KR20070048251A - Optimized liquid-phase oxidation - Google Patents

Abstract

According to the present invention, an optimization process and apparatus for carrying out liquid phase oxidation of oxidizing compounds effectively and economically are described. This liquid phase oxidation is carried out in bubble column reactors which provide a very effective reaction at relatively low temperatures. If the oxidative compound is para-xylene and the product from the oxidation reaction is crude terephthalic acid (CTA), the CTA product is purified by more economical techniques than those used in the formation of CTA by conventional high temperature oxidation processes. And can be separated.

Description

Optimized Liquid Oxidation Method {OPTIMIZED LIQUID-PHASE OXIDATION}

The present invention relates generally to a liquid phase catalytic oxidation process of aromatic compounds. One aspect of the invention is directed to a method in which partial oxidation of a dialkyl aromatic compound (eg para-xylene) can be partially oxidized to produce a crude aromatic dicarboxylic acid (eg crude terephthalic acid), which can then be purified and separated. . Another aspect of the invention relates to an improved bubble column reactor that provides a more effective and economical liquid phase oxidation process.

Liquid phase oxidation is used in a variety of existing commercial processes. For example, current oxidation of aldehydes to acids (eg, propionaldehyde to propionic acid), cyclohexane to adipic acid, and alkyl aromatic compounds to alcohols, acids or diacids. Liquid phase oxidation is used. A particularly important commercial oxidation process in the last category (oxidation of alkyl aromatic compounds) is the liquid phase catalytic partial oxidation of para-xylene to terephthalic acid. Terephthalic acid is an important compound with a variety of uses. The main use of terephthalic acid is as feedstock in the production of polyethylene terephthalate (PET). PET is a well known plastic that is used in large quantities around the world to manufacture products such as bottles, fibers and packaging.

In a typical liquid phase oxidation process, including partial oxidation of para-xylene to terephthalic acid, a liquid feed stream and a gaseous oxidant stream are introduced into the reactor and produce a multiphase reaction medium in the reactor. The liquid feed stream introduced into the reactor contains one or more oxidative organic compounds (eg para-xylene), while the gaseous oxidant stream contains molecular oxygen. At least a portion of the molecular oxygen introduced into the reactor as a gas is dissolved in the liquid phase of the reaction medium so that the oxygen can be used for the liquid phase reaction. If the liquid phase of the multiphase reaction medium contains insufficient concentrations of molecular oxygen (ie, certain portions of the reaction medium are "oxygen-deficient"), undesirable side reactions generate impurities and / or the intended reaction There may be a delay in speed. If the liquid phase of the reaction medium contains too little oxidizing compound, the reaction rate may be undesirably low. In addition, if the liquid phase of the reaction medium contains an excess of oxidizing compound, further undesirable side reactions may produce impurities.

Conventional liquid phase oxidation reactors are provided with agitation means for mixing the multiphase reaction medium contained therein. In order to promote the dissolution of the molecular oxygen into the liquid phase of the reaction medium and to maintain a relatively uniform concentration of dissolved oxygen in the liquid phase of the reaction medium, to maintain a relatively uniform concentration of the oxidizing organic compound in the liquid phase of the reaction medium, The reaction medium is shaken.

Often, for example, by means of mechanical shaking means in a vessel such as a continuous stirred tank reactor (CSTR), the reaction medium to be liquid oxidized is shaken. Although CSTR can completely mix the reaction medium, CSTR has a number of disadvantages. For example, CSTRs have a relatively high capital cost because they require expensive motors, fluid-sealed bearings and drive shafts, and / or complex stirring mechanisms. In addition, rotating and / or vibrating mechanical components of conventional CSTRs require regular maintenance. The effort and downtime associated with such maintenance increases the operating cost of the CSTR. However, despite regular maintenance, mechanical shaking systems used in CSTRs are prone to mechanical failure and may require replacement over a relatively short period of time.

Bubble column reactors provide a preferred alternative to CSTR and other mechanical shake oxidation reactors. Bubble column reactors shake the reaction medium without requiring expensive and unreliable mechanical equipment. Bubble column reactors typically include a rectangular granular reaction zone containing the reaction medium. The reaction medium in the reaction zone is shaken primarily by the natural buoyancy of the gas bubbles emerging through the liquid phase of the reaction medium. This natural buoyancy shake provided in the bubble column reactor reduces the capital and maintenance costs associated with a mechanical shake reactor. In addition, since there are substantially no moving mechanical parts involved in the bubble column reactor, an oxidation system is provided which causes less mechanical failure than a mechanical shake reactor.

When performing liquid partial partial oxidation of para-xylene in a conventional oxidation reactor (CSTR or bubble column), the product collected from the reactor is typically a slurry comprising crude terephthalic acid (CTA) and mother liquor. CTA contains a relatively high level of impurities (eg, 4-carboxybenzaldehyde, para-toluic acid, fluorenone and other colors) which make it unsuitable as a feedstock for PET production. Thus, CTA produced in a conventional oxidation reactor is typically subjected to a purification process to convert CTA into purified terephthalic acid (PTA) suitable for PET production.

One typical purification method for converting CTA to PTA includes the following steps: (1) replacing the mother liquor of the CTA-containing slurry with water, (2) heating the CTA / water slurry to dissolve the CTA in water, (3) hydrogenating the CTA / water solution with a catalyst to convert impurities into compounds that can be more preferably and / or easily separated; and (4) precipitation of the resulting PTA from the hydrogenated solution through multiple crystallization steps. (5) separating the crystallized PTA from the remaining liquid. Conventional purification methods of this type are effective but can be very expensive. Industrial factors contributing to the high cost of conventional CTA purification processes are, for example, the thermal energy required to promote the dissolution of CTA in water, the catalyst required for hydrogenation, the hydrogen stream required for hydrogenation, and the yield caused by the hydrogenation of some terephthalic acid. It contains a number of vessels required for loss and multistage crystallization. Therefore, it is desirable to provide a CTA product that can be purified without the need for dissolution, hydrogenation and / or multistage crystallization in water promoted by heat.

Purpose of the Invention

It is therefore an object of the present invention to provide a more effective and economical liquid phase oxidation reactor and method.

It is another object of the present invention to provide a more effective and economical reactor and method for the liquid catalytic catalytic oxidation of para-xylene to terephthalic acid.

Another object of the present invention is to provide a bubble column reactor which promotes an improved liquid phase oxidation reaction while reducing the production of impurities.

It is yet another object of the present invention to provide a more effective and economical system for producing PTA, wherein crude terephthalic acid (CTA) is prepared through liquid phase oxidation of para-xylene and then purified from pure terephthalic acid (PTA). will be.

A further object of the present invention is a CTA product which can be purified without oxidizing para-xylene and without the need for heat-promoted dissolution of CTA in water, hydrogenation of dissolved CTA and / or multistage crystallization of hydrogenated PTA. It is to provide a bubble column reactor for producing a.

It should be noted that the scope of the present invention as defined in the appended claims is not limited to methods or apparatus that can achieve all of the objects listed above. Rather, the scope of the invention as claimed may encompass various systems that do not accomplish all of the above-listed purposes or any of the above-listed purposes. Those skilled in the art will readily recognize additional objects and advantages of the present invention upon reviewing the following detailed description and the accompanying drawings.

Summary of the Invention

One embodiment of the present invention comprises the steps of (a) introducing a feed stream comprising an oxidative compound into the reaction zone of a bubble column reactor; (b) introducing a first oxidant stream comprising molecular oxygen through the one or more bottom oxidant openings into the reaction zone; And (c) introducing a second oxidant stream comprising molecular oxygen through the one or more upper oxidant openings into the reaction zone, wherein the reaction zone has a maximum diameter (D), wherein the upper oxidant opening is the lower oxidant opening. Above about 1D above, each of the first and second oxidant streams is directed to a method containing less than 50 mole percent molecular oxygen.

Another embodiment of the invention comprises the steps of: (a) introducing a predominantly liquid feed stream comprising para-xylene into the reaction zone of a bubble column reactor; (b) introducing a predominantly gaseous first oxidant stream into the reaction zone through the one or more lower oxidant openings; (c) introducing a predominantly gaseous second oxidant stream into the reaction zone through the one or more upper oxidant openings, wherein the reaction zone has a maximum diameter (D), and the upper and lower oxidant openings are at least about 1D. Spaced vertically from each other); (d) oxidizing at least a portion of para-xylene in the liquid phase of the multiphase reaction medium contained in the reaction zone to form crude terephthalic acid; And (e) oxidizing at least a portion of the crude terephthalic acid in a second oxidation reactor to produce purer terephthalic acid.

Another embodiment of the present invention is directed to a bubble column reactor for predominantly reacting a gaseous stream with a predominantly liquid stream. The bubble column reactor includes a vessel shell, one or more liquid openings, and a plurality of gas openings. The vessel shell defines the reaction zone. The reaction zone has a typical lower end and a conventional upper end spaced from each other by a maximum axis distance (L). The reaction zone has a maximum diameter (D) and an L: D ratio of at least about 3: 1. One or more liquid openings introduce a liquid stream into the reaction zone. The plurality of gas openings introduces a gaseous stream into the reaction zone. At least two of the gas openings are axially spaced from each other by about 1D or more. At least one of the gas openings is typically located within about 0.25L of the lower end.

Preferred embodiments of the invention are described in detail below with reference to the accompanying drawings.

1 is a view of the invention, specifically illustrating the introduction of feed, oxidant and reflux streams into the reactor, the presence of a multiphase reaction medium in the reactor and the collection of gases and slurries from the reactor top and bottom (bottom), respectively. A side view of an oxidation reactor constructed in accordance with an embodiment.

FIG. 2 is an enlarged partial side view of the bottom of the bubble column reactor taken along line 2-2 of FIG. 3, specifically illustrating the location and configuration of the oxidant sparger used to introduce the oxidant stream into the reactor. to be.

FIG. 3 is a top plan view of the oxidant sparger of FIG. 2 specifically showing the oxidant openings on top of the oxidant sparger.

FIG. 4 is a bottom plan view of the oxidant sparger of FIG. 2 specifically showing the oxidant openings in the oxidant sparger bottom.

FIG. 5 is a partial side view of the oxidant sparger taken along line 5-5 of FIG. 3, specifically illustrating the orientation of the oxidant openings above and below the oxidant sparger.

6 is an enlarged side view of the bottom of the bubble column reactor, specifically illustrating a system for introducing a feed stream into a reactor at a plurality of vertically spaced locations.

FIG. 7 specifically illustrates how the feed introduction system shown in FIG. 6 distributes the feed stream into the preferred radial feed zone FZ and more than one azimuthal quadrant Q ₁ , Q ₂ , Q ₃ , Q ₄ . It is a partial top plan view taken along the line 7-7 of FIG.

FIG. 8 is a partial top plan view similar to FIG. 7 but showing another means for discharging the feed stream into the reactor using a bayonet tube, each having a plurality of small feed openings.

9 feed streams in multiple vertically spaced locations without requiring multiple vessel penetrations, specifically showing that the feed distribution system can be at least partially supported on the oxidant sparger. Is an isometric view of another system for introducing into the reaction zone.

FIG. 10 is a side view of the single-through feed distribution system and oxidant sparger shown in FIG. 9.

FIG. 11 is a partial top plan view taken along line 11-11 of FIG. 10, further showing a single-through feed distribution system supported on an oxidant sparger.

12 is an isometric view of another oxidant sparger with all oxidant openings located at the bottom of the ring member.

FIG. 13 is a top plan view of another oxidant sparger of FIG. 12.

FIG. 14 is a bottom plan view of another oxidant sparger of FIG. 12 specifically illustrating the location of the bottom opening for introducing an oxidant stream into the reaction zone.

FIG. 15 is a partial side view of the oxidant sparger taken along line 15-15 of FIG. 13, specifically illustrating the orientation of the lower oxidant opening.

16 is a side view of a bubble column reactor with an internal degassing vessel installed near the bottom outlet of the reactor.

FIG. 17 is an enlarged fragmentary side view of the bottom of the bubble column reactor of FIG. 16 taken along line 17-17 of FIG. 18, illustrating in detail the configuration of an internal degassing vessel located at the bottom outlet of the bubble column reactor.

FIG. 18 is a partial top plan view taken along line 18-18 of FIG. 16 specifically showing the vortex breaker disposed in the degassing vessel.

FIG. 19 is a side view of a bubble column reactor equipped with an external degassing vessel, showing how a portion of the degassed slurry exiting the bottom of the degassing vessel can be used to flush a deinventorying line connected to the bottom of the reactor.

20 is a side view of a bubble column reactor equipped with a hybrid internal / external degassing vessel for separating the gas phase of the reaction medium collected from the high side position of the reactor.

21 is a side view of a bubble column reactor with another hybrid degassing vessel installed near the bottom of the reactor.

FIG. 22 is an enlarged fragmentary side view of the bottom of the bubble column reactor of FIG. 21, specifically illustrating the use of another oxidant sparger using an inlet conduit to receive an oxidant stream through the bottom head of the reactor.

FIG. 23 illustrates in detail another means for introducing an oxidant stream into the reactor through a plurality of openings in the reactor bottom head (optionally using an impingement plate to more evenly distribute the oxidant stream in the reactor). An enlarged partial side view similar to FIG.

FIG. 24 is a side view of a bubble column reactor using an internal flow conduit to aid in improved dispersion of oxidative compounds by recycling a portion of the reaction medium from the top of the reactor to the bottom of the reactor.

FIG. 25 is a side view of a bubble column reactor using an external flow conduit to assist in improved dispersion of oxidative compounds by recycling a portion of the reaction medium from the top of the reactor to the bottom of the reactor.

FIG. 26 is an oxidative property in an oxidation reactor, specifically illustrating an extractor using the incoming liquid feed to draw the reaction medium into the eductor and discharge the mixture of feed and reaction medium into the reaction zone at high speed. Partial side view of a horizontal extractor that can be used to improve the dispersion of a compound.

27 is an extractor that mixes a liquid feed and an inlet gas, draws the reaction medium into the extractor using the mixed two-phase fluid, and discharges the mixture of the liquid feed, the inlet gas and the reaction medium at high speed into the reaction zone. Particularly shown is a partial side view of a vertical extractor that can be used to improve the dispersion of oxidizable compounds in an oxidation reactor.

FIG. 28 is a side view of a bubble column reactor containing a multiphase reaction medium, specifically showing the reaction medium being theoretically divided into 30 horizontal slices of equal volume to quantify specific gradients in the reaction medium.

FIG. 29 is a side view of a bubble column reactor containing a multiphase reaction medium, specifically showing the first and second separate 20% continuous volumes of the reaction medium having substantially different oxygen concentrations and / or oxygen consumption rates.

FIG. 30 contains or consists of a multiphase reaction medium, with or without optional mechanical shaking, specifically showing that the vessel contains distinct 20% continuous volumes of reaction medium having substantially different oxygen concentrations and / or oxygen consumption rates. Side view of two stacked reaction vessels that are not supported.

FIG. 31 contains or consists of a multiphase reaction medium, with or without optional mechanical shaking, specifically showing that the vessel contains distinct 20% continuous volumes of reaction medium having substantially different oxygen concentrations and / or oxygen consumption rates. Side view of three side-by-side reaction vessels that are not supported.

32 is a side view of a staged velocity bubble column reactor with a wide bottom reaction zone and a narrow top reaction zone.

33 is a side view of a bubble column reactor equipped with granular partition walls for adding granular surface areas in contact with the reaction medium.

FIG. 34 is a cross-sectional view taken along lines 34-34 of FIG. 33, specifically showing that the partition wall is a planar member that divides the reaction zone into two substantially identical zones.

35 is a side view of a bubble column reactor equipped with a short granular partition wall for adding a granular surface area in contact with the reaction medium.

36 is a side view of a bubble column reactor equipped with short, curved granular partition walls for adding granular surface areas in contact with the reaction medium.

FIG. 37 is a cross-sectional view taken along line 37-37 of FIG. 36, specifically showing that the curved granular partition wall is an approximately S-shaped member that divides a portion of the reaction zone into two substantially identical zones.

38 is a side view of a bubble column reactor equipped with a short granular inner member for adding a granular surface area in contact with the reaction medium.

FIG. 39 is a cross-sectional view taken along lines 39-39 of FIG. 38, specifically showing that the granular inner member has an “X” shape and the edge of the inner member does not extend far to the reactor sidewall.

40 is a side view of a bubble column reactor equipped with differently configured granular inner members for adding granular surface areas in contact with the reaction medium.

FIG. 41 is a cross-sectional view taken along line 41-41 of FIG. 40 specifically showing a configuration of a granular member having an “X” shape and dividing a portion of the reaction zone into four substantially identical quadrants.

FIG. 42 is a cross-sectional view taken along line 42-42 of FIG. 40, specifically illustrating another configuration of the granular member that divides a portion of the reaction zone into eight substantially wedge-shaped zones.

43 is a side view of a bubble column reactor equipped with a plurality of helical inner members for adding a granular surface area in contact with the reaction medium.

FIG. 44 is a cross-sectional view taken along line 44-44 of FIG. 43, specifically illustrating the shape of one of the spiral-shaped inner members.

45 is a side view of a bubble column reactor equipped with a plurality of baffles each including a plurality of cylinder bars for contacting the reaction medium.

FIG. 46 is an enlarged isometric view of the baffle of FIG. 45 detailing how the cylindrical bars of adjacent baffles are rotated 90 ° relative to each other.

FIG. 47 is a cross-sectional view taken along line 47-47 of FIG. 45, specifically illustrating one of the baffles.

48 is a side view of a bubble column reactor equipped with a plurality of baffles each including a plurality of L-shaped members for contacting the reaction medium.

FIG. 49 is an enlarged side view of the baffle of FIG. 48 detailing how the L-shaped members of adjacent baffles are rotated 90 ° relative to each other.

50 is a cross-sectional view taken along line 50-50 of FIG. 48, specifically illustrating one of the baffles.

FIG. 51 is a side view of a bubble column reactor equipped with a single monolithic cylindrical diamond-shaped baffle for contact with the reaction medium.

FIG. 52 is an enlarged side view of the monolithic baffle of FIG. 51.

FIG. 53 is a cross-sectional view taken along lines 53-53 of FIG. 51 showing the cylindrical characteristics of the monolithic baffle.

FIG. 54 is a side view of a bubble column reactor having a plurality of vertically spaced inlets for introducing an oxidant stream into the reaction zone. FIG.

55A and 55B illustrate crude terephthalic acid (CTA) prepared according to one embodiment of the present invention, specifically showing that each CTA particle is a low density high surface area particle composed of a plurality of loosely bound CTA sub-particles. ) Enlarged view of particles.

56A and 56B are fabricated in a conventional manner, specifically showing that conventional CTA particles have larger particle size, lower density, and lower surface area than the CTA particles of the present invention shown in FIGS. 55A and 55B. This is an enlargement of the CTA.

57 is a simplified process flow diagram of a prior art method for preparing purified terephthalic acid (PTA).

58 is a simplified process flow diagram of a method for making a PTA in accordance with an embodiment of the present invention.

FIG. 59 is a table summarizing various operating parameters of the bubble column oxidation reactor, in which specific parameters of the operating parameters were adjusted in accordance with the description in the Examples section.

One embodiment of the present invention relates to liquid phase partial oxidation of oxidizable compounds. This oxidation is preferably carried out in the liquid phase of the multiphase reaction medium contained in one or more shake reactors. Suitable shake reactors include, for example, bubble-shake reactors (eg bubble column reactors), mechanical shake reactors (eg continuous stirred tank reactors) and flow shake reactors (eg jet reactors). . In one embodiment of the present invention, liquid phase oxidation is carried out in a single bubble column reactor.

As used herein, the term "bubble column reactor" refers to a reactor for catalyzing a chemical reaction in a multiphase reaction medium, which shakes the reaction medium mainly by upward movement of gas bubbles through the reaction medium. As used herein, the term “shaking” refers to an operation that disperses the reaction medium causing fluid flow and / or mixing. As used herein, the terms "mostly", "mainly" and "predominantly" mean more than 50%. As used herein, the term "mechanical shaking" refers to shaking of a reaction medium caused by physical movement in or within the reaction medium of the rigid or flexible element (s). For example, mechanical agitation can be provided by rotation, vibration and / or vibration of an internal stirrer, paddle, vibrator or acoustic reflector positioned in the reaction medium. As used herein, the term “flow shaking” refers to the shaking of the reaction medium caused by the high speed injection and / or recycle of one or more fluids in the reaction medium. For example, flow shake can be provided by nozzles, ejectors and / or extractors.

In a preferred embodiment of the invention, less than about 40% of the shaking of the reaction medium of the bubble column reactor during the oxidation is provided by mechanical and / or fluid shaking, more preferably less than about 20% of the shaking is mechanical and / or Provided by flow shake, most preferably less than 5% of the shake is provided by mechanical and / or flow shake. Preferably, the amount of mechanical and / or fluid agitation imparted to the multiphase reaction medium during oxidation is less than about 3 kilowatts per cubic meter of reaction medium, more preferably less than about 2 kilowatts, and most preferably less than 1 kilowatt.

1, the preferred bubble column reactor 20 is shown to include a vessel shell 22 having a reaction zone 24 and a separation zone 26. Reaction zone 24 defines an internal reaction zone 28, while separation zone 26 defines an internal separation zone 30. The predominantly liquid feed stream is introduced into reaction zone 28 through feed inlets 32a, 32b, 32c, 32d. A predominantly gaseous oxidant stream is introduced into reaction zone 28 through oxidant sparger 34 located at the bottom of reaction zone 28. The liquid feed stream and the gaseous oxidant stream cooperate to form the multiphase reaction medium 36 in the reaction zone 28. Multiphase reaction medium 36 includes a liquid phase and a gas phase. More preferably, the multiphase reaction medium 36 comprises a three phase medium having solid, liquid and gaseous components. The solid component of the reaction medium 36 is preferably precipitated in the reaction zone 28 as a result of the oxidation reaction carried out in the liquid phase of the reaction medium 36. Bubble column reactor 20 includes a slurry outlet 38 located near the bottom of reaction zone 28 and a gas outlet 40 located near the top of separation zone 30. Slurry effluent comprising the liquid and solid components of the reaction medium 36 is collected from the reaction zone 28 through the slurry outlet 38, while the predominantly gaseous effluent is separated zone through the gas outlet 40. It is collected from 30.

The liquid feed stream introduced into bubble column reactor 20 through feed inlets 32a, 32b, 32c, 32d preferably comprises an oxidizing compound, a solvent and a catalyst system.

The oxidative compound present in the liquid feed stream preferably comprises one or more hydrocarbyl groups. More preferably, the oxidizing compound is an aromatic compound. Even more preferably, the oxidizable compound is an aromatic compound having one or more attached hydrocarbyl groups or one or more attached substituted hydrocarbyl groups or one or more attached heteroatoms or one or more attached carboxylic acid functional groups (-COOH). Even more preferably, the oxidizable compound is an aromatic compound having one or more attached hydrocarbyl groups or one or more attached substituted hydrocarbyl groups, each of which has 1 to 5 carbon atoms. Even more preferably, the oxidizable composition is an aromatic compound each containing exactly one carbon atom and having exactly two attached groups composed of a methyl group and / or a substituted methyl group and / or no more than one carboxylic acid group. Even more preferably, the oxidizing compound is para-xylene, meta-xylene, para-tolualdehyde, meta-tolualdehyde, para-toluic acid, meta-toluic acid and / or acetaldehyde. Most preferably, the oxidizing compound is para-xylene.

A "hydrocarbyl group" as defined herein is one or more carbon atoms bonded only to hydrogen atoms or other carbon atoms. A "substituted hydrocarbyl group" as defined herein is one or more carbon atoms bonded to one or more heteroatoms and one or more hydrogen atoms. "Heteroatoms" as defined herein are all atoms other than carbon and hydrogen atoms. Aromatic compounds as defined herein preferably comprise aromatic rings having at least 6 carbon atoms, even more preferably only carbon atoms as part of the ring. Suitable examples of such aromatic rings include, but are not limited to, benzene, biphenyl, terphenyl, naphthalene and other carbon-based condensed aromatic rings.

Suitable examples of oxidizable compounds include aliphatic hydrocarbons (eg, alkanes, branched alkanes, cyclic alkanes, aliphatic alkenes, branched alkenes and cyclic alkenes); Aliphatic aldehydes (eg, acetaldehyde, propionaldehyde, isobutyraldehyde and n-butyraldehyde); Aliphatic alcohols (eg, ethanol, isopropanol, n-propanol, n-butanol and isobutanol); Aliphatic ketones (eg, dimethyl ketone, ethyl methyl ketone, diethyl ketone and isopropyl methyl ketone); Aliphatic esters (eg, methyl formate, methyl acetate, ethyl acetate); Aliphatic peroxides, peracids and hydroperoxides (eg, tert-butyl hydroperoxide, peracetic acid and di-tert-butyl hydroperoxide); Aliphatic compounds having groups which are combinations of said aliphatic species with other heteroatoms (e.g., hydrocarbons, aldehydes, alcohols, ketones, esters, peroxides, peracids and / or hydropers combined with sodium, bromine, cobalt, manganese and zirconium) Aliphatic compounds comprising one or more molecular segments of oxide); Various benzene rings, naphthalene rings, biphenyls, terphenyls and other aromatic groups with one or more attached hydrocarbyl groups (eg toluene, ethylbenzene, isopropylbenzene, n-propylbenzene, neopentylbenzene, para-xylene , Meta-xylene, auto-xylene, all isomers of trimethylbenzene, all isomers of tetramethylbenzene, pentamethylbenzene, hexamethylbenzene, all isomers of ethyl-methylbenzene, all isomers of diethylbenzene, ethyl- All isomers of dimethylbenzene, all isomers of dimethylnaphthalene, all isomers of ethyl-methylnaphthalene, all isomers of dimethylnaphthalene, all isomers of dimethylbiphenyl, all isomers of ethyl-methylbiphenyl, and diethylbi All isomers of phenyl, stilbenes and stilbenes, fluorene and one or more moieties having at least one attached hydrocarbyl group The hydro-car fluorene, anthracene and the anthracene-diphenyl ethane has, and diphenyl ethane and one or more attachment dihydro car invoking having at least one attached dihydro car invoking having a pray); Various benzene rings, naphthalene rings, biphenyls, terphenyls and other aromatic groups having one or more attached hydrocarbyl groups and / or one or more attached heteroatoms that may be linked to other atoms or groups of atoms (e.g., phenol, All isomers of methylphenol, all isomers of dimethylphenol, all isomers of naphthol, all isomers of benzyl methyl ether, all isomers of bromophenol, all isomers of bromotoluene, including bromobenzene, alpha-bromotoluene, dibromobenzene , Cobalt naphthenate, and all isomers of bromobiphenyl); Various benzene rings, naphthalene rings, biphenyls, terphenyls and other aromatic groups having one or more attached hydrocarbyl groups and / or one or more attached heteroatoms and / or one or more attached substituted hydrocarbyl groups (eg , All isomers of benzaldehyde, bromobenzaldehyde, all isomers of brominated tolualdehyde, including all isomers of alpha-bromotolualdehyde, all isomers of hydroxybenzaldehyde, all isomers of bromo-hydroxybenzaldehyde, All isomers of benzene dicarboxaldehyde, all isomers of benzene tricarboxaldehyde, all isomers of para-tolualdehyde, meta-tolualdehyde, auto-tolualdehyde, all isomers of toluene dicarboxaldehyde, all isomers of toluene tricarboxaldehyde , Toluene All isomers of lacarboxaldehyde, all isomers of dimethylbenzene dicarboxaldehyde, all isomers of dimethylbenzene tricarboxaldehyde, all isomers of dimethylbenzene tetracarboxaldehyde, all isomers of trimethylbenzene tricarboxaldehyde , All isomers of ethyltolualdehyde, all isomers of trimethylbenzene dicarboxaldehyde, all isomers of tetramethylbenzene dicarboxaldehyde, hydroxymethyl-benzene, hydroxymethyl-toluene, all of isomers of hydroxymethyl-bromotoluene All isomers, all isomers of hydroxymethyl-tolualdehyde, all isomers of hydroxymethyl-bromotolualdehyde, all isomers of benzyl hydroperoxide, benzoyl hydroperoxide, tolyl methyl-hydroperoxide and methylphenol Butyl-all isomers of hydroperoxides); Various benzene rings, naphthalenes, having one or more attached selected groups, which means hydrocarbyl groups and / or attached heteroatoms and / or substituted hydrocarbyl groups and / or carboxylic acid groups and / or peroxy acid groups Rings, biphenyls, terphenyls and other aromatic groups (e.g. benzoic acid, para-toluic acid, meta-toluic acid, auto-toluic acid, all isomers of ethylbenzoic acid, all isomers of propylbenzoic acid, all isomers of butylbenzoic acid , All isomers of pentylbenzoic acid, all isomers of dimethylbenzoic acid, all isomers of ethylmethylbenzoic acid, all isomers of trimethylbenzoic acid, all isomers of tetramethylbenzoic acid, all isomers of pentamethylbenzoic acid, diethylbenzoic acid, all of All isomers, all isomers of benzene tricarboxylic acid, all isomers of methylbenzene dicarboxylic acid, All isomers of dimethylbenzene dicarboxylic acid, all isomers of methylbenzene tricarboxylic acid, all isomers of bromobenzoic acid, all isomers of dibromobenzoic acid, all isomers of bromotoluic acid, including alpha-bromotoluic acid, tolyl acetic acid, hydroxy All isomers of benzoic acid, all isomers of hydroxymethyl-benzoic acid, all isomers of hydroxytoluic acid, all isomers of hydroxymethyl-toluic acid, all isomers of hydroxymethyl-benzene dicarboxylic acid, all isomers of hydroxybromobenzoic acid Isomers, all isomers of hydroxybromotoluic acid, all isomers of hydroxymethyl-bromobenzoic acid, all isomers of carboxy benzaldehyde, all isomers of dicarboxy benzaldehyde, all isomers of perbenzoic acid, hydroperoxymethyl-benzoic acid, Hydroperoxymethyl All isomers of hydroxybenzoic acid, all isomers of hydroperoxycarbonyl-benzoic acid, all isomers of hydroperoxycarbonyl-toluene, all isomers of methylbiphenyl carboxylic acid, all isomers of dimethylbiphenyl carboxylic acid, methylbiphenyl All isomers of dicarboxylic acids, all isomers of biphenyl tricarboxylic acids, all isomers of stilbenes with one or more attached selected groups, all isomers of fluorenone with one or more attached selected groups, all of naphthalene with one or more attached selected groups Isomers, benzyl, all isomers of benzyl with one or more attached selected groups, benzophenones, all isomers of benzophenone with one or more attached selected groups, anthraquinones, all isomers of anthraquinones with one or more attached selected groups, one or moreAll isomers of diphenylethane with attached selected groups, benzocomarin, and all isomers of benzocomarin with one or more attached selected groups).

If the oxidizing compound present in the liquid feed stream is typically a solid compound (ie, a solid at standard temperature and pressure), it is preferred to substantially dissolve the oxidizing compound in the solvent when introducing it into the reaction zone 28. It is preferable that the boiling point of the oxidizing compound at atmospheric pressure is about 50 ° C or higher. More preferably, the boiling point of the oxidizable compound is about 80 to about 400 ° C, most preferably 125 to 155 ° C. The amount of oxidizing compound present in the liquid feed is preferably about 2 to about 40% by weight, more preferably about 4 to about 20% by weight, most preferably 6 to 15% by weight.

Note that the oxidative compound present in the liquid feed may comprise a mixture of two or more different oxidative chemicals. These two or more different chemicals may be mixed and fed into the liquid feed stream or may be fed separately into multiple feed streams. For example, an oxidizing compound comprising para-xylene, meta-xylene, para-tolualdehyde, para-toluic acid and acetaldehyde can be fed to the reactor via a single inlet or multiple separate inlets.

The solvent present in the liquid feed stream preferably comprises an acid component and a water component. The solvent is preferably present in the liquid feed stream at a concentration of about 60 to about 98 weight percent, more preferably about 80 to about 96 weight percent, most preferably 85 to 94 weight percent. The acid component in the solvent is preferably an organic low molecular weight monocarboxylic acid having mainly 1 to 6 carbon atoms, more preferably 2 carbon atoms. Most preferably, the acid component of the solvent is mainly acetic acid. Preferably, the acid component constitutes at least about 75%, more preferably at least about 80%, most preferably 85 to 98% by weight of the solvent, with the remainder being primarily water. Solvents introduced into bubble column reactor 20 are, for example, para-tolualdehyde, terephthalaldehyde, 4-carboxybenzaldehyde (4-CBA), benzoic acid, para-toluic acid, para-toluic acid aldehyde, alpha-bromo-para -May contain minor amounts of impurities such as toluic acid, isophthalic acid, phthalic acid, trimellitic acid, polyaromatic compounds and / or suspended particulates. Preferably, the total amount of impurities in the solvent introduced into the bubble column reactor 20 is less than about 3 weight percent.

The catalyst system present in the liquid feed stream is preferably a homogeneous liquid catalyst system capable of promoting oxidation (including partial oxidation) of oxidizable compounds. More preferably, the catalyst system comprises one or more polyvalent transition metals. Even more preferably, the polyvalent transition metal comprises cobalt. Even more preferably, the catalyst system comprises cobalt and bromine. Most preferably, the catalyst system comprises cobalt, bromine and manganese.

If cobalt is present in the catalyst system, the amount of cobalt present in the liquid feed stream may be such that the concentration of cobalt in the liquid phase of the reaction medium 36 is from about 300 to about 6,000 ppmw, more preferably from about 700 to about 4,200 ppmw, Most preferably, it is maintained at 1,200 to 3,000 ppmw. When bromine is present in the catalyst system, the amount of bromine present in the liquid feed stream is such that the concentration of bromine in the liquid phase of the reaction medium 36 is about 300 to about 5,000 ppmw, more preferably about 600 to about 4,000 ppmw, most Preferably it is desirable to maintain at 900 to 3,000ppmw. If manganese is present in the catalyst system, the amount of manganese present in the liquid feed stream is such that the concentration of manganese in the liquid phase of the reaction medium 36 is from about 20 to about 1,000 ppmw, more preferably from about 40 to about 500 ppmw, most preferred. Preferably it is to be maintained at 50 to 200ppmw.

The concentrations of cobalt, bromine and / or manganese in the liquid phase of reaction medium 36, described above, are expressed based on time-average and volume-average. As used herein, the term “time-average” refers to the average of ten or more measurements measured equally over successive times of 100 seconds or more. As used herein, the term “volume-average” refers to the average of ten or more measurements measured at uniform three-dimensional intervals throughout a particular volume.

The weight ratio (Co: Br) of cobalt to bromine in the catalyst system introduced into the reaction zone 28 is preferably from about 0.25: 1 to about 4: 1, more preferably from about 0.5: 1 to about 3: 1, most Preferably 0.75: 1 to 2: 1. The weight ratio (Co: Mn) of cobalt to manganese in the catalyst system introduced into the reaction zone 28 is preferably about 0.3: 1 to about 40: 1, more preferably about 5: 1 to about 30: 1, most Preferably from 10: 1 to 25: 1.

The liquid feed stream introduced into bubble column reactor 20 is for example toluene, ethylbenzene, para-tolualdehyde, terephthalaldehyde, 4-carboxybenzaldehyde (4-CBA), benzoic acid, para-toluic acid, para Small amounts of impurities such as toluic acid aldehyde, alpha bromo para-toluic acid, isophthalic acid, phthalic acid, trimellitic acid, polyaromatic compounds and / or suspended particulates. When bubble column reactor 20 is used to prepare terephthalic acid, meta-xylene and auto-xylene are also considered impurities. Preferably, the total amount of impurities in the liquid feed stream introduced into bubble column reactor 20 is less than about 3 weight percent.

FIG. 1 illustrates an embodiment in which the oxidizing compound, solvent and catalyst system are mixed together and introduced into bubble column reactor 20 as a single feed stream, but in another embodiment of the present invention, It may be introduced separately into the bubble column reactor (20). For example, a pure para-xylene stream may be fed into bubble column reactor 20 through an inlet separate from solvent and catalyst inlet (s).

The predominantly gaseous oxidant stream introduced into bubble column reactor 20 through oxidant sparger 34 comprises molecular oxygen (O ₂ ). Preferably, the oxidant stream comprises about 5 to about 40 mole percent, more preferably about 15 to about 30 mole percent, most preferably 18 to 24 mole percent molecular oxygen. It is preferred that the remainder of the oxidant stream consists mainly of a gas or gases such as nitrogen which are inert to oxidation. More preferably, the oxidant stream consists essentially of molecular oxygen and nitrogen. Most preferably, the oxidant stream is anhydrous air comprising about 21 mole percent molecular oxygen and about 78 to about 81 mole percent nitrogen. In another embodiment of the present invention, the oxidant stream may comprise substantially pure oxygen.

