CA2159388A1 - Colorable non-sticky resin, non-sticky prepolymerized catalyst, and processes for making them - Google Patents

Abstract

A colorable resin particle having an outer shell of a non-sticky polymer and an inner core of a sticky polymer produced in a gas phase fluidized bed reactor at or above the sticking temperature of the sticky polymer using a non-sticky prepolymerized catalyst and processes for producing the colorable resin and the non-sticky prepolymerized catalyst.

Description

D-17229 21~ 9388 COLOR~ABLE NON-STICKY RESIN, NON-STICKY
PREpoLylvl~R~ n CATALYST, AND PROC~ ;; FOR
MAKING THEM

Field of the Invention The invention relates to elastomeric ethylene/propylene rubbers (EPRs) such as ethylene/propylene copolymers (EPMs) and ethylene/propylene/diene terpolymers (EPDMs), which EPRs have a crystallinity of less than about 16 percent by weight, and their preparation. More particularly, the invention relates to a colorable non-sticky resin, a non-sticky prepolymerized catalyst and processes for m~king them.

R~k~round of the Invention EPRs are produced co_mercially in solution and slurry processes with soluble vanadiu_ catalysts. These processes are expensive to run because they require solvent removal and ste~m stripping steps.
The production of-EPRs in a fluidized bed in a gas phase reaction process is advantageous because of the absence of large volumes of solvent or liquid monomer, the high heat removal capacity of the fluidized bed, the granular nature of the polymer which facilitates purging and post treatment, and the ability to control reaction concentrations over a wide range of conditions without limitations imposed by the solubility of reactants and catalysts.
However, a gas phase reaction process requires that the granular resin product be free-flowing and non-sticky. EPRs are often referred to as sticky polymers because they agglomerate forming sticky, granular resin bed particles under polymerization conditions or upon the cess~tion of gas phase polymerization. A sticky polymer has been defined as a polymer which, although particulate at tempe~a~ es below the sticking or softening temperature, agglomerates at temperatures at or above the sticking te~ el~u~e. The st;~king or softening temperature has been defined as the tempe~ e at which fluidization ceases due to the agglomeration of particles of polymer in a fluidized bed. The agglomeration may be spontaneous or occur on short periods of settling.
Also, a polymer may be inherently sticky due to its chemic~l or mech~nir~l properties or pass through a sticky phase during the production cycle. Sticky polymers have also been lefe~led to as non-free flowing polymers because of their t~n-lency to comr~ct into aggregates of much larger size than the original particles and not to flow out of the relatively small openings in the bottom of the product h~rge tanks or purge bins. Polymers of this type can show acceptable ~uidity in a gas phase fluidized bed reactor; but, once motion ceases, the additional mechanical force provided by the fluidizing gas p~sing through the distributor plate is insufficient to break up the aggregates which form, and the bed will not re-fluidize.
Stickiness is even more critical with EPRs having a crystalline content of less than 15 percent by weight. Commercially desirable EPMs and EPDMs contain about 20 to about 55 percent by weight propylene and have a Mooney viscosity of about 20 to about 120.
EPDMs additionally coTlt~in about 2 to about 15 percent by weight of a non-conjugated diene to further contribute to sti~kiness. Further, EPRs are practically amorphous and they have glass transition temperatures of minus 50C to minus 60C. At tempe.a~u~es above the glass transition temperature, EPMs and EPDMs are rubbers whose viscosity decreases, like all rubbers, exponentially with increases in temperature. The viscosity decrease with rising temperatures is a major obstacle in fluidized bed production of EPR because agglomeration increases as particle surface viscosity decreases. At tempelatu~es above about 30C, amorphous EPM particles become so sticky that fl~ li7.e~ bed polymerization cannot be carried out reliably.
Particles co~ ;sing resin of lower moleclll~r weight or lower Mooney viscosity are stickier than particles comprising resin of higher molecular weight or higher Mooney viscosity, and EPDM particles particles are even stickier than EPM due to the presence of soluble dienes.
Industry has generally solved this problem of sticky polymers by avoiding polymerization at teml.eldtures at or above the stirkin~
temperature of the polymer. Such polymeri7Ation processes are disclosed, for e~Ample, in U.S. Patent Nos. 5,087,522 and 5,208,303 and WO 88/02379.
U.S. Patent Nos. 4,994,534 and 5,304,588 describe a method for overc.~.. .; ..g the tendency of EPRs to become sticky during - polymerization. In these patents, EPRs are produced in a fl~ i7e-1 bed using an alpha olefin polymerization catalyst such as a transition metal catalyst of vanadium and/or lit~i~ and an inert particulate material. The EPRs are produced by polymerizing at or above the softening or st,icking temperature of the polymer product. W~ile this procedure is e~ ~ely effective and permits operation of the fluid bed above the sticking tempelatule of the polymer, it generally results in the final polymer product cont~inin~ a relatively large amount (about 10 to 50 percent based on the total weight of the final polymer product) of the inert particulate material.
- Use of inert parti~llAte materials as fluirli7Ation aids in the amounts needed for fluid bed operations can also impart undesirable properties (merh~nicAl or melt compollntling difficulties) to the final product mAking them unsuitable for some end uses. For e~Ample, silica can be highly abrasive to the mi~ing eqllipm~nt normally used with EPDM resins and can retard sulfur vulcAni7~tion.
Carbon black always produces a black product, so that it cannot be used in markets in which a colorable resin is desired, such as rubber merhAnicAl goods for some automotive use, oil viscosity i~provers, or molded articles for consumer use. Furthermore, the inert particulate materials require storage vessels, feeding de-vices, special treAtrne~t to ensure dFyness, and the use of additional co-catalyst beyond that required by the catalyst, all of which impose additional costs on the process.

D-17229 2 1 S 9 3 8 ~

U.S. Serial No. 029,821 filed March 11, 1993 describes another method for overco...;.~g the sticky tendency of EPRs at or above the sticking tempelatule of the product resin by using a non-sticky prepolymer prepared by a prepolymeri7~tion conducted in a h~ ne slurry of ethylene and optionally a comonomer such that the comnn~m~r content of the prepolymer ranges from about 0 to about 15 weight percent based upon the total weight of the monomers.
Acco~ g to the described process an inert particulate material can be incorporated into the prepolymer or introduced direcl~ly into the fluidized bed reactor independently of the prepolymer. Attempts to prepare EPR compositions high in propylene and low in molecular weight by this procedure results in a significant amount of prepolymer residue (up to 20 percent by weight based on the weight of the final polymer product) rçm~inin~ in the product to prevent agglomeration during polymeri~tion of the final polymer resin. Or, alternatively, substantial quantities of inert particulate material are added to the reactor during polymerization of the resin product. This procedure can result in lln~lesirable levels of inert particulate material or catalyst residue in the product that can tlimini~h final polymer product properties. Furthermore, the prepolymer residues in the product contain sufficient material that is still cryst~lline at temperatures as high as 120C such that properties of the final product can be adversely affected and end use applications limited.
Thus, a process which produces a non-agglomerating or non-sticky colorable resin, without feeding an inert particulate material to the polymerization reactor, having little or no inert particulate material in the final product, and still pelll~i~illg operation of a gas phase reactor above the sticking temperatwe of the polymer being made, would be hig~ly desirable. There is also an on-going need for amorphous, colorable sticky polymers, such as EPRs, produced at temperatures at or above their sticking temye~a~ es in a gas phase polymerization, while at the same time reducing or elimin~ting the use of inert particulate matter. By colorable is meant that the resin is -capable of producing an end product which accepts and/or exhibits a color other than black.

Brief Description of the Drawings Figure 1 is a schematic diagram of a catalyst particle and a final resin product particle. ~igure 2 is a graphic depiction of the ratio of ethylene infrared signal to propylene infrared signal across a cross section of a final resin product particle as ~ cllcsed in F~mple 19 herein.

S--mms-ry of the Invention Acco~ gly, the present invention provides a resin particle having (A) an outer shell that is at least about 80% by weight of a non-sticky polymer having (a) 10 to 90 mole % ethylene and having at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms, and (b) a flow index of less than 20 decigrams/minute; and (B) an inner core of at least about 90% by weight of a sticky polymer;
said resin having at least about 1 percent by weight of said non-sticky polymer.
There is provided in another embo~iiment the above-mentioned non-sticky prepolymerized catalyst having a prepolymer portion and a catalyst portion wherein (1) the prepolymer portion has (a) 10 to 90 mole % ethylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms; and (b) a flow index of less than 20 decigrams/minute; and wherein (2) the ratio of the prepolymer portion to the catalyst portion is 25:1 to 1000:1.
Still another embo-liment provides a process for the production of a resin having (A) an outer shell of a non-sticky polymer and (B) an .

inner core of a sticky polymer, which process comprises contacting ethylene, at least one alpha olefin having 3 to 18 carbon atoms, and optionally at least one diene, in a gas phase flni-li7.~ bed in the presence of hydrogen, at a tempe~atule at or above the st;cking tempela~ule of the sticky polymer under polymeri7.~tion conditions with (I) the above-ment;oned non-sticky prepolymerized catalyst, (II) a co-catalyst, and (III) optionally a promoter, wherein the amount of non-sticky polymer is sllffiçi~nt to es.~nti7.lly ~ veut agglomeration of the fluidized bed and of the sticky polymer.
Still other embolliment,s provide three prepolymerizing processes for preparing the non-sticky prepolymerized catalyst particles. One is a prepolymerization process for producing the above-mentioned non-sticky prepolymerized catalyst comprises cont~cting a prepolymerization catalyst in a slurry of inert solvent with ethylene and at least one alpha olefin, and optionally a diene, such that (i) the tempela~-lre ofthe slurry is m~int~ined such that the prepolymer portion of the non-sticky prepolymerized catalyst is insoluble in the slurry;
(ii) the total feed rate of the ethylene and the alpha olefin is less than or equal to 500 grams of ethylene and alpha olefin per gram of catalyst per hour;
(iii) the ratio of ethylene to alpha olefin is m~int~ined at a constant ratio of less than or e~ual to 9:1;
(iv) the process is termin~te~ when the ratio of the prepolymer portion to catalyst portion of the non-sticky prepolymerized catalyst is 25:1 to 1000:1, by evaporating the solvent and unreacted ethylene and alpha olefin; and (v) optionally an inert part~ te~l material is added to the slurry i~nediately prior to the termin~1;ng step.
Another prepolymerization process for producing the above-mentioned non-sticky prepolymerized catalyst comprises con~cting a prepolymerization catalyst precursor in a slurry of liquid ethylene and propylene, and optionally a diene, with a co-catalyst such that D-17229 21S 9~88 (i) the rate of prepolymeri7.~ion is less than or equal to 500 grams ethylene and propylene per gram catalyst precursor per hour;
(ii) the slurry is m~int~inerl at a constant pressure of at least 300 psia;
(iii) the process is termin~t~l when the ratio of the prepolymer portion to catalyst portion of the non-sticky prepolymerized catalyst i~
25:1 to 1000:1 by purging the unreacted ethylene and alpha olefin; and (iv) optioIlally an inert particulate material is added to the slurry im~ tely prior to the terrnin~ting step.
A third prepolymeri7~tion process for producing the above-mentioned non-sticky prepolymerized catalyst comprises charging a prepolymeri7~tion catalyst, ethylene, and an alpha olefin, and optionally a diene, in a gas phase stirred reactor (i) at a temperature below the sti-~king temperature of the prepolymer portion of the non-sticky prepolymerized catalyst, (ii) such the total feed rate of the ethylene and the alpha olefin is less than or equal to 500 grams ethylene and alpha olefin per gram of catalyst per hour; and such t,hat (iii) the ratio of ethylene to alpha olefin is maintained at a constant ratio of less than or equal to 9:1;
(*) terminating the process by purging unreacted ethylene and alpha olefin; and (v) optionally an inert particulate material is added initially to the reactor.

Detailed Description of the Invention The resin composition of the present invention is produced by polymerizing ethylene, at least one alpha olefin, and a polymerization co-catalyst in the presence of a non-sticky prepolymerized catalyst particles in a fluidized bed operated in the gas phase at or above the sticking temperature of the final sticky resin polymer product being produced. Optionally the polymerization can include a diene, a chain transfer agent, and a promoter.

The non-sticky prepolymerized catalyst particles can be prepared using any of t,he following prepoly_eri7~tion processes: (1) a gas phase prepolymerization, (2) a slurry prepolymerization using an inert diluent, or (3) a slurry prepolymeri7~t;on using a mo~omer as a diluent. These prepolymçri7.~tion processes employ ethylene, at least one alpha olefin, and a prepolymeri7.~tion catalyst (which includes a catalyst precursor, co-catalyst, and opt;on~l promoter and/or other modifiers). Optionally these prepolymeri7.~t;0n processes can include a diene, a chain transfer agent, and an inert particulate material.

