PROCESS

FIELD OF THE INVENTION

The present invention relates to a method to remove one or more contaminant from a gas stream, preferably a hydrocarbon gas steam, using a solid adsorbent and regenerating said adsorbent by an improved pressure swing adsorption process.

BACKGROUND OF THE INVENTION

The present invention relates to a method of removing a component or components from a gas stream by adsorption onto a solid adsorbent with regeneration of the adsorbent at intervals.

In such methods, the gas stream is fed in contact with a solid adsorbent to adsorb the component to be removed which gradually builds-up in the adsorbent. The concentration of the removed component in the adsorbent will gradually rise. The concentration of the removed gas component in the adsorbent will not be uniform but will be highest at the upstream end of the adsorbent bed and will tail off progressively through a mass transfer zone in the adsorbent. If the process is conducted indefinitely, the mass transfer zone will progressively move downstream in the adsorbent bed until the component which is to be removed breaks through from the downstream end of the bed. Before this occurs, it is necessary to regenerate the adsorbent.

In the pressure swing adsorption (PSA) system adsorbent regeneration is done by stopping the flow into the adsorbent of gas to be treated, depressurising the adsorbent and, usually, by passing through the bed counter-current to the product feed direction a flow of a regenerating gas, usually at a lower pressure than the gas to be treated and low in its content of the component adsorbed on the bed.

An alternative procedure is known as temperature swing adsorption (TSA). In TSA, to achieve regeneration heat is supplied to desorb the adsorbed gas component. To this end the regenerating gas used is heated for a period to produce a heat pulse moving through the bed counter-current to the normal feed direction. This flow of heated regenerating gas is followed by a flow of cool regenerating gas which continues the displacement of the heat pulse through the bed toward the upstream end. TSA generally requires greater than 2 beds due to the requirement of longer heating and/or cooling periods. TSA is characterized by an extended cycle time as compared to PSA.

Each procedure has its own characteristic advantages and disadvantages. TSA is energy intensive because of the need to supply heat to the regenerating gas. The temperatures needed for the regenerating gas are typically sufficiently high, e.g. 150°C to 200°C, as to place demands on the system engineering which increase costs. The high temperature used in a TSA system give rise to a need for the use of insulated vessels, a purge preheater and an inlet end precooler and generally the high temperatures impose a more stringent and costly mechanical specification for the system. In operation, there is extra energy cost associated with using the purge preheater and the additional cooling step.

In PSA, a short cycle time is typically used in order to avoid the heat pulse leaving the adsorbent bed which requires frequent depressurisation of the bed, during which the unadsorbed feed gas in the adsorber is vented off and lost ("switch loss"). In addition, it is usual to use two adsorbent beds, with one being on-line while the other is regenerated. The depressurisation and regeneration of one bed may take place during the short time for which the other bed is on-line.

USP 4,857,084 discloses a PSA process that regenerates under vacuum pressures.

USP 5,779,768 discloses a TSA process wherein purified gas is heated (up to 300°C) to a temperature above that of the feed gas and used alone or with a make-up gas to regenerate the adsorbent. Regeneration with the heated regenerating gas is followed by a cooling step.

USP 5,855,650 discloses a PSA process using a first heated low pressure regenerating gas, preferably nitrogen, followed by a second non-heated low pressure regenerating gas, again preferably nitrogen, to regenerate adsorbent use to purify air wherein the regenerating gas is provided from an outside source, not usage of the purified gas.

USP 7,066,986: discloses variations of a TSA separation process wherein the nitrogen, oxygen, and argon components from a purified air are cryogenically separated then the nitrogen component is heated and used as a regenerating gas for the adsorbent used top purify the air.

USP 8,734,571 discloses a PSA process using a using a first heated (20°C to 80°C) low pressure regenerating gas, preferably nitrogen, and then a second regenerating gas at a lower temperature than the first regenerating gas to regenerate adsorbent use to purify air wherein the first and second regenerating gases are provided from an outside source, not usage of the purified gas.

US Publication No. 2014/0338425 discloses a PSA process using a heated (140°C to 220°C) low pressure regenerating gas, preferably nitrogen, to regenerate adsorbent use to purify air wherein the regenerating gas is provided from an outside source, not usage of the purified gas.

There remains a need for an improved PSA adsorbent regeneration process. SUMMARY OF THE INVENTION

The present invention is such an improved PSA regeneration process.

The present invention is a method for removing at least one component from a hydrocarbon gas stream by a pressure swing adsorption process having an adsorbent regeneration phase, said method comprising the steps of: (i) passing the hydrocarbon gas stream having a first temperature and a first pressure in a first direction in contact with an adsorbent to adsorb the component from the hydrocarbon gas stream on the adsorbent providing a component loaded adsorbent and a component lean hydrocarbon gas stream, (ii) ceasing passing said hydrocarbon gas stream in contact with said adsorbent, (iii) lowering the pressure of a portion of the component lean hydrocarbon gas stream to a second pressure, which is less than the first pressure, for use as a regenerating gas, (iv) heating a portion of the component lean hydrocarbon gas stream to a second temperature for use as a regenerating gas raising the temperature of the regenerating gas to a temperature above that of said component lean hydrocarbon gas stream, (v) passing said lower pressure/heated regenerating gas in a second direction opposite to said first direction to partially desorb said hydrocarbon gas stream component from said adsorbent forming a hydrocarbon gas stream component rich regenerating gas and partially regenerated adsorbent, (vi) ceasing passing said lower pressure/heated regenerating gas in contact with said adsorbent and (vii) repeating steps (i) to (vi), wherein step iii) may proceed or follow step iv) and there is no cooling the adsorbent step between step vi) and step vii).

In one embodiment of the method of the present invention disclosed herein above, step (ii) further comprises reducing the gas pressure over said adsorbent and wherein said gas pressure is restored prior to or at the commencement of repeating step (i). In one embodiment of the method of the present invention disclosed herein above, there are at least two vessels containing said adsorbent and said method is operated on each vessel with the steps so phased between the vessels that at least one vessel is in step (i) whilst another is in steps (v).

In one embodiment of the method of the present invention disclosed herein above, the second temperature of the regenerating gas is 1°C to 80°C hotter than the first temperature of the hydrocarbon gas stream.

