The molecular weight distribution of a polymer affects its properties, in particular its mechanical strength and processing properties. Mechanical strength is to a large extent determined by the high molecular weight fraction, whereas extrudability is determined by the low molecular weight fraction. As a result, polyolefins having improved mechanical and processing properties may be obtained if the molecular weight distribution is tailored to the end use of the polymer.

Several chromium catalysts are known for use in the preparation of polyolefins. For example, polymerisation of a-olefins such as ethylene by means of a catalyst comprising a cyclopentadienyl chromium complex is described in DE-A-19630580. Such catalysts produce polyethylene having a broad molecular weight distribution which is well suited for making moulded articles, blown films, extruded pipes, etc. However, the properties of such products could be significantly improved if it were possible to specifically tailor the properties, especially the molecular weight distribution, of the polyolefin through the use of specific catalyst materials, e. g. a blend of catalyst materials exhibiting different polymerisation properties. In this way, greater control over the properties of the resultant polymers may be achieved.

One proposed solution to this problem is the use of olefin polymerisation catalysts derived from iron and

cobalt complexes bearing 2,6-bis (imino) pyridyl ligands as described by Britovsek et al., Chem. Commun. 1998, 849-850. However, there is an ongoing need to identify other catalyst materials capable of providing polyolefins having new and/or improved performance parameters, and it is highly desirable to produce new olefin polymerisation catalysts which have different polymerisation properties and which may be blended for use in a tailor made catalyst system. It is an object of the present invention to provide novel catalysts useful in such'catalyst systems.

Viewed from one aspect the invention provides an olefin polymerisation catalyst comprising at least one metal complex comprising a catalytically effective group 6 metal, preferably chromium, and a ligand capable of forming at least two electron donor bonds and at least one further single atom to single atom bond (e. g. a covalent sigma bond, a covalent n bond, a carbene: metal bond etc.) to said metal.

As used in respect of the invention described herein, the term"polymerisation"includes both homopolymerisation and copolymerisation.

In a further aspect the invention provides a process for the polymerisation of an olefin, preferably an a-olefin, for example a C26 a-olefin, in particular ethylene or propylene, or a mixture of olefins, by means of a catalyst comprising one or more metal complexes as defined above.

Ligands present in such metal complexes may be any organic ligands capable of forming bonds to the catalytically effective group 6 metal as defined above, and are conveniently tri-or tetra-dentate, preferably tridentate. Preferred ligands include tridentate "pincer"ligands, for example comprising at least one carbon or nitrogen atom able to form a sigma bond with the group 6 metal, and two or more nitrogen or phosphorus donor atoms capable of coordinating to the

metal by electron transfer. Especially preferred are tridentate NCN-and PCP-type pincer ligands, such as those described by Ohff et al., J. Am. Chem. Soc. 1997, 119,11687-11688 and by Lee et al. Organometallics 1998, 17 (1), 1-3. Typically such ligands will comprise a 5- or 6-ring membered, preferably unsaturated, carbocyclic or heterocyclic group in which at least two of the ring atoms are substituted by organic groups carrying donor atoms, preferably nitrogen or phosphorus donor atoms.

Any remaining ring atoms may be substituted, e. g. by halogen atoms,'alkyl, aryl or aralkyl groups.

More generally, metal complexes useful in catalysts in accordance with the invention include those of formula (I): in which M is chromium, molybdenum or tungsten, preferably chromium; Y is CH, N or an optionally substituted, saturated or unsaturated, 5-or 6-ring membered carbocyclic or heterocyclic group, preferably an aromatic group, in each case substituted by Z, preferably in the o, o'- positions; or Y is a carbene moiety, preferably a cyclic carbene, e. g. a heterocyclic carbene, substituted by Z; each Z is independently-(CRl2) n-NR2R3,-(CRl2) n-PR2R3, - (CR1)-AsR2R3 ~ (CR12) n-SbR2R3,-(CRl2) p-CR =NR, or a carbene moiety, preferably a cyclic carbene, e. g. a heterocyclic carbene; Ru ils hydrogen, an optionally substituted alkyl group, preferably Cl6 alkyl, e. g. methyl, an optionally substituted 5-to 10-ring membered carbocyclic or heterocyclic group, e. g. an aromatic 5-or 6-ring membered monocyclic or 8-to 10-ring membered bicyclic

group (e. g. phenyl, naphthyl, Cl6 alkyl substituted phenyl, optionally substituted ferrocenyl etc.), or two RI groups together with the carbon atom or atoms to which they are attached form an optionally substituted carbocyclic group; R2 and R3 are each independently hydrogen, an optionally substituted alkyl group, preferably Cl6 alkyl, e. g. methyl, an optionally substituted 5-or 6- ring membered carbocyclic or heterocyclic group, e. g. an aromatic group, or R2 and R3 together with the intervening heteroatom form an optionally substituted heterocyclic ring; R'is hydrogen or an optionally substituted alkyl group, preferably Cl6 alkyl, e. g. methyl, or an optionally substituted 5-or 6-ring membered carbocyclic or heterocyclic group, preferably an aromatic group, e. g. phenyl or optionally substituted ferrocenyl (e. g.

