Synthesis, characterization and reactivity of amido titanium and zirconium complexes

..,.,loumal

OlurganQ. .... m e t a l l i c

t:taemistry

ELSEVIER

Journal of O~anometallic Chemistry 525 ( !996) 81- 87

Synthesis, characterization and reactivity of amido titanium and zirconium complexes Helmut Mack, Moris S. Eisen * i)cparmwnt of Chemisto,. Technion-i.~'raelln.~tin,e of Technoh,gy. ttaifi~ 32000. Israel

Received 26 February 1996; revised 20 May 1996

Abstract

The polymerization of ethylene and propylene was studied by usi~g a :;cries of mone- .:nd spirocyc!ic ¢oordin:;tively un,,a~ur:~te.:le:~rly transition metal amides as pre-catalysts. Amido complexes were prepared either by metathesis reaction of Cp.ZrCI, and the dilithium salt of the bisamido ligand (MesSiN-(CH~):,-NSiM%)2 -. by tmnsamination reaction of Zr(NMe2)4 with two equivalents of the diamine ligand (Me:~SiNH-(CH~)a-NHSiMe 0, and by metathesis reaction of ZrCi a • 2THF or TiCi.t with equimolar amounts of the dilithium salt (Me ~SiN-(CH ~)2-NSiMe02 . The complexes obtained were characterized by standard spectroscopic technt~laes. "Cationic' polymerization catalysts were generated from the early transition-metal amides with methylalumoxane. Polymerization activity is dependent on catalyst and co-catalyst concentrations. gcvu',rd.~: Group 4i Metallocenes; Titanium: Zirconium; Polymerization; ot-olefins

I. Introduction

Currently, there is considerable imerc.~t in cationic d u C p : M ( R ) ' (M ~ Ti, Zr, I11") complexes its alternatives tO traditional Zieglcro~Natta catalysts in olefin poly~ merizatiun: fi~r recent reviews of olefin polymerization catalysts, see Refs. [i,2]; for representative recent isoo tactic polypropylene work, see Refs. [2.o5]: for representative recent syndiotactic polypropylene work, see Refs. [2,6]; Ibr recent thermoplastic elastomers work. see Refs. [2,7]; tot" recent theoretical studies, see Refs. [8-1 I]: for recent metallocene ion pairing studies, see Refs. [ ! ,2,4,5.12,13]; tot' recent polyethylene work, see Refs. [I 4,15]; for other recent metaliocene studies, see Ref. [16]• These species are highly reactive duc to (i) the electrophilicity of the d o cationic center, (it) the highly pol.'u'ized M - C bonds, which are innately reactive. (iii) the stntcturai bent metallocene, which restricts coordinatiort of substratcs cis to the M - R group, and (iv) the steric, electronic and chiral properties of the metal center, which may be tailored by modifying the Cp ligands. These properties lead to a rich insertion and or-bond metathcsis chemistry. K.'mlinsky el al. [! 7]. have

• Correspolldingauthor.

pioneered a new generation of bridged cationic Group (IV) metallocenes A (Fig. I), that can be used to produce high-density i~lyethylene and both, isotactic and syndiotactic polyolefins, tlniformity of t11o active sites in these metallocene-based systems leads to greater unilormity el tile polymer microstruclure, allowing in° corporation ~,f COomonom,.rs witl° regularity, its well as pro, luci:~ narrow molecular weight distributioa phiso tics. P,ecently, tile constrained geometry catalyst of the hall.sandwich metallocene, B (Fig. I)~4or theoretical studies see Refs. [9-11,18], produced ix)lymers with side branches well beyond the traditional (7:~ to C~ o¢-olefins. This long chain branching and the narrow molecular weight of tile polymers, that is characteristic l't~r metalloceneocatalyzed reactions, allows tile production of polymers which posses highly desirable i~rlor o mance properties with acceptable processability [19]. In conventional catalytic systems, methylalurnoxane (MAO) is typically used as the activator of metallocene dihalides. Alternatively, tile reaction of thc dialky~ rnctallocenes with triphenylcarbenium salts of weakly ccx)t'dinating anions, as B(C~,I:~).i~ or B((7~,Fs)~. in non-coordinative solvents leads qu.'mtitatively to cationic come plexes [i.2,4,5.12.13]. Thus, tile cationic Cp: M(R)' nlay be expected to establish a series of equilibria by interacting with arty available nucleophile, such as tile

0()22-328X/96/SI5.00 Copyright © 1996 Elsevier Science S.A. All rights reserved. Pll S0022-328X(96)06515- I

82

!1. MacL M.S. Eisen / Journal of Organometallic Chemistry 525 (1996) 81-87

Me2Si

~s~X

!

the effect of the amido ligand either as a counterion or as a spectator group when these amido complexes are reacted in polymerizations in the presence of MAO. These new catalytic precursors differ from the known metallocene complexes in the electron density around the metal center influencing the catalytic activity in the pc!ymerization of a-olefins.

