RU2796340C2 - Recycling method for obtaining ethylene glycol - Google Patents

Abstract

FIELD: ethylene glycol production.

SUBSTANCE: present invention relates to a liquid phase process for the production of ethylene glycol. The process includes the following steps: obtaining an aqueous ethylene oxide feed stream containing ethylene oxide and water; combining the catalytic hydration feed stream with the recycled split stream to form a first reactor inlet stream; feeding the inlet stream of the first reactor to the inlet of the first adiabatic reactor, where the inlet of the first adiabatic reactor has a first inlet temperature; reacting ethylene oxide and water in the presence of a first ion exchange resin catalyst in a first adiabatic reactor to produce an effluent containing water, ethylene glycol and unreacted ethylene oxide; further combining the effluent stream with the recirculated feed stream to form a combined stream containing water, ethylene glycol and unreacted ethylene oxide; feeding the combined stream to the inlet of the second adiabatic reactor, where the inlet of the second adiabatic reactor has a second inlet temperature; reacting ethylene oxide and water in the presence of a second ion exchange resin catalyst in a second adiabatic reactor to produce a second reactor effluent containing water, ethylene glycol and unreacted ethylene oxide; compressing the second effluent from the reactor; dividing the second reactor effluent into a recycle stream and a front stream; dividing the recirculated stream into a recirculated split stream and a feeding recirculated stream; feeding the front stream in the form of a tubular reactor feed stream to a non-catalytic tubular reactor to obtain a product stream in which the total concentration of diethylene glycol (DEG), triethylene glycol (TEG) and higher glycols in the product stream is higher than the total concentration of DEG, TEG and higher glycols in the front stream.

EFFECT: process maintains low resin swelling and excellent selectivity while reducing process complexity and increasing versatility.

18 cl, 3 dwg

Description

CROSS-REFERENCE TO SUBSTITUTE APPLICATION

This invention claims priority to U.S. Provisional Application No. 62/664,706, filed April 30, 2018, the entire content and disclosure of which is incorporated herein by reference.

FIELD OF TECHNOLOGY

This invention relates to a process for the oxidation of ethylene to ethylene oxide and the hydrogenation of ethylene oxide to ethylene glycol.

BACKGROUND OF THE INVENTION

The role of ethylene glycol in society has increased significantly since soon after its first synthesis, the properties of the molecule became the subject of a fierce dispute between two giants of early organic chemistry, Adolf Wurz and Hermann Kolbe. In particular, Würz and Kolbe disputed the functionality and chemical formula of ethylene glycol among the still emerging notions of alcohol homologues, which were the key to Kolbe's great theories on chemical structure. The laboratory soon turned into a proxy war of the growing industrial and technical power of the growing rivalry between Germany and France, with both countries devoting significant resources to scientific research, and Wurtz's magic in chemical synthesis giving France a significant advantage. The dispute was only ended by the use of arms when Bismarck's German Confederation annexed Wurtz's Alsatian homeland in the Franco-Prussian War and thus effectively turned the international dispute into an internal one.

Today, interest in ethylene glycol is more peaceful, but all the more competitive, because ethylene glycol is one of the most widely produced organic chemicals. Ever since large-scale industrial production of ethylene glycol began on the eve of World War I, the dramatic increase in the use of internal combustion engines for motor cars and other vehicles has stimulated demand for ethylene glycol as coolant and antifreeze. Since then, the growth of ethylene glycol production has only accelerated, so that by 2017 the estimated global production of ethylene glycol exceeded 25 billion tons.

Ethylene glycol is usually produced as one of the many derivatives of ethylene oxide, and although other production methods are available, the majority is produced from ethylene oxide by a liquid phase non-catalytic thermal hydration process. Since ethylene oxide reacts more easily with ethylene glycols than with water, it is inevitable that mixtures of monoethylene glycol (MEG) as well as higher glycol content by-products such as diethylene glycol (DEG), triethylene glycol (TEG) and even higher ethylene glycols will appear. Although these higher glycols have significant economic value, many manufacturers and plant operators would like to avoid their formation because the end user market for these products is not as well developed and it can be difficult to find industrial customers who need them and sell these higher glycols to them. .

To suppress the reaction between the glycol product and ethylene oxide, and thereby reduce the formation of these higher glycols, conventional non-catalytic hydration is carried out with an amount of water that is much higher than the stoichiometric amount of water for hydration of ethylene oxide to ethylene glycol, for example, 15-40 moles of water per mole of ethylene oxide. This addition of excess water is effective in balancing the kinetically favorable parallel reaction between the glycol product and ethylene oxide, which, as mentioned above, competes with the hydration of ethylene oxide to monoethylene glycol. However, while effective in suppressing the formation of higher ethylene glycols, the use of a large excess of water compared to ethylene oxide presents a problem for the plant operator in removing these large excesses of unreacted water, since such removal is energy intensive and requires evaporation/distillation facilities. Accordingly, there has been intensive research into alternatives to the standard thermal hydration of ethylene oxide for the production of ethylene glycol.

One early example of an alternative to traditional thermal hydration is the homogeneous catalytic hydration of ethylene oxide to monoethylene glycol. The earliest examples of this approach involved the homogeneous catalysis of sulfuric acid and its associated salts (see Othmer, D.F. and Thakar, M.S., Glycol Production - Hydration of Ethylene Oxide. Ind. Eng. Chem. 1958, 50, 1235). EP 0,123,700 describes the modification of these earlier generations of acid catalysts by treating them with, for example, ethylamines to partially neutralize them, in the hope of improving the selectivity of the hydration reaction to monoethylene glycol. Since then, other salts have been proposed for homogeneous systems, such as quaternary phosphonium salts as described in US Pat. No. 4,160,116, and metallate and bicarbonate salts as described in US Pat. No. 7,683,221. Increasingly creative combinations of organics such as EDTA and Salen compounds have also been proposed as homogeneous catalysts (see Hal, J. W., Ledford, J. S., and Zhang, X., Catalysis Today 123 (2007), 310-315).

Homogeneous catalyst systems are often used in a two-stage process for the production of ethylene glycol, see for example US Patent No. 4,519,875, in which ethylene oxide is first reacted with carbon dioxide to produce ethylene carbonate, which is then hydrolyzed to ethylene glycol, where the same catalyst is usually used in both stages. . Following this pioneering patent, continuous research has led to incremental improvements in the two step process. For example, in US Pat. No. 5,763,691, the carbonization reaction is catalyzed in ethylene oxide absorbate in the presence of a halogenated organic phosphonium salt carbonization catalyst. Additional research has greatly expanded the range of known catalysts; see, for example, macrocyclic chelating compounds ("crown ethers"), described in US patent No. 7,453,015.

Although homogeneous catalysts have improved the selectivity of the ethylene glycol hydration reaction over non-catalytic hydration, homogeneous catalysts have disadvantages that more than offset these improvements. In particular, the glycol product solution contains a dissolved or suspended homogeneous catalyst—essentially causing catalyst contamination of the product, which greatly degrades the quality of the product, and is also very aggressive, as is the case for example with acid catalysts. Thus, an additional step of separating the homogeneous catalyst from the glycol product solution is necessary, adding cost and complexity to the process. For example, although US Pat. No. 5,763,691, cited above, attempts to simplify the two-stage production of ethylene glycol through the reaction of ethylene oxide and carbon dioxide in the absorbate, the patent still requires a separate distillation stage to remove the carbonization catalyst.

This complexity of the method brings with it the additional disadvantage that it is not very versatile and flexible. For example, the process lacks the flexibility to operate partially catalytically, i.e. in mixed catalytic and non-catalytic modes, which would allow the production of more higher glycols if the operator so desires, as well as to adapt the plant in case of some disruptions in logistics and operation. when it cannot operate in catalytic mode, such as when sufficient catalyst is not available. The two-stage hydration process also lacks the versatility to be used in both new plants and retrofits. The two-stage hydration method can only be used in new plants and cannot be used in retrofits, because in the event of an upgrade, all "internal interfaces" would need to be removed and replaced, making such an upgrade too costly. The disadvantage of the two-stage hydration method is the need to inventory concentrated ethylene oxide (concentration of more than 70%) for the production of intermediate ethylene carbonate in the first stage of the method. The carbonization reaction in the first step is carried out by bubbling carbon dioxide into such a concentrated ethylene oxide liquid solution at high temperature and pressure (for example, more than 1.6 MPa). Large volumes of concentrated ethylene oxide require significant additional measures (e.g., storage in a coolant liquid) to prevent reactions or events that could result in ignition, combustion, deflagration, detonation, or explosion of any gas stream, but especially those containing typical concentrations of ethylene oxide.

