Ethylene bis(2-indenyl) zirconocenes: A new class of diastereomeric metallocenes for the (Co)Polymerization of α-olefins

3436

Organometallics 2001, 20, 3436-3452

Eth ylen e Bis(2-in d en yl) Zir con ocen es: A New Cla ss of

Dia ster eom er ic Meta llocen es for th e (Co)P olym er iza tion

of r-Olefin s

Colin J . Schaverien,*^,†Rene´ Ernst,^†Peter Schut,^†and Tiziano Dall’Occo^‡

Shell Research and Technology Centre Amsterdam, Postbus 38000,

1030 BN Amsterdam, The Netherlands, and Basell Polyolefins, Centro Ricerche G. Natta,

P.le G. Donegani 12, 44100 Ferrara, Italy

Received February 27, 2001

Sequential additions of NaOEt and benzyl chloride (BzCl) to diethyl adipate in THF

afforded EtOOCCH(Bz)CH₂CH₂CH(Bz)COOEt (3) in high yield via successive Dieckmann

condensations. This was converted to CH₂CH₂(2-indene)₂(2) using standard synthetic

methodology, from which 436 g of CH₂CH₂(2-indenyl)₂ZrCl₂(1) was prepared. Incorporation

of a 1-alkyl substituent gave rac and meso isomers CH₂CH₂(1-R-2-indenyl)₂ZrCl₂(R ) Me,

12r , 12m ; R ) Et, 14r , 14m ). Methylation gave rac- and meso-CH₂CH₂(1-Me-2-indenyl)₂-

ZrMe₂(15r , 15m ) and CH₂CH₂(2-indenyl)₂ZrMe₂(16). CH₂CH₂(4-Ph-2-indene)₂(20) was

prepared similarly to 2 using 2-phenyl benzylbromide and diethyladipate/NaOEt, from which

rac- and meso-CH₂CH₂(4-Ph-2-indenyl)₂ZrCl₂(19r and 19m ) were prepared. Alkylation of

19m gave meso-CH₂CH₂(4-Ph-2-indenyl)₂ZrMe₂(23m ). The 4-phenyl metallocenes rac- and

meso-CH₂CH₂(1-Me-4-Ph-2-indenyl)₂ZrCl₂(25r and 25m ) were prepared via 20. C₂-symmetric

CH₂CH₂(1,3-Me₂-2-indenyl)₂ZrCl₂(29) was prepared to compare its polymerization behavior

with 1. These zirconocenes displayed activities of up to 4320 kg PE/(g Zr‚h) for the

polymerization of ethylene (9.6 bar, 80 °C, hexane) to linear PE. The activity was much

lower (2-20 kg/(g Zr‚h); liquid propylene, 50 °C) for the polymerization of propylene to low

molecular weight atactic polypropylene. Alkyl groups in the 1- and/or 3-positions led to a

significant reduction of 2,1-insertions from 9.4% in the polymer from 1/MAO, and 9.9% from

19/MAO, to 0.1-0.4%. Attempts to prepare an isospecific catalyst were unsuccessful;

substituent variation i.e., 4-phenyl in 19, did not provide effective enantiomorphic site control.

Addition of H₂has a dramatic activation effect (>100-fold) in propylene polymerization and

afforded saturated propylene oligomers with activities up to 6240 kg/(g Zr‚h). The low polymer

molecular weights allowed detailed NMR analysis of the end groups, which enabled the

chain transfer and metallocene activation mechanisms to be identified. Copolymerization

of ethylene and propylene (in liquid propylene at 50 °C with 6 mol % ethylene) afforded

copolymers with activities of 1300-12500 kg/(g Zr‚h). Reactivity ratios (r_E) 41-98, r_P

)

0.0025-0.0084, r_Er_P) 0.1-0.7) indicate a very high preference for sequential ethylene

insertions. Copolymerization of ethylene and 1-hexene (4.4 bar ethylene, heptane, 70 °C)

gave copolymers with activities of up to 1100 kg/(g Zr‚h) using 12r /MAO. The copolymer

microstructure showed highly preferential sequential ethylene incorporation: r_E) ca. 200,

r_H) 0.004.

In tr od u ction

metallocenes could exercise enantiomorphic site control

in the polymerization of propylene and afford isotactic

polypropylene (i-PP).⁴Extensive investigation of sub-

stituent effects has resulted in highly active, highly

stereospecific homogeneous catalysts for the polymeri-

zation of propylene to high molecular weight, highly

Unbridged zirconocene dichlorides,¹when activated

by a cocatalyst, particularly methylaluminoxane (MAO),

are very efficient catalysts for the polymerization of

ethylene. By linking two indenyl ligands Brintzinger²

and Ewen³showed that the resulting chiral bridged

(4) (a) Ewen, J . A.; Haspeslagh, L.; Elder, M. J .; Atwood, J . L.;

Zhang, H.; Cheng, H. N. In Transition Metals and Organometallics as

Catalysts for Olefin Polymerization; Kaminsky, W., Sinn, H., Eds.;

Springer-Verlag: Berlin, 1987; p 281. (b) Kaminsky, W.; Rabe, O.;

Schauwienold, A.-M.; Schupfner, G. U.; Hanss, J .; Kopf, J . J . Orga-

nomet. Chem. 1995, 497, 181. (c) Spaleck, W.; Ku¨ber, F.; Winter, A.;

Rohrmann, J .; Bachmann, B.; Antberg, M.; Dolle, V.; Paulus, E. F.

Organometallics 1994, 13, 954. (d) Ewen, J . A.; Elder, M. J . In Ziegler

Catalysts; Fink, G., Mulhaupt, R., Brintzinger, H. H., Eds.; Springer-

Verlag: Berlin, 1995; p 99. (e) Stehling, U.; Diebold, J .; Kirsten, R.;

Ro¨ll, W.; Brintzinger, H. H.; J u¨ngling, S.; Mu¨lhaupt, R.; Langhauser,

F. Organometallics 1994, 13, 964.

* To whom correspondence should be addressed. E-mail:

Colin.J .Schaverien@opc.shell.com.

^†Shell Research and Technology Centre Amsterdam.

^‡Basell Polyolefins.

(1) Kaminsky, W.; Miri, M.; Sinn, H.; Woldt, R. Makromol. Chem.

Rapid Commun. 1983, 4, 417. Ewen, J . A.; Welborn, H. C. Eur. Patent

129369 B1 priority date J une 6, 1983, to Exxon. Sinn, H.; Kaminsky,

W. Adv. Organomet. Chem. 1980, 18, 99.

(2) Kaminsky, W.; Ku¨lper, K.; Brintzinger, H.-H.; Wild, F. R. W. P.

Angew. Chem., Int. Ed. Engl. 1985, 24, 507.

(3) Ewen, J . A. J . Am. Chem. Soc. 1984, 106, 6355.

Publication on Web 07/30/2001

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3437

isotactic PP.⁴These metallocenes⁵usually contain two

indenyl groups with a bridge (normally CH₂CH₂, SiMe₂,

or CMe₂) linking the 1,1′-positions. Ligand modifica-

tions, with retention of a bridged bis(cyclopentadienyl)-

type framework, have afforded catalysts for syndiotactic-

PP,⁶hemiisotactic-PP,⁷and atactic-PP,⁸whereas

unbridged 2-arylindenyl zirconocenes have yielded

elastomeric-PP.⁹Increasing attention is also being paid

to variations in the bridge position in the conventional

bis(1-indenyl) ligand framework.¹⁰

Sch em e 1

Halterman^11aand Bosnich^11bfirst prepared chiral

titanocenes with bis(indenyl) ligands bridged at the

2-position. They devised a conceptually elegant way to

obtain a single metallocene diasteromer by using rigid,

intrinsically chiral binaphthyl and 6,6′-dimethylbiphen-

yl bridges, while Nantz¹²has focused on ethylene-

bridged bis(2-indenyl) titanocenes. Recently, Nantz has

reported^12cthe TiCl₄-mediated reductive coupling of

2-(hydroxymethyl)indenes for the preparation of a range

of such ethylene-bridged bis(2-indenes), from which the

corresponding titanocenes were prepared. The parent

titanocene CH₂CH₂(2-indenyl)₂TiCl₂, as well as rac- and

meso-CH₂CH₂(1-R-2-indenyl)₂TiCl₂(R ) Me,^12bbenzyl^12b),

were prepared and the X-ray crystal structures of rac-

CH₂CH₂(1-R-2-indenyl)₂TiCl₂(R ) Me,^12bi-Bu,^12ci-Pr^12c),

CH₂CH₂(2-tetrahydroindenyl)₂TiCl₂,^12band CH₂CH₂(5-

Me-2-tetrahydroindenyl)₂TiCl₂^12cdetermined. These ti-

tanocenes were used as catalysts for the epoxidation of

substituted olefins.¹²

Resu lts a n d Discu ssion

The reported synthesis of CH₂CH₂(2-indene)₂(2) was

of low yield (15%)^12aand, although convenient to

perform, only suitable for preparing small quantities (ca.

5 g) of 2. We have developed three new routes to 2 from

which ethylene-bridged bis(2-indenes) can be prepared.

Two are described¹⁴in the Supporting Information.

Although synthetically viable, they are, however, only

suitable for preparing a maximum of 10 g (route 1) or

50 g (route 2) of 2. A new, simple, and efficient route

was devised that uses inexpensive starting materials.

Sequential addition¹⁵of NaOEt, benzyl chloride (BzCl),

NaOEt, and benzyl chloride to diethyl adipate in THF,

then washing with water to remove NaCl, and subse-

quent removal of excess BzCl and BzOEt (from the

reaction of BzCl with NaOEt) under vacuum afforded

ca. 95% pure dibenzyl ester 3 in 92% yield. 3 was

isolated as a viscous yellow-brown oil, which solidifies

on standing for several weeks at 20 °C to a light beige

powder (Scheme 1). For the exact methodology, see the

Experimental Section.

We report here a family of chiral ethylene-bridged

bis(2-indenyl) zirconocenes for the polymerization of

ethylene, of propylene, and for the highly efficient

copolymerization of ethylene/propylene and ethylene/

1-hexene. Some preliminary results have been reported

in patent applications.¹³

The impurities are typically 2-3% of 4 and <1% of

the cyclopentanone 5. This approach relies on successive

Dieckmann condensations. Although the Dieckmann

reaction is well established¹⁶and has been used for

preparing alkyl-substituted cyclopentanones,¹⁷we are

not aware of other examples of its use to prepare 2,5-

dialkyladipates. A proposed mechanism is shown in

Scheme 2.

(5) Horton, A. D. Trends Polym. Sci. 1994, 2, 158. Brintzinger, H.

H.; Fischer, D.; Mulhaupt, R.; Rieger, B.; Waymouth, R. M. Angew.

Chem., Int. Ed. Engl. 1995, 34, 1143. Kaminsky, W. J . Chem. Soc.,

Dalton Trans. 1998, 1413.

(6) Ewen, J . A.; J ones, R. L.; Razavi, A.; Ferrara, J . D. J . Am. Chem.

Soc. 1988, 110, 6255.

(7) (a) Ewen, J . A.; Elder, M. J .; J ones, R. L.; Haspeslagh, L.; Atwood,

J . L.; Bott, S. G.; Robinson, K. Makromol. Chem., Macromol. Symp.

1991, 48/ 49, 253. (b) Farina, M.; Di Silvestro, G.; Sozzani, P. Prog.

Polym. Sci. 1991, 16, 219. (c) Llinas, G. H.; Day, R. O.; Rausch, M. D.;

Chien, J . C. W. Organometallics 1993, 12, 1283.

(8) Resconi, L.; J ones, R. L.; Albizzati, E.; Camurati, I.; Piemontesi,

F.; Guglielmi, F.; Balbontin, G. Polym. Prepr. 1994, 35, 663. Resconi,

L.; J ones, R. L.; Rheingold, A. L.; Yap, G. P. A. Organometallics 1996,

15, 998.

(9) Hauptman, E.; Waymouth, R. M.; Ziller, J . W. J . Am. Chem. Soc.

1995, 117, 11586. Llinas, G. H.; Dong, S.-H.; Mallin, D. T.; Rausch,

M. D.; Lin, Y.-G.; Winter, H. H.; Chien, J . C. W. Macromolecules 1992,

25, 1242.

(10) Schaverien, C. J .; Ernst, R.; Schut, P.; Skiff, W. M.; Resconi,

L.; Barbassa, E.; Balboni, D.; Dubitsky, Y. A.; Orpen, A. G.; Mercan-

delli, P.; Moret, M.; Sironi, A. J . Am. Chem. Soc. 1998, 120, 9945.

Halterman, R. L.; Combs, D.; Khan, M. A. Organometallics 1998, 17,

3900.

(11) (a) Halterman, R. L.; Ramsey, T. M. Organometallics 1993, 12,

2879. (b) Ellis, W. E.; Hollis, T. K.; Odenkirk, W.; Whelan, J .;

Ostrander, R.; Rheingold, A. L.; Bosnich, B. Organometallics 1993, 12,

4391.

(12) (a) Nantz, M. H.; Hitchcock, S. R.; Sutton, S. C.; Smith, M. D.

Organometallics 1993, 12, 5012. (b) Hitchcock, S. R.; Situ, J . J .; Covel,

J . A.; Olmstead, M. M.; Nantz, M. H. Organometallics 1995, 14, 3732.

(c) Palandoken, H.; Wyatt, J . K.; Hitchcock, S. R.; Olmstead, M. M.;

Nantz, M. H. J . Organomet. Chem. 1999, 579, 338.

De-esterification of 3 with KOH (3 equivs) in MeOH/

H₂O (4:1 vol/vol) at 90 °C for 3 h afforded the dicar-

(14) Route 1: Reaction of the diGrignard of o-C₆H₄(CH₂Cl)₂with

0.5 equiv of diethylsuccinimate gave the diol in 65-75% yield. However,

the diGrignard has to be prepared in dilute solution. See: Lappert,

M. F.; Martin, T. R.; Raston, C. L.; Skelton, B. W.; White, A. H. J .

Chem. Soc., Dalton Trans. 1982, 1959. Similar methodology has been

used previously to afford 2-indenes. See ref 11b as well as Eur. Patent

Appl. 727446 to Idemitsu priority date Oct 26, 1993, and Witte, P.;

Lal, T. K.; Waymouth, R. M. Organometallics 1999, 18, 4147. Dehydra-

tion with p-toluenesulphonic acid in refluxing toluene yielded 2 (36%),

1,1′,4,4′-bis(indanyl)furan (38%), and a double bond isomer of 2, C₆H₄-

(CH₂)₂CdCHCH₂(2-indene) (2%). Route 2: Treatment of 300 g of

PhCH₂CH(COOEt)₂with NaH at 50 °C for 2 h gave Na[PhCH₂C-

(COOEt)₂], which was quenched with 2 equiv, 450 g of 1,2-dibromo-

ethane to afford PhCH₂C(COOEt)₂CH₂CH₂Br. This was isolated and

reacted with an additional 1 equiv of Na[PhCH₂C(COOEt)₂] to yield

307 g, 49% of the tetraester as a white powder. De-esterification, then

decarboxylation at 180 °C afforded 7. See Supporting Information for

full experimental details.

(15) Direct reaction of diethyladipate with 2 equiv of NaH in the

presence of 2 equiv of PhCH₂Cl afforded a complex mixture of products,

including 3. NB: Benzyl chloride does not react with NaH. See: Cristol,

S. J .; Ragsdale, J . W.; Meek, J . S. J . Am. Chem. Soc. 1949, 71, 1863.

(16) (a) Vogel’s Textbook of Practical Organic Chemistry, 5th ed.;

Longman Scientific & Technical: Harlow, U.K., 1989; p 1095. March,

J . Advanced Organic Chemistry, 4th ed.; J ohn Wiley & Sons: New

York, 1992; p 492.

