Olefin P olym er iza tion Stu d y

Li J ia, Xinmin Yang, Charlotte L. Stern, and Tobin J . Marks*

Department of Chemistry, Northwestern University, 2145 Sheridan Road,

Evanston, Illinois 60208-3113

Received October 18, 1996^X

The synthesis and properties of two soluble, weakly coordinating derivatives of the tetrakis-

(perfluoroaryl)borate anion B(4-C₆F₄TBS)₄^-and B(4-C₆F₄TIPS)₄^-(TBS ) tert-butyldimeth-

ylsilyl and TIPS ) triisopropylsilyl) are reported. Reaction of the trityl salts of the above

anions with a variety of zirconium and thorium L₂MMe₂complexes in benzene or toluene

affords the cationic ion-paired methyl complexes L₂MMe⁺X^-or the corresponding hydrido

complexes L₂MH⁺X^-(L₂) bis(cyclopentadienyl)- or cyclopentadienylamido-type ligand) when

the reaction is carried out under dihydrogen. The solid state structure of the complex

(Me₅Cp)₂ThMe⁺B(C₆F₅)₄^-has been characterized by X-ray diffraction. The B(C₆F₅)₄-based

-

zirconocenium methyl complexes L₂MMe⁺are unstable at room temperature with respect

to, among other factors, intramolecular C-H activation of the ligand framework. In general,

-

the thermal stabilities of the B(C₆F₄TBS)₄- and B(C₆F₄TIPS)₄-derived complexes are greater

-

than those of the corresponding B(C₆F₅)₄- and MeB(C₆F₅)₃-derived analogues. The relative

-

coordinative tendencies of MeB(C₆F₅)₃, B(C₆F₅)₄, B(C₆F₄TBS)₄, and B(C₆F₄TIPS)₄are

estimated from the solution spectroscopic information and the structural dynamics of the

-

ion-pairs and follow the order MeB(C₆F₅)₃> B(C₆F₄TBS)₄≈ B(C₆F₄TIPS)₄> B(C₆F₅)₄.

The coordination of the neutral metallocene precursors to the cationic metallocenes is found

to compete with counteranion coordination. Arene solvent coordination to the zirconium

constrained geometry cation [(Me₄Cp)SiMe₂(N^tBu)]ZrMe⁺is also observed when B(C₆F₅)₄

-

is the counteranion. (1,2-Me₂Cp)₂ZrMe⁺B(C₆F₄TBS)₄^-undergoes slow decomposition under

an inert atmosphere to afford [(1,2-Me₂Cp)₂ZrF]₂(µ-F)⁺B(C₆F₄TBS)₄, which has been

-

characterized by X-ray diffraction. The olefin polymerization activity and thermal stability

of the zirconocene catalysts reaches a maximum when B(C₆F₄TBS)₄and B(C₆F₄TIPS)₄

-

are used as counteranions. The polymerization activity of the zirconium constrained

geometry complex also reaches a maximum in aromatic solvents when B(C₆F₅)₄^-is used as

the counteranion, apparently due to solvent coordination.

In tr od u ction

Cationic group 4 metallocene complexes have been the

subject of intensive research activity over the past

decade because of their extraordinary characteristics as

olefin polymerization catalysts.^1,2Recent observations

suggest that many of the properties of this class of

highly active and selective homogeneous catalysts and

the polymeric products derived therefrom are closely

connected with the nature of the cation-anion tight ion-

pairing (A).^3,4Despite numerous elegant efforts to

“engineer” the cationic portion of such catalysts,-⁵far

less attention has been devoted to optimizing the

charge-compensating anion (X^-) and to understanding

the role of the ion-pairing in catalytic performance. As

R = methyl or hydride; Cp = cyclopentadienyl-type ligand

part of our continuing efforts to study the anion-related

structure-reactivity aspects of this class of compounds,

we were particularly interested in (fluoroaryl)borate

(2) For some leading references in d⁰metal complex-catalyzed olefin

polymerization, see: (a) Coates, G. W.; Waymouth, R. M. Science 1995,

267, 217-219. (b) Leclerc, M. K.; Brintzinger, H. H. J . Am. Chem. Soc.

1995, 117, 1651-1652. (c) Pellecchia, C.; Pappalardo, D.; Oliva, L.;

Zambelli, A. J . Am. Chem. Soc. 1995, 117, 6593-6594. (d) Giardello,

M. A.; Eisen, M. S.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1995,

117, 12114-12129. (e) Kaminsky, W.; Engehausen, R.; Ko¨pf, J . Angew.

Chem., Int. Ed. Engl. 1995, 34, 2273-2275. (f) Spaleck, W.; Ku¨ber, F.;

Winter, A.; Rohrmann, J .; Bochmann, B.; Antberg, M.; Dolle, V.;

Paulus, E. F. Organometallics 1994, 13, 954-963. (g) Razavi, A.;

Atwood. J . L. J . Am. Chem. Soc. 1993, 115, 7529-7530. (h) Chien, J .

C. W.; Rausch, M. D.; Lin, G. Y.; Winter, H. H. J . Am. Chem. Soc.

1991, 113, 8569-8570. (i) Chien, J . C. W.; Tsai, W. M.; Rausch, M. D.

J . Am. Chem. Soc. 1991, 113, 8570-8571. (j) Resconi, L.; Bossi, S.;

Abis, L. Macromolecules 1990, 23, 4489-4491 (k) Ewen, J . A.; J ones,

R. L.; Razavi, A.; Ferrara, J . D. J . Am. Chem. Soc. 1988, 110, 6255-

6256. (l) Ewen, J . A. J . Am. Chem. Soc. 1984, 106, 6355-6364.

^XAbstract published in Advance ACS Abstracts, J anuary 15, 1997.

(1) For recent reviews, see: (a) Bochmann, M. J . Chem. Soc., Dalton

Trans. 1996, 225-270. (b) Brintzinger, H. H.; Fischer, D.; Mu¨lhaupt,

R.; Rieger, B.; Waymouth, R. M. Angew. Chem., Int. Ed. Engl. 1995,

34, 1143-1170. (c) Kaminsky, W. Catal. Today 1994, 20, 257-271.

(d) Mo¨hring, R. C.; Coville, N. J . J . Organomet. Chem. 1994, 479, 1-29.

(e) Marks, T. J . Acc. Chem. Res. 1992, 25, 57-65. (f) J ordan, R. F.

Adv. Organomet. Chem. 1991, 32, 325-387.

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 843

anions and the ion-pairing behavior of complexes de-

rived from them. We previously reported the successful

strategy of using strong organo-Lewis acids such as

B(C₆F₅)₃to generate ion-paired cationic zirconocene

complexes (eq 1), and we have studied their spectro-

scopic and structural properties, solution structural

dynamic behavior, and catalytic performance in detail.^4a,b

ization catalytic consequences. We first report that

changing the metallocene metal center from zirconium

to thorium leads to isolation of the crystallographically

characterizable, ion-paired complex (Me₅Cp)₂Th-

-

Me⁺B(C₆F₅)₄and that changing from the sandwich

ligand framework (A) to a sterically more open half-

sandwich (“constrained geometry”) ligand framework

(B) also leads to the isolation of ion-paired complexes.

toluene

-

L₂ZrMe₂+ B(C₆F₅)₃

8 L₂ZrMe⁺MeB(C₆F₅)₃

(1)

-

However, studies of related borate-based (B(C₆F₅)₄

)

zirconocene complexes were encumbered by the dif-

ficulty of isolating such complexes. Our attempted

syntheses employing the reactions shown in eqs 2 and

3 invariably resulted in insoluble, oily materials con-

taining a number of unidentified species. Nevertheless,

-

In a complementary effort to modify B(C₆F₅)₄with

lipophilic, sterically shielding, π-electron withdrawing

protective groups, B(C₆F₄TBS)₄and B(C₆F₄TIPS)₄

-

L₂ZrMe₂+ Ph₃C⁺B(C₆F₅)₄^toluene8

-

(TBS ) tert-butyldimethylsilyl and TIPS ) triisoprop-

ylsilyl) have been synthesized. These counteranions are

expected to exhibit enhanced solubility in nonpolar

solvents and modified metallocene coordination tenden-

cies compared to B(C₆F₅)₄^-. We report a broad class of

highly reactive cationic Zr and Th metallocene catalysts

prepared using these anions.⁶Several B(C₆F₅)₄^--based

cationic zirconocene hydride complexes are also re-

ported, and the complexity of eqs 2 and 3 is argued to

arise from, among other factors, intramolecular C-H

-

L₂ZrMe⁺B(C₆F₅)₄(2)

L₂ZrMe₂+ ⁿBu₃NH⁺B(C₆F₅)₄^toluene8

-

L₂ZrMe⁺B(C₆F₅)₄^-+ CH₄+ ⁿBu₃N (3)

these mixtures, when freshly prepared, exhibit some of

the highest reported olefin polymerization activities to

date.^2d,i,j,4j,lClearly, such transformations warrant ad-

ditional scrutiny, and additional examples of robust,

weakly coordinating anions capable of stabilizing such

highly electrophilic cations would be desirable to better

understand ion-pairing behavior and the consequences

on reactivity.

activation of the ligand framework. We next evaluate

the relative coordinative tendencies of MeB(C₆F₅)₃

-

,

B(C₆F₅)₄^-, B(C₆F₄TBS)₄^-, and B(C₆F₄TIPS)₄with re-

spect to metallocene cations. In addition, the coordina-

tion of the neutral L₂MMe₂metallocene precursors with

the cationic metallocenes is studied and shown to

compete significantly with borate counteranion coordi-

nation. Arene solvent coordination to the constrained

geometry cation (B) is observed only when B(C₆F₅)₄^-is

the counterion. In addition, the relative thermal sta-

-

In the present contribution, we present a full account

of our efforts to better define the nature of metalloce-

nium-(fluoroaryl)borate ion-pairing and the polymer-

(3) (a) Chien, J . C. W.; Song, W.; Rausch, M. D. J . Polym. Sci., Part

A: Polym. Chem. 1994, 32, 2387-2393. (b) Vizzini, J . C.; Chien, J . C.

W.; Babu, G. N.; Newmark, R. A. J . Polym. Sci., Part A: Polym. Chem.

1994, 32, 2049-2056. (c) Giardello, M. A.; Eisen, M. S.; Stern, C. L.;

Marks, T. J . J . Am. Chem. Soc. 1993, 115, 3326-3327. (d) Herfert, N.;

Fink, G. Makromol. Chem. 1992, 193, 773-778.

bilities and olefin polymerization catalytic properties of

the B(C₆F₅)₄^--, B(C₆F₄TBS)₄^--, and B(C₆F₄TIPS)₄

-

stabilized cationic complexes are investigated, and it is

seen that they are generally more thermally stable and

active for ethylene and propylene polymerization than

the corresponding MeB(C₆F₅)₃^--derived catalysts.

(4) For leading references in chemisty of d⁰cationic group 4 and

thorium metallocenes, see: (a) Yang, X.; Stern, C. L.; Marks, T. J . J .

Am. Chem. Soc. 1994, 116, 10015-10031. (b) Deck, P. A.; Marks, T. J .

J . Am. Chem. Soc. 1995, 117, 6128-6129. (c) Wu, Z.; J ordan, R. F.;

Petersen, J . L. J . Am. Chem. Soc. 1995, 117, 5867-5868. (d) Boch-

mann, M.; Lancaster, S. J .; Robinson, O. B. J . Chem. Soc., Chem.

Commun. 1995, 20, 2081-2082. (e) Erker, G.; Ahlers, W.; Frohlich, R.

J . Am. Chem. Soc. 1995, 117, 5853-5854. (f) Siedle, A. R.; Newmark,

R. A. J . Organomet. Chem. 1995, 119, 119-125. (g) J ia, L.; Yang, X.;

Stern, C. L.; Marks, T. J . Organometallics 1994, 13, 3755-3757. (h)

Sishta, C.; Hathorn, R. M.; Marks, T. J . J . Am. Chem. Soc. 1992, 114,

1112-1114. (i) Yang, X.; King, W. A.; Sabat, M.; Marks, T. J .

Organometallics 1993, 12, 4254-4258. (j) Hlatky, G. G.; Eckman, R.

R.; Turner, H. W. Organometallics 1992, 11, 1413-1416. (k) Ewen, J .

A.; Elder, M. Eur. Pat. Appl. EP 427697, 1991. (l) Siedle, A. R.;

Newmark, R. A.; Lamanna, W. M.; Schroepfer, J . N. Polyhedron 1990,

9, 301-308. (m) Turner, H. W.; Hlatky, G. G. PCT Int. Appl. WO 88/

05793 (Eur. Pat. Appl. EP 211004, 1988).

Exp er im en ta l Section

Ma ter ia ls a n d Meth od s. All operations were performed

with rigorous exclusion of oxygen and moisture in flamed

Schlenk-type glassware on a dual-manifold Schlenk line or

interfaced to a high-vacuum line (10^-5Torr) or in a dinitrogen-

filled, Vacuum Atmospheres glovebox with a high-capacity

atmosphere circulator (1-2 ppm O₂). Argon (Matheson,

prepurified), ethylene (Matheson, CP), propylene (Matheson,

PP), and dihydrogen (Linde) were purified by passage through

a supported MnO oxygen-removal column and a Davison 4 A

molecular sieve column. Hydrocarbon solvents (toluene, pen-

tane) were distilled under nitrogen from Na/K alloy. Ether

solvents (THF, Et₂O) were distilled under nitrogen from

sodium benzophenone ketyl. All solvents were stored in vacuo

over Na/K in Teflon-valved bulbs. 1-Hexene (Aldrich, 99%)

was dried over Na/K for 1 h at room temperature, freeze-

pump-thaw-degassed, and vacuum-transferred into a Teflon-

(5) For leading references in zirconium catalysts with non-cyclo-

pentadienyl ligands, see: (a) Flores, J . C.; Chien, J . C. W.; Rausch, M.

D. Organometallics 1995, 14, 1827-1833. (b) Bowen, D. E.; J ordan,

R. F.; Rogers, R. D. Organometallics 1995, 14, 3630-3635. (c) Black,

D. G.; Swenson, D. C.; J ordan, R. F.; Rogers, R. D. Organometallics

1995, 14, 3539-3550. (d) Tjaden, E. B.; Swenson, D. C.; J ordan, R. F.;

Petersen, J . L. Organometallics 1995, 14, 371-386. (e) van der Linden,

A.; Schaverien, C. J .; Meijboom, N.; Ganter, C.; Orpen, A. G. J . Am.

Chem. Soc. 1995, 117, 3008-3021. (f) Shapiro, P. J .; Cotter, W. D.;

Schaefer, W. P.; Labinger, J . A.; Bercaw, J . E. J . Am. Chem. Soc. 1994,

116, 4623-4640. (g) Quan, R. W.; Bazan, G. C.; Kiely, A. F.; Schaefer,

W. P.; Bercaw, J . E. J . Am. Chem. Soc. 1994, 116, 4489-4490. (h)

Devore, D. D. Eur. Pat. Appl. 514828, 1991. (i) Stevens, J . C.; Timmers,

F. J .; Wilson, D. R.; Schmidt, G. F.; Nickias, P. N.; Rosen, R. K.; Knight,

G. W.; Lay, S. Y. Eur. Pat. Appl. 416815, 1990.

(6) For preliminary communications of part of these results, see:

(a) J ia, L.; Yang, X.; Ishihara, A.; Marks, T. J . Organometallics 1995,

14, 3135-3137. (b) Yang, X.; Stern, C. L.; Marks, T. J . Organometallics

1991, 10, 840-842.