Prepolymerization Catalysts The prepolymerization catalysts employed in prepolymerization processes of this invention are composed of a catalyst precursor, co-catalyst, and optionally a promoter.
Catalyst precursor compounds that can be used in the prepolymerization processes of the present invention include transition metal compounds from Groups IIB-VIII of the Periodic Table of the Elements. Among the yrefel~ed transition metal compounds are compounds from Groups IVB-VIB. Catalyst precursors can include vanadium compounds, titanium compounds, chromium compounds, and metallocenes. These co,ll~oullds may be supported or unsupported. Each transition metal cu~oulld is generally employed along with co-catalyst and promoter which are associated with that particular catalyst precursor. r~efelled among these precursors are vanadium and titanium precursors.
Vanadium compounds which can be used to practice the prepolymerization processes of the present invention are vanadium salts or the reaction product of a vanadium salt and an electron donor.
Of course, ~2~ es of these compounds may also be used. Non-limiting, illustrative ~Y~mples of these compounds are as follows:
A. vanadyl trihalide, alkoxy halides and ~lkom~les such as VOCl3, VOC12(0R) wherein R is an alkyl having 1 to 12 carbon atoms, and VO(OCxHy)3 wherein ~ is 1 to 12 and y is x+3.

g B. vanadium tetrahalide and vanadillm alkoxy h~lirles such as VCl4 and VCl3(0R) wherein R is an alkyl having 1 to 12 carbon atoms.
C. vanadium and vanadyl acetylaceton~tes and chloracetyl acetonates such as V(AcAc)3 and VOCl2(AcAc) wherein (AcAc) is an acetyl acetonate.
D. vanadium trihalides and alkoxy halides, such as VCl3 and VC12(0R) wherein R is an alkyl having 1 to 12 carbon atoms.
The electron donor, if used with the vanadium compound, is an organic Lewis base in which the vanadium compound is soluble. It can be an alkyl ester of an aliphatic or aromatic carboxylic acid, an ~liph$ltic ketone, an aliphatic amine, an aliphatic alcohol, an alkyl or cycloalkyl ether, or ....~ures thereof, each electron donor having 2 to 20 carbon atoms. F~mrles include ethylacetate, butyl acetate, ethyl ether, dibutyl ether, methyl acetate and tetrahy~oÇuldn.
Modifiers can also optionally be used with these vanadium catalyst systems. A modifier can have the formula AlR(3 a)xa or BX3 or SiX4 wherein each R is independently an alkyl radical having 1 to 14 carbon atoms; each X is independently chlorine, bfo..~ e or iodine;
and _ is an integer from 0 to 2. Preferred modiSers include alkylaluminum mono and dichlorides, BCl3, and the trialkylaluminums. h',~mples include diethylal.. ;.ll.. chloride, triethylaluminum and boron trichloride. The molar ratio of modifier to vanadillm is in the range of about 1:1 to about 10:1.
Co-catalysts utilized with vanadium compounds consist essentially of an alkyl aluminum halide having the formula AlR(3 a)Xa~ wherein each R is independently alkyl having 1 to 14 carbon atoms, each X is independently chlorine, bro~ e or iodine, and a is 1 or 2, or a trialkylaluminum compound having the formula AlR3, wherein R is the same as above. Alkylalnminum halides falling within the above formula include alkylaluminum mono and dichlorides wherein each alkyl radical has 1 to 6 carbon atoms. h~mrles include diethylalllminllm chloride, ethylalllminllm dichloride, ethylal.. i~ m sesquichloride, dimethylaluminum chloride, methylalllminum dichloride, diisobutyl alllmin~lm chloride and isobutylalu~inum dichloride. ~mrles of trialkylalnminllm cv~ oullds include trihexylalllminllm, trimethylaluminum, triethylal.. ;.. l triisobutylaluminum and trioctylalu . .-; . . . - . . .
The optional promoter llt;li~er3 with vanadium catalyst precursors can be a chlorinated ester having at least 2 chlorine atoms or a perchlorinated ester. Suitable esters are Cl3CCOOC2H6 (ethyl trichloroacetate), Cl3CCOOCH3 (methyl trichloroacetate), CCl3CCl=CClCOOC4Hg (butyl perchlorocrotonate), and Cl2C=CClCCl2COOC4Hg (butyl perchloro-3-buteno~t~). The promoter can also be a saturated aliphatic hyJl oca~bon of formula RyCX(4 y wherein R is hydrogen or an unsubstituted or halogen-substituted alkyl radical having 1 to 6 carbons, each X is independently fluorine, chlorine, b~ o..lllle or iodine and y is an integer from 0 to 2. F~mples include methylene dichloride, 1,1,1-trichloroethane, chlorvfo~ , CFCl3, hexachloroethane, and F2ClCCCl3 (1,1-difluorotetrachloroethane).
The promoter can also be a saturated aliphatic halocarbon having the formula C3XaFbHC wherein each X is indepentlently chlorine, brv~ le or iodine, _ is an integer from 6 to 8, _ and c are integers from 0 to 2, and a+b+c equals 8. ~mples include hexachlorol,l ulJane~
heptachloropropane and octachlolo~lul ane. These saturated halocarbon promoters are mentioned in U.S. Patent No. 4,892,853. In addition, the promoter can also be an unsaturated aliphatic halocarbon such as perchloropropylene or any unsaturated halocarbon having a CX3 group attached to a C=C group wherein each X is indepen-lçntly chlorine, bromine or iodine or a haloalkyl substituted aromatic hydrocarbon wherein the haloalkyl substituent has at least 3 halogen atoms, such as trichlorotoluene or trichloro~ylene.
Supporting the catalyst precursor is preferred, and when the catalyst precursor is supported the ~le~elled support is silica. Other suitable supports are inorganic o~ides such as al~ ;..u..- phosphate, alllmin~, silica/alumina ~ es, silica modified with an organoalln.. ;... compound such as triethylall,.. ;.. ~.. , and silica modified with diethylzinc. ~mrles of polymeric supports are cross-linked poly~ylelle and poly~lol ylene. A typical support is a solid, particulate, porous material essentially inert to the prepolymerization.
It is used as a dry powder having an average particle size of about 10 to about 250 microns and preferably about 30 to about 100 microns; a surface area of at least about 200 square meters per gram and lJl efelably at least 250 square meters per gram; and a pore size of at least about 100 An~vl-ls and l.lere~ably at least about 200 Angstroms. Generally, the amount of support used is that which will provide about 0.1 to about 1.0 millimole of transition metal per gram of support. Impregnation of the above-mentioned catalyst precursor transition metal compound into a silica support is ~ccomplished by mi~ing the compound and silica gel in an ~l~lo~l;ate solvent in which the compound is soluble followed by solvent removal under reduced pressure. Spray-drying technology can also be used to generate well-shaped catalyst precursors having little or no silica or other inorganic solids content.
The preferred vanadium co~l,oullds are V(AcAc)3, VCl3, VOCl3, VCl4, and VO(OR)3 wherein R is a hy~oc~llon radical, ~lefe~ably a C1 to C1o aliphatic or aromatic hydrocarbon radical such as ethyl, phenyl, iso~lo~yl, butyl, propyl, n-butyl, i-butyl, t-butyl, hexyl, cyclohegyl, naphthyl, and so forth and electron donor complexes of these vanadium compounds. Such catalysts are described, for example, in U.S. Patent Nos. 4,508,842, 5,342,907 and 5,332,793.
Suitable titanium compounds for use in the present invention include catalyst precursors having the formula MgaTi(OR)bXC(ED)d wherein R is indepçn-lçntly an :~liph~tic or aromatic LyLvcarbon radical having 1 to 14 carbon atoms or COR' wherein R' is an aliphatic or aromatic hydrocarbon radical having 1 to 14 carbon atoms; X is independently chlorine, bl ~ ~ille or iodine; ED is an electon donor; _ is 0.5 to 56; k is an integer from 0 to 2; _ is 2 to 116; and d is 2 to 85.
Titanium compounds which are useful in preparing these precursors D-17229 21~ 9388 ,, have the formula Ti(OR)bXe wherein R, X and ~ are as defined above, e is an integer from 1 to 4, and b+e is 3 or 4. ~mples include TiCl3, TiCl4, Ti(OC2H5)2Br2, Ti(OC6Hs)Cl3 and Ti(OCOCH3)C13. These titanium-based catalysts and their method of ~le~al alion are disclosed more fully in U.S. Patent No. 4,302,565.
Electron donors, modifiers, SU1J~)O1 ls, co-catalysts and promoters can be used with titanium-based catalysts in the same manner as for vanadium, and are the same as those described above for vanadium.
The particular combination of electron donor, modifier, co-catalyst, and promoter are chosen from those known to those skilled in the art of polymerization to be most efficacious for the particular catalyst precursor.
Chromium compounds which are suitable for use in the present invention include chromyl chloride (CrO2Cl2), chl Oll~i.. 2-ethylh~no~te, cLon~iu~ acetylacetonate (Cr(AcAc)3), and the like which are disclosed in, for ~mple, U.S. Patent Nos. 3,242,099 and 3,231,550.
Metallocenes which can be employed in the invention include catalyst compositions comprising at least one metallocene catalyst, at least one co-catalyst, and particulate filler material having an average particle size of less than about 10 micrometers. The particulate filler material is unreactive with both the metallocene catalyst and the co-catalyst.
One useful class of metallocene catalysts are organometallic compounds cont~inin~ at least one cyclopentadienyl group bonded to a Group IIIB to VIII metal, such as mono, di-, and tricyclopentadienyls and their derivatives of these transition metals.
A preferred metallocene catalyst cont~inin~ at least one cyclopentadienyl group bonded to a Group IIIB to VIII metal has the formula:

(CsRn)yR'z(C5Rm)MY(x-y-1) (1) _ wherein M is a metal from Groups IIIB to VIII of the Periodic Table;
(CsRn) and (CsRm) are independently cyclopentadienyl or substituted cyclopentadienyl groups bonded to M; each R is independently hydrogen or a hydrocarbyl radical such as alkyl, alkenyl, aryl, alkylaryl, or arylalkyl radical cont~ining from 1 to 20 carbon atoms, or two carbon atoms are joined together to form a C4-C6 ring; each R' is~
C l-C4 substituted or unsubstituted alkylene radical, a dialkyl or diaryl germanillm or silicon, or an alkyl or aryl phosphine or amine radical bridging two (CsRn) and (CsRm) rings; each Y is independently a hydrocarbyl radical such as aryl, alkyl, alkenyl, alkylaryl, or arylalkyl radical having from 1-20 carbon atoms, hydrocarboxy radical having from 1 to 20 carbon atoms, or halogen; n and m are each 0, 1, 2, 3, or 4;
zisOorl,andzisOwhenyisO;yisO,l,or2;xisl,2,3,or4 depending upon the valence state of M; and x-y 2 1.
Illustrative, but non-limiting ç~mples~ of metallocene catalysts of Formula 1 are dialkyl metallocenes such as bis(cyclopentadienyl)titanium dimethyl, bis(cyclopentadienyl)zilconiu~ dimethyl; the mono alkyl metallocenes such as bis(cyclopentadienyl)~,lcol~ium methyl chloride; the trialkyl metallocenes such as cyclopentadienyl titanium trimethyl; silicon, phosphine, amine or carbon bridged metallocene compounds such as isu~,u~ylidene(cyclopentadienyl)(fluorenyl)zirconium dichloride, diphenylmethylene(cyclopentadienyl)(fluorenyl) zirconium dichloride, rac-ethylenebis(indenyl);6ilcw~ dichloride.
A more l~refelled metallocene catalyst cont~ining at least one cyclopentadienyl group bonded to a Group IIIB to VIII metal is a bridged metallocene having the formula:

Rm~

Q~,~

H4 n (2 wherein:
Q is a bridging linkage selected from R"2C \, R"2Si . R"2Ge \, and -C2R"4-wherein each R" moiety is independently H or an alkyl group, or two R"
moieties are joined to form a ring structure. Plefel ably, when an R"
moiety is an alkyl group, it cont~in~ 3 to 8 carbon atoms, and when two R" moieties are joined to form a ring structure with the atom or atoms to which they are respectively attached, a 5 or 6-membered ring is formed. The subscripts m and n are each 0, 1, 2, 3, or 4, and the sum of ~a and g is l,.efe~ably 2 to 6. The metal M is a Ti, Zr, or Hf atom, preferably Zr. Each ~ is independ ently H, an alkyl group, or a halogen atom.
In bridged metallocenes, the cyclic 7~-bonded moieties may bear one or more substituents R. Each R moiety is indepçnrlently an alkyl, cycloalkyl, alkenyl, cycloalkenyl, phenyl, alkyl-substituted phenyl, or a phenyl-substituted alkyl group, or two adjacent R groups on a given ring are joined to form a second ring. Preferably, each R moiety is independently an alkyl or cycloalkyl group of 3 to 8 carbon atoms, an alkenyl group of 2 to 8 carbon atoms, a cycloalkenyl group of 5 to 8 - 2t ~ 9:~88 carbon atoms, phenyl, an alkyl-substituted phenyl group in which the alkyl group cont~in~ 3 to 8 carbon atoms, a phenyl-substituted alkyl group in which the alkyl portion cont~in~ 1 to 4 carbon atoms, or two adjacent R groups on a given ring are joined and together with the carbon atoms to which they are respectively attached form a saturated or lln~aturated 4, 5, or 6-membered ring.
Illustrative but non-limiting ç~mples of bridged metallocenes of Formula (2) that may be used as the metallocene catalyst are rac-ethylenebis(indenyl)zirconium dichloride and rac-ethylenebis(4,5,6,7-H-tetrahydroindenyl);Lircollium dichloride.
Another class of useful metallocene catalysts are constrained geometry metallocenes as described in PCT Publication No. W0 93/08221.
Preferred constrained geometry metallocenes have the formula:

Z y..
(3) Cp' M

(X)n' wherein M' is a metal of Groups IIIB-VIII or the Lanthanide series of the Periodic Table; Cp* is a cyclopentadienyl or substituted cyclopentadienyl group bound in an h6 bonding mode to M'; Z is a moiety comprising boron or a member of Group IVa of the Periodic Table, and optionally sulfur or oxygen, said moiety having up to 20 non-hydrogen atoms, and optionally Cp* and Z together form a fused ring system; each X is independently an anionic ligand group or neutral Lewis base ligand group having up to 30 non-hydrogen atoms;
.