(CsH4) Fe (CsHs)); Q is an organic or inorganic group; m is 2 or 3, preferably 2; n is 0,1,2 or 3, preferably 1; p is 0,1,2 or 3, preferably 0.

A wide range of substituents may be present in the ligands of such metal complexes of the invention, with different substituents affecting their final properties as catalysts. Typical substituents which may be present include halogen atoms, amide, sulphoxy, ferrocenyl, alkyl, aryl and aralkyl groups optionally substituted by alkyl, amino, amido, hydroxy, alkoxy or carboxy groups.

Examples of particular substituents include methyl, ethyl, isopropyl, n-propyl, t-butyl, n-butyl, phenyl and benzyl.

Cyclic carbene moieties which may be present in complexes of the invention include any cyclic carbene capable of coordinating to the metal. Typical such carbenes are heterocyclic with the C: providing one ring atom, and preferably comprise an unsaturated ring. In

general, the atoms adjacent to the C: will be substituted, preferably with bulky substituents containing up to 30 non-hydrogen atoms, preferably at least 4 non-hydrogens, e. g. containing 4 to 12 carbon atoms. More preferably, such carbenes comprise a 5-ring membered, preferably mono-unsaturated, heterocyclic group which contains 2,3 or 4 ring nitrogens, two of which are optionally substituted, and 1,2 or 3 ring carbons, one of which (the C: atom) is adjacent to at least one ring nitrogen and is unsubstituted, with any remaining ring'carbon optionally being substituted.

Thus for example the carbenes may comprise the following skeletal structures (IIa) or (IIb) where each X may independently represent N or an optionally substituted CH group.

Carbenes of skeletal formula (IIa), especially where the ring nitrogens are substituted by bulky structures, by Z groups or by-LYZ groups (where L is a bond or linker group), and more especially where each X is carbon, are particularly preferred.

The carbenes of formula (IIa) or (IIb) are sometimes referred to as Arduengo carbenes (as opposed to the Fischer and Schrock carbenes which are more commonly encountered in publications relating to organometallic complexes). Arduengo carbenes tend to be more stable-if they dissociate from a metal complex they usually have sufficiently large half lives to re- enter the metal's coordination sphere. Such stability facilitates the synthesis of substituted derivatives, allowing greater freedom to modify the electronic and steric properties of the carbene. Moreover, Arduengo

carbenes tend to bind efficiently to metals whether in low or high oxidation states, as opposed to Fischer and Schrock carbenes which favour low and high oxidation states respectively; this property of Arduengo carbenes is advantageous in polymerization catalysis where oxidation state changes may occur. Arduengo carbenes are efficient 2-electron donor ligands (comparable to P (CH3) 3 or P (C6Hll) 3) with no tendency to act as n acceptors-again in contrast to Fischer and Schrock carbenes-and may be used in place of phosphine ligands.

The strong metal: Arduengo carbene bond strengths (comparable to or greater than metal: phosphine bond strengths) mean that the complexes are thermally robust.

Accordingly such carbenes have good catalyst lifetimes and thermal stabilities.

Many such carbenes of formula (IIa) or (IIb) are already known as ligands, e. g. compounds having the following skeletal structures (i. e. omitting ring substituents): (where n is from 1 to 6).

The carbene ring nitrogens and ring carbons may

carry a very large range of substituents, with different substitution patterns resulting in variations in the properties of the resulting catalyst.

Thus, for example, substituents may be selected from halogen atoms, non-carbon oxyacid groups and derivatives thereof, and optionally substituted alkyl and aryl groups, e. g. such groups substituted by groups selected from alkyl, aryl, amino, hydroxy, alkoxy, oxo, oxa, carboxy, thia, sulphur oxyacid and halo groups and combinations thereof. Examples of particular ring substituents include for example methyl, ethyl, iso- propyl, t-butyl, n-butyl, cyclohexyl, phenyl, benzyl, methylbenzyl, mesityl, methylnaphthyl, ethoxyethyl, diphenylmethyl, benzyl, ethylaminoethyl, diethylamino- methyl, chlorophenyl, adamantyl, dihydroimidazol- ylidinylmethyl, dihydropyrazolylidinylmethyl, 2,6- diisopropylphenyl and dihydrotriazolylidinylmethyl groups.