Referring again to FIG. 1, bubble column reactor 20 is preferably equipped with a reflux distributor 42 located above the upper surface 44 of the reaction medium 36. Reflux distributor 42 may be operated by any droplet forming means known in the art to introduce droplets of the predominantly liquid reflux stream into separation zone 30. More preferably, the reflux distributor 42 produces a droplet of spray downwards towards the upper surface 44 of the reaction medium 36. Preferably, the downward spray of the droplets covers (ie, affects correspondingly) at least about 50% of the maximum horizontal cross-sectional area of the separation zone 30. More preferably, the spray of droplets covers at least about 75% of the maximum horizontal cross-sectional area of the separation zone 30. Most preferably, the spray of droplets covers at least 90% of the maximum horizontal cross-sectional area of the separation zone 30. This downward liquid reflux spray can help prevent bubbles from generating at or above the top surface 44 of the reaction medium 36 and also entrains any upwardly moving gas flowing towards the gas outlet 40. It can also help to separate liquid or slurry droplets. In addition, liquid reflux may also contain particulates and possibly precipitated compounds present in the gaseous effluent collected from separation zone 30 via gas outlet 40 (eg, dissolved benzoic acid, para-toluic acid, 4-CBA, Terephthalic acid and catalytic metal salts). It is also possible to adjust the composition of the gaseous effluent collected through the gas outlet 40 by distillation by introducing reflux droplets into the separation zone 30.

The liquid reflux stream introduced into bubble column reactor 20 through reflux distributor 42 is preferably introduced into bubble column reactor 20 through feed inlets 32a, 32b, 32c, and 32d. It has approximately the same composition as the solvent component of. Thus, it is preferred that the liquid reflux stream comprises an acid component and water. The acid component of the reflux stream is preferably a low molecular weight organic monocarboxylic acid having 1 to 6 carbon atoms, more preferably 2 carbon atoms. Most preferably, the acid component of the reflux stream is acetic acid. Preferably, the acid component comprises at least about 75%, more preferably at least about 80%, most preferably 85 to 98% by weight of the reflux stream, with the remainder being water. Since the reflux stream typically has a composition that is substantially the same as the solvent of the liquid feed stream, when such substrate refers to the "total solvent" introduced into the reactor, this "total solvent" refers to both the solvent portion of the reflux stream and the feed stream. Contains different.

During the liquid phase oxidation in bubble column reactor 20, the feed, oxidant and reflux streams are continuously continuously into reaction zone 28, while the gas and slurry effluent streams are collected substantially continuously from reaction zone 28. It is preferable to introduce. As used herein, the term “substantially continuously” means a period of at least 10 hours that is disconnected in less than 10 minutes. During the oxidation, at least about 8,000 kg, more preferably about 13,000 to about 80,000 kg, even more preferably about 18,000 to about 50,000 kg, most preferably 22,000 to 30,000 oxidizing compounds (eg para-xylene) per hour It is preferred to introduce substantially continuously into the reaction zone 28 at a rate of kg. While it is desirable that the flow rates of the incoming feed, oxidant and reflux stream remain substantially unchanged, one embodiment of the present invention provides a pulsed feed of the incoming feed, oxidant and / or reflux stream to improve mixing and mass transfer. Note that the variation is considered. When the feeds, oxidants and / or reflux streams are introduced by pulsed fluctuations, their flow rates are from about 0 to about 500% of the steady state flow rates mentioned herein, more preferably the steady state flow rates mentioned herein. It is preferred to vary within about 30 to about 200% of the most preferably 80 to 120% of the steady-state flow rates mentioned herein.

The average space-time velocity (STR) of the reaction in bubble column oxidation reactor 20 is the mass of oxidative compound fed per unit volume of reaction medium 36 per unit time (e.g., of para-xylene fed per m3 per hour). kg number). In conventional use, the amount of oxidizing compound that has not been converted to the product prior to calculating the STR is typically subtracted from the amount of the oxidizing compound in the feed stream. However, many of the preferred oxidizing compounds (eg para-xylene) herein are typically high in conversion and yield, and it is convenient to define this term as mentioned above. It is generally desirable to carry out the reaction with a high STR, especially for capital costs and operational inventory reasons. However, conducting reactions at increasingly higher STRs can affect the quality or yield of partial oxidation. Bubble column reactor 20 has an STR of an oxidizing compound (eg, para-xylene) of about 25 kg / m 3 / hour to about 400 kg / m 3 / hour, more preferably about 30 kg / m 3 / hour to about 250 kg / m 3 / Particularly useful in the case of hours, even more preferably from about 35 kg / m 3 / hour to about 150 kg / m 3 / hour, most preferably from 40 kg / m 3 / hour to 100 kg / m 3 / hour.

Oxygen-STR in bubble column reactor 20 is defined as the weight of molecular oxygen consumed per unit volume of reaction medium 36 per unit time (eg, kg number of molecular oxygen consumed per m 3 per hour). Especially for reasons of capital cost and consumption due to oxidation of the solvent, it is usually preferred to carry out the reaction with high oxygen-STR. However, carrying out the reaction with increasingly higher oxygen-STR ultimately reduces the quality or yield of partial oxidation. Without being bound by a particular theory, it appears to be related to the rate of delivery of molecular oxygen, possibly from the gaseous phase into the liquid and then into the bulk liquid, at the interface surface region. If the oxygen-STR is too high, the amount of dissolved oxygen in the bulk liquid phase of the reaction medium may be too low.

Total-average-oxygen-STR is defined herein as the weight of all oxygen consumed in the total volume of reaction medium 36 per unit time (eg, kg number of molecular oxygen consumed per m 3 per hour). Bubble column reactor 20 has a total-average-oxygen-STR of about 25 kg / m 3 / hour to about 400 kg / m 3 / hour, more preferably about 30 kg / m 3 / hour to about 250 kg / m 3 / hour, even more preferred. Preferably about 35 kg / m 3 / hour to about 150 kg / m 3 / hour, most preferably 40 kg / m 3 / hour to 100 kg / m 3 / hour.

During oxidation in bubble column reactor 20, the ratio of the mass flow rate of the total solvent (from both the feed stream and the reflux stream) to the mass flow rate of the oxidizing compound entering the reaction zone 28 is from about 2: 1 to about 50: 1, more preferably from about 5: 1 to about 40: 1, most preferably from 7.5: 1 to 25: 1. Preferably, the ratio of the mass flow rate of solvent introduced as part of the feed stream to the mass flow rate of solvent introduced as part of the reflux stream is from about 0.5: 1 to the absence of reflux stream flow, more preferably about 0.5: 1. To about 4: 1, even more preferably about 1: 1 to about 2: 1, most preferably 1.25: 1 to 1.5: 1.

During liquid phase oxidation in bubble column reactor 20, it is desirable to introduce an oxidant stream into bubble column reactor 20 in an amount that provides molecular oxygen that slightly exceeds the stoichiometric oxygen demand. The amount in excess of the molecular oxygen required to achieve the best results with a particular oxidizing compound affects the overall economics of liquid phase oxidation. During liquid phase oxidation in bubble column reactor 20, the ratio of the mass flow rate of the oxidant stream to the mass flow rate of the oxidative organic compound (eg, para-xylene) entering the reactor 20 is from about 0.5: 1 to about 20: 1. More preferably from about 1: 1 to about 10: 1, most preferably from 2: 1 to 6: 1.

Referring again to FIG. 1, the feed, oxidant and reflux stream introduced into bubble column reactor 20 cooperate to form at least a portion of the multiphase reaction medium 36. The reaction medium 36 is preferably a three phase medium including solid phase, liquid phase and gas phase. As mentioned above, the oxidation of the oxidizing compound (eg para-xylene) takes place predominantly in the liquid phase of the reaction medium 36. Thus, the liquid phase of reaction medium 36 includes dissolved oxygen and oxidative compounds. Because of the exothermic nature of the oxidation reactions occurring in bubble column reactor 20, some of the solvents (eg, acetic acid and water) introduced through feed inlets 32a, 32b, 32c, 32d are boiled / vaporized. Therefore, the gas phase of the reaction medium 36 of the reactor 20 consists mainly of the unresolved, unreacted portion of the vaporized solvent and oxidant stream. Certain prior art oxidation reactors use heat exchange tubes / fins to heat or cool the reaction medium. However, such heat exchange structures may be undesirable for the reactors and methods of the present invention described herein. Thus, it is preferred that bubble column reactor 20 is substantially free of surfaces that are in contact with reaction medium 36 and exhibit a time-averaged heat flux greater than 30,000 watts per square meter.

The concentration of oxygen dissolved in the liquid phase of the reaction medium 36 is the dynamic equilibrium between the rate of mass transfer from the gas phase and the rate of consumption of reactants in the liquid phase (ie Despite being a factor of the feed rate and affecting the upper concentration of dissolved oxygen, it is not simply set by this partial pressure). The amount of dissolved oxygen varies locally, higher near the bubble interface. In total, the amount of dissolved oxygen depends on the equilibrium of supply and demand factors in different regions of the reaction medium 36. Temporarily, the amount of dissolved oxygen depends on the uniformity of gas and liquid mixing related to the rate of chemical consumption. In designing to roughly match the supply and demand of dissolved oxygen in the liquid phase of the reaction medium 36, the time-average and volume-average oxygen concentrations in the liquid phase of the reaction medium 36 are higher than about 1 ppm molar, more Preferably from about 4 to about 1,000 ppm mol, even more preferably from about 8 to about 500 ppm mol, most preferably from 12 to 120 ppm mol.

The liquid phase oxidation reaction carried out in the bubble column reactor 20 is preferably a precipitation reaction to produce a solid. More preferably, at least about 10% by weight of the oxidative compound (eg, para-xylene) introduced into the reaction zone 28 is liquid phase oxidation carried out in the bubble column reactor 20 in the reaction medium 36. (Eg, crude terephthalic acid particles). Even more preferably, liquid phase oxidation causes at least about 50% by weight of the oxidative compound to produce a solid compound in the reaction medium 36. Most preferably, at least 90% by weight of the oxidizable compound produces a solid compound in the reaction medium 36 by liquid phase oxidation. Preferably, the total amount of solids in the reaction medium 36 is higher than about 3 weight percent based on time-average and volume-average. More preferably, the total amount of solids in the reaction medium 36 is maintained at about 5 to about 40 weight percent, even more preferably about 10 to about 35 weight percent, most preferably 15 to 30 weight percent. It is preferred that a substantial portion of the oxidation product (eg, terephthalic acid) produced in the bubble column reactor 20 remains as a solid in the reaction medium 36 without remaining dissolved in the liquid phase of the reaction medium 36. The amount of solid phase oxidation product present in the reaction medium 36 is at least about 25% by weight of the total oxidation product (solid and liquid) in the reaction medium 36, more preferably about 75% of the total oxidation product in the reaction medium 36. Or at least about 95% by weight of the total oxidation product in the reaction medium 36. The numerical ranges described above for the amount of solids in the reaction medium 36 apply to substantially steady state operation of the bubble column reactor 20 over substantially continuous time, starting and shutting down the bubble column reactor 20. Or does not apply to sub-optimal operation. The amount of solids in the reaction medium 36 is determined by gravimetric method. In this gravimetric method, representative portions of the slurry are collected from the reaction medium and weighed. Under conditions that effectively maintain the overall solid-liquid distribution present in the reaction medium, the glass liquid is solids by sedimentation or filtration effectively, with no loss of precipitated solids, but with less than about 10% of the initial liquid mass remaining in the solid portion. Remove from part. The liquid remaining in the solid is evaporated until it is effectively dried without sublimation of the solid. Weigh the remaining solid portion. The ratio of the weight of the solid portion to the weight of the original slurry portion is typically the fraction of solids expressed as a percentage.

Precipitation reactions carried out in bubble column reactor 20 may cause contamination (ie, solid buildup) on the surface of certain rigid structures in contact with reaction medium 36. Thus, in one embodiment of the present invention, it is preferred that bubble column reactor 20 substantially comprises no internal heat exchange, agitation, or baffle structures in the reaction zone 28 (since such structures are susceptible to contamination). If there is an internal structure in the reaction zone 28, it is desirable to avoid an internal structure having an outer surface comprising a substantial amount of upward facing surface (upward facing) (these facing upward planar surfaces are contaminated). Because it is very easy). Therefore, if any internal structure is present in the reaction zone 28, less than about 20% of all of the upward exposed outer surface area of this internal structure consists of a substantially planar surface that is inclined from less than about 15 ° from horizontal. It is preferable.

Referring again to FIG. 1, the physical configuration of bubble column reactor 20 helps to optimize the oxidation of oxidizable compounds (eg, para-xylene) with minimal generation of impurities. It is preferred that the rectangular reaction zone 24 of the vessel shell 22 comprises a substantially cylindrical main body portion 46 and a lower head 48. The upper end of the reaction zone 28 is defined by a horizontal plane 50 extending across the top of the cylindrical main body portion 46. The lower end 52 of the reaction zone 28 is defined by the lowest inner surface of the lower head 48. Typically, the lower end 52 of the reaction zone 28 is located proximate to the opening for the slurry outlet 38. Therefore, the rectangular reaction zone 28 defined in the bubble column reactor 20 is the maximum measured from the top end 50 to the bottom end 52 of the reaction zone 28 along the extension axis of the cylindrical main body portion 46. Has a length "L". The length "L" of the reaction zone 28 is preferably about 10 to about 100 m, more preferably about 20 to about 75 m and most preferably 25 to 50 m. The reaction zone 28 typically has a maximum diameter (width) "D" equal to the maximum internal diameter of the cylindrical main body portion 46. The maximum diameter "D" of the reaction zone 28 is about 1 to about 12 m, more preferably about 2 to about 10 m, even more preferably about 3.1 to about 9 m, most preferably 4 to 8 m. In a preferred embodiment of the invention, reaction zone 28 has a length to diameter “L: D” ratio of about 6: 1 to about 30: 1. Even more preferably, reaction zone 28 has an L: D ratio of about 8: 1 to about 20: 1. Most preferably, reaction zone 28 has an L: D ratio of 9: 1 to 15: 1.

As disclosed above, reaction zone 28 of bubble column reactor 20 contains a multiphase reaction medium 36. The reaction medium 36 has a bottom end that coincides with the lower end 52 of the reaction zone 28 and an upper end located on the upper surface 44. The upper surface 44 of the reaction medium 36 is defined along a horizontal plane that is cut through the reaction zone 28 in a vertical position where the contents of the reaction zone 28 transition from the gaseous-continuous state to the liquid-continuous state. . The upper surface 44 is preferably located in a vertical position where the local time-averaged gas hold-up of the thin horizontal slice of the contents of the reaction zone 28 is 0.9.

The reaction medium 36 has a maximum height "H" measured between its upper and lower ends. The maximum width "W" of the reaction medium 36 is typically equal to the maximum diameter "D" of the cylindrical main body portion 46. During liquid phase oxidation in bubble column reactor 20, it is preferred that H be maintained at about 60 to about 120%, more preferably about 80 to about 110%, and most preferably 85 to 100% of L. In a preferred embodiment of the present invention, reaction medium 36 has a height-to-width "H: W" ratio of greater than about 3: 1. More preferably, reaction medium 36 has an H: W ratio of about 7: 1 to about 25: 1. Even more preferably, reaction medium 36 has an H: W ratio of about 8: 1 to about 20: 1. Most preferably, reaction medium 36 has an H: W ratio of 9: 1 to 15: 1. In one embodiment of the present invention, L = H and D = W so that the various dimensions or ratios provided for L and D herein also apply to H and W, and vice versa.

The relatively high L: D and H: W ratios provided in accordance with embodiments of the present invention may contribute to some important advantages of the system of the present invention. As discussed in more detail below, higher L: D and H: W ratios, and certain other features discussed below, may be attributed to molecular oxygen and / or oxidizing compounds (eg, para-xylene) in the reaction medium 36. It has been found that it can promote an advantageous vertical gradient in terms of the concentration of. In contrast to the conventional knowledge of favoring well-mixed reaction media with relatively uniform concentrations overall, vertical stepping of oxygen and / or oxidative compound concentrations has been found to promote more effective and economical oxidation reactions. By minimizing the oxygen and oxidative compound concentration near the top of the reaction medium 36, it can help to avoid the loss of unreacted oxygen and unreacted oxidizing compound through the upper gas outlet 40. However, when the concentrations of oxidizing compound and unreacted oxygen are low throughout the reaction medium 36, the rate and / or selectivity of oxidation is reduced. Therefore, it is desirable that the concentration of molecular oxygen and / or oxidizing compound be significantly higher near the bottom of the reaction medium 36 than near the top of the reaction medium 36.

In addition, the high L: D and H: W ratios allow the pressure at the bottom of the reaction medium 36 to be significantly greater than the pressure at the top of the reaction medium 36. This vertical pressure gradient is the result of the height and density of the reaction medium 36. One advantage of this vertical pressure gradient is that due to the elevated pressure at the bottom of the vessel, higher oxygen solubility and more mass transfer can be achieved than would otherwise be achieved at comparable temperatures and overhead pressures in shallow reactors. . Therefore, the oxidation reaction can be carried out at a lower temperature than necessary in the shallower vessel. When bubble column reactor 20 is used to partially oxidize para-xylene to produce crude terephthalic acid (CTA), the ability to operate at lower reaction temperatures with the same or better oxygen mass transfer rate has many advantages. Has For example, low temperature oxidation of para-xylene reduces the amount of solvent burned during the reaction. As discussed in more detail below, low temperature oxidation also promotes the production of small, high surface area, loosely bound and easily dissolved CTA particles, which are large and low surface area produced by conventional high temperature oxidation processes, It can be purified more economically than dense CTA particles.

During the oxidation in the reactor 20, the time-averaged and volume-averaged temperatures of the reaction medium 36 are from about 125 to about 200 ° C, more preferably from about 140 to about 180 ° C, most preferably 150 to 170 ° C. Is preferably maintained. The overhead pressure over the reaction medium 36 is preferably maintained at about 1 to about 20 barg, more preferably about 2 to about 12 barg, most preferably 4 to 8 barg. Preferably, the pressure difference between the top of the reaction medium 36 and the bottom of the reaction medium 36 is about 0.4 to about 5, more preferably the pressure difference is about 0.7 to about 3 bar, most preferably The pressure difference is 1 to 2 bar. While it is generally desirable for the overhead pressure above the reaction medium 36 to be maintained at a relatively constant value, one embodiment of the present invention provides an overload to facilitate improved mixing and / or mass transfer in the reaction medium 36. Consider pulse head fluctuations. If the overhead pressure fluctuates pulsed, the pulsed fluctuating pressure is about 60 to about 140% of the steady state overhead pressure cited herein, more preferably about the steady state overhead pressure cited herein. It is preferably from 85 to about 115%, most preferably from 95 to 105% of the steady state overhead pressure recited herein.

An additional advantage of the high L: D ratio of the reaction zone 28 is that this may contribute to an increase in the average surface speed of the reaction medium 36. The terms "surface velocity" and "surface gas velocity" as used herein in connection with the reaction medium 36 are divided by the horizontal cross-section of the reactor at a specific height, and the volumetric flow rate of the gas phase of the reaction medium 36 at this height of the reactor. It is called. The increase in surface velocity provided by the high L: D ratio of reaction zone 28 promotes local mixing of reaction medium 36 and increases gas retention. The time-averaged surface velocity of the reaction medium 36 at 1/4, 1/2 and / or 3/4 of the height of the reaction medium 36 is preferably greater than about 0.3 m / sec, more preferably About 0.8 to about 5 m / sec, even more preferably about 0.9 to about 4 m / sec, most preferably 1 to 3 m / sec.

Referring again to FIG. 1, separation zone 26 of bubble column reactor 20 is simply an expanded portion of vessel shell 22 located just above reaction zone 24. Separation zone 26 reduces the rate of up-moving gas phase in bubble column reactor 20 as gas phase rises above top surface 44 of reaction medium 36 to approach gas outlet 40. This reduction in the upward velocity of the gas phase aids in the easy removal of the liquids and / or solids entrained in the upward flow gaseous phase, thus reducing the undesirable loss of certain components present in the liquid phase of the reaction medium 36.

The separation zone 26 preferably comprises a transition wall 54 which is approximately frustoconical, a wide sidewall 56 which is approximately cylindrical and an upper head 58. The narrow lower end of the transition wall 54 is connected to the top of the cylindrical main body portion 46 of the reaction zone 24. The wide top end of the transition wall 54 is connected to the bottom of the wide sidewall 56. The transition wall 54 extends upwardly and outwardly from its narrow lower end at an angle of about 10 to about 70 degrees in the vertical, more preferably about 15 to about 50 degrees in the vertical, and most preferably 15 to 45 degrees in the vertical. It is desirable to be. Where the top of the reaction zone 24 has a diameter smaller than the overall maximum diameter of the reaction zone 24, the wide sidewall 56 generally has a reaction zone 24, although X may actually be smaller than D. Has a maximum diameter "X" greater than the maximum diameter "D". In a preferred embodiment of the invention, the ratio "X: D" of the diameter of the wide sidewall 56 to the maximum diameter of the reaction zone 24 is from about 0.8: 1 to about 4: 1, most preferably from 1.1: 1 to 2: 1. The upper head 58 is connected to the top of the wide sidewall 56. The upper head 58 is preferably an approximately elliptical head member that defines a central opening that allows gas to exit the separation zone 30 through the gas outlet 40. Alternatively, the upper head 58 may be of any shape, including conical. The separation zone 30 has a maximum height "Y" measured from the top 50 of the reaction zone 28 to the top of the separation zone 30. The ratio “L: Y” of the length of the reaction zone 28 to the height of the separation zone 30 is preferably from about 2: 1 to about 24: 1, more preferably from about 3: 1 to about 20: 1, Most preferably 4: 1 to 16: 1.

Referring now to FIGS. 1-5, the location and configuration of the oxidant sparger 34 is discussed in more detail. 2 and 3 show that the oxidant sparger 34 can include a ring member 60, a transverse member 62 and a pair of oxidant inlet conduits 64a, 64b. For convenience, these oxidant inlet conduits 64a and 64b may enter the vessel at a position higher than the ring member 60 and then bend downward as shown in FIGS. 2 and 3. Alternatively, the oxidant inlet conduits 64a and 64b may enter the vessel below the ring member 60 or on a substantially flat horizontal plane as the ring member 60. Each oxidant inlet conduit 64a, 64b includes a first end connected to individual oxidant inlets 66a, 66b formed in the vessel shell 22 and a second end variably connected to the ring member 60. The ring member 60 preferably consists of a conduit, more preferably of a plurality of straight conduit sections, most preferably of a plurality of straight pipe sections that are strongly connected to one another to form a tubular polygonal ring. Preferably, the ring member 60 consists of at least three straight pipe sections, more preferably 6 to 10 pipe sections and most preferably 8 pipe sections. Thus, when the ring member 60 consists of eight pipe sections, it has a roughly octagonal structure. The cross member 62 preferably consists of a substantially straight pipe section which is variably connected between the opposing pipe sections of the ring member 60 and extends diagonally between the pipe sections. The pipe section used for the cross member 62 preferably has a diameter substantially the same as the pipe section used to form the ring member 60. The pipe section constituting the oxidant inlet conduits 64a, 64b, ring member 60 and cross member 62 is greater than about 0.1 m, more preferably about 0.2 to about 2 m, most preferably 0.25 to 1 m It is preferred to have a nominal diameter which is As best shown in FIG. 3, the ring member 60 and the cross member 62 each provide a plurality of upper oxidant openings 68 for releasing the oxidant stream upward into the reaction zone 28. As best shown in FIG. 4, the ring member 60 and / or the transverse member 62 can provide one or more lower oxidant openings 70 for releasing the oxidant stream downward into the reaction zone 28. . Lower oxidant openings 70 may also be used to release liquids and / or solids that have entered the ring member 60 and / or the transverse member 62. In order to prevent the buildup of solids within the oxidant sparger 34, the liquid stream can be continuously or periodically passed through the sparger 34 to flush any accumulated solids.

Referring again to FIGS. 1 through 4, during oxidation in bubble column reactor 20, oxidant streams are pushed into oxidant inlet conduits 64a and 64b through oxidant inlets 66a and 66b, respectively. The oxidant stream is then conveyed to the ring member 60 through the oxidant inlet conduits 64a and 64b. After the oxidant stream enters the ring member 60, the oxidant stream is distributed throughout the interior volume of the ring member 60 and the cross member 62. The oxidant stream is then pushed out of the oxidant sparger 34 and into the reaction zone 28 through the upper and lower oxidant openings 68, 70 of the ring member 60 and the cross member 62.

The outlets of the upper oxidant opening 68 are laterally spaced apart from each other and are located at substantially the same height in the reaction zone 28. Therefore, the outlet of the upper oxidant opening 68 is generally located along a substantially horizontal plane defined by the top of the oxidant sparger 34. The outlets of the lower oxidant opening 70 are laterally spaced apart from one another and are located at substantially the same height in the reaction zone 28. Thus, the outlet of the lower oxidant opening 70 is generally located along a substantially horizontal plane defined by the bottom of the oxidant sparger 34.

In one embodiment of the present invention, oxidant sparger 34 has at least about 20 upper oxidant openings 68 formed therein. More preferably, oxidant sparger 34 has about 40 to about 800 upper oxidant openings formed therein. Most preferably, oxidant sparger 34 has 60 to 400 upper oxidant openings 68 formed therein. The oxidant sparger 34 preferably has approximately one or more lower oxidant openings 70 formed therein. More preferably, oxidant sparger 34 has about 2 to about 40 lower oxidant openings 70 formed therein. Most preferably, the oxidant sparger 34 has 8 to 20 lower oxidant openings 70 formed therein. The ratio of the number of the upper oxidant opening 68 to the lower oxidant opening 70 of the oxidant sparger 34 is preferably from about 2: 1 to about 100: 1, more preferably from about 5: 1 to about 25: 1, most preferably 8: 1 to 15: 1. The diameters of substantially all of the upper and lower oxidant openings 68, 70 are preferably substantially the same so that the ratio of the volume flow rate of the oxidant stream exiting the upper opening 68 and the lower opening 70 is upper oxidant opening 68. And for the relative number of lower oxidant openings 70 is substantially the same as the ratio described above.

5 shows the direction of oxidant release from the top and bottom oxidant openings 68, 70. With respect to the top oxidant opening 68, it is preferred that at least a portion of the top oxidant opening 68 release the oxidant stream at an angle "A" that is tilted from the vertical. The percentage of the upper oxidant opening 68 inclined from the vertical by the angle "A" is about 30 to about 90%, more preferably about 50 to about 80%, even more preferably 60 to 75%, most preferably about 67% is preferable. The angle "A" is preferably about 5 to about 60 degrees, more preferably about 10 to about 45 degrees, most preferably 15 to 30 degrees. For the lower oxidant opening 70, it is preferred that substantially all the lower oxidant opening 70 is located near the bottom (bottom) of the ring member 60 and / or the transverse member 62. Thus, any liquids and / or solids that may inadvertently enter the oxidant sparger 34 may be easily released from the oxidant sparger 34 through the lower oxidant opening 70. Preferably, the lower oxidant opening 70 discharges the oxidant stream downward at a substantially vertical angle. Here, the upper compensator opening can be any opening that releases the oxidant stream in an upward direction (ie, at an angle greater than horizontal), and the lower oxidant opening is in an approximately downward direction (ie, horizontal). At any angle).

In many conventional bubble column reactors containing a multiphase reaction medium, substantially all of the reaction medium located under the oxidant sparger (or other mechanism for introducing the oxidant stream into the reaction zone) has very low gas retention values. . As is known in the art, "gas retention" is simply the volume fraction of the multiphase medium in the gaseous state. The zone of low gas retention in the medium may also be referred to as the "degassed zone". In many conventional slurry bubble column reactors, a significant portion of the total volume of the reaction medium is located below the oxidant sparger (or other mechanism for introducing the oxidant stream into the reaction zone). Thus, a significant portion of the reaction medium present in the bottom of the conventional bubble column reactor is degassed.

It has been found that minimizing the amount of degassed zone in the reaction medium that is oxidized in a bubble column reactor can minimize the occurrence of certain types of undesirable impurities. The degassed zone of the reaction medium contains a relatively small amount of oxidant bubble. The presence of this small volume of oxidant bubbles reduces the amount of molecular oxygen that can be used to dissolve into the liquid phase of the reaction medium. Thus, the liquid phase in the degassed zone of the reaction medium has a relatively low molecular oxygen concentration. This oxygen-depleted degassed zone of the reaction medium tends to promote undesirable side reactions rather than the desired oxidation reaction. For example, when partial oxidation of para-xylene produces terephthalic acid, undesirable amounts of benzoic acid and coupled aromatic rings (known as fluorenone and anthraquinone) due to insufficient oxygen utilization in the liquid phase of the reaction medium. Presenting highly undesirable colored molecules).

According to one embodiment of the invention, liquid phase oxidation is carried out in a bubble column reactor constructed and operated in such a way that the volume fraction of the reaction medium having a low gas retention value is minimized. Minimization of this degassed zone can be quantified by theoretically dividing the total volume of the reaction medium into 2,000 separate horizontal layers of uniform volume. Except for the highest and lowest horizontal slices, each horizontal slice is a separate volume whose sides are bounded by the side walls of the reactor and whose top and bottoms are bounded by an imaginary horizontal plane. The highest horizontal strata are bounded at their bottom by an imaginary horizontal plane and bounded at their top by the top surface of the reaction medium. The lowest horizontal strata are bounded at their top by an imaginary horizontal plane and at their bottom by the lower end of the vessel. After theoretically dividing the reaction medium into 2,000 separate horizontal partitions of the same volume, the time-average and volume-average gas retention of each horizontal partition can be determined. When using this method of quantifying the amount of degassed zone, the number of horizontal strata having a time-average and volume-average gas retention of less than 0.1 is less than 30, more preferably less than 15, even more preferably 6 It is preferred that there are less than two, even more preferably less than four, most preferably less than two. It is preferred that the number of horizontal slices having a gas retention of less than 0.2 is less than 80, more preferably less than 40, even more preferably less than 20, even more preferably less than 12 and most preferably less than 5. . It is preferred that the number of horizontal slices having a gas retention of less than 0.3 is less than 120, more preferably less than 80, even more preferably less than 40, even more preferably less than 20 and most preferably less than 15. .

1 and 2, it has been found that placing the oxidant sparger 34 at the bottom of the reaction zone 28 provides several advantages, including a reduction in the amount of degassed zone in the reaction medium 36. . Given the height “H” of the reaction medium 36, the length “L” of the reaction zone 28 and the maximum diameter “D” of the reaction zone 28, the majority of the oxidant stream (ie, greater than 50 wt%). It is preferred to be introduced into reaction zone 28 within about 0.025H, 0.022L and / or 0.25D of lower end 52 of this reaction zone 28. More preferably, most of the oxidant stream is introduced into reaction zone 28 within about 0.02H, 0.018L, and / or 0.2D of lower end 52 of reaction zone 28. Most preferably, most of the oxidant stream is introduced into reaction zone 28 within 0.015H, 0.013L and / or 0.15D of the lower end 52 of reaction zone 28.

In the embodiment shown in FIG. 2, the vertical distance “Y ₁ ” between the lower end 52 of the reaction zone 28 and the outlet of the upper oxidant opening 68 of the oxidant sparger 34 is about 0.25H, 0.022 Below L and / or 0.25D, substantially all of the oxidant stream enters the reaction zone 28 within about 0.25H, 0.022L and / or 0.25D of the lower end 52 of the reaction zone 28. More preferably, Y ₁ is less than about 0.02H, 0.018L and / or 0.2D. Most preferably, Y ₁ is less than 0.015H, 0.013L and / or 0.15D and greater than 0.005H, 0.004L and / or 0.06D. 2 shows a tangential position where the bottom edge of the cylindrical main body portion 46 of the container shell 22 is connected with the top edge of the elliptical lower head 48 of the container shell 22. Alternatively, the lower head 48 may be of any shape, including conical, and the tangent is still defined as the bottom edge of the cylindrical main body portion 46. The vertical distance “Y ₂ ” between the tangent 72 and the top of the oxidant sparger 34 is preferably at least about 0.0012H, 0.001L and / or 0.01D; More preferably about 0.005H, 0.004L and / or 0.05D or more; Most preferably 0.01H, 0.008L and / or 0.1D or more. The vertical distance “Y ₃ ” between the lower end 52 of the reaction zone 28 and the outlet of the lower oxidant opening 70 of the oxidant sparger 34 is preferably about 0.015H, 0.013L and / or 0.15D under; More preferably less than about 0.012H, 0.01L and / or 0.1D; Most preferably less than 0.01H, 0.008L and / or 0.075D and greater than 0.003H, 0.002L and / or 0.025D.

In a preferred embodiment of the present invention, the opening for releasing the oxidant stream and the feed stream into the reaction zone is configured such that the amount (by weight) of the oxidant or feed stream released from the opening is directly proportional to the open area of the opening. . Thus, for example, if 50% of the cumulative open area defined by all oxidant openings is located within 0.15D of the bottom of the reaction zone, 50% by weight of the oxidant stream enters the reaction zone within 0.15D of the bottom of the reaction zone and , And vice versa.

In addition to the benefits provided by minimizing the degassed zone (ie, the zone with low gas retention) in the reaction medium 36, it has been found that oxidation can be improved by maximizing the gas retention of the entire reaction medium 36. Reaction medium 36 preferably has a time-average and volume-average gas retention of at least about 0.4, more preferably from about 0.6 to about 0.9, most preferably from 0.65 to 0.85. Several physical and operating properties of bubble column reactor 20 contribute to the high gas retention rates discussed above. For example, for a given reactor size and flow of oxidant stream, the high L: D ratio of the reaction zone 28 makes the diameter smaller, which increases the surface velocity in the reaction medium 36, which in turn Increase gas retention. It is also known that the actual diameter and L: D ratio of the bubble column affect the average gas retention at a given constant surface velocity. In addition, minimization of the degassed zone, especially at the bottom of the reaction zone 28, contributes to increased gas retention values. In addition, the overhead pressure and mechanical configuration of the bubble column reactor can affect operational stability at high surface velocities and gas retention values disclosed herein.

In addition, the inventors have found the importance of operating at an optimized overhead pressure to obtain increased gas retention and increased mass transfer. According to Henry's Law effect, operation at lower overhead pressures that reduce the solubility of molecular oxygen appears to reduce the rate of mass transfer of molecular oxygen from gas to liquid. In mechanical shake containers, this is typically the case since the aeration level and mass transfer rate are governed by the shaker design and overhead pressure. However, in the bubble column reactor according to a preferred embodiment of the present invention, lower overhead pressure is used to increase the surface velocity in the reaction medium 36 and again by allowing a certain mass of the gaseous oxidant stream to occupy more volume. Methods have been found to increase gas retention and rate of delivery of molecular oxygen.

The equilibrium between bubble agglomeration and collapse, on the one hand, reduces the internal circulation rate of the liquid phase and may tend to require very large separation zones, and on the other hand lower gas retention and more from the oxidant stream to the liquid phase. Very large bubbles that exhibit low mass transfer rates are very complex phenomena that cause less tendency. With respect to the liquid phase, its composition, density, viscosity and surface tension, among other factors, are known to interact in a very complex manner, producing very complex results even in the absence of a solid phase. For example, laboratory researchers have found it useful to limit whether "water" is tap water, distilled water or deionized water, even for very simple water-air bubble towers, for observations for reporting and evaluation. In the case of complex mixtures and solid phase additions in the liquid phase, the degree of complexity is higher. Among other things, the surface unevenness of individual particles of a solid, the average size of the solids, the particle size distribution, the amount of solids to the liquid phase, and the ability of the liquid to wet the surface of the solid all result in any foaming behavior and natural convective flow pattern. It is important for interaction with the liquid and oxidant streams in determining whether or not.

Therefore, the ability of the bubble column reactor to function usefully while exhibiting high surface velocities and high gas retentions as disclosed herein may include, for example, (1) the composition of the liquid phase of the reaction medium; (2) the amount and type of solids precipitated, both of which can be adjusted by reaction conditions; (3) the amount of oxidant stream fed to the reactor; (4) overhead pressure affecting reaction temperature through volume flow of oxidant stream, bubble stability and energy balance; (5) the reaction temperature itself which affects the fluid properties, the properties of the precipitated solids and the specific volume of the oxidant stream; And (6) appropriate selection of the geometry and mechanical details of the reaction vessel, including the L: D ratio.

Referring again to FIG. 1, by introducing a liquid feed stream into the reaction zone 28 at a plurality of vertically spaced locations, an oxidizable compound (eg, para-xylene) can be improved in the reaction medium 36. It turned out. Preferably, the liquid feed stream is introduced into reaction zone 28 through at least three feed openings, more preferably at least four feed openings. As used herein, the term “feed opening” refers to the opening that discharges the liquid feed stream into the reaction zone 28 for mixing with the reaction medium 36. It is preferred that at least two of the feed openings are spaced perpendicularly to each other at least about 0.5D, more preferably at least about 1.5D, most preferably at least 3D. However, the highest feed opening is about 0.75H, 0.65L and / or 8D or less, more preferably 0.5H, 0.4L and / or 5D or less, most preferably 0.4H, 0.35L and / or from the lowest oxidant opening. Preferably, they are spaced vertically by 4D or less.

