TECHNICAL BACKGROUND Polyolefins, such as polyethylene and polypropylene, are important items of commerce, and many methods have been developed for their production. Commonly a transition metal compound is used as a polymerization catalyst, and recently there has been great interest on the use of late transition metal complexes (Group 8 to Group 10 metals, IUPAC designa- tion) as such polymerization catalysts. While some of these catalysts display excellent productivity of polymer (amount of polymer produced per unit of catalyst) in various types (homogeneous, slurry and gas phase for instance) of polym- erization processes, others display relatively low produc- tivity and/or the lifetime of the active catalyst is shorter than that desired in a typical commercial polymerization process.

The cause of these shortcomings for those polymeriza- tion catalysts has not been clear, so it has been difficult to rectify them. Generally speaking, while varying polym- erization conditions such as temperature, pressure of mono- mer (if it is a gas), concentration of polymerization cata- lyst, variation of cocatalyst (such as alkylaluminum com- pounds), etc., can result in modest improvements in polymer productivity in some instances, often a desired level of productivity is not reached. Therefore methods of making

these types of polymerizations more productive are being sought.

US4710552, US5110928 and US5210160 report the use of various halogenated compounds as additives in the polymeri- zation of olefins using Ziegler early transition metal po- lymerization catalysts, principally to improve the process- ability of the polyolefins produced. No mention is made of late transition metal catalysts.

WO00/50470 mentions in Example 148 the use of ethyl phenyldichloroacetate in combination with a certain iron containing catalyst in the polymerization of ethylene.

There is no mention of any improvement in polymer productiv- ity because of use of the ester.

SUMMARY OF THE INVENTION This invention concerns a process for the polymeriza- tion of an olefin, comprising the step of contacting, under polymerization conditions, said olefin with an olefin coor- dination polymerization catalyst comprising a complex of a Group 8 to Group 10 metal, wherein an oxidizing agent is present during at least a portion of said contacting.

This invention also concerns a process for improving the productivity of an olefin coordination polymerization catalyst comprising a complex of a Group 8 to Group 10 metal, in a process for producing a polyolefin by contacting an olefin with said polymerization catalyst under conditions to polymerize said olefin, said process comprising the step having an oxidizing agent present during at least a portion of the contacting of said olefin and said polymerization catalyst.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the polymerization processes and catalyst composi- tions described herein, certain groups may be present.

A"hydrocarbyl group"is a univalent group containing only carbon and hydrogen. As examples of hydrocarbyls may be mentioned unsubstituted alkyls, cycloalkyls and aryls.

If not otherwise stated, it is preferred that hydrocarbyl groups herein contain 1 to about 30 carbon atoms.

By"saturated"hydrocarbyl is meant a univalent radical that contains only carbon and hydrogen, and contains no car bon-carbon double bonds, triple bonds or aromatic groups.

By"substituted hydrocarbyl"herein is meant a hydro- carbyl group that contains one or more (types of) substitu- ents that do not substantially interfere with the operation of the polymerization catalyst system. Suitable substitu- ents in some polymerizations may include some or all of halo, ester, keto (oxo), amino, imino, carboxyl, phosphite, phosphonite, phosphine, phosphinite, thioether, amide, ni- trile, and ether. Preferred substituents when present are halo, ester, amino, imino, carboxyl, phosphite, phosphonite, phosphine, phosphinite, thioether, and amide. Which sub- stituents are useful in which polymerizations may in some cases be determined by reference to US5880241 (incorporated by reference herein for all purposes as if fully set forth).

If not otherwise stated, it is preferred that substituted hydrocarbyl groups herein contain 1 to about 30 carbon at- oms. Included in the meaning of"substituted"are chains or rings containing one or more heteroatoms, such as nitrogen, oxygen and/or sulfur, and the free valence of the substi- tuted hydrocarbyl may be to the heteroatom. In a substi- tuted hydrocarbyl, all of the hydrogens may be substituted, as in trifluoromethyl.

By" (substituted) hydrocarbylene"is meant a group analogous to hydrocarbyl, except the radical is divalent.

"Alkyl group"and"substituted alkyl group"have their usual meaning (see above for substituted under substituted

hydrocarbyl). Unless otherwise stated, alkyl groups and substituted alkyl groups preferably have 1 to about 30 car- bon atoms.

By"aryl"is meant a monovalent aromatic group in which the free valence is to the carbon atom of an aromatic ring.

An aryl may have one or more aromatic rings which may be fused, connected by single bonds or other groups.

By"substituted aryl"is meant a monovalent aromatic group substituted as set forth in the above definition of "substituted hydrocarbyl". Similar to an aryl, a substi- tuted aryl may have one or more aromatic rings which may be fused, connected. by single bonds or other groups ; however, when the substituted aryl has a heteroaromatic ring, the free valence in the substituted aryl group can be to a het- eroatom (such as nitrogen) of the heteroaromatic ring in- stead of a carbon.

By"phenyl"is meant the C6H5-radical, and a phenyl moiety or substituted phenyl is a radical in which one or more of the hydrogen atoms is replaced by a substituent group (which may include hydrocarbyl). Preferred substitu- ents for substituted aryl include those listed above for substituted hydrocarbyl, plus hydrocarbyl.

By" (inert) functional group"herein is meant a group other than hydrocarbyl or substituted hydrocarbyl that is inert under the process conditions to which the compound containing the group is subjected. The functional groups . also do not substantially interfere with any process de- scribed herein that the compound in which they are present may take part in. Examples of functional groups include some halo groups (for example fluoro and some unactivated chloro) ether such as-OR31 wherein R31 is hydrocarbyl or sub- stituted hydrocarbyl. In cases in which the functional group may be near a metal atom, the functional group should

not coordinate to the metal atom more strongly than the groups in those compounds are shown as coordinating to the metal atom, that is they should not displace the desired co- ordinating group.

By an"active halocarbon"is meant a compound that con- tains carbon and halogen, and optionally hydrogen, and may contain inert functional groups (other than halogen) and, preferably when present in the polymerization process, in- creases the productivity of the polymerization catalyst by at least 10 percent, based on a similar polymerization with- out the active halocarbon present.

By a"neutral bidentate ligand"is meant a bidentate ligand that no charge on the ligand (is not ionic in a for- mal sense if not coordinated to the transition metal).

By a"neutral tridentate ligands meant a tridentate ligand that has no charge on the ligand.

By a"neutral monodentate ligand"is meant a monoden- tate ligand that no charge on the ligand.

By an"monoanionic bidentate ligand"is meant a biden- tate ligand that has one negative charge on the ligand (is ionic in a formal sense if not coordinated to the transition metal).

By a"monoanionic tridentate ligand"is meant a triden- tate ligand that has one negative charge on the ligand.

By a"monoanionic monodentate ligand"is meant a mono- dentate ligand that has one negative charge on the ligand.

By"halogen"is meant any one or more of fluorine, chlorine, bromine or iodine, and chlorine, bromine and io- dine are preferred. Which of chlorine, bromine or iodine is preferred in any particular situation will depend on the particular polymerization process being run, the other in- gredients in that process, and which agent is used.

Preferred ligands herein are neutral bidentate ligands, as described in further detail below.

By an"activator","cocatalyst"or a"catalyst activa- tor"is meant a compound that reacts with a transition metal compound to form an activated catalyst species. This tran- sition metal compound may be added initially, or may be formed in situ, as by reaction of a transition metal com- pound with an oxidizing agent. A preferred catalyst activa- tor is an"alkyl aluminum compound", that is, a compound which has at least one alkyl group bound to an aluminum atom. Other groups such as, for example, alkoxide, hydride and halogen, may also be bound to aluminum atoms in the com- pound. Another useful activator is an alkylzinc compound.

"Noncoordinating ions" (or"weakly coordinating ions") are sometimes useful in the polymerization processes de- scribed herein. Such anions are well known to the artisan, see for instance W. Beck., et al., Chem. Rev., vol. 88, p.

1405-1421 (1988), and S. H. Strauss, Chem. Rev., vol. 93, p.

927-942 (1993), both of which are hereby included by refer- ence. Relative coordinating abilities of such noncoordinat- ing anions are described in these references, Beck at p.

1411, and Strauss at p. 932, Table III. Useful noncoordi- nating anions include, for example, SbF6, BAF, PF6, B (C6F5) 4-, or BF4, wherein BAF is tetrakis [3, 5-bis (trifluoromethyl) phenyl] borate.

A neutral Lewis acid or a cationic Lewis or Bronsted acid whose counterion is a weakly coordinating anion may also be present as part of the catalyst system, for example as a cocatalyst (or catalyst activator). By neutral Lewis acid"is meant a compound that is a Lewis acid capable of abstracting an anion from a late transition metal com- pound to form a weakly coordination anion. The neutral Lewis acid is originally uncharged (i. e., not ionic). Suit-

able neutral Lewis acids include SbF5, Ar3B (wherein Ar is aryl), and BF3. By a cationic Lewis acid is meant a cation with a positive charge such as Ag+, H+, and Na+.

