Dehydration processes utilizing cationic zeolite RHO

Abstract

Disclosed are processes and systems for the removal of water from a feed stream utilizing swing adsorption processes including an adsorbent bed comprising an adsorbent material which is a cationic zeolite RHO. The cationic zeolite RHO comprises at least one, preferably two, metal cations selected from Group 1 and 2 elements (new Group 1-18 IUPAC numbering). The swing adsorption processes and systems utilizing the cationic zeolite RHO have an adsorption selectivity for water and are useful in selective dehydration of commercial feed streams. The cationic zeolite RHO additionally has an exceptionally high water adsorption stability for use in feed streams with wet acid gas environments operating under cyclic swing adsorption conditions.

Claims

1. A process for removing water from a feed stream, the process comprising performing a swing adsorption process by: a) performing an adsorption step, wherein the adsorption step comprises passing a feed stream comprising water through an adsorbent bed unit comprising at least one adsorbent bed wherein water is selectively separated from the feed stream to form a product stream which has a lower molar fraction of water than the feed stream, wherein the adsorbent bed comprises an adsorbent material which is a cationic zeolite RHO comprising at least one metal cation; b) interrupting the flow of the feed stream; c) performing a desorption step, wherein the desorption step comprises removing at least a portion of the water from the adsorbent material; and d) repeating the steps a) to c) for at least one additional cycle.

2. The process of claim 1, wherein the cationic zeolite RHO is a RHO framework zeolite comprising at least one metal cation selected from the Group 1 and Group 2 elements (new Group 1-18 IUPAC numbering), and the ratio of average atomic metal cation charge to the atomic Al plus Ga content in the crystal as measured by either XRF or AA/ICP is from 0.7 (atomic charges per atom) to 1.5 (atomic charges per atom).

3. The process of claim 2, wherein the cationic zeolite RHO comprises an aluminosilicate, a gallosilicate, or a combination thereof.

4. The process of claim 1, wherein the at least one metal cation is selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

5. The process of claim 4, wherein the at least one metal cation is Cs.

6. The process of claim 1, wherein the cationic zeolite RHO comprises at least two metal cations selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

7. The process of claim 6, wherein one of the at least two metal cations is Cs.

8. The process of claim 1, wherein the cationic zeolite RHO comprises at least two metal cations selected from NaCs, KCs, and LiCs.

9. The process of claim 1, wherein the cationic zeolite RHO further comprises a non-metallic cation selected from H.sup.+, NH4.sup.+, and a combination thereof.

10. The process of claim 9, wherein the cationic zeolite RHO comprises an aluminosilicate and the cationic zeolite RHO has a ratio of the average non-metallic cation charge to the atomic Al plus Ga content is less than 0.3 (atomic charges per atom).

11. The process of claim 1, wherein the feed stream is air.

12. The process of claim 11, wherein the cationic zeolite RHO is NaCs—RHO.

13. The process of claim 1, wherein the feed stream comprises hydrocarbons.

14. The process of claim 13, wherein the feed stream further comprises an acid gas.

15. The process of claim 14, wherein the cationic zeolite RHO is NaCs—RHO.

16. The process of claim 15, wherein the water content of the feed stream is at least about 100 parts per million (ppm) mole fraction and the water content of the product stream is less than about 10 ppm mole fraction.

17. The process of claim 16, wherein the product stream comprises at least 90 wt % of the hydrocarbons present in the feed stream.

18. The process of claim 17, wherein the feed stream is natural gas.

19. A swing adsorption system for removing water from a feed stream, the system comprising: a) at least one adsorbent contactor containing an adsorbent bed; b) a feed stream inlet fluidly connected to the at least one adsorbent contactor; and c) a product stream outlet fluidly connected to the at least one adsorbent contactor; wherein the adsorbent bed comprises an adsorbent material which is a cationic zeolite RHO comprising at least one metal cation.

20. The system of claim 19, wherein the adsorbent bed is a parallel channel contactor.

21. The system of claim 19, wherein the mass of cationic zeolite RHO in the adsorbent bed is more than 20 wt % of the total materials making up the adsorbent bed.

22. The system of claim 19, wherein the swing adsorption system is configured to perform steps comprising: a) perform an adsorption step, wherein the adsorption step comprises passing a feed stream comprising water through the adsorbent bed to form a product stream; b) interrupting the flow of the feed stream; c) performing a desorption step, wherein the desorption step comprises removing at least a portion of the water from the adsorbent material; and d) repeating the steps a) to c) for at least one additional cycle.