Not surprisingly, given the aforementioned disadvantages of both non-catalytic and homogeneous catalytic hydration systems, heterogeneous catalysis using solid or supported catalysts has also been explored. In search of the benefits of higher catalytic hydration selectivity without the aforementioned disadvantages, researchers have turned their attention to heterogeneous catalyst systems. Because heterogeneous catalysts do not adsorb into liquid phase reaction mixtures or product solutions, do not contaminate product solutions, do not require a separation step, and promote continuous operation of the hydration reactor, they are of considerable interest.

A variety of solid materials have been proposed as heterogeneous hydration catalysts. This list includes zeolites such as those described in US Pat. No. 3,028,434, as well as amorphous aluminosilicates, solid bases, and supported acid catalysts such as those described in US Pat. No. 4,165,440. Ion exchange resins have also been proposed, a sulfonic acid-based partially amine-neutralized styrene-divinylbenzene resin is described in EP 0 123 709; quaternary phosphonium ion exchange resin catalysts, see PCT publication WO 2000/35840; and strong base ion exchange resins, summarized in US Pat. No. 6,211,419, were among the first ion exchange resins proposed to catalyze the reaction of water and ethylene oxide. More recent publications have expanded the scope to a wide range of anion exchange resins, see for example Chinese Patent No. 1208135.

Although excellent results have been achieved using heterogeneous ion exchange resin catalysts, their use also has certain disadvantages. In particular, the ion exchange resin has limited resistance to heat, so that under typical reaction conditions, the resin swells for long periods of time, resulting in deactivation. Swelling, which results from internal polymerization in the resin layer and the overall accumulation of high molecular weight by-products, is a serious problem that is a significant limiting factor in the service life of an ion exchange resin. Resins typically undergo some reversible swelling during operation due to exposure to solvents; however, with careful handling, this swelling can be reduced and the life of the ion exchange resin extended. However, exposure of the ion exchange resin to ethylene oxide can cause more problematic and irreversible swelling of the resin, especially when such exposure occurs at higher hydration reactor temperatures, which, as mentioned above, may be outside the temperature range that the resin can satisfactorily tolerate.

Swelling is not only a problem in itself, but can also reduce efficiency and complicate the catalytic hydration process; for example, when excessive swelling requires changing the resin and thus shutting down the catalytic hydration unit in which it is contained. Various attempts have been made to address these problems, including attempts to find resins that are less likely to swell under normal operating conditions, and to develop process modifications in an attempt to improve performance in addition to reducing swelling. For example, European Patent EP1828086 describes the use of "protective beads" designed to remove impurities that would conceivably cause the resin to swell, but such attempts have been ineffective.

Rather than protecting resins with a protective layer in an attempt to minimize swelling, another more successful approach, described in PCT Publication WO2008/150338, does not aim to protect the resin, but to facilitate the removal of an expired ion exchange resin through previously unsuggested work. an up-flow catalytic reactor that expands the catalyst bed to allow size separation of the catalyst particles and thereby selectively remove swollen particles. Another approach is shown in US Pat. No. 6,160,187 which states that swelling reduction can be achieved under adiabatic reaction conditions and using an adiabatic reactor. However, the examples contained in the "187" patent do not actually show any improvement in the swelling characteristics of the resins, nor has there been any subsequent development of the adiabatic regime, which has been neglected.

Accordingly, although there have been several promising developments in certain aspects of heterogeneous catalytic hydration, there remains a significant need for further improvements in catalytic hydration under adiabatic process conditions.

BRIEF DESCRIPTION OF THE INVENTION

The present invention relates to a liquid phase process for the catalytic hydration of ethylene oxide for the production of monoethylene glycol (MEG), comprising: obtaining an ethylene oxide-water solution containing ethylene oxide and water; combining the catalytic hydrogenation feed stream with the recycled split stream to form a first reactor inlet stream; feeding the inlet stream of the first reactor to the inlet of the first adiabatic reactor, where the inlet of the first adiabatic reactor has a first inlet temperature; reacting ethylene oxide and water in the presence of a first ion exchange resin catalyst in a first adiabatic reactor to produce an effluent containing water, ethylene glycol and unreacted ethylene oxide; further combining the extraction of the outgoing stream with the recycled feed stream to obtain a combined stream containing water, ethylene glycol and unreacted ethylene oxide; feeding the combined stream to the inlet of the second adiabatic reactor, where the inlet of the second adiabatic reactor has a second inlet temperature; reacting ethylene oxide and water in the presence of a second ion exchange resin catalyst in a second adiabatic reactor to produce a second reactor effluent containing water, ethylene glycol, and unreacted ethylene oxide; compressing the second effluent from the reactor; dividing the second reactor effluent into a recycle stream and a front stream; dividing the recirculated stream into a recirculated split stream and a feeding recirculated stream; feeding the front stream as a tubular reactor feed stream to a non-catalytic tubular reactor to produce a product stream in which the total concentration of DEG, TEG and higher glycols in the product stream is higher than the total concentration of DEG, TEG and higher glycols in the front stream.

BRIEF DESCRIPTION OF MULTIPLE DRAWING TYPES

The foregoing summary, as well as the following detailed description of the preferred embodiments of the invention, will be better understood when read in conjunction with the accompanying drawings. To illustrate the invention, the drawings show the presently preferred embodiments. However, it should be understood that the invention is not limited to the precise devices and instruments shown. On the drawing:

Figure 1 shows a process flow diagram for obtaining a stream of concentrated ethylene oxide in accordance with this invention.

Figures 2-3 show process flow diagrams for ethylene glycol production options in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

All parts, percentages and ratios used herein are by volume unless otherwise indicated. All pressures are absolute. All documents cited here are incorporated by reference.

By "water" is meant any kind of water suitable for use in chemical and petrochemical processing, including deionized, demineralized, industrial, drinking and distilled water. Steam condensate and condensate obtained from the dehydration and evaporation part of the glycol production process can also be used.

The present invention provides an improved adiabatic catalytic hydration process in which catalyst swelling is reduced, process flexibility is greatly improved, and the process is configured to advantageously be used to upgrade a non-catalytic process to a catalytic one. A particular advantage of this invention is that a number of adiabatic reactors are used. Typically, the non-catalytic hydration reaction is carried out in a single adiabatic reactor, regardless of yield or maximum reactor temperature. Such operation can be particularly detrimental to the ion exchange resins of this invention, as temperatures outside the temperature range indicated for such resins can increase swelling and reduce catalyst life, as described above.

Instead of using such single-stage adiabatic reactors and their attendant disadvantages (such as high temperature in the presence of a catalyst), the present invention uses multi-stage adiabatic reactors arranged in a row. Since no heat is removed in the adiabatic reactor, the temperature of the effluent is higher than the temperature of the feed, and the heat of reaction must be removed from the effluent by a cooler at each stage. This is particularly important to prevent the aforementioned catalyst swelling and to prevent excessive non-catalytic reaction of glycol products with ethylene oxide to higher glycols. In the present invention, these problems are solved by a catalytic hydration reactor unit with two or more adiabatic reactors in series with recirculation and cooling of the reactor effluent to control the reactor temperature. External refrigeration removes heat of reaction from the effluent stream, lowering its temperature so that it can be recycled to one or more reactor streams so that reactor temperatures do not exceed a target maximum value. Moreover, from a process point of view, cooling with external heat exchangers is easier and simpler than setting the appropriate mode for isothermal reactors, and can be adjusted to absorb normal fluctuations in the process and maintain the performance of catalytic reactors. Thus, the use of a range of adiabatic reactors with such external coolers improves process versatility and heat removal efficiency. A second advantage is that the process allows the conventional non-catalytic to catalytic hydration process to be upgraded or converted, as described in more detail below.

The present invention achieves these advantages while maintaining the high selectivity of prior catalytic hydration systems for monoethylene glycol. In each bed of an adiabatic reactor, ethylene oxide and water react in the presence of a heterogeneous catalyst (such as an ion exchange resin) to form monoethylene glycol, with at least 98% of the ethylene oxide converted to monoethylene glycol, with only a small proportion converted to higher glycols such as diethylene glycol or triethylene glycol.