844 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

valved storage flask. Deuterated solvents were purchased

from Cambridge Isotope Laboratories (all g99 atom % D),

degassed and dried over Na/K alloy, and stored in resealable

flasks. 1-Bromo-2,3,5,6-tetrafluorobenzene and bromopenta-

fluorobenzene (Aldrich) were dried over molecular sieves. tert-

Butyldimethylsilyl (TBS) triflate (98%), triisopropylsilyl (TIPS)

triflate (98%), BCl₃(1.0 M in hexane), and ⁿBuLi (1.6 M in

hexane) (Aldrich) were used as received. Triphenylcarbenium

tetrakis(pentafluorophenyl)borate was received as a gift from

Austin Chemicals Corp. and used as received.

The complexes (Me₅Cp)₂ThMe₂,⁷Cp₂ZrMe₂,⁸(1,2-Me₂Cp)₂-

ZrMe₂,⁹(Me₅Cp)₂ZrMe₂,¹⁰[(Me₄Cp)SiMe₂(N^tBu)]ZrMe₂,¹¹and

[(Me₄Cp)SiMe₂(N^tBu)]ZrMe⁺MeB(C₆F₅)₃^{- 11}were synthesized

following established procedures.

P h ysica l a n d An a lytica l Mea su r em en ts. NMR spectra

were recorded on either Varian Unity Plus 400 (FT, 400 MHz,

¹H; 100 MHz, ¹³C), Gemini 300 (FT, 300 MHz, ¹H; 75 MHz,

¹³C; 150 MHz, ¹⁹F), or VXR-300 (FT 300 MHz, ¹H; 75 MHz,

¹³C) spectrometers. Chemical shifts for ¹H and ¹³C spectra

were referenced using internal solvent resonances and are

reported relative to tetramethylsilane. ¹⁹F NMR were refer-

enced to external CFCl₃. NMR experiments on air-sensitive

samples were conducted in Teflon valve-sealed sample tubes

(J . Young). Elemental analyses were performed by Oneida

Research Service, Inc., Whitesboro, NY. GPC analyses of the

polymer samples were performed at Akzo-Nobel Corporation

Laboratories, Dobbs Ferry, NY.

Syn th esis of HC₆F ₄Si^tBu Me₂(1a ). In a 500 mL Schlenk

flask, C₆HBrF₄(14.8 g, 64.6 mmol) was dissolved in diethyl

ether (300 mL) and cooled to -78 °C. ⁿBuLi (40 mL, 1.6 M in

hexanes) was then added dropwise over 0.5 h while the

solution was stirred. Precipitation was immediately observed

upon ⁿBuLi addition. The reaction mixture was stirred for an

additional 2 h before TBS triflate (17.0 g, 64.6 mmol) was

injected into the suspension by a syringe. The flask was then

slowly warmed to room temperature over the period of 8 h,

and the resulting suspension was suction filtered through a

frit. After the solvent was removed at 25 °C under reduced

pressure, the nonvolatile liquid residue was distilled and a

colorless liquid product was collected (bp 45 °C/0.8 mmHg).

Yield: 14.0 g (82%). ¹H NMR (CCl₂D₂, 20 °C): δ 0.40 (“t”, 6H),

0.93 (s, 9H), 7.10 (m, 1H). ¹⁹F NMR (CCl₂D₂, 20 °C): δ -128.2

(m), -132.4 (m). ¹³C NMR (CCl₂D₂, 20 °C): δ -3.7, 18.1, 26.5,

108.0 (t, J _C-F) 42 Hz), 116.5 (t, J _C-F) 48 Hz), 146.8 (d, J _C-F

) 264 Hz), 150.4 (d, J _C-F) 262 Hz).

mL) was layered on top of the solution. After 2 days, the

recrystallized product was collected as large colorless crystals

by filtration. Yield: 3.9 g (74%). ¹H NMR (C₆D₆, 20 °C): δ

0.20 (s, 24H), 0.80 (m, 12H), 0.82 (s, 36H), 3.05 (q, J ) 7.1 Hz,

8H). ¹⁹F NMR (C₆D₆, 20 °C): δ -129.4 (m), -133.8 (m).

-

Syn th esis of Li(Et₂O)_n⁺B(C₆F ₄SiⁱP r ₃)₄(2b). The reac-

tion was carried out following a procedure similar to that for

n

2a above, using 1b (6.8 g, 22.2 mmol), BuLi (10.5 mL, 2.5 M

in hexanes), and BCl₃(5.6 mL, 1.0 M in hexanes). Excess

diethyl ether (∼800 mL) was needed to completely dissolve

all of the product. The volume of the solution was reduced to

50 mL, and the white powder product was isolated after

filtration and drying under vacuum (without further purifica-

tion) in 72% yield (5.9 g). ¹H NMR (CDCl₃, 20 °C): δ 0.80 (t,

J ) 7.1 Hz, 12H), 1.08 (d, J ) 7.5 Hz, 72H), 1.52 (“p”, J ) 7.5

Hz, 12H), 3.05(q, J ) 7.1 Hz, 8H). ¹⁹F NMR (CDCl₃, 20 °C):

δ -131.3 (m), -132.1 (m).

-

Syn th esis of (C₆H₅)₃C⁺B(C₆F ₄Si^tBu Me₂)₄(3a ). Com-

pound 2a (3.8 g, ∼3.1 mmol) and (C₆H₅)₃CCl (1.14 g, 4.0 mmol)

were loaded into a 500 mL Schlenk flask and suspended in

pentane (250 mL). The orange suspension was stirred for 6 h

at room temperature and then suction filtered through a frit.

The orange solid which was collected was then dissolved in

CH₂Cl₂, and the resulting suspension was filtered through a

Celite-covered frit to remove LiCl. The volume of the filtrate

was reduced to 50 mL, and pentane was added until the first

precipitation was observed. The solution was then slowly

cooled to -78 °C and stored for 12 h, and the solid product

was collected by rapid filtration. The orange crystalline

product was isolated in 86% yield (3.5 g). ¹H NMR (CCl₂D₂,

20 °C): δ 0.31 (s, 24H), 0.88 (s, 36H), 7.64 (d, J ) 6.9 Hz, 6H),

7.85 (t, J ) 6.9 Hz, 6H), 8.25 (“t”, J ) 6.9 Hz, 3H). ¹⁹F NMR

(CCl₂D₂, 20 °C): δ -129.0 (m), -132.5 (m). ¹³C NMR (CCl₂D₂,

20 °C): δ -3.8, 17.8, 26.4, 108.4 (t, J _C-F) 42 Hz), 131.0, 140.2,

143.0, 144.0, 148.4 (d, J _C-F) 262 Hz), 148.6 (d, J _C-F) 262

Hz), 211.3. Anal. Calcd for C₆₇H₇₅BF₁₆Si₄: C, 61.55; H, 5.78.

Found: C, 61.83; H, 5.61.

-

Syn th esis of (C₆H₅)₃C⁺B(C₆F ₄SiⁱP r ₃)₄(3b). This syn-

thesis was carried out following a procedure similar to that

for 3a above, using compound 2b (8.04 g, ∼5.6 mmol) and

(C₆H₅)₃CCl (1.7 g, 5.9 mmol). The orange crystalline product

was isolated in 82% yield (6.8 g). ¹H NMR (CDCl₃): δ 1.09 (d,

J ) 7.5 Hz, 72H), 1.51 (“p”, J ) 7.5 Hz, 12H), 7.60 (d, J ) 6.9

Hz, 6H), 7.85 (t, J ) 6.9 Hz, 6H), 8.25 (“t”, J ) 6.9 Hz, 3H).

¹⁹F NMR (CDCl₃, 20 °C): δ -131.3 (m), -132.1 (m). ¹³C NMR

(CDCl₃, 20 °C): δ 12.8, 18.8, 108.6 (t, J _C-F) 42 Hz), 130.6,

Syn th esis of HC₆F ₄SiⁱP r ₃(1b). This synthesis followed a

procedure similar to that for 1a above, using C₆HBrF₄(15.1

g, 65.9 mmol), ⁿBuLi (41 mL, 1.6 M in hexanes), and TIPS

triflate (21.0 g, 66.0 mmol). The product was distilled and

collected at 62 °C/2 mmHg. Yield: 17.6 g (87%). ¹H NMR

(CCl₂D₂, 20 °C): δ 0.98 (d, J ) 7.5 Hz, 72H), 1.50 (“p”, J ) 7.5

Hz, 12H), 7.10 (m, 1H). ¹⁹F NMR (CDCl₃, 20 °C): δ -128.2

(m), -132.4 (m).

139.2, 142.4, 144.0, 147.8 (d, J _C-F) 26 Hz), 147.9 (d, J _C-F

)

260 Hz), 211.3. Anal. Calcd for C₇₉H₉₉BF₁₆Si₄: C, 64.30; H,

6.76. Found: C, 64.57; H, 6.89.

-

Syn th esis of Cp ₂Zr Me⁺B(C₆F ₄Si^tBu Me₂)₄(4a ). Com-

pound 3a (390 mg, 0.30 mmol) and Cp₂ZrMe₂(82 mg, 0.32

mmol) were loaded into a 50 mL reaction flask in the glovebox.

The flask was attached to the vacuum line and evacuated, and

toluene (20 mL) was condensed in the flask under vacuum.

The flask was then backfilled with argon and warmed to room

temperature, and the reaction mixture was stirred for 4 h. A

white slurry formed during this period. After the slurry was

collected by filtration and dried under vacuum, the pale yellow

product was isolated. Yield: 292 mg (75%). ¹H NMR (C₆D₆,

60 °C): δ 0.23 (s, 24H), 0.56 (s, 3H), 0.85 (s, 36H), 5.75 (s,

10H). ¹⁹F NMR (C₆D₆, 60 °C): δ -128.2 (b), -130.6 (b). ¹³C

NMR (C₆D₄Cl₂, 20 °C): δ -4.4, 17.2, 25.8, 49.6, 111.6 (t, J _C-F

) 42 Hz), 114.8, 119.8, 147.0 (d, J _C-F) 262 Hz), 149.8 (d, J _C-F

) 261 Hz). Anal. Calcd for C₅₉H₇₃BF₁₆Si₄Zr: C, 54.49; H, 5.66.

Found: C, 54.30; H, 5.35.

Syn th esis of Li(Et₂O)_n⁺B(C₆F ₄Si^tBu Me₂)₄^-(2a ). In a 500

mL Schlenk flask, 1a (5.4 g, 20.4 mmol) was dissolved in

diethyl ether (300 mL) and cooled to -78 °C. ⁿBuLi (13.0 mL,

1.6 M in hexanes) was then added dropwise over 0.5 h while

the solution was stirred. After an additional 2 h, BCl₃(4.2

mL, 1.0 M in hexanes) was injected by a syringe and the

reaction mixture was warmed to room temperature over the

period of 16 h. The resulting suspension was then suction

filtered through a Celite-covered frit. The volume of the

mother liquid was next reduced to 75 mL, and pentane (250

(7) Fagan, P. J .; Manriquez, J . M.; Maatta, E. A.; Seyam, A. M.;

Marks, T. J . J . Am. Chem. Soc. 1981, 103, 6650-6667.

(8) Samuel, E.; Rausch, M. D. J . Am. Chem. Soc. 1973, 95, 6263-

6267.

(9) Smith, G. M. Ph.D. Thesis, Northwestern University, Evanston,

IL, 1985.

(10) Manriquez, J . M.; McAlister, D. R.; Sanner, R. D.; Bercaw, J .

E. J . Am. Chem. Soc. 1978, 100, 2716-2724.

(11) Lanza, G.; Fragala, I. L.; Fu, P.-F.; Marks, T. J .; Wilson, D. R.;

Rudolph, P. R. Manuscript in preparation.

-

Syn th esis of Cp ₂Zr Me⁺B(C₆F ₄SiⁱP r ₃)₄(4b). This syn-

thesis was carried out following a procedure similar to that

for 4a above, using compound 3b (405 mg, 0.27 mmol) and

Cp₂ZrMe₂(68 mg, 0.27 mmol). The pale yellow product was

isolated in 83% yield (332 mg). ¹H NMR (C₆D₆, 60 °C): δ 0.54

(s, 3H), 1.01 (d, J ) 7.5 Hz, 72H), 1.43 (“p”, J ) 7.5 Hz, 12H),

5.75 (s, 10H). ¹⁹F NMR (C₆D₆, 60 °C): δ -130.1 (b), -131.4

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 845

(b). ¹³C NMR (C₆D₄Cl₂, 20 °C): δ 12.0, 19.2, 111.6 (t, J _C-F

42 Hz), 114.8, 120, 146.6 (d, J _C-F) 259 Hz), 150 (d, J _C-F

)

Hz), 167 (b). Anal. Calcd for C₄₄H₃₁BF₂₀Zr: C, 50.73; H, 3.00.

Found: C, 51.69; H, 2.99.

Syn t h esis of [(Me₄Cp )SiMe₂(N^tBu )]Zr Me⁺B(C₆F ₄Si-

^tBu Me₂)₄^-(7a ). Compound 3a (362 mg, 0.28 mmol) and [(Me₄-

Cp)SiMe₂(N^tBu)]ZrMe₂(103 mg, 0.28 mmol) were loaded into

a 50 mL reaction flask in the glovebox. The flask was attached

to the vacuum line and evacuated, and toluene (30 mL) was

condensed into the flask under vacuum. The flask was then

backfilled with argon, warmed to room temperature, and

stirred for 4 h. The orange slurry gradually turned to a

transparent colorless solution during this time period. The

volume of toluene was next reduced to 10 mL, and 20 mL of

hexanes was condensed into the flask at -78 °C. A clear

solution formed at room temperature, and the solution was

slowly cooled to -78 °C and kept at this temperature for 12 h.

The pale yellow crystalline product was isolated by quick

filtration through a course frit and was dried under vacuum

overnight. Yield: 123 mg (32%). ¹H NMR (C₆D₆, 60 °C): δ

0.21 (s, 3H), 0.23 (s, 24H), 0.28 (s, 3H), 0.35 (s, 3H), 0.86 (s,

36H), 1.23 (s, 9H), 1.49 (s, 3H), 1.83 (s, 3H), 1.97 (s, 3H), 2.05

(s, 3H). ¹³C NMR (C₆D₆, 20 °C): δ -4.4 (TBS), 5.7 (SiMe₂),

9.0 (SiMe₂), 11.3 (Me₄Cp), 12.4 (Me₄Cp), 13.2 (Me₄Cp), 14.9

(Me₄Cp), 17.2, 18.7 (CMe₃), 25.8 (TBS), 30.8 (CMe₃), 57.2 (Zr-

Me), 102.9 (Cp), 108.0 (t, J _C-F) 42 Hz, C₆F₅), 131.9 (Cp), 132.1

(Cp), 132.4 (Cp), 133.6 (Cp), 148.9 (d, J _C-F) 262 Hz, C₆F₄),

262 Hz). Anal. Calcd for C₇₁H₉₇BF₁₆Si₄Zr: C, 58.06; H, 6.66.

Found: C, 58.32; H, 6.79.

Syn th esis of (1,2-Me₂Cp )₂Zr Me⁺B(C₆F ₄Si^tBu Me₂)₄^-(5a ).

Compound 3a (390 mg, 0.30 mmol) and (1,2-Me₂Cp)₂ZrMe₂(92

mg, 0.30 mmol) were loaded into a 50 mL reaction flask in

the glovebox. The flask was attached to the vacuum line and

evacuated, and toluene (20 mL) was condensed into the flask

under vacuum. The flask was then backfilled with argon and

warmed to room temperature. The reaction mixture was

stirred for 6 h, during which time a pale yellow solution

formed. The volume of the solution was reduced to ∼5 mL,

and 20 mL of pentane was condensed into the flask to

precipitate the product. After this was collected by filtration

and dried under vacuum, the off-white product was isolated

in 78% yield (317 mg). ¹H NMR (C₆D₆, 20 °C): δ 0.21(s, 24H),

0.34 (s, 3H), 0.83 (s, 36H), 1.37 (s, 6H), 1.61 (s, 6H), 5.00 (b,

2H), 5.69 (b, 2H), 5.97 (t, 2H). ¹³C NMR (C₆D₆, 20 °C): δ -3.9,

12.5, 17.7, 26.4, 45.9, 108.1 (t, J _C-F) 42 Hz), 110.1, 111.8,

119.8, 133.5, 147.7 (d, J _C-F) 262 Hz), 148.0 (d, J _C-F) 262

Hz). Anal. Calcd for C₆₃H₈₁BF₁₆Si₄Zr: C, 55.78; H, 6.02.

