JPS6116250B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for selectively separating ethylbenzene and P-xylene from a mixture of xylene isomers using a novel selective adsorption separation method.

Conventionally, various methods have been proposed as selective adsorption separation methods using adsorbents. When separating each isomer from a xylene isomer mixture, P-xylene
In many cases, only PX is separated and recovered as a product. In this case, O-xylene OX, m-xylene MX, and ethylbenzene EB are used as non-adsorptive components, and these three components are passed through an isomerization reaction tower, and only PX is retained as an adsorbent component in zeolite to obtain the product. Method (Method 1), or method (Method 2) in which EB is removed by ultra-precision distillation before being supplied to the zeolite separation tower, OX and MX are passed through the isomerization reaction tower, and PX is recovered as an adsorbed component. .

On the other hand, a method for simultaneously obtaining EB or OX in addition to PX is to separate and recover OX from the stream flowing out from the isomerization reaction tower and the raw material xylene stream by distillation, and then use MX as a non-adsorbable component in the isomerization reaction tower. and retain PX and EB as adsorbent components in the first separation column, and further separate the mixture of PX and EB recovered from the first separation column in the second separation column,
A method (Method 3) of producing a product containing both PX and EB has been disclosed (Special Publication No. 52-29300,
−29730).

In methods 1 and 2 of the above three methods,
Since OX, MX, and EB are non-adsorbable components and PX is an adsorbent component, the selection coefficient that determines the difficulty of separation is K.
^PX _EB , K ^PX _{MX (OX)} , etc. (MX and OX usually exhibit almost the same adsorption force, so they are represented by MX), and the smallest value among them determines the degree of separation. In addition, in method 3, the selection coefficient that determines the degree of separation in the first separation column is K ^EB _{MX (OX)} , and in the second separation column K ^P
X _EB governs the degree of separation. Therefore, prior art method 1
And as in the case of method 2, in order to include EB in the MX, (OX) band, K ^PX _{MX (OX)} 〓K ^PX _EB (
K ^EB _MX 〓1)
For example, since a zeolite with a relatively low potassium content is used, the separation has to be carried out under such conditions that the selection coefficient is also low. Furthermore, in method 3, K ^EB _MX(OX) 〓K ^PX _MX (
K ^PX _EB 〓1)
It is necessary to separate them under the condition of a small selection coefficient. All of these conventional methods have the disadvantage that they do not utilize the large area of K ^PX _MX and must perform separation with relatively low selectivity of K ^PX _EB and K ^EB _MX . When the separation is carried out under small selectivity,
The travel distance required to obtain the desired degree of separation is large, and the amount of zeolite used is enormous.
Since the traveling distance is long, the degree of intrusion of the developer into the xylene adsorption zone increases, resulting in a disadvantage that the concentration of xylene recovered is low and the distillation energy required to recover the developer is large.

An object of the present invention is to eliminate the above-mentioned drawbacks of the prior art and provide an advantageous separation method for ethylbenzene and P-xylene by employing a separation method different from the above-mentioned conventional separation methods.

That is, the present invention provides at least one of O-xylene and m-xylene, ethylbenzene and P
- xylene isomer mixture containing xylene
Ethylbenzene and P-xylene are selected using a zeolite of phasiasite type with a SiO ₂ /Al ₂ O ₃ ratio (mol) of 4 or more and in which 60% or more of the exchangeable cation sites are replaced with potassium as an adsorbent. In the method of adsorbing and separating O-xylene (and/or m-
xylene), ethylbenzene, and P-xylene in the order of adsorption separation, and within the adsorption zone, bands A to E containing the following isomer components are formed in this order in the development direction, and these are separated and recovered at the exit of the separation column. A: o-xylene (and/or m-xylene) B: o-xylene (and/or m-xylene) and ethylbenzene C: o-xylene (and/or m-xylene) ethylbenzene and p-xylene D : Ethylbenzene and p-xylene E: p-xylene p-xylene is obtained as a product from E, and the mixture recovered from D is further selectively adsorbed and separated in the same manner in a second separation column filled with the above zeolite. Together, ethylbenzene and p-xylene are obtained as products, and each mixture recovered from A to C can be further separated, recycled to the feedstock, converted to ethylbenzene and p-xylene by an isomerization operation, or a combination thereof. Provided is a method for separating ethylbenzene and p-xylene, which is characterized in that each isomer is obtained as a product through operations.