Although it is desirable to introduce the liquid feed stream in a plurality of vertical positions, the majority of the liquid feed stream is introduced into the reaction medium 36 if it is introduced into the reaction medium 36 and / or the bottom half of the reaction zone 28. It has been found to distribute the oxidizing compound in an improved manner. Preferably, at least about 75% by weight of the liquid feed stream is introduced into the bottom half of reaction medium 36 and / or reaction zone 28. Most preferably, at least 90% by weight of the liquid feed stream is introduced into the bottom half of reaction medium 36 and / or reaction zone 28. It is also desirable to introduce at least about 30% by weight of the liquid feed stream into the reaction zone 28 within about 1.5D of the lowest vertical position that introduces the oxidant stream into the reaction zone 28. This lowest vertical position that introduces the oxidant stream into the reaction zone 28 is typically the bottom of the oxidant sparger. However, various other configurations for introducing an oxidant stream into the reaction zone 28 are also contemplated by preferred embodiments of the present invention. Preferably, at least about 50% by weight of the liquid feed is introduced within about 2.5D of the lowest vertical position that introduces the oxidant stream into the reaction zone 28. Preferably, at least about 75% by weight of the liquid feed stream is introduced within about 5D of the lowest vertical position that introduces the oxidant stream into the reaction zone 28.

Each feed opening defines an open area for discharging the feed. Preferably, at least about 30% of the cumulative open area of all feed inlets is located within about 1.5D of the lowest vertical position that introduces the oxidant stream into the reaction zone 28. Preferably, at least about 50% of the cumulative open area of all feed inlets is located within about 2.5D of the lowest vertical position that introduces the oxidant stream into the reaction zone 28. Preferably, at least about 75% of the cumulative open area of all feed inlets is located within about 5D of the lowest vertical position that introduces the oxidant stream into the reaction zone 28.

Referring again to FIG. 1, in one embodiment of the invention, the feed inlets 32a, 32b, 32c, 32d are simply a series of vertically aligned openings present along one side of the vessel shell 22. These feed openings preferably have substantially similar diameters of less than about 7 cm, more preferably from about 0.25 to about 5 cm, most preferably from 0.4 to 2 cm. Bubble column reactor 20 is preferably equipped with a system for regulating the flow rate of the liquid feed stream from each feed opening. This flow control system preferably includes separate flow control valves 74a, 74b, 74c, 74d for each individual feed inlet 32a, 32b, 32c, 32d. In addition, at least a portion of the liquid feed stream has a high inlet surface velocity of at least about 2 m / sec, more preferably at least about 5 m / sec, even more preferably at least about 6 m / sec, most preferably 8 to 20 m / sec. It is preferable to install a flow control system in the bubble column reactor 20 which can be introduced into the furnace reaction zone 28. As used herein, the term "inlet surface velocity" refers to the time-averaged volume flow rate of the feed stream exiting the feed opening divided by the area of the feed opening. Preferably, at least about 50% by weight of the feed stream is introduced into reaction zone 28 at a high inlet surface velocity. Most preferably, substantially all of the feed streams are introduced into reaction zone 28 at high inlet surface velocities.

Referring now to FIGS. 6 and 7, another system for introducing a liquid feed stream into reaction zone 28 is shown. In this embodiment, the feed stream is introduced into reaction zone 28 at four different heights. At each height, individual feed distribution systems 76a, 76b, 76c, 76d are installed. Each feed distribution system 76 includes a main feed conduit 78 and a manifold 80. Each manifold 80 has two or more outlets 82, 84 connected with individual insertion conduits 86, 88 extending into the reaction zone 28 of the vessel shell 22. Each insertion conduit 86, 88 provides a separate feed opening 87, 89 to discharge the feed stream into the reaction zone 28. Feed openings 87 and 89 preferably have substantially similar diameters of less than about 7 cm, more preferably between about 0.25 and about 5 cm, most preferably between 0.4 and 2 cm. Positioning the feed openings 87, 89 of each feed distribution system 76a, 76b, 76c, 76d opposite in diameter to introduce feed stream into the reaction zone 28 in opposite directions. desirable. It is also desirable to orient the feed openings 86, 88 positioned opposite to each other relative to the diameter of the adjacent feed distribution system 76 by rotating them 90 ° with respect to each other. In operation, the liquid feed stream exits main feed conduit 78 and then enters manifold 80. Manifold 80 evenly distributes the feed stream for simultaneous introduction on opposite sides of reactor 20 through feed openings 87 and 89.

FIG. 8 shows another configuration where bayonet tubes 90 and 92 are installed in each feed distribution system 76 rather than insertion conduits 86 and 88 (shown in FIG. 7). Bayonet tubes 90, 92 protrude into reaction zone 28 and include a plurality of small feed openings 94, 96 for discharging the liquid feed into reaction zone 28. It is preferred that the small feed openings 94, 96 of the bayonet tubes 90, 92 have substantially the same diameter of less than about 50 mm, more preferably about 2 to about 25 mm, most preferably 4 to 15 mm. .

9-11 illustrate another feed distribution system 100. Feed distribution system 100 introduces a liquid feed stream in a plurality of vertically spaced and laterally spaced locations without requiring multiple through holes in the sidewall of bubble column reactor 20. Feed introduction system 100 generally includes a single inlet conduit 102, a header 104, a plurality of granular distribution conduits 106, a lateral support mechanism 108, and a vertical support mechanism 110. . Inlet conduit 102 penetrates the sidewall of main body portion 46 of vessel shell 22. Inlet conduit 102 is variably connected to header 104. Header 104 evenly distributes the feed stream received from inlet conduit 102 to each of the granular distribution conduits 106. Each distribution tube 106 has a plurality of vertically spaced feed openings 112a, 112b, 112c, 112d for discharging the feed stream into the reaction zone 28. Lateral support mechanism 108 is connected to each dispensing tube 106 and inhibits relative lateral movement of the dispensing tube 106. The vertical support mechanism 110 is preferably connected to the lateral support mechanism 108 and also to the top of the oxidant sparger 34. Vertical support mechanism 110 substantially suppresses vertical movement of distribution tube 106 in reaction zone 28. It is desirable for the feed opening 112 to have substantially the same diameter of less than about 50 mm, more preferably about 2 to about 25 mm, most preferably 4 to 15 mm. The vertical spacing of the feed openings 112 of the feed distribution system 100 shown in FIGS. 9-11 may be substantially the same as described above in connection with the feed distribution system of FIG. 1.

It has been found that the flow pattern of the reaction medium in many bubble column reactors can allow for non-uniform azimuth distribution of the oxidizing compound in the reaction medium, especially when the oxidizing compound is introduced mainly along one side of the reaction medium. As used herein, the term “azimuth angle” refers to the angle or spacing around the granular elongation axis of the reaction zone. As used herein, "granular" means within 45 ° of vertical. In one embodiment of the invention, a feed stream containing an oxidizing compound (eg para-xylene) is introduced into the reaction zone through feed openings spaced at a plurality of azimuths. Feed openings spaced apart in these azimuths can help prevent excessively high oxidative compound concentrations and too low oxidative compound concentrations in the reaction medium. The various feed introduction systems shown in FIGS. 6-11 are examples of systems that provide adequate azimuth spacing of feed openings.

Referring again to FIG. 7, in order to quantify the spaced apart introduction of the liquid feed stream into the reaction medium into the reaction medium, the reaction medium is approximately equal volume of granular azimuth quadrant “Q ₁ , Q ₂ , Q ₃ , Q ₄ ”. Theoretically divided by These azimuthal quadrants “Q ₁ , Q ₂ , Q ₃ , Q ₄ ” are defined by a pair of imaginary intersecting orthogonal vertical planes “P ₁ , P ₂ ” extending beyond the maximum vertical dimension and the maximum radial dimension of the reaction medium. . When the reaction medium is contained in a cylindrical vessel, the intersections of the virtual intersecting vertical planes P ₁ , P ₂ approximately coincide with the vertical center lines of the cylinders, and each azimuth quadrant "Q ₁ , Q ₂ , Q ₃ , Q ₄ " It is generally a wedge-shaped vertical volume having the same height as that of the reaction medium. It is desirable to release a significant amount of the oxidizing compound into the reaction medium through feed openings located in at least two different azimuthal quadrants.

In a preferred embodiment of the invention, up to about 80% by weight of the oxidizable compound is released into the reaction medium through a feed opening that can be located in one of the azimuthal quadrants. More preferably, up to about 60% by weight of the oxidizing compound is released into the reaction medium through a feed opening that can be located in one of the azimuthal quadrants. Most preferably, up to 40% by weight of the oxidizing compound is released into the reaction medium through a feed opening that can be located in one of the azimuthal quadrants. When the azimuth quadrant is azimuth-oriented such that the maximum possible amount of oxidizing compound is released into one of the azimuth quadrants, these variables of the azimuth distribution of the oxidative compound are measured. For example, in order to determine the azimuth distribution in the four azimuth quadrants, when the entire feed stream is discharged into the reaction medium through two openings spaced azimuthally at 89 ° from each other, the azimuth quadrant has two feed openings. 100% by weight of the feed stream is discharged into the reaction medium in one of the azimuth quadrants since they can be azimuth-oriented in such a way that they are all located in one of the azimuth quadrants.

In addition to the advantages associated with the proper azimuth spacing of the feed openings, it has been found that the proper radial spacing of the feed openings in the bubble column reactor may also be important. It is desirable to release a significant amount of the oxidative compound introduced into the reaction medium through a feed opening spaced radially inwardly from the vessel sidewall. Therefore, in one embodiment of the present invention, a significant amount of the oxidative compound enters the reaction zone through a feed opening located in the "preferred radial feed zone" spaced inward from the granular sidewalls defining the reaction zone.

Referring again to FIG. 7, the preferred radial feed zone “FZ” centers on the reaction zone 28 and measures the outer diameter “D _o ” of 0.9D, where “D” is the diameter of the reaction zone 28. Can take the shape of a theoretical granular cylinder. Therefore, the outer ring "OA" with a thickness of 0.05D is defined as being between the preferred radial feed zone (FZ) and the sidewall interior defining the reaction zone (28). It is preferred that little or no oxidizing compound is introduced into the reaction zone 28 through the feed opening located in this outer ring (OA).

In other embodiments, it is preferred that little or no oxidizing compound is introduced into the center of the reaction zone 28. Therefore, as shown in FIG. 8, the preferred radial feed zone FZ has the shape of a theoretical granular ring having an outer diameter D _O of 0.9D and an inner diameter D _I of 0.2D centered on the reaction zone 28. Can be taken. Thus, in this embodiment, the inner cylinder IC with a diameter of 0.2D is the "erased part" of the center of the preferred radial feed zone FZ. It is preferred that little or no oxidizing compound is introduced into the reaction zone 28 through the feed opening located in this inner cylinder (IC).

In a preferred embodiment of the present invention, regardless of whether the preferred radial feed zone is cylindrical or cyclic as described above, a significant amount of oxidizing compound is introduced into the reaction medium 36 through a feed opening located in the preferred radial feed zone. do. More preferably, at least about 25% by weight of the oxidizing compound is released into the reaction medium 36 through a feed opening located in the desired radial feed zone. Even more preferably, at least about 50% by weight of the oxidizing compound is released into the reaction medium 36 through a feed opening located in the desired radial feed zone. Most preferably, at least 75% by weight of the oxidizing compound is released into the reaction medium 36 through a feed opening located in the desired radial feed zone.

Although the theoretical azimuth quadrant and theoretical preferred radial feed zones shown in FIGS. 7 and 8 have been described in relation to the distribution of the liquid feed stream, proper azimuth distribution and radial distribution of the gaseous oxidant stream may also provide certain advantages. It was found. Therefore, in one embodiment of the invention, the description of the azimuth and radial distributions of the liquid feed stream described above also applies to the manner of introducing the gaseous oxidant stream into the reaction medium 36.

Referring now to FIGS. 12-15, another oxidant sparger 200 is shown generally comprising a ring member 202 and a pair of oxidant inlet conduits 204, 206. The oxidant sparger 200 of FIGS. 12-15 is similar to the oxidant sparger 34 of FIGS. 1-11 but has three major differences: Not including; (2) the top of the ring member 202 has no opening for releasing the oxidant upwardly; (3) The oxidant sparger 200 has many more openings at the bottom of the ring member 202.

As best shown perhaps in FIGS. 14 and 15, the bottom of the oxidant sparger ring 202 provides a plurality of oxidant openings 208. The oxidant opening 208 is positioned such that at least about 1% of the total open area defined by the oxidant opening 208 is below the centerline 210 (FIG. 15) of the ring member 202 (the center line 210 is a ring member). Preferably located at the height of the volume center of 202). More preferably, at least about 5% of the total open area defined by all oxidant openings 208 is located below the centerline 210, and at least about 2% of the total open area moves the oxidant stream within about 30 ° of vertical. It is usually defined by openings 208 that emit downward. Even more desirably, at least about 20% of the total open area defined by all oxidant openings 208 is located below the centerline 210, and at least about 10% of the total open area is typically within 30 ° of the oxidant stream. It is defined by the opening 208 which releases downward. Most preferably, at least about 75% of the total open area defined by all oxidant openings 208 is located below the centerline 210, and at least about 40% of the total open area is typically within 30 ° of the oxidant stream. It is defined by the opening 208 which releases downward. The fraction of the total open area defined by all oxidant openings 208 located above the centerline 210 is preferably less than about 75%, more preferably less than about 50%, even more preferably less than about 25%. Most preferably less than 5%.

As shown in FIGS. 14 and 15, the oxidant opening 208 includes a downward opening 208a and an oblique opening 208b. The downward opening 208a is configured to typically release the oxidant stream downwards within about 30 ° vertically, more preferably within about 15 ° vertically and most preferably within 5 ° vertically. The inclined opening 208b ejects the oxidant stream typically outwardly and downwardly at an angle “A” that is about 15 to about 75 ° from vertical, more preferably angle A is about 30 to about 60 ° from vertical, most Preferably the angle A is 40 to 50 ° from vertical.

It is preferred that substantially all oxidant openings 208 have approximately the same diameter. The diameter of the oxidant opening 208 is preferably about 2 to about 300 mm, more preferably about 4 to about 120 mm, most preferably 8 to 60 mm. The total number of oxidant openings 208 in the ring member 202 is selected to meet the low pressure drop criteria described in detail below. Preferably, the total number of oxidant openings 208 generated in the ring member 202 is about 10 or more, more preferably the total number of oxidant openings 208 is about 20 to about 200, most preferably The total number of oxidant openings 208 is 40 to 100.

12-15 illustrate very specific configurations of the oxidant sparger 200, it is noted that various oxidant sparger configurations may be used to achieve the advantages described herein. For example, the oxidant sparger does not necessarily have to have the octagonal ring member structure shown in FIGS. 12 and 13. Rather, the oxidant sparger may be constructed of any structure of the flow conduit (s) using a plurality of spaced openings to discharge the oxidant stream. The size, number and direction of release of the oxidant opening in the flow conduit is preferably within the aforementioned range. In addition, the oxidant sparger is preferably configured to provide an azimuth and radial distribution of the molecular oxygen described above.

Regardless of the specific configuration of the oxidant sparger, it is desirable to configure and operate the oxidant sparger in a manner that minimizes the pressure drop associated with the release of the oxidant stream from the flow conduit (s) through the oxidant opening into the reaction zone. From the time-averaged static pressure of the oxidant stream inside the flow conduit at the oxidant openings 66a, 66b of the oxidant sparger, the height at which half of the oxidant stream is introduced above a certain vertical position and half of the oxidant stream is introduced below the vertical position. This pressure drop is calculated by subtracting the time-averaged static pressure of the reaction zone at. In a preferred embodiment of the invention, the time-average pressure drop associated with the release of the oxidant stream from the oxidant sparger is less than about 0.3 MPa, more preferably less than about 0.2 MPa, even more preferably less than about 0.1 MPa, most Preferably it is less than 0.05 Mpa. Under the preferred operating conditions of the bubble column reactor described herein, the pressure of the oxidant stream within the flow conduit (s) of the oxidant sparger is preferably from about 0.35 to about 1 MPa, more preferably from about 0.45 to about 0.85 MPa, most preferably Is 0.5 to 0.7 MPa.

As mentioned above in connection with the oxidant sparger configuration shown in FIGS. 2-5, the oxidant sparger is flushed continuously or periodically with a liquid (eg, acetic acid, water and / or para-xylene). It may be desirable to prevent the oxidant sparger from contaminating with solids. With such a liquid flush, it is desirable to pass a liquid effective amount (ie, not a trace of liquid droplets naturally present in the oxidant stream) at least once per day for more than one minute through the oxidant sparger out of the oxidant opening. Do. When the liquid is continuously or periodically released from the oxidant sparger, the time-average ratio of the mass flow rate of the liquid through the oxidant sparger to the mass flow rate of molecular oxygen through the oxidant sparger is about 0.05: 1 to about 30: 1, Or about 0.1: 1 to about 2: 1, or 0.2: 1 to 1: 1.

In one embodiment of the present invention, a significant amount of oxidative compound (eg para-xylene) can be introduced into the reaction zone through an oxidant sparger. In this configuration, it is preferable to release the oxidizing compound and the molecular oxygen from the oxidant sparger through the same opening of the oxidant sparger. As indicated above, the oxidizing compound is typically a liquid in STP. Thus, in this embodiment, the biphasic stream can be discharged from the oxidant sparger, where the liquid phase comprises an oxidizing compound and the gas phase comprises molecular oxygen. However, it should be appreciated that at least a portion of the oxidizable compound may be gaseous when released from the oxidant sparger. In one embodiment, the liquid phase released from the oxidant sparger predominantly consists of oxidizing compounds. In other embodiments, the liquid phase exiting the oxidant sparger has a composition substantially the same as the feed stream described above. If the liquid phase exiting the oxidant sparger has a composition that is substantially the same as the feed stream, this liquid phase may comprise a solvent and / or catalyst system in the amounts and ratios described above in relation to the composition of the feed stream.

In one embodiment of the invention, it is preferred that at least about 10% by weight of all oxidative compounds introduced into the reaction zone are introduced into the reaction zone through the oxidant sparger, more preferably at least about 40% by weight of the oxidizing compounds The oxidant sparger is introduced into the reaction zone and most preferably at least 80% by weight of the oxidizing compound is introduced into the reaction zone through the oxidant sparger. If all or part of the oxidizing compound is introduced into the reaction zone through the oxidant sparger, preferably at least about 10% by weight of all the molecular oxygen introduced into the reaction zone is introduced through the same oxidant sparger, more preferably At least about 40% by weight of molecular oxygen is introduced into the reaction zone through the same oxidant sparger, and most preferably at least 80% by weight of molecular oxygen is introduced into the reaction zone through the same oxidant sparger. If a significant amount of oxidizing compound is introduced into the reaction zone through the oxidant sparger, it is desirable to place one or more temperature sensing devices (eg, thermocouples) in the oxidant sparger. These temperature sensors can be used to help ensure that the temperature in the oxidant sparger is not dangerously high.

Referring now to FIGS. 16-18, bubble column reactor 20 is shown including an internal degassing apparatus 300 disposed at the bottom of reaction zone 28 near slurry outlet 38. It has been found that impurity-generating side reactions occur at a relatively high rate during the degassing of the reaction medium 36. As used herein, “degassing” refers to separating the gas phase from the multiphase reaction medium. If the reaction medium 36 is highly aerated (> 0.3 gas retention), impurity generation is minimal. Even when the reaction medium 36 is highly degassed (<0.01 gas retention), impurity generation is minimal. However, if the reaction medium is partially aerated (gas retention of 0.01 to 0.3), undesirable side reactions are promoted and more impurities are produced. The degassing vessel 300 solves this and other problems by minimizing the volume of reaction medium 36 that is partially degassed and minimizing the time it takes to degas the reaction medium 36. Substantially degassed slurry is produced from the bottom of the degassing vessel 300 and exits the reactor 20 through the slurry outlet 38. The substantially degassed slurry preferably contains less than about 5% by volume, more preferably less than about 2% by volume, and most preferably less than 1% by volume of gas phase.

In FIG. 16, bubble column reactor 20 is shown to include a level regulator 302 and a flow control valve 304. The level regulator 302 and the flow control valve 304 cooperate to keep the reaction medium 36 at a substantially constant height in the reaction zone 28. The level regulator 302 senses the height of the upper surface 44 of the reaction medium 36 (eg, by differential pressure level sensing or by nucleation level sensing) and at the height of the reaction medium 36. In response to generating an adjustment signal 306. Flow control valve 304 receives a control signal 306 to adjust the flow rate of the slurry through the slurry outlet conduit 308. Therefore, the flow rate of the slurry exiting the slurry outlet 38 is the maximum slurry volume flow rate F _max when the height of the reaction medium 36 is too high and the minimum slurry volume flow rate when the height of the reaction medium 36 is too low. Can change between (F _min ).

In order to remove the solid phase oxidation product from the reaction zone 28, some must first be passed through the degassing vessel 300. The degassing vessel 300 is a low turbulent flow that can naturally raise the gas phase of the reaction medium 36 from the liquid and solid phases of the reaction medium 36 as liquids and solids flow down toward the slurry outlet 38. -turbulence provides the internal volume. The gas phase rises from the liquid phase and the solid phase by the natural upward buoyancy of the gas phase in the liquid phase and the solid phase. When using the degassing vessel 300, the process of transferring the reaction medium 36 from the fully-aerated three-phase medium to the fully-degassed two-phase slurry is fast and efficient.

Referring now to FIGS. 17 and 18, the degassing vessel 300 generally includes a granular sidewall 308 that defines a degassing zone 312 therebetween. Preferably, sidewall 308 extends upwards within about 30 ° of vertical, more preferably within about 10 ° of vertical. Most preferably, sidewall 308 is substantially vertical. Degassing zone 312 is separated from reaction zone 28 and has a height “h” and a diameter “d”. The upper end 310 of the sidewall 308 is open to receive the reaction medium from the reaction zone 28 into the interior volume 312. The lower end of the sidewall 308 is variably connected to the slurry outlet 38 through the transition zone 314. In certain cases, such as when the opening of the slurry outlet 38 is large or the diameter "d" of the sidewall 308 is small, the transition zone 314 may be eliminated. As best shown in FIG. 18, the degassing vessel 300 may also include a vortex breaker 316 disposed in the degassing zone 312. Vortex prevention device 316 can be any structure that can be operated to suppress the generation of vortices as the solid and liquid phases flow down toward slurry outlet 38.

In order to properly separate the gaseous phase from the solid and liquid phases in the degassing vessel 300, the height “h” and the horizontal cross-sectional area of the inner degassing zone 312 are carefully selected. The height “h” and the horizontal cross-sectional area of the internal degassing zone 312 are substantially before the bubbles reach the bottom outlet of the degassing vessel 300 even when the maximum amount of slurry is collected (ie, even when the slurry is collected by F _mzx ). Sufficient distance and time should be provided to allow all bubble volumes to rise from the solid and liquid phases. Therefore, the cross-sectional area of the degassing zone 312 is such that the maximum downward velocity (V _dmax ) of the liquid and solid phase through the degassing zone 312 is substantially greater than the natural ascent rate of vapor phase bubble V _u through the liquid and solid phase. It is preferable to make the cross section to be low. The maximum downward velocity (V _dmax ) of the liquid and solid phase through the degassing zone 312 occurs at the maximum slurry volumetric flow rate F _max discussed above. The natural rate of rise (V _u ) of the bubbles through the liquid and solid phases depends on the size of the bubbles; Since the whole of the original bubble volume present in the reaction medium (36) is substantially larger than 0.5cm, the natural rise velocity (V _{u0 .5)} of the bubble diameter of 0.5cm with the liquid phase and the solid phase the cut-off (cut-off) Can be used as a value. Preferably, the cross-sectional area V _dmax V _u0 is such that less than about 75% cross-section and, even more preferably V _dmax of _0.5 to a stripping zone (312) is less than about 40% of V _{u0 0.5,} and most preferably is V _dmax is less than about 20% of V _{u0 .5.}

The downward velocity of the liquid and solid phase in the degassing zone 312 of the degassing vessel 300 is defined as the volume flow rate of the degassed slurry through the slurry outlet 38 divided by the minimum cross-sectional area of the degassing zone 312. The downward velocity of the liquid and solid phase in the degassing zone 312 of the degassing vessel 300 is preferably less than about 50 cm / sec, more preferably less than about 30 cm / sec, most preferably less than 10 cm / sec.

Now, although the granular sidewall 308 of the degassing vessel 300 is shown to have a cylindrical structure, as long as the wall defines an internal volume having an appropriate volume, cross-sectional area, width "d" and height "h". Note that sidewall 308 may include a plurality of sidewalls that form various structures (eg, triangles, squares, or ellipses). In a preferred embodiment of the invention, "d" is from about 0.2 to about 2 m, more preferably from about 0.3 to about 1.5 m, most preferably from 0.4 to 1.2 m. In a preferred embodiment of the present invention, "h" is about 0.3 to about 5 m, more preferably about 0.5 to about 3 m, most preferably 0.75 to 2 m.

In a preferred embodiment of the invention, the sidewall 308 is substantially vertical such that the horizontal cross-sectional area of the degassing zone 312 is substantially constant along the entire height "h" of the degassing zone 312. Preferably, the maximum horizontal cross-sectional area of the degassing zone 312 is less than about 25% of the maximum horizontal cross-sectional area of the reaction zone 28. More preferably, the maximum horizontal cross-sectional area of the degassing zone 312 is about 0.1 to about 10% of the maximum horizontal cross-sectional area of the reaction zone 28. Most preferably, the maximum horizontal cross-sectional area of the degassing zone 312 is 0.25-4% of the maximum horizontal cross-sectional area of the reaction zone 28. Preferably, the maximum horizontal cross-sectional area of the degassing zone 312 is from about 0.02 to about 3 m 2, more preferably from about 0.05 to about 2 m 2 and most preferably from 0.1 to 1.2 m 2. The volume of the degassing zone 312 is preferably less than about 5% of the total volume of the reaction medium 36 or the reaction zone 28. More preferably, the volume of the degassing zone 312 is about 0.01 to about 2% of the total volume of the reaction medium 36 or the reaction zone 28. Most preferably, the volume of the degassing zone 312 is 0.05 to about 1% of the total volume of the reaction medium 36 or the reaction zone 28. The volume of the degassing zone 312 is preferably less than about 2 m 3, more preferably about 0.01 to about 1 m 3, most preferably 0.05 to 0.5 m 3.

Referring now to FIG. 19, bubble column reactor 20 is shown as including an external degassing vessel 400. In this configuration, the aerated reaction medium 36 is collected from the reaction zone 28 through the high opening on the side of the vessel shell 22. The collected aerated medium is transported to the degassing vessel 400 through the outlet conduit 402 to separate the gas phase from the solid and liquid phases. The separated gas phase exits the degassing vessel 400 through the conduit 404, while the substantially degassed slurry exits the degassing vessel 400 through the conduit 406.

In FIG. 19, the outlet conduit 402 is shown to be approximately straight and horizontal and perpendicular to the vessel shell 22. It is just one convenient configuration; The outlet conduit 402 may be different in any respect, provided that it advantageously connects the bubble column reactor 20 to an external degassing vessel 400. Returning to conduit 404, it is useful to connect this conduit at or near top degassing vessel 400 to control safety issues associated with stagnant gas pockets containing oxidizing compounds and oxidants. In addition, conduits 402 and 404 may advantageously include flow disconnect means such as valves.

When the reaction medium 36 is collected from the reactor 20 via a high outlet as shown in FIG. 19, the bottom outlet 408 in the bubble column reactor 20 near the bottom 52 of the reaction zone 28. Is preferably installed. The bottom outlet 408 and the bottom conduit 410 connected to it can be used to eliminate (ie, empty) the reactor 20 during shutdown. Preferably, at least one lower outlet 408 is at the bottom third of the height of the reaction medium 36, more preferably at the bottom quarter of the reaction medium 36, most preferably in the reaction zone 28 Provide at the lowest point of).

With respect to the high slurry collection and degassing system shown in FIG. 19, the lower conduit 410 and outlet 408 are not used to collect the slurry from the reaction zone 28 during oxidation. It is known in the art that solids tend to settle by specific gravity in non-aerated and otherwise agitated portions of the slurry, including stagnant flow conduits. In addition, the precipitated solids (eg terephthalic acid) may tend to solidify into large aggregates by continuous precipitation and / or crystalline reorganization. Therefore, to prevent clogging of the lower flow conduit 410, a portion of the slurry degassed from the bottom of the degassing vessel 400 is used to continuously or intermittently flush the lower conduit 410 during normal reactor 20 operation. You can. A preferred means of providing such slurry flush to conduit 410 is to periodically open valve 412 in conduit 410 and pass a portion of the degassed slurry through conduit 410 and also through the lower opening 408 to the reaction zone. (28) to flow into. Even when the valve 412 is fully or partially open, only a portion of the degassed slurry flows back through the lower conduit 410 into the reaction zone 28. The remaining portion of the degassed slurry that is not used to flush the bottom conduit 410 is transported away from the reactor 20 through the conduit 414 for subsequent processing (eg, purification).

During normal operation of bubble column reactor 20 over a significant time (eg,> 100 hours), the total degassing resulting from the bottom of degassing vessel 400 is the amount of degassed slurry used to flush bottom conduit 410. It is preferred that less than 50%, more preferably less than about 20%, most preferably less than 5% by weight of the prepared slurry. In addition, the mean mass flow rate of the degassed slurry used to flush the bottom conduit 410 over a significant time is less than about 4 times the mean mass flow rate of the oxidizing compound into the reaction zone 28, more preferably the reaction zone ( 28) less than about 2 times the average mass flow rate of the oxidizing compound into, more preferably less than the average mass flow rate of the oxidizing compound into the reaction zone 28, and most preferably the average of the oxidizing compound into the reaction zone 28 It is preferably less than 0.5 times the mass flow rate.

Referring again to FIG. 19, the degassing vessel 400 includes a substantially granular, preferably cylindrical sidewall 416 that defines the degassing zone 418. Degassing zone 418 has a diameter "d" and a height "h". The height "h" is measured as the vertical distance between the bottom of the sidewall 416 and the location where the aerated reaction medium enters the degassing vessel 400. The height “h”, diameter “d”, area and volume of the degassing zone 418 are preferably substantially as described above with respect to the degassing zone 312 of the degassing vessel 300 shown in FIGS. 16-18. same. Degassing vessel 400 also includes an upper region 420 created by extending sidewall 416 over the degassing zone 418. The upper zone 420 of the degassing vessel 400 can be any height, but it preferably extends upwards to or above the level of the reaction medium 36 in the reaction zone 28. The upper zone 420 ensures a space to allow the gas phase to properly separate from the liquid and solid phases before exiting the degassing vessel 400 through the conduit 404. Although conduit 404 is now shown returning the separated gas phase to the separation zone of reactor 20, conduit 404 may otherwise be connected to vessel shell 22 at any height above outlet conduit 402. It may be. Optionally, conduit 404 is connected to gas outlet conduit 40 such that the separated gaseous phase from degassing vessel 400 merges with the overhead vapor stream removed from conduit 40 and is sent down for further processing. can do.

Referring now to FIG. 20, bubble column reactor 20 is shown to include a hybrid inner-outer degassing vessel 500. In this configuration, a portion of the reaction medium 36 is collected from the reaction zone 28 through a relatively large high opening 502 in the side wall of the vessel shell 22. The collected reaction medium 36 is transported through a relatively large L-shaped tube 504 and enters the top of the degassing vessel 500. In FIG. 20, the L-shaped conduit 504 is shown perpendicularly connected to the sidewall of the vessel shell 22 and includes a natural transition at an angle of about 90 °. It is just one convenient configuration; The L-shaped conduit 504 may vary in any respect, provided that it usefully connects the external degassing vessel 500 and bubble column reactor 20 as described. In addition, the L-shaped conduit 504 may advantageously include flow disconnect means such as a valve.

In the degassing vessel 500, the gas phase moves upward while the solid and liquid phases move downward. The upwardly moving gas enters the L-shaped conduit 504 again and exits through the opening 502 back to the reaction zone 28. Thus, in the opening 502, opposite flows of incoming reaction medium 36 and outgoing separated gas may occur. The degassed slurry exits the degassing vessel 500 through the conduit 506. The degassing vessel 500 includes a substantially granular, preferably cylindrical sidewall 508 that defines the degassing zone 510. Degassing zone 510 has a height "h" and a diameter "d". It is preferred that the high opening 502 and the L-shaped conduit 504 have a diameter equal to or greater than the diameter “d” of the degassing zone 510. The height “h”, diameter “d”, area and volume of the degassing zone 510 are preferably substantially as described above with respect to the degassing zone 312 of the degassing vessel 300 shown in FIGS. 16-18. same.

19 and 20 illustrate embodiments of bubble column reactor 20 in which solid product (eg, crude terephthalic acid) produced in reaction zone 28 is collected from reaction zone 28 via a high outlet. Collecting the aerated reaction medium 36 at a position higher than the bottom of the bubble column reactor 20 avoids accumulation and stagnation of the poorly aerated reaction medium 36 at the bottom 52 of the reaction zone 28. It can help According to another aspect of the present invention, the concentration of oxygen and oxidizing compounds (eg para-xylene) in the reaction medium 36 near the top of the reaction medium 36 is preferably lower than near the bottom. Therefore, collecting the reaction medium 36 at a high location can increase yield by lowering the amount of unreacted reactant collected from the reactor 20. In addition, when bubble column reactor 20 is operated with a high STR and a gradient of the chemical composition disclosed herein, the temperature of reaction medium 36 varies significantly in the vertical direction. In this situation, the temperature of the reaction medium 36 typically has a local minimum near the lower and upper ends of the reaction zone 28. Near the lower end, the minimum is related to the evaporation of the solvent near the point where all or part of the oxidant is received. Near the upper end, the minimum is again due to the evaporation of the solvent, but in this case the evaporation of the solvent is due to the pressure drop in the reaction medium. In addition, other local minimums may occur between the upper and lower ends wherever additional feed or oxidant is received into the reaction medium. Thus, there is one or more temperature maximums caused by the exotherm of the oxidation reaction between the lower and upper ends of the reaction zone 28. The collection of reaction medium 36 at a higher location at a higher temperature can be particularly advantageous when subsequent processing takes place at higher temperatures, since the energy costs involved in heating the collected medium for subsequent processing are reduced.

Therefore, in a preferred embodiment of the present invention, especially when subsequent feeding is at a higher temperature, the reaction medium 36 is characterized in that at least 50% by weight of the liquid feed stream and / or gaseous oxidant stream enters the reaction zone 28. Collected from bubble column reactor 20 through high outlet (s) located higher than location (s). More preferably, reaction medium 36 is bubble column reactor through a high outlet (s) located substantially above the position (s) at which all liquid feed streams and / or gaseous oxidant streams enter reaction zone 28. It is collected from 20. Preferably, at least 50% by weight of the solid and liquid components collected from the bubble column reactor 20 are collected through the high outlet (s). More preferably, substantially all solid and liquid components collected from bubble column reactor 20 are collected through high outlet (s). Preferably, the high outlet (s) are located at least about 1D higher than the lower end 52 of the reaction zone 28. More preferably, the high outlet (s) are located at least about 2D higher than the lower end 52 of the reaction zone 28. Most preferably, the high outlet (s) are located at least 3D higher than the lower end 52 of the reaction zone 28. Given the height "H" of the reaction medium 36, the high outlet (s) is between about 0.2H and about 0.8H, more preferably between about 0.3H and about 0.7H, most preferably 0.4H and 0.6 It is preferably located vertically between H. In addition, the temperature of the reaction medium 36 at the high outlet from the reaction zone 28 is preferably at least 1 ° C. higher than the temperature of the reaction medium 36 at the lower end 52 of the reaction zone 28. . More preferably, the temperature of the reaction medium 36 at the high outlet of the reaction zone 28 is about 1.5 to about 16 ° C. more than the temperature of the reaction medium 36 at the lower end 52 of the reaction zone 28. high. Most preferably, the temperature of the reaction medium 36 at the high outlet of the reaction zone 28 is 2-12 ° C. higher than the temperature of the reaction medium 36 at the lower end 52 of the reaction zone 28.

Referring now to FIG. 21, bubble column reactor 20 is shown to include another hybrid degassing vessel 600 located at the bottom of reactor 20. In this configuration, the degassed reaction medium 36 is collected from the reaction zone 28 through a relatively large opening 602 at the lower end 52 of the vessel shell 22. The opening 602 defines the open upper end of the degassing vessel 600. In the degassing vessel 600, the gas phase moves upward while the solid and liquid phases move downward. The upwardly moving gaseous phase may reenter the reaction zone 28 through the opening 602. Therefore, in the opening 602, opposite flows of incoming reaction medium 36 and outgoing separated gas can occur. The degassed slurry exits the degassing vessel 600 through conduit 604. The degassing vessel 600 includes a substantially granular, preferably cylindrical sidewall 606 that defines the degassing zone 608. Degassing zone 608 has a height "h" and a diameter "d". Preferably, the opening 602 has a diameter equal to or greater than the diameter “d” of the degassing zone 608. The height “h”, diameter “d”, area and volume of the degassing zone 608 are preferably substantially as described above with respect to the degassing zone 312 of the degassing vessel 300 shown in FIGS. 16-18. same.

