RU2368594C2 - Method of isomerising normal butane to isobutane and device for realising said method - Google Patents

Abstract

FIELD: chemistry. ^ SUBSTANCE: invention relates to a method of isomerising normal butane containing initial material, containing at least 50 wt % normal butane to an isomerate which contains isobutane, involving: (a) isomerisation of the initial material under isomerisation conditions, involving presence of isomerisation catalyst, until an output isomerisation stream is obtained, containing normal butane, but with less concentration than in the initial isomerisation material; (b) distillation of at least part of the output isomerisation stream until a low-boiling fraction is obtained, containing isobutane and light paraffins, where at least 80 wt % of the low-boiling fraction is isobutane, and a high-boiling fraction, containing normal butane and at least 10 wt % isobutane; (c) bringing at least part of the fraction containing normal butane from stage (b) into contact with medium at the retentate side of a permeation-selective membrane, with efficiency index of flow of C4 permeate equal to at least 0.01 and pressure difference between media on both sides of the membrane, which allows for obtaining a retentate fraction containing at least 80 wt % isobutane, and for obtaining a permeate fraction at the permeate side, after passage through the membrane, with high concentration of normal butane; and (d) tapping off retentate fraction from stage (c). ^ EFFECT: simplification of method. ^ 10 cl, 2 dwg

Description

FIELD OF THE INVENTION

This invention relates to improved methods for isomerizing normal butane to isobutane using an isobutane distillation column in combination with a membrane that is selectively permeable to normal butane in comparison to isobutane, and to improved methods using a membrane to recover at least a portion of isobutane contained in the isomerization feedstock containing normal butane. Methods for isomerizing normal butane to isobutane are widely used in practice. In the isomerization process, the reaction proceeds in the direction of thermodynamic equilibrium. Thus, the isomerate will still be characterized by a significant concentration of normal butane, usually corresponding to a range of molar ratios between isobutane and normal butane from 1.2: 1 to 2: 1. The desired isobutane product, usually characterized by a degree of purity equal to at least 80, often at least 90 or more, for example, in the range from 95 to 99 mol% of isobutane, is obtained by distillation (isobutane distillation column ), which provides a relatively pure isobutane product in the form of a distillate and a fraction containing normal butane, which is recycled to the isomerization reactor.

Since the boiling points of normal butane and isobutane are relatively close, and a relatively pure isobutane product is desirable, the isobutane distillation column is usually operated at a high reflux ratio. Thus, the power consumption for heating in the isobutane distillation column is an essential component of the operating costs in the butane isomerization process, and the power consumption for heating becomes even more significant when isobutane product streams with a higher degree of purity are required. In accordance with this, a search is made for improved methods for isomerizing normal butane, which are characterized by improved fixed capital and operating costs.

State of the art

Isolation of linear paraffins from branched paraffins, for example, the separation of normal butane and isobutane, was proposed, but so far the membranes still have to find a scope of practical, commercial application. US 5,069,794 describes microporous membranes containing crystalline molecular sieve material. Column 8, lines 11 and further describe the potential applications of membranes, including the separation of linear and branched paraffins. See also US Pat. No. 6,090,289 for a layered composite containing molecular sieves that could be used as a membrane. Among the potential divisions in which the membrane can be used, and which are described starting from column 13, line 6 of the patent, is the separation of normal paraffins from branched paraffins. US 6156950 and US 6338791 discuss separation methods due to differences in permeability, which can be used to isolate normal paraffins from branched paraffins, and describes specific separation schemes due to isomerization. US 2003/0196931 A1 describes a two-stage isomerization process for improving the quality of hydrocarbon feeds containing 4 to 12 carbon atoms. The use of zeolite membranes has been proposed as a suitable method for isolating linear molecules. See, for example, paragraphs 0008 and 0032. US 6818333 describes thin zeolite membranes that are claimed to have n-butane permeability of at least 6 · 10 ⁻⁷ mol / m ² · sec · vPa, and selectivity for the conversion of n-butane to isobutane equal to at least 250.

Due to the volumes of feed materials containing normal butane that are processed in industrial butane isomerization plants, to achieve the desired isolation of linear paraffins, it would be necessary to have large membrane surface areas. For example, ZSM-5 / silicalite (MFI) membranes (sieving membrane), available from NGK Insulators, Ltd., Japan, which exhibit linear paraffin permeability selectivity compared to branched paraffins, under operating conditions have permeabilities ranging from 0.1 to 1.0 milligram mol per second per square meter with a pressure difference in the range of 15 to 500 kPa. Thus, the cost of commercializing such a membrane separation system using these membranes or membranes of the type described in US 6818333 makes it uncompetitive in comparison with an adsorption separation system or with separation by distillation method, such as using a column for distillation of isobutane.

Authors Bourney, et al. WO 2005/049766 recently described a process for producing high octane gasoline using a membrane to remove, inter alia, n-pentane from an isomerized stream obtained from a distillate of an isohexane distillate column. The running run from the isohexane stripping column acts as a stripping fluid on the membrane permeate side. The mixture of permeate and stripping fluid is recycled to the isomerization reactor. The results of computer simulations based on the use of a membrane in the form of an MFI on alumina, shown in Example 1 of the publication, indicate that 5,000 square meters of membrane surface are required to remove 95% by weight of distillate from the distillation column to distill off isohexane. With the consumption of the source material supplied to the membrane separator (75000 kg / h, if it contains 20.6 weight percent of n-pentane), the value of the n-pentane flow used in the simulation, apparently, is of the order of 0.01 grams mol / m ² · sec at 300 ° C.

For the purposes of the following discussion of the invention, the following membrane properties are defined.

Microporous

The terms “microporous” and “microporosity” refer to pores characterized by effective diameters ranging from 0.3 to 2 nanometers.

Mesoporous

The terms “mesoporous” and “mesoporosity” refer to pores characterized by effective diameters ranging from 2 to 50 nanometers.

Macroporous

The terms “macroporous” and “macroporosity” refer to pores characterized by effective diameters greater than 50 nanometers.

Nanoparticle

Nanoparticles are particles characterized by a basic size reaching up to 100 nanometers.

Molecular sieves

Molecular sieves are materials characterized by the presence of microporosity, and they can be amorphous, partially amorphous or crystalline and can be zeolite, polymer, metal, ceramic or carbon.

Screening membrane

A screening membrane is a composite membrane containing a continuous or discrete medium for selective separation, including a molecular sieve barrier layer. The barrier layer is a structure that is present to selectively block the flow of fluid in the membrane. In a continuous sieving membrane, the molecular sieves themselves form a continuous layer, which they strive to make free from defects. The continuous barrier layer may contain other materials, such as those that would be present in the case of membranes with mixed matrices. A discrete screening membrane is a discrete prefabricated molecular sieve barrier layer structure in which free spaces or voids are present between particles or regions of molecular sieves. These spaces or voids may contain other solid material or may be filled with it. Particles or regions of molecular sieves represent a barrier layer. Separation using sieving membranes can be based on the steric properties of the components to be separated. Other factors can also affect permeability. One is the presence of sorption ability or its absence in the component and in the material of molecular sieves. Another is the interaction between the shared components in the microporous structure of molecular sieves. For example, in the case of some zeolite molecular sieves, the presence of, say, n-hexane molecules in the pore can prevent the penetration of 2-methylpentane into this pore to a greater extent than another n-hexane molecule. Thus, zeolites, which, apparently, do not show a high selectivity for the separation of normal and branched paraffins solely from the point of view of molecular size, in practice can ensure the achievement of increased selectivity in relation to separation.

Permeate C ₄ Flow Efficiency Index

The permeability of the screening membrane, that is, the speed at which a given component passes through a given membrane thickness, often varies depending on conditions such as temperature and pressure, absolute and differential. Thus, for example, different penetration rates can be determined when the absolute pressure on the permeate side is 1000 kPa, and then when this pressure is 5000 kPa, while all other parameters, including the pressure difference, are kept constant. Accordingly, a C ₄ permeate flow efficiency index is used herein to describe screening membranes. The C ₄ permeate flow efficiency index for a given membrane is determined by measuring the speed (gram moles per second) at which substantially pure normal butane (preferably at least 95 wt.% Normal butane) penetrates the membrane at about 150 ° C at a pressure on the retentate side of 1000 kPa absolute pressure and a pressure on the permeate side of 100 kPa absolute pressure. The C ₄ permeate flow efficiency index reflects the penetration rate per square meter of surface area on the retentate side, but it does not lead to a unit thickness of the membrane. Thus, the efficiency index of the flow of C ₄ permeate for a given membrane will be expressed in units of gram moles of normal butane penetrating per second through a square meter of membrane surface area on the retentate side.