M = Omup IV metals X =el. Me A B Fig. !. Uniform site pre-catalysts for olefin polymerization.

solvent, the anion, neutral metal alkyls, or the olefin substrate prior to polymerization [20]. In the absence of a coordinating solvent the cation-anion interactions are the most important of these interactions and have a dramatic influence on the catalytic activity. Pioneering work by Arnold and co-workers. Floriani and co-workers and Jordan and co-workers has shown the analogy between Cp2M and (N4-macrocycle)M- and (N20,macroeycle)M-compounds: for (N4=macrocycle)MR'2 and (N.,=macrocycle)M(R')+ ( M ~ G r o u p 4) complexes, see Ref. [21]: for (N,O,-chelate)MR 2 ( M = Group 4 and N~O,=chelate = acen. salen, salophen) complexes, see Ref. [22]. Despite the fact that cationic complexes of the macrocycles can be achieved in good yields, the reactivity of these complexes toward the polymerization of olefins is not observed. Our efforts are aimed to improved the catalytic polymerization pcrfom~anc¢ of the cationic Group (IV) corn° plexes by the incorporation of alternative o~donor Iig o ands, focusing on the steric congestion and enhancing the Lewis acidity at the cationic metal center, Thus, tl~e possibility of changing the chlorine or the cyclol~ntadienyl ancillary ligands in early transition metal complexes by araido Iigands (R~N °) pl~vides a way to change the electron-donating anionic equivalents and a possibility for tailoring steric hindrance. In this contribution, we report the preparation of a series of soluble mono- and spirocyclic amido zirconium complexes that constitute a new class of homogeneous polymerization catalysts. The complext,, prepared allow us to compare

2. Results and discussion

Bisamido monocyclic zirconium metallocene complex 1 was obtained by the reaction of equimolar amounts of (M%SiN-(CH2):-NSiM%)Li 2, obtained by the double deprotonation of the N,N'-bis(trimethylsilyi)ethylenediamine with n-BuLi in hexane, with zirconocenedichloride (CP2ZrC! 2) in THE This yielded a yellowish slurry and afforded, after evaporation, extraction into Et20 and cooling to -50°(2, orange crystals of the corresponding bis(amido)bis(v/5-cyclopen tadienyi)zirconium chelating complex 1 in 75% yield, as shown in Scheme 1. The IH NMR spectrum of complex 1 in benzene-d6 at 25 °C exhibits a resonance Cp ringlet at 8 -- 6.14 ppm compared with the Cp resonance 6 - 6.47ppm of zirconocene dichloride at the same temperature in the same solvent. A similar shift pattern is also obsel~ed for the I~C{H} NMR resonances of the Cp groups (1:112.9, Cp,ZrCl~: 116.0ppm). These shifts to higher field valties in 1 are a consequence of the inclx~ased electron density at the Cp ancillary ligands due In the electron donor cap'lcity of the chelating bis(amido) ligand compared with the two chlorine atoms in zirconocenc dichloride. Thus, we can cs~ci the Zr=N bond to be less strong than the Zr=CI bond, due to this ionic character (vide infra), and should allows us to activate the metallocene amido complex in the presence of Lewis acids. The 114 and L~C(H} NMR resonances related to the ethylene bridge of the bisamido ligand in I appear as singlets at 6 = 3.29ppm and t~ = 54,5 ppm respectively. Comparison with the corresponding ethylene ringlet shifts H 8 ~ 2.48 and ~C{H} ~$~ 45.4ppm off the flx~ediamine (Me~SiNH-CHz-CIol,-NHSiMe~)

it,.Si

i Zr"

I.THF/hcx~ulc

SiMc

÷

3 1

Scheme I, Synthesisroute to the monocyclicbismnidozitconocenecomplex I.

H. Mack. M.S. Eisen / Journal of Organomemllic Chemistry 525 ( 1996J 81-87

83

SiMe3 I

~

i) Me2NH

2

n-BuLi/hexane Me2X "20°C

it) Z I r C l a ( T H F ) 2

. . . . t~NMe2

~"

THF/hcxane

/

z~r~

hl-'t

- 4

I

Nil SiMe a

NMe2

SiMe 3

~

HNMe2

THF/hexane

Me2N

StMe 3

RT

tea N -.,.-~

N/ I

]