In one embodiment of the method of the present invention disclosed herein above, the second pressure of the regenerating gas is 1 to 45 psia.

In one embodiment of the method of the present invention disclosed herein above, step (v) further comprises the step of passing said lower pressure regenerating gas in a second direction opposite to said first direction to partially desorb said hydrocarbon gas stream component from said adsorbent forming a hydrocarbon gas stream component rich regenerating gas and partially regenerated adsorbent.

In one embodiment of the method of the present invention disclosed herein above, the adsorbent is silica gel, alumina, silica-alumina, zeolites, activated carbon, polymer supported silver chloride, copper-containing resins, porous cross-linked polymeric adsorbents, pyrolized macroporous polymers, or mixtures thereof.

In one embodiment of the method of the present invention disclosed herein above, the adsorption media is a porous cross-linked polymeric adsorbent, a pyrolized macroporous polymer, or mixtures thereof.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a flow diagram of one embodiment of the adsorption process according the present invention.

FIG. 2 illustrates a flow diagram of another embodiment of the adsorption process according the present invention.

FIG. 3 is a plot of the working capacity of butane at various regeneration temperatures and various regeneration pressures for an adsorbent of the present invention. DETAILED DESCRIPTION OF THE INVENTION

The present invention is an improved process to remove one or more components from a gas stream, for example gas streams from a refinery operation, petrochemical operation, or other operations, preferably the gas stream is a natural gas steam. The present process is particularly suitable for gas streams comprising mixtures of two or more of methane, ethane, propane, butane, and/or heavier hydrocarbons. The gas stream may further comprise gasses common to gas streams such as, but not limited to, carbon dioxide (CO ₂ ), hydrogen sulfide (H ₂ S), sulfur dioxide (SO ₂ ), carbon disulfide (CS ₂ ), hydrogen cyanide (HCN), carbonyl sulfide (COS), mercaptans, ethylene, propylene, butenes, and the like.

Raw natural gas comes from three types of wells: oil wells, gas wells, and condensate wells. Natural gas that comes from oil wells is typically termed "associated gas". This gas can exist separate from oil in the formation (free gas), or dissolved in the crude oil (dissolved gas). Natural gas from gas and condensate wells, in which there is little or no crude oil, is termed "non-associated gas". Gas wells typically produce raw natural gas by itself, while condensate wells produce free natural gas along with a semi-liquid hydrocarbon condensate. Whatever the source of the natural gas, once separated from crude oil (if present) it commonly exists in mixtures with other hydrocarbons; principally ethane, propane, butane, and pentanes and to a lesser extent heavier hydrocarbons.

Raw natural gas often contain a significant amount of impurities, such as water or acid gases, for example carbon dioxide (CO ₂ ), hydrogen sulfide (H ₂ S), sulfur dioxide (SO ₂ ), carbon disulfide (CS ₂ ), hydrogen cyanide (HCN), carbonyl sulfide (COS), or mercaptans as impurities. The term "natural gas feedstream" as used in the method of the present invention includes any natural gas source, raw natural gas, flare gas, or raw natural gas that has been treated one or more times to remove water and/or other impurities.

The terms "natural gas liquids" (NGL) and "ethane plus" (C ₂ +) refer broadly to hydrocarbons having two or more carbons such as ethane, propane, butane, and possibly small quantities of pentanes or heavier hydrocarbons. Preferably, NGL have a methane concentration of 5 mol percent or less.

The term "methane-rich" refers broadly to any vapor or liquid stream, e.g., after separation from which ethane plus amounts have been recovered or removed. Thus, a methane- rich stream has a higher concentration of Ci than the concentration of Ci in associated and non- associated natural gas. Preferably, the concentration increase of Ci is from removal of at least a portion of one or more heavier hydrocarbons. When described in terms of the component being removed, a gas stream from which a component, for example one or more C ₂ + hydrocarbon, is removed is referred to as "component lean" gas stream.

As used herein, the term "hydrocarbon gas stream" refers to any raw or treated natural gas stream comprising methane and one or more additional component, said components include, but are not limited to, carbon dioxide (CO ₂ ), hydrogen sulfide (H ₂ S), sulfur dioxide (S0 ₂ ), carbon disulfide (CS ₂ ), hydrogen cyanide (HCN), carbonyl sulfide (COS), mercaptans, ethane, propane, butane, pentanes, and heavier hydrocarbons.

The process of the present invention removes one or more said component from a hydrocarbon gas stream by a pressure swing adsorption process, preferably thermally enhanced pressure swing adsorption process.

Now referring to the drawings, as shown in FIG. 1 a feed gas, for example a hydrocarbon gas, to be purified having a first pressure and a first temperature is supplied through inlet 1 to an inlet manifold 2 containing inlet control valves 3 and 19 to which is connected a pair of adsorbent bed containing vessels 20 and 40. The inlet manifold is bridged downstream of the control valves 3 and 19 by a venting manifold 18 containing venting valves 5 and 17 which serve to close and open connections between the upstream end of respective adsorbent vessels 20 and 40 and a discharge 50. Each of the two adsorbent beds 20 and 40 contains an adsorbent.

Passing the hydrocarbon gas through the adsorbent vessel comprising adsorbent provides a component lean hydrocarbon gas stream having a first pressure. The apparatus has an outlet 11 connected to the downstream ends of the two adsorbent vessels 20 and 40 by an outlet manifold 10 containing outlet control valves 9 and 13. The outlet manifold 10 is bridged by a regenerating gas manifold 35 containing regenerating gas control valves 7 and 15. Bridging the outlet manifold 10 and the regenerating gas manifold 35 is line 31 containing a pressure drop valve 30 and a heater 32 (FIG. 1) between the pressure drop valve 30 and the outlet 11 and/or a heater 34 (FIG. 2) between the pressure drop valve 30 and the regenerating gas manifold 35. Purified gas, or product gas (e.g., the component lean hydrocarbon gas stream), passes through the outlet 11. When purification is complete and prior to regeneration of the adsorbent in the same vessel, the flow of the hydrocarbon gas stream in contact with said adsorbent is stopped.