One of the features of the process of the present invention is that in the first separation column packed with zeolite, which supplies a mixture feed containing at least three isomer components of OX (and/or MX), EB and PX, conventional separation The purpose of this method is to adopt an adsorption/separation pattern for each component that is different from that used in the conventional method. FIGS. 1 (1) to (3) show the distribution state of the adsorption zone of each component in the direction of development (arrow) of the developing agent in the above-mentioned conventional methods 1 to 3, respectively. Figures 2 (1) to (3) show the distribution of adsorption zones used in the present invention. As is clear from the comparison between Figures 1 and 2, unlike the conventional method, the EB
The adsorption zone of is located between the adsorption zones of MX (or OX) and PX. By using such a separation pattern and employing the above separation operations, isomerization reactions, etc., efficient separation of EB and PX can be achieved.

As shown in Figure 2 (1), when each zone is separated and recovered at a point when a considerable amount of zone C, which has a low degree of separation and contains all of the raw material components, exists, the mixture recovered from zone C is It is desirable to recycle and reuse it as the feedstock for the first separation column.

OX and/or MX recovered from zone A can be used as a product as is if necessary, but in order to increase the yield of EB and PX, at least a part of it can be passed through an isomerization reaction column. It is preferable to convert at least a part of it into EB and/or PX, and to recycle and reuse the effluent containing the components converted in the isomerization reaction column as the feedstock for the first separation column.

The mixture recovered from zone B is subjected to selective adsorption separation in a third separation column packed with zeolite to obtain EB as a product, and OX and/or MX to A.
can be combined with the mixture recovered from the zone and fed to the isomerization reactor. It is also possible to immediately send the mixture recovered from zone B to the isomerization reaction column together with the components recovered from zone A without separating it.

In order to realize the method of the present invention, the adsorbents and separation conditions that have the highest selective adsorption power may be selected and used in the order of PX, EB, and MX (or OX). In particular, as an adsorbent with a high selective adsorption coefficient K ^PX _{MX (OX) between PX and MX (OX)} , phosiasite type zeolite in which 60% or more, preferably 90% or more of the cation sites are replaced with potassium, more preferably It is preferable to use phosiasite type zeolite or the like which has been acid-treated by the method described in Japanese Patent Application No. 53-4244. At this time, K ^PX _EB , K ^EB _{MX (O
X)} etc. are K
Since it will not improve to the same level as ^PX _MX , (K ^PX _E
B ,K
^EB _MX ≪K ^PX _MX ) The separated peak of EB is exactly MX
It will be located in the center of the separated peaks of (OX) and PX. By dividing this peak containing EB into two or three equal parts, and reseparating the mixed zone B of MX(OX) and EB and the mixed zone D of EB and PX in the third separation column and the second separation column, PX and EB as a product at the same time, and compared to conventional methods, the separation efficiency of the entire system can be increased and separation costs can be reduced. That is, the first
By performing separation under conditions that maximize the separation coefficient K ^PX _MX of the separation column, the desired degree of separation can be achieved with a small amount of zeolite packed, and by recovering the product with a high xylene isomer concentration, the distillation of the developing agent can be achieved. Recovery costs can be reduced through separation. As a result of this improvement, the second and third separation towers
Although the EB collection process is involved and the number of devices increases slightly, overall PX and EB can be separated at a lower cost than before.