D-17229 21~9388 n' is 1, 2, 3, or 4 and is 2 less than the valence of M'; and Y" is an anionic or nonionic ligand group bonded to Z and M' COLU1 l;sing ~itrogen, phosphorus, oxygen, or sulfur and having up to 20 non-hydrogen atoms, optionally ~' and Z together form a fused ring system.
Illu~ ative, but non-limiting e~P.mples, of l.refelred constrained geometry metallocenes include (tert-butyl~mido)(tetramethyl-h5-cyclopentadienyl)-1,2-et~ne~liylzirconillm dichloride and (methylamido)(tetramethyl-h~-cyclopentadienyl)-1,2-ethanediylli~ ium dichloride.
The co-catalyst is capable of activating the metallocene catalyst, and may be one of the following: (a) branched or cyclic oligomeric poly(hydrocarbylaluminum o~ide)s which contain repeating units of the general formula -(Al(R"')O)-, where R"' is hydrogen, an alkyl radical cont~ining from 1 to about 12 carbon atoms, or an aryl radical such as a substituted or unsubstituted phenyl or naphthyl group; (b) ionic salts of the general formula [A+][BR*4-], where A+ is a cationic Lewis or Bronsted acid capable of abstracting an alkyl, halogen, or hydrogen from the bridged metallocene catalyst, B is boron, and R* is a substituted aromatic hydrocarbon, preferably a perfluorophenyl radical; and (c) boron alkyls of the general formula BR*3, where R* is as defined above.
Preferably, the co-catalyst employed with the metallocene catalyst is a branched or cyclic oligomeric poly(hydrocarbylaluminum oxide). More preferably, the co-catalyst is an all1mino~ne such as methylalllmino~ne (MAO) or modified methylalllminn~ne (MMAO).
Alllminc-~nes are well known in the art and comprise oligomeric linear alkyl aluminoxanes represented by the fo~nula:

R"' Al-O AIR"'2 I

R"' s and oligomeric cyclic alkyl alllmino~nes of the formula:

D-17229 2159~88 -Al-O-R"' ~ p wherein ~ is 1 to 40, preferably 10 to 20; p is 3 to 40, preferably 3 to 20;
and R"' is an alkyl group c~nt~inin~ 1 to 12 carbon atoms, l ~efe~ably methyl or an aryl radical such as a substituted or unsubstituted phenyl or naphthyl radical.
The metallocene catalyst composition may optionally co~t~in one or more second catalysts. These second catalysts include for example any Ziegler-Natta catalysts cont~ining a metal from Groups IV(B), V(B), or VI(B) of the Periodic Table. Suitable activators for Ziegler-Natta catalysts are well known in the art and may also be included in the catalyst composition.
The filler material is selected from organic and inorganic compounds that are inert to the co-catalyst and the metallocene catalyst. F~mples include alumina, titanium dio~ide, poly~ylelle, rubber modified poly~ylelle, polyethylene, poly~ ylene, m~neSium chloride, and silicon dioxide such as fumed silica. Such fillers may be used individually or in cnmhin~tions. The filler material has an average particle size of less than about 10 micrometers, preferably less than about 1 micrometer, and most preferably has an average particle size in the range of about 0.001 to about 0.1 micrometers.

Alpha Olefins All three of these prepolymerization processes and the polymerization process for m~king the final resin composition employ ethylene and at least one alpha olefin have 3 to 18 carbon atoms.
These alpha olefins can be linear or branched. Illu~Lla~ive, but non-limiting ç~mples of these alpha olefins, include propylene, butene-1, isobutene, pentene-1, octene, decene-1, dodecene-1, tetradecene-1, hexadecene-1, octadecene-1, and mixtures thereof. Preferred alpha D-17229 21~9388 olefins have 3 to 12 carbon atoms. These alpha olefins are preferably selected from the group consisting of propylene, butene-1, pentene-1, hexene-1, 4-methyl-1-pentene and octene-1. Most l,lerelably the alpha olefins have 3 to 6 carbon atoms and of these propylene, butene-1 and he~ene-1 are especially preferred.

Dienes Sticky polymers produced in the invention can optionally contain non-conjugated dienes. Both the non-sticky prepolymerized catalyst and the final resin composition can contain one or more of such dienes.
It is 1,l efe-l ed that dienes not be inco.l,o.ated into the non-sticky prepolymerized catalyst since they have a tendency to exacerbate ~gglomeration. Preferably all or most of the diene is added during the polymerization of the final resin to produce a sticky EPDM. These non-conjugated diene monomers used to produce EPDMs may be straight chain, branched chain or cyclic hydrocarbon dienes having from about 5 to about 15 carbon atoms. F~mples of suitable non-conjugated dienes are straight chain acyclic dienes such as 1,4-hexadiene, 1,5-hç~tliene, 1,7-octadiene, 1,9-decadiene and 1,6-octadiene. Illu~LLa~ive branched chain acyclic dienes include such as 5-methyl-1,4-hexadiene, 3,7-dimethyl-1,6-octadiene, 3,7-~limethyl-1,7-octadiene and mixed isomers of dihydromyricene and dihydrocinene.
Single ring alicyclic dienes can include, for e~mple, 1,3-cyclopentadiene, 1,4-cycloh~ ie~e, 1,5-cycloctadiene and 1,5-cyclododecadiene. Illustrative multi-ring alicyclic fused and bridged ring dienes such as tetrahydroindene, methyl tetrahydroindene, dicyclopentadiene, bicyclo(2,2,1)-hepta-2,5-diene, alkenyl, alkylidene, cycloalkenyl and cycloalkylidene norbornenes such as 5-methylene-2-norbornene, 5-propenyl-2-norbornene, 5-iso-1,lvl)ylidene-2-norbornene, 5-(4-cyclopentenyl)-2-norbornene, 5-cyclohexylidene-2-norbornene, 5-vinyl-2-norbornene and norborn~tliene can be employed in the process of the present inVçntion Dienes which are especially ~lef~,led include 1,4-hexadiene, and 5-ethylidene-2-norbornene. While a portion of one D-17229 21~ 9388 or more of these dienes can be incorporated into the non-sticky prepolymerized catalyst, ~are~dbly the diene, or bulk of the amount of diene, is incorporated into the resin particle composition in the gas phase polymerization described below. The amount of diene in the prepolymerized catalyst ranges from about 0 to about 6 percent by weight based upon the weight of the prepolymer portion of the catalys~,~
and amount of diene in final resin composition is about 0 to about 15 percent by weight based upon the weight of the sticky polymer in the resin.

Chain Transfer AFents A chain transfer agent, such as hydrogen or a metal alkyl (e.g., diethyl zinc) are normally used in polymerizations to control molecular weight. They can be employed during the prepolymerization processes and the polymerization process of this invention. Such transfer agents are well known in the art and are used in the normal manner in the polymerization of the resin product herein. When a chain transfer agent is ~sed in a prepolymerization process, it is 1~ afe, ably added after prepolymerization is initiated and gradually increased in - concentration to produce the desired non-sticky prepolymerized catalyst.

Inert Particulate Material ~ ptionally, inert particulate material, also referred to as a fluidization aid, can be employed in the prepolymerization process of the present invention. When employed in the gas phase prepolymerization process, the inert particulate material 1.l afel ably is added in the initial charge to aid in bed f~uidization. In slurry prepolymerization processes, it is ~lafe~led to add the fluidization aid immediately prior to or simultaneously with evaporating the solvent and/or e-x-cess monomers. Suitable inert particulate materials are disclosed, for ex~mrle~ in U.S. Patent Nos. 4,994,634 and 5,304,688.
They include carbon black, silica, talc, and clay. r~efe~lad materials are silica, talc and clay with silica being most preferred. When employed, the inert particulate material is about 0.05 to about 50 weight percent of the non-sticky prepolymerized catalyst particle, preferably about 0.05 to about 20 weight percent In the final resin composition, when employed, the inert particulate material ranges from about 0.05 to about 30 wt~o based upon the weight of the final resin product, and preferably is about 0.3 to about 10 wt%. Most preferably, inert particulate material is not employed in the present mvention.

Methods for Prepar;nF the Non-sticky Prepolymerized Catalyst (Prepolymerization Processes) The non-sticky prepolymerized catalyst of the present invention must be prepared in such a way that it is non-sticky itself, provides agglomeration protection for the sticky polymer in the resin product, and doesn't detract from the properties of the final resin product. This is ~ccompli.~hed by prepolymerizing a catalyst precursor under controlled conditions to produce a prepolymerized catalyst having particles of a specific composition. The particles can be prepared in either (1) a gas phase prepolymerization, (2) a slurry prepolymerization using an inert hydrocarbon as a diluent, or (3) a slurry prepolymerization using liquified monomers as a diluent by controlling the rate of prepolymerization, controlling the monomer feed composition, and controlling the quantity of the prepolymer portion produced.
The prepolymerization is conducted in a stirred gas phase reactor, a fluidized bed gas phase reactor or a stirred slurry reactor, as al ~l ol~l;ate. The reactor is operated in batch mode, or, if continuous, in a uniform residence time mode.

Rate of Prepolymerization In order to produce non-sticky prepolymerized catalyst particles with most of the prepolymer portion on the outside of the particle, and - 21~9~88 most of the catalyst portion within the core of the particle, it is desirable that the prepolymerization be conducted under conditions that limit the rate of the prepolymerization. Since the rate of a polymerization is well-known to be a function of co-catalyst concentration, one way to ~ccompli~h this is to limit the amount of co-catalyst available. By feeding only the ~ mlll~ level of co-catalyst needed to sustain a reaction, the co-catalyst feed rate will be rate-limiting, and the reaction rate will be controlled by this feed rate.
While this level will vary from one catalyst system to another, it will normally be from about V100 to about 1/2 of the normal amount of co-catalyst needed for full catalyst activity, or a mole ratio of co-catalyst to transition metal in the catalyst precursor of about 0.1:1 to about 10:1.
Since the rate of a polymeri7.~tion is also well-known to depend on the ethylene partial pressure, another way to control the rate of polymerization is to control the ethylene partial pressure at a low value. Ethylene partial pressures in the range of 2 to 200 psia can be used to limit the production rate to the desired value.
Still another way to control the rate of a polymerization is to limit the flow of monomers to the reactor to the desired production rate. The ethylene partial pressure will then change as catalyst activity changes either from one catalyst system to another or in the course of polymerization, but the production rate will be fixed by the feed rate.
The prepolymerization rate should be limited to about 1 to about 500 grams of total monomers (ethylene plus alpha olefin) per gram of catalyst per hour, preferably about 5 to about 300 grams, most preferably about 10 to about 200 grams.