The ratio of the effectiveness of the flow of C ₄ permeates

The ratio of the flow efficiencies of C ₄ permeates for a given sieving membrane is the ratio between the flow efficiency index

With ₄ permeate (n-butane) and the iso-C ₄ permeate flow efficiency index, where the iso-C ₄ permeate flow efficiency index is determined in the same way as the C ₄ permeate flow efficiency index, but when using substantially pure isobutane (preferably at least 95 wt.% isobutane).

SUMMARY OF THE INVENTION

This invention improves the isomerization methods of normal butane for feedstocks containing normal butane, where the methods use distillation (isobutane distillation column) to recover the purified isobutane product from the isomerization effluent. The methods of this invention include the integration of membrane separation with an isobutane distillation column to reduce power consumption in the reboiler of the isobutane distillation column and to be able to eliminate bottlenecks in the isobutane distillation column or to have a reduced diameter and number of distillation plates for a given isobutane product and its degree of purity.

In many cases, distillation is used to produce a fraction containing normal butane intended for recycling to isomerization reactors. In accordance with this invention, the isobutane distillation column is operated at a reflux ratio low enough to produce a fraction containing normal butane, which is characterized by a significant concentration of isobutane, often equal to at least 10, and sometimes at least 25 wt.% isobutane, and a lower boiling fraction characterized by an isobutane content of at least 80, preferably at least 90, and sometimes in the range from 95 to 99 wt.%.

Thus, the power consumption for heating and, thus, the operating costs for distillation can be significantly reduced for a given capacity for the isobutane product and its degree of purity. For example, the mass ratio between reflux and the feed material may in some cases be reduced by at least 20%. Typically, the ratios between the amounts of reflux and the feed feed for such distillation columns correspond to an order of ratios ranging from 2.5 to 4: 1. When using the methods of this invention, the ratio between the amounts of reflux and the feed material can be reduced to a ratio in the range from 0.5: 1 to 3: 1, preferably from 0.5: 1 to 3: 1, say, from 0.5 to 2 : 1, and more preferably, to a ratio in the range of 0.75: 1 to 1.3: 1 for the same isobutane performance and purity.

In the methods of the invention, the distillation fraction containing normal butane is contacted with the medium on the retentate side of the selectively permeable membrane to produce normal butane enriched permeate and a retentate containing at least 80% by weight of isobutane. Since the volume of the fraction containing normal butane is only part of the effluent from the isomerization reactor, the required membrane surface area will be less than what would be required if membrane separation was used to isolate isobutane from the effluent from the isomerization reactor. In addition, despite the presence of isobutane in the feed to the membrane, its concentration in the stream, which is a fraction of the effluent from the isomerization reactor, is negligible. Thus, the penetration of a certain amount of isobutane through the membrane can be reconciled, since the absolute amount of isobutane in the stream sent for recycling to the isomerization reactor will have little effect on achieving a high degree of single conversion of normal butane in fresh feedstock. Consequently, suitable membranes for use include not only those that are highly selective for the isolation of normal butane from isobutane, but also those that exhibit a high C ₄ permeate flow efficiency index even with a concomitant low C flow efficiency ratio ₄ permeates.

In one broad aspect of the methods of this invention related to the use of distillation to obtain a product containing purified isobutane, they include:

and less than in the starting material fed to isomerization;

(b) distilling at least a portion, preferably at least 90% (mass.), and most preferably substantially the total amount, of the isomerization effluent to obtain (i) a lower boiling fraction containing isobutane and lighter paraffins, where at least 80 preferably at least 90 wt.% of the lower boiling fraction are isobutane, and (ii) a higher boiling fraction containing normal butane, which contains normal butane and at least 10 wt. .% isobutane;

(c) introducing at least a portion, preferably at least 90 wt.%, and most preferably, substantially the entire amount, of a fraction containing normal butane, from step (b), selectively contacting the medium on the retentate side a permeable membrane under conditions including a membrane surface area and a pressure difference on both sides of the membrane sufficient to obtain a retentate fraction containing at least 80 preferably at least 90 wt.% isobutane, and to obtain after passing through the membrane on side n rmeata permeate fraction, characterized by an increased concentration of normal butane, said permeate fraction preferably comprises at least 90 wt% of the normal butane contained in the input into contact with a membrane fraction containing normal butane.; and

(d) selecting from the retentate fraction from step (c) preferably for combining with the lower boiling fraction from step (b).

The feedstock for isomerization may be feedstock containing normal butane, or it may be a stream obtained from feedstock containing normal butane, from which at least a portion of any amount of isobutane contained therein has been removed, for example, by distillation or membrane separation, including membrane separation according to another broad aspect of the present invention. Thus, a feedstock containing normal butane may contain as little as 30 wt.% Normal butane, but still be suitable for use in the methods of this invention. In preferred embodiments of the methods of the invention, at least a portion, preferably at least 90 wt.%, And most preferably substantially all, of the permeate fraction from step (c) is bypassed to step (a).

Preferably, the lower boiling fraction from step (b) contains at least 25, preferably from 30 to 95, and in some cases from 30 to 80% by weight of isobutane isomerization effluent. In the general case, for a given ratio between the amounts of reflux and the feed material, an increase in the degree of purity of isobutane in the product will lead to a decrease in the concentration of isobutane in the lower boiling fraction.

In preferred embodiments of the invention, at least 50% by weight of isobutane contained in the normal butane containing fraction contacted with the membrane remains in the retentate fraction. The concentration of normal butane in the permeate fraction is often at least 50, preferably at least 70 wt.%.

Another broad aspect of the methods of the present invention relates to conducting at least a portion of the normal butane containing feedstock by membrane separation to remove isobutane and thereby obtaining a higher concentration of normal butane in the feed fed to the isomerization reactor. Often, starting materials containing normal butane contain significant amounts of isobutane, for example, from 5 or 10 wt.% Or more up to 50 or 60 wt.%. Since isomerization is determined by the achievement of equilibrium, the presence of isobutane in the feed to isomerization will adversely affect the degree of conversion of normal butane to isobutane, and will also lead to an increase in the consumption of feedstock through the isomerization reaction zone. Thus, usually such an isobutane-containing feedstock is distilled to remove at least a portion of the isobutane. For example, the feedstock fed to the isomerization unit can be introduced into the isobutane distillation column. Despite the fact that the isobutane distillation column does indeed provide isobutane extraction from the feedstock and thus increases the concentration of normal butane in the fluid supplied to the isomerization reaction zone, the isobutane distillation column must not only have a size corresponding to processing additional volume fluid, but also the isobutane stripping column will also increase the power consumption for heating.

In accordance with one aspect of the invention, membrane separation is used to remove isobutane from the feed. In the broadest terms of this aspect of the invention, distillation of the isomerization effluent is not essential. Broad aspects of the isomerization methods of normal butane of this aspect of the invention include: (a) introducing at least a portion, preferably at least 90 wt.%, And most preferably, substantially the entire amount of the feedstock containing normal butane, and containing at least 5, often at least 10 wt.% isobutane, in contact with the medium on the retentate side of the selectively permeable membrane under conditions including a membrane surface area and a pressure difference on both sides of the membrane, sufficient to obtain a retentate fraction containing at least 80, preferably at least 90 wt.% isobutane, and to obtain, after passing through the membrane on the permeate side, a permeate fraction characterized by an increased concentration of normal butane, wherein said permeate fraction preferably contains at least 80 wt.% normal butane contained in the feedstock contacted with the membrane; and (b) exposing the permeate fraction to isomerization conditions, including the presence of an isomerization catalyst, to obtain an isomerization effluent stream containing normal butane but with a concentration lower than the concentration in the feedstock.

In one embodiment, the retentate from step (a) is combined with the isomerization effluent, and thus, it can provide an isobutane product of an appropriate degree of purity even without distillation to remove normal butane. Alternatively, the retentate itself can be used as a product stream or it can be combined with the lower boiling fraction obtained by distillation.

Preferably, the isomerization effluent stream is enriched to produce a product stream containing at least 80, preferably at least 90 wt.% Isobutane, and a stream containing normal butane, at least some of which may or may not be sent recycling for isomerization. Often, the retentate fraction contains at least 50, preferably at least 60, substantially up to 100% by weight of isobutane contained in the feed.

Preferably, in both broad aspects of the invention, the membrane is a sieving membrane characterized by a C ₄ permeate flow efficiency index of at least 0.01, more preferably at least 0.02, and a C ₄ permeate flow efficiency ratio corresponding to at least 1.25: 1, more preferably at least 1.3: 1, and often an appropriate ratio in the range of 1.35: 1 to 5: 1 or 6: 1.