In many of those instances in which the transition metal compound does not contain an alkyl or hydride group already bonded to the metal, the neutral Lewis acid or a cationic Lewis or Bronsted acid may also alkylate or add a hydride to the metal, i. e., causes an alkyl group or hydride to become bonded to the metal atom, or a separate compound is added to add the alkyl or hydride group. It is preferred that a neutral Lewis acid which can alkylate or add hydride to the metal, or a combination of a Lewis acid and a com- pound which can alkylate or add hydride to the metal, be present, and either of these (single or multiple compounds) can be considered an activator. By"hydride"is meant a single hydrogen atom covalently bound to the transition metal. No particular charge of the transferred hydrogen is implied, i. e., the Lewis acid may formally protonate the metal (transfer H+), but the product is referred to as a hy- dride.

A preferred neutral Lewis acid which can alkylate the transition metal is a selected alkyl aluminum compound, such as R93Al, R92AlCl, R9AlCl2, (R9AlCl) 2°, and"R9AlO" (alkylalu- minoxanes), wherein R9 is alkyl containing 1 to 25 carbon at- oms, preferably 1 to 4 carbon atoms. Suitable alkyl alumi- num compounds include methylaluminoxane (which is an oli- gomer with the general formula [MeAlO] n), (C2H5) 2AlCl, C2H5AlCl2, [ (CH3) 2CHCH2AlCl] 20 and [ (CH3) 2CHCH213Al. Preferred alkylaluminum compounds have at least one atom of an element more electronegative than carbon (on Pauling's electronega- tivity scale) attached to the aluminum atom. Such elements include halogen, especially chlorine, and oxygen. Another

useful alkylating agent is a dialkyl zinc, such as diethyl zinc.

Preferred metals for the transition metal complex herein are Ni, Pd, Fe and Co, Ni and Fe are especially pre- ferred, and Ni is particularly preferred.

Late transition metal complexes useful as polymeriza- tion catalysts herein, as well as combinations of two or more late transition metal catalyst, and combinations of late transition metal catalysts with other types of cata- lysts, are described in US5714556, US5852145, US5880241, US5929181, US5932670, US5942461, US5955555, US6060569, US6103658, US6174975, W096/37522, W097/23492, W097/48735, W098/30612, W098/37110, W098/38228, W098/40420, W098/42664, W098/42665, W098/47934, W099/49969, W099/41290, W099/51550, WO00/50470, JP-A-09255712, JP-A-09255713, JP-A-11158213, JP-A-11180991, JP-A-11209426, EP-A-0893455 and EP-A-0924223, all of which are hereby included by reference for all pur- poses as if fully set forth. Unless otherwise stated herein, polymerization catalysts for the purposes of the present invention also include catalysts that produce olefin oligomers.

As described in the above-incorporated publications, suitable catalysts are complexes of a late transition metal (Group 8 to Group 10, IUPAC designation) with an organic li- gand. Preferred ligands are mono-or bidentate ligands, es- pecially neutral mono-or bidentate ligands. A specific preferred such organic ligand is of the formula (I)

wherein : R13 and R16 are each independently hydrocarbyl or substi- tuted hydrocarbyl, provided that the atom bound to the imino nitrogen atom has at least two carbon atoms bound to it ; and R14 and Rl5 are each independently hydrogen, hydrocarbyl, substituted hydrocarbyl, or R14 and R15 taken together are hy- drocarbylene or substituted hydrocarbylene to form a carbo- cyclic ring.

As examples of when R14 and R15 are each independently a substituted hydrocarbyl may be mentioned when R14 is-YR17Rl8 and R15 is-ZR19R20, wherein Y and Z are each independently nitrogen, oxygen or sulfur and R17 and R19are each independ- ently hydrocarbyl, or substituted hydrocarbyl or taken to- gether form a ring, and R'8 and R20 are each independently hy- drogen, hydrocarbyl, or substituted hydrocarbyl, provided that when Y is oxygen or sulfur R18 is not present, and when Z is oxygen or sulfur R20 is not present.

Preferably the late transition metal catalyst is not an iron compound and/or the late transition metal is not coor- dinated to a tridentate ligand. More preferably this tri- dentate ligand is not a bisimine of a 2, 6-diacylpyridine or a 2, 6-pyridinedicarboxaldehyde, and most preferably this tridentate ligand is not There are many different ways of preparing active po- lymerization catalysts of transition metal coordination com- pounds as described herein, many of which are described in

the previously incorporated references (see, for example, US5714556, US5880241, US6103658 and WO00/06620), and those so described are applicable herein."Pure"compounds which themselves may be active polymerization catalysts may be used, or the active polymerization catalyst may be prepared in situ by a variety of methods.

For instance, olefins may be polymerized by contacting, at a temperature of about-100°C to about +200°C, a first compound W, which is a neutral Lewis acid capable of ab- stracting an anion to form a weakly coordinating anion ; or a cationic Lewis or Bronsted acid whose counterion is a weakly coordinating anion ; a second compound such as (II) and one or more polymerizable olefins'wherein : M is an appropriate transition metal ; m is the oxidation state of M minus 1 ; R13 through R16 are as defined above, each Q is independently a monóanion, preferably alkyl, hydride, chloride, iodide, or bromide ; and S is a monoanion, preferably alkyl, hydride, chloride, iodide, or bromide.

In this instance it is preferred that W is an alkyl aluminum compound. Other methods for preparing active po- lymerization catalyst will be found in the previously incor- porated references, and in the Examples herein.

The polymerization processes described herein may be run in a"normal"manner as described in the various refer- ences listed above for the various late transition metal complexes, with the oxidizing agent present during at least

a portion of the polymerization. It is particularly pre- ferred that the oxidizing agent be present (or added) con- tinuously or essentially continuously (for example added pe- riodically, particularly with little time between individual additions) during the polymerization while the components are present, for example, in the appropriate reactor. It is believed that the beneficial effect of the oxidizing agent is enhanced if it is present during most or all of the time the polymerization is taking place. The oxidizing agent not only usually enhances productivity of the catalyst, it also often increases the lifetime of the polymerization catalyst (which of course may also increase productivity depending on how long the polymerization is run).

The polymerization may be run in any of the known ways, for example, it may be a batch, semibatch or continuous po- lymerization which may be run as a liquid slurry, solution or gas phase polymerization. By"gas phase"is meant that the olefin monomer (s) is transported to the reactive polym- erization site (i. e., to contact with the catalyst particle) as a gas (except perhaps to diffuse through some polyolefin surrounding the active site), for example in a fluidized bed reactor. Other additives normally present in such polymeri- zations may also be present herein. For example chain transfer agents such as hydrogen may be present. Such polymerizations are well known in the art, see for instance B. Elvers, et al., Ed., Ullmann's Encyclopedia of Industrial Chemistry, 5th Ed., Vol. A21, VCH Verlagsgesellschaft mbH, Weinheim, 1992, p. 496-514 and 518-531, which is hereby in- cluded by reference.

The polymerization processes herein may be run in the presence of various liquids, particularly aprotic organic liquids. The catalyst system, monomer (s), and polymer may be soluble or insoluble in these liquids, but obviously

these liquids should not prevent the polymerization from oc- curring. Suitable liquids include alkanes, cycloalkanes, selected halogenated hydrocarbons, and aromatic hydrocar- bons. Specific useful solvents include hexane, cyclohexane, toluene, benzene, heptane, isooctane, methylene chloride, and 1, 2, 4-trichlorobenzene.

The olefin polymerizations herein may also initially be carried out in the"solid state"by, for instance, support- ing the transition metal compound on a substrate such as silica or alumina, or a polymeric material, activating if necessary with one or more cocatalysts, and contacting with the olefin (s). Alternatively, the support may first be con- tacted (reacted) with one or more cocatalysts (if needed) such as an alkylaluminum compound, and then contacted with an appropriate transition metal compound. The support may also be able to take the place of a Lewis or Bronsted acid, for instance an acidic clay such as montmorillonite, if needed. Another method of making a supported catalyst is to start a polymerization or at least make a transition metal complex of another olefin or oligomer of an olefin such as cyclopentene on a support such as silica or alumina. These "heterogeneous"catalysts may be used to catalyze polymeri- zation in the gas phase or the liquid phase.

The process of the present invention can be also used as last step of a multistep process such as described in the WO00/53646, which is incorporated by reference herein for all purposes as if fully set forth. In this process a poly- mer previously prepared with a different catalyst system is impregnated with the olefin polymerization catalyst system herein and then one or more olefins are polymerized accord- ing to the process of the present invention. The polymer of the first steps range from 10 to 70% of the total polymer

obtained in the multistep process, preferably from 10 to 60%, more preferably 20 to 50%.

Other details concerning general polymerization condi- tions may be had by referring to the previously incorporated references.