(13) Schaverien, C. J .; Ernst, R.; van Loon, J . D.; Dall’Occo, T. Eur.

Patent Appl. 941997 to Montell Technology Company, priority date

March 9, 1998. Dall’Occo, T.; Baruzzi, G.; Schaverien, C. J . Eur. Patent

Appl. 942011 to Montell Technology Company, priority date March 9,

1998.

(17) Ring opening of the cyclopentanone ring has been shown to give

monoalkyl adipic esters in good yield. Nicole, L.; Berlinguet, L. Can.

J . Chem. 1962, 40, 353.

3438 Organometallics, Vol. 20, No. 16, 2001

Schaverien et al.

Sch em e 2

Sch em e 4

of 10 at 250-260 °C (melt) gave 2¹²in moderate yield

(42%) (Scheme 4).

In the thermolysis of 10, a byproduct was formed. On

boxylate dianion. After evaporation of MeOH, the

colorless aqueous solution was washed with ether to

remove the organic impurities formed in the synthesis

of 3 (Scheme 3). Acidification of 3 with concentrated HCl

in water at 10 °C precipitated 7 as a white powder,

which was conveniently isolated by filtration and wash-

ing with water. After drying to constant weight 7 was

isolated in 95% yield.

1

the basis of mass spectrometry, H-¹H COSY, ¹³C-¹H

HETCOR, and ¹³C-¹³C INADEQUATE 2D NMR tech-

niques, this was identified as 11. Only one (double bond)

regioisomer is present. Instead of eliminating a proton

to form 2,¹²intramolecular electrophilic attack of the

formed double bond on the transitory positive charge

occurs (Scheme 5).

Sch em e 5

Sch em e 3

Treatment of 7 with excess SOCl₂(55 °C, 8 h) gave

acid chloride 8 as a viscous light brown oil. This was

used directly in the Friedel-Crafts reaction (2.6 equiv

AlCl₃in CH₂Cl₂at 20 °C). After stirring for 16 h at 20

°C, the reaction mixture was carefully poured onto

crushed ice. This afforded bis(indanone) 9 as a white

powder in 90% overall yield from 7. Reduction of 9 with

NaBH₄(1.1 equiv) at 10 °C under air in THF/MeOH

afforded diol 10. Many different acid/solvent combina-

tions were investigated for the dehydration of 10 to 2.

No reaction occurred on refluxing 10 in toluene with

p-toluene sulfonic acid under Dean-Stark conditions (as

determined by NMR spectroscopy), although two prod-

ucts were observed by GC(MS) of the reaction mixture,

both with mass 258, indicating dehydration on the GC

column. This suggested that thermolysis of 10 might

be an efficient dehydration method. Indeed, thermolysis

UFF molecular modeling¹⁸shows 11 is in the lowest

energy configuration, i.e., with the protons attached to

carbons d and f in a cis conformation. Bis(indene) 2¹²

could be separated from 11 by washing with CH₂Cl₂.

Deprotonation of 2 with 2 equiv of n-BuLi in ether at

-5 °C cleanly afforded Li₂[CH₂CH₂(2-indenyl)₂]. Addi-

tion of Li₂[CH₂CH₂(2-indenyl)₂] in ether, cooled to -78

°C, to ZrCl₄in CH₂Cl₂, also cooled to -78 °C, and then

warming to 20 °C afforded 1, which was separated from

LiCl, and any polymeric products, by Soxhlet extraction

(18) Rappe´, A. K.; Casewit, C. J .; Colwell, K. S.; Goddard, W. A.,

III; Skiff, W. M. J . Am. Chem. Soc. 1992, 114, 10024. Castonguay, L.

A.; Rappe´, A. K. J . Am. Chem. Soc. 1992, 114, 5832. Rappe´, A. K.;

Goddard, W. A., III. J . Phys. Chem. 1991, 95, 3358.

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3439

Sch em e 6

Sch em e 7

with CH₂Cl₂. This afforded 436 g of zirconocene 1¹⁹as

a bright yellow powder in 75% yield (Scheme 6). As

expected, the ¹H NMR spectrum in CD₂Cl₂of 1 displays

equivalent cyclopentadienyl protons at δ 6.41 ppm, and

the ethylene bridge protons are a singlet at δ 3.31 ppm.

In the ¹³C NMR spectrum, these resonate at δ 103.1 and

31.1 ppm, respectively.

This route uses inexpensive starting materials and

concentrated solutions allowing the use of ordinary

laboratory glassware. No crystallization or distillation

purification steps are involved, and workup can be

performed under air. Only the treatment of 8 with AlCl₃,

and conversion of 2 to 1, must be performed under an

inert atmosphere.

with MeI and was isolated as a mixture of the three

double bond isomers. As shown by Nantz for titano-

cenes,¹²the prochiral ligands CH₂CH₂(1-R-2-indene)₂

afford a mixture of rac and meso diastereomers in the

corresponding zirconocenes. From 13, rac- and meso-

CH₂CH₂(1-Me-2-indenyl)₂ZrCl₂(12r and 12m ) were

prepared. In 12r , the indenyl 3-H proton resonates at

δ 6.16 ppm (¹H NMR) and the indenyl 3-carbon at δ 97.9

ppm (¹³C NMR). In 12m , these resonances are at δ 6.71

and 101.0 ppm, respectively. The 1-ethyl derivatives rac-

and meso-CH₂CH₂(1-Et-2-indenyl)₂ZrCl₂(14r and 14m )

were prepared in an analogous fashion. See the Experi-

mental Section. The rac and meso metallocene isomers

12r , 12m and 14r , 14m could be completely separated

by crystallization.

Metallocene 1 is achiral, possesses C₂-symmetry,²⁰

and would not be expected to exert enantiomorphic site

control in the polymerization of R-olefins. As exemplified

crystallographically by Nantz,¹²and in contrast to the

rac and meso isomers of ethylene-bridged bis(1-indenyl)

metallocenes, the spatial environment of the bis(2-

indenyl) framework remained almost identical in both

the rac and meso titanocene isomers.¹²To induce enan-

tiomorphic site control, a difference in energy between

the two diastereomers (∆∆G) is required to allow effec-

tive discrimination between the two propylene enantio-

faces. This enables the substituent effect (size and

position) in the CH₂CH₂(2-indenyl)₂framework to be

separated from the intrinsic influence of different rac

and meso ligand spatial environments, found in bridged

bis(1-indenyl)₂metallocenes, in inducing enantiomorph-

ic site control in R-olefin polymerization. To prepare rac

and meso isomers, two approaches were used: (i) alkyl

substitution in the cyclopentadienyl ring. This was

achieved by either deprotonation of 2 and RX quench

or, preferably, reaction of 9 with RMgX followed by

dehydration and (ii) substitution in the benzene ring.

This was achieved by successive Dieckmann condensa-

tions of diethyladipate with the appropriately substi-

tuted benzyl halide.

Due to the absence of a suitable stereochemical probe,

these metallocene dichlorides could not be unequivocally

distinguished as the rac and meso isomers. Therefore

rac- and meso-CH₂CH₂(1-Me-2-indenyl)₂ZrMe₂(15r and

15m ) were prepared by addition of MeLi to 12r and to

12m , respectively. As expected, 15r has a single Zr-Me

resonance at δ -1.02 ppm (¹H) and at δ 40.1 ppm (¹³C).

In contrast, 15m gives rise to two equal intensity Zr-

Me signals, at δ -0.38 and -1.82 ppm (¹H), and at δ

44.1 and 37.0 ppm (¹³C). See Scheme 7. As is commonly

observed in metallocene chemistry, the numerical aver-

age of the meso isomer resonances is very close to that

1

of the single resonance of the rac isomer in both the H

Bis(indene) CH₂CH₂(1-Me-2-indene)₂(13)¹²was pre-

pared by deprotonation of 2 with n-BuLi and quenching

and ¹³C NMR spectra and to that in CH₂CH₂(2-indenyl)₂-

ZrMe₂(16) (Zr-Me at δ -1.02 ppm).

4-Phenyl-substituted CH₂CH₂(4-Ph-2-indenyl)₂ZrCl₂

(19) was prepared, an analogue of the archetypal

catalyst for isospecific propylene polymerization rac-

Me₂Si(2-Me-4-Ph indenyl)₂ZrCl₂.^4c2,5-Di(2-phenylben-

zyl)diethyladipate (18) was prepared similarly to 2.

Addition of NaOEt to diethyladipate in THF, followed

by 2-phenylbenzyl bromide, afforded 2-(2-phenylbenzyl)-

diethyladipate (17) in 92% yield (GC) as a viscous oil.

This was used directly to prepare 18 in another NaOEt/

2-phenylbenzylbromide cycle (Scheme 8). De-esterifica-

tion of 18 using KOH in MeOH/H₂O (in 73% overall

(19) Ca u tion : The mutagenic potential of 1 was determined by its

effects on two histidine-requiring strains of Salmonella typhimurium

(TA 98 and TA 100) in the absence, and presence, of a rat liver

metabolizing system (S9 mix) using the plate-incorporation method

(“Ames test”). This showed 1 to be moderately mutagenic (>5-fold

increase in mean number of revertant colonies (at dosages > 556 µg/

plate)) compared to the spontaneous reversion rate observed in DMSO.

Ames, B. N.; McCann, J .; Yamasaki, E. Mutat. Res. 1975, 31, 347.

Maron, D. M.; Ames, B. N. Mutat. Res. 1983, 113, 173. We speculate

that the flat indenyl rings allow intercalation.

(20) Zirconocenes 1, 16, and 29 have C₂-symmetry due to the zigzag

conformation of the ethylene bridge. The X-ray crystal structure of CH₂-

CH₂(2-tetrahydroindenyl)₂TiCl₂showed slight distortion from idealized

C₂-symmetry.^12b

3440 Organometallics, Vol. 20, No. 16, 2001

Schaverien et al.

Sch em e 8

Sch em e 9

which consists of two double bond isomers. A byproduct

was consistently observed. On the basis of the similarity

of its H NMR spectrum to that of 11, it was tentatively

1

identified as the multicyclic molecule in Scheme 9. 20

was further purified by column chromatography. This

also allowed separation of the double bond isomers.

Treatment of 20 with 2 equiv of n-BuLi in ether gave

the dianion, which was reacted with ZrCl₄in toluene.

Workup afforded 19 as a 1:1 mixture of rac and meso

isomers (Scheme 10).

yield from diethyl adipate), preparation of the acid

chloride with excess SOCl₂(100% yield), and intramo-

lecular Friedel-Crafts acylation with AlCl₃afforded 21

in 89% isolated yield. ¹³C NMR data (see Experimental

Section) of the diacid, acid chlorides, 21 and 22 in

Scheme 8 indicated two diastereomers for each com-

pound. The ortho-phenyl substitution pattern in 18

allowed formation of only one (regio)isomer of bis-

(indanone) 21 after Friedel-Crafts ring closure. In

principle, this synthetic methodology could be extended

to other (ortho-)substituted benzyl halides in order to

prepare other ethylene-bridged bis(2-indenes) from di-

ethyladipate.

The meso isomer 19m could be completely separated

from 19r by selective crystallization. However, the best

separation achieved for 19r was a 4:1 mixture of rac

and meso isomers. The assignment of 19m as the meso

isomer was confirmed by reacting 19m with MeLi to

1

afford meso-CH₂CH₂(4-Ph-2-indenyl)₂ZrMe₂(23m ). H

NMR spectroscopy showed two equal intensity Zr-Me

resonances at δ -1.83 and 0.11 ppm.

Bis(indene) 24 was prepared straightforwardly in 69%

isolated yield from bis(indanone) 21 by reaction with

MeMgBr in a ether/THF mixture and dehydration with

p-MeC₆H₄SO₃H at 20 °C. This facile dehydration is in

marked contrast to the dehydration of 10 or 22, which

required elevated temperatures. It is probable that the

Reduction of 21 with LiAlH₄gave diol 22 in 91% yield.

Thermolysis of 22 for 20 min at 345 °C was found to be

the best method for the formation of 20 (in 87% yield),

Sch em e 10

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3441

Sch em e 11

Sch em e 12

1-alkyl group stabilizes the formed cation. The absence

of multicyclic byproducts, as was observed for dehydra-

tion of 10 and 22, can be ascribed to steric hindrance,

which inhibits intramolecular attack of the formed

double bond on the transient carbocation. The route

described here to give 24 and 27 (vide infra) is much

preferred over the alternative deprotonation of 20/RX

quench, which would lead to various mutual orienta-

tions of 1-R and 4-Ph groups (Scheme 11).

Given the close congruence of these metallocenes with

known titanocenes,¹²and our emphasis on (co)polym-

erization, (co)polymer characterization, end group analy-

sis, and structure-property relationships (vide infra),

many of these zirconocenes were not characterized by

1

elemental analysis. In lieu, H NMR spectra are repro-

duced in the Supporting Information.

P olym er iza tion Resu lts

CH₂CH₂(1-Me-4-Ph-2-indenyl)₂ZrCl₂(25) was formed

as a 1:1 ratio of rac and meso isomers. Although 25m

could be completely separated from the rac isomer, total

separation of 25r by crystallization was unsuccessful.

The best result, obtained by extraction with and crystal-

lization from hexane, was a 4:1 25r :25m ratio.

The steric bulk of the alkyl groups in the 1- (or 3-)

position (R ) H, 19; R ) Me, 25) has no discernible effect

on the rac:meso ratio. In contrast, Nantz observed¹²that

the diastereomeric ratios in the titanocenes CH₂CH₂(1-

R-2-indenyl)₂TiCl₂(R ) Me, rac:meso ) 1.1:1; R )

benzyl, rac:meso ) 3.6:1) and in their hydrogenated

counterparts CH₂CH₂(1-R-2-tetrahydroindenyl)₂TiCl₂

(R ) i-Bu, rac:meso ) 4:1; R ) i-Pr, rac:meso ) 2:1; R

) Ph, rac:meso ) 4.3:1) were much more sensitive to

the steric bulk of the substituent. This is presumably

due to the enhanced mutual repulsion of these 1- or

3-substituents in the meso form of the compacter titano-

cenes.

CH₂CH₂(1,3-Me₂-2-indenyl)₂ZrCl₂(29) is expected to

be more active in the polymerization of propylene

because of the anticipated suppression of 2,1-insertions,

as determined by UFF molecular modeling¹⁸(Scheme

12).

Deprotonation of 13 with 2 equiv of n-BuLi and

subsequent quenching with MeI gave bis(indene) 30 as

mixture of double bond isomers as well as 1,1-gem-

dialkylated product(s). Purification was accomplished

by reaction of 30 with n-BuLi to afford Li₂[30], followed

by washing with hexane to remove the more soluble

gem-dimethylated impurities, in which the (di)anion

cannot be formed. This gave 30 in 24% isolated yield.

Deprotonation of 30 with n-BuLi afforded Li₂[30], which

reacted with ZrCl₄in toluene to give 29 as an orange

powder in 10% isolated yield.