The invention also relates to apparatus for implementing the methods of the invention. In a broad aspect, apparatus for isomerizing a feedstock containing normal butane includes:

a. an isomerization reactor (106) adapted to perform isomerization of normal butane to produce an isomerate containing isobutane and having an inlet and an outlet for isomerate;

b. a distillation column (116) having an inlet through fluid in communication with the outlet of the isomerization reactor (106), an outlet channel for a lower boiling fraction (118) and an outlet channel for a higher boiling fraction (122), wherein the distillation column (116) is adapted to distill at least a portion of the isobutane-containing isomerate to obtain a lower boiling fraction, in comparison with the isobutane-enriched isomerate, and a higher boiling fraction, in comparison with an isomerate enriched with normal butane, wherein said fraction contains isobutane; and

c. a membrane separator (124) having an inlet on the side of the feed source through a fluid in communication with the channel for the higher boiling fraction (122), an outlet on the side of the feed inlet through a fluid in communication with the line (118) from the outlet of the distillation column for a lower boiling fraction (116), and the outlet for permeate, through the fluid in communication with the inlet of the isomerization reactor (106).

Brief Description of the Drawings

Figure 1 shows a schematic representation of the methods of this invention using a stabilization column in front of the isobutane distillation column.

Figure 2 shows a schematic representation of the methods of this invention, wherein the normal butane-containing feedstock is subjected to membrane separation to obtain an isobutane-containing effluent that is combined with the isomerization effluent to form a product without the need for distilling off isobutane from normal butane.

Detailed description

ISOMERIZATION

In the methods of this invention, any suitable starting material containing normal butane can be used. The feedstock can be obtained from any suitable source for use, including the following, but not limited to: fractionation or liquefaction of gaseous or under normal conditions liquid petroleum materials, from synthetic paraffins, such as from Fischer-Tropsch processes, cracking of higher molecular weight hydrocarbons and the like like that. The composition of the feedstock will depend on the source of feedstock. Preferred feedstocks will contain at least 30, say, from 50 to substantially 100% by weight of normal butane. Other components that may be present include isobutane, butenes, and lighter and heavier hydrocarbons. In general, lighter hydrocarbons are present in amounts of less than 20, preferably less than 5 wt.%, And heavier hydrocarbons are present in amounts of less than 10, preferably less than 5 wt.%. In the event that the feed contains C ₅ or more heavy hydrocarbons, then it is usually preferred that the feed be subjected to a typical physical process of chemical technology, such as distillation (for example, in an isobutane distillation column), to reduce the concentration C ₅ and heavier hydrocarbons to a level less than 1, preferably less than 0.5 wt.%.

The feedstock, together with the material sent for recycling, if any, recovered from the isomerization reaction effluent, is passed as feed to isomerization in one or more isomerization zones. The feedstock and the material sent for recycling are usually mixed before being introduced into the isomerization zone, but if desired, they can be introduced separately. Alternatively, the feedstock may be combined with the isomerization effluent in the isobutane distillation column, and then transferred to isomerization zones in combination with the material sent for recycling. This latter technique serves to extract from the feedstock at least a portion of the isobutane contained therein. In each case, the total feed to the isomerization zone is referred to herein as the feed to isomerization. Material sent for recycling can come in the form of one or more flows. As discussed below, material sent for recycling contains normal butane.

The concentration of normal butane in the feed to isomerization will depend not only on the concentration of normal butane in the feed, but also on its concentration in the material sent for recycling, if available, and the relative amount of material sent for recycling, in comparison with the feedstock, which can be in a wide range. Often the starting material supplied for isomerization is characterized by a concentration of normal butane equal to at least 50, say, in the range from 60 to 100, preferably from 75 to 90 wt.%.

In the isomerization zone, the feed to the isomerization is subjected to isomerization conditions including the presence of an isomerization catalyst, preferably in the presence of a limited amount of hydrogen. Isomerization of normal butane is generally considered a first order reversible reaction. Thus, the effluent of the isomerization reaction will be characterized by a higher concentration of isobutane and a lower concentration of normal butane in comparison with what is the case for the starting material fed to the isomerization. In preferred embodiments of the invention, the conditions for isomerization are sufficient to isomerize at least 20, preferably 30 to 60% by weight of normal paraffins in the combined feed and recyclable material. In general, the conditions for isomerization can achieve at least 70, preferably at least 75, say, 75 to substantially 100 percent of the equilibrium values for the C ₄ paraffins present in the isomerization feed. In many cases, the isomerization reaction effluent is characterized by a mass ratio between isobutane and normal butane corresponding to at least 1.2: 1, preferably corresponding to a ratio in the range from 1.4 to 2: 1.

For broad aspects of the methods of this invention, the isomerization catalyst is not critical, and any suitable isomerization catalyst can be used. Suitable isomerization catalysts include acid catalysts that use chloride to maintain the desired activity. The isomerization catalyst may be amorphous, for example based on amorphous alumina, or zeolite. The zeolite catalyst will usually still contain an amorphous binder. The catalyst may contain a platinum group metal on chloride-treated alumina.

Contacting in isomerization zones can be carried out using the catalyst in a fixed bed system, a moving bed system, a fluidized bed system, or in a batch type operation. A fixed bed system is preferred. Reagents can be brought into contact with a bed of catalyst particles when flowing from bottom to top, top to bottom or in the radial direction. When brought into contact with the catalyst particles, the reactants can be in the liquid phase, the mixed vapor-liquid phase or the vapor phase, with excellent results obtained when applying the present invention to a mainly vapor-phase operation. The isomerization zone may be located in one reactor or in two or more separate reactors equipped with suitable devices that maintain the desired isomerization temperature at the entrance to each zone. Preferred to provide improved isomerization are two or more reactors arranged in sequence, due to the temperature control of the individual reactors and due to the partial replacement of the catalyst without stopping the process.

Isomerization conditions in the isomerization zone include reactor temperatures, typically in the range of 40 to 250 ° C. In general, lower reaction temperatures are preferred, which favors the production of equilibrium mixtures with a maximum concentration of isobutane and minimizes cracking of the feed material to form lighter hydrocarbons. In the present invention, temperatures in the range of 100 to 200 ° C. are preferred. The operating pressures in the reactor are generally in the range from 100 kPa to 10 MPa absolute pressure, preferably from 0.5 to 4 MPa absolute pressure. The hourly space velocities of the liquid are in the range of 0.2 to 25 volumes of isomerizable hydrocarbon feed material per hour per volume of catalyst, with a range of 0.5 to 15 h ^-1 being preferred.

Hydrogen is mixed or stored in the combined feedstock and materials sent to the isomerization zone and recycled to obtain a molar ratio between hydrogen and the hydrocarbon feed material in the range of 0.01 to 20. Hydrogen can be supplied entirely from an external system for the method, or it can be replenished with hydrogen sent for recycling to the original feed material after separation from the isomerization reactor effluent. Light hydrocarbons and small amounts of inert materials such as nitrogen and argon may be present in hydrogen. Water must be removed from the hydrogen supplied from a system external to the process, preferably using an adsorption system known in the art. In a preferred embodiment, the molar ratio of hydrogen to hydrocarbon in the reactor effluent is equal to or less than 0.05, which generally eliminates the need for hydrogen to be recycled from the reactor effluent to the feed.

Particularly when a chloride-treated catalyst is used for isomerization, the isomerization effluent is contacted with the sorbent to remove any chloride components, or distilled or stripped to remove chloride components as well as light fractions.

PROCESSING OF RAW MATERIALS

In those aspects of the present invention in which the feedstock is subjected to membrane separation, the preferred feedstock contains at least 5% by weight of isobutane. At least a portion of preferably at least 80, and more preferably at least 90 wt.% Of the feedstock is subjected to membrane separation prior to the introduction of isomerization into the reaction zone. The feedstock can be subjected to membrane separation immediately before being fed to the isomerization unit, or it can be fed to the isomerization unit located upstream of the membrane separator used to extract isobutane from a stream containing normal butane. For example, the feedstock may be fed to an isobutane distillation column or to a portion between the isobutane distillation column and a membrane separator used to recover isobutane from the fraction containing normal butane sent to the isomerization reaction zone for recycling to the isomerization reaction zone.

The feedstock is contacted with the medium on the retentate side of the selective membrane to obtain a retentate fraction of an isomerization effluent stream that contains at least 80 wt.% Isobutane, and to obtain, after passing through the membrane on the permeate side, a permeate fraction characterized by an increased concentration normal butane. To achieve the desired separation at a suitable penetration rate, a pressure drop is maintained on the membrane. The membrane can be of any suitable type, including diffusion and screening, and can be made from inorganic, organic, or composite materials. In the case of diffusion membranes, the driving force is the difference in partial pressures between the media on the sides of the retentate and permeate. In the case of sieving membranes, regardless of partial pressures or concentrations, a drop in absolute pressure becomes an essential component of the driving force. Preferred membranes are sieving membranes characterized by a C ₄ permeate flow efficiency index of at least 0.01 and a C ₄ permeate flow efficiency ratio corresponding to at least 1.25: 1. Further sieving membranes are discussed in more detail.