Although not wishing to be bound by theory, it is be- lieved that the oxidizing agents used herein function by oxidizing lower valent transition metal compounds to higher. valent compounds, such as Ni [I] and/or Ni [0] compounds to Ni [II] compounds. It is further believed that the more ac- tive form of the polymerization catalyst is a (relatively) higher valent form of the transition metal compound, and that during the polymerization the transition metal is re- duced (in an unwanted side reaction) to a lower valent form which is less active or inactive as a polymerization cata- lyst. The oxidizing agent thus oxidizes the lower valent form of the transition metal to the higher valent, more ac- tive form. This regenerates an active polymerization cata- lyst, thereby increasing the productivity and apparent rate of polymerization of the polymerization catalyst.

An"oxidizing agent"within the meaning of the present invention, therefore, is an agent capable of oxidizing the transition metal used in the polymerization catalyst, under polymerization conditions, from a lower valent state to the higher valent state to result in a re-activated polymeriza- tion catalyst. Thus the oxidizing agent should be a suffi- ciently strong oxidizing agent (as measured for instance by electrode potentials) to oxidize the appropriate transition metal to the desired oxidation state. The oxidizing agent, however, should not cause significant unwanted side reac- tions to produce large amounts of byproducts and/or be de- stroyed before it can carry out the desired oxidation. In addition the oxidizing agent should have some way of con-

tacting the lower valent transition metal atoms, for example be soluble in a liquid medium used in the polymerization, or be volatile enough to be added to a gas phase polymerization process.

The oxidizing agent may be any means suitable for ac- complishing the above-stated purpose. Preferred, however, are chemical oxidizing agents such as organic or inorganic compounds with the requisite properties. Preferred of the chemical oxidizing agents are one-electron oxidants. Useful oxidizing agents include, for example, iodine and halocar- bons such as perfluoroalkyl iodides, CI4, CHI3, CH2I2, ICH2CH2I, and trityl iodide, preferably iodine. Another pre- ferred oxidizing agent is a benzylic or allylic bromide or chloride such as a, a, a-trichlorotoluene and allyl chloride.

Other useful classes of oxidizing agents and/or indi- vidual oxidizing agents include, for example : NO, NO2, N-bromosuccinimide and °2i metal cations such as Fe+3, Cu+2, Ag+2, and ferricinium cations ; iminium radical cations such as tris (4-bromophneyl) iminium hexachloroantimonate ; and halogens and pseudohalogens such as BrCN, IBr and ICl.

One type of halocarbon which is useful has the formula wherein : T1 is a hydrocarbyl or substituted hydrocarbyl group containing at least one halogen bonded to a carbon atom ; preferably T'contains 2 or more halogen atoms, more prefera- bly 3 or more halogen atoms ; T2 is hydrogen, hydrocarbyl, substituted hydrocarbyl or a functional group, optionally containing one or more halo- gen bonded to a carbon atom.

Preferred agents formula (V) are those of the formula (VI) : Rl9-C (0)-OR" (VI) wherein : R19 is selected from the group consisting of hydrocarbyl or substituted hydrocarbyl wherein at least one hydrogen atom bonded to a carbon atom is replaced with a halogen atom ; and R20 is selected from the group consisting of R19 or hydrocar- byl or substituted hydrocarbyl. Preferably 2 or more, more preferably 3 or more of the hydrogens in R19 are replaced by halogen. R19 is preferably selected from the group consisting of linear or branched, saturated or unsaturated Cl-C20 perhalo- alkyl, C3-C20 perhalocycloalkyl, C6-C20 perhaloaryl, C7-C20 per- haloalkylaryl and C7-C20 perhaloarylalkyl radical optionally containing heteroatoms belonging to groups 13 or 15-16 of the Periodic Table of the Elements wherein"perhalo"means that all the hydrogens bonded to the carbon atoms of the corre- spondent hydrocarbon radical are replaced with halogen atoms.

R20 is preferably selected form the group consisting of a C1- C1o alkyl, C6-C20 aryl and C7-C20 alkylaryl.

More preferably the compound of formula (VI) has for- mula (VII), (VIII) or (IX) (VIII) (IX) wherein : R° has the meaning given above ;

each R23 is independently selected from the group con- sisting of halogen and R19, preferably each R23 is independ- ently selected from the group consisting of halogen trichlo- romethyl, perchloroethyl, perchloropropyl, perchlorobutyl ; more preferably R23 is halogen ; each R24 is independently selected from the group con- sisting of hydrogen, halogen and R19 ; preferably each R24 is independently halogen, more preferably chlorine ; R25 is selected from the group consisting of phenyl, thienyl, furyl, pyrollyl, pyridyl radicals ; preferably R25 is phenyl or a phenyl radical substituted by one or more halo- gen atoms, preferably chlorine ; X is halogen, preferably chlorine ; Y is selected from the group consisting of hydrogen, halogen, C1-C20 alkyl or C6-C20 aryl, C7-C20 alkylaryl and C7-C20 arylalkyl, and optionally containing halogen atoms ; prefera- bly Y is chlorine ; R26 is selected from the group consisting of linear or branched, saturated or unsaturated C1-C20 perhaloalkyl, C3-C20 perhalocycloalkyl, C6-C20 perhaloaryl, C7-C20 perhaloalkylaryl and C7-C20 perhaloarylalkyl radical optionally containing het- eroatoms belonging to groups 13 or 15-16 of the Periodic Ta- ble of the Elements wherein the suffix"perhalo"means that all the hydrogens bonded to the carbon atoms of the corre- spondent hydrocarbon radical are replaced with halogen atoms, preferably R26 is selected from the group consisting of tri- chloromethyl, perchloroethyl, perchloropropyl and perchloro- butyl.

Non limitative examples of compounds of formula (VI) are :

In (V) and any of its preferred forms it is preferred that the halogen present is chlorine and/or bromine, more preferably chlorine. Especially preferred compounds (VI) are esters of perchlorocrotonic acid or trichloroacetic acid.

Other halocarbons useful in the current process are found in US4710552, US5112928 and US5210160, which are in- corporated by reference herein for all purposes as if fully set forth. For example, mentioned in these patents are com-

pounds such as allyl chloride, allyl bromide, crotyl chlo- ride, crotyl bromide, propargyl chloride, propargyl bromide, a-chlorotoluene (benzyl chloride), a-bromotoluene, a, a, a- trichlorotoluene, a, a-dichlorotoluene, trichloroacetyl chlo- ride, trichloromethyl vinyl ketone, and other specific com- pounds that have been previously mentioned herein.

The syntheses of the active halocarbons are well known in. the art, and many of them are commercially available.

The chemical oxidizing agent may be introduced into the reactor in any suitable manner. For example in a solution or liquid slurry polymerization it may introduced as a solu- tion in a suitable liquid. Since very little oxidizing agent is required, this additional stream will usually in- troduce very little"extra"liquid into the reactor. In a gas phase reaction the oxidizing agent may be introduced as a vapor, for example as a vapor in ethylene if ethylene is a monomer. This will be particularly useful when elevated temperatures (above room temperature) are present in the po- lymerization system, since at elevated temperatures the oxi- dizing agents have higher vapor pressures.

Iodine is preferred and, by iodine herein, it is meant iodine (I2) itself, as well as compounds or combinations of compounds that readily generate iodine, or any chemically equivalent form of iodine such as the triiodide anion (I3-).

For example the compound KI3 is soluble in some organic sol- vent, and may be used in place of I2. Another preferred agent is a benzylic or allylic bromide or chloride such as a, a, a-trichlorotoluene and allyl chloride.

Chemical oxidizing agents are, of course, chemically reactive, and may interact with other ingredients in the po- lymerization process or even the process equipment. While small amounts of such reactions may not adversely affect the polymerization, substances which may rapidly react with the

chemical oxidizing agents may negate their effectiveness.

For example it is believed that some activators such as al- kylaluminum compounds and dialkylzinc compounds react with iodine to form alkyl iodides and aluminum or zinc iodides.

Alkylaluminum compounds that react relatively rapidly with iodine are believed to include trialkylaluminum compounds such as trimethylaluminum and triisobutylaluminum. Alu- minoxanes such as methylaluminoxane and dialkylzinc com- pounds such as diethylzinc react at somewhat slower but still appreciable rates, while alkylaluminum compounds that already contain aluminum halide groups react more slowly.

Therefore alkylaluminum compounds such as dialkylaluminum chlorides, alkylaluminum dichlorides, alkylaluminum ses- quichlorides and alkylhaloaluminoxanes such as [R1AlCl] 2° wherein R1 is alkyl such as methyl, ethyl, propyl and isobu- tyl are preferred, alkylaluminum compounds, alkylaluminum sesquichlorides and [RAICIO are more preferred, and al- kylaluminum sesquichlorides and [RAlCljgO are especially preferred. Even when a compound such as methylaluminoxane or diethylzinc is used, an improvement in productivity is seen, particularly when the concentration of methylaluminox- ane or diethylzinc is kept as low as possible, consistent with obtaining a reasonably rapid polymerization rate.