In one embodiment of the process of the present invention (FIG. 1), an inlet for regenerating gas is provided at 31 wherein a portion of the purified gas at a first pressure passes through a heater 32 and then through pressure drop valve 30 to provide a heated portion of purified gas having a second pressure lower than the first pressure. The heated portion of purified gas with a second pressure passes into the regenerating gas manifold 35 and passes through either valve 7 to enter vessel 20 or through valve 15 to enter vessel 40.

In another embodiment of the process of the present invention (FIG. 2), an inlet for regenerating gas is provided at 31 wherein a portion of the purified gas at a first pressure passes through the pressure drop valve 30 and then through a heater 34 to provide a heated portion of purified gas having a second pressure lower than the first pressure. The heated portion of purified gas with a second pressure passes into the regenerating gas manifold 35 and passes through either valve 7 to enter vessel 20 or through valve 15 to enter vessel 40.

The operation of the valves may be controlled by suitable programmable timing and valve opening means as known in the art, not illustrated.

In one embodiment of the operation of the present process, a hydrocarbon gas at a first temperature and pressure comprising at least one component to be removed is fed through inlet 1 to the inlet manifold 2 and passes through one of the two vessels containing adsorbent. Starting from a position in which the hydrocarbon gas is passing through open valve 3 into and in a first direction through adsorbent vessel 20, and through open valve 9 to the outlet 11, valve 19 in the inlet manifold 2 will just have been closed to cut-off vessel 40 from the feed of the hydrocarbon gas for purification. Valve 13 will just have closed also. At this stage valves 5 and 7 are closed. Vessel 20 is thus on-line for a first amount of time and bed 40 is to be regenerated. After the hydrocarbon gas has passed through the adsorbent bed, a component loaded adsorbent and a component lean hydrocarbon gas stream are formed.

When the end of the adsorption period is reached, valve 3 and 9 are closed and valve 5 is opened allowing the depressurization of vessel 20. After a second amount of time valve 7 and pressure drop valve 30 are opened to allow for the flow of the regenerating gas through vessel 20.

When the regeneration step is complete, valves 5 and 7 are closed and valve 3 is opened allowing feed gas to repressurize vessel 20 for a third amount of time. After repressurization of vessel 20 valve 9 is opened and the adsorption step is repeated for a first amount of time.

Although the hydrocarbon gas enters the vessel 20 at a first temperature, it is often warmed by heat generated as the component is adsorbed onto the adsorbent in the vessel 20. The component lean hydrocarbon gas emerging from the outlet 11 often has a slightly warmer temperature than when entering the vessel 20. In one embodiment of the present invention (FIG. 1), to commence regeneration of bed 40, a portion of the component lean hydrocarbon gas is heated in heater 32 and pressure drop valve 30 is opened to pass the heated portion of the component lean hydrocarbon gas providing a heated component lean hydrocarbon gas having a lower pressure than the heated component lean hydrocarbon gas entering the pressure drop valve 30. Said heated/lower pressure component lean hydrocarbon gas, i.e., the regenerating gas, passes through regeneration manifold 35, wherein valve 7 is closed and valve 15 is opened, into vessel 40.

In another embodiment of the present invention (FIG. 2), to commence regeneration of bed 40, pressure drop valve 30 is opened and a portion of the component lean hydrocarbon gas passes through the pressure drop valve 30 to provide a portion of the component lean hydrocarbon gas having a lower pressure than the portion of the component lean hydrocarbon gas prior to passing through the pressure drop valve 30. The portion of the component lean hydrocarbon gas having a lower pressure passes into the heater 34 and is heated. Said heated/lower pressure component lean hydrocarbon gas, i.e., the regenerating gas, passes through regeneration manifold 35, wherein valve 7 is closed and valve 15 is opened, into vessel 40.

In both embodiments, valve 17 is kept open whilst valve 15 is opened to commence a flow of regenerating gas in a second direction opposite to said first direction. Although the regenerating gas enters the vessel 40 at the selected elevated temperature, it is cooled by giving up heat to desorb the adsorbed component of the adsorbent in the vessel 40. Since the heat pulse is retained and consumed in the system, the exit purge gas typically emerges from the discharge outlet 50 in a cooled state. Progressively, a heat wave moves through the adsorbent as the adsorbed component is cleared forming a component rich hydrocarbon gas stream regenerating gas and regenerated adsorbent. The lower pressure regenerating gas flows through the vessel 40 and removes the adsorbed component and partially regenerates the adsorbent in the vessel 40. At the end of the allotted regeneration period, valve 15 is closed to end the flow of regenerating gas and, after the closing of valve 17. Valve 19 is opened to allow repressurization. After an amount of time valve 13 is opened and vessel 40 is back on line. At this time valves 3 and 9 are closed. Once the passing of the hydrocarbon gas stream in contact with said adsorbent vessel 20 has been stopped, the adsorbent in vessel 20 may then be depressurized and regenerated in a similar manner and the whole sequence continued with the vessels being on-line, depressurising, regenerating, repressurising, and going back on-line in phased cycles of operation. In one embodiment of the process of the present invention, there is no step wherein the adsorbent is cooled between the last step of regenerating the adsorbent in vessel 40, i.e., ceasing passing the lower pressure/heated regenerating gas in contact with the adsorbent and the first step of using vessel 40 to remove one or more component from the

hydrocarbon gas stream, i.e., the step of passing the hydrocarbon gas at a first temperature and pressure comprising at least one component to be removed through inlet 1 to the inlet manifold 2 and passes through vessel 40 containing regenerated adsorbent.

In one embodiment of the process of the present invention, the lower pressure regenerating gas is heated for only a portion of the regeneration step, i.e., not for the complete regeneration step.

In one embodiment of the present invention, the process comprises more than two adsorption/regeneration beds wherein one or more participate in adsorption at the same time.

In one embodiment of the present invention, the process further comprises the step of pressure equalization at the beginning of the depressurization/repressurization cycle where valves 17 and 5 are kept open at the same time to allow the gas from the previously adsorbing vessel to help pressurize the previously regenerating vessel.

The adsorbent used in the apparatus and method described above may be of several kinds. Each of the adsorbent vessels may contain a single type of adsorbent or may contain more than one type of adsorbent. In one embodiment of the present invention, the vessels may comprise a layered bed configuration containing an upstream layer of a first adsorbent followed by a downstream layer of a second adsorbent. In another embodiment of the present invention, a vessel may comprise a layered bed configuration wherein every bed contains one or more adsorbents blended together. In another embodiment of the present invention, a vessel may comprise a layered bed configuration wherein one or more adsorbents may be present in one or more beds.