The chromatographic separation method of the present invention is to form a xylene adsorption zone by supplying a developing agent and then a xylene isomer mixture to a separation column filled with zeolite, and then moving the xylene adsorption zone by supplying the developing agent again. and at the separated material acquisition port,
A method is preferably adopted in which the total amount is divided into fractions according to the composition and collected. The separation status of each xylene isomer component at the separation material acquisition port varies depending on the zeolite selectivity, developer, development conditions, amount of adsorbed xylene, and travel distance, but for a certain zeolite, developer, and development conditions. In this case, the degree of separation is approximately determined by the amount of xylene adsorbed initially and the distance traveled. In other words, the elution curve representing the xylene concentration distribution at the separation port of xylene isomers is
MX moves forward to the intermediate component in the direction of eluent movement.
If EB and the adsorbed component PX are located at the rear, but there are many xylene isomers to be supplied now, or the moving distance is short, the same 4
Unseparated parts of the components remain. On the other hand, if the amount of xylene supplied is small or the distance traveled is large, the
As shown in Figure 2, the separation progresses too much and a component area containing only EB appears. In the former case, the unseparated zone C will need to be recycled together with the raw material xylene, and in the latter case, the amount of zeolite used will increase due to excessive separation. A preferred separation pattern is the second
As shown in Figure 3, it is desirable to determine the amount of xylene supplied and the moving distance so that the rear end of MX or OX and the front end of PX are substantially in contact with each other, thereby performing separation without any waste.

The chromatographic separation method in the third separation column and the second separation column for re-separating zone B or D obtained from the first separation column is, first, two-component separation, and
Selectivity K ^A _B (K ^PX _EB , K ^EB _{MX (OX)} ) is K
Since ^PX is lower than _MX , the separation pattern at the time of obtaining the separated product is different from that of the first separation column. In this case as well, the determination is made taking into consideration the isomer yield per hour, 0.8 average concentration, etc., but since the boundary area where the two components are only partially separated is relatively large, the peaks of each component are separated until they are completely separated. It is not necessarily advantageous to do so, and generally a method is adopted in which the two-component coexistence zone (three zones) is left and recycled to the raw material.
However, since the amount of C8 to be separated is smaller than that in the first separation column, the degree of freedom in the separation method is large, and it is not necessarily essential to divide it into the three zones mentioned above.

A specific combination of separation processes involves passing only the non-adsorbable components MX and OX, or including EB contained in the MX and OX zones, to an isomerization reaction column.
A closed system is adopted in which it is converted into PX and/or EB and recycled back to the first separation column.

In order to easily understand the method of the present invention, specific embodiments will be explained with reference to drawings. FIG. 3 shows a case where only the components recovered from the above-mentioned zone A are passed through the isomerization reaction column, and FIG. shows the case where components in both zones A and B are passed through the isomerization reaction column.

In FIGS. 3 and 4, the first separation tower 3, the second separation tower 22, and the third separation tower 21 are all adsorption separation towers filled with zeolite, and a xylene isomer mixture as a raw material is supplied from the upper part, Each component can be obtained separately from the lower acquisition port. The isomerization reaction tower 35 is OX
It is a reaction system that can convert a stream containing PX and MX, or a part of OX, MX, and EB to PX or EB.

