MXPA97002293A - Process to control a depolimerization reactor of fluidized battery of phase gase - Google Patents

Abstract

The present invention relates to control of a gas phase polymerization reactor is performed with the product melt index and the reaction temperature, and the partial pressure of the reagent of

Description

PROCESS TO CONTROL A FLUJ BED POLYMERIZATION REACTOR. GLYSE PHASE FIELD OF THE INVENTION The invention relates to a process for controlling the operation of a gas-phase fluidized-bed reactor to reduce the time and volume of transient operation when switching from one type of product to another or controlling fluctuations in the manufacture of gas. constant state.

BACKGROUND OF THE INVENTION The fluidized bed technology in olefin polymerization reactors currently used can be adjusted to produce a wide variety of products. This is particularly the case for polyethylene manufacture. It is not unusual to demand that a system produce resins that can be used in injection-molded, blow-molded, rotomoulding, wire-coating, tubing and pipe and film products. The fluidized bed technology can be used to make a wide variety of polyolefin products, e.g., homopolymers and copolymers of polyethylene, polypropylene, to C4-C12 α-olefins; ethylene-propylene-diene monomer (EPDM), polybutadiene, polyisoprene and other rubbers. In general, polymer products made by a given reactor system use the same reagents but in different ratios and at different temperatures. Each of these polymer products can be made with a number of different properties, or grades, of resin. Each grade of polymer product has a narrow limit on its properties, eg, density and melt index. The length of time a reactor is used to make a particular type of polymer depends on the market demand for the product. Some products can be made for weeks without change. Other products are made for shorter periods. Unfortunately, industrial reactors require time to adjust to new conditions (eg, temperature, reagent pressures and reagent ratios) and produce material in the interim that is constantly changing but not within the properties (v. ., melt index and density) of either the old or the new product. New products can not be made instantaneously and require a quantifiable period of transition to be adjusted to the new, desired conditions. Similarly, reactors - operating at fixed conditions, that is, at "constant state" - may experience fluctuations that may result in the production of "inferior" material. This lower quality material represents an economic loss and is minimized - desirably. In general, industrial control systems for gas phase, fluidized bed polymerization reactors are designed to allow operators to control the area allowing operators to select a desired melting point and density index. The correlations of these properties are usually well known to operators and those in the field for particular reactor design and used catalyst. The prior art has designed a number of methods for reducing the transient, lower material. These methods typically invoke some combination of adjustment of the automatic flow / ratio controllers to a new value either at or above the value finally desired ("marked transition" and "overmodulated"), eliminating fully the reactive gas (" inventor depleted "), reducing the catalyst level (" low bed "), and adding a non-reactive gas (" nitrogen addition "). DE 4,241,530 describes using an extermination gas to stop a polymerization reaction, blowing the gas inventory for that reaction out of the reactor, and reconstructing a new gas inventory for a new product. This method reduces the transition material. The costs associated with pulling the gas inventory and rebuilding a new inventory are too high for commercial transitions between closely related grades. In this way, most transitions between grades of the same material are made by adjusting the reaction conditions. Mcauley et al. ("Optimal Grade Transitions in a Gas Phase Polyethylene Reactor," AIChEJ., Vol 38, No. 10: 1992, pp. 1564-1576) describes three manual transition strategies, in-tensile in work for gas-phase polyethylene reactors. The first is an adjustment to the controls to overmodulate the melt index and density values. Hydrogen feed does not and the comonomer feeds are increased to fill the designated properties. The actual desired set point values are directed when the sensors indicate that the desired product is being produced. The second is an increase in temperature and management of slow ventilation to move the melting rate of the product produced. The third is a drop in the catalyst level of the lower bed while the residence time of bed resin is maintained at a constant to redjust the lower production. Debling et al., "Dyna ic Modeling of Product Grade Transitions for Olefin Polymerization Processes," AIChE J., vol. 40, no. 3: 1994, p. 506-520) compares the operation of transmission of different types of polyethylene reactors. The article describes seven manual, labor intensive, separate transition strategies: (1) mark the final goal transition; (2) • deplete the gas inventory and simple marking transition; (3) low bed and simple marking transition; (4) depleting the gas inventory and overmodulating the melt index and the release ratio; (5) low bed, depletion of gas inventory and overmodulation transition; (6) low bed and overmodulated transition; and (7) deplete gas inventory, overmodulate and bring nitrogen addition.