With membrane separations, the pressure drop is often in the range from 0.1 to 10, preferably from 0.2 to 2 MPa. In practice, a fraction containing normal butane will be brought into contact with the medium on the retentate side of the membranes without additional compression, which minimizes the cost of fixed capital and operating costs. The temperature of the membrane separation will partly depend on the nature of the membrane and the temperature of the feedstock. Often the temperature is in the range from 25 ° C to 150 ° C. Membrane separation conditions may ensure that a liquid or vapor or mixed phase is present on the retentate side of the membrane (the terms “mixed phase” used herein mean a mixed liquid and vapor phase). Permeate can form a vapor, liquid or mixed phase. If the fluid on the retentate side of the membrane is a liquid phase, then permeate can form a liquid, vapor or mixed phase.

Any suitable selectively permeable membrane may be used in the apparatus and methods of the present invention. Preferred membranes are screening membranes. The membranes used in the methods of the present invention are highly permeable, that is, characterized by a flow index C _{6 of} permeate equal to at least 0.01. The membranes can take any suitable shape, such as for hollow fibers, sheets, and the like, which can be assembled to form a separator module, such as membranes including hollow fiber bundles or flat plates or spirally wound sheets. The physical structure of the membranes during assembly in the separator module should make it possible to obtain a pressure drop on the membrane sufficient to achieve the desired permeability. In the case of membranes comprising hollow fibers, the medium of the high pressure side (retentate side) will usually be outside the hollow fiber. The permeate flow can be co-directional, counter-directional, or cross-directional with respect to the fluid flow on the retentate side of the membrane. A sufficient membrane surface area is provided such that, under stationary conditions, the desired concentration of isobutane is obtained in the retentate fraction.

WASTE FLOW ISOMERIZATION AND REMOTE

The isomerization effluent is subjected to one or more separation operations to remove components such as hydrogen, lower alkanes and, especially in the case of chloride-treated catalysts, halogen compounds.

In accordance with aspects of the present invention, where distillation is used to produce an isobutane product, the isomerization effluent is bypassed either directly or by incorporating the above typical intermediate physical processes of chemical technology into a distillation assembly (which is called an isobutane distillation column for the purpose of the discussion that follows). ) in order to obtain a lower boiling fraction containing isobutane and a higher boiling fraction containing normal butane. Since isomerization conditions tend to produce lighter components such as methane, ethane and propane, these components are either removed in a stabilization column or removed as a distillate from the isobutane distillation column. Where light fractions will be substantially removed from the isomerization effluent prior to introduction of isobutane into the distillation column, the lower boiling isobutane fraction may be a distillate. The feed to the isobutane stripping column may form a vapor, liquid, or mixed phase.

Most often, the isobutane distillation column is adapted to produce a stream containing normal butane as a side stream and to produce a bottoms stream containing C ₅ components such as normal pentane and isopentane. The isobutane distillation column may be a packed or plate column, and is usually operated at a column top pressure in the range of 100 to 1000 kPa (gauge pressure) and bottoms temperature in the range of 60 to 170 ° C. The ratio between the amounts of reflux and the feed in the isobutane distillation column will affect the concentration of isobutane in the fraction containing normal butane. In the general case, from the viewpoint of reducing the power consumption for heating in the isobutane distillation column, lower reflux numbers are preferred. Thus, reflux numbers are preferred (ratios between the amounts of reflux and the starting feed when calculated per mole) in the range of 0.5: 1 to 3: 1. When the reflux numbers are reduced, the lower boiling fraction for a given degree of isobutane purity will contain a smaller portion of isobutane, which will be fed to the isobutane distillation column. Similarly, to achieve higher isobutane concentrations in the lower boiling effluent, higher reflux numbers will be required while keeping all other parameters identical.

The lower boiling fraction from the isobutane stripping column typically contains at least 80, and preferably at least 90, wt.% Of isobutane. For some operations carried out further down the process, fractions with a higher degree of purity may be desirable, and the methods of the present invention when carrying out economical operations can provide fractions with a content level greater than 95 or 97 wt.%, And even reaching up to 99 wt.%, while still obtaining benefits in the form of reduced power consumption for heating in the isobutane distillation column.

When using the isobutane distillation column, one or more other fractions can be obtained, including a fraction containing normal butane. This fraction containing normal butane also contains isobutane due to the reduced ratios between the amounts of reflux and the feed material allowed when using membrane separation in accordance with this invention. Often, the fraction containing normal butane will contain at least 10, preferably from 20 to 50% by weight of isobutane, with the balance amount essentially being normal butane. If there is a column for the distillation of isobutane having a bottoms stream, C ₅ hydrocarbons will be removed at this point, otherwise they will be additional components in the stream containing normal butane.

At least a portion of preferably at least 80, and more preferably at least 90 wt.%, To substantially the total amount of the fraction containing normal butane is contacted with the medium on the retentate side of the selective membrane. from the isobutane distillation column. A retentate fraction is obtained which contains at least 80, and preferably at least 90% by weight of isobutane, and after passing through the membrane on the permeate side, a permeate fraction is obtained, characterized by an increased concentration of normal butane.

Although the distillation operation with a reduced amount of reflux will result in a smaller amount of isobutane being obtained as a lower boiling fraction (at a given concentration of isobutane in the lower boiling fraction), the membrane separation corresponding to this aspect of the invention will lead to recovery in the fraction retentate of additional isobutane, which may be characterized by a degree of purity suitable for commercial use or for combining with lower boiling point fraction from the isobutane distillation column. In one embodiment of the invention, the retentate may be characterized by an increased concentration of isobutane in comparison with the concentration in the lower boiling fraction, and the retentate is used to enrich the combined stream of retentate and lower boiling fraction.

The pressure drop is maintained on the membrane in order to achieve the desired separation at suitable penetration rates. The membrane can be of any suitable type, including diffusion and screening, and can be made from inorganic, organic, or composite materials. In the case of diffusion membranes, the driving force is the difference in partial pressures between the media on the sides of the retentate and permeate. In the case of sieving membranes, regardless of partial pressures or concentrations, a drop in absolute pressure becomes an essential component of the driving force. Preferred membranes are sieving membranes characterized by a C ₄ permeate flow efficiency index of at least 0.01 and a C ₄ permeate flow efficiency ratio corresponding to at least 1.25: 1. Screening membranes are discussed in more detail below.

With membrane separations, the pressure drop is often in the range from 0.1 to 10, preferably from 0.2 to 2 MPa. In practice, a fraction containing normal butane will be brought into contact with the medium on the retentate side of the membranes without additional compression, which minimizes the cost of fixed capital and operating costs. The temperature of the membrane separation will partly depend on the nature of the membrane and the temperature of the fraction. Thus, in the case of polymer-containing membranes, the temperatures should be sufficiently low so that an undesirable adverse effect would not be exerted on the membrane strength. In most cases, the separation temperature is the temperature of the fraction from the isobutane distillation column. Often the temperature is in the range of 25 ° C to 150 ° C. Thus, the conditions for conducting membrane separation can lead to a liquid or vapor or mixed phase on the retentate side of the membrane. Permeate can form a gas, liquid or mixed phase. If the fluid on the retentate side of the membrane is a liquid phase, then permeate can form a liquid, gas, or mixed phase.

The membranes may take any suitable shape, such as for hollow fibers, sheets, and the like, which can be assembled to form a separator module, such as membranes including hollow fiber bundles or flat plates or spirally wound sheets. The physical structure of the membranes during assembly in the separator module should make it possible to obtain a pressure drop on the membrane sufficient to achieve the desired permeability. In the case of membranes comprising hollow fibers, the medium of the high pressure side (retentate side) will usually be outside the hollow fiber. The permeate flow can be co-directional, counter-directional, or cross-directional with respect to the fluid flow on the retentate side of the membrane. A sufficient membrane surface area is provided such that, under stationary conditions, the desired concentration of isobutane is obtained in the retentate fraction.

At least a portion of the permeate fraction is recycled to the isomerization step.

RESISTING MEMBRANES

Preferred sieving membranes can be of various types, for example molecular sieve, porous ceramic, metal, polymer or carbon membranes, or composite membranes having a highly porous polymer, metal, molecular sieve, ceramic or carbon carrier with a thin sieving layer or barrier layer (molecular sieves) for example zeolite, polymer, metal, ceramic or carbon, characterized by the presence of microporosity.