Found: C, 55.56; H, 6.01.

-

Syn th esis of (1,2-Me₂Cp )₂Zr Me⁺B(C₆F ₄SiⁱP r ₃)₄(5b).

This synthesis was carried out following a procedure similar

to that for 5a above, using compound 3b (420 mg, 0.28 mmol)

and (1,2-Me₂Cp)₂ZrMe₂(87 mg, 0.28 mmol). The off-white

product was isolated in 78% yield (332 mg). ¹H NMR (toluene-

d₈, 20 °C): δ 0.37 (s, 3H), 1.08 (d, J ) 7.5 Hz, 72H), 1.43 (s,

6H), 1.51 (“t”, J ) 7.5 Hz, 12H), 1.73 (s, 6H), 5.08 (b, 4H), 5.75

(b, 4H), 6.01 (t, J ) 1.0 Hz, 2H). ¹³C NMR (toluene-d₈, 20 °C):

δ 12.0, 12.8, 19.2, 47.0, 109.1 (t, J _C-F) 42 Hz), 110.1, 113.8,

121.4, 149.8 (d, J _C-F) 262 Hz), 150.7 (d, J _C-F) 262 Hz). Anal.

Calcd for C₇₅H₁₀₅BF₁₈Si₄Zr: C, 59.07; H, 6.94. Found: C, 59.34;

H, 6.94.

150.0 (d, J _C-F) 262 Hz, C₆F₄). Anal. Calcd for C₆₄H₉₀BNF₁₆

-

Si₅Zr: C, 54.16; H, 6.39; N, 0.99. Found: C, 54.46; H, 5.93; N,

0.62.

t

+

i

-

4

Syn th esis of [(Me Cp)SiMe₂(N Bu )]Zr Me B(C₆F₄Si P r ₃)₄

(7b). Compound 3b (412 mg, 0.28 mmol) and [(Me₄Cp)-

SiMe₂(N^tBu)]ZrMe₂(104 mg, 0.28 mmol) were loaded into a

50 mL reaction flask in the glovebox. The flask was connected

to the vacuum line and evacuated, and benzene (30 mL) was

condensed into the flask under vacuum. The flask was next

backfilled with argon, warmed to room temperature, and

stirred for 4 h. The orange slurry gradually became a

transparent, colorless solution during this time period. Ben-

zene was then removed in vacuo, and 20 mL of hexanes was

condensed into the flask at -78 °C. The slurry was stirred at

room temperature for 1 h and was filtered. The off-white

product was isolated by filtration and dried under vacuum

overnight. Yield: 332 mg (75%). ¹H NMR (C₆D₆, 60 °C): δ

0.16 (s, 3H), 0.23 (s, 3H), 0.32 (s, 3H), 1.05 (d, J ) 7.6 Hz,

72H), 1.22 (s, 9H), 1.47 (b, 12H), 1.54 (s, 3H), 1.82 (s, 3H),

1.97 (s, 3H), 2.04 (s, 3H). ¹³C NMR (C₆D₆, 20 °C): δ 4.0 (TIPS),

5.7 (SiMe₂), 9.0 (SiMe₂), 11.3 (Me₄Cp), 12.3 (Me₄Cp), 13.3 (Me₄-

Cp), 14.9 (Me₄Cp), 18.7 (TIPS), 30.8 (CMe₃), 40.2 (CMe₃), 57.2

(Zr-Me), 102.6 (Cp), 108.0 (t, J _C-F) 42 Hz, C₆F₅), 131.9 (Cp),

132.1 (Cp), 132.4 (Cp), 133.6 (Cp), 149.7 (d, J _C-F) 262 Hz),

150.0 (d, J _C-F) 262 Hz). Anal. Calcd C₇₆H₁₁₄BNF₁₆Si₅Zr: C,

57.48; H, 7.24; N, 0.88. Found: C, 57.86; H, 6.62; N, 0.46.

Syn th esis of [(Me₄Cp)SiMe₂(N^tBu )]Zr Me⁺B(C₆F₅)₄^-‚C₆H₆

(7c). This synthesis was carried out following a procedure

Syn th esis of (1,2-Me₂Cp )₂Zr H⁺B(C₆F ₅)₄^-(5c). This syn-

thesis was carried out following a procedure similar to that

for 5a above, except that the reaction was carried out under

-

1.0 atm of H₂using compound Ph₃C⁺B(C₆F₅)₄(258 mg, 0.28

mmol) and (1,2-Me₂Cp)₂ZrMe₂(87 mg, 0.28 mmol). The off-

white product was isolated in 78% yield (212 mg). ¹H NMR

(C₆D₆/THF-d₈, 20 °C): δ 1.71 (s, 12H), 5.71 (d, J ) 2.8 Hz, 4H),

5.78 (t, J ) 2.8 Hz, 2H), 9.4 (s, 1H). ¹⁹F NMR (C₆D₆, 60 °C):

δ -132.3 (b), -163.5 (t, J ) 20 Hz), -167.2 (b). Anal. Calcd

for C₃₈H₁₉BF₂₀Zr: C, 47.66; H, 2.00. Found: C, 47.11; H, 1.60.

Syn th esis of (Me₅Cp)₂Zr H⁺B(C₆F₄Si^tBu Me₂)₄^-(6a). This

synthesis was carried out following a procedure similar to that

for 5a above, except that the reaction was carried out under

1.0 atm of H₂using 3a (400 mg, 0.31 mmol) and (Me₅Cp)₂ZrMe₂

(120 mg, 0.31 mmol). Yield: 318 mg (72%). ¹H NMR (C₆D₆,

20 °C): δ 0.25 (s, 24H), 0.87 (s, 36H), 1.63 (s, 30H), 7.90 (b,

1H). ¹³C NMR (C₆D₆, 20 °C): δ -3.9, 11.2, 17.7, 26.5, 110.1

(t, J _C-F) 42 Hz), 122.5, 148.1 (d, J _C-F) 262 Hz), 148.2 (d,

J _C-F) 262 Hz). Anal. Calcd for C₆₈H₉₁BF₁₆Si₄Zr: C, 57.24;

H, 6.43. Found: C, 56.87; H, 6.46.

-

similar to that for 7b above, using Ph₃C⁺B(C₆F₅)₄(440 mg,

0.477 mmol) and [(Me₄Cp)SiMe₂(N^tBu)]ZrMe₂(176 mg, 0.477

mmol). Yield: 413 mg (84%). ¹H NMR (C₆D₆, 20 °C): δ -0.45

(s, 3H), 0.13 (s, 3H), 0.23 (s, 3H), 0.78 (s, 9H), 1.37 (s, 3H),

1.47 (s, 3H), 1.58 (s, 3H), 1.65 (s, 3H). ¹⁹F NMR (C₆D₆, 20 °C):

-

Syn th esis of (Me₅Cp )₂Zr H⁺B(C₆F ₄SiⁱP r ₃)₄(6b). This

synthesis was carried out following a similar procedure to that

δ -132.2 (b), -162.2 (t, J ) 19 Hz), -166.2 (d, J ) 16 Hz). ¹³

C

for 6a above, using compound 3b (368 mg, 0.25 mmol) and

1

(Me₅Cp)₂ZrMe₂(98 mg, 0.25 mmol). Yield: 311 mg (78%). H

NMR (C₆D₆, HMQC, 20 °C): δ 5.1 (SiMe₂), 10.2 (SiMe₂), 10.4

(Me₄Cp), 11.1 (Me₄Cp), 12.5 (Me₄Cp), 15.4 (Me₄Cp), 32.2 (Zr-

Me), 33.6 (CMe₃). Anal. Calcd for C₄₆H₃₆BNF₂₀SiZr (7c‚

C₆H₆): C, 49.60; H, 3.35; N, 1.26. Found: C, 49.05; H, 3.45;

N, 1.26.

NMR (C₆D₆, 20 °C): δ 1.12 (d, J ) 7.5 Hz, 72H), 1.50 (“p”, J )

7.5 Hz, 12H), 1.72 (s, 30H), 7.96 (b, 1H). ¹³C NMR (toluene-

d₈, 20 °C): δ 12.0, 12.5, 18.8, 109.8 (t, J _C-F) 42 Hz), 111.9,

148.2 (d, J _C-F) 262 Hz), 148.0 (d, J _C-F) 262 Hz). Anal. Calcd

for C₈₀H₁₁₅BF₁₆Si₄Zr: C, 60.46; H, 7.33. Found: C, 60.56; H,

6.98.

Syn th esis of (Me₅Cp )₂Zr H⁺B(C₆F ₅)₄^-(6c). This synthesis

was carried out following a procedure similar to that for 6a

above. The reagents Ph₃C⁺B(C₆F₅)₄^-(286 mg, 0.31 mmol) and

(Me₅Cp)₂ZrMe₂(120 mg, 0.31 mmol) were used. Yield: 220

mg (68%). ¹H NMR (C₆D₆, 20 °C): δ 1.38 (s, 30H), 7.41 (b,

1H). ¹⁹F NMR (C₆D₆, 20 °C): -133.9 (b), -161.0 (t, J ) 20

-

Syn th esis of (Me₅Cp )₂Th Me⁺B(C₆F ₄Si^tBu Me₂)₄(8a ).

This synthesis was carried out following a procedure similar

to that for 5a above, using 3a (368 mg, 0.28 mmol) and (Me₅-

Cp)₂ThMe₂(150 mg, 0.28 mmol). Yield: 72% (318 mg). ¹H

NMR (C₆D₆, 20 °C): δ 0.25 (s, 24H), 0.53 (s, 3H), 0.87 (s, 36H),

1.75 (s, 30H). ¹⁹F NMR (C₆D₆, 20 °C): -128.2 (b), -130.4 (b).

¹³C NMR (C₆D₆, 20 °C): δ -3.7, 10.8, 17.8, 26.5, 77.0, 109.1

(t, J _C-F) 42 Hz), 129.2, 148.0 (d, J _C-F) 262 Hz), 151.0 (d,

846 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

J _C-F) 258 Hz). Anal. Calcd for C₆₈H₉₃BF₁₆Si₄Th: C, 52.20;

H, 6.28. Found: C, 52.35; H, 5.84.

ing to Zr⁺-Me could not be used because it overlaps with the

Si-Me signal. The sum of the concentrations of the three

species equals the sum of the initial concentrations of 5a and

(1,2-Me₂Cp)₂ZrMe₂. The concentration of each species in the

solution was then determined from the total concentration, and

the concentration ratio was deduced from eq 4. Each of the

three samples was then diluted to 1.6 mL, and the above

measurement was repeated.

Stu d y of Solu tion Dyn a m ic Beh a vior of 5a , 5b, a n d 7c

Usin g Va r ia ble Tem p er a tu r e NMR Sp ectr oscop y. The

solution structural dynamic processes involving the title

compounds are all in the slow exchange limit at temperatures

below 55 °C. The rate constant for the exchange process at

each temperature was determined using modified Bloch

eq 5,¹²where ∆W ) W - W_oand W is the line width at half-

Syn th esis of (Me₅Cp )₂Th Me⁺B(C₆F ₄SiⁱP r ₃)₄^-(8b). This

synthesis was carried out following a procedure similar to that

for 5a above, using 3b (412 mg, 0.28 mmol) and (Me₅-

Cp)₂ThMe₂(150 mg, 0.28 mmol). Yield: 382 mg (78%). ¹H

NMR (C₆D₆, 20 °C): δ 0.59 (s, 3H), 1.09 (d, J ) 7.4 Hz, 72H),

1.51 (“p”, J ) 7.4 Hz, 12H), 1.75 (s, 30H), 7.96 (b, 1H). ¹⁹F

NMR (C₆D₆, 20 °C): -128.6 (b), -132.2 (b). ¹³C NMR (C₆D₆,

20 °C): δ 10.7, 12.1, 18.4, 77.1, 109.8 (t, J _C-F) 42 Hz), 129.5,

148.2 (d, J _C-F) 262 Hz), 150.0 (d, J _C-F) 261 Hz). Anal. Calcd

for C₈₁H₁₁₇BF₁₆Si₄Th: C, 55.59; H, 6.74. Found: C, 55.35; H,

6.52.

-

Syn th esis of (Me₅Cp )₂Th Me⁺B(C₆F ₅)₄(8c). This syn-

thesis was carried out following a similar procedure to that

for 5a above, using Ph₃C⁺B(C₆F₅)₄^-(300 mg, 0.325 mmol) and

(Me₅Cp)₂ThMe₂(173 mg, 0.325 mmol) with benzene as the

solvent. Yield: 303 mg (78%). ¹H NMR (C₆D₆, 20 °C): δ 0.34

(s, 3H), 1.54 (s, 12H). ¹⁹F (C₆D₆, 20 °C): -132.6 (b, s, 8F),

-162.11 (t, J _F-F) 21.1 Hz, 4F), -166.0 (m, 8F). ¹³C (CP-

k ) π(∆W)

(5)

height of the signals of interest at each individual temperature,

and W_ois the natural line width at half-height, which was

measured at 20 °C and confirmed by observing no change of

the line width at 10 °C. The same measurements were

undertaken on samples of various concentrations over a 4-fold

range, and no concentration dependence of the processes was

detected.

MAS): δ 10.1 (Me₅Cp), 78.1 (Th-Me). Anal. Calcd for C₄₅H₃₃

BF₂₀Th: C, 45.17; H, 2.78. Found: C, 44.53; H, 2.88.

-

Syn th esis of [(Me₅Cp)₂Th Me]₂(µ-Me)⁺B(C₆F₅)₄^-(9). Com-

pound 8c (103 mg, 0.086 mmol) and (Me₅Cp)₂ThMe₂(69 mg,

130 mmol) were loaded into a 25 mL reaction flask in the

glovebox. The flask was then transferred to the vacuum line

and evacuated, and benzene (10 mL) was condensed into the

flask. The reaction mixture was stirred for 4 h at room

temperature. The resulting slurry was then filtered through

a frit, and the colorless crystalline product was isolated after

drying under vacuum. Yield: 105 mg (71%). ¹H NMR

(toluene-d₈, -27 °C): δ -1.22 (s, 3H), 0.05 (s, 6H), 1.72 (s,

30H). ¹³C NMR (CPMAS, 20 °C): δ 10.9 (Me₅C₅), 56.6

(bridging Me), 76.5 (terminal Me), 127.3 (C₅Me₅). Anal. Calcd

for C₆₇H₆₉BF₂₀Th₂: C, 46.54; H, 4.02. Found: C, 46.31; H, 4.04.

Solu tion P h a se Th er m a l Sta bilities of Com p lexes 4-8.

1

A C₆D₆solution of each complex was monitored by H and ¹⁹F

NMR spectroscopy at 70 and 100 °C. Complexes 4 and 8

decomposed rapidly at 70 °C on a time scale of ∼0.5 h.