Control of Composition Another important aspect of the prepolymeri7.~tion process involves controlling the composition of the prepolymer portion of the prepolymerized catalyst being produced. This composition is controlled D-17229 215 938 ~
_ by the balance between the monomers fed to the reactor and the monomers consumed in the prepolymerization. As long as the amount of monomers consumed in the prepolymerization is much greater than the amount of monoIners that remain unpolymeri7.etl at the end of the prepolymerization, mass b~lAnse requires that the average composition of-the prepolymer portion closely reflect the average composition of the monomer feed during the prepolymeri7.~tion Thus controlling the feed composition provides a simple method for controlling the average composition of the prepolymer.
In addition, controlling the feed composition provides a method to introduce a composition gradient into the prepolymer portion of the prepolymerized catalyst. It is well-known in the art that olefin polymerization catalysts react more rapidly with ethylene than higher alpha-olefins. Thus, when a constant feed composition is supplied to a catalyst, the first material polymerized will have a composition that has a higher ethylene content than the composition being fed.
However, because the average composition over time must reflect the composition of the feed, as indicated above, the composition of the prepolymer portion being produced during the prepolymerization gradually becomes more enriched in alpha olefin as the prepolymerization proceeds. This generates a composition gradient in alpha olefin content.
Prepolymerization in liquified mnnomers requires a L~l ell~ -means of control of composition. The composition of the monomers in the slurry medium is determined by the respective vapor pressures of the monomers, according to Henry's Law. Thus, in any volume of liquified ethylene/propylene mi~ e, the amount of ethylene dissolved is a function only of the pressure of ethylene over the slurry. If this pressure is held constant, the conce,ltlal~ion of ethylene in the slurry will be constant, and the composition of the prepolymer portion of the prepolymerized catalyst produced will be constant. If it is desired to change that composition during the prepolymerization, it is neces~ry only to adjust the ethylene pressure that is m~int~inerl over the slurry.

~ D-17229 2159388 Gas Phase Prel~olymerization Process An initial bed is normally charged to a gas phase reactor for dispersing a catalyst charge and maint~ining agitation. When used, the bed can be a granular polymer, an inert particulate material such as sodium chloride granules, an inorganic oxide, such as Davison 9~5 silica, or an inert particulate material as described above, such as HiSil 233 silica available from PPG. Of course any bed material added to the reactor must be thoroughly dry and free from water or moisture. This is Accompli.qhed using well known techniques such as purging with hot nitrogen gas or calcin~t;on A catalyst precursor is charged to the reactor cor t~inin~ the bed.
A co-catalyst and promoter are preferably added separately neat or as a solution in an inert solvent, such as isopentane. The prepolymerization catalyst is then contacted with ethylene and at least one alpha olefin or at least one alpha olefin and diene, as previously described. The ethylene, alpha olefin, and optional diene can each be added separately to the reactor, or as a ~lu.e. They may be added continuously or intermittently both as individual components or as the ll~ix l~`~. .
Whether added as a ~ e or separately, the ethylene and alpha olefin are fed to the reactor in a controlled ratio. The ratio should be essentially the same as the desired ratio of ethylene to alpha olefin in the non-sticky prepolymerized catalyst to be produced. For ex~mple, if it is desired that the prepolymerized catalyst cont~in 20 mole ~
propylene, then the feed should contain 20 mole % propylene. The ethylene and alpha olefin should be fed such that the ethylene is less than about 90 mole % of the total ~mount of ethylene and alpha olefin in the feed, ~,efe.ably less than about 85 mole % of the total amount of ethylene and alpha olefin in the feed.
Optionally, a chain transfer agent, can also be employed.
Hydrogen is the 1~l efel, ed chain transfer agent. When used, it can be added at any time during the prepolymerization. However, it is D-17229 215938~

preferable that no chain transfer agent be present at the initiation of prepolymerization. Rather, it is added after prepolymeri7~ion has begun and gradually increased until the conçentration in the reactor is such as to instantaneously produce prepolymer having a flow index up to 20 decigr~ms per minute, preferably up to 5 decigrams/minute, most preferably 2 or less.
Throughout the prepolymerization process, the temperature is maintained below the stiçking temperature of the prepolymer portion of the non-sticky prepolymerized catalyst. This tempe~dl,ule will depend on the composition of the non-sticky prepolymerized catalyst, and the degree of crystallinity of the prepolymer portion. Typically, the temperature of the prepolymerization is less than about 80C, and preferably less than about 50C.
The prepolymerization is conducted under conditions that limit the rate of the prepolymerization. This rate can be controlled by using individually or in comhin~tion any of the methods set forth above.
E'lefelable are limiting the amount of co-catalyst or limiting the monomer feed rate.
When the weight ratio of the resulting prepolymer portion to catalyst portion is about 25:1 to about 1000:1, l.lefelably 50:1 to about 500:1, and most preferably about 100:1 to about 300:1, the prepolymerization is terminated by purging the residual monomçrs from the reactor, thereby leaving the non-sticky prepolymerized catalyst as a free-flowing solid powder to be removed from the reactor.

Slurry Process in an Inert Hydrocarbon Diluent.
An inert hydrocarbon is charged to a stirred slurry reactor as the slurry medium. The inert hydrocarbon can be any saturated, linear or branched, h~dlocdll)on having 2 to 8 carbon atoms.
Illustrat*e but non-limiting ex~mples include ethane, propane, n-butane, isobutane, n-pentane, isopentane, n-hexane, isohç~ne, n-heptane, n-octane or isooctane. P~efelably the diluent has 2 to 4 carbon atoms, and most preferably the diluent is ethane, propane, isobutane, or ~Lxlu~es thereof. The amount of hydroc~l,oll used for the slurry is such that the amount of mnnom~rs dissolved in the hydrocarbon is small comp~red to the amount of non-sticky prepolymerized catalyst produced. However, the amount of-diluent must be sufficient to allow çffi~içnt mi~ing of the slurry and to provide sufficient heat transfer capability to remove the heat of prepolymerization. Thus, the amount of diluent employed is about 0.001 to about 1 liters per gr~m of non-sticky prepolymerized catalyst to be produced, preferably about 0.002 to about 0.1 liters per gram of non-sticky prepolymerized catalyst, and most preferably about 0.003 to 0.01 liters per gram of non-sticky prepolymerized catalyst.
A catalyst precursor is charged to the reactor cont~ining the slurry medinm. A co-catalyst and promoter are preferably added separately neat or as a solution in an inert solvent, such as isopentane.
The prepolymerization catalyst is then cont~cted with ethylene and at least one alpha olefin or at least one alpha olefin and diene as previously described. The ethylene, alpha olefin, and optional diene can each be added separately to the reactor, or as a mi~lu e. They may be added coll~hluously or inter_ittently both as individual components or as the ~i~ e.
Whether added as a ~ e or separately, the ethylene and alpha olefin are fed to the reactor in a controlled ratio. The ratio should be essentially the same as the desired ratio of ethylene to alpha olefin in the non-sticky prepolymerized catalyst particle to be produced. For e~mple, if it is desired that the prepolymerized catalyst particle contain 20 mole % propylene, then the feed should cont~in 20 mole %
propylene. The ethylene and alpha olefin should be fed such that the ethylene is never less than about 90 mole % of the total amount of ethylene and alpha olefin in the feed, preferably less than about 85 mole % of the total amount of ethylene and alpha olefin in the feed.
Optionally, a chain transfer agent can also be employed.
Hydrogen is the preferred chain transfer agent. When used, it can be added at any time during the prepolymerization. How~ver, it is -preferable that no chain transfer agent be present at the initiation of prepolymerization. Rather, it is added after prepolymeri7~ion has begun and gradually increased until the concentration in the reactor headspace is such as to inst~nt~n~ously produce polymer having a flow index up to 20 decigrams per minntR, preferably up to 5 decigrams/minute, most l~rerel dbly 2 or less.
Throughout the prepolymerization process, the tempela~Lue is maintained low enough such that the non-sticky prepolymerized catalyst is not appreciably soluble in the inert saturated hydrocarbon solvent. This temperature will depend on the composition of the prepolymer portion of the non-sticky prepolymerized catalyst, the degree of crystallinity of the prepolymer portion being made, and type of solvent employed for the slurry. Typically, the tempelatule of the prepolymerization is less than about 80C, and preferably less than about 50C.
The prepolymerization is conducted under conditions that limit the rate of the prepolymerization. This rate can be controlled by using individually or in combination any of the three methods set forth previously. r~efe.dble are limiting the amount of co-catalyst and optional promoter or limiting the ethylene and alpha olefin feed rate.
When the weight ratio of the resulting prepolymer portion to catalyst portion is about 26:1 to about 1000:1, ~fefelably 60:1 to about 600:1, and most preferably about 100:1 to about 300:1, the prepolymerization is terminated by evaporating the hydrocarbon and purging the residual monomers from the reactor, thereby leaving the non-sticky prepolymerized catalyst as a free-flowing solid powder to be removed from the reactor.

Slurry Process ;n T~i~uified Mon- mers Liquified propylene is charged to a stirred slurry reactor as the slurry medium. A catalyst precursor is charged to the reactor cont~ining the slurry medium.

D-17229 21~ 9 3 8 ~

Ethylene is charged to the reactor to a desired pressure such that the ethylene/propylene composition of the slurry produces prepolymerized catalyst of at least 10 mole percent propylene, preferably 16 mole percent. Of course, the pressure required is depçn~çnt upon the tempe,a~ule ofthe prepolymerization, the composition of the non-sticky prepolymerized catalyst desired, and the specific catalyst employed. In general, howt~vel, the pressure ranges from about 150 psia to about 1000 psia, 1,l efe, ably about 300 psia to about 750 psia, and most p,efe,dbly about 400 psia to about 500 psia, and is readily calculable by those skilled in the art. The slurry of liquified ethylene/propylene and catalyst ~ ecursor is then contacted with co-catalyst and optional promoter. Ethylene is fed on tiem~nd to the reactor during the prepolymerization in order to m~int~in the desired partial pressure of ethylene in the reactor.
Optionally, a chain transfer agent, can also be employed.
Hydrogen is the preferred chain transfer agent. When used, it can be added at any time during the prepolymerization. How~v~" it is preferable that no chain transfer agent be present at the initiation of prepolymeri7.s~tion. l?~t~lçr, it is added after polymeri7.~tioIl has begun and gradually increased until the concentration in the reactor is such as to inst~n~n~ously produce polymer having a flow index up to 20 decigrams per minute, preferably up to 5 decigrams/minute, most preferably 2 or less.
Throughout the prepolymerization process, the tempe,~tule is maintained low enough such that the non-sticky prepolymerized catalyst particle is not appreciably soluble in the slurry medium. This temperature will depend on the composition of the non-sticky prepolymerized catalyst particle and the degree of cryst~llinity of the particle being made. Typically, the tempe,dlule of the prepolymerization is less than about 80C, and l~,efe~ably less than about 50C.
The prepolymerization is cr ntlll~te-l under conditions that limit the rate of the prepolymerization. The rate of the prepolymeri7sltion can only be limited in this process by the method of limiting the supply of cocatalyst and optional promoter.
When the weight ratio of the resulting prepolymer portion to catalyst portion is about 25:1 to about 1000:1, preferably 50:1 to about 500:1, and most l,.eîelably about 100:1 to about 300:1, the prepolymerization is termin~te-l by evaporating and purging the residual monomers from the reactor, thereby leaving the non-sticky prepolymerized catalyst as a free-flowing solid powder to be removed from the reactor.