Thus, the present invention provides an improved process scheme that includes a catalytic hydration reactor block containing adiabatic reactors in series, which maintains low resin swelling and excellent selectivity, and reduces process complexity and increases versatility.

This catalytic hydration reactor bank (also “reactor bank”) obtained in accordance with the present invention as described above is shown as 25 in Figure 2 and is described in more detail below. The reactor block 25 is loaded with, among other feed streams, a liquid cold feed stream through conduit 40 which is highly enriched in ethylene oxide and which is supplied at a pressure high enough to keep the ethylene oxide in the liquid. Reactor bank 25 includes a final adiabatic reactor for non-catalyzed thermal hydrolysis of EO 26, which provides complete conversion of the remaining unconverted EO (present in the last adiabatic reactor effluent from the series of reactors 22) to ppm levels, significantly reducing the total volume of catalytic reactor of a number of reactors 22.

Any suitable means of obtaining such a liquid feed stream that is highly enriched in ethylene oxide is applicable to this inventive process for the catalytic production of ethylene glycol. One suitable means for this is the straight two-stage evaporative absorber-stripper unit shown in Figure 1. Figure 2 shows the communications between the absorber-stripper unit 1 and the reactor bank 25, which are obtained in accordance with the present invention. The process for producing ethylene oxide with reference to this direct two-stage evaporator-absorber-desorber unit is described in more detail below.

Ethylene oxide is produced by continuously contacting an oxygen-containing gas with an olefin, preferably ethylene, in the presence of an ethylene oxide (“epoxidation”) catalyst (described in more detail below). The oxygen may be supplied to the reaction in substantially pure molecular form or in a mixture such as air. By way of example, conventional reactant feed mixtures under operating conditions may contain from about 0.5% to about 45%, preferably from about 5% to about 30% ethylene and from about 3% to about 15% oxygen, with the balance being relatively inert materials, including substances such as carbon dioxide, water, inert gases, other hydrocarbons and reaction moderators described here. Non-limiting examples of inert gases include nitrogen, argon, helium, and mixtures thereof. Non-limiting examples of other hydrocarbons include methane, ethane, propane, and mixtures thereof. Carbon dioxide and water are by-products of the epoxidation process, as well as common impurities in feed gases. Both have an adverse effect on the catalyst, so the concentrations of these components are usually kept at a minimum.

As noted previously, one or more reaction moderators are also present in the reaction, non-limiting examples of which include organic halogenated compounds such as C _{1 -C 8} halohydrocarbons; especially preferred are chloride-containing moderators such as methyl chloride, ethyl chloride, ethylene dichloride, vinyl chloride or mixtures thereof. Controlling the level of chloride concentration is especially important for ruthenium-containing catalysts.

As stated above, a conventional process for epoxidizing ethylene involves the vapor phase oxidation of ethylene with molecular oxygen in the presence of an epoxidation catalyst, in a fixed bed tubular reactor. Typically, commercial fixed bed ethylene oxide reactors are in the form of a plurality of parallel tubes (suitably sheathed) approximately 0.7-2.7 inches OD and 0.5-2.5 inches ID and 15-53 feet long, each filled and packed with catalyst. The initial reaction mixture (described above) is introduced into these tubes and the resulting reactor effluent contains ethylene oxide, unused reactants and by-products.

For purposes of illustration only, the following are conditions that are often used in modern commercial ethylene oxide reactors: gas hourly space velocity (GOSV) 1500-10000 h ^-1 , reactor inlet pressure 150-400 psi. inch g, coolant temperature 180-315°C, oxygen conversion 10-60%, and EO production (output) 7-20 lb. EO/cfm catalyst/h Feedstock consumption at the reactor inlet after start-up is complete and during normal operation is typically (in % vol) 1-40% ethylene, 3-12% O ₂ ; 0.3%-20%, preferably 0.3-5%, more preferably 0.3-1% CO ₂ ; 0-3% ethane, some of one or more chloride moderators as described herein; and the remainder of the raw material is argon, methane, nitrogen, or mixtures thereof.

Typical operating conditions for the ethylene epoxidation process include temperatures ranging from about 180°C to about 330°C, and preferably from about 200°C to about 325°C, and more preferably from about 225°C to about 280°C. Operating pressure can vary from about atmospheric pressure to about 30 atmospheres, depending on the desired mass flow and performance. Higher pressures may be used within the scope of this invention. Residence times in commercial reactors are typically on the order of about 2 to about 20 seconds.

The reactor effluent exits the reactor outlet and preferably flows into an EO wash column where the reactor effluent is contacted with a lean recycled aqueous solution to absorb ethylene oxide as well as several "light" solutes contained in the reactor effluent.

In figure 1, the liquid from the wash column (hereinafter referred to as the stream of "water from the enrichment cycle") flows through the pipeline 1 and is fed into the flash drum 5. The stream of water from the enrichment cycle is an aqueous stream containing from 1.0 mol% to 4% mol ethylene oxide with some dissolved gases such as methane, ethylene and others. This water stream from the enrichment cycle is fed under pressure, preferably from about 0.4 MPa to about 0.6 MPa, into the flash drum 5. The absorber 9 is permanently and directly fixed to the top of the flash drum 5, and the flash drum has an opening for connection to the absorber 9. In this invention, the pressure in both the flash drum 5 and the absorber 9 is maintained at a level sufficient so that when water from the enrichment cycle enters the flash drum 5, the "light" gases that are dissolved in the water from the enrichment cycle , which have a volatility greater than that of ethylene oxide, were washed out of the water from the enrichment cycle upon entering the evaporation drum 5, and the vapor of light gases is directed by the pressure difference between the evaporation drum 5 and the absorber 9 upwards through the hole into the absorber 9 with the formation of the upper layers of the absorber 9, and the dissolved ethylene oxide remains substantially solubilized in the upgrade water stream in the bottoms of the flash drum 5. These more volatile "light" gases include the aforementioned inert, unreacted or by-product gases dissolved in the upgrade water. (Furthermore, traces of ethylene oxide can also be vaporized with the more volatile light gases in the upgrade water stream. Small amounts of ethylene oxide that vaporize from the upgrade water and rise to absorber 9 with the dissolved light gas vapor can be recovered by the downstream wash stream ( not shown) to absorb and recover upward moving ethylene oxide).

As stated above, the absorber 9 is directly and permanently attached to the top surface of the flash drum 5, for example by welding. Absorber 9 is attached to flash drum 5 in the same manner as in a flash column and reboiler configuration well known to one skilled in the art. More specifically, the absorber 9 is attached to the flash drum 5 along the longitudinal axis of the flash drum 5, but is preferably attached in the middle of the longitudinal axis. The cross section of the absorber 9 matches the shape of the opening in the flash drum 5 over which the absorber 9 is attached and which allows communication between the flash drum 5 and the absorber 9. The cross section of the absorber and the shape of the hole in the flash drum 5 have the same matching shape and this shape can be any shape between a circle and an ellipse, between eccentricity 0 and eccentricity 1. The opening in the flash drum 5 and the absorber 9 is positioned so that the center of the hole is collinear with the center of the absorber 9. The configuration of the aperture in the flash drum 5 and the shape of the absorber 9 and their respective positions provides proper fluid communication between the two elements. The absorber 9 is preferably flat welded to the inner diameter of the hole made in the flash drum 5 so that the outer diameter of the absorber 9 is flush with the diameter of the hole in the flash drum 5.

In order to achieve a balance between washing out the more volatile "light" solutes while maintaining high ethylene oxide solubility in enriched cycle water, the pressure in the flash drum 5 is maintained at about 0.1 MPa to 0.3 MPa, and the pressure in the absorber 9, s to which there is a direct connection, only slightly below the pressure in the flash drum 5. The temperature of the gases in the upper part of the absorber 9 is from about 30°C to about 45°C, and the temperature in the flash drum 5 is from about 50°C to about 60°C .