Complexes 5, 6, 7a , and 7b do not undergo noticeable

decomposition over the course of 1 h at 100 °C and only give

evidence of decomposition at higher temperatures (130 °C) or

on prolonged heating at 100 °C. Compound 7c is less stable

and has a half-life of ∼0.5 h at 80 °C. These decomposition

processes are complicated, and it was not possible to charac-

terize the products in detail. However, a toluene solution of

5a stored at room temperature for several weeks resulted in

precipitatation of crystallographically characterizable single

-

Syn th esis of [(1,2-Me₂Cp )₂Zr Me]₂(µ-Me)⁺B(C₆F ₄TBS)₄

(10). Compound 5a (200 mg, 0.147 mmol) and a stoichiometric

excess of (1,2-Me₂Cp)₂ZrMe₂(80 mg, 0.260 mmol) were loaded

into a 25 mL reaction flask in the glovebox. The flask was

then connected to the vacuum line and evacuated, and toluene

(10 mL) was condensed into the flask. The reaction mixture

was stirred for 4 h at room temperature. The resulting slurry

was then filtered through a frit, and the colorless crystalline

product was isolated after drying under vacuum. Yield: 103

mg (42%). ¹H NMR (C₆D₄Cl₂, 20 °C): δ -1.47 (s, 3H), -0.23

(s, 6H), 0.13 (s, 24H), 0.72 (s, 36H), 1.61 (s, 12H), 1.80 (s, 12H),

5.38 (t, J ) 2.8 Hz), 5.58 (“t”, 8H). ¹⁹F (C₆D₆, 20 °C): -128.6

(b), -132.2 (b). ¹³C NMR (C₆D₄Cl₂/C₆D₆, 20 °C): δ -6.0, 10.8,

11.1, 15.4, 20.5, 24.1, 39.9, 104.5, 106.2 (t, J _C-F) 48 Hz), 109.3,

112.5, 147.1 (d, J _C-F) 262 Hz), 147.2 (d, J _C-F) 262 Hz). Anal.

Calcd for C₇₉H₁₀₅BF₁₆Si₄Zr₂: C, 55.59; H, 6.74. Found: C,

55.35; H, 6.52.

Equ ilibr a tion betw een 5a , (1,2-Me₂Cp )₂Zr Me₂, a n d 10.

The three title species are in equilibrium in benzene solution,

and the equilibrium constant was measured at 19 °C by ¹H

NMR, using the following procedure. Solutions of 5a (10.0 mg,

0.0073 mmol in 1.00 mL of C₆D₆) and (1,2-Me₂Cp)₂ZrMe₂(6.0

mg, 0.0195 mmol in 2.65 mL of C₆D₆) were prepared in the

glovebox. In the next step, 0.1 mL + 0.1 mL, 0.1 mL + 0.2

mL, and 0.2 mL + 0.1 mL of the above two solutions were

loaded into each of three NMR tubes. Additional C₆D₆was

then added to make the total liquid volume of each sample

0.8 mL. The concentration ratio of each species in the solution

was determined using eq 4, where I_aand I_bare the integated

-

crystals of [(1,2-Me₂Cp)₂ZrF]₂(µ-F)⁺B(C₆F₄TBS)₄(11).

¹³C Scr a m blin g betw een 8a a n d (Me₅Cp )₂Th (¹³CH₃)₂.

A mixture of the title compounds was shaken together in C₆D₆.

After 1 h at room temperature, ¹H NMR spectroscopy revealed

that the ¹³C labels were evenly distributed among the Th-

Me positions.

Eth ylen e P olym er iza tion Exp er im en ts. Ethylene po-

lymerizations were carried out in flamed 250 mL round-bottom

flasks, using 100 mL of toluene as the solvent under 1.0 atm

ethylene pressure and following the procedures described

previously.^4aThese processes are designed to assure rapid

mixing of the monomer with the catalyst solution and to

minimize mass transport effects on the measured rates.

P r op ylen e P olym er iza tion Exp er im en ts. Propylene

polymerizations were carried out in flamed 100 mL round-

bottom flasks, using 40 mL of toluene as the solvent under

1.0 atm propylene pressure and following the procedures

described previously.^2d

-

F or m a tion of (Me₅Cp )₂Th (η³-a llyl)⁺B(C₆F ₅)₄(12).

A

toluene-d₈solution of 8c (∼2 mg) in a NMR tube was prepared

in the glovebox, evacuated at -78 °C on the high-vacuum line,

and exposed to 1.0 atm of propylene at 0 °C. After 15 min at

0 °C, the ¹H NMR spectrum showed that 8c was quantitatively

converted to an allylic complex, with elimination of NMR-

detectable methane. ¹H NMR (toluene-d₈, 25 °C): δ 6.12 (m,

1H), 3.06 (d, J ) 13.8 Hz, 2 H), 2.53 (d, J ) 8.8 Hz, 2H), 1.45

(s, 30H).

X-r a y Cr ysta llogr a p h ic Stu d ies of Com p lexes 8c a n d

11. Crystals of 8c suitable for diffraction studies were grown

by slow diffusion of pentane into a saturated benzene solution

[5a ]:[(1,2-Me₂Cp)₂ZrMe₂]:[10] ) [n(I_a/2 + I_b) - I_b]:I_a/2:I_b

(4)

values of the signals at δ -0.36 and -1.27 ppm, corresponding

to neutral terminal Zr-Me₂and bridging Zr(µ-Me)Zr groups,

respectively, and n is the initial mole ratio of 5a and (1,2-Me₂-

Cp)₂ZrMe₂. The direct integral value of the signal correspond-

(12) (a) Sandstrom, J . Dynamic NMR Spectroscopy; Academic

Press: New York, 1982; pp 77-92. (b) Kaplan, J . I.; Fraenkel, G. NMR

of Chemically Exchanging Systems; Academic Press: New York, 1980;

pp 71-128.

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 847

-

Ta ble 1. Su m m a r y of th e Cr ysta l Str u ctu r e Da ta for (Me₅Cp )₂Th Me⁺B(C₆F ₅)₄(8c) a n d

-

[(1,2-Me₂Cp )₂Zr F ]₂(µ-F )⁺B(C₆F ₄TBS)₃(11)

complex

8c

11

formula

ThC₄₅H₃₃BF₂₀‚(C₆H₆)_2.5

Zr₂Si₄F₁₉C₈₈BH₈₇

triclinic

P1h

15.840(6)

15.979(7)

20.79(1)

crystal system

space group

a, Å

b, Å

c, Å

monoclinic

P2₁/c

18.390(7)

16.444(4)

18.731(6)

R, deg

90

99.20(3)

90

5592(3)

96.56(4)

â, deg

101.22(3)

106.26(3)

4875(4)

γ, deg

V, Å³

Z

4

2

1.23

d(calcd), g/cm³

crystal size, mm

color, habit

diffractometer

temp, °C

1.653

0.4 × 0.4 × 0.4

colorless, transparent

Enraf-Nonius, CAD4

-120

0.05 × 0.03 × 0.06

yellow, transparent

Enraf-Nonius, CAD4

-120

3.37

0.878-0.978

µ, cm^-1

transmission factors range

radiation

28.62

0.411-0.256 (numerical)

graphite monochromator; Mo KR:λ ) 0.710 69 Å

scan type

θ-2θ

ω-θ

2.8-45.9

13 976 (13 532, 0.072)

2θ range, deg

intensities (unique, R_i)

intensities >2.58σ(I)

intensities >3.00σ(I)

no. of params

2-42 (+h,+k,(l)

6609 (5989, 0.015)

4074

6197

607

672

R

0.035

0.037 (r ) 0.010)

1.03

0.09

0.111

1.26

_wfor w ) 1/s²(F_o) + rF_o

2

max density in DF map, e/Å³

of 8c at room temperature, and crystals of 11 suitable for

diffraction studies were retrieved from a decomposed solution

of 5a in toluene. Both were mounted on thin glass fibers, after

protecting with a layer of Paratone oil (Exxon, degassed at

100 °C for 10 h under high vacuum). Data were collected on

an Enraf-Nonius CAD4 diffractometer at -120 °C. Final cell

dimensions were obtained by a least-squares fit to the auto-

matically centered settings for 25 reflections. Three reference

reflections monitored during data collection for each crystal

showed no significant variations. Intensity data were all

corrected for absorption, anomalous dispersion, and Lorentz

and polarization effects. The space group choice for each

complex was unambiguously determined.¹³Crystallographic

data are summarized in Table 1.

The structure of complex 8c was solved by Patterson

methods. The correct thorium atom position was automati-

cally determined from a Patterson map (SHELX-86).¹⁴Sub-

sequent least-squares difference Fourier calculations (SHELX-

76) revealed atomic positions for the remaining non-hydrogen

atoms. The hydrogen atoms were included as fixed contribu-

tors in “idealized” positions. The hydrogen atoms of the

benzene solvent molecules were not included in the structure

refinement. The non-hydrogen atoms were refined with aniso-

tropic thermal coefficients, and common isotropic thermal

parameters were varied for hydrogen atoms. The highest

peaks in the final difference Fourier map were in the vicinity

of the thorium position. A final analysis of the variance

between observed and calculated structure factors showed a

slight dependence on sin θ.

observed reflections (I > 3.00σ(I)) and 607 variable parameters.

All calculations were performed using the TEXSAN crystal-

lographic software package of Molecular Structure Corp.

Resu lts a n d Discu ssion

The goal of this study was to understand the role of

the charge-compensating counteranion in stabilizing

and activating cationic Zr and Th metallocene com-

plexes, the solution ion-pairing properties associated

with this interaction, and, ultimately, to probe for “ideal”

situations in which the ion-paired complexes exhibit

optimal olefin polymerization activity, solubility, and

thermal stability. It will be seen that several broad new

classes of ion-paired, cationic Zr and Th complexes with

a variety of counteranions can be efficiently synthesized.

Comparison of the spectroscopy, solution dynamic prop-

erties, coordinative behavior, and polymerization activ-

ity of these complexes facilitates a considerably deeper

understanding of cation architecture-anion architec-

ture-reactivity relationships.

I. Ca tion ic Ion -P a ir ed Alk yl- a n d Hyd r id ozir -

con iu m an d Alkylth or iu m (Flu or oar yl)bor ate Com -

p lexes. Syn th esis a n d Sp ectr oscop y. A. Syn th e-

-

sis of F u n ction a lized Bor a te An ion s B(C₆F ₄TBS)₄

a n d B(C₆F ₄TIP S)₄. The quest for noncoordinating

anions has been a long standing challenge in inorganic

chemical research.¹⁵Due to the extremely high elec-

trophilicity of the Lewis base-free cationic zirconocene

and thorocene complexes, an effective counteranion

must necessarily be chemically robust and exceptionally

resistant to electrophilic attack, in addition to other

recognized requirements for weakly coordinating an-

ions.¹⁵In fact, only a small number of anions stabilize

The structure of 11 was solved by direct methods and

expanded using Fourier techniques. The carbon and boron

atoms were refined isotropically. The remaining non-hydrogen

atoms were refined anisotropically. The hydrogen atoms were

not included in the structure factor calculations. The final

cycle of full-matrix least-squares refinement was based on 6197

(13) Cromer, D. T.; Waber, J . T. International Tables for X-ray

Crystallogrophy; The Kynoch Press: Birmingham, England, 1974; Vol.

IV, p 149.

(14) (a) Sheldrick, G. M. SHELXS-86. In Crystallographic Comput-

ing; Sheldrick, G. M., Kruger, C., Goddard, R., Eds.; Oxford University

Press: Oxford, 1985; pp 175-189. (b) Sheldrich, G. M. SHELXS-76 A

Program for Crystal Structure Determination; University Chemical

Laboratory: Cambridge, England, 1976.

(15) (a) J elinek, T.; Baldwin, P.; Scheidt, W. R.; Reed, C. A. Inorg.

Chem. 1993, 32, 1982-1990. (b) Strauss, S. H. Chem. Rev. 1993, 93,

927-942. (c) Seppelt, K. Angew. Chem., Int. Ed. Engl. 1993, 32, 1025-

1027. (d) Bochmann, M. Angew. Chem., Int. Ed. Engl. 1992, 31, 1181-

1182.

848 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

Sch em e 1

the present highly electrophilic cations. On the basis

of our earlier success using the (fluoroaryl)borate

MeB(C₆F₅)₃as the counteranion for zirconocenium

or alkide/hydide abstractors (e.g., Ph₃C⁺X^{- 2i}and

B(C₆F₅)₃^4a). We chose the methide/hydride abstracting

triphenylcarbenium salt because the product Ph₃CMe

is soluble in saturated hydrocarbon solvents, hence, it

is readily removed from the metallocenium product.

A series of cationic ion-paired zirconium and thorium

(fluoroaryl)borate complexes (4a ,b, 5a -c, 7a -c, and

8a -c) was prepared in high yield by reaction of the

corresponding neutral dimethylmetal compounds with

the appropriate triphenylcarbenium borate salt in an

aromatic hydrocarbon solvent at room temperature (eq

6). All of these compounds were characterized by

-

ions^4a,band observations that reactions of the trityl or

ammonium salts of B(C₆F₅)₄^-with neutral zirconocene

dialkyls afford highly active olefin polymerization

catalysts,^2d,i,j,4j,lsilyl-functionalized (fluoroaryl)borate

anions 3a and 3b were designed. A high level of aryl

group fluorination was retained to ensure delocaliztion/

neutralization of the negative charge. The sterically

encumbered TBS and TIPS trialkylsilyl functionalities,

which are efficacious protecting groups in organic

synthesis,¹⁶were chosen to sterically screen the inner

sphere of the anion, to enhance solubility, and to further

delocalize the negative charge via the established π-ac-

ceptor properties of silyl arene substituents.¹⁷Syntheti-

cally, nucleophilic substitution reactions on silicon occur

readily and provide a reliable access to the silyl-

functionalized fluoroarene groups. The synthetic routes

to the triphenylcarbenium salts 3a ,b are shown in

Scheme 1. Silyltetrafluorobenzenes 1a ,b are prepared

from the reaction of the corresponding silyltriflates with

LiC₆F₄H generated in situ from BrC₆F₄H in diethyl

ether.¹⁸Arylation of BCl₃with the fluroaryl lithium

reagents formed in situ¹⁹from 1 yields the lithium salts

of borates 2a and 2b, which are then converted by

metathesis to triphenylcarbenium salts 3a ,b. Isolation

of poorly soluble 2b requires a large quantity of diethyl

ether and is not a neccesary step. Alternatively, after

washing with a small amount of ether, the crude

mixture of 2b + LiCl can be used directly in the next

reaction with LiCl removal postponed to the subsequent

step. A single-crystal structure of the anion B(C₆-

RT

LL′MMe₂+ 3a -c

8 LL′MMe⁺X^-+ Ph₃CMe (6)

4a ,b, 5a ,b,

7a -c, 8a -c

C₆H₆

-

3a , Ph₃C⁺B(C₆F₄TBS)₄

;

-

3b, Ph₃C⁺B(C₆F₄TIPS)₄

;

-

3c, Ph₃C⁺B(C₆F₅)₄

-

4a , Cp₂ZrMe⁺B(C₆F₄TBS)₄

;

-

4b, Cp₂ZrMe⁺B(C₆F₄TIPS)₄

;

-

5a , (1,2-Me₂Cp)₂ZrMe⁺B(C₆F₄TBS)₄

5b, (1,2-Me₂Cp)₂ZrMe⁺B(C₆F₄TIPS)₄

;

-

;

-

7a , [(Me₄Cp)SiMe₂(N^tBu)]ZrMe⁺B(C₆F₄TBS)₄

;

-

7b, [(Me₄Cp)SiMe₂(N^tBu)]ZrMe⁺B(C₆F₄TIPS)₄

;

-

7c, [(Me₄Cp)SiMe₂(N^tBu)]ZrMe⁺B(C₆F₅)₄

;

-

8a , (Me₅Cp)₂ThMe⁺B(C₆F₄TBS)₄

8b, (Me₅Cp)₂ThMe⁺B(C₆F₄TIPS)₄

;

-

F₄TBS)₄paired with the cation [(1,2-Me₂Cp)₂ZrF]₂(µ-

-

F)⁺has been obtained and will be discussed later.