Non-sticky Prepolymerized Catalyst Composition The non-sticky prepolymerized catalyst of the present invention is comprised of particles that have a prepolymer portion and a catalyst portion. The prepolymer portion has up to 90 mole % ethylene, preferably about 45 to about 85 mole % ethylene, and most ~refel ably about 60 to about 85 mole % ethylene. Also the prepolymer portion contains at least 10 mole % of one or more alpha olefins as described previously. Preferably, the prepolymer portion contains about 15 to 55 mole % alpha olefins, and most ~refel ably about 15 to 40 mole %. The flow index of the polymer portion is less than 20 decigrams/minute, preferably less than about 5 decigrams/minute, and most l~l efel ably is less than 2 decigrams/minute. The ratio of the polymer portion to the catalyst portion is 25:1 to 1000:1, preferably 50:1 to 500:1, and most preferably is 100:1 to 300:1.
Further, the non-sticky prepolymerized catalyst of this invention has particles having a shell and a core such that the prepolymer portion is substantially contained in the shell of the non-sticky prepolymerized catalyst particle, and the catalyst portion is subst~nl;iqlly con~in~-l in the core of the non-sticky prepolymerized catalyst particle. The shell of the non-sticky prepolymerized catalyst particle should cont~in at least about 75 percent by weight of the prepolymer portion, and the core of the non-sticky prepolymerized catalyst particle should cont~in at least about 8~ percent by weight of the catalyst portion of the particle.
In general prepolymerized catalysts are known in the art.
While they have been used, for ~mple, to protect against the hot spots that can sometimes occur in the gas phase polymerization of ethylene such catalysts are not known to provide adequate protection under more severe conditions, such as polymerizations above the sticking tempelatu~e of the polymer being made.
In the present invention it has been found that in order for the prepolymer portion of a prepolymerized catalyst to provide agglomeration protection to a sticky polymer above its sticking temperature, it must have three properties: (1) it must itself be non-sticky; (2) it must be elastic, to allow for the growth and e~ qn~ion of the sticky resin that it is to cont~in; and (3) its residues in the final resin product must not detract from the desired properties of the final sticky polymer product. Acco~ gly, the composition of the prepolymerized catalyst includes not just the rhemir~l composition, but also the molecular weight, the viscosity, the crystalline content, the form or morphology and other characteristics of the complete sticky polymer resin composition, and it is this specific comhin~tion of features that is unique and not one alone. A granular prepolymerized catalyst particle that has these characteristics will allow the production of a sticky polymer in a gas phase reaction at or above the sticking temperature of the sticky polymer.
The st;ckiness of a particle is known to be related to its viscosity. Crystallinity and molecular weight influence this viscosity.
Therefore, in general, polymers of high molecular weight are more viscous and, hence, less sticky than polymers of low molecular weight.
In addition, in general, polymers of high crystallinity are more viscous and less sticky than polymers of low crystallinity. Acco~ gly, in the present invention, the prepolymer portion of the non-sticky prepoly neri~ed catalyst should have a very high molecular weight, or should be highly cryst~lline, i.e. have a low content of alpha olefin, or D-17229 21S 9~88 both. However, highly crystalline polymers are not highly elastic, and prepolymers having a low co~tent of alpha olefin do not provide good agglomeration protection. Further, cryst~llinity is 1m~le~irable in a substantially amorphous final resin product.
- Theferole, the prepolymerized catalyst composition of the present invention has a prepolymer portion of low crystallinity, or high alpha olefin content, in comhin~t;on with high molecular weight.
This provides the required elasticity as well as the non-sticky characteristic; and the low cryst~11inity ensures that the non-sticky polymer will not detract from the desirable ~ro~ lies of the final resin product.
If necessary, the non-sticky characteristic of the prepolymerized catalyst can be further Pnh~ncetl in several optional ways. Since the viscosity of the prepolymer portion near the surface of the prepolymerized catalyst particle has the greatest effect on the stickiness of that particle, a gradient in composition in the prepolymer portion can be used. By gradient is meant that the prepolymer portion near the outer surface of the prepolymerized catalyst particle is of higher viscosity than the average viscosity of the prepolymer portion, and the material away from the surface of the particle is of lower viscosity than the average. Such a gradient r~n involve, individually or in comhin~tion, a gradient in the molecular weight of the prepolymer portion or a gradient in the alpha-olefin content of the prepolymer portion. Alternat*ely, the viscosity r~n be further enhanced by ~linE an inert particulate material, as described above, to the surface of the prepolymerized catalyst.
Another aspect of the prepolymerized catalyst particle composition is the ratio of the prepolymer portion to the catalyst portion. If this ratio is small, less than about 25 grams per gram, then the prepolymer portion will not be large enough to ~1 ~vellt agglomeration when the final resin product is fully grown. On the other hand, if the ratio is too high, greater than about 1000 gr~ms prepolymer portion per gram catalyst portion, then the resi~

prepolymer portion in the final product will be excessively high and will interfere with resin properties. The l .efel~ed ratio will be different for different catalyst systems, but will normally be in the range of about 50 to about 500 and usually in the range of about 100 to about 300 grams prepolymer portion per gr m catalyst portion.
Still another aspect of the prepolymerized catalyst particle composition is the morphology of the particle. For the prepolymer portion to cont~in the sticky polyrner produced in the product polymerization, the prepolymer portion must be subst~nti~lly located on the surface of the prepolymerized catalyst particle, and the catalyst portion, should be in the core of the particle. Ideally, the non-sticky prepolymerized catalyst particle should consist of a shell which is entirely prepolymer portion and a core which is entirely catalyst portion. Of course, it will be obvious to those skilled in the art that this is an ideal that can only be appro~rh~-l in the limit. The exact morphology will be closer or further from this ideal depending on the original morphology of the catalyst precursor before prepolymerization and on how the prepolymerization is conducted. In any prepolymerization process, some of the prepolymer portion will be produced in the core of the catalyst particle, and some of the catalyst portion will be dispersed throughout the shell of the prepolymer portion. Despite such variation, the prepolymerized catalyst particle should have a granular form in which the prepolymer portion is more highly concentrated in the shell of the particle, and in which the catalyst portion is more highly cor cen~rated in the core of the particle.
A ~rhem~tic diagram of a cross-section of the resin particle is depicted in Figure 1.

Polymerization of Resin Particle Composition The polymerization is con-lllcted in the gas phase, preferably in a fluidized bed made up of particulate resin. The bed is usually made up of the same granular resin that is to be produced in the reactor.
The gas phase reactor can be the fllli~li7.e~ bed reactor described in U.S.

Patent Nos. 4,482,687; 4,994,534; 5,304,588 or another collv~.,tional gas phase reactor. The fluidized bed reactor can be operated at a temperature in the range of about 0C to about 100C and is ylafelably operated in the range of about 20C to about 70C. A superficial velocity of about 1 to about 4.5 feet per secon~l and ~lafeYdbly about 1.5 to about 3.5 feet per æecond can also be used in the fluidized bed. The total reactor pressure can be in the range of about 150 to about 600 psia and is preferably in the range of about 250 to about 500 psia. The ethylene partial pressure can be in the range of about 25 psi to about 350 psi and is preferably in the range of about 80 psi to about 250 psi.
The gaseous feed streams of ethylene, alpha-olefin, and hydrogen are preferably fed to the reactor recycle line while diene, when employed, is preferably fed directly to the fluidized bed reactor to ~nh?nce miYin~
and dispersion. Feeding liquid streams into the reactor recycle line can cause a rapid buildup of a fouling layer resulting in very poor reactor operation. Polymer composition can be varied by ch~nging the alpha-olefin/ethylene molar ratio in the gas phase and the diene concçnt~ation in the fluidized bed. The product is continuously chs~rged from the reactor as the bed level builds up with polymerization. The production rate is controlled by adjusting the catalyst feed rate and the ethylene partial pressure.
The polymerization is conducted using procedures well-est~hli~hed in the art. The catalyst employed is the non-sticky prepolymerized catalyst of the invention, and the cocatalyst and optional promoter are chosen from those ayylol,l;ate for the particular catalyst. They may or may not be the same as those employed in the prepolymerization process. For eY~mrle, when producing a yrefelled sticky polymer, such as an EPDM, using a ~lerellad prepolymerized catalyst, such as prepolymerized V(acetylacetonate)3 catalyst, diethylal.l...i..llm chloride is used as the cocatalyæt and ethyl trichloroacetate is used as the promoter. The molar ratio of propylene to ethylene is in the range of about 0.2:1 to about 4.5:1 and is preferably in the range of about 0.35: to about 3:1 The propylene/ethylene molar ratio iB adjusted to control the level of propylene incorporated into the terpolymer. The molar ratio of hydrogen to ethylene i6 in the range of about 0.0001:1 to about 0.3:1 and is l.lef~ably in the range of about 0.0005:1 to about 0.1:1. The hydrogen/ethylene molar ratio is adjusted to control averàge molecular weights. The level of diene in the bed is in the range of about 1 to about 15 percent by weight based on the weight of the bed and is l,le~lably in the range of about 2 to about 10 percent by weight. The resi-lçnce time of the ~ule of resin and liquid in the fluidized bed r~n be in the range of about 1.5 to about 8 hours and is 1., efel ably in the range of about 2 to about 5 hours. The final EPM or EPDM
product can cont~in the following amounts of comonnmers: about 35 to about 80 percent by weight ethylene; about 18 to about 50 l elc~llt by weight propylene; and about 0 to about 15 l.erce~ by weight diene.
The crystallinity, also in weight percent, can be in the range of about zero (essçn1;~l1y amorphous) to about 15 percent by weight (nearly amorphous). The Mooney viscosity can be in the range of about 10 to about 150, ~refe~ably about 30 to about 100. Of course, it is to be understood that production of a di~el~llt sticky polymer, or use of a di~e-ellt prepoly_erized catalyst will cause these ratios and leveis and amounts to change. Such changes are well-understood by ~hose skilled in the art of polymerization.

Resin Particle Composition The novel resin particle of the instant il~ tion includes a central core substantially comprised of a sticky polymer and an outer shell subs~n~;~l1y comprised of a non-sticky polymer of high mnlec~ r weight and high comonomer content. A s-h~m~t;c diagram of a cross-section of the resin particle is depicted in Figure 1.
During the course of a polyme~7~tinn, the bed co...l l;ses three kinds of particles: newly introduced catalyst particles, ~ Wlll~ resin particles and fully formed resin particles. To ~re~ t ~g~lomeration, each of these kinds of particles must be p~ev~l~ted ~om a~glomerating.

First, normal catalyst particles introduced into a reactor under conditions that produce a sticky polymer would cause other particles to stick to the growing unprotected particle, and lead to rapid agglomeration of the entire bed. In the process of the present invention, howevel-, the prepolymerized catalyst particles that are introduced to the reactor are comprised of a high viscosity, non-sticky prepolymer surface or shell with active catalyst sites concentrated in the interior of the particle. Under polymerization conditions, sticky polymer is prerlomin~ntly made in the interior of the particle, the surface rçmAin~ non-sticky, and agglomeration does not take place.
Second, the growing resin particle, even if protected initially by a non-sticky coating, can be brittle and fracture, leading to sticky polymer finlling its way to the surface and, again, to ~glomeration of the bed.
This can happen, for instance, if the prepolymer portion of the prepolymerized catalyst has a high degree of crystallinity, such as an ethylene pre-homopolymer. In the process of the present invention, however, the prepolymerized catalyst particles are comprised of a prepolymer portion of high comonomer c~ntent Hence, in the growing resin particle, the shell is elastic enough to expand rather than fracture. Third, the fully formed resin particle is fully grown and e~pAn-led. But even if provided with an initial elastic non-sticky coating, if the e~pAn-led shell becomes too thin, it can break or the sticky polymer being produced can easily diffuse through to the surface, and again the bed vlill agglomerate. This can happen, for instance, if the prepolymer has a thin shell or is sufficently irregular to have thin spots in the shell. In the process of the present invention, however, the initial extent of prepolymerization is at least 50, preferably at least 100 grams of prepolymer per grArn of catalyst precursor. Such a high degree of prepolymerization ensures that the final resin product v~ill retain a sufficiently thick shell to provide Agglomeration protection, even for those particles that may be irregularly shaped. In addition, the shell will remain a smaller fraction of the final product on a weight percent basis th~n a shell of the s~me thickness grown from a prepolymer of lower initial loading.
In the process of the present illvq..~ ;on, then, the prepolymerized catalyst particles are comprised of an elastomeric prepolymer portion that can e~p~n~ as the catalyst particle grows to form the final resin product. Thus, there should be little prepolymer in the core of the particle. However, it is always possible, and in fact does happen, that n~cent particles aggregate together before growing subst~nt;s~lly.
This can happen when some prepolymerized catalyst particles do not have an adequate coating of prepolymer, for instance. When this aggregate is fully grown, its core is bound to have some non-sticky polymer that was brought in during the agglomeration phase.
Therefore, the typical values for the amount of sticky polymer in the core of the resin particle are at least 90% by weight, but less than 100% by weight.
On the other hand, as the resin particle grows, the non-sticky shell thins and becomes more porous. Sticky polymer molecules will diffuse toward the surface, and the shell will become a mi~ re of sticky polymer and non-sticky polymer. The interface of the sticky polymer and the non-sticky polymer will normally have a bigh concentration of sticky polymer while the surface region v~ill normally consist essentially of non-sticky polymer and remain non-sticky.
Consequently, the typical average values of non-sticky polymer in the outer shell are higher than 80% by weight.
Although such particles may not be entirely colorless, since they can be devoid of carbon black or other prominent color or color bodies, they are capable of acc~plillg and/or displaying color, or are colorable.
The resin composition of the present invention is therefore a colorable resin particle having (A) an outer shall that is at least about 80% by weight of a non-sticky polymer having (a) up to 90 mole % etbylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms; and (b) a flow inde~ of less than 20 decigrams/minute; and (B) an inner core of at least about 90% by weight of a sticky polymer. P~efelably the reæin particles are produced by a gas phase fluidized bed polymerization process at a tempeldlule at or above the sticking temperature of the resin and the resin particles C~JIItS~;II at least about 1 percent by weight of the non-sticky polymer.
mples of sticky polymers, which can be produced by the invention include ethylene/propylene rubbers and ethylene/~lo~lylene/diene termon--mer rubbers, high ethylene content propylene/ethylene block copoly_ers, poly (l-butene3 (when produced under certain react;on conditions), very low density (low modulus) polyethylenes, i.e., ethylene butene rubbers or hexene containing terpolymers, an ethylene alpha-olefin copolymer having a density of 880 kg/m3 to 915 kg/m3, ethylene/propylene/ethyli-len~norbornene, and ethylene/~lolJylene he~ ene terpolymers of low density.
The invention is filrther illustrated by the following e~mples.