The oxidizing agent is used in an amount (or is present in such a manner) effective to oxidize the transition metal used in the polymerization catalyst, under polymerization conditions, from a lower valent state to the higher valent state. Preferred amounts are sufficient to achieve an at least 10% increase in the productivity of the catalyst (measured in terms of kg polymer/g transition metal).

For chemical oxidizing agents, the molar ratio of oxi- dizing agent (such as iodine) to transition metal in the re- actor feed (s) may vary depending on the particular polymeri-

zation system used, for example the activator present, but a generally useful range is a molar ratio of additive to late transition metal of about 5 to about 2000, preferably about 50 to about 1000. The activator may also be continuously or intermittently added to a batch or semibatch reaction.

Preferably also the activator to agent ratio should be no less than about 1, more preferably at least about 2.

Any olefin monomer that may usually be (co) polymerized by the late transition metal catalyst may also be polymer- ized in the process described herein. Copolymers of two or more olefins, and copolymers with other types of polymeri- zable monomers (for example, carbon monoxide) are also in- cluded herein. Which active polymerization catalysts will polymerize which olefins (not all catalysts will polymerize all olefins or combinations of olefins) will also be found in the above listed references. Monomers useful herein in- clude ethylene, propylene, other a-olefins of the formula R2CH=CH2, wherein R2 is n-alkyl containing 2 to about 20 car- bon atoms, cyclopentene, a styrene, a norbornene, and an olefin of the formula H2CH=CHR3Z, wherein R3 is alkylene or a covalent bond, preferably- (CH2) n-wherein n is an integer of 1 to 20 or a covalent bond, more preferably a covalent bond, and Z is a functional group, preferably-CO2X, wherein X is hydrogen, hydrocarbyl, especially alkyl, or substituted hy- drocarbyl. Preferred monomers are ethylene, propylene and cyclopentene, and ethylene is especially preferred. Also preferred are copolymers in which ethylene is a monomer.

As indicated above, the oxidizing agents may chemically interact with the process equipment, so appropriate materi- als of construction should be used for the reactors used in the polymerization process herein. Thus the addition of any particular oxidizing agent may require the use of materials

of construction that are resistant not only to the oxidizing agent but any products of reaction of that compound.

In the Examples the following abbreviations are used : n-BPCC-n-butyl perchlorocrotonate ETA-ethyl trichloroacetate DSC-differential scanning calorimetry GPC-gel permeation chromatography IBACO-isobutylchloroaluminoxane I. V.-intrinsic viscosity MMAO-modified (with butyl groups) methylaluminoxane Mn-number average molecular weight Mw-weight average molecular weight PE-polyethylene RT-room temperature Tm-melting point In the Examples, all pressures gauge pressures. Except where otherwise noted, the nickel compound used was (III) Another nickel compound used was

Example 1 A 600-mL stirred autoclave was loaded with 250 mL dry isooctane containing 1. 5 mL 1, 3-dichloro-1, 3- diisobutylalumoxane (0. 35 M in hexane) under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The solvent was saturated with ethylene by stirring at 350 kPa/60°C for 10 min.

A solution of 80 mg iodine was made up in 250 mL dry, deaerated isooctane. The autoclave was vented and the io- dine solution was pumped into the autoclave at 0. 3 mL/min (50 equivalents of iodine per hour based on Ni) using a high-pressure liquid chromatography pump via a feed line through one of the autoclave head ports. After 15 sec of iodine solution feed, a solution of 0. 3 mg of the Ni com- pound in 3 mL methylene chloride was injected via syringe through a head port. The autoclave was immediately pres- sured to 1. 03 MPa with ethylene and was stirred in a con- stant-temperature water bath at 60°C for 120 min as ethylene polymerized. The iodine feed was then stopped and the auto- clave was vented. The viscous polymer suspension was di- luted with acetone and filtered, and the resulting powder

was oven-dried to yield 42. 3 g (1529 kg polyethylene/g Ni) polyethylene powder.

Comparative Example 1 A 600-mL stirred autoclave was loaded with 200 mL dry isooctane containing 2. 5 mL 1, 3-dichloro-1, 3- diisobutylalumoxane (0. 35 M in hexane) under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The solvent was saturated with ethylene by stirring at 70 kPa/60°C for 10 min. The ethylene was vented and 0. 5 mg of (III) in 3 mL methylene chloride was injected via syringe through a head port. The autoclave was immedi- ately pressured to 1. 03 MPa with ethylene and was stirred in a constant-temperature water bath at 60°C for 120 min as ethylene polymerized. The autoclave was then vented and the resulting suspension of polymer powder was diluted with ace- tone, filtered and oven-dried to yield 25. 9 g (561 kg poly- ethylene/g Ni) polyethylene powder. This control example with no added iodine shows the great improvement afforded by addition of. iodine to the polymerizations described in Exam- ple 1.

Comparative Example 2 A 600-mL stirred autoclave was loaded with 200 mL dry isooctane containing 2 mL 1, 3-dichloro-1, 3- diisobutylalumoxane (0. 35 M in hexane) and 10 mg iodine un- der nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The solvent was satu- rated with ethylene by stirring at 350 kPa/60°C for 10 min.

The ethylene was vented and 0. 5 mg of (III) in 3 mL methyl- ene chloride was injected via syringe through a head port.

The autoclave was immediately pressured to 1. 03 MPa with ethylene and was stirred in a constant-temperature water bath at 60°C for 120 min as ethylene polymerized. The auto- clave was then vented and the resulting suspension of poly-

mer powder was diluted with acetone, filtered and oven-dried to yield 27. 7 g (600 kg polyethylene/g Ni) polyethylene pow- der. This control example of batch addition of iodine shows the great improvement afforded by continuous iodine addition to the polymerizations described in Example 1.

Example 2 A 600-mL stirred autoclave was loaded with 250 mL dry isooctane containing 2. 5 mL 1, 3-dichloro-1, 3- diisobutylalumoxane (0. 35 M in hexane) under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The solvent was saturated with ethylene by stirring at 350 kPa/70°C for 10 min.

A solution of 80 mg iodine was made up in 250 mL dry, deaerated isooctane. The autoclave was vented and the io- dine solution was pumped into the autoclave at 0. 5 mL/min (50 equivalents of iodine per hour based on Ni) using a high-pressure liquid chromatography pump via a feed line through one of the autoclave head ports. After 15 sec of iodine solution feed, a solution of 0. 5 mg of the Ni com- pound in 3 mL methylene chloride was injected via syringe through a head port. The autoclave was immediately pres- sured to 1. 03 MPa with ethylene and was stirred in a con- stant-temperature water bath at 70°C for 120 min as ethylene polymerized. The iodine feed was then stopped and the auto- clave was vented. The viscous polymer suspension was di- luted with acetone and filtered, and the resulting powder was oven-dried to yield 46. 0 g (993 kg polyethylene/g Ni) polyethylene powder.

Example 3 A 600-mL stirred autoclave was loaded with 200 mL dry isooctane under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The isooctane was saturated with ethylene by stirring at 70 kPa for 10 min

as the autoclave was cooled in an ice bath to 0°C. The eth- ylene was vented and a solution of 320 mg iodine in 100 mL dry, deaerated isooctane was pumped into the autoclave at 0. 2 mL/min via a feed line through one of the autoclave head ports. After 1 min of iodine solution feed, a slurry of 100 mg silica-supported (III) in 5 mL isooctane was injected via syringe through a head port. This supported catalyst was made by contacting a toluene solution of (III) with silica- supported methylalumoxane ; 100 mg of the catalyst contained 2 mg of (III) and 18 wt% aluminum in the form of methylalu- moxane. The autoclave was immediately pressured to 1. 03 MPa with ethylene and was stirred at 0-3°C/1. 03 MPa for 30 min as ethylene polymerized. Then the autoclave was heated in a hot water bath to quickly raise the internal temperature to 60°C and the polymerization was continued at 1. 03 MPa/60°C for 3. 5 hr more. The iodine feed then was stopped, and the autoclave was vented. The polymer suspension was diluted with acetone, suction-filtered and oven-dried to yield 45. 9 g (249 kg polyethylene/g Ni) fine polymer beads.

Comparative Example 3 A 600-mL stirred autoclave was loaded with 200 mL dry isooctane under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The isooctane was saturated with ethylene by stirring at 70 kPa for 10 min as the autoclave was cooled in an ice bath to 0°C. The eth- ylene was vented and a slurry of 100 mg silica-supported (III) in 5 mL isooctane was injected via syringe through a head port. This is a sample of the same catalyst batch used in Example 3. The autoclave was immediately pressured to 1. 03 MPa with ethylene and was stirred at 0-3°C/1. 03 MPa for 30 min as ethylene polymerized. Then the autoclave was heated in a hot water bath to quickly raise the internal temperature to 60°C and the polymerization was continued at

1. 03 MPa/60°C for 3. 5 h more. The autoclave was vented.

The polymer suspension was diluted with acetone, suction- filtered and oven-dried to yield 34. 2 g (185 kg polyethyl- ene/g Ni) fine polymer beads. This control example with no added iodine shows the great improvement afforded by iodine addition to the polymerizations described in Example 3.