B. Zir con iu m a n d Th or iu m Bor a te Com p lexes.

Zr and Th metallocenium borate ion pairs are typically

synthesized using redox reagents (e.g., Ag⁺X^-and

Fc⁺X^-),²⁰protonolytic reagents (e.g., R₃NH⁺X^-),^4m,20,22

;

-

8c, (Me₅Cp)₂ThMe⁺B(C₆F₅)₄

standard spectroscopic methods and analytical tech-

niques (see Experimental Section for data). The anions

B(C₆F₄TBS)₄^-and B(C₆F₄TIPS)₄^-were found to impart

good solubilities to the ion-pairs, a property which

renders characterization straightforward and the study

of the ion-pair structure in solution possible. An

interesting example is complex 7a , which to our sur-

prise, is readily soluble in saturated hydrocarbon sol-

vents such as pentane. This property also unfortunately

renders isolation of 7a relatively difficult and lowers the

(16) (a) Kunz, H.; Waldmann, H. In Comprehensive Organic Syn-

thesis; Trost, B. M., Fleming, I., Winterfelt, E., Eds.; Pergamon Press:

Oxford, 1991; Vol. 6, pp 652-657. (b) Lalonde, M.; Chan, T. H.

Synthesis 1985, 817-845.

(17) (a) Zhang, S. Z.; Zhang, X. M.; Bordwell, F. G. J . Am. Chem.

Soc. 1995, 117, 602-606. (b) Wetzel, D. M.; Brauman, J . I. J . Am.

Chem. Soc. 1988, 110, 8333-8336. (c) Schleyer, P. von R.; Clark, T.;

Kos, A. J .; Spitznagel, G. W.; Rohde, C.; Arad, D.; Houk, K. N.; Rondan,

N. G. J . Am. Chem. Soc. 1984, 106, 6467-6475. (d) Hopkinson, A. C.;

Lien, M. H. J . Org. Chem. 1981, 46, 998-1003.

(18) Tamborski, C.; Soloski, E. J . J . Org. Chem. 1966, 31, 743-745.

(19) (a) Massey, A. G.; Park, A. J . J . Organomet. Chem. 1964, 2,

245-250. (b) Massey, A. G.; Park, A. J . J . Organomet. Chem. 1966, 5,

218-225.

(20) J ordan, R. F.; Bajgur, C. S.; Dasher, W. E.; Rheingold, A. L.

Organometallics 1987, 6, 1041-1051.

(21) (a) Shannon, R. D. Acta Crystallogr., Sect. A 1976, 32, 751-

767. (b) Huheey, J . E. Inorganic Chemistry, 3rd ed.; Harper and Row:

New York, 1983; pp 258-259.

(22) Lin, Z.; Le Marechal, J . F.; Sabat, M.; Marks, T. J . J . Am. Chem.

Soc. 1987, 109, 4127-4129.

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 849

preparative yield, even though eq 6 is determined to be

1

-

quantitative by H NMR. In contrast, the B(C₆F₅)₄

-

derived ion pairs often have poor solubilities. Thus,

solution ¹³C NMR data for complex 8c could not be

obtained due to poor solubility in benzene. Instead,

CPMAS techniques were employed to obtain the solid

state ¹³C NMR spectrum. Compound 7c also has very

limited solubility in benzene, and ¹³C spectra could only

be obtained (for nuclei with attached protons) by two-

dimensional HMQC NMR techniques. Note that iso-

lated solid complex 7c invariably contains 1 equiv of

aromatic solvent judging from NMR integration (con-

firmed by elemental analysisssee Experimental Sec-

tion). The Zr:solvent molecule stoichiometry remains

1:1 even after prolonged pumping under high vacuum.

These results suggest coordination of the arene solvent

molecule to the cation (for additional discussion, see

section II). A characteristic NMR spectroscopic feature

of all of the methyl ion pairs is that the ¹³C resonance

of the M-Me⁺group is shifted considerably downfield

compared to the corresponding neutral precursor M-Me₂

groups. The only exception is the 7c Zr-Me signal,

which is likely influenced by the diamagnetic anisotropy

of the coordinated arene.

Attempts to isolate analogous cationic metallocene

complexes via reaction of 3a or 3b with (Me₅Cp)₂ZrMe₂

and of 3c with Cp₂ZrMe₂, (1,2-Me₂Cp)₂ZrMe₂, or (Me₅-

Cp)₂ZrMe₂result in complicated oily mixtures of un-

identified species (for discussion of the stabilities, see

section IV). However, exposure of these mixtures to

dihydrogen leads to cationic ion-paired hydrides (5a ,

6a -c) in high yields (eq 7). These complexes exhibit

F igu r e 1. Perspective ORTEP drawing of the molecular

structure of complex (Me₅Cp)₂ThMe⁺B(C₆F₅)₄^-(8c). Ther-

mal ellipsoids are drawn at the 50% probability level.

RT, 1 atm H₂

+

L₂MMe₂+ 3a -c

8

^-+ Ph₃CMe

L₂MH X

-CH₄

5c, 6a -c

(7)

Numerous attempts were made during the course of

this study to grow single crystals of complexes 4-8

suitable for X-ray analysis. However, all crystals

obtained suffered from severe twinning and disorder

except for 8c, which has been crystallographically

characterized. The single-crystal diffraction study of 8c

reveals a structure consisting of loosely associated (Me₅-

Cp)₂ThMe⁺cations and tetrahedral B(C₆F₅)₄^-counter-

anions (Figure 1). Important bond distances and bond

angles of 8c derived from the final refinement are

summarized in Table 2. For each molecule of 8c, there

are 2.5 solvent (benzene) molecules present in the unit

cell. However, their locations rule out any direct

interactions with the thorium ion.

-

3a , Ph₃C⁺B(C₆F₄TBS)₄

;

-

3b, Ph₃C⁺B(C₆F₄TIPS)₄

;

-

3c, Ph₃C⁺B(C₆F₅)₄

-

5c, (1,2-Me₂Cp)₂ZrH⁺B(C₆F₅)₄

6a , (Me₅Cp)₂ZrH⁺B(C₆F₄TBS)₄

6b, (Me₅Cp)₂ZrH⁺B(C₆F₄TIPS)₄

;

-

;

-

;

-

6c, (Me₅Cp)₂ZrH⁺B(C₆F₅)₄

straightforward ¹H NMR spectra with one exception.

The reaction product of 3c with Cp₂ZrMe₂yields an

immediate precipitate upon exposure to dihydrogen; the

product is insoluble in aromatic hydrocarbon solvents

It can be seen from Figure 1 that one aryl ring of the

anion approaches the cationic metal center in the

equatorial girdle plane between the two cyclopentadi-

enyl rings, with two fluorine atoms (F18, F19) in closest

proximity to the metal cation (Th-F18 ) 2.757(4) Å,

Th-F19 ) 2.675(5) Å). These Th‚‚‚F distances are

considerably longer than the sums of relevant Th⁴⁺and

1

and only barely soluble in THF. The H NMR spectrum

of the product in THF-d₈is too complex to assign with

confidence. Compounds 6a ,b were characterized by

standard spectroscopic methods and elemental analysis

(see Experimental Section for data). Compound 5c is

insoluble in aromatic solvents, including 1,2-dichlo-

F

^-ionic radii (2.28 Å)²¹and typical Th-O (THF) dative

1

robenzene. The H NMR spectrum of 5c was recorded

bond distances, after correcting for differences in F/O

covalent radii (2.56 Å).²²The Th‚‚‚F distances in 8c are

comparable to the corresponding Zr‚‚‚F(arene) distances

using THF-d₈as the solvent, revealing features similar

to those of an analogous cationic zirconium hydride C

having two molecules of trans-coordinated PMe₃.²⁰The

¹³C NMR spectra of compounds 5c and 6c could not be

obtained due to poor solubility. All of the above cationic

zirconocene hydrides exhibit characteristic downfield-

-

in (Me₅Cp)₂ZrH⁺HB(C₆F₅)₃(D),^4aafter correcting for

differences in Th/Zr ionic radii (2.64 and 2.75 Å).

Furthermore, the C43-F18 and C44-F19 bond lengths

in 8c (1.325(6) and 1.344(7) Å, respectively) are slightly

greater or indistinguishable from the average C-F bond

shifted H NMR signals of the Zr-H⁺group, in agree-

1

-

ment with analogous cationic metallocene hydrides.^4a,20

length in the B(C₆F₅)₄anion (1.32 Å). However, the

850 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

Ta ble 2. Selected Bon d Len gth s (Å) a n d An gles

(d eg) for Com p lex 8c

those found in analogous thorium cationic complexes

containing other counterions and/or Lewis bases (in

(Me₅Cp)₂ThMe(THF)₂⁺, Th-Me ) 2.49(1) Å, Th-

C_ring(av) ) 2.80 Å;²¹in (Me₅Cp)₂ThMe(THF)⁺, Th-Me

) 2.433(7) Å, Th-C_ring(av) ) 2.763 Å;^4cin [(Me₅Cp)₂-

Bond Lengths

Th-C1

Th-C2

Th-C3

Th-C4

Th-C5

Th-C11

Th-C12

Th-C13

Th-C14

Th-C15

Th-C21

Th-F18

Th-F19

B-C27

B-C28

B-C39

B-C42

F1-C22

F2-C23

F3-C24

2.725(8)

2.774(7)

2.766(8)

2.729(8)

2.728(8)

2.791(8)

2.762(8)

2.757(8)

2.761(8)

2.751(8)

2.399(8)

2.757(4)

2.675(5)

1.70(1)

1.68(1)

1.70(1)

1.294(6)

1.332(9)

1.346(9)

F3-C24

F4-C25

F5-C26

F6-C29

F7-C30

F8-C31

F9-C32

F10-C33

F11-C34

F12-C35

F13-C36

F14-C37

F15-C38

F16-C40

F17-C41

F18-C43

F19-C44

F20-C45

1.346(9)

1.318(6)

1.305(9)

1.306(7)

1.326(8)

1.353(6)

1.330(7)

1.305(7)

1.312(7)

1.349(7)

1.350(7)

1.311(8)

1.296(7)

1.319(7)

1.281(7)

1.325(6)

1.344(7)

1.358(7)

ThMe⁺]₂Fe(B₉C₂H_{11 2}^2-, Th-Me ) 2.480(3) Å, Th-

)

C_ring(av) ) 2.817 Å^4e). Such shortening of the Th-Me

and Th-C_ring(av) contacts doubtless reflects increased

electron deficiency and coordinative unsaturation at the

metal center.

II. Ion -P a ir ed Alk yl- a n d Hyd r id ozir con iu m a n d

Alk ylth or iu m (F lu or oa r yl)bor a te Com p lexes. Co-

or d in a tion of An ion s, Neu tr a l Meta l Dia lk yls, a n d

Solven ts. Due to the high Lewis acidity of the cationic

species, weak occupancy of the open coordination site

at the metal center normally occurs and logically affects

the catalytic performance of these highly unsaturated

complexes.^3,4Three types of common coordination will

be discussed: anion coordination, neutral metal alkyl

coordination, and solvent coordination.

An ion Coor d in a tion . In nonpolar hydrocarbon

solvents such as toluene, which are commonly used for

olefin polymerization, the cation-anion interaction is

doubtless an important modulator of reactivity in this

class of ion-paired complexes.⁴Hence, assessing the

coordinative ability of the present noncoordinating

anions in this type of cationic, unsaturated environment

Bond Angles

Th-C1-C6

Th-C2-C7

Th-C3-C8

Th-C4-C9

Th-C5-C10

Th-C11-C16

Th-C12-C17

Th-C13-C18

119.3(5)

119.0(6)

121.7(5)

124.3(5)

116.8(5)

124.1(6)

119.5(6)

120.5(5)

Th-C14-C19

Th-C15-C20

C27-B-C28

C27-B-C42

C27-B-C39

C28-B-C42

C28-B-C39

C39-B-C42

123.4(5)

120.7(6)

115.1(6)

113.4(6)

100.9(6)

100.0(6)

113.7(6)

114.4(6)

is of great interest. Evaluation of the relative coordi-

Cp_{ring-centroid}-Th-Cp_{ring-centroid}

140.1(3)

-

native ability of the anions B(C₆F₄TBS)₄^-, B(C₆F₄TIPS)₄

,

B(C₆F₅)₄^-, and Me(C₆F₅)₃^-was probed by measuring the

energetics of dynamic solution structural reorganization

processes by NMR.

corresponding C-F bond lengths in D (1.379(6) and

1.396(5) Å) are conspicuously greater than the average

C-F bond distance (1.35 Å) in the HB(C₆F₅)₃^-anion of

D. All other C-F distances in 8c and D are unexcep-

tional when compared to a typical C(sp²)-F σ-bond in

other organofluorine molecules.²³These data clearly

show that despite the strong Lewis acidity of the Th

cation, the cation-anion coordinative interaction is

rather weak, in agreement with NMR observations in

solution (vide infra). In addition, the cation-anion

interaction in 8c appears to be somewhat weaker than

that in D, judging from the fact that the anion structure

is less perturbed (also in agreement with solution NMR

results^4a).

A unique molecular dynamic process observable in

(1,2-Me₂Cp)₂ZrMe⁺X^-complexes is ion pair reorganiza-

tion/symmetrization of the disymmetric ground state

structure indicated by permutation of the diastereotopic

ring methyl groups (eq 8). It is physically reasonable

that the activation energetic parameters of such a

process measure the barriers to essentially complete (or

at least to the same degree) ion-pair dissociation (pre-

sumably within a solvent cage). Studies in this labora-

tory previously quantified these parameters by dynamic

- 4b,f

NMR techniques for (1,2-Me₂Cp)₂ZrMe⁺MeB(C₆F₅)₃

.

The same measurement was performed in the present

study on complexes 5a ,b. The rate constants for this

process were determined by line shape analysis of the

1

variable-temperature H NMR spectra (Figure 2) using

the appropriate modified Bloch equation.¹²The activa-

tion enthalpies and entropies (Table 3) were derived

from a least-squares Eyring analysis (Figure 3) accord-

ing to eq 9, where k_Bis the Boltzmann constant, and h

The (Me₅Cp)₂ThMe⁺cation of 8c adapts the usual

bent-sandwich configuration, and the ring centroid-

Th-ring centroid angle (149.1(3)°) is not unusual for

complexes containing (Me₅Cp)₂M (M ) lanthanide,

actinide, or group 4 metals) moities.^4ac,24However, the

Th-Me and Th-C_ring(av) distances in 8c (2.399(8) and

2.754 Å, respectively) are significantly shorter than

ln(k/T) ) -(∆H^q/R)(1/T) + [∆S^q/R + ln(k_B/h)] (9)

is Planck’s constant. The activation enthalpies for (1,2-

Me₂Cp)₂ZrMe⁺B(C₆F₄TBS)₄^-(5a) and (1,2-Me₂Cp)₂ZrMe⁺

B(C₆F₄TIPS)₄^-(5b) ion-pair reorganization are signifi-

cantly lower than that for (1,2-Me₂Cp)₂ZrMe⁺MeB-

(C₆F₅)₃^-, arguing that the strength of the anion coor-

dination is less than that in (1,2-Me₂Cp)₂ZrMe⁺-

(23) Trotter, J . In The Chemistry of the Carbon-Halogen Bond; Patai,

S., Ed.; Wiley: London, 1973; Chapter 2.

(24) (a) J eske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks,

T. J . J . Am. Chem. Soc. 1985, 107, 8111-8118. (b) Schumann, H.;

Meese-Marktscheffel, J . A.; Esser, L. Chem. Rev. 1995, 95, 865-986.

(c) Schaverien, C. J . Adv. Organomet. Chem. 1994, 36, 283-362.