Referring now to FIG. 22, the bubble column reactor 20 of FIG. 21 is shown to include another oxidant sparger 620. The oxidant sparger 620 includes a ring member 622 and a pair of inlet conduits 624 and 626. The ring member 622 preferably has substantially the same configuration as the ring member 202 described above with respect to FIGS. 12-15. Inlet conduits 624 and 626 extend upward through openings in the lower head 48 of vessel shell 22 and provide an oxidant stream to ring member 622.

Referring now to FIG. 23, the bubble column reactor 20 of FIG. 21 is shown to include spargerless means for introducing an oxidant stream into the reaction zone 28. In the configuration of FIG. 23, an oxidant stream is provided to reactor 20 via oxidant conduits 630 and 632. Oxidant conduits 630 and 632 are connected to individual oxidant openings 634 and 636 at the lower head 48 of the vessel shell 22. An oxidant stream is introduced directly into reaction zone 28 through oxidant openings 634 and 636. Optional impingement plates 638 and 640 may be provided to deflect the flow of the oxidant stream after it has entered the reaction zone 28.

As mentioned above, it is preferred that the oxidation reactor be constructed and operated in such a way as to avoid zones with high concentrations of oxidant compounds in the reaction medium (because such zones can produce impurities). One way to improve the initial dispersion of oxidizing compounds (eg para-xylene) in the reaction medium is to dilute the oxidizing compounds with a liquid. The liquid used to dilute the oxidizing compound may be derived from a portion of the reaction medium that is located far away from the position (s) where the oxidizing compound is supplied to the reaction zone. This liquid from the distant portion of the reaction medium may be circulated through a flow conduit disposed inside and / or outside the main reaction vessel to a position close to the inlet position of the oxidizing compound.

24 and 25 illustrate two preferred methods for circulating a liquid from a distant portion of the reaction medium to a location close to the inlet of the oxidizing compound using an inner conduit (FIG. 24) or an outer conduit (FIG. 25). Preferably, the length of the flow conduit from the inlet (ie, the opening (s) into which the liquid enters the conduit) to the outlet (ie, the opening (s) from which the liquid is discharged from the conduit) is longer than about 1 m, more preferably Longer than about 3 m, even more preferably longer than about 6 m, most preferably longer than 9 m. However, the actual length of the conduit becomes less relevant when the liquid is obtained from a separate vessel located directly above or next to the vessel from which the oxidative compound feed is initially released. Liquid from any separate vessel containing at least a portion of the reaction medium is a preferred source for the initial dilution of the oxidizing compound.

Whatever the source, it is desirable for the liquid flowing through the conduit to have an unchanging concentration of oxidizing compound lower than the reaction medium immediately adjacent one or more outlets of the conduit. In addition, it is preferred that the liquid flowing through the conduit has a concentration of oxidizing compound in the liquid phase of less than about 100,000 ppmw, more preferably less than about 10,000 ppmw, even more preferably less than about 1,000 ppmw, most preferably less than 100 ppmw. The concentration is then measured before the addition of the oxidative compound feed and any separate solvent feed to the conduit. If the increase in oxidative compound feed and the optional solvent feed are measured after addition, the combined liquid stream entering the reaction medium is less than about 300,000 ppm, more preferably less than about 50,000 ppm and most preferably less than 10,000 ppm It is preferred to have a concentration of heavy oxidizing compound.

It is desirable to maintain the flow through the conduit at a sufficiently low rate so that the circulating liquid does not suppress the desired overall gradient of oxidizing compound in the reaction medium. In this regard, the ratio of the mass flow rate of the liquid flowing through the conduit to the mass flow rate of the liquid flowing through the conduit in the reaction zone in which the increase in oxidizing compound is first released is greater than about 0.3 minutes, more preferably greater than about 1 minute, even more Preferably from about 2 minutes to about 120 minutes, most preferably from 3 minutes to 60 minutes.

There are a number of means for forcing a liquid to flow through the conduit. Preferred means include specific gravity, all types of extractors using gas or liquid or both as motive fluids, and all types of mechanical pumps. When using an extractor, one embodiment of the invention is a feed (liquid or gas) of an oxidizing compound, a feed (gas) of an oxidant, a feed (liquid) of a solvent and a pumped source (slurry) of the reaction medium. At least one fluid selected from the group consisting of is used as the motif fluid. In another embodiment, two or more fluids selected from the group consisting of a feed of an oxidizable compound, a feed of an oxidant and a feed of a solvent are used as the motif fluid. Another embodiment uses a combination of a feed of an oxidizable compound, a feed of an oxidant, and a feed of a solvent as the motif fluid.

Appropriate diameters or diameters of the circulation conduit may vary depending on the amount and nature of the material being transported, and the energy and capital costs that can be used to force flow movement. It is preferred that the minimum diameter of these conduits is greater than about 0.02 m, more preferably about 0.06 to about 2 m, most preferably 0.12 to 0.8 m.

As indicated above, it is desirable to regulate the flow through the conduit to a specific desired range. There are a number of means known in the art that influence this control by setting up appropriate fixed geometries during fabrication of the flow conduit. Other preferred embodiments include geometries that can vary during operation, including valves of all kinds and descriptions, including manual operation and motorized operation by any means with or without feedback control loops from the sensing elements. Is to use. Another preferred means of controlling the flow of dilution liquid is to change the energy input between the inlet and outlet of the conduit. Preferred means include changing the flow rate of one or more motif fluids to the extractor, changing the energy input to the pump driver, and changing the height difference when using a force by density difference or specific gravity. These preferred means can be used in combination of all together.

The conduits used for the circulation of the liquid from the reaction medium can be of any type known in the art. One embodiment utilizes conduits made fully or partially using conventional piping materials. Other embodiments utilize conduits made fully or partially using reaction vessel walls as part of the conduits. The conduits can be made completely within the boundaries of the reaction vessel (FIG. 24), or the conduits can be made completely outside the reaction vessel (FIG. 25), or the conduits include a zone within the reaction vessel and a zone outside the reaction vessel. You may.

The inventors contemplate that it may be desirable to have several conduits and conduits of various designs in order to move the liquid through the conduits, especially in large reactors. It may also be desirable to provide multiple outlets at multiple locations on one or both of the conduits. Certain aspects of the design balance a desired overall gradient of unchanging concentration of oxidizing compound with the desired initial dilution of the oxidizing compound feed according to another aspect of the present invention.

24 and 25 show a design using a degassing vessel both connected to a conduit. This degassing vessel ensures that the portion of the reaction medium used to dilute the oxidizing compound entering is a substantially degassed slurry. However, it is noted that the liquid or slurry used to dilute the oxidizing compound entering may be in aerated and degassed form.

Dilution of the oxidative compound feed with liquid flowing through the conduit is particularly useful in bubble column reactors. In addition, in a bubble column reactor, the superior advantage of initial dilution of the oxidant compound feed can be achieved without adding the oxidant compound feed directly into the conduit, although the outlet of the conduit is located close enough to the location of the addition of the oxidizing compound. In this embodiment, the outlet of the conduit is most within about 27 conduit outlet diameters, more preferably within about 9 conduit outlet diameters, even more preferably within about 3 conduit outlet diameters of the nearest addition location of the oxidizing compound. It is preferably located within one conduit outlet diameter.

It has been found that a flow extractor may be useful for initial dilution of an oxidizing compound feed in an oxidized bubble column reactor according to one embodiment of the present invention without using conduits to obtain dilution liquid from a distant portion of the reaction medium. In this case, the extractor is located in the reaction medium and has an open path from the reaction medium to the throat of the extractor, where low pressure is caused in the adjacent reaction medium. Examples of two possible extractor configurations are shown in FIGS. 26 and 27. In a preferred embodiment of these extractors, the closest location for supplying the oxidizing compound is within about 4 m, more preferably within about 1 m and most preferably within 0.3 m of the neck of the extractor. In other embodiments, the oxidative compound is fed under pressure as a motif fluid. In another embodiment, the solvent or oxidant is fed under pressure as an additional motif fluid with the oxidizing compound. In another embodiment, the solvent and oxidant are fed under pressure as additional motive fluid together with the oxidizing compound.

The inventors believe that it may be desirable to have a plurality of extractors of various designs located at various locations in the reaction medium, especially in larger reactors. The details of the design balance the desired initial dilution of the oxidizable compound feed with the overall preferred gradient of the unchanging concentration of the oxidizable compound according to another aspect of the present invention. In addition, the inventors believe that the outflow flow plume from the extractor may be oriented in any direction. When using a plurality of extractors, each extractor can be oriented separately in any direction.

As mentioned above, the specific physical and operating characteristics of bubble column reactor 20 described above with reference to FIGS. 1 to 27 are characterized by the pressure, temperature and reactants (ie, oxygen and oxidizing compounds) of reaction medium 36. Provide a vertical gradient. As discussed above, these vertical gradients can provide a more efficient and economical oxidation method compared to conventional oxidation processes that favor a well-mixed reaction medium of relatively uniform pressure, temperature and reactant concentration throughout. . The vertical gradients of oxygen, oxidative compounds (eg, para-xylene) and temperatures made possible by using an oxidation system according to an embodiment of the invention are now discussed in more detail.

Referring now to FIG. 28, in order to quantify the gradient of reactant concentration present in reaction medium 36 during oxidation in bubble column reactor 20, the total volume of reaction medium 36 is equal to 30 separate horizontal separations of the same volume. Can theoretically be divided into 28 illustrates the concept of dividing the reaction medium 36 into 30 separate horizontal layers of equal volume. Except for the highest and lowest horizontal slices, each horizontal slice is an individual volume bounded at the top and bottom by an imaginary horizontal plane and having side boundaries by the walls of the reactor 20. The highest horizontal layer is bottom bounded by an imaginary horizontal plane and top bounded by an upper surface of the reaction medium 36. The lowest horizontal layer is bounded at the top by the virtual horizontal plane and at the bottom by the bottom of the vessel shell. After theoretically dividing the reaction medium 36 into 30 separate horizontal slices of equal volume, the time-averaged and volume-averaged concentrations of each horizontal slice can be determined. An individual horizontal slice that has a maximum concentration of all 30 horizontal slices can be referred to as a "C-maximum horizontal slice". An individual horizontal slice that has a minimum concentration of all horizontal slices positioned above the C-maximal horizontal slice and positioned above the C-maximal horizontal slice can be referred to as a "C-minimal horizontal slice". The vertical concentration gradient can then be calculated as the ratio of the concentration in the C-maximal horizontal separation to the concentration in the C-minimal horizontal separation.

Regarding the quantification of the oxygen concentration gradient, when the reaction medium 36 is theoretically divided into 30 separate horizontal slices of the same volume, the O ₂ -maximum horizontal slices all have a maximum oxygen concentration of 30 horizontal slices. It is confirmed that the O ₂ -minimal horizontal slice has a minimum oxygen concentration of the horizontal slice located higher than the O ₂ -maximum horizontal slice. The oxygen concentration of the horizontal partition is measured in the gas phase of the reaction medium 36 on a time-averaged and volume-averaged molar (wet) basis. O ₂ - Maximum horizontal oxygen concentration vs. O ₂ of the partitioned About 2 ratio of the oxygen concentration in the minimum horizontal partitioned: 1 to about 25: 1, more preferably from about 3: 1 to about 15: 1, most preferably from 4 It is preferable that it is 1: 1-10: 1.

Typically, the O ₂ -maximum horizontal separation is located near the bottom of the reaction medium 36, while the O ₂ -minimum horizontal separation is located near the top of the reaction medium 36. Preferably, the O ₂ -minimal horizontal slice is one of five highest horizontal slices of 30 individual horizontal slices. Most preferably, the O ₂ -minimal horizontal slice is the highest slice of 30 individual horizontal slices as shown in FIG. 28. Preferably, the O ₂ -maximum horizontal slice is one of the ten lowest horizontal slices of the 30 individual horizontal slices. Most preferably, the O ₂ -maximum horizontal slice is one of the five lowest horizontal slices of 30 separate horizontal slices. For example, FIG. 28 shows the O ₂ -maximum horizontal separation as the third horizontal separation from the bottom of the reactor 20. O ₂ - and at least O ₂ - is the maximum horizontal partitioned vertical gap of about 2W or more between, more preferably at least about 4W, and most preferably preferably at least 6W. O ₂ - and at least O ₂ - is the maximum perpendicular distance between the horizontal partitioned about 0.2H, more preferably at least about 0.4H or greater and most preferably is not less than 0.6H.

The time-average and volume-average oxygen concentrations (wet basis) of the O ₂ -minimal horizontal slice are from about 0.1 to about 3 mol%, more preferably from about 0.3 to about 2 mol%, most preferably from 0.5 to 1.5 mol% to be. The time-average and volume-average oxygen concentrations of the O ₂ -maximum horizontal partitions are preferably about 4 to about 20 mole percent, more preferably about 5 to about 15 mole percent, most preferably 6 to 12 mole percent. . The time-averaged oxygen concentration (dry basis) in the gaseous effluent discharged from the reactor 20 through the gas outlet 40 is preferably from about 0.5 to about 9 mol%, more preferably from about 1 to about 7 mol%. And most preferably 1.5 to 5 mol%.

Since the oxygen concentration decreases significantly toward the top of the reaction medium 36, it is desirable to reduce the oxygen demand at the top of the reaction medium 36. Thus reduced near the top of the reaction medium 36 by creating a vertical gradient with respect to the concentration of the oxidizing compound (eg para-xylene) (the minimum concentration of the oxidizing compound is located near the top of the reaction medium 36). Oxygen demand can be achieved.

With regard to quantifying the oxidative compound (eg para-xylene) concentration gradient, when the reaction medium 36 is theoretically divided into 30 separate horizontal separations of the same volume, the OC-maximum horizontal separation is 30 It is found that it has a maximum oxidizing compound concentration of both horizontal slices, and the OC-minimum horizontal slice is found to have a minimum oxidizing compound concentration of horizontal slice located above the OC-maximum horizontal slice. The oxidizing compound concentration in the horizontal partition is measured in the liquid phase on a time-averaged and volume-averaged mass fraction. The ratio of the oxidizing compound concentration of the OC-maximum horizontal layer to the oxidizing compound concentration of the OC-minimal horizontal layer is greater than about 5: 1, more preferably greater than about 10: 1, even more preferably greater than about 20: 1, Most preferably, it is 40: 1-1000: 1.

Typically, the OC-maximum horizontal separation is located near the bottom (bottom) of the reaction medium 36, while the OC-minimum horizontal separation is located near the top of the reaction medium 36. Preferably, the OC-min horizontal slice is one of the 5 highest horizontal slices of 30 separate horizontal slices. Most preferably, the OC-minimum horizontal slice is the highest slice of 30 separate horizontal slices as shown in FIG. 28. Preferably, the OC-max horizontal slice is one of the 10 highest horizontal slices of 30 separate horizontal slices. Most preferably, the OC-maximum horizontal slice is one of the five lowest horizontal slices of 30 separate horizontal slices. For example, FIG. 28 shows the OC-maximum horizontal separation as the fifth horizontal separation from the bottom (bottom) of the reactor 20. Preferably, the vertical spacing between the OC-minimum horizontal separation and the OC-maximum horizontal separation is at least about 2W, where "W" is the maximum width of the reaction medium 36. More preferably, the vertical spacing between the OC-min horizontal slice and the OC-max horizontal slice is at least about 4W, most preferably at least 6W. Given the height "H" of the reaction medium 36, the vertical spacing between the OC-minimum horizontal separation and the OC-maximum horizontal separation is at least about 0.2H, more preferably at least about 0.4H, most preferably 0.6H. It is preferable that it is above.

The time-average and volume-average oxidizing compound (eg, para-xylene) concentrations in the liquid phase of the OC-minimal horizontal layer are preferably less than about 5,000 ppmw, more preferably less than about 2,000 ppmw, even more preferably about 400 ppmw Below, most preferably 1 to 100 ppmw. The time-averaged and volume-averaged oxidizing compound concentrations in the liquid phase of the OC-maximum horizontal partition are preferably about 100 to about 10,000 ppmw, more preferably about 200 to about 5,000 ppmw, most preferably 500 to 3,000 ppmw.

Although it is desirable for the bubble column reactor 20 to provide a vertical multiple of the oxidizing compound concentration, it is also desirable to minimize the volume percent of the reaction medium 36 having a concentration of the oxidizing compound in the liquid phase above 1,000 ppmw. Preferably, the time-average volume percentage of reaction medium 36 having a concentration of oxidizing compound in the liquid phase higher than 1,000 ppmw is less than about 9%, more preferably less than about 6%, most preferably less than 3%. Preferably, the time-average volume percentage of reaction medium 36 having a concentration of oxidizing compound in the liquid phase higher than 2,500 ppmw is less than about 1.5%, more preferably less than about 1%, most preferably less than 0.5%. Preferably, the time-average volume percentage of reaction medium 36 having an oxidizing compound concentration in the liquid phase above 10,000 ppmw is less than about 0.3%, more preferably less than about 0.1%, most preferably less than 0.03%. Preferably, the time-average volume percentage of reaction medium 36 having a concentration of oxidizing compound in the liquid phase higher than 25,000 ppmw is less than about 0.03%, more preferably less than about 0.015%, most preferably less than 0.007%. The inventors have found that the volume of reaction medium 36 having a high level of oxidizing compound need not belong to one contiguous volume. At various times, the chaotic flow pattern in the bubble column reaction vessel simultaneously produces two or more consecutive but separate portions of reaction medium 36 having high levels of oxidative compounds. At each time used for time averaging, all of these continuous but separate volumes greater than 0.0001% by volume of the total reaction medium are added together to determine the total volume having a high level of oxidizing compound concentration in the liquid phase.

In addition to the concentration gradients of the oxygen and oxidizing compounds discussed above, it is desirable that a temperature gradient be present in the reaction medium 36. Referring again to FIG. 28, this temperature gradient is similar to the concentration gradient by theoretically dividing the reaction medium 36 into 30 separate horizontal layers of the same volume and measuring the time-average and volume-average temperatures of each layer. Can be quantified. The horizontal slice with the lowest temperature among the lowest 15 horizontal slices may be referred to as the T-minimum horizontal slice, and the horizontal slice with the maximum temperature of all slices above the T- minimum horizontal slice is the "T- Maximum horizontal separation ". It is preferred that the temperature of the T-maximal horizontal slice is at least about 1 ° C. higher than the temperature of the T-minimum horizontal slice. More preferably, the temperature of the T-maximal horizontal slice is about 1.25 to about 12 ° C. higher than the temperature of the T-minimum horizontal slice. Most preferably, the temperature of the T-maximal horizontal slice is 2-8 ° C. higher than the temperature of the T-minimum horizontal slice. The temperature of the T-maximal horizontal separation is preferably about 125 to about 200 ° C, more preferably about 140 to about 180 ° C, most preferably 150 to 170 ° C.

Typically, the T-maximal horizontal separation is located near the center of the reaction medium 36, while the T-minimum horizontal separation is located near the bottom of the reaction medium 36. Preferably, the T-min horizontal slice is one of the 10 lowest horizontal slices of the 15 lowest horizontal slices. Most preferably, the T-min horizontal slice is one of the five lowest horizontal slices of the 15 lowest horizontal slices. For example, FIG. 28 shows the T-minimum horizontal separation as the second horizontal separation from the bottom of the reactor 20. Preferably, the T-maximal horizontal slice is one of 20 intermediate horizontal slices of 30 separate horizontal slices. Most preferably, the T-maximal horizontal slice is one of 14 intermediate horizontal slices of 30 separate horizontal slices. For example, FIG. 28 shows the T-maximal horizontal separation as the 12th horizontal separation from the bottom of the reactor 20 (ie, one of the middle ten horizontal separations). It is preferred that the vertical spacing between the T-minimum horizontal slice and the T-maximal horizontal slice is at least about 2W, more preferably at least about 4W, most preferably at least 6W. It is preferred that the vertical spacing between the T-minimum horizontal slice and the T-maximal horizontal slice is at least about 0.2H, more preferably at least about 0.4H, most preferably at least 0.6H.

As discussed above, where there is a vertical temperature gradient in the reaction medium 36, the reaction medium 36 is at a higher position where the temperature of the reaction medium is at its highest, particularly when further processing of the collected product at a higher temperature. It may be desirable to collect. Therefore, when the reaction medium 36 is collected from the reaction zone 28 through one or more high outlets as shown in FIGS. 19 and 20, the high outlet (s) are located near the T-maximal horizontal separation. It is preferable. Preferably, the high outlet is in ten horizontal slices of the T-maximal horizontal slice, more preferably in five horizontal slices of the T-maximal horizontal slice, and most preferably in two horizontal slices of the T-maximal horizontal slice. Located.

It is now noted that many of the features of the invention described herein can be used in multiple oxidation reactor systems, rather than systems using only a single oxidation reactor. In addition, certain features of the invention described herein may be utilized in machine- shake and / or flow-shake oxidation reactors rather than bubble-shake reactors (ie bubble column reactors). For example, the inventors have found certain advantages associated with stepping / changing the oxygen concentration and / or the rate of oxygen consumption throughout the reaction medium. The advantages achieved by the stepping of the oxygen concentration / consumption in the reaction medium can be achieved whether the entire volume of the reaction medium is contained in a single vessel or in multiple vessels. In addition, the advantages achieved by the stepping of the oxygen concentration / consumption of the reaction medium can be achieved regardless of whether the reaction vessel (s) are machine-shake, flow-shake and / or bubble-shake.

One way to quantify the oxygen concentration of the reaction medium and / or the degree of stage of consumption is to compare two or more separate 20% continuous volumes of the reaction medium. These 20% continuous volumes need not be limited to any particular shape. However, each 20% continuous volume must form a volume within the same limits of the reaction medium (ie each volume is "continuous"), and the 20% continuous volumes must not overlap one another (ie, the volumes must be "stars" Game "). 29 to 31 show that these separate 20% continuous volumes can be located in the same reactor (FIG. 29) or multiple reactors (FIGS. 30 and 31). Note that the reactor shown in FIGS. 29-31 can be a machine-shaking, flow-shaking and / or bubble-shaking reactor. In one embodiment, it is preferred that the reactor shown in FIGS. 29-31 is a bubble-shaking reactor (ie bubble column reactor).

Referring now to FIG. 29, reactor 20 is shown to contain reaction medium 36. The reaction medium 36 comprises a first separate 20% continuous volume 37 and a second separate 20% continuous volume 39.

Referring now to FIG. 30, a multiple reactor system is shown comprising a first reactor 720a and a second reactor 720b. Reactors 720a and 720b cooperate to contain the total volume of reaction medium 736. The first reactor 720a contains the first reaction medium portion 736a, while the second reactor 720b contains the second reaction medium portion 736b. The first separate 20% continuous volume 737 of the reaction medium 736 is shown as defined within the first reactor 720a while the second separate 20% continuous volume of the reaction medium 736 is defined. 739 is shown as being defined within the second reactor 720b.

Referring now to FIG. 31, a multiple reactor system is shown to include a first reactor 820a, a second reactor 820b, and a third reactor 820c. Reactors 820a, 820b, and 820c cooperatively contain the entire volume of reaction medium 836. The first reactor 820a contains a first reaction medium portion 836a; Second reactor 820b contains a second reaction medium portion 836b; The third reactor 820c contains a third reaction medium portion 836c. The first distinct 20% continuous volume 837 of the reaction medium 836 is shown as defined within the first reactor 820a; A second separate 20% continuous volume 839 of the reaction medium 836 is shown as defined within the second reactor 820b; A third separate 20% continuous volume 841 of the reaction medium 836 is shown defined in the third reactor 820c.

Quantifying the stepping of oxygen utilization efficiency in the reaction medium by mentioning 20% continuous volume of the reaction medium richest in the molar fraction of oxygen in the gas phase and 20% continuous volume of the reaction medium lacking the molar fraction of oxygen in the gas phase. can do. In a separate 20% continuous volume of gas phase of the reaction medium containing the highest concentration of oxygen in the gas phase, the time-averaged and volume-average oxygen concentration (wet basis) is preferably from about 3 to about 18 mole percent, more preferably Is about 3.5 to about 14 mol%, most preferably 4 to 10 mol%. In a separate 20% continuous volume of gas phase in the reaction medium containing the lowest concentration of oxygen in the gas phase, the time-average and volume-average oxygen concentration (wet basis) is preferably from about 0.3 to about 5 mole percent, more preferably Is about 0.6 to about 4 mol%, most preferably 0.9 to 3 mol%. In addition, the ratio of the time-average and volume-average oxygen concentration (wet basis) of the richest 20% continuous volume of the reaction medium, as compared to the 20% continuous volume deficient in the reaction medium, is preferably about 1.5: 1. To about 20: 1, more preferably about 2: 1 to about 12: 1, most preferably 3: 1 to 9: 1.

The staged rate of oxygen consumption in the reaction medium can be quantified with the oxygen-STR described above. Oxygen-STR has been described above in a broad sense (ie, in terms of average oxygen-STR of the entire reaction medium), but oxygen-STR is also used locally in order to quantify the rate of oxygen consumption rate throughout the reaction medium. (Ie, part of the reaction medium) may be contemplated.

The inventors have found that it is very useful to vary the oxygen-STR throughout the reaction medium in general harmony with the preferred gradients disclosed herein with respect to the pressure of the reaction medium and the mole fraction of molecular oxygen in the gas phase of the reaction medium. Thus, the ratio of the first separate 20% continuous volume of oxygen-STR of the reaction medium to the second separate 20% continuous volume of oxygen-STR is about 1.5: 1 to about 20: 1, More preferably about 2: 1 to about 12: 1, most preferably 3: 1 to 9: 1. In one embodiment, the "first separate 20% continuous volume" is located closer than the "second separate 20% continuous volume" to the position where molecular oxygen is first introduced into the reaction medium. This large gradient of oxygen-STR may be applied to any other type of reaction vessel in which the partial oxidation reaction medium is contained in the bubble column oxidation reactor or in which a gradient is produced at the pressure of the gas phase of the reaction medium and / or the mole fraction of molecular oxygen. For example, in a mechanical shake vessel having a plurality of vertically placed stirring zones achieved by using a plurality of impellers with strong radial flow, which may be reinforced by a generally horizontal baffle assembly (here, significant ash of the oxidant flow Although mixing can take place within each vertically placed stirring zone and some remixing of the oxidant flow can take place between adjacent vertically placed stirring zones, the oxidant flow rises substantially up from the feed near the bottom of the reaction vessel. It is preferable regardless of whether it contains. In other words, when there is a gradient in the gas phase pressure and / or the mole fraction of molecular oxygen in the reaction medium, the inventors have found that it is desirable to produce a similar gradient in the chemical demand for dissolved oxygen by the means disclosed herein. It was.

A preferred means of causing the local oxygen-STR to change is to adjust the concentration gradient of the oxidizing compound in accordance with another disclosure of the present invention by adjusting the feed site of the oxidizing compound and controlling the mixing of the liquid phase of the reaction medium. Another useful means of causing the local oxygen-STR to change is by causing a local temperature change and by changing the local mixing of the catalyst and solvent components (e.g., introducing additional gas to evaporate in certain portions of the reaction medium). By causing cooling by adding a solvent stream containing a large amount of water to reduce the activity in a particular portion of the reaction medium).

As discussed above in connection with FIGS. 30 and 31, at least a portion, preferably at least 25%, more preferably at least 50%, most preferably at least 75% of the molecular oxygen coming from the first reaction vessel is additional Multiple reaction vessels delivered to one or more subsequent reaction vessels in order to consume an increment, preferably at least 10%, more preferably at least 20%, most preferably at least 40% of the molecular oxygen leaving the first / upstream reaction vessel. The partial oxidation reaction can be usefully carried out. When using this series flow of molecular oxygen from one reactor to another reactor, the first reaction vessel is of higher reaction strength than at least one of the subsequent reaction vessels, preferably from about 1.5: 1 to about 20: 1, more Preferably vessel-average-oxygen-STYR in the first reaction vessel from about 2: 1 to about 12: 1, most preferably 3: 1 to 9: 1 versus vessel-average- in the subsequent reaction vessel. Preference is given to operating at the ratio of oxygen-STR.

As discussed above, it may or may not be different from the first reaction vessel of any type (eg, bubble column, machine-shaking, remix, internal multistage, plug flow, etc.) and the first reaction vessel. All subsequent types of subsequent reaction vessels are useful for the in-line flow of molecular oxygen into subsequent reaction vessels according to the present invention. Means for reducing the vessel-average-oxygen-STR in subsequent reaction vessels may advantageously include reduced temperatures, reduced concentrations of oxidizing compounds, and reduced reaction activity (eg, reduced cobalt concentrations) of certain mixtures of catalyst components and solvents. , Increased water concentration and the addition of catalytic retardants such as small amounts of copper ions).

Upon flowing from the first reaction vessel to the subsequent reaction vessel, the oxidant stream is treated by any means known in the art, such as compression or pressure reduction, cooling or heating, and removal or addition of any amount or any type of material. can do. However, using a reduced vessel-average-oxygen-STR in the subsequent reaction vessel, the absolute pressure at the top of the first reaction vessel is less than about 2.0 MPa, more preferably less than about 1.6 MPa, most preferably 1.2 MPa. It is particularly useful when less than. In addition, using a reduced vessel-average-oxygen-STR in the subsequent reaction vessel, the ratio of the absolute pressure of the upper portion of the first reaction vessel to the absolute pressure of the upper portion of the at least one subsequent reaction vessel is from about 0.5: 1 to It is particularly useful when it is 6: 1, more preferably about 0.6: 1 to about 4: 1, most preferably 0.7: 1 to 2: 1. Reducing the pressure below these lower limits in subsequent vessels excessively reduces the utilization efficiency of molecular oxygen, and increasing the pressure above these upper limits further increases the cost compared to a fresh supply of oxidant.

When using a serial flow of molecular oxygen to a subsequent reaction vessel with a reduced vessel-mean-oxygen-STR, a fresh feed stream of oxidizing compound, solvent and oxidant may be flowed into the subsequent reaction vessel and / or the first reaction vessel. Can be. When present, the liquid and solid flow of the reaction medium can flow in either direction between the reaction vessels. All or part of the gaseous phase exiting the first reaction vessel and entering the subsequent reaction vessel, if present, may flow separately or in combination with a portion of the liquid or solid phase of the reaction medium from the first reaction vessel. If present, the flow of the product stream, including the liquid and solid phases, can be withdrawn from the reaction medium in any reaction vessel of the system.

Referring again to FIGS. 1 to 29, in one embodiment of the present invention, the oxidized bubble tower reactor 20 produces significantly higher production than conventional oxidized bubble tower reactors, especially conventional bubble column reactors used to produce terephthalic acid. Have speed. In order to provide increased production rates, the size of the bubble column reactor 20 must be increased. However, the natural convective multiphase fluid flow kinetics in these expanded bubble column reactors is considerably different from the flow kinetics in smaller conventional reactors. For the proper functioning of a high production rate expanded bubble column reactor, certain design and operating parameters were found to be important.

When bubble column 20 is a high production rate expanded oxidized bubble column reactor according to one embodiment of the invention, the height "H" of reaction medium 36 is preferably at least about 30 m, more preferably about 35 to about 70 m, most preferably 40 to 60 m. The density and height of the reaction medium 36 in the expanded bubble column reactor 20 cause a significant pressure difference between the top of the reaction medium 36 and the bottom of the reaction medium 36. Preferably, the pressure difference between the top and bottom of the reaction medium 36 is at least about 1 bar, more preferably at least about 1.4 bar, most preferably 1.6 to 3 bar. The maximum width “W” of the reaction medium 36 is preferably at least about 2.5 m, more preferably from about 3 to about 20 m, even more preferably from about 3.25 to about 12 m, most preferably from 4 to 10 m. The H: W ratio of the reaction medium 36 is preferably at least about 6: 1, more preferably from about 8: 1 to about 20: 1, most preferably 9: 1 to 15: 1. The total volume of reaction medium 36 and / or reaction zone 28 is preferably at least about 250 m 3, more preferably at least about 500 m 3, most preferably at least 1,000 m 3.

During operation of the expanded bubble column reactor 20, the time-averaged surface velocity of the gas phase of the reaction medium 36 at 1/4 height, 1/2 height and / or 3/4 height is from about 0.8 to about 5 m / s. Seconds, more preferably about 0.9 to about 3 m / sec, most preferably 1 to 2 m / sec. When expanded oxidized bubble column reactor 20 is used to produce terephthalic acid via partial oxidation of para-xylene, para-xylene is at least about 11,000 kg / hour, more preferably from about 20,000 to about 100,000 kg / Preference is given to feeding reaction zone 28 at a rate of time, most preferably 30,000 to 80,000 kg / hour. The terephthalic acid production rate of the expanded bubble column 20 is preferably at least about 400 tons / day, more preferably at least about 700 tons / day, most preferably 750 to 3,000 tons / day. Other design and operating parameters of the expanded oxidized bubble column reactor 20 may be substantially the same as described above for the first time with reference to FIGS.

It has been found that varying the horizontal cross-sectional area of the reaction zone of the bubble column reactor along the height of the reactor can help to improve the fluid flow kinetics of the reaction medium, particularly in the expanded oxidized bubble column reactor design discussed above. Referring now to FIG. 32, in one embodiment of the present invention, the vessel shell 22 of the oxidized bubble column reactor 20 includes a wide lower zone 23, a narrow upper zone 25, and a transition zone 27. do. Preferably, the lower and upper zones 23, 25 are substantially cylindrical in shape and aligned along a common central granular axis. The transition zone 27 may have a number of suitable shapes (eg, horizontal planar shape, 2: 1 ellipse shape, hemispherical shape, etc.). Preferably, the transition zone 27 is a roughly truncated conical member that transfers the container shell 22 from the wide bottom section 23 to the narrow top section 25. The lower section 23 of the vessel shell 22 defines a wide lower reaction zone 29. The upper section 25 of the vessel shell 22 defines a narrow upper reaction zone 31. The transition zone 27 defines a transition zone located between the lower reaction zone 29 and the upper reaction zone 31. The lower reaction zone 29, the upper reaction zone 31 and the transition zone cooperate to form the entire reaction zone of the bubble column reactor 20 containing the multiphase reaction medium 36.

In a preferred embodiment of the invention, the maximum horizontal cross-sectional area of the reaction medium 36 in the lower reaction zone 29 is at least about 10% greater than the minimum cross-sectional area of the reaction medium 36 in the upper reaction zone 31. More preferably, the maximum horizontal cross-sectional area of reaction medium 36 in lower reaction zone 29 is about 25 to about 210% greater than the minimum horizontal cross-sectional area of reaction medium 36 in upper reaction zone 31. Most preferably, the maximum horizontal cross-sectional area of reaction medium 36 in lower reaction zone 29 is 35 to 160% greater than the minimum horizontal cross-sectional area of reaction medium 36 in upper reaction zone 31.

As shown in FIG. 32, the lower reaction zone 29 has a maximum diameter “D ₁ ” greater than the minimum diameter “D _u ” of the upper reaction zone 31. Preferably, D _I is at least about 5% greater than D _u . More preferably, D ₁ is about 10 to about 100% greater than D _u . Most preferably, D ₁ is 15-50% greater than D _u . 32 also shows that the lower reaction zone 29 has a maximum height "L _l ", the upper reaction zone 31 has a maximum height "L _u ", and the transition zone has a maximum height "L _t ". do. Note that the height of the portion of the reaction medium 36 contained in the lower reaction zone 29 is L _l and the height of the portion of the reaction medium 36 contained in the transition zone is L _t . In one embodiment, the height of the portion of the reaction medium 36 located in the upper reaction zone 31 is L _u . In certain cases, however, the height of reaction medium 36 in upper reaction zone 31 may be less than L _u . In other cases, the overall height of the reaction medium 36 may extend higher than the upper end 50 of the upper reaction zone 31 (ie, the total height of the reaction medium 36 is the sum of L ₁ + L _t + L _u) . Greater than). Preferably, the overall height of the reaction medium 36 is within 50%, 25% or 10% of L _u measured up or down from the upper end 50 of the upper reaction zone 31. Preferably, the oxidized bubble column reactor 20 has an L _l of about 0.05: 1 to about 5: 1, more preferably about 0.1: 1 to about 2.5: 1, most preferably 0.15: 1 to 1.5: 1. _{Has the} ratio: L _u . Preferably, the oxidized bubble column reactor 20 has an L _l : D _l ratio of at least about 0.5: 1, more preferably from about 1: 1 to about 10: 1, most preferably from 1.5: 1 to 8: 1. Has Oxidized bubble column reactor 20 preferably has a L _u : D _u ratio of at least about 2: 1, more preferably from about 2.5: 1 to about 20: 1, most preferably from 3: 1 to 15: 1. Have In a preferred embodiment of the present invention, L ₁ is at least about 2 m, more preferably L ₁ is about 4 to about 50 m, most preferably 6 to 40 m. Preferably, L _u is at least about 4 m, more preferably from about 5 to about 80 m, most preferably from 10 to 60 m. Preferably, D ₁ is about 2 to about 12 m, more preferably about 3.1 to about 10 m, most preferably 4 to 9 m.