It should be understood of course that the vessels 20 and 40 can each if desired be separated into smaller vessels arranged in series and references to "layers" of adsorbent above include arrangements in which the separate adsorbents are placed to separate vessels arranged in series.

Suitable adsorbents for use in the process of the present invention are solids having a microscopic structure. The internal surface of such adsorbents is preferably between 100 to 2000 m ² /g, more preferably between 500 to 1500 m ² /g, and even more preferably between 1000 to 1300 m ² /g. The nature of the internal surface of the adsorbent in the adsorbent bed is such that light hydrocarbons (C ₂ and C ₃ ) and heavier hydrocarbons (C ₄ +) are adsorbed. Suitable adsorbent media include materials based on silica, silica gel, alumina or silica- alumina, zeolites, activated carbon, polymer supported silver chloride, copper-containing resins. Most preferred adsorbent media is a porous cross-linked polymeric adsorbent or a partially pyrolized macroporous polymer. Preferably, the internal surface of the adsorbent is non-polar.

In one embodiment, the process of the present invention uses an adsorbent media to extract one or more component, for instance NGLs, from a natural gas stream. The mechanism by which the macroporous polymeric adsorbent extracts the NGLs from the natural gas stream is a combination of adsorption and absorption; the dominating mechanism at least is believed to be adsorption. Accordingly, the terms "adsorption" and "adsorbent" are used throughout this specification, although this is done primarily for convenience. The invention is not considered to be limited to any particular mechanism.

When an adsorbent media has adsorbed any amount of C ₂ + hydrocarbons it is referred to as "loaded". Loaded includes a range of adsorbance from a low level of hydrocarbons up to and including saturation with adsorbed hydrocarbons.

The term "macroporous" is used in the art interchangeably with "macroreticular" and refers in general to pores with diameters of 500 A or greater. "Mesopores" are characterized as pores of between 50 Angstroms and larger but less than 500 angstroms. "Micropores" are characterized as pores of less than 50 Angstroms. The engineered distribution of these types of pores gives rise to the desired properties of high adsorption capacity for NGLs and ease of desorption of NGLs under convenient/practical chemical engineering process modifications (increase in temperature or reduced pressure [vacuum]). The process giving rise to the distribution of micropores, mesopores and macropores can be achieved in various ways, including forming the polymer in the presence of an inert diluent or other porogen to cause phase separation and formation of micropores by post cross-linking.

In one embodiment, the adsorbent media of the process of the present invention is a macroporous polymeric adsorbent of the present invention is a post cross-linked polymeric synthetic adsorbents engineered to have high surface area, high pore volume and high adsorption capacities as well as an engineered distribution of macropores, mesopores and micropores. Preferably, the macroporous polymeric adsorbent useful in the process of the present invention is hypercrosslinked and/or methylene bridged having the following characteristics: a BET surface area of equal to or greater than 500 m ² /g and preferably equal to or greater than 1,000 m ² /g, and having a particle size of 300 microns to 1500 microns, preferably 500 to 1200 microns.

Examples of monomers that can be polymerized to form macroporous polymeric adsorbents useful are styrene, alkylstyrenes, halostyrenes, haloalkylstyrenes, vinylphenols, vinylbenzyl alcohols, vinylbenzyl halides, and vinylnaphthalenes. Included among the substituted styrenes are ortho-, meta-, and para-substituted compounds. Specific examples are styrene, vinyltoluene, ethylstyrene, t-butylstyrene, and vinyl benzyl chloride, including ortho-, meta-, and para-isomers of any such monomer whose molecular structure permits this type of isomerization. Further examples of monomers are polyfunctional compounds. One preferred class is polyvinylidene compounds, examples of which are divinylbenzene, trivinylbenzene, ethylene glycol dimethacrylate, divinylsulfide and divinylpyridine. Preferred polyvinylidene compounds are di- and trivinyl aromatic compounds. Polyfunctional compounds can also be used as crosslinkers for the monomers of the first group.

One preferred method of preparing the polymeric adsorbent is by swelling the polymer with a swelling agent, then crosslinking the polymer in the swollen state, either as the sole crosslinking reaction or as in addition to crosslinking performed prior to swelling. When a swelling agent is used, any pre-swelling crosslinking reaction will be performed with sufficient crosslinker to cause the polymer to swell when contacted with the swelling agent rather than to dissolve in the agent. The degree of crosslinking, regardless of the stage at which it is performed, will also affect the porosity of the polymer, and can be varied to achieve a particular porosity. Given these variations, the proportion of crosslinker can vary widely, and it is not restricted to particular ranges. Accordingly, the crosslinker can range from 0.25% of the polymer to 45%. Best results are generally obtained with 0.75% to 8% crosslinker relative to the polymer, the remaining (noncrosslinking) monomer constituting from 92% to 99.25% (all percentages are by weight).

Other macroporous polymeric adsorbents useful in the practice of this invention are copolymers of one or more monoaromatic monomers with one or more nonaromatic monovinylidene monomers. Examples of the latter are methyl acrylate, methyl methacrylate and methylethyl acrylate. When present, these nonaromatic monomers preferably constitute less than 30% by weight of the copolymer.

The macroporous polymeric adsorbent is prepared by conventional techniques, examples of which are disclosed in various United States patents. Examples are USP

4,297,220; 4,382,124; 4,564,644; 5,079,274; 5,288,307; 4,950,332; and 4,965,083. The disclosures of each of these patents are incorporated herein by reference in their entirety. For polymers that are swollen and then crosslinked in the swollen state, the crosslinking subsequent to swelling can be achieved in a variety of ways, which are further disclosed in the patents cited above. One method is to first haloalkylate the polymer, and then swell it and crosslink by reacting the haloalkyl moieties with aromatic groups on neighboring chains to form an alkyl bridge. Haloalkylation is achieved by conventional means, an example of which is to first swell the polymer under non-reactive conditions with the haloalkylating agent while including a Friedel-Crafts catalyst dissolved in the haloalkylating agent. Once the polymer is swollen, the temperature is raised to a reactive level and maintained until the desired degree of haloalkylation has occurred. Examples of haloalkylating agents are chloromethyl methyl ether, bromomethyl methyl ether, and a mixture of formaldehyde and hydrochloric acid. After haloalkylation, the polymer is swelled further by contact with an inert swelling agent. Examples are dichloroethane, chlorobenzene, dichlorobenzene, ethylene dichloride, methylene chloride, propylene dichloride, and nitrobenzene. A Friedel-Crafts catalyst can be dissolved in the swelling agent as well, since the catalyst will be used in the subsequent crosslinking reaction. The temperature is then raised to a level ranging from 60°C to 85 °C in the presence of the catalyst, and the bridging reaction proceeds. Once the bridging reaction is complete, the swelling agent is removed by solvent extraction, washing, drying, or a combination of these procedures.