In FIG. 3, a raw material stream 1 of a mixture of four xylene isomers is collected from a product stream 36 flowing out from an isomerization reaction column 35 and a stream 6 of an intermediate component C separated in a first separation column into a developer recovery column 11. After removing the developing agent, it is combined with stream 19 and fed to the first separation column 3. After that, each developer recovery tower 9, 1
Streams 37 of the developing agent recovered from 0, 11, 12, 29, and 30 are alternately fed, the xylene adsorption zone is moved and expanded, and the xylene isomers are divided into five zones A to E at the column outlet. Collect the entire amount. First, A is supplied to the developer recovery column 12, and after removing the developer, a stream 17 of OX and MX is sent to the isomerization reaction column 35. After the developer is removed from the B stream 5 in the developer recovery column 9, the MX, OX, and EB streams 18 are again separated in the third separation column 21 in the same manner. The C stream 6 is recycled to the feed xylene stream via the developer recovery column 11. After the developer is removed from the D stream 7 in the developer recovery column 10, the PX/EB stream 20 is sent to the second separation column 22. Flow 8 of E merges with flow 28, which will be described later. In the third separation column 21, MX (OX) and EB are separated in the same manner as in the first separation column using the supplied 18 consisting of MX, OX, and EB and the recovered developing agent 38,
It is divided into three bands, F, G, and H, and collected. MX,
F consisting of OX is combined with the above-mentioned stream 4 as a stream 23, passes through a developing agent recovery column 12, and is sent to an isomerization reaction column 3.
G flow 2 is sent to 5 and consists of MX, OX, and EB.
4 is combined with 5 supplied to the developer recovery column 9 and recycled, and the H stream 25 consisting of EB is separated in the second separation column 22 and becomes the I stream 26 consisting of EB.
The developer is removed by the developer recovery tower 29, and the EB is converted into a stream 33 to be used as an EB product. Similarly, a stream 2 consisting of PX and EB is sent to the second separation column 22.
The I stream 26 consisting of EB is divided into three bands I, J, and K by 0 and the developer 39, and the I stream 26 consisting of EB is converted into an EB product together with 25 as described above, and the J stream 27 consisting of PX and EB is divided into the developing agent recovery column 10. It is combined with the entering stream 7 and recycled as a raw material for the second separation column 22,
The K stream 28 consisting of PX is combined with the E stream 8 flowing out from the first separation column, and the developer is removed in a developer recovery column 30, resulting in a PX product as a stream 34. Recovered developer 13, 14, 15, 1
6, 31, 32 are collected and become 37, 3 respectively
8,39 is fed to separation columns 3, 21, 22 as streams. Reused as a developing agent.

In the splitting system in which intermediate component C is not present in the first separation column (FIG. 2-(3)), stream 6, developer recovery column 11, and effluents 15 and 19 are excluded, but the rest is the same.

In the separation process in which components A and B obtained in the first separation column are sent together to the isomerization reaction system (see Figure 4), the separation column,
Although the number of developer recovery towers can be reduced by one each, EB cannot obtain the entire amount as a product.

That is, mixed xylene feed stream 1, isomerization reaction column 3
The product stream flowing out from 5 and the stream 19 after removing the developer in the developer recovery column 11 from the stream 6 of the intermediate component C separated in the first separation column are combined and supplied to the first separation column 3. Ru. Thereafter, the developer stream 37 recovered from each developer recovery tower 10, 11, 12, 29, 30 is alternately supplied with xylene to the first separation tower, and the xylene adsorption zone is moved and developed. At the tower outlet, the data is divided into 5 bands A to E, of which 4 bands including both A and B are obtained.
The flow is (MX, OX, EB), and the developer recovery tower 1
After removing the developing agent in step 2, it is sent to an isomerization reaction tower. The C stream 6 passes through a developer recovery column 11 to remove the developer, and is then recycled to raw material xylene through a stream 19.

Similarly to FIG. 3, in zone D (PX, EB), the developer is removed as stream 7 in the developer recovery tower 10, and then the stream 20 containing PX and EB is sent to the second separation tower 22.
sent to. Zone E (PX) is combined with PX obtained from the second separation column, which will be described later. In the second separation column, the three bands I, J, and K including EB, EB+PX, and PX are obtained separately. Stream 26 of I containing EB has the developer removed by developer recovery column 29 to obtain product EB as stream 33. Zone J, containing PX, EB, joins stream 7 as stream 27 and joins the feed stream of the second separation column. Band K is flow 28
The product PX is combined with the obtained PX8 from the first separation column described above as a stream 34, and after the developer is removed in a developer recovery column 30, the product PX is obtained as a stream 34.