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 851

Ta ble 3. Activa tion En er getic P a r a m eter s for th e Ion P a ir Reor ga n iza tion P r ocess in (1,2-Me₂Cp )₂Zr Me⁺X^-

Com p lexes

T_c(coalescence

temperature), °C

X^-

∆H^q, kcal/mol

∆S^q, eu

∆G_coal^q, kcal/mol^a

ref

-

MeB(C₆F₅)₃

24(1)

18.6(8)

19.2(4)

17(2)

11(1)

10(1)

18.3(2)

15(1)

15.8(7)

80(1)

56(1)

65(1)

4b

-

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

this work

-

a

q

∆G_coalat the coalecence temperature.

F igu r e 3. Eyring plot for solution dynamic ion pair

reorganization processes in complexes 5a and 5b.

tightly with (1,2-Me₂Cp)₂ZrMe⁺than do B(C₆F₄-

-

TBS)₄and B(C₆F₄TIPS)₄^-. This comparison is likely

to be valid for the ion-pairing in complexes of the other

cations discussed in this work, as will be argued below.

The ¹⁹F NMR chemical shift positions and line shapes

within the fluoroaryl groups are affected little by the

coordination in the MeB(C₆F₅)₃^--derived ion-paired L₂-

ZrMe⁺complexes since the cation-anion interaction is

largely through a Zr⁺-Me-B^-contact (E).^4aHowever,

-

interactions between the cations and B(C₆F₄TBS)₄

,

B(C₆F₄TIPS)₄^-, and B(C₆F₅)₄^-significantly influence the

borate ¹⁹F NMR spectra, which is presumably a conse-

quence of direct metal-fluorine interactions. Compari-

son of the ¹⁹F NMR spectra of ion-paired (fluoroaryl)-

borate anions to those of the corresponding triphen-

ylcarbenium borates, in which the carbocations adopt

a planar trigonal geometry and do not interact signifi-

cantly with the anion,²⁵provides useful information

about the cation-anion interaction. In the (Me₅-

Cp)₂ThMe⁺system, the fact that the ¹⁹F NMR spectrum

of 8c shows no substantial chemical shift differences

F igu r e 2. (a) Variable temperature ¹H NMR spectra of

complex 5a in toluene-d₈solution. (b) Variable temperature

¹H NMR spectra of complex 5b in toluene-d₈solution. The

¹H signal line broadening of the two diastereotopic cyclo-

pentadienyl methyl groups (H_dand H_e, δ 1.61 and 1.37 ppm

in 5a , δ 1.73 and 1.43 ppm in 5b) is measured as a function

of temperature. These spectra also illustrate the high

thermal stability of these complexes.

-

from that of Ph₃C⁺B(C₆F₅)₄at room temperature

-

suggests that coordination of anion B(C₆F₅)₄to Th is

weak and labile. In fact, only minor line broadening of

the three ¹⁹F signals in 8c is observed at temperatures

as low as -80 °C. In contrast, the interaction of anions

-

MeB(C₆F₅)₃(Table 3). The activation entropies for

5a ,b are also somewhat lower (Table 3), possibly reflect-

ing less ordered ground states in the more loosely

associated ion pairs of 5a ,b. On the basis of these

(25) (a) Nenitzescu, C. D. In Carbonium Ions; Olah, G. A., Schleyer,

P. von R., Eds.; Wiley: New York, 1968, Vol. 1, pp 2-7. (b) For a crystal

structure of a trityl salt with the counteranion perfluorobis(4,4′-

biphenyl)borate, see: Ishihara, A.; Marks, T. J . Manuscript in prepara-

tion.

-

results, it is clear that MeB(C₆F₅)₃associates more

852 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

Ta ble 4. Selected Zr -CH₃¹³C NMR Ch em ica l Sh ift

-

B(C₆F₄TBS)₄and B(C₆F₄TIPS)₄with Th appears to

be relatively strong and causes severe line broadening

of the ¹⁹F signals at room temperature, due to relatively

slow permutation of the fluoroaryl groups. In the (Me₅-

Da ta for Zir con iu m a n d Th or iu m Meta llocen es

complex

δ, ppm

ref

(1,2-Me₂Cp)₂ZrMe₂

38.4^a

42.3^a

44.6^a

45.9^a

47.0^a

68.5^a

71.8^a

77.0^a

77.4^a

78^b

9

Cp)₂ZrH⁺system, ¹⁹F line broadening of all three

(1,2-Me₂Cp)₂ZrMe(THF)⁺B(C₆F₄TBS)₄

this work

4a

this work

6

-

(1,2-Me₂Cp)₂ZrMe⁺MeB(C₆F₅)₃

-

anions, B(C₆F₄TBS)₄^-, B(C₆F₄TIPS)₄^-, and B(C₆F₅)₄

,

-

(1,2-Me₂Cp)₂ZrMe⁺B(C₆F₄TBS)₄

-

is observed at room temperature; however, B(C₆F₄TBS)₄

and B(C₆F₄TIPS)₄are affected more than B(C₆F₅)₄

(1,2-Me₂Cp)₂ZrMe⁺B(C₆F₄TIPS)₄

-

(Me₅Cp)₂ThMe₂

(Me₅Cp)₂ThMe⁺B(C₆H₅)₄

22

-

judging from the absolute line widths at half-height of

the signals corresponding to ortho and meta fluorines.

The anion coordination is also strong in [(Me₄Cp)-

(Me₅Cp)₂ThMe⁺B(C₆F₄TBS)₄

this work

4c

-

(Me₅Cp)₂ThMe⁺B(C₆F₄TIPS)₄

-

(Me₅Cp)₂ThMe⁺B(C₆F₅)₄

-

SiMe₂(N^tBu)]ZrMe⁺B(C₆F₄TBS)₄(7a ) and [(Me₄Cp)-

(Me₅Cp)₂ThMe⁺{[(C₆F₅)₂B]₂CH^tBu}(µ-H)^-

80^b

-

SiMe₂(N^tBu)]ZrMe⁺B(C₆F₄TIPS)₄(7b), judging from

a

b

In C₆D₆, 20 °C. CPMAS solid state NMR, 20 °C.

the room temperature ¹⁹F spectrum of 7a which exhibits

a very broad envelope and that of 7b which exhibits two

sharp peaks on top of a broad envelope. In sharp

contrast, the cation-anion interaction in [(Me₄Cp)-

SiMe₂(N^tBu)]ZrMe(C₆H₆)⁺B(C₆F₅)₄^-(7c) is replaced by

coordination of an aromatic solvent molecule (vide infra)

and the ¹⁹F NMR spectrum is identical to that of the

free anion even at -17 °C (very dilute solution; spectra

at still lower temperatures could not be obtained due

to solubility limitations), indicating the absence of

significant cation-anion interactions. Interestingly, the

¹⁹F spectra of the ion pairs 5a ,b, 6a ,b, 7a ,b, and 8a

proceed through a similar series of complicated signal

broadening and sharpening on lowering the tempera-

ture in toluene-d₈from room temperature to -90 °C.

In all cases, the slow exchange limit spectra exhibit

similar complex patterns with 12 signals from δ -115

to -155 ppm, having nonequivalent intensities, sug-

gesting similar contact ion-pairing modes in the static

structures. In addition, complex 7a is soluble in pen-

tane, suggesting a tightly associated ion pair, and ¹H

NMR signal line broadening corresponding to the hy-

pothetical dynamic process of eq 10 is not observed in

toluene-d₈at temperatures as high as 120 °C/400 MHz.

cationic species results in µ-Me-bridged bimetallic com-

plexes (eq 11). These compounds are usually much less

soluble than the corresponding neutral and ion-paired

compounds, a property that can be utilized in their

syntheses. Compound 9 is isolated as a crystalline

-

precipitate when (Me₅Cp)₂ThMe⁺B(C₆F₅)₄(8c) and

(Me₅Cp)₂ThMe₂are mixed in benzene. The zirconocene

analogue of 9 was proposed by Bochmann et al. on the

basis of NMR studies but was not isolated in a pure

state.²⁶We also find that reaction of (1,2-Me₂Cp)₂-

-

ZrMe⁺B(C₆F₄TBS)₄(5a ) with a stoichiometric excess

of (1,2-Me₂Cp)₂ZrMe₂affords crystalline complex 10 in

benzene. Structurally diagnostic spectral characteris-

tics of 9 and 10 are the large ¹³C-¹H coupling constants

of the µ-Me functionality (130.1 and 132.8 Hz, res-

pectively).^26-28In both cases, the ¹⁹F NMR spectra of

the bimetallic compounds are identical to those of the

corresponding trityl salt, indicating a weak cation-

anion interaction in solution. Unlike 9, which is stable

in benzene, redissolving 10 in benzene-d₆leads instan-

taneously to partial dissociation of the bimetallic spe-

cies, and precursors 5a and (1,2-Me₂Cp)ZrMe₂are

detected coexisting with 9. The equilibrium constant

for eq 11 is determined to be 6.3(3) × 10²L/mol at 19

°C over an 8-fold concentration range, corresponding to

∆G ) -3.7(1) kcal/mol. When 1,2-dichlorobenzene-d₄

is used as the solvent, the equilibrium is shifted

dramatically to the right, rendering the species on the

left side of the equation undetectable. The shift of the

equilibrium is apparently due to the ability of the more

polar 1,2-dichlorobenzene-d₄to stabilize the more charge-

separated structure. In addition, the bridging and

terminal methyl groups of complex 9 are in rapid

Taken together, the spectroscopic information argues

that B(C₆F₄TBS)₄^-and B(C₆F₄TIPS)₄^-coordinate more

strongly than B(C₆F₅)₄^-, although it does not differenti-

ate among the steric and dative aspects of the interac-

tions.

On the basis of these observations, the approximate

relative coordinating abilities of the (fluoroaryl)borate

anions with respect to the present metallocene cations

-

can be ranked as MeB(C₆F₅)₃> B(C₆F₄TBS)₄

≈

B(C₆F₄TIPS)₄^-> B(C₆F₅)₄^-. Interestingly, a correlation

is also seen between this ranking and the ¹³C chemical

shifts of the M-Me⁺groups. That is, the weaker the

anionic donor, the further downfield the signal is

displaced (Table 4), presumably also, among other

factors, reflecting the progression in the electron-diffi-

ciency/cationic character of the metal centers. Solvent

coordination and neutral metal alkyl coordination will

be discussed below. Both of these phenomena offer

further supporting information for evaluating relative

anions coordinative abilities.

1

exchange on the H NMR time scale at room tempera-

ture, but can be resolved below -27 °C. However, the

corresponding groups are in slow exchange in complex

(26) (a) Bochmann, M.; Lancaster, S. J . Angew. Chem., Int. Ed. Engl.

1994, 33, 1634-1637. (b) See also: Turner, H. W.; Hlatky, G. G.;

Eckman, R. R. U.S. Patent 5,470,927, 1995.

(27) (a) Stern, D.; Sabat, M.; Marks, T. J . J . Am. Chem. Soc. 1990,

112, 9558-9575. (b) Evans, W. J .; Chamberlain, L. R.; Ulibarri, T. A.;

Ziller, J . W. J . Am. Chem. Soc. 1988, 110, 6423-6432. (c) Busch, M.

A.; Harlow, R.; Watson, P. L. Inorg. Chim. Acta 1987, 140, 15-20. (d)

Holton, J .; Lappert, M. F.; Pearce, R.; Yarrow, P. I. W. Chem. Rev.

1983, 83, 135-201.

(28) Ozawa, F.; Park, J . W.; Mackenzie, P. B.; Schaefer, W. P.;

Henling, L. M.; Grubbs, R. H. J . Am. Chem. Soc. 1989, 111, 1319-

1327.

Neu tr a l Meta llocen e Alk yl Coor d in a tion . Coor-

dination of neutral metal alkyl compounds to the

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 853

10 at room temperature. This difference presumably

reflects the larger ionic radius of Th⁴⁺, which renders

the transition state for methyl group permutation

sterically more accessible.

interaction. The ¹H spectrum of 7c in toluene-d₈

indicates nonlabile occupancy of the coordination site

on Zr adjacent to the methyl group, as significant line

broadening of the diastereotopic Cp and silyl methyl

groups is not observed at temperatures below 80 °C. At

temperatures above 80 °C, broadening of the diaste-

reotopic methyl signals is observed, presumably result-

ing from dynamic symmetrization processes (e.g., eq 12,

possibly assisted by other solvent molecules). At still

When the (fluoroaryl)borate counteranions are varied,

a drastically different behavior than that observed

between 5a or 8c and the corresponding neutral met-

allocene dialkyls is observed. In contrast to 8c,

-

(Me₅Cp)₂ThMe⁺B(C₆F₄TBS)₄(8a ) and (Me₅Cp)₂Th-

Me⁺B(C₆F₄TIPS)₄^-(8b) do not form H NMR-detectable

1

bimetallic species with (Me₅Cp)₅ThMe₂in benzene.

However, when 8a and 8b are mixed with (Me₅Cp)₅Th-

(¹³CH₃)₂in C₆D₆, rapid (within 1 h) ¹³C isotopic scram-

bling is observed between cationic and neutral Th

species, suggesting that eq 9 is operative but lies further

to the left in both cases. Unlike 5a , (1,2-Me₂Cp)₂ZrMe⁺-

-

MeB(C₆F₅)₃does not form a detectable bimetallic

species with (1,2-Me₂Cp)₂ZrMe₂. Although similar iso-

topic scrambling is observed upon mixing (1,2-Me₂-

Cp)₂Zr(¹³CH₃)₂and (1,2-Me₂Cp)₂ZrMe⁺MeB(C₆F₅)₃^-, a

different mechanism involving methide abstraction to

form the neutral borane may be operative in this case^4a,b

and is beyond the scope of this discussion. These

results, as well as those in eq 11, can be explained in

terms of competition between the (fluoroaryl)borate

anions and the neutral metal alkyls for the metallocene

cations. For (1,2-Me₂Cp)₂ZrMe⁺paired with counter-

anion MeB(C₆F₅)₃^-, anion coordination is stronger than

binding of (1,2-Me₂Cp)₂ZrMe₂and a µ-Me bimetallic

higher temperatures, irreversible decomposition occurs,

rendering it difficult to accurately determine the rate

of the dynamic process. The rate can be roughly

estimated as ∼62 s^-1at the coalescence temperature

(94 °C)¹²of one set of diastereotopic cyclopentadienyl

methyl groups, corresponding to ∆G^q≈ 24.9 kcal/mol.

1

In addition, the Zr⁺-Me group H NMR signal (δ -0.45

ppm) and ¹³C signal (δ 32.2 ppm) in [(Me₄Cp)Si-

Me₂(N^tBu)]₂ZrMe(C₆H₆)⁺B(C₆F₅)₄^-(7c) are at a surpris-

ingly high field compared to the other cationic group 4

and Th complexes (Table 4), all of which have low field

M⁺-Me resonances. Especially relevent are analogous

complexes 7a (¹H, δ 0.21 ppm; ¹³C, δ 56.9 ppm) and 7b

(¹H, δ 0.23 ppm; ¹³C, δ 57.2 ppm). Such a dramatic

chemical shift difference is reasonably ascribed to the

diamagnetic anisotropy of the coordinated arene ring.