1. Eth ylen e. Activation of these 2,2′-bridged metal-

locenes with excess MAO or the aluminoxane of tri-

isobutylaluminum (TIBA) or trisooctylaluminum (TIOA)²¹

afforded highly active catalysts for the polymerization

of ethylene. Methylene-bridged CH₂(2-indenyl)₂ZrCl₂

(33)²²was also evaluated for comparison purposes, as

well as the archetypal C₂-symmetry metallocene rac-

CH₂CH₂(1-indenyl)₂ZrCl₂(rac-EBIZrCl₂). For conditions

and details see Table 1. Aluminoxanes tetraisobutyl-

aluminoxane (TIBAO) and tetraisooctylaluminoxane

(TIOAO) (freshly prepared from TIBA or TIOA with 0.5

equiv of water at 0 °C in hexane) afford catalysts at least

as active as those obtained with MAO as cocatalyst.

The activity of 1 and 12r are higher than, and the

activity of 12m is similar to, that of rac-EBIZrCl₂

(entries 13, 14), and much higher than that of 33

(entries 11, 12). Low molecular weight polyethylene

(LVN ) 0.2-0.7 dL/g) is produced with a relatively

narrow molecular weight distribution, M_w/M_n) ca. 2.5.

The presence of a 1-Me group provides rac and meso

metallocene isomers with different behavior. Precatalyst

12m is less active and affords a polymer of lower

molecular weight than either 1 or 12r . The use of

TIOAO as cocatalyst, instead of MAO, gives a further

increase in molecular weight (see entries 1 and 2 for 1

and entries 5 and 7 for 12r ) and an increase in activity.

(21) Dall’Occo, T.; Galimberti, M.; Resconi, L.; Albizzati, E.; Pennini,

G. WO Patent Appl. 96/2580 to Montell Technology Company, priority

date J uly 20, 1994.

(22) Van Beek, J . A. M.; de Vries, J .; Arts, H. J .; Persad, R.; van

Doremaele, G. H. J . WO 94/11406 to DSM, priority date November

11, 1992. van Doremaele, G. H. J .; van Beek, J . A. M.; Postema, R. A.

J . WO 95/10546 to DSM, priority date October 11, 1993. See also:

Resconi, L. WO 00/29415 to Montell Technology Company, priority date

November 18, 1998.

3442 Organometallics, Vol. 20, No. 16, 2001

Ta ble 1. Eth ylen e P olym er iza tion P a r a m eter s a n d P olym er Ch a r a cter iza tion ^a

Schaverien et al.

amount

(µmol)

yield

(g)

activity

(kg/(g Zr‚h))

LVN

(dL/g)

M_w× 10^-3

f

entry

metallocene

cocatalyst

(g/mol)^f

M_w/M_n

1^b

2

1

12r

12m

33

rac-EBIZrCl₂

0.24

0.25

0.48

0.22

0.74

1.20

0.21

MAO

2.88

3.25

94

530

860

4320

1470

810

1070

1350

560

380

460

130

27

0.42

0.54

0.56

0.47

0.46

0.59

0.67

0.36

0.37

0.17

0.2

TIOAO

MAO

TIBAO

TIOAO

MAO

TIBAO

TIOAO

MAO

TIOAO

MAO

3^c

4^d

5

23

2.6

64

2.75

3.66

4.61

1.92

1.28

1.57

1.51

1.54

1.31

1.18

6

7

8

9

10

11

12^e

13

14

32

18

2.3

2.5

410

370

1.38

1.60

TIOAO

a

Polymerization conditions: 200 mL glass autoclave, 100 mL of hexane with Al/Zr molar ratio of 1000:1. 10 min at 80 °C under 4.6 bar

ethylene. ^bPolymerization time 15 min. In 1 L of hexane in a 2.3 L autoclave for 30 min, 9.6 bar ethylene. Al/Zr molar ratio 2000 ^dIn 1

L of hexane in a 2.3 L autoclave for 60 min, 9.6 bar ethylene under 0.2 bar H₂. Al/Zr ) 2000. Polymerization time 30 min. GPC data.

c

e

f

This increase is most pronounced for 1/TIOAO (entry

3) compared to 1/MAO (entry 1) and 12r /TIOAO (entry

7), compared to 12r /MAO (entry 5). Addition of H₂(0.2

bar) leads to a decrease in activity²³and the expected

decrease in polymer molecular weight (compare entries

3 and 4, Table 1). This H₂-induced reduction of activity

in ethylene polymerizations has been previously ob-

served in both Ti-based Ziegler-Natta catalysts^24a,band

metallocenes.^24cThe exact reason for this has, however,

not been unequivocally established, although it has been

attributed^24bto slow chain reinitiation after chain trans-

fer to hydrogen caused by M-Et being stabilized by a

relatively strong â-agostic interaction. Single-component

ethylene polymerization catalysts with â-agostic stabi-

lized M-Et^24d,ehave been isolated, and the observed^24d

rates of ethylene insertion into (C₅Me₅)₂Sc-R followed

Sc-H . Sc-n-Pr > Sc-Me > Sc-Et.

The polyethylene obtained is essentially linear, as

given by DSC measurements (mp (T_m2) ) 133 °C and

enthalpy of fusion of 233 J /g, for 1/MAO). Due to the

low molecular weight and relatively narrow molecular

weight distribution of the polyethylene produced, met-

allocene 1, in conjunction with a Ziegler-Natta catalyst,

has been used to prepare broad molecular weight

distribution HDPE having a very low xylene soluble

polymer content.¹³

insertions was observed. Similarly, CH₂CH₂(4-Ph-2-

indenyl)₂ZrCl₂(19) also showed a very low activity (5

kg/(g Zr‚h)) and gave (presumably)²⁵atactic-PP pos-

sessing 9.9% 2,1-insertions. Alkyl substitution in either

the 1- and/or 3-positions suppresses propylene 2,1-

insertions. The 1-alkyl-substituted metallocenes 12r /

MAO, 12m /MAO, and 14m /MAO polymerize propylene

with activities of 16, 15, and 11 kg polymer/(g Zr‚h),

respectively, to afford essentially a-PP having fewer 2,1-

insertions (2.7, 0.4, and 0.9%, respectively). The effect

of alkyl groups in the 1- and/or 3-positions is exemplified

by 1,3-dimethyl-substituted 29, activated by MAO,

which afforded a-PP with only 0.2% 2,1-insertions and

a higher activity (60 kg/(g Zr‚h)).

Both rac and meso isomers of 12, 14, 19, and 25

polymerize propylene to give essentially a-PP, as re-

flected by the triad distributions in Table 2. A 1-alkyl

substituent, although inhibiting 2,1-insertions, provides

insufficient steric bulk to afford enantiomorphic site

control. Slightly enhanced polymer isotacticity (mm )

62%) was achieved in the polymerization with 19r /MAO/

H₂, although it is uncertain if this is energetically or

statistically significant. Given the suppression of 2,1-

insertions by 1- and/or 3-alkyl substituents (vide supra),

the combination of a 1-methyl and a 4-phenyl group in

CH₂CH₂(1-Me-4-Ph-2-Ind)₂ZrCl₂(25r ) gave a synergetic

effect with an increase of activity to 1350 kg/(g Zr‚h).

25m /MAO gave atactic-PP with moderate activity (280

kg/(g Zr‚h)).

2. P r op ylen e. These zirconocenes were also tested

for the polymerization of propylene (liquid propylene,

50 °C, with and without H₂). The results are given in

Table 2, as well as the characterization of the resulting

polymers and end group analysis. Metallocene 1, when

activated by methylaluminoxane (MAO), exhibited very

low activity (1.9 kg/(g Zr‚h)) for the polymerization of

propylene to low molecular weight (presumably)²⁵atac-

tic-PP. An unusually high percentage (9.4%) of 2,1-

To compare the effect of bridge length, the known

bridged (2-indenyl)metallocenes 1,1′-biphenyl(2-indenyl)₂-

11

22

ZrCl₂(31), S(2-indenyl)₂ZrCl₂(32), and CH₂(2-

22

indenyl)₂ZrCl₂(33) were prepared. These 2-indenyl

metallocenes were much less active than their ethylene-

bridged counterparts. See Table 2.

Sa tu r a ted En d Gr ou p s. The formation of a low

molecular weight polymer allowed comprehensive iden-

tification of the end groups, providing information on

the chain transfer mechanisms. As well as n-propyl and

vinylidene end groups characteristic of 1,2-insertion into

Zr-H/â-H elimination to the metal or â-H transfer to

monomer, strong signals corresponding to isobutyl end

groups were also observed, indicating that a major

termination pathway is chain transfer to Al, particularly

in the absence of H₂. This is commensurate with the

high Al:Zr ratio and the low activity. Thus, even in the

(23) The polymerization time of entry 3 was 30 min, while that of

entry 4 was 60 min. We cannot rule out that the apparent reduction

in activity is due to a nonlinear polymerization activity/decay pro-

file.

(24) (a) Kissin, Y. V. J . Mol. Catal. 1989, 56, 220 (b) Kissin, Y. V.;

Mink, R. I.; Nowlin, T. E.; Brandolini, A. J . Metalorganic Catalysts

for Synthesis and Polymerization; Kaminsky, W., Ed.; Springer-

Verlach: Berlin, 1999; p 60. (c) Reddy, S. S.; Sivaram, S. Prog. Polym.

Sci. 1995, 20, 309. (d) Burger, B. J .; Thompson, M. E.; Cotter, W. D.;

Bercaw, J . E. J . Am. Chem. Soc. 1990, 112, 1566. (e) Tanner, M. J .;

Brookhart, M.; DeSimone, J . M. J . Am. Chem. Soc. 1997, 119, 7617.

(25) We assume the polypropylene is atactic, although the high level

of 2,1-regioerrors rendered accurate triad analysis meaningless in the

polymer from 1/MAO and 19/MAO.

Ta ble 2. P r op ylen e P olym er iza tion Da ta , P olym er Micr ostr u ctu r e, a n d En d Gr ou p An a lysis^a

polymerization conditions

polymer regio- and steroeoregularity

2,1- mm mr rr

(mol %) (mol %) (mol %) (mol %) n-propyl n-butyl i-Bu DMB^gvinylidene allyl 4-butenyl 2-butenyl vinylidene

polymer end groups^c

amt Zr

activity

H₂

cis/trans

internal

b

compd (µmol) Al/Zr ratio (kg/(g Zr‚h)) (mol %)

M_n

1

4.0

2.0

3.75

0.5

4.0

0.5

4.0

0.5

4.0

0.5

4

0.5

4

0.5

4

0.5

4

1000

10000

5400

40200

5000

40200

5000

40000

5000

40000

5000

40000

5000

40000

5000

40000

5000

1.9

530

16

6240

15

4160

7

1320

11

0.0

2-3

0.0

2-3

0.0

2.5-3.1

0.0

2.4-3.6

0.0

2.0-2.5

0

2.7

0.0

2.7

0.0

2.7

0.0

2.7

0.0

2-4

0.0

2-3

3200

400

2700

850

3600

1800

9.4

0.3

2.7

n.d.

0.4

0.3

d

n.d.

20

n.d.

18.9

22.0

d

0.42

0.89

0.34

0.82

0.2

0.0

0.9

0.32

0.57

0.0

1.25 0.00

0.2 0.00

1.17 0.00

0.056

0.01

0.17

0.11

0.6

0.07

0.00

0.01

0.14

0.00

0.06

0.00

n.d.

49.1

n.d.

49.9

46.0

n.d.

30.9

n.d.

31.2

32.0

12r

12m

14r

14m

19r ^e

25r ^e

25m

29

0.5

1.2

trace

n.d

0.83

0.18

0.89 trace

0.09

insufficient polymer sample

1600

3300

1700

3200

1100

15500

1000

15100

1000

5500

1900

12000

1700

800

0.5

0.9

0.3

9.9

0.2

0.4

0.1

<0.1

0.2

0.5

1.4

1.2

2.0

1.2

36

17

23

d

38

51

46

d

26

32

31

d

16.2

31.6

27.9

28.9

27.1

29.2

22.8

5.1

0.78

0.20

0.82

0.60

0.95

0.35

0.94

0.38

0.93

0.0

0.8

0.6

0.9

0.97

1.04

0.48

0.01

0.13

0.25

0.68

0.03

0.04

0.0

0.05

0.5

0.2

0.05

0.04

0.70 trace

1.25 0.22

0.81 0.08

0.47 0.29

0.26 0.07

0.74 0.00

0.73 0.00

1.13 0.00

0.91 trace

1.91 0.00

1.14 trace

0.03

0.3

0.01

0.02

0.00

0.05

0.003

0.15

0.02

0.14

0.03

0.002

0.00

0.02

0.00

0.01

0.20

0.00

0.05

0.00

n.d.

0.00

0.03

0.09

0.00

0.01

0.00

0.006

0.00

n.d.

0.00

0.01

0.00

0.02

0.00

0.43

0.00

0.21

0.00

0.003

0.00

n.d.

5080

5

0.12

0.03

0.04

0.24

0.28

0.11

0.09

0.06

0.01

0.00

0.77

0.73

330

1350^f

1080

280

900

60

5040

31

114

27

62.2

21.1

27.8

21.9

26.3

19.0

40.3

84.1

44.2

12

21.6

47.4

44.3

49.2

46.6

51.8

36.8

10.7

33.9

49

29

31

33

0.5

4.0

0.5

4.0

2.0

40000

5000

40195

5000

0.9

1.2

n.d.

21.9

39

42

n.d.

0.00

n.d.

0.00

0.19 0.00

33

10000

400

700

13

45

a

Performed in a 5 L reactor containing 1.6 kg of liquid propylene at 50 °C for 1 h unless otherwise stated, cocatalyst ) MAO. Total amount of MAO kept constant at 20 mmol in each

experiment except in the first entry. ^bFrom ¹H NMR. ^cSum of end groups normalized to 2. The high percentage of 2,1-regiodefects renders triad analysis meaningless. Polymerizations performed

d

e

with 4:1 rac:meso ratio. Polymerization time ) 30 min. ^gDMB ) 2,3-dimethylbutyl. Under the conditions used, S(2-indenyl)₂ZrCl₂(32)/MAO was inactive.

f

3444 Organometallics, Vol. 20, No. 16, 2001

Schaverien et al.

1

Sch em e 13. Un sa tu r a ted En d Gr ou p s; H NMR

Ch em ica l Sh ifts Rela tive to Low est F ield Sign a l of

1,2,4-Tr ich lor oben zen e a t δ 7.32 p p m

is strongly inhibited due to the higher steric hindrance

of a Zr-CHMeCH₂-P moiety, compared to Zr-CH₂-

CHMe-P. The dosing of hydrogen in the polymerization

of propylene by both Ziegler-Natta³²and metallocene³³

catalysts leads to an activity enhancement due to the

reactivation of these dormant sites. For the polymeri-

33a

zation of propylene by rac-EBIZrCl₂

and rac-Me₂Si-

(2-Me-4-Ph-indenyl)₂ZrCl₂,^33bdosing of H₂resulted in

an increase of activity by 3-fold and 10-fold, respectively.

Notably, when the polymerization of propylene is

performed under H₂(2-3 mol % in the autoclave gas

cap as determined by on-line monitoring), a dramatic

increase (100-1000-fold) in catalyst activity was ob-

served. This afforded saturated propylene oligomers

with activities of 6-10 tons/(g Zr‚h). In addition, the

polymer derived from 1/MAO/H₂and 19/MAO/H₂showed

a marked reduction in embedded 2,1-insertions: from

9.4% to 0.3% and from 9.9% to 0.2%, respectively. As

well as n-propyl (from 1,2-insertion into Zr-H initiator)

and isobutyl (from chain transfer to H₂after a 1,2-

insertion) end groups, the other significant chain ter-

mination pathway is chain transfer to H₂after a 2,1-

insertion giving rise to n-butyl end groups.

