CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims the benefit of U.S. Provisional Application number 63/114,708 filed November 17, 2020, entitled “Gas Phase Reactor Startup Process”, the entirety of which is incorporated by reference herein.

FIELD OF THE INVENTION

[0002] The present invention relates to the startup process for gas phase reactors, and in particular to the startup process of a gas phase reactor when using metallocene catalysts.

BACKGROUND

[0003] When starting up a gas phase reactor with metallocene catalysts — transition metal compounds comprising at least one cyclopentadienyl ligand and/or ligand(s) isolobal to cyclopentadienyl — there is an increased tendency, compared to Ziegler-Natta catalysts, of reactor sheeting and fouling, especially in the expanded zone of the reactor. Operation of most reactor systems is critically dependent upon good mixing for uniform reactor conditions, heat removal, and effective catalyst performance. The process must be controllable and capable of a high production rate. In general, the higher the operating temperature, the greater the capability to achieve high production rate. Because polymerization reactions are typically exothermic, heat transfer out of the reactor is critical to avoid such problems as particle agglomeration and runaway reactions. However, as the operating temperature approaches and exceeds the melting point of the polyolefin product, the particles of polyolefin become tacky and melt. For example, non-uniform fluidization of the bed can create “hot spots,” which in turn can cause the newly-formed polymer particles to become tacky due to elevated temperatures in the hot spots. A predominant time during the operation of a reactor for such problems to occur is at startup, when the polymerization reaction within the reactor is initiated.

[0004] An interplay of forces may result in particles agglomerating with adjacent particles, and may lead to sheeting and other forms of reactor fouling, especially in the expanded zone of the fluidized bed gas phase reactor. In agglomeration, the particles stick together, forming agglomerated particles that affect fluid flow and may be difficult to remove from the system. In sheeting, tacky particles gather on a surface of the reactor system, such as the wall of the reactor vessel, forming a sheet of polymer particles. Progressive cycles in this process may eventually result in the growth of the sheet and its falling into the fluid bed. These sheets can interrupt fluidization, circulation of gas and withdrawal of the product from the reactor, and may require a reactor shutdown for removal.

[0005] Many factors influence the propensity for sheeting and other fouling phenomena, of which one is the type of catalyst. For example, metallocene catalysts allow the production of polyolefins with unique properties such as narrow molecular weight distributions and narrow chemical compositions. These properties in turn result in improved structural performance in products made with these polymers. However, while metallocene catalysts have yielded polymers with improved characteristics, they have presented particular challenges when used in fluidized bed reactors, in particular, in relation to sheeting and fouling in other portions of the reactor system, such as the distributor plate and the cooler.

[0006] Various methods for controlling sheeting and other forms of reactor upset have been developed. These methods often involve monitoring the static charges near the reactor wall in regions where sheeting is known to develop and introducing a static control agent into the reactor when the static levels fall outside a predetermined range. For example, US 4,803,251 and US 5,391,657 disclose the use of various chemical additives in a fluidized bed reactor to control static charges in the reactor. The static charge in the reactor is typically measured at or near the reactor wall at or below the site where sheet formation usually occurs, using static voltage indicators such as voltage or current probes or electrodes. However, these approaches not only add to the cost of the process but also to complexity of process control.

[0007] What is needed is a method of starting up a gas phase reactor, either in open bed (or “seedbed”) configuration or recirculating startup configuration, when using metallocene and/or other high activity single-site catalysts.

[0008] Other related publications include US 4,588,790; US 5,385,991; US 6,753,390; US 7,774,178; US 7,910,668; US 8,273,834; US 8,383,739; US 8,598,283; US 9,540,460; US 10,207,237; and US 10,221,254.

SUMMARY OF THE INVENTION

[0009] Disclosed is a gas phase reactor startup process comprising (consisting of, consisting essentially of) providing a reactor having a vertical cylindrical portion from a distributor plate at the base to a neck at the top and comprising a fluidized bed zone therebetween, wherein above the neck and connected thereto is a frustoconical portion closed at the top, the vertical cylindrical portion having a length from the distributor plate to the neck; providing a bed of polyolefin granules within the fluidized bed zone; wherein the polyolefin granules fill the fluidized bed zone to a resting bed level within a range from 75, or 80% to 85, or 90% of the length of the vertical cylindrical portion prior to fluidization with cycle gas; flowing cycle gas to fluidize the polyolefin granules and form a fluidized bed having a fluidized bed level greater than the resting bed level; and injecting metallocene catalyst into the fluidized bed zone.

[0010] Further disclosed is a method of transitioning the fluidized bed zone from a dry mode (no condensing agent) of operation to a condensed mode (adding condensing agent until reaching or past the dew point of the system) of operation, wherein it is desirable that the temperature of the cycle gas (as measured at the cycle gas cooler outlet) is lowered below the dewpoint of said cycle gas (“dewpoint cross”), going from an initial cycle gas temperature to a condensed-mode cycle gas temperature, over a particular time period. Preferably, this dewpoint cross from initial to condensed-mode temperature is accomplished over a time period that is within a range from 30, or 35 minutes to 40, or 45, or 50, or 60 minutes.

BRIEF DESCRIPTION OF THE DRAWING

[0011] The Figure is a partial cutaway representation of a gas phase reactor.