Example 4 A 600-mL stirred autoclave was loaded with 200 mL dry isooctane containing 0. 5 mL modified methylalumoxane (Akzo MMAO-3A ; 1. 7 M in hexane ; contains about 30% isobutyl groups) under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The solvent was saturated with ethylene by stirring at 350 kPa/60°C for 10 min.

A solution of 80 mg iodine was made up in 250 mL dry, deaerated isooctane. The autoclave was vented and the io- dine solution was begun pumping into the autoclave at 0. 3 mL/min (50 equivalents of iodine per hour based on Ni) using a high-pressure liquid chromatography pump via a feed line through one of the autoclave head ports. After 15 sec of iodine solution feed, a solution of 0. 5 mg (III) in 3 mL methylene chloride was injected via syringe through a head port. The aluminum : nickel molar ratio was 1000. The auto- clave was immediately pressured to 1. 03 MPa with ethylene and was stirred in a constant-temperature water bath at 60°C for 120 min as ethylene polymerized. The iodine feed was then stopped and the autoclave was vented. The viscous polymer suspension was diluted with acetone and filtered, and the resulting powder was oven-dried to yield 21. 7 g (984, 300 turnovers ; 470 kg polyethylene/g Ni) polyethylene powder.

Example 5 This polymerization was conducted identically to Exam- ple 4, except 0. 25 mL MMAO-3A was used (Al : Ni=500). The polymer yield was 30. 2 g (1, 370, 000 turnovers ; 654 kg poly- ethylene Ni).

Example 6 This polymerization was conducted identically to Exam- ple 5, except a solution of 80 mg iodine in 125 mL isooctane was fed to the polymerization at 0. 5 mL/min (I2 : Ni=100/hr).

The polymer yield was 39. 0 g (1, 769, 000 turnovers ; 845 kg polyethylene/g Ni).

Examples 4, 5 and 6 demonstrate the manner in which to balance the relative quantities of activator (e. g., aluminum alkyl) and iodine.

Example 7 A 600-mL stirred autoclave was loaded with 200 mL dry isooctane containing 2. 0 mL 1, 3-dichloro-1, 3- diisobutylalumoxane (0. 35 M in hexane) under nitrogen, and the nitrogen was displaced by pressuring with ethylene and venting 3 times. The solvent was saturated with ethylene by stirring at 350 kPa/60°C for 10 min.

A solution of 60 mg methyl iodide was made up in 120 mL dry, deaerated isooctane. The autoclave was vented and the methyl iodide solution was begun pumping into the autoclave at 0. 5 mL/min (130 equivalents of methyl iodide per hour based on Ni) using a high-pressure liquid chromatography pump via a feed line through one of the autoclave head ports. After 15 sec of methyl iodide solution feed, a solu- tion of 0. 5 mg (III) in 3 mL methylene chloride was injected via syringe through a head port. The aluminum : nickel molar ratio was 1800. The autoclave was immediately pressured to 1. 03 MPa with ethylene and was stirred in a constant- temperature water bath at 60°C for 120 min as ethylene po-

lymerized. The methyl iodide feed was then stopped and the autoclave was vented. The viscous polymer suspension was diluted with acetone and filtered, and the resulting powder was oven-dried to yield 33g (1, 492, 000 turnovers ; 712 kg polyethylene/g Ni) polyethylene powder.

Examples 8 and 9 In each of these examples 1. 6 mg (2. 5 jjmol) of (III) was used. An iodine stock solution of 80 mg in 20 ml of methylene chloride was made up and used (31 pmol/2 mL). The MMAO-3A solution in heptane contained 6. 8 weight percent Al, and IBCAO (isobutylchloroaluminoxane) solution in toluene was 0. 34 M. A 3. 8 (1 gal) stirred autoclave was used.

Granular NaCl (1000 g) was added to the reactor, and the NaCl dried at 150°C under N2 for 24 h, then purged with Ar for 50 min at 100°C, then purged with ethylene for 10 min.

The autoclave was cooled to 20°C, and then vented with Ar and then 4 times with ethylene. The scavenger was added with 1 L of propane and the contents stirred for 15 min. Then, separately, (III) was contacted with the cocatalyst in 4 mL methylene chloride and allowed to react for 10 min. This solution was then added to the reactor with 250 mL propane, stirred 2 min, the propane vented to <70 kPa, while the autoclave setpoint was still at 20°C. The ethylene was added to a pressure of 700 kPa, and these conditions maintained for 10 min. Then (time = 0) the temperature setpoint was increased to 60°C, 2 mL of the iodine stock solution was added and the ethylene pressure was increased to 2. 41 MPa.

At 15, 30 and 45 min 2 mL of the iodine stock solution was injected. After 60 min the polymerization was terminated.

The polymer was isolated by washing with 80°C water 4 times in a blender. Other detail of these Examples are given in Table 1.

Table 1 Example 8 9 Scavenger IBCAO MMAO-3A (mL) 3. 3 1. 2 (mmol) 2. 3 2. 2 (Al/Ni) 900 880 Cocatalyst I BCAO I BCAO (mL) 0. 4 0. 4 (mmol) 0. 3 0. 3 (Al/Ni) 100 100 PE 45 97 kg PE/g Ni 307 661

The productivities (kg PE/g Ni, PE is polyethylene) in Examples 9 and 10 are about 40-50% higher than those ob- tained in similar polymerizations without iodine present.

Example 10 A 600 mL stirred autoclave reactor was dried at 130°C under vacuum, purged with nitrogen pressure (1. 72 MPa) three times, and cooled under pressure. The nitrogen was dis- placed with ethylene (1. 03 MPa) by pressuring and venting the reactor three times. The reactor was charged with 200 mL dry isooctane containing 0. 46 mL (0. 16 mmol) 1, 3- dichloro-1, 3-diisobutylaluminoxane (0. 34 M in toluene), heated to 60°C, and saturated with ethylene while stirring at 1000 rpm.

A solution of iodine in dry isooctane (0. 66 mM) under nitrogen was pumped into the autoclave at 0. 20 mL/min (20 equivalents of iodine per hour based on Ni) using a high- pressure liquid chromatography pump. After 30 sec of iodine solution feed, a solution of 0. 25 mg (0. 39 pmol) of the Ni compound (III) dissolved in 3 mL dichloromethane and diluted with 7 mL toluene was injected into the reactor using a 0. 1 MPa overpressure of nitrogen. The reactor was maintained at 1. 03 MPa and 60°C in a constant-temperature water bath for 120 min. The ethylene uptake was monitored using calibrated

mass flow meters and the pressure drop from a gas reservoir.

The polymerization was terminated by stopping the iodine feed and venting the reactor pressure. The viscous polymer suspension was diluted with acetone. The polymer was col- lected by vacuum filtration and dried at 70°C in a nitrogen- purged vacuum oven to yield 17. 9 g (782 kg polyethylene/g Ni) polyethylene powder. Additional Examples that follow this procedure are shown in Table 2.

Table 2 Polymerizations using nickel complex (III), catalyst activa- tors, and reactivators Example (III) Ral AI : Ni RI RI : Ni Time Yield M Equiv. equiv./hr min kg/g Ni 10 2 (iBuAICI) 20 800 12 20 120 782 CE4 2 (iBuAfCI) 20 800--120 478 11 2 Et3Al2Cl3 800 I2 20 120 1167 CE5 2 Et3Al2CI3 800--120 438 12 2 Et3Al2Cl3 800 I2 20 225 1694 13 2 Et3Ai2CI3 800 12 20 90 947 14 2 iBu2AICI 800 12 20 120 849 15 2 EtAICI2 800 12 20 120 497 16 2 Et3AI2CI3 800 ICH2CH2i 20 120 584 17 2 Et3Al2Cl3 800 CH2I2 20 120 673 18 2 Et3Al2Cl3 800 CHI3 20 120 711 19 2 Et3Al2Cl3 800 CI4 10 120 683 20 2 Et3AI2CI3 800 C i4 20 120 1002 21 2 Et3AI2CI3 800 Cl4 40 120 857 Example 22 The procedure of Example 10 was modified by replacing the 1, 3-dichloro-1, 3-diisobutylaluminoxane with 0. 17 mL (0. 16 mmol) ethylaluminum sesquichloride (0. 91 M in toluene) and the isooctane with cyclohexane as solvent. The polym- erization was terminated after 90 min by stopping the iodine feed and venting the reactor pressure. The polymer was

fully dissolved in the cyclohexane until it was precipitated by diluting with acetone to yield 21. 67 g (947 kg polyethyl- ene/g Ni) polyethylene powder. The melt index (190°C, 2160 g) was 18. The branch content was 26. 9 CH3/1000 CH2 by 1H NMR (500 MHz, C12CDCDC12, 120°C). The peak melting tempera- ture was 102°C (107 J/g) by DSC on the second heating cycle.