In -Situ Hyd r ogen Gen er a tion a n d Con su m p tion .

In some of these polymerizations performed with added

H₂, obvious discrepancies between isolated and calcu-

lated³⁴polymer yields were observed. In these instances,

we attribute the higher calculated yield to the hydro-

genation of propylene to propane³⁴(determined and

quantified by on-line GC sampling) concomitant with

the polymerization of propylene. This indicates that in

polymerizations performed without added H₂this is a

viable pathway for low equilibrium levels of hydrogen

generated in-situ (determined and quantified by on-line

GC sampling) by allylic activation to be consumed by

propylene hydrogenation. An alternative pathway for

H₂consumption is chain transfer to H₂(i.e., detection

of n-Bu end groups (vide supra) in the absence of

external hydrogen (see Table 2).

absence of external H₂, the majority of the end groups

are saturated (Table 2). The high percentage of 2,1-

regioerrors in the polymer obtained from 1/MAO and

19/MAO has consequences for the end group analysis.

A small amount of n-butyl end groups are observed (in

the absence of external H₂). These arise from 2,1- (1,3-)

insertion, then transfer to Al or, alternatively, by

consumption of in-situ-generated H₂(vide infra). Trans-

fer to Al would be consistent with the large Al:Zr ratio.

Resonances due to the 2,3-dimethylbutyl end group²⁶

were also observed (2,1-insertion into Zr-H followed by

1,2-insertion into Zr-i-Pr).

Un sa tu r a ted En d Gr ou p s. A variety of unsaturated

end groups were observed. Small amounts of allyl end

groups (from â-Me elimination) were observed through-

out. No resonances due to an isobutenyl end group (δ

5.0 ppm) were observed.²⁷The resonance is also ob-

scured by that of the higher field allyl multiplet.

Relatively high levels of cis or trans (we assume cis)

2-butenyl end groups were observed. These are formed

by â-H transfer to monomer after a 2,1-insertion.²⁸As

previously described by Resconi,^29d,30athe broad singlet³¹

at δ 5.20 ppm (¹H NMR) and ascribed to the 4-butenyl

end group^29dis only observed in conjunction with the

broad singlet at δ 5.42 ppm (2-butenyl), indicating that

it, too, is associated with 2,1-insertions. Indeed, inspec-

tion of Table 2 shows that both end groups are most

prevalent from 1/MAO and 19/MAO (Scheme 13).

Effect of Ad d ed Hyd r ogen . Kinetic data³²has

indicated that chain propagation after a 2,1-regiodefect

Hydrogen evolution has been observed during olefin

polymerizations²⁹and is supported theoretically.^29ePro-

pylene insertion into a Zr-allyl species has been

recently studied,³⁵and propylene oligomerization by

[(C₅Me₅)₂MMe(THT)]BPh₄(M ) Zr, Hf; THT ) tetrahy-

drothiophene)³⁶showed slow deactivation by formation

(26) Randall, J . C.; Ruff, C. J .; Vizzini, J . C.; Speca, A. N.; Burkhardt,

T. J . In Metalorganic Catalysts for Synthesis and Polymerization;

Kaminsky, W., Ed.; Springer-Verlag: Berlin, 1999; p 601. Moscardi,

G.; Piemontesi, F.; Resconi, L. Organometallics 1999, 18, 5264.

(27) Spectra measured in 1,2,4-trichlorobenzene. In C₂D₂Cl₄, which

can contain acidic impurities, isomerization to the isobutenyl end group

has been observed.^30a

(33) Metallocenes: (a) Tsutsui, T.; Kashiwa, N.; Mizuno, A. Mak-

romol. Chem. Rapid Commun. 1990, 11, 565. (b) Aulbach, M.; Bach-

mann, B.; Ku¨ber, F.; Spaleck, W.; Winter, A. Proceeding of Metallocenes

’95; April 1995 in Brussels, Belgium; Schotland: Skillman, NJ ; p 321.

(c) Busico, V.; Corradini, P.; Cipullo, R. Makromol. Chem. Rapid

Commun. 1993, 14, 97. (d) Busico, V.; Cipullo, R.; Chadwick, J . C.;

Modder, J . F.; Sudmeijer, O. Macromolecules 1994, 27, 7538. (e)

Spaleck, W.; Ku¨ber, F.; Bachmann, B.; Fritze, C.; Winter, A. J . Mol.

Catal. 1998, 128, 279. (f) Carvill, A.; Tritto, I.; Locatelli, P.; Sacchi, M.

C. Macromolecules 1997, 30, 7056. (g) Lin, S.; Kravchenko, R.;

Waymouth, R. M. J . Mol. Catal. 2000, 158, 423.

(34) During propylene polymerizations the gas cap was sampled and

analyzed by on-line GC at 1 min intervals. The propane/propylene ratio

was determined, making use of the natural abundance (typically 0.3

mol %) of propane in the propylene feed. From the change in propane/

propylene ratio as a function of time propylene conversion could be

determined. The calculated polypropylene yield is therefore given by

propylene intake × (1 - R₀/R_t) where R₀) propane/propylene ratio at

t ) 0; R_t) propane/propylene ratio at time t during the polymerization.

The calculated yield is compared with the actual weight of polymer

isolated from the autoclave. A calculated yield significantly larger than

the actual yield is ascribed to propane formation by hydrogenation of

propylene during the polymerization.

(28) Resconi, L.; Fait, A.; Piemontesi, F.; Colonnesi, M.; Rychlicki,

H.; Ziegler, R. Macromolecules 1995, 28, 8, 6667.

(29) (a) Horton, A. D.; Van Baar, J . F. Personal communication. (b)

Karol, F. J .; Kao, S.; Wasserman, E. P.; Brady, R. C. New. J . Chem.

1997, 21, 797. (c) Wasserman, E.; Hsi, E.; Young, W.-T. Polym. Prepr.

Am. Chem. Soc., Div. Polym. Chem. 1998, 39, 425. (d) Resconi, L.;

Piemontesi, F.; Camurati, I,; Sudmeijer, O.; Nifant’ev, I. E.; Ivchenko,

P. V.; Kuz’mina, L. G. J . Am. Chem. Soc. 1998, 120, 2308. (e) Margl,

P. M.; Woo, T. K.; Blo¨chl, P. E.; Ziegler, T. J . Am. Chem. Soc. 1998,

120, 2174.

(30) (a) Resconi, L.; Camurati, I.; Sudmeijer, O. Top. Catal. 1999,

7, 145. (b) Resconi, L. J . Mol. Catal. A: Chem. 1999, 146, 167 (c) Sacchi,

M. C.; Carvill, A.; Forlini, F.; Tritto, I.; Locatelli, P. Recent Res. Dev.

Macromol. Res. 1999, 4, 57.

(31) In C₂D₂Cl_4,this resonance appears as a triplet at δ 5.21 ppm.^29d

(32) Ziegler-Natta: Chadwick, J . C.; Morini, G.; Albizzati, E.;

Balbontin, G.; Mingozzi, I.; Christofori, A.; Sudmeijer, O.; van Kessel,

G. M. M. Macromol. Chem. Phys. 1996, 197, 250.

(35) Lieber, S.; Prosenc, M.-H.; Brintzinger, H.-H.Organometallics

2000, 19, 377.

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3445

Ta ble 3. Eth ylen e/P r op ylen e Cop olym er iza tion P a r a m eter s a n d Cop olym er Ch a r a cter iza tion ^a

amt Zr

(µmol)

Al/Zr

ratio

polym time

(min)

yield

(g)

activity

(kg/(g Zr‚h))

LVN

(dL/g)

propylene

incorp (mol %)

metallocene

1

0.5

0.94

1.0

0.5

40000

21500

20400

40180

40230

40450

21000

45

17

33

65

60

44.5

260

250

316

570

6.3

1300

10 660

4992

6400

12 500

138

0.47

0.71

0.53

0.93

0.63

2.22

13.8

17.9

24.1

26.9

23.3

52.9

45

12r

12m

14r

14m

31

0.5

0.97

32

∼1

a

EP copolymerizations were performed at 50 °C in a 25 L reactor containing 7.5 kg of liquid propylene and with 5.5-6.2 mol % ethylene

as measured in the liquid phase. Total amount of MAO cocatalyst was kept constant at 20 mmol in each experiment.

of meth(allyl) species. Addition of H₂restored catalyst

activity. As suggested by others,^29,30hydrogen formation

can be ascribed to formation of internal unsaturations

(internal vinylidene at δ 4.80 ppm) in the polymer. See

Table 2. The mechanism of allylic activation and H₂

evolution has been previously reported,^29b,30a,bas well

as end group characterization. This suggests propylene

insertion into a Zr(η¹-allyl) initiator generated by allylic

activation of methyl, rather than methylene groups,

Cop olym er iza tion Resu lts

Eth ylen e/P r op ylen e. Copolymerization of ethylene

and propylene (liquid propylene, 50 °C, ca. 6 mol %

ethylene) afforded low molecular weight (LVNs ) 0.47-

0.93 dL/g) EP copolymers with very high activities

(1300-12 500 kg/(g Zr‚h)). For conditions see Table 3.

The activities are also higher than those obtained in the

polymerization of ethylene. This increase in activity, the

“positive comonomer effect”,^24c,33g,37is often observed in

copolymerizations of ethylene and another R-olefin

(often propylene or 1-hexene), compared to the ethylene

30

with liberation of H₂(Scheme 14). Indeed, we did not

detect isobutenyl end groups (Table 2).

polymerization activity, despite reactivity ratio r_E

.

Sch em e 14. P r op osed Mech a n ism for In ter n a l

Vin ylid en e F or m a tion

r_R-olefin(vide infra). This phenomenon is not completely

understood. It has been postulated³⁷that this may be

due to easier monomer diffusion in the copolymer, with

respect to the more crystalline polyethylene, or to

changes in rate constants for ethylene insertion in the

presence of R-olefin, or to an increase in the number of

active sites.

The EP copolymers were analyzed by ¹H and ¹³C

NMR spectroscopy. Despite these copolymerizations

being performed in liquid propylene (>10 M), with ca.

6 mol % ethylene, much more ethylene (73.1-86.1 mol

%) than propylene is incorporated. This is also reflected

in their ethylene and propylene reactivity ratios (vide

infra). There is also little variation in copolymer “blocki-

ness” (Table 4). Analysis of the saturated end groups

by ¹³C NMR spectroscopy showed only sequences of

n-hexyl (and possibly longer)³⁸of the initiating alkyl

group. This suggests that initiation occurs exclusively

by sequential ethylene insertion. Given the much higher

activities observed for EP copolymerization compared

to propylene polymerization, ethylene seems to fulfill a

similar activating role to H₂in the polymerization of

propylene. One possibility is insertion into a transient

Zr-allyl species to reinitiate propagation, although we

have no evidence to support this. The unsaturated end

groups were predominantly vinyl with relatively little

vinylidene (¹H NMR), indicating that chain transfer to

monomer occurs mainly after ethylene insertion. As in

the polymerization of ethylene and propylene, the EP

(co)polymer molecular weights are all low (LVN < 1 dL/

As found for rac-Me₂C(3-t-Bu-indenyl)₂ZrCl₂/MAO³⁰

and observed here for various 2,2′-bridged metallocenes

(Table 2), addition of external H₂suppresses the forma-

tion of both internal vinylidene and 2- and/or 4-butenyl

end groups.

Field desorption MS analysis of the low molecular

weight polymers from 19r /MAO and 29/MAO showed

peaks corresponding to saturated polymer as well as

masses corresponding to significant amounts of mono

and double unsaturations. The polymer from 29/MAO/

H₂, measured as a reference for the invasiveness of the

technique, showed, as expected for a polymerization

performed under hydrogen, only masses attributable to

saturated polymer.

(37) (a) Chien, J . C. W.; Nozaki, T.J . Polym. Sci. Part A: Polym.

Chem. 1993, 31, 227. (b) Herfert, N.; Montag, P.; Fink, G. Makromol.

Chem. 1993, 194, 3167. (c) Lehtinen, C.; Lo¨fgren, B. Eur. Polym J .

1997, 33, 115. (d) Naga, N.; Ohbayashi, Y.; Mizunuma, K. Makromol.

Rapid Commun. 1997, 18, 837. (e) Koivuma¨ki, J .; Fink, G.; Seppa¨la¨,

J . V. Macromolecules 1994, 27, 6254. (f) Lehmus, P.; Ha¨rkki, O.; Leino,

R.; Luttikhedde, H. J . G.; Na¨sman, J . H.; Seppa¨la¨, J . V. Macromol.

Chem. Phys. 1998, 199, 1965. (g) Spitz, R.; Pasquet, V.; J oly, J . F.

Makromol. Chem., Macromol. Symp. 1991, 47, 95. (h) Soga, K.;

Yanagihara, H.; Lee, D. Makromol. Chem. 1989, 190, 995.

(36) Eshuis, J . J . W.; Tan, Y. Y.; Meetsma, A.; Teuben, J . H.;

Renkema, J .; Evens, G. G. Organometallics 1992, 11, 362.

(38) n-Alkyl end groups longer than n-hexyl cannot be differentiated

from each other by NMR spectroscopy.

3446 Organometallics, Vol. 20, No. 16, 2001

Ta ble 4. Distr ibu tion of Eth ylen e a n d P r op ylen e Block s in EP Cop olym er ^a

Schaverien et al.

entry

metallocene

E

P

PP

PE

EE

PPP

PPE

EPE

PEP

PEE

EEE

r_E

r_P

r_Er_P

1

2

3

4

5

1

86.1

82.1

75.9

46.7

55.0

13.8

17.9

24.0

52.9

45.0

2.0

4.0

2.1

23.4

27.7

43.6

30.0

74.3

68.3

54.1

31.4

0.2

2.2

0.4

3.7

5.7

3.7

10.3

11.4

19.6

4.7

2.3

2.4

6.6

5.3

21.2

22.3

30.2

19.1

61.9

56.0

39.1

22.7

98.1

78.1

41.6

43.4

11.8

0.0068

0.0084

0.0025

0.237

0.67

0.66

0.104

10.29

0.31

12r

12m ^b

31^c

32

37.7

28.0

18.8

0.026

a

b

All percentages in mol %. 0.1 mol % 2,1-regiodefects present, hence % E + % P * 100%. ^c0.4 mol % 2,1-regiodefects present, hence

% E + % P * 100%.

F igu r e 1. Ethylene content in copolymer as a function of the feed ratio in liquid propylene at 50 °C. H1 ) rac-Me₂Si(2-

Me-4-Ph-indenyl)₂ZrCl₂/MAO.

g). There does not appear to be a correlation of activity

with the percentage of incorporated ethylene or with

copolymer molecular weight. An increase in EP copoly-

mer molecular weight with increasing ethylene incor-

over propylene with r_E) 6.61, r_P) 0.06 and r_E) 4.23,

r_P) 0.12, respectively. The highly isospecific and

sterically encumbered metallocene rac-Me₂Si(2-Me-4-

4c

Ph-indenyl)₂ZrCl₂gives high levels of R-olefin incor-

poration even at relatively low R-olefin concentrations.⁴²

In contrast, sterically hindered (C₅Me₅)₂ZrCl₂/MAO⁴²

shows a very high preference for consecutive ethylene

insertions with r_E) 250.