Th er m a l Sta bility. Good stability at high temper-

atures (e.g., g100 °C) is a crucial prerequisite for the

application of metallocene complexes in industrial scale

olefin polymerization technologies. We previously stud-

ied the thermal decomposition of MeB(C₆F₅)₃^--based

cationic metallocene complexes and identified four major

thermal decomposition pathways in toluene-d₈(cation/

anion ligand redistribution, intramolecular C-H activa-

tion, fluoride abstraction, and intermolecular C-H

activation).^4aThe thermal stability of the ion-paired

complexes was found to be a sensitive function of the

cyclopentadienyl ancillary ligand subtituents. In the

present investigation, the roles of both the ancillary

ligand and the counteranion in stabilizing the cations

were studied. The zirconocene complexes with unsub-

stituted cyclopentadienyl ligands (4a ,b) and the (Me₅-

Cp)₂Th complexes (8a -c) exhibit moderate thermal

stability at room temperature and decompose rapidly

at 70-80 °C in benzene. In the case of 4a , the major

thermolysis pathway is aryl group redistribution be-

tween cation and anion (eq 13), resulting in a neutral

metallocene compound, which can be crystallized from

the reaction mixture at -78 °C .³⁰The other decom-

-

complex is not detected. However, when B(C₆F₄TBS)₄

is the counteranion, the coordinating strength of the

anion and the binding of (1,2-Me₂Cp)₂ZrMe₂are com-

parable and 9 is in equilibrium with 5a and (1,2-Me₂-

-

Cp)₂ZrMe₂. For (Me₅Cp)₂ThMe⁺paired with B(C₆F₅)₄

,

anion coordination is much weaker than that of (Me₅-

Cp)₂ThMe₂and sizable concentrations of the bimetallic

species are present in benzene solution. In contrast,

-

when B(C₆F₄TBS)₄is the counteranion, anion coordi-

nation is stronger than that of (Me₅Cp)₂ThMe₂and

detectable quantities of a µ-Me bimetallic complex are

not formed. These observations demonstrate again that

relative coordination ability follows the order

-

MeB(C₆F₅)₃> B(C₆F₄TBS)₄≈ B(C₆F₄TIPS)₄

>

B(C₆F₅)₄^-. NMR scale reactions of either 7a or 7c with

the corresponding neutral precursors gave no evidence

of dimeric µ-Me species.

Solven t Coor d in a tion . Although significant coor-

dination of arene ligands to cationic metallocene species

has not been observed, it is an important interaction in

complexes such as (Me₅Cp)ZrMe₂⁺MeB(C₆F₅)₃^-, which

exists as a toluene adduct both in solution and in the

solid state.²⁹In the present study, arene coordination

to the constrained geometry Zr cation [(Me₄Cp)SiMe₂(N^t-

Bu)]ZrMe⁺is observed when the counteranion is

B(C₆F₅)₄^-, as noted in section II. In addition to the

analytical evidence already presented, other spectro-

scopic evidence also supports the coordination of arenes

to this type of ion pair. The ¹⁹F spectrum of 7c in

toluene-d₈displays three sharp peaks in the fluoroaro-

Cp₂ZrMe⁺B(C₆F₄TBS)₄^{70-80 °C}8

-

toluene

Cp₂ZrMe(C₆F₄TBS) + B(C₆F₄TBS)₃(13)

matic region with coupling constants and chemical shifts

-

identical to those of the free anion in Ph₃C⁺B(C₆F₄)₄

,

position product in this case is presumably a neutral

consistent with the absence of significant cation-anion

(29) (a) Gillis, D. J .; Tudoret, M.-J .; Baird, M. C. J . Am. Chem. Soc.

1993, 115, 2543-2545. (b) Lancaster, S. J .; Robinson, O. B.; Bochmann,

M.; Coles, S. J .; Hursthouse, M. B. Organometallics 1995, 14, 2456-

2462.

(30) NMR spectroscopic characterization: ¹H NMR (C₆D₆, 25 °C):

δ 0.19 (t, J _H-F) 4.2 Hz, 3H, Zr-Me), 0.35 (s, 6H, Si-Me), 0.91 (s, 9H,

Si-CMe₃), 5.85 (s, 10H, Cp). ¹⁹F NMR (C₆D₆, 60 °C): δ -128.2 (m, 2F,

C₆F₄), -132.6 (m, 2F, C₆F₄).

854 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

borane , which was not characterized in detail. Such

aryl abstraction processes from borate anions by cationic

Zr species have been previously reported by serveral

groups.^4a,31The other B(C₆F₄TBS)₄^--, B(C₆F₄TIPS)₄^--,

and B(C₆F₅)₄^--based Zr complexes (5, 6, and 7a -c)

-

exhibit greater thermal stability than the MeB(C₆F₅)₃

-

based analogues and are generally stable in toluene at

100-110 °C for at least 1 h without significant decom-

position. However, complex 7c decomposes rapidly at

100 °C in toluene. After prolonged heating, the decom-

position products in this case are complex mixtures that

could not be unambiguously characterized.

It is interesting that the cationic methylzirconocenes

having B(C₆F₅)₄^-as the charge-compensating anion are

not isolable at room temperature. In contrast, hydrides

-

(Me₅Cp)₂ZrH⁺B(C₆F₅)₄(5c) and (1,2-Me₂Cp)₂ZrH⁺B-

(C₆F₅)₄^-(6c), prepared by exposing the mixtures derived

from attempted syntheses of the cationic methyl com-

plexes to H₂, exhibit good thermal stability at 100 °C.

One reasonable explanation is that C-H activation

processes involving the cyclopentadienyl ligand and

Zr⁺-Me fragment intervene to afford η⁵,η¹complexes

and that the (fluoroaryl)borate anion is not directly

involved. Such a process appears more likely if the

cation is extremely electron-deficient and may, in

principle, be avoidable if the cation is less unsaturated.

In other words, a certain degree of weak coordination

is required to prevent the cation from undergoing self-

decomposition. In accordance with this hypothesis, the

-

MeB(C₆F₅)₃anion, which is the most coordinating

anion identified in this work, stabilizes all four of the

methylzirconium cations discussed here. However, (1,3-

^tBu₂Cp)₂ZrMe⁺, in which the highly congested ligand

F igu r e 4. Perspective ORTEP drawing of the molecular

-

environment may weaken MeB(C₆F₅)₃coordination

-

structure of complex [(1,2-Me₂Cp)₂ZrF]₂(µ-F)⁺B(C₆F₄TBS)₃

and in which a proximate C-H functionality is present,

(11). Thermal ellipsoids are drawn at the 35% probability

level: (a) cation, (b) anion.

t

undergoes clean BuC-H metallation to afford an η⁵,η¹

“tuck-in” cation (eq 14), which is inert with respect to

product),^4athe methyl groups connected to Zr are

completely replaced by F in this case. The crystal

structure of 11 consists of separated, discrete [(1,2-Me₂-

Cp)₂ZrF]₂(µ-F)⁺cations and B(C₆F₄TBS)₄^-anions (Fig-

ure 4). Important bond distances and bond angles of

11 are summarized in Table 5. Unlike previously

characterized [(1,2-Me₂Cp)₂ZrMe]₂(µ-F)⁺, which has a

linear Zr-F-Zr configuration, the present Zr1-F1-Zr2

angle is 159.0(6)°. The Zr-(µ-F) distances in 11 are

approximately the same as those in [(1,2-Me₂Cp)₂ZrMe]₂-

(µ-F)⁺(Zr1-F1 ) 2.11(1) Å and Zr2-F1 ) 2.11(1) Å vs

2.108(2) and 2.118(2) Å in the latter). The present

terminal Zr-F bond lengths (Zr1-F3 ) 1.92(1) Å and

Zr2-F2 ) 1.94(1) Å) are shorter than those in Cp₂ZrF₂

(1.98(1) Å),³⁰presumably reflecting the electron-defi-

cient cationic character of 11. The two (1,2-Me₂Cp)₂ZrF

fragments in 11 are crystallographically nearly identical

(e.g., Zr1-C_ring(av) ) 2.51 Å and Zr2-C_ring(av) ) 2.49

Å). The large borate anion adopts a normal tetrahedral

geometry, and within the accuracy of the present

determination, the average B-C(aryl) distance in 11 is

comparable to or slightly shorter than the corresponding

average bond distance in 8c. The C-F distances in the

four silyl-functionalized fluoroaryl ligand are unexcep-

tional C(sp²)-F distances.²³

ethylene polymerization or oligomerization.^4aIn com-

-

parison, the anions B(C₆F₄TBS)₄and B(C₆F₄TIPS)₄

stabilize three of the four methylzirconium cations but

do not stabilize the sterically most hindered (Me₅-

Cp)₂ZrMe⁺. Intramolecular C-H activation and other

decomposition processes are also likely to be detrimental

to catalytic activity (the polymerization activity of

B(C₆F₅)₄^--derived catalysts erodes significantly upon

storage at room temperature, vide infra) since the

constrained geometries of the η⁵,η¹compounds may not

support rapid multiple CdC insertions.^4a

Upon prolonged standing at 25 °C over the course of

several weeks, a solution of 5a in toluene decomposed

to yield single crystals of [(1,2-Me₂Cp)₂ZrF]₂(µ-F)⁺B-

-

(C₆F₄TBS)₄(11), which were characterized by X-ray

diffraction. Unlike previously reported [(1,2-Me₂-

-

Cp)₂ZrMe]₂(µ-F)⁺MeB(C₆F₅)₃(also a decomposition

In summary, the identified metallocenium fluoro-

borate decomposition pathways can be divided into two

categories: (i) decompositions involving the anions, e.g.,

the aryl transfer reactions such as eq 13 and fluoride

(31) (a) Horton, A. D.; Orpen, A. G. Organometallics 1991, 10, 3910-

3918. (b) Bochmann, M.; J aggar, A. J .; Nicholls, J . C. Angew. Chem.,

Int. Ed. Engl. 1990, 29, 780-781. (c) Lin, Z. Ph.D. Dissertation,

Northwestern University, Evanston, IL, 1988.

(32) Bush, M. A.; Sim, G. A. J . Chem. Soc. A 1971, 2225-2229.

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 855

Ta ble 5. Selected Bon d Len gth s (Å) a n d An gles

(d eg) for 11

anion effects on polymerization activity. The pure,

isolated organometallic salts were employed in all of the

experiments except those of entries 15-17, in which the

unstable B(C₆F₅)₄^--derived catalysts were generated in

situ by mixing the neutral dimethylzirconocenes with

Bond Lengths

Zr1-F1

Zr1-F3

Zr2-F1

Zr2-F2

Zr1-C15

Zr1-C16

Zr1-C17

Zr1-C18

Zr1-C19

Zr1-C22

Zr1-C23

Zr1-C24

Zr1-C25

Zr1-C26

Zr2-C1

Zr2-C2

Zr2-C3

Zr2-C4

Zr2-C5

Zr2-C8

Zr2-C9

Zr2-C10

Zr2-C11

Zr1-C12

2.11(1)

1.92(1)

2.11(1)

1.94(1)

2.48(2)

2.55(2)

2.51(2)

2.46(2)

2.49(2)

2.53(2)

2.51(2)

2.50(2)

2.48(2)

2.44(2)

2.56(2)

2.54(2)

2.48(2)

2.51(2)

2.50(2)

2.55(2)

2.49(2)

2.47(2)

2.54(2)

F4-C30

F5-C31

F6-C33

F7-C34

F8-C42

F9-C44

F10-C45

F11-C46

F12-C54

F13-C55

F14-C57

F15-C58

F16-C66

F17-C67

F18-C69

F19-C70

B-C29

1.37(2)

1.33(2)

1.36(2)

1.34(2)

1.37(2)

1.38(2)

1.36(2)

1.37(2)

1.35(2)

1.33(2)

1.34(2)

1.35(2)

1.66(3)

1.66(2)

1.67(3)

1.66(3)

1.91(2)

1.89(2)

1.90(2)

-

Ph₃C⁺B(C₆F₅)₄in a small volume of toluene (∼2 mL).

-

The freshly generated B(C₆F₅)₄catalysts exhibit high

activity; however, aging significantly erodes the perfor-

mance (entry 20, Table 6), in agreement with the

thermal stability studies (vide supra). In comparison

to the previously reported MeB(C₆F₅)₃^--derived zir-

conocene catalysts,^4athe corresponding B(C₆F₅)₄^--,

B(C₆F₄TBS)₄^--, and B(C₆F₄TIPS)₄^--derived zirconocene

catalysts exhibit significantly higher ethylene polym-

erization activities under identical reaction conditions.

Although the exact quantity of active species could not

-

be determined for the zirconocene + Ph₃C⁺B(C₆F₅)₄

-

B-C44

based catalysts (due to the thermal instability), the

corresponding B(C₆F₅)₄^--, B(C₆F₄TBS)₄^--, and B(C₆F₄-

TIPS)₄^--derived zirconocene catalysts have similar over-

all activities and the differences among them are

generally within experimental error (entries 5-10 and

15-17, Table 6). However, comparison of the activity

of the thorocene (8a -c) and the constrained geometry

(7a -c) catalyst classes, both of which are formally more

coordinatively unsaturated than the zirconocenes, re-

veals that the freshly prepared B(C₆F₅)₄^--derived cata-

lysts exhibit comparable or only slightly greater reac-

B-C53

B-C65

Si1-C32

Si2-C44

Si3-C56

Si4-C68

Bond Angles

F1-Zr2-F2

F1-Zr1-F3

Zr1-F1-Zr2

C29-B-C53

C41-B-C53

119.3(5)

119.0(6)

158.0(6)

100(1)

C65-B-C53

C29-B-C41

C29-B-C65

C41-B-C65

113(1)

116(1)

112(1)

102(1)

112(1)

-

tivity than the analogous B(C₆F₄TBS)₄and B(C₆-

Cp_{ring-centroid}-Zr1-Cp_{ring-centroid}

Cp_{ring-centroid}-Zr2-Cp_{ring-centroid}

131(1)

-

F₄TIPS)₄catalysts (entries 11-14, 18, and 19, Table

6). Note that the MeB(C₆F₅)₃^--derived constrained

geometry catalyst is almost completely inactive for

ethylene polymerization (entry 4, Table 6) at room

temperature. The activity differences among the cata-

lysts having identical cationic fragments and variable

counterions doubtless reflects the relative coordinative

abilities of the anions and tightness of the ion-pairing.

In addition to the “anion effects” on the ethylene

polymerization activities, the data in Table 6 also reveal

noteworthy properties of the cations. Note that the

abstraction and (ii) cation self-decomposition, e.g., C-H

activation reactions such as eq 14.

III. Ca ta lytic Olefin P olym er iza tion . The results

of ethylene polymerization experiments, conducted in

a manner so as to minimize mass transport effects,^4a

are shown in Table 6. The results with zirconocene

catalysts are first discussed, then those with thorocene

and constrained geometry catalysts. The data will be

analyzed with a view toward differentiating cation and

Ta ble 6. Eth ylen e P olym er iza tion P er for m a n ce of Ion -P a ir ed Ca tion ic Gr ou p 4 a n d Th or iu m Com p lexes

w ith Va r iou s An ion s^a

[cat.],

mM

rxn

time, s

polymer

yield,^bg

activity, ×10⁶g PE/

mol of M‚atm‚h^c

M_w,^d×10⁵

M_w/M_n

d

entry no.

cation

anion

Cp₂ZrMe^{+ e}

MeB(C₆F₅)₃

0.32

0.22

0.40

40

66

62

0.65(5)

1.1(1)

0.58(5)

<0.01

4.5(4)

6.8(7)

3.8(4)

1.24

5.21

2.55

2.0

1.4

2.0

-

1

2

3

4

(1,2-Me₂Cp)₂ZrMe^{+ e}

(Me₅Cp)₂ZrMe^{+ e}

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺MeB(C₆F₅)₃

900

<0.001

-

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

Cp₂ZrMe⁺

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

0.14

0.10

0.28

0.25

0.20

0.14

0.23

0.20

62

60

10

20

1.4(1)

1.5(2)

1.9(8)

1.6(7)

0.9(2)

0.8(2)

0.91(4)

0.84(8)

0.25(4)

0.32(5)

2.1(2)

3.2(7)

1.1(2)

0.78(4)

1.4(1)

1.0(2)

5.7(6)

6.2(6)

50(20)

42(20)

11(2)

8.78

5.65

8.29

6.02

21.6

12.8

3.36

1.70

0.087

0.068

9.87

7.03

5.32

0.986

0.090

1.72

1.32

1.57

1.99

1.78

2.06

1.7

2.87

2.52

1.38

1.61

1.66

1.93

3.13

Cp₂ZrMe⁺

-

(1,2-Me₂Cp)₂ZrMe⁺

(Me₅Cp)₂ZrMe⁺

(Me₅Cp)₂ThMe⁺

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

-

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

20

10(2)

-

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

600

720

60

22

20

60

240

60

0.92(4)

1.0(1)

0.052(9)

0.060(9)

6.4(5)

28(6)

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺B(C₆F₄TBS)₄

-

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺B(C₆F₄TIPS)₄

-

Cp₂ZrMe^{+ f}

B(C₆F₅)₄

-

(1,2-Me₂Cp)₂ZrMe^{+ f}

(Me₅Cp)₂ZrMe^{+ f}

(Me₅Cp)₂ThMe⁺

-

10(2)

3.6(2)

0.91(1)

3.1(6)

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺B(C₆F₅)₄

(1,2-Me₂Cp)₂ZrMe^{+ g}

B(C₆F₅)₄

a

b

25 °C, 1 atm of ethylene, in 100 mL of toluene. Average yield of g3 runs; the number in parenthesis is the standard deviation in the

yield. ^cThe number in the parenthesis is the standard deviation in the activity. By GPC, relative to polystyrene standards. ^eData of ref

d

4d. ^fCatalyst generated by in situ reaction of the corresponding dimethylzirconocene with Ph₃C⁺B(C₆F₅)₄^-in toluene (2-3 mL) for 5 min.

