PROCESS FOR THE PURIFICATION OF INERT GAS BACKGROUND OF THE INVENTION The present invention relates to a process for the purification of an inert gas containing impurities formed from organic compounds. The present invention also relates to a process for the purification of an inert gas recycled from a polymerization reactor, and particularly a solid state polycondensation reactor (SSP) for aromatic polyester resins. Polymer resins are molded into a variety of useful products. One of these polymeric resins is the polyethylene terephthalate (PET) resin. It is well known that aromatic polyester resins, particularly TEP, copolymers of terephthalic acid with low proportions of isophthalic acid and polybutylene teletalate are used in the production of beverage containers, films, fibers, packages and tire ropes. U.S. Pat. No. 4,064,112 Bl discloses a polycondensation or solid state polymerization (SSP) process for the production of PET resins. While for fibers and films the intrinsic viscosity of the resin should generally be between 0.6 and 0.75 dl / g, higher values are needed to mold materials such as containers and tire fibers. Higher intrinsic viscosities, such as those greater than 0.75 dl / g, can only be obtained with difficulty directly by means of the polycondensation of molten TEP, commonly called the molten phase process. The SSP process forces the polymerization to a greater degree, thereby increasing the molecular weight of the polymer by heating and removing the reaction products. Polymers with higher molecular weight have higher mechanical strength and other useful properties for the production of containers, fibers and films, for example. An SSP process starts with polymer granules in an amorphous state. U.S. Pat. No. 4,064,112 Bl discloses the crystallization and heating of the granules in a crystallizing vessel under stirring at a density of 1,403 to 1,415 g / cm 3 and a temperature of between 230 and 245 ° C before entering the SSP reactor. Otherwise, the granules, which are sticky, tend to stick together. The SSP reactor may consist of a cylindrical reactive section containing a vertical moving bed into which the polymer granules are introduced from above, and a frusto-conical dispensing section in the base to supply the product granules. The polycondensation reactor typically operates at temperatures between 210 and 220 ° C. Various reactions occur during the polycondensation of PET. The main reaction that increases the molecular weight of PET is the elimination of the ethylene glycol group: TEP-COO-CH2-CH2-OH + HO-CH2-CH2-OOC-TEP? TEP-COO-CH2-CH2-OOC-TEP + HO-CH2-CH2-OH An inert gas, such as nitrogen, is passed through the polymerization reactor to remove impurities from the developing polymer. The impurities present in the inert gas stream used in the production of polyethylene terephthalate in an SSP process generally include water and organics such as aldehydes and glycols, typically acetaldehyde and ethylene glycol, and glycol oligomers. In addition, the volatile impurities include low molecular weight TEP oligomers, such as the cyclic TEP trimer. Water is removed from the inert gas stream before being recycled to the SSP, because it can precipitate an inversion of the polymerization process. Organic impurities are removed to strengthen the polymer product, and ensure that impurities do not contaminate the compatibility of the final product with its use. It is especially important to prevent organic impurities from leaking from a resin container to the beverage it contains. These impurities are removed from the polymer granules and accumulate in the inert gas stream. Organic impurities are present in the inert gas stream to be purified, in quantities, defined as methane equivalent, between 2,000 and 3,000 ppm or more. U.S. Pat. No. 5,708,124 Bl discloses maintaining the mass flow rate ratio of the inert gas against solid mass flow rate of polymer TEP to less than 0. 6 in an SSP reactor. It is also well known that polyamide resins, and among these particularly PA6, PA6, 6, PA11, PA12 and their copolymers, have various applications in the fiber and flexible packaging sectors, and in the production of articles manufactured by technology. blowing and extrusion. Although the relative viscosity of the resin for fibers is low, at between 2.4 and 3.0, higher relative viscosities, between 3.2 and 5.0, are required for articles produced by blowing and extrusion technologies. The relative viscosity is increased to more than 3.0 by an SSP process that operates at temperatures between 140 and 230 ° C, depending on the types of polyamide used. U.S. Pat. No. 4,460,762 Bl describes an SSP process for a polyamide, and different methods for accelerating this reaction. An SSP process for polyamide resins is also described in the article "Nylon 6 Polyme ization in the Solid Sta te", by RJ Gaymans et al., Journal of Applied Polymer Science, Vol. 27, 2515-2526 (1982), which notes the use of nitrogen as a heating and purge aid. The reaction is carried out at 145 ° C.