Despite this wide variety of available schemes there is a continuing need and desire to reduce the amount of inferior material produced during the transition to a new product grade or during continuous state manufacturing.

SUMMARY OF THE INVENTION An object of the invention is to provide a method for reducing the amount of inferior material produced during the grade transition or during the continuous state fabrication. Another object of the invention is to provide a method - to reduce the transition time and the volume of transient material when changing from a polymeric product to another similar chemical product but with different properties. In accordance with these and other objects of the invention which will become apparent hereinafter, a process according to the invention is used in a reactor with adjustable set point values of melt index, product reaction temperature, partial pressures of reagent and levels of catalyst in the reactor producing a polymerization product in a fluidized bed of catalyst when changing from a first product made at a first temperature and first set of conditions to a second product made at a second temperature and set of conditions, either in transition between products or during continuous state manufacture of a specific product that undergoes "within grade" fluctuations, and - comprises the steps of: (a) comparing the reaction temperature of first product and the second product reaction temperature, change-the set point of product reaction temperature to the temperature of rea tion of the second product if said second product reaction temperature is lower than the reaction temperature of the first product, (b) adjusting the fixed point of melting index which is -0-150% higher or 0-70% lower than the desired value of index-fusion of second product, (c) adjust the fixed point of reaction temperature which is: 1 - 15 e C above the desired reaction temperature of second product if the value of melt index of second The product is greater than the melt index value of the first product, or 0- 159 C lower than the actual reaction temperature of the second product if the melt index of the second product is lower than the melt index of the first product. (d) adjust a set point of partial pressure of or product rate limiting reagent that is 0.07-1.76 kg / cm gauge or less than the partial pressure of the first product regimen irrigating reagent if the second index value of The merger is greater than the melt index value of the first product, or more than the limiting reagent partial pressure-product regime of the second product if the melt index value of the second product is less than the melt index value. of second product; (f) changing the fixed point of the melting index to the desired value of the second product melt index; (g) changing the fixed point of the product reaction temperature to a value that is: (i) 0- 159 C higher than the desired second product reaction temperature if the melt index value of second product is higher than the melt index value of first product, or (ii) 0-159C lower than the desired second product reaction temperature if the melt index value of second product is lower than the index value of fusion of first product; (h) changing the setpoint of the partial pressure of the regime limit to a value that is: (i) Manometric 0-1.76 kg / cm either lower than the limiting partial pressure of the desired second product regime if the value of Fusion indicium according to do is higher than the first melt index value, or (ii) 0-1.76 kg / cm metric hand above the limiting partial pressure of second product regimen if the index value of - second melt it is infeiror that the first index value of fu-; e (i) changing the set point of reaction temperature and the set point values of partial pressure-of regime-limiting reagent to the desired reaction temperature of second product and partial pressure value of second rate limiting reagent when the The reaction product exhibits an average melt index value within acceptable limits of the second product melt index value. With the present procedure, the transient time and the amount of inferior material produced during that transient period are reduced through the control of the set point of reaction temperature as well as the point pressure established from the concentration of monomer limiting the reaction temperature. regime. The control system can further reduce the pull time by providing the removal of hydrogen in the reactant gas. The control procedure of the invention is well suited for automated control over the polymerization reaction system utilizing computer monitoring of the product properties and setpoint value adjustments.

BRIEF DESCRIPTION OF THE DRAWINGS Figures 1 and 2 are flow charts illustrating the steps of the prior art used to conduct a manual transition from a first product to a second product. Figures 3-5 are flow charts that compare manual control to an automated transition process that links to an existing dynamic model of the reactor system.