The characteristics and advantages and disadvantages of sending only band A to the isomerization system as shown in Figure 3 and sending both bands A and B to the isomerization area as shown in Figure 4 are as follows: Since only MX and OX are sent to the isomerization reaction tower, the amount of feed is small, so the heat load is small. Furthermore, since EB is not included as a raw material, there is no need to use an expensive noble metal supported catalyst. Furthermore, although it has the advantage of being able to use all the EB produced by the isomerization reaction with the raw material xylene as a product, it requires three adsorption separation towers and six developer recovery towers.

In the latter method, some of the EB is mixed into the isomerization reaction tower, resulting in a slightly larger amount of feed and an increased heat load. Further, the isomerization catalyst requires a noble metal supported catalyst, and a portion of the EB is circulated through the first separation column and the isomerization system, resulting in a decrease in production. On the other hand, the number of adsorption separation towers will be two, one less than the former, and the number of developer recovery towers will also be five.

Comparison of these two methods is from an economic point of view, because the second and third separation towers and the accompanying developer recovery tower are small, so it is possible to obtain a large amount of EB without increasing the construction cost. Generally speaking, the former method can be said to be somewhat advantageous. In addition, the choice between the two is determined by environmental factors such as demand for EB.

In Figures 3 and 4, the first, second, third
The developing agent used in the separation tower was a common substance, but
It goes without saying that, in general, the developing agent used in each separation column may be different. However, if a common developer is used, it is possible to allow a relatively large concentration of developer to be mixed into the xylene isomer, which is used as a raw material for the next column, when removed in the developer recovery column. It was found that the developing agent used in the first separation column does not cause deterioration of the zeolite in the second separation column or other adverse effects, and is economical and has excellent operability. For example, if the developer recovery column is a distillation column, even if a few percent of the developer is mixed into the xylene product flowing out from the top or bottom of the column, it will not affect the second and third separation columns. is small and can reduce distillation costs.

Preferred specific examples of developing agents used in carrying out the method of the present invention include the following: R-O-R' (R, R': n having 3 to 5 carbon atoms excluding the butyl group) alkyl group), cyclic ethers such as furan, 2-methylfuran, hexamethylene oxide, and cineole; methylene chloride, chloroform, ethyl chloride,
1,2 dichloropropane, 1,1,2,2-tetrachloroethane, 1,2,3-trichloropropane, 1-chloropentane, m-dibromobenzene,
Halogen-substituted hydrocarbons such as p-chlorotoluene; p-ethyltoluene, p-cymene, p-diethylbenzene, m-diethylbenzene, cumene, t-
Butylbenzene, alkyl-substituted aromatic hydrocarbons such as toluene, ethyl sulfide, etc.

In the chromatographic separation method of the present invention, 1
The amount of xylene isomers adsorbed at the time is determined by the selectivity between xylene isomers, the raw material composition, and the zeolite.
Since it varies depending on factors such as C8 isomer adsorption capacity, migration distance, zeolite packing density, and temperature, it is necessary to always consider the optimum C8 isomer supply amount for the system. Very standard conditions e.g. PX20%, EB20
%, 40% MX, and 20% OX using a 10 m Y-type zeolite column in which 95% potassium was substituted.
When separating at 100℃, the xylene adsorption amount W _x (ml/g-zeolite) per time is 0.10 to 0.5 ml/g-zeolite per unit weight of the zeolite used (total zeolite to which the C8 adsorption band has moved). Selected by range. If the selectivity K ^A _B between isomers decreases, _W Other operating conditions will also vary W _x to some extent, but generally W _x is selected between 0.04 and 2 ml/g zeolite. Within this range, the amount of C8 isomer to be supplied to obtain the separation pattern as shown in (3) in Figure 2 is within a very limited range, and the amount to be supplied is usually determined by model experiments, but under standard conditions it is 0.10 to 0.50 ml. /g Zeolite is often used.