The molecular weights were Mn of 22, 941 and Mw of 45, 022 by GPC (1, 2, 4-trichlorobenzene, 135°C, universal calibration as polyethylene).

Example 23 The procedure of Example 10 was modified by replacing the 1, 3-dichloro-1, 3-diisobutylaluminoxane with 0. 17 mL (0. 16 mmol) ethylaluminum sesquichloride (0. 91 M in tolu- ene) and the solution of iodine with a solution of tetraio- domethane in dry isooctane (0. 66 mM) that was pumped at 0. 20 mL/min (20 equivalents per hour based on Ni) to yield 22. 95 g (1002 kg polyethylene/g Ni) polyethylene powder. The so- lution of CI4 was prepared in near darkness and the transfer lines protected from light. The melt index (190°C, 2160 g) was 28. 6. The branch content was 33. 1 CH3/1000 CH2 by 1H NMR (500 MHz, C12CDCDC12, 120°C). The peak melting temperature was 98°C (95 J/g) by DSC on the second heating cycle. The molecular weights were Mn of 21, 109 and Mw of 41, 358 by GPC (1, 2, 4-trichlorobenzene, 135°C, universal calibration as polyethylene).

Comparative Example 6 A 600 mL stirred autoclave reactor was dried at 130°C under vacuum, purged with nitrogen pressure (1. 72 MPa) three times, and cooled under pressure. The nitrogen was dis- placed with ethylene (1. 03 MPa) by pressuring and venting the reactor three times. The reactor was charged with 200 mL dry isooctane containing 0. 17 mL (0. 16 mmol) ethylalumi-

num sesquichloride (0. 91 M in toluene), heated to 70°C, and saturated with ethylene while stirring at 1000 rpm.

A solution of 0. 25 mg (0. 38 jmol) of the Ni compound (IV) dissolved in 3 mL dichloromethane and diluted with 7 mL isooctane was injected into the reactor using a 0. 1 MPa overpressure of nitrogen. The reactor was maintained at 1. 03 MPa and 70°C in a constant-temperature water bath. The ethylene uptake was monitored using calibrated mass flow me- ters and the pressure drop from a gas reservoir. The polym- erization was terminated after 45 min by venting the reactor pressure because the ethylene uptake had ceased. The poly- mer was fully dissolved in the isooctane until it was pre- cipitated by diluting with acetone. The polymer was col- lected by vacuum filtration and dried at 70°C in a nitrogen- purged vacuum oven to yield 3. 68 g (165 kg polyethylene/g Ni) polyethylene powder. The melt index (190°C, 2160 g) was 5. 7. The branch content was 66. 7 CH3/10 0 0 CHZ by 1H NMR (500 MHz, C12CDCDC12, 120°C). The peak melting temperature was 62°C (42 J/g) by DSC on the second heating cycle. The mo- lecular weights were Mn of 38, 292 and Mw of 63, 831 by GPC (1, 2, 4-trichlorobenzene, 135 °C, universal calibration as polyethylene).

Example 24 A 600 mL stirred autoclave reactor was dried at 130°C under vacuum, purged with nitrogen pressure (1. 72 MPa) three times, and cooled under pressure. The nitrogen was dis- placed with ethylene (1. 03 MPa) by pressuring and venting the reactor three times. The reactor was charged with 200 mL dry isooctane containing 0. 17 mL (0. 16 mmol) ethylalumi- num sesquichloride (0. 91 M in toluene), heated to 70°C, and saturated with ethylene while stirring at 1000 rpm.

A solution of iodine in dry isooctane (0. 66 mM) under nitrogen was pumped into the autoclave at 0. 20 mL/min (20

equivalents of iodine per hour based on Ni) using a high- pressure liquid chromatography pump. After 30 sec of iodine solution feed, a solution of 0. 25 mg (0. 38 mol) of the Ni compound (IV) dissolved in 3 mL dichloromethane and diluted with 7 mL isooctane was injected into the reactor using a 0. 1 MPa overpressure of nitrogen. The reactor was main- tained at 1. 03 MPa and 70°C in a constant-temperature water bath for 120 min. The ethylene uptake was monitored using calibrated mass flow meters and the pressure drop from a gas reservoir. The polymerization was terminated by stopping the iodine feed and venting the reactor pressure. The poly- mer was fully dissolved in the isooctane until it was pre- cipitated by diluting with acetone. The polymer was col- lected by vacuum filtration and dried at 70°C in a nitrogen- purged vacuum oven to yield 9. 88 g (443 kg polyethylene/g Ni) polyethylene powder. The melt index (190°C, 2160 g) was 5. 2. The branch content was 72. 1 CH3/1000 CH2 by lH NMR (500 MHz, C12CDCDC12, 120°C). The peak melting temperature was 61°C (42 J/g) by DSC on the second heating cycle. The mo- lecular weights were Mn of 41, 455 and Mw of 70, 502 by GPC (1, 2, 4-trichlorobenzene, 135°C, universal calibration as polyethylene).

Example 25 The procedure of Example 24 was modified by charging 350 kPa of hydrogen to the reactor from a 10. 1 mL addition cylinder after charging the isooctane to yield 6. 81 g (305 kg polyethylene/g Ni) polyethylene powder. The melt index (190°C, 2160 g) was 15. The branch content was 62. 2 CH3/1000 CH2 by 1H NMR (500 MHz, Cl2CDCDCl2, 120°C). The peak melting temperature was 59°C (29 J/g) by DSC on the second heating cycle. The molecular weights were Mn of 27, 009 and Mw of 50, 741 by GPC (1, 2, 4-trichlorobenzene, 135°C, universal calibration as polyethylene).

Example 26 The procedure of Example 24 was modified by charging 690 kPa of hydrogen to the reactor from an 10. 1 mL addition cylinder to yield 5. 56 g (249 kg polyethylene/g Ni) polyeth- ylene powder. The melt index (190°C, 2160 g) was 34. The branch content was 78. 1 CH3/1000 CH2 by 1H NMR (500 MHz, C12CDCDC12, 120°C). The peak melting temperature was 57°C (16 J/g) by DSC on the second heating cycle. The molecular weights were Mn of 23, 970 and Mw of 45, 700 by GPC (1, 2, 4- trichlorobenzene, 135°C, universal calibration as polyethyl- ene).

Example 27 A 600 mL stirred autoclave reactor was dried at 130°C under vacuum, purged with nitrogen by pressuring (1. 72 MPa) and venting the reactor three times, and cooled under nitro- gen pressure. The nitrogen was displaced with ethylene by pressuring (450 kPa) and venting the reactor three times.

The reactor was charged with 200 mL dry isooctane containing 0. 46 mL (79 J, mol, 100 equiv. Zn per Ni) of a solution of di- ethylzinc, Et2Zn, in toluene (0. 17 M), cooled to 0°C, and the isooctane was saturated with ethylene (450 kPa) while stir- ring at 1000 rpm.

A solution of 0. 5 mg (0. 79 p, mol) of the Ni complex (III) and 1. 26 mg (1. 57 pmol, 2 equiv. per Ni) of N, N- dimethylanilinium tetrakis (pentafluorophenyl) borate, PhNMe2H B (C6F5) 4, which were dissolved in 1 mL dichloromethane and diluted with 9 mL toluene, was injected into the reactor us- ing a 0. 1 MPa overpressure of nitrogen. After allowing 5 min to initiate the polymerization at 0°C, a solution of io- dine in dry isooctane (1. 31 mM) under nitrogen was pumped into the autoclave at 0. 40 mL/min (40 equivalents of iodine per hour based on Ni) using a high-pressure liquid chroma- tography pump and the reactor was heated rapidly to 60°C in

a constant-temperature water bath as the ethylene pressure was increased to 1. 03 MPa. The ethylene uptake was moni- tored using calibrated mass flow meters and the pressure drop from a gas reservoir. The polymerization was termi- nated after 120 min by stopping the iodine feed and venting the reactor pressure. The viscous polymer suspension was diluted with acetone. The polymer was collected by vacuum filtration and dried at 70°C in a nitrogen-purged vacuum oven to yield 21. 47 g (463 kg polyethylene/g Ni) polyethyl- ene powder. The melt index (190°C, 2160 g) was 6. 0. The branch content was 29. 1 CH3/1000 CH2 by 1H NMR (500 MHz, Cl2CDCDCl2, 120°C). The peak melting temperature was 95°C (104 J/g) by DSC on the second heating cycle. The molecular weights by GPC (1, 2, 4-trichlorobenzene, 135 °C, universal calibration as polyethylene) were Mn of 19, 200 and Mw of 61, 700. Additional Examples that followed this procedure are shown in Table 3.