DETAILED DESCRIPTION The present invention is directed to a method for reducing the volume of lower material in an olefin polymerization reactor employing a rate-limiting olefin gas. Specifically, the volume of inferior material produced when changing from a first product to a second product grade or fluctuations within a designated "grade" of durable product is reduced by adjusting the fixed point of temperature as well as the fixed point. of partial pressure on the regime limiting reagent. The combination of these two controls accelerates the speed with which the reactor moves towards the production of the desired product. The volume of inferior material, of course, is directly related to the time it takes for the reactor to change from a first set of conditions (temperature, partial pressure, monomer ratios, etc.) to a second set of desired conditions. The air pressure of the regime limiting reagent is a clue. The concentration of hydrogen in the gaseous reactor inventory can also be used to reduce the transition time. Because the hydrogen ends the polymerization reaction, even small amounts of hydrogen in the reactor can have a pronounced impact on the degree of the average product melt index. The concentration of hydrogen can be adjusted by bleeding a relatively small amount, e.g., 1-8% by weight, preferably about 3-6% by weight, of invantant gas or passing a proportion of the total inventory - of gas on hydrogenation catalyst in a satellite system with a fixed or fluidized bed. The hydrogenation will convert some amount of the olefin to non-reactive alkane which would constitute a diluent. The present control process can be carried out in a charity of appropriate reactor equipment to perform catalytic, gas phase, fluidized bed polymerization. One or more reactors may be used in sequence or in parallel, usually, said reactors shall be designed for commercial operation and shall have appropriate controls that allow adjustable steady-state values for melt index, product reaction temperature, reagent ratios, Feeding ratios, reagent partial pressures and catalyst levels in the reactor The most preferred reactor is sold under the trademark UNIP0L and is available from Union Carbide Corporation, Danbury, Connectj_cut. US Nos. 4,302,565 and 4,482,687, the teachings of which are incorporated herein by reference.Any polymerization catalyst that can be used in the reactor can be controlled with a process-control sequence of the present invention.The appropriate catalysts include those of metals of transition, Ziegler-Natta, rne talocene and rare earth components. or it can be soluble, insoluble, sustained or unsupported. The polymers that can be produced using the method of the present invention are generally olefin polymers. The exemplary products include omopol ethylene; propylene homopolymers; copolymers of ethylene and at least one alpha-olefin of C-C12 'terpolymers of ethylene and at least one alpha-olefin of C3C12 and a diene. Examples of specific product polymers that can be made include ethylene-propylene copolymer, ethylene-butene copolymer, ethylene-ethyl acetate copolymer and ethylene-propylene-diene rubbers, propylene copolymers and at least one alpha olafin of C.-C. ^ (such as propylene-butene copolymer and propylene-hexane copolymer), polybutadiene and polyisoprene. The reaction conditions and catalysts used in the controlled process are those conventionally used to make the designated products. In general, the inventory of gas in the reactor will be completely replaced up to about 5 times per hour (* 5 GHSV), preferably approximately 0.5-3 GHSV. The inventory of catalyst in the fluidized bed changes every 2-4 hours (0.25-0.5 CSV). The control process of the invention utilizes the set point partial pressure of the rate limiting reagent to accelerate the transition to the desired product. For any specific product, the regime limiting reagent will be known to those experienced in the field. In general, however, ethylene is the regime-limiting reagent when polyewned with oxtene, hexene, octene, or a combination of butene and hexene; propylene is the rate-limiting reagent when polymerized with butene, hexene, octene or a combination of butene and hexene; ethylene is the regime-limiting reagent when polymerized with propylene when it is made of ethylene-propyle (not EPR) rubber. For homopolymers, the rate-limiting reagent is the monomer concentration. The invention is conveniently described herein with reference to ethylene-hexene copolymers in which ethylene is the rate-limiting reagent. The transition generally starts from an initial operating condition in which a first product exhibiting a melt index value of first product (or any other e_x pressure of a change in molecular weight, eg, flow rate, - flow relationships and related normal methods) is done at a first fixed point of product reaction temperature and a first fixed point of regime limiting reagent concentration. The second desired product will have different values of melting index and reagent feed ratio which are higher or lower. Changes in temperature and values - partial pressure of regime limiting reagent are used - to reduce the time required for the catalyst bed to start producing the new product. Importantly, the transition can occur between constant state operating conditions or within a subset of conditions within upper and lower limits acceptable for a particular constant state manufacturing process. Certain changes of fixed point are applicable to change and control the transition from one product to another. Depending on the specific system and the degree of computer control available over the process, operators can change the fixed points for melt index, reaction temperature, hydrogen feed rate, regimen feed rate regimen, and / or partial pressure, feeding regime - of comonomer and / or partial pressure, and level of catalyst in the reactor. Generally, a change in the desired melt index depends on the rate at which hydrogen is introduced into the system and is to provide the change in the fixed point of reaction temperature. Conversely, a partial pressure of regime-limiting reagent and hydrogen removal rate are inversely proportional to the change in melt index. Another manifested form, an increase in melt index value is accompanied by an increase in temperature, but a decrease in partial pressure of regime-limiting reagent and rate of fu- trogen removal. For the control process of the invention, the set temperature point can be changed in advance of time with a change in partial pressure of ethylene and fixed point of hydrogen feed rate to cause a change in the average melt index of the reaction product - inside a catalyst bed. A change in the fixed point of the melting point, regardless of whether it is directly adjusted or calculated, should be approximately 0-150% higher than the fixed target point for the new product if the melting index of the second product is higher than the first product or 0-70% lower if the melting index of the second product is lower than the melting index of the first product. The change in melt index will be accompanied by a change in hydrogen concentration. The average melt index can be determined by measurements on samples that are removed and tested or by automated, on-line sampling. Actual sampling can be done every 2-4 hours. In online samplers they will perform these tests 3-4 times per hour. The control over the fixed points of temperature and rate-limiting reagent can then be used to accelerate the rate of transition to a new product without exceeding the desired equilibrium reaction conditions within the catalyst bed. After the new fixed point of fusion index is adjusted, the fixed temperature point is changed to a value within the scale of about 1 - 15 s C from the fixed point of -meta for the constant state manufacture of the new product - (above the old fixed point if the melting index it is adjusting, or below the old fixed point if the melting index is decreasing) and changing the fixed point of partial ethylene pressure to a value within the scale of approximately 0.07-1.76 kg / cm2 anometric point fixed goal (below-the old fixed point if the fuson index is increased, or above the old fixed point if the melt index is decreasing). The actual implementation of such changes will depend on the existant material, but changes can occur through discrete step increments or a smoothly varying change. When the fixed point values of temperature and ethylene partial pressure are reached, the reactor is maintained within _ + 10%, preferably within + 5% of the fixed point values until the reaction product within the bed of catalyst starts to be made at the target fusion value for the new product. The total reactor pressure may also be allowed to fluctuate within upper and lower limits generally designated around + 20% and preferably around within + 10%. The additional control over the target reaction temperature and the fixed points of regime limiting partial pressure will increase the rate of transition and will reduce the amount of time necessary to start making product within acceptable limits 9es devir, melt index and density) . it should be remembered that the gaseous inventory in a polymerization reactor can be changed to the new target composition much more rapidly than the catalyst bed can start to produce the new product. For example, the gas composition can be adjusted in about 15 minutes, but the reactor will usually not be producing the new product for 2-6 hours even when controlled by the process of the invention. This inertia through the bed is quantifiable through appropriate, known differential mass transfer equations that are used to generate computer models of the reaction system. In this way, the changes in temperature and partial pressure of the regime limiting reagent act as forces on the reaction system to cause the change. When the product melt index, as reflected by sampling, is at a point where the inertia of the bed suggests a change or is within acceptable upper and lower limits of the desired product, the fixed points are changed back to the fixed points of constant state goal for the new product, so that the reactor can start to achieve constant state operation with a consistent product grade of polymer. The technique refers to this reduction in the regime of changes with the "return" phase of a transition, that is, a return to constant state. During the return phase, the temperature and the partial pressure of the rate limiting reagent are adjusted to constant state goal values to reduce the force of change on the bed and approach the constant state. The fixed point reaction temperature is readjusted to a value that is 0-15 QC either lower than the target point if the melt index is rising to the new target value or higher than the target value if the target index Fusion is decreasing towards the new metal value. Similarly, the ethylene partial pressure is adjusted to 0-1.76 lg / cm gauge below the target value if the melting index is rising to the new target value or to 0-1.76 kg / cm gauge per gage. of the target value if the index of -fusion is decreasing towards the new metal value. Because the rate of change in the product melt index is towards making it slow and approaching a value within acceptable upper and lower limits around a target index value, instant melt index readings should be monitored frequently and carefully. When the melt index of the reaction product falls within upper and lower acceptable limits, the fixed points of temperature and ethylene partial pressure are adjusted to the target values. The reactor is then controlled to maintain steady state operation. In some cases, the operator's experience or computer models can be used to determine the time to change the fixed points by recognizing the risk that fixed-point changes that are too early or too late will prolong the return time, will increase the transition time and will increase the amount of lower material produced.

Transition Control with Computer Model Additional reductions in the volume of transition and less intense work in constant state manufacturing control can be achieved by controlling fixed temperature points, partial pressure of regime limiting reagent and hydrogen removal from of an existing computer model of the -reactor. This model can be used as an expert database that will control the transition strategy and the critical food programming. Such a system will usually involve a series of "IF-THEN" manifestations and conditional branching looking for detailed answers from a database of previously identified responses based on the specific system being used. The details of generating a computer model of a reactor transition and steady-state operation properties are within the existing level of experience of practitioners in the field, so that an extensive discussion of the techniques and relationships is not provided. mathematical See, Ignizio, Introduction to Expert Systems: The Development and Implementation of Rule-Based Expert Systems, McGraw-Hill (1991) whose exposition is incorporated herein by reference. An expert control system is particularly useful - to automate temperature control, rate limiting reagent and hydrogen removal values and the timing of any changes. For ease of explanation, the expert system is described with reference to a transition to a product with a higher melt index that requires an underestimated fixed point of temperature ("underpulsed") and fixed points of regime limiting reagent and removal of In fact, the hydrogen removal regime generally adjusts to "0" to retain all the excess hydrogen, the same principles and importance will amplify to the opposite case, that is, transition to an excess of hydrogen. Lower melting index that requires a fixed point value of underestimated temperature and partial pressure of over-rated rate limiting reagent and fixed values of hydrogen removal at the beginning of the transition The same control principles and controls used for the transition between degrees also are used to control constant state manufacturing by monitoring product properties and correcting or the conditions if it appears that the product properties are beginning to move from a desired medium to one of the extremes of the acceptable limits. The automated supervision of the product produced with a cadro of -data of the upper and lower acceptable limits as well as a middle-set of property values 9v.gr., the region in a graph of volume of product-properties where at least 50% of the product volume must fall) will allow the computer to take the corrective value before inferior material is produced. The system will monitor the product properties and, if necessary, readjust the fixed points from initial values to new values as a way of forcing the reactor to make material "in grade" of consistent quality. The use of a computer has a number of advantages - over manual operation. As a minimum, automated controls reduce the need for competent human beings to take corrective action and make the fixed point process easier. Importantly, computerized control allows an operator to directly enter the new melting point and density index values. With an appropriate database that correlates these values with the reactor, the catalyst, feed ratios, etc., the computer will interpret the necessary overshoot / sub pulse of temperature, partial pressures of reagent and reagent ratios as well as regime values of hydrogen removal. As noted in the foregoing, the retention of an overdriven value for too long or too-short increases the transition time and the volume of inferior material that occurs. With a control system linked to a precise computer model of the reactor by a series of appropriate "yes-then" events, the timing and dune of any over-pulsed or underestimated fixed-point values can be controlled automatically. This is particularly useful for having the model running on the same computer as the control system so that the system data can be written more or less continuously to a common data frame that is read by the control system for changes in points. fixed and information of duration updated. Such a model-controlled transition is able to adjust fixed-points from the reaction product information of the model more frequently than is feasible with manual adjustments based on property measurements still online. The transition times and lower material in this way are reduced to a minimum.

Figures The invention can also be described with reference to the appended figures. Figures 1 and 2 illustrate a typical flow chart of the prior art for manual control of the transition from a first product to a second product of higher melting density and density. The reactor is operating initially in a constant state making a first product. Then a decision is made to switch to product 2 with a new "recipe", ie, reagent ratios, reaction temperature hydrogen / monomer ratios, etc. The operator then allows the level of the catalyst bed to fall to a low level of transition state. Four operations are then performed more or less simultaneously. First, the catalyst bed is maintained at its transition state level, low, for 2 hours and then it gradually increases to its normal production, normal, second, a new fixed temperature point is entered to make the second product towards the temperature control panel and it is left to change to the new fixed point and remains constant. Thirdly, and finally, the operator selects new ratios of hydrogen to ethylene, the "hydrogen ratio", and for the comonomers that "overshoot" the desired value, that is, the value of the fixed point is per mass. of the goal value, constant state for the new product. The selection of the amount of overdrive is done by experience and is maintained at that fixed point for approximately 2 hours. The relationships are then readjusted to the target values for the new product and maintained at that value until the reaction product exhibits the desired values of melt index and density. The time elapsed for the transition is approximately 12 hours. Figures 3-5 illustrate a flow chart of the process control of the invention. The initial stages are similar: the reactor is operating initially at a constant state, -the level of the bed is reduced to a transition level and is maintained there for 2 hours. Unlike the previous process, when the temperature of the new product simply adjusted, the temperature of the second product is checked to see if it is above or below the temperature of the first product. If the second temperature is lower, the fixed property points are not changed, and the reaction temperature "slides" downward-in stepped units or a smooth gradient. If the second temperature is higher than the first temperature, the fixed property points are changed at the same time the new fixed temperature point is entered. This analysis is particularly helpful in the operation of the reactor. Because the resins are made almost close to their point of tackiness, operating the reactant at converging temperatures during the transition helps prevent adhesion. In the expert process 2, the computer model is used to adjust the temperature, the partial pressure of the regime limiting reactor (eg, partial pressure of ethylene or "C2PP"), and the rate of hydrogen removal. (eg, slow ventilation) values of undershoot or overshoot and dura- tion of the fixed-point values of the melt index and density regi- lated to the computer model in relation to the initial fixed-condition conditions. Changes in reaction product and reactor conditions are monitored and corrected to maintain the optical transition tray from the reactor model. The fixed points are changed as necessary to minimize the transition time. A computer controlled system can also be applied as a quality control system to monitor the manufactured product and guide the transitions back to a desired scale of constant state operation, ie the first product (product in deviation) and the second product (goal product) are within acceptable limits of the same degree of particular desired product. Such systems can be controlled by the principles described herein to change the applicable fixed points and force the system to produce products within acceptable limits of the desired products. The present invention can be explained with reference to the following examples.

EXAMPLES Example 1 The control method of the invention was applied to a gas phase polymerization reaction between ethylene and hexen in a fluidized bed reactor. The transition from a first grade of product to a second grade of product required the following changes: an increase in melt index of 56%, -an increase in density of 0.023 g / cm, and a decrease in temperature of 20QC. The catalyst was changed from a titanium-based liquid to a dry catalyst based on titanium, but the level of the bed remained constant. The transition was started by changing a number of fixed point control values. The melt index was adjusted to 25% above the target value for the desired new grade. The fixed density push was adjusted to 0.002 g / cm below the target value for the new grade. The temperature was dropped in increments at 29C above the target temperature to make the new grade steady state. The partial pressure of ethylene p was reduced by 3.23 kg / cm gauge. The molar ratios of hydrogen to ethylene (H / C2) and hexene to ethylene (Cg / C2) were calculated. continuously from these new fixed points. Large amounts of hydrogen and hexer were added to the reactor. after 5 hours, the average melting rate of the metal was at the metal grade. At this point, the fixed point of the melting point was changed to the target value of the final product. The fixed point of temperature was lowered the remaining 2eC to the target reaction temperature of the final product. The partial pressure of ethylene moved up 0.21 kg / c metric hand. Slow ventilation was opened for 45 minutes to allow excess hydrogen from the reactor inventory. After 6 hours, the average density of the bed is within acceptable limits of the density of met ^ i. The fixed density point was then changed to the target value for the final nro-final. The melt index remained on its metal, so that all controls were changed to constant state mode.

Example 2 Traneicir control, manual indicative of! The method of the prior art was applied to the gas phase reaction of Example 1. For the products, it was applied to the products. ini ia '»fi? It was a dry cement, based on titanium. The transition began by increasing the rate of hydrogen fed to the reactor until an instantaneous melt index of 25% above the final target value is reached. The hexene alimejite was opened up until the density was at the metal vilor. The fixed temperature point was dropped to the final target value. The reactor pressure remained substantially constant. Large amounts of hexene and hydrogene were added. After the envelope impulse impulse was fired, the relation of W was maintained. C 2 - From "" and - "-" similar, the value of Cg / C2 was maintained when the density falls within the range of the target value After 5 hours, the fixed point was changed of H2 / C2 to a calculation of the final value needed to cause the melt index to fall within the target limits No changes or manipulations were made to either the temperaure or the slow ventilation.The partial pressure of ethylene was controlled for If the total pressure of the bed was within the target limits, the H2 / C2 and Cg / C2 controls were adjusted to constant state.This old control method produced approximately double the amount of material less than the process of -example 1.

Example 3 The transition control process of the invention was applied to a fluidized bed with the following product changes: (a) Change in melt index: 71%; (b) Change in density: -0.011 g / cm; and (c) a decrease in temperature by 2SC. The type of catalyst and the bed level were not changed to make the new product grade. The transition was made with the following changes: (a) Fusion index adjustment to 23% less than the target product value; (b) Density fixed point adjustment to 0.0002 g / cm over the target product density; (c) Maintain the fixed temperature point at the same value as needed for the initial product; (d) Increase the partial pressure of ethylene by 0.42 2 kg / cm gauge while continuously calculating fixed points of H2C2 and Cg / C2 based on the fixed point values of melt index and density; and (e) Open venting to release 5% of the reactor gas inventory per hour. after 2.5 hours, the melt index and density reached their set point values. Slow ventilation was then closed. The established points of fusion index and density were maintained. The H2 / C2 and Cg / C2 ratios were also maintained. The melt index was then adjusted to the value of the new product. The temperature was then reduced 2SC to the metal reaction temperature. The partial pressure of ethylene was increased by 0.21 kg / cm gauge. Slow ventilation was opened for 45 minutes to release 8% of the gas inventory per hour, eliminating excess hydrogen from the reactor. After 6 hours, the average density of the product - which is being made in the bed was within acceptable limits of the metal values. The established density point was then set at the value of the new product. All controls -changed to constant state.

Example 4 The control method of the invention was applied to a reactor operating at a constant state. The constant state product has a fixed point of target melt index of 104 kg / min and an acceptable product melt index of + 10 kg / min around that fixed point. The desired product property distribution is to have 50% of the product melt index within the range of 99-109 kg / min. in one period in manufacturing, the product melt index was raised to a value greater than 109 dg / min as measured - by an inline melt index device. The automatic control system changed the fixed point of fusion index to 99 dg / min and calculated that the reactor contained an excess of hydrogen. Slow ventilation was opened to release 8% of the reactor gas inventory per hour.

After 45 minutes, slow ventilation was closed. The melt index, as measured by the on-line sensor, reverted to within the 50% central grade region. The set point of the melt index was returned to 104 dg / min. No inferior resin was produced during the controlled period,

Claims (11)

CLAIMS:

1. A process for controlling a gas phase polymerization reaction in a reactor when changing from a first product made to a first set of conditions to a second product made to a second set of conditions, the process comprising the steps of: a) compare the reaction temperature of the first product and the reaction temperature of the second product, change the set point of product reaction temperature to the second product reaction temperature if the reaction temperature of the product is lower than the temperature of - reaction of the first product, (b) adjusting a set point of melting index - which is 0-150% higher or 0-70% lower than the desired melting index value of the second product; (c) adjusting the set point of reaction temperature which is: 1-159C above the reaction temperature - desired of the second product if the melt index value of -second product is higher than the melt index value of the second product. first product, or 1-159C below the reaction temperature of the second actual product if the melt index of the second product is lower than the melt index of the first product; (d) adjust a set point of partial pressure of product rate limiting reagent that is: 0.07-1.76 kg / cm gauge either below the partial pressure of first product rate limiting reagent if the melt index value of the second product is greater than the value of the first product melt, or above the partial pressure of the first product rate limiting reagent if the melt index value of the second product is lower than the index value of the first product. fusion of first product; (e) maintaining the fixed point of melting index, the set temperature temperature and the fixed point values of the partial pressure of the rate limiting reagent until the polymerization product exhibits an average melt index and a density of the product. average with an acceptable scale of the desired value of the second product melt index and the density value of the second product; (f) changing the set point of the melt index - to the desired value of the melt index of the second product; (g) changing the set point of reaction temperature of the product to a value that is: (i) 0- 159C above the desired reaction temperature of the second product if -the melt index value of the second product is greater than the melt index value of the first product, or (ii) 0-159C below the desired reaction temperature of the second product if the melt index value of the second product is lower than the melt index value of the first product. product; (h) changing the set point of partial pressure from the rate moderator to a value that is: (i) 0-1.76 kg / cm manometric below the partial pressure limiting regime of the second desired product if the value of second melt index is higher than the first melt index value, 9 (ii) 0-1.76 p kg / c gauge above the limiting partial pressure of second product regime if the second melt index value is less than the first melt index value; * e (i) change the set point of reaction temperature and the fixed point values of the partial pressure of the rate limiting reagent to the reaction temperature of the second product and the partial pressure value of the second rate limiting reagent when the product of The reaction exhibits an average melt index value within acceptable limits of the second product melt index value.

2. A process according to claim 1, wherein the melt index of the first product is higher than the melt index of the second product and the process further comprises: removing hydrogen from the gas inventory inside the reactor.

3. A process according to claim 2, where the hydrogen is removed: ventilating 1-8% by weight per hour of the gas inventory.

4. A process of comfort with claim 2, wherein the hydrogen is removed by passing at least a portion of the gas inventory on a hydrogenation catalyst.

5. A process according to claim 1, wherein the regime limiting reagent is ethylene.

6. A process as in claim 1, wherein the first product and the second product are within acceptable limits of the same desired product, and the polymerization reaction is in constant state manufacture.

7. A process as in claim 6, wherein the first product is selected from the group consisting of ethylene homopolymers; propylene homopolymers; ethylene grouts and at least one C-C12 alpha-olefin; ter-polymers of ethylene, and at least one alpha-olefin of C-C 12 and a diene.

8. A process as in claim 7, wherein the first product is selected from the group consisting of ethylene-propylene copolymers, ethylene-butene copolymers, ethylene-hexane copolymers, and rubbers. of eti lenpropi leno-dieno, copolymers of propylene and at least one alpha olefin -of C4-C12, copolymers of propi lenhexane, polybutadiene and polyiso preño.

9. A process as in claim 1, wherein the first product and the second product are of different product grades, and the polymerization reaction is to change conditions of making the first product to conditions for making the second product.

10. A process as in claim 1, wherein the first product is selected from the group consisting of homopolymers of ethylene; propylene homopolymers; copolymers of ethylene and at least one C3-C12 alpha olefin; ethylene ter-polymers, and at least one C3-C.2 alpha olefin and a diene.

11. A process as in claim 10, wherein the first product is selected from the group consisting of ethylene-propylene copolymers, ethylene-butene copolymers, ethylene-hexane copolymers, and rubbers. eti lenpropi leno-diene, copolymers of propylene and at least one alpha olefin -of C4C12, polybutadiene and polyisoprene.