DETAILED DESCRIPTION

[0012] It has been found that at least one or more steps can be taken in the startup phase of a gas phase reactor operation to produce polyolefins to reduce or eliminate reactor upsets such as polyolefin melting, sheeting, and agglomeration formation that results in inoperability of the reactor, especially when using high activity catalysts such as metallocenes.

[0013] As provided herein, after startup and upon reaching a steady state, the reactor can operate to perform polymerization using a typical continuous gas phase fluidized bed process. For example, the polymerization reactor can be a fluidized bed reactor that is operated to produce polyolefin polymers. The polymerization reactor can be a staged reactor where two or more reactors are employed in series, where a first reactor can produce, for example, a high molecular weight (or low melt flow rate) component and a second reactor can produce a low molecular weight (or high melt flow rate) component. In operation, the polymerization medium can be fluidized by the continuous flow of the gaseous monomer and diluent.

[0014] In a continuous gas phase fluidized bed reactor, the polymerization reactor comprises a fluidized bed of solid phase material. At startup, the seed bed comprising polymer granules is loaded into the polymerization reactor. Liquid or gaseous feed streams of a primary monomer and hydrogen together with a liquid or gaseous comonomer (and optionally inert or carrier gas such as nitrogen, which may also be referred to as a diluent) are combined and then introduced into the fluidized bed, often via an upstream recycle gas line. The fluidized bed reactor for performing a continuous gas phase process typically comprises a reaction zone and a so-called velocity reduction zone. The reaction zone comprises a bed of growing polymer particles, formed polymer particles, and a minor amount of catalyst particles (collectively sometimes referred to herein as “solid phase material”) fluidized by the continuous flow of the gas comprising gaseous monomer and/or comonomer(s) and potentially other gaseous compounds (e.g., induced condensing agent ICA (such as pentane, in particular isopentane or iC5), hydrogen, and/or diluent/carrier gas/inert gas such as nitrogen) to remove heat of polymerization through the reaction zone. Re-circulated gases (recycled gas) can be compressed and cooled to form liquids that increase the heat removal capacity of the circulating gas stream when readmitted to the reaction zone. This method of operation is referred to as “condensed mode.”

[0015] A suitable rate of gas flow into the fluidized bed reactor can be readily determined. The flow rates of monomer and circulating gas into the polymerization reactor is approximately equal to the rate that polymer product and unreacted monomer are withdrawn. In any embodiment, the cycle gas circulating rate (gas entering and leaving the bed) is within a range from 1000 ton/hour to 1400, or 200, or 2500, 3000, or 5000 ton/hour, and the reactor production rate is within a range from 20, or 40 ton/hour to 60, or 80, or 100, or 200 ton/hour (noting 1 ton is a metric ton, or 1000 kg). The composition of the gas passing through the reactor (and, hence, the bed) can be adjusted to maintain a steady state gaseous composition within the bed or ‘reaction zone’. Gas leaving the reaction zone is passed to the velocity reduction zone where most entrained particles settle back into the dense phase zone. Gas is compressed in a compressor and passed through a heat exchanger wherein the heat of polymerization is removed, and the gas is returned to the reaction zone.

[0016] The “cycle gas” can comprise any number of components and the identity and amount of components can change (be controlled) by the operator depending upon the desired compositions. The cycle gas can comprise inert gas such as nitrogen, monomers such as ethylene or propylene, and comonomers such as butene, hexene or octene, and/or hydrogen. The cycle gas can also comprise condensing agents such as isopentane or isobutane, used for additional heat removal from the polymerization system occurring within the fluidized bed zone. [0017] To maintain a constant reactor temperature, the temperature of circulating gas can be continuously adjusted up or down to accommodate any changes in the rate of heat generation due to the polymerization. The fluidized bed can be maintained at a constant height by withdrawing a portion of the bed at a rate equal to the rate of formation of particulate product. Polymer product can be removed semi-continuously via a series of valves into a fixed volume chamber, which is simultaneously vented back to the reactor for efficient removal of the product. At the same time, a significant portion of the unreacted gases are recycled into the reactor. Polymer product is purged to remove entrained hydrocarbons and can be treated with a small steam of humidified nitrogen to deactivate any trace quantities of residual catalyst.

[0018] Furthermore, the reactor temperature of the fluidized bed reactor can range from 30, or 40, or 50°C to 85, or 90, or 95, or 100, or 110, or 120°C. In general, the reactor temperature is operated at the highest temperature that is feasible, taking into account the sintering temperature of the polymer product within the reactor.

[0019] As described herein, the reactor used in connection with the present methodologies can be operated to produce homopolymers of olefins, for example, ethylene or propylene, and/or copolymers, terpolymers, and the like, of olefins, particularly ethylene, and at least one other olefin. For example, the polymerization reactor can produce polyethylenes. Such polyethylenes can be homopolymers of ethylene and interpolymers of ethylene and at least one a-olefin wherein the ethylene content is at least 50%, or 60%, or 70%, or 80% by weight of the total monomers involved. Exemplary olefins that can be utilized in the reactor are ethylene, propylene, 1 -butene, 1 -pentene, 1 -hexene, 1 -heptene, 1 -octene, 4-methylpent-l-ene, 1 -decene, 1 -dodecene, 1- hexadecene and the like. Also utilizable herein are polyenes such as 1,3 -hexadiene, 1,4-hexadiene, cyclopentadiene, dicyclopentadiene, 4-vinylcyclohex-l-ene, 1,5-cyclooctadiene, 5-vinylidene-2- norbornene and 5-vinyl-2-norbomene, and olefins formed in situ in the polymerization medium. When olefins are formed in situ in the polymerization medium, the formation of polyolefins containing long chain branching can occur.