On the other hand, the supply amount of the developing agent is also influenced by each of the above factors, and more particularly, it is greatly influenced by the diffusion situation at the interface and the type of the developing agent. In order to minimize the amount of zeolite used and perform efficient separation, one xylene adsorption zone and the next
In the separation column, a separation pattern is obtained in which the rearmost part of the interface of the PX peak and the leading part of the interface of the MX (OX) peak of the next xylene adsorption zone just touch or partially overlap (within the allowable spec). is most preferable. When selecting the amount of developer to be supplied, according to this method, it is desirable to leave a portion where no xylene is adsorbed and only the developer is adsorbed, so that the zeolite is not wasted.
In this case, the amount of developer supplied is determined by multiplying the amount of xylene supplied by 1-C8 isomer average concentration/C8 isomer average concentration (usually the C8 isomer average concentration is in the range of 20-90%). .

The setting of the travel distance of the C8 isomer adsorption zone, that is, the length of the column, is determined by the C8 isomer concentration at the time of product acquisition and the C8 isomer separation amount yield per unit time. In other words, in general, when the travel distance is increased, the product C8 isomer concentration increases up to a certain distance, and then remains constant or tends to decrease slightly (the distance at which the C8 isomer concentration becomes maximum is the smaller K ^A _B is). on the other hand, the time to obtain the product increases and the yield decreases. These two contradictory
The travel distance is ultimately determined by economic calculations based on model experiments, taking into account factors, but is generally set at a travel distance of 2 to 50 meters.

The development speed of the C8 isomer adsorption zone or the flow rate of the developing solution can be determined arbitrarily, but in systems with relatively large K ^A _B , considerable separation is achieved faster than the spread of the interface, so the higher the flow rate, the faster the unit time. The amount of separation per unit increases. However, the pressure loss of the zeolite layer increases, and it is not economical to increase it significantly.
The optimum flow rate can be considered in a system where K ^A _B is small. The superficial velocity is usually chosen between 0.5 m/h and 80 m/h.

The operating temperature is preferably as high as possible from the viewpoint of adsorption/desorption rate, and relatively low as it is from the standpoint of adsorption capacity.

The temperature is generally selected within the range of 20° to 200°C, preferably 40 to 160°C. When developing a low-boiling-point developing agent above its boiling point, it is necessary to maintain the entire column system at a level higher than the sum of its vapor pressure and xylene vapor pressure, and the inlet pressure of the liquid is further reduced by the pressure loss of the zeolite packed bed. is added.

The separation column used in the present invention is a container that can be filled with zeolite to form a fixed bed,
Usually, a cylindrical column is used, which has a liquid supply and intake port at one or more locations in the vessel. Preferably, a uniform dispersion plate and a uniform collection plate for the liquid are provided at the liquid inlet and the liquid acquisition port of the tower, respectively, to prevent liquid-liquid mixing, vortex flow,
The structure is designed to prevent remixing of separated substances due to such occurrences. Furthermore, when the diameter of the packed column and the length of the packed bed (column length) are large, separation is performed not only by one column but also by multiple columns. At this time, the liquid outlet of one column is connected with the liquid inlet of the next column by piping, and the liquid is obtained from the last column.

In addition, it is necessary to install flow path switching valves (three-way valve, four-way valve, multi-way valve, etc.) for liquid switching between C8 and developing agent liquid, or for dividing sampling during product acquisition, and furthermore, it is necessary to install flow path switching valves (three-way valve, four-way valve, multi-way valve, etc.) for liquid switching between C8 and developing agent liquid, and for liquid flow piping and separation tower. It is also desirable to install a C8 isomer and/or developing agent concentration detector in a part of the tank and use it for detecting the C8 isomer adsorption zone, operating the switching valve, supplying raw materials, and extracting the product.