Membranes can be continuous or discrete. A discrete membrane includes a collection of small particles forming a microporous barrier layer, while a continuous membrane includes a continuous layer of microporous barrier layer. The membranes can be obtained from one material, or they can be composites, including a microporous barrier layer and a carrier and, optionally, a different structure. As the thickness of the screening layer decreases as a thin continuous barrier layer is obtained, difficulties in obtaining a defect-free layer will increase. Since high selectivity is not required in the methods of this invention, membranes may have minor defects. Typically, continuous membranes are made by depositing on a meso / macroporous structure or by growing a continuous thin layer of a microporous barrier layer on it. Discrete clusters formed by particles of a microporous barrier layer of nanometer size make it possible to obtain very small penetration thicknesses, but with the potential for bypass. Discrete membranes use a meso / macroporous structure with which a microporous barrier layer is combined.

Examples of the zeolite barrier layer include small pore molecular sieves such as SAPO-34, DDR, A1PO-14, A1PO-17, A1PO-18, A1PO-34, SSZ-62, SSZ-13, zeolite 3A, zeolite 4A, zeolite 5A, zeolite KFI, H-ZK-5, LTA, UZM-9, UZM-13, ERS-12, CDS-1, phillipsite, MCM-65 and MCM-47; medium pore molecular sieves such as silicalite, SAPO-31, MFI, BEA and MEL; large pore molecular sieves such as FAU, OFF, NaX, NaY, CaY, 13X and zeolite L; and mesoporous molecular sieves such as MCM-41 and SBA-15. Several types of molecular sieves, such as A, X, L, OFF, MFI, and SAPO-34, are available in colloidal form (nanoscale particles). Zeolites may or may not be subjected to metal exchange.

Other types of screening materials include carbon sieves; polymers such as PIM (intrinsic microporosity polymers) such as those described by McKeown, et al., Chem. Commun., 2780 (2002); McKeown, et al., Chem. Eur. J., 11: 2610 (2005); Budd, et al., J. Mater. Chem., 13: 2721 (2003); Budd, et al., Adv. Mater., 16: 456 (2004) and Budd, et al., Chem. Commun., 230 (2004); polymers into which porosity is introduced using blowing agents, such as poly (alkylene oxide), polyvinylpyrrolidone; cyclic organic host matrices such cyclodextrins, calixarenes, crown ethers and spherands; microporous organometallic scaffolds such as MOF-5 (or IRMOF-1); glass, ceramic and metal forms into which microporosity has been introduced.

Composite membranes use a meso / macroporous structure. The meso / macroporous structure is used in one or more functions depending on the type of membrane. It can be a carrier for a composite membrane, it can be an integral part in the preparation of a microporous barrier layer, it can be a structure on which or in which a microporous barrier layer is placed. The meso / macroporous structure may be continuous or discrete, and thus, the meso / macroporosity may be channels through the material of the meso / macroporous structure, or it may be formed between particles that form the meso / macroporous structure. Examples of the latter are AccuSep ™ inorganic filtration membranes available from Pall Corp. having a zirconia layer on a porous metal support, where zirconia is present in the form of spherical crystals.

The meso / macroporous structure preferably defines channels or pores with diameters ranging from 2 to 500, preferably from 10 to 250, more preferably from 20 to 200 nanometers, and is highly permeable. In more preferred embodiments, the efficiency index of the course of the C ₄ permeate of the meso / macroporous structure is at least 1, and most preferably at least 10 and sometimes at least 1000. The meso / macroporous structure can be isotropic or anisotropic. Meso- / macropores can be relatively straight or winding.

The meso / macroporous structure may consist of an inorganic, organic or mixed inorganic and organic material. The choice of material will depend on the conditions of the separation, as well as on the type of meso- / macroporous structure formed. The material of the meso / macroporous structure may be the same or different in comparison with the molecular sieve material. Examples of the compositions of the porous structures include metal, alumina such as alpha alumina, gamma alumina and transition aluminas, molecular sieves, ceramics, glass, polymer and carbon. In preferred embodiments, prior to forming the barrier layer or barrier layer precursor, defects in the substrate are corrected. In yet another embodiment, the substrate can be treated with a silica sol to partially occlude the pores and facilitate deposition of the barrier layer or barrier layer precursor. Silicon dioxide particles will still provide sufficient free space between their internodes, which will achieve high permeability. Another technique consists in applying a coating of siloxane rubber or another polymer to the carrier, which will allow high permeability, but will lead to the occluding of defects in the carrier or in the barrier layer.

If the meso / macroporous structure does not fulfill this function, then the membrane may contain a porous carrier for the meso / macroporous structure. The porous carrier is usually selected based on strength, tolerance of the conditions for the intended separation and porosity.

AccuSep ™ inorganic filtration membranes available from Pall Corp. and similar types of meso / macroporous structures are particularly advantageous since the meso / macroporous structure can be thin, thereby avoiding excessive thicknesses of the grown molecular sieves. In addition, zirconia is relatively inert with respect to the solutions of the zeolite precursors and the conditions for synthesis and calcination, which makes it a preferred meso / macroporous structure for these types of sieving membrane.

High permeability is achieved using at least one of the following methods: firstly, the use of pores larger than those necessary for the passage of a normal alkane; and secondly, the use of an extremely thin porous layer. If high permeability is achieved when less selective micropores of larger size are used in the microporous barrier layer, proper separation can be achieved. Often the pores of these types of membranes are characterized by an average pore diameter greater than 5.0 Å (average value for length and width), say, in the range from 5.0 to 7.0 or 8 Å. Preferably, the structures are characterized by an aspect ratio (length to width) of less than 1.25: 1, for example, a corresponding ratio in the range of 1.2: 1 to 1: 1. As for membranes containing molecular sieves, examples of structures are USY, ZSM-12, SSZ-35, SSZ-44, VPI-8 and cancrinite. In some cases, a penetrating molecule in a micropore can increase selectivity. For example, a normal hydrocarbon in a pore can reduce the rate at which a branched hydrocarbon can enter a pore in comparison with another normal hydrocarbon.

High permeability can also be obtained by using a very thin microporous barrier layer in either a continuous or discrete membrane. The microporous barrier layer can optionally be selected from sieving structures having micropores that are substantially impermeable to the fragment that needs to be stored in the medium on the retentate side. In the general case, pores for these types of membranes are characterized by an average pore diameter reaching up to 5.5 Å, for example, in the range from 4.5 to 5.4 Å. The aspect ratio for the pores of these membranes can vary within wide limits and usually ranges from 1.5: 1 to 1: 1. As for membranes containing molecular sieves, examples of structures are ZSM-5, silicalite, ALPO-11, ALPO-31, ferrierite, ZSM-11, ZSM-57, ZSM-23, MCM-22, NU-87, UZM- 9 and CaA.

Membranes containing a discrete cluster of particles forming a microporous barrier layer are characterized in that the barrier layer is characterized by a main size smaller than 100 nanometers, and the microporous barrier layer is combined with a meso / macroporous structure defining pores for the flow of a fluid, where the barrier layer placed to obstruct the flow of fluid through the pores of the meso- / macroporous structure. The molecular sieve barrier layer is “combined” with the meso / macroporous structure when it is placed on or in the structure, linking it to the structure or not. Thus, particles or islands of molecular sieves of nanometer size are used as barrier layers in membranes. A discrete microporous barrier layer is placed to obstruct the flow of fluid through the fluid flow paths defined by the meso / macroporous structure. The barrier layer may at least partially occlude the opening of the fluid flow channel of the meso / macroporous structure and / or the region within the fluid flow passage. Due to the small size of the particles or islands forming a discrete accumulation of microporous barrier layer, a certain selectivity for separation can be achieved despite the discreteness.

When choosing the components of the sieving membranes, the size and configuration of the molecular sieve particles and the size and configuration of the meso / macropores in the meso / macroporous structure will usually be taken into account. In the case of more spherical particles of molecular sieves, such as silicalite, it is preferable to choose a meso / macroporous structure having pores that are close to the same effective diameter. Thus, particles of molecular sieves, when placed in the pores or partially in the pores of the meso / macroporous structure, will form the minimum free space of the cavities for the bypass to occur. Greater flexibility exists in the case of plates and particles of molecular sieves of irregular shape, since they can overlap with little or no cavity space. In some cases, a combination of molecular sieve configurations may be desirable. For example, spherical molecular sieves can be pulled into the pores of the meso- / macroporous structure, followed by the introduction of smaller plate-like particles of molecular sieves. Additional functions are that the sphere serves as a carrier for the plate-like particles, and the plate-like particles overlap, reducing the likelihood of a bypass. Although molecular sieves are likely to be characterized by different compositions, and thus exhibit different sizes and configurations of microporosity, improved separation without excessive loss of permeability will be an advantage.