39a

poration has been previously reported for rac-EBIZrCl₂

and rac-Me₂Si(1-tetrahydroindenyl)₂ZrCl₂/MAO.^39b

11

1,1′-Biphenyl(2-indenyl)₂ZrCl₂(31) gave a higher

molecular weight EP copolymer with significantly more

(52.9 mol %) propylene incorporation than the ethylene-

bridged bis(2-indenyl) zirconocenes (Table 3). S(2-

indenyl)₂ZrCl₂²²(32) was much less active and gave only

traces (<1 g) of copolymer, but also with higher pro-

pylene incorporation (45 mol %) than the ethylene-

bridged zirconocenes.

The data in Table 4 are also given in Figure 1, in

which the ethylene content (f_E) of the EP copolymer is

displayed as a function of the feed (ethylene/propylene

in mol %) ratio (F_E) and of the ethylene and propylene

reactivity ratios (r_E, r_P). The lines give f_Eas a function

of F_Eas calculated according to the Mayo-Lewis copo-

lymerization equation,⁴⁴and the displayed points in

Figure 1 give f_Eas determined directly by ¹³C NMR

spectroscopy.

Eth ylen e/1-Hexen e. Copolymerizations were also

performed with 1-hexene as R-olefin in heptane at 70

°C under 4.4 bar ethylene to form LLDPE. The details

are shown in Table 5. As expected, the incorporation of

1-hexene increases with increasing [1-hexene], and

there is a commensurate decrease in both copolymer

crystallinity, as given by ∆H_fusion, and melting point,

entries 1-5. Using 1/TIOAO as catalyst and a 50:50 vol/

vol 1-hexene:heptane ratio, a maximum of 9.9 wt %

The ethylene and propylene distributions, as well as

the reactivity ratios, are given in Table 4.

Ethylene (r_E) and propylene (r_p) reactivity ratios⁴⁰

were calculated⁴¹for the EP copolymers from 1, 12r ,

and 12m and are compared with 31, 32, and rac-

EBIZrCl₂by fitting normalized experimental ¹³C NMR

peak areas to a single-site first-order Markov model. The

single-site Markov model is completely defined by the

ethylene and propylene reactivity ratios r_Eand r_p,

respectively. Each of the 2,2′-bridged metallocenes has

a clear preference for ethylene with r_E. 1 and r_P, 1.

In comparison, rac-EBIZrCl₂^39aand rac-Me₂Si(1-indenyl)₂-

37c,d

ZrCl₂

have a much smaller preference for ethylene

(42) Miri, M.; Hetzer, D.; Miles, A.; Pecak, M.; Riscili, B. In

Metalorganic Catalysts for Synthesis and Polymerization; Kaminsky,

W., Ed.; Springer-Verlag: Berlin, 1999; p 509.

(39) (a) Dro¨gemuller, H.; Heiland, K.; Kaminsky, W. In Transition

Metals and Organometallics as Catalysts for Olefin Polymerization;

Kaminsky, W.; Sinn, H., Eds.; Springer-Verlag: Berlin, 1988; p 303.

(b) Sugano, T.; Endo, J .; Takahama, T. In Science and Technology in

Catalysis; Izumi, Y., Arai, H., Iwamoto, M., Eds.; Elsevier: New York,

1994; p 371.

(40) Odian, G. In Principles of Polymerization, 3rd ed.; Wiley &

Sons: New York, 1991.

(41) The reactivity ratios were determined by using a first-order

Markov model (single site) with the Levenspiel Marquadt algorithm.

See: Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T.

Numerical Recipes in Pascal. The Art of Scientific Computing; Cam-

bridge University Press: Cambridge, 1990.

(43) Ewen, J . A. In Catalytic Polymerization of Olefins; Keii, T., Soga,

K., Eds.; Kodansha: Tokyo, 1986; p 271. Zambelli, A.; Grassi, A.;

Galimberti, M.; Mazzocchi, R.; Piemontesi, F. Makromol. Chem. Rapid

Commun. 1991, 12, 523.

2

(44) Mayo-Lewis equation: f_E) F_E(1 - F_E) + r_EF_E²/{r_EF_E+ 2F_E(1

- F_E) + r_P(1 - F_E)²} where F_E) x/1+ x where f_E) ethylene mole

fraction in EP copolymer, F_E) ethylene mole fraction in the reaction

medium (= x/1 + x), r_E) ethylene reactivity ratio, r_p) propylene

reactivity ratio, x ) feed ratio (mol/mol), i.e., molar ratio of ethylene

to propylene (x = [E]/[P]). Mayo, F. R.; Lewis, F. M. J . Am. Chem. Soc.

1944, 66, 1594.

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3447

Ta ble 5. Eth ylen e/1-Hexen e Cop olym er iza tion P a r a m eter s a n d Cop olym er Ch a r a cter iza tion ^a

amount

(µmol)

1-hexene

(mL)

time

(min)

yield

(g)

activity

(kg/(g Zr‚h))

LVN

(dL/g)

1-hexene

incorp (wt %)

∆H_fusion

(J /g)

entry

metallocene

mp (°C)

1

2

3

4

5

6

7

8

9

1

12r

12m

0.18

0.20

0.18

0.20

0.24

0.11

0.22

0.44

0.25

0

5

15

30

20

30

20

30

20

1.65

2.25

2.62

3.18

3.88

3.72

5.60

3.20

6.46

3.2

396

497

653

703

356

1093

1097

470

316

440

0.60

0.54

0.60

0.57

0.49^b

0.59

0.58^c

0.30^d

0.33

1.03

0.0

1.7

2.9

4.0

9.9

5.4

16.4

9.9

31.2

28.2

233

188

162

152

86

154

90

139

132.9

124.2

120.9

118.4

116.0

116.7

95.3

10

15

50

15

50

15

50

15

108.9

12m

rac-EBIZrCl₂

amorphous

44

10

68.0

a

Cocatalyst was TIOA/H₂O in all cases. Al:Zr ) 1000. Polymerizations in heptane, total volume ) 100 mL, ethylene pressure 4.4 bar.

^bM_w) 22 000, M_w/M_n) 2.6. M_w) 33 000, M_w/M_n) 2.2. ^dM_w) 18 000, M_w/M_n) 2.3.

c

Ta ble 6. Distr ibu tion of Eth ylen e a n d Hexen e Block s in EH Cop olym er ^a

entry

metallocene

E

H

HH

HE

EE

HHH HHE EHE HEH HEE

EEE

n_E

r_E

r_H

r_Er_H

4

5

6

7

8

9

10

1

12r

12m

98.63

96.47

98.1

93.8

96.5

86.8

1.37 0.00

3.53 0.00

2.80 97.2

7.10 92.9

0.00

0.4

0.00

0.8

1.37 0.00

3.53 0.00

2.90 95.73 70.4

7.05 89.42 27.2

182

235

144

b

1.9

6.2

0.00

0.2

3.5

12.3

6.8

96.5

87.5

93.2

73.9

77.3

1.7

5.9

0.00

0.6

0.1

2.1

1.7

3.7

11.2

6.8

94.6

81.9

89.8

63.9

67.9

56.1

15.2

28.4

128 0.004 0.51

72

52 0.004 0.21

18 0.018 0.32

3.5

13.2

11.6

0.00

0.4

0.5

3.3

b

25.7

22.2

12.3

10.6

21.8

18.8

6.75

7.97

rac-EBIZrCl₂88.4

1.0

a

All percentages in mol %. [E] ) [EEE] + [HEE] + [HEH]; [H] ) [HHH] + [HHE] + [EHE]. ^bDue to the total absence of adjacent

hexene units ([HH] ) 0), r_Hand hence the product r_Er_Hcould not be calculated.

1-hexene incorporation was obtained. The ethylene/1-

hexene copolymer is low molecular weight (LVN )

0.49-0.60 dL/g for 1/TIOAO), which is practically

independent of the level of 1-hexene incorporation. GPC

analysis confirms that the copolymer is of low molecular

weight (M_w) 18-33 000) and shows a narrow molec-

ular weight distribution, with M_w/M_n) 2.2-2.6. Notably

both 12r /TIOAO and 12m /TIOAO afford copolymers

with significantly higher 1-hexene incorporation than

1/TIOAO, reaching 31.2 wt % (entry 9, Table 5). The

copolymerization activity of this class of metallocenes

is higher than rac-EBIZrCl₂for both ethylene polym-

erization and ethylene/1-hexene copolymerization. The

reactivity toward R-olefins is, however, less, as reflected

by their respective reactivity ratios (Tables 4-6). rac-

EBIZrCl₂(entry 10, Table 5) affords copolymers with

ca. 3-7 times higher comonomer content.

metallocenes are less open than rac-EBIZrCl₂(r_E) 18;

r_H) 0.018) (entry 10, Table 6), which incorporates more

1-hexene. With both the 2,2′-bridged metallocenes and

rac-EBIZrCl₂, the comonomer distribution, evaluated on

the basis of r_Er_His in the range 0.2-0.5, indicating a

tendency to an alternate distribution.

Su m m a r y

A robust and convenient synthetic route to ethylene-

bridged bis(2-indenyl) zirconocenes has been developed.

This proceeded via successive Dieckmann condensations

to afford 2,5-dibenzyl-substituted adipates 3 and 18 as

the key intermediates. Incorporation of 1-alkyl and/or

4-phenyl substituents provided rac and meso metal-

locene isomers.

After activation with excess MAO, TIBAO, or TIOAO,

they were highly active for the polymerization of eth-

ylene to low molecular weight linear PE (activities up

to 4320 kg PE/(g Zr‚h)) and for the copolymerization of

ethylene with either propylene (activities of 1.3-12.5

tons copolymer/(g Zr‚h)) or 1-hexene (activities up to

1100 kg/(g Zr‚h)). The preference for ethylene incorpora-

tion was reflected in the high ethylene (r_E. 1) and the

very low propylene (r_P) or 1-hexene (r_H) reactivity ratios

with r_P) 0.002-0.008 and r_H) 0.004. In contrast, they

displayed poor activity (10-20 kg PP/(g Zr‚h)) for the

polymerization of propylene. The low activity was

(partially) attributed to the uncommonly high percent-

age 2,1-insertions from 1 (9.4%) and from 19 (9.9%) and

to consumption of the in-situ-generated hydrogen from

allylic activation. Analogous to adding ethylene, per-

forming the polymerization of propylene under H₂led

to a >100-fold increase in activity and afforded satu-

rated propylene oligomers with activities up to 6240 kg/

(g Zr‚h).

The diad and triad distributions in ethylene/1-hexene

copolymerization were determined for entries 4-10 of

Table 5. These are shown in Table 6.

The reactivity ratios r_Eand r_Hfor ethylene/1-hexene

copolymerization were determined from the peak inten-

sities in the copolymer ¹³C NMR spectrum and calcu-

lated using the equations⁴⁵r_E) 2[EE]/[EH]Ì and r_H

)

2[HH]Ì/[EH], where [EE], [HE], [HH] are the intensi-

ties of the diad sequences and Ì is the concentration

ratio of ethylene to hexene in the reactor. This yields

r_E) 182 (entry 4) and 235 (entry 5). These high r_E

values evidence the high reactivity of 1 toward ethylene

compared to 1-hexene and the difficulties in inserting

high amounts of comonomer. Lower r_Evalues, but of

the same order of magnitude, are obtained with 1-meth-

yl-substituted 12r and 12m . The presence of a substitu-

ent in the 1-position of the indenyl ligand, particularly

in 12m , presumably for steric reasons, facilitates ap-

proach of the R-olefin to the active site compared to 1.

These reactivity ratios suggest that the 2,2′-bridged

Exp er im en ta l Section

Gen er a l P r oced u r es. 1,1′-Biphenyl(2-indenyl)₂ZrCl₂¹¹(31)

22

and S(2-indenyl)₂ZrCl₂(32) were prepared according to the

22

(45) Cheng, H. C. Polym. Bull. 1991, 26, 325.

literature, and CH₂(2-indenyl)₂ZrCl₂(33) was prepared by

3448 Organometallics, Vol. 20, No. 16, 2001

Schaverien et al.

modification⁴⁶of the reported synthesis. Tris(2,4,4-trimethyl-

pentyl)aluminum (TIOA) and tris(2-methylpropyl)aluminum

(TIBA) were purchased from WITCO and used as a 1 M

solution in heptane. Diethyl adipate and 2-phenylbenzyl

bromide were purchased from Aldrich and used as received.

Claisen alkali solution was prepared by slowly adding 1050 g

(18.75 mol) of KOH pellets to 750 mL of water in a 3 L

Erlenmeyer flask with stirring. Ca u tion : Solution becomes

hot. After allowing the aqueous KOH solution to cool, it was

diluted with MeOH to a total volume of 3 L. Elemental

analyses were performed at Analytische Laboratorien, Lindlar,

Germany. Given the close congruence of these zirconocenes

with related titanocenes,¹²only a limited number of elemental

analysis were performed. ¹H NMR spectra of the zirconocenes

are given in the Supporting Information.

NMR. The ligands and metallocenes were characterized

using a Varian Gemini 300 (¹H NMR at 300 MHz, ¹³C at 75.4

MHz) unless otherwise stated. The solvents are as indicated,

and the measurements were performed at 25 °C. Polypropylene

and the EP copolymers were characterized by ¹³C NMR at

125.4 MHz on a Bruker 500. The samples were dissolved in

1,2,4-trichlorobenzene with some 1,4-C₆D₄Cl₂added as lock in

5 mm NMR tubes. The measurements were at 130 °C with a

70° pulse and a relaxation delay of 15 s. ¹H NMR spectra were

determined at 120 °C in 1,2,4-trichlorobenzene/1,4-C₆D₄Cl₂on

the Bruker 500 (512 scans, 5 s delay) and the chemical shifts

of the end groups referenced to the lowest field signal of 1,2,4-

trichlorobenzene at δ 7.32 ppm. The 1-hexene units in the

ethylene/1-hexene copolymers were determined at 120 °C on

a Bruker AC-200 spectrometer operating at 50 MHz. The

powder polymer samples were dissolved in C₂D₂Cl₄to give an

8% w/v solution. Peak assignments and composition calcula-

tions were according to Cheng.⁴⁵The triad and diad distribu-

and water, respectively (O₂level ∼0.3 ppm, H₂O level 0.5 ppm).

Each column is ca. 3 m × 15 cm. Nitrogen and propylene

feedstreams were continuously monitored at ppm level. AMS

and Systech analyzers were used to determine the oxygen

content. MCM analyzers were used to determine the water

content. RGA3 (reduction gas analyzer; Trace Analytical) was

used to determine the CO content (ppb range). The propane/

propylene ratio in the gas cap was determined at 1 min

intervals using an Interscience GC. Hydrogen was measured

by an Orbisphere hydrogen analyzer at 15 s intervals during

the polymerizations.

Propylene polymerizations were performed in a 5 L jacket-

cooled steel autoclave equipped with mechanical stirrer and

loaded with 1.6 kg of liquid propylene. In the premix (in

toluene), 500 molar equiv of MAO was used and the premix

stirred in a drybox at 20 °C for a known time, between 1.5 h

and overnight. The scavenger MAO was injected into the auto-

clave containing the liquid propylene, and subsequently (5 min

later) the premix added. The reactor was held at 30 °C for 1

min and then the temperature raised with a temperature gra-

dient of 4 °C/min to 50 °C (pressure 19.5 bar). Polymerization

was continued for a further 60 min at 50 °C and then stopped

by rapidly venting the excess propylene to the incinerator.