Table 3 Polymerizations using nickel complex (III), catalyst activa- tor, N, N-dimethylanilinium tetrakis (pentafluorophenyl) borate, and iodine Example (III) RxM M : Ni B : Ni 12 : Ni Time Yield M Equiv. equiv. equiv./hr min kg/g Ni 27 4 Et2Zn 100 2 40 120 463 28 4 Et2Zn 100 2 20 120 439 29 4 Et2Zn 100 2 80 90 351 CE7 4 Et2Zn 100 2-90 328 30 4 iBu3AI 100 2 5 120 317 CE8 4 iBu3AI 100 2-120 295 CE9 4 iBu3AI 800 2 20 90 137 Comparative Example 10 The procedure of Example 27 was modified by omitting the iodine feed. The ethylene uptake ceased after 85 min, so the polymerization was terminated by venting the reactor

pressure after 90 min. The viscous polymer suspension was diluted with acetone. The polymer was collected by vacuum filtration and dried at 70°C in a nitrogen-purged vacuum oven to yield 15. 23 g (328 kg polyethylene/g Ni) polyethyl- ene powder. The melt index (190°C, 2160 g) was 5. 3. The branch content was 30. 0 CH3/1000 CH2 by 1H NMR (500 MHz, C12CDCDC12, 120°C). The peak melting temperature was 96°C (113 J/g) by DSC on the second heating cycle. The molecular weights by GPC (1, 2, 4-trichlorobenzene, 135°C, universal calibration as polyethylene) were Mn of 26, 600 and Mw of 74, 400X.

Examples 31-39 and Comparative Examples 11-16 Thermal analysis : DSC measurements were made using a DSC Mettler. The instrument is calibrated with indium and tin standards. A weighed sample (5-10 mg) was sealed into an aluminum pan, heated to 200°C and kept at that temperature for 5 min to allow a complete melting of all the crystal- lites. The sample was cooled at 20°C/min to 0°C. After standing 5 minutes at 0°C, the sample was heated to 200°C at a rate of 10°C/min. In this second heating run, the peak temperature was assumed as melting temperature (Tm) and the area as the melting enthalpy (AHf).

Intrinsic viscosity was measured in tetrahydronaph- thalene (THN) solution obtained by dissolving the polymer powder at 135°C for 1 h.

Modified methylalumoxane MMAO-3A, containing 7 wt% Al in heptane, was purchased from Akzo Nobel. 1, 3-Dichloro-1, 3- diisobutylaluminoxane (IBCAO) (0. 34 M Al in toluene) was sup- plied by Aldrich.

Example 31 A 250 ml glass autoclave, provided with magnetic stir- rer, temperature indicator and ethylene feed line, was puri- fied and fluxed with ethylene at 35°C. n-Heptane (100 ml)

and 0. 88 ml (1. 6 mmol) of MMAO were introduced into the re- actor at RT. The temperature was then raised to 50°C and the ethylene pressure to 520 kPa in order to allow the mono- mer to dissolve into the solvent. The catalyst system was separately prepared by introducing 0. 18 ml (0. 4 mmol) of MMAO in 2 ml of n-heptane and then adding 1 ml of a 2 mM so- lution of (IV) in CH2Cl2 (0. 002 mmol, Al/Ni = 200 mol/mol).

The catalyst solution was stirred for 5 min and then intro- duced into the autoclave after venting ethylene until the pressure reaches 0 kPa. The reactor was closed and pressur- ized again with ethylene to 520 kPa. The total pressure was kept constant by feeding ethylene. After 15 min, the reac- tor was vented and 0. 10 mmol of n-BPCC, diluted in 2 ml of n-heptane, was introduced into the polymerization. The pressure was raised again, immediately after the n-BPCC in- jection, and the polymerization proceeded for 15 more min.

It was stopped by cooling, venting the reactor and introduc- ing 1 ml of methanol. The polymer was coagulated with ace- tone/HCl, filtered and dried in vacuum at 60°C. Polymer (6. 7 g) was recovered. Polymerization data are summarized in Table 4.

Example 32 The procedure described in Example 31 was repeated with the only difference that ETA was used instead of n-BPCC.

Polymer (7. 5 g) was recovered. Polymerization data are sum- marized in Table 4.

Comparative Example 11 The procedure described in Example 31 was repeated with the only difference that the polymerization was carried on for 30 min, without any interruption and no n-BPCC was added. Polymer (5. 9 g) was recovered. Polymerization data are summarized in Table 4.

Table 4 Example Additive Time of in-Yield Activity jection (g) (kg PE/g (min) Ni) 31 n-BPCC 15 6. 7 75 32 ETA 15 7. 5 64 CE11 None 5. 9 51

Example 33 Preparation of a homopolymer of propylene [step (I)] : In a 100 ml glass flask, treated beforehand in nitrogen at 90°C for three h, 0. 0098 g of a solid component containing titanium, prepared according to the procedure of Example 3 of EP-A-395083, 0. 76 g of triethylaluminum (TEAL) and 0. 313 g of cyclohexylmethyldimethoxysilane (CMMS) were brought into contact in 10 ml of hexane for 5 min. The mixture was then fed into a 4-1 steel autoclave, which had been treated beforehand with nitrogen at 90°C for 3 h. Feed was effected at 30°C in a propylene atmosphere. 1-L of H2 and 1. 2 kg of propylene were then introduced and the autoclave was heated to 70°C. Polymerization was effected for two h, followed by degassing in a stream of N2 at 70°C for 1 h. Spherical poly- mer (238 g) with the following characteristics were ob- tained : MIL=3. 5 ; porosity (% voids) = 24%.

Treatment of the homopolymer with a deactivating agent [step (II) (a)] : After degassing propylene, the same reactor was charged with 1000 ml of hexane moistened with 0. 513 g of water. It was left in contact with the polymer at 50°C for 30 min, under a nitrogen atmosphere. The liquid was removed by siphoning and some washings with cycles of vac- uum/nitrogen were carried out at RT.

Treatment of the deactivated homopolymer with MMAO/ (III) mixture [stage (II) (b)] and ethylene polymeriza-

tion [stage (III)] : 70-g of porous deactivated polypropylene (70 g) from stage (II) (a) were loaded into a 2. 6 L stainless steel reactor. Liquid propane (500 ml) and 3. 7 ml of MMAO solution (6. 8 mmol of Al ; Al/Ni = 2000 mol/mol) diluted in 2. 3 ml of n-heptane, were introduced into the reactor, to scavenge liquid propane. The reactor was maintained at 25°C, under a total pressure of 1. 08 MPa. At the same time, 2. 16 mg (3. 4 pmol) of (IV) was dissolved in 6. 0 ml of toluene in a glass vessel, and 0. 62 ml of MMAO solution (0. 34 mmol of Al ; Al/Ni = 100 mol/mol) was then added. The mixture was stirred for 2 min at RT and then loaded into the reactor with a nitrogen overpressure. After 2-3 min of stirring, the pro- pane was flashed off in a few minutes, while maintaining the temperature at 20-30°C. Then, ethylene was fed all at once to a total pressure of 2. 48 MPa ; ethylene concentration was about 1. 17 mol/L. The polymerization was carried out at 60°C for 4 h, keeping the pressure constant by feeding ethylene.

During the polymerization, a 0. 017 M solution of n-BPCC in n- heptane, was injected with a ethylene. overpressure, according to the following scheme : 3. 0 ml at the 18th minute 3. 5 ml at the 45th minute 3. 5 ml at the 75th minute 3. 5 ml at the 180th minute.

The overall n-BPCC/Ni molar ratio was 68. After stop- ping the polymerization reaction by venting off ethylene, the reactor was cooled to RT and 247 g of polymer was ob- tained, in the form of free-flowing spheroid particles, con- taining 71. 6 wt% of ethylene polymer. A total of 177 g of ethylene polymer was produced, corresponding to an activity of 890 kg PE/g Ni, having Tm = 131. 8°C (experimental) and I. V. =1. 97 dl/g (calculated). Polymerization data are summa- rized in Table 5.

Example 34 The procedure described in Example 33 was repeated with the only difference that the polymerization was stopped af- ter 2 h instead of 4 and the n-BPCC solution was 0. 05 M and was injected according to the following scheme : 3. 0 ml at the 10th minute 3. 5 ml at the 25th minute 3. 5 ml at the 40th minute The overall n-BPCC/Ni molar ratio was 150. A total of 235 g of polymer was obtained in the form of free-flowing spheroid particles, containing 70 wt% of ethylene polymer.

A total of 165 g of ethylene polymer was produced, corre- sponding to an activity of 827 kg PE/g Ni. Polymerization data are summarized in Table 5.

Example 35 The procedure described in Example 34 was repeated with the only difference that ethyl trichloroacetate (ETA) was used as oxidant, instead of n-BPCC. ETA solution (0. 05 M in n-heptane) was injected according to the following scheme : 3. 0 ml at the 10th minute 3. 5 ml at the 25th minute 3. 5 ml at the 40th minute The overall ETA/Ni molar ratio was 150. A total of 173 g of polymer was obtained, in the form of free-flowing sphe- roid particles, containing 59 wt% of ethylene polymer. A total of 103 g of ethylene polymer were produced, correspond- ing to an activity of 518 kg PE/g Ni. Polymerization data are summarized in Table 5.