23. The system of claim 19, wherein the cationic zeolite RHO is a RHO framework zeolite comprising at least one metal cation selected from the Group 1 and Group 2 elements (new Group 1-18 IUPAC numbering), and the ratio of average atomic metal cation charge to the atomic Al plus Ga content in the crystal as measured by either XRF or AA/ICP is from 0.7 (atomic charges per atom) to 1.5 (atomic charges per atom), and the cationic zeolite RHO comprises an aluminosilicate, a gallosilicate, or a combination thereof.

24. The system of claim 23, wherein the at least one metal cation is selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

25. The system of claim 24, wherein the at least one metal cation is Cs.

26. The system of claim 23, wherein the cationic zeolite RHO comprises at least two metal cations selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

27. The system of claim 26, wherein one of the at least two metal cations is Cs.

28. The system of claim 23, wherein the cationic zeolite RHO comprises at least two metal cations selected from NaCs, KCs, and LiCs.

29. The system of claim 23, wherein the cationic zeolite RHO further comprises a non-metallic cation selected from H.sup.+, NH4.sup.+, and a combination thereof.

30. The system of claim 29, wherein the cationic zeolite RHO comprises an aluminosilicate and the cationic zeolite RHO has a ratio of the average non-metallic cation charge to the atomic Al plus Ga content is less than 0.3 (atomic charges per atom).

Description

BRIEF DESCRIPTION OF THE FIGURES

(1) FIGS. 1A and 1B are field emission scanning electron microscopy (SEM) images of synthesized NaCs—RHO zeolite crystals shown at two (2) different magnification levels as noted.

(2) FIG. 2 is a schematic of a testing apparatus and arrangement used in the stability screening tests.

(3) FIG. 3 is a graph showing initial water uptake at 35° C. after sample regenerated at 400° C. for a fresh NaCs—RHO sample, the partial pressure of water is ˜0.0042 bar.

(4) FIG. 4 is a graph showing a NaCs—RHO sample weight and the temperature changes as a function of time for 100 adsorption/desorption temperature cycles under wet CO.sub.2 atmosphere conditions.

(5) FIG. 5 is a graph showing the water uptake at 35° C. after sample regenerated at 400° C. for the NaCs—RHO sample after the 100 adsorption/desorption temperature cycles shown in FIG. 4. The partial pressure of water is ˜0.0042 bar.

(6) FIG. 6 is a graph showing the results measuring H.sub.2O capacity vs. the number of adsorption/desorption temperature swing cycles under wet CO.sub.2 atmosphere conditions of comparative testing of a Zeolite 3A and a NaCs—RHO zeolite.

(7) FIG. 7 is a graph showing the results measuring H.sub.2O capacity loss for three (3) LTA zeolites (3A, 4A, and 5A) after 60 adsorption/desorption cycles under the same adsorption cyclic operating conditions and wet CO.sub.2 atmosphere conditions except desorption with wet CO.sub.2 instead of dry CO.sub.2 in previous cycle testing. Wet CO.sub.2 desorption conditions accelerated degradation as compared to dry CO.sub.2 desorption condition.

(8) FIG. 8 is a diagram of the water breakthrough on a NaCs—RHO zeolite packed adsorbent bed.

(9) FIG. 9 is a graph showing the test results showing the water capacity loss for a Zeolite 3A (LTA) subjected to 3000 adsorption/desorption temperature swing cycles under wet CO.sub.2 atmosphere conditions.

(10) FIG. 10 is a graph showing the test results showing the water capacity loss for a NaCs—RHO zeolite subjected to 3000 adsorption/desorption temperature swing cycles under wet CO.sub.2 atmosphere conditions.

(11) FIG. 11 is a graph showing the water capacity as a function of temperature and pressures for NaCs—RHO zeolite.

(12) FIG. 12 is a graph showing the comparison of working capacity for pressure swing cycles between 3A and NaCs—RHO.

(13) FIG. 13 is a graph showing the comparison of water capacity at high temperature for regeneration to clean the adsorbate between 3A and NaCs—RHO.

(14) FIG. 14 is a graph showing the results comparing H.sub.2O capacity vs. the number of adsorption/desorption temperature swing cycles under wet CO.sub.2 atmosphere conditions of LiCs—RHO, CaNH.sub.4—RHO, NH.sub.4—RHO, and NaCs—RHO zeolite.

(15) FIG. 15 is a PXRD pattern for a synthesized NaCs—RHO disclosed herein.