Catalyst aged for 3 h at 25 °C after generation by in situ reaction of (1,2-Me₂Cp)₂ZrMe₂with Ph₃C⁺B(C₆F₅)₄in toluene (2-3 mL).

g

-

856 Organometallics, Vol. 16, No. 5, 1997

J ia et al.

Ta ble 7. P r op ylen e P olym er iza tion P er for m a n ce of Ion -P a ir ed Ca tion ic Gr ou p 4 Com p lexes w ith Va r iou s

An ion s^a

[cat.], rxn time, polymer activity, ×10⁵g PP/

yield,^bg

mol of M‚atm‚h^c

M_w,^d×10³M_w/M_n

d

entry no.

cation

anion

mM

min

Cp₂ZrMe⁺

MeB(C₆F₅)₃

0.49

0.47

60

15.5(7)

5.1(5)

2.3(2)

0

8.1(7)

2.8(3)

1.4(1)

0

-

1

2

3

4

(1,2-Me₂Cp)₂ZrMe⁺

(Me₅Cp)₂ZrMe⁺

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺MeB(C₆F₅)₃

300

-

5

6

7

8

9

10

11

12

Cp₂ZrMe⁺

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

0.38

0.37

0.33

0.35

0.31

60

50

66

70

300

420

11(1)

7.5(8)

9.0(6)

12.9(1)

11.0(9)

1.3(3)

1.2(1)

0.16(1)

3.29

2.52

5.19

6.03

5.15

2.61

3.59

4.28

1.86

1.69

2.15

2.21

2.12

1.7

Cp₂ZrMe⁺

13.6(9)

16.2(1)

12.0(9)

2.0(5)

1.8(1)

1.1(1)

2.8(2)

-

(1,2-Me₂Cp)₂ZrMe⁺

(Me₅Cp)₂ZrH⁺

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

-

B(C₆F₄TBS)₄

B(C₆F₄TIPS)₄

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺B(C₆F₄TBS)₄

35

63

2.16

2.72

-

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺B(C₆F₄TIPS)₄

-

13

14

15

16

Cp₂ZrMe^{+ e}

B(C₆F₅)₄

0.39

40

11.2(1)

9.9(8)

1.3(1)

6.3(1)

10.8(1)

10.2(8)

1.3(1)

2.30

7.60

2.63

3.64

1.64

2.31

1.74

2.20

(1,2-Me₂Cp)₂ZrMe^{+ e}

(Me₅Cp)₂ZrMe^{+ e}

0.39

0.29

-

(Me₄Cp)SiMe₂(N^tBu)ZrMe⁺B(C₆F₅)₄

8.2(1)

a

b

25 °C, 1 atm of propylene, in 40 mL of toluene. Average yield of two runs; the number in parenthesis is the standard deviation in

the yield. ^cThe number in the parenthesis is the standard deviation in the activity. By GPC relative to polystyrene standards. ^eGenerated

d

by in situ reaction of the corresponding dimethylzirconocene with Ph₃C⁺B(C₆F₅)₄

.

-

polymerization activities of the (1,2-Me₂Cp)₂ZrMe⁺sys-

tems exceed those of the other cationic systems having

the same counteranions, with the effect most pro-

nounced for poorly coordinating anions (e.g., entries 7,

8, and 16 vs 5, 6, 9, 10, 15, and 17 in Table 6). This

(1,2-Me₂Cp)₂ZrR⁺effect has also been observed in the

case of other weakly coordinating anions developed in

this laboratory.^25b,33The constrained geometry catalysts

7a -c exhibit significantly lower polymerization activi-

ties than the metallocene complexes. Although this

probably reflects tight cation-anion coordination in the

case of 7a and 7b (entries 13 and 14, Table 6), it is likely

an intrinsic property of the constrained geometry cation

in aromatic solvents for 7c (entry 19, Table 6) because

an arene is coordinated to the cation rather than

has been observed with analogous lanthanide com-

plexes.²⁴

The present data evidence marked sensitivity of olefin

polymerization rates to details of cation-anion pairing,

with increased activity correlating with evidence for

weakened ion-pairing. Maximum activities are ob-

served for B(C₆F₄TBS)₄^-- and B(C₆F₄TIPS)₄^--based

-

pairs, with modification as B(C₆F₅)₄not further en-

hancing activity and significantly eroding thermal

stability. At present, the energetic profile for olefin

activation and insertion in such ion-pairs has not been

completely defined,^1a,35especially when the counteran-

ion is included. In principle, either olefin coordination^4f

(Figure 5a) or insertion (Figure 5b) could be rate-

limiting.^1a-cQuantum chemical analyses³⁵at several

levels argue that olefin coordination is exothermic for

the naked cations (∼20 kcal/mol), subsequent insertion

proceeds with little or no barrier (∼0-7 kcal/mol), and

it is highly exothermic. Addition of the counteranion

complicates this picture; however, it seems reasonable

that olefin activation would be more anion-sensitive

than insertion. Surprisingly little experimental activa-

tion energetic data relating to these reaction coordinates

is available, with E_a∼ 14 kcal/mol for (Me₄Cp)SiMe₂(N^t-

Bu)TiCl₂/MAO-mediated ethylene-styrene copolym-

erization^36aand ∼7 kcal/mol for (neomenthylC₅H₄)₂-

ZrCl₂/MAO-mediated ethylene polymerization.^36bPre-

liminary efforts in this laboratory³⁷to estimate ethylene

polymerization activation energetics for various zir-

-

B(C₆F₅)₄(vide supra). The polymeric products of the

reactions were characterized using ¹H and ¹³C NMR

spectroscopy and exibit the characteristic signatures of

highly linear polyethene, except those that are produced

by 7a -c (entries 13, 14, and 19, Table 6), which possess

branching units³⁴consistent with literature reports.^5h,i

The results of propylene polymerization experiments

are shown in Table 7. The activity differences among

the zirconocene catalysts as a function of anion are not

as dramatic as in ethylene polymerization. The intrinsic

steric and electronic characteristics of the cation appear

to play a more important role in monomer coordination

and insertion at the catalytic center for propylene than

for ethylene polymerization, and the role of the particu-

lar anion is less distinctive. However, the activity

differences among the constrained geometry complexes

with various anions are amplified, and strong anion

effects on polymerization activity are observed (entries

1

4, 11, 12, and 16, Table 7). As assayed by H and ¹³C

NMR,³²the polymerization products are all highly linear

atactic polypropylenes derived from regioselective 1,2-

insertion. Interestingly, the Th complexes do not cata-

lyze propylene polymerization. Instead, catalyst 8c

reacts with propylene stoichiometrically to afford an η³-

allyl complex and methane (eq 15), identical to what

(35) For relevant theoretical studies see: (a) Lohrenz, J . C. W.; Woo,

T. K.; Fan, L.; Ziegler, T. J . Organomet. Chem. 1995, 497, 91-104. (b)

Yoshida, T.; Koga, N.; Morokuma, K. Organometallics 1995, 114, 746-

758. (c) Bierwagen, E. P.; Bercaw, J . E.; Goddard, W. A., III. J . Am.

Chem. Soc. 1994, 116, 1481-1489. (d) Woo, T. K.; Fan, L.; Ziegler, T.

Organometallics 1994, 13, 432-433. (e) Weiss, H.; Ehrig, M.; Ahlrichs,

R. J . Am. Chem. Soc. 1994, 116, 4919-4928. (f) Castonguay, L. A.;

Rappe, A. K. J . Am. Chem. Soc. 1992, 114, 5832-5842.

(36) (a) Sernetz, F. G.; Mu¨lhaupt, R.; Waymouth, R. M. Macromol.

Chem. Phys. 1996, 197, 1071-1083. (b) Chien, J . C. W.; Razavi, A. J .

Polym. Sci., Part A: Polym. Chem. 1988, 26, 2369-2380.

(33) Chen, Y.-X., Marks, T. J . Manuscript in preparation.

(34) Bovey, F. A. Chain Structure and Conformation of Macromol-

ecules; Academic: New York, 1982; pp 78-91.

(37) Ishihara, A.; J ia, L.; Marks, T. J . Manuscript in preparation.

Cationic Metallocene Polymerization Catalysts

Organometallics, Vol. 16, No. 5, 1997 857

robust anions are required in this case. The second type

of decomposition channel appears to be blocked only by

employing strongly coordinating Lewis bases, such as

the counteranions. However, as noted above, Lewis

base coordination also suppresses catalytic activity.

Therefore, anion Lewis basicity properties must be

balanced sterically and electronically with cation Lewis

acidity properties to obtain ion-paired catalysts with

optimal stability and activity. One example of a catalyst

which may be nearer optimal is (1,2-Me₂Cp)₂Zr-

Me⁺B(C₆F₄TBS)₄^-, which has both good thermal stabil-

-

ity and polymerization catalytic activity. The B(C₆F₅)₄

-

based systems are representative examples of catalysts

in which the anion coordination is apparently too weak

to stabilize the highly electrophilic cations. Of course,

these correlations are presently empirical, and further

cation-anion pair elaboration and characterization is

needed to better define the boundary conditions.

F igu r e 5. Plausible energetic profiles for olefin activation

and insertion at metallocenium ion pairs.

Sch em e 2

Con clu sion s

Two silyl-functionalized/protected derivatives of the

tetrakis(perfluoroaryl)borate anion, B(C₆F₄TBS)₄^-and

B(C₆F₄TIPS)₄^-(TBS ) tert-butyldimethylsilyl and TIPS

) triisopropylsilyl) have been synthesized, and a series

of stable, highly reactive Zr and Th ion-paired methyl

and hydride catalysts have been isolated using these

anions. In contrast, the analogous B(C₆F₅)₄^--based

zirconocene methyl complexes are not stable at room

temperature; however, B(C₆F₅)₄^--based zirconocene hy-

dride complexes can be isolated. The relative coordi-

native ability of the series of fluoroarylborates with

respect to metallocene cations has been evaluated on

-

conocenes paired with MeB(C₆F₅)₃vs B(C₆F₄TBS)₄

reveal ∆H^q≈ 2-10 kcal/mol, with the former anion

exhibiting the greatest cation sensitivity (∼10 kcal/mol

for Cp₂ZrR⁺vs ∼2 kcal/mol for (Me₅Cp)₂ZrR⁺), while

results with the latter anion are more monotonic (∼4

kcal/mol for Cp₂ZrR⁺vs ∼7 kcal/mol for (Me₅Cp)₂ZrR⁺).

Importantly, if ion-pair reorganization processes such

as those in eqs 8 and 10 represent the extreme in the

energetic requirements for ion-pair separation (19-24

kcal/mol; Table 3) then either this degree of dissociation

the basis of spectroscopic and reactivity data and follows

-

the approximate order MeB(C₆F₅)₃^-> B(C₆F₄TBS)₄

≈

B(C₆F₄TIPS)₄^-> B(C₆F₅)₄^-. In cases of weakly coordi-

natiing anions, the neutral metallocenes compete with

the anions for the cationic metallocenes, forming dimeric

µ-Me complexes. Arene solvent coordination to the

constrained geometry cation [(Me₄Cp)SiMe₂(N^tBu)]-

ZrMe⁺is observed only when weakly coordinating

B(C₆F₅)₄is the counterion. The thermal stabilities of

the B(C₆F₄TBS)₄^-- and B(C₆F₄TIPS)₄^--based complexes

have been studied and are found to be generally greater

is not required for olefin activation (since ∆H^q

polymerization

< ∆H^q_{eq 8,10}) or the energetic demands of ion pair

dissociation are smaller when the σ-alkyl group is a

polymer chain and/or are partially compensated by

concurrent olefin complex formation (Figure 5b) or

insertion (Figure 5a) in a concerted transition state.

IV. F u r th er Discu ssion . Ca ta lyst Th er m a l Sta -

bility a n d Activity. For efficient catalytic processes,

the ideal situation is that the catalyst be both highly

active and thermally stable. It is therefore important

to understand how these two properties are related.

Since the present catalysts arguably exist in a metal

alkyl form at most stages in the catalytic cycle (Scheme

2), the stability of the metal alkyls is of greater interest

than that of the corresponding hydrides. Metal methyl

compounds can be regarded as working models of alkyls.

As discussed previously, the thermal decomposition of

the ion-paired methylzirconium complexes can be clas-

sified as follows: (i) decomposition involving the coun-

teranion or (ii) cation self-decomposition. The first

decomposition mode can be retarded or even prevented

by either exogenous Lewis base coordination, which is

not desirable because it will suppress catalytic activ-

ity,^1,2or by using robust anions. As an example, the

-

than those of the MeB(C₆F₅)₃analogues. The polym-

erization activity of the zirconocene catalysts reaches a

-

maximum when B(C₆F₄TBS)₄and B(C₆F₄TIPS)₄are

counteranions, and the polymerization activity of the

Zr constrained geometry catalyst reaches a maximum

in aromatic solvents due to arene coordination when

-

B(C₆F₅)₄is the counteranion.

Ack n ow led gm en t. This research was supported by

the U.S. Department of Energy (Grant DE-FG02-

86ER13511). L.J . thanks Akzo-Nobel Chemicals for a

predoctoral fellowship. We thank Dr. D. A. Kershner

of Akzo-Nobel for GPC analyses.

Su p p or tin g In for m a tion Ava ila ble: Complete X-ray

experimental details and tables of bond lengths, angles, and

positional parameters for the crystal structures of 8c and 11

(50 pages). Ordering information is available on any current

masthead page.

stability of the (1,2-Me₂Cp)₂ZrMe⁺system improves

-

significantly when protected B(C₆F₄TBS)₄

and

B(C₆F₄TIPS)₄^-are used as charge-compensating anions

instead of MeB(C₆F₅)₃^-. However, no obvious stabiliza-

tion of the Cp₂ZrMe⁺system is observed and still more

OM960880J

Article Doi

Cationic metallocene polymerization catalysts based on tetrakis(pentafluorophenyl)borate and its derivatives. probing the limits of anion noncoordination via a synthetic, solution dynamic, structural, and catalytic olefin polymerization study

DOI: 10.1021/om960880j

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om960880j

Full text of DOI:10.1021/om960880j

Products guided by the article

R&D Labs maybe for 167172-29-4

Relevant to this article

Hot Product