It is also known that the molecular weight of polycarbonate can be increased by an SSP process.
The polyamides and polycarbonates in development also emit organic impurities that must be purged by a stream of inert gas that must then be purified. European patent EP 0 222 714 Bl discloses a method for producing polyethylene terephthalate and polyethylene isophthalate with very low generation of acetaldehyde to reduce the amount of necessary purification of inert gas. The conventional method used for the purification of an recycled inert gas stream from an SSP process includes an oxidation step to convert the organic impurities into C02, and a drying step to remove the water formed in the polymerization process and the oxidation step . The oxidation step is carried out with oxygen or a gas containing oxygen, such as air, using an oxygen concentration of no more than a slight excess of the stoichiometric amount as regards the organic impurities. The oxidation step is controlled in accordance with U.S. Pat. No. 5,612,011 Bl, so that the inert gas stream in the outlet contains an oxygen concentration of not more than 250 ppm, and preferably in accordance with U.S. Pat. No. 5,547,652 Bl, so that the inert gas stream in the outlet contains an oxygen concentration of not more than 10 ppm. These patents disclose that a deoxidation step previously required to reduce oxygen with hydrogen between the oxidation and drying steps is not necessary. Conventionally the oxidation reaction is carried out at a temperature between 250 and 600 ° C, the inert gas stream circulating on a catalyst bed formed of a support coated with platinum, or platinum and palladium. The low oxygen content present in the inert gas stream leaving the oxidation section allows it to be recycled to the SSP process after the drying step. Moreover, higher oxygen concentrations in the recycled inert gas stream present the risk of oxidation reactions that degrade the polymer product, for example, by "yellowing" the product. Japanese publication 20885/71 discloses a method of reconstituting the inert gas used in the polycondensation or solid state polymerization of linear polyesters, which comprises putting the gas in contact with a metal oxide at between 150 and 300 ° C. The organic reaction products contained in the inert gas are oxidized in water and carbon dioxide. However, since the metal oxide loses its activity, it must be heated in the presence of air in a batch process. Accordingly, this publication does not refer to a continuous gas catalytic purification process.
The last step of inert gas purification is a drying step carried out by circulating the gas on a silica gel, molecular filters or other beds of drying materials. In this step, the water, stripped of polymer granules by the inert gas stream, and generated in the oxidation step, is removed. After this step, the inert gas is recycled to the SSP process. Small traces of oxygen, when present in the recycled stream of inert gas, do not cause polymer oxidation or degradation effects. Even if the amount of oxygen in the oxidation reactor is stoichiometric or a little higher, it is possible to reduce the organic impurities to acceptable levels, such as less than 10 ppm defined as methane equivalent. An article by E.V. Kuznetsova et al., Entitled "Purification of Industrial Vapor-Gas Discharges and Nastewaters by Vapor-Phase Catalytic Oxidation" reveals the use of platinum and other metal catalysts for the oxidation of organic substances in water vapor from a stream of wastewater. The article indicates that when the temperature drops below 250 ° C, the degree of conversion of hydrocarbon substances is little less than incomplete for the aluminum-copper oxide catalyst. The platinum or platinum and palladium catalyst previously used in the purification of an inert gas from a polymerization process should be carried out at between 250 and 600 ° C to ensure adequate oxidation of the hydrocarbon impurities from the nitrogen gas stream when essentially stoichiometric amounts of oxygen are used. The higher temperature used in the reaction zone requires relatively more expensive equipment and operations to preheat the impure inert gas stream that is fed into the oxidation zone. further, higher equipment and operating costs are necessary to recover heat from the oxidation step. Accordingly, an object of the present invention is to provide a catalyst that oxidizes almost all organic impurities from an inert purge stream of a polymerization reactor with essentially stoichiometric amounts of oxygen at lower temperatures. SUMMARY OF THE INVENTION It was unexpectedly discovered that catalysts with 0.1 to 2.0% platinum weight, where platinum is in a reduced state, almost completely oxidize the organic impurities of a polymerization reaction with an essentially stoichiometric amount of oxygen at much higher temperatures. lower than those previously practiced, specifically at less than 250 ° C. Additional objects, embodiments and details of the present invention can be obtained from the following detailed description of the present invention.
BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic view of a polymerization flow scheme, in which the present invention can be used. Figure 2 is a graph comparing the behavior of a conventional catalyst and a catalyst of the present invention. DETAILED DESCRIPTION OF THE INVENTION A detailed description of a preferred method for carrying out the process is provided, in the context of a polycondensation or solid state polymerization (SSP) process. However, the present invention can be used in other types of polymerization processes, such as for polyamide and polycarbonate, in which impurities are purged from the polymerization with inert gas and the impurities are oxidizable, especially to carbon dioxide and water. The polyester resins usable in the SSP process are polycondensation products of aromatic bicarboxylic acid, particularly terephthalic acid or its esters with diols of 1 to 12 carbon atoms, such as ethylene glycol, 1,4-dimethylolcyclohexane and 1,4-butanediol. Preferred resins are polyethylene terephthalate (PET) and polybutylene terephthalate. Polyester resins usable in the SSP process may also include elastomeric polyester resins, including segments derived from polyethylene glycol, and copolyesters containing up to 20% units derived from different bicarboxylic aromatic acids of terephthalic acid, such as isophthalic acid. The resins to be subjected to SSP may contain a resin enhancing additive, to accelerate the SSP reaction. The preferred catalyst compounds are the dianhydrides of aromatic tetracarboxylic acids, and particularly pyromellitic dianhydride. The improving agent is generally used in an amount between 0.05 and 2% by weight. Conventional additives such as stabilizers, dyes, fire retardants and nucleants may also be present in the resin. Polyester resins useful for improving an SSP process can also be materials produced from recycled containers that have been washed, crushed and dried. Typically, the recycled material is remelted and crushed before being sent to the SSP process. The polyamide resins usable in the process of the present invention include polyamide 6 derived from caprolactam, polyamide 6,6 obtained from hexamethylenediamine and adipic acid, polyamide 11 obtained from aminoundecanoic acid, and 12-polylaurylacetone copolyamides 6/10 and 10/12 , in addition to metaxylene diamine polyamides. Polycarbonates can also be used in the process of the present invention. Referring to Figure 1, the SSP process of polyester to which the present invention can be applied comprises feeding amorphous and transparent polyester granules, with an intrinsic viscosity generally ranging from 0.57 to 0.65 dl / g, to a hopper 12 for a line 10. The intrinsic viscosity or molecular weight of the initial material is not important for the practice of the present invention. Generally, the SSP process can be performed effectively with feeds within a wide range of values. For example, techniques for using an initial material with a degree of polymerization of up to 2-40 are disclosed in U.S. Pat. Nos. 5,540,868 Bl, 5,633,018 Bl and 5,744,074 Bl, which contemplate passing eventually through the SSP process to raise the molecular weight enough to make useful resins. In addition, the initial intrinsic viscosity in the case of post-consumer recycled material can be at levels of more than 0.65 dl / g. The hopper 12 feeds the granules by a line 14 and a control valve 16 to a fluidized bed pre-crystallizer 18. The pre-crystallizer 18 operates at 170 ° C and a pressure of 10.3 kPa to obtain a crystallinity of 35% in the granules Of polyester. The polyester granules are then fed from the pre-crystallizer 18, by a line 20 and a control valve 22 to a first crystallizer 24. If a larger capacity is needed, a second crystallizer 28 which the first crystallizer 24 feeds can be used. on line 26.