DETAILED DESCRIPTION

(16) This invention includes a cationic zeolite RHO and its use in swing adsorption processes for dehydration of process feedstreams, particularly feedstreams comprising water and CO.sub.2. In particular, is disclosed a cationic form of zeolite RHO utilized in pressure swing adsorption processes (PSA), temperature swing adsorption processes (TSA), partial pressure swing adsorption processes (PPSA), rapid temperature swing adsorption (RTSA), rapid cycle pressure swing absorption (RCPSA), rapid cycle partial pressure swing adsorption (RCPPSA), or a combination thereof, which may be collectively referred to herein as “swing adsorption processes” unless further defined. The term “rapid cycle swing adsorption processes” refer to rapid temperature swing adsorption (RTSA), rapid cycle pressure absorption (RCPSA), rapid cycle partial pressure swing adsorption (RCPPSA), or a combination thereof. The term “rapid cycle swing adsorption processes” will include such processes as mentioned wherein the total cycle time or period for the rapid cycle swing adsorption processes to go through a full cycle, such as feed/product step(s), desorption step(s), purge step(s) and repressurization step(s) and back to the next feed/product step(s), is a period greater than 1 second and less than 600 seconds. In preferred embodiments, total cycle time or period for the rapid cycle swing adsorption processes is greater than 2 seconds and less than 300 seconds.

(17) These swing adsorption processes may be used to separate gases of a gas mixture because different gases tend to fill the micropore of the adsorbent material to different extents Conventional PSA, TSA, or such techniques are generally operated with cycle times (particularly adsorption steps or cycles) of sufficient duration to allow the adsorption of the components to come to near equilibrium conditions (i.e., allowing the adsorbent to selectively adsorb the amount of one component relative to another simply by the inherent equilibrium selectivity of the adsorbent at adsorption conditions). In most rapid cycle adsorption processes the kinetics of the adsorbent are fast enough so that some components of the feed equilibrate with the adsorbent material. However, in many “rapid cycle” swing adsorption processes, the adsorption cycle is operated at a short enough duration so that some components of the feed do not equilibrate with the adsorbent material (i.e. are predominantly excluded from the adsorbent). As such, all components in a multicomponent feed stream do not reach equilibrium loadings in the adsorbent and the competitive separation does not come to equilibrium conditions, but instead operates in a “kinetic” separation regime. In this mode, cycle times and process conditions (particularly in the adsorption step) are designed to take advantage of the relative kinetic adsorption rates (or “diffusivity” rates) between various components (i.e., some components may adsorb “faster” than other components) to facilitate and improve the separation efficiency of the rapid cycle processes. In some other rapid cycle processes the kinetics of the transport do not provide discrimination between different species and selectivity is achieved from the competitive equilibrium adsorption isotherm which is related to the relative adsorption strength of different molecules. For an example related to the dehydration processes herein it is preferred that the kinetic adsorption rate of water be fast enough to reach equilibrium water loadings in the adsorbent within the time allotted for the adsorption step. It is even more preferred that the kinetics of the water adsorption process be fast enough to reach equilibrium water loadings in the adsorbent within one-fifth of the time allotted for the adsorption step. It is also preferred that the equilibrium selectivity for water in the adsorbent be greater than for any of the other components in the feed stream. In a more preferred embodiment the kinetic selectivity for water uptake is faster than for other components in the feed stream. In a most preferred embodiment, the kinetic uptake rate of the other components in the feed stream does not allow them to reach equilibrium loadings in the time of the adsorption step. If a gas mixture, such as natural gas containing water (or water vapor), is passed under pressure through a vessel containing an adsorbent material that is more selective towards water vapor than it is for methane, at least a portion of the water vapor is selectively adsorbed by the adsorbent material, and the gas exiting the vessel is enriched in methane. As such, the adsorbent would be considered to have a “selectivity” (or “greater selectivity”) for water over methane which can come from either equilibrium loading (competitive adsorption), kinetics (relative adsorption rates) or combinations of these effects. Before the adsorbent material reaches the end of its capacity to adsorb water vapor it is switched from an adsorption step (or “cycle”) to a desorption step. Desorption can be accomplished by raising the temperature of the adsorbent (TSA), purging the adsorbent with a dry stream (PPSA), reducing the pressure of the adsorbent (PSA) or by combinations of these methods. Once the adsorbent has gone through a desorption step it is ready for another adsorption step. Other additional steps such as depressurization, purging, repressurization, reheating, or cooling may alternatively be included in the overall process steps. The combination of the overall steps from the beginning of one adsorption step to the next adsorption step may be referred to as the “total cycle” or the “swing adsorption process cycle”. Such cycles would also apply in the case of both conventional swing adsorption processes and rapid cycle swing adsorption processes.