In cases where Y in formula (I) is a carbene moiety, this is preferably a N, N-heterocyclic ("Arduengo") carbene in which the N atoms are each substituted by a group Z. The remaining ring atoms are optionally substituted, e. g. by groups Rl. Where two R1 groups are on neighbouring ring atoms, they may together form an optionally substituted carbocyclic group, preferably an aromatic group, e. g. a fused benzene ring.

Particular examples of suitable carbenes include compounds of formulae (IIIa) to (IIIj) wherein n is from 1 to 6; each Z'represents a group Z or one represents a group-LZY and the other represents a group as defined for R4; and R7, which may be the same or different, is hydrogen or a group as defined for R1.

Particularly suitable Arduengo carbenes include 1,3-disubstituted-imidazolin-2-ylidenes, e. g. compounds of formula

wherein R'is methyl, phenyl, naphthyl, 2,6-dimethyl- phenyl, 2,6-diisopropylphenyl, 2,6-di-t-butylphenyl, mesityl or ferrocenyl.

Unless otherwise specified, alkyl groups or alkylene moieties referred to herein may conveniently contain 1 to 10, more preferably 1 to 6 carbon atoms and may be linear or branched. Likewise, unless otherwise

specified, aryl groups are homocyclic or heterocyclic and preferably contain 5 to 7 ring atoms per ring, with such rings containing 0,1,2,3 or 4 ring heteroatoms selected from O, N and S, preferably 0,1,2 or 3 N atoms, and with the groups containing a total of 5 to 16 ring atoms. The ring atoms may be substituted, e. g. by alkyl groups, preferably C16 alkyl groups such as methyl, isopropyl or t-butyl, by other groups listed above or by fused saturated or unsaturated rings.

Examples of typical aryl groups include phenyl, naphthyl, mesityl, 2,6-diisopropylphenyl, 2,6-di-t- butylphenyl and 2,6-di-t-butyl-4-methylphenyl.

Similarly, unless otherwise specified, cycloalkyl and aralkyl groups may conveniently contain 5 to 7 ring atoms per ring, for example as in benzyl.

Examples of suitable carbenes and carbene-metal complexes and procedures for their synthesis are described in the literature, e. g. by Ofele, K., J.

Organomet. Chem. 1968,12,42-43; Wanzlick, H. W. et al., Angew. Chem., Int. Ed. Engl. 1968,7,141-142; Ofele, K. et al., Angew. Chem., Int. Ed. Engl. 1970,9,739-740; Schonherr, H. J. et al., Chem. Ber. 1970,103,1037-1046; Schönherr, H. J. et al., Liebigs Ann. Chem. 1970,731, 176-179; Luger, P. et al., Acta Cryst., Sect. B 1971, B27,2276-2279; Ofele, K. et al., Chem. Ber. 1972,105, 529-540; Ofele, K. et al., Z. Naturforsch. 1973,28B, 306-309; Kreiter, C. G. et al., Chem. Ber. 1976,109, 1749-1758; Ofele, K. et al., Z. Naturforsch. 1976,31B, 1070-1077; Krist, H. G. Dissertation, Technische Universität München, 1986; Bonati, F. et al., J.

Organomet. Chem. 1989,375,147-160; Arduengo, A. J., III et al., J. Am. Chem. Soc. 1991,113,361-363; Bonati, F. et al., J. Organomet. Chem. 1991,408,271-280; Arduengo, A. J., III et al., J. Am. Chem. Soc. 1992,114, 5530-5534; Herrmann, W. A. et al., Chem. Ber. 1992,125, 1795-1799; Britten, I. F. et al., Acta Cryst., Sect. C 1992, C48,1600-1603; Mihailos, D. Dissertation,

Technische Universität Munchen, 1992; Arduengo, A. I., III et al., Organometallics 1993,12,3405-3409; Arduengo, A. I. et al., J. Organomet. Chem. 1993,462, 13-18; Ofele, K. et al., J. Organomet. Chem. 1993,459, 177-184; Kuhn, N. et al., Synthesis 1993,561-562; Arduengo, A. I., III et al., J. Am. Chem. Soc. 1994,116, 4391-4394; Arduengo, A. I., III et al., J. Am. Chem. Soc.

1994,116,7927-7928; Schumann, H. et al., Angew. Chem., Int. Ed. Engl. 1994,33,1733-1734; Kuhn, N. et al., J.

Organomet. Chem. 1994,470, C8-C11; Herrmann, W. A. et al., J. Organomet. Chem. 1994,480, C7-C9; Schumann, H. et al., Chem. Ber. 1994,127,2369-2372; Dias, H. V. R. et al., Tetrahedron Lett. 1994,35,1365-1366; Gridnev, A. A., et al., Synth. Commun. 1994,24,1547-1555; Herrmann, W. A., Organometallics 1995,14,1085-1086; Herrmann, W. A. et al., Angew. Chem., Int. Ed. Engl.

1995,34,2371-2374; Ofele, K. et al., J. Organomet.