The top components of the absorber 9 are returned via conduit 12 to the pre-installation reactor block (not shown) as a possible recycle to the reactor. Water from the enrichment cycle flows as bottoms from flash drum 5 through pump 14 and through an optional heat exchanger 20 (which preheats the stream) and into stripper 15. Stripper 15 is pressurized by controlling the back pressure created by removing vapors from flash condenser 19. The temperature at the top of the stripper 15 should be from about 25°C to about 40°C, and at the bottom of the stripper 15 is from about 120°C to about 160°C. The water stream from the enrichment cycle may be preheated in the heat exchanger 20 to reduce the load required to operate the stripper 15. The pressure in the stripper 15 is maintained in the range from about 0.2 MPa to about 0.6 MPa. The stripper 15 may also employ a steam jet pump system well known to those skilled in the art, where the steam is supplied from: (1) steam generated throughout the ethylene oxide production system, or (2) medium or high pressure steam supplied from external sources. /ZGU; or a combination of the two sources.

As shown in Figure 2, the pressurized stream from the top of the stripper flows through conduit 30 to condenser 16 and then to flash drum 17 which separates the scrubbed light gases from the concentrated aqueous ethylene oxide liquid stream that is fed to pump 18. Outgoing from pump 18 divided into two streams: one stream as a return stream back to the stripper 15, and one liquid aqueous stream of pure product concentrated in ethylene oxide. The washed light gases from drum 17 flow to a flash condenser 19 to further recover ethylene oxide from the washed steam stream and control the amount of dissolved light gases in the concentrated ethylene oxide liquid solution 41. Compared to a highly dilute aqueous solution found at a comparable location in a conventional process ethylene oxide recovery unit , the pressurized overhead stream of the stripper is substantially enriched in ethylene oxide, containing at least about 40 mole %, preferably at least about 50 mole % ethylene oxide, with the remainder being primarily water. Thus, the pressurized stream from the top of the stripper produces both high concentrations of ethylene oxide and is under high pressure. This makes the stream extremely useful for further processing to ethylene glycols via catalytic hydrolysis on a very economical basis, since there is no need to further enrich the stream or pressurize the following processing steps. Another preferred option of the process of the present invention is that the bottoms of stripper 15, which is a lean aqueous cycle solution containing some ethylene oxide (from about 1 molar ppm to about 50 molar ppm), and which is also free of light components as well as impurities, can preferably be used in a variety of other thermal energy management schemes throughout the plant, depending on needs and conditions.

Under these conditions of streams with higher than normal concentrations of ethylene oxide, operators will continue to comply with the appropriate safety standards that are always followed when producing, processing or storing ethylene oxide. As always, precautions must be taken to prevent reactions or events that could result in ignition, combustion, deflagration, detonation or explosion of any gas stream, especially those containing ethylene oxide in concentrations higher than typical. Accordingly, to prevent such events, safety devices may be used to relieve or reduce unwanted pressure buildup in the separation process, reaction or systems, or elsewhere in the ethylene oxide plant, both before and after what is shown. The separation in stripper 15 is so efficient that nearly all of the ethylene oxide in the enrichment water that enters stripper 15 is successfully separated and recovered from the enrichment water and exits stripper 15 as a clean liquid overhead stream, while only very a small portion of the EO leaves stripper 15 in other streams.

The pressurized overhead stripper stream obtained in stripper 15 flows at elevated pressure through conduit 40 from stripper unit 1 to reactor bank 25 as described above and shown in Figure 2. In accordance with the present invention, reactor bank 25 comprises two or more reactors. . 2, a reactor bank 25 is shown with a series of two adiabatic catalytic reactors in series, each with a reactor stream cooler, where each reactor stream cooler is capable of receiving a hot recycle effluent from the second reactor via pump 19. More specifically, as shown, in Figure 2, and in more detail in Figure 3, the reactor bank 25 shows a series of reactors 22 of two in-line adiabatic fixed-bed downflow reactors 38 and 40, each with a reactor flow heat exchanger cooler 39 and 43, respectively, and recycle pump 19. The reactors are referred to as the first adiabatic reactor 38 and the second adiabatic reactor 40, but in this description, these adiabatic reactors 38 and 40 may also be referred to as "reactors".

As described above, external cooling is necessary because the reaction of water with ethylene oxide is highly exothermic, so it is necessary to remove the heat of reaction from the product or effluents if the reaction takes place in an adiabatic reactor, as in the present invention, to avoid excess temperature in the reactor bed, which can reduce the life of a temperature sensitive catalyst such as ion exchange resins. This is different from isothermal conditions where the heat of reaction is removed directly from the reactor. Keeping the ion exchange resin catalyst at temperatures that the resin can withstand is of course critical to reduce swelling, prevent degradation, and maximize catalyst life. Under proper operation, the rise in adiabatic temperature can be controlled at a level that maintains good efficiency and service life.

Finally, downstream of the row of reactors 22 is a conventional non-catalytic adiabatic tubular reactor 26 (this reactor, of course, does not contain a catalyst).

As indicated above, the concentrated ethylene oxide-water solution in conduit 40, obtained from the cooled process stream in condenser 16, is relatively cold and is therefore mixed with water stream 36, which is heated in heat exchanger 35, to form mixed stream 41, then combined with recirculated split stream 57 and cooled at a temperature high enough to carry out the reaction of catalytic hydration of ethylene oxide to monoethylene glycol. The heat exchanger 39 brings the ethylene oxide-water solution to a temperature of 70°C to 100°C. The heat exchanger 35 also provides heat to reach a temperature high enough to start the ethylene oxide hydration reaction when a hot recycle stream from the reactor is not yet available.

The ethylene oxide-water solution 41 will preferably contain from about 5 mole % to about 10 mole % ethylene oxide. Thus, the ethylene oxide-water solution 41 contains a significant stoichiometric excess of water to ethylene oxide, as compared to that required for the hydration of ethylene oxide to ethylene glycol. On a molar basis, the ethylene oxide-water solution will have a water:ethylene oxide molar ratio of from about 5:1 to about 15:1, preferably from about 7:1 to about 12:1; and although this is the above stoichiometric excess, as stated above, it is significantly lower than the amounts used in the art for non-catalytic hydration. And thus, since the feed stream contains less water, the final product also has correspondingly reduced amounts of water to be removed from the product, as described below.

The ethylene oxide-water solution stream 41 is fed to the aforementioned reactor bank 25 where it is combined with the recycled cut off stream 57 to form the first feed stream 58. into the reactor bank 25, the resulting first feed stream 58 will have a higher ratio of water to ethylene oxide compared to the ratio of water to ethylene oxide in the ethylene oxide-water solution stream 41, preferably a molar ratio of from 40:1 to about 10:1, preferably from about 30:1 to about 20:1, water:ethylene oxide. This is indicated for the variant with two consecutive reactors shown in figure 2, since the amount of ethylene oxide gradually decreases with the production of more and more monoethylene glycol, and the ratio of water to ethylene oxide in subsequent reactors increases. This is a significant improvement over conventional catalytic and non-catalytic hydration. By recirculating a portion of the reactor effluent to each reactor (streams 61 and 57), a high water:ethylene oxide ratio in the feed stream to each reactor can be obtained while maintaining a low overall water:ethylene oxide ratio in product stream 88 at the end of the reactor bank being fed. to evaporators to remove water. Thus, the reaction of ethylene oxide and water is catalyzed both in the presence of an ion exchange resin and with a significant excess of water to ethylene oxide. This promotes the production of monoethylene glycol and suppresses the formation of the higher glycol homologue, thus both effects are advantageous. And yet this is retained by obtaining an end product that is much more concentrated in ethylene glycol and therefore requires much less energy in the evaporation unit. For example, in accordance with this invention, the end product of the block of reactors, the combined product from the reactor 88, may have a total concentration of glycols (mono-, di-, tri - and trace higher) 20 wt%. or even higher, while conversely, in a conventional plant, the reactor effluent fed to the evaporation unit has a total concentration of glycols of about 12% by weight, thereby reducing the amount of heat load for evaporation of excess water by a factor of at least 40 %.

Having illustrated the general scheme of operation of the reactor block and the advantages of the present invention, the method according to the present invention will now be described in more detail with reference to Figure 2.

As noted above, streams 41 and 57 are combined and cooled in a heat exchanger 39 to form a feed stream 37 to the first reactor 38 and thus the temperature of the reactor 38 is controlled through the temperature control of the stream 37 while the temperature in the subsequent reactor 40 is controlled by using the heat exchanger. 43.