The pore size distribution and related properties of the finished adsorbent can vary widely and no particular ranges are critical to the invention. In most applications, best results will be obtained at a porosity (total pore volume) within the range of from 0.5 to 1.5 cc/g of the polymer. A preferred range is 0.7 to 1.3 cc/g. Within these ranges, the amount contributed by macropores (i.e., pores having diameters of 500 A or greater) will preferably range from 0.025 to 0.6 cc/g, and most preferably from 0.04 to 0.5 cc/g. The surface area of the polymer, as measured by nitrogen adsorption methods such as the well-known BET method, will in most applications be within the range of 150 to 2100 m ² /g, and preferably from 400 to 1400 m ² /g. The average pore diameter will most often range from 10 A to 100 A.

The form of the macroporous polymeric adsorbent is likewise not critical and can be any form which is capable of containment and contact with a flowing compressed air stream.

Granular particles and beads are preferred, ranging in size from 50 to 5,000 microns, with a range of 500 to 3,000 microns particularly preferred. Contact with the adsorbent can be achieved by conventional flow configurations of the gas, such as those typically used in fluidized beds or packed beds. The adsorbent can also be enclosed in a cartridge for easy removal and replacement and a more controlled gas flow path such as radial flow. The macroporous polymeric adsorbent can function effectively under a wide range of operating conditions. The temperature will preferably be within any range which does not cause further condensation of vapors or any change in physical or chemical form of the adsorbent. Preferred operating temperatures are within the range of from 5°C to 100°C, and most preferably from 10°C to 80°C. The pressure of the natural gas stream entering the adsorbent bed can vary widely as well, preferably extending from 2 psig (115 kPa) to 1000 psig (7000 kPa). The pressure will generally be dictated by the plant unit where the product gas will be used. A typical pressure range for some applications is from 100 psig (795 kPa) to 300 psig (2170 kPa). The residence time of the natural gas stream in the adsorbent bed will most often range from 10 seconds to 15 minutes and preferably from 1 minute to 5 minutes. Finally, the relative humidity can have any value up to 100%, although for convenience, the preferred range of relative humidity is 0.1% to 98%.

The macroporous polymeric adsorbents useful in the process of the present invention described herein above can be used to separate ethane, propane, butane, pentane, and heaver hydrocarbons from mixed gases. Preferably, the macroporous polymeric adsorbents useful in the process of the present invention adsorb equal to or greater than 60 cm ³ STP of propane per gram of sorbent at 35°C and 500 mmHg of propane. Preferably, the adsorbents useful in the process of the present invention adsorb equal to or greater than 60 cm ³ STP of n-butane per gram of sorbent at 35°C and 100 mmHg of n-butane.

In another embodiment, the process of the present invention uses a pyrolized macroporous polymeric adsorbent media to extract NGLs from a natural gas stream.

Pyrolized macroporous polymeric adsorbent media are well known, for instance see USP 4,040,990, incorporated by reference herein in its entirety. Partially pyrolyzed particles, preferably in the form of beads or spheres, produced by the controlled decomposition of a synthetic polymer of specific initial porosity. In a preferred embodiment, the pyrolyzed particles are derived from the thermal decomposition of macroreticular ion exchange resins containing a macroporous structure.

In general pyrolysis comprises subjecting the starting polymer to controlled

temperatures for controlled periods of time under certain ambient conditions. The primary purpose of pyrolysis is thermal degradation while efficiently removing the volatile products produced.

The maximum temperatures may range from 300°C to up to 900°C, depending on the polymer to be treated and the desired composition of the final pyrolyzed particles. Higher temperature, e.g., 700°C and higher result in extensive degradation of the polymer with the formation of molecular sieve sized pores in the product. Most desirably, thermal decomposition (alternatively denoted "pyrolysis" or "heat treatment") is conducted in an inert atmosphere comprised of, for example, argon, neon, helium, nitrogen, or the like, using beads of macroreticular synthetic polymer substituted with a carbon- fixing moiety which permits the polymer to char without fusing in order to retain the macroreticular structure and give a high yield of carbon. Among the suitable carbon-fixing moieties are sulfonate, carboxyl, amine, halogen, oxygen, sulfonate salts, carboxylate salts and quaternary amine salts. These groups are introduced into the starting polymer by well-known conventional techniques, such as those reactions used to functionalize polymers for production of ion exchange resins. Carbon-fixing moieties may also be produced by imbibing a reactive precursor thereof into the pores of macroreticular polymer which thereupon, or during heating, chemically binds carbon-fixing moieties onto the polymer. Examples of these latter reactive precursors include sulfuric acid, oxidizing agents, nitric acid, Lewis acids, acrylic acid, and the like.

Suitable temperatures for are generally within the range of 300°C to 900°C, although higher temperatures may be suitable depending upon the polymer to be treated and the desired composition of the final pyrolyzed product. At temperatures above 700°C the starting polymer degrades extensively with the formation of molecular sieve sized pores in the product, i.e., 4 to 6 A average critical dimension, yielding a preferred class of adsorbents. At lower temperatures, the thermally-formed pores usually range from 6 A to as high as 50 A in average critical size. A preferred range of pyrolysis temperatures is between 400°C and 800°C. As will be explained more fully hereinafter, temperature control is essential to yield a partially pyrolyzed material having the composition, surface area, pore structures and other physical characteristics of the desired product. The duration of thermal treatment is relatively unimportant, providing a minimum exposure time to the elevated temperature is allowed.