32 also shows that the bubble column reactor 20 has a separation zone 26 located above the upper reaction zone 31. Separation zone 26 defines separation zone 30. As shown in FIG. 32, the separation zone 30 has a maximum height "Y" and a maximum width "X". It is preferred that the oxidized bubble column reactor 20 has an X: D _l ratio of about 0.8: 1 to about 4: 1, most preferably 1.1: 1 to 2: 1. Oxidized bubble column reactor 20 preferably has a L _u : Y ratio of about 1: 1 to about 10: 1, most preferably 2: 1 to 5: 1. Oxidized bubble column reactor 20 preferably has a L ₁ : Y ratio of about 0.4: 1 to about 8: 1, most preferably 0.6: 1 to 4: 1. The transition zone 27 of the vessel shell 22 has a maximum diameter “D ₁ ” adjacent the lower zone 23 and a minimum diameter “D _u ” adjacent the upper zone 25. Oxidized bubble column reactor 20 preferably has an L _t : D _l ratio of about 0.05: 1 to about 5: 1, most preferably 0.1: 1 to 2: 1.

The vertically-variable horizontal cross-sectional area of the oxidizing bubble column reactor 20 shown in FIG. 32 is the upper reaction zone having higher surface gas velocity and higher gas retention than the portion of reaction medium 36 contained in the lower reaction zone 29. A portion of reaction medium 36 contained in 31 is provided. Preferably, the time-averaged surface gas velocity of the portion of the reaction medium 36 contained in the upper reaction zone 31 at half the height of the reaction medium in the upper reaction zone 31 is the reaction in the lower reaction zone 29. It is at least about 15% greater than the time-averaged surface gas velocity of the portion of reaction medium 36 contained in lower reaction zone 29 at half the height of the medium. More preferably, the portion of the reaction medium 36 contained in the upper reaction zone 31 is the reaction medium 36 contained in the lower reaction zone 29 at half the height of the reaction medium in the lower reaction zone 29. ) Has a time-averaged surface gas velocity at half height of the reaction medium in the upper reaction zone 31, which is about 25 to about 210% greater than the time-averaged surface gas velocity of the. Most preferably, the time-averaged surface gas velocity of the portion of the reaction medium 36 contained in the upper reaction zone 31 at half the height of the reaction medium in the upper reaction zone 31 is lower in the lower reaction zone 29. 35-160% greater than the time-averaged surface gas velocity of the portion of reaction medium 36 contained in lower reaction zone 29 at half the height of the reaction medium. Preferably, the time-average and volume-average gas retention of the portion of reaction medium 36 contained in the upper reaction zone 31 is determined by the time-average and portion of the reaction medium 36 portion contained in the lower reaction zone 29. At least about 4% greater than the volume-average gas retention. More preferably, the time-average and volume-average gas retention rates of the portion of reaction medium 36 contained in the upper reaction zone 31 are the time-average of the portion of reaction medium 36 contained in the lower reaction zone 29. And from about 6 to about 80% greater than the volume-average gas retention. Most preferably, the time-average and volume-average gas retention of the portion of the reaction medium 36 contained in the upper reaction zone 31 is the time-average of the portion of the reaction medium 36 contained in the lower reaction zone 29. And 8-70% greater than volume-average gas retention.

While FIG. 32 illustrates a very specific two stage cylindrical sidewall bubble column design, it should be appreciated that many other designs may fall within the scope of this embodiment of the present invention. For example, one or more inclined sidewalls and / or a plurality of stepped-diameter sidewall zones may form a narrow top and wide bottom zone of the reactor. In either case, the claims, not the drawings, obtain the integers of this embodiment.

As mentioned above, in some cases it may be desirable to use larger bubble column reactors to provide higher production rates for a single reactor. However, as the bubble column reactor increases in size, the fluid flow behavior of the multiphase reaction medium contained therein differs significantly from the fluid flow behavior of smaller reactors. It has been found that the changed fluid flow behavior of this larger bubble column reactor can be reduced by contacting the multiphase reaction medium contained in the bubble column reactor with additional granular surfaces. Thus, FIGS. 33-44 illustrate various ways of providing additional granular surface area to the reaction zone of an expanded bubble column reactor. Each of the bubble column reactors shown in FIGS. 33-44 includes one or more granular internal members contained in the reaction zone of the bubble column reactor. This granular inner member is added in addition to the granular pressure-containing sidewall of the reactor. As mentioned above, the reaction medium contained in the bubble column reactor has a maximum height "H" and a maximum width "W". The minimum width of the reactor medium that occurs above the height of H / 5 is referred to herein as "Wmin". In a preferred embodiment of the invention, the granular surface area of the bubble column reactor in contact with the reaction medium is preferably greater than about 3.25 W min H, more preferably from about 3.5 W min H to about 20 W min H, even more preferably from about 4 W min H to about 15 W min H, And most preferably from 5 W min H to 10 W min H. This granular surface area in contact with the reaction medium includes the area of all granular surfaces including the granular surface of the pressure-containing reactor sidewalls and the granular surface of any internal member present in the reaction zone. As used herein, the term "upright" means less than 45 ° from vertical. The granular surface in contact with the reaction medium in the bubble column reactor is preferably at an angle within 30 ° from vertical, more preferably at an angle within 15 ° from vertical, even more preferably at an angle within 5 ° from vertical and most preferably substantially. Extends vertically.

Further, the total amount of granular surface area in contact with the reaction medium due to the non-pressure-containing inner member is at least about 10% of the total amount of granular surface area in contact with the reaction medium due to the pressure-containing sidewall of the vessel. More preferably, the exposed total granular surface area in contact with the reaction medium and presented by the inner member ranges from about 15 to about 600% of the exposed total granular surface area in contact with the reaction medium and presented by the pressure-containing sidewall, even more preferred. Preferably in the range of about 25 to about 400%, and most preferably in the range of 35 to 200%. The presence of additional particulate surface area in bubble column oxidation can have a lower H: W ratio than is possible in conventional bubble column reactors with little or no added particulate surface area. Thus, the ratio of H: W of the bubble column reactor using additional granular surface area is preferably in the range of about 3: 1 to about 20: 1, more preferably in the range of about 3.5: 1 to about 15: 1 and most preferably Is 4: 1 to 12: 1.

Referring now to the embodiment shown in FIGS. 33 and 34, the oxidizing bubble column reactor 20 is a single partition wall 33 extending radially from one side of the side wall 46 to the opposite side of the side wall 46. It may include. Partition wall 33 is disposed above sparger 34 such that substantially all of the oxidant stream is introduced below the bottom of partition wall 33. Thus, about half of the gaseous phase of the reaction medium 36 comprising the undissolved portion of the oxidant stream flows upward on both sides of the partition wall 33. Preferably, about half of the feed stream is introduced through feed inlet 32a on one side of partition wall 33 and the other half of the feed stream is on the other side of partition wall 33. Is introduced via 32b. The height of the partition wall 33 is substantially the same as the height of the cylinder side wall 46.

Partition wall 33 divides reaction zone 28 into approximately two halves, and reaction medium 36 is disposed on each side of partition wall 33. Substantially all of the granular surface area of the reactor 20 in contact with the reaction medium 36 comes from the internal exposed surface of the sidewall 46 and the external exposed surface of the partition wall 33. If no partition wall 33 is present in the reaction zone 28, substantially all of the granular surface area in contact with the reaction medium 36 will originate from the side wall 46 of the pressure-containing vessel shell 22. The partition wall 33 affects the fluid flow rheology of the reaction medium 36 and provides an additional surface area that enlarges the oxidized bubble column reactor 20 without significantly negatively affecting reactor performance.

In the embodiment shown in FIG. 35, the oxidized bubble column reactor 20 is shown to include a truncated partition wall 35. The partition wall 35 of FIG. 35 is similar to the partition wall 33 of FIGS. 33 and 34, but the partition wall 35 of FIG. 35 has a height that is much smaller than the overall height of the reaction medium 36. In the configuration shown in FIG. 35, it is preferred that substantially all of the feed stream and the oxidant stream are introduced into the reaction zone 28 below the bottom of the partition wall 35. The top of the partition wall 35 is preferably spaced a considerable distance from the top surface 44 of the reaction medium 36. In this configuration, two half reaction media 36 disposed on both sides of the partition wall 35 may be mixed with each other above and below the partition wall 35.

Referring to the embodiment illustrated in FIGS. 36 and 37, the oxidizing bubble column reactor 20 adds a significant amount of particulate surface area to the reactor 20 without the need for multiple additional internal members in the reaction zone 28. It may include a non-flat partition wall (37). Similar to the partition walls 33 and 35 of FIGS. 33 to 35, the partition walls 37 of FIGS. 36 and 37 are joined and extend between radially opposed inner sidewall surfaces of the side walls 46.

In the embodiment shown in FIGS. 38 and 39, the oxidizing bubble column reactor 20 generally includes a granular inner member 41 having an X-type structure. The outer vertical edges of the inner member 41 are spaced inward from the inner surface of the side wall 46 such that the reaction medium 36 can flow between the quadrants formed by the X-type inner member 41. The various exposed outer granular surfaces of the inner member 41 add a significant amount of surface area in contact with the reaction medium 36.

40 to 42 show that a portion of the reaction zone 28 is divided into four vertical quadrants through the inner member 53 and another portion of the reaction zone 28 is eight vertical wedges through the inner member 55. An embodiment divided into zones is shown. As shown in FIG. 40, the reaction zone 28 alternates vertically between the division into four vertical quadrants through the inner member 53 and the eight wedge shaped zones through the inner member 55.

In FIGS. 43 and 44, the oxidized bubble column reactor 20 is generally a plurality of spiral inner members 61a, 61b, 61c and 61d and a vertical X-shaped partition member 63 disposed over the spiral member 61. It is shown to include. The helical member 61 has a sloped outer surface that induces a vortex flow pattern in the upward flow portion of the reaction medium 36. It is preferable to arrange the inclined directions of the helical members 61 such that the helical members 61 adjacent to each other cause the vortices of the reaction medium 36 to flow generally in opposite directions. Thus, when the helical member 61a rotates the reaction medium clockwise as the reaction medium 36 rises in the reaction zone 28, the helical member 61b (disposed immediately above the spiral member 61a) is upward. Vortex the reaction medium 36 which moves in a counterclockwise direction. Vortex of the reaction medium 36 due to the vertical inner partition member 63 and the additional granular surface area is added to the oxidized bubble column reactor 20 and the gas phase rises towards the upper surface 44 of the reaction medium 36. It can act to minimize turbulence.

Regardless of which structure of FIGS. 33-44 is used in the oxidized bubble column reactor 20, a manner in which a significant amount of molecular oxygen and oxidizing compound is introduced into the reaction zone 28 below a granular inner member or a substantial portion of the members. As such, it is preferred to introduce the oxidant stream and the feed stream into the reaction zone 28. Preferably, at least 50% by weight of all molecular oxygen and oxidative compounds introduced into the reaction zone 28 are introduced into the reaction zone 28 below the internally exposed or at least 50% of the granular exposed external surface area of the internal members, More preferably, at least about 75% by weight of the molecular oxygen and oxidizing compounds are introduced into the reaction zone below the inner member or at least 50% of the granular exposed outer surface area of the inner members, even more preferably about At least 90% by weight is introduced into the reaction zone below the granular exposed outer surface area of at least 50% of the inner member or inner members, and most preferably almost all amounts of molecular oxygen and oxidizing compound are 50 of the inner member or inner members. It is introduced into the reaction zone below the granular exposed external surface area of at least%. It is also preferred that the oxidant stream and the feed stream are introduced in such a way that a substantial amount of the gas phase of the reaction medium 36 flows upward in all aspects of the additional exposed outer surface area provided by the inner member or members. In addition, it is more preferred that the oxidant and feed streams are introduced into the reaction zone 28 according to the radial or azimuthal distribution scheme described above.

Although certain conventional oxidation reactors may utilize a heat exchange surface that contacts the reaction medium in a manner similar to the inner member (s) described herein, the inner member (s) of the present invention heat or react the reaction medium to a significant extent. It should be noted that cooling is undesirable. Accordingly, it is desirable that the thermal flux of the exposed (ie contacting reaction medium) granular surface of the inner member (s) described herein is less than about 30,000 watts per square meter.

45-53 illustrate embodiments of the invention in which the oxidized bubble column reactor 20 has one or more baffles in contact with the multiphase reaction medium 36 to improve oxidation while forming minimal impurities. Baffles are particularly useful in the enlarged bubble column reactor design described above. In each of the baffle bubble column reactors 20 shown in FIGS. 45-53, it is preferred that the baffles or baffles have an open area in the range of about 10 to about 90%. As used herein, "% open area of baffle" means the minimum percentage of horizontal cross-sectional area of the reaction zone that is open (not filled by the structure of the baffle) in the vertical position of the baffle. More preferably, the open area of the baffles or baffles shown in FIGS. 45-53 ranges from about 25 to about 75%, most preferably from 35 to 65%.

An important feature of the baffles shown in FIGS. 45-53 is their antifouling ability. As noted above, the oxidized bubble column reactor 20 is preferably used in the field of oxidation where precipitation occurs where solids form in the reaction medium 36 during oxidation. Baffles having a substantial amount of planar surface area near the horizontal upward face tend to be contaminated in reactors operating under precipitation conditions. If the baffle is contaminated, solids will be deposited on the upward facing surface of the baffle, and as the amount of solid deposited on the baffle increases, fragments of precipitated solids may separate from the baffle and fall to the bottom of the reactor. These pieces of fallen solids can be deposited at the bottom of the reactor and can cause a number of problems including, for example, suppression of slurry release at the bottom of the reactor.

In view of the foregoing, in one embodiment it is preferred that the baffles or baffles used in the oxidizing bubble column reactor 20 have no outer surface on the upward facing plane (e.g., the baffles are made of piping material having a circular cross section). Can be configured). Unless otherwise defined herein, an upward facing surface is a surface having a vertical vector that projects above the horizontal. In another embodiment, less than about 50% of the total external surface area exposed to the baffle or baffles (ie, in contact with reaction medium 36) of the baffle or baffles is less than 30 °, or 20 °, or even less than 10 ° from horizontal. As long as it comes from an inclined substantially planar surface, some substantially planar surface may be used. More preferably, less than about 35% of the baffle or up-facing exposed total external surface area of the baffles comes from a substantially planar surface that is inclined to less than 30 °, or 20 °, or even less than 10 ° from horizontal. Most preferably, less than about 25% of the baffle or up-facing exposed total external surface area of the baffles comes from a substantially planar surface that is inclined to less than 30 °, or 20 °, or even less than 10 ° from horizontal. It is also desirable for the baffle or the up-facing exposed outer surface of the baffles to have a substantially smooth finish that further resists contamination. Preferably, at least a substantial amount of the up-facing exposed outer surface of the baffle or baffles has a surface roughness of less than about 125 microns RMS, more preferably less than about 64 microns RMS, and most preferably less than 32 microns RMS. Particularly useful are electropolished finishes and smooth "2B" mechanical polished finishes.

In order to provide the optimum effect, in addition to the antifouling design of the baffles or baffles shown in FIGS. 45-53, the baffle (s) is preferably spaced from the upper and lower ends of the reaction medium 36. In a preferred embodiment of the invention, the baffles or baffles are spaced at least 0.5W and / or 0.05H from the upper and lower ends of the reaction medium 36, wherein "W" is the maximum width of the reaction medium 36, "H" is the maximum height of the reaction medium 36. More preferably, the baffle or baffles are spaced at least 1 W and / or 0.1 H from both the upper and lower ends of the reaction medium 36. Most preferably, the baffle or baffles are spaced at least 1.5 W and / or 0.15 H from both the upper and lower ends of the reaction medium 36. The presence of baffles or baffles in the oxidized bubble column reactor 20 may have a lower H: W ratio than is possible in a similar, non-baffle reactor. Thus, the ratio of H: W of the baffle bubble column reactor is preferably from about 3: 1 to about 20: 1. More preferably about 3.5: 1 to about 15: 1, and most preferably 4: 1 to 12: 1.

Referring now to Figures 45-47, the oxidized bubble column reactor 20 is shown to include a plurality of vertically spaced baffles 71a, 71b, 71c, and 71d. Preferably, the oxidizing bubble column reactor 20 comprises 2 to 20 vertically spaced baffles, most preferably 3 to 8 vertically spaced baffles. Preferably, each baffle 71 comprises a plurality of separate baffle members 73 in the longitudinal direction. In this embodiment, each of the separate baffle members 73 has a substantially cylindrical exposed outer surface in contact with the reaction medium 36. In the embodiment shown in FIGS. 45-47, the baffles 71a, 71b, 71c and 71d are rotated relative to each other so that the separate baffle members 73 of the neighboring baffles 71 extend substantially perpendicular to each other.

48-50 in detail, another baffle 81a, 81b, 81c and 81d is shown to include a plurality of separate baffle members 83 in the longitudinal direction. In this embodiment, each of the baffle members 83 is formed of an L-sectional member and generally has an exposed outer surface of the inverted V-shaped upward facing. The structure of the exposed outer surface of the separate baffle member 83 helps to prevent contamination with precipitated solids in the reaction medium 36. The number, space and orientation of the angled iron baffle member 83 may be substantially the same as described for the cylindrical baffle member 73 of FIGS. 45-47.

Referring now to FIGS. 51-53 in detail, the oxidative bubble column reactor 20 has one monolithic baffle 91 in which two oppositely extending vertical cones 93a and b generally have a form in which they are bonded to one another at the base. Is shown to include. The exposed upward facing outer surface slope of the monolithic baffle 91 prevents the baffle 91 from being contaminated with solids precipitated in the reaction medium 36.

The various baffle configurations shown in FIGS. 45-53 are illustrative only, and many other baffle configurations also fall within the scope of the present invention. It should also be noted that the baffle configurations shown in FIGS. 45-53 can be used in combination.

While heat exchange tubes that contact the reaction medium in a manner similar to the baffles described herein may be used in certain conventional oxidation reactors, it should be noted that the baffles of the present invention are not desirable to heat or cool the reaction medium to a significant extent. Accordingly, it is desirable that the heat flux of the exposed (ie in contact with the reaction medium) outer surface of the baffles disclosed herein is less than about 30,000 watts per square meter.

Referring to FIG. 54, in one embodiment of the present invention, the oxidant stream is passed through the plurality of vertically spaced oxidant inlets 98a, 98b, 98c and 98d to the reaction zone 28 of bubble column reactor 20. Is introduced. The introduction of an oxidant stream at a number of vertical points can adjust the temperature and / or oxygen concentrations and / or liquid and gas flow patterns in various portions of the reaction medium 36. It may also be useful to introduce oxygen at various vertical points into the reaction medium to control flammability, reaction rate and selectivity in various parts of the reaction medium 36. Additional control may be provided by adjusting the flow rate of the oxidant stream through the oxidant inlets 98a, 98b, 98c and 98d through control valves 99a, 99b, 99c and 99d.

54 briefly shows a number of feed inlets 98a, 98b, 98c and 98d penetrating the vessel shell at a number of vertically spaced locations, but a number of internal oxidant dispersing means without multiple vessel penetrations. It should be understood that it can be used to introduce the oxidant stream at various heights of the reaction zone 28. For example, bubble column reactor 20 may have two or more vertically spaced apart structures each having a similar structure to sparger 34 shown in FIGS. 2-5 and / or sparger 200 shown in FIGS. Can include a sparger. Alternatively, the oxidant stream may be introduced into the reaction zone 28 at multiple heights in a manner similar to that described above to introduce the feed stream into the reaction zone 28 at multiple heights. Thus, a distribution system similar to that shown in FIGS. 6-10 can be used to distribute the oxidant stream into the reaction zone 28.

Regardless of the structure of the system used to introduce the oxidant stream into the reaction zone 28, the oxidant stream comprising molecular oxygen is preferably introduced into the reaction zone 28 through a plurality of oxidant openings, wherein the oxidant opening Two or more of which are spaced vertically from each other by about 1 W and / or 0.15 H or more, where "W" is the maximum width of the reaction medium 36 and "H" is the maximum height of the reaction medium 36. More preferably, two or more of the oxidant openings are spaced vertically from each other by about 2 W and / or 0.25 H or more. Most preferably, two or more of the oxidant openings are spaced vertically from each other by at least 3W and / or 0.4H. As described above, it is preferred that the oxidant stream introduced into the reaction zone 28 through the vertically spaced oxidant openings has a concentration substantially equal to oxygen. Preferably, all of the vertically spaced oxidant streams comprise less than about 50 mole percent molecular oxygen, more preferably less than about 40 mole percent molecular oxygen, even more preferably about 18 to about 24 mole percent molecular oxygen. do. Most preferably, the oxidant stream is air.

At least about 10 mole percent, or at least about 20 mole percent, or even at least 40 mole percent of all molecular oxygen introduced into the reaction zone 28 is about 0.5 W and / or 0.05 of the bottom portion 52 of the reaction zone 28. It is preferred to be introduced into the reaction zone 28 in H. At least about 10 mole percent, or at least about 20 mole percent, or even at least 40 mole percent of all molecular oxygen introduced to the reaction zone 28 is about 0.25 W and / or 0.025 of the bottom portion 52 of the reaction zone 28. More preferably, it is introduced into the reaction zone 28 in H. At least about 10 mole percent, or at least about 20 mole percent, or even at least 40 mole percent of all molecular oxygen introduced to the reaction zone 28 is about 1.5 W and / or 0.15 above the bottom 52 of the reaction zone 28. More preferably, it is introduced into the reaction zone 28 at a position or positions vertically spaced above H. At least about 10 mole percent, or at least about 20 mole percent, or even at least 40 mole percent of all molecular oxygen introduced to the reaction zone 28 is at least 2 W and / or 0.2 H from the bottom 52 of the reaction zone 28. More preferably, it is introduced into the reaction zone 28 at a vertically spaced position or locations.

As suggested above, the oxygen concentration in the liquid phase of the reaction medium 36 at various heights can be adjusted by introducing an oxidant stream at multiple vertically spaced locations. As described above, the reaction medium 36 may have an oxygen concentration gradient that changes vertically. A portion of the oxidant stream is preferably introduced into the reaction zone 28 at a height at which the oxygen concentration in the reaction medium 36 falls below 60% of the maximum oxygen concentration in the reaction medium 36 or at a predetermined location above this height. Do. In order to define the location of the high oxidant opening (s) used to introduce the oxidant stream at or above the low oxygen concentration, the reaction medium 36 may theoretically be divided into 30 separate horizontal layers of equal volume. And the oxygen concentration of each horizontal slice can be measured on a time-averaged and volume-averaged molar (wet) basis. The horizontal separation with the highest oxygen concentration is called the O ₂ -maximum horizontal separation. O ₂ - O ₂ with 60% less than the oxygen concentration of the oxygen concentration of the maximum horizontal partitioned - first horizontal partitioned above the maximum level is partitioned O ₂ - is referred to as partitioned threshold level. It is preferred that the O ₂ -maximum horizontal slice is one of the ten horizontal slices of the bottom, most preferably one of the five horizontal slices of the bottom. It is preferred that the O ₂ -critical horizontal slice is one of the middle 20 horizontal slices, most preferably one of the middle 14 horizontal slices. The O ₂ -critical slice is at least two horizontal slices above the O ₂ -maximum horizontal slice, more preferably at least 4 horizontal slices above the O ₂ -maximum horizontal slice, and most preferably above the O ₂ -maximum horizontal slice. It is preferred to be located at six or more horizontal layers of. At least a portion of the oxidant stream is preferably introduced into the reaction zone 28 through oxidant opening (s) located at or above the O ₂ -critical horizontal separation. Preferably 5%, 10%, or even 25% by weight of all molecular oxygen introduced into the reaction zone 28 is introduced into the reaction zone 28 in or above the O ₂ -critical horizontal separation.

In addition, the temperature of reaction medium 36 at various heights can be adjusted by introducing an oxidant stream at multiple vertically spaced locations. As described above, the reaction medium 36 may have a temperature gradient that changes vertically. A portion of the oxidant stream is introduced into the reaction zone 28 at a height at which the temperature of the reaction medium 36 increases undesirably above the lowest temperature at the bottom half of the reaction medium 36 or at any height at this height. It is preferable. To define the location of the high oxidant opening (s) used to introduce the oxidant stream, the reaction medium 36 can theoretically be divided into 30 separate horizontal layers of equal volume, the temperature of each horizontal layer being time -Average and volume-average. The horizontal slice with the lowest temperature of the bottom 15 horizontal slices may be referred to as the T-lowest horizontal slice. Undesirably a first layer above the T-lowest horizontal slice with a high temperature may be referred to as a T-critical horizontal slice. The temperature of the T-critical horizontal separation is at least about 1 ° C. higher than the temperature of the T-lowest horizontal separation, more preferably at least about 2 ° C. higher than the temperature of the T-lowest horizontal separation, most preferably the T-lowest horizontal. It is preferable that it is temperature higher 3 degreeC or more than the temperature of a partition. It is preferred that the T-lowest horizontal slice is one of the bottom ten horizontal slices, most preferably one of the bottom five horizontal slices. It is preferred that the T-critical horizontal slice is one of 20 middle slices, most preferably one of 14 middle horizontal slices. The T-critical slice is at least two horizontal slices on top of the T-lowest horizontal slice, more preferably at least four horizontal slices on top of the T-lowest horizontal slice, and most preferably six on top of the T-lowest horizontal slice. It is preferable to be located in the above horizontal division. At least a portion of the oxidant stream is preferably introduced into reaction zone 28 through oxidant opening (s) located at or above the T-critical separation. Preferably, 5%, 10%, or even 25% by weight of all molecular oxygen introduced into the reaction zone 28 is introduced into the reaction zone 28 in or above the T-critical horizontal separation.

In addition to the vertical distribution of the oxidant stream just described above, it is preferred that the oxidant stream is distributed azimuthally and radially in the manner described above.

In addition, radial and directional distribution of the oxidant stream introduction can be used to reconstruct the natural convective flow pattern in the reaction medium of the oxidizing bubble column. The normal pattern of natural convective flow resets itself into two or three column diameters, usually open above the high inlet of the oxidant, but there is local disturbance to provide important controllability in the vertical stepping of the oxidant and oxidizing compound. Conceptually, the azimuthal and radial localization of the oxidant stream can be adjusted to provide more or less hydraulic baffles of the reaction medium.

In a preferred embodiment using a high oxidant inlet to provide a hydraulic baffle, the inlet is used to increase the vertical step by reducing the liquid end-end mixing as follows. Ascending injection of an oxidant stream results in at least about 60%, more preferably in flows other than 60%, more preferably 70%, most preferably 80% of the cross-sectional area closest to the horizontal center of the reaction medium at the height of the inlet. It is preferred to release at least about 80%, most preferably substantially all. At the local location of this outer diameter oxidant inlet, the increased presence of oxidant flowing near the wall of the reaction vessel will resist and delay time-averaged flow of liquid or slurry down the wall. In a more preferred embodiment, the flow opening area of said and high outer diameter oxidant inlet will mostly face horizontally or horizontally upwards, preferably at least 50% of the open area, more preferably at least 70%, most preferably More than 90% will point horizontally or above a horizontal position.

In a more preferred embodiment that uses a high oxidant inlet to provide a hydraulic baffle, the inlet is used to increase the vertical stage by reducing the liquid end-end mixing as follows. The high inlet of the oxidant stream is located within 70%, more preferably 60%, most preferably 50% of the cross-sectional area closest to the horizontal center of the reaction medium at the height of the inlet, oriented in the horizontal position and below the horizontal position. It is preferred to release at least about 60%, more preferably at least about 80% and most preferably substantially all of the flow. At the local location of this inner diameter oxidant inlet, the horizontal or downward motion of the incoming oxidant stream will resist and delay the time average flow of the reaction medium flowing upward through the core of the reaction vessel. In the most preferred embodiment, more than 50%, more preferably more than 70%, and most preferably more than 90% of the flow opening area of the high inner diameter oxidant inlet is within about 60 ° from the vertical, more preferably about Downwards within 45 °, most preferably within 30 °.

Referring again to FIGS. 1-54, according to a preferred embodiment disclosed herein, oxidation is preferably performed in bubble column reactor 20 under conditions very different from conventional oxidation reactors. When bubble column reactor 20 is used to liquid phase partial oxidation of para-xylene to crude terephthalic acid (CTA) according to a preferred embodiment disclosed herein, the local reaction intensity combined with the liquid flow pattern in the reaction medium, The local evaporation intensity and the spatial profile of the local temperature and the preferred relatively low oxidation temperature contribute to the formation of CTA particles with significant and beneficial properties.

Basic CTAs prepared according to one embodiment of the present invention are shown in FIGS. 55A and 55B. FIG. 55A is an enlarged view of the basic CTA particles at 500 magnification, and FIG. 54B is a view enlarged at 2000 magnifications by zooming on one of the basic CTA particles. As best shown in FIG. 55B, each basic CTA particle typically forms a small mass of multiple CTA subparticles such that the basic CTA particles have a relatively high surface area, high porosity, low density and good solubility. Basic CTA particles typically have an average particle size of about 20 to about 150 microns, more preferably about 30 to about 120 microns, and most preferably 40 to 90 microns. Basic CTA subparticles typically have an average particle size of about 0.5 to about 30 microns, more preferably about 1 to about 15 microns, most preferably 2 to 5 microns. The relatively high surface area of the basic CTA particles shown in FIGS. 55A and 55B can be quantified using the Braunauer-Emmett-Teller (BET) surface area measurement method. Preferably, the basic CTA particles have an average BET surface area of at least about 0.6 m 2 / g. More preferably, the basic CTA particles have an average BET surface area of about 0.8 m 2 / g to about 4 m 2 / g. Most preferably, the basic CTA particles have an average BET surface area of about 0.9 m 2 / g to about 2 m 2 / g. The physical properties (eg, particle size, BET surface area, porosity and solubility) of the basic CTA particles formed by the optimal oxidation process of a preferred embodiment of the present invention are more effective and / or as described further below in connection with FIG. 57. Economical methods can be used to purify CTA particles.

Average particle size provided above is measured using polarization microscopy and image analysis. The device used for particle size analysis was a Nikon E800 optical microscope with a 4x Plan Flour NA 0.13 objective, a Spot RT digital camera and a personal computer operated Image Pro Plus. V4.5.0.19 Includes image analysis software. The particle size analysis method comprises the steps of (1) dispersing the CTA powder in mineral oil; (2) providing a microscope slide / cover slip of the dispersion; (3) illuminating the slide using a polarization microscope (cross polarization conditions—particles appear as bright objects on a black background); (4) collecting different images for each sample preparation (field size = 3 x 2.25 mm, pixel size = 1.84 microns / pixel); (5) performing image analysis with Image Pro Plus software; (6) sending particle measurements to a spreadsheet; And (7) performing statistical characterization on the spreadsheet. "Perform phase analysis with Image Pro Plus software" of step 5 includes: (a) setting a phase threshold to detect white particles on a dark background; (b) generating a binary image; (c) operating the open filter once to filter the pixel noise; (d) measuring all particles in the image; And (e) recording the average diameter measured on each particle. Image Pro Plus software measures the average diameter of each particle at 2 degree intervals and defines it as the number average diameter length of the particles passing through the particle center. “Perform statistical characterization in the spreadsheet” of step 7 involves calculating the volume-weighted average particle size as follows. The volume of each of the n particles in the sample is obtained by using pi / 6 * d _i ³ when the particles are spherical, multiplying the volume of each particle by its diameter to obtain pi / 6 * d _i ⁴ , and pi / 6 * sum for all particles in the sample of d _i ⁴ value, sum the volume of all particles in the sample, sum the sum for all n particles in the sample of (pi / 6 * d _i ⁴ ) (pi / 6 * d _i Calculated by calculating the volume-weighted particle diameter by dividing by the sum for all n particles in the sample of ³ ). As used herein, “average particle size” refers to the volume-weighted average particle size measured according to the test method described above, and is also referred to as D (4,3) in the following equation:

Step 7 also includes finding such a particle size where any fraction of the total sample volume is less than that particle size. For example, D (v, 0.1) is 10% smaller, 90% larger particle size, and D (v, 0.5) is half larger and half smaller particles. Size, D (v, 0.9) is the particle size where 90% of the total sample volume is smaller. Step 7 also includes calculating a D (v, 0.9) -D (v, 0.1) value defined herein as a “particle size spread”; Step 7 involves calculating the particle size spread divided by D (4,3), which is defined as “particle size relative spread”.

In addition, the D (v, 0.1) of the CTA particles measured above is preferably in the range of about 5 to about 65 microns, more preferably in the range of about 15 to about 55 microns, and most preferably in the range of 25 to 45 microns. . The D (v, 0.5) of the CTA particles measured above is preferably in the range of about 10 to 90 microns, more preferably in the range of about 20 to about 80 microns, and most preferably in the range of 30 to 70 microns. The D (v, 0.9) of the CTA particles measured above is preferably in the range of about 30 to about 150 microns, more preferably in the range of about 40 to about 130 microns, and most preferably in the range of 50 to 110 microns. The particle size relative spread is preferably in the range of about 0.5 to about 20 microns, more preferably in the range of about 0.6 to about 1.5 microns and most preferably in the range of 0.7 to 1.3 microns.

The BET surface area values provided above are measured by Micromeritics ASAP2000 (commercially available from Micromeritics Instrument Corporation, Norcross, GA, USA). In the first step of the measurement process, 2-4 g of sample of the particles are weighed and dried at 50 ° C. The sample is then placed on an analytical gas manifold and cooled to 77 ° K. The nitrogen absorption isotherm is measured by exposing the sample to a known volume of nitrogen gas at a minimum of 5 equilibrium pressures to measure the decrease in pressure. The equilibrium pressure is suitably in the range P / Po of 0.01 to 0.20, where P is the equilibrium pressure and Po is the vapor pressure of liquid nitrogen at 77 ° K. The isotherms thus obtained are then plotted according to the following BET equation 2:

Where

V _a is the volume of gas absorbed by the sample at P,

V _m is the volume of gas required to cover the entire surface of the sample with a single layer of gas,

C is a constant.

From this plot, V _m and C are determined. Thereafter, V _m is converted to the surface area using the cross-sectional area of nitrogen at 77 ° K by the following equation:

Where

σ is the cross section of nitrogen at 77 ° K,

T is 77 ° K,

R is the gas constant.

As suggested above, CTAs formed in accordance with one embodiment of the present invention exhibit superior dissolution properties as compared to conventional CTAs prepared by other processes. This improved dissolution rate allows the CTA of the present invention to be purified by more efficient and / or more effective purification processes. The description below relates to how the dissolution rate of CTA is quantified.

The rate at which a known amount of solids are dissolved in a known amount of solvent in the shaken mixture can be measured by various protocols. As used herein, a measurement method called "timed dissolution test" is defined as follows. An atmospheric pressure of about 0.1 MPa is used throughout the time dissolution test. The ambient temperature used throughout the time dissolution test is about 22 ° C. In addition, the solids, solvents, and all dissolution apparatus maintain complete thermal equilibrium at this temperature before testing begins, and there is no appreciable heating or cooling of the beaker or its contents while dissolution proceeds. 250 g of a solvent portion of a new HPLC analytical grade tetrahydrofuran (having a purity of at least 99.9%, hereinafter referred to as "THF") have a generally uninsulated, smooth surface and are generally cylindrical cleaned KIMAX tall form) in a 400 ml glass beaker (Kimble Part No. 14020, Kimble / Kontes, Vineland, NJ). Teflon-coated magnetic stir bar (VWR Part No. 58948-230, approximately 1 inch long, octagonal cross-section with 3/8 inch diameter, VWR International, 19380 West Chester, Pa.) So that it naturally sits at the bottom. The sample is agitated at 800 revolutions per minute using the trademark Variomag multipoint 15 magnetic stirrer (H & P Laboratore AG, Auberscheilsheim, Germany). The stirring begins only 5 minutes before the solids are added and continues for at least 30 minutes after the solids are added. Up to 250 mg of solid or purified sample of TPA particles is weighed with a non-stick sample weighing pan. At the onset time at which t is zero, all of the metered solids are poured into the stirred THF all at once, and at the same time the timer starts. If normal, THF quickly wets the solid to form a well shaken dilution slurry within 5 minutes. Thereafter, samples of this mixture are obtained at the following times when t is measured from 0 in minutes: 0.08, 0.25, 0.50, 0.75, 1.00, 1.50, 2.00, 2.50, 3.00, 4.00, 5.00, 6.00, 8.00 , 10.00, 15.00, and 30.00. Each small sample is obtained from a well shaken dilution mixture using a new disposable syringe (5 ml, REF 30163, manufactured by Becton, Dickinson and Co., 07417 Franklin Lake, NJ). . Upon collection from the beaker, approximately 2 ml of clear liquid sample was used with a new syringe filter (Pall Corporation, 11548 East Hills, NY, USA) manufactured by a 25 mm diameter and 0.45 micron gelman GHP acrodisk GF ( Rapid release into new labeled glass sample vials via Gelman GHP Acrodisc GF®). The duration of each of the operations of filling the syringe, placing in the filter, and releasing into the sample vial is exactly less than about 5 seconds, and this time interval is appropriately initiated and terminated within about 3 seconds before and after each target sampling time. Within about 5 minutes after every operation of filling the sample vial, the sample vial is capped and maintained at approximately constant temperature until the next chemical analysis is performed. After a final sample was obtained 30 minutes after t was zero, all 16 samples were analyzed to determine the amount of dissolved TPA using the HPLC-DAD method outlined elsewhere herein. However, in this test, both calibration standards and reported results are based on mg of TPA dissolved per gram of THF solvent (hereinafter referred to as “ppm in THF”). For example, if all 250 mg of solids were very pure TPA and the entire amount was completely dissolved in 250 g of THF solvent before obtaining a particular sample, the precisely measured concentration would be about 1,000 ppm in THF.