(18) While not limiting, the swing adsorption processes herein preferably further include the use of an adsorbent comprising a cationic zeolite RHO, wherein the feed stream is comprised of water, and optionally, other components such as hydrocarbons, CO.sub.2, nitrogen (N.sub.2), and/or hydrogen sulfide (H.sub.2S); and further wherein at least a portion of the water is preferentially removed from the feed stream of the swing adsorption processes wherein the swing adsorption processes produces a product stream wherein the term preferentially removal of water (or the like) means that the wt % of water in the product stream (based on the total product stream) is less than the wt % of water in the feed stream (based on the total feed stream). This is equivalent to the statement that the mole % of water in the product stream (based on the total product stream) is less than the mole % of water in the feed stream (based on the total feed stream). Embodiments of the invention are particularly applicable to swing adsorption processes that rigorously dehydrate the feed stream. Rigorous dehydration is achieved when the product stream from the swing adsorption process contains less than 10 ppm (mole fraction) of water, preferable less than 1 ppm (mole fraction) of water, and even more preferably less than 0.1 ppm (mole fraction) of water. Embodiments of the invention may also be utilized for the removal of water from such feed streams may be required to meet specifications and process requirements for such things as pipeline specifications, cryogenic applications, dehydration for air separation processes including nitrogen purification/production as well as O.sub.2/Ar separation, and miscellaneous intermediate process steps, particularly in the oil and gas industry.

(19) Zeolite LTA has been a favored zeolite for use in dehydration and rigorous dehydration processes due to its high H.sub.2O uptake capacity. However, it has been discovered herein that while zeolite LTA has a high capacity for water as performed in slow cycle testing, that LTA, when repeatedly thermally cycled under wet CO.sub.2 conditions (i.e., a stream comprising water and CO.sub.2), that the water adsorption capacity can quickly and significantly deteriorate. While water adsorption capacity is an important consideration in the selection of an adsorbent material in a swing adsorption process, significant deterioration of the adsorbent capacity cannot be tolerated as, to be used in an economically justifiable commercial application, particularly in rapid swing adsorption processes, the adsorbent materials must be able to be operated in processes comprising thousands to hundreds of thousands of cycles between adsorbent change-outs.

(20) The invention discloses the use of a cationic zeolite RHO as dehydration adsorbent that possesses significant stability for feed gas (or “feed streams”) which comprise acidic gas such as CO.sub.2. It has been discovered, that when subjected to numerous swing adsorption cycles, that common LTA molecular sieve adsorbents such as 3A, 4A and 5A (or “zeolite A”) show continuous degradation particularly during the temperature swing cycles for removing H.sub.2O especially with combinations of CO.sub.2 present in the feed stream. The cationic zeolite RHO shows much better stability and much less degradation for dehydration applications with feed containing wet CO.sub.2 (also known as an “acid gas” or “wet acidic gas”). This is surprising because the structure of zeolite RHO is related to Linde type A zeolites such as 3A, 4A and 5A [Robson, H. E., Shoemaker, D. P., Ogilvie, R. A. and Manor, P. C., Adv. Chem. Ser., 121, 106-115 (1973)].

(21) All of the descriptions of zeolite RHO refer to the IZA structure code of RHO which will include the other structure types such as the berrylophosphate, zinc berrylophosphates, LZ-214 or other UOP equivalents, cobalt phosphates, manganese aluminophosphates, magnesium aluminophosphates, and minerals such as pahasapaite. In addition to the aluminosilicate version of the zeolite, the gallosilicate version, ECR-10, may be utilized in preferred embodiments of the cationic zeolite RHO herein. Forms of these with cations to balance charge in the framework are preferred. In particular the aluminosilicate and gallosilicate versions of the cationic zeolite RHO are preferred and most preferred are the aluminosilicate versions of the cationic zeolite RHO.

(22) In many syntheses of the aluminosilicate versions of zeolite RHO that use inorganic cations, the crystallization can result in other impurities which form as a function of the gel composition or the kinetics of the reaction. Zeolite RHO with impurity phases from syntheses with inorganic cations or other synthesis routes (including those that make the gallosilicate version) can still be useful. Impurity phases may be present from 0 to 50% (mass), preferably, 0 to 25% (mass), more preferably, 0 to 10% (mass), and even more preferably, 0 to 2% (mass) as measured by XRD, SEM, rotation electron diffraction, or electron diffraction. Examples of impurity phases for the aluminosilicate versions of zeolite RHO include faujasite, chabazite, pollucite, zeolite P, gismodine, amorphous components, and dense phases, i.e. structures with the largest window containing 6 T-atoms or lower, such as Cs or Na non-porous aluminosilicates, quartz, analcite, sodalite, and tridymite.