Ethylene/propylene copolymerizations were performed in a

25 L reactor loaded with 7.5 kg of liquid propylene and with

an ethylene/propylene feed ratio in the range 5.5-6.1 mol %

ethylene (liquid phase) depending on the copolymerization. The

feed ratio was kept constant throughout the polymerization

by adding ethylene on demand. The total amount of MAO was

kept constant at 20 mmol (11.0 g of a 4.84 wt % Al solution in

toluene). The premix was with a Zr:MAO ratio of 1:500. The

scavenger MAO was injected into the autoclave and 5 min later

the premix added at 50 °C. The copolymerizations were

essentially isothermal, with typical initial reactor exotherms

of 1-5 °C being observed. The copolymerization was continued

at 50 °C for the time stated and then killed by the injection of

2.5 mL of MeOH. The excess ethylene and propylene were

vented to the incinerator, and the copolymer was isolated.

Monomer distributions were calculated as described.⁴⁹

Ethylene/1-hexene copolymerizations were performed at 70

°C similar to that described for ethylene polymerization except

that heptane was used as solvent. The total volume was 100

mL, and the amount of 1-hexene, in the feed, is given in Table

5.

tions as well as the kinetic parameters r_Eand r_P

were

H

or

calculated from ¹³C NMR analysis.⁴⁷Ethylene/propylene co-

polymer limiting viscosity numbers (LVNs) were measured at

135 °C in decalin, which was inhibited with 2,6-di-tert-butyl-

4-methylphenol (1 g/L). Polyethylene and ethylene/1-hexene

copolymers were measured at 135 °C in tetrahydronaphtha-

lene in the presence of 0.5 wt % 2,6-di-tert-butyl-4-methylphe-

nol. The polymer viscosity-averaged molecular weights (M_v)

can be obtained from the experimentally determined LVN

using the correlations reported.⁴⁸

P olym er iza tion Meth od ology. Ethylene polymerizations

were performed in a 200 mL glass autoclave equipped with

magnetic stirrer, temperature indicator, and ethylene-feed

line. The autoclave was flushed with ethylene, and 90 mL of

hexane was then added at 20 °C. The catalyst premix was

prepared by adding to 10 mL of hexane in the following

order: the cocatalyst, then in the case of TIOA or TIBA 0.48

equiv of H₂O, and finally after 5 min stirring the appropriate

metallocene dissolved in the minimum amount of toluene.

After another 5 min stirring, this premix was introduced into

the autoclave. The temperature was increased to 80 °C and

kept constant throughout the polymerization. The autoclave

was pressurized to 4.6 bar and the total pressure kept constant

by supplying ethylene on demand. After the time indicated

the polymerization was interrupted by removing from the

heating bath, immediately venting the reactor, and adding 1

mL of MeOH. The thus obtained polymer was washed with

acidic methanol, then with methanol, and dried at 60 °C under

vacuum. Propylene (gas, 9 bar) was purified by passing it over

a BTS column containing a copper catalyst (BASF R3-11), as

well as a column containing a mixed bed of molecular sieves

(3 Å (bottom), 4 Å (mid part), and 13X (top)) to remove oxygen

P r ep a r a tion of 3 fr om Dieth yl Ad ip a te. NaOEt (487 g,

6.87 mol) was suspended in 3.54 kg of THF in a 10 L cylindrical

double-walled glass reactor equipped with overhead stirrer and

oil heating/cooling bath. This suspension was heated to 60 °C,

whereupon 1170 g (5.73 mol) of diethyl adipate was added

during 1 h. The reaction mixture was stirred for 16 h at 60

°C, and 820 g (6.41 mol, 1.12 equiv) of benzyl chloride was

added during 3.75 h. This was stirred for another 3.5 h at 60

°C and then allowed to cool. At 23 °C, 487 g (6.87 mol) of

NaOEt was added during 2 h and the reaction then slowly

warmed to 60 °C and stirred for 16 h. Then 879 g (6.87 mol)

of benzyl chloride was added during 1.75 h. The reaction

mixture was stirred at 65 °C for 6 h and then cooled to 30 °C,

whereupon an extra 100 g of NaOEt was added to effect

complete conversion of dibenzyl-substituted cyclopentanone 6

to 3. After 5 h at 60 °C, 207 g (1.62 mol) of benzyl chloride

was added during 1.75 h to convert unreacted NaOEt to

PhCH₂OEt. After 2 h at 60 °C, 224 g of EtOH was added. The

reaction mixture was allowed to cool to 25 °C and 2.8 L of

dilute HCl (pH ) 2) added to neutralize the bases. Toluene

(1.5 L) was added, and the organic and aqueous layers were

separated by draining out of the base of the reactor. The

solvents were removed on a rotary evaporator and excess

benzyl chloride and benzyl ethyl ether removed on a Schlenk

(46) See Supporting Information for the synthesis of 33.

(47) Uozumi, T.; Soga, K. Macromol. Chem. 1992, 193, 823.

(48) For PE, [η ] ) 3.8 × 10^-4M_v^0.725; for a-PP, [η ] ) 1.85 ×

10^-4M_v^0.74; for i-PP, [η ] ) 1.93 × 10^-4M_v^0.74. Moraglio, G.; Gianotti,

G.; Bonicello, U. Eur. Polym. J . 1973, 9, 623. Pearson, D. S.; Fetters,

L. J .; Younghouse, L. B.; Mays, J . W. Macromolecules 1988, 21, 478.

(49) Doi, Y.; Ohnishi, R.; Soga, K.Makromol. Chem. Rapid Commun.

1983, 4, 169.

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3449

line at 100 °C. This gave 2035 g (92% yield based on

diethyladipate) of ca. 95% pure 3 as a viscous oil that solidifies

on prolonged standing. The impurities are ca. 3% 4 and

cyclopentanone 5.

three-necked round-bottom flask in a heating mantel. These

were heated to 250-265 °C with a N₂purge to flush away the

evolved water. The diol melts at 180-200 °C. Each thermolysis

was monitored regularly by GC. The thermolysis was stopped

when conversion of 10 was complete. This was typically after

2-3 h and gave a melted “glass”. The best conversions were,

as determined by GC, a 2:11 ratio of 3:1, in which 2 was 60%

of the total products. After thermolysis, the five fractions were

combined by adding ca. 500 mL of CH₂Cl₂to each 1 L flask

with stirring and scratching to solubilize the melted glass. The

CH₂Cl₂-insolubles were isolated by filtration and were, sur-

prisingly, 2 by GC analysis. These were washed with a little

pentane to give 114 g of 98.95% pure (by GC) 2 as a white

powder. The CH₂Cl₂was removed under vacuum from the CH₂-

Cl₂-soluble fraction to give a slightly oily orange powder. This

was washed with pentane to give 252 g of 2 as a 96.0% pure,

light beige, powder. Total ) 366 g, 42% yield. 2: ¹H NMR

(CDCl₃): δ 7.5-7.1 (m, 8H), 6.63 (s, 2H), 3.39 (s, 4H, CH₂),

2.86 (s, 4H, CH₂) ppm. ¹³C NMR (CDCl₃): δ 150.1, 145.9, 143.4,

127.1, 126.7, 124.2, 123.8, 120.5, 41.5 (CH₂), 31.0 (CH₂) ppm.

11: ¹H NMR (C₆D₆, 400 MHz): δ 7.50 (t, 7.43 Hz, 2H), 7.27

(t, 2H), 7.12 (t, 2H), 7.03 (t, 7.43 Hz, 1H), 6.93 (t, 7.43 Hz,

1H), 3.97 (d, 5.4 Hz, 1H, f), 2.99 (dd, 6.60, 8.76 Hz, 1H, c),

2.87 (s, 2H, e), 2.40 (d, 15.37 Hz, 1H, c′), 2.26 (m, 1H, d ), 1.98

(m, 2H, a ), 1.35 (m, 2H, b) ppm. ¹³C NMR (CDCl₃, 125 MHz):

δ 24.8 (C_a), 25.5 (C_b), 38.0 (C_c), 38.3 (C_d), 40.1 (C_e), 42.9 (C_f),

118.3 (C_g), 123.1 (C_h), 123.4 (C_i), 124.6 (C_j), 124.7 (C_k), 125.7

(C_l), 125.8 (C_m), 125.8 (C_n), 136.3 (C_o), 141.9 (C_p), 142.1 (C_q),

142.6 (C_r), 145.0 (C_s), 146.3 (C_t) ppm. Assignments were from

¹³C-¹H HETCOR and ¹³C-¹³C INADEQUATE 2D NMR

spectra.

De-ester ifica tion of 3 to Dia cid 7. Compound 3 was

divided into three batches: 680 g (1.78 mol) of 3 was added to

a 3 L three-necked round-bottom flask equipped with an

overhead stirrer; 3 equiv (5.34 mol, 850 mL) of Claisen alkali

solution was added and the mixture heated to 90 °C. After 3

h GC shows that complete conversion had occurred. The

mixture was cooled and the MeOH removed under vacuum.

The aqueous solution was then washed with 800 mL ether to

remove organic impurities (from the synthesis of 3). The

viscous aqueous dicarboxylate solution was poured into a 5 L

beaker, diluted with an additional 1 L of water, and cooled to

10 °C. Concentrated HCl was slowly added until pH 1-2 was

reached. Diacid 7 precipitated and was isolated by filtration

and washed with 300 mL of water. Drying in a vacuum oven

at 70 °C to constant weight gave 546 g, 94% of diacid 7. ¹H

NMR (CD₃OD): δ 7.10 (m, 10H), 2.90 (m, 2H), 2.70 (m, 4H),

1.60 (m, 4H) ppm.

P r ep a r a tion of Acid Ch lor id e 8. To 546 g (1.674 mol) of

diacid 7 in a 2 L pear-shaped flask was added 13.75 mol (1004

mL) of SOCl₂. The resulting suspension was stirred for 16 h

at 20 °C. Gas evolution was rapid, but controlled. After heating

to 60 °C for 4 h, a homogeneous solution was obtained. ¹H

NMR spectroscopy showed complete conversion. The SOCl₂

was removed under vacuum to afford a viscous oil. This was

dissolved in 300 mL of toluene and then the toluene removed

under vacuum at 50 °C, to ensure complete removal of residual

1

SOCl₂. This afforded 8 as a viscous light brown oil. H NMR

(CDCl₃): δ 7.4-7.2 (m, 10H), 3.14 (m, 4H), 2.87 (m, 2H), 1.80

(m, 4H) ppm. ¹³C NMR (CDCl₃): δ 176.1 (CdO), 137.1 (C),

129.1 (CH), 128.9 (CH) 127.3 (CH), 58.5 (CH), 58.4 (CH), 37.7

(CH₂), 37.5 (CH₂), 28.3 (CH₂), 28.23 (CH₂) ppm. Two (diaste-

reo)isomers were observed.

P r ep a r a tion of Bis(in d a n on e) 9. To 537 g (4.02 mol, 30%

excess) of AlCl₃suspended in 700 mL of CH₂Cl₂in a 3 L three-

necked flask equipped with a large dropping funnel and kept

under nitrogen was slowly (during 3 h) added 571 g of 8

(prepared from 508 g of 7) dissolved in 1050 mL of CH₂Cl₂.

There was no significant temperature rise. Ca u tion : Copious

HCl evolution. After stirring for 16 h at 20 °C an orange

suspension was obtained. This was carefully poured onto

crushed ice. The yellow CH₂Cl₂layer was separated and the

water layer extracted with 4 × 250 mL of CH₂Cl₂. Upon drying

P r ep a r a tion of CH₂CH₂(2-in d en yl)₂Zr Cl₂(1). To 146 g,

0.566 mol of 2 in a 2 L three-necked round-bottom flask under

N₂was added 1 L of ether, and the mixture was cooled to -5

°C. Then 475 mL of a 2.5 M n-BuLi solution in hexane (2.1

equiv) was added over 2.5 h. A beige slurry of dianion was

obtained. The internal solution temperature rose to and was

maintained at 2-4 °C, by adjusting the rate of addition.

Ca u tion : ca. 1.2 mol (ca. 25 L) butane (bp -0.5 °C) is

liberated. After the addition was complete, the reaction was

with anhydrous MgSO₄

a clear red-orange solution was

obtained. This was filtered and the solvent removed in a

vacuum to afford a slightly sticky white solid. This was washed

with 500 mL of pentane to give 420 g, 93% of 9 as a white

powder. ¹H NMR (CDCl₃): δ 7.7-7.3 (m, 8H), 3.3 (dd, J ) 17.4,

8.1, 2H), 2.85 (m, 2H), 2.63 (m, 2H), 2.1 (m, 2H), 1.60 (m, 2H)

ppm. ¹³C NMR (CDCl₃): δ 208.6 (CdO), 153.8, 136.7, 134.8,

127.4 (CH), 126.6 (CH), 123.9, 47.5 (CH), 32.7 (CH₂), 29.3 (CH₂)

ppm. Two (diastereo)isomers in ca. 1:1 ratio were observed by

GC and ¹³C NMR spectroscopy.

P r ep a r a tion of Diol 10. 9 (414 g, 1.427 mol) was sus-

pended in 2.5 L of THF and 1 L of MeOH in a 5 L beaker.

This was cooled in an ice bath to 10 °C. NaBH₄(60 g, 1.58

mol) was added slowly during 3 h to this stirred suspension,

resulting in immediate gas evolution. This was stirred for 16

h at 20 °C, and then THF and MeOH were removed on a rotary

evaporator to give a viscous beige slurry. Water (2 L) was

added and the solution acidified using dilute HCl to pH 3 to

give a beige powder that was isolated by filtration. Drying to

constant weight in a vacuum oven at 60 °C gave 413 g of 10.

¹H NMR (CDCl₃): δ 7.4-7.05 (m, 8H), [5.13 (m), 5.05 (m), 4.87

(m), total 2H], 3.1 (m, 1H), 2.9 (m, 1H), 2.75 (m, 1H), 2.5 (m,

1H), 2.4-1.4 (m, total 8H) ppm.

1

stirred for a further 30 min. H NMR spectroscopy (THF-d₈)

showed clean conversion to Li₂[CH₂CH₂(2-indenyl)₂]. The

solution was reduced to ca. 1 L and cooled to -78 °C. A

suspension of 132 g (0.566 mol) of ZrCl₄in 500 mL of CH₂Cl₂

in a 2 L three-neck round-bottom flask, also cooled to -78 °C,

was added to the solution of the cooled dianion. This rapidly

gave a bright yellow suspension with considerable heat evolu-

tion. The reaction mixture was allowed to warm slowly to 20

°C and stirred for a further 16 h at 20 °C. All solvents were

then removed under vacuum to give a light yellow powder.

Pure 1 was separated from LiCl, and possible polymeric

products were separated by Soxhlet extraction with CH₂Cl₂.

This gave 197 g, 83% yield of 1. In a separate reaction, a

further 210 g of 2 was converted into 239 g of 1 (70% yield). A

total of 436 g of 1 was prepared from 356 g of 2. Li₂[CH₂CH₂-

1

P r ep a r a tion of CH₂CH₂(2-in d en e)₂(2). Quantities (64,

110, 234, 308, 300 g) of 10 were placed (separately) in a 1 L

(2-indenyl)₂]: H NMR (THF-d₈): δ 7.00 (dd, J ) 5.7, 3, 4H),

1

6.20 (dd, J ) 5.7, 3, 4H), 5.70 (s, 4H), 3.22 (s, 4H) ppm. 1: H

3450 Organometallics, Vol. 20, No. 16, 2001

Schaverien et al.