Temperature and pressure control are important parts of the process according to the present invention. The hydration reaction in the present invention is carried out as a liquid phase process. In the present invention, by "liquid phase process" is meant that the feed streams to the reactor, reactor product or reactor effluents, and the reactants within the reactor are maintained in the liquid phase. Accordingly, the temperature and pressure in each reactor is such that as the streams are fed into and enter each reactor, they are maintained in the liquid phase. Thus, the temperature and pressure in each reactor is controlled to keep the contents of the reactor in the liquid phase. The choice of temperature is largely determined by the considerations of the above paragraphs. Thus, at the following temperature ranges in the present invention, when the contents of the reactor are in liquid form, the pressure will be from about 7 atm. up to about 15 atm., preferably from about 8 atm. up to about 13 atm.

When choosing a temperature, two competing requirements must be balanced. First, as noted above, the inlet temperature of the present invention must be high enough to drive the ethylene oxide hydration reaction so that the conversion of ethylene oxide to monoethylene glycol is significant. In the present invention, the percentage of conversion of ethylene oxide to monoethylene glycol in the first reactor should be at least about 50%, preferably at least about 60% of the total ethylene oxide reacted. Thus, the temperature at the entrance to the first reactor 38 should be in the range from about 50°C to about 90°C, preferably in the range from about 70°C to about 85°C. The percentage conversion of ethylene oxide to monoethylene glycol in the second reactor 40 should be at least about 10% of the total EO reacted, preferably at least about 20% of the total EO reacted. Thus, the inlet temperature to the second adiabatic reactor 40 should be in the range of about 70°C to about 110°C, preferably in the range of about 80°C to about 95°C. However, the temperature in the reactor (and therefore at the outlet of the reactor) should not be too high, because, as mentioned above, a significant increase in temperature inside the reactor causes swelling and destruction of the ion exchange resin. The high temperatures in the reactor are the result of an exothermic hydration reaction, so this heat of reaction must be removed to contain any temperature increase. In isothermal operations of the prior art, the heat of reaction may be removed, for example, by absorption into the body cooler. This is not possible with the adiabatic reactors of the present invention, which have significantly fewer process complications but are correspondingly less versatile in their ability to remove heat. However, it has been found in the present invention that the temperature rise can still be at least partially controlled within a range suitable for stable operation of the ion exchange resin catalyst by excess water in each reactor, which acts as a heat sink. Thus, although one of the objectives of the present invention is to reduce the amount of water used in the production of ethylene glycol, nevertheless, at least a minimum amount of water is necessary to contain the temperature increase in the reactors resulting from the exothermic reaction. As described above, both of these objectives are accomplished in the present invention by recycling portions of the effluent from catalytic reactor 74 to the same reactors 38 and 40. With this inventive configuration, progressively higher water:ethylene oxide ratios are produced in successive reactors, keeping the exotherm at bay so that the outlet temperature of each reactor is from about 90°C to about 120°C, preferably from about 95°C to about 110°C.

Despite the success of the present invention in controlling the temperature increase in the reactor, it may nevertheless be necessary to further reduce the temperature of the reactor outlet stream to achieve the acceptable inlet temperature range mentioned above. To do this, the combined streams 58 and 62, including streams 57 and 61, are passed through heat exchangers 39 and 43. This allows for heat integration, since a stream (not shown) from outside the reactor block can be heated in any of the heat exchangers due to indirect heat exchange with the outgoing a reactor stream from one or more reactors arranged in series.

As noted above, at higher temperatures almost complete conversion of ethylene oxide to monoethylene glycol occurs, at lower inlet temperatures not all of the ethylene glycol will react, and detectable amounts of ethylene oxide will appear and remain unreacted ("breakthrough") in the reactor effluent. In certain situations, it may be necessary to lower the inlet temperature in order to reduce ethylene oxide conversion to provide more ethylene oxide breakthrough, when, for example, it is desirable to extend the life of the ion exchange resin catalyst or in other situations, as described below.

In addition to the above parameters and operating conditions, other factors that may be considered during process operation include, but are not limited to, reactor and process configuration, ion exchange resin catalyst conversion rate and selectivity, and reactant flow rate (RFR).

With the above considerations in mind, feed stream 41 is combined with recycled split stream 57 in a volume ratio of stream 41 to stream 57 of about 0.5:1 to about 2:1. The resulting first feed stream 58, after being cooled to a sufficient reaction temperature in heat exchanger 39, is fed to reactor 38 where ethylene oxide is reacted with water in the presence of an ion exchange resin catalyst or catalyst bed to form monoethylene glycol. In adiabatic reactors in accordance with the present invention, for example reactor 38, mixing ethylene oxide with excess water at or near the catalytic temperature, in the presence of a catalyst, causes ethylene oxide hydration which is highly selective for monoethylene glycol such that at least about 95%, preferably , at least about 98% of the ethylene oxide is converted to monoethylene glycol, and only a small proportion is converted to higher glycols, such as diethylene glycol or triethylene glycol (percent conversion indicated above). In addition to the monoethylene glycol product, each effluent from the reactors of this invention contains water and ethylene oxide (excess amounts are available for recycling to the same reactor or for further addition to the next reactor) and trace amounts of other components, including ethylene oxide and higher glycol homologues.

Leaving reactor 38, effluent 70 is combined with recycled effluent 61 in a volume ratio of stream 70 to stream 61 of about 0.5:1 to about 2:1. The combined stream 62 is then passed through heat exchanger 43 where it is cooled to form stream 81 for feeding to reactor 40. In the second adiabatic reactor 40, as in reactor 38, ethylene oxide reacts with water with high selectivity and conversion so that essentially the effluent the second reactor 73, like the first reactor effluent, contains unreacted water and ethylene oxide, as well as monoethylene glycol and the trace amounts mentioned above. Second reactor effluent 73 is removed from second adiabatic reactor 40, compressed in pump 19, and then divided into front stream 76 and recycle stream 75, which is further divided into recirculated split stream 57 and recirculated feed stream 61. Essentially, the only difference between 70 and 73 is that the latter contains more monoethylene glycol because it contains the amount of monoethylene glycol produced in both reactors 38 and 40 and therefore contains significantly less ethylene oxide.

The first feed stream 58 tons of the combined stream 62 is passed through the respective heat exchangers 39 and 43 to obtain an acceptable temperature streams. In this case, heat exchanger 43 acts as heat exchanger 39, cooling the reactor feed stream to sufficiently control the catalytic reaction.

In addition to the series of reactors described above, including two downflow fixed bed reactors in series, the reactor bank 25 also includes a conventional tubular reactor 26 into which a forward stream 76 can be fed. In the present invention, this tubular reactor 26 is configured and operated so that impart previously unknown levels of versatility and controllability to the catalytic hydration process scheme, which is described in more detail below.

First, in addition to improving the conversion and selectivity of the reaction of ethylene oxide and water to MEG, the present invention can also be used and configured to increase the versatility and adaptability of the hydration process, allowing the process operator to produce di- and triethylene glycol. This can be done by physically bypassing one or more catalytic reactors 38 and 40 (this bypass itself is not shown) and thus removing one or more catalytic reactors from the circulation of the feed stream. Ethylene oxide that "breaks through" or remains unreacted after bypassing one or more of the aforementioned catalytic reactors is sent to a non-catalytic tubular reactor 26 to thermally convert ethylene oxide and water to ethylene glycol. Because tubular reactor 26 is non-catalytic and is entirely dependent on thermal conversion, front stream 76 is preheated in heat exchanger 55 so that tubular reactor feed stream 77 can fall within the effective temperature range for the non-catalytic hydration of ethylene oxide to monoethylene glycol, preferably the tubular inlet temperature. reactor 26 is from about 130°C to about 160°C.

Of course, this non-catalytic thermal conversion is significantly less selective for MEG, producing significant amounts of higher grade glycols such as DEG and TEG, so that by increasing the amount of ethylene oxide that is converted to ethylene glycol in the tubular reactor, reactor product 88 contains less MEG and more DEG and TEG than would be produced if only catalytic reactors were used in reactor bank 25. This gives the operator a versatile and flexible ethylene glycol production process, as mono-, di- and triethylene glycol production can be adjusted by the operator to meet demand . In the extreme case, all catalytic reactors are bypassed so that only the last non-catalytic tubular reactor 26 remains, and all hydration will take place in this reactor, thereby rendering the catalytic process non-catalytic.