SiMe 3 SiMe 3

Scheme 2. Synthesispathwayto complex2. shows the expected lower field shifts in 1 caused by the attachment of the bisamido ligand to the Lewis acid zirconium atom. The t H and taC{H} NMR signals of the methyl resonances of the two Me3Si groups in I appear at 8 = 0 . 0 7 p p m and 8 = l.Oppm respectively with a similar downfield trend, compared with the corresponding signals, 8 -- - 0.14 ppm and 6 = - 0.2 ppm, of the free diamine; (recently, a similar cyclic tetraphenyl bisamido zirconium complex was obtained from Zr(ll), see Ref. [23]). Furthermore, structural data tbr early transition-metal amides show that the nitrogen atom of the amido ligand generally has a planar three-coordinate environment with short M-N distances due to possible P~r "* d,r electronic interactions [24]. The spirocyclic tetraamido zirconium complex 2 was synthesized in 82% yield at room temperature via the transanlination reaction of N , N ' ° b i s ( t r i m e t h y i silyl)elhylettedianfine will) ZttNMe2) 4 it)THF, the latter obtained by the metathesis reaction of ZtCI4.2THF with flmr etltfvalenl~ of LiNMe,, as shown in Scheme 2. Recryslalli/aliol) of COlllplex 2 ill Ilexatle lit =~75 °C afl'ordcd tile almlylically pure crystalline product; for the analogous Ti complex, see Ref. [25]. The electron density transfer fronl the two bisamido ligands towards the electrophilic zirconium atom in complex 2 is ascertained by the chemic:d shifts of the

( ci

.,,~

SMe ~ I N• L| 2 N • L~

)Me3

\M..,,,,

ethylene groups which ap~ar at 6 = 3.74ppm and 6 = 54.1 ppm in the I}1 and t C{H} NMR spectra respectively. A similar trend for the Me3Si groups in complex 2 (6 'H =0.07, 6 '3C=0.3ppm) is observed. These results, compared with those obtained for complex 1, show a similar electronic environment between the Cp and the bisamido ancillary iigands. Thus, if activation of the amido iigands in complex 2 with Lewis acids yielding the active cationic complexes is possible, the complexes obtained ate expected to have a similar reactivity to complex I [20]. Alternatively, complex 2 is also accessible by treatment of two equivalents of the dilithiated N,N'-disilylated ethylenediamine with ZtCI4. 2THF in THF and recrystallization from hexane at - 4 0 ° C with an overall yield of 54%, as shown in Scheme 3. Metathesis reaction of one equivalent of the dilithiated diamine with ZtCI 4 • 2THF or TiCI,~ forms tile monocyclic complexes 3 or 4 respectively. Fractional crystallization from cold hexane affords colorless mio crocrystals of 3 and orange microcrystals of 4 in 42% and 58% yield accordingly. Complex 3 shows a large electron density transfer from the bisamido ligand towards the metal atom, its indicated by the NMR s~ctra (6 'H ~ 3.61, 6 t~Clt ~ 69.3). For tile Ti analog 4, the NMR signals are in agreement with the electrophilicity

THF/hexane

SiM,~ N - I°,

,

_

(

Shie 3 I N- L,

,

MCI.4 •

2'1"!]I:-

THF/hexane

~S iM¢:1

?.

/ N

/

M¢~Si

Me3Si

M '''*l~ N-

SiM¢3

M=Zr(2)

M=Zr (3) M=Ti (4) Scheme 3. Synthesisroute to complexes2, 3 and 4.

84

!t. Muck. M.S. Ei.~'en/ Journal of Organometallic Chemistry 525 (1996) 8t-87

-® SiMe 3

-

"AI~ O-

SiMe3

-

N

,~.,,~.~'~"~l A,"~

toluene

_Z:" +

N

A

$iMe 3

SiMe3

i

1

Scheme 4. Formation of the active cationic complex of I

of the metal, inducing a downfield shift for both methylene and TMS groups respectively (8 ~H = 4.53, 0.25; ~C{H} ~ 53.8. - 1.6ppm). Although purification of metal amides by sublimation is a well-known experimental procedure [24,25]. intended purification of complex 2 by sublimation at 60°C/10-3Torr fi'om the reaction mixture leads to colorless decomposition crystals of the bis(~-bis(trimethylsilyl)amidoJdilithium compound [(TMS) 2NLi. THF] z. as confirmed by X-ray crystallography [26] (for the similar diethyl ether adduct. see Ref. [27]. Comparison of the thermodynamic bond disruption energies in metallocene zirconium dichloride and complex 1 (D(zroct) ~ 491 kJ mol ~t. D(zr=n)-355 kJ tool =t) and out" previous observation that early= late phosphidoo and arsenido-bridged heterobimetallic complexes can be activated by strong Lewis acids to produce the methyl 'cationic' complexes [28]. suggests that. in a similar fllshion, it would be possible to heterolytically activate complexes 1=4 by MAO and obtain active species for the polymerization of a,oolet'ins. The activation of complex i with MAO (Scheme 4) Table

leads to the active catalyst for the polymerization of a-olefins and according to the L~C{H}NMR spectra, to Cp, ZrMe + [ 1,2,4,5, ! 2,13]. Presumably, the formation of the cationic complex is formed by the double metathesis of the amido ligands and the methyl groups of the MAO in a similar way as found for metallocene phosphido ligands [29]. Comparison of the activity of the Cp.,ZrCI.,-MAO system with the I-MAO system shows a slight increase in the activity by using the amido ancillary ligand instead of the chlorine ligands under the same conditions [30]. The catalytic activity of complex I, with ethylene (Table !), shows the same trend as found for the analog Cp:ZrCH~-MAO. For polypropylene, the proton and carbon spectroscopic analysis of the polymers reveal only vinyl/iso-propyl end-groups with no -inylidene/n-propyl end-groups. Polymers containing these end=groups may be tbrmed from at least three different theoretical mechanisms. The first, involves an allylic C=It activation of propylene [13], the second. involves a /J-mclhyl elimination [131, and thirdly, a /]°hydrogen elimination li'ont a polymer chain in which