NMR (C₆D₆): δ 7.5 (dd, 4H), 6.95 (dd, 4H), 5.85 (s, 4H), 2.52

(s, 4H) ppm. ¹H NMR (CD₂Cl₂): δ 7.5 (dd, J ) 9.7, 4.8, 4H),

7.17 (dd, J ) 9.7, 4.8, 4H), 6.41 (s, 4H), 3.31 (s, 4H) ppm. ¹³C

NMR (CD₂Cl₂): δ 140.6 (2C), 129.4 (4C), 126.1 (4CH), 125.5

the toluene solution was separated by centrifugation and the

remaining solid extracted with 3 × 25 mL of toluene. The

toluene extracts were combined, concentrated, and crystallized

at -35 °C to afford 72 mg of 14r . The toluene-insoluble

precipitate from the reaction mixture was extracted with CH₂-

Cl₂, and the CH₂Cl₂removed in a vacuum to give a yellow

powder. This was washed with 2 × 5 mL of pentane to afford

300 mg of 14m .

(4CH), 103.1 (4CH), 31.1 (2 CH₂) ppm. Anal. Calcd for C₂₀H₁₆

-

ZrCl₂: C, 57.40; H, 3.85. Found: C, 57.29; H, 4.00.

P r ep a r a tion of CH ₂CH₂(1-Me-2-in d en e)₂(13). To a

suspension of 10 g (38.7 mmol) of 2 in 100 mL of THF was

added 32 mL of n-BuLi (2.5 M in hexane) in 15 min at 0 °C.

After stirring for 30 min at 20 °C the red-brown solution was

cooled to 0 °C, and a solution of 6.0 mL of MeI in 50 mL of

THF (also cooled to 0 °C) was added. The reaction mixture

was stirred for 2 h at 20 °C and then the solvent was removed

in vacuo. The sticky yellow-orange solid was dissolved in 200

mL of ether and extracted three times with saturated NaHCO₃

and saturated NaCl. The ether was removed in vacuo to afford

13 as a mixture of two double bond isomers in a ca. 10:1 ratio.

Yield: 9.7 g, 88%. NMR data for major double bond isomer.

¹H NMR (CDCl₃): δ 7.45-7.10 (m, 8H), 6.53 (s, 2H), 3.43-

3.30 (m, 2H, CHMe), 2.90-2.60 (m, 4H, CH₂-bridge), 1.36 (d,

7.5 Hz, 6H, Me) ppm. ¹³C NMR (CDCl₃): δ 154.7 (C), 148.8

(C), 144.0 (C), 126.5 (CH), 124.8 (CH), 123.9 (CH), 122.5 (CH),

120.0 (dCH), 46.2 and 46.0 (CHMe), 28.1 (CH₂-bridge), 15.8

(Me) ppm.

Li₂[CH₂CH₂(1-Et-2-indenyl)₂]: ¹H NMR (THF-d₈): δ 7.09

(dd, J ) 4.8, 4.2, 2H), 6.99 (dd, J ) 4.8, 4.2, 2H), 6.23 (m, 4H),

5.65 (s, 2H), 3.28 (s, 4H, CH₂), 2.96 (q, J ) 7.5, 4H, CH₂Me),

1.29 (t, 7.5, 6H, CH₂Me) ppm. 14r : ¹H NMR (C₆D₆): δ 7.43 (d,

J ) 9, 4H), 6.97 (t, J ) 6.6, 2H), 6.86 (t, J ) 6.6, 2H), 5.65 (s,

2H, dCH), 3.01 (dq, J ) 14.7, 7.5, 2H), 2.77 (d, J ) 9, 2H),

2.45 (dq, J ) 14.7, 7.5, 2H), 2.35 (d, J ) 8.4, 2H), 0.95 (t, J )

7.5, 6H) ppm. ¹³C NMR (THF-d₈): δ 137.3 (C), 130.84 (C),

128.75 (C), 127.0 (CH), 126.0 (CH), 125.5 (CH), 124.0 (CH),

122.75 (C-Me), 98.5 (CH), 27.4 (CH₂), 20.7 (CH₂), 16.1 (Me)

1

ppm. 14m : H NMR (CD₂Cl₂): δ 7.51 (m, 2H), 7.37 (m, 2H),

7.20-7.07 (m, 4H), 6.73 (s, 2H, dCH), 3.50-3.28 (m, 4H, CH₂

bridge), 3.05 (dq, J ) 14.7, 7.5, 2H), 2.80 (dq, J ) 14.7, 7.5,

2H) 1.14 (t, J ) 7.5, 6H) ppm. ¹³C NMR (CD₂Cl₂): δ 135.0 (C),

129.9 (C), 125.6 (C), 125.0 (CH), 124.3 (CH), 123.0 (CH), 122.9

(CH), 122.1 (C-Me), 100.8 (CH), 28.1 (CH₂-bridge), 20.0 (CH₂),

14.8 (Me) ppm.

Syn th esis of r a c- a n d m eso-CH₂CH₂(1-Me-2-in d en yl)₂-

Zr Cl₂(12r a n d 12m ). A 2.5 M n-BuLi hexane solution (4.1

mL) was added to 1.4 g of 13 in 50 mL of ether cooled to 0 °C.

After 45 min the ether was removed in a vacuum and the

dianion suspended in 80 mL of toluene at 20 °C in the drybox.

ZrCl₄(1.15 g, 4.93 mmol) as a slurry in 10 mL of toluene was

added, whereupon the reaction mixture changed from yellow

to dark brown. After stirring for 21 h at 20 °C, the toluene

solution was separated by centrifugation and the remaining

solid extracted with 3 × 25 mL of toluene. The toluene extracts

were combined, concentrated, and crystallized at -35 °C to

afford 300 mg of the rac isomer (12r ). The toluene-insoluble

precipitate from the reaction mixture was extracted with CH₂-

Cl₂, and the CH₂Cl₂removed in a vacuum to give a yellow

powder. This was washed with 2 × 5 mL of pentane to afford

230 mg of pure meso isomer (12m ). 12r : ¹H NMR (CD₂Cl₂): δ

7.54-7.34 (m, 4H), 7.21-7.10 (m, 4H), 6.16 (s, 2H, dCH),

3.63-3.45 (m, 2H, CH₂bridge), 3.20-3.03 (m, 2H, CH₂bridge),

2.48 (s, 6H, Me) ppm. ¹³C NMR (CD₂Cl₂): δ 137.7 (C), 129.8

(C), 128.9 (C), 126.2 (CH), 125.6 (CH), 125.3 (CH), 123.5 (CH),

116.3 (C-Me), 97.9 (CH), 27.5 (CH₂-bridge), 11.8 (Me) ppm.

Anal. Calcd for C₂₂H₂₀ZrCl₂: C, 59.18; H, 4.51. Found: C,

Syn th esis r a c- an d meso-CH₂CH₂(1-Me-2-in den yl)₂Zr Me₂

(15r a n d 15m ). Zirconocenes 12r and 12m (ca. 20 mg) were

suspended in C₆D₆in different NMR tubes. MeLi, as a white

powder, made by removing the ether from a MeLi solution,

was added to each NMR tube in the drybox. Occasional

shaking of the NMR tube gave within 2 h clean conversion to

15r and 15m , respectively. 15r : ¹H NMR (C₆D₆): δ 7.37 (m,

4H), 6.95 (m, 4H), 5.50 (s, 2H, dCH), 2.63-2.3 (m, 4H, CH₂

bridge), 2.09 (s, 6H, Me), -1.02 (s, 6H, Zr-Me) ppm. ¹³C NMR

(C₆D₆): δ 129.6 (C), 125.2 (CH), 124.8 (C), 123.4 (CH), 122.6

(CH), 110.0 (5 ring C-Me), 97.6 (5-ring CH), 40.1 (Zr-Me), 26.8

(CH₂-bridge), 10.8 (Me) ppm. 15m : ¹H NMR (C₆D₆): δ 7.42

(d, 2H), 7.30 (d, 2H), 7.04 (t, 2H), 6.9 (t, 2H), 5.93 (s, 2H, dCH),

2.6-2.3 (m, 4H, CH₂bridge), 2.03 (s, 6H, Me), -0.38 (s, 3H,

Zr-Me), -1.82 (s, 3H, Zr-Me) ppm. ¹³C NMR (C₆D₆): δ 129.2,

127.4, 125.6, 125.0, 123.7, 123.2, 121.9, 108.6 (5 ring C-Me),

97.8 (5 ring CH), 44.1 (Zr-Me), 37.0 (Zr-Me), 27.5 (CH₂-bridge),

11.6 (Me) ppm.

Syn th esis of CH₂CH₂(2-in d en yl)₂Zr Me₂(16). Zirconocene

1 was suspended in toluene and cooled to -78 °C. Two

equivalents of solid MeLi was added and the reaction mixture

allowed to warm to room temperature. Filtration and removing

the toluene in vacuo gave 16. ¹H NMR (C₆D₆): δ 7.33 (m, 4H),

6.8 (m, 4H), 5.73 (s, 4H, dCH), 2.29 (m, 4H, CH₂bridge), -1.02

(s, 6H, Zr-Me) ppm.

1

58.94; H, 4.57. 12m : H NMR (CD₂Cl₂): δ 7.50 (m, 2H), 7.38

(m, 2H), 7.24-7.06 (m, 4H), 6.71 (s, 2H, dCH), 3.55-3.25 (m,

4H, CH₂bridge), 2.44 (s, 6H, Me) ppm. ¹³C NMR (CD₂Cl₂): δ

137.1 (C), 130.3 (C), 128.1 (C), 126.3 (CH), 125.8 (CH), 125.3

(CH), 123.6 (CH), 116.6 (C-Me), 101.0 (CH), 28.9 (CH₂-bridge),

11.79 (Me) ppm.

P r ep a r a tion of CH₂CH₂(1-Et-2-in d en e)₂. 2 (1.36 g, 5.26

mmol) was dissolved in 25 mL of THF and cooled to -10 °C.

n-BuLi (5.5 mL) was added and the orange-brown solution

stirred for 2 h at 20 °C. This was added via a cannula to a

solution of EtI (1.1 mL, 7.1 mmol) dissolved in 10 mL of THF

at -10 °C. After 2 h the THF was removed under vacuum,

and ether added. This ether solution was washed three times

with saturated NaHCO₃and with saturated NaCl. After drying

over MgSO₄, 1.46 g of CH₂CH₂(1-Et-2-indene)₂was obtained

as a yellow oil. ¹H NMR and GC spectroscopy showed the three

possible double bond isomers.

Syn th esis of r a c- a n d m eso-CH₂CH₂(1-Et-2-in d en yl)₂-

Zr Cl₂(14r a n d 14m ). n-BuLi (4.5 mL, 7.2 mmol) was added

to 1.16 g (3.5 mmol) of CH₂CH₂(1-Et-2-indene)₂suspended in

50 mL of ether at -10 °C. After 45 min the ether was removed

under vacuum and the dianion suspended in 80 mL of toluene

at 20 °C. ZrCl₄(1.16 g, 4.93 mmol) as a slurry in 10 mL of

toluene was added, whereupon the reaction mixture changed

from yellow to dark brown. After stirring for 21 h at 20 °C,

Syn th esis of 2,5-Di(2-ph en ylben zyl)h exan e Diacid (18).

A solution of 43.8 mL (0.22 mol) of diethyladipate in 220 mL

of THF was added to a three-neck flask equipped with water

cooler. NaOEt (16.3 g, 0.24 mol) was added in small portions

at 20 °C. The reaction mixture was stirred at 60 °C for 16 h.

After cooling to 20 °C, 48.6 g (0.20 mol) of 2-(bromomethyl)-

biphenyl was added dropwise. This was stirred at 60 °C for

16 h to give 92% (by GC) of 2-(2-phenylbenzyl)diethyladipate

(17) as a viscous oil. NaOEt (14.9 g, 0.22 mol) and 2-(bromo-

methyl)biphenyl (44.4 g, 0.18 mol) were then added succes-

sively. Workup involved adding dilute HCl to neutralize the

base, and the THF was separated from the aqueous layer. The

aqueous layer was extracted three times with CH₂Cl₂. The

combined THF and CH₂Cl₂extracts were evaporated to dry-

ness to give 2,5-di(2-phenylbenzyl)diethyladipate (18) as a very

viscous oil. KOH (180 mL of 6.25 M) in H₂O/MeOH was added

and the reaction mixture refluxed for 16 h. The methanol was

removed on the rotary evaporator and the aqueous layer

washed with 2 × 200 mL of ether. The alkaline aqueous layer

was acidified at 0 °C with 37% HCl to pH 1. A white/yellow

powder precipitated, which was filtered off and washed with

Ethylene Bis(2-indenyl) Zirconocenes

Organometallics, Vol. 20, No. 16, 2001 3451

water and a little hexane. It was thoroughly dried to afford

77.0 g, 73% yield.

heated to 345 °C for 20 min under a slow stream of N₂. After

cooling, this afforded 20 as a yellow glass, which was isolated

as a yellow powder by dissolving in CH₂Cl₂and removing the

CH₂Cl₂in vacuo. Yield: 8.8 g, 87%. The product was isolated

as a mixture of two double bond isomers in a ca. 1:1 ratio.

Column chromatography on silica was used for additional

purification. Double bond isomer 1: ¹H NMR (CDCl₃): δ 7.59-

7.14 (m, 16H), 6.64 (s, 2H, CH), 3.46 (s, 4H, CH₂), 2.82 (s, 4H,

CH₂) ppm. ¹³C NMR (CDCl₃): δ 149.9 (C), 146.0 (C), 141.2 (C),

140.5 (C), 137.5 (C), 128.4 (2 × CH), 127.0 (2 × CH), 126.7

(CH), 124.5 (CH), 119.2 (CH), 41.1 (CH₂), 30.6 (CH₂) ppm. The

other double bond isomer has an olefinic CH at δ 6.76 ppm.

Syn th esis of r a c/m eso-CH₂CH₂(4-P h -2-in d en yl)₂Zr Cl₂

(19r , 19m ). To a solution of 465 mg (1.13 mmol) of 20 in 20

mL of ether was added dropwise 0.95 mL of n-BuLi (2.5 M in

hexane) at -78 °C. The reaction mixture was allowed to warm

to 20 °C. The ether was removed in vacuo to afford the dianion

as a light yellow powder. The dianion was suspended in

toluene and cooled to -20 °C, and 0.28 g of ZrCl₄was added.

After stirring for 16 h at 20 °C, the reaction mixture was

centrifuged. The remaining powder was washed once more

with toluene (20 mL) and extracted with CH₂Cl₂to separate

the product from LiCl. This afforded a mixture of rac and meso

isomers as a yellow powder. By crystallization from toluene,

19m could be completely separated from 19r . Isolated 19r was

contaminated with 20% 19m . Li₂[20]: ¹H NMR (THF-d₈): δ

7.74 (d, J ) 7.2, 4H), 7.28 (t, J ) 7.8, 4H), 7.10 (t, J ) 7.2,

2H), 7.05 (d, J ) 7.2, 2H), 6.35 (m, 4H), 6.08 (d, J ) 2, 2H,

CH), 5.85 (d, J ) 2, 2H, CH), 3.22 (s, 4H, CH₂) ppm. 19r : ¹H

NMR (C₆D₆): δ 7.70-6.90 (m, 16H), 6.37 (d, J ) 2.1, 2H, CH),

5.53 (d, J ) 2.1, 2H, CH), 2.48 (br s, 4H, CH₂-bridge) ppm. ¹H

NMR (CD₂Cl₂): δ 7.76-7.22 (m, 16H), 6.70 (d, J ) 2.1, 2H,

CH), 6.35 (d, J ) 2.1, 2H, CH), 3.39 (m, 4H, CH₂-bridge) ppm.