In the crystallizers 24, 28, the granules are optionally preheated or cooled in some cases to an SSP reaction temperature while they are subjected to mechanical agitation to prevent the granules from sticking together. The granules leaving the crystallizer will have a crystallinity of 45%. Crystallizing the PET granules before polycondensation prevents the granules from sticking during the SSP reaction. The granules leaving the crystallizers are fed by a line 30 and a control valve 32 to a mobile bed polymerization reactor bonded 34 which can be suitably operated at 150-240 ° C, and preferably at 210-220 ° C. for TEP. The granules move by gravity through the moving bed for 12 to 36 hours to produce crystalline and opaque granules with an intrinsic viscosity of 0.75 dl / g or more, depending on the application to the polyester granules. The granules are removed from the reactor 34 by a line 36. An inert gas without oxygen, typically nitrogen, purges the polymerization reactor, the crystallizers and the pre-crystallizer to remove impurities produced by the granules. The inert gas is supplied by a line 38 and is distributed to the polymerization reactor 34 by a distributor 40. The ratio of inert gas mass flow rate to polymer mass flow rate preferably should not exceed 0.6 in the reactor 34 if the product is TEP. A gas line 42 removes inert gas with impurities from the reactor 34 and is divided into a recycle line 44 and a crystallisation line 46. The crystallisation line 46 supplies the inert gas to the second crystallizer 28, and a line 48 supplies the inert gas. from the second crystallizer 28 to the first crystallizer 24. A line 50 supplies inert gas with impurities to the pre-crystallizer 18, and a line 52 supplies inert gas with impurities to join with the recycling line 44. The inert gas recycled in a combined line of Recycled 53 is preferably at a temperature between 200 and 220 ° C. The combined recycling line 53 passes the inert gas with impurities through a filter 54. After the recycled stream of inert gas is filtered, air is injected through a line 57 to a line 56 that leaves the filter 54. The mixture of air and Inert gas is conveyed by a line 59 through a heater (not shown), if necessary to obtain the desired oxidation reaction temperature, to an oxidation reactor 58, where the organic impurities are burned by circulating the stream over a catalyst bed, including the catalyst containing reduced platinum in accordance with the present invention. Oxygen is injected via line 57 in essentially stoichiometric amounts to ensure complete combustion of the organic impurities in the oxidation reactor 58, tolerating a maximum surplus of no more than 250 ppm, preferably no more than 100 ppm, and preferably no more than 10 ppm. ppm of oxygen at the outlet of the reactor 58. The oxidation reactor 58 can be operated with these conversions at temperatures of less than 350 ° C. However, since operation at lower temperatures is more economical, the oxidation reactor 58 will preferably operate at less than 250 ° C, which is the temperature without heating of the inrush current on line 59. Accordingly, eliminates the need for in-line heater 59. A line 60 extracts the effluent from oxidation reactor 58 which contains only nitrogen, carbon dioxide, water and traces of oxygen. The carbon dioxide content stabilizes at some level due to the losses through the SSP plant, and acts as if it were an inert gas, due to its chemical inertness. The gaseous stream exiting on line 60 can be circulated through a heat exchanger (not shown) to recover heat, or to condense and discard part of the water by cooling the effluent from the oxidation reactor at 10 to 15 ° C. The cooling step can be omitted because the economy may not require heat recovery from an inert gas stream of relatively lower temperature, in which case condensation itself occurs. No condensation collector optional equipment is shown in the drawings. The gaseous stream is supplied by line 60 to a dryer 62 which preferably operates at 200 ° C. The dryer 62 preferably contains molecular filters to adsorb the water. The effluent from the dryer 62 is transported by line 64 to a heater 66 after filtering (not shown) the possible particles derived from the molecular filters. The heater 66 heats the gaseous stream to a temperature commensurate with that of the reactor 34, and recycles the gaseous stream to the reactor 34 via the line 38. The regeneration of the molecular filter bed is carried out in accordance with known methods, operating for example in a closed circuit with a hot nitrogen stream (not shown). The catalyst used in the oxidation reactor 58 is a platinum and palladium catalyst, or essentially only platinum of up to 5.0% by weight, suitably 0.1 to 2.0% by weight, and preferably 0.2 to 0.8% by weight of the catalyst, based on the total weight of the catalyst product in which the platinum is in an essentially reduced state. We have found that providing a catalyst with platinum in a reduced state allows the oxidation of organic impurities to occur at much lower temperatures than previously practiced; that is, at 250 ° C or less. By the term "essentially reduced" we mean that at least 70% of the weight of the platinum in the catalyst is metallic platinum, as opposed to platinum oxide. Suitably, at least 90% of the weight of platinum in the catalyst is metallic platinum and, preferably, all of the platinum in the