(23) The zeolite RHO to be used in the current application requires that the charge in the framework, generated from insertion of tetrahedral aluminum or gallium species into a tetrahedrally coordinated silica network, be balanced by monovalent, divalent, or trivalent cations. Preferably these cations are an alkali cation, alkaline earth cation, rare earth cation, or mixtures thereof which will be referred to as metal cations. Nonmetal cations are taken to be H+ and NH4.sup.+. For RHO zeolites useable in the current application, the ratio of average atomic “metal” cation charge to the atomic Al plus Ga content (where Al is required, but Ga is optional, i.e., Al content is greater than 0 and Ga content can be 0 or greater than 0) in the crystal as measured by XRF (X-ray Fluorescence) or AA/ICP (Atomic Adsorption/Inductively Coupled Plasma) must be equal to 0.7 (atomic charges per atom) or greater, preferably 0.9 (atomic charges per atom) or greater and even more preferably 0.95 (atomic charges per atom) or greater. It is also preferred that there not be a great excess of “metal cations” in the zeolite RHO materials. As such the ratio of average atomic “metal” cation charge to the atomic Al plus Ga content in the crystal as measured by XRF or AA/ICP must be less than 1.5 (atomic charges per atom), preferably less than 1.1 (atomic charges per atom) and even more preferably less than 1.01 (atomic charges per atom). In these systems the value of this ratio should always be close to 1.0 unless the materials are not washed well. Monovalent metal cations have one atomic charge per cation and examples include Li, Na, K, Rb, and Cs. Divalent metal cations have two atomic charges per cation and examples include Mg, Ca, Sr, and Ba. The amount of nonmetallic cations (H.sup.+ or NH4.sup.+) balancing the charge should be minimized. It is preferred that the ratio of the average nonmetallic cation charge to the atomic Al plus Ga content be less than 0.3 (atomic charges per atom), preferably less than 0.1 (atomic charges per atom) and even more preferably less than 0.05 (atomic charges per atom). For aluminosilicate versions of RHO, a criteria for having a low amount of nonmetallic cations can be stated as a having nonmetallic ion content of 0.2 or less of the exchange capacity of the material (exchange capacity is defined as the total moles of all cations/moles of aluminum). In a more preferred embodiment aluminosilicate versions of RHO have nonmetallic ion content of 0.1 or less of the exchange capacity of the material, and in an even more preferred embodiment aluminosilicate versions of RHO have nonmetallic ion content of 0.05 or less of the exchange capacity of the material. It should be noted the H.sup.+ nonmetallic cations could be generated from protons transported into the crystal from exchange solutions, low pH water washes or exchange solutions where the pH <7.0, or the use of NH.sub.4 ion exchange followed by the removal of NH.sub.3 upon heat treatment or desorption in the process.

(24) In the synthesis of zeolite RHO using exclusively inorganic cations to form the structure, i.e. Na, Cs, the bulk composition of resulting zeolite RHO product ranges from SiO.sub.2/Al.sub.2O.sub.3 ratios of 4 to 14 (Si/Al of 2 to 14), with the more preferred ranges being from 5 to 7 (Si/Al of 2.5 to 3.5). Upon the use of other structure directing agents, i.e. quaternary ammonium containing molecules (mono, di, or tri quaternary), the SiO.sub.2/Al.sub.2O.sub.3 ratio is larger ranging from infinite to 40. In the case of the SDA (structure directing agent) containing syntheses, the SDA and cations will balance the charge in the structure. Upon decomposition of the SDA, a proton will balance the charge in the framework which may exceed the values above. Upon the exchange with monovalent, divalent, or trivalent cations the above exchange capacity limitations could be reached.

(25) Individual zeolite RHO crystal sizes that provide fast kinetic for water transport range from 0.05 microns to 40 microns, preferably from 0.1 microns to 15 microns and even more preferably from 0.2 microns to 5 microns. As-synthesized crystals can be intergrown or clumped and the size specification denoted herein applies to individual crystallites. The kinetic water transport rate is set by the crystal size and the diffusion constant for water.

(26) As utilized herein, the term “cationic zeolite RHO” (or equivalently “cationic RHO zeolite”) as is defined in its broadest terms is a RHO framework zeolite with at least one metal cation selected from the Group 1 and Group 2 elements (new Group 1-18 IUPAC numbering), with a ratio of average atomic metal cation charge to the atomic Al plus Ga content in the crystal as measured by XRF or AA/ICP is from 0.7 (atomic charges per atom) to 1.5 (atomic charges per atom). As an example, if the 1 Cs.sup.+ cation per aluminum atom with no Ga atom, the average atomic metal cation charge is 1. Alternatively, if there is 0.5 Ca.sup.2+ cation per aluminum atom with no Ga atom, then the average atomic metal cation charge is 1. In preferred embodiments, the cationic zeolite RHO comprises at least one metal cation selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba. In a preferred embodiment, the at least one metal cation is selected from Li, Na, K, and Cs, and more preferably, the at least one metal cation is Cs. In other preferred embodiments, the cationic zeolite RHO comprises at least two metal cations selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba. In a preferred embodiment, one of the at least two metal cations is Cs. In another preferred embodiment, the cationic zeolite RHO comprises at least two metal cations selected from the combinations of NaCs, KCs, and LiC.