A wide range of pyrolyzed resins may be produced by varying the porosity and/or chemical composition of the starting polymer and also by varying the conditions of thermal decomposition. In general, the pyrolyzed resins useful in the process of the invention have a carbon to hydrogen ratio of 1.5 : 1 to 20 : 1, preferably 2.0 : 1 to 10 : 1, whereas activated carbon normally has a C/H ratio much higher, at least greater than 30 : 1 (Carbon and Graphite Handbook, Charles L. Mantell, Interscience Publishers, N.Y. 1968, p. 198). The product particles contain at least 85% by weight of carbon with the remainder being principally hydrogen, alkali metals, alkaline earth metals, nitrogen, oxygen, sulfur, chlorine, etc., derived from the polymer or the functional group (carbon-fixing moiety) contained thereon and hydrogen, oxygen, sulfur, nitrogen, alkali metals, transition metals, alkaline earth metals and other elements introduced into the polymer pores as components of a filler (may serve as a catalyst and/or carbon-fixing moiety or have some other functional purpose).

The pore structure of the final product must contain at least two distinct sets of pores of differing average size, i.e., multimodal pore distribution. The larger pores originate from the macroporous resinous starting material which preferably contains macropores ranging from between 50 to 100,000 A in average critical dimension. The smaller pores, as mentioned previously, generally range in size from 4 to 50 A, depending largely upon the maximum temperature during pyrolysis. Such multimodal pore distribution is considered a novel and essential characteristic of the composition for use in the process of the invention.

The pyrolyzed polymers useful in the process of the present invention have relatively large surface area resulting from the macroporosity of the starting material and the smaller pores developed during pyrolysis. In general the overall surface area as measured by nitrogen adsorption ranges between 50 and 1500 m ² /gram. Of this, the macropores will normally contribute 6 to 700 m ² /gram, preferably 6 to 200 m ² /g, as calculated by mercury intrusion techniques, with the remainder contributed by the thermal treatment. Pore-free polymers, such as "gel" type resins which have been subjected to thermal treatment in the prior art do not contribute the large pores essential to the adsorbents useful in the process of the invention nor do they perform with the efficiency of the pyrolyzed polymers described herein.

The duration of pyrolysis depends upon the time needed to remove the volatiles from the particular polymer and the heat transfer characteristics of the method selected. In general, the pyrolysis is very rapid when the heat transfer is rapid, e.g., in an oven where a shallow bed of material is pyrolyzed, or in a fluidized bed. To prevent burning of the pyrolyzed polymer, normally the temperature of the polymer is reduced to not more than 400°C, preferably not more than 300°C, before the pyrolyzed material is exposed to air. The most desirable method of operation involves rapid heating to the maximum temperature, holding the temperature at the maximum for a short period of time (in the order of 0 to 20 minutes) and thereafter quickly reducing the temperature to room temperature before exposing the sample to air. Suitable products for use in the process of the invention have been produced by this preferred method by heating to 800°C and cooling in a period of 20 to 30 minutes. Longer holding periods at the elevated temperatures are also satisfactory, since no additional decomposition appears to occur unless the temperature is increased.

Activating gases such as CO ₂ , NH ₃ , (¾, H ₂ 0 or combinations thereof in small amounts tend to react with the polymer during pyrolysis and thereby increase the surface area of the final material. Such gases are optional and may be used to obtain special characteristics of the adsorbents.

The starting polymers which may be used to produce the pyrolyzed resins useful in the process of the invention include macroreticular homopolymers or copolymers of one or more monoethylenically or polyethylenically unsaturated monomers or monomers which may be reacted by condensation to yield macroreticular polymers and copolymers. The macroreticular resins used as precursors in the formation of macroreticular heat treated polymers are not claimed as new compositions of matter in themselves. Any of the known materials of this type with an appropriate carbon-fixing moiety is suitable. The preferred monomers are those aliphatic and aromatic materials which are ethylenically unsaturated.

Examples of suitable monoethylenically unsaturated monomers that may be used in making the granular macroreticular resin include: esters of acrylic and methacrylic acid such as methyl, ethyl, 2-chloro ethyl, propyl, isobutyl, isopropyl, butyl, tert -butyl, sec-butyl, ethylhexyl, amyl, hexyl, octyl, decyl, dodecyl, cyclohexyl, isobornyl, benzyl, phenyl, alkylphenyl, ethoxymethyl, ethoxyethyl, ethoxypropyl, propoxymethyl, propoxyethyl, propoxypropyl, ethoxyphenyl, ethoxybenzyl, ethoxycyclohexul, hydroxyethyl, hydroxypropyl, ethylene, propylene, isobutylene, diisobutylene, styrene, ethylvinylbenzene, vinyltoluene,

vinylbenzylchloride, vinyl chloride, vinyl acetate, vinylidene chloride, dicyclopentadiene, acrylonitrile, methacrylonitrile, acrylamide, methacrylamide, diacetone acrylamide, functional monomers such as vinylbenzene, sulfonic acid, vinyl esters, including vinyl acetate, vinyl propionate, vinyl butyrate, vinyl laurate, vinyl ketones including vinyl methyl ketone, vinyl ethyl ketone, vinyl isopropyl ketone, vinyl n-butyl ketone, vinyl hexyl ketone, vinyl octyl ketone, methyl isopropenyl ketone, vinyl aldehydes including acrolein, methacrolein, crotonaldehyde, vinyl ethers including vinyl methyl ether, vinyl ethyl ether, vinyl propyl ether, vinyl isobutyl ether, vinylidene compounds including vinylidene chloride bromide, or bromochloride, also the corresponding neutral or half-acid half-esters or free diacids of the unsaturated dicarboxylic acids including itaconic, citraconic, aconitic, fumaric, and maleic acids, substituted acrylamides, such as N-monoalkyl, -Ν,Ν-dialkyl-, and N- dialkylaminoalkylacrylamides or methacrylamides where the alkyl groups may have from one to eighteen carbon atoms, such as methyl, ethyl, isopropyl, butyl, hexyl, cyclohexyl, octyl, dodecyl, hexadecyl and octadecyl aminoalkyl esters of acrylic or methacrylic acid, such as β- dimethylaminoethyl, β-diethylaminoethyl or 6-dimethylaminohexyl acrylates and

methacrylates, alkylthioethyl methacrylates and acrylates such as ethylthioethyl methacrylate, vinylpyridines, such as 2-vinylpyridine, 4-vinylpyridine, 2-methyl-5-vinylpyridine, and so on.