When obtaining high permeability in a discrete microporous barrier layer in the case of using zeolite molecular sieves, it is important to obtain small particles. In many zeolites, grain particles are available whose main size is less than 100 nanometers. Most molecular sieves are obtained using organic templates, which will need to be removed to gain access to the cells. Typically, this removal is carried out by calcination. As discussed below, calcination can be carried out when the molecular sieves containing the template are placed in the macropore in such a way that avoids excessive agglomeration simply due to the limitation of the number of particles located in the vicinity. Another technique to prevent agglomeration of zeolite particles during calcination is to silylate the surface of the zeolite, for example, using aminoalkyltrialkoxysilane, aminoalkylalkyl dialkoxysilane or aminoalkyl dialkylalkoxysilane. The degree of silylation required will depend on the size of the zeolite and its composition, as well as on the conditions used for calcining. Generally, 0.1 to 10 millimoles of silane are used per gram of zeolite.

There are various techniques for preparing molecular sieve particles on or in a meso / macroporous support in a manner that provides at least partial occluding of the mesos or macropores in the support. The particular technique used will depend on the size and configuration of the molecular sieve particles, the size and configuration of the meso / macropores in the meso / macroporous structure, and the desired placement of the molecular sieves in or on the meso / macroporous structure.

The meso / macroporous structure can be wetted with a solution or suspension of nanometer-sized molecular sieves, especially when molecular sieves are placed on the surface of the meso / macroporous structure to occlude at least a portion of the pore openings. The concentration of molecular sieves in the suspension should be low enough so that after drying the resulting layer of molecular sieves would not be excessively thick. Advantageously, during coating on the meso / macroporous structure, at least an insignificant pressure drop is maintained such that there is a motive force for drawing molecular sieves into any pores in the meso / macroporous structure that have not been occluded . Typically, the suspension will be an aqueous suspension, although it may be advantageous to use suspensions in alcohols and other relatively inert liquids at a concentration in the range from 2 to 30, say, from 5 to 20 weight percent. In the case of using the pressure difference, the pressure difference in the General case will be in the range from 10 to 200 kPa. One or more molecular sieve coatings may be used, preferably during drying between coatings. Drying is usually carried out at elevated temperatures, for example, in the range from 30 ° C to 150 ° C, for a period of time from 1 to 50 hours. A vacuum may be used to facilitate drying. In the case of using zeolites as molecular sieves, the fixation of molecular sieves on the meso / macroporous structure can in some cases contribute to calcination, for example, at temperatures in the range from 450 ° C to 600 ° C. Calcination can also be used to agglomerate molecular sieve particles and thus reduce the number of cavities and their size. It goes without saying that in the broad aspects of the present invention, calcination is not an essential point, and it is necessary only when, for example, the template is placed in micropores.

If a discrete cluster of nanoscale molecular sieve particles is placed outside the pores of the meso / macroporous structure, it may be desirable to bond at least a portion of the particles to the surface of the structure. There are several ways to achieve this. For example, the surface of a structure can be functionalized with hydroxyl groups or other fragments that will react with zeolite molecular sieves. In the case of polymeric molecular sieves, the surface can be functionalized using fragments that react, such as addition or condensation, with functional fragments on the polymer. These techniques are well known in the art for other applications.

Similar production techniques can be used if it is desired to include at least a portion of the molecular sieve particles in the pores of the meso / macroporous structure. To enter the meso- / macropores, the molecular sieve particles must be of the proper size. The pressure difference can be used to draw the particles of the barrier layer into the pores, or ultrasound can be used to facilitate the entry of the particles of the barrier layer into the pores of the meso / macroporous carrier. The depth of molecular sieve particles in the pores of the meso / macroporous structure should not be so large as to unnecessarily reduce the permeability. Often, any molecular sieves deposited on the surface are removed, for example, by washing.

If desired, zeolite molecular sieves can be grown locally in the pores of the meso / macroporous structure to form a discrete membrane. Synthesis may result in discrete particles or islands among another structure, such as a meso / macroporous structure, or among other particles.

An example of the use of other particles in the preparation of discrete membranes of zeolite molecular sieves includes the presence of silicon dioxide in or on the meso / macroporous structure, which may have a particle size in the range of 5 to 20 nanometers. Silicon dioxide, due to the presence of active hydroxyls on its surface, is used as a nucleation center for a solution of a precursor forming a zeolite, and zeolite layers can be grown on or between silica particles.

As nucleation centers, materials other than silica particles can be used, including other molecular sieves or seed crystals of the same zeolite. The surface of the meso / macroporous structure can be functionalized to provide selective placement of zeolite growth sites. Some zeolites exhibit characteristics of self-nucleation, and, thus, they can be used in the absence of nucleation centers. Examples of these zeolites are FAU and MFI. In these situations, it may be desirable before the precursor solution is brought into contact with the mesoporous / macroporous structure to withstand the precursor solution under conditions of zeolite formation for a period of time sufficient to initiate zeolite growth.

For example, one way to obtain a barrier layer is to place a liquid precursor of zeolite molecular sieves on a meso / microporous structure. The precursor is allowed to crystallize under hydrothermal crystallization conditions, after which the membrane is washed and heated to remove residual organic material. The molecular sieve material remains mainly in the pores of the porous substrate and occludes them.

Molecular sieves may refer to any suitable combination of elements to provide the desired porous structure. As structural elements of molecular sieves, such as molecular sieves based on silicon dioxide-alumina, including zeolites; silicalite; A1PO; SAPO and borosilicates, used aluminum, silicon, boron, gallium, tin, titanium, germanium, phosphorus and oxygen. The precursor includes the above elements, usually in the form of oxides or phosphates, together with water and an organic structuring agent, which is usually a polar organic compound such as tetrapropylammonium hydroxide. Other excipients such as amines, ethers and alcohols may also be used. The mass ratio between the materials of the polar organic compound and the structural element is generally in the range from 0.1 to 0.5 and will depend on the specific structural elements used. In order to obtain thin layers of molecular sieves in the membranes, it is generally preferred that the precursor solution be enriched in water. For example, in the case of molecular sieves based on silica-alumina, the molar ratio between water and silica should correspond to at least 20: 1, and in the case of aluminophosphate molecular sieves, the molar ratio should correspond to at least 20 moles of water per one mole of aluminum.

The crystallization conditions often represent a temperature in the range from 80 ° C to 250 ° C at pressures in the range from 100 to 1000, often from 200 to 500 kPa absolute pressure. The crystallization time is limited so as not to get an excessively thick layer of molecular sieves. In general, the crystallization time is less than 50, say, lasting from 10 to 40 hours. Preferably, the time is sufficient to obtain crystals, but less than what is required to obtain a layer of molecular sieves with a thickness equal to 200 nanometers, say, in the range from 5 to 50 nanometers. Crystallization can be carried out in an autoclave. In some cases, microwave heating can lead to crystallization in a shorter period of time. After that, the membrane is washed with water, and then calcined at a temperature in the range from 350 ° C to 550 ° C to remove any organics.

The production of particles with sizes smaller than 100 nanometers is problematic, especially in the case of some zeolite molecular sieve materials. In addition, even when using seed crystals, the particle size may be large as desired. Another implementation option when producing a membrane including a discrete barrier layer is the synthesis of zeolite in open areas between particles (substrate particles), characterized by a main size of less than 100 nanometers. Accordingly, the main size of the microporous barrier layer may be less than 100 nanometers. Particles of the substrate are used as the nucleation center in the preparation of the zeolite and, thus, they are selected from materials with the ability to nucleate in the growth of the zeolite. Examples of such materials are silicon dioxide, in particular, silicon dioxide, characterized by a main size in the range of 5 to 50 nanometers, and other zeolites, characterized by a main size of less than 100 nanometers. The use of colloidal silicon dioxide as substrate particles is particularly suitable for use in the preparation of the microporous A1PO barrier layer.

The growth of zeolite on the particles of the substrate can occur before or after the use of the particles of the substrate upon receipt of the composite membrane.

In a favorable case, the growth of zeolite on the particles of the substrate occurs during the pulling of the solution for synthesis through the composite. This technique contributes to the passage of growth not only in the form of a layer on top of particles, but also in interstitials between particles. As the zeolite growth progresses, the pressure drop increases, and the pressure drop can be used as an indicator of the proper passage of the zeolite formation.