EXAMPLES

h~Y~mrle 1 This e~mple illustrates prepolymerized catalyst preparation in a gas phase reactor using minim~l ~mounts of alkyl to limit the polymerization rate, and a gradient in propylene content A catalyst precursor was prepared from vanadium tris(acetylacetonate) and a silica support cont~ining 0.47 millimnles of vanadium per gram of precursor. A 50-liter, stirred, jacketed gas-phase reactor equipped with purified flow-controlled monomer feeds, a closed- loop, tempered-water tempelalure control system, and a vent/make-up system for pressure control was used to conduct the prepolymerization. The reactor was hçate-l to 35 C and charged with about 500 g of dehydrated Davison 955 silica as a start-up bed for dispersing catalyst and maint~qinin~ agitation, and not as a fluidization aid. Ethylene and propylene monomers were charged in a 0.5 molar ratio of propylene to ethylene until an ethylene partial pressure of 60 psia was reached, followed by 50 g of the catalyst precursor. Aliquots of ethyl tnchloroacetate and diethylaluminum chlonde were fed in the amount nece6s~ry to initiate and sustain reaction, and propylene and ethylene were fed in a molar ratio of 0.25 as necessary to m~int?~in pressure. As a consequence the molar ratio of propylene to ethylene in the reactor increased from its initial 0.5 value. The reaction was terminated when about 5000 g of monnm~r had been fed to the reactor, the reactor was cooled and purged with nitrogen, and the product discharged inertly (without exposure to air) and stored under nitrogen. Granular, free-flowing prepolymer was obtained, although the reactor walls were fouled with agglomerated resin.
Analysis showed the prepolymer to contain 20.0 wt% propylene by nuclear m~gnetiC re~on~nce (NMR) analysis, 212 ppm vanadium and 2.9 wt% silicon by inductively coupled pl~ (ICP) analysis, and an unmeasurably low melt flow index measured under ASTM-1238, Condition F, at 190C and 21.6 kg.

~mrles 2 5 The following ~mples illustrate the excellent gas phase operability of prepolymerized catalyst with no fouling or agglomeration, with no fluidization aid added to the gas phase polymerization, and with low levels of non-sticky polymer in the final resin product so that product properties, such as cure, remain normal.

mrl~ 2 A 1-liter stirred autoclave reactor was initially purged with nitrogen and heated to an internal tempeLa~ure of 100C for at least 15 minllteS under a slow, continuous purge of nitrogen. The reactor was cooled to 85C and about 200 g of sodium chloride, dried under vacu-lm at 115C for at least 12 hours, was taken from the vacuum oven while hot and added to the reactor through a 0.~-inch port under a nitrogen flow. The salt bed was stirred and purged with nitrogen for a additional 15 minutes. The reactor jacket was then cooled to 20C.

A quantity of the prepolymerized catalyst from ~ mrle 1 cont~ining a~l(,~ately 0.03 millimole vanadillm was weighed into a glass addition tube under nitrogen. Diethylaluminum chloride (DEAC) in a 240:1 DEAC/V mole ratio based on the catalyst charge and hey~(hloloplol~ene (HCP) in a 5:1 HCP/V mole ratio based on the catalyst charge were mixed with 1 gram of Davison 955 silica to aid in dispersion and added to the reactor under nitrogen. The catalyst charge was added to the reactor from the addition tube and the 0.5-inch port was tightly capped. The reactor was purged briefly with nitrogen through the vent line, sealed, and the stirring speed increased to 300 rpm.
An initial quantity of 4 ml of ENB was added to the reactor. At the same time, a mixture of ethylene, propylene and hydrogen with a C3/C2 molar ratio of 1.20 and an H2/C2 ratio of 0.15 was fed to the reactor at an ethylene flow rate of 2.0 liters per minute until the reactor pressure reached 120 psig. At this point, the hydrogen feed was turned off, the C3/C2 molar ratio in the feed was adjusted to a value of 0.23, the ENB/C2 molar feed was adjusted to a rate of 0.02 and the jacket tempçra~ul e was adjusted to bring the internal reactor tempe~a~u,e to 35C. Additional HCP was fed to the reactor at a rate of about 0.13 millimr~les per mole of C2 consumed and monomers were fed on ~çm~n~ to maintain pressure. The reaction was continued for 150 minutes and then terminated by stopping the flow of monomers and purging the reactor with nitrogen.
The reactor was cooled and opened to take out the ..-; x (- ., e of salt and polymer product. The salt was washe* out with water to obtain about 160 grams of granular resin. The product was free-flowing and not agglomerated and the reactor walls were not fouled.
The residual amounts of prepolymerized catalyst in the resin, and the catalyst productivity, were determined by mass b~l~nce and the polymer composition was determined by NMR (n~ le~r m~ tiC
resonance) analysis. The properties are set forth in Table 1.

~mrle 3 F,~mple 2 was repeated except the DEAC/V ratio was increased to 400:1 and the H2/C2 ratio was decreased to 0.075. The re~ct;on was continued for 115 minutes and 175 grams of granular resin were recovered. The product was free-flowing and not agglomerated and the reactor walls were not fouled. Additional properties are set forth in Table 1.

~mrle 4 F,~mple 2 was repeated except the DEAC/V ratio was increased to 580:1. The reaction was contimled for 105 minutes and 160 grams of granular resin were recovered. The product was free-flowing and not agglomerated and the reactor walls were not fouled. Additional properties are set forth in Table 1.

h~Y~mrle 5 F.~mple 2 was repeated except the DEAC/V ratio was increased to 580:1 and the H2/C2 ratio was decreased to 0.10. The reaction was continued for 100 minutes and 148 grams of granular resin were ~ recovered. The product was free-flowing and not agglomerated and the reactor walls were not fouled. Additional ~lo~el lies are set forth in Table 1.

Table 1 Example Agglom- Activity Prepolymer Wt% C3 Wt% Melt Mooney Cure eration Residue ENB Flow (est.) 2 None 2020 4% 23.74.9 8 46 29.8 3 None 2850 4% 24.33.9 2 70 30.5 4 None 2800 4% 23.03.2 90 21 25.2 - 5 None 2750 5% 23.53.9 17 53 30.3 -Notes to Table:
1. Activity = the amount of resin produced in grams per millimole of vanadium per hour;
2. Wt% C3 and wt ~o ENB = the percentages of propylene and ethylidene norbornene, respectively, in the resin as determined by C13 NMR;
3. Melt flow = the melt index as determined under ASTM-1238, Condition F at 190C and 21.6 kilograms;
4. Mooney (est) = Mooney viscosity as estimated using an oscillating disk rheometer (ODR) from a linear correlation of gum Mooney viscosity under standard conditions (M(L) (minimum torque resistance)1~4 at 125C] with M(L) measured in ASTM D-3568 formula no. 1 using an ODR at 160C and a 1 arc at 100 cpm;
5. Cure = the difference between the m~x;~ torque, M(H), and the ini_um torque, M(L), following ASTM D-2084 test methods for the ODR, where Formula No. 1 of ASTM D-3568 is used following Procedure 6.1.2 for a miniature internal mixer and Practice D-3182.

mr)les 6-8 These examples illustrate prepolymerized catalyst preparation in a gas phase reactor using controlled monomer flow to li_it the polymerization rate, a gradient in propylene content and a gradient in molecular weight.

mrle 6 A catalyst precursor was prepared from vanadium tris(acetylacetonate) and a silica support Cont~ininF 0.47 millimnles of vanadium per gram of precursor. A 4-liter, stirred, jacketed gas-phase reactor equipped with purified, flow-controlled monomer feeds, and a steam/water temperature control system was used to conduct the prepolymeriz~ation. The reactor waæ baked out at 100C and then cooled to 30C and charged with about 25 g of dehydrated HiSil 233 silica as a start-up bed for dispersing catalyst and ms~intpining agitation, as well as serving as an inert particulate material fluidization aid. T_is was followed by about 2 g of the catalyst precursor, about 30 millim~lles oftriethylal.. ;.. - chloride and about 15 millimoles of heY~-hloro~ro~ e. Propylene and et_ylene were then fed in a molar ratio of 0.50 until the reactor pressure re~che-l 50 psig.
Monomers were then fed in a molar ratio of propylene to ethylene of 0.25 at a rate of 1 standard liter per minute, and reactor pressure was allowed to vary. The reaction was termin~te~ when about 250 g of monomer had been fed to the reactor, the reactor was cooled and purged with nitrogen, and the product discharged inertly and stored under nitrogen. Granular, free-flowing prepolymer was obtained.
Analysis showed the prepolymer to contain 25.1 wt% propylene by nuclear magnetic resonance, and 161 ppm vanadium by inductively coupled pl~m~ analysis. It had an lmme~urable melt flow index measured at ASTM-1238, Condition F.

~mrle 7 - A catalyst precursor was prepared, as described above, from vanadillm tris(acetylacetonate) and a silica support cont~ining 0.47 millimoles of vanadium per gram of precursor. A 4-liter, stirred, jacketed gas-phase reactor equipped with purified, flow-controlled monomer feeds, and a steam/water te~ ,e,atule control system was used to conduct the prepolymerization. The reactor was baked out at 100C and cooled to 30C and charged with about 25 g of dehydrated HiSil 233 silica as a start-up bed for dispersing catalyst and maint~ining agitation, as well as serving as the inert particulate material fluidization aid. This was followed by about 2.2 g of the catalyst precursor, about 25 millimoles of diethyl~l.. ;.. l. chloride and about 15 millimoles of he~hlolo~lo~ene. Propylene and ethylene were fed in a molar ratio of 0.25 at a rate of 0.30 st~n~l~rd liters per minute to initiate reaction. The rate was slowly increased to 1 standard liter per minute over 30 minutes, and reactor pressure was allowed to vary. Simultaneously, 200 st~n~rd cubic centimeters of hydrogen were added slowly over 30 ...; . . ~ s. The re~ction was termin~ted when about 200 g of monomer had been fed to the reactor, the reactor was cooled and purged with nitrogen, and the product discharged inertly and stored under nitrogen. Granular, free-flowing prepolymerized catalyst was obtained. Analysis showed the prepolymerized catalyst to contain 300 ppm vanadium by inclllctively coupled pl~m~ analysis. It had a melt flow index of 0.42 measured at ASTM-1238, Condition F.

mrl~ 8 A catalyst precursor was prepared, as described above, from vanadium tris(acetylacetonate) and a silica support cont~ining 0.47 millimoles of vanadillm per gram of precursor. A 50-liter, stirred, jacketed gas-phase reactor equipped with purified, flow-controlled mo~omer feeds, a closed-loop, tempered-water tempelalule control system, and a vent/make-up system for pressure control was used to conduct the prepolymeri7.~tio~ The reactor was he~te~ to 35 C and charged with about 500 g of dehydrated HiSil 233 silica as a start-up bed for dispersing catalyst and maint~ining agitation, as well as serving as a fluidization aid. This was followed by about 50 g of the catalyst precursor, about 250 millimoles of diethylalumi,lu.l, chloride and about 35 millimoles of hexachlo~o~,o~ene. Propylene and ethylene were then fed in a molar ratio of 0.25 to initiate reaction, and hydrogen was added slowly until the H2/C2 ratio reached 0.002. Hydrogen feed was stopped. Monomers were fed at a rate of 3 lb/hr and reactor pressure was allowed to va2y. The propylene to ethylene ratio in the reactor increased from its initial value of 0.25, and the hydrogen to ethylene ratio increased from its initial value of 0Ø The reaction was termin~te~ when about 5000 g of monomer had been fed to the reactor, the reactor was cooled and purged with nitrogen, and the product discharged inertly and stored under nitrogen. Granular, free-flowing prepolymer was obt~ine-1, and the reactor walls were not fouled.

D-17229 21593~8 Analysis showed the prepolymerized catalyst to contain 23.4 wt%
propylene by nuclear magnetic reson~nce, 215 ppm vanadillm and 1.8 wt~o silicon by inductively coupled p~m~ analysis, and a melt flow index of 0.2 g per 10 min measured at ASTM-1238, Condition F.