When the CTA according to the present invention is subjected to the timed dissolution test described above, a sample obtained when one minute has elapsed from the time t is 0 is dissolved at a concentration of at least about 500 ppm in THF, more preferably at least 600 ppm in THF. It is preferable. In the sample obtained when 2 minutes have elapsed from the time point at which t is 0, it is preferable that the CTA according to the present invention is dissolved at a concentration of about 700 ppm or more in THF, more preferably at a concentration of 750 ppm or more in THF. In the sample obtained when 4 minutes have elapsed since the time t is 0, it is preferable that the CTA according to the present invention is dissolved at a concentration of about 840 ppm or more in THF, more preferably at a concentration of 880 ppm or more in THF.

The inventors have found that a relatively simple negative expontial growth model is useful to account for the time dependence of the entire data set from a complete timed dissolution test, despite the complexity of the particle sample and the dissolution process. The equation referred to below as the "timed dissolution model" is as follows:

Where

t = time in minutes;

S = solubility in ppm in THF at time t;

exp = exponential with base of natural logarithms;

A, B = regression constant in ppm in THF, where A is mainly related to the rapid dissolution of smaller particles in a very short time, and A + B is mainly related to the total amount of dissolution at the end of the specified test time. Related);

C = regression time constant in reciprocal minute (min- ¹ ).

The regression constant is adjusted to minimize the sum of the squares of the errors between the actual data points and the corresponding model values, which method is generally referred to as a "least squared" fit. A preferred software package for performing this data regression is JMP Release 5.1.2 (JMP Software, SAS Institute Inc., Carrie SAS Campus Drive, 27513, NC, USA).

When the CTA according to the invention is tested in a timed dissolution test and applied to the timed dissolution model described above, a time constant of CTA greater than about 0.5 / min, more preferably about 0.6 / min, and most preferably 0.7 / min It is preferred to have "C".

56A and 56B show conventional CTA particles produced by conventional high temperature oxidation processes in a continuous stirred tank reactor (CSTR). FIG. 56A shows conventional CTA particles magnified 500 times, and FIG. 56B shows CTA particles zoomed 2000 times. Visually comparing the CTA particles of the present invention shown in FIGS. 55A and 55B with the conventional CTA particles shown in FIGS. 56A and 56B show that conventional CTA particles have higher density, smaller surface area, and more than the CTA particles of the present invention. Low porosity and larger particle size. Indeed, the conventional CTA shown in FIGS. 56A and 56B has an average particle size of about 205 microns and a BET surface area of about 0.57 m 2 / g.

57 shows a conventional process for preparing purified terephthalic acid (PTA). In a conventional PTA process, para-xylene is partially oxidized in a mechanically stirred high temperature oxidation reactor 700. The slurry comprising CTA is collected from reactor 700 and then purified in purification system 702. The PTA product of purification system 702 is introduced into separation system 706 to separate and dry the PTA particles. Purification system 702 represents a significant portion of the costs associated with producing PTA particles by conventional methods. Purification system 702 generally includes a water addition / exchange system 708, a dissolution system 710, a hydrogenation system 712, and three separate crystallization vessels 704a, 704b, and 704c. In the water addition / exchange system 708, a significant amount of mother liquor is replaced with water. After adding water, water / CTA slurry is introduced into dissolution system 710 where the water / CTA mixture is heated until the CTA particles are completely dissolved in water. After CTA dissolution, the CTA solution in water is hydrogenated in hydrogenation system 712. Thereafter, the hydrogenated effluent from hydrogenation system 712 is treated by three crystallization steps in crystallization vessels 704a, 704b, and 704c, and the PTA is separated in separation system 706.

58 illustrates an improved process for preparing PTA using bubble column oxidation reactor 800 constructed in accordance with one embodiment of the present invention. The initial slurry comprising solid CTA particles and liquid mother liquor is collected from reactor 800. Typically, the initial slurry may contain solid CTA particles in the range of about 10-50% by weight, with the remainder being liquid mother liquor. Solid CTA particles present in the initial slurry are typically at least about 400 ppmw 4-carboxybenzaldehyde (4-CBA), more typically at least about 800 ppmw 4-CBA and most typically 4- in the range of 1,000 to 15,000 ppmw. Contains CBA. The initial slurry collected from reactor 800 is introduced into purification system 802 to reduce the concentration of 4-CBA and other impurities present in the CTA. A purer / purified slurry is prepared in the purification system 802 and separated and dried in the separation system 804 to 4-CBA of less than about 400 ppmw, more preferably 4-CBA of less than about 250 ppmw, and most preferably Pureer solid terephthalic acid particles are prepared comprising 4-CBA in the range of 10-200 ppmw.

The purification system 802 of the PTA manufacturing system shown in FIG. 58 provides a number of advantages over the purification system 802 of the prior art system shown in FIG. Preferably, purification system 802 generally includes a liquid exchange system 806, a cracker 808, and a single crystallizer 810. In liquid exchange system 806, at least about 50% by weight of the mother liquor present in the initial slurry is exchanged with a fresh exchange solvent to form a solvent-exchanged slurry comprising CTA particles and exchange solvent. Solvent-exchanged slurry exiting the liquid exchange system 806 is introduced into a cracker (or second oxidation reactor) 808. In the cracker 808, the second oxidation reaction is conducted at a slightly higher temperature than that used in the initial / first oxidation reaction carried out in the bubble column reactor 800. As discussed above, any impurities trapped in the CTA particles due to the large surface area, small particle size, and low density of the CTA particles produced in the reactor 800 are decomposed 808 without complete dissolution of the CTA particles in the cracker 808. Can be used for oxidation. Thus, the temperature in cracker 808 may be lower than many similar prior art processes. The second oxidation reaction carried out in the cracker 808 preferably reduces the concentration of 4-CBA in CTA to at least 200 ppmw, more preferably at least about 400 ppmw, and most preferably from 600 to 6,000 ppmw. Preferably, the second oxidation temperature in cracker 808 is about 10 ° C. or more than the first oxidation temperature in bubble column reactor 800, more preferably about 20 to about 80 above the first oxidation temperature in reactor 800. 30 ° C. or higher, most preferably 30-50 ° C. or higher above the first oxidation temperature in reactor 800. The second oxidation temperature is preferably in the range of about 160 to about 240 ° C, more preferably in the range of about 180 to about 220 ° C, most preferably in the range of 190 to about 210 ° C. The purified product from the cracker 808 requires a single crystallization step in crystallizer 810 prior to separation in separation system 804. Suitable second oxidation / degradation techniques are described in more detail in US Patent Application Publication No. 2005/0065373, the entire disclosure of which is hereby expressly incorporated by reference.

Terephthalic acid (eg, PTA) prepared by the system shown in FIG. 58 is preferably at least about 40 microns, more preferably in the range of about 50 to about 2,000 microns, and most preferably in the range of 60 to 200 microns It is formed of PTA particles having a size. The PTA particles preferably have an average BET surface area of less than about 0.25 m 2 / g, more preferably in the range of about 0.005 to about 0.2 m 2 / g, and most preferably in the range of 0.01 to 0.18 m 2 / g. The PTA produced by the system shown in FIG. 58 is suitable for use as feedstock in PET production. Typically, PET is prepared by esterifying terephthalic acid with ethylene glycol followed by polycondensation. Preferably, terephthalic acid prepared by one embodiment of the present invention is used as a feed to the pipe reactor PET process described in US patent application Ser. No. 10 / 013,318, filed Dec. 7, 2001, and the entire disclosure thereof. The contents are incorporated herein by reference.

CTA particles having the preferred form disclosed herein are particularly useful in the oxidative degradation process described above for reducing the amount of 4-CBA. In addition, these preferred CTA particles offer advantages in a wide variety of other post-processes, including dissolution and / or chemical reaction of the particles. These additional post-processes include reactions that react with one or more hydroxyl containing compounds to form ester compounds, particularly those that react CTA with methanol to form dimethyl terephthalate and impurity esters; Reaction with one or more diols to form ester monomers and / or polymeric compounds, particularly reaction of CTA with ethylene glycol to form polyethylene terephthalate (PET); And reactions that are complete or partially soluble in solvents including, but not limited to, water, acetic acid, and N-methyl-2-pyrrolidone, wherein the process is more pure terephthalic acid. And additional treatments including, but not limited to, reprecipitation of and / or selective chemical reduction of carbonyl groups except for carboxylic acid groups. Notably, substantial dissolution of CTA is included in a solvent comprising water, combined with partial hydrogenation to reduce aldehydes, particularly 4-CBA, fluorenone, phenone, and / or anthraquinone.

CTA particles having the desirable properties disclosed herein include, but are not limited to, mechanical breakdown of non-conforming CTA particles and complete or partial dissolution of non-conforming CTA particles from CTA particles (nonconforming CTA particles) that do not conform to the desirable properties disclosed herein; It is contemplated that it may be prepared by a method including, but not limited to, subsequent full or partial reprecipitation.

According to one embodiment of the invention, there is provided a process of partially oxidizing an oxidative aromatic compound to one or more aromatic carboxylic acid types, wherein the purity of the solvent portion of the feed (ie, "solvent feed") and the oxidizing compound portion of the feed (Ie, "oxidative compound feed") is controlled within the specific ranges specified below. With other embodiments of the present invention, this allows the purity of the liquid phase of the reaction medium and, if present, the solid and combined slurry (ie, solid and liquid) phases to be adjusted within the specific preferred ranges outlined below.

With regard to the solvent feed, it is known to oxidize the oxidative aromatic compound (s) to produce aromatic carboxylic acids, wherein the solvent feed introduced into the reaction medium is a mixture of acetic acid and water with analytical purity, which is a laboratory scale and Often used at pilot scale. It is also known that oxidative aromatic compounds are oxidized to aromatic carboxylic acids, in which the solvent exiting the reaction medium is separated from the produced aromatic carboxylic acid and then circulated back to the reaction medium as a solvent feed for reasons of production costs. do. Recycling of these solvents causes certain impurity feeds and process byproducts to accumulate in the recycled solvent over time. Various means for purifying recycled solvent prior to reintroduction to the reaction medium are known in the art. In general, purifying recycled solvent to a higher degree incurs significantly higher production costs than purifying to a lower degree by similar means. One embodiment of the present invention understands the preferred range of many impurities present in the solvent feed (many of which have been considered to be generally harmless so far) in order to find the optimal balance between overall production cost and overall product purity. It is about defining.

A "recycled solvent feed" is defined herein as a solvent feed comprising at least about 5% by weight of material that has already passed through a reaction medium containing at least one oxidative aromatic compound that is partially oxidized. For reasons of solvent inventory and operating time in the production unit, at least once per day during the operation of some of the recycled solvent, more preferably at least once per day during the continuous operation of at least 7 days, and most preferably 30 days It is preferred to pass through the reaction medium at least once per day for at least continuous operation. For economic reasons, at least about 20%, more preferably at least about 40%, even more preferably at least about 80%, and most preferably 90% by weight of the solvent feed fed to the reaction medium of the present invention It is preferred that at least% is recycled solvent.

For reasons of reaction activity and taking into account the metallic impurities remaining in the oxidation product, the inventors have found that the concentration of the selected polyvalent metal in the recycled solvent feed is preferably within the range specified below. The concentration of iron in the recycled solvent is preferably less than about 150 ppmw, more preferably less than about 40 ppmw, and most preferably 0 to 8 ppmw. The concentration of nickel in the recycled solvent is preferably less than about 150 ppmw, more preferably less than about 40 ppmw, and most preferably 0 to 8 ppmw. The concentration of chromium in the recycled solvent is preferably less than about 150 ppmw, more preferably less than about 40 ppmw, and most preferably 0 to 8 ppmw. The concentration of molybdenum in the recycled solvent is preferably less than about 75 ppmw, more preferably less than about 20 ppmw, and most preferably 0 to 4 ppmw. The concentration of titanium in the recycled solvent is preferably less than about 75 ppmw, more preferably less than about 20 ppmw, and most preferably 0 to 4 ppmw. The concentration of copper in the recycled solvent is preferably less than about 20 ppmw, more preferably less than about 4 ppmw, and most preferably 0 to 1 ppmw. Typically, other metallic impurities are also present in the recycled solvent, which generally have lower levels than one or more of the metals listed above. By adjusting the metals listed above within the desired range, other metallic impurities can be maintained at a suitable level.

Such metals may occur as impurities in any process feeds that are introduced (eg, incoming oxidizing compounds, solvents, oxidants, and catalyst compounds). Alternatively, the metal may arise as a corrosion product from any process unit in contact with the reaction medium and / or recycled solvent. Means for adjusting metals to the disclosed concentration ranges include suitable specifications and monitors of various feed purity, and many commercial grade titanium and stainless steels, such as grades and high-bolybdenum stainless steels known as duplex stainless steels, Appropriate use of the constituent material is not limited to this.

We also found a preferred range for the aromatic compounds selected in the recycled solvent. This includes aromatic compounds precipitated and dissolved in recycled solvent.

Surprisingly, even products precipitated from partial oxidation of para-xylene (eg TPA) are contaminants that must be managed in recycled solvent. Because there is a preferred range of levels of solids in the reaction medium, any precipitated product in the solvent feed is subtracted directly from the amount of oxidizable compound that can be fed at once. In addition, supplying high levels of precipitated TPA solids in recycled solvent adversely affects the properties of particles formed in the precipitating oxidation medium, resulting in downstream operation (e.g., product filtration, solvent cleaning, oxidative degradation of crude product, further processing). In the dissolution of the crude product, etc.). Another undesirable characteristic of precipitated solids in recycled solvent feeds is that they often contain very high amounts of precipitated impurities as compared to the concentration of impurities in solids in the TPA slurry in which many recycled solvents are produced. Presumably, the high levels of impurities observed in the solids suspended in the recycled filtrate, whether intentional or ambient loss, and the nucleation time for precipitation of certain impurities from the recycled solvent and / or cooling of the recycled solvent May be related. For example, highly colored and undesirable concentrations of 2,6-dicarboxyfluorenone are much higher in solids present in recycled solvents at 80 ° C. than those observed in TPA solids separated from recycled solvents at 160 ° C. High level was observed. Similarly, the concentration of isophthalic acid was observed at much higher levels in the solids present in the recycled solvent compared to the levels observed in the TPA solids from the reaction medium. When reintroduced into the reaction medium, the exact manner in which certain precipitated impurities trapped in the recycled solvent will behave differently. This probably depends on the relative solubility of the impurities in the liquid phase of the reaction medium, the manner of stacking the impurities precipitated in the solid, and the local rate of TPA precipitation in which the solid is first reintroduced into the reaction medium. Thus, the inventors have found it useful to control the amount of certain impurities in the recycled solvent as described below, whether the impurities are present in the solvent recycled in dissolved form or the particles trapped therein. .

The amount of precipitated solids present in the recycled filtrate is determined by gravimetric method as follows. Representative samples are collected from the solvent feed fed to the reaction medium while the solvent flows into the reaction medium in the conduit. A useful sample size is about 100 g contained in a glass container having an internal volume of about 250 ml. While flowing to the sample vessel before it is at atmospheric pressure, the recycled filtrate is cooled to below 100 ° C., in order to limit solvent evaporation for a short time interval before sealing in the glass vessel. After placing the sample under atmospheric pressure, the glass container is immediately sealed. The sample is then cooled to about 20 ° C using air at about 20 ° C without forced convection. Upon reaching about 20 ° C., the sample is maintained for at least about 2 hours under the above conditions. Next, the sealed container is shaken strongly until a visually uniform distribution of solids is formed. Immediately thereafter, the magnetic stir bar is placed in the sample vessel and rotated at a speed sufficient to maintain a uniform distribution of solids effectively. A 10 ml aliquot of the mixed liquid with suspended solids is collected using a pipette and weighed. Subsequently, the liquid material is substantially separated from this aliquot at a temperature of about 20 ° C. by vacuum filtration without loss of solids. The wet solid filtered from the aliquot is then substantially dried without sublimation of the solid and then the weight of the dried solid is measured. Typically, the ratio of the weight of dried solids to the weight of the original aliquot of the slurry is the fraction of solids expressed as a percentage, referred to herein as the amount of recycled filtrate in the precipitated solid at 20 ° C.

The inventors surprisingly find that aromatic compounds (e.g., ice phthalic acid, benzoic acid, phthalic acid, 2,5,4'-tricarboxybiphenyl) dissolved in the liquid phase of the reaction medium and comprising aromatic carboxylic acids lacking non-aromatic hydrocarbyl groups It was found to be harmful. It has been found that these compounds nevertheless perform many detrimental reactions, although the chemical activity of these compounds is much reduced in the present reaction medium compared to oxidative compounds having non-aromatic hydrocarbyl groups. Therefore, it is advantageous to adjust the amount of the compound in the preferred range in the liquid phase of the reaction medium. This leads to a preferred range of compounds selected from recycled solvent feeds and a preferred range of precursors selected from feeds of oxidative aromatic compounds.

For example, in the liquid partial oxidation of para-xylene to terephthalic acid (TPA), the inventors found that when meta-substituted aromatic compounds are present in very low amounts in the reaction medium, they are largely colored and undesirable impurities. It has been found that 2,7-dicarboxyfluorenone (2,7-DCF) is substantially undetectable in the reaction medium and in the product flow path. The inventors have also found that 2,7-DCF is formed in almost direct proportion when isophthalic acid impurities are present in increasing amounts in the solvent feed. We have found that 2,7-DCF is also formed almost directly in proportion when meta-xylene impurities are present in the feed of para-xylene. Moreover, even if the solvent feed and the oxidative compound feed are free of meta-substituted aromatics, some isophthalic acid is present during typical partial oxidation reactions of very pure para-xylene, especially when benzoic acid is present in the liquid phase of the reaction medium. The inventors have found that this is formed. Spontaneously-generated isophthalic acid can be increased over time in commercial units using recycled solvents because it has a higher solubility than TPA in solvents comprising acetic acid and water. Thus, the amount of isophthalic acid in the solvent feed, the amount of meta-xylene in the oxidative aromatic compound feed and the rate of spontaneous production of isophthalic acid in the reaction medium are all appropriately balanced with each other and also consume any isophthalic acid. I think it's in balance. It has been found that isophthalic acid performs an additional consumption reaction in addition to the formation of 2,7-DCF as described below. In addition, the inventors have found that when setting the appropriate range for the meta-substituted aromatic species in the partial oxidation of para-xylene to TPA, another consideration has to be taken into account. Other highly colored and undesirable impurities, such as 2,6-dicarboxyfluorenone (2,6-DCF), are largely associated with dissolved para-substituted aromatic species that are always present with para-xylene supplied for liquid phase oxidation. It seems to be related. Therefore, the inhibition of 2,7-DCF is most considered equally with the amount of other colored impurities produced.

For example, the inventors have found that in the liquid partial oxidation of para-xylene to TPA, trimellitic acid is formed when the amount of isophthalic acid and phthalic acid increases in the reaction medium. Trimellitic acid is a trivalent carboxylic acid which branches the polymer while producing PET from TPA. For many PET applications, the degree of branching should be controlled at low levels, so trimellitic acid should be controlled at low levels in purified TPA. In addition to the induction of trimellitic acid, the presence of meta- and auto-substituted species in the reaction medium also forms other tricarboxylic acids (eg, 1,3,5-tricarboxybenzene). Moreover, increasing the presence of tricarboxylic acids in the reaction medium also increases the formation of tetracarboxylic acids (eg, 1,2,4,5-tetracarboxybenzene). Control of the combined production of all aromatic carboxylic acids having two or more carboxylic acid groups is one in setting the preferred levels of meta-substituted and auto-substituted species in the recycled solvent feed, the oxidative compound feed and the reaction medium in accordance with the present invention. Becomes an element of.

For example, in the liquid partial oxidation of para-xylene to TPA, the inventors have found that the production of carbon monoxide and carbon dioxide is also direct if the amount of some dissolved aromatic carboxylic acids lacking non-aromatic hydrocarbyl groups increases in the liquid phase of the reaction medium. It was found to increase. This increased production of carbon oxides results in yield loss of both the oxidizing agent and the oxidizing compound, and many of the aromatic carboxyls produced together are then regarded as impurities on the one hand and commercially valuable on the other. Thus, adequate removal of relatively soluble carboxylic acids lacking non-aromatic hydrocarbyl groups from the recycled solvent, in addition to suppression of the production of highly undesirable impurities such as various fluorenones and trimellitic acids, can lead to oxidative aromatic compounds and oxidizing agents. Provides economic value in preventing yield loss.

For example, in the liquid partial oxidation of para-xylene to TPA, the inventors have found that the formation of 2,5,4'-tricarboxybiphenyl in appearance is inevitable. 2,5,4'-tricarboxybiphenyls may be prepared by the coupling of two aromatic rings, possibly to decarboxylation or decarbonylation of dissolved para-substituted aromatic species and aryl radicals (possibly para-substituted aromatic species). Is an aromatic tricarboxylic acid formed by a bond). Fortunately, 2,5,4'-tricarboxybiphenyl is typically produced at lower levels than trimellitic acid and does not significantly increase the difficulty due to branching of polymer molecules during the production of PET. However, the inventors have found that, in accordance with a preferred embodiment of the present invention, elevated levels of 2,5,4'-tricarboxybiphenyl in reaction media involving the oxidation of alkyl aromatic compounds result in highly colored undesirable 2,6. It has also been found to increase the level of DCF. Increased 2,6-DCF is likely formed with loss of water molecules by ring closure from 2,5,4'-tricarboxybiphenyl, but the exact reaction mechanism is not known for certain. If 2,5,4'-tricarboxybiphenyl, which is more soluble than TPA in solvents containing acetic acid and water, increases too much in recycled solvent, the rate of conversion to 2,6-DCF will be unacceptable.

For example, in the liquid partial oxidation of para-xylene to TPA, the inventors have found that the chemical activity of the reaction medium when aromatic carboxylic acids (eg, isophthalic acid) lacking non-aromatic hydrocarbyl groups are present in the liquid phase at a sufficient concentration It has been found that it generally inhibits slowly.

For example, in the liquid partial oxidation of para-xylene to TPA, the inventors have found that precipitation is very often non-ideal (ie, non-equilibrium) with respect to the relative concentrations of other species in the solid and liquid phases. . Perhaps this is believed to be because the precipitation rate at the desired space-time reaction rate is so great that it results in non-ideal co-precipitation or even closure of impurities. Thus, if it is desirable to limit the concentration of certain impurities (eg trimellitic acid and 2,6-DCF) in the crude TPA because of the configuration of the downflow unit operation, these in the reaction medium as well as their concentration in the solvent feed It is also desirable to control the speed of.

For example, the inventors have found that benzophenone compounds (eg, 4,4 ') prepared during partial oxidation of para-xylene, although the benzophenone compounds are not highly colored in the TPA itself, such as fluorenone and anthraquinone. -Dicarboxybenzophenone and 2,5,4'-tricarboxybenzophenone) have been found to have undesirable results in PET reaction media. Therefore, it is desirable to limit the presence of benzophenone and to select precursors in recycled solvent and oxidative compound feed. In addition, the inventors have found that the presence of elevated levels of benzoic acid results in elevated production rates of 4,4′-dicarboxybenzophenones, even if benzoic acid is introduced in the recycled solvent or formed in the reaction medium. It became.

In this regard, the inventors have found and quantified a surprising array of reactions for aromatic compounds lacking non-aromatic hydrocarbyl groups present in the liquid partial oxidation of para-xylene to TPA. To summarize a single case of benzoic acid, the increase in benzoic acid in the reaction medium of certain embodiments of the present invention greatly colored and greatly increased the production of undesirable 9-fluorenone-2-carboxylic acid, and 4,4'-dicarboxy Greatly increase the level of biphenyl, increase the level of 4,4'-dicarboxybenzophenone, gently inhibit the chemical activity of the intended oxidation of para-xylene, and reduce the level of carbon oxides and subsequent yield loss It was found to increase. The inventors have found that an increase in the level of benzoic acid in the reaction medium also increases the production of isophthalic acid and phthalic acid, the increasing level of which is preferably controlled in a low range according to a similar embodiment of the present invention. Since some of the inventors considered using benzoic acid instead of acetic acid as the main component of the solvent, the number and importance of reactions involving benzoic acid is perhaps even more surprising (see, for example, US Pat. No. 6,562,997). In addition, the present inventors have found that benzoic acid is a very important ratio for the formation of It was found that during the oxidation it is spontaneously-generated.

On the other hand, the inventors have found that the addition of the recycled solvent composition to the aromatic reaction medium and the presence of the oxidative aromatic compound in that both the oxidative aromatic compound and the aromatic reaction intermediate have non-aromatic hydrocarbyl groups and are relatively soluble in the recycled solvent Little value was found for accommodation. In general, these compounds are introduced or produced in the reaction medium at a rate substantially greater than that present in the recycled solvent; The rate of consumption of these compounds in the reaction medium is so great that it adequately inhibits their increase in recycled solvent to retain one or more non-aromatic hydrocarbyl groups. For example, during partial oxidation of para-xylene in a multiphase reaction medium, para-xylene evaporates to a limited extent with a large amount of solvent. If this evaporated solvent is withdrawn from the reactor as part of the exhaust gas and condensed and collected as recycle solvent, a substantial portion of the evaporated para-xylene is also condensed there. There is no need to limit the concentration of the para-xylene in the recycle solvent. For example, if the solvent separates from the solid when the slurry exits the para-xylene oxidation reaction medium, this collected solvent contains dissolved para-toluic acid at a concentration similar to that present when removed from the reaction medium. something to do. Although it may be important to limit the current concentration of para-toluic acid in the liquid phase of the reaction medium (see below), para-toluic acid has a relatively good stability and production of para-toluic acid in the reaction medium Since it has a low mass flow rate compared to, it is not necessary to adjust para-toluic acid separately in this part of the recycled solvent. Similarly, we find aromatic compounds with methyl substituents (e.g. toluic acid), aromatic aldehydes (e.g. terephthalaldehyde), aromatic compounds with hydroxy-methyl substituents (e.g. 4-hydroxymethylbenzoic acid) and one The concentration in the recycle solvent of the brominated aromatic compound (eg alpha-bromo-para-toluic acid) bearing the above non-aromatic hydrocarbyl groups is discharged from the reaction medium resulting from the partial oxidation of xylene according to a preferred embodiment of the present invention. Little reason has been found to be limited to concentrations below those originally observed in the liquid phase. Surprisingly, the inventors have also found that there is no need to adjust the concentration of selected phenols inherently produced during the partial oxidation of xylene in the recycle solvent, which means that these compounds are reacted at a much higher rate than in the presence of the recycled solvent. Because it is created and destroyed within. For example, the inventors have reported that 4-hydroxybenzoic acid acts as an important poison in similar reaction media, although it is reported by others that higher amounts of para-xylene, 1 kg, in the natural presence of the recycled solvent. When 4-hydroxybenzoic acid is injected together at a ratio of at least 2 g of sugar 4-hydroxybenzoic acid, it has been found that 4-hydroxybenzoic acid only has a relatively small effect on chemical activity in a preferred embodiment of the present invention (pattenheimer (W). Partenheimer, Catalysis Today 23 (1995) p. 81].

Thus, as will now be described, there are many reactions and many considerations in determining the preferred range of various aromatic impurities in the solvent feed. This is described in terms of the total weight average composition of all solvent streams fed to the reaction medium for a fixed period of time, preferably one day, more preferably one hour, and most preferably one minute. For example, one solvent feed flows substantially 40ppmw of isophthalic acid at a flow rate of 7 kg / min, and another second solvent feed flows substantially 2000 ppmw of isophthalic acid at a flow rate of 10 kg / min. If there is no other solvent feed stream flowing to the reaction medium, the total weight average composition of the solvent feed is calculated by the following equation:

It is noticeable that the weight of any oxidant compound feed or any oxidant feed, possibly mixed with the solvent feed before being introduced into the reaction medium, is not taken into account in calculating the total weight average composition of the solvent feed.

 Table 1 below sets forth preferred values for certain components in the solvent feed introduced into the reaction medium. The solvent feed components listed in Table 1 are as follows: 4-carboxybenzaldehyde (4-CBA), 4,4'-dicarboxycistbene (4,4'-DCS), 2,6-dicarboxylic cyantra Quinone (2,6-DCA), 2,6-dicarboxyfluorenone (2,6-DCF), 2,7-dicarboxyfluorenone (2,7-DCF), 3,5-dicarboxyfluorenone ( 3,5-DCF), 9-fluorenone-2-carboxylic acid (9F-2CA), 9-fluorenone-4-carboxylic acid (9F-4CA), total fluorenones (including other fluorenones not separately listed) 4,4'-dicarboxybiphenyl (4,4'-DCB), 2,5,4'-tricarboxybiphenyl (2,5,4'-TCB), phthalic acid (PA) De (IPA), benzoic acid (BA), trimellitic acid (TMA), 2,6-dicarboxybenzocomarin (2,6-DCBC), 4,4'-dicarboxybenzyl (4,4'-DCBZ) , 4,4'-dicarboxybenzophenone (4,4'-DCBP), 2,5,4'-tricarboxybenzophenone (2,5,4'-TCBP), terephthalic acid (TPA), precipitated at 20 ° C Aromatic carboxyl lacking solid and non-aromatic hydrocarbyl groups . Table 1 below provides the preferred amounts of these impurities in the CTA prepared according to embodiments of the present invention.

Many other aromatic impurities are also typically present in the recycled solvent at levels much lower than and / or in proportion to one or more of the disclosed aromatic compounds. Typically, other aromatic impurities will be maintained at appropriate levels due to the method of adjusting the disclosed aromatic compounds to the desired range.

When bromine is used in the reaction medium, it is known that many ionic and organic forms of bromine are present in dynamic equilibrium. These various forms of bromine have different stability as soon as they leave the reaction medium and move through the various unit operations associated with the recycled solvent. For example, alpha-bromo-para-toluic acid may persist as it is under some conditions, or may be rapidly hydrolyzed under other conditions to form 4-hydroxymethylbenzoic acid and hydrogen bromide. In the present invention, at least about 40 wt%, more preferably at least about 60 wt%, most preferably at least about 80 wt% of the total bromine present in the total solvent feed introduced into the reaction medium is ionic bromine, alpha Preference is given to at least one of -bromo-para-toluic acid and bromoacetic acid.

Although the importance and value of adjusting the total average purity of the solvent feed to the disclosed desired range of the present invention has not been discovered and / or disclosed so far, suitable means for controlling the solvent feed purity are known in the art. It will be similar to already known methods. First, any solvent evaporated from the reaction medium typically has a suitable purity, provided that no liquid or solid from the reaction medium is trapped in the evaporated solvent. As disclosed herein, the injection of reflux solvent droplets into the exhaust gas separation space above the reaction medium limits the capture appropriately; Recycled solvents of suitable purity in terms of aromatics can be condensed from the exhaust gas. Second, more difficult and expensive purification of the recycled solvent feed is generally a solvent taken in liquid form from the reaction medium, and a solvent (eg, a solid that is in continuous contact with the liquid and / or solid phase of the reaction medium collected from the reaction vessel). Recycled solvents obtained from concentrated and / or washed filters, recycled solvents obtained from centrifuges where solids are concentrated and / or washed, recycled solvents obtained from crystallization processes, and the like. However, means for carrying out the necessary purification of such recycled solvent streams using one or more of the previously described means are also known. With regard to controlling the precipitated solids in the recycled solvent within the specified ranges, suitable control means include gravity sedimentation, mechanical filtration using filter paper on rotary belt filters and rotary drum filters, mechanical filtration using fixed filter media in pressure vessels. , Hydro-cyclones, centrifugation, and the like, but are not limited thereto. With regard to adjusting the dissolved aromatic species in the recycled solvent to the defined range, the means for controlling includes the means described in US Pat. No. 4,939,297 and US Patent Application Publication No. 2005-0038288, which are incorporated herein by reference. It is not limited to this. However, none of these prior inventions found nor disclosed the desired level of purity in the total solvent feed as disclosed herein. Rather, the prior inventions only provide a means for purifying a partial selective stream of recycled solvent without eliciting an optimal value of the total weight average solvent composition supplied to the reaction medium.

As for the purity of the feed of oxidizing compounds, it is known that certain amounts of isophthalic acid, phthalic acid, and benzoic acid are present in the purified TPA used to produce the polymer, which is acceptable at low levels. Moreover, it is known that these species are relatively more soluble in many solvents and can be advantageously removed from the purified TPA by the crystallization process. However, from the embodiments of the invention disclosed herein, the amount control of several relatively soluble aromatic species, mainly isophthalic acid, phthalic acid, and benzoic acid, in the liquid reaction medium results in the polycyclic and colored aromatic compounds produced in the reaction medium. It has now been found that it is of great importance for controlling the level of, controlling compounds having two or more carboxylic acid functional groups per molecule, controlling the reaction activity in the partial oxidation reaction medium, and controlling the yield loss of oxidizing agents and aromatic compounds.

It is known in the art that isophthalic acid, phthalic acid and benzoic acid are formed in the reaction medium as follows. The meta-xylene impure feed is oxidized to good conversion to produce IPA. The auto-xylene impure feed is oxidized to good conversion to form phthalic acid. Ethylbenzene and toluene impure feeds are oxidized to good conversion to form benzoic acid. However, the inventors have found that significant amounts of isophthalic acid, phthalic acid and benzoic acid are also formed by means other than oxidation of meta-xylene, auto-xylene, ethylbenzene and toluene in a reaction medium comprising para-xylene. I found out. Such other intrinsic chemical modalities probably include decarbonylation, decarboxylation, reconstitution of transition states, and the addition of methyl and carbonyl radicals to aromatic rings.

Many factors are involved in determining the preferred range of impurities in an oxidative compound feed. If the purity requirements of the oxidized product are very stringent, any impurities in the feed will probably result in a direct yield loss and the cost of purification of the product (e.g., in the reaction medium for partial oxidation of para-xylene, Toluene and ethylbenzene, typically found in para-xylene, form benzoic acid, most of which is usually removed from commercially available TPA). When partial oxidation products of impure feeds participate in additional reactions, considering the cost of feed purification, factors other than mere yield loss and removal should be appropriately considered (e.g., para In the reaction medium for the partial oxidation of xylene, ethyl benzene forms benzoic acid, which in turn continues to form highly colored 9-fluorenone-2-carboxylic acid, isophthalic acid, phthalic acid, and increased carbon oxides. do). If the reaction medium spontaneously forms additional amounts of impurities by chemical mechanisms not directly related to the impure feed, the analysis is still more complicated (e.g., in the reaction medium for partial oxidation of para-xylene). Benzoic acid is also spontaneously produced from para-xylene itself). In addition, the downstream flow treatment of the crude oxidation product can affect the consideration of the purity of the desired feed. For example, the costs of removing direct impurities (benzoic acid) and subsequent impurities (isophthalic acid, phthalic acid, 9-fluorenone-2-carboxylic acid, etc.) to appropriate levels may be the same and may differ from each other, It may differ from the elimination requirements of highly unrelated impurities (eg, 4-CBA, an incomplete oxidation product in the oxidation of para-xylene to TPA).

When para-xylene is fed together with a solvent and an oxidant to the reaction medium for the partial oxidation to produce TPA, the range of purity of the feeds disclosed below for para-xylene is preferred. This range is more preferred in TPA-generating processes having a post-oxidation step that is removed from reaction medium impurities (eg, catalytic metals) other than oxidants and solvents. This range includes the removal of 4-CBA from CTA (oxidation for conversion of 4-CBA to TPA by conversion of CTA to dimethyl terephthalate and impurity ester followed by distillation of methyl ester of 4-CBA by distillation. Even more preferred for the TPA production process, by the decomposition process, by a hydrogenation process for converting 4-CBA to para-toluic acid; then, they are separated by a partial crystallization process. This range is most preferred in the TPA production process to remove additional 4-CBA from CTA by oxidative degradation methods for converting 4-CBA to TPA.

Compared with other inherent chemistries, using the new knowledge of the relative amounts of aromatic compounds formed directly from the oxidation of impurity feeds and the preferred range of recycled aromatic compounds, the impurities supplied to the partial oxidation process for TPA production An improved range of impurities has been found for para-xylene. Table 2 below provides the preferred values for the amounts of meta-xylene, auto-xylene, and ethylbenzene + toluene in the para-xylene feed.