Alternatively, instead of physically bypassing the reactors, a "thermal bypass" can be implemented by lowering the inlet temperature to one or more catalytic reactors to the required low level to reduce the reaction between ethylene oxide and water and thus reduce the amount of monoethylene glycol formed. in reactors, which means more "breakthrough" of unreacted ethylene oxide seen at the outlets of catalytic reactors. This ethylene oxide is then converted to ethylene glycol in a tubular reactor as described above.

The tubular reactor may also play a "finishing" role. Even when the row of reactors 22 is operating at the normal catalytic efficiency designed to maximize the conversion of ethylene oxide to ethylene glycol, however, a small amount of unreacted ethylene oxide will remain unreacted after all reactors in row 22, and this ethylene oxide will be a "breakthrough" in the forward stream 76. In this Invention row 22 operates normally such that the concentration of ethylene oxide in front stream 76 will not exceed about 1% mol. This remaining ethylene oxide can then be converted to monoethylene glycol in tubular reactor 26. The use of a tubular reactor as the final reactor can provide significant economic benefits by reducing the amount of catalyst required. The low conversion of ethylene oxide achieved in the tubular reactor is carried out at a very high excess of water, so that the selectivity for monoethylene glycol to higher glycols is very close to that of catalytic reactors. The amount of catalyst required and the volume of the catalytic reactor are therefore significantly reduced while maintaining a very high overall selectivity to monoethylene glycol.

Finally, the tubular reactor 26 can also be used as a backup reactor in situations where one or more of the catalytic reactors fail or must otherwise be taken out of service due to depletion, plugging of the resin bed or nozzle, or other difficulties commonly seen in bed operation. ion exchange resin. When this happens, the failed catalytic reactor must be taken out of service. This means that there will be an excess of unreacted ethylene oxide in reactor row 22, which is undesirable. This excess ethylene oxide breakthrough can be converted to ethylene glycol in tubular reactor 26. Thus, the presence of tubular reactor 26 in the present invention provides considerable flexibility and operability for the process operator. In contrast to the prior art, where the inability to replace or repair the ion exchange resin catalyst or reactor is a serious disadvantage of solid/heterogeneous catalyst systems, in the present invention the reactors can be taken offline for such repairs without interrupting operation.

Although two reactors and two heat exchangers are shown in the reactor block of Figures 2 and 3, this is for purposes of illustration only. More or fewer reactors may be used depending on specific needs and requirements. Increasing the number of reactors has the advantage of reducing the exotherm in each reactor, thereby reducing the amount of irreversible temperature-induced swelling that the ion exchange resin catalyst experiences within each reactor. In addition, increasing the number of reactors also increases ethylene oxide dilution, and since ethylene oxide destroys most ion exchange resins, increased dilution is expected to prolong resin life.

The primary disadvantages of more reactors are cost and complexity, including additional capital costs for reactors, pumps, heat exchangers, and other equipment, as well as operating costs for large volumes of ion exchange resin catalyst, as well as additional piping, instrumentation, and operational complexity.

The alkylene glycol in the combined product of the reactor 88 can then be recovered using an evaporation battery or vacuum distillation, or any other suitable method known to those skilled in the art. Recovery methods can be combined for different stages of recovery. For example, an evaporator battery can be used to remove water from the reactor effluent, while vacuum distillation can be used to dry the glycol. Indeed, one of the advantages of the present invention is that complexity can be reduced in the plant's post-glycol stack of reactors because the amount of heat load required to boil off excess water has been reduced as noted above. Thus, although prior art plants often include evaporation systems having multiple effects or stages (see, for example, EP No. 2121646 B1), the present invention requires fewer stages.

As discussed above, this invention may include one or more ion exchange resin catalysts. Ion-exchange resins have a polymer matrix that contains ion exchange centers filled with ionic functional groups on the surface. Ion exchange resins are generally distinguished into cationic and anion exchange resins, although other types of ion exchange resins are available. For a more complete description of cation and anion exchange resins, see de Dardel, F. and Arden, TV Ion Exchangers in Ullman's Encyclopedia of Industrial Chemistry (2005).

Suitable polymeric matrices for the ion exchange resin catalyst include polystyrene matrix, polyacrylic matrix, polyalkylamine resin, as well as other polymeric materials. Preferably, the polymeric matrix is cross-linked with divenylbenzene sufficiently to increase operating power, but without increasing the density of the ion exchange material to such an extent that the ion exchange material becomes too physically hard and too chemically resistant to chemical processing. Preferably, the matrix is a copolymer of styrene and divinylbenzene.

Attached to the regions of the polymer matrices described above are ionic stationary groups which determine whether the resin functions as a cationic or anionic ion exchange resin. In solutions, the positive or negative charge of stationary groups is compensated by ions with the opposite charge, which are called here functional groups.

Strongly acidic cationic ion exchange resins typically include sulfonic groups as stationary groups, in turn attached to the styrene-divenylbenzene polymer matrix. Examples of strongly acidic sulfonic acid cation exchange resins include Amberlite IR 120, Dowex HCR, Lewatit S 100 and Amberlyst 15, among others.

Cationic ion exchange resins may also include a polymeric material that is a copolymer obtained by addition polymerization of acrylic or methacrylic acid and divinylbenzene, such as described in US patent No. 3,957,698. Other suitable polymeric materials for cation exchange resins include a polymeric matrix obtained by crosslinking a polyvinylpyridine resin with divenylbenzene. Such materials are available under the brand name Reillex HPQ. Sulfonated phenolic polymer resins are also suitable cationic ion exchange resins.

Weak acid cationic ion exchange resins typically include carboxyl groups as stationary groups. Examples of weak acid cation exchange resins include, Amberlite IRC 86, Dowex Mac-3, Lewatit CNP, among others.

In the present invention, the strongly basic anion exchange resins contain quaternary ammonium stationary groups. They are further classified into type I, obtained by the reaction of trimethylamine with a styrene-divenylbenzene copolymer after chloromethylation, and type II, obtained by the reaction of a styrene-devenylbenzene copolymer with dimethylethanolamine. Suitable examples of such Type I resins include Lewatit MP 500 available from Lanxess and Amberlyst 26 and Amberlite IRA 402 and IRA 410 available from Dow. Suitable examples of strongly basic type II resins include, for example, Purolite A510S (Purolite Corporation).

Weak base anion exchange resins typically include polyacrylic resins that have received stationary groups by reaction with a polyfunctional amine to form anion exchange resins such as the weak base tertiary ammonium resins Amberlite IRA 67 and Amberlyst 21 (available from Dow). Of particular note, this ion exchange resin can then be further treated with chloromethane or dimethyl sulfate to form a Type I strong base quaternary amine resin Amberlite IRA 458 (Dow). Weak base anion exchange resins may also include a free base group as a stationary group, such as Amberlite IRA-67 resin (Dow).

In a particular embodiment of this invention, the strongly basic anion exchange resin contains a "bonding" group of 3-7 bonding atoms, in particular 3-5 carbon atoms between the quaternary ammonium stationary group and the benzene group of the polymer matrix/material. An example of such a link chain in a strong base resin is illustrated in US Pat. No. 5,945,568 and was produced under the name Diaion TSA1200 (Mitsubishi Chemical). As stated in the '568 patent, the linking group that links the stationary quaternary ammonium group to the benzene group of the polymer matrix is not particularly limited as long as it is long enough. Suitable examples of the linking group are an alkylene group or an alkyleneoxyalkylene group. A preferred example for use in this invention is an anion exchange resin which contains, as a substrate, a polymer of a vinylaromatic compound, and which has such a structure that the quaternary ammonium group is linked to the aromatic group through a linking group having a chain length of at least 3.

Without being limited by theory, it is believed that in this invention, this linking group prevents nucleophilic attack on the ammonium stationary group from the benzene ring, which would otherwise lead to accelerated deactivation and degradation of the ion exchange resin catalyst and reduced catalyst stability and life. .

A variety of ions are suitable for use as a functional group in the anionic resins of this invention and may be selected from the group consisting of bicarbonate, carbonate, hydroxide, bisulfite, carboxylate, citrate and metalate, and molybdate anions. These functional groups can be used with any of the stationary groups and polymeric material mentioned above.