The zeolite used in the present invention needs to have high selectivity between xylene isomers, especially K
Regarding ^PX _{MX (OX)} , it is desirable to have at least 4 or more. In order to improve K ^PX _{MX (OX)} , firstly, the type of zeolite, secondly metal ion substitution, and thirdly
It is necessary to optimize activation conditions, etc. The zeolite preferably used is a so-called phasiasite type zeolite, which is a crystalline aluminosilicate and is characterized by β,26-hedron-() as a polyhedral cage type. In addition to natural phosiasite,
There are two types of synthetic products: so-called X-type and Y-type. According to the research results of the present inventors,
K ^PX _MX is high in Y-type synthesized zeolite with a large SiO ₂ /Al ₂ O ₃ ratio, and once synthesized, pretreatment
The highest K ^PX _MX was obtained with the modified faujasite in which the SiO ₂ Al ₂ O ₃ ratio was further increased. Specifically, as disclosed in Japanese Patent Application No. 53-4244, a method is used in which the synthesized Y-type zeolite is treated with an inorganic acid or an organic acid. It is desirable that the SiO ₂ /Al ₂ O ₃ ratio is at least 4.0 or more.

It has been known for a long time that the cation species occupying the exchangeable cation sites of the zeolite are related to the selectivity of K ^PX _MX , etc. In particular, potassium is the main component, and the remainder is sodium, lithium, rubidium, and cesium. , magnesium, calcium, strontium, barium, nickel,
It is replaced with at least one selected from the group consisting of alkali metals such as copper, silver, manganese, cadmium, thallium, and lanthanum, alkaline earth metals, and others. Of these, at least 60% potassium,
Zeolite containing preferably 90% or more with the remainder substituted with lithium, barium, thallium, or cesium is most preferably used.

The zeolite thus prepared is K ^PX _EB ,
^KEB _MX was also quite high, and the separation efficiency in the second and third separation columns was also sufficiently satisfactory.

The isomerization method itself used in the present invention can be a well-known method in detail. For example, for isomerization between xylene isomers, an isomerization reaction tower is filled with a catalyst mainly composed of silica-alumina or, if necessary, a catalyst in which precious metals and/or base metals are supported by vapor deposition or other methods using silica-alumina as a carrier. This is accomplished by feeding a catalytic reaction bed with the xylene isomer feed stream. Mutual isomerization between xylene isomers is carried out in the presence of the above catalyst,
The reaction is easily carried out with or without heating,
In principle, the composition of the main component after isomerization is determined by the equilibrium between the four isomers at the reaction temperature, regardless of the composition of the raw material feed stream, but in reality it is influenced by the composition of the feed stream. The degree to which a product stream with a composition deviating from equilibrium can be obtained mainly depends on the catalyst used,
It is determined by the composition ratio of ethylbenzene.
In the case of the present invention, the feed stream consisting of OX, MX and
From the feed stream consisting of OX, MX, EB, as much as possible
It is desirable to set conditions so that a product stream with a high composition of PX and EB can be obtained.

Although not shown in Figures 3 and 4 above, in addition to these main equipment, there is a distillation tower and adsorption separation for removing impurities other than xylene isomers contained in the product stream from the isomerization reaction tower. Of course, incidental equipment such as a distillation column for removing impurities in the raw material xylene, an adsorption separation column for removing impurities in the recirculation line of the developing agent, and a product purification column may be installed as necessary. Any known means can be used as these means.

By implementing the present invention, compared to the prior art,
It has the following advantages.

(i) In the present invention, since most of the supplied xylene is separated into its components, the amount of zeolite used can be significantly reduced compared to the prior art in which a large amount of unseparated xylene zone is circulated.

(ii) Since separation can be carried out under operating conditions where K ^PX _MX is maximum, even if the second and third separation columns are included, the zeolite filling amount, proportional costs, particularly developer recovery cost and construction cost can be reduced.

(iii) In addition to reducing construction costs, it is possible to reduce costs by acquiring EB as part or all of the product. It is also possible to arbitrarily adjust the production amount of EB.