(27) In particular, the cationic zeolite RHO can be useful in removing water from feed streams (dehydration) wherein the water content is at least about 10 ppm (by volume or mole fraction), preferably at least about 100 ppm (by volume or mole fraction), and even more preferably, and even more preferably at least 250 ppm (by volume or mole fraction). As previously mentioned, in a preferred embodiment the swing adsorption process in which the zeolite RHO is utilized to rigorously dehydrate the feed stream producing a product stream that contains less than 10 ppm (mole fraction) of water, more preferable less than 1 ppm (mole fraction) of water, and even more preferably less than 0.1 ppm (mole fraction) of water. In another preferred embodiment the feed stream contains CO.sub.2. In this embodiment, the CO.sub.2 content of the feed stream is at least about is at least about 50 ppm (by volume or mole fraction), more preferably the CO.sub.2 content of the feed stream is at least about is at least about 100 ppm (by volume or mole fraction) of CO.sub.2, and even more preferably the CO.sub.2 content of the feed stream is at least about is at least about 2500 ppm (by volume or mole fraction). In some embodiments, the feed stream may also comprise hydrocarbons, and the method herein are utilized to dehydrate the hydrocarbon containing feed stream. In particular embodiments, the hydrocarbon may be natural gas. The feed stream may also be comprised of methane, ethane or a combination thereof. In particular embodiments, the feed stream may contain at least 50 wt % hydrocarbons, or more preferably, at least 90 wt % hydrocarbons. It has been discovered that the selectivity of zeolite RHO is such that the swing adsorption dehydration process can be conducted using this zeolite to remove water from a hydrocarbon containing feed stream in a manner such that at least 90 wt % of the hydrocarbons present in the feed stream remain present in the product gas from the swing adsorption process, more preferably at least 95 wt % of the hydrocarbons present in the feed stream remain present in the product gas from the swing adsorption process, and even more preferably, at least 98 wt % of the hydrocarbons present in the feed stream remain present in the product gas from the swing adsorption process. This can be achieved while dehydrating the hydrocarbon containing feed stream to less than about 10 ppm mole fraction, whether or not acid gas conditions are present in the feed stream. It also been found that the cationic zeolite RHO possesses very rapid water kinetics which make it particularly advantageous for use as an adsorbent in rapid cycle swing dehydration processes.

(28) In performing swing adsorption processes the zeolite RHO crystals are incorporated into a contactor that can be in the form of a structured contactor or unstructured (pelletized) contactor. Several patents teach ways of incorporating zeolite crystals into contactors usable in a swing adsorption processes. When the zeolite is formulated into a contactor for a swing adsorption process, it may be bound together or held together in a coating with inorganic oxides, metals, other zeolites, other microporous materials such as MOFs, carbons, or polymers. In some instances the crystals are bound into the form of a pellet. In other instances the crystals may be coated onto the surface of a monolith with the aid of a binding agent. In other instances the crystals are grown on the surface of a monolith. In other instances the crystals are extruded with a binding agents to form a monolithic structure. In preferred embodiments the mass of cationic zeolite RHO in the adsorbent bed of the contactor or contactors used in the swing adsorption dehydration process is more than 2 wt %, preferably more than 20 wt % and even more preferably greater than 50 wt % of the total adsorbent bed materials.

(29) It has been found (as shown by the testing in the examples herein) that molecular sieve zeolite A which is commonly used for swing adsorption processes for dehydration degrades very rapidly under wet acidic atmospheres (i.e., a feedstream containing with CO.sub.2 and water), especially in temperature swing adsorption (TSA) cycles, including rapid cycle temperature swing adsorption (RCTSA) cycles. The discoveries and disclosures herein provide for an alternative dehydration adsorbent (a cationic zeolite RHO) with superior hydrothermal stability for wet CO.sub.2 feed as compared to zeolite A.