FY~mFles 9-11 These e~mples illustrate the excellent gas phase operability of prepolymerized catalyst with no fouling or agglomeration, no additional fll~ tion aid, and low levels of non-sticky polymer in the resin product, even at high propylene contents and low molecular weights where the polymer produced will have m~imum stickiness.

FYs-mrl~ 9 mple 2 is repeated except the prepolymerized catalyst of mrle 6 contsining 0.05 millimole vanadium was used and the H2/C2 ratio used was 0.20. The reaction was continlle~ for 160 minutes ad 143 gr~ms of granular resin were recovered. The product was free-flowing and not agglomerated and the reactor walls were not fouled.
Additional properties are set forth in Table 2.

~YS~mrle 10 mple 7 was repeated with an H2/C2 charge ratio of 0.076, a C3/C2 charge ratio of 2.1 and a C3/C2 feed ratio of 0.38. The reaction was continued for 190 minutes and 189 grams of granular resin were recovered. The product was free-flowing and not ~gglomerated and the reactor walls were not fouled. Additional properties are set forth in Table 2.

~mrle 11 mple 8 was repeated with an HCPIV charge ratio of 10:1.
The re~ct,ion was co..~ ;....ed for 108 minutes and 90 grams of granular resin were lecoveled. The product was free-flowing and not ~gglomerated and the reactor walls were not fouled. Additional properties are set forth in Table 2.

Table 2 Example Agglom-Activit3r Prepolymer wt% C3 wt% Melt Mooney Cure eration Residue ENB Flow (est.) 9 None950 8% 27.12.5 3.5 76 21.4 None1080 6% 36.23.4 1.5 92 24.3 11 None530 9% 36.93.5 33 32 24.0 12 Severe790 0% 23.00.0 7.5 13 Severe1750 0% 25.60.0 27 -- --Notes to Table:
1. Activity = the amount of resin produced in grams per millimole of vanadium per hour;
2. Wt% C3 and wt % ENB = the percentages of propylene ad ethylidene norbornene, respectively, in the resin as determined by C13 NMR;
3. Melt flow = the melt index as deter_ined under ASTM-1238, Condition F at 190C and 21.6 kilograms;
-4. Mooney (est) = Mooney viscosity as estim~t~d using an oscillating disk rheometer (ODR) from a linear correlation of gum - Mooney viscosity under standard conditions [M(L) (minimum torque resistance) 1+4 at 125C] with M(L) measured in ASTM D-3668 For_ula No. 1 using an ODR at 160C and a 10 arc at 100 cpm;
5. Cure = the dilTe~ellce between the m~mmllm torque, M(H), and the 111;11;1111~111 torque, M(L), following ASTM D-2084 test methods for the ODR, where Formula No. 1 of ASTM D-3568 is used following Procedure 6.1.2 for a miniature internal miYer and Practice D-3182.

Comparative ~Y~mples 12-13 The following eY~mrles are comr~rative ~y~mr~es that illustrate that gas phase polymerizations using a non-prepolymerized catalyst exhibit severe fouling and agglomeration even under EPM
polymerization conditions, which are the least sticky conditions due to the lack of liquid diene in the reactor.

Comparative h~Y~mrle 12 A supported catalyst precursor was ~ aled as for ~Y~mple 1, but not prepolymerized. A polymerization was conducted in a .~imil~r fashion to ~ mple 3, but this ull~lelJolymerized catalyst cont~ining al.~.o~;...~t,ely 0.035 millimole vanadium was used, no ENB was used, and the H2/C2 ratio was reduced to 0.04. The reaction was continued for 190 minutes and about 100 grams of resin-were produced. However the resin was in the form of sheets and chunks and the walls were heavily fouled. The resin was not free-flowing and had to be scraped from the reactor. Additional properties are set forth in Table 2.
omparative ~y~mrle 13 mple 12 is repeated except ethyl trichloroacetate was used as the promoter instead of h~Y~rhlol o~lo~ene. The reaction was continued for 106 minutes and 116 grams of resin were produced.
However the resin was in the form of sheets and chunks and the walls were heavily fouled. The resin was not free-fiowing and had to be scraped from the reactor. Additional properties are set forth in Table 2.

mrle 14 This e~mple illustrates production of a non-sticky prepolymerized catalyst with an EPR prepolymer portion in a slurry of the inert diluent, isobutane, using the method of limiting monomer flow to control the rate of production of the prepolymer portion.

A catalyst precursor was prepared as in U. S. Patent No.
4,~08,842 from vanadium trichloride, tetrahydrofuran, silica and diethylaluminum chloride. A l.~-liter, stirred, doublejacketed slurry reactor equipped with purified, flow-controlled monomer feeds and a l~nu~llg external fluid tempeldlule control system was used to conduct the prepolymeri7.Ation The reactor was baked out at 100C
and then cooled to 25C and charged with 1.0 grams of the catalyst precursor described above. The reactor was then sealed and charged under pressure with 650 mL of dry isobutane. To the resulting slurry was then added 7.0 mL of a 25 % by weight solution of triisobutylaluminum in h~ne, 1.0 mL of a 1.0 mole per liter solution of 1,1-difluorotetrachloroethane in hexane, and 50 st~nrl~rd cubic centimeters of hydrogen gas. The tempeldlu~e was then controlled at 25C + 1 C and monomer flows were initiated at a rate of 1.0 standard liter per minute of ethylene and 0.25 standard liters per mimlt4 of propylene. The pressure of the reactor was allowed to vary according to the monomer feed rate and polymerization rate. Polymerization was continued for 1.0 hour, at which time the reactor was depressurized, allowing all of the isobutane to evaporate from the reactor. The resulting powder was then removed from the reactor and stored under an inert atmosphere. The total non-sticky prepolymerized catalyst removed from the reactor was 105 grams. It was entirely free-flowing and had an overall propylene content of 25.1 percent by weight (as measured by infrared spectroscopy) and a flow index of 0.33 decigramstminute (ASTM-1238, Condition F, at 190C and 21.6 kg).
The reactor was entirely unfouled and was ready for reuse with minim~l cleanup.

F~Ys~mrle~ 15-17 These examples illustrate prepolymerization in slurry in liquified ethylene and propylene, using the method of limiting the co-catalyst feed in order to limit the production rate.

~mrle 15 A catalyst ~eculsor was prepared as in U. S. Patent No.
4,508,842 from vanadium trichloride, tetrahy~L.Jrul~, silica and D-17229 ~i5 9388 diethylal~ , chloride. A 1-liter, stirred, jacketed slurry reactor equipped with purified, flow-controlled monomer feeds and flowing mi~ed hot and cold water for heat removal and t~mre~alule control was used to conduct the prepolymerization. The reactor was baked out at 100C and then cooled to 25C, and charged with 0.50 grams of the catalyst precursor described above as a 10 % by weight slurry in mineral oil. The reactor was sealed and charged under pressure with 500 mL of liquid propylene. The temperature was controlled at 20C +
1 C and ethylene was introduced to a pressure of 420 psig, and continually fed on rlçm~n~l throughout the prepolymerization to maintain this pressure. Diethylaluminum chloride (25 % by weight in he~r~ne) was then added to the reactor by syringe in 0.15 mL aliquots until the inception of a low level of polymerization, represented by a flow of ethylene of about 1 st~n-3~rd liter per minute into the reactor and a 5 C tempelal~e difference across the wall of the reactor. Two such aliquots of diethylalu~ chloride were necess~ry. The prepolymerization was allowed to continue until about 50 gr~ms of prepolymerized catalyst had been produced as judged by the amount of heat evolved in the prepolymerization. The reactor was then vented and the mo~omers allowed to evaporate from the reactor. The resulting non-sticky prepolymerized catalyst, 51.9 grams, was removed from the reactor and stored under an inert atmosphere. It was an entirely free-flowing powder, cont~ined 25.8 % by weight propylene (as measured by infrared spectroscopy) and had an lmme~urably low flow index (ASTM-1238, Condition F, at 190C and 21.6 kg). The reactor was entirely unfouled and could be reused with minim~l cleanup.

F~Y~mrle 16 Prepolymerization as in F.~nple 15 was carried out, except using 0.2 grams of catalyst precursor and at a pressure of 400 psig, and producing 44.1 gra_s of non-sticky prepolymerized catalyst. The prepolymer portion had a propylene content of 28.9 % by weight (as D-17229 21S9~88 measured by infrared spectroscopy), was entirely free-flowing and was removed from the reactor under non-inert conditions in air.

~Ys~mrle 17 - The prepolymerization F~mple 16 was repeated, except using a catalyst precursor ~ Ja,ed from vanadium acetylacetonate, silica and diethylal.... -;.. ..chloride and slurried in mineral oil, and using diethylal~l.. ;... m chloride as the co-catalyst in place of triisobutylaluminum. From 0.1 grams of catalyst precursor was obtained 17.8 grams of granular, free-flowing non-sticky prepolymerized catalyst with a propylene contçnt of 28.0 ~o by weight as measured by infrared spectroscopy.

.Ys~mples 18 and 19 These examples illustrate the shell and core composition of the resin particle.

~m~le 18 A prepolymerized catalyst was prepared as in F~mple 8, except that about 500 g of dehydrated carbon black were used as the ~ lup bed for dispersing catalyst and m~int~ining agitation. This made the outside of the prepolymerized catalyst particle black. The prepolymerized catalyst was isolated and an EPM polymerization was conducted in a simil~r m~nnçr to F~mrle 3, but using this carbon black-cont~ining prepolymer, an H2/C2 ratio of 0.04 and no ENB. The reaction was continued for 72 minutes to produce 40 grams of product resin. The product was free-flowing and not agglomerated, and the reactor walls were not fouled. -Randomly selected particles from the sample were embedded in Buehler epoxy and allowed to cure overnight at room tempela~e. The embedded particles were then ~ectioned with a ~ monri blade using a Reirher-Jung Ultracut E cryogenic _icrotome operating at minus 60 C to expose the region most closely representing a diametrical cross section of the particle. These cross-- ~- 2159388 sectioned samples were then ç~mined in an Olympus Vanox light microscope and a JEOL JSM-35C ss~nning electron microscope. All of the particles e~qmined cont~in~d a layer of carbon black ~ o~ ing an outer shell of non-sticky prepolymer and a core of sticky EPM
rubber. There was no carbon black in the interior of the particles. The thin carbon black layer served as an e~cellent marker to show that the~
prepolymerized catalyst particle had expanded with the ~lVWillg resin particle and non-sticky polymer had rem~ined on the surface to form the product composition of the instant invention.

mple 19 A prepolymerized catalyst was prepared as in F'~mple 8, without carbon black, isolated, and an EPM poly_erization was conducted in a fiimil~r m~nner to ~mple 18 except the polymerization was allowed to continue for 187 minutes to produce 113 grams of product resin. R~n~omly selected particles were isolated, embedded, and cross-sectioned as in ~mple 18 in order to reveal their cross-sectional morphology. Electron micrographs of the cross-sections were obt~ine~l All of the particles e~qmined revealed an outer shell of non-sticky polymer surrollnliing an inner core of soft, sticky polymer.
In addition, to further characterize the product resin, one of the cross-sectioned particles was e~mined using a Bio-Rad UMA300A
sc~nning infrared microscope, and the ethylene content of the particle determined as a function of the distance from the surface by measuring the ratio of the 719 cm-1 absorption due to ethylene to the 1178 cm-1 absorption due to propylene as described in ASTM Method D-3900.
Spectra were recorded in steps of 25 microns for the first four measur~ments on either side of the cross-sectioned particle and in steps of 50 rnicrons across the center of the particle. The results, shown in Figure 2, show the ethylene content decreasing sharply across an outer shell of non-sticky polymer of app~,x;f .~t~ly 100 microns, rem~ining constant at the lowest ethylene content across a across an outer shell of non-sticky polymer of a~ o~ ately 100 microns, rem~ining constant at the lowest ethylene content across a central core of amorphous sticky polymer, then rising sharply again as the shell is reached on the other side.
.
F.~mrles 20-21 These ey~mples illustrate the production of a non-sticky prepolymerized catalyst and an EPDM resin made from it in which a fluorescent alpha olefin comonomer is incorporated in the prepolymer portion to mark the location of the prepolymer portion, and e~min?.tion of the particles using fluorescenre microscopy.