I

A¢livily data for the l)olytt~erization of ethylene and propylene with amido complexc,~ I =4 Run

Catalyst

I a (retool)

T ("C)

Olefm r,

MAO ~ (tnmol)

Polymer (rag)

I 2 3 4 5 6 7 8 9 I0 II 12 13 14 15 16 17

I [ I I I 2 2 ~3 3 ,I 4 4 4 I I I

3.54 3,54 3.54 3.54 3,54 3.02 3,02 3,02 5.49 5.49 5.49 4,07 4,b? 46? 3,54 ?,54 3,54

25 25 25 ~) 60 25 25 25 25 25 25 25 25 25 25 25 25

ethyle,e ethyk, n¢ ethylea~¢ ethylene ethylene ethylene ethylene ethyk, l~e eth>k',e ethylene cth',~lene ethyleue cth}len~: cthylctlc propene prolx'ne protx'ue

0,186 0,373 0.559 O.lS6 t),373 0,175 (),350 0,525 O,31t~ 0.636 0.954 0,270 0,541 0,811 O, 186 0,373 0,559

269 8~ 769 320 377 131 45 39 47(~ 3Sl 374 69S

577 536 322 588 361

CaI/MAO ratio I:I(X)) i:2(HX) 1:3(XX) I: IO(X) I:20(X) I:IO(X) 1:2(XX) 1:3(XX) I:1000 1:20(X) i:3000 I:IOIX) 1:2(}(X) 1:3000 I: I O(Kl 1:2000 1:30(X)

Activity d 2.28 x 7,53 x 6.52 x 2.71 x 3,20 x 3.01 × 1.24 × I,(18 x 7.23 x 5.78 x 5.68 x 1,24 x 1.03 X 9,50 x 1,09 x 1,99 X 1,22 x

I0 ~' !0 ~' IO¢' I0" It) ~' I0' I0 I0' I() ~ I0; I() ~ I0' I0 "~ I()' I ()~' I0 ~'

IIY'

" In 50ml toluel~e, e' Atmospheric pressure, " MAO, solvent rcmo~ed from a 20 wt,g solution m toluene (wltco) at 25 ~C/i0 ~' Ton'. d Grams total polymer/( tool Zr (tool I ~~) h arm),

H. Mack, M.S. Ei.~en/ Jemrnal of Organometallie Chemistry .525 (1996) 81-87

the last inserted monomer inserts in a "2-1' fashion. The last mechanism in our case can be rejected since no vinylidene/n-propyl end-groups are observed. The activation of the spirocycle zirconium complex 2 by MAO also leads to the formation of an active catalyst for the polymerization of ethylene, presumably the ten-electron cationic monocyclic zirconium methyl complex. In addition, the activation of complex 3 by MAO affords the same intermediate complex, despite the bond disruption energies, as shown by the IaC{H} NMR spectroscopy. The cationic complexes show signals at 50.1 ppm, 69.2ppm (broad), and a multiplet between 0-3ppm for the Zr-CH~, (N-CH2) 2, and AI-CH,a/TMS groups respectively. It is important to point out that similar chemical shifts appear for the Zr-CH~ signal in other metallocene cationic systems [1,2,4,5,12,13]. Table I shows the similarity in activity of complexes 2 and 3 (entries 6 - I ! ) for the polymerization of ethylene, with a slightly higher activity towards the chloride ancillary ligand. In addition, complexes 2 and 3 were found to be less reactive than complex 1 (entries 1-3). Furthermore, comparison of the relationship between activity and co-catalyst concentration shows that, in contrast to complex I and other metallocene complexes, in which higher activity is found with large catalyst:MAO ratios [14], complexes 2-4 show the opposite trend. Thus, higher MAO concentrations lowered the activity. This behavior is simihu" to that found in benzamidinate zitvonium complexes [31] and in earlylate heterobimetallic phosphine* and arsenide-bridged complexes [29]. This activity/MAO behavior is I~lieved to be a result of the large coordinative unsaturao tion and clectrophilicity of the metal center which drives the active 'cationic' complexes Io coordinate with the solvent or MAO [2]. The analogous Ti complex 4 shows slightly higher polymerization activity and a less pronounced effect with MAO. We have shown that, in addition to the known methyl and chloride ancilhu'y ligands in metallocene complexes, amido ligands can also be activated by strong Lewis acids, such as MAO, to produce similarly active cationic complexes. Furthermore, highly coordinative unsaturated monocyclic and spirocyclic amido complexes can also be activated for the polymerization of ethylene, as with the metallocenes although with lower activity. Albeit the use of B(C01~)~ as co-catalyst with complex I has shown no polymerization activity, the reactivity of complex I with other Lewis acid borane salts is currently being investigated.

3. Experimental section All manii~ulations of air-sensitive materials were pet'formed with the rigorous exclusion of oxygen and mois-

85

ture in flamed Schlenk-type glassware on a dual manifold Schlenk line, or interfaced to a high vacuum (10 -5 Torr) line, or in a nitrogen-filled 'Vacuum Atmospheres' glove box with a medium capacity recirculator (1-2 ppm O z ). Argon, ethylene, propylene and nitrogen gases were purified by passage through an MnO oxygen-removal column and a Davison 4 A molecular sieve column. Ether solvents (THF-d8) were distilled under argon from sodium benzophenone ketyl. Hydrocarbon solvents (toluene-ds, benzene-d6) were distilled under nitrogen from Na-K alloy. All solvents for vacuum-line manipulations were stored in vacuo over Na-K alloy in resealable bulbs. NMR spectra were recorded on Bruker AM 200 and Bruker AM 400 spectrometers. Chemical shifts for ~I-1 NMR and L~C NMR ate referenced to internal solvent resonances and are reported relative to tetramethyisilane. The NMR e×periments were conducted in Teflon valve-sealed tubes (J-Young) after vacuum transfer of the liquids in a high-vacuum line. n-BuLi (solution in hexane), Cp, ZtC! 2 and the ethylenediamine were purchased from Aldrich and used as-received. The dilithiated amine MeaSiNLi-(CH2) 2NLiSiM% was prepared from the freshly distilled respective diamine with n-BuLi in hexane [32]. LiNMe 2 was prepared from the con'esponding condensed amine and n-BuLi and used without fitrther purification. Zt(NMe,) 4 [32] was prepared by published procedures.

3. !. Synthesis of Cp2ZrlNSiMej-(CH2 ),=NSiMej ] ! Into a glove-box, a I(X)ml glass vessel w.is charged with 1.46g (5.0retool) of Cp:ZrCI, and i.08g (5.0 retool) of Me~SiNLi=(CII 2),=NLiSiMe~. 50 ml of THF were vacuum transferred into tile vessel at = 78'C on a high vacuum line; the temperature was slowly raised to ambient tem~rature and the solution stirred for 41,1h. The THF was then removed and 30ml of Et20 were added. After 2411 stirring, the precipilaled LiCI was filtered off through a C4 flit and the pale orange filtrate wits kept overnight at -50°C. Orange crystals were isolated while cold in 75% yield (I.59 g, 3.11retool, M = 423.86) and dried under high vacuum. Anal. Found: C, 51.04; H, 7.42; N, 6.12. C~xlt~,N2Si2Zr. Cale.: C. 50.00, H, 7.55; N, 6.6[)%. ~H NMR (2(X)Mloiz, Col)r,, 6ppm): 6.14 (s, 1011, Cp), 3.29 (s, 4H. Clt,=CH,). (I.07 (s, 18H, Si(CH,):~). ~C NMR (20OMHz, C01)~,, 6 ppm): I ! 2.9 (Cp), 54.5 (CH 2-Cil 2), 1.0 (Si(CH ~)~).

3.2 Synthesis of ZrlNSiMG-(CH: ),,-NSiMej ]: 2 (a) Into a glove-box, a 100ml glass vessel was charged with 1.89g (5.0mmoi) of ZrCI~. 2TItF and 2.16 g ( 10.0 mmol) of Me~SiNLi-(CH 2)2=NLiSiMe3 ' 50 ml of THF were vacuum transferred into the vessel tit - 7 8 ° C on a high vacuum line; the temperature was slowly raised to ambient temperature and the solution

86

H. Mack. M.,~ EL~'en/ Journal ¢~ Organometallic Chemistry 525 (1996) 81-87

stirred for 24h. The solvent was then removed and 30ml of hexane was added. After 24h stirring, the precipitated LiCI was filtered off through a C4 frit and the colorless filtrate was kept overnight at -50°C. Colorless crystals were isolated while cold in 54% yield (I.34g, 2.7retool, M-'-496.12) and dried under high vacuum. Anal. Found: C, ~.96; H, 8.89; N, 10.77. Ct6H44N4Si4Zr. Calc.: C, 38.74; H, 8.94; N, 11.29%. tH N ~ (200MHz, C6D6, 8 ppm!: 3.74 (s, 8H, CH 2e l l 2 ) . 0.07 (s, 36H, Si(CH3)3). " C NMR (20OMHz, CoD6. 8 ppm): 54.1 (CH2-CH2), 0.0 (Si(CH3)3). (b) Into a glove-box, 1.34g (5.0retool) of Zr(NMe 2)4 were weighed in a 100ml glass vessel. 50ml of THF and 2.04 g ( 10.0 mmoi) of Me3SiNH-(CH 2)2-NHSiMe3 were tr~sferred into the vessel at - 7 8 " C on a high vacuum line. The temperature was slowly raised to ambient temperature and the solution stirred for 24h. After reflux for I h the solvents were removed and 30ml of hexane was added. Cooling to -50"C, colorless crystals were formed which were filtered off while cold (82% yield) and dried under high vacuum.

3.3. Synthesis of CIzZr[NSiMe~-(CH," ),-NSiMe~ i 3 In a procedure similar to that used for 2, 1.89g (5.0retool) of ZrCI 4 . 2THF were treated with 1.08g (5.0retool) of Me~SiNLi~(CH~)2-NLiSiMe~, to give 0,77 g (2.1 retool, 42%) of colorles: complex 3 (M 364,57), Anal. Found: C, 22.56; H, 5.79; N, 7.28. CsH~CI~N~Si2Zr. Calc.: C, 21.97; H, 6.04; N, 7.68%. *H NMR (200MHz, C~D6, 8 ppm): 3.61 (s, 4H, CHa-CHa), 0.09 (s, 18H, Si(CH:~)~). )~C NMR (200MH~, C~,D~,, ~ ppm): 69.3 (CtI~CH~), 0,3 (Si(CH ~)~).

3,4, Synthesis of CI:TiiNSiMe~fCH: ),~NSiMes 1 4 In a procedure similar to that used for 3, 2.20ml (20,0retool) of TiCI4 were treated with 4.32g (20.0retool) of Me)SiNLi~(CH,)~-NLiSiMe~, to give 3.7g (ll.6mmol, 58%) of orange complex 4 (M 321.23). Anal. Found: C, 29.28; H, 6.90; N, 8.72. C~Hz~CI~N,.Si,Ti. Calc.: C, 29.91; H, 6.90; N, 8.70%. tH NMR (2~lVIHz, C~,D6, 8 ppm): 4.53 (s, 4H, CH , CH,), 0.25 (s, 18H, Si(CHa)a). t'~C NMR (200MH'z, C6D6, ~5ppm): 53.8 (CH,-CH,), - !.6 (Si(CHa)a).

3.5. Ethylene luglymerizution cxperbuents l'hese experiments were conducted in a I(X)ml flamed round-bottom reaction flask attached to a high°vacuum line. In a typical experiment, 1.5 mg of the catalyst I and 205rag of MAO (Zr:A! ~ I:1~)0) were charged into the flask containing a magnetic stir bar. The reaction vessel was connected to a high-vacuum line, pumped down and back-filled three times with argon. Then the flask was re-evacuated and a measured quan-

tity of toluene (40 ml) was vacuum transferred into the reaction flask from Na-K. After temperature equilibration, the gaseous ethylene was admitted to the vessel through a gas purification column. The gas pressure was continuously maintained at 1.0atm with a mercury manometer. Rapid stirring of the solution was initiated and after a measured time interval (2min) the polymerization was quenched by injecting a mixture of methanol-HCi. The polymeric product was collected by filtration, washed with hexane and acetone and dried under vacuum for several horns.

3.6. Propylene polymerization experiments This was performed in a similar manner to the ethylene polymerization. The polymerization was quenched after 5 min by adding a mixture of methanolHCi. The resulting oil was removed from the vessel. washed three times with CH2CI 2 and then the combined organic phases were dried over MgCI2 for several hours. After filtration, the organic solvents were removed under vacuum and the remaining colorless, viscous oil was dried under vacuum for 12 h.

Acknowledgements This research was supported by The Israel Science Foundation administered by The Israel Academy of Sciences and Humanities under Contract 76/94~1, and the Ministry of Science and the Arts, Ministry of Nicdersachssen, Germany, It.M. thanks the MaxPlanck-Gesellschaft for a MINERVA i~stdoctot~d I~1° Iowship.

References {I] (a) P.C. M6hring and J.N. Coville. J. Organmnct, Clwm., 479 (1994) I. (b) W, Kaminsky, C, zal. T~M,y, 20 (1994) 257. (c) R,P, Quirk (ed,), lhm,~ition Metal Cat, lyzcd Polymvricazi,ns. Cambridge University Press, Cambridge, 1988. (d) W, Kamino sky and tl. Sinn (eds,), 7?an,~itiem Mct, l,~.and Organomctallit:~ .Ibr Cut, ly.~t.~ fi : Of@in P, lymcri:~atiem, Springer, New York. 1988. (e) T. Keii and K, Soga (eds.), Cm, lyti~ Polymc~rt,:,t/m~ ,1" Ol¢:lbt,~. Elsevier, New York, 1986. (f) R.F. Jordan. P.K. Bradley, R,E. IoaPointe and D,F, Taylor. Nor J. Cho,.. 14 (19o0) 499, (g) R,F, Jordan, Adv, Orgammwt. Chem.. 32 (1991) 325, (h)T.J. Marks, Ace. Chem. Rc,~., 25 (1992) 57. [2] H,H. Brlatzmger, D, Fischer, R, Miilhaupt. B. Rieger and R.M. Waymouth, Angcw, C'hon, Int. FM. F,ngL. 34 (1995) 1143. [3] (a) W. Spaleck, F, Kiiber, A. Winter, J. Rohrmann. B. Bochmann, M, Antberg, V. Dolle and E.F. Paulus, Orgammwtedlics, 13 (1994) 954. (b) A. Razavi and J.L. Atwood, J, Am. Chore, Soc.. 115 (1993) 7529. [4] M,A. Giat~ello, M.S. Eisen, C.L. Stem and T.J. Marks. J. Am. Chem. Sot.. 117 (i995) 12114. [5] M.A, Giardello, M.S. Eisen, C.L. Stem and T.J. Marks, J. Am. Chcm, Sot,, 115 (1993) 3326.

H. Mack, M.S. Eisen / Journal of Organometallic Chemistry 525 (1996) 81-87

[6] (a) A. Razavi and J.L. Atwood, J. Organomet. Chem., 459 (1993) i 17. (b) M. Bochmann and SJ. Lancaster, J. MakromoL Chem. Rapid Commun., 14 (1993) 807. (c) ,I.A. Ewen, M.J. Elder, R.L. Jones, L. Haspeslagh and ,I.L. Atwood, MakromoL Chem. MakromoL Syrup., 4 8 - 4 9 (1991) 253. [7] (a) E. Hauptman, R.M. Waymouth and J.W. Ziller, J. Am. Chem. Sot'., 117 (1995) 11586. (b) G.W. Coat.es and R.M. Waymouth, &'ience, 267 (1995) 217. (c) B. Rieger, G. Jany, R. Fawzi and M. Steimann, Organometallics, 13 (1994) 647. (d) G.H. Llinas, S.H. Dong, D.T. Mallin, M.D. Rausch, Y.G. Lin, H.H. Winter and J.C.W. Chien, Macromolecules, 113 (1992) 1242. [8] (a) T. Yoshida, N. Koga and K. Morokuma, Organometailics, 14 (1995) 746. (b) L. Fan, D. Harrison, T.K. Woo and T. Ziegler, Organometallics, 14 (1995) 2018. (f) E.P. Bier,vagen, .I.E. Bercaw and W.A. Goddard, J. Am. Chem. Sot., !16 (1994) 1481. (g) G. Guerra, L. Cavallo, G. Moscardi, M. Vacatello and P. Corradini, d. Am. Chem. Soc., 116 (1994) 2988. (h) R. Meier, G.HJ. van Doremaele, S. larlori and F. Buda, J. Am. Chem. Sot., 116 (1994) 7274. [9] T.N. Doman, T.K. Hollis and B. Bosnich, J. Am. Chem. Sot., !17 (1995) 1352. [10] T.K. Woo, L. Fan and T. Ziegler, Organometallics, 13 (1994) 432. [I 1] T.K. Woo, L. Fan and T. Ziegler, Organometailic.~', 13 (1994) 2252. [12] P.A. Deck and T.,I. Marks, J. Am. Chem. Soe., !!7 (1995) 6128. [13] (a)L. Jia, X. Yang, A. Ishihara and T.J. Marks, Organometallies, 14 (1995) 3135. (b) X. Yang, C.L. Stem and T.J. Marks, J. Am. Chem. Soc., 116 (1994) 10015. (c) M. Boehmann, S.J. Lancaster, M.B. Hursthouse and K.M.A. Malik, Organtmaemilies, 13 (1994) 2235. (d) J.C.W. Chien, W. Song and M.D. Rausch, J. Polym. &'i. Part A: Polym. Chem., 32 (1994) 2387. [14] C. Janiak, U. Versteeg, K.C.H. Lange, R. Weimann and E. Itahrl, ,I. Organomet. Chem., 501 (1995) 219. [15] (a) G. Erker, M. Aulbach, C. Kr|lger and S. Werner, J. Organomet. Chem., ,I50 (1993) I. (b) M. Farina, G. Disilvestro and P. Sottanill, Macromoh','uk's, 26 (1993)946. ( c ) W . Km, lnsky, A. Bark and R. Sleigcr, .I. Mol. ('utul., 74 (1992) 1119. (d) J.C.W, Chien, G.II. L!inas, M.D. Rausch, Y.G. Lin, ILH. Winter, J.L. Atwood and S.G. Bolt, J. Polym. Svt. Part A: Poly,t. Chem.. 30 (1992) 2601. [16] (a) W. Kaminsky, Ang('|t'. Makron=oL Chem., 223 (1994) I01. (b) F. Langhauser, J. Kerth, M. Kersting, P. K~lle, D. bilge and P, Mi~ller, Angew. Makromol. Chem., 223 (1994) 155. (c) M.L.H. Green and N. ishihara, J. Chem. Soc. Dahon Trans., (1994) 173. (d) H.G. All, W. Mdius and S.J. Palackel, J. Organmnet. Chem., 472 (1994) 113. (e) G. Erker, C. Mollenkopf, M. Grehl, R. Frtihlich, C. Krtiger, R. Noe and M. Riedel, Organometallics, 13 (1994) 1950. (f) J.A. Canisll and H.W. Turner (Exxon) Patent Int. Appl. W092/12162, July 23, 1992. [17] W. Kaminsky, K. Kulpcr, IoI.H. Brintzinger and P.W.R.F. Wild, Angtz, w. Chem. int. Ed. Engl., 24 (1985) 507.

87

[18] (a) J.C. Stevens, FJ. Timmers, D.R. Wilson, G.F. Schmidt, P.N. Nickias, R.K. Rosen, G.W. Knight and S. Lai, European Patent Application EP416815A2, March 13, 1991. (b) J.A. Canich, PCT Application WO 91/04257, April 4, 1991. (c) D.D. Deyore, European Patent Application EP514828Al, November 25, 1992. (d) S.Y. Lai, ,I.R. Wilson, R. Collier, G.W. Knight and J.C. Stevens, PCT Application WO93/08221, April 29, 1993. (e) D.D. Devore, F.J. Timmers, D.L. Hasha, R.K. Rosen, T.J. Marks, P.A. Deck and C.L. Stem, Organometallics, 14 (1995) 3132. (0 P.J. Shapiro, W.D. Cotter, W.P. Schaefer, J.A. Labinger and .I.E. Bercaw, J. Am. Chem. Sot., 116 (1994) 4623 and references cited therein. [19] (a) Plast. Technol., Sept. (1992) 25. (b) Chem. Bus., Oct (1992) 15. (c) Mod. ~)last., Jan. (1993) 57. (d) Chemicalweek, Feb. (1995) 8. [20] (a) M. Bochmann and SJ. Lancaster, J. Organomet. Chem., 434 (1992) CI. (b) M. Bochmann and AJ. Jaggar, J. Organomet. Chem., 424 (1992) C5. [21] (a) H. Brand and J. Arnold, Angew. Chem. hu. Ed. Engl., 33 (1994) 95. (b) H. Brand and J. Arnold, Organometallics, 12 (1993) 3655. (c) ,i. Arnold, S.E. Johnson, C.B. Knobler and M.F. Hawthorne, J. Am. Chem. Sot., il4 (1992) 3996. (d) R. Uhrhammer, D.G. Black, F.G. Gardner, J.D. Olsen and R.F. Jordan, J. Am. Chem. Sot., 115 (1993) 8493. [22] (a) E.B. Tjaden, D.C. Swenson, R.F. Jordan and J.L. Petersen, Organometallics, 14 (1995) 371. (b) F. Cobrazza, E. Solari, C. Floriani, A.G. Chiesa-Villa and C. Guastini, J. Chem. Soc. Dalum Trans., (1990) 1335 and references cited therein. [23] C. Lefeber, P. Arndt, A. Tillack, W. Baumann, R. Kempe, V.V. Burlakov and U. Rosenthal, Organometailics, 14 (1995) 3090. [24] (a) E. Brady, J.R. Telford, G. Mitchell and W. Lukens, Acta Crystalh~gr. Sect. C:, 51 (1995) 558. (b) F.G.N. Cloke. P.B. Hitchcock and J.B. Love, J. Chem. Soe. Dalum Trans., (1995) 25. (c) Z. Duan and J.G. Verkade, lnorg. Chem., 34 (1995) 431 I. (d) D.J. Brauer, H. Biirger, E. Essig and W. Geschwandto ner, J. Or~tanmnct. Chem., 190 (1980) 343. (e) M. Schlingmann and U. Wannagat, Z. Anorg. Allg. Chem., 419 (1976) 115. (f) P.J. Walsh, A.M. Bar,roger and R.G. Bergman, J. Am. Chem. Sin'., 114 (1992) 1708. (g) R.A. Ander~n, D.B. Beach and W.L. Jolly, Inorg. Clwm., 24 (1985) 4741. [25] W.A. Herrmann, M. Denk, R.W. All~ach, J. Behnl and E. Ilerdlwcck, Chcm. Bcr., 12,I ( 1991 ) 083. [26] H. Muck and M.S. Eisen, in preparation. [27] M. Lapperl, M.J. Slade, A. Singh, J. Atwt~l, R.D. Roger~ ~md R. Shakir, .I. Am. Chem. Sot'., 105 (1983)302. [28] F. Lindenberg, T. Shribman, J. Sider, E. Hey°Hawkins and M.S. Eisen, J. Organomet. Chem., ill press. [29] E. ttey.Hawkins, M.F. Lappert, J.b. Atwood and S.G. Boll, J. Chem. Sot'. Dalum Trans., ( i 991 ) 939. [30] (a) G. Henrici-Olive and S. Olive, Chem. li,¢hnoL. (1981) 746. (b) H. Sinn, W. Kaminsky, H.-J. Vollmer and R. Woldl. Angew. Chem. h~t. Fd. Engl., 19 (1980) 390. [31] D. Herskovics-Korine ~,nd MS. Eisen, J. Organomct. Chcm.. 503 (1995) 307. [32] H. Mack, Dis.wrtation, University of Mtmich. 1995.