Those skilled in the art will recognize that the impurities in impure para-xylene can have the greatest impact on the reaction medium after partial oxidation product has accumulated in the recycled solvent. For example, injecting 400 ppmw, the upper limit of the most preferred range of meta-xylene, will immediately produce about 200 ppmw of isophthalic acid in the liquid phase of the reaction medium when operated with about 33 weight percent solids in the reaction medium. . This is compared to the inflow of 400 ppmw, the upper limit of the most preferred range of isophthalic acid in the recycled solvent, after typical solvent evaporation to cool the reaction medium, with an amount of about 1,200 ppmw isophthalic acid in the liquid phase of the reaction medium. That is, the effect of meta-xylene, auto-xylene, ethylbenzene and toluene impurities in the feed of impure para-xylene may be greatest on the accumulation of partial oxidation products over time in the recycled solvent. To present that. Thus, the above range for impure para-xylene feeds is at least half per day during the operating days of any partial oxidation reaction medium in a particular manufacturing unit, more preferably at least 3/4 per day during the last at least seven days of operation It is preferred that the mass mean of the impure para-xylene feed composition in the continuous operation for at least 30 days is maintained within the above preferred range.

Means for obtaining impure para-xylene with preferred purity are already known in the art and include, but are not limited to, distillation, partial crystallization at subatmospheric temperatures and molecular sieve methods using selective pore size adsorption. It is not limited to. However, the preferred range of purity specified herein is more difficult and expensive than that characteristically implemented by commercially available suppliers of para-xylene at its upper end; At the lower end of the preferred range, the effects of the combined production of impurities themselves from the para-xylene itself and the impurity consumption reaction in the reaction medium are found to be more important than those of impurity feeding of impurities in the para-xylene. Excessive expensive purification of para-xylene to be fed to the oxidation reaction medium is avoided.

When the xylene-containing feed stream contains selected impurities such as ethylene-benzene and / or toluene, the oxidation of these impurities may result in the production of benzoic acid. As used herein, the term "impurity-generated benzoic acid" means benzoic acid derived from a source other than xylene during oxidation of xylene.

As described herein, some of the benzoic acid produced during the oxidation of xylene is derived from xylene itself. In addition to producing any amount of benzoic acid that may be impurity-generated benzoic acid, the production of benzoic acid from xylene is evident. Without being bound by theory, benzoic acid is derived from xylene in the reaction medium when various intermediate oxidation products of xylene spontaneously decarbonylated (carbon monoxide loss) or decarboxylated (carbon dioxide loss) to form aryl radicals. I think. The aryl radicals can then desorb hydrogen atoms from one of the many available raw materials in the reaction medium to produce spontaneously produced benzoic acid. Whatever the chemical mechanism, as used herein, the term “spontaneously-generated benzoic acid” means benzoic acid derived from xylene during xylene oxidation.

As disclosed herein, when para-xylene is oxidized to produce terephthalic acid (TPA), the production of spontaneously produced benzoic acid causes a loss of yield of para-xylene and a loss of oxidant yield. In addition, the presence of spontaneously-generated benzoic acid in the liquid phase of the reaction medium is associated with an increase in many undesirable side reactions which mainly involves the production of largely colored compounds called mono-carboxy-fluorenones. Spontaneously-generated benzoic acid also affects the undesirable accumulation of benzoic acid in the recycled filtrate, which in turn raises the concentration of benzoic acid in the liquid phase of the reaction medium. Therefore, while it is desirable to minimize the formation of spontaneously-generated benzoic acid, it is appropriate to simultaneously consider the impurity-generated benzoic acid, factors affecting the consumption of benzoic acid, factors related to the reaction selectivity of other subjects, and the overall economics. .

The inventors have found that spontaneously-generated benzoic acid can be adjusted to low levels by appropriately selecting temperature, xylene distribution and oxygen availability in the reaction medium during oxidation. While not wishing to be bound by theory, the lower temperature and improved oxygen availability seem to inhibit the decarbonylation rate and / or the decarboxylation rate to avoid yield loss of spontaneously-generated benzoic acid. Sufficient oxygen availability seems to make the aryl radicals a more acceptable product (particularly hydroxybenzoic acid). The distribution of xylene in the reaction medium also affects the balance between conversion of aryl radicals to benzoic acid or conversion of aryl radicals to hydroxybenzoic acid. Whatever the chemical mechanism, we have stringent reaction conditions sufficient to oxidize a significant amount of hydroxybenzoic acid with carbon monoxide and / or carbon dioxide that is readily removed from the oxidation product, even if sufficient to reduce the production of benzoic acid. Was found.

In a preferred embodiment of the invention, the oxidation reactor is constructed and operated in such a way that the formation of spontaneously-generated benzoic acid is minimized and the oxidation of hydroxybenzoic acid to carbon monoxide and / or carbon dioxide is maximized. If an oxidation reactor is used to oxidize para-xylene to terephthalic acid, it is preferred that para-xylene comprises at least about 50% by weight of total xylene in the feed stream introduced into the reactor. More preferably, para-xylene comprises at least 75% by weight of total xylenes in the feed stream. Still more preferably, para-xylene comprises at least 95% by weight of total xylenes in the feed stream. Most preferably, para-xylene makes up almost all of the total xylenes in the feed stream.

When a reactor is used to oxidize para-xylene to terephthalic acid, it is desirable to maximize the production rate of terephthalic acid while minimizing the production rate of spontaneously-generated benzoic acid. The ratio of the production rate (weight) of terephthalic acid to the production rate (weight) of spontaneously-produced benzoic acid is preferably at least about 500: 1, more preferably at least about 1,000: 1 and most preferably at least 1,500: 1. As can be seen below, it is desirable to measure the rate of production of spontaneously-generated benzoic acid when the concentration of benzoic acid in the liquid phase of the reaction medium is less than 2,000 ppmw, more preferably less than 1,000 ppmw and most preferably less than 500 ppmw. This is because such low concentrations moderately inhibit the rate of conversion of benzoic acid to other compounds.

When combining the spontaneously-produced benzoic acid and the impurity-produced benzoic acid, the ratio of the production rate (weight) of terephthalic acid to the production rate (weight) of total benzoic acid is preferably about 400: 1 or more, more preferably about 700: At least 1, and most preferably at least 1,100: 1. As can be seen below, when the concentration of benzoic acid in the liquid phase of the reaction medium is less than 2,000 ppmw, more preferably less than 1,000 ppmw, and most preferably less than 500 ppmw, the spontaneously-generated benzoic acid and the impurity-generated benzoic acid It is desirable to measure the combined rate of production, because such low concentrations will moderately lower the rate of reaction for conversion of benzoic acid to other compounds.

As described herein, elevated concentrations of benzoic acid in the liquid phase of the reaction medium increase the formation of many other aromatic compounds, several of which are harmful impurities to TPA; As described herein, elevated concentrations of benzoic acid in the liquid phase of the reaction medium increase the formation of carbon oxide gas, which indicates loss of yield of oxidant and aromatic compounds and / or solvents. Moreover, the inventors have found that the formation of other aromatic compounds and carbon monoxide is increased in significant amounts from reactions in which some of the benzoic acid molecules themselves are converted, in contrast to benzoic acid, which is not consumed by itself and catalyzes other reactions. Thus, "effective production of benzoic acid" is defined as the time-average weight of all benzoic acid exiting the reaction medium for the same period minus the time-average weight of all benzoic acid entering the reaction medium. Effective production of the benzoic acid is often positive by the rate of production of impurity-generated benzoic acid and the rate of production of spontaneously-generated benzoic acid. However, the inventors have found that carbon oxides of benzoic acid increase as the concentration of benzoic acid increases in the liquid phase of the reaction medium, as measured when other reaction conditions, including temperature, oxygen availability, STR, and reaction activity, remain adequately constant. And the rate of conversion to several other compounds has increased almost proportionally. Thus, if the concentration of benzoic acid in the liquid phase of the reaction medium is large enough, perhaps due to an increase in the concentration of benzoic acid in the recycled solvent, the conversion of the benzoic acid molecule to another compound (e.g. carbon oxide) is equal to or greater than the chemical production of the new benzoic acid molecule. Can be large. In such cases, the effective production of benzoic acid can be nearly zero or even negative. We believe that when the effective production of benzoic acid has a positive value, the ratio of the production rate (weight) of terephthalic acid in the reaction medium to the effective production rate of benzoic acid in the reaction medium is preferably greater than about 700: 1, more preferably. Preferably greater than about 1,100: 1, most preferably greater than 4,000: 1. We believe that when the effective production of benzoic acid has a negative value, the ratio of the production rate (weight) of terephthalic acid in the reaction medium to the effective production rate of benzoic acid in the reaction medium is preferably greater than about 200: (-1), More preferably greater than about 1,000: (-1), most preferably greater than 5,000: (-1).

We also found a preferred range for the composition of the slurry (liquid + solid) collected from the reaction medium and the solid CTA portion of the slurry. The composition of the preferred slurry and the preferred CTA was surprisingly good and useful. For example, purified TPA produced from the preferred CTA by oxidative decomposition has sufficiently low total impurities and colored impurities such that the purified TPA is a PET fiber without hydrogenation of additional 4-CBA and / or colored impurities. Suitable for a wide range of uses and PET packaging applications. For example, a preferred slurry composition provides a liquid phase of the reaction medium with a relatively low concentration of important impurities, which inhibits the production of other impurities even more undesirable as disclosed herein. In addition, the preferred slurry compositions aid in the continued processing of the liquid from the slurry to be suitably pure recycled solvent in accordance with another embodiment of the present invention.

CTAs prepared according to one embodiment of the present invention contain fewer impurities of the selected type than CTAs prepared by conventional processes and apparatuses, primarily those using recycled solvents. Impurities that may be present in the CTA include: 4-carboxybenzaldehyde (4-CBA), 4,4'-dicarboxycistbene (4,4'-DCS), 2,6-dicarboxylic cyantraquinone (2 , 6-DCA), 2,6-dicarboxyfluorenone (2,6-DCF), 2,7-dicarboxyfluorenone (2,7-DCF), 3,5-dicarboxyfluorenone (3,5 -DCF), 9-fluorenone-2-carboxylic acid (9F-2CA), 9-fluorenone-4-carboxylic acid (9F-4CA), 4,4'-dicarboxybiphenyl (4,4'-DCB), 2,5,4'-tricarboxybiphenyl (2,5,4'-TCB), phthalic acid (PA), isophthalic acid (IPA), benzoic acid (BA), trimellitic acid (TMA), para-toluic acid ( PTAC), 2,6-dicarboxybenzocomarin (2,6-DCBC), 4,4'-dicarboxybenzyl (4,4'-DCBZ), 4,4'-dicarboxybenzophenone (4,4 '-DCBP), 2,5,4'-tricarboxybenzophenone (2,5,4'-TCBP). Table 3 below shows the preferred amounts of impurities in the CTA prepared according to embodiments of the present invention.

It is also preferred that CTAs prepared according to embodiments of the present invention have a reduced amount of color compared to CTAs prepared by conventional processes and apparatuses, primarily those using recycled solvents. Thus, CTAs prepared according to one embodiment of the invention preferably have a transmission at 340 nm of at least about 25%, more preferably at least about 50% and most preferably at least 60%. In addition, it is preferred that CTAs prepared according to one embodiment of the present invention have a transmission at 400 nm of at least about 88%, more preferably at least about 90%, and most preferably at least 92%.

The test of transmittance is to determine colored light absorbing impurities present in TPA or CTA. Herein, this test relates to measurements made on a portion of a solution prepared by dissolving 2.00 g of anhydrous solid TPA or CTA in 20.0 ml of dimethyl sulfoxide (DMSO) of analytical or higher grade. A portion of the solution was made of quartz, a Helma semi-micro flow cell with a light path of 1.0 cm and a volume of 0.39 ml, PN 176.700 (Hellma USA; 11803, New York, USA) Location on Plane View Skyline Drive 80). Agilent 8453 Diode Array Spectrophotometer is used to measure the transmission of different wavelengths of light passing through the charged flow cell (Agilent Technologies; Palo Alto Page, California, 94303, USA) Mill Road 395). After appropriately correcting the absorbance from the background, including but not limited to the cells and solvents used, the results of the transmittance characterized by the fraction of incident light passing through the solution are recorded directly by the machine. The transmittance values (%) at light wavelengths of 340 nm and 400 nm are particularly useful for distinguishing pure TPA from the many impurities typically found.

Preferred ranges of the various aromatic impurities in the slurry (solid + liquid) phase of the reaction medium are shown in Table 4 below.

The preferred composition of this slurry is that the reaction medium is usefully avoiding experimental challenges associated with precipitation of additional liquid components from the reaction medium into the solid phase component, sampling of liquids and solids, and changes in analytical conditions during sampling. To achieve the desired composition of the liquid phase.

Many other aromatic impurities are typically present on the slurry of the reaction medium and in the CTA of the reaction medium in varying amounts even at lower levels and / or in proportion to at least one of the disclosed aromatic compounds. Adjusting the disclosed aromatic compounds to the desired ranges will keep other aromatic impurities at appropriate levels. This advantageous composition for solid CTA obtained directly from the slurry phase and slurry in the reaction medium may be possible by operating according to the embodiments of the invention described herein for partial oxidation of para-xylene to TPA.

Determination of the concentrations of solvents, recycled solvents, CTA, slurries from the reaction medium, and low levels of components in PTA is carried out using liquid chromatography methods. Two interchangeable embodiments are described below.

The method, referred to herein as HPLC-DAD, includes high pressure liquid chromatography (HPLC) combined with a diode array detector (DAD) to provide separation and quantification of several molecular species in a given sample. The instrument used for this measurement is a Model 1100 HPLC with DAD provided by Agilent Technologies (Palto Alto, Calif.), Other suitable instruments are also commercially available and are available from other suppliers. As is known in the art, both elution time and detector reaction are calibrated using known compounds present in known amounts, compounds and amounts corresponding to those occurring in substantially unknown samples.

The method, referred to herein as HPLC-MS, includes high pressure liquid chromatography (HPLC) combined with a mass spectrometer (MS) to provide separation, identification and quantification of several molecular species in a given sample. Instruments used for this measurement are Alliance HPLC and ZQ MS provided by Waters Corp (Milford, Mass., USA), with other suitable instruments also commercially available and from other suppliers. As is known in the art, both the elution time and the reaction of the mass spectrometer are calibrated using known compounds present in known amounts, compounds and amounts corresponding to those occurring in substantially unknown samples.

Another embodiment of the invention relates to aromatics, on the one hand, with respect to the production of carbon dioxide and carbon monoxide (collectively carbon oxides; CO _x ), on the other hand, while maintaining appropriately balanced suppression of harmful aromatic impurities It involves the partial oxidation of oxidizing compounds. Typically, the carbon oxides are discharged in the exhaust gas from the reaction vessel, which ultimately corresponds to the detrimental loss of solvents and oxidizable compounds, including the preferred oxidizing derivatives (eg acetic acid, para-xylene and TPA). The inventors have found a lower limit for the production of carbon oxides, and as described below, it is likely that a lot of harmful aromatic impurities are produced below this lower limit, and thus the total conversion level is so inferior that there is no economic usefulness. The inventors have also found an upper limit of carbon oxides, above which the production of carbon oxides continues to increase, with little added value provided by the reduction of the production of harmful aromatic impurities.

The inventors have found that reducing the liquid phase concentration of the aromatic oxidizing compound feed and aromatic intermediate species in the reaction medium during the partial oxidation of the aromatic oxidizing compound lowers the rate of formation of harmful impurities. Such harmful impurities include bound aromatic rings and / or aromatic molecules containing more than the desired number of carboxylic acid groups (e.g., harmful impurities in the oxidation of para-xylene are 2,6-dicarboxylic cyanquinone, 2, 6-dicarboxyfluorenone, trimellitic acid, 2,5,4'-tricarboxybiphenyl, and 2,5,4'-benzophenone). Aromatic intermediate species are derived from feeds of oxidative aromatic compounds and still include aromatic compounds containing non-aromatic hydrocarbyl groups (e.g., in the oxidation of para-xylene, aromatic intermediate species are para-tolualdehyde, Terephthaldehydro, para-toluic acid, 4-CBA, 4-hydroxymethylbenzoic acid and alpha-bromo-para-toluic acid). Feeds of aromatic oxidizing compounds and aromatic intermediate species containing non-aromatic hydrocarbyl groups are already described herein for dissolved aromatic species (eg, isophthalic acid) that lack non-aromatic hydrocarbyl groups when present in the liquid phase of the reaction medium. It is likely to form harmful impurities in a manner similar to that described in.

In connection with this need for higher reaction activity to inhibit the formation of harmful aromatic impurities during the partial oxidation of oxidative aromatic compounds, the inventors have found that the undesirable following results increase the production of carbon oxides. Importantly, such carbon oxides are believed to exhibit a yield loss of oxidizing compounds and oxidants (but not solvents). Obviously, substantial and sometimes significant amounts of carbon oxides are derived from oxidizable compounds and derivatives thereof rather than from solvents; Often oxidizable compounds require a higher cost per carbon unit than in solvents. Importantly, it is also contemplated that the preferred product, the carboxylic acid (eg TPA), is also peroxidized to carbon oxide when it is present in the liquid phase of the reaction medium.

It is also important to consider that the present invention relates to reactions and reactant concentrations in the liquid phase of the reaction medium. This is in contrast to some prior art, which is directly related to the production of precipitated solid forms of aromatic compounds containing non-aromatic hydrocarbyl groups. In particular, in the partial oxidation of para-xylene to TPA, certain prior inventions relate to the amount of 4-CBA precipitated in the solid phase of CTA. However, the inventors have found that, when using the same conditions of temperature, pressure, catalyst, solvent composition and space-time reaction rate of para-xylene, partial oxidation is carried out in a well-mixed autoclave or according to the invention. It was found that the ratio of 4-CBA in the solid phase to 4-CBA in the liquid phase changed to 2: 1 or more, depending on whether oxygen and para-xylene were carried out in the staged reaction medium. In addition, the inventors found that the ratio of 4-CBA in the solid phase to 4-CBA in the liquid phase mixed well according to the space-time reaction rate of para-xylene in other similar specifications of temperature, pressure, catalyst, and solvent composition. It has been found that it can vary from 2: 1 or higher in the reaction medium or staged medium. Additionally, 4-CBA in solid CTA does not appear to be due to the formation of harmful impurities, and 4-CBA in solid phase can be collected and simply oxidized to high yields of TPA (e.g., By oxidative decomposition of the CTA slurry, as described in); On the other hand, the removal of harmful impurities is more difficult and expensive than the removal of solid 4-CBA, and the production of carbon oxides represents a sustained yield loss. Therefore, it is important to recognize that this aspect of the invention relates to the liquid phase composition in the reaction medium.

We believe that the production of carbon oxides in the conversion of commercial applications, whether derived from solvents or from oxidizing compounds, is dependent on the temperature of the reaction medium as measured by temperature, metal, halogen, pH, and overall reaction activity. It has been found that despite a wide range of changes in certain combinations of concentrations of water used to obtain levels, it is strongly related to the level of overall reaction activity. The inventors have found that it is useful in partial oxidation of xylene to assess the overall reaction activity level using the medium height of the reaction medium, the bottom of the reaction medium and the liquid phase concentration of toluic acid at the top of the reaction medium.

Therefore, increasing the reaction activity to minimize the generation of harmful impurities and lowering the reaction activity to minimize the production of carbon oxides takes place at the same time in balance. In other words, if the total production of carbon oxides is too low, harmful impurities are formed at excessive levels, and vice versa.

In addition, the inventors have found that the solubility of the desired carboxylic acid (e.g. TPA) and the presence of other dissolved aromatic species lacking relative reactivity and non-aromatic hydrocarbyl groups play an important post function in this balance of carbon oxides and harmful impurities. It was found to cause. Typically, the desired product, the carboxylic acid, is dissolved in the liquid phase of the reaction medium even when present in solid form. For example, at a preferred range of temperatures, TPA is dissolved in a reaction medium comprising acetic acid and water at levels ranging from about 1,000 ppmw to more than 1 wt%, and solubility increases with increasing temperature. Oxidative aromatic compound feeds (e.g. para-xylene), aromatic reaction intermediates (e.g. para-toluic acid), the desired product aromatic carboxylic acid (e.g. TPA), and aromatic species lacking non-aromatic hydrocarbyl groups (e.g. Despite the difference in reaction rate for the formation of several harmful impurities from isophthalic acid), the presence and reactivity of the latter two groups result in the former two groups: oxidative aromatic compound feeds and aromatic reaction intermediates. With respect to the further suppression of the area of recovery, there is a reduction area. For example, in the partial oxidation of para-xylene to TPA, the dissolved TPA in the liquid phase of the reaction medium is 7,000 ppmw, the dissolved benzoic acid is 8,000 ppmw, the dissolved isophthalic acid is 6,000 ppmw, dissolved under the given conditions. If the phthalic acid added is 2,000 ppmw, the additional reduction value of the total harmful compound begins to decrease as the reaction activity increases to inhibit the liquid phase concentrations of para-toluic acid and 4-CBA below similar levels. That is, the presence and concentration in the liquid phase of the reaction medium of the species of aromatic substance lacking non-aromatic hydrocarbyl groups are hardly altered by the increase in the reaction activity, the presence of which is due to the presence of intermediates in the reaction to suppress Serves to expand the reduction zone of recovery to reduce the concentration.

Thus, in one embodiment of the present invention, the lower end is caused by low reaction activity and excessive formation of harmful impurities and the upper end is caused by excessive carbon loss, but more than those already found and disclosed as commercially useful. It provides carbon oxides with low levels of preferred ranges. Therefore, the formation of the carbon oxide is preferably controlled as follows. The ratio of total carbon oxides (moles) produced to oxidative aromatic compounds (moles) supplied is preferably greater than about 0.02: 1, more preferably greater than about 0.04: 1, even more preferably greater than about 0.05: 1, And most preferably greater than 0.06: 1. At the same time, the ratio of the total carbon oxide (moles) produced to the supplied oxidative aromatic compound (moles) is preferably less than about 0.24: 1, more preferably less than about 0.22: 1, even more preferably about 0.19: 1 Less than 1, and most preferably less than 0.15: 1. The ratio of the total carbon dioxide (moles) produced to the supplied oxidative aromatic compound (moles) is preferably greater than about 0.01: 1, more preferably greater than about 0.03: 1, even more preferably greater than about 0.04: 1, And most preferably greater than 0.05: 1. At the same time, the ratio of the total carbon dioxide (moles) produced to the supplied oxidative aromatic compound (moles) is preferably less than about 0.21: 1, more preferably less than about 0.19: 1, even more preferably about 0.16: 1 Less than 1, and most preferably less than 0.11: 1. The ratio of the total carbon monoxide (moles) produced to the supplied oxidative aromatic compound (moles) is preferably greater than about 0.005: 1, more preferably greater than about 0.010: 1, even more preferably greater than about 0.015: 1, And most preferably greater than 0.020: 1. At the same time, the ratio of the total carbon monoxide (moles) produced to the supplied oxidative aromatic compound (moles) is preferably less than about 0.09: 1, more preferably less than about 0.07: 1, even more preferably about 0.05: 1. Below, and most preferably less than 0.04: 1.

The amount of carbon dioxide in the dry exhaust gas from the oxidation reactor is preferably greater than about 0.10 mol%, more preferably greater than about 0.20 mol%, even more preferably greater than about 0.25 mol%, and most preferably greater than 0.30 mol%. to be. At the same time, the amount of carbon dioxide in the dry exhaust gas from the oxidation reactor is preferably less than about 1.5 mol%, more preferably less than about 1.2 mol%, even more preferably less than about 0.9 mol%, and most preferably 0.8 mol. Less than%. The amount of carbon monoxide in the dry exhaust gas from the oxidation reactor is preferably greater than about 0.05 mol%, more preferably greater than about 0.10 mol%, even more preferably greater than about 0.15 mol%, and most preferably greater than 0.18 mol%. to be. At the same time, the amount of carbon monoxide in the dry exhaust gas from the oxidation reactor is preferably less than about 0.60 mol%, more preferably less than about 0.50 mol%, even more preferably less than about 0.35 mol%, and most preferably 0.28 mol Less than%.

The inventors believe that an important factor for reducing the production of carbon oxides in this preferred range is the improvement of the purity of the recycled filtrate and the feed of the oxidizing compound, thus reducing the concentration of aromatic compounds deficient in non-aromatic hydrocarbyl groups in accordance with the present disclosure. It has been found that reducing the concentration of aromatic compounds at the same time reduces the formation of carbon oxides and harmful impurities. Another factor is to improve the distribution of para-xylene and oxidant in the reaction vessel according to the present disclosure. Another factor that enables this preferred level of carbon oxide is to operate using gradients of the reaction medium as disclosed herein for pressure, temperature, concentration of oxidizing compounds in the liquid phase and oxidants in the gas phase. Another factor that enables this preferred level of carbon oxide is to operate within the scope of the present disclosure, which is desirable for space-time reaction rates, pressure, temperature, solvent composition, catalyst composition, and mechanical form of the reaction vessel.

An important advantage when operating within the preferred range of carbon oxide formation is that the use of molecular oxygen can be reduced, even if not stoichiometric. In spite of the good stepping of the oxidizing compound and the oxidizing compound according to the invention, when calculated for the feed of the oxidizing compound alone, excess oxygen remains above the stoichiometric value, resulting in partial loss of carbon oxides and excess molecular oxygen Is provided to suppress the formation of harmful impurities. In particular, when xylene is a feed of an oxidizing compound, the feed ratio of molecular oxygen (weight) to xylene (weight) is preferably greater than about 0.91: 1.00, more preferably greater than about 0.95: 1.00, and most preferably Is greater than 0.99: 1.00. At the same time, the feed ratio of molecular oxygen (weight) to xylene (weight) is preferably less than about 1.20: 1.00, more preferably less than about 1.12: 1.00, and most preferably less than about 1.06: 1.00. In particular, for xylene feeds, the time-averaged amount of molecular oxygen in the dry exhaust gas from the oxidation reactor is preferably greater than about 0.1 mol%, more preferably greater than about 1 mol%, and most preferably greater than 1.5 mol%. to be. At the same time, the time-averaged amount of molecular oxygen in the dry exhaust gas from the oxidation reactor is preferably less than about 6 mol%, more preferably less than about 4 mol%, and most preferably less than 3 mol%.

Another important advantage when operating within the preferred range of carbon oxide formation is that less aromatic compounds are converted to carbon oxides and other less valuable forms. This advantage introduces the sum of the moles of aromatic compounds exiting the reaction medium over a continuous period of time (preferably 1 hour, more preferably 1 day, and most preferably 30 days) into the reaction medium. It is evaluated using the value divided by the sum of the number of moles of all aromatic compounds. This ratio is then referred to as the “molar survival ratio” of the aromatic compound through the reaction medium and expressed as a percentage value. If all incoming aromatics are discharged through the reaction medium as aromatics (most often in the oxidized form of the incoming aromatics), the molar survival ratio has a value of up to 100%. If exactly 1 is converted to carbon oxides and / or other non-aromatic compounds (eg, acetic acid) per 100 aromatic compounds introduced while traveling through the reaction medium, the molar survival ratio is 99%. In particular, when xylene is the main feed of the oxidative aromatic compound, the molar survival ratio of the aromatic compound through the reaction medium is preferably greater than about 98%, more preferably greater than about 98.5%, and most preferably 99.0% Is less than. At the same time, and in order that there is sufficient overall reaction activity, when xylene is the main feed of the oxidative aromatic compound, the molar survival ratio of the aromatic compound through the reaction medium is preferably less than about 99.9%, more preferably about 99.8 Less than%, and most preferably less than 99.7%.

Another aspect of the invention relates to the production of methyl acetate in a reaction medium comprising acetic acid and one or more oxidative aromatic compounds. Such methyl acetate is relatively volatile compared to water and acetic acid, so unless additional cooling or other unit operation is used to collect and / or destroy it before it is released back to the surrounding environment, Tends to be processed. Thus, the formation of methyl acetate takes operating and capital costs. Probably, methyl acetate is the methyl radical from acetic acid decomposition combined with oxygen to form methyl hydroperoxide first, followed by decomposition to form methanol, and finally the resulting methanol reacts with the remaining acetic acid to form methyl acetate. By forming. Whatever the chemical route, the inventors have found that if the rate of methyl acetate production is too low, the production of carbon oxides is too low and the production of harmful aromatic impurities is too high. If the production rate of methyl acetate is too high, the production of carbon oxides is also unnecessarily high resulting in yield loss of solvents, oxidizing compounds and oxidants. When using the preferred embodiments described herein, the production ratio of the resulting methyl acetate (mole) to the supplied oxidative aromatic compound (mole) is preferably greater than about 0.005: 1, more preferably greater than about 0.010: 1. And most preferably greater than 0.020: 1. At the same time, the production ratio of the resulting methyl acetate (mole) to the supplied oxidative aromatic compound (mole) is preferably less than about 0.09: 1, more preferably less than about 0.07: 1, even more preferably about 0.05: 1. Below, and most preferably less than 0.04: 1.

While certain embodiments of the invention may be further described by the following examples, these examples are for illustration of the invention only, and are not intended to limit the scope of the invention unless specifically indicated otherwise. Should be understood.

Example  1 to 10

Example 1 is a planned example of a bubble column oxidation reactor for oxidizing para-xylene in a liquid phase in a three phase reaction medium. The bubble column reactor of Example 1 has an industrial design with a para-xylene feed rate of 7,000 kg per hour. Examples 2 to 10 are planned examples of bubble column oxidation reactors operating at 7 times greater performance than the reactor of Example 1. 59 is a table summarizing the different variables of the bubble column oxidation reactor in Examples 1-10.

Example  One

In this example, a bubble column reaction vessel having a vertical cylindrical zone having an inner diameter of 2.44 m was used. The height of the cylindrical zone was 32 m from the lower tangent (TL) to the upper TL of the cylindrical zone. The vessel is equipped with 2: 1 elliptic heads at the top and bottom of the cylindrical zone. The height from the bottom of the reaction medium to the top of the cylindrical zone was about 32.6 m and the overall height of the reaction vessel was about 33.2 m. The operating level was about 25.6 m at the bottom of the reaction medium.

Para-xylene was introduced into the reactor at a constant rate of 7,000 kg per hour. A filtrate solvent containing mainly acetic acid was initially introduced with mixing with para-xylene at a constant rate of 70,000 kg per hour. The feed was dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross-sectional area inside a radius (inner diameter) of 0.45 ^* . The concentration of the catalyst component in the filtrate solvent was determined so that the composition in the liquid phase of the reaction medium was 1,800 ppmw cobalt, 1800 ppmw bromine and 100 ppmw manganese. A separate reflux solvent stream was introduced in droplet form into the gas-separation zone above the operating level of the reaction medium at a constant rate of 49,000 kg per hour and distributed over essentially all cross-sectional areas of the separation zone. There were almost no catalyst components in this reflux solvent. The amount of water combined in the filtrate solvent feed and the reflux solvent feed was such that the concentration of water in the liquid phase of the reaction medium was 6% by weight. The feed rate of air through the oxidant injection distributor similar to that shown in Figures 12-15 is 35,000 kg / hour, and all of the oxidant inlet holes are located below the lower TL of the cylindrical zone. The operating pressure of the reaction vessel overhead gas continued to be 0.52 MPa. The reaction vessel was operated in a substantially adiabatic manner to raise the temperature of the feed into which the heat of reaction was introduced and to evaporate most of the incoming solvent. The operating temperature measured near the middle height of the reaction medium was about 160 ° C. The reaction medium comprising crude terephthalic acid (CTA) was removed at a constant rate from the reaction vessel side at a height of 15 m using an external degassing vessel.

The ratio of L: D was 13.4, and the ratio of H: W was 10.5. The volume occupied by the reaction medium was about 118 m 3 and the reaction vessel contained about 58,000 kg of slurry. The ratio of slurry mass to feed rate of para-xylene was about 8.3 hours. The space time reaction intensity was about 59 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas discharged from the reaction vessel was about 0.12 MPa. The surface velocity of the gas phase at the middle height of the reaction medium was about 0.8 m / sec. The area of the granular surface in contact with the reaction medium was about 197 m 2, which was about 3.16 * Wmin * H. The ratio of the volume of reaction medium to the granular surface area in contact with the reaction medium was about 0.60 m. Under the conditions of Example 1, the decomposition of acetic acid in the reaction medium was estimated to be about 0.03 kg per kg of CTA produced. This is a significant cost in terms of overall production costs.

A useful measure of the end-terminal flow rate in the bubble column reaction vessel is the maximum time-averaged upward velocity on the slurry and oxidant. In the cylindrical bubble column reaction vessel, this maximum upward velocity occurred near the vertical axis of symmetry of the cylindrical zone. Gas-Liquid Recirculation Model of Gupta ( Churn- turbulent bubble columns : experiments and modeling ; Gupta, Puneet; Washington Univ., St. Louis, MO, USA. Avail. UMI, Order No. DA3065044; (2002), 357 pp. From: Diss. Abstr. Int., B 2003, 63 (9), 4269.Dissertation written in English. Using a method derived from CAN 140: 130325 AN 2003: 424187 CAPLUS) and a dedicated database, the maximum time-averaged upward velocity of the gas phase near the median height of the reaction medium was about 3.1 m / sec. Similarly calculated, the maximum time-averaged upward velocity of the slurry near the median height of the reaction medium was about 1.4 m / sec.

Another measure of the end-terminal flow rate in the bubble column reaction vessel is the maximum time-averaged downward velocity of the slurry in the portion of the reaction vessel located away from the central core. In the cylindrical reaction vessel, the maximum downward velocity occurs in an area outside the 0.35 ^* radius (inner diameter) from the vertical axis of symmetry of the cylindrical zone. Using a method derived from the Gupta's gas-liquid recycle model and a dedicated database, the maximum time-averaged downward velocity of the slurry in the outer annulus near the median height of the reaction medium was about 1.4 m / sec.

Example  2

In this example, para-xylene was introduced into the bubble column reactor at an increasing rate of 49,000 kg per hour, seven times larger than Example 1. Surface gas velocities, often considered as important rate increasing parameters for the bubble column, remained about the same as in Example 1 by increasing the cross-sectional area of the reaction vessel about seven times larger than in Example 1. The ratios of H: W and L: D, which are often considered as important rate increasing parameters for the bubble column, also remained about the same as in Example 1.

Another feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 40 m using an external degassing vessel.

The bubble column reaction vessel contained a vertical cylindrical zone with an internal diameter of 6.46 m. The ratio of L: D remained the same as in Example 1, and the height from the bottom of the reaction medium to the top of the cylindrical zone was therefore 86.3 m. The top and bottom of the cylindrical zone were equipped with 2: 1 elliptic heads and the overall height of the reaction vessel was about 88.0 m. The feed was also dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross section inside a radius (inner diameter) of 0.45 ^* . Air was also supplied through an oxidant injection distributor similar to that shown in FIGS. 12-15, with all of the oxidant inlet holes located below the lower TL of the cylindrical zone. The reflux solvent was dispensed as droplets over essentially all cross sections of the separation zone.

The ratio of H: W of the reaction medium remained about the same as in Example 1, and therefore the operating level was about 67.8 m of reaction medium. For this reason, the separation height is the sum of the heights of about 18.5 m in the cylindrical zone and about 1.6 m in the upper oval head. This separation height was exceeded by about 10 meters. Thus, adjusting the vessel size to a constant L: D puts an excessive cost on the mechanical plant (e.g. pressure vessels, stiffeners from material and wind loads, structural steel, process and plant piping and excessive for equipment and electrical cables). cost).

The volume occupied by the reaction medium was about 2,200 m 3 and the reaction vessel contained about 1,100,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 22 hours, a significant increase over Example 1. The space time reaction intensity was significantly reduced compared to Example 1, with about 22 kg of para-xylene supplied per 1 m 3 of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel rose to about 0.33 MPa. The granular surface area in contact with the reaction medium was about 1,393 m 2, which was about 3.18 * Wmin * H. The ratio of the volume of the reaction medium to the granular surface area in contact with the reaction medium is about 1.58 m, which is a significant increase compared to Example 1.

Thus, adjusting the dimensions of the reaction medium to keep both surface velocity and the ratio of H: W approximately constant between this example and Example 1 resulted in very large changes in reaction conditions. After all, this change is very undesirable. This example had the positive effect of producing less undesirable colored byproducts per unit CTA (e.g., more on the concentration of more diluted oxidizing compounds, the mass transfer rate per unit volume of molecular oxygen from the gas phase to the liquid phase). Low requirements etc.). However, there has been a severe economic loss related to the amount of pressure and power required for acetic acid cracking and air supply to the bottom of the bubble column reaction vessel. The decomposition of acetic acid is approximately proportional to the mass of acetic acid in the reaction medium when the operating temperature and the composition of the liquid phase are kept approximately constant, and when para-xylene is fed with excess molecular oxygen. Since the mass of acetic acid in the reaction vessel is much larger than the amount of CTA produced in this example, it was estimated that the decomposition of acetic acid was increased to about 0.09 kg per kg of CTA produced. In addition, in this embodiment, the air compressor must be used to deliver air to the reaction medium at a pressure of 0.85 MPa, while in Example 1 the delivery pressure was 0.64 MPa. Depending on the typical air demand of 245,000 kg / hour air delivery rate and the various compression and transfer efficiencies, the additional power requirement for the higher transfer pressure in this example was still about 3,000 kilowatts. Thus, adjusting the reaction medium of this example to approximately constant surface gas velocity and H: W ratio is economically unacceptable despite the good quality of the CTA expected.