[0020] In the production of polyethylene or polypropylene, comonomers can be present in the polymerization reactor. When present, the comonomer can be present at any level with the ethylene or propylene monomer that will achieve the desired weight percent incorporation of the comonomer into the finished granules. [0021] In addition, hydrogen gas is often used in olefin polymerization to control the final properties of the polyolefin. For some types of catalyst systems, it is known that increasing concentrations (partial pressures) of hydrogen increase the melt flow (MF) and/or melt index (MI) of the polyolefin generated. The MF or MI can thus be influenced by the hydrogen concentration. The amount of hydrogen in the polymerization can be expressed as a mole ratio relative to the total polymerizable monomer, for example, ethylene, or a blend of ethylene and hexene or propene. The amount of hydrogen used in some polymerization processes is an amount necessary to achieve the desired MF or MI of the final polyolefin granule.

[0022] The processes described herein are particularly useful when the polymerization process is catalyzed by a metallocene catalyst. As used herein, “metallocene” or “metallocene catalyst” refers to transition metal compounds comprising at least one cyclopentadienyl ligand and/or ligand(s) isolobal to cyclopentadienyl. “Isolobal” means radical molecules wherein the number, symmetry properties, approximate energy and shape of the frontier orbitals that participate in bonding to the transition metal center, especially Group 4 metals such as titanium, zirconium and hafnium, and the number of electrons in them, are similar or identical; preferably, isolobal ligands are those with a similar electronic bonding structure (ligand to metal) as C5-cyclopentadienyl anions. Non-limiting examples of ligands isolobal to the cyclopentadienyl group are cyclopentaphenanthrenyl, indenyl, benzindenyl, fluorenyl, octahydrofluorenyl, cyclooctatetraenyl, cyclopentacyclododecene, phenanthrindenyl, 3,4-benzofluorenyl, 9- phenylfluorenyl, 8-H-cyclopent[a]acenaphthylenyl, 7-H-dibenzofluorenyl, indeno[l,2-9]anthrene, thiophenoindenyl, thiophenofluorenyl, hydrogenated versions thereof (e.g., 4, 5,6,7- tetrahydroindenyl), substituted versions thereof (e.g., n-propylcyclopentadiene), and heterocyclic versions thereof. “Substituted” versions thereof include those having alkyl and/or aryls bound to one or more positions along the rings, and “heterocyclic” versions thereof include rings having one or more carbons substituted by a heteroatom, preferably sulfur, nitrogen, oxygen, silicon and/or phosphorous. Two or more cyclopentadienyl groups may be bridged with a divalent alkylene or silyl bridging group or other bridging group as is well known in the art.

[0023] The metallocene catalyst or catalyst precursor must also be combined with at least one “activator” to effect polymerization of the a-olefin monomers such as ethylene, wherein the activator preferably comprises a methalumoxane compound or a non-coordinating borate anion and a bulky organic cation. In any embodiment, the non-coordinating borate anion comprises a tetra(perfluorinated C6 to C14 aryl)borate anion and substituted versions thereof; most preferably the non-coordinating borate anion comprises a tetra(pentafluorophenyl)borate anion or tetra(perfluoronaphthyl)b orate anion.

[0024] Preferably the bulky organic cation is selected from the following structures (A) and (B): wherein each R group is independently hydrogen, a C6 to C14 aryl (e.g., phenyl, naphthyl, etc.), a Cl to CIO or C20 alkyl, or substituted versions thereof; and more preferably at least one R group is an C6 to C14 aryl or substituted versions thereof.

[0025] In any embodiment, the bulky organic cation is a reducible Lewis Acid, especially a trityl-type cation (wherein each “R” group in (A) is aryl) capable of extracting a ligand from the catalyst precursor, where each “R” group is an C6 to C14 aryl group (phenyl, naphthyl, etc.) or substituted C6 to C14 aryl, and preferably the reducible Lewis acid is triphenyl carbenium and substituted versions thereof.

[0026] Also, in any embodiment, the bulky organic cation is a Bronsted acid capable of donating a proton to the catalyst precursor, wherein at least one “R” group in (B) is hydrogen. Exemplary bulky organic cations of this type in general include ammoniums, oxoniums, phosphoniums, silyliums, and mixtures thereof; preferably ammoniums of methylamine, aniline, dimethylamine, diethylamine, N-methylaniline, diphenylamine, trimethylamine, triethylamine, N,N-dimethylaniline, methyldiphenylamine, pyridine, p-bromo-N,N-dimethylaniline, and p-nitro- N,N-dimethylaniline; phosphoniums from triethylphosphine, triphenylphosphine, and diphenylphosphine; oxoniums from ethers, such as dimethyl ether diethyl ether, tetrahydrofuran, and dioxane; and sulfoniums from thioethers, such as diethyl thioethers and tetrahydrothiophene, and mixtures thereof.

[0027] The catalyst precursor preferably reacts with the activator upon their combination to form a “catalyst” or “activated catalyst” that can then effect the polymerization of monomers. The catalyst may be formed before combining with monomers, after combining with monomers, or simultaneous therewith. [0028] As used herein, the term “startup” refers to the process of beginning the polymerization process of converting olefins to form a polyolefin within a reactor starting from no polymerization until the reaction reaches a steady state within the reactor; wherein “startup” includes the process where fresh polyolefin granules are provided to the reactor as a seedbed bed followed by the introduction of monomers and catalyst to begin the polymerization process (which may be referred to as a “seedbed startup”)’ or wherein the polyolefin granules from a previous shutdown of the reactor are provided, that is, are already present in the reactor such as from a previous polymerization process (e.g., from a situation wherein a previous polymerization process was halted by stopping the introduction of catalyst and/or monomer in the reactor, and/or a compound such a carbon dioxide or carbon monoxide was added to quench the polymerization reactor), and then restarted by adding catalyst and monomer and thus the already present polyolefin granules act as a “circulating restart” of the existing bed of polyolefin granules.

[0029] In any embodiment the polyolefin granules have a bulk density in the range from 300, or 350 kg/m ³ to 450, or 500, or 550 kg/m ³ .

[0030] Thus, in any embodiment is a gas phase reactor startup process comprising providing a reactor capable of sustaining a continuous fluidized bed polymerization process and having a vertical cylindrical portion from a distributor plate to a neck, wherein above the neck and connected thereto is a frustoconical portion closed at the top, the vertical cylindrical portion having a length from the distributor plate to the neck defining a fluidized bed zone and the frustoconical portion defining an expanded zone; providing a bed of polyolefin granules within the fluidized bed zone; wherein the polyolefin granules fill the fluidized bed zone to a resting bed level within a range from 75, or 80% to 85, or 90% of the length of the vertical cylindrical portion prior to fluidization with cycle gas; flowing cycle gas to fluidize the polyolefin granules and form a fluidized bed having a fluidized bed level greater than the resting bed level; and injecting metallocene catalyst into the fluidized bed zone.

[0031] Preferably, cycle gas is always flowing in order to fluidize the bed of polyolefin granules. The composition of the cycle gas can vary, for instance, the cycle gas may be 100% nitrogen while drying the bed down, or be a specified ratio of ethylene, comonomer, hydrogen, condensing agent, and nitrogen when starting catalyst injection and running the reactor in polymerization mode. [0032] In some embodiments, the seedbed of polyolefin granules could be purged with cycle gas comprising (or consisting of) dry inert gas that, when fed to the system, is at a temperature that can be as low as that of the surrounding weather conditions, such as within a range from 0, or 10, or 15, or 18°C to 25, or 30, or 40°C prior to injecting the metallocene catalyst (noting that the dry inert gas, when combined with circulating cycle gas, may increase in temperature, and therefore the just-provided temperature range is of dry inert gas prior to mixing with any cycle gas). Preferably, during the purging of the polyolefin granules the bed of polyolefin granules is at a temperature within a range from 50, or 60, or 70°C to 80, or 90, or 100°C. Preferably, monomer and comonomer is added to the cycle gas after purging the polyolefin granules with a cycle gas comprising nitrogen. Also preferably, metallocene catalyst is injected after purging the polyolefin granules, and most preferably after flowing cycle gas comprising monomers and optional comonomers; preferably once the fluidized bed is at steady state; more preferably, catalyst injection is initiated with cycle gas flowing, the reactor pressure rising, and the bed level is rising due to increasing pressure.

[0033] Preferably, cycle gas is always flowing in order to fluidize the bed, such as before drying the seedbed with nitrogen, during nitrogen drying, after nitrogen drying, as well as during the concentration build, during catalyst injection, during dry mode, during dewpoint approach, during steady state, and after killing the reactor and keeping the bed in standby for a circulating restart.

[0034] Preferably, when purging out moisture from a seedbed, the temperature of the cycle gas is elevated to within a range from 70, or 80°C to 90 or 95°C. If purging out carbon monoxide (CO) after a circulating kill and preparing to restart, the temperature may also be elevated when proceeding to a concentration build. In all gases, the incoming nitrogen is typically at a temperature within a range from 0, or 10, or 15°C to 20, or 25, or 30, or 35°C, and the cycle gas loop flow is heated to the desired temperature.

[0035] In any embodiment, during drying of the polyolefin granules (or also during purging of kill agents from the bed in the case of circulating restart) there is cycle gas flow. The amount varies with reactor size but could be within a range from 800, or 900 tons/hour to 1100, or 1200, or 2000 tons/hour tons/hour. A small amount of inert purge gas may then be added (for instance, at ambient temp) at a rate of 5 to 20 ton/hour. Further, an equivalent amount of cycle gas is preferably vented from the system to maintain pressure. Leaving the 5 to 20 ton/hour of cycle gas in place carries the impurities (such as water for a seedbed startup or CO kill agent for a circulating restart) out of the system. Preferably there is simultaneous flow of cycle gas and purge inert gas, most preferably nitrogen. The cycle gas is preferably heated (i.e., the cooler/heater circulates hot water instead of cold water) during this step and maintains the bed temperature preferably within a range from 75, or 80 °C to 85, or 90 °C thereby overcoming the heat loss to the cooler purge gas. [0036] The gas phase reactor is typically run at pressure above atmospheric pressure, such as at least 20 psig. In any embodiment the metallocene catalyst is injected when the fluidized bed zone is at a pressure within a range from 200, or 220 psig to 265, or 300 psig. Pressure is generated within the closed reactor by the introduction of gases, or “cycle gas”, therein. Cycle gas can include, but is not limited to, monomer and comonomer, condensing agent such as iso-pentane or iso-butane, inert gases such as nitrogen, and other gases or liquids that exist as a gas at the reactor conditions. It is desirable to inject the metallocene catalyst into the reactor under certain desirable conditions and in a desirable manner. In both a seedbed startup and a recirculating startup or recirculating restart, the level of the fluidized bed rises as cycle gas flows from the distributor plate upwards. In a seedbed startup, the metallocene catalyst is preferably injected when the fluidized bed level is within a range from 90, or 95% to 98, or 100, or 102, or 105% from the length from the distributor plate. Described another way, the metallocene catalyst is inj ected when the fluidized bed level is within a range from 1, or 0.8, or 0.6, or 0.5, or 0.4 meters below the neck of the reactor to 0.4, or 0.5, or 0.6, or 0.8, or 1 meters above the neck of the reactor.

[0037] In some embodiments, the level of the fluidized bed at the desired point of time for catalyst injection may vary depending on conditions such as the level of condensing agent present in the cycle gas and the nature of the startup — that is, whether the startup is a “seedbed startup” as opposed to a “recirculating startup.” For instance, a substantial amount of condensing agent may be present in the cycle gas for recirculating restarts in particular, due e.g., to previous condensed-mode operation of the reactor during a previous production campaign; the presence of this condensing agent may result in the bed being at lower levels (e.g., perhaps even 80% from the length of the distributor plate) at the desired point of time of catalyst injection. In any embodiment, once the catalyst injection has started, the bed level is allowed to increase with rising pressure and polyolefin formation and control to the target bed level for the first 2 to 3 bed turnovers (BTOs) under moderated conditions (e.g., 0.8 m height above the neck). [0038] Certain aspects of the process described herein can be readily envisioned with respect to the Figure. The figure depicts a partial cutaway view of a gas phase reactor 100 having a fluidized bed zone 102 which resides within a vertical cylindrical portion 104 defined by the area from the distributor plate 108 to the neck 106. The full distance from the distributor plate 108 to the neck 106 is the length of the vertical cylindrical portion, or length 112. The fluidized bed zone 102 contains polyolefin granules such as granules of polyethylene or polypropylene and has a bed height or bed level exemplified by bed level 118. Most preferably, for startup with metallocene catalyst, the steady state operation level is within a range from 0.3 to 0.8 m above the neck 106. The Figure shows a preferred representation of the bed level in a pre-fluidized state (e.g., at 85% of the length 112). Above this bed can exist cycle gas, as well as some polyolefin particles or granules that may be lifted from the top of the fluidized bed. Above the neck 106, within the reactor is the expanded zone 110 where such lifted particles and granules experience a larger volume and thus will have a tendency to experience lower drag forces and drop back into the bed 102. Gasses and very small particles will exit the reactor in the cycle gas flow out line 116 where the cycle gas may be compressed and/or cooled and recycled back into the reactor through the cycle gas flow in line 114.

[0039] In any embodiment the exemplary granular polyolefin bed level 118 can be controlled by the operator of the reactor through computer or manual means to effect the bed level along the length 112 and/or in the expanded zone 110 above the neck 106. Certain bed levels are preferred for certain stages of the process from startup to first introduction of metallocene catalyst to steady state operation. In any embodiment the reactor may be operated at or above the neck in steady state operation.

[0040] In any embodiment, in seedbed startup, metallocene catalyst is initially injected at a rate of 0.4, or 0.5 kg/hr/m to 2 kg/hr/m (relative to the reactor inside diameter of the fluidized bed zone, in m, which is preferably straight and vertical to the horizon), preferably for an initial injection period that is within the range from 30, or 35 minutes to 40, or 60 minutes; and wherein the catalyst is injected at a progressively increasing rate after the initial injection period. The progressive increase may be continuous or stepwise, and may entail increasing catalyst injection rate from 2 kg/hr/m up to 4, 5, 6, 7, or 8 kg/hr/m over the course of the progressive rate increase (such that 2 kg/hr/m is the injection rate immediately following the initial injection period; and 4, 5, 6, 7 or 8 kg/hr/m is the injection rate at the end of the progressive rate increase). The period of progressive rate increase preferably lasts until the reactor is moved into condensed mode operation. Put in other words, the period of progressive rate increase may start at the end of the initial inj ection period, and run until reactor temperature enters the zone where dewpoint approach is +/- 5°C (that is, such that the cycle gas cooler outlet temperature is within +/- 5°C of the cycle gas dewpoint). In some cases, however, the period of progressive rate increase may last even during the period where cycle gas cooler outlet temperature is within the range of +/- 5°C of the cycle gas dewpoint, and end once the reactor temperature reaches the steady state for condensed mode operation (e.g., temperature within the range from -5°C to -8°C of cycle gas dewpoint). For a recirculating restart, the same general procedure may be followed for catalyst injection (e.g., initial injection period, followed by period of progressive rate increase), except that during the initial injection period, catalyst injection rate is preferably within the range from 0.2, or 0.25 kg/hr/m to 1 kg/hr/m (e.g., catalyst injection rate may be about half the rate used for seedbed startup, during the initial injection period). Catalyst rate is then raised to similar rates over the course of the period of progressive rate increase (e.g., from the rate used during initial injection period; up to 4, 5, 6, 7 or 8 kg/hr/m).

[0041] In any embodiment, the fluidized bed zone is maintained with a cycle gas flow upwards from the distributor plate of the reactor at a space velocity within a range from 0.4 m/s to 0.6 m/s for an initial flow period during drying and/or purging of impurities (e.g., CO) from the reactor. Most preferably, these space velocities are sustained (at least prior to catalyst feeding) in an open seedbed mode of startup.

[0042] The fluidized bed zone may then be maintained with a cycle gas flow upwards from the distributor plate of the reactor at a space velocity within a range from 0.5 m/s to 0.7 m/s when the metallocene catalyst is first injected into the fluidized bed zone. Desirably, the average rate may be in increments and adjusted up and down or may be a steady flow, or a combination of the two. Preferably at steady state operation the fluidized bed zone is maintained with a cycle gas flow upwards from the distributor plate of the reactor at a space velocity within a range from 0.7 m/s to 0.9 m/s at steady state operation.

[0043] As mentioned before, the cycle gas is responsible for creating fluidization of the bed of polyolefin granules and typically includes at least monomers and condensing agent. More particularly the cycle gas comprises ethylene, nitrogen, hydrogen, comonomer, and at least one condensing agent; alternatively, wherein the cycle gas comprises ethylene, nitrogen, hydrogen, comonomer, and one, two, or even more condensing agents. There can be minor amounts of other chemical species that are either impurities in these added components or are side reaction products produced in the reactor and which may accumulate to measureable levels.

[0044] Preferably, the condensing agent, like most feeds (ethylene, comonomer, recovered liquids, hydrogen, and/or nitrogen) is injected into the cycle gas line 114 (see the Figure) located below the distributor plate 108, and not into the fluidized bed zone. Agents such as continuity additive and catalysts, potentially also or instead aluminum alkyl (e.g., when operating with a nonmetallocene catalyst) are injected into the fluidized bed zone 102, preferably the lower portion (lower 25, or 20, or 15, or 10%).

[0045] In any embodiment, the cycle gas is cooled prior to entering the fluidized bed; preferably the cycle gas is cooled below its dewpoint to operate the reactor in condensed mode; wherein condensed mode is initiated then maintained by injecting a condensing agent into the fluidized bed zone. A cooler/heater may be used to control the temperature of the cycle gas. Preferably, the condensing agent is pumped into the cycle gas line with the other feeds either upstream or downstream, but typically downstream, of the cooler. Strategies for controlling cycle gas and/or reactor temperature, particularly in view of melt initiation temperatures (MIT) and delta-MIT (difference between reactor temperature and MIT) are described in U.S. Pat. No. 7,774,178, incorporated by reference herein.

[0046] In any embodiment the fluidized bed zone is transitioned to a condensed mode over a given time period. The transition to condensed mode involves moving the dewpoint approach (here, “dewpoint approach” is a temperature delta: the cycle gas cooler outlet temperature minus the dewpoint of the cycle gas) over a temperature range that crosses the dew point of the cycle gas. Put in other words, the cycle gas cooler outlet temperature is changed from an initial temperature greater than the cycle gas dewpoint (e.g., +5°C relative to dewpoint), to a final temperature, or a condensed-mode temperature, lower than the cycle gas dewpoint (e.g., -5°C relative to dewpoint). According to some embodiments, this crossing of the dewpoint (“dewpoint cross”) from +5°C to -5°C is carried out over a particular time range, such as within the range from 30, or 35 minutes to 40, or 45, or 50, or 60 minutes. The cycle gas dewpoint, it will be understood, may be affected by various factors, such as the presence of ICA (which lowers the dewpoint). The condensing agent preferably is added before the dewpoint of the system (cycle gas) is reached (that is, before this “dewpoint cross” is carried out). [0047] In any embodiment the timing of the transition to condensed mode does not correspond to how long ago condensing agent was added or initial catalyst injection started, but may depend on other factors. For instance, the reactor may start up in dry mode and operate in dry mode for a day and then transition into condensed mode. But when a decision is made to transition the polymerization process into condensed mode it preferably is done such that the +5 °C to - 5 °C dewpoint approach zone is crossed in 30 to 50 minutes. In any embodiment, from the initial catalyst injection rate, the catalyst injection rate is increased to achieve a dewpoint cross from +5 °C to -5 °C of the dewpoint approach in the time period. Further, once the reactor is operating in steady state condensed mode with the dewpoint approach falling and reaching -5 °C to -8 °C, one can reduce catalyst injection in increments of at least 0.3, or 0.4, or 0.5 kg. Most preferably, these rates are sustained in recirculating startup mode or open seedbed mode.

[0048] In any embodiment, the catalyst is injected through one or more points located in the first 5% or 10%, to 25% or 30 %, of the length 112 from the distributor plate.

[0049] In any embodiment, once hydrogen is fed to the reactor it is continuously fed to the fluidized bed at a rate of at least 0.01 kg/hr; preferably at a rate within a range from 0.01 kg/hr to 3, 4, 5, or 6 kg/hr; there are some metallocenes that require rates greater than 6 kg/hr, such as within a range from 6 kg/hr to 10, 12, or 16 kg/hr. Most preferably, the flow of hydrogen into the gas phase reactor is not stopped once it is started, and is desirably kept at a rate of at least 0.01 kg/hr. In certain embodiments, there preferably is no upper limit on hydrogen rate so long as rate is maintained at or above the minimum rate; for instance, certain grades may require more and more hydrogen feed at certain points of production. It is surprising to observe that concentration of hydrogen in the reactor is not important, so much as the fact that some amount of hydrogen is continuously fed.

[0050] In any embodiment, the fluidized bed temperature is maintained using the MIT and delta-MIT control strategies described in U.S. Pat. No. 7,774,178 (and particularly the description spanning column 7, line 24 through column 14, line 12) incorporated by reference herein.

[0051] Described more particularly are desirable steps and conditions for the startup of a gas phase reactor when metallocene catalysts are used for the polymerization of olefins, especially ethylene and optionally comonomers (such as C3 to C12 comonomers, e.g., butene, pentene, hexene, octene). Desirably, the seedbed is first purged to remove residual water and oxygen impurities, typically with the circulation of a nitrogen purge at elevated temperature. Then the feeds to the reactor are started (e.g., ethylene, optional comonomer, hydrogen, condensing agent), and continued until a target concentration of these constituents in the reactor for startup is reached. During this continued feed, the reactor pressure rises. Catalyst injection is preferably initiated when reactor pressure is 230 to 250 psig and the fluidized bed level has expanded and is at or within 0.45 to 0.55 m of the neck. If the bed level is too low, then reactor pressure is allowed to rise to a level greater than 230 to 250 psig, and once bed level comes up due to bed expansion to a level of 0.45 to 0.55 m below the neck, catalyst injection is then initiated. This increased pressure and/or bed level can be accomplished, e.g., by raising space velocity (e.g., to 0.5, 0.6, 0.7, 0.8, 0.9 m/s). However, if space velocity is increased too much, catalyst entrainment may result, potentially leading to fouling issues. Accordingly, in some embodiments, space velocity is raised to less than 0.75 m/s, such as less than 0.7, or less than 0.65 or even less than 0.6 m/s. Furthermore, catalyst is preferably injected while reactor pressure remains below 260 psig.

[0052] If bed level is greater than 0.5 m above the neck at 230 to 250 psig, then polyolefin granules are removed from the reactor (e.g., using product discharge system) to bring the bed level into the desired range, after which catalyst injection is commenced.

[0053] It is preferable to inject catalyst at a low rate over a period of time (e.g., 20, 30, 40, 50, 60, or even 70 minutes) to minimize entrainment.

[0054] It is typical to start the reactor by injecting condensing agents to achieve condensing mode upon steady state operation. But even in dry mode, there can be liquid-fed condensable components in the cycle gas (e.g., induced condensing agent or ICA, such as iso-pentane, and comonomer such as butene and/or hexene). When the heat of polymerization is generated at a fast enough rate (e.g., sufficient catalyst and therefore sufficient reaction is taking place), the cooling system preferably reduces the temperature of the cycle gas going back to the reactor in order to maintain reactor bed temperature control. Depending on the concentration of the condensables in the cycle gas and the degree of cooling, the cycle gas may become cold enough that the condensables start to condense, that is, their dewpoint has been reached. That is the transition from dry mode to condensed mode.

[0055] The startup process with a fresh seedbed or “open seedbed” may have particular features. A controlled startup procedure when starting with an open seedbed may be used to cross quickly into condensed mode to prevent startup reactor fouling, and/or to maintain moderate space velocity at start-up, per discussion above regarding space velocity as related to bed level for catalyst injection. The procedure in such instances preferably includes the following elements: (1) maintaining a space velocity in the bed of, e.g., 0.4, 0.5, or 0.6 to 0.65, 0.70, 0.75, or 0.8 m/s (preferably within the range from 0.4 to 0.8 m/s, such as 0.5 to 0.7 m/s) for the catalyst injection period (e.g., from 20, 25, 30, 35, 40, or 45 to 50, 55, 60, 65, or 70 minutes), and then increase space velocity to line out at the steady state target (discussed below) before transitioning to a condensed mode operation; (2) from the initial catalyst injection rate, increase catalyst to achieve a dewpoint cross of +5 °C to -5 °C dewpoint approach during first phase of catalyst injection (with suitable timeframes for this phase per above discussion); and (3) once the reactor is operating in condensed mode with the dewpoint approach falling and reaching -5 °C to -8 °C, the catalyst injection rate is preferably reduced to avoid overheating. Catalyst injection rate can be reduced in increments (e.g., desired amount of kg/s, such as 1 kg/s, 0.9, 0.8, 0.7, 0.6, 0.5, 0.4, or 0.3 kg/s increments). According to some embodiments, reactor operation in condensed mode can continue with steady state operation. In steady state, the space velocity may be the same as start-up, or optionally it may be increased relative to start-up (e.g., from 0.71, 0.72, 0.75, or 0.80 m/s to 0.76, 0.77, 0.78, 0.79, 0.80, 0.85, or 0.90 m/s, provided the high end of the range is greater than the low end - for instance, suitable range of space velocity at steady-state operation per some embodiments may be within the range from 0.70 to 0.90 m/s).