Preparation #1 of NaCs—RHO Zeolite

(30) A zeolite NaCs-Rho was synthesized by the following preparation technique. 39.5 g of NaOH was dissolved in 25 g of water. 43.1 g of 50% CsOH was added with stirring to the NaOH solution resulting in a clear solution. 25.5 g of C-31 Alcoa (Al(OH).sub.3, aluminum trihydrate) was added to the Na—Cs solution. This mixture was brought to a mild boil with stirring to dissolve all of the Al(OH).sub.3. A clear Na—Cs—Al solution resulted which was then cooled to room temperature. 272.6 g of Ludox, HS-40 (colloidal silica) was mixed with 25 g of water in a Teflon bottle. The Na—Cs—Al solution was slowly added to the colloidal silica solution with stirring. The final mixture was brought up to 500 g with water and stirring. This final mixture was white and pasty. The mixture was aged at approximately ambient conditions for 4 days. The resulting thick mixture was shaken then placed in an 85° C. oven for 4 days under static conditions. The sample was filtered, washed then dried at 115° C.

(31) A powder x-ray analysis (PXRD) was taken of the white powder using a D8 Bruker Endeavor analyzer and the PXRD pattern was analyzed using Jade 9 commercial software. The PXRD pattern matched that of NaCs—RHO. The PXRD pattern for the synthesized NaCs—RHO is shown in FIG. 15. The peak intensities resulting from the PXRD are shown in Table 1 below.

(32) TABLE-US-00001 TABLE 1 Peak PXRD intensities for NaCs-RHO 2-Theta d(Å) Height (CPS) Height % 8.548 10.2929 321.8 100 14.87 5.9527 141.1 43.8 21.074 4.2122 19.1 5.9 22.772 3.9018 49.7 15.4 25.864 3.442 184.6 57.4 27.282 3.2662 182.7 56.8 28.64 3.1143 102.2 31.8 29.936 2.9824 94.6 29.4 31.191 2.8652 53.4 16.6 33.566 2.6677 69.7 21.7 37.931 2.3701 18.3 5.7 38.949 2.3105 18.9 5.9 40.925 2.2034 16.2 5 43.76 2.067 19.4 6 49.041 1.8561 13.5 4.2

(33) A compositional analysis confirmed the following composition of the zeolite crystals: Al: 8.00 wt %, Cs: 24.90 wt %, Na: 4.03 wt %, Si: 24.70 wt %. Si/Al: 2.97, Na/Cs: 0.94, (Na+Cs)/Al: 1.22. SEM images using a Hitachi S-4500 field emission scanning electron microscope (FE-SEM) indicated that the crystals size ranged from 0.3 to 0.45 μm. and the images of the crystals at different magnification levels (magnification scales as shown in figures) are shown in FIGS. 1A & 1B, respectively. In other preferred embodiments, the crystal size of the cationic RHO zeolite may be from about 0.3 to about 5.0 μm, or about 0.3 to about 3.0 μm, or about 0.3 to about 1.0 μm. Other NaCs—RHO zeolite samples were synthesized using alternate alumina sources (instead of the Al(OH).sub.3 illustrated in this example) such as aluminum foil, Al (foil), and sodium aluminate, NaAlO.sub.2. NaCs—RHO was also fabricated using varying inductions times ranging from 0-7 days and temperatures ranging from 85-100° C. as well as in static systems and stirred systems up to 250 RPM. The stirred systems tended to produce somewhat larger NaCs—RHO crystals as compared to the static systems. When RHO seeds were added to a stirred reaction the morphology of the crystals changed. All of the samples utilizing these alternate synthesis methods showed good stability in dehydration studies.

EXAMPLE 1

Stability Screening

(34) The stability screening test was been carried out using a commercial thermal gravimetric analysis (TGA) instrument 201. A schematic of the testing apparatus and arrangement is shown in FIG. 2. A feed stream of carbon dioxide (CO.sub.2) 205, nitrogen (N.sub.2) 210, and water (H.sub.2O) 215 (via an entrained N.sub.2 stream 220), was controlled at a set flow rate via mass flow controllers MFC1, MCF2 and MCF3 as shown. A sample of the zeolite to be tested was placed in a sample holder 225 within the oven 230 wherein the temperature could be controlled. The feed stream mixture was passed through the oven 230 containing the sample holder 225 and then vented 235. Depending on whether the sample was being simulated in an adsorption or desorption step, the sample weight changed accordingly and was continuously monitored through the microbalance 235. The oven temperature was been controlled in the way to mimic temperature swing cycles, i.e., using a high temperature for desorption/regeneration and low temperature for adsorption. The oven had the capability to increase temperature to 400° C. in approximately 30 seconds, but took about 20 mins to cool down to 35° C. The sample was continuously exposed to various temperature cycles, and the performance of the sample was been evaluated by one point H.sub.2O uptake at 30° C., which has been carried out before and after cycle treatments with in-situ regeneration at 400° C. for 30 mins. By comparing the H.sub.2O uptake capacity measured at beginning and the final capacity after nth temperature swing cycles, the degradation of adsorbents was evaluated for the wet CO.sub.2 feed. The feed gas (or feedstream) was a wet CO.sub.2 stream formed by mixing CO.sub.2 with wet N.sub.2 stream (see FIG. 2). All the testing are carried by running water concentration (5000 ppm at 1 bar) with CO.sub.2 (0.7 bar CO.sub.2 balanced with N.sub.2) and temperature cycles from 35° C. to 400° C. These operating conditions were chosen to expedite the degradation process using severe conditions to evaluate stability in relatively short time run, instead of months or years operation time.