F.~s~mrle 20 A catalyst precursor was prepared as in U. S. Patent No.
4,508,842 from vanadium trichloride, tetrahy~orul~l, silica and diethylall~...illl.... chloride. A 1.5-liter, stirred, doublejacketed gas-phase reactor equipped with purified, flow-controlled monomer feeds, and a refluxing external fluid tempela~ule control system was used to conduct the prepolymeri7.~t;on The reactor was baked out at 100C
and then cooled to 25C and charged with 0.33 grams of the catalyst precursor described above. The reactor was sealed and charged under pressure with 650 mL of dry isobutane. To the resulting slurry was added 1.0 mL of 1-(7-octene-1-yl)-2-phenylindole (prepared in a manner ~imil~r to that described in Heaney, H. and Ley, S. V. J. Chem. Soc., Perkin I, 1973, p. 499), 15.0 mL of a 25 % by weight solution of triisobutylalllminum in he~ne, 2.0 mL of a 60 % by weight solution of 1,1-difluoro-1,2,2,2-tetrachloroethane in he~ne, and 25 st~ntl~rd cubic centimeters of hydrogen gas. The temperature was controlled at 25C
+ 1C and monomer flows were initiated at a rate of 0.33 st~n~l~rd liter per minute of ethylene and 0.17 standard liters per minute of propylene. The pressure of the reactor was allowed to vary accol Lng to the monomer feed rate and polymerization rate. Prepolymerization was continued for 140 minlltes, at which time the reactor was -depressurized, allowing all of the isobutane to evaporate to the atmosphere. The resulting powder was removed from the reactor under an inert atmosphere of dry nitrogen, washed with 1500 mL of dry, deo~y~ellated isopentane to remove excess unpolymerized 1-(7-octene-1-yl)-2-phenylindole, and stored under an inert atmosphere.
The total non-sticky prepolymerized catalyst rem~ining after washing was 19.4 gr~ms. It was entirely free-flowing and had an overall propylene content of 21.0 percent by weight (as measured by infrared spectroscopy) and a flow index of 0.33 decigrams/minute (ASTM-1238, Condition F, at 190C and 21.6 kg). The reactor was entirely unfouled and was ready for reuse with minim~l cleanup.
Randomly selected particles from the siq~nple are embedded in Buehler epoxy and allowed to cure overnight at room tempeldlule. The embedded particles are then sectioned with a ~ mond blade using a Reicher-Jung Ultracut E cryogenic microtome operating at minus 60C
to expose the region most closely representing a diametrical cross section of the particle. Then a 1 to 3 micrometer slice is remove with a mon~ blade and mounted on a microscope slide. These cross-sectioned samples are P~mined in an Olympus Vanox light microscope fitted with a reflected-light ultraviolet excitation with barrier high-pass filters cutting off at either 435 nm or 455 nm for fluorescence measurement. Results indicate a core-shell particle morphology in the prepolymerized catalyst particles.

mrle 21 The non-sticky prepolymerized catalyst prepared as in Example 20 was oxygenated under an atmosphere of dry o~y~ell as described in U. S. Patent No. 5,322,793. Excess oxygen was removed from the prepolymerized catalyst by evacuation, and the prepolymer catalyst was stored under an atmosphere of dry nitrogen. An EPDM
polymerization was conducted following the techniques of F'.~mple 2, but using 0.05 millimoles of the above prepolymerized catalyst.
I)iethylaluminum chloride (DEAC) in a 400:1 DEAC/V mole ratio based - ~2 -on the catalyst charge and ethyltrichloroacetate (ETCA) in a 10:1 ETCA~V mole ratio based on the catalyst charge were used. An H2/C2 ratio of 0.01 was used, and ETCA and ENB were continually fed to the polymerization. The polymerization was continued for 176 minutes to yield 114 grams of product resin. The product was free-flowing and not ~glomerated and the reactor walls were not fouled. The resin contained 19 wt% C3 and 2.~ wt~Zo ENB by nuclear magnetic resonance analysls.
R~n-lomly selected particles from the sample are embedded in Buehler epoxy and allowed to cure overnight at room temperature. The embedded particles are then sectioned with a ~ mon~l blade using a Reicher-Jung Ultracut E cryogenic microtome operating at minus 60C
to expose the region most closely representing a diametrical cross section of the particle. Then a 1 to 3 micrometer slice is remove with a di~mond blade and mounted on a microscope slide. These cross-sectioned samples are ~mined in an Olympus Vanox light microscope fitted with a reflected-light ultraviolet excitation with barrier high-pass filters cutting off at either 435 nm or 455 nm for fluorescence measurement. Results indicate a core-shell particle morphology in the final resin particles.

Claims (20)

1. A resin particle having (A) an outer shell that is at least about 80% by weight of a non-sticky polymer having (a) 10 to 90 mole % ethylene and having at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms, and (b) a flow index of less than 20 decigrams/minute; and (B) an inner core of at least about 90% by weight of a sticky polymer;
said resin containing at least about 1 percent by weight of said non-sticky polymer.

2. The resin particle of Claim 1 wherein said non-sticky polymer is about 1 to about 50 percent by weight of said resin.

3. The resin particle of Claim 2 wherein said non-sticky polymer is about 1 to about 15 percent by weight of said resin.

4. The resin particle of Claim 1 wherein said sticky polymer is:
(a) an ethylene propylene rubber (b) an ethylene propylene diene termonomer rubber (c) an ethylene alpha-olefin copolymer having a density of 880 kg/m3 to 915 kg/m3.

5. The resin particle of Claim 4 wherein said ethylene propylene diene termonomer rubber is an ethylene propylene 5-ethylidene-2-norbornene termonomer rubber.

6. A non-sticky prepolymerized catalyst comprising a prepolymer portion and a catalyst portion wherein (1) the prepolymer portion has (a) 10 to 90 mole % ethylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms; and (b) a flow index of less than 20 decigrams/minute; and wherein (2) the ratio of the prepolymer portion to the catalyst portion is 26:1 to 1000:1.

7. The non-sticky prepolymerized catalyst of Claim 6 wherein the prepolymerized catalyst has particles having a shell and core such that the prepolymer portion is substantially located in the shell and the catalyst portion is substantially located in the core.

8. The non-sticky prepolymerized catalyst of Claim 7 wherein the shell contains about 75 percent by weight of the prepolymer and the core contains about 85 percent by weight of the catalyst portion.

9. The non-sticky prepolymerized catalyst of Claim 8 wherein the core is about 1 to 50 volume percent of the catalyst and the shell is about 50 to 99 volume percent of the catalyst.

10. The non-sticky prepolymerized catalyst of Claim 6 wherein the catalyst portion contains a catalyst precursor selected from the group consisting of A. vanadyl trihalide, alkoxy halides and alkoxides such as VOCl3, VOCl2(OR), and VO(OCxHy)3 wherein x is 1 to 12 and y is x+3;
B. vanadium tetrahalide and vanadium alkoxy halides such as VCl4 and VCl3(OR);
C. vanadium and vanadyl acetylacetonates and chloracetyl acetonates such as V(AcAc)3 and VOCl2(AcAc) wherein (AcAc) is an acetylacetonate; and D. vanadium trihalides and alkoxy halides such as VCl3 and VO(OR)3; and wherein R is an alkyl having 1 to 12 carbon atoms.

11. The non-sticky prepolymerized catalyst of Claim 6 wherein the catalyst portion contains a catalyst precursor having the formula MgaTi(OR)bXc(ED)d wherein R is independently an aliphatic or aromatic hydrocarbon radical having 1 to 14 carbon atoms or COR' wherein R' is an aliphatic or aromatic hydrocarbon radical having 1 to 14 carbon atoms; X is independently chlorine, bromine or iodine; ED is an electron donor; a is 0.5 to 56; b is an integer from 0 to 2; c is 1 to 116; and d is 2 to 85.

12. The non-sticky prepolymerized catalyst of Claim 6 wherein the catalyst portion comprises a metallocene.

13. The non-sticky prepolymerized catalyst of Claim 6 wherein the prepolymer portion contains about 0.05 to about 10 percent of an inert particulate material selected from the group consisting of carbon black, talc, clay, and silica.

14. The non-sticky prepolymerized catalyst of Claim 13 wherein the inert particulate material is silica.

15. The non-sticky prepolymerized catalyst of Claim 6 wherein the prepolymer portion contains a diene.

16. A process for the production a resin having (A) an outer shell of a non-sticky polymer and (B) an inner core of a sticky polymer, which process comprises contacting ethylene, at least one alpha olefin having 3 to 18 carbon atoms, and optionally at least one diene, in a gas phase fluidized bed in the presence of hydrogen, at a temperature at or above the sticking temperature of the sticky polymer under polymerization conditions with (I) a non-sticky prepolymerized catalyst comprising a prepolymer portion and a catalyst portion wherein (1) the prepolymer portion has (a) 10 to 90 mole % ethylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms, and (b) a flow index of less than 20 decigrams/minute; and wherein (2) the ratio of the prepolymer portion to the catalyst portion is 25:1 to 1000:1;
(II) a co-catalyst; and (III) optionally a promoter;
and wherein the amount of non-sticky polymer is sufficient to essentially prevent agglomeration of the fluidized bed and of sticky polymer.

17. The process of Claim 16 wherein the alpha olefin is an alpha olefin having 3 to 6 carbon atoms; the diene is 1,4-hexadiene or 5-ethylidene -2-norbornene; the non-sticky prepolymerized catalyst comprises a vanadium precursor; the co-catalyst is selected from the group consisting of diethylaluminum chloride, triethylaluminum, triisobutylaluminum, trimethylaluminum, diisobutylaluminum chloride, and dimethylaluminum chloride; and the promoter is selected from the group consisting of ethyl trichloroacetate, hexachloropropylene, chloroform, butyl perchlorocrotonate, and 1,1-difluorotetrachloroethane.

18. A prepolymerization process for producing the non-sticky prepolymerized catalyst having a prepolymer portion and a catalyst portion wherein (1) the prepolymer portion has (a) 10 to 90 mole %
ethylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms, and (b) a flow index of less than 20 decigrams/minute; and wherein (2) the ratio of the prepolymer portion to the catalyst portion is 25:1 to 1000:1, which process comprises contacting a prepolymerization catalyst in a slurry of inert solvent with ethylene and at least one alpha olefin, and optionally a diene, such that (i) the temperature of the slurry is maintained such that the prepolymer portion of the non-sticky prepolymerized catalyst is insoluble in the slurry;

(ii) the total feed rate of the ethylene and the alpha olefin is less than or equal to 500 grams of ethylene and alpha olefin per gram of catalyst per hour;
(iii) the ratio of ethylene to alpha olefin is maintained at a constant ratio of less than or equal to 9:1;
(iv) the process is terminated when the ratio of the prepolymer portion to catalyst portion of the non-sticky prepolymerized catalyst is 25:1 to 1000:1 by evaporating the solvent and unreacted ethylene and alpha olefin; and (v) optionally an inert particulate material is added to the slurry immediately prior to the terminating step.

19. A prepolymerization process for producing the non-sticky prepolymerized catalyst having a prepolymer portion and a catalyst portion wherein (1) the prepolymer portion has (a) 10 to 90 mole %
ethylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms, and (b) a flow index of less than 20 decigrams/minute; and wherein (2) the ratio of the prepolymer portion to the catalyst portion is 25:1 to 1000:1, which process comprises contacting a prepolymerization catalyst precursor in a slurry of liquid ethylene and propylene, and optionally a diene, with a co-catalyst such that (i) the rate of polymerization is less than or equal to 500 grams ethylene and propylene per gram catalyst precursor per hour;
(ii) the slurry is maintained at a constant pressure of at least 300 psia;
(iii) the process is terminated when the ratio of the prepolymer portion to catalyst portion of the non-sticky prepolymerized catalyst is 25:1 to 1000:1 by purging the unreacted ethylene and alpha olefin; and (iv) optionally an inert particulate material is added to the slurry immediately prior to the terminating step.

20. A prepolymerization process for producing the non-sticky prepolymerized catalyst having a prepolymer portion and a catalyst portion wherein (1) the prepolymer portion has (a) up to 10 mole %
ethylene and at least 10 mole % of one or more alpha olefins having 3 to 18 carbon atoms, and (b) a flow index of less than 20 decigrams/minute; and wherein (2) the ratio of the prepolymer portion to the catalyst portion is 25:1 to 1000:1, which process comprises charging a prepolymerization catalyst, ethylene, and an alpha olefin, and optionally a diene, in a gas phase stirred reactor (i) at a temperature below the sticking temperature of the prepolymer portion of the non-sticky prepolymerized catalyst, (ii) such that the total feed rate of the ethylene and the alpha olefin is less than or equal to 500 grams ethylene and alpha olefin per gram of catalyst per hour;
(iii) the ratio of ethylene to alpha olefin is maintained at a constant ratio of less than or equal to 9:1;
(iv) terminating the process by purging unreacted ethylene and alpha olefin; and (v) optionally an inert particulate material is added initially to the reactor.