¹³C NMR (CD₂Cl₂): δ 140.5, 140.4, 139.6, 130.3, 129.1, 128.5,

127.6, 126.2, 125.9, 124.5, 103.9 (CH), 100.4 (CH), 30.8 (CH₂)

ppm. 19m : ¹H NMR (C₆D₆): δ 7.62-6.85 (m, 16H), 6.39 (d, J

) 2.1, 2H, CH), 5.95 (d, J ) 2.1, 2H, CH), 2.41 (m, 4H, CH₂-

bridge) ppm. ¹H NMR (CD₂Cl₂): δ 7.64-7.19 (m, 16H), 6.78

(d, J ) 2.1, 2H, CH), 6.41 (s, J ) 2.1, 2H, CH), 3.37 (s, 4H,

CH₂-bridge) ppm.

2-(2-P h en ylben zyl)dieth yladipate (17). ¹H NMR (CDCl₃):

δ 7.48-7.16 (m, 18H), 4.08 (q, 2H, CH₂), 3.97 (q, 2H, CH₂),

2.85 (d, 2H, CH₂), 2.37 (m, 1H, CH), 2.11 (t, 2H, CH₂), 1.45-

1.28 (q, 2H, CH₂and m, 2H, CH₂), 1.21 (t, 3H, Me), 1.08 (t,

3H, Me) ppm. ¹³C NMR (CDCl₃): δ 175.2 (CO₂Et), 173.1 (CO₂-

Et), 142.2 (C), 141.6 (C), 136.6 (C), 130.1 (CH), 129.7 (CH),

129.1 (CH), 128.1 (CH), 127.2 (CH), 126.9 (CH), 126.3 (CH),

60.14 (CH₂Me), 60.0 (CH₂Me), 46.0 (CH₂), 35.6 (CH₂), 33.8

(CH₂), 31.4 (CH₂), 22.4 (CH₂), 14.2 (CH₂Me), 14.05 (CH₂Me)

ppm.

2,5-Di(2-p h en ylben zyl)h exa n e Dia cid . ¹H NMR (THF-

d₈): δ 10.45 (s, 2H, OH), 7.39-7.05 (m, 18H), 2.76 (m, 2H,

CH₂), 2.60 (m, 2H, CH₂), 2.22 (m, 2H, CH), 1.30-1.00 (m, 4H,

CH₂) ppm. ¹³C NMR (THF-d₈): δ 176.5 (CO₂H), 143.7 (C), 143.2

(C), 138.5 (C), 131.0 (CH), 130.8 (CH), 130.4 (2 × CH), 129.1

(2 × CH), 128.2 (CH), 127.9 (CH), 127.0 (CH), 47.2 and 47.0

(CH), 36.3 and 36.2 (CH₂), 30.8 (CH₂) ppm. Two isomers were

observed.

2,5-Di(2-p h en ylben zyl)h exa n oyl Dich lor id e. To 77.0 g

(0.16 mol) of 2,5-di(2-phenylbenzyl)hexane diacid in a 1 L flask

equipped with a reflux condenser was added 150 mL (2.06 mol)

of SOCl₂. The white suspension was warmed to 50 °C and

became a yellow solution with concomitant evolution of HCl

and SO₂. After gas evolution had finished, excess thionyl

chloride was removed under vacuum. Toluene (100 mL) was

then added, and this was removed under vacuum to ensure

complete removal of SOCl₂. The acid chloride was isolated as

1

a viscous beige oil. Yield: 82.9 g, 100%. H NMR (CDCl₃): δ

7.55-7.20 (m, 18H), 3.07 (m, 2H, CH₂), 2.85 (dd, 2H, CH₂, J

) 15 Hz), 2.64 (m, 2H, CH), 1.36 (m, 2H, CH₂), 1.17 (m, 2H,

CH₂) ppm. ¹³C NMR (CDCl₃): δ 175.9 (COCl), 142.3 (C), 141.1

(C), 134.6 (C), 130.5 (CH), 129.9 (CH), 129.1 (2 x CH), 128.6

(2 x CH), 127.8 (CH), 127.4 (CH), 127.2 (CH), 57.25 and 57.08

(CH), 34.9 (CH₂), 28.2 (CH₂) ppm. Two isomers were observed.

Syn th esis of 1,2-Di(4-p h en yl-2-in d a n oyl)eth a n e (21).

To a suspension of 56.0 g (0.42 mol) of AlCl₃in 40 mL of CH₂-

Cl₂was added 82.9 g (0.16 mol) of 2,5-di(2-phenylbenzyl)-

hexanoyl dichloride in 250 mL of CH₂Cl₂at 20 °C. The reaction

mixture was poured onto ice 5 min after addition was complete.

The CH₂Cl₂layer was separated from the water layer and the

water layer extracted with 2 × 200 mL of CH₂Cl₂. The

combined CH₂Cl₂layers were washed with saturated NaHCO₃.

The CH₂Cl₂was removed in vacuo and the resulting powder

washed with ca. 50 mL of hexane. This afforded 63.3 g, 89%

yield of 21. ¹H NMR (CDCl₃): δ 7.86-7.10 (m, 16H), 3.36 (dd,

2H), 2.86 (m, 2H, CH), 2.63 (m, 2H, CH₂), 2.07 (m, 2H, CH₂),

1.55 (m, 2H, CH₂) ppm. ¹³C NMR (CDCl₃): δ 208.3 (CdO),

151.0 (C), 140.3 (C), 139.0 (C), 137.1 (C), 134.9 (CH), 128.6 (2

x CH), 128.4 (2 × CH), 128.0 (CH), 127.7 (CH), 122.8 (CH),

47.8 and 47.2 (CH), 32.7 and 32.4 (CH₂), 29.5 and 28.6 (CH₂)

ppm. Two isomers were observed.

Syn th esis of 1,2-Di(4-p h en yl-2-in d a n ol)eth a n e (22). 21

(20.0 g, 45.2 mmol) in 310 mL of THF was slowly added during

20 min via a cannula to a suspension of 1.85 g (48.7 mmol) of

LiAlH₄(1.09 equiv) in 80 mL of THF in a 1 L flask cooled to

0 °C. After the addition was complete, the reaction mixture

was stirred for 45 min at 20 °C. THF was removed in vacuo,

ice and dilute HCl were added, and the suspension was

acidified to pH 1. The suspension was filtered, washed with

water, and dried to give 18.3 g of 22 as an off-white powder.

Yield: 91%. ¹H NMR (CDCl₃): δ 7.50-7.10 (16H), 4.85 (m,

2H, CH-OH), 3.11 (dd, J ) 18, 2H, CH₂), 2.55 (m, 2H, CH),

2.22-1.40 (m, 8H, 3 × CH₂, 2 × OH) ppm. ¹³C NMR (THF-

d₈): δ 149.0 (C), 142.3 (C), 140.1 (C), 139.2 (C), 129.6 (2 ×

CH), 129.3 (2 × CH), 128.4 (CH), 127.8 (2 × CH), 123.8 (CH),

81.9 (C-OH), 56.9 and 52.6 (CH), 37.0 (CH₂), 33.4 and 33.2

(CH₂) ppm. Two isomers were observed.

Syn th esis of m eso-CH₂CH₂(4-P h -in d en yl)₂Zr Me₂(23m ).

Excess MeLi was added to 30 mg of 19m in C₆D₆to cleanly

1

afford 23m after 1 h at 20 °C. H NMR (C₆D₆): δ 7.48-6.89

(m, 16H), 6.24 (d, J ) 2.1, 2H, CH), 5.90 (d, J ) 2.1, 2H, CH),

2.28 (m, 4H, CH₂), 0.11 (s, 3H, Zr-Me), -1.83 (s, 3H, Zr-Me)

ppm.

P r ep a r a tion of CH₂CH₂(1-Me-4-P h -2-in d en e)₂(24). To

a solution of 67.8 mL of MeMgBr (3 M in ether) in 100 mL of

ether cooled to 0 °C was added 15 g (33.9 mmol) of 21 in 150

mL of THF in 45 min. A white/beige precipitate formed. The

reaction mixture stirred for 2 h at 20 °C and then was poured

onto ice, acidified to pH 1, and extracted with ether. The

organic layer was washed with saturated NaCl solution and

the solvent removed in vacuo. The diol was dissolved in CH₂-

Cl₂, p-toluene sulfonic acid was added, and the mixture was

stirred overnight at 20 °C. The solution was dried over MgSO₄

and filtered, and the solvent was removed in vacuo. Yield: 13.1

g, 69%. ¹H NMR (CD₂Cl₂): δ 7.6-7.2 (m, 16H), 3.43 (d, 4H,

CH₂, 5-ring), 2.69 (s, 4H, CH₂bridge), 2.08 (s, 6H, CH₃) ppm.

P r epar ation of CH₂CH₂(1-Me-4-P h -2-in den yl)₂Zr Cl₂(25).

To a solution of 0.54 g (1.23 mmol) of 24 in 20 mL of ether

was added 1.1 mL of n-BuLi (2.5 M in hexane) at -78 °C. The

reaction mixture was allowed to warm to 20 °C and the color

changed to orange-red. The reaction mixture was stirred for 2

h at 20 °C, after which the solvent was removed in vacuo. The

sticky product was washed with hexane, and the orange

dianion was used without further purification. To a suspension

of the dianion in 10 mL of toluene cooled to -30 °C was added

0.29 g of ZrCl₄. The reaction mixture was stirred for 16 h at

20 °C. The orange suspension was filtered and the filtrate

concentrated. ¹H NMR spectroscopy showed a 1:1 ratio of rac/

Syn th esis of CH₂CH₂(4-P h -2-in d en e)₂(20). 22 (11.0 g,

24.6 mmol) was added to a 1 L three-neck flask. This was

3452 Organometallics, Vol. 20, No. 16, 2001

Schaverien et al.

meso isomers. Washing with hexane gave a small quantity of

25r , albeit in a 4:1 rac:meso ratio. Crystallization from a CH₂-

Cl₂/hexane mixture gave 94% 25m . 25r : ¹H NMR (CD₂Cl₂): δ

7.7-7.1 (m, 16H), 6.44 (s, 2H, CH), 3.59-3.14 (m, 4H, CH₂

bridge), 2.34 (s, 6H, Me) ppm. 25m : ¹H NMR (CD₂Cl₂): δ 7.7-

7.1 (m, 16H), 6.90 (s, 2H, CH), 3.50-3.20 (m, 4H, CH₂bridge),

2.48 (s, 6H, Me) ppm. rac and meso assignments were based

on comparison with the cyclopentadienyl CH shift in 12r and

12m and the characteristic CH₂CH₂bridge pattern for the rac

and meso isomers.

yellow sticky solid washed with hexane to extract the gem-

disubstituted bis(indene). Isolated yield of 30 was 2.3 g, 24%.

P r ep a r a tion of CH₂CH₂(1,3-Me₂-2-in d en yl)₂Zr Cl₂(29).

To 1.47 g (4.68 mmol) of 30 in 30 mL of ether at -78 °C was

added 3.9 mL of n-BuLi (2.5 M in hexane). After 30 min the

reaction mixture was allowed to warm to 20 °C. The orange-

brown reaction mixture was filtered and the resulting powder

washed with hexane to give Li₂[30]. To a suspension of light

yellow Li₂[30] in 30 mL of toluene and cooled to -30 °C was

added a suspension of 0.95 g (4.08 mmol) of ZrCl₄in 10 mL of

toluene. The reaction mixture was allowed to warm to 20 °C,

whereupon the color changed to orange. The mixture was

stirred overnight and filtered, and the precipitate was ex-

tracted with 3 × 10 mL of toluene. The solvent was evaporated

in vacuo, and the solid was washed with hexane to afford 206

mg, 10.5% of 29 as an orange powder. Li₂[30]: ¹H NMR (THF-

d₈): δ 7.03 (dd, 4H), 6.25 (dd, 4H), 3.01 (s, 4H, CH₂bridge),

2.43 (s, 12H, CH₃) ppm. 29: ¹H NMR (CD₂Cl₂): δ 7.45 (dd,

4H), 7.16 (dd, 4H), 3.50 (s, 4H, CH₂bridge), 2.59 (s, 12H, CH₃)

ppm.

P r ep a r a tion of CH₂CH₂(1-i-P r -4-P h -2-in d en e)₂(27). In

a 250 mL flask containing 40 mL of a 2.0 M (9 equiv) i-PrMgCl

in ether solution and cooled to 0 °C was added dropwise during

15-20 min a solution of 4.0 g (9.04 mmol) of 21 in 80 mL of

THF. This was stirred for 90 min and then poured onto ice.

The organic layer was separated and the aqueous layer

extracted with CH₂Cl₂. The combined organic layers were

washed with a saturated NaCl solution and then dried over

MgSO₄to give the diol. This was dehydrated by adding 1.5 g

of p-toluene sulfonic acid to a CH₂Cl₂solution and stirring for

20 h at 20 °C. This was dried with MgSO₄to give a clear yellow

solution, which was filtered through a plug of silica to give

290 mg of 27. The low yield is due to the large excess of p-TsOH

Ack n ow led gm en t. We thank Olof Sudmeijer (SRT-

CA) and Isabella Camurati (Basell, Ferrara) for polymer

NMR measurements and assignments, J os Rijsemus

and Max van de Pas for the preparation of 2 kg of 3,

trainees Milroy van der Bor, Oscar Dofferhoff, and

Andre´ Blaauw for their experimental contributions, J an-

Dirk van Loon for UFF calculations, and Andrew Horton

for useful discussions regarding H₂generation and end

groups. We also wish to thank Basell Polyolefins,

Hoofddorp, for permission to publish this work.

1

inadvertently used. H NMR (CDCl₃): δ 7.52-7.13 (m, 16H),

3.36 (s, 4H, CH₂), 3.18 (septet, 2H, CH), 2.67 (s, 4H, CH₂-

bridge), 1.36 (d, 12H, Me) ppm. ¹³C NMR (CDCl₃): δ 145.5,

142.7 (C), 141.4 (C), 141.0 (C), 140.6 (C), 137.6, 128.5 (2 × CH),

128.3 (2 × CH), 126.9 (CH), 126.5 (CH), 124.2 (CH), 119.5 (CH),

40.4 (CH₂), 26.7 (CH), 21.1 (Me) ppm.

P r ep a r a t ion of CH₂CH₂(1,3-Me₂-2-in d en e)₂(30). To a

suspension of 9.7 g (33.9 mmol) of 13 in 100 mL of THF was

added 28.5 mL of n-BuLi (2.5 M in hexane) in 10 min at 0 °C.

After stirring for 90 min at 20 °C the red-brown solution was

cooled to 0 °C and 5.3 mL of MeI in 50 mL of THF (also cooled

to 0 °C) was added. The reaction mixture was stirred for 2 h

at 20 °C. Subsequently the solvent was removed in vacuo and

the sticky yellow-orange solid dissolved in 200 mL of ether

and extracted three times with saturated NaHCO₃and

saturated NaCl. The ether was removed in vacuo and the

Su p p or tin g In for m a tion Ava ila ble: Experimental de-

tails for the two alternative synthetic routes to 1, modified

1

synthesis of 33, and H NMR spectra of the zirconocenes. This

material is available free of charge via the Internet at

http://pubs.acs.org.

OM010160Z

Article Doi

Ethylene bis(2-indenyl) zirconocenes: A new class of diastereomeric metallocenes for the (Co)Polymerization of α-olefins

DOI: 10.1021/om010160z

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om010160z

Full text of DOI:10.1021/om010160z

Products guided by the article

R&D Labs maybe for 243146-42-1

Relevant to this article

Hot Product