Chem. 1995,498,1-14; Herrmann, W. A. et al., J.

Organomet. Chem. 1995,501, Cl-C4; Kuhn, N. et al., Inorg. Chim. Acta 1995,238,179-181; Herrmann, W. A. et al., Chem. Eur. J. 1996,2,772-780; Herrmann, W. A. et al., Chem. Eur. J. 1996,2,1627-1636; Herrmann, W. A. et al., Angew. Chem., Int. Ed. Engl. 1996,35,2805-2807; Herrmann, W. A. et al., J. Organomet. Chem. 1996,520, 231-234; Herrmann, W. A. et al., Organometallics 1997, 16,682-688; Herrmann, W. A. et al., Organometallics 1997,16,2209-2212; Herrmann, W. A. et al., Organometallics 1997,16,2472-2477; Herrmann, W. A. et al., Angew. Chem., Int. Ed. Engl. 1997,36,1049-1067; Herrmann, W. A. et al., J. Organomet. Chem. 1997,530, 259-262; Kocher, C. et al., J. Organomet. Chem. 1997, 532,261-265; Arduengo et al., J. Am. Chem. Soc., 1997, 119,12742; Hermann et al., Angew. Chem. Int. Ed. Engl.

1997,36,2162; Kacher et al., J. Organomet. Chem. 1997, 532,26; Gardiner et al., J. Organomet. Chem. 1999,572, 239; J. Organomet. Chem. 1999,572,177; Herrmann et al., J. Organomet. Chem. 1997,547,357; Wang et al.,

Organometallics 1998,17,972; Liu et al., Organometallics 1998,17,993; Herrmann et al., Organometallics 1998,17,2162; Weskamp et al., Angew.

Chem. Int. Ed. Engl. 1998,37,2490; Green et al., J.

Organomet. Chem. 1998,554,175; Herrmann et al., J.

Organomet. Chem. Arduengo et al., Chemie Unserer Zeit 1998,32,6; Voges et al., Organometallics 1999,18,529; Abernethy et al., J. Am. Chem. Soc. 1999, 121,2329; Huang et al., J. Am. Chem. Soc. 1999.121, 2624; Wang et al., Organometallics 1999,18,1216; and McGuinness et al., Organometallics 1999,18,1596.

Other suitable carbenes may be synthesised analogously.

Besides the unsaturated carbene ligands, ring saturated carbenes may also be used. Examples are described by, for example, Denk et al., Angew. Chem.

Int. Ed. Engl. 1997,36,2607 and Sellmann et al., J.

Organomet. Chem. 1997,541,291.

Particularly suitable carbenes include: 1,3-dimethylimidazolin-2-ylidene; 1,3-diisopropyl- imidazolin-2-ylidene; 1,3-di-n-butylimidazolin-2- ylidene; 1, 3-di-t-butylimidazolin-2-ylidene; 1,3-di- trimethylsilylimidazolin-2-ylidene; 1,3-dibenzyl- imidazolin-2-ylidene; 1,3-dicyclohexylimidazolin-2- ylidene; 1,3-diphenylimidazolin-2-ylidene; 1,3-bis (2,6- diisopropylphenyl) imidazolin-2-ylidene; 1,3-bis (2,6-di- t-butylphenyl) imidazolin-2-ylidene; 1,3-di-(1-naphthyl)- imidazolin-2-ylidene, 1,3-di- (l-anthracyl) imidazolin-2- ylidene; 1-methyl-3-isopropylimidazolin-2-ylidene; 1-n- butyl-3-isopropylimidazolin-2-ylidene; 1-t-butyl-3- isopropylimidazolin-2-ylidene; 1-trimethylsilyl-3- isopropylimidazolin-2-ylidene; 1-benzyl-3-isopropyl- imidazolin-2-ylidene; 1-cyclohexyl-3-isopropyl- imidazolin-2-ylidene; 1-phenyl-3-isopropylimidazolin-2- ylidene; 1- (2, 6-diisopropylphenyl)-3-isopropyl- imidazolin-2-ylidene; 1- (2,6-di-t-butylphenyl)-3- isopropylimidazolin-2-ylidene; 1-mesityl-3-isopropyl-