In the present invention, it is preferable that the ion exchange resin catalyst belongs to the group of strongly basic Type I anion exchange resins, more preferably, strongly basic Type I anion exchange resins with a bicarbonate or monocitrate functional group, and most preferably, strongly basic Type I anion exchange resins with a bicarbonate or monocitrate functional group. with a link group.

In addition to the ion exchange materials mentioned above, any other suitable cation or anion exchange resin may be used, such as the cation and anion exchange resins mentioned above in de Dardel, F. and Arden, TV 2008, Ion Exchanger , in Ullman's Encyclopedia of Industrial Chemistry .

In addition to the above ion exchange material, a combination of a homogeneous and a heterogeneous catalyst, such as a solid catalyst on which a soluble metal catalyst is adsorbed, can also be used. In this case, a solid catalyst is first prepared with a metal catalyst adsorbed thereon. Then, during the liquid phase hydration reaction described above, the metal catalyst is desorbed from the solid catalyst and can effectively catalyze the reaction. Then the metal catalyst must be separated from the reaction products and can be reused for re-adsorption on the solid catalyst. The solid catalyst in this case may be an ion exchange resin.

The above paragraphs relate to the selection of a suitable ion exchange resin catalyst for use in the catalytic reactors in accordance with the present invention. With respect to the operation of the catalytic reactors containing the ion exchange resin, in Figures 2 and 3 the feed flows from the top of the reactor down through each of the reactors. This is called the "downflow" mode, but in the present invention, the reactors can be operated in "downflow" or "upflow" modes. Downflow processes have the advantage of increasing the catalyst density within the reactor and thus reducing the size and cost of the reactors themselves. Downflow operation also minimizes the non-catalytic reaction of water and ethylene oxide to ethylene glycol. However, during downflow operation, the selectivity of the catalyst can be degraded, both due to local inhomogeneities and impurities that can form and be trapped in the catalyst bed, and due to "canalization" that develops in the catalyst bed.

As can be expected, in upflow operation, the feed stream moves from the bottom of the reactor to the top. PCT Publication No. WO2008/150338A1 reports an increase in the stability and service life of an ion exchange resin in upflow operating mode. In the present invention with liquid phase reactants and solid catalyst, upflow operation can give a moderate degree of fluidization of the bed during upflow operation, which increases the proportion of voids between adjacent resin particles, allowing to reduce inhomogeneities and impurities that are otherwise embedded and trapped in the catalyst bed. during downflow operation and degrade catalyst performance. US Pat. No. 6,160,187 has previously underestimated the performance of upflow catalytic reactors because the resulting bed fluidization supposedly causes more catalyst attrition and reduced monoethylene glycol selectivity due to longitudinal agitation. However, it has been found in the present invention that in the liquid phase process of the present invention these criticisms are unrealistic because they are more likely to occur in gas/solid catalyst systems where there are much higher velocities, large differences in densities (between the gaseous reactant and the solid catalyst, according to compared to liquid reactant and solid catalyst) and thus much greater bed expansion than in the liquid/solid catalyst system of the present invention. Thus, upflow operation remains a suitable mode of operation in accordance with the present invention.

Silver-based epoxidation catalyst.

The silver-based epoxidation catalyst includes a support and at least a catalytically effective amount of silver or a silver-containing compound; an activating amount of rhenium or a rhenium-containing compound is also optionally present; an activating amount of one or more alkali metals or alkali metal-containing compounds is also optionally present. The substrate used in accordance with this invention may be selected from a wide variety of solid refractory substrates, which may be porous and may provide a preferred porous structure. It is well known that alumina is useful as a catalyst support for olefin epoxidation and is the preferred support.

Regardless of the nature of the support used, it is usually in the form of particles, large pieces, small pieces, granules, rings, spheres, wagon wheel, baffled hollow cylinders and the like, the size of which is suitable for use in a fixed bed epoxidation reactor. The substrate particles will preferably have equivalent diameters in the range of about 3 mm to about 12 mm, and more preferably in the range of about 5 mm to about 10 mm. (The equivalent diameter is the diameter of a sphere having the same ratio of outer surface (i.e., excluding surface in the pores of the particle) to volume as the particles of the substrate used.) Suitable substrates are available from Saint-Gobain Norpro Co., Sud Chemie AG , Noritake Co., CeramTec AG and Industrie Bitossi S.p.A. Without being limited to the specific compositions and formulations contained herein, additional information on support compositions and methods for preparing supports can be found in US Patent Publication No. 2007/0037991.

To prepare a catalyst for oxidizing an olefin to an olefin oxide, a catalytically effective amount of silver is then applied to the surface of a support having the above characteristics. In one embodiment, the catalytically effective amount of silver is from 10% by weight to 45% by weight. The catalyst is prepared by impregnating the support with a silver compound, complex or salt dissolved in a suitable solvent sufficient to cause precipitation of the silver precursor compound onto the support. Preferably, an aqueous solution of silver is used.

An activating amount of the rhenium component, which may be a rhenium-containing compound or a rhenium-containing complex, may also be applied to the substrate before, simultaneously with, or after the silver is deposited. The rhenium promoter may be present in an amount of from about 0.001 wt%. up to about 1% wt., preferably from about 0.005% wt. up to about 0.5% of the mass. and more preferably, about 0.01% of the mass. up to about 0.1% of the mass. calculated on the weight of the entire catalyst, including the substrate, in terms of metallic rhenium.

Other components that may also be applied to the substrate before, simultaneously with or after the deposition of silver and rhenium are activation amounts of an alkali metal or a mixture of two or more alkali metals, as well as optional activation amounts of a Group IIA alkaline earth metal component or a mixture of two or more components a Group IIA alkaline earth metal, and/or a transition metal component, or mixtures of two or more transition metal components, all of which may be in the form of metal ions, metal compounds, metal complexes, and/or metal salts, dissolved in a suitable solvent. The support may be impregnated simultaneously or in separate stages with different catalyst promoters. A particular combination of support, silver, alkali metal promoter(s), rhenium component and optional additional promoter(s) in accordance with this invention provides an improvement in one or more catalytic properties over the same combination of silver and support in the absence or use of only one of promoters.

As used herein, the "activating amount" of a particular catalyst component refers to the amount of that component that works effectively to improve the catalytic performance of a catalyst compared to a catalyst that does not contain that component. The exact concentrations used will, of course, depend on the desired silver content, the nature of the support, the viscosity of the liquid, and the solubility of the particular compound used to deliver the promoter to the impregnation solution, among other factors. Examples of catalytic properties include, among others, functionality (resistance to degradation), selectivity, activity, conversion, stability, and yield. One skilled in the art will appreciate that one or more individual catalytic properties may be improved by an "activating amount" while other catalytic properties may or may not be improved or may even be reduced.

Suitable alkali metal promoters may be selected from lithium, sodium, potassium, rubidium, cesium, or combinations thereof, with cesium being preferred, and combinations of cesium with other alkali metals being particularly preferred. The amount of alkali metal deposited or present on the support must be an activating amount. Preferably, the amount varies from about 10 ppm. up to about 3000 ppm, more preferably from about 15 ppm. up to about 2000 ppm and even more preferably from about 20 ppm. up to about 1500 ppm and particularly preferably from about 50 ppm. up to about 1000 ppm mass from the entire catalyst, in terms of metal.

Suitable alkaline earth metal promoters include elements from Group IIA of the Periodic Table of the Elements, which may be beryllium, magnesium, calcium, strontium and barium, or combinations thereof. Suitable transition metal promoters may include elements from Groups IVA, VA, VIA, VIIA and VIIIA of the Periodic Table of the Elements and combinations thereof.

The amount of alkaline earth metal promoter(s) and/or transition metal promoter(s) deposited on the support is the activating amount. The transition metal promoter may typically be present in an amount of from about 0.1 micromoles per gram to about 10 micromoles per gram, preferably from about 0.2 micromoles per gram to about 5 micromoles per gram.