Example  3

This example scales the process of Example 1 using surface velocity and space-time response intensity. In short, natural convective flow patterns result in poor product quality because they naturally form a vertically inferior reaction profile.

In this example, the feed rate of para-xylene was about 49,000 kg per hour, seven times greater than Example 1. Surface gas velocities remained about the same as in Example 1, but the ratios of L: D and H: W were not the same. Instead, the STR remained about the same as in Example 1. This gives approximately the same column base pressure and decomposition ratio of acetic acid as in Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was 0.52 MPa and the operating temperature was about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 15 m using an external degassing vessel.

The bubble column reaction vessel includes a vertical cylindrical zone having an internal diameter of 6.46 m, with approximately constant gas phase surface velocity compared to Examples 1 and 2. To maintain the same STR as in Example 1, the operating level was slightly changed to about 26.1 m of reaction medium. The height from the lower TL to the upper TL of the cylindrical zone was 32 m, the same as in Example 1, which gave approximately the same clearance separation height between the upper and over gas outlets of the reaction medium. At the top and bottom of the cylindrical section are mounted 2: 1 ellipsoid heads. The height from the bottom of the reaction medium to the top of the cylindrical zone was about 33.6 m and the overall height of the reaction vessel was about 35.2 m. The feed was also dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross section inside a radius (inner diameter) of 0.45 ^* . Air was also supplied through an oxidant injection distributor similar to that shown in FIGS. 12-15, with all of the oxidant inlet holes located below the lower TL of the cylindrical zone. The reflux solvent was partitioned in droplet form over essentially all cross sections of the separation zone.

The ratio of H: W in the reaction medium was greatly reduced to 4.0. L: D of the reaction vessel was greatly reduced to 5.2. The volume occupied by the reaction medium was about 828 m 3 and the reaction vessel contained about 410,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 8.3 hours. The STR was about 59 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.13 MPa. The granular surface area in contact with the reaction medium was about 546 m 2, which was about 3.24 * Wmin * H. The ratio of the volume of reaction medium to the granular surface area in contact with the reaction medium was about 1.52 m. Under the conditions of this example, the decomposition of acetic acid in the reaction medium was preferably restored to the lower level of Example 1, which was about 0.03 kg per kg of CTA produced.

However, in combination with the smaller H: W ratio, the larger diameter of the reaction vessel used in this example resulted in very undesirable changes in flow rate, mixing and staging in the reaction medium. This significantly increased the loss of para-xylene due to the formation of overhead exhaust gases and undesirable colored byproducts. In short, even if the surface velocity remains constant, the larger axial flow rates generated by natural convection forces in the larger diameter bubble column. Using the method derived from Gupta's gas-liquid recycle model and a dedicated database, the maximum time-average upstream velocity of the gas phase near the median height of the reaction medium was about 3.9 m / sec. Similarly, the maximum time-averaged upward velocity of the slurry near the median height of the reaction medium was about 2.2 m / sec. Similarly, the maximum time-averaged downward velocity of the slurry in the outer annulus near the median height of the reaction medium was about 2.3 m / sec.

 Since the height of the reaction medium hardly changed between this example and Example 1, this increased time-average vertical velocity greatly reduced the end-to-end mixing time compared to Example 1. This increased the amount of para-xylene moving towards the top of the vessel prior to oxidation, which was undesirable. This results in an undesirable loss of the yield of para-xylene exiting the top of the reaction vessel with the exhaust gas, and the molecular oxygen dissolved in the nearer top of the reactor where the mole fraction of molecular oxygen is relatively diminished in the gas phase. More movement was required. Moreover, in this example the increased time-averaged downward velocity of the slurry in the region towards the vessel wall caused more and larger vapor bubbles to pull down against natural buoyancy in the gravitational field. This undesirably increased the recycle of the gaseous phase in which the molecular oxygen was partially extinguished, thus reducing the usefulness of dissolved oxygen in this region. Among other things, the reduced availability of dissolved oxygen in the various parts of the reaction medium greatly increases the formation ratio of undesirable colored byproducts in this example, compared to Example 1, Elevated levels prevented the use of PET in many applications.

Thus, Examples 2 and 3 illustrate the inadequacy of the prior art of designing large-scale oxidized bubble towers using primarily surface gas velocities (Ug), L: D ratios, and average space-time velocities (STR) of the reaction. .

Example  4

In this example, the pressure-containing member of the bubble column reaction vessel is the same as in Example 3, but gives a vertical drag by adding a granular surface in the reaction medium to establish a reaction stage profile that is more similar to Example 1 Thereby, the quality of the product and the yield of para-xylene were recovered, but there was no increase in the decomposition ratio of acetic acid as in Example 2.

The feed rate of para-xylene was also about 49,000 kg per hour, seven times larger than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was about 160 ° C. measured near the middle height of the reaction medium. The reaction medium containing CTA was removed at a constant rate from the side of the reaction vessel at a height of 15 m using an external degassing vessel.

The bubble column reaction vessel contains a vertical cylindrical zone having an internal diameter of 6.46 m. The height from the lower TL to the upper TL of the cylindrical zone was 32 m and the operating level was about 26.3 m of reaction medium. At the top and bottom of the cylindrical section are mounted 2: 1 ellipsoid heads. The height from the bottom of the reaction medium to the top of the cylindrical zone was about 33.6 m and the overall height of the reaction vessel was about 35.2 m. The feed was also dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross section inside a radius (inner diameter) of 0.45 ^* . Air was also supplied through an oxidant injection distributor similar to that shown in FIGS. 12-15, with all of the oxidant inlet holes located below the lower TL of the cylindrical zone. The reflux solvent was partitioned in droplet form over essentially all cross sections of the separation zone.

The bubble column reaction vessel further included two orthogonal planar surfaces located with cross lines coinciding with the vertical axis of symmetry of the cylindrical zone. These planar surfaces conveniently comprise metal plates having the same type and surface finish as those used in the cylindrical zone of the reaction vessel. Each planar surface extends above 20 meters starting at a low height of 3 meters above the lower TL. Each of the two planar surfaces extends essentially horizontally in all directions to the walls of the cylindrical zone (ie each width is equal to the inner diameter) and bears from the cylindrical zone. The thickness and bearing of the planar surface are designed to withstand the various forces that can occur under normal and conducted operating conditions. Because of the volume occupied by the metal plate, the operating level was adjusted slightly upward to 26.3 m at the bottom of the reaction medium to keep the STR the same as Example 1.

Thus, in this example, the reaction medium was formed into small volumes of 20 m divided into four equal sizes from the total height of the reaction medium. These four small volumes are interchanged with each other both below and above the surface. Because of the relatively uniform distribution of oxidant and para-xylene below the lower end of the planar surface, each of the four small volumes has similar gas phase surface velocities and similar reaction intensity profiles. Four small volumes can be thought of as analogous to four smaller size bubble column reaction vessels within the shell of one pressure-containing vessel.

The ratio H: W of the reaction medium was 4.1. L: D of the reaction vessel was 5.2. The volume occupied by the reaction medium was about 828 m 3 and the reaction vessel contained about 410,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 8.3 hours. The STR was about 59 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.13 MPa. The granular surface area in contact with the reaction medium was about 1,066 m 2, which was about 6.29 * Wmin * H. The ratio of the volume of reaction medium to the granular surface area in contact with the reaction medium was about 0.78 m. This value is an intermediate value between Example 1 and Example 3, which is usefully closer to Example 1. Under the conditions of this example, the decomposition of acetic acid in the reaction medium was preferably restored to the lower level of Example 1, which was about 0.03 kg per kg of CTA produced.

The maximum time-average up and down rates of the gas phase and slurry were reduced in this example compared to Example 3. This provides a useful improvement in the vertical profile of para-xylene and improves the usefulness of dissolved oxygen in the liquid phase near the granular wall surface. In addition, these changes resulted in an improved yield of para-xylene and reduced formation of undesirable colored byproducts in this Example compared to Example 3.

Example  5

In this example, the pressure-containing member of the reaction vessel is the same as in Example 3, but by adding a non-polluting baffle member in the reaction medium to reset the reaction staged profile more similarly to Example 1, whereby The quality and yield of para-xylene were recovered, but there was no increase in the decomposition ratio of acetic acid as in Example 2.

The feed rate of para-xylene was about 49,000 kg per hour, seven times greater than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was 0.52 MPa and the operating temperature was also about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 15 m using an external degassing vessel.

The bubble column reaction vessel contains a vertical cylindrical zone having an internal diameter of 6.46 m. The height from the lower TL to the upper TL of the cylindrical zone was 32 m and the operating level was about 26.1 m of reaction medium. At the top and bottom of the cylindrical section are mounted 2: 1 ellipsoid heads. The height from the bottom of the reaction medium to the top of the cylindrical zone was about 33.6 m and the overall height of the reaction vessel was about 35.2 m. The feed was also dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross section inside a radius (inner diameter) of 0.45 ^* . Air was also supplied through an oxidant injection distributor similar to that shown in FIGS. 12-15, with all of the oxidant inlet holes located below the lower TL of the cylindrical zone. The reflux solvent was partitioned in droplet form over essentially all cross sections of the separation zone.

The bubble column reaction vessel further comprises a horizontal baffle assembly positioned in the bubble column at a height of 12 meters above the lower TL of the reaction vessel. The baffle assembly is located about 13.6 m or 2.1 * D at the bottom of the reaction medium. The baffle assembly consisted of fifteen respective baffle members. Each baffle member consisted of an extruded or assembled L-shape, wherein the two L-shapes had a width of 0.15 m and the interior angle between the two legs was 90 °. The L-shapes are all arranged horizontally parallel to each other, with the corners pointing upwards and located at the same height. The two distal edges of each L-shape are all under a corner pointing up at the same height. Seen at one end, each member appears as an inverted V-shaped. Thus, a baffle assembly that includes an upwardly facing planar surface that slopes below 5 ° from horizontal is not effective. The gap between the lower edge of each member and the edge of the adjacent member thereof was 0.21 m. The longest member has a length substantially the same as the inner diameter of the cylindrical vessel and extends across the vessel in the radial direction from wall to wall. The other 14 individual baffle members all naturally had shorter lengths. Both baffle members are supported at each end by extending to and attaching to the cylinder wall in all directions. Thus, the open area at the height of the baffle assembly was about 16 m 2, which corresponds to about 50% of the cross-sectional area of the reaction vessel at this height. The baffle member is designed to withstand the various forces that can occur under normal and varied operating conditions. The member consisted of the same metal as the piping component used in the air sparger assembly, with the metal being properly selected to resist corrosion and erosion. However, the surface of the baffle member was polished to 125 RMS or finer surface finish. Despite the precipitation of about 76,000 kg of CTA per hour in the reaction vessel, the baffle assembly was not excessively contaminated or debris of solid mass was falling.

The ratio of H: W in the reaction medium was 4.0. L: D of the reaction vessel was 5.2. The volume occupied by the reaction medium was about 828 m 3 and the reaction vessel contained about 410,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 8.3 hours. The STR was about 59 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.13 MPa. The granular surface area in contact with the reaction medium was about 546 m 2, which was about 3.24 * Wmin * H. The ratio of the volume of reaction medium to the granular surface area in contact with the reaction medium was about 1.52 m. Under the conditions of this example, the decomposition of acetic acid in the reaction medium was preferably restored to the lower level of Example 1, which was estimated to be about 0.03 kg per kg of CTA produced.

The effect of the horizontal baffle assembly is to hinder the vertical velocity of the gas phase and slurry in the reaction vessel. This has the advantage that the progression of para-xylene towards the upper surface of the reaction medium is delayed, reducing the yield loss of para-xylene in the overhead exhaust gas. In addition, the formation of undesirable colored byproducts in this example was reduced in comparison with Example 3 due to the stepping of the molecular oxygen and oxidizing compounds.

Example  6

In this embodiment, the reaction vessel is designed according to the invention to have a very high gas surface velocity and gas retention. By using a smaller D, higher L: D is possible without using an excessively high reaction vessel and without excessive decomposition of the acetic acid solvent.

The feed rate of para-xylene was about 49,000 kg per hour, seven times greater than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was also about 160 ° C. measured near the middle height of the reaction medium. The reaction medium containing CTA was removed at a constant rate from the side of the reaction vessel at a height of 15 m using an external degassing vessel.

The bubble column reaction vessel contains a vertical cylindrical zone having an inner diameter of 5.00 m. The height from the bottom TL to the top TL of the cylindrical zone was 70 m. At the top and bottom of the cylindrical section are mounted 2: 1 ellipsoid heads. The height from the bottom of the reaction medium to the top of the cylindrical zone was about 71.3 m and the overall height of the reaction vessel was about 72.5 m. The operating level was about 61.3 m at the bottom of the reaction medium. The feed was dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross-sectional area inside a radius (inner diameter) of 0.45 ^* . Air was also supplied through an oxidant injection distributor similar to that shown in FIGS. 12-15, with all of the oxidant inlet holes located below the lower TL of the cylindrical zone. The reflux solvent was partitioned in droplet form over essentially all cross sections of the separation zone.

The ratio H: W of the reaction medium was 12.3. L: D of the reaction vessel was 14.3. The volume occupied by the reaction medium was about 1,190 m 3 and the reaction vessel contained about 420,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 8.7 hours. The STR was about 41 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.21 MPa. The granular surface area in contact with the reaction medium was about 975 m 2, which was about 3.18 * Wmin * H. The ratio of the volume of reaction medium to the granular surface area in contact with the reaction medium was about 1.22 m. The relatively small D value resulted in the surface velocity of the gas phase at the intermediate height of the reaction medium which is about 1.7 times the surface velocity used in Examples 1-5. The gas retention at the median height of the reaction medium was greater than 0.6. Under the conditions of Example 6, it was evaluated that the decomposition of acetic acid in the reaction medium is preferably reduced to less than about 0.03 kg per kg of CTA produced. This is due to the reduced amount of slurry in the reaction vessel (more particularly the amount of acetic acid) compared to Example 3.

In this example, the ratio of H: W favored reduced end-end mixing and beneficial stepping of molecular oxygen and oxidative compounds. However, the axial velocity was higher than in Example 1, accelerating end-end mixing for a given H: W. Advantageously, the low STR reduced the volume requirement for the transfer of molecular oxygen from gas to liquid; Increased gas retention improved the performance for the transfer of molecular oxygen from gas to liquid. In conclusion, the production level of undesirable colored by-products was evaluated to be similar to Example 1.

Example  7

In this example, the reaction vessel was designed to have a higher gas surface velocity and gas retention in accordance with the present invention. Extended and installed separation zones were used to limit the capture of slurry at overhead.

The feed rate of para-xylene was also about 49,000 kg per hour, seven times larger than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was also about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 28 m using an external degassing vessel.

The bubble column reaction vessel includes a vertical cylindrical zone having an internal diameter of 4.60 m. The height from the bottom TL to the top TL of the cylindrical zone was 60 m. At the upper end of the cylindrical zone, the conical zone diverged to an inner diameter of 7 meters while rising to a height of 2 meters. Thus, the slope of the conical wall was about 31 ° from vertical. On top of the conical section a gas separation cylindrical section with an internal diameter of 7 m was installed. The height of the upper cylindrical zone was 7 m. The vessel is equipped with a 2: 1 ellipse head at the top and bottom. Thus, the combined height of the reaction vessels was about 71.9 m. The operating level was about 61.2 m at the bottom of the reaction medium and was located near the binding site of the cylinder main body and the diverging conical zone. The feed was dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross-sectional area inside a radius (inner diameter) of 0.45 ^* . Air was also supplied through an oxidant injection distributor similar to that shown in FIGS. 12-15, with all of the oxidant inlet holes located below the bottom TL of the lower cylindrical zone. The reflux solvent was dispensed in droplet form over essentially all cross sectional areas of the enlarged separation zone.

The ratio H: W of the reaction medium and L: D of the reaction vessel were 13.3. The ratio of X: D of the reaction vessel was 1.5. The ratio of L: Y of the reaction vessel was 5.7. The volume occupied by the reaction medium was about 1,000 m 3 and the reaction vessel contained about 320,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 6.5 hours. The STR was about 49 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.19 MPa. The granular surface area in contact with the reaction medium was about 896 m 2, which was about 3.19 * Wmin * H. The ratio of the volume of the reaction medium to the granular surface area in contact with the reaction medium was about 1.12 m. The relatively small D value resulted in the surface velocity of the gas phase at the intermediate height of the reaction medium which is about twice the surface velocity used in Examples 1-5. However, the surface velocity of the gas phase in the enlarged separation zone was reduced to 0.85 times the surface velocity used in Examples 1-5. The gas retention at the median height of the reaction medium was about 0.7. Under the conditions of this example, it was estimated that the decomposition of acetic acid in the reaction medium is preferably reduced to less than about 0.03 kg per kg of CTA produced. This is due to the reduced amount of slurry in the reaction vessel (more particularly the amount of acetic acid) compared to Example 3. It was estimated that the level of undesirable colored byproducts was lower than in Example 6 due to the improved staged and high gas retention.

Example  8

In this example, the reaction vessel was the same as in Example 7, but the operating level was raised to about 63.2 m at the bottom of the reaction medium and located near the binding site of the conical zone and the enlarged gas separation cylindrical zone. This provides several advantages over controlling the level in the cylindrical main body section, including a reduction in the tendency for the top of the reaction medium to become foamy and circulate very poorly.

The feed rate of para-xylene was about 49,000 kg per hour, seven times greater than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was also about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 28 m using an external degassing vessel.

The ratio H: W of the reaction medium was about 13.7, and the ratio L: D of the reaction vessel was 13.3. The volume occupied by the reaction medium was about 1,060 m 3 and the reaction vessel contained about 330,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 6.8 hours. The STR was about 46 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.20 MPa. The granular surface area in contact with the reaction medium was about 953 m 2, which was about 3.39 * Wmin * H. The ratio of the volume of reaction medium to the granular surface area in contact with the reaction medium was about 1.11 m. The surface velocity of the gas phase at the median height of the reaction medium was about twice the surface velocity used in Examples 1-5. The gas retention at the median height of the reaction medium was about 0.7. Under the conditions of this example, it was estimated that the decomposition of acetic acid in the reaction medium preferably decreased to less than about 0.03 kg per kg of CTA produced. This is due to the reduced amount of slurry in the reaction vessel (more particularly the amount of acetic acid) compared to Example 3. It was estimated that the level of by-products of undesirable coloring was lower than in Example 6 because of the improved stepping and high gas retention.

Example  9

In this embodiment, the pressure-containing member of the reaction vessel was the same as in Example 7, but the inner member used to introduce the oxidant and para-xylene was changed to have a plurality of inlets, each of which was importantly vertically separated.

The feed rate of para-xylene was about 49,000 kg per hour, seven times greater than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 28 m using an external degassing vessel.

The bubble column reaction vessel contained the same oxidant distributor in the bottom head of the reaction vessel as in Example 8. However, only 70% of the total gaseous oxidant stream was introduced through the bottom distributor. The remaining 30% of the gaseous oxidant was introduced through the high position oxidant injection distributor. This flow ratio was imparted by a flow control loop using a control valve and a flow transmitter conveniently disposed on the external compressed air supply conduit of the reaction vessel. The high position oxidant distributor included horizontal mitered rectangular flow conduits rather than the octagonal conduits used for the low oval heads. The rectangular flow conduit included a nominal 14 inch Schedule 10S piping material. The distance from the center of one side to the center of the other side was 1 m. The high position oxidant distributor includes about 60 discharge holes for the gaseous oxidant, all of which have a diameter of 0.03 m and are located near the bottom of the conduit of about 14 m above the bottom of the reaction medium. This high position oxidant distributor has two or more useful functions. First, the oxidant flow injected down into the reaction medium interferes with the rising axial velocity profile along the vertical axis of symmetry of the cylindrical zone. This confers a hydraulic baffle function that delays the diffusion of para-xylene into the upper region of the reaction medium, which is associated with overhead yield loss and reduced demand for dissolved oxygen in the upper region. At several meters above the high oxidant inlet, the natural convective flow pattern is reconstructed to raise itself along the central axis of symmetry, but hydraulic baffle is nevertheless valid. Second, most of the heat of reaction is removed from the reaction medium through evaporation of the solvent, and most of this evaporation occurs near the oxidant feed point. The vertical profile of the temperature in the reaction medium is adjusted by vertically separating the introduction position of the gaseous oxidant stream portion.

The bubble column reaction vessel contained two para-xylene injection distributors similar to Example 8. A lower para-xylene injection distributor was placed to release nearly 50% of the liquid feed introduced over the cross-sectional area inside the 0.45 ^* (inner diameter) radius at a low location of 2m above the bottom of the reaction medium. An upper para-xylene injection distributor was placed to release 50% of the liquid feed supplied over a cross-sectional area inside a radius of 0.45 ^* (inner diameter) at a high height of 10m above the bottom of the reaction medium. Its flow ratio was imparted by a flow control loop using a control valve and a flow distributor conveniently disposed on the supply conduits for the liquid feed outside the reaction vessel.

In this example, the operating level rose to about 63.7 m at the bottom of the reaction medium and was located in the enlarged gas separation cylindrical zone just above the diverging conical zone. The ratio of H: W of the reaction medium was about 13.8, and the ratio of L: D of the reaction vessel was 13.3. The volume occupied by the reaction medium was about 1,070 m 3 and the reaction vessel contained about 340,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 6.9 hours. The STR was about 46 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.20 MPa. The granular surface area in contact with the reaction medium was about 975 m 2, which was about 3.47 * Wmin * H. The ratio of the volume of the reaction medium to the granular surface area in contact with the reaction medium was about 1.10 m. The surface velocity of the gas phase at the median height of the reaction medium was about twice the surface velocity used in Examples 1-5.

Example  10

In this example, the reaction vessel was designed to have three different cylinder diameters at different heights, with the smallest diameter at the middle height. This configuration has the advantage of providing a relatively large liquid phase for the initial dilution and reaction of the oxygen-rich para-xylene in the lower cylindrical zone where the liquid feed stream and the gaseous oxidant are first introduced; In the intermediate cylindrical zone where the consumption of molecular oxygen is increased, it offers the advantage of providing a relatively higher gas retention and rate of mass transfer of the gas into the liquid; In the upper cylindrical zone, which is the gas separation zone, the advantage is a relatively reduced gaseous velocities that limit the capture of slurry in the overhead exhaust gas.

The feed rate of para-xylene was about 49,000 kg per hour, seven times greater than Example 1. The other feed flow was increased to a ratio of 7: 1 as above for Example 1. The composition of the feed was the same as in Example 1, giving the same concentrations of water, cobalt, bromine and manganese in the liquid phase of the reaction medium as in Example 1. The operating pressure of the reaction vessel overhead gas was also 0.52 MPa and the operating temperature was about 160 ° C. measured near the middle height of the reaction medium. The reaction medium comprising CTA was removed at a constant rate from the side of the reaction vessel at a height of 28 m using an external degassing vessel.

The bubble column contained three vertical cylindrical zones with different diameters. The lowest cylindrical zone had an internal diameter of 6.46 and the gas phase surface velocity in this zone was about the same as the surface velocity of Example 1. The height from the bottom TL to the top TL of the lower cylindrical zone was 8 m. At the upper end of the lower cylindrical zone, the conical zone diverged to an inner diameter of 4.5m, rising to a height of 1m. Thus, the slope of this conical wall was about 44 ° from vertical. The intermediate cylindrical zone above the lower conical zone has an inner diameter of 4.5 m, in which the gas phase surface speed is twice the surface speed of the lowest cylindrical zone. The height of the middle cylindrical zone was 45 m. At the upper end of the middle cylindrical zone, the conical zone diverged to an internal diameter of 7 meters, rising to a height of 2 meters. The slope of this conical wall was about 32 ° from vertical. Above the upper conical diverging zone is a gas separation cylindrical zone with an internal diameter of 7 m. The height of the upper cylindrical zone was 7 m. The vessel is equipped with a 2: 1 ellipse head at its top and bottom. Therefore, the combined height of the reaction vessels was about 66.4 m. The operating level was about 57.6 m at the bottom of the reaction medium and was located near the junction of the diverging conical zone and the upper cylindrical zone. The feed was also dispensed into the reaction vessel near a height of 2 m above the bottom of the reaction medium using a horizontal distributor assembly designed to release the feed substantially uniformly over a cross section inside a radius (inner diameter) of 0.45 ^* . Air was supplied through an oxidant injection distributor similar to that shown in Figures 12-15, with all of the oxidant inlet holes located below the lower TL of the cylindrical zone in the lowest position. The reflux solvent was dispensed in droplet form over essentially all cross sectional areas of the upper cylindrical zone.

The volume occupied by the reaction medium was about 1,080 m 3 and the reaction vessel contained about 400,000 kg of slurry. The ratio of slurry mass to para-xylene feed rate was about 8.1 hours. The STR was about 45 kg para-xylene fed per cubic meter of reaction medium per hour. The pressure difference from the bottom of the reaction medium to the overhead exhaust gas exiting the reaction vessel was about 0.20 MPa. The granular surface area in contact with the reaction medium was about 944 m 2, which was about 3.64 * Wmin * H and 2.34 * Wmax * H. The ratio of the volume of the reaction medium to the granular surface area in contact with the reaction medium was about 1.14 m. The ratio of L _l : D _l was about 1.5: 1. The ratio of L _u : D _u was about 10: 1. The ratio of L _l : L _u was about 0.21: 1. The ratio of X: D _l was about 1.1: 1. The ratio of L _u : Y was about 4.2: 1. The ratio of L _t : D _l was about 0.15: 1.

Due to the larger diameter at the base of the reactor, a large amount of slurry is provided near the introduction zone of the feed para-xylene, where a diluent of the feed is provided to mix with the liquid flow to avoid bound and colored aromatic impurities. Is very important. This larger diameter also places a larger amount of reaction medium under a greater head pressure from the slurry to facilitate mass transfer of oxygen partial pressure and molecular oxygen from the gas to the liquid. Rectangles of smaller diameter, middle cylindrical zones provide reactant stages and high retention rates, thereby improving the conformity of the mass transfer feed from the rising gas phase with the reaction demands of dissolved oxygen, resulting in continued consumption of oxygen And its partial pressure is reduced.

Although the present invention has been described in detail with particular reference to preferred embodiments, it will be understood that various changes and modifications may be made within the spirit and scope of the invention.

Claims (51)

A method comprising oxidizing an oxidizing compound in a liquid phase of a multiphase reaction medium contained in a reaction zone of an initial oxidation reactor, When the total volume of the reaction medium is theoretically divided into 2,000 separate horizontal slices of the same volume, the oxidation is carried out in such a way that less than 120 horizontal slices have a gas retention of less than 0.3 on a time-averaged and volume-averaged basis. How to be. The method of claim 1, The total volume of the reaction medium has a gas retention of at least about 0.4 on a time-averaged and volume-averaged basis. The method of claim 1, Less than 80 horizontal partitions have a gas retention of less than about 0.3 on a time-averaged and volume-averaged basis. The method of claim 1, Less than 40 horizontal layers have a gas retention of less than about 0.3 on a time-average and volume-average basis, and the total volume of reaction medium has a gas retention in the range of about 0.6 to about 0.9 on a time-average and volume-average basis. How to have. The method of claim 1, The initial oxidation reactor is a shake reactor. The method of claim 1, The initial oxidation reactor is a bubble column reactor. The method of claim 6, Degassing a portion of the reaction medium in a degassing zone separate from the reaction zone to provide a substantially degassed slurry comprising less than about 5% by volume of gas, the degassing in the solid and liquid phases of the reaction medium. Method mainly caused by the natural buoyancy of the gas phase of the reaction medium. The method of claim 7, wherein The degassing zone defines between one or more granular sidewalls of the degassing vessel, wherein the maximum horizontal cross-sectional area of the degassing zone is less than about 25% of the maximum horizontal cross-sectional area of the reaction zone. The method of claim 8, Wherein at least a portion of the degassing zone is located in the reactor. The method of claim 8, Wherein both degassing zones are located outside of the reactor. The method of claim 6, The reaction medium has a maximum height (H), a maximum width (W), and an H: W ratio of greater than about 3: 1. The method of claim 11, The H: W ratio ranges from about 8: 1 to about 20: 1. The method of claim 11, Introducing a predominantly gaseous oxidant stream comprising molecular oxygen into the reaction zone, wherein a majority of the molecular oxygen enters the reaction zone within about 0.25 W of the bottom of the reaction zone. The method of claim 13, Wherein most of the molecular oxygen enters the reaction zone within about 0.2 W and about 0.02 H at the bottom of the reaction zone. The method of claim 13, When the reaction zone is theoretically divided into four vertical quadrants of equal volume by a pair of intersecting vertical planes, molecular oxygen is introduced in such a way that less than about 80% by weight of molecular oxygen flows into one of the vertical quadrants in the reaction zone. Entering the reaction zone. The method of claim 13, At least a portion of the reaction zone is defined by at least one granular sidewall of the reactor, at least about 25% by weight of molecular oxygen enters the reaction zone at one or more locations spaced inward at least 0.05D from the granular sidewall and the reaction zone is at most Method with diameter (D). The method of claim 13, A first portion of the oxidant stream is introduced into the reaction zone through the one or more lower oxidant openings, a second portion of the oxidant stream is introduced into the reaction zone through the one or more lower oxidant openings and the upper oxidant opening is at least 1 W above the lower oxidant opening. How to be located. The method of claim 17, Wherein each of the first and second portions of the oxidant stream contains less than about 50 mole percent molecular oxygen. The method of claim 17, Wherein at least about 10 mole percent of the total amount of molecular oxygen introduced to the reaction zone is introduced through the lower oxidant opening, and at least about 10 mole percent of the total amount of molecular oxygen introduced to the reaction zone is introduced through the upper oxidant opening. The method of claim 17, Wherein the upper oxidant opening is at least about 2D above the lower oxidant opening, wherein each of the first and second portions of the oxidant stream contains less than 40 mole percent molecular oxygen. The method of claim 1, The oxidizable compound is an aromatic compound. The method of claim 1, The oxidizing compound is para-xylene. The method of claim 1, At least about 10% by weight of the oxidizable compound is oxidized to form a solid in the reaction medium. The method of claim 1, Wherein the oxidation is carried out in the presence of a catalyst system comprising cobalt. The method of claim 24, The catalyst system further comprises bromine and manganese. The method of claim 1, Terephthalic acid is formed in the reaction medium by oxidation in an initial oxidation reactor, and further comprising oxidizing at least a portion of terephthalic acid in a second oxidation reactor. The method of claim 26, Wherein the oxidation in the second oxidation reactor is carried out at an average temperature of at least about 10 ° C. higher than the oxidation of the initial oxidation reactor. The method of claim 26, Oxidation in the second oxidation reactor is carried out at an average temperature of about 20 ° C. to about 80 ° C. higher than the average temperature of the initial oxidation reactor, and oxidation in the initial oxidation reactor is carried out at an average temperature in the range of about 140 ° C. to about 180 ° C. And oxidation in the second oxidation reactor is carried out at an average temperature in the range of about 180 ° C to about 220 ° C. The method of claim 1, Oxidation forms crude terephthalic acid particles in the reaction medium, A representative sample of the crude terephthalic acid particles contains (i) less than about 12 ppmw of 4,4-dicarboxy cistilbene (4,4-DCS), and (ii) contains less than about 800 ppmw of isophthalic acid (IPA), (iii) contains less than about 100 ppmw 2,6-dicarboxyfluorenone (2,6-DCF), and (iv) has one or more of features having a transmittance (% T ₃₄₀ ) at ₃₄₀ nm greater than about 25 Way. (a) introducing a feed stream comprising para-xylene into the reaction zone of the bubble column reactor; (b) introducing an oxidant stream comprising molecular oxygen into said reaction zone, said reaction zone having a maximum diameter (D), with the majority of molecular oxygen entering said reaction zone within about 0.25D of the bottom of said reaction zone; Becoming; And (c) oxidizing at least a portion of the para-xylene in the liquid phase of the three-phase reaction medium contained in the reaction zone, wherein at least about 10% by weight of para-xylene is oxidized to form crude terephthalic acid particles in the reaction medium. Method comprising the steps of: The method of claim 30, Wherein the reaction medium has a maximum height (H) and most of the molecular oxygen enters the reaction zone within about 0.025H of the bottom of the reaction zone. The method of claim 31, wherein Wherein most of the molecular oxygen enters the reaction zone within about 0.2D and about 0.02H of the bottom of the reaction zone. The method of claim 31, wherein Wherein most of the molecular oxygen enters the reaction zone within about 0.15D and about 0.015H at the bottom of the reaction zone. The method of claim 30, Degassing a portion of the reaction medium in a degassing zone separate from the reaction zone to provide a substantially degassed slurry comprising less than about 5% by volume of gas, the degassing being in the solid phase and Method mainly caused by the natural buoyancy of the gas phase in a three-phase reaction medium in the liquid phase. The method of claim 34, wherein The degassing zone defines between one or more granular sidewalls of the degassing vessel, wherein the maximum horizontal cross-sectional area of the degassing zone is less than about 25% of the maximum horizontal cross-sectional area of the reaction zone. The method of claim 30, The reaction medium comprises from about 5 to about 40 weight percent solids on a time-averaged and volume-averaged basis. The method of claim 30, When the total volume of the reaction medium is theoretically divided into 2,000 separate horizontal slices of the same volume, the oxidation is carried out in such a way that less than 120 horizontal slices have a gas retention of less than 0.3 on a time-averaged and volume-averaged basis. How to be. The method of claim 30, Oxidizing at least a portion of the crude terephthalic acid particles in a second oxidation reactor. The method of claim 30, Representative samples of crude terephthalic acid particles (i) contain less than about 12 ppmw of 4,4-dicarboxylstilbene (4,4-DCS), (ii) contain less than about 800 ppmw of isophthalic acid (IPA), and iii) containing less than about 100 ppmw 2,6-dicarboxyfluorenone (2,6-DCF), and (iv) having at least one of the features having a transmittance (% T ₃₄₀ ) at ₃₄₀ nm greater than about 25 . A bubble column reactor for predominantly reacting a liquid stream and a gas phase stream, A reaction extending with a central axis, having a typical lower end and a conventional upper end spaced from each other by a maximum axis distance (L), having a maximum diameter (D), and a ratio of L: D of about 6: 1 or greater A vessel shell defining a zone; And One or more gas openings for discharging the gaseous stream to the reaction zone, Wherein most of the cumulative open area defined by all of the gas openings is located within about 0.25D of a typical lower end of the reaction zone, Bubble column reactor. The method of claim 40, A bubble column reactor wherein a majority of the cumulative open area defined by all of the gas openings is located within about 0.022 L of the typical lower end of the reaction zone. The method of claim 40, A bubble column reactor wherein substantially all of the cumulative open area defined by both gas openings is located within about 0.25D and about 0.022L of the typical lower end of the reaction zone. The method of claim 40, A bubble column reactor wherein at least two of the gas openings are axially spaced from each other by at least about 1D. The method of claim 40, Further comprising at least one liquid opening for discharging the liquid stream to the reaction zone, wherein at least about 30% of the cumulative opening area defined by all of the liquid openings is located closest to the conventional lower end. Bubble column reactor located within 1.5D. The method of claim 44, A bubble column reactor wherein at least two of the liquid openings are spaced apart from each other by at least about 0.5D. The method of claim 40, A bubble column reactor in which L ranges from about 20 to about 75 meters, D ranges from about 2 to about 10 meters and the L: D ratio ranges from about 8: 1 to about 20: 1. The method of claim 40, A degassing vessel further defining an inlet and outlet, said inlet being in fluid communication with said reaction zone, said outlet being typically located below said inlet, said degassing vessel at least partially extending between said inlet and outlet And a conventional granular sidewall defining a degassing zone, said degassing zone having a maximum horizontal cross-sectional area of less than about 25% of the maximum horizontal cross-sectional area of said reaction zone. The method of claim 47, Bubble column reactor wherein the degassing zone is at least partially located in the vessel shell. The method of claim 47, A bubble column reactor wherein the degassing zone is located entirely outside the vessel shell. The method of claim 40, When the reaction zone is theoretically divided into four vertical quadrants of equal volume by a pair of intersecting vertical planes, less than about 80 weight percent of the cumulative open area defined by all of the gas openings is located in a common one of the vertical quadrants. Bubble column reactor in which the gas openings are arranged in such a way. The method of claim 40, At least a portion of the reaction zone is defined by at least one granular sidewall of the reactor, and at least about 25% of the cumulative open area defined by all of the gas openings is attributable to a gas opening spaced inwardly at least 0.05D from the granular sidewall. Bubble column reactor.

Images