In the case of copolymers containing ethylthioethyl methacrylate, the products can be oxidized to, if desired, the corresponding sulfoxide or sulfone.

Polyethylenically unsaturated monomers which ordinarily act as though they have only one such unsaturated group, such as isoprene, butadiene, and chloroprene, may be used as part of the monoethylenically unsaturated category.

Examples of polyethylenically unsaturated compounds include: divinylbenzene, divinylpyridine, divinylnaphthalenes, diallyl phthalate, ethylene glycol diacrylate, ethylene glycol dimethacrylate, trimethylolpropanetrimethacrylate, divinylsulfone, polyvinyl or polyallyl ethers of glycol, of glycerol, of pentaerythritol, of diethyleneglycol, of monothio or dithio- derivatives of glycols, and of resorcinol, divinylketone, divinylsylfide, allyl acrylate, diallyl maleate, diallyl fumarate, diallyl succinate, diallyl carbonate, diallyl malonate, diallyl oxalate, diallyl adipate, diallyl sebacate, divinyl sebacate, diallyl tartrate, diallyl silicate, triallyl tricarballylate, triallyl aconitate, triallyl citrate, triallyl phosphate, N,N'-methylenediacrylamide, N,N'-methylenedimethacrylamide, Ν,Ν'-ethylenediacrylamide, trivinylbenzene,

trivinylnaphthalenes, and polyvinylanthracenes.

A preferred class of monomers of this type is aromatic ethylenically unsaturated molecules such as styrene, vinyl pyridine, vinyl naphthalene, vinyl toluene, phenyl acrylate, vinyl xylenes, and ethylvinylbenzene.

Examples of preferred polyethylenically unsaturated compounds include divinyl pyridine, divinyl naphthalene, divinylbenzene, trivinylbenzene, alkyldivinylbenzenes having from 1 to 4 alkyl groups of 1 to 2 carbon atoms substituted in the benzene nucleus, and alkyltrivinylbenzenes having 1 to 3 alkyl groups of 1 to 2 carbon atoms substituted in the benzene nucleus. Besides the homopolymers and copolymers of these poly( vinyl) benzene monomers, one or more of them may be copolymerized with up to 98% (by weight of the total monomer mixture) of (1) monoethylenically unsaturated monomers, or (2) polyethylenically unsaturated monomers other than the poly(vinyl)benzenes just defined, or (3) a mixture of (1) and (2). Examples of the alkyl-substituted di- and tri-vinyl-benzenes are the various vinyltoluenes, the divinylethylbenzene, 1,4-divinyl- 2,3,5,6-tetramethylbenzene, 1,3,5-trivinyl- 2,4,6-trimethylbenzene, 1,4-divinyl, 2,3,6-triethylbenzene, l,2,4-trivinyl-3,5-diethylbenzene, 1,3,5-tri vinyl -2-methylbenzene.

Most preferred are copolymers of styrene, divinylbenzene, and ethylvinylbenzene. Examples of suitable condensation monomers include: (a) aliphatic dibasic acids such as maleic acid, fumaric acid, itaconic acid, 1,1-cyclobutanedicarboxylic acid, etc.; (b) aliphatic diamines such as piperazine, 2-methylpiperazine, cis, cis-bis (4-aminocyclohexyl) methane, metaxylylenediamine, etc.; (c) glycols such as diethylene glycol, Methylene glycol, 1,2- butanediol, neopentyl glycol etc.; (d) bischloroformates such as cis and trans- 1 ,4-cyclohexyl bischloroformate, 2,2,2,4-tetramethyl-l,3-cyclobutyl bischloroformate and bischloroformates of other glycols mentioned above, etc. ; (e) hydroxy acids such as salicylic acid, m- and p-hydroxy- benzoic acid and lactones, derived therefrom such as the propiolactones, valerolactones, caprolactones, etc.; (f) diisocyanates such as cis and trans-cyclopropane- 1,2 -diisocyanate, cis and trans-cyclobutane- 1-2 -diisocyanate etc.; (g) aromatic diacids and their derivatives (the esters, anhydrides and acid chlorides) such as phthalic acid, phthalic anhydride, terephthalic acid, isophthalic acid, dimethylphthalate, etc.; (h) aromatic diamines such as benzidine, 4,4'- methylenediamine, bis(4-aminophenyl) ether, etc. ; (i) bisphenols such as bisphenol A, bisphenol C, bisphenol F, phenolphthalein, recorcinol, etc.; (j) bisphenol bis(chloroformates) such as bisphenol A bis(chloroformate), 4,4' -dihydroxybenzophenone bis(chloroformate) etc.; (k) carbonyl and thiocarbonyl compounds such as formaldehyde, acetaldehyde, thioacetone acetone, etc.; (1) phenol and derivatives such as phenol, alkylphenols, etc.; (m) polyfunctional cross-linking agents such as tri or poly basic acids such as trimellitic acid, tri or polyols such as glycerol, tri or polyamines such as diethylenetriamine; and other condensation monomers and mixtures of the foregoing.

Ion exchange resins produced from aromatic and/or aliphatic monomers provide a preferred class of starting polymers for production of porous adsorbents. The ion exchange resin may also contain a functional group selected from cation, anion, strong base, weak base, sulfonic acid, carboxylic acid, oxygen containing, halogen and mixtures of the same. Further, such ion exchange resins may optionally contain an oxidizing agent, a reactive substance, sulfuric acid, nitric acid, acrylic acid, or the like at least partially filling the macropores of the polymer before heat treatment.

The synthetic polymer may be impregnated with a filler such as carbon black, charcoal, bonechar, sawdust or other carbonaceous material prior to pyrolysis. Such fillers provide an economical source of carbon which may be added in amounts up to 90% by weight of the polymer.