[0056] The startup process with a prior generated seedbed, or “recirculation startup” (e.g., after the reactor has been employed in a previous production campaign, the production campaign stopped, and it is desired to restart). Preferably in such situations, the reactor may be left in a partial circulation mode to keep the seed bed from settling entirely, in some instances left in full circulation to maintain fluidization of the bed. A controlled startup or “re-start” procedure is desirably used for circulating restarts such that the reactor crosses quickly into condensed mode to prevent startup fouling, and a moderate space velocity in the bed is maintained for catalyst feeding as discussed previously. The procedure preferably include these elements: (1) maintaining a space velocity within a range from a low of 0.50, 0.55, 0.60, 0.65, or 0.70 m/s to a high of 0.65, 0.70, 0.75, 0.80, 0.85, or 0.90 m/s, provided the high end of the range is greater than the low end of the range (e.g., a range of 0.65 to 0.75 m/s) for at least the first 20, 25, 30, 35, 40, 45, 50, or 55 minutes of catalyst injection to minimize entrainment and then adjust space velocity to line out at the steady state target before transitioning to condensed mode, (2) from the initial catalyst injection rate, increase catalyst to achieve a dewpoint cross of +5 °C to -5 °C dewpoint approach in 25, 30, 35 or 40 to 45, 50, or 55 minutes; (3) once the reactor is operating in condensed mode with the dewpoint approach falling and reaching -5 °C to -8 °C, catalyst injection is preferably reduced (optionally in increments as discussed previously), and (4) maintaining a bed level (118 in the Figure) within a range of 0.75 to 0.85 m greater than the neck level for 1, 2, 3, 4, or 5 BTOs (e.g., for at least 2 BTOs), which may aid in scrubbing the reactor above the neck level.

[0057] Further, in some embodiments, it is desired to be at steady state conditions (e.g., steady state superficial velocity, where steady state space velocity is greater than at start-up) prior to carrying out a dewpoint cross (i.e., moving the reactor to condensed mode) as described above.

[0058] The reactor cycle gas has a certain dewpoint: the temperature at which the cycle gas first begins to condense at a given pressure. Dewpoint cannot be measured directly; however, it can be estimated thermodynamically. In a typical gas phase reactor, the dewpoint can be estimated for a given cycle gas continuously based on process measurements (cycle gas composition and system pressure) using an equation of state. Dewpoint approach (DEWAPP) is defined as the cycle gas cooler outlet temperature minus the cycle gas dewpoint. In dry mode operation, DEWAPP is greater than zero: the cycle gas cooler outlet temperature is higher than the cycle gas dewpoint. In condensed mode operation, DEWAPP is less than zero: the cycle gas cooler outlet temperature is lower than the cycle gas dewpoint and thus the condensable components in the cycle gas appear as a liquid. The precise transition from dry mode to condensed mode occurs when DEWAPP is equal to zero.

[0059] The DEWAPP zone is defined as the region where 5 °C > DEWAPP > - 5 °C. When operating a reactor in dry mode and transitioning into condensed mode, the DEWAPP zone is crossed. On a reactor startup, DEWAPP becomes progressively less due to the increasing concentration of condensing agent in the cycle gas (increases the cycle gas dewpoint) and the increased production rate due to catalyst loading (requires the cooler outlet temperature to be lower in order to control the reactor bed temperature to target). For metallocene startup, the operator can specify a target rate of transitioning from dry mode to condensed mode in terms of how fast it is desired to cross the DEWAPP zone. According to some embodiments, crossing the DEWAPP zone in less time may help avoid plate fouling (e.g., crossing DEWAPP zone in a maximum of 65, 60, 55, 50, 45, or even 40 minutes). On the other hand, a minimum amount of time may also be desired in order to avoid polymerization rate and heat generation from increasing too quickly (e.g., at least 15, 20, 25, 30, 35, or 40 minutes). [0060] Once DEWAPP < - 5 °C is reached and the reactor is comfortably in condensed mode, the startup focus shifts to maintaining adequate temperature control. Typically, elevated levels of catalyst have been used to drive the reactor through the DEWAPP zone and this may overwhelm the steady state capacity of the cooling system. Therefore, when DEWAPP reaches - 8 °C, catalyst injection rate is proactively reduced in multiple increments (again, with example increments noted above).

[0061] As noted previously, it is common to add hydrogen to the reactor as this is used to regulate the molecular weight of the forming polymer. Preferably, hydrogen is continually flowed at suitable rate for target MI as compared to flow rate of other reactor inputs (e.g., monomer/comonomer). For instance, according to various embodiments, hydrogen is maintained at or above 0.01, 0.02, or 0.03 kg/hr flow rate (again, depending upon reactant flow rates).

[0062] As used herein, the phrase “consisting essentially of’ means that there may be minor apparatus features present, such as valves, heaters, coolers, and pumps that facilitate the operation of the claimed apparatus or cycle but are not essential to the operation of such apparatus or cycle. Likewise, as it relates to a process claim, “consisting essentially of’ does not exclude minor features such as a valve operation, heating/cooling, and pumping of gases, fluids and/or solids that are not essential to the process as claimed: the polymerization of olefins catalyzed with metallocene catalyst to form polyolefin granules with reduced or eliminated reactor fouling, upsets, and/or the formation of agglomerations that halt commercial operations.

[0063] All publications, patents and patent applications mentioned in this Specification are indicative of the level of skill those skilled in the art to which this invention pertains and are herein incorporated by reference to the same extent as if each individual publication patent, or patent application was specifically and individually indicated to be incorporated by reference.

[0064] The invention being thus described, it will be apparent that the same may be varied in many ways. Such variations are not to be regarded as a departure from the spirit and scope of the invention, and all such modification as would be apparent to one skilled in the art are intended to be included within the scope of the following claims.