WO2006130310A1

WO2006130310A1 - Process for transitioning a reactor from a ziegler-natta catalyzed polymerization to a polymerization catalyzed by another type of catalyst system

Info

Publication number: WO2006130310A1
Application number: PCT/US2006/017497
Authority: WO
Inventors: Michael D. Turner; Guadalupe G. Leal
Original assignee: Union Carbide Chemicals & Plastics Technology Corporation
Priority date: 2005-05-27
Filing date: 2006-05-05
Publication date: 2006-12-07
Also published as: TW200706546A; AR055325A1

Abstract

A process for transitioning a continuous olefin polymerization reaction, catalyzed by a Ziegler-type catalyst, to a polymerization reaction catalyzed by another type of catalyst system. In this process, the introduction of the transition metal halide component of the Ziegler-type catalyst and the electron donor of the Ziegler-type catalyst into the reactor are discontinued, and the level of each of these components is reduced, using a catalyst inventory reduction step or steps, before and/or after the discontinuation of the catalyst components. The catalyst inventory reduction is eventually followed by the introduction of at least one reversible poison, such as CO2, and/or at least one irreversible poison into the reactor, and the circulation of the poison(s) within the reactor. The reactor is purged with an inert gas. Following this process, another catalyst system (or second catalyst system) is introduced into the reactor, and polymerization initiated.

Description

PROCESS FOR TRANSIΗONING A REACTOR FROM A ZIEGLER-NATTA

CATALYZED POLYMERIZATION TO A POLYMERIZATION CATALYZED BY

ANOTHER TYPE OF CATALYST SYSTEM

REFERENCE TO PRIOR APPLICATION

This application claims the benefit of Provisional Application No. 60/685,553, filed on May 27, 2005, incorporated herein, in its entirety, by reference.

BACKGROUND OF THE INVENTION The invention relates to processes for transitioning a continuous polymerization reactor from a Ziegler-Natta catalyzed reaction to a reaction catalyzed by another type of catalyst.

In the production of olefin polymers in a commercial reactor, it is often necessary to convert from one type of catalyst system, producing polymers having certain properties and characteristics, to another catalyst system, capable of producing polymers with different properties and characteristics. For example, when converting from a Ziegler-Natta or Ziegler-type catalyst to a chromium-based catalyst, it has been found that the components of the Ziegler-type catalyst act as poisons for the chromium- based catalyst, and that the chromium-based catalyst is ineffective in promoting polymerization if it is introduced directly into the polymerization reactor in the presence of the Ziegler-type catalyst. This is true even if a large excess of the chromium-based catalyst is employed.

Since direct conversion from a Ziegler-type catalyst to a chromium-based catalyst has proven unsuccessful, one method to effect such change of catalyst has been by halting the polymerization reaction initiated by the Ziegler type catalyst completely, emptying the reactor to rid it of the original catalyst system, and then re-initiating polymerization with the new catalyst system. Furthermore, in order to reinitiate polymerization with the new catalyst, the recharged reactor first had to be purged to low levels of oxygen and water, scavengers had to be employed to remove other catalyst poisons, and prolonged induction periods had to be used after the new catalyst was introduced into the reactor. These steps were required before polymerization would begin again. Consequently, such catalyst conversions were time consuming and costly, ordinarily requiring about four days or more of reactor shutdown in a commercial operation, before polymerization could once again be re-initiated with the new catalyst. Another current transition procedure from Ziegler-type catalyst to a chromium catalyst uses silica injection to bind the free triethylaluminum (or other aluminum alkyl) — electron donor complexes, typically present in almost all second generation Ziegler type catalysts, and which inhibit chromium catalyst activity. The surface silanols on the silica are theorized to bind these complexes, and this binding effectively scavenges the aluminum alkyl and electron donor complexes to levels conducive for chromium catalyst initiation and good reactor operability. However, commercial experience with this complex transition has not been reproducible, and periodically, this transition mechanism results in reactor shutdowns. Systematic signs of reactor operability problems, observed when using this procedure, have been failure of the transition due to static induced sheeting, high levels of static cold banding during silica injection leading to bed segregation and sheeting; additional off grade due to gels caused by the silica, and temperature excursions within the fluidized bed, evidenced by thermocouple activity and plate thermowell activity upon reaction initiation, leading to sheeting and agglomerate formation. In addition, the use of silica as an adsorbent of the TEA1/THF complex does not completely bind all of the alkyl in the reaction system, and the remaining small amounts of aluminum alkyls can result in extremely high chromium activity that can also lead to reactor operability problems.

Akzo Nobel has shown that residual metal alkyls can be deactivated using gaseous carbon dioxide (CO₂). The CO₂ is advantageous, in that, as a gas, it can permeate into all the reactor interstices, including the nozzles, cycle piping, compressor, plate, and other parts. In addition, the CO₂ will not condense on any of the reactor surfaces, which would results in difficult purging of the material. While the CO₂ is efficient at scavenging the aluminum alkyl cocatalyst from the reaction system, it does not affect free electron donor, unbound to aluminum alkyl.

U.S. Patent Nos. 5,442,019; 5,672,665; 5,747,612 and 5,753,786 disclose processes for transitioning between a Ziegler-Natta catalysts system to a metallocene catalyst system. These transitions primarily operate by discontinuing the introduction of the Ziegler-Natta catalyst, typically by initially reducing TEAL (triethylaluminum) concentration, into the reactor, followed by the introduction of a catalyst killer. See also U.S. 5,672,666, in which a deactivating agent is used to inhibit catalyst activity. U.S. Patent 4,460,755 discloses a method of converting an olefin polymerization catalyzed by a Ziegler-type catalyst to a polymerization catalyze by a chromium-based catalyst, which involves adding a hydroxyl-containing compound to the reactor containing the Ziegler-type catalyst. These transitions primarily operate by discontinuing the introduction of the components of the Ziegler catalyst system into the polymerization reactor, and adding a hydroxyl-containing silica that reacts with and absorbs components of the Ziegler catalyst system. U.S. Patent 4,875,941 discloses a method for the treatment of organometallic-contaminated equipment with gaseous carbon dioxide.

International Publication No. WO 2004/060929 discloses processes for transitioning between incompatible polymerization catalysts, and more specifically between polymerization reactions using silyl-chromate catalyst systems and mixed metallocene/Ziegler Natta or metallocene catalyst systems. These transitions primarily operate by discontinuing the introduction of the first catalyst system into the reactor, followed by maintaining polymerization conditions before introducing the second catalyst system. International Publication No. WO 2004/060921 discloses processes for transitioning among polymerization reactions using Ziegler-Natta catalyst systems, metallocene catalyst systems and chromium-based catalyst systems. These transitions primarily operate by first discontinuing the introduction of the first catalyst system into the reactor, followed by lowering the height of the bed of polymer particles in the reactor. International Publication No. WO 2004/085488 discloses a method for transitioning catalysts, in which a catalyst killer is introduced into the polymerization reactor in an amount sufficient to terminate the first polymerization reaction, and then introducing a second catalyst system to the polymerization reactor in the presence of a portion of the catalyst killer, which is present in an amount sufficient to activate the second catalyst system.

International Publication No. WO 2004/060922 discloses a process for transitioning from a first polymerization reaction to a second polymerization reaction, in which each reaction uses a catalyst system incompatible with the other, and where the transition involves conducting multiple gas phase polymerization reactions, and forming a substantially contaminant-free seedbed. International Publication No. WO 2004/060938 discloses a process for transitioning from a first gas phase polymerization catalyst system to a second polymerization catalyst system, incompatible with the first, and wherein the transition involves removing the contents of the first polymerization reaction from the gas-phase reactor, while maintaining a closed system, and introducing a substantially contaminant free seedbed into the gas-phase reactor. International Publication No. WO 00/58377 discloses a process for changing between two incompatible polymerization catalysts, which involves stopping the first polymerization, removing substantially all of the polymer from the reactor, rapidly purging with nitrogen, and adding a seedbed of polymer particles to the reactor. There is a need for an efficient process and less energy intensive process for converting a Ziegler-type catalyst system to another type of catalyst system, which effectively removes all catalyst poisons associated with the Ziegler-type system. Such a process should consume less time, energy and cost, as compared to current processes. These and other issues have been satisfied by the following invention.

SUMMARY OF THE INVENTION

In accordance with the present invention, it has now been found that a continuous olefin polymerization reaction, catalyzed by a Ziegler-type catalyst, containing a transition metal halide, an electron donor and an activator (or cocatalyst) component, can be converted effectively into a reaction, catalyzed by another catalyst system (second catalyst), such as a chiOmium-based catalyst, without the need of emptying and re-charging the polymerization reactor, and in substantially shorter periods of time than previously required. The conversion is effectuated by one of the following processes A), B) or C): A) (1) reducing catalyst inventory in a reactor,

(2) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor,

(3) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and

(4) purging the reactor with an inert gas; or B) (1) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into a polymerization reactor,

(2) reducing catalyst inventory in the reactor, (3) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and

(4) purging the reactor with an inert gas; or

C) (1) reducing catalyst inventory in a reactor,

(3) further reducing catalyst inventory in the reactor,

(4) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and

(5) purging the reactor with an inert gas; and where for each of the three processes, A), B) and C), the catalyst inventory comprises the Ziegler-type transition metal halide and the electron donor used for the Ziegler-type catalyst; and the reversible and irreversible poisons will deactivate an activator (or cocatalyst) component of the Ziegler-type catalyst

Each process A), B) and C) is specifically disclosed herein. It is noted that for each process, a reversible poison or poisons may be used, or an irreversible poison or poisons may be used, or combinations thereof may be used. All of these embodiments in regard to the use of these poisons are specifically disclosed herein.

The reduction in catalyst inventory is performed by one or more of the following steps: a) continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the comonomer to monomer ratio in the reactor; b) continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the monomer partial pressure in the reactor, c) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; d) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and decreasing the hydrogen to monomer ratio in the reactor; e) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and adjusting reaction temperature to increase catalyst activity; f) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing residence time; g) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the comonomer to monomer ratio in the reactor; h) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the monomer partial pressure in the reactor; i) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, increasing the comonomer to monomer ratio in the reactor and increasing the monomer partial pressure in the reactor; j) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and decreasing the hydrogen to monomer ratio in the reactor; k) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and adjusting reaction temperature to increase catalyst activity; and 1) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing residence time.

It is noted that the catalyst reduction process may take place using one step from steps a) through 1), or a combination of such steps. Thus, each of the above steps a) through 1) is specifically disclosed herein, and combinations of two or more of such steps are specifically disclosed herein. It is also noted that some polymerizations may require the use of more than one comonomer, and that such polymerizations are also covered in the inventive transition procedures disclosed herein.

In one aspect, for each process A), B), or C), at least one reversible poison is introduced into the reactor, hi a further aspect, the at least one reversible poison is carbon dioxide.

In another aspect, each process A), B)₅ or C), further comprises circulating the said poison or poisons within one or more auxiliary pieces of equipment, prior to purging the reactor with an inert gas. In a further aspect, the one or more auxiliary pieces of equipment are selected from the group consisting of one or more vent recovery streams, one or more product discharge systems, and combinations thereof. In yet another aspect, for each process A), B), or C)₅ after purging the reactor with an inert gas, at least one reversible poison and/or at least one irreversible poison is introduced into the reactor, and circulated within the reactor. In a further aspect, each process further comprises circulating said poison or poisons within one or more auxiliary pieces of equipment, prior to purging the reactor with an inert gas, and/or circulating said poison or poisons within one or more auxiliary pieces of equipment after purging the reactor with an inert gas. In yet a further aspect, the one or more auxiliary pieces of equipment are selected from the group consisting of one or more vent recovery streams, one or more product discharge systems, and combinations thereof.

As discussed above, the process of transition comprises one of the three processes. Thus, in one aspect of the invention the process for transitioning comprises process A). In another aspect, the process for transitioning comprises process B). In yet another aspect, the process for transitioning comprises process C). Each process, A), B) and C) of the invention may include one or more aspects and/or embodiments, all as described herein (in the prior and subsequent texts).

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic of a fluid bed reactor system that can be used to polymerize olefins. Figure 2 represents temperature (⁰C) versus time profiles of PI (Process Interface) data for Blank Isopentane for the First THF Injection.

Figure 3 represents temperature (⁰C) versus time profiles of PI data for the Second THF Injection. Figure 4 represents temperature (⁰C) versus time profiles of PI data for the First

UCAT™ A to UCAT™ G Transition.

Figure 5 represents potential at the reactor wall (volts) versus time profile of a static response of the CO₂ injection during First UCAT™ A to UCAT™ G Transition.

Figure 6 represents temperature (⁰C) versus time profiles of PI data for the Second UCAT™ A to UCAT™ G Transition.

Figure 7 represents potential at the reactor wall (volts) versus time profile of a static response of the CO₂ injection during the Second UCAT™ A to UCAT™ G Transition.

Figure 8 represents temperature (⁰C) versus time profiles of PI data for the Third UCAT™ A to UCAT™ G Transition.

Figure 9 represents potential at the reactor wall (volts) versus time profile of a static response of the CO₂ injection during the Third UCAT™ A to UCAT™ G Transition.

Figure 10 represents temperature (⁰C) versus time profiles of PI data for the Fourth UCAT™ A to UCAT™ G Transition.

Figure 11 represents potential at the reactor wall (volts) versus time profile of a static response of the CO₂ injection during the Fourth UCAT™ A to UCAT™ G Transition.

Figure 12 represents profiles for reaction rate trajectories for Transitions #2-4. Each profile represents a normalized polymerization rate (polymerization rate at time "t" divided by polymerization rate at "t_f (time final)").

Figure 13 represents temperature (⁰C) versus time profiles of PI data for the UCAT™ A to UCAT™ B Transition.

Figure 14 represents potential at the reactor wall (volts) versus time profile of a static response of the CO₂ injection during the UCAT™ A to UCAT™ B Transition. DETAILED DESCRIPTION OF THE INVENTION

The invention provides for a transition process from a Ziegler-Natta catalyst to another catalyst type. In this process, the introduction of the transition metal halide component of the Ziegler-type catalyst and the electron donor of the Ziegler-type catalyst into the reactor are discontinued, and the level of each of these components is reduced using a catalyst inventory reduction step or steps before and/or after the discontinuation of the Ziegler components. The catalyst inventory reduction is eventually followed by the introduction of at least one reversible poison, such as CO₂, and/or at least one irreversible poison into the reactor, and the circulation of the poison(s) within the reactor. Finally, the reactor is purged with an inert gas. Following this process, another catalyst system (or second catalyst system) is introduced into the reactor, and the second reaction is initiated. This second catalyst system may include any catalyst type which is susceptible to inactivation or excess activation by the activator or cocatalyst component of the Ziegler-type catalyst, such as triethylaluminum, and/or the electron donor component of the Ziegler-type catalyst, such as tetrahydrofuran.

The processes of the invention include process A), B) and C), as described below:

A) (1) reducing catalyst inventory in a reactor, (2) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor, (3) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and (4) purging the reactor with an inert gas; or

B) (1) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into a polymerization reactor, (2) reducing catalyst inventory in the reactor,

(3) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and (4) purging the reactor with an inert gas; or

C) (1) reducing catalyst inventory in a reactor,

(3) further reducing catalyst inventory in the reactor,

(4) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and (5) purging the reactor with an inert gas; and where for all three processes, A), B) and C), the catalyst inventory comprises the Ziegler-type transition metal halide and the electron donor used for the Ziegler-type catalyst; and the reversible and irreversible poisons are used to deactivate an activator (or cocatalyst) component of the Ziegler-type catalyst. Preferably, the transition process for each process A), B) and C) begins with the respective step (1) in each process.

The catalyst inventory reduction provides for a decrease in the levels of Ziegler- Natta transition metal halide component and Ziegler-Natta electron donor, remaining in the reactor, by one or more of the following steps: a) continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the comonomer to monomer ratio in the reactor; b) continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the monomer partial pressure in the reactor, c) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; d) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and decreasing the hydrogen to monomer ratio in the reactor; e) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and adjusting reaction temperature to increase catalyst activity; f) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing residence time; g) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the comonomer to monomer ratio in the reactor; h) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the monomer partial pressure in the reactor; i) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; j) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and decreasing the hydrogen to monomer ratio in the reactor; k) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and adjusting reaction temperature to increase catalyst activity; and

1) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing residence time. Each of these steps, or a combination of these steps, results in the complete or nearly complete consumption of the Ziegler-Natta transition metal halide remaining in the reaction system or auxiliary equipment. The catalyst inventory reduction also reduces the electron donor concentration in the reactor, such as tetrahydrofuran (THF), which is a poison to chromium catalysts and to metallocene catalysts. For each of the three processes A), B) and C), the cocatalyst feed is discontinued before the introduction of the at least one reversible poison and/or the at one irreversible poison into the reactor.

In one embodiment, the process comprises: (1) reducing catalyst inventory in the reactor, (2) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor, (3) further reducing catalyst inventory in the reactor, (4) introducing at least one reversible poison into the reactor, and circulating the poison within the reactor, and

(5) purging the reactor with an inert gas; and wherein, for the first and second catalyst inventory reduction steps, the catalyst inventory is reduced by continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or cocatalyst) component of the Ziegler-type system to the reactor, and increasing the comonomer to monomer ratio in the reactor; and where the activator (or cocatalyst) feed is discontinued into the reactor before the introduction of the reversible poison into the reactor. In another embodiment, the process comprises:

(1) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor,

(2) reducing catalyst inventory in the reactor,

(3) introducing at least one reversible poison into the reactor, and circulating the poison or poisons within the reactor, and

(4) purging the reactor with an inert gas; and wherein, the catalyst inventory is reduced by continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; and where the activator (or cocatalyst) feed is discontinued into the reactor before the introduction of the reversible poison into the reactor.

In another embodiment, for each process, A), B) and C), the hydrocarbon feed (monomer, optional comonomer and hydrogen) is continued after the discontinuation of the transition metal halide component and the electron donor component of the Ziegler- type catalyst into the reactor.

In another embodiment, for each process A), B), or C), at least one reversible poison is introduced into the reactor. In a further embodiment, the reversible poison is carbon dioxide.

In another embodiment, each process A), B), or C)₅ further comprises circulating said poison or poisons within one or more auxiliary pieces of equipment, prior to purging the reactor with an inert gas. In a further embodiment, the one or more auxiliary pieces of equipment are selected from the group consisting of one or more vent recovery streams, one or more product discharge systems, and combinations thereof. In yet a further embodiment, at least one reversible poison is introduced into the reactor, and preferably this reversible poison is carbon dioxide.

In another embodiment, for each process A), B)₅ or C), after purging the reactor with an inert gas, at least one reversible poison and/or at least one irreversible poison is introduced into the reactor and circulated within the reactor. In a further embodiment, said poison or poisons are circulated within one or more auxiliary pieces of equipment, prior to purging the reactor with an inert gas, and/or circulated within one or more auxiliary pieces of equipment after purging the reactor with an inert gas. In yet a further embodiment, the one or more auxiliary pieces of equipment are selected from the group consisting of one or more vent recovery streams, one or more product discharge systems, and combinations thereof. In a further embodiment, at least one reversible poison is introduced into the reactor. In yet a further embodiment, the reversible poison is carbon dioxide. The introduction and circulation of one or more poisons to the activator component to Zeigler-type catalyst, will result in the scavenging of residual cocatalyst, for example triethylaluminum (TEAl), remaining in the reaction system. For example, CO₂ is very effective in scavenging small amounts of residual TEAL Small amounts of TEAl, if left in the reactor, can lead to "hot spots" of activity upon the introduction of a chromium catalyst system, and this, in turn, can result in poor reactor operability.

In a preferred embodiment, carbon dioxide is used as a reversible poison. Other effective reversible poisons, include, but are not limited to, carbon monoxide, water and aliphatic alcohols, such as methanol, ethanol, propanol, isopropanol. Water can also be used as an irreversible poison, depending on the reaction conditions employed. Any compound or compounds sufficient to bind to, or scavenge, the activator (or cocatalyst) at issue, under the transitioning conditions at issue, may be used in the processes of the invention. Additional irreversible poisons include oxygen, glycols, phenols, ethers, carbonyl compounds, esters, amines, nitriles, nitrous compounds, carbonyl sulfide and mercaptans. Each poison may be used separately, or in combination with another suitable poison or poisons.

The purge removes any residual electron donor component of the Ziegler-type catalyst, such as, tetrahydrofuran (THF), from the reactor. The purge can also remove the activator poison, such as CO₂, from the reactor. In a Ziegler-type catalyst that uses TEAl (triethylaluminum) and THF, it is noted that once the TEAl has been removed from the reactor bed, the unbound THF remaining, and can be removed from the reactor by purging. In addition, THF has a normal boiling point of 66°C, making it relatively easier to purge; thus, THF can be purged from a reactor in a reasonable time frame (hours not days).

The pressure of the inert purge gas and the duration of the purge must each be of sufficient amount to effectively reduce residual amount of catalyst electron donor and CO₂ in the reactor. These parameters are chosen based on the catalyst systems being transitioned, the catalyst inventory reduction steps employed, the purge gas, and the reactor conditions. Typically, for a transition from a Ziegler-type catalyst system to a chromium-based catalyst system, a gas change out of preferably less than 30 reactor volumes, more preferably less than 20 reactor volumes, and most preferably between 10 and 15 reactor volumes, is sufficient to reduce THF and CO₂ to levels that will not inhibit an operable reaction by the second catalyst system.

In one embodiment, the purging step is performed using a flow purge, at moderate reaction pressures sufficient to operate the cycle gas compressor. In another embodiment, the purging step is performed by pressure purging the reactor.

Once the reactor has been purged, the introduction of the second catalyst system into the reactor provides for a polymerization that initiates smoothly, typically within 4 hours, usually within 2 hours. Transition periods from the first catalyst system to the second catalyst system will depend on the catalysts systems employed, the reactor size, reactor configuration, and feed and purge capacities, such as nitrogen capacity. Typically, transitions from a Ziegler-type catalyst system to a chromium-based catalyst system may occur in 36 hours or less, and preferably in 24 hours or less. The change from a substantially inert environment to a reactive environment is known as "building conditions," or a "condition build." Once satisfactory polymerization conditions are achieved, the second catalyst system can be added to the reactor.

In a gas phase reactor, during the transition, the level of the polymer bed remains constant or relatively constant. In addition, in a preferred embodiment of the invention, the reaction system is closed and the polymer bed is neither removed nor lowered at any time. While in the preferred embodiment, the processes of the invention specifically addresses the transition between a Ziegler-type catalyst system to a chromium-based catalyst system, it is within the scope of the invention that the processes apply to any transition from a Ziegler-type catalyst to another type catalyst system. For example, transitioning from a Ziegler-type catalyst to a metallocene catalyst system, or a constrained geometry catalyst, or transitioning from a two incompatible Ziegler-Natta catalysts, are also within the scope of the invention. In these cases, the first catalyst system is a poison(s) to the second catalyst system, and the poison(s) of the first catalyst system can be effectively removed from the reactor. The inventive process may also be use to effect the transition of one or more

Ziegler-type catalysts to one or more chromium-type catalysts or one or more metallocene-type catalysts or one or more constrained geometry catalysts. The process may also be used in the transition from one or more Ziegler-type catalysts to one or more other Ziegler-type catalysts. In this situation, the one or more poisons from the first Ziegler catalyst system, containing one or more of such catalysts, inhibit reaction of one or more catalysts from the second catalyst system, and the one or more poisons of the first catalyst system can be effectively removed from the reactor.

It has been found that the processes of the invention can be preformed within 24 hours or less. It has also been found that the catalyst inventory reduction steps increase catalyst productivity, and that such steps can effectively reduce catalyst components within a relatively short period of time, for example, within 2 to 4 hours. After such time, lower amounts of catalyst poison(s) and shorter purge periods are needed for the transitioning to the next catalyst system, in comparison to a process that does not perform a catalyst inventory reduction step. The processes of the invention also reduce the formation of high molecular weight gels that typically form in the transition from a Ziegler-type catalyst to a chromium based catalyst.

The processes of the invention eliminate the issues associated with the silica injection, as discussed above, and also significantly reduce the amount of time to transition from an initial Ziegler-type catalyst to a second catalyst system. The processes also provide increased flexibility to produce Ziegler-type catalyzed resins and chromium-type catalyzed resins on the same reactor train, while reducing potential reactor downtime, reducing off grades, reducing high molecular weight gels, and increasing profitability.

Representative catalysts for each category are discussed below. Additional catalyst structures for each category are also included within the scope of the invention.

Ziegler-Type Catalysts

Ziegler type catalysts are typified by a transition metal halide supported on magnesium chloride. Generally, a procatalyst; that is a formulation which requires chemical activation to become an active polymerization catalyst, can be described by the generic formulation:

(MCl₂)_xTi(OR)_aCly(ED)_z ,

where: x typically is >1 and < 100, a is >0 and <2, a+y is 3 or 4, z is from 0 to 10 or more,

ED is an electron donor,

M is a transition metal, and

R is an aliphatic or aromatic radical.

Although other transition metals, such as vanadium, zirconium and hafnium, can be used in these catalysts, mixtures of transition metals can also be utilized, such as Ti and Hf, Ti and V, Ti and Zr. However, titanium is the predominant transition metal utilized in the vast majority of commercial Ziegler type catalysts. The "procatalyst" is then further activated by contact with an aluminum compound with reducing power (for example, compounds such as AlR_nCl_3-11, where R is alkyl group of 1 to 20 carbons, and n ranges from 1.5 to 3). The most typically used aluminum compounds are trialkyl aluminums and dialkyl aluminum halides. This activation may be partial, prior to introduction into the reactor, or completed fully within the polymerization reactor.

The catalyst may be supported further on a substrate, prepared by precipitation, crystallization or even spray dried, however they all share similar chemistry.

Examples of procatalyst compositions and methods of making such procatalyst compositions are described in: U.S. Patent Nos. 5,487,938; 5,290,745; 5,247,032; 5,247,031; 5,229,342; 5,153,158; 5,151,399; 5,146,028; 5,106,806; 5,082,907; 5,077,357; 5,066,738; 5,066,737;5,034,361; 5,028,671; 4,990,479; 4,927,797; 4,829,037; 4,816,433; 4,547,476; 4,540,679; 4,460,701; 4,442,276; the entire contents of each of these patents are incorporated herein, in their entirety, by reference.

Most preferred transition metal compounds are titanium halides and haloalcoholates having 1 to 8 carbon atoms per alcoholate group. Examples of such compounds include: TiCl₄, TiBr₄, TiI₄, TiCl₃, Ti(OCH₃)Cl₃, Ti(OC₂H₅)Cl₃, Ti(OC₄H₉)Cl₃, Ti(OC₆H₅)Cl₃, Ti(OC₆H₁₃)Br₃, Ti(OC₈H₁₇)Cl₃, Ti(OCH₃)₂Br₂, Ti(OC₂Hs)₂Cl₂, Ti(OC₆H₁₃)₂Cl₂, Ti(OC₈H₁₇)₂Br₂, Ti(OCH₃)₃Br, Ti(OC₂Hs)₃Cl, Ti(OC₄Hg)₃Cl, Ti(OC₆H₁₃)₃Br, and Ti(OC₈Hn)₃Cl. Mixtures of titanium compounds can be employed if desired.

The magnesium compounds include magnesium halides, such as MgCl₂, MgBr₂, and MgI₂ . Anhydrous MgCl₂ is a preferred compound.

The electron donor is an organic Lewis base, liquid at temperatures in the range of 0°C to 200°C, in which the magnesium and titanium compounds are soluble. The electron donor can be an alkyl ester of an aliphatic or aromatic carboxylic acid, an aliphatic ketone, an aliphatic amine, an aliphatic alcohol, an alkyl or cycloalkyl ether, or mixtures thereof, each electron donor having 2 to 20 carbon atoms. Among these electron donors, the preferred are alkyl and cycloalkyl ethers having 2 to 20 carbon atoms; dialkyl, diaryl, and alkylaryl ketones having 3 to 20 carbon atoms; and alkyl, alkoxy, and alkylalkoxy esters of alkyl and aryl carboxylic acids having 2 to 20 carbon atoms. The most preferred electron donor is tetrahydrofuran. Other examples of suitable electron donors are methyl formate, ethyl acetate, butyl acetate, ethyl ether, dioxane, di-n-propyl ether, dibutyl ether, ethanol, 1-butanol, ethyl formate, methyl acetate, ethyl anisate, ethylene carbonate, tetrahydropyran, and ethyl propionate. While an excess of electron donor is used initially to provide the reaction product of titanium compound and electron donor, the reaction product finally contains 1 to 20 moles of electron donor per mole of titanium compound, and preferably 1 to 10 moles of electron donor per mole of titanium compound.

The procatalyst activator compound used in the partial pre-activation step can be one compound or a mixture of two different compounds. Each compound can have the formula M(Rn )X (3-n) wherein M is Al or B; each X is, independently, chlorine, bromine, or iodine; each R is, independently, a saturated aliphatic hydrocarbon radical having 1 to 14 carbon atoms, provided that when M is Al, n is 1 to 3, and when M is B, n is 0 to 1.5. Examples of the R radical are methyl, ethyl, n-butyl, isobutyl, n-hexyl and n-octyl. Examples of n when M is aluminum, are 1 , 1. 5, 2 and 3. Examples of n when M is boron, are 0, 1 or 1.5. Preferred activator compounds include diethyl aluminum chloride, triethyl aluminum, tri-n-hexyl aluminum, dimethyl aluminum chloride, and tri-n-octyl aluminum. Particularly preferred activator compounds are: a sequential mixture of tri-n-hexylaluminum and diethylaluminum chloride; a sequential mixture of triethylaluminurn and diethylaluminum chloride; a sequential mixture of diethylaluminum chloride and tri-n-hexylaluminum; a sequential mixture of diethylaluminum chloride and triethylaluminum; and either diethyl aluminum chloride or tri-n-hexyl aluminum.

The partially activated catalyst can function as a polymerization catalyst, but at greatly reduced and commercially unsuitable catalyst productivity. Complete activation in the polymerization reactor by additional cocatalyst is required to achieve full activity. Alternately, the catalyst may be fully activated in the polymerization reactor.

The cocatalyst, generally a hydrocarbyl aluminum cocatalyst, can be represented by the formula R₃Al or R₂AIX, wherein each R is independently alkyl, cycloalkyl, aryl, or hydrogen; at least one R is hydrocarbyl; and two or three R radicals can be joined to form a heterocyclic structure. Each R, which is a hydrocarbyl radical, can have 1 to 20 carbon atoms, and preferably has 1 to 10 carbon atoms. X is a halogen, preferably chlorine, bromine, or iodine. Examples of hydrocarbyl aluminum compounds are as follows: triisobutylaluminum, tri-n-hexylaluminum, di-isobutyl-aluminum hydride, dihexylaluminum hydride, di-isobutyl-hexylaluminum, isobutyl dihexylaluminum, trimethylaluminum, triethylaluminum, tripropylaluminum, triisopropylalummum, tri-n- butylaluminum, trioctylaluminum, tridecylaluminum, tridodecylaluminum, tribenzylaluminum, triphenylaluminum, trmaphthylaluminum, tritolylaluminum, dibutylaluminum chloride, diethylaluminum chloride, diisobutylaluminum chloride, and ethylaluminum sesquichloride. Cocatalyst is introduced in a sufficient amount to provide, in addition to the molar ratio of activator to titanium mentioned above, a cocatalyst to titanium mole ratio of 10 : 1 to 100 : 1 , preferably 20 : 1 to 50 : 1. This amount will complete the activation of the precursor. In those cases where it is desired to support the procatalyst, silica is the preferred support. Other suitable supports are inorganic oxides, such as aluminum phosphate, alumina, silica/alumina mixtures, silica modified with an organoaluminum compound, such as triethylaluminum, and silica modified with diethyl zinc. A typical support is a solid, particulate, porous material, essentially inert to the polymerization. It is used as a dry powder having an average particle size of 10 to 250 microns, and preferably 30 to 100 microns; a surface area of at least 200 square meters per gram, and preferably at least 250 square meters per gram; and a pore size of at least 100 angstroms and preferably at least 200 angstroms. Generally, the amount of support used, is that which will provide 0.1 to 1.0 millimole of titanium per gram of support, and preferably 0.4 to 0.9 millimole of titanium per gram of support. Impregnation of the above mentioned catalyst precursor into a silica support can be accomplished by mixing the precursor and silica gel in the electron donor solvent, or other solvent, followed by solvent removal under reduced pressure. The resultant solid catalyst may then be converted into a free flowing slurry with an aliphatic hydrocarbon as described in U.S. Pat. No. 5,290,745 and European Patent Application 771 820, or may be used as a dry powder. Both of the patent and application are incorporated in their entirety, herein by reference.

As noted, the precursor may be partially activated before polymerization. Activation is completed in the reactor via the cocatalyst. The cocatalyst is preferably added separately neat, or as a solution in an inert solvent, such as isopentane, to the polymerization reactor at the same time as the flow of ethylene is initiated.

These catalysts contain significant amounts of organic species which are bound to the catalyst residues, as well as significant quantities of halogens, particularly chlorides which can cause mold staining and corrosion. The organics must be deactivated and purged out of the polymer before the polymer is suitable for commercial applications. The halogen species likewise need to be sequestered, or rendered inactive, by various additives such as calcium or zinc stearate, zinc oxide . Water is frequently added to the polymers produced using these catalyst systems to react with the organic residues of the catalyst system and any reactive halogens (that is, free TiCIx for example) to allow these materials to be purged from the polymer prior to use. Removal of these air and water reactive residues to levels on the order of 100 ppm, remaining in the polymer, is required by governmental regulations.

Chromimn-Type Catalysts

Both chromium oxide based catalysts and silyl chromate based catalysts are useful in the invention.

The chromium oxide catalysts may be CrO₃, or any compound convertible to CrO₃ under the activation conditions employed. Compounds convertible to CrO₃ are disclosed in USP Nos. 2,825,721; 3,023,203; 3,622,251; and, 4,011,382 (the entire disclosures of which patents are incorporated herein by reference), and include chromic acetyl acetone, chromic chloride, chromic nitrate, chromic acetate, chromic sulfate, ammonium chromate, ammonium dichromate, or other soluble salts of chromate.

The silyl chromate catalysts are characterized by the presence of at least one group of Formula I below:

wherein R, in each occurrence, is a hydrocarbyl group having from 1 to 14 carbon atoms. Among the preferred compounds, having the group of Formula I are the bis- trihydrocarbylsilylchromates of Formula II below:

R O R

R- -Si- -O- -Cr- -O- -Si- -R (H)

R O R

where R is defined as above. R can be any hydrocarbon group, such as an alkyl, alkaryl, aralkyl or an aryl radical containing from 1 to 14 carbon atoms, preferably from 3 to 10 carbon atoms. Illustrative thereof, are methyl, ethyl, propyl, iso-propyl, n-butyl, iso-butyl, n-pentyl, iso-pentyl, t-pentyl, hexyl, 2-methyl-pentyl, heptyl, octyl, 2- ethylhexyl, nonyl, decyl, hendecyl, dodecyl, tridecyl, tetradecyl, benzyl, phenethyl, p methyl-benzyl, phenyl, tolyl, xylyl, naphthyl, ethylphenyl, methylnaphthyl, dimethylnaphthyl, . Illustrative of the preferred silylchromates, but by no means exhaustive or complete of those which can be employed in this process, are such compounds as bis-trimethylsilylchromate, bis-triethylsilylchromate, bis- tributylsilylchromate, bis-triisopentylsilylchromate, bis-tri-2-ethylhexylsilylchromate, bis-tridecylsilylchromate, bis-tri(tetradecyl)silylchromate, bis-tribenzylsilylchromate, bis-triphenethylsilylchromate, bis-triphenylsilylchromate, bis-tritolylsilylchromate, bis- trixylylsilylchromate, bis-trinaphthylsilylchromate, bis-triethylphenylsilylchromate, bis- trimethylnaphthylsilylchromate, polydiphenylsilylchromate, polydiethylsilylchromate . Examples of such catalysts are disclosed, for example, in USP Nos. 3,324,101; 3,704,287; and 4,100,105, the disclosures of which patents are incorporated herein by reference in their entireties.

The chromium based catalysts of the current invention are deposited onto conventional catalyst supports or bases, for example, inorganic oxide materials. The inorganic oxide materials which may be used as a support, in the catalyst compositions of the present invention, are porous materials, having a high surface area, for example, a surface area in the range of 50 to 1000 square meters per gram, and a particle size of 20 to 200 microns. The inorganic oxides which may be used, include silica, alumina, thoria, zirconia, aluminum phosphate and other comparable inorganic oxides, as well as mixtures of such oxides. Where both chromium oxide-based catalysts and silylchromate-based catalysts are employed together in this invention, each catalyst is deposited on a separate support.

Chromium Oxide Based Catalysts

Processes for depositing the chromium species on supports are known in the art, and may be found in the previously incorporated U.S. patents. The chromium compound is usually deposited on the support from solutions thereof, and in such quantities as to provide, after the activation step, the desired levels of chromium in the catalyst. Modifying materials, such as titanium and fluoride, are generally added prior to the activation. After the compounds are placed on the supports, and are activated, there results a powdery, free-flowing particulate material.

Generally, catalysts are prepared by using commercially available silica, to which a chrome source has been added. The silica substrate may be treated with a titanium ester (titanium tetraisopropylate or titanium tetraethoxide are typically used) either, after the Cr compound is deposited, or prior to this deposition. The support is generally pre-dried at 150-200⁰C to remove physically adsorbed water. The titanate may be added as a solution to a slurry of the silica in isopentane solvent or directly into a fluidized bed of support. If added in slurry form, the slurry is dried. Generally, the Cr compound, which is convertible to Cr+6, has already been added to the support. The support is then converted into active catalyst by calcination in air, at temperatures up to 1000°C.

During activation, the titanium is converted to some type of surface oxide. The chromium compound (generally chromium (III) acetate) is converted to a Cr+6 oxide of some kind. Fluoriding agents may also be added during the activation process to selectively collapse some pores in the support, modifying the molecular weight response of the catalyst. The activated catalyst may also be treated with reducing agents prior to use, such as carbon monoxide in a fluidized bed, or other reducing agents, such as aluminum alkyls, boron alkyls, lithium alkyls . Catalysts of this type are described in numerous patents, such as WO

2004/094489, EP,0640625, USA 100, 105, and the references cited within these references. Each of these references is incorporated herein, in its entirety, by reference. For example, a useful catalyst is a supported chromium-titanium catalyst (or titanated chrome oxide catalyst) which is substantially non-spherical or irregular in shape, and has a broad particle size distribution, with at least 75 percent of its pore volume ranging in pore size from 200 to 500 Angstroms. Activation of the supported chromium oxide catalyst can be accomplished at nearly any temperature up to about its sintering temperature. The passage of a stream of dry air or oxygen through the supported catalyst during the activation aids in the displacement of the water from the support and converts, at least partially, chrome species to Cr⁺6. Activation temperatures from 300⁰C to 900⁰C, for periods from greater than 1 hour to as high as 48 hours, are acceptable. Well dried air or oxygen is used and the temperature is maintained below the sintering temperature of the support.

Preferred conditions utilize a temperature from 300°C to 900°C, preferably from 700⁰C to 850⁰C, for at least two hours, preferably from 5 hours to 15 hours. The chromium compound, titanium compound and fluorine compound, if used, are deposited on the support, in such quantities as to provide, after the activation step, the desired levels of chromium, titanium and fluorine in the catalyst.

Preferred Compounds Preferred chromium compounds which may be used, include CrOs, or any compound of chromium which is ignitable to CrO₃, under the activation conditions employed. Chromium compounds other than CrO₃ which may be used, are disclosed in U.S. Pat. Nos. 2,825,721 and 3,622,521 (the disclosures of these patents are incorporated herein, in their entirety, by reference), and include chromic acetyl acetonate, chromic nitrate, chromic acetate, chromic chloride, chromic sulfate, and ammonium chromate.

Water soluble compounds of chromium, such as CrO₃, are the preferred compounds for use in depositing the chromium compound on the support from a solution of the compound. Organic solvent soluble chromium compounds may also be used.

The titanium compounds which may be used, include all those which are ignitable to TiO₂ under the activation conditions employed, especially those disclosed in U.S. Pat. Nos. 3,622,521 and 4,011,382 (the disclosures of these patents are hereby incorporated by reference). These compounds include those having the structures (R%Ti(OR')_m

(RO)_mTi(OR')_n and

TiX₄ where m is 1, 2, 3 or 4; n is 0, 1, 2 or 3; and m + n = 4;

R is a C₁ to C₁₂ alkyl, aryl or cycloalkyl group, or combinations thereof, such as aralkyl, alkaryl, ;

R¹ is R, cyclopentadienyl, or C₂ to C₁₂ alkenyl groups, such as ethenyl, propenyl, isopropenyl, butenyl ; and X is chlorine, bromine, fluorine or iodine.

The titanium compounds would thus include titanium tetrachloride, titanium tetraisopropoxide and titanium tetrabutoxide. The titanium compounds are conveniently deposited on the support from a hydrocarbon solvent solution thereof. The titanium (as Ti) is present in the catalyst, with respect to the Cr (as Cr), in a mole ratio of 0 to 180, and preferably of 4 to 35.

The fluorine compounds which may be used, include HF, or any compound of fluorine which will yield HF under the activation conditions employed. Fluorine compounds other than HF which may be used, are disclosed in U.S. Pat. No. 4,011,382. These compounds include ammonium hexafluorosilicate, ammonium tetrafluoroborate, and ammonium hexafluorotitanate. The fluorine compounds are conveniently deposited on the support from an aqueous solution thereof, or by dry blending the solid fluorine compounds with the other components of the catalyst prior to activation.

The support employed for the catalyst are porous, inorganic oxide materials, having a high surface area, that is, a surface area in the range of 50 to 1000 square meters per gram, and an average particle size of 10 to 200 microns. The inorganic oxides which may be used, include silica, alumina, thoria, zirconia and other comparable inorganic oxides, as well as mixtures of such oxides. Silica, silica alumina, aluminum phosphate, silica titania and silica aluminum phosphates are preferred support compounds. Particularly preferred supports are microspheroidal particles of surface area 200 to 500 square meters per gram, a pore diameter of 100 to 500 Angstroms, and an average particle size of 20 to 100 microns (for example, Grade 952 MS, 957HS, 957 silica available from Davison Chemical Division, W. R. Grace and Company, and Ineos EP30X, Ineos EP30XA available from Ineos Corporation and similar silica grades available from Philadelphia Quartz).

Activation of the supported catalyst can be accomplished at nearly any temperature, up to about its sintering temperature. The passage of a stream of dry air or oxygen through the supported catalyst during the activation, aids in the displacement of the water from the support. Normally, the activated catalyst component is employed in the reactor in an amount of from 0.005 weight percent to 0.2 weight percent of the weight of polymer produced. Silyl Chromate Catalysts

wherein R, in each occurrence, is a hydrocarbyl group having from 1 to 14 carbon atoms. Among the preferred compounds having the group of Formula I are the bis- trihydrocarbylsilylchromates of Formula II below:

O R

R- -Sl- -O- -O- -Si- -R (H)

R O R

where R is defined as above. R can be any hydrocarbon group such as an alkyl, alkaryl, aralkyl or an aryl radical containing from 1 to 14 carbon atoms, preferably from 3 to 10 carbon atoms. Illustrative thereof, are methyl, ethyl, propyl, iso-propyl, n-butyl, iso- butyl, n-pentyl, iso-pentyl, t-pentyl, hexyl, 2-methyl-pentyl, heptyl, octyl, 2-ethylhexyl, nonyl, decyl, hendecyl, dodecyl, tridecyl, tetradecyl, benzyl, phenethyl, p-methyl- benzyl, phenyl, tolyl, xylyl, naphthyl, ethylphenyl, methylnaphthyl, dimethylnaphthyl, .

Illustrative of the preferred silylchromates, but by no means exhaustive or complete of those which can be employed in this process, are such compounds as bis- trimethylsilylchromate, bis-triethylsilylchromate, bis-tributylsilylchromate, bis- triisopentylsilylchromate, bis-tri-2-ethylhexylsilylchromate, bis-tridecylsilylchromate, bis-tri(tetradecyl)silylchromate, bis-tribenzylsilylchromate, bis- triphenethylsilylchromate, bis-triphenylsilylchromate, bis-tritolylsilylchromate, bis- trixylylsilylchromate, bis-ttinaphthylsilylchromate, bis-triethylphenylsilylchromate, bis- trimethylnaphthylsilylchromate, polydiphenylsilylchromate, polydiethylsilylchromate . Especially preferred compounds are bis-triphenylsilylchromate, bis-tritolylsilylchromate and bis-triadamantylsilylchromate. In order to be an active polymerization catalyst, the silylchromate must be supported on an inorganic support followed by an optional, but preferred reduction reaction, to convert some (or all) of the Cr+6 species to lower valence states.

The support employed for the catalyst are porous, inorganic oxide materials, having a high surface area, that is, a surface area in the range of 50 to 1000 square meters per gram, and an average particle size of 10 to 200 microns. The inorganic oxides which may be used, include silica, alumina, thoria, zirconia and other comparable inorganic oxides, as well as mixtures of such oxides. Silica, silica alumina, aluminum phosphate, silica titania and silica aluminum phosphates are preferred support compounds. Particularly preferred supports are microspheroidal particles of surface area 200 to 500 square meters per gram, a pore diameter of 100 to 300

Angstroms, and an average particle size of 20 to 100 microns (for example, Grade 952 MS, 955 silica available from Davison Chemical Division, W. R. Grace and Company, and Ineos EP30 available from Ineos Corporation and similar silica grades available from Philadelphia Quartz.) The support employed for the catalyst must be partially dehydrated prior to attempting to support the silylchromate. The partial dehydration is typically carried out in a fluidized bed dehydrator, using nitrogen or air as the fluidizing gas. Dehydration temperatures of 300⁰C to 800⁰C may be used. Dehydration takes place over a period of 1 to 48 hours, typically from 1.5 to 8 hours. Preferred dehydration temperatures range from 350 to 600⁰C.

The chromate compound is then deposited on the dehydrated support material via suitable methods. The most typical method is to place the dehydrated support in suspension in an inert aromatic or aliphatic hydrocarbon (such as isopentane, hexane, heptane, toluene, mixtures of hydrocarbons ) at temperatures and pressures such that the solvent remains a liquid, adding the chromate compound to the suspended support, and allowing the chromate compound to deposit on the support. It is theorized that the chromate compound reacts with the remaining surface hydroxyl groups on the support surface, leading to a supported cupport-chromate material. Typical reaction conditions comprise reaction temperatures from 10-100⁰C and reaction duration from 1 to 48 hours. Preferred temperature ranges are 35-8O⁰C₃ most preferred 35-6O⁰C. Preferred reaction times are 2 to 24 hours, more preferable 4 to 20 hours, most preferable 6 to 10 hours. Preferred solvents are isopentane, mixed pentane isomers, n-hexane, hexane isomers and heptane and heptane isomers. Most preferred solvents are saturated C5 and C6 hycrocarbons.

Following completion of the deposition reaction, the reaction product is further treated with an aluminum alkyl compound to partially or completely reduce the Cr+6 species to a lower valence state. This may be done immediately after the deposition reaction has completed, or at a later time, although sequential reaction is highly preferred. Suitable aluminum alkyl compounds are described by the formula:

AlCRV_xtfORV where x ranges from zero to 2; R¹ and R² may be the same or different, and are alkyl, alkaryl or aromatic radicals containing from 1 to 25 carbons.

Preferred R groups are methyl, ethyl, n-propyl, n-butyl, isobutyl, n-pentyl, n-hexyl, n-octyl, tolyl, and adamantyl. X is preferably 0 to 1.5, most preferably 1.0. The most preferred R group is ethyl.

The reduction reaction occurs, as indicated above, in a solvent which may be the same or different as that used for the deposition. Preferably, the same solvent is used and the reactions are sequential. Temperature and pressure are controlled to maintain the solvent in a liquid state, until drying of the suspension is begun. Typical reaction temperatures and times are 35 to 9O⁰C and 1 to 48 hows, respectively. Preferred reaction time is between 12 and 24 hours, at temperatures between 45 and 75⁰C. Following completion of the reduction reaction, the catalyst is dried under nitrogen or vacuum to a free flowing solid, and stored under essentially poison free nitrogen until use.

Single Site Catalysts Single site catalysts, such as metallocene catalysts and constrained geometry catalysts may be used in the practice of the invention. Generally, such catalyst compounds include half and full sandwich compounds having one or more π-bonded ligands including cyclopentadienyl-type structures or other similar functioning structure such as pentadiene, cyclooctatetraendiyl and irnides. Typical compounds are generally described as containing one or more ligands capable of π-bonding to a transition metal atom, usually, cyclopentadienyl derived ligands or moieties, in combination with a transition metal selected from Group 3 to 8, preferably 4, 5 or 6 or from the lanthanide and actinide series of the Periodic Table of Elements.

Some metallocene-type catalyst compounds are described in, for example, U.S. Patents: 4,530,914; 4,871,705; 4,937,299; 5,017,714; 5,055,438; 5,096,867; 5,120,867; 5,124,418; 5,198,401; 5,210,352; 5,229,478; 5,264,405; 5,278,264; 5,278,119; 5,304,614; 5,324,800; 5,347,025; 5,350,723; 5,384,299; 5,391,790; 5,391,789; 5,399,636; 5,408,017; 5,491,207; 5,455,366; 5,534,473; 5,539,124; 5,554,775; 5,621,126; 5,684,098; 5,693,730; 5,698,634; 5,710,297; 5,712,354; 5,714,427; 5,714,555; 5,728,641; 5,728,839; 5,753,577; 5,767,209; 5,770,753 and 5,770,664; European publications: EP-A-O 591 756; EP-A-O 520 732; EP-A-O 420 436; EP-A-O 485 822; EP-A-O 485 823 ; EP-A-O 743 324; EP-A-O 518 092; and PCT publications: WO 91/04257; WO 92/00333; WO 93/08221; WO 93/08199; WO 94/01471; WO 96/20233; WO 97/15582; WO 97/19959; WO 97/46567; WO 98/01455; WO 98/06759 and WO 98/011144. Additional metallocene-type catalysts are described in U.S. Patents 5,442,019 and 5,672,666. All of these references are incorporated herein, in their entirety, by reference.

Suitable catalysts for use herein, preferably include constrained geometry catalysts as disclosed in U.S. Patent Nos. 5,272,236; 5,278,272; and 5,132,380, which are each incorporated in their entirety by reference. The monocyclopentadienyl transition metal olefin polymerization catalysts taught in U.S. Patent No. 5,026,798, the teachings of which are incorporated herein by reference, are also suitable as catalysts of the invention.

The foregoing catalysts may, as described in U.S. 5,278,272, comprise a metal coordination complex comprising a metal of groups 3-10 or the Lanthanide series of the Periodic Table of the Elements, and a delocalized π- bonded moiety substituted with a constrain-inducing moiety, said complex having a constrained geometry about the metal atom, such that the angle at the metal between the centroid of the delocalized, substituted π- bonded moiety, and the center of at least one remaining substituent is less than such angle in a similar complex, containing a similar π- bonded moiety, lacking in such constrain-inducing substituent, and provided further that for such complexes comprising more than one delocalized, substituted x-bonded moiety, only one thereof, for each metal atom of the complex, is a cyclic, delocalized, substituted π- bonded moiety. The catalyst further comprises an activating cocatalyst.

Preferred catalyst complexes correspond to the structure I below:

Structure I .

In structure I, M is a metal of group 3-10, or the Lanthanide series of the Periodic Table of the Elements;

Cp* is a cyclopentadienyl or substituted cyclopentadienyl group bound in an η5 bonding mode to M; Z is a moiety comprising boron, or a member of group 14 of the Periodic Table of the Elements, 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;

X independently each occurrence is an anionic ligand group or neutral Lewis base ligand group having up to 30 non-hydrogen atoms; n is 0, 1, 2, 3, or 4 and is 2 less than the valence of M; and

Y is an anionic or nonanionic ligand group, bonded to Z and M, comprising nitrogen, phosphorus, oxygen or sulfur, and having up to 20 non-hydrogen atoms, optionally Y and Z together form a fused ring system.

More specific complexes are described in U.S. Patent 5,278,272, incorporated herein by reference.

Specific compounds include: (tert-butylamido) (tetramethyl-η5 - cyclopentadienyl)- 1,2- ethanediylzirconiurn dichloride, (tert-butylamido)(tetramethyl- η5 -cyclopentadienyl) 1 ,2- ethanediyltitanium dichloride, (methylamido)(tetramethyl-η5 - cyclopentadienyl)- 1,2- ethanediylzirconium dichloride, (methylamido) (tetramethyl-η5 cyclopentadienyl)-l,2-ethanediyltitanium dichloride, (ethylamido)(tetramethyl-η5 - cyclopentadienyty-methylenetitanium dichloro, (tertbutylamido)dibenzyl(tetramethyl-η5 -cyclopentadienyl) silanezirconium dibenzyl, (benzylamido)dimethyl(tetramethyl-η5- cyclopentadienyl)silanetitanium dichloride, (phenylphosphido) dimethyl(tetramethyl η5 -cyclopentadieny^silanezirconium dibenzyl, (tertbutylamido)dimethyl(tetramethyl-η5 - cyclopentadienyl) silanetitanium dimethyl.

The complexes may be prepared by contacting a derivative of a metal, M, and a group I metal derivative or Grignard derivative of the cyclopentadienyl compound in a solvent, and separating the salt byproduct. Suitable solvents for use in preparing the metal complexes are aliphatic or aromatic liquids, such as cyclohexane, methylcyclohexane, pentane, hexane, heptane, tetrahydrofuran, diethyl ether, benzene, toluene, xylene, ethylbenzene, etc., or mixtures thereof.

Suitable cocatalysts include polymeric or oligomeric aluminoxanes, especially methyl aluminoxane, as well as inert, compatible, noncoordinating, ion forming compounds. So-called modified methyl aluminoxane (MMAO) is also suitable for use as a cocatalyst. One technique for preparing such modified aluminoxane is disclosed in U.S. Pat. No. 5,041,584, the teachings of which are incorporated herein by reference. Aluminoxanes can also be made as disclosed in U.S. Pat. Nos. 5,542,199; 4,544,762; 5,015,749; and 5, 041,585, the entire specification of each of which is incorporated herein by reference. Preferred cocatalysts are inert, noncoordinating, boron compounds.

Polymerization Reactor and General Conditions

While transformation of a polymerization reaction catalyzed by a Ziegler-type catalyst system into another type of catalyst system may be effected in either gas phase, slurry or solution polymerizations, it is preferably effected in a gas phase polymerization, such as a stirred bed reactor or a gas fluidized bed reactor. A schematic of a typical reactor is provided in Figure 1. The reactor 1 consists of a reaction zone 2 and a velocity reduction zone 3. One skilled in the art will recognized that the dimensions and particular reactor configuration, and feed rates, etc., will vary, and will depend, in part, on the polymerization production scale and reaction components. The reaction zone 2 comprises a bed of growing polymer particles, formed polymer particles, and a minor amount of catalyst particles fluidized by the continuous flow of polymerizable and modifying gaseous components in the form of make-up feed and recycle gas through the reaction zone. To maintain a viable fluidized bed, the superficial gas velocity through the bed must exceed the minimum flow required for fluidization, and preferably is at least 0.2 feet per second above the minimum flow. Ordinarily the superficial gas velocity does not exceed 5.0 feet per second, and most usually no more than 2.5 feet per second is sufficient.

It is essential that the bed always contains particles to prevent the formation of localized "hot spots" and to entrap and distribute the particulate catalyst throughout the reaction zone. Fluidization is achieved by a high rate of gas recycle to and through the bed, typically on the order of about 50 times the rate of feed of make-up gas. The pressure drop through the bed is equal to, or slightly greater than, the mass of the bed divided by the cross-sectional area. It is thus dependent on the geometry of the reactor.

Make-up gas is fed to the bed at a rate equal to the rate at which particulate polymer product is withdrawn. The composition of the make-up gas is determined by a gas analyzer 5 positioned above the bed. The gas analyzer determines the composition of the gas being recycled and the composition of the make-up gas is adjusted accordingly to maintain an essentially steady gaseous composition within the reaction zone. To insure complete fluidization, the recycle gas, and, where desired, part of the make-up gas are returned over gas recycle line 6 to the reactor at point 7 below the bed. At this point there is a gas distribution plate 8 above the point of return to aid in fluidizing the bed.

The portion of the gas stream which does not react in the bed constitutes the recycle gas which is removed from the polymerization zone, preferably by passing it into a velocity reduction zone 3 above the bed, where entrained particles are given an opportunity to drop back into the bed.

The recycle gas is then compressed in a compressor 9 and passed through a heat exchanger 10 wherein it is stripped of heat of reaction before it is returned to the bed. The temperature of the bed is controlled at an essentially constant temperature under steady state conditions by constantly removing heat of reaction. The recycle is then returned to the reactor at its base 7 and to the fluidized bed through distribution plate 8. The compressor 9 can also be placed downstream of the heat exchanger 10. Catalyst is injected at point 13, and resin is discharged at the resin discharge port 14. There are typically two resin discharge ports per reactor. Components 15-17 are part of the resin discharge system, and components 18-21 represent a portion of the vent recovery system. Polymer product removed at valve 16 is transferred to post reaction, purging and compounding unit(s).

The distribution plate 8 plays an important role in the operation of the reactor. The fluidized bed contains growing and formed particulate polymer particles, as well as catalyst particles. As the polymer particles are hot and possible active, they must be prevented from settling; for if a quiescent mass is allowed to exist, any active catalyst contained therein may continue to react and cause fusion. Diffusing recycle gas through the bed at a rate sufficient to maintain fluidization throughout the bed is, therefore, important. The distribution plate 8 serves this purpose and may be a screen, slotted plate, perforated plate, a plate of the bubble cap type . The elements of the plate may all be stationary, or the plate may be of the mobile type disclosed in U.S. Pat. No.

3,298,792. Whatever its design, it must diffuse the recycle gas through the particles at the base of the bed to keep the bed in a fluidized condition, and also serve to support a quiescent bed of resin particles when the reactor is not in operation. The mobile elements of the plate may be used to dislodge any polymer particles entrapped in or on the plate.

Any gas, inert to the catalyst and reactants, may also be present in the gas stream. Hydrogen may be used as a chain transfer agent in amounts determined by the desired polymer composition and properties.

It is necessary to operate the fluid bed reactor at a temperature below the sintering temperature of the polymer particles to insure that sintering does not occur. While temperatures of from 30°C to 15O⁰C are suitable, temperatures of 60°C or 75°C to 115°C are preferred, and temperatures of 70°C or 80⁰C to 110°C are most preferred.

The fluid bed reactor may be operated at pressures of up to 3600 IcPa (IkPa = 0.145 psi), but is preferably operated at pressures of from 1700 kPato 3100 kPa, with operation at the higher pressures in such ranges favoring heat transfer, since an increase in pressure increases the unit volume heat capacity of the gas. Generally, temperatures from 70°C to 110°C and pressures from 1700 kPa to 3100 kPa may be employed for both Ziegler catalyzed polymerizations and those catalyzed with chromium-based catalysts.

The catalyst employed in the fluidized bed is preferably stored for service in a reservoir under an inert gas blanket, and is introduced into the reactor at a rate equal to its consumption. An inert gas can be used to carry the catalyst into the bed. Preferably, the catalyst is injected at a point in the bed where good mixing of polymer particles occurs. Injection into the viable bed aids in distributing the catalyst throughout the bed, and tends to preclude the formation of localized spots of high catalyst concentration which may result in the formation of "hot spots" .

The activator or cocatalyst component of a Ziegler-type polymerization, is preferably added to the reaction system downstream from heat exchanger. Thus, the activator component may be fed into the gas recycle system through line from a dispenser. The production rate of the reactor is controlled by the rate of catalyst injection and the partial pressure of monomer, for example, the partial pressure of ethylene (C2PP). The production rate may be increased by increasing the rate of catalyst injection and/or the partial pressure of monomer, and decreased by reducing the rate of catalyst injection and/or the partial pressure of monomer. It has also been found, as discussed above, that the catalyst productivity of a Ziegler-type catalyst can be increase by using one or more of the catalyst inventory reduction steps a) through 1) as discussed above.

The monomer composition and monomer concentration employed will, of course, depend upon the desired polymer composition and properties. Both catalysts types (first catalyst and second catalyst) of the transition are ordinarily employed to homopolymerize ethylene (or another olefin base), or copolymerize ethylene (or another olefin base) with at least one other alpha-olefm, which typically contain from 3 to 8 carbon atoms. Alpha-olefins include, but are not limited to, propylene, 1-butene, 1- pentene, 1-hexene, 1-heptene, 1-octene, 4-methyl-l-pentene, 1-decene and combinations thereof. Preferably, such alpha-olefins are employed in an amount sufficient to achieve a concentration of from 0 to 30 mol percent in the copolymer. DEFINITIONS

The terms "catalyst system," or "catalyst" as used herein, refer to the one or more components of a catalyst complex or catalyst mixture.

The term "polymer," as used herein, refers to a polymeric compound prepared by polymerizing monomers, whether of the same or a different type. The generic term polymer thus embraces the term homopolymer, usually employed to refer to polymers prepared from only one type of monomer, and the term interpolymer as defined hereinafter.

The term "interpolymer," as used herein, refers to polymers prepared by the polymerization of at least two different types of monomers. The generic term interpolymer thus includes copolymers, usually employed to refer to polymers prepared from two different types of monomers, and polymers prepared from more than two different types of monomers.

The term "inert gas," as used herein, refers to any gas inert to the catalyst and reactants at issue. Typically, such term refers to nitrogen and helium, but may also refer to unreactive aliphatic hydrocarbons.

The term "skin thermocouple," as used herein, refers to thermocouples placed at the wall of the reaction vessel.

The terms "static level" and "static pattern," as used herein, respectively refer to the static voltage in the reactor bed and the physical appearance of the static voltage trace.

The term "crossover," are used herein, refers to the time when the temperature of the inlet reactor equals (or crosses over) the temperature of the reactor bed.

TEST PROCEDURES

The density of the ethylene homopolymers and interpolymers is measured in accordance with ASTM D-792. Melt index (I₂) is measured in accordance with ASTM D- 1238, condition

190°C/2.16 kg (formerly known as "Condition (E)"). FI₂₁ measurements were made in accordance with ASTM D-1238, condition F (190°C/21.6 kg).

The molecular weight distribution of the ethylene interpolymers used in the present invention can be determined by gel permeation chromatography (GPC) using the following procedure.

The chromatographic system may consist either of a Polymer Laboratories Model PL-210 or a Polymer Laboratories Model PL-220. The column and carousel compartments are operated at 14O⁰C. The columns are three Polymer Laboratories 10- micron Mixed-B columns. The solvent is 1,2,4 trichlorobenzene. The samples are prepared at a concentration of 0.1 grams of polymer in 50 milliliters of solvent. The solvent used to prepare the samples contains 200 ppm of butylated hydroxytoluene (BHT). Samples are prepared by agitating lightly for 2 hours at 16O⁰C. The injection volume is 100 microliters and the flow rate is 1.0 milliliters/minute.

A fifth-order polynomial fit of the calibration of the GPC column set is performed with 21 narrow molecular weight distribution polystyrene standards with molecular weights ranging from 580 to 8,400,000, arranged in 6 "cocktail" mixtures with at least a decade of separation between individual molecular weights. The standards are purchased from Polymer Laboratories (UK). The polystyrene standards are prepared at 0.025 grams in 50 milliliters of solvent for molecular weights equal to or greater than 1,000,000, and 0.05 grams in 50 milliliters of solvent for molecular weights less than 1,000,000. The polystyrene standards are dissolved at 80 degrees Celsius with gentle agitation for 30 minutes. The narrow standards mixtures are run first, and in order of decreasing highest molecular weight component to minimize degradation. The polystyrene standard peak molecular weights are converted to polyethylene molecular weights using the following equation (as described in Williams and Ward, J.^'Polym. Sci., Polym. Let., 6, 621 (1968)):

Mpolyethylene = A x (Mpolystyrene)^B, where M is the molecular weight, A has a value of 0.4315 and B is equal to 1.0.

Polyethylene equivalent molecular weight calculations are performed using Viscotek TriSEC software Version 3.0.

All patents and applications cited herein are incorporated by reference. The processes of the invention are more fully described by the following examples, which are provided for the purpose of illustrating the invention, and are not to be construed as limiting the scope of the invention.

EXPERIMENTAL

Throughout these experiments the following catalyst systems were used. Each catalyst described by the designation UCAT is trademarked and is property of Union Carbide Corporation, subsidiary of The Dow Chemical Company.

UCAT™ A - Ziegler-Natta catalyst system (transition metal halide = titanium chloride complex, activator = TEAl, electron donor = THF) UCAT™ G - Silyl Chromate catalyst system UCAT™ B - chromium oxide catalyst system

The UCAT™ G catalyst was represented by the UCAT™ G 500 series catalyst producing a 0.953 g/cc density, 37 dg/min FI₂₁, Resin B (ethylene/ 1-hexene interpolymer).

The UCAT™ A catalyst was represented by the UCAT™ A 2000 series catalyst producing Resin A (ethylene/ 1-hexene interpolymer), 0.954 g/cc density, 20 dg/min I₂. The UCAT™ B catalyst was represented by the UCAT™ B 400 series catalyst producing a 0.920 g/cc density, 0.65 dg/min I₂, Resin C (ethylene/ 1-hexene interpolymer).

Reaction condition and product properties are summarized in Table 1.

Table 1 - Reaction Condition Summary

The following reactions were preformed in a gas fluidized bed reactor, similar to that as shown in Figure 1. The reactor dimensions were 0.337 meter in diameter, with a straight section height of 1.52 meter.

Experiment 1 - Severity of Tetrahydrofuran (THF) as Catalyst Poison

The severity of THF as a poison to UCAT™ G, and the ability to purge the THF out of the reactor to level at which UCAT™ G could sustain a reaction, were examined. To perform this test a sample cylinder was cleaned and filled with isopentane. To test the purity of the sample cylinder and the isopentane supply, prior to the start of the THF testing, a "blank" was prepared, by filling the cylinder with 13O g of purified isopentane. The cylinder was then connected to the reactor running UCAT™ G, via an injection tube at the three-foot level, and pressured into the reactor using 550 psig high- pressure high-purity nitrogen. Upon injection of the isopentane, the reaction rate was reduced by approximately 50 percent, but fully recovered in 30 minutes (see Figure 1). This indicated that impurities entered the reactor during the injection. Based on the quick recovery of the reaction rate and the installation set-up of the injection tubes, it was concluded that the reduction in the reaction rate was most likely due to air in the line, despite the nitrogen purge. There did not appear to be present a significant impurity in either the cylinder or isopentane supply. THF Injection Test #1

A reactor running high density injection molding resin (Resin A) typically has catalyst (Ziegler-Natta) productivity on the order of 3500-4000 pound resin per pound catalyst, based on residual Ti levels. Based on the composition of the catalyst, this productivity ratio corresponds to a value of 35 ppmw (parts per million weight) of THF in the reactor bed as a catalyst component. Free THF may be present in the gas phase and/or absorbed in the resin, and the ratio of free THF to absorbed THF will equilibrate over time. Due to the difficulty in measuring this amount of THF with the current analyzer setup, 50-60 ppmw of THF was injected into the bed to account for the free and absorbed THF. The tested sample cylinder, as described above, was filled with 1.9 cc of THF, and topped off with approximately 130 g of isopentane. The content of the cylinder was pressured into the reactor, using high pressure, and high purity nitrogen, as discussed above. The result was an immediate and complete kill of the UCAT™ G reaction (see Figure T). Catalyst feed was continued for an additional 20 minutes, with no sign of reaction.

The reactor was vented down to 90 psig and the cycle water temperature adjusted to maintain approximately 90-93⁰C. The 90 psig is the minimum reactor pressure at which the cycle gas compressor should be operated, and the 93 °C temperature is the maximum achievable bed temperature, based on the expansion bellows limitation in the cycle water system. Nitrogen was fed to the reactor to produce a vent rate of 45 pph. This vent rate simulated the purge rate of the reactor where this transition from a Ziegler-type catalyst system to a chromium catalyst system would be tested. The reactor was purged for three hours and 20 minutes (from 17:40 to 21 :00). After this time (at 21 :00), the reactor conditions were built to their pre-kill levels, and catalyst was initiated approximately one hour and 50 minutes later (at 22:50). Catalyst was fed at full rate (9 shots/min) for approximately 2 hours and 25 minutes (until 00:15), with little sign of reaction. The reactor was again blown down to 90 psig and purged as above for three hours (from 01:00 to 04:00). Reactor conditions were built, and catalyst feed was initiated one hour later (on at 05:00). Rates slowly came on, with inlet temperature and bed temperature crossover one hour and 50 minutes later (at 06:50). The reactor was returned to its previous pre-kill baseline conditions. This first test shows that, with less 12 hrs of purging, the level of THF in the reactor system can be reduced to the point that UCAT G can establish a reaction.

THF Injection Test #2 A second THF injection test was performed, in an attempt to verify the purging results. The procedure was the same as above, except that a continuous 8 hour purge was completed prior to building of conditions and catalyst initiation. Prior to the start of the test, the reactor was lined out on a high density blow molding resin (Resin B). The tested sample cylinder was charged with 1.9 cc of THF and a balance of approximately 13O g of isopentane. The sample was pressured into the reactor using high-pressure, high-purity nitrogen (at 19:20) (see Figure 3). This amount of THF was equivalent to 50-60 ppmw in the bed. The reaction was immediately killed as before. The reactor was then blown down to 90 psig, the bed temperature set to 93⁰C₅ and purged with nitrogen to produce a nitrogen vent of 45 pph for 8 hours. After the purge was completed, the reactor was brought back up to the high density blow molding resin conditions, and catalyst initiated about one hour and ten minutes later (at 05:10). The polymerization reaction initiated with the start of the catalyst feed, and the inlet and bed temperature crossover occurred approximately 2 hours and ten minutes later (at 07:20). This approximately 2 hour time frame for temperature crossover was similar to the time for crossover for Test # 1. This test verifies that 50-60 ppmw of THF can be purged from the reactor, to a level that is conducive to a UCAT™ G reaction, after approximately 8 hours.

Experiment 2 -Transitions from UCAT™ A2000 to UCAT™ G500 (Chromium! Transition #1 from Resin A to Resin B

The reactor was lined out on Resin A produced with UCAT™ A2000 catalyst. The average residual metals, measured via X-ray were 3.3 ppmw Ti and 87 ppmw Al for a molar Al/Ti ratio of 47. Based on the residual Al measurement, there was approximately 368 ppmw of TEAl in the bed. This amount of TEAl was determined when the TEAl feed was shut off. The calculated bed weight was 68 lbs, resulting in a TEAl inventory of 0.025 lbs. The deactivation of TEAl with CO₂ used an amount of CO₂ equal to 20 times the mass of the TEAL Therefore, 0.50 lbs of CO₂ was used to neutralize the TEAl at these conditions.

The transition was initiated at 22:20, when the catalyst feed (supported transition metal halide and THF) was discontinued, and the reaction allowed to die off (see Figure 4). The TEAl feed to the reactor and the hydrocarbon flows (monomer, comonomer and hydrogen) were continued during this time. The inlet and bed temperatures crossed over approximately one hour and 40 minutes later (at 0:00), at which time the TEAl and hydrocarbon flows were shut off. A 500 mpph (millipound per hour) feed of CO₂ was initiated at the same time the TEAl and hydrocarbon flows were shut off. The result was a rapid kill of the remaining reaction rate, as CO₂ is a poison to Ziegler catalysts. No change in the skin thermocouples, static level (see Figure 5) or static pattern was observed during the CO₂ feed. The CO₂ feed was continued for one hour (until 01 :00). The CO₂ was allowed to circulate in the reaction system for an additional hour. The reactor was vented down to 90 psig, the bed temperature was set at 94°C, and the reactor was purged with 53 pph (pounds per hour) of nitrogen for 9.5 hours.

After completing the purge step (at 12:00), the reactor was brought up to Resin B conditions, and the catalyst was initiated at full rate (9 shots/min) at 13:45. Approximately four hour and 40 minutes later (by 16:15), a low grade reaction was observed, but crossover had not been achieved. At this time, all feeds were stopped, the reactor was blown down and reactor conditions were rebuild. Once these steps were completed, and catalyst was initiated at 18:00. The polymerization reaction was initiated, and crossover was achieved at 19:00. From an observation of temperature trends over an extended period, it appear that the blow down did not accelerated the initiation of reaction.

Transition #2 from Resin A to Resin B

The reactor was lined out, producing Resin A using UCAT™ A 2000 catalyst as before. At 10:00, the transition was started by shutting off the catalyst feed (transition metal halide and THF) while maintaining TEAl and hydrocarbon feeds (see Figure 6). Approximately 10-15 minutes later the reaction rate began to decline. At that time the C6/C2 molar ratio was increased from 0.020 to 0.023. In addition, the ethylene partial pressure was increased towards a target value of 160 psi, but the target was overshot, and the partial pressure climbed to a maximum of 169 psi, before slowly falling back toward 160 psi. These increases were done in an attempt to maintain the reaction rate for as long as possible. The longer the reaction continued, the more resin would be produced, and the more THF would be removed from the reactor via resin discharges, and the more the transition metal halide would be consumed. Crossover of the inlet and bed temperatures occurred at 12:15. At that time, the TEAl and hydrocarbon feeds were shut off, and a 500 mpph feed of CO₂ initiated. The remaining reaction was killed with addition of the CO₂. There were no changes in the behavior of the static (see Figure 7) or the skin thermocouples during the CO₂ injection. With the remaining reaction dead, the bed temperature was decreased to 93⁰C. The CO₂ was fed for one hour and then circulated for an additional hour. After the circulation was completed, the reactor was blown down to 90 psig and purged with 40 pph of N₂ for 16 hours.

The purge was completed at 07:00, and the reactor was brought up to Resin B conditions. The bed temperature was held at 93 °C, until the inlet and bed temperatures crossed over, and then the bed temperature was ramped to 110°C. Catalyst was initiated at 08 :40 with noticeable reaction within half an hour. Crossover occurred at 10:15, and the bed temperature was raised to 110°C by 12:45. The reaction rate lined out at final conditions by 15:00. A verification run was performed next.

Transition #3 from Resin A to Resin B

The reactor was lined out on Resin A at the start of this transition. At 19:10 the catalyst feed to the reactor was shut off, while maintaining the TEAl and hydrocarbon feeds (see Figure 8). At 19:30, the reaction rate began to decay, and the C6/C2 ratio was increased to 0.025, from 0.020, and the ethylene partial pressure was increased to 164 psi. The bed and inlet temperatures cross over at 21 :05, at which time the TEAl and hydrocarbon feeds were discontinued, and a flow of 500 mpph of CO₂ was started to the reactor. For the first time during these transitions, the static pattern in the bed changed (see Figure 9). Upon addition of the CO₂ to the reactor, the static began to draw a band from the -40 V baseline down to -300 V for approximately 30 minutes, after which the static narrowed, producing intermittent spikes to -100 V for the remainder of the purge. Previous purges saw similar static spikes without issue. The - 300 V static level upon addition of CO₂ did not have an effect on the reactor skin TCs (thermocouples). The reaction was completely dead by 21:10, and the bed temperature lowered to 93°C at 21 :20. The CO₂ feed was discontinued at 22:05, and circulated for an additional hour. At 23:05 the reactor was blown down to 90 psig and purged with 40 pph nitrogen for 16 hours. At 17 : 10 the purge was completed, and the reactor brought to Resin B conditions. Catalyst was initiated at 18:25 with crossover obtained at 20:00. The bed temperature was left at 93°C, until crossover occurred, at which time the bed temperature was ramped to 110°C. The bed temperature reached 110°C at 22:25. Reaction reached steady state by 01 :00. The results of this transition mirror the previous transition well, suggesting the procedure, is an effective means of transitioning from a Ziegler-type system to a chromium-based catalyst system.

Transition #4 from Resin A to Resin B

Prior to the start of the test, the reactor was lined out on Resin A. At 10:20 the catalyst feed to the reactor was shut off, while maintaining the TEAl and hydrocarbon feeds (see Figure 10). At 10:35 the reaction rate began to decay, and the ethylene partial pressure was increased to 164 psi. The C6/C2 ratio was increased to 0.032, from 0.020. The bed and inlet temperatures crossed over at 12: 15, at which time the TEAl and hydrocarbon feeds were discontinued, and a flow of 500 mpph of CO₂ was started to the reactor. For the second time during these transitions, the static pattern in the bed changed (see Figure 11). The static response during this transition was quite similar to the previous transition. Again, the skin thermocouples did not change in their response. The observed -300 V level of static is not unusual on a UNIPOL™ reactor (Union Carbide Corporation, subsidiary of The Dow Chemical Company). In addition, this response to the CO₂ injection is of minor concern, as the skin thermocouples did not deviate during the increased static level, thus indicating no reaction or resin build-up at the reactor wall. The reaction was completely dead by 12:20, and the bed temperature was lowered to 93°C at 12:30. The CO₂ feed was discontinued at 13:15, and the CO₂ was circulated for an additional hour. At 14:20, the reactor was blown down to 90 psig and purged with 40 pph nitrogen for 6 hours.

At 21 : 10, the purge was completed, and the reactor was brought to Resin B conditions. Catalyst was initiated at 00:25, with crossover obtained at 02:25. The bed temperature ramp began at 05:00 and was completed by 07:20. Reaction reached steady state by the end of the temperature ramp. Typically the bed temperature is ramped to 11O⁰C at crossover, but was delayed in this attempt, to see if the reaction rate trajectory changed. Figure 12 shows the reaction trajectories for transitions #2, #3, and #4. The delay in ramping the temperature during transition #4 appeared to stall the progress of the reaction rate at 80 percent of the final rate. Transition #4 used a purge time of only 6 hours versus 16 hours for transitions #2 and #3, and thus a higher level of residual THF would be present in the reactor for transition #4. At lower temperatures, the residual THF or other impurities in the system may be adsorbed on the walls of the reactor or absorbed in the resin. Therefore, as the temperature is ramped up, the THF or impurities are desorbed and act as a poison to the chromium catalyst. The level of these components would be greater as a result of the shorter purge time. However, a short delay in the time for the reaction rate to reach steady state is preferable, to an additional 10 hours of purge time. As the reaction rate initiates and reaches steady state, resin and impurities are discharged from the reactor; thus an increase in this reaction time, removes more impurities from the reactor and reduces the time to effectuate a sufficient purge of the reactor.

Experiment 3 - Transition from Resin A (Z/N Catalyst) to Resin C (Chromium) The four transitions prior to this experiment have been from UCAT™ A 2000 to

UCAT™ G 500. This final test was performed to test the adequacy of the procedure defined in Experiment 2 for the initiation reaction of the UCAT™ B catalyst family, specifically the UCAT™ B 400 catalyst.

Prior to the start of the test, the reactor was lined out on Resin A. At 14:05, the catalyst feed to the reactor was shut off, while maintaining the TEAl and hydrocarbon feeds (see Figure 13). At 14:25, the reaction rate began to decay, and the C6/C2 ratio was increased to 0.029, from 0.020, and the ethylene partial pressure was increased to 165 psi. The bed and inlet temperatures crossed over at 16:00. The reaction rate was allowed to decay for 45 minutes to reduce the catalyst inventory (transition metal halide and THF) as much as possible, based on the results of transitions #2-4, as discussed in Experiment 2. At 16:45, the TEAl and hydrocarbon feeds were discontinued, and a flow of 500 mpph of CO₂ was started to the reactor. For the third time during these transitions, the static pattern in the bed changed (see Figure 14). Again, the response to the CO₂ injection was static briefly down to -300 V, with no change in the skin thermocouple activity. The reaction was completely dead by 16:50, and the bed temperature lowered to 93 °C at the same time. The CO₂ feed was discontinued at 17:45, and the CO₂ was circulated for an additional hour and a half. At 19:15, the reactor was blown down to 90 psig and purged with 45 pph nitrogen for 6 hours. At 19:20, a computer control loop shed, and the cycle water temperature was held constant. This undetected issue resulted in the bed temperature not controlled at 93 °C. For the majority of the 6-hour purge, the bed temperature was 78⁰C. This lower temperature resulted in a less efficient purge of the THF .

At 02:50, the purge was completed, and the reactor brought up to Resin C conditions. Catalyst feed was initiated at 04:30, at 2 shots/min, with little sign of reaction. At 06:20, the catalyst feed was increased to 4 shots/min, and by 06:45, reaction began to come on strong. Crossover was obtained at 07: 15. Reaction reached steady state by 10:25 at a catalyst feed rate of 4.25 shots/min.

Thus, the transition from UCAT™ A 2000 to UCAT™ B 400 can be completed by performing a catalyst inventory reduction, a CO₂ passivation and circulation, and a purge step of reasonable duration, approximately 6 hours.

Conclusion

The general procedure as modified during the five transitions is presented below. One skilled in the art would recognized that modifications in this procedure are possible to achieve operable catalyst activation and reactor conditions with the catalyst systems used above and with other catalyst systems. Stop catalyst feed. As rates begin to fall off, increase C2PP (ethylene partial pressure) toward 165 psi to extend the consumption of transition metal halide and THF.

Increasing the C6/C2 ratio would help maintain reaction longer and thereby lower the

THF level and expend more of the Ziegler-type catalyst. However, maintaining aim grade would likely outweigh the benefit of the C6/C2 increase. Continue to burn catalyst until rates are at least two-thirds of the lined out production rate, or until control of the gas composition ratios is no longer possible. Feed a total amount of COj equivalent to 20 times the mass of TEAL in the bed over a period one hour. Circulate for an additional hour.

Blow down reactor to 90 psig and set bed temperature to 93⁰C₅ and purge with maximum nitrogen flow rate possible for at least 6 hrs. Build conditions for Cr reaction, and start catalyst per normal startup procedures.

The initial THF injection tests verified that THF in concentrations of 50-60 ppmw were capable of poisoning and completely killing a chromium catalyzed reaction system. This concentration of THF would not be atypical for a UC AT™ A catalyzed high density resin. Additionally, it was determined that using a purge flow and reactor bed temperature comparable to a commercial facility, that it was possible to purge the THF from the reactor to a level that facilitated chromium reaction in a time frame of approximately 8 hours.

Four transitions from UCAT™ A 2000 to UCAT™ G 500 were performed, following the THF injection study. These transitions showed that via a combination of a catalyst inventory reduction, a CCh injection and circulation, and a purge of at least 6 hours, it was possible to transition from UCAT™ A 2000 to UCAT™ G 500 with little difficulty. Reducing catalyst inventory by increasing the ethylene partial pressure was not performed during Transition #1, and the second polymerization failed to establish reaction after a 12 hour purge. The introduction of the catalyst inventory reduction step lowered the purge time to 6 hours. An added benefit of this step is that the majority of the UCAT™ A 2000 catalyst life would be expended, thereby minimizing the formation of high molecular weight (HMW) gels that are generated on restart due to a low grade reaction under UCAT™ G 500 reactor conditions. The only negative response was a 30-60 minute banding of static from baseline to -300 V during the CO₂ injection into the reaction system. However, this static level did not result in cold banding in the reaction, system, and no ill effects were observed.

A final transition from UCAT™ A 2000 to UCAT™ B 400 was performed to ensure that inventive procedure would translate from the UCAT™ G 500 catalyst system to the UCAT™ B 400 system. The response of the catalyst and reaction system was similar to the commercial response of the catalyst system. This test confirmed that the procedure was capable of a successful transition from UCAT™ A 2000 to UCAT™ B 400. This transition procedure from UCAT A/J to UCAT B/G, as developed herein, addresses the major issues associated with the current silica injection procedures. The main improvements are the reduced probability of a reactor continuity incident and the reduced probability of gels from either the silica or high molecular weight gels from active UCAT A/J at UCAT B/G reactor conditions. The combination of these two improvements will decrease the possibility of a reactor train stop or additional offgrade from the production unit at a cost savings.

Finally, TEAl passivation of a reactor has been known to scavenge impurities from the system. Unfortunately, TEAl passivation of a UCAT B/G reactor is not applicable due to the incompatibility of the UCAT B/G catalysts with aluminum alkyls. Typically a reactor is baked out at high temperature to drive off impurities. However, a TEAl passivation, followed by the use of CO₂ to scavenge the residual aluminum alkyl from the reaction system can dramatically shorten the time required to obtain an impurity lean system. This procedure has the ability to greatly improve the efficiency of the transition from a Ziegler-type catalyst system to a chromium-based catalyst system.

Claims

WHAT IS CLAIMED:

1. A process for transitioning a continuous olefin polymerization reaction catalyzed by a Ziegler-type catalyst, comprising a transition metal halide, an electron donor component and an activator (or cocatalyst), to a polymerization catalyzed by another type of catalyst (second catalyst), said process comprising a process selected from the group consisting of A), B) and C):

A) (1) reducing catalyst inventory in a reactor, (2) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor,

(3) introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor, and (4) purging the reactor with an inert gas;

B) (1) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into a reactor,

(4) purging the reactor with an inert gas; and

C) (1) reducing catalyst inventory in a reactor,

(3) further reducing catalyst inventory in the reactor,

(5) purging the reactor with an inert gas; and wherein, for each of process, A), B) and C), the catalyst inventory comprises the Ziegler-type transition metal halide and the electron donor used for the Ziegler-type catalyst; and wherein, the reversible and irreversible poisons will deactivate the activator (or cocatalyst) component of the Ziegler-type catalyst; and wherein, for each process, step (2) is performed after step (1), step (3) is performed after step (2), and so on.

2. The process of Claim 1, wherein, for each process A), B) or C), the reduction in catalyst inventory is performed by one or more steps selected from the group consisting of: a) continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the comonomer to monomer ratio in the reactor; b) continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the monomer partial pressure in the reactor, c) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; d) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and decreasing the hydrogen to monomer ratio in the reactor; e) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and adjusting reaction temperature to increase catalyst activity; f) continuing both hydrocarbon feed (monomer, optional comonomer and hydrogen) and the feed of activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing residence time; g) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the comonomer to monomer ratio in the reactor; h) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing the monomer partial pressure in the reactor; i) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; j) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and decreasing the hydrogen to monomer ratio in the reactor; k) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and adjusting reaction temperature to increase catalyst activity; and 1) continuing hydrocarbon feed (monomer, optional comonomer and hydrogen) to the reactor, and increasing residence time.

3. The process of Claim 1, wherein for each process A), B) and C), the hydrocarbon feed (monomer, optional comonomer and hydrogen) is continued after the discontinuing of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor.

4. The process of Claim 2, wherein for each process A), B) and C), the cocatalyst feed is discontinued before the introduction of the at least one reversible poison and/or the at one irreversible poison into the reactor.

5. The process of Claim 2, wherein the process is process C), and comprises the following steps:

(1) reducing catalyst inventory in the reactor, (2) discontinuing the introduction of the transition metal halide component and the electron donor component of the Ziegler-type catalyst into the reactor,

(3) further reducing catalyst inventory in the reactor,

(4) introducing at least one reversible poison into the reactor, and circulating said poison within the reactor, and (5) purging the reactor with an inert gas; and

wherein, for the first and second catalyst inventory reduction steps, the catalyst inventory is reduced by continuing both hydrocarbon feed (monomer, optional comonomer, and hydrogen) and feed of the activator (or co-catalyst) component of the Ziegler-type catalyst to the reactor, and increasing the comonomer to monomer ratio in the reactor; and wherein the cocatalyst feed is discontinued into the reactor before the introduction of the at least one reversible poison into the reactor.

6. The process of Claim 2, wherein the process is process B), and comprises the following steps:

(2) reducing catalyst inventory in the reactor,

(3) introducing at least one reversible poison into the reactor, and circulating said poison within the reactor, and

(4) purging the reactor with an inert gas; and

wherein the catalyst inventory is reduced by continuing hydrocarbon feed

(monomer, optional comonomer and hydrogen) to the reactor, and increasing the comonomer to monomer ratio in the reactor, and increasing the monomer partial pressure in the reactor; and wherein the cocatalyst feed is discontinued into the reactor before the introduction of the at least one reversible poison into the reactor.

7. The process of Claim 2, wherein the second catalyst is a chromium-based catalyst.

8. The process of Claim 2, wherein the second catalyst is a metallocene catalyst.

9. The process of Claim 2, wherein the second catalyst is a constrained geometry catalyst.

10. The process of Claim 1 , wherein polymerization is conducted in a gas phase fluid bed reactor.

11. The process of Claim 2, wherein, for each process A), B) and C), at least one reversible poison is introduced into the reactor, and wherein the polymerization in the reactor is allowed to continue for 12 hour or less, after the transition metal halide and electron donor components have been discontinued, and before the at least one reversible poison has been introduced into the reactor.

12. The process of Claim 11 , wherein the at least one reversible poison is carbon dioxide.

13. The process of Claim 12, wherein the amount of carbon dioxide introduced to the reactor is at least 20 times the mass of the activator (or cocatalyst) component of the Ziegler catalyst present in the reactor.

14. The process of Claim 12, wherein the carbon dioxide is fed into the reactor for at least one hour, and is then circulated within the reactor for at least one hour.

15. The process of Claim 2, wherein, for each process A), B) and C), the inert gas is nitrogen.

16. The process of Claim 15 , wherein the purging step is performed by flow purging or pressure purging the reactor.

17. The process of Claim 2, wherein, for each process A), B) and C), the activator (or co-catalyst) component of the Ziegler-type catalyst is a trialkylaluminum compound, and the electron donor is tetrahydrofuran.

18. The process of Claim 17, wherein the activator (or co-catalyst) component of the Ziegler-type catalyst is triethyl aluminum.

19. The process of Claim 18, wherein, for each process A)₅ B) and C), at least one reversible poison is introduced into the reactor.

20. The process of Claim 19, wherein the at least one reversible poison is carbon dioxide.

21. The process of Claim 2, further comprising, for each process A), B), or C), circulating said poison or poisons within one or more auxiliary pieces of equipment, prior to purging the reactor with an inert gas.

22. The process of Claim 21 , wherein the one or more auxiliary pieces of equipment are selected from the group consisting of one or more vent recovery streams, one or more product discharge systems, and combinations thereof.

23. The process of Claim 2, wherein for each process A), B), or C), at least one reversible poison is introduced into the reactor.

24. The process of Claim 23, wherein the at least one reversible poison is carbon dioxide.

25. The process of Claim 21, wherein for each process A), B), or C), at least one reversible poison is introduced into the reactor.

26. The process of Claim 25, wherein the at least one reversible poison is carbon dioxide.

27. The process of Claim 2, further comprising, for each process A), B), or C), after purging the reactor with an inert gas, introducing at least one reversible poison and/or at least one irreversible poison into the reactor, and circulating said poison or poisons within the reactor.

28. The process of Claim 27, further comprising circulating said poison or poisons within one or more auxiliary pieces of equipment, prior to purging the reactor with an inert gas, and/or circulating said poison or poisons within one or more auxiliary pieces of equipment after purging the reactor with an inert gas.

29. The process of Claim 28, wherein the one or more auxiliary pieces of equipment are selected from the group consisting of one or more vent recovery streams, one or more product discharge systems, and combinations thereof.

30. The process of Claim 27, wherein for each process A), B), or C), at least one reversible poison is introduced into the reactor.

31. The process of Claim 30, wherein the at least one reversible poison is carbon dioxide.

32. The process of Claim 28, wherein for each process A), B), or C), at least one reversible poison is introduced into the reactor.

33. The process of Claim 32, wherein the at least one reversible poison is carbon dioxide.

34. The process of Claim 2, wherein the process for transitioning comprises process A).

35. The process of Claim 2, wherein the process for transitioning comprises process B).

36. The process of Claim 2, wherein the process for transitioning comprises process C).