The silver solution used to impregnate the support may also contain an optional solvent or complexing/solubilizing agent known in the art. A variety of solvents or complexing/solubilizing agents can be used to solubilize the silver to the desired concentration in the impregnation medium. Suitable complexing/solubilizing agents include amines, ammonia, oxalic acid, lactic acid, and combinations thereof. Amines include alkylenediamine containing from 1 to 5 carbon atoms. In one preferred embodiment, the solution contains an aqueous solution of silver oxalate and ethylenediamine. The complexing/solubilizing agent may be present in the impregnating solution in an amount of from about 0.1 to about 5.0 moles per mole of silver, preferably from about 0.2 to about 4.0 moles, and more preferably from about 0.3 to about 3 .0 mole for every mole of silver.

When a solvent is used, it may be an organic solvent or water and may be polar or substantially or completely non-polar. In general, the solvent must have sufficient solvating power to dissolve the components of the solution. At the same time, it is preferred that the solvent be chosen so as to avoid excessive influence on or interaction with the solvated promoters. Organic based solvents containing from 1 to about 8 carbon atoms per molecule are preferred. Mixtures of several organic solvents or mixtures of organic solvents with water may be used, provided such mixed solvents function as desired herein.

The concentration of silver in the impregnating solution is typically from about 0.1% by weight up to the maximum solubility provided by the specific solvent/solubilizing agent combination used. It is generally very convenient to use solutions containing from 0.5% to about 45% by weight silver, with concentrations of 5 to 35% by weight silver being preferred.

The impregnation of the selected substrate is carried out by any of the conventional methods; e.g. excess solution impregnation, moisture capacity impregnation, spray application, etc. Typically, the substrate material is placed in contact with the silver containing solution until a sufficient amount of the solution is absorbed by the substrate. Preferably, the amount of silver-containing solution used to impregnate the porous substrate is not more than is necessary to fill the pores of the substrate. A single impregnation or a series of impregnations with or without intermediate drying may be used, depending in part on the concentration of the silver component in the solution. Impregnation methods are described, for example, in US Pat. Techniques known in the art for pre-precipitation, co-precipitation and post-precipitation of different promoters can be used.

After the substrate has been impregnated with the silver compound, i.e., the silver precursor, the rhenium component, the alkali metal component, and optional other promoters, the impregnated substrate is fired for a time sufficient to convert the silver compound to active silver species and to remove volatile components from the impregnated substrate to obtain catalyst precursor. The firing may be performed by heating the impregnated substrate, preferably gradually, to a temperature in the range of about 200° C. to about 600° C. under a pressure in the range of 0.05 to about 3.5 MPa. In general, the higher the temperature, the shorter the required heating period. A wide range of heating periods has been proposed in the art; for example, US Pat. No. 3,563,914 discloses heating for less than 300 seconds and US Pat. No. 3,702,259 discloses heating for 2 to 8 hours at 100°C to 375°C, typically for about about 8 hours. However, it is only important that the heating time correlates with the temperature so that all of the silver contained is converted to active silver species. For this purpose, continuous or stepped heating can be used.

During firing, the impregnated substrate may be exposed to a gas atmosphere containing an inert gas or an inert gas mixture with from about 10 ppm. up to 21% by volume oxygen-containing oxidizing component. For the purposes of this invention, an inert gas is defined as a gas that does not substantially react with a catalyst or catalyst precursor under the conditions chosen for firing. Further information on the manufacture of the catalyst can be found in the aforementioned US Patent Publication No. 2007/0037991.

For illustrative purposes only, the following conditions are often used in modern commercial ethylene oxide reactor plants: gas hourly space velocity (GSP) 1500-10000 h ^-1 , pressure at the reactor inlet from 1 MPa to 3 MPa, coolant temperature 180 -315°C, oxygen conversion rate 10-60% and EO production rate (operating rate) 100-350 kg EO/m ³ catalyst/h and ethylene oxide, DEO concentration change from about 1.5% to about 4.5% . The feedstock composition at the reactor inlet after startup is complete and during normal operation typically contains (in % by volume) 1-40% ethylene, 3-12% O ₂ ; 0.2%-10%, preferably 0.2%-6%, more preferably 0.2%-5% CO ₂ ; 0-5% ethane, some of one or more chloride moderators as described herein; and the rest of the raw material is argon, methane, nitrogen, or mixtures thereof.

Claims (33)

1. Liquid-phase method for producing ethylene glycol, including: obtaining an aqueous ethylene oxide feed stream containing ethylene oxide and water; combining the catalytic hydration feed stream with the recycled split stream to form a first reactor inlet stream; feeding the inlet stream of the first reactor to the inlet of the first adiabatic reactor, where the inlet of the first adiabatic reactor has a first inlet temperature; reacting ethylene oxide and water in the presence of a first ion exchange resin catalyst in a first adiabatic reactor to produce an effluent containing water, ethylene glycol and unreacted ethylene oxide; further combining the effluent stream with the recirculated feed stream to form a combined stream containing water, ethylene glycol and unreacted ethylene oxide; feeding the combined stream to the inlet of the second adiabatic reactor, where the inlet of the second adiabatic reactor has a second inlet temperature; reacting ethylene oxide and water in the presence of a second ion exchange resin catalyst in a second adiabatic reactor to produce a second reactor effluent containing water, ethylene glycol and unreacted ethylene oxide; compressing the second effluent from the reactor; dividing the second reactor effluent into a recycle stream and a front stream; dividing the recirculated stream into a recirculated split stream and a feeding recirculated stream; feeding the front stream in the form of a tubular reactor feed stream to a non-catalytic tubular reactor to obtain a product stream in which the total concentration of diethylene glycol (DEG), triethylene glycol (TEG) and higher glycols in the product stream is higher than the total concentration of DEG, TEG and higher glycols in the front stream. 2. Liquid phase process according to claim 1, wherein the temperature at the inlet of the non-catalytic tubular reactor is between 130°C and 160°C. 3. Liquid phase process according to claim 1, wherein the inlet temperature to the first adiabatic reactor is between 50°C and 90°C. 4. Liquid phase process according to claim 1, wherein the outlet temperature of each adiabatic reactor is between 90°C and 120°C. 5. Liquid phase process according to claim 1, wherein the molar ratio of water:ethylene oxide in the first inlet stream of the reactor is from 40:1 to 10:1. 6. Liquid phase process according to claim 5, wherein said water:ethylene oxide molar ratio in the first reactor inlet stream is between 30:1 and 20:1. 7. Liquid-phase method according to claim 1, carried out continuously. 8. The liquid phase process of claim 1 wherein the total water:ethylene oxide molar ratio in the aqueous ethylene oxide feed stream is between 5:1 and 15:1. 9. The liquid phase process of claim 1, wherein the total water:ethylene oxide molar ratio in the aqueous ethylene oxide feed stream is between 7:1 and 12:1. 10. Liquid phase process according to claim 1, wherein the concentration of ethylene oxide in the forward flow is not more than 1 mole%. 11. Liquid phase process according to claim 1, wherein the percentage of conversion of ethylene oxide to monoethylene glycol in the first adiabatic reactor is at least 50%. 12. The liquid phase process of claim 1, wherein the aqueous ethylene oxide feed stream is produced according to the following steps: obtaining a stream of water from the enrichment cycle containing ethylene oxide, methane, ethylene and other dissolved light gases; separating, in the flash drum, the dissolved light gas vapor from the water stream from the enrichment cycle; upwardly directing the dissolved light gas vapor through an opening in the flash drum in fluid communication with an absorber mounted on the top surface of the flash drum to form a vapor stream from the top of the absorber; heating water from the enrichment cycle, obtained from the bottoms of the flash drum, and pumping it into the desorber; And separation into: (1) an enriched liquid stream from the top of the desorber containing at least 40 mol.%; and (2) a lean aqueous cycle solution in the stripper bottoms containing 5 to 50 molar ppm ethylene oxide. 13. The liquid phase process of claim 12, wherein said rich liquid stream from the top of the stripper contains at least 50 mole % ethylene oxide. 14. The liquid phase process of claim 1, wherein the ion exchange resin catalyst is a type I strongly basic anion exchange resin. 15. The liquid phase process of claim 1 wherein the ion exchange resin catalyst has a bicarbonate or monocitrate functional group. 16. The liquid phase process of claim 1 wherein the ion exchange resin catalyst comprises a linking group. 17. The liquid phase process of claim 1, wherein the first reactor inlet and the second reactor inlet are substantially free of carbon dioxide. 18. Liquid phase process according to claim 1, wherein the inlet temperature to the second adiabatic reactor is between 70°C and 110°C.

Images