The starting polymers, when ion exchange resins, may optionally contain a variety of metals in their atomically dispersed form at the ionic sites. These metals may include iron, copper, silver, nickel, manganese, palladium, cobalt, titanium, zirconium, sodium, potassium, calcium, zinc, cadmium, ruthenium, uranium and rare earths such as lanthanum. By utilizing the ion exchange mechanism it is possible for the skilled technician to control the amount of metal that is to be incorporated as well as the distribution.

Although the incorporation of metals onto the resins is primarily to aid their ability to serve as catalytic agents, useful adsorbents may also contain metal.

Synthetic polymers, ion exchange resins whether in the acid, base or metal salt form are commercially available. According to the invention there is also provided an adsorption process for separating components from a gaseous or liquid medium which comprises contacting the medium with particles of a pyrolyzed synthetic polymer. For example it has been discovered that a styrenedivinylbenzene based strongly acidic exchange resin pyrolyzed from any of the forms of Hydrogen, Iron (III), Copper (II), Silver (I) or Calcium (II) can decrease the concentration of vinylchloride in air preferably dry air from initial concentration of 2 ppm to 300,000 ppm to a level of less than 1 ppm at flow rates of 1 bedvolume/hour to 600 bedvolume/min. preferably 10 to 200 bedvolume/minute.

The partially pyrolyzed macroporous polymer adsorbent useful in process of the present invention disclosed herein above are able to adsorb greater than 10 cm ³ STP of ethane per gram of sorbent at 35°C and 200 mmHg of ethane and greater than 20 cm ³ STP of propane per gram of sorbent at 35°C and 100 mmHg of propane.

In the process of the present invention, the adsorption of hydrocarbons by the adsorbing media is a reversible process. The practice of removing volatiles from a loaded adsorption media can be accomplished by any suitable means, typically by reducing the pressure over the media, heating, or the combination of reduced pressure and heating. In either case the desired outcome is to re-volatilize the trapped vapors, and subsequently remove them from the adsorbent media so that it can be reused to capture additional volatiles.

Periodic regeneration preferably takes place while a second set of the first and second adsorbents is used to continue the purification process, each set of the two adsorbents being on-line in the purification process and being regenerated in alternation.

In one embodiment, the process further comprises reducing the gas pressure over the adsorbent prior to the purging step and then restoring the pressure prior to or at the commencement of repeating the purification step.

Preferably, the regeneration of the first and second adsorbents (e.g., in vessels 20 and 40) comprises passing heated regenerating gas (purge gas) counter currently through the second and first adsorbents for a period from 10 seconds to 100 min, preferably 1 min to 20 min, more preferably 2 min to 10 min.

The temperature of the feed gas is from 0°C to 80°C. Preferably, the temperature of the feed gas is equal to or greater than 0°C, more preferably equal to or greater than 10°C, even more preferably equal to or greater than 20°C. Preferably, the temperature of the feed gas is equal to or less than 80°C, more preferably equal to or less than 60°C, even more preferably equal to or less than 40°C.

The pressure of the feed gas is typically between 15 psia and 1,000 psia. Preferably, the pressure of the feed gas is equal to or greater than 15 psia, more preferably equal to or greater than 17 psia. Preferably, the pressure of the feed gas is equal to or less than 1,000 psia, more preferably equal to or less than 500 psia. The temperature of the heated regenerating gas is from 30°C to 100°C. Preferably, the temperature of the heated regenerating gas is equal to or greater than 30°C, more preferably equal to or greater than 35°C, even more preferably equal to or greater than 40°C. Preferably, the temperature of the heated regenerating gas is equal to or less than 100°C, more preferably equal to or less than 80°C, even more preferably equal to or less than 70°C.

The temperature of the regenerating gas is 1°C to 80°C greater than the temperature of the feed gas. Preferably, the temperature of the regenerating gas is equal to or greater than 1°C, more preferably equal to or greater than 5°C, even more preferably equal to or greater than 10°C greater than the temperature of the feed gas. Preferably, the temperature of the regenerating gas is equal to or less than 100°C, more preferably equal to or less than 80°C, even more preferably equal to or less than 60°C greater than the temperature of the feed gas.

The pressure of the regenerating gas is typically between 1 psia and 45 psia. The pressure of the regenerating gas is preferably equal to or greater than 1 psia, preferably equal to or greater than 7 psia, more preferably equal to or greater than 15 psia. The pressure of the regenerating gas is preferably equal to or less than 45 psia, preferably equal to or less than 35 psia, more preferably equal to or less than 25 psia.

EXAMPLES

Results for Example 1 are created using a rate based process simulator developed using laboratory thermodynamic and mass transfer data. Equilibrium and rate data are entered into computer software and regressed to obtain values that correspond to parameters in the applicable mathematical equations which enable the software to calculate performance of the adsorption system. The simulator performs rigorous momentum, heat and mass balances at various points throughout the adsorbent bed.

For Example 1 a feed gas comprising 95 percent methane and 5 percent butane having a pressure of 350 psig and a temperature of 100°F is passed through a porous cross- linked polymeric adsorbent media having a high surface area equal to or greater than 1 ,000 m ² /g made from a macroporous copolymer of a monovinyl aromatic monomer and a crosslinking monomer, where the macroporous copolymer has been post-crosslinked in the swollen state in the presence of a Friedel-Crafts catalyst. FIG. 3 shows the working capacity of butane versus regeneration temperature at different regeneration pressures for Example 1. In this simulation the adsorption step proceeded for 15 minutes at a flow rate of 5000 mol/hr in a vessel 18 inch diameter by 13.5 feet tall. Adsorbtion was followed by depressurization and then regeneration with a 500 mol/hr purge rate for 15 minutes according to the flow scheme shown in Figure 2.

Afterwards the bed was repressurized and the cycle repeated. The cycle was repeated in the software over 20 times to approach a cyclic steady state behavior. The effect on working capacity with varying temperatures and regeneration pressures during the purge step is demonstrated. Purge temperature is important since working capacity affects system sizing and cost. The difference in butane capacity at the end of the adsorption step and the butane capacity at the end of the regeneration step is the working capacity of butane.

In Example 2, the equilibrium adsorption values for methane and butane for the adsorbent used in Example 1 are determined at three partial pressures and two temperatures which show that the adsorbent preferentially adsorbs butane over methane and that the addition of heat will enhance regeneration. The equilibrium adsorption values are shown in Table 1.

Table 1