(35) The NaCs—RHO crystals were tested over multiple cycles under these wet CO.sub.2 conditions. The initial water uptake for the NaCs—RHO sample was measured before the cycle experiments. FIG. 3 shows the sample weight and the temperature changes as a function of time for the fresh NaCs—RHO sample. The sample was regenerated at 400° C. for 10 mins with 25 sccm (standard cm.sup.3/min) flow of nitrogen passing through, and then cooled down to 35° C. A 5 sccm flow of nitrogen was added through a water bubbler at room temperature to measure water uptake. Balance of flow in the TGA was 2 sccm with N.sub.2, so the partial pressure of water was about 0.0042 bar. The water uptake on the fresh NaCs—RHO was measured at 0.082 mg based on the 0.573 mg fresh sample weight. This amounted to a water uptake of 14.66% by weight of zeolite under these conditions.

(36) This NaCs—RHO sample was then run through 100 cycles under wet CO.sub.2 atmosphere with temperature cycling from 35 to 400° C., with a typical full adsorption/desorption cycle lasting approximately 30 minutes. During adsorption step, the sample was exposed to a mix of flow including 20 sccm CO.sub.2, 2 sccm dry N.sub.2 and 5 sccm N.sub.2 bubbling through a water bubbler to saturated at room temperature. During desorption step, the sample was exposed to mixed gases of 20 sccm CO.sub.2 and 2 sccm N.sub.2 at 400° C. FIG. 4 shows all temperature cycling and related weight change following temperature change. The temperature and sample weight of the first cycle and the last cycle has been marked in the plot. The uptake is 0.082 mg for wet CO.sub.2 in the first cycle, and it is same for the last cycle of 0.083 mg. This suggests no degradation of the NaCs—RHO sample under multiple wet CO.sub.2 atmosphere cycles. Furthermore, the water uptake was measured again at the end of the 100 cycles to give accurate evaluation of water capacity for the sample. The results are shown in FIG. 5. Here, it can be seen that the water uptake at the end of the 100 cycles was 14.61%, which was essentially the same as the initial water uptake 14.66% (i.e., within 1% deviation). Thus it was discovered that no degradation occurred for the NaCs—RHO after 100 cycles under wet CO.sub.2 atmosphere conditions.

(37) Surprised by the results from the initial testing of the NaCs—RHO zeolite under wet CO.sub.2 atmosphere conditions, this testing was continued for more cycles. This testing procedure provides a quick way to evaluate material stability at expedited conditions. By controlling the same operating conditions and testing procedures, commercial 3A, 4A and 5A LTA zeolite crystals were tested and evaluated as comparison to the NaCs—RHO under the same wet CO.sub.2 atmosphere conditions and cycles noted for the NaCs—RHO testing above. The results of this comparative testing for the NaCs—RHO and the Zeolite 3A are shown in FIG. 6. As can be seen, the Zeolite 3A has a higher initial water capacity than the NaCs—RHO zeolite. However, as can be seen in FIG. 7, the Zeolite 3A is not stable over continued adsorption/desorption cycles under the wet CO.sub.2 atmosphere conditions. In FIG. 6 it can be seen that the Zeolite 3A lost a significant amount of its initial water capacity over only 220 cycles. As can be seen in FIG. 6, the Zeolite 3A showed a clear trend to a near linear decrease in capacity as a function of the # of adsorption/desorption cycles. As such, the water adsorption of the LTA zeolites are expected to significantly deteriorate over a short life span of adsorption cycles conditions under wet CO.sub.2 cycle conditions.

(38) In contrast, as can be seen in FIG. 6, the NaCs—RHO zeolite sample was subjected to a total of 300 adsorption/desorption cycles under wet CO.sub.2 atmosphere conditions and showed no significantly measurable loss of water capacity when subjected to the same adsorption cycles and wet CO.sub.2 atmosphere conditions.

(39) Based on this data, is estimated that the NaCs—RