MXPA98007724A - Apparatus and improved process for a reaction of policondensac - Google Patents

Abstract

An improved apparatus and process for the production of a polyester or other condensation polymer is exposed. In particular, the polymerization is carried out in a reaction vessel equipped with a specially designed agitator which exposes the partially molten polymer by charging the reaction vessel with inert gas flowing through the vessel. The agitator comprises a plurality of elements that drive a portion of a molten polymer in the reaction vessel and generate films of the molten polymer which extend in planes that are parallel to the shaker shafts and to the gas flow through the vessel. reaction. In a preferred process, a melt of bis (2-hydroxyethyl) terephthalate, and its lower molecular weight oligomers, obtained by esterification of terephthalic acid or transesterification of dimethyl terephthalate with ethylene glycol, is contacted with an inert gas at about atmospheric pressure to remove reaction by-products and facilitate polymerization

Description

APPARATUS AND PROCESS FOR A REACTION OF POLYCONDENSATION FIELD OF THE INVENTION An improved apparatus and process for the production of a polyester or other condensation polymer is exposed. In particular, the polymerization is carried out in a reaction vessel equipped with a specially designed agitator which exposes the molten polymer within the reaction vessel of inert gas flowing through the vessel. The agitator comprises a plurality of elements that drive a portion of a molten polymer in the reaction vessel and generate films of the molten polymer which extend in planes that are parallel to the shaker shafts and to the gas flow through the vessel. reaction.

TECHNICAL BACKGROUND It is known to produce polyester of aromatic dicarboxylic acids or their esters such as dimethyl terephthalate (DMT), and glycols. This has been done by step-melt polymerization of the aromatic dicarboxylic acid dihydroxy ester, or weight oligomers REF. : 28127 molecular lower thereof, under successively higher vacuum conditions. In order for the polymerization to continue to the degree needed for most commercial applications, the condensation of by-products, especially ethylene glycol, must be removed from the reaction system to voids as much as 1-3 mmHg. Such processes require highly expensive vacuum equipment, multi-stage steam ejectors to create vacuum, and N2 purge seals and flanges to minimize air leakage in the system. The condensate from the steam ejectors and the organic by-products of the system produce a wastewater stream that requires treatment and contributes to volatile organic emissions to the air. The present invention relates to a less expensive polymerization process that can be carried out at atmospheric pressure.

Processes at atmospheric pressure that employ an inert gas have been discussed in the prior art, but these suffer from one or more drawbacks such as (1) the amount of inert gas used is too large to be economical; (2) reactor size could not be feasible for commercial scale operation; (3) the inert gas velocities could be too high to be feasible for commercial scale operation; or (4) the contact between the inert gas and the molten polymer in the reactor is inadequate or non-uniform.

Due to such drawbacks, the processes currently employed for commercial production of polyester continue to be carried out under high vacuum. An object of the present invention is to provide further improvement in a process, at about atmospheric pressure, for production of polyesters in continuous or intermittent form, particularly polyethylene terephthalate, of high molecular weight. In another aspect of the present invention, there is disclosed an improved apparatus that could be employed in the reaction process involving mass transfer of a volatile by-product into an inert gas.

BRIEF DESCRIPTION OF THE INVENTION The present invention relates to a process for making polyesters of aromatic dicarboxylic acids and glycols in a molten state in which an inert gas is used to help remove a by-product of volatile condensation, wherein the improvement comprises, using a reactor vessel. horizontally arranged cylindrical partially filled with a polymerization reaction mass in the form of a melt, this reactor is equipped with the following: a) a reactor inlet for introducing a polymerizable feed into the reactor vessel; b) a gas inlet to introduce an inert gas at or near one end of the reactor vessel and a gas outlet to remove the inert gas at or near an opposite end of the reactor vessel, thereby resulting in the flow of gas in front of the reaction mass in the reactor vessel; c) means for maintaining the reaction mass in the molten state; Y d) an agitator rotating on its axis during the operation, the agitator comprises a plurality of elements that are arranged longitudinally to carry a portion of the melt as said elements move through the reaction mass, the elements are positioned that said elements generate films, the planes of the films are parallel to the central axis of the agitator and to the flow of inert gas which is predominantly in the axial direction; Y e) an exit from the reactor to remove the polymer product from the reactor vessel.

The agitator according to the present invention is different from the agitators used in conventional vacuum processes, which essentially consist of rotating discs or grids. The agitators of the prior art generate films that are perpendicular to the axis of the reactor vessel.

In a preferred embodiment of the present process, the polymerization is carried out at atmospheric pressure. A dihydroxy ester of an aromatic dicarboxylic acid, or polymerizable oligomer of lower molecular weight thereof, is polymerized to a product with a higher degree of polymerization (DP), preferably in the presence of a polyester polymerization catalyst, wherein the byproducts of The polymerization is removed from the system by means of an inert gas. This higher degree of polymerization is useful in bottles, fibers and films. This process provides an improved method for producing linear aromatic polyesters, especially poly (ethylene terephthalate) (PET) The aromatic dicarboxylic acid used in the production of PET is terephthalic acid (TPA). The process could involve the production of poly (ethylene terephthalate) from terephthalic acid and ethylene glycol (EG) by esterification, or from dimethyl terephthalate (DMT) and ethylene glycol by a transesterification step, followed by polycondensation. The process is carried out at approximately atmospheric pressure or below, whereby high vacuum equipment is avoided and possible air pollution that causes decomposition of the product and gel formation is eliminated. First the terephthalic acid is esterified or the dimethyl terephthalate is transesterified with ethylene glycol to produce bis (2-hydroxyethyl) terephthalate or its lower molecular oligomers, which are then contacted in molten form with an inert gas. The volatile reaction by-products are removed with an inert gas, so that the polymerization is preferably completed in less than about 5 hours, more preferably less than 3 hours, of contact time while the reagents are maintained at an appropriate temperature to maintain them in a molten form to produce polyethylene terephthalate.

The above processes are preferably carried out in the presence of a polyester polymerization catalyst. However, a catalyst is not needed for the esterification step if the initiator material is terephthalic acid. In a preferred embodiment of the invention, a single stream of inert gas is recirculated through a polymer terminator step, a polycondensation step and a step wherein the ethylene glycol is recovered for reuse in the process.

The invention also relates to a novel apparatus described above for carrying out polycondensation or another reaction in which a volatile by-product is removed by mass transfer from a melt to an inert gas stream.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 depicts a schematic drawing of one embodiment of an apparatus that is suitable for carrying out the polymerization of the invention, wherein the material having a lower degree of polymerization is converted to the material having a higher degree of polymerization.

Figure 2 represents a schematic drawing of a rotary agitator frame embodiment.

Figure 3 illustrates a rotating shaker frame comprising an additional internal concentric "cage" formed by another group of stirring elements.

Figures 4a, b, c and d illustrate cross-sectional and isometric views of a stirrer that employs gratings as agitators for the generation of the film surface.

Figures 5a, b and c illustrate cross-section and side views of an agitator assembly consisting of concentric cylindrical wire separators.

Figure 6 illustrates concentric octagonal wire separators that can be employed in an agitator assembly.

These figures are for the purpose of schematic illustration and are not drawn to scale.

DETAILED DESCRIPTION OF THE INVENTION The polymerization according to the present process can be carried out in a container, or more than one physically separate series vessel, wherein the reaction mass is polycondensed to some degree of polymerization in a vessel and then transferred to another vessel for polymerization additional . The number of containers could depend on the mechanical considerations related to the handling of the polymer melt as its viscosity increases with the degree of polymerization, heat requirements introduced to volatilize the by-products of the reaction and cost.

Preferably, a single container could be used to convert a prepolymer to the final product having the desired degree of polymerization (DP).

The process of the present invention could be carried out intermittently or continuously. Intermittent production could be preferred to prepare specialty polymers when the required production is not very large and strict quality control is required particularly with respect to the additives. For large-scale production for product applications, such as molded resin, raw material, yarn, the effective cost to carry out the above steps is continuously increased where the reactants are fed substantially continuously into the processing vessel and the products are they remove substantially continuously. The product feed and removal rate is coordinated to maintain a substantially stable amount of the reactants in the reaction vessels while the inert gas flows countercurrent with respect to the flow of the melt.

If two or more vessels are used in series to carry out the polycondensation, it is preferred that a single inert gas stream is used which flows countercurrent with respect to the flow of the melt in the process, e.g. ex. the inert gas leaving a final stage of polymerization is carried out by means of the preceding stage and finally through a stage wherein the ethylene glycol is recovered for reuse and the inert gas is recirculated to the final stage of polymerization.

The preparation of higher molecular weight polyesters by melt polymerization from polymerizable monomers and / or oligomers is a well known process, see for example N.P. Cheremisinoff, Ed., Handbook of Polymer Science and Technology, Vol. 1, Marcel Dekker, Inc., New York, 1989, pages 87-90 H. Mark, et al., Ed., Encyclopedia of Polymer Science and Technology, 2nd Ed., Vol. 12, John Wiley & Sons, New York, 1988, pages 43-46, 130-135 and 217-225, all of which are included here by reference. The necessary conditions for these polymerizations, which in general are already known to the expert, are applicable to the polymerization here, modified as needed and described herein, to make various polyesters. These known features include points such as process temperatures and polymerization catalysts (any).

As an example, polyethylene terephthalate (PET) is made in this process by first reacting terephthalic acid (TPA) or dimethyl terephthalate (DMT) with ethylene glycol (EG). If the DMT is the initiator material, an appropriate transesterification catalyst such as zinc acetate or manganese is used for the reaction. In a preferred processDMT / TPA (trans) esterified is polymerized as a melt at atmospheric pressure or below by contacting the melt with an inert gas stream (eg, but not limited to, N2 or C02) to remove condensation by-products , mainly, ethylene glycol. Preferably, the inert gas is preheated to about the polymerization temperature or below, before its introduction into the polymerization equipment. It is preferred that the velocity of the inert gas through the polymerization equipment is in the range of 0.2 to 3 ft / sec, more preferably 0.3 to 1.5 ft / sec. The vapor leaving the polymerization (which contains the ethylene glycol by-product) could be treated to recover the ethylene glycol to recycle it to the esterification stage or for other uses. Then the inert gas stream could be cleaned and recirculated. Thus, the global process could operate as a closed cycle system which avoids environmental contamination and integrates the purification of ethylene glycol and its recirculation in the process.

The amount of inert gas flow should be sufficient to bring the ethylene glycol to be removed at a partial pressure of ethylene glycol below the equilibrium partial pressure of ethylene glycol with the reaction mass at the operating temperature. The operating temperature during the polycondensation is kept high enough to maintain the reaction mass in a molten state. Preferably the temperature range is approximately 270 ° C to 300 ° C. The polymerization equipment is designed so that the interfacial area between the melt and the inert gas is at least 20 square feet, preferably at least about 30 square feet, per cubic foot of melt, and that this surface area is frequently renewed. Under these process conditions, the high degree of polymerization useful for fibers and films and other uses can usually be achieved in less than 5 hours of residence time, preferably in less than 3 hours of residence time.

To more reliably produce good product quality of the desired degree of polymerization, the polymerization should preferably be completed in a reasonably short period such as less than 5 hours, preferably less than about 3 hours. This means that the overall residence time of the polymerization mass in the process is preferably about 5 hours or less, more preferably about 3 hours or less. The polymerization is considered complete when the desired degree of polymerization (DP) for a particular application is reached. For more common applications, such as fibers, the DP should be at least 50, preferably at least 60, and more preferably at least 70. By "degree of polymerization" is meant the average number of repeating units in the polymer, for example for the poly (ethylene terephthalate), the average number of ethylene terephthalate units in a polymer molecule. Exposure of the molten polymeric mass at high operating temperatures for prolonged periods can cause chain breakage and decomposition reactions with the result that the product fades and / or a high degree of polymerization is not achieved. If the speed of the inert gas is too low, the polymerization takes a long time. If the speed is too high this can lead to the entrance of the reaction mass in the gas. In a continuous form of operation, high inert gas velocities in a countercurrent direction can also hinder the flow of melt through the equipment. Also, higher speeds may require larger amounts of gas flow without substantially increasing the effectiveness of the polymerization.

The amount of inert gas flow used to remove ethylene glycol or other evolving volatile byproduct is high enough that the partial pressure of ethylene glycol or other by-product in the gas, at any point in the process, is low, preferably very low , that the equilibrium partial pressure of ethylene glycol with the melt at this point. The larger amounts of gas flow generally increase the polymerization rate but the increase is not proportionately larger. Therefore, very large amounts of gas are usually not necessary or desirable as large amounts could increase the size and cost of the recirculation equipment. Very large amounts may also require larger size polymerization equipment to maintain gas velocity in the desired range.

In the continuous mode of this invention, wherein the inert gas flows countercurrent to the flow of the molten reaction mass, the actual rates of polymerization can be achieved with about 0.3-0.7 pounds of N2 per pound melt (equivalent to about 2 to 5 moles of inert gas per mole of polymer repeat unit) provided that the inert gas velocity is at least about 0.2 ft / sec, preferably at least about 0.3 ft / sec. The flow of N2, however, should preferably be at least 0.2 lb / lb of polymer (equivalent to 1.5 moles of inert gas per mole of polymer repeat unit). However, larger amounts of gas flow may be needed to obtain the preferred gas velocities.

In the process of this invention, the reagent is maintained in a molten state, e.g. ex. , above its melting point, which for example is approximately 260-265 ° C for PET. At temperatures above 300 ° C, decomposition reactions frequently cause discoloration of the product which interferes with the quality of the product. For PET, the reaction mass should preferably be maintained from about 270 ° C to about 300 ° C.

For the polycondensation to continue, the ethylene glycol or other volatile by-product generated must be removed from the reaction mass by the inert gas. This removal is facilitated if there is a high surface area between the melt and the gas phase. To complete the polymerization in a reasonably short period, the surface area should be at least about 20 ft2 / ft3 of the melt, preferably at least about 30 ft2 / ft3 of the melt. A larger surface area is preferred to increase the polymerization rate. The reaction equipment for contacting the melt and the inert gas should also be designed to frequently renew the interfacial area and mix the molten polymer. This is particularly important as the degree of polymerization increases and the melt becomes more viscous.

The polymerization rate can also be increased by using an appropriate polymerization catalyst, particularly where a high interfacial area for inert gas-melt contact is provided. The increase in overall velocity, however, is not proportional to the catalyst concentration as the removal of ethylene glycol begins to limit the overall polymerization rate.

The catalyst could also increase the rates of decomposition reactions. An effective concentration of catalyst for a group of reaction conditions, such as temperature, gas flow, velocity and surface area, is such that it gives the greatest improvement in the polymerization rate without substantial decomposition. The optimum concentration of catalysts of various species are known in the art, or can be determined by experimentation. In general it would be in the range of a few parts per million parts per million of the polymer, such as about 5-300 parts per million.

The catalysts for facilitating polymerization are any or more polyester polymerization catalysts known in the prior art to catalyze such polymerization processes, such as, but not limited to, antimony, germanium and titanium compounds. Antimony trioxide (Sb2Os) is especially an effective catalyst that could be introduced, for convenience, as a solution of glycollate in ethylene glycol. Examples of such catalysts are found in U.S. 2,578,660, U.S. 2,647,885 and U.S. 2,789,772, which are incorporated herein by reference.

The polymers that can be produced by the present process include those derived from one or more aromatic dicarboxylic acids and one or more aliphatic or cycloaliphatic glycols. By an aromatic dicarboxylic acid is meant a dicarboxylic acid in which two carboxyl groups are each linked directly to a carbon atom of an aromatic ring. The aromatic dicarboxylic acid could otherwise be substituted with one or more other groups which do not interfere with the polymerization, such as alkyl groups, chlorine groups, alkoxy groups, etc. Examples of aromatic dicarboxylic acids used include terephthalic acid, isophthalic acid and 2,6-naphthalene dicarboxylic acid.

By an aliphatic or cycloaliphatic glycol is meant a compound containing 2 to 20 carbon atoms of the formula R OH), wherein R 1 is a divalent aliphatic or cycloaliphatic radical. If an aliphatic radical could contain one or more cycloaliphatic groups, and if a cycloaliphatic radical could contain one or more alkyl or alkylene radicals. It is preferred that R1 contains from 2 to 8 carbon atoms. The glycols used include ethylene glycol, 1,3-propanediol, 1,4-butanediol, 1,6-hexanediol and 1,4-bis (hydroxymethyl) cyclohexane. Preferred glycols are ethylene glycol, 1,4-butanediol, and 1,3-propanediol, and ethylene is especially preferred.

The preferred combinations of aromatic dicarboxylic acids and glycols are (the polymer produced is in square brackets) terephthalic acid and ethylene glycol [poly (ethylene terephthalate), terephthalic acid and 1,3-propanediol [poly (1,3-propylene terephthalate)] , terephthalic acid and 1,4-butanediol [poly (1,4-butylene terephthalate)], 2,6-naphthalene dicarboxylic acid and ethylene glycol [poly (ethylene 2, 6-naptoate], and a combination of terephthalic and isophthalic acid and ethylene glycol [copoly (ethylene isophthalate / erephthalate).] Poly (ethylene terephthalate) and / or poly (1,3-propylene terephthalate) are especially preferred products.The polymers can be made by polymerization of various polymerizable ethers and / or oligomers described herein.

Dihydroxy esters of various aromatic dicarboxylic acids could be used in the processes described herein. These are onomeric compounds that can polymerize a polymer. Examples of such compounds are bis (2-hydroxyethyl) terephthalate, bis (3-hydroxyethyl) terephthalate, bis (4-hydroxybutyl) terephthalate, bis (2-hydroxyethyl) naptoate, bis (2-hydroxyethyl) isophthalate, bis [2- ( 2-hydroxyethoxy) ethyl isophthalate, bis [(4-hydroxymethylcyclohexyl) methyl] terephthalate, bis [(4-hydroxymethyl-cyclohexyl) methyl] isophthalate, and a combination of bis (4-hydroxybutyl) terephthalate and its oligomers. Mixtures of these monomers and oligomers could also be used to produce copolymers.

By a "polymerizable oligomer" is meant any oligomeric material that can polymerize to a polyester. This oligomer could contain low molecular weight polyester and varying amounts of monomer. For example, the reaction of dimethyl terephthalate or terephthalic acid with ethylene glycol, when carried out to remove methyl ester or carboxylic groups usually produces a mixture of bis (2-hydroxyethyl) terephthalate, low molecular weight polymers (oligomers) of bis (2-hydroxyethyl) terephthalate and oligomers of mono (2-hydroxyethyl) terephthalate (containing carbonyl groups). This type of material is referred to herein as "polymerizable oligomer".

The process could be used to produce various polyesters such as poly. { ethylene teref alato), poly (propylene teref alato), poly (1,4-butylene terephthalate), poly (ethylene naptoate), poly (ethylene isof alato), poly (3-oxa-l, 5-pentadyl terephthalate), poly (3-oxa-l, 5-pentadyl isophthalate), poly [1,4-bis (oxymethyl) cyclohexyl terephthalate] and poly [1,4-bis (oxymethyl) -cyclohexyl isophthalate]. Poly (ethylene terephthalate) is a particularly important commercial product.

This process avoids high vacuum polymerization processes characteristic of conventional art. The advantages of the process are often a simpler flow pattern, and / or lower operating costs and / or avoid steam ejectors, hot wells and atmospheric emissions. The process also has environmental advantages due to the elimination of volatile organic and discharge of wastewater. In addition, the polymerization is carried out in an inert environment. Therefore, there is often less decomposition and gel formation which results in better product quality. Ethylene glycol and inert gas (eg, N2 or C02) could be continuously recirculated.

In a preferred embodiment of the process for making PET, an oligomer leaving the esterifier is prepolymerized to a degree of polymerization (DP) of about 15-30 and this prepolymer is fed to a terminator to polymerize it at a higher DP of between about 50 and 150, preferably about 60 to about 120 and more preferably about 70 to about 90. The terminator is maintained at a temperature greater than about 260 ° C but not much higher to cause decomposition of the polymer. A temperature range of about 270 ° C to 300 ° C is preferred. The polymerization product is continuously removed from the terminator. An inert gas, preferably nitrogen, is heated in a heater at a temperature from about 280 ° C to 320 ° C and is introduced into the terminator to flow countercurrent to the polymer flow direction to remove volatile reaction byproducts, mainly ethylene glycol. Preferably, the nitrogen is used in a closed cycle and all the process equipment to clean and recirculate the nitrogen is operated at atmospheric pressure (or above, as necessary to ensure the flow of nitrogen through the equipment in the cycle). The amount of inert gas introduced into the system is sufficient for the partial pressure of the by-products to be kept below the equilibrium pressure of the by-products with the melt to provide for the continuous polymerization. The amount of inert gas could be as small as approximately 0.3-0.7 pounds per pound of PET produced.

FIG. 1 illustrates a mode of a reactor or terminator that is suitable for carrying out the polymerization of the invention, especially for producing high viscosity polymers having a degree of polymerization found in a terminator. The reactor comprises a stirred horizontal cylindrical reaction reaction vessel 1. The reactor housing 2 is conveniently constructed with a cylindrical body (shell) and end plates 4 and 6 that close the ends of the cylindrical body. A jacket of the reactor 8 through which a heat transfer material is passed around the cylindrical body. An exemplary heat transfer material is the DowthermE heat transfer fluid, commercially available from Dow Chemical (Michigan). Other heating methods known in the art could be used, such as heating with hot oil, high pressure steam or electric heating. A reactor inlet 14 for introducing a prepolymer feed into the reactor is shown at one end of the reactor, a reactor outlet 16 for discharging the product from the reactor vessel is shown at the opposite end of the reactor.

The DMT or esterified TPA, or oligomers of lower molecular weight or prepolymers thereof, is continuously introduced as stream 3 at one end of the reactor vessel. A preheated inert gas, such as nitrogen, is continuously introduced as stream 7 at the other end, to provide countercurrent flow to the polymer stream. The nitrogen stream 9 carrying the reaction by-product vapors, mainly etne glycol, salts of the reaction vessel as stream 11. The reaction mass flows as the molten polymer mass as stream 5. The polymerized product, polyetne terephthalate , is removed as the current 15. The flow rates of the streams 3 and 15 are coordinated to be equivalent to one another and controlled to provide the desired cessation of the melt in the terminator, usually about 1 to 3 hours, which is equivalent to a melt level of about 1/4 to 1/3 of the diameter of the container. The amount of nitrogen introduced into the system is sufficient for the partial pressure of the evolution of the reaction by-products to be kept lower than the equilibrium pressure of the by-products in, for example, poly (etne) terephthalate (PET) melt. , to provide adequate driving force to remove the etne glycol from the melt in the gas stream. The diameter of the vessel is designed so that the surface velocity of the inert gas stream is in the desired range.

In one embodiment of the process, the use of the Dowtherm heat transfer fluid or other heating means is eliminated using the same preheated nitrogen stream for heating. In this embodiment the nitrogen stream is first brought through the heating jacket 8 in FIG. 1 to maintain the reactor wall above the melting point of the reaction mass, and then fed as stream 7 to the vessel. of reaction.

The reaction vessel in FIG. 1 is equipped with an agitator 20 connected via the drive shaft 18 to an impeller 22 so that the agitator can be rotated at a controlled speed. The mechanical design of the agitator is such that (a) the walls of the container are cleaned (b) a large interfacial area of at least 20 ft2 / ft3 of the melt is preferably created greater than 30 ft2 / ft3; (c) the surface area is frequently renewed; and (d) good mixing is provided.

It is also preferred that only one agitator is used in each container, and that this agitator comprises approximately 75 percent or more of the internal length of the container.

Although shown horizontally arranged, it is possible for the reactor vessel to be pssicted to a degree to facilitate the flow of the reaction melt.

The agitator can have several designs, so long as they provide the desired surface area that is parallel to the flow of inert gas during use, which is predominantly axial along the longitudinal central axis of the reaction vessel. In one embodiment, shown in Figure 2, the agitator 20 comprises vanes 21 attached to the rotating ends 23 and 24 which rotate during use, to form a rotating frame.

A central shaft attached to the agitator in the reactor vessel could extend outside the housing of the reactor vessel where it is attached to a motor or impeller to provide rotation to the agitator at an appropriate speed.

The frame could comprise at least two groups of a plurality of arms extending radially from the longitudinal axis of the reaction vessel. Each pair of arms can support a blade, which is suitable as an elongated blade. The edge of the pallet could be tilted for a better cleaning of the inner surface of the reaction vessel. The blades or blades of the blades could be set at an angle to move an appropriate amount of melt as it moves through the molten liquid, to spill the melt to generate films that can last through most of the rotation outside of the fusion bath. If the angle is such that the space between the blade and the cylindrical wall is too narrow, the blades will only carry a small amount of the melt which could be quickly reduced by actuation through the gap between the blades and the blade. Cylindrical wall, and not enough to the left to generate films or be cleaned on the interior cylindrical wall. If the angle is too large, otherwise, not necessarily more molten mass will be transported around.

The number of blade vanes and the number of arms attached thereto at a point along the length of the reaction vessel could vary. Large diameter containers would have more blades in general. Also, there could be more vanes near the feed end, where the melt is less viscous, and less near the final product where the melt is very viscous. The blades could, for example, be from 2 to 32 in number, preferably 4 to 12 in number.

The blade frame assembly is a mechanically strong construction to withstand the torque required to move through the viscous polymer mass and transport it. In one embodiment, the cross rods are joined between the blade vanes for mechanical reinforcement.

As the vanes or vanes move out of the melting pool of the reaction mass, molten polymer spills both the films that last for a short distance and the surface tension begins to gradually melt the film along with the thickest currents that have much smaller surface area. It has been observed that the films last for about 1/2 inch when the DP is approximately 30-40, approximately 1 inch to approximately DP 50 and approximately 2 inches to DP 60-80. Therefore, to maximize the surface area, the additional longitudinal elements are located under the blades, at appropriate distances, during which the melt can fall and continue to spill out like films. It is advantageous to maintain the spacing between the narrow elements near the feed end where the melt is very fluid and easily distributed in thin films, and to increase the spacing towards the product end where the melt is very viscous and flowing like thick films. If the spacing is too narrow, the viscous melt would stagnate between the elements and would not generate the desired surface area. Thus, the spacing could be as small as 1/2 inch near the feed end and 2-4 inches near the end of the product. The spacing can be optimized for a given reaction vessel diameter and rotation speed. The longitudinal elements could be rectangular bars, rods, wires, grids or metal sheets punched or cut to form desired spacing meshes. These could be arranged to form a "cage" or a plurality of concentric "cages" as shown in Figure 3.

Alternatively, as shown in Figure 4, the elements could be arranged in a rectangular geometry, these rectangles are parallel to one another and extend longitudinally, again retaining the larger spacing at the viscous end and smaller at the less viscous end. The agitator could thus be constructed in sections that appear to be "stacked" or "superimposed" on rectangular mounts. These sections could be installed in the step agitator frame, p. ex. , the plane of a section could be rectangular for those of the next section to keep the inert gas well distributed and to minimize passing (travel through) the melt making the route more stormy.

In Figure 4, the elements are meshed grids, but these could be of other configurations such as rods or perforated sheets of metal. In this type of agitator, the molten material raised by the blades during its run through the melting bath in the lower one, and subsequently spilled by the blades, flows along the rectangular elements to generate surface area.

The agitator is rotated at a speed (rpm) that maximizes the generation of surface area and provides frequent surface renewal. The more rapid renewal of the surface is advantageous to increase the transfer coefficient of volatile products from the reaction melt to the inert gas but the rotation that is too fast may result in the viscous molten polymer remaining as "globules" between the elements and, in fact, decrease the surface renovation. To achieve a reasonably good transfer coefficient it is preferred that the surface be renewed at least once per minute. The speed of the agitator for the generation of surface area is also important. If the rotation is too slow, not enough molten mass rises from the melting bath, or spills very early, and all the elements do not generate films. If the rotation is too fast the melt could be removed as "globules" and not flow effectively to generate surface area. The transfer rate of volatile byproducts, and therefore the speed at which the DP of the polymer increases is proportional to the transfer coefficient (k) and the surface area (a). The rotation speed, or revolutions per minute (rpm) for a given agitator geometry and the diameter of the vessel could be optimized to maximize the product k x a. Preferably, the agitator is rotated at about 1 to 60 rpm, more preferably at about 1-30 rpm and more preferably at about 2-18 rpm.

To illustrate a "cage" type construction in detail, one embodiment of an agitator is shown in Figure 5 in which the elements are wires, the circumferential spacing of which varies along the length of the reactor vessel. The spacing is narrower at the feed end and wider at the discharge end. The agitator is divided into sections, and a plurality of concentric "cages" can exist in each section, the number of which could vary from section to section.

The surface area in this type of configuration is generated in two ways, first by filming the melt circumferentially on the "cage" and, second, by draining the melt from the elements of a "cage" below to a diameter smaller of the lower separator. The spacing and rpra are optimized to obtain good coverage and circumferential drainage at all points along the length of the agitator. The transport of the "globules" of the melt is minimized as previously discussed. At the preferred 2-12 rpm, the spacing near the feed end could be as narrow as 1/2 inch and, near the end of the product, could be 2-3 inches. A) Yes, it is preferable to have more concentric "cages" near the feed end and less at the discharge end. The surface area generated per unit length is, therefore, greater near the feed end and decreases along the length toward the end of product as the number of "cages" decreases. To compensate for this, the longer spacing sections could be made proportionally longer. In this way, the surface generated at each spacing, and hence the increase in DP at each spacing, is approximately the same.

The surface area created in the reactor is equal to the sum of (A) the cleaned surface in the inner wall of the reactor, (B) the surface area of the melt of the melting bath, (C) the surface area of the elements of the agitator and those of the films of the melt generated as the agitator rotates. The area of the film is going to be multiplied by 2 to count the available surface area by mass transfer from both sides of the films.

As the reactor size increases, the contributions to the surface area of (A) and (B) decrease in relation to that of (C). Thus, for large, commercial-scale terminators, the majority of the surface area is from the films generated by the agitator elements and the area due to (A) and (B) could be neglected for design purposes. For example, a reactor 7 feet in diameter X 29 feet in length, designed to generate 15,000 square feet of surface area, the contribution of (A) and (B) is less than 4%.

In the calculation of the surface area, which could be generated with an agitator assembly to be considered, it is first assumed that an optimal combination of the agitator RPM and the spacing element is selected to maximize the generation of films, e.g. ex. , in the grille and wire type "cage" agitators, the grids and the circumferential area of the "cages" are completely covered with the melt. The surface area of the film is twice the area covered by counting the two sides of the films. Preferably, the reactor is designed for a larger area to compensate for less than full coverage during operation under sub-optimal conditions.

The overall agitator for the reactor is conveniently constructed in sections or "wound pieces" that could be held together by appropriate means. The manufacture of the agitator in wound parts offers the flexibility to provide different spacings or other variations depending on the particular application or conditions of use.

Such sectionalized manufacture of the agitator also allows the insertion of screens, for example discs and toroidal tubes that contribute to the inert gas distribution and improve the contact between the inert gas and the reaction mass. This also compartmentalizes the reactor longitudinally so that when it is operated continuously it acts as a number of reactors in series and the methods of developing a plug or an intermittent reactor.

The length and spacing of each Section can be conveniently determined by following the equations in which L is the total length of the agitator, N is the number of sections desired. The length of the first section (at the feed end) is given by the following equation: where X = fold increase number in the DP which is equal to DP of the product / DP of the feed.

For subsequent sections, the length of the nth section is preferably defined as follows: n = .L1 where pn is the distance or spacing of the rods in the nth section and pl is the spacing in the first section. The parameter pn refers to pl by the following equation. pn = - (.r-l) (n-l) pl N-l For concentric "cages" in a given section, the spacing between the consecutive separations equals the distance.

The length of each section as previously calculated could be rounded to a convenient figure for manufacturing, such that: Ll + L2 + L3 ... LN = L The rods selected for this construction are of appropriate diameter and have adequate mechanical stress to withstand the cutting forces of the viscous polymer melt. The rods could be 1/16"diameter near the feed end and thickness gauge, eg, 3/16" diameter, near the viscous product end. The transverse rods could be welded circumferentially at appropriate distances, e.g. ex. , from 3 to 5 times the distance or rod spacing, by mechanical stress.

For ease of fabrication, a large rectangular metal grating of desired distance and distance of cross-section rods could be first constructed and then rolled into a "spiral", instead of constructing individual "cages", while maintaining the separation between consecutive winding. of the spiral near it according to the distance between consecutive concentric "cages", p. ex. , approximately equal to the spacing of the rod.

The "cages" do not necessarily need to be cylindrical. For ease of manufacture, these could be geometries such as hexagonal, octagonal, etc. Figure 6 shows an octagonal assembly of rod spacings as seen from the end of the agitator. The rectangular sections of the metal grids 30 are joined to the radial arms 33 of the rotating end. Such geometries allow metal grids to be cut or made into rectangular sections that can be welded to the radial arms.

The reaction vessel and the agitator are constructed of an appropriate material that has the proper mechanical stress at the operating temperature and which, to produce a quality product, is not easily corrosive or reactive with the reaction mass to contaminate the product . Stainless steel is an appropriate material that has the required properties.

The surface area needed to achieve a given degree of polymerization (DP) can be estimated, as a first approximation, using the following simple equation which has been found to hold when the polymerization is carried out under intermittent flow conditions or plug and A large amount of inert gas is used: DP - DP ° = kat In this equation: DP = the DP of the desired product DP ° = DP of the prepolymer or oligomer feed a = surface area in square feet t = residence time or unemployment time in hours k = overall transfer coefficient for transfer of volatile condensation byproducts, mainly ethylene glycol, from the melt to the inert gas. The units are foot / hr.

The transfer coefficient, k, depends on several factors, such as temperature, surface renewal speed, catalyst concentration and inert gas velocity. Under the conditions of Example 1, its value was found to be approximately 0.79 ft / hr.

Thus, to polymerize a DP prepolymer of 20 to a DP product of 80 in 2 hours of residence time, the required surface area, using this value for k, can be calculated as: a = DP-DP ° kt 80"20 = 38 ft2 / ft3 of melt 0.79x2 For continuous polymerization, the reactor is preferably designed to provide a greater surface area, such as 50-75 ft2 / ft3 of melt for the previous example, to compensate for using less inert gas flow, e.g. ex. , 0.3-0.7 lb N2 / lb of melt, or for deviations from the melt flow of the ideal plug flow. Surface area greater than the calculated one also allows flexibility of operation. If the reactor has a smaller area, the stoppage time would need to be proportionally greater than 2 hours. The agitator configurations described herein can provide the required surface areas.

For operation of the polymerization reaction continuously, it is desirable that the residence time distribution of the melt flow be reduced, e.g. ex. , closer to the plug flow, and purging is avoided. The purge can potentially occur around the elements of the linear vanes and the agitator, particularly when the melt is not highly viscous.

Also the reactor could be divided longitudinally into a number of compartments by introducing screens such that the melt flows from one compartment to the next and thus the reactor works as several smaller reactors in series. A convenient way to achieve this is by inserting along the length of the agitator rings or toroidal tubes with an outside diameter equal to that of the agitator. This inner diameter of the toroidal tubes is such that the reactor operates at the desired level. The inner diameter could be about 0.7 times the outside diameter. Disks could also be inserted between the toroidal tubes to form a toroidal-disc-toroidal tube configuration, to keep the flow of inert gas well distributed and improve contact with the melt by forcing it to go through the toroidal tube and then around the disc, etc. The screens are dimensioned such that the velocity of the gas through, around or between these is not too high to cause entry or thrust of the melt in the direction of inert gas flow.

Similarly, another embodiment of the agitator comprises partial discs or partial rings installed such that the inner edges are staggered 180 °, e.g. ex. , alternating screens faced in opposite directions, so that the inert gas zigzags past as well as swirling, creating greater turbulence and more effective contact with the molten mass as they turn.

The process of this invention could also be carried out for intermittent polyester preparation wherein an intermittent low molecular weight oligomer reactor is charged to the polymerization equipment and contacted with the inert gas as described until it is reached the high degree of polymerization desired. The oligomer is prepared by esterification as described, except that it could also be prepared intermittently either in a separate vessel in the same polymerization vessel. The contact equipment of the gas and the melt could be similar to that described for the continuous mode of this invention except that it is not necessary to vary the spacing between the elements of the agitator along the length of the container. Also, compartmentalization is not required to develop plug flow. The spacing of the agitator elements should be chosen to accommodate the viscosity and flow characteristics of the high molecular weight end product. For intermittent preparation it is advantageous to adjust the speed of the agitator as the viscosity of the melt increases. Initially, when the viscosity is low, the agitator could operate as high as 100 rpm but until completion of the polymerization a low speed of about 1 to 20 rpm, preferably 2-12 rpm, is desirable. Intermittent production is appropriate for economic reasons when preparing relatively small quantities of polyester or when strict control of additive concentrations is required for product quality considerations. When the quantities to be prepared are very small, it may be more economical not to provide equipment for recirculation of the inert gas, or ethylene glycol, and discharge it to the atmosphere after making it non-toxic to the environment by known methods such as washing it thoroughly with water and disposing it out of water in an environmentally safe way.

The invention can also be carried out in a semi-intermittent manner wherein the polymerization equipment is fed intermittently, the reaction mass is polymerized to a greater degree, and the product is discharged intermittently.

EXAMPLE 1 This example illustrates the polymerization at a pilot scale in a polymerization reactor according to the present invention. The reactor consists of a 6-inch-diameter nominal diameter glass tube. This has a glass tube of diameter 8 inches of similar length inside with the help of end plates to form an annular space around the reactor and serves as a heating jacket. The heating medium was air heated to 295-300 ° C which was introduced into the annular space at one end and flowed out of the other end.

The agitator consists of two terminal pieces each with four arms in the shape of a cross. Each pair of arms maintains a length of approximately 20"in length, blade width or blade width. The rings were mounted within this frame, each a few inches inside the ends to maintain four more paddles, such that the 8 paddles formed a "cage" of diameter slightly smaller than the 6"diameter of the reactor so that it could rotate freely inside the reactor.The axes were attached to the two final cross pieces which could be rotated inside the supports provided in the center of each end plate of the reactor.The stirrer was turned by the use of a motor having a reducer of Gears attached to the shaft at one end of the reactor The temperature of the molten polymer and the inert gas was monitored by placing thermocouples inserted into the reactor at each of its two ends.

The reactor was charged with 9 pounds of approximately 20 DP prepolymer obtained from a commercial plant where TPA was esterified with ethylene glycol and prepolymerized at a DP of about 20. This contained about 200 ppm of antimony as a catalyst. The charge was made by feeding the solid prepolymer through a melt extruder which melted the prepolymer and heated it to about 280-295 ° C. The agitator was turned at 12 rpm, and N2 was preheated to approximately 295 ° C was flowed through the reactor at a rate of 0.57 ft / sec based on the free cross-section of the reactor. Since the reactor was filled approximately 30% with molten mass the contact velocity was approximately 0.82 feet / sec. The N2 was introduced at one end and discharged into the atmosphere from the other end. Thus, the reactor was essentially at atmospheric pressure. The temperature of the reaction mass was maintained at about 280 ° C by controlling the temperature of the hot air in the annulus. The polymerization was continued under these conditions for two hours. Polymer samples were taken every half hour and analyzed for DP by gel permeation chromatography (CPG). The average GP number was found to be approximately 36, 52, 68 and 80, then 1/2, 1, 1-1 / 2 and 2 hours of polymerization, respectively. These DP values when plotted against time were adjusted to a straight line.

DP - DP ° = (ka) t with a slope = ka of 30 hr "1 The reactor was estimated to provide above the average of 4.58 ft2 of film area which for 9 lb of melt is translated to an "a" value equal to 38 ft2 / ft3 of melt. The value of k was in this way 30/38 = 0.79 ft / hr.

Initially, when the melt was DP 20, it spilled from the agitator blades like streams but after a few minutes it started to become viscous and fell like films that extend approximately 1 / 4-1 / 2"from the As the polymerization proceeds at a higher DP * the film becomes more pronounced.The melt extends as 3 / 4-1"films from the blades and to the end of the spilled films extended 1- 1-1 / 2"Thus, the larger surface area could have been generated if the additional elements had been placed on the agitator, under the paddles, on which the melt could fall and drip like films. Using hot air in the annular heating jacket, the preheated N2 could have been first passed through the annulus and then fed to the reactor.

EXAMPLE 2 This example illustrates a design of a prototype terminator according to the present invention to be operated continuously at a rate of 100 to 150 lb / hr. It will be fed continuously with a DP prepolymer of 20 prepared in an upstream esterifier and a prepolymerizer. The reactor is designed to produce PET product of approximately DP 80 useful for fiber spinning or lamella production. The reactor is 9 feet long and has a diameter of 18 inches. It has a heating jacket heated with Dowtherm steam. "It is equipped with an agitator 7.5 feet in length to leave approximately 9 inches of space on each end for feed and discharge nozzles.The agitator has parts or end plates that extend each to 8 arms which are joined to the axes for rotation.To each pair of arms a vane vane is attached 1-1 / 2"wide at a 45 ° angle to the inner wall of the reactor. They are kept within this frame of eight blades, parts of concentric "cages" of varied distances made of stainless steel rods such that, starting from the feed end, there is a length section of 9 inches of distance of 1/2"(and spacing between consecutive concentric" cages "), then length of 18" of "cages" away from 1", followed by "cages" of length 27"that are at a distance of 1-1 / 2" and finally cages of concentric length of 36"of distance of 2" where the concentric "cages" are separated 2". The toroidal tubes and discs are inserted alternately between the wound parts so that the reactor is compartmentalized to act as 4 reactors in series.The agitator can be rotated at 3-12 rpm.The wound parts of each of the distances of 1/2", 1", 1-1 / 2"and 2" can provide approximately 57 ft2 of surface area for a total of 228 ft2 of surface area The reactor is operated with approximately 300 lb or 4 ft3 of stop melt. This translates to an average surface area of 57 ft2 / ft3 of the melt which is about 50% more than would be required if carried out in a plug flow reactor.N2 flows countercurrent to the flow of the melt at 90 to 120 lb / hr The surface velocity of gas under the condition n operation about 1 atmosphere and 285 ° C, based on a free cross section, is from 0.36 to 0.48 ft / sec.

It is noted that in relation to this date, the best method known to the applicant to carry out the aforementioned invention, is that which is clear from the present description of the invention.

Having described the invention as above, the content of the following is claimed as property.

Claims (24)

1, A process for making a polyester of one or more aromatic dicarboxylic acids and one or more glycols in a molten state, in which an inert gas is used to assist in the removal of a by-product of volatile condensation, characterized in that the improvement comprises , employ a horizontally disposed cylindrical reactor vessel partially filled with a polymerization reaction mass in the form of a melt, which is equipped with the following: a) a reactor inlet for introducing a polymerizable feed into the reactor vessel; b) a gas inlet to introduce an inert gas into or near one end of the reactor vessel and a gas outlet to remove the inert gas at or near an opposite end of the reactor vessel, thereby resulting in the flow of gas passes in front of the reaction mass in the reactor vessel; c) means for maintaining the reaction mass in the molten state; and d) a stirrer rotating on its axis during operation, the agitator comprises a plurality of longitudinally disposed elements for transporting a portion of the melt as the elements move through the reaction mass, the elements are positioned as such. that said elements generate films, the planes of the films are parallel to the central axis of the agitator and the flow of inert gas which is predominantly in the axial direction; Y e) an exit from the reactor to remove the polymer product from the reactor vessel.

2. The process of Claim 1 for the production of higher molecular weight PET, the process is carried out at atmospheric pressure or above, characterized in that it comprises contacting DHET or its lower molecular weight oligomers, in the molten form, with an inert gas, which flows in the process countercurrent to the flow of the reaction melt, so that the ethylene glycol and other reaction byproducts are continuously stirred and where the PET product is continuously stirred with an equal stop time to less of about 5 hours.

3. The process of Claim 1, characterized in that the process is carried out at atmospheric pressure.

4. The process of Claim 1, characterized in that the surface area of the films is at least 30 square feet per cubic foot of melt.

5. The process of Claim 1, characterized in that the elements are selected from the group consisting of rectangular bars, rods, wires, grids or perforated or cut sheets.

6. The process of Claim 1, characterized in that the agitator further comprises blades for cleaning one or more internal walls of the reactor vessel.

7. The process of Claim 1, characterized in that the agitator elements are spaced in a circular or polygonal geometry, in cross section, at one or more radial distances from the axis of the agitator to the cages in a concentric manner.

8. The process of Claim 1, characterized in that the polyester is poly (ethylene terephthalate).

9. The process of Claim 1, characterized in that the inert gas is selected from the group consisting of nitrogen, dioxide and combinations thereof.

10. The process of Claim 1, characterized in that the agitator is rotated from 2 to 12 rpm.

11. The process of Claim 1, characterized in that it also comprises screens.

12. The process of Claim 11, characterized in that the screens comprise toroidal tubes and discs, spaced longitudinally along the reactor vessel.

13. The process of Claim 1, characterized in that the distance between the elements of the agitator is increased along the length from the feed end to the product end of the reactor.

14. The process as mentioned in Claim 1, characterized in that the agitator comprises at least about 75 percent of the internal length of the reactor vessel.

15. The process of Claim i, characterized in that the polyester is poly (ethylene terephthalate).

16. The process of Claim 2, characterized in that the inert gas flows at a velocity of 0.2 to 3 ft / sec and a gas-liquid interfacial surface is at least about 20 ft2 / ft3 of the melt.

17. The process of Claim 2, characterized in that the agitator elements are 1/2 to 1 inch apart from one another near the feed end of the reactor, the spacing is incremented step by step along the length of the agitator and the elements are 1.5 to 4 inches from each other near the end of the reactor product.

18. The process of Claim 2, characterized in that the feed is an oligomer of bis (2-hydroxyethyl) terephthalate of at least about DP of 5 and polymerization to the final product DP is carried out by contacting the reaction melt with the gas inert in a single reaction vessel.

19. The process of Claim 2, characterized in that the inert gas is selected from N2 and C02.

20. The process of Claim 2, characterized in that the inert gas is preheated to about the polymerization temperature or below the polymerization temperature or above the polymerization temperature before being brought into contact with the melt.

21. The process of Claim 1, characterized in that the polyester is poly (ethylene terephthalate), poly (1,3-propylene terephthalate), poly (1,4-butylene terephthalate), poly (ethylene 2, 6-naptoate), or copol i (ethylene isophthalate / terephthalate).

22. The process of Claim 2, characterized in that the polymerization temperature is from about 270 ° C to about 300 ° C.

23. An apparatus for implementing a condensation polymerization in a molten state, this apparatus, is characterized in that it comprises a cylindrical reactor arranged horizontally equipped with the following: a) a reactor inlet for introducing a polymerizable feed into the reactor vessel; b) a gas inlet to introduce an inert gas into or near one end of the reactor vessel and a gas outlet to remove the inert gas at or near an opposite end of the reactor vessel, thereby resulting in the flow of gas passes in front of the reaction mass in the reactor vessel; c) means for maintaining the reaction mass in the molten state; Y d) a stirrer that is adapted to rotate on its axis during operation, the agitator comprises a plurality of elements that are arranged longitudinally to transport a portion of the melt in the lower part of the reactor as the elements move through the melt, the elements are positioned such that said elements generate films, the planes of the films are parallel to the longitudinal axis of the reactor vessel; wherein the films that are thus generated provide the majority of the interfacial surface area in the reactor, and e) an exit from the reactor to remove the polymer product from the reactor vessel.

24. The apparatus of claim 22, characterized in that the agitator further comprises flanges along its length to improve the distribution of inert gas and / or reduce the purge of the reaction mass of a continuous operation.