Comparative Example 12 The procedure described in Example 34 was repeated with the only difference that no n-BPCC was added during the po- lymerization run. A total of 145 g of polymer composition was obtained, in the form of free-flowing spheroid particles,

containing 52 wt% of ethylene polymer. A total of 75 g of ethylene polymer was produced, corresponding to an activity of 357 kg PE/g Ni. Polymerization data are summarized in Ta- ble 5.

Example 36 The procedure described in Example 34 was repeated with the difference that only 26 g of porous polypropylene (in- stead of 70 g) and 6. 48 mg (0. 010 mmol) of (IV) (instead of 2. 16 mg) were introduced into the reactor. The Al/Ni molar ratio was still 100 in the precontact solution. l MMAO (10. 2 ml) was used to scavenge liquid propane, instead of 6. 8 mmol. n-BPCC solution (0. 05 M in n-heptane) was injected according to the following scheme : 3. 0 ml at the 20th minute 3. 5 ml at the 40 th minute 3. 5 ml at the 60th minute The overall n-BPCC/Ni molar ratio was 50. A total of 147 g of polymer was obtained in the form of free-flowing spheroid particles, containing 82. 3 wt% of ethylene polymer.

A total of 121 g of ethylene polymer was produced, corre- sponding to an activity of 203 kg PE/g Ni. The PE had Tm = 129°C (experimental) and I. V. =1. 86 dl/g (calculated). Polym- erization data are summarized in Table 5.

Comparative Example 13 The procedure described in Example 36 was repeated with the only difference that n-BPCC was added during the polym- erization run. A total of 70 g of polymer was obtained in the form of free-flowing spheroid particles, containing 62. 9 wt% of ethylene polymer. A total of 44 g of ethylene polymer was produced, corresponding to an activity of 74 kg PE/g Ni.

Polymerization data are summarized in Table 5.

Example 37 The procedure described in Example 36 was repeated with the only difference that the polymerization temperature was 70°C instead of 60°C. n-BPCC solution (0. 05 M in n-heptane) was injected according to the following scheme : 3. 0 ml at the 5th minute 3. 5 ml at the 15th minute 3. 5 ml at the 30th minute The overall n-BPCC/Ni molar ratio was 150. A total of 189 g of polymer was obtained, in the form of free-flowing spheroid particles, containing 63 wt% of ethylene polymer.

A total of 119 g of ethylene polymer was produced, corre- sponding to an activity of 598 kg PE/g Ni. Polymerization data are summarized in Table 5.

Comparative Example 14 The procedure described in Example 37 was repeated with the only difference that no n-BPCC was added during the po- lymerization run. A total of 100 g of polymer was obtained, in the form of free-flowing spheroid particles, containing 31 wt% of ethylene polymer. A total of 31 g of ethylene polymer was produced, corresponding to an activity of 156 kg PE/g Ni.

The PE had a Tm = 126°C (experimental) and I. V. =3. 41 dl/g (calculated). Polymerization data are summarized in Table 5.

Example 38 The procedure described in Example 34 was repeated with the only difference that 1, 3-dichloro-1, 3- diisobutylaluminoxane (IBCAO) was used as the cocatalyst, in- stead of MMAO. n-BPCC solution (0. 017 M in n-heptane) was injected according to the following scheme : 3. 0 ml at the 15th minute 3. 5 ml at the 30th minute 3. 5 ml at the 50th minute

The overall n-BPCC/Ni molar ratio was 50. A total of 160 g of polymer as obtained, in the form of free-flowing spheroid particles, containing 56 wt% of ethylene polymer.

A total of 90 g of ethylene polymer was produced, correspond- ing to an activity of 452 kg PE/g Ni. Polymerization data are summarized in Table 5.

Comparative Example 15 The procedure described in Example 38 was repeated with the only difference that no n-BPCC was added during the po- lymerization run. A total of 130 g of polymer was obtained, in the form of free-flowing spheroid particles, containing 45 wt% of ethylene polymer. At total of 59 g of ethylene poly- mer was produced, corresponding to an activity of 296 kg PE/g Ni. Polymerization data are summarized in Table 5.

Example 39 The procedure described in Example 34 was repeated with the only difference that (III) was used instead of (IV) and the polymerization temperature was 50°C, instead of 60°C. n-BPCC solution (0. 05 M in n-heptane) was injected according to the following scheme : 3. 0 ml at the 5th minute 3. 5 ml at the 15th minute 3. 5 ml at the 30th minute The overall n-BPCC/Ni molar ratio was 150. A total of 222 g of polymer was obtained in the form of free-flowing spheroid particles, containing 68 wt% of ethylene polymer.

A total of 152 g of ethylene polymer was produced, corre- sponding to an activity of 762 kg PE/g Ni. Polymerization data are summarized in Table 5.

Comparative Example 16 The procedure described in Example 39 was repeated with the only difference that no n-BPCC was added during the po- lymerization run. A total of 124 g of polymer was obtained

in the form of free-flowing spheroid particles, containing 43. 5 wt% of ethylene polymer. A total of 54 g of ethylene polymer was produced, corresponding to an activity of 271 kg PE/g Ni, having Tm = 128°C (experimental) and I. V. =5. 63 dl/g (calculated). Polymerization data are summarized in Table 5.

Table 5 Addi-time of injec-Activity Example [Ni] tive/Ni tion T t (kg PE/g (mol/mol) (min) Ni) 33 2. 7 68 18, 45, 75, 180 60 4 890 34 2. 7 150 10, 25, 40 60 2 827 35 2. 7 150 10, 25, 40 60 2 518 CE12 2. 7 60 2 377 36 23 50 20, 40, 60 60 2 203 CE13 23 60 2 74 37 2. 7 150 5, 15, 30 70 2 598 CE14 2. 7 70 2 156 38 2. 7 50 15, 30, 50 60 2 450 CE15 2. 7 60 2 300 39 2. 8 150 5, 15, 30 50 2 762 CE16 2. 8=502271 * ppm of the Ni complex Examples 40-42 and Comparative Example 17 Catalyst Preparation The supported catalyst was made by contacting a toluene solution of 41. 3 mg of (III) with 1. 0 g silica supported methylalumoxane containing a nominal alumi- num content of 14 weight percent (Albermarle Corp., Baton Rouge, LA, USA). The resulting suspension was decanted and dry pentane was added. This suspension was also decanted and a second addition of pentane was made. After a final decanting, the solids were dried under vacuum. Assuming all of (III) became attached to the silica, the catalyst con- tained 3. 7 µg/mg of supported catalyst composition.

Example 40 A solution of 69. 3 mg a, trichlorotoluene was made up in 236 mL dry toluene and transferred to a high-pressure

liquid chromatography pump. A 500-mL stirred autoclave was fitted with addition tubes, one containing 1521 mg of sil- ica-supported methylalumoxane as solvent scavenger suspended in 3 ml dry toluene, and the other containing 112 mg of the supported catalyst suspended in 2 ml of dry cyclohexane.

The autoclave was loaded with 250 mL dry cyclohexane.

The scavenger was pressure-transferred into the solvent and the pressure vented to zero. The stirrer was started and the autoclave heated to 60°C. The solvent was saturated with ethylene by stirring at 1. 38 MPa/60°C for 2 min. The cata- lyst was pressure-injected into the autoclave to initiate polymerization. After 1 min, pumping of the trichlorotolu- ene solution was begun at 0. 64 mL/min via a feed line through one of the autoclave head ports. Temperature was maintained at 60°C for 120 min as ethylene polymerized.

The trichlorotoluene feed was then stopped and the autoclave was vented. The polymer suspension was filtered, rinsed with acetone and the resulting powder was oven-dried to yield 74. 3 g polyethylene powder. Productivity was about 179 kg PE/g Ni.

Comparative Example 17 The procedure of Example 40 was followed, except that 100. 1 mg of supported catalyst was used, and trichlorotolu- ene solution was pumped during the polymerization. After a 2 h polymerization time, the suspension was filtered, rinsed and dried to yield 16. 2 g of polymer. This control example with no added trichlorotoluene shows the great improvement afforded by addition of the oxidant to the polymerization described in Example 40. Productivity was about 44 kg PE/g Ni.

Example 41 The procedure of Example 40 was followed, except the prepared solution contained 29. 6 g of ethyl trichloroacetate

in 104 ml toluene. Also, 51 mg of the supported catalyst was charged and after 1 min, and the pumping rate of the ethyl trichloroacetate solution was 0. 62 ml/min. After a 2 h polymerization time, the polymer suspension was filtered and rinsed. The resulting powder was oven-dried to yield 17. 3 g polyethylene powder. Productivity was about 92 kg PE/g Ni.

Example 42 The procedure of Example 40 was followed, except the prepared solution contained 22. 7 g 1, 2-dibromoethane in 79 ml toluene. Also, 53 mg of the supported catalyst was charged and the pumping rate of dibromoethane solution was 0. 61 ml/min. After a 2 h polymerization, 10. 9 g of dried polyethylene was isolated. Productivity was about 56 kg PE/g Ni.