[Brief explanation of the drawing]

FIG. 1 shows typical examples of adsorption/separation patterns used in various conventional methods. FIG. 2 shows various aspects of the adsorption/separation pattern used in the present invention. FIGS. 3 and 4 show flow sheets of specific process examples used in carrying out the present invention.

Claims (1)

[Claims] 1. A xylene isomer mixture containing at least one of O-xylene and m-xylene, ethylbenzene and P-xylene is mixed with a phodiasite type zeolite having a SiO ₂ /Al ₂ O ₃ molar ratio of 4 or more. In a method for selectively adsorbing and separating ethylbenzene and P-xylene using a zeolite in which 60% or more of the exchangeable cation sites are replaced with potassium as an adsorbent, first the above-mentioned zeolite is charged into a first separation column filled with the above-mentioned zeolite. Development is carried out by alternately supplying a xylene isomer mixture and a developing agent, and O-xylene (and/or m-xylene), ethylbenzene, and P-xylene are adsorbed and separated in this order, and the following isomers are present in the adsorption zone. Zones A to E containing the components are formed in this order in the development direction, and these are separated and recovered at the exit of the separation column. P-xylene is obtained as a product from E and recovered from D The resulting mixture was further selectively adsorbed and separated in the same manner in a second separation column filled with the above zeolite to obtain ethylbenzene and P-xylene as products, and each mixture recovered from A to C was further separated. A method for separating ethylbenzene and p-xylene, characterized in that each isomer is obtained as a product through separation, recycling to the feedstock, conversion of ethylbenzene and p-xylene by an isomerization operation, or a combination of these operations. . Claim 1 wherein the mixture recovered from the 2C zone is recycled and reused as the first separation column feedstock.
The method described in section. 3. Pass at least a portion of the O-xylene (and/or m-xylene) recovered from zone A through an isomerization reaction column to convert at least a portion of it into ethylbenzene and/or para-xylene, and perform an isomerization reaction. The effluent containing the components converted in the column is
The method according to claim 1 or 2, wherein the method is recycled and reused as a feedstock for a separation column. 4 The mixture recovered from zone B is selectively adsorbed and separated in a third separation column filled with zeolite to obtain ethylbenzene as a product, and O-xylene (and/or m-xylene) is separated from zone A. 4. A method as claimed in claim 3, in which the mixture recovered from the zone is combined. 5. The method according to claim 3, wherein the mixture recovered from zone B is sent to an isomerization reaction column together with the components recovered from zone A. 6. The method according to any one of claims 1 to 5, wherein the mixture of the C band is separated into the components of the B and D bands, and the divided recovery is performed with the C band substantially eliminated. Method. 7. The method according to claim 6, wherein the partial recovery is performed at a point where the O-xylene (and/or m-xylene) adsorption zone in the B zone borders the P-xylene adsorption zone in the D zone. . 8 Claim 1, in which the developing agent is separated and recovered by distillation from a mixture of xylene isomers and developing agent that is separated and recovered from the separation column, and then is supplied to an isomerization reaction or other separation column, or is made into a product. ~The method according to any one of Items 7. 9. The method according to any one of claims 4 to 8, wherein a common developing agent is used in the first to third separation columns. 10. The method according to claim 1, 6, 7, or 9, using a developer whose selective adsorption rate K for P-xylene is 0.5 to 1.5. 11. The method according to claim 10, wherein the developing agent is a chain ether consisting of R-O-R' (R, R': alkyl group having 3 to 5 carbon atoms excluding n-butyl group). 12 At the time of xylene isomer acquisition, the developing agent is supplied so that the most distal end of the trailing interface of zone E and the distal end of the distal interface of zone A of the next xylene adsorption zone substantially touch or partially overlap. 12. The method according to claim 10 or 11, which sets the amount. 13 Change the amount of developing agent supplied at one time (gr) to the weight of xylene supplied at one time (gr) x 1 - average _C8 isomer concentration (wt%) / average _C8 isomer concentration (wt%) The first claim characterized in that
The method described in Section 2.