catalyst is reduced. Although without wishing to be bound by any theory, we believe that reducing platinum to a metal with a valence state of zero, four or two in the oxidized state, provides a significantly greater oxidant activity to the catalyst at lower temperatures. Additionally, the platinum is distributed on the outermost surface of the support comprising a shell around the support, thereby giving it greater activity. Thus, more than 90% of the weight of the platinum in the catalyst is present at less than 100 μm depth of the catalyst surface. This is very important for the oxidation reaction, of highly limited diffusion. The support for the catalyst can be an alumina made using the teachings of U.S. Pat. Nos. 4,108,971 Bl or 4,301,033 Bl, both incorporated herein by reference. It may be preferable to use a catalyst support with higher bulk density, such as more than 0.3, and up to 0.7 g / cc, in accordance with the teachings of U.S. Pat. No. 4,301,033 Bl, since we think that the amount of platinum supported is proportional to the apparent density of the support. It is contemplated that the effectiveness of the catalyst be increased with promoters and additives. Additionally, if the catalyst is made to contain more reduced platinum, the reaction temperatures may decrease further. In the context of the present invention, other metals of the platinum group of the second and third rows of Group VIII of the periodic table of elements can also advantageously operate in a state of reduced valence. Finally, it is contemplated that the catalyst of the present invention can be regenerated in situ by removal of carbon with known methods. The following examples are provided by way of illustration, and not to limit the present invention. EXAMPLE I The catalyst of the present invention was prepared in the following manner. A solution of 100 milliliters was mixed with 4.5% by weight of chloroplatinic acid, which contained 20% by weight of platinum and 0.69% by weight of thiomalic acid in deionized water, and was stirred for 1 hour in a flask. The solution was aged for 48 hours at room temperature, and then adjusted to a volume of 600 milliliters with deionized water. 500 milliliters of activated alumina spheres with an apparent density of 0.40 g / cc, a crush resistance of 8.0 kg, an average diameter of 3.2 mm and a BET surface area of 165 m / g in the flask of aged solution were poured. of chloroplatinic acid. The flask was connected to a rotary evaporator and immersed in a boiling water bath to evaporate the solution under conditions of slight vacuum for 4 hours under rotation. The impregnated spheres were then dried at 150 ° C for 2 hours in an air-flow oven, and oxidized at 500 ° C for 1 hour in the same oven. The oxidized spheres were then charged in a reduction reactor and heated to 200 ° C under nitrogen flow for 1 hour. The flow of nitrogen to hydrogen was changed and heated at 500 ° C for 1 hour. Then the reactor was cooled to less than 200 ° C, and the flow of hydrogen to nitrogen was changed. The prepared catalyst, ready to be charged to the oxidation reactor, had a platinum content of 0.45% by weight, and all the platinum was in the reduced state. EXAMPLE II In Figure 2 the behavior of the new catalyst with reduced platinum is shown, compared to the behavior of a conventional platinum catalyst, not reduced. Figure 2 is a graph of the percent conversion of hydrocarbons against temperature. Each catalyst was tested in a fully integrated SSP demonstration, where 907 kg per day of base TEP resin were improved from an initial intrinsic viscosity of 0.58 dl / g to a final intrinsic viscosity of between 0.80 and 0.81 dl / g. The SSP process comprised a reaction section in which 43 to 45 kg / hr of granular base resin was upgraded to 210 ° C with 25 kg / hr of nitrogen. The nitrogen used was then electrically heated and sent to one of two test reactors containing conventional catalyst and the catalyst of the present invention. The velocity of the gas corresponded to a gas hourly space velocity (VEGH) of 5000 hr "" 1 in the catalytic reactors. The gas leaving the reactor was cooled by heat exchange at 60 ° C, dehydrated, sent to a desiccant dryer containing molecular filters, and dried to a gas dew point of less than -60 ° C. In all cases, the reaction was controlled to near stoichiometric levels of less than 10 ppm oxygen by the addition of clean, dry air to the reaction vessel by the control of an oxygen analyzer. An online hydrocarbon analyzer was used to track the conversion. The PET base resin was a TEP and isophthalate copolymer suitable for containers. The properties were the following:
Base resin
The catalyst of the present invention achieves a better hydrocarbon conversion, and at lower temperatures, than the conventional catalyst without reduced platinum. The graph indicates that the catalyst of the present invention achieves a hydrocarbon conversion of up to 10 ppm to just over 210 ° C, and a hydrocarbon conversion of up to 1 ppm to just over 240 ° C. The conventional catalyst requires temperatures of just under 290 and 310 ° C, respectively, to obtain the same conversion of hydrocarbons. The catalyst of the present invention achieved a complete conversion without hydrocarbon remnants at 270 ° C. Therefore, the catalyst can be operated at temperatures of more than 250 ° C if a higher conversion is required, or when the activity decreases with use.