Polymeric molecular sieves can be synthesized in the meso / macroporous structure. One way to synthesize small polymer molecular sieves is to functionalize the nanoparticles and / or the meso- / macroporous structure with a group that can react with the oligomer, such as through a condensation or addition reaction. For example, functional groups may provide a hydroxyl, amine, anhydride, dianhydride, aldehyde moiety, an amino acid moiety, a carboxyl, amide, nitrile or olefin moiety to undergo an addition or condensation reaction with a reactive oligomer moiety. Suitable oligomers for use may have molecular weights in the range of 30,000 to 500,000 or more and may be reactive polysulfone oligomers; poly (styrenes), including styrene-containing copolymers; cellulosic polymers and copolymers; polyamides; polyimides; polyethers; polyurethanes; polyesters; acrylic and methacrylic polymers and copolymers; polysulfides, polyolefins, in particular vinyl polymers and copolymers; polyallyl; poly (benzimidazole); polyphosphazines; polyhydrazides; polycarbodiids and the like.

The synthesis of molecular sieves, whether inorganic or organic, locally can be carried out under suitable conditions. A preferred technique involves carrying out the synthesis while simultaneously drawing a reagent solution, for example, a precursor solution or an oligomer solution, through a meso / macroporous structure. This technique ensures the advantage of directing the reagent solution into cavities that have not been occluded, and also limits the growth rate of molecular sieves, because once the molecular sieves have occluded meso- or macropore, the fresh reagent will lose its ability to enter the reaction center.

Molecular sieves on membranes comprising a polymeric carrier or on the polymeric carriers themselves can also be pyrolyzed in a vacuum oven to form a carbon membrane. In the case of such membranes containing molecular sieves, the porous structure of the carbon carrier is preferably characterized by a diameter sufficient to minimize resistance to the flow of fluids during separation on the structure of molecular sieves. The pyrolysis temperature will depend on the nature of the polymer carrier and will be lower than the temperature at which the porosity will unnecessarily decrease. Examples of polymeric carriers include polyimides, polyacrylonitrile, polycarbonates, ether-based polyether ketones, ether-based polyethersulfones and polysulfones, and before pyrolysis, the carriers have pores or openings ranging in size from 2 to 100, preferably from 20 to 50 nanometers.

Continuous membranes can be obtained using any suitable technique. Typically, the thickness of the microporous barrier layer will correlate with the duration of deposition or growth of the microporous barrier layer on the meso / macroporous structure. A microporous barrier layer can be obtained by reducing the pore size of the ultrafiltration membrane (effective pore diameters in the range from 1 to 100 nanometers) or microfiltration membrane (effective pore diameters in the range from 100 to 10,000 nanometers) by, for example, applying an organic or inorganic coating to the channel is either on the inner surface, or preferably at least partially in the vicinity of the channel opening. The precipitated material serves to obtain a localized reduction of pores or holes through the carrier to a size that will enable the desired sieving without unduly reducing the diameter of the remaining porous structure in the carrier. Examples of vapor deposited materials include silanes, paraxylene, alkylene imines and alkylene oxides. Another technique for reducing pore size is to deposit a layer of coke on the meso / macroporous structure. For example, a carbonizable gas, such as methane, ethane, ethylene or acetylene, can be brought into contact with the structure at a temperature high enough to lead to coke formation. Preferred porous supports are ultrafiltration membranes characterized by pore sizes ranging from 1 to 80, preferably from 2 to 50 nanometers.

In the case of zeolite continuous membranes, one preparation procedure involves contacting the surface of the mesoporous / macroporous structure with a molecular sieve precursor and growing molecular sieves for a period of time sufficient to obtain the desired film thickness. For the synthesis of molecular sieves, the procedure described above can be used. In some cases, it may be desirable to occlude, for example, using wax, meso- / macropores of the carrier to prevent excessive growth of zeolite in these pores. Subsequently, the wax can be removed.

Various techniques are available to improve the selectivity of highly permeable membranes. There are numerous techniques for correcting defects in continuous or discrete membranes. Since in order to be suitable for use in many fields of application, membranes do not have to demonstrate high values of flow efficiency ratios C ₄ permeates, any technique that will increase the resistance to flow through defects will be used to improve membrane performance. For example, regardless of how the membrane is prepared, a coating formed from a silica sol can be used to occlude openings in the internodes between crystals of molecular sieves or the remaining large pores in the carrier.

Another technique for occluding large pores is to ensure that there is a large reactive molecule on one side of the barrier layer that is unable to penetrate the micropores of the barrier layer and a crosslinker on the other side. Major defects and to some extent minor defects become filled with large reactive molecules, which are fixed due to crosslinking. After this, the unreacted component formed by the large molecule, as well as the non-reactive crosslinker, can be removed. The large molecule may be an oligomer or a large molecule.

In the case of discrete membranes, at least in the part of the cavities between the particles or islands of the microporous barrier layer and between the microporous barrier layer and the meso / microporous structure, a solid phase can be provided.

One characteristic technique for improving the selectivity of a screening membrane is to agglomerate adjoining particles of molecular sieves to reduce or essentially eliminate cavities between particles and between particles and walls of a porous structure in a meso / macroporous structure. Since the particles are characterized by nanometer sizes, and the number of particles adjacent to each other can be relatively small, agglomeration can occur while maintaining the desired permeate flow rates. In the case of polymer molecular sieves that are thermoplastic, agglomeration can occur as a result of heating to a temperature at which the agglomeration occurs, but so high that they lose either their microporous structure or their ability to provide the desired occlusion of the meso or macropores of the meso - / macroporous structure. Agglomeration can also be carried out as a result of calcination of zeolite molecular sieves. Calcination tends to agglomerate small zeolite particles, especially particles that are not silylated or otherwise treated to reduce the tendency to agglomerate. The temperature and duration of calcination will depend on the nature of the zeolite molecular sieves. Typically, temperatures in the range of 450 ° C. to 650 ° C. are used for a period of time from 2 to 20 hours.

The agglomeration technique can be used in relation to particles of molecular sieves that are on the surface of the meso / macroporous structure, as well as those that are in the pores of the structure. Most preferably, agglomeration is used when the molecular sieve particles are located inside the meso- or macropores of the meso- / macroporous structure so that the main size of the agglomerate is less than 200, preferably less than 100 nanometers. Agglomeration can be carried out in the presence or absence of a pressure difference across the membrane. Preferably, the pressure difference is used to help reduce the cavities through which the fluid can bypass the molecular sieves.

Another characteristic technique in which the cavity determines a discrete accumulation of particles of the barrier layer is to at least partially occlude at least a portion of the cavities using solid material in them. Preferably, the solid material is a polymeric or inorganic material. The solid material may simply be located in the cavity, or may adhere, or bind to molecular sieves or a meso / macroporous structure. The solid material may be a particle or oligomer, which can be preformed, and then introduced into the cavity, or they can be molded in place.

In one aspect, the solid material, together with the particles of the microporous barrier layer, forms a “cementitious composition”. The cementitious composition is preferably a suitable polymeric material that can withstand the effects of separation conditions. Such polymers include polysulfones; poly (styrenes), including styrene-containing copolymers; cellulosic polymers and copolymers; polyamides; polyimides; polyethers; polyurethanes; polyesters; acrylic and methacrylic polymers and copolymers; polysulfides, polyolefins, in particular vinyl polymers and copolymers; polyallyls; poly (benzimidazole); polyphosphazines; polyhydrazides; polycarbodiids and the like. Preferred polymers are those which are characterized by the presence of porosity, such as PIM (see document WO 2005/012397), and polymers into which porosity is introduced using pore-forming agents. These polymers have pores that can have a basic size of 0.3 or more, preferably at least 1 nanometer, and thus make it possible for the fluid to flow in and out of the particles of the barrier layer.

There is no need for cementing all particles. Often, the average thickness of the cementitious layer is less than 100 nanometers, and preferably not greater than the main particle size. If you use too much cementitious composition, the result may be a membrane with a mixed structure, and permeability will be unnecessarily impaired. Thus, the mass ratio between the amounts of particles of the barrier layer and the cementitious composition is often in the range from 1: 2 to 100: 1, preferably from 3: 1 to 30: 1.

The cementitious composition and particles can be mixed, for example, to obtain a suspension, and then placed in combination with a microporous structure, or the filing can be carried out after the deposition of particles. The polymer can be obtained locally in the region containing the particles of the barrier layer. Particles of the barrier layer may be inert with respect to polymerization or may have active sites for fixing the polymer. For example, the particles can be functionalized using a reactive group that can bind to a polymer or to a monomer undergoing polymerization, say, by a condensation or addition mechanism, such as discussed above.

The problem is that the cementitious compound occludes the micropores of molecular sieves. In the case of a highly porous polymer such as PIM, the effect of occluding can be reduced. Often the amount of polymer used for the cementitious composition, and its molecular weight, and configuration are such that not enough is present to encapsulate all particles of the polymer molecular sieves. Often the mass ratio between the amounts of polymer and molecular sieves is in the range from 0.01: 1 to 0.3: 1. The weight average molecular weight of the polymer is sometimes in the range from 20,000 to 500,000, preferably from 30,000 to 300,000.

The cementitious composition may be different from the polymer. For example, if the molecular sieves are zeolite, then silicon tetraalkoxide can be reacted with the zeolite and, as a result of hydrolysis between the particles of the molecular sieves, a framework or mass of silicon dioxide can be obtained. Typically, a dilute aqueous solution of silicon tetraalkoxide is used to ensure distribution, for example, containing from 0.5 to 0.25 weight percent silicon tetraalkoxide. The functionalization of zeolite with silicon tetraalkoxide is also suitable for use in the preparation of a crosslinking center with an organic polymer, especially those containing functional groups such as hydroxyl, amine, anhydride, dianhydride, aldehyde groups or amine acid groups, which may form covalent bonds with organosilicon alkoxide. In addition, using the techniques described above, the same or different zeolite can be grown between zeolite particles and zeolite particles and a meso / macroporous structure.

Another approach to reducing the efficiency of the bypass is to use particles with two or more sizes when producing a layer including a barrier layer. If, for example, the particles of the microporous barrier layer are generally spherical at a nominal basic size of 60 nanometers, then the areas between the particles can be of a significant size and make bypass possible. The inclusion of configurationally compatible particles in these regions can impede the flow of the fluid and, as a result, result in the directing of most of the fluid to the particles of the barrier layer for selective separation. The configuration of the particles of the barrier layer will depend on the type of particles used in the barrier layer. Particles of microporous zeolite molecular sieves, characterized by a basic size of less than 100 nanometers, are likely to have a certain configuration due to their crystalline structure. Some zeolites tend to have a plate type configuration, while others, such as A1PO-14, have a rod-like structure. Similarly, particles of polymer, ceramic, glass and carbon molecular sieves can have configurations that will not be easy to change. Thus, the configuration of open regions between particles can vary widely.

Sometimes, configurationally compatible particles are selected to achieve at least partial occlusion of the region. Thus, in the case of spherical particles of the barrier layer, rod-like or much less configurationally compatible particles may be desirable. Configurationally compatible particles can have any composition suitable for use with a given size and operation conditions. Particles can be polymeric, including oligomeric; carbon; and inorganic, such as colloidal silicon dioxide, zeolite, alumina and the like.

Referring to FIG. 1, it can be said that the butane-containing feedstock is fed to the isomerization unit via line 102. As shown, the feedstock is introduced into line 122 between the isobutane stripping column 116 and the membrane separator 124. Alternatively, the feedstock the feed can be introduced into the isomerization unit through an isobutane distillation column 116, as described below. Hydrogen is supplied via line 104 and combined in line 126 with a permeate fraction containing normal butane. The combined isomerization feed stream is passed to the isomerization reactor 106. The effluent of the isomerization reactor 106 is sent via a line 108 to a stabilization column 110. In the stabilization column 110, light fractions and chloride-containing components are removed as distillate via line 112. The light fractions can be used for any suitable purpose, including taking into account their energy value. The bottoms stream from the stabilization column 110 is passed through line 114 to the isobutane 11 distillation column. A distillate is obtained from line 118 from the isobutane distillation column 116. The bottoms stream from the isobutane stripping column 116 is removed through line 120. The side stream containing normal butane from the isobutane stripping column 116 is passed through line 122 to the retentate side of the membrane separator 124. The isobutane enriched stream from separator 124 is removed through line 128 and combined with the distillate in line 118. The permeate fraction is recycled through line 126 to the isomerization reactor 106.

If we turn to figure 2, we can say that the feedstock containing normal butane is fed through line 202 to the retentate side of the membrane separator 204. Permeate, characterized by an increased concentration of normal butane, is passed through line 206 to the isomerization reactor 208 to obtain waste the isomerization stream, which is taken through line 210. A stabilization column is not shown to remove chloride components and light fractions from the isomerization effluent. The retentate from the membrane separator 204 is passed through line 212 to combine with the isomerization effluent to produce an isobutane product. As an example, we can say that under the assumption that the starting material contains 50 wt.% Normal butane and 50 wt.% Isobutane and assuming that the retentate stream contains 75% isobutane of the feedstock with a purity of 95 wt.% And equilibrium is achieved during isomerization distribution in the form of 60 wt.% isobutane to 40 wt.% normal butane, the combined product stream will have an isobutane concentration of approximately 75 wt.%.

Claims (10)

1. The method of isomerization of a feedstock containing normal butane, fed to isomerization, containing at least 50 wt.% Normal butane, to obtain an isomerate containing isobutane, including:
(a) isomerization of the feed starting material under isomerization conditions, including the presence of an isomerization catalyst, to obtain an isomerization effluent containing normal butane but with a concentration lower than that in the feed to isomerization;
(b) distilling at least a portion of the isomerization effluent to obtain a lower boiling fraction containing isobutane and lighter paraffins, where at least 80 wt.% of the lower boiling fraction are isobutane, and a higher boiling fraction containing normal butane, which contains normal butane and at least 10 wt.% isobutane;
(c) introducing at least a portion of the fraction containing normal butane from step (b) into contact with a retentate-side medium of a selectively permeable membrane having a flow efficiency index C _{4 of} permeate equal to at least 0.01, and the pressure difference between the media on both sides of the membrane, providing a retentate fraction containing at least 80 wt.% isobutane, and to obtain, after passing through the membrane on the permeate side, a permeate fraction having an increased concentration of normal butane; and
(d) selection from step (c) of the retentate fraction. 2. The method according to claim 1, in which at least a portion of the permeate fraction is recycled to step (a) as a portion of the starting material fed to isomerization. 3. The method according to claim 2, in which the distillation of stage (b) involves the use of a ratio between the amounts of reflux and the feed material in the range from 0.5: 1 to 3: 1. 4. The method according to claim 2, in which the concentration of normal butane in the permeate fraction from step (c) is at least 50 wt.%. 5. The method according to claim 1, in which the membrane is a sieving membrane, characterized by a flow efficiency index of C ₄ permeate equal to at least 0.01, and a ratio of flow efficiencies of C ₄ permeate corresponding to at least 1, 25: 1. 6. The method according to claim 5, in which the sieving membrane is characterized by an average pore diameter in the range from 5.0 to 7.0

. 7. The method according to claim 5, in which the sieving membrane is characterized by an average pore diameter in the range from 4.5 to 5.4

. 8. The method according to claim 2, in which the feedstock containing normal butane is fed to at least one position selected from step (b) and the side of the membrane retentate from step (c). 9. A method of isomerizing normal butane, including:
distillation of at least a portion of the normal butane contained in the feed, including isobutane, providing a lower boiling fraction containing isobutane and a higher boiling fraction containing normal butane and at least 10 wt.% isobutane;
(a) contacting at least a portion of the higher boiling fraction containing normal butane with the medium on the retentate side of a selectively permeable membrane under conditions including a flow efficiency index of C ₄ permeate of at least 0.01, the ratio the permeate flow efficiencies of at least 1.25: 1 and the pressure difference on both sides of the membrane from 10 to 200 kPa to obtain a retentate fraction containing at least 80 wt.% isobutane, and to obtain after passing through the membrane to side permeate fraction lane EATA having an increased concentration of normal butane, said permeate fraction contains at least 80 wt% of the normal butane contained in the feedstock contacted with the membrane.; and
(b) exposing the permeate fraction to isomerization conditions, including the presence of an isomerization catalyst to produce an isomerization effluent containing normal butane, with a concentration lower than that in the feedstock. 10. Installation for isomerization of feedstock containing normal butane, in accordance with paragraphs.1-9, including:
(a) an isomerization reactor adapted to carry out isomerization of normal butane to produce an isomerate containing isobutane and having an inlet and an outlet for isomerate;
(b) a distillation column having an inlet through fluid in communication with the outlet of the isomerization reactor, an outlet channel for a lower boiling fraction and an outlet channel for a higher boiling fraction, wherein said distillation column is adapted to distill at least , parts of isobutane-containing isomerate to obtain a lower boiling fraction, in comparison with isobutane-enriched isomerate, and a higher boiling fraction, in comparison with about enriched with normal butane, wherein said fraction contains isobutane; and
(c) a membrane separator having an inlet on the side of the feed source through a fluid in communication with the channel for the higher boiling fraction, an outlet on the side of the feed inlet through a fluid in communication with the line from the outlet of the distillation column for lower boiling fraction, and the outlet for permeate, through a fluid in communication with the inlet of the isomerization reactor.

Images