imidazolin-2-ylidene; 1-(l-naphthyl)-3-isopr imidazolin-2-ylidene; 1-(l-anthracyl)-3-isopropyl- imidazolin-2-ylidene;1-methyl-3- (2,6-diisopropyl- phenyl) imidazolin-2-ylidene; 1-isopropyl-3- (2,6- diisopropylphenyl) imidazolin-2-ylidene; 1-n-butyl-3- (2,6-diisopropylphenyl) imidazolin-2-ylidene; 1-t-butyl- 3- (2,6-diisopropylphenyl) imidazolin-2-ylidene; 1- trimethylsilyl-3- (2,6-diisopropylphenyl) imidazolin-2- ylidene; 1-benzyl-3- (2, 6-diisopropylphenyl) imidazolin-2- ylidene; 1-cyclohexyl-3- (2, 6-diisopropylphenyl)- imidazolin-2-ylidene;1-phenyl-3- (2, 6-diisopropyl- phenyl) imidazolin-2-ylidene; 1- (2, 6-di-t-butylphenyl)-3- (2,6-diisopropylphenyl) imidazolin-2-ylidene; 1-mesityl- 3- (2,6-diisopropylphenyl) imidazolin-2-ylidene; 1- (1- naphthyl)-3- (2,6-diisopropylphenyl) imidazolin-2-ylidene; 1- (l-anthracyl)-3- (2,6-diisopropylphenyl) imidazolin-2- ylidene; 1-methyl-3-mesitylimidazolin-2-ylidene; 1- isopropyl-3-mesitylimidazolin-2-ylidene; 1-n-butyl-3- mesitylimidazolin-2-ylidene; 1-t-butyl-3-mesityl- imidazolin-2-ylidene; 1-trimethylsilyl-3-mesityl- imidazolin-2-ylidene; l-benzyl-3-mesitylimidazolin-2- ylidene; 1-cyclohexyl-3-mesitylimidazolin-2-ylidene; 1- phenyl-3-mesitylimidazolin-2-ylidene; 1- (2,6- diisopropylphenyl)-3-mesitylimidazolin-2-ylidene; 1- (2,6-di-t-butylphenyl)-3-mesitylimidazolin-2-ylidene; 1- (1-naphthyl)-3-mesitylimidazolin-2-ylidene and 1- (1- anthracyl)-3-mesitylimidazolin-2-ylidene.

In formula (I) Y is preferably a 5-or 6-membered carbocyclic or heterocyclic ring substituted by two Z groups in the o, o'-positions. Suitable heterocyclic rings may contain up to three heteroatoms, i. e. N, S and/or O atoms, preferably N atoms. Representative 5- membered rings include cyclopentyl, cyclopentenyl, cyclopentadienyl, pyrrole and pyrrolidine.

Representative 6-membered rings include phenyl, 4- pyridyl, cyclohexyl, cyclohexenyl and cyclohexadienyl.

Preferably Y is an optionally substituted phenyl group.

Rl is preferably hydrogen. R2 and R3 are each preferably hydrogen atoms or C16 alkyl groups, more preferably Cl4 alkyl groups, e. g. methyl, ethyl, n- propyl, isopropyl or t-butyl. Where R and R3 are linked to form a heterocyclic group, this is preferably pyrrole. Z is particularly preferably-(CH2) n-NR2R3.

Q may be a hydrocarbyl group, e. g. an alkyl, cycloalkyl, aryl or aralkyl group, a carbonyl group, an alkoxy group, a carboxyl group or an amide group.

Alternatively, Q may be an inorganic group, such as a halogen, nitrate, sulphate or oxide. Preferably Q is a halogen, for example chloride or bromide, especially preferably chloride.

Complexes particularly suitable for catalytic use in accordance with the invention include the following compounds of formulae (Ia) to (If):

in which M is chromium; R2, R3 and R4 each independently represent hydrogen atoms or C16 alkyl groups, e. g. methyl or t-butyl; n is 1,2 or 3 and X is chloride.

Examples of ligands suitable for use in complexes of the invention, together with methods for their synthesis, are described in the literature, e. g. by Ohff et al., J. Am. Chem. Soc. 1997,119,11687-11688; Lee et al., Organometallics 1998,17 (1), 1-3; and Britovsek et al., Chem. Commun. 1998,849-850. The synthesis techniques described in these documents may be used analogously to prepare other ligands suitable for use in the invention.

Preferred for use in the invention are the chromium (III) complexes with the following ligands: 2,6-bis (dimethylaminomethyl) phenyl, 2,6-bis (diethylaminomethyl) phenyl, 2,6-bis (diethylaminoethyl) phenyl, 2,6-bis (dimethylaminoethyl) phenyl, 2,5-bis (dimethylaminomethyl) cyclopentyl, 2,5-bis (diethylaminomethyl) cyclopentyl, 2,5-bis (diethylaminoethyl) cyclopentyl, 2,5-bis (dimethylaminoethyl) cyclopentyl, 3- (1,5-dimethylamino) pentyl, 3- (1,5-diethylamino) pentyl, 4- (1,7-dimethylamino) heptyl, 4- (1,7-diethylamino) heptyl, bis (2-dimethylaminoethyl) amide, bis (2-diethylaminoethyl) amide, bis (2-dimethylaminopropyl) amide,

bis (2-diethylaminopropyl) amide, 2,6-bis (dimethylphosphinemethyl) phenyl, 2,6-bis (diethylphosphinemethyl) phenyl, 2,6-bis (diethylphosphineethyl) phenyl, 2,6-bis (dimethylphosphineethyl) phenyl, 2,5-bis (dimethylphosphinemethyl) cyclopentyl, 2,5-bis (diethylphosphinemethyl) cyclopentyl, 2,5-bis (diethylphosphineethyl) cyclopentyl, 2,5-bis (dimethylphosphineethyl) cyclopentyl, 3- (1, 5-dimethylphosphine) pentyl, 3- 5-diethylphosphine) pentyl 4- (1,7-dimethylphosphine) heptyl, 4- (1,7-diethylphosphine) heptyl, bis (2-dimethylphosphineethyl) amide, bis (2-diethylphosphineethyl) amide, bis (2-dimethylphosphinepropyl) amide, bis (2-diethylphosphine propyl) amide, and 2,6-bis (imino) pyridyl.

The metal complex for use in the invention may comprise other organic or inorganic groups in addition to those provided by Q. For example, the metal complex may comprise other metals which may be selected from the group 1 metals, especially lithium. In particular, alkali metal halides, e. g. lithium halides such as LiCl may be bound into the ligand sphere of the complex.

Typically, the complex will also comprise one or more solvating ligands capable of stabilising the complex although in general this will depend on the process used to prepare the catalyst. Suitable solvating ligands include Lewis base or electron pair donor ligands and may be selected from nitrogen-, oxygen-, phosphorus-and sulphur-containing compounds, such as ethers, thioethers, esters, ketones, aldehydes, alcohols, amines, amides, nitriles and phosphines. Examples of suitable solvents include tetrahydrofuran (THF), dioxane, pyridine, dimethylpyridine, dimethylformamide, N-methylformamide, aniline, acetonitrile, ethylacetate,

diethylether, diglyme, methyl ethyl ketone, acetaldehyde, dimethylsulphoxide, trimethylamine and triphenylphosphine. THF is particularly preferred as an electron pair donor solvent.

Especially preferably, catalysts for use in the invention may comprise the following metal complex: Metal complexes useful in accordance with the invention may be produced using known methods of synthesis. For example, these may be prepared by reacting a compound containing the desired ligand, such as an organoalkali metal compound, e. g. an organolithium compound, or a corresponding Grignard reagent, with a salt of a group 6 metal, e. g. a chromium salt, in a hydrocarbon solvent. Such methods and the products obtained by such methods form further aspects of the invention.

Suitable salts of group 6 metals for use in preparing the compounds of the invention include the halides, e. g. chromium di-and tri-halides, especially CrCl2 and CrCl3. Preferred methods for preparing the compounds of formula iso comprise reaction of chromium (III) chloride with an organoalkali metal compound in which the organic radical is the group YZ2. Typically, reaction may be carried out at ambient temperature in a Lewis base or electron pair donor solvent, especially preferably THF. A solvent other than the Lewis base solvent may, however, be present.

Where Y is N or a carbene, the complexes of the invention may be prepared by reacting the carbene or Z2N ligand with a coordinatively unsaturated group 6 metal

compound or with a complex of the group 6 metal which contains ligands displaceable by the carbene or Z2N- ligand, optionally in the presence of a base. See for example Bildstein et al., J. Organomet. Chem. 1999,572, 177.

The catalyst complexes herein described may be used in combination with other conventional olefin polymerisation catalysts, typically transition metal organometallic or coordination compounds capable of polymerising a-olefins, optionally in the presence of a suitable co-catalyst. Such catalyst mixtures form a further aspect of the invention. Particularly preferred are metallocene catalysts. As used herein, the term metallocene is used to refer to all catalytically active metal: n-ligand complexes in which a metal is complexed by one, two or more rl-ligands. The use of twin p-ligand metallocenes and single p-ligand"half metallocenes"is particularly preferred. The metal in such complexes is preferably a group 4,5,6,7 or 8 metal or a lanthanide or actinide, especially a group 4,5 or 6 metal, particularly Zr, Hf or Ti. The ri-ligand preferably comprises a cyclopentadienyl ring, optionally with a ring carbon replaced by a heteroatom (e. g. N, B, S or P), optionally substituted by pendant or fused ring substituents and optionally linked by bridge (e. g. a 1 to 4 atom bridge such as (CH2) 2, C (CH3) 2 or Si (CH3) 2) to a further optionally substituted homo or heterocyclic cyclopentadienyl ring. The ring substituents may for example be halo atoms or alkyl groups optionally with carbons replaced by heteroatoms such as O, N and Si, especially Si and O, and optionally substituted by mono or polycyclic groups such as phenyl or naphthyl groups.

Examples of such homo or heterocyclic cyclopentadienyl ligands are well known in the art (see e. g. EP-A-416815, W096/04290, EP-A-485821, EP-A-485823, US-A-5276208 and US-A-5145819).

Preferred metallocene catalysts for use in the

invention include those comprising at least one optionally substituted cyclopentadienyl ring, preferably those comprising a transition metal atom selected from group 3 to group 10, preferably a group 4,5 or 6 metal, particularly Zr, Hf or Ti. Suitable co-catalysts for use in combination with these catalysts are aluminoxanes and Lewis acid compounds.

For use in olefin polymerisation, the metal complexes of the invention are preferably used in combination with a co-catalyst or catalyst activator.

Preferred co-catalysts include alkyl aluminium compounds, in particular aluminoxanes. Suitable aluminoxanes include Cl-lo alkyl aluminoxanes, e. g. methyl aluminoxane, and aluminoxanes in which the alkyl groups comprise isobutyl groups optionally together with methyl groups, e. g. isobutyl aluminoxane. Methyl aluminoxane is especially preferred for use as a co-catalyst. Alkyl aluminium compounds, such as aluminoxanes, may be used as the sole co-catalyst or alternatively may be used with other co-catalysts. Thus, besides or in addition to alkyl aluminium compounds other catalyst activators such as boron containing compounds, transition metal compounds (e. g. halogenide compounds), magnesium compounds, and group 3 organometallic compounds, e. g. aluminium or boron based compounds, may be used. Such materials may be solids, liquids or may be in solution in a liquid phase of the catalyst material which may be in solution, a solid, a dispersion, a suspension, a slurry, etc. Particular mention may be made of cation complex forming catalyst activators such as the silver and boron compounds known in the art. What is required of such activators is that they should react with the n- liganded complex to yield an organometallic cation and a non-coordinating anion (see for example the discussion on non-coordinating anions J-in EP-A-617052 (Asahi)).

Aluminoxane co-catalysts are described by Hoechst in WO-A-9428034. These are linear or cyclic oligomers

having up to 40, preferably 3 to 20,- [Al (R") O]- repeat units (where R"is hydrogen, C1l0 alkyl, preferably methyl, or C6_le aryl or mixtures thereof).

Polymerisation according to the invention may be performed using standard polymerisation techniques, e. g. gas phase, slurry phase or liquid phase polymerisation and using conventional polymerisation reactors, e. g. loop reactors, gas phase reactors, or stirred tank reactors, or any combination thereof.

The polymerisation processes of the invention may be carried out in a single reactor or in a series of two or more reactors. Each polymerization stage may be effected using conventional procedures, e. g. as a slurry, gas phase, solution or high pressure polymerization. Slurry polymerization is preferably effected, e. g. in a tank reactor or more preferably a loop reactor. Preferably however the polymerization process uses a series of two or more reactors, preferably loop and/or gas phase reactors, e. g. a combination of loop and loop, gas phase and gas phase or most preferably loop and gas phase reactors. In such reactors, the (major) monomer may also function as a solvent/carrier as well as a reagent, or alternatively a non-polymerizable organic compound, e. g. a C310 alkane, for example propane or isobutane, may be used as a solvent/carrier. Where this is done, the volatile non- reacted or non-reactive materials will desirably be recovered and reused, especially where gas phase reactors are used.

Typical reaction conditions for loop and gas phase reactors are: loop-temperature 60-110°C, pressure 30- 70 bar, mean residence time 30-80 minutes; and gas phase -temperature 60-110°C, pressure 10-25 bar, mean residence time 20-300 minutes.

The polymerisation process of the invention is typically conducted in the presence of a solvent, for example a linear or cyclic saturated hydrocarbon such as

propane, butane, pentane, hexane, heptane, octane, cyclohexane or methylcyclohexane; an aromatic hydrocarbon such as benzene, toluene, xylene or ethylbenzene; or a chlorinated aliphatic or aromatic hydrocarbon such as chloroform, methylene chloride, dichloroethane, trichloroethane, tetrachloroethane, chlorobenzene or dichlorobenzene. Of these compounds, aromatic hydrocarbons, in particular toluene, and alkanes, in particular propane or butane, are preferred.

Any olefin or mixture of olefins may be used in the process of the'invention, for example optionally substituted C230 a-olefins. C28 a-olefins and mixtures thereof, e. g. C2-or C3-olefins, are particularly preferred. The process of the invention is particularly suitable for the polymerisation of a-olefins such as ethylene, propylene, 1-butene, 1-pentene, 1-hexene, 3- <BR> <BR> methyl-1-butene,3-methyl-1-pentene,4-methyl-1-pentene, and 1-octene, especially preferably ethylene.

In the polymerisation process of the invention more than one olefin monomer may be used. It is preferred that comonomers are used in a minor amount, e. g. 0.5 to 40%, preferably up to 10% by weight, relative to the total monomer weight, with the major monomer (e. g. ethylene) making up the major amount, e. g. 60 to 99.5%, preferably 90 to 99.5% by weight. Suitable comonomers include other linear or branched Calo a-olefins, but may also be more bulky monomers containing unsaturated carbon carbon bonds and, for example, up to 20 carbon atoms, preferably up to 16 carbons, e. g. up to 14. Such comonomers may thus be mono or polycyclic, fused ring or unfused compounds containing one or more, e. g. 1,2 or 3, unsaturated carbon carbon bonds. Examples of suitable bulky comonomers include norbornene, norbornadiene, dicyclopentadiene, 1,5-cyclooctadiene and cyclooctene.

The temperature of the polymerisation reaction is typically in the range of from 0 to 300°C, preferably

from 60 to 120°C. The pressure employed for the olefin or olefins is typically from 1 to 3000 bars, preferably from 5 to 20 bars. The residence time is generally from 1 minute to 20 hours, preferably from 0.5 to 6 hours.

The metal complexes of the invention (or mixtures thereof) are preferably loaded onto solid supports prior to use in polymerisation reactions. Such supports preferably comprise porous substrates, e. g. inorganic oxides such as silica, alumina, silica-alumina or zirconia, inorganic halides such as magnesium chloride, or porous polymer particles, e. g. acrylate polymer particles or styrene-divinylbenzene polymer particles which optionally carry functional groups such as hydroxy, carboxyl etc. Particulate supports preferably have particles sizes in the range 5 to 60 Hm and may advantageously have porosities in the range 1 to 3 ml/mg. The metal complexes may be loaded onto the support before or, more preferably, after it has been reacted with a co-catalyst. Desirably inorganic supports are heat treated (calcined) before being loaded with the complex.

When used in combination with an alkyl aluminium co-catalyst, preferred molar ratios of Al: group 6 metal, e. g. Al: Cr, are 100-500: 1.

Polymeric olefin products produced by the polymerisation process of the invention form a further aspect of the invention.

The invention will now be described further by way of the following non-limiting Examples, where NCN denotes 2,6-bis (dimethylaminomethyl) phenyl.

Example 1 Preparation of f2.6-bis (dimethylaminomethyl) phenyl1 chromium dichloride Preparation of (NCN) CrCl3Li (THF) z: a THF solution of (2,6-bis (dimethylaminomethyl) phenyl-lithium (NCN-Li) was added dropwise to a THF solution of chromium (III) chloride tetrahydrofuran complex, CrCl3 (THF) 3. The solution was stirred for 2 hours and after workup a 500 yield of the compound (NCN) CrCl3Li (THF) 2 (1) (Scheme 1) was isolated. The compound was characterised by 1H-NMR, mass spectrometry, elemental analysis and X-ray. lH-NMR (THF-d8) show characteristic very broad signals at 5-17 and 5-8 which can be assigned to the methyls and methylenes at the nitrogen atoms. The mass spectrum shows a mass at 313, with a characteristic isotope distribution, which is assigned for the fragment LiCl (THF) 2. Other masses were observed at 278,277,191, 148 and 105.

Elemental analysis for CzoH3sCl3CrLiN202 (500.8054) : Calc.: C 47.97; H 7.04; Cl 21.24; Cr 10.38; N 5.59.

Found: C 45.40; H 6.59; Cl 21.17; Cr 10.30/10.44; N 5.63.

X-ray analysis confirmed the structure shown in Scheme 1.

SCHEME I

The magnetic susceptibility of (1) was measured by Evans NMR method in THF-d8 with benzene as internal standard.

The paramagnetic shift of the benzene signal indicated the magnetic moment, eff, as 3.83 which is in accordance with three unpaired electrons.

Example 2 Polymerisation with f2,6-bis (dimethylaminomethyl) phenyll chromium dichloride Polymerisation tests were carried out using the complex ligand (1) prepared in Example 1, activated with methyl aluminoxane, as a catalyst for polymerisation of ethylene.

Two polymerisation tests were carried out in glass reactors each containing an atmospheric pressure of ethylene in toluene at 30°C. Each reactor was connected to an ethylene reservoir whereby ethylene could be added through a mass flow meter to maintain a constant pressure inside each reactor. 0.05 mmol of the complex (1) was used in each test. Methyl aluminoxane was used as co-catalyst with Al/Cr molar ratios of 100 and 400 in the two reactors respectively; reaction between the complex (1) and a toluene solution of methyl aluminoxane (7.4 wts) was carried out in separate bottles for approximately 2 minutes under continuous stirring before the catalyst solutions were introduced to the reactors.

After adding the catalyst, a low, steady ethylene flow was observed, the average consumption of ethylene being 4 ml/min and 10 ml/min respectively for the two reactors. After about one hour, the polymerisation was terminated by addition of acidic methanol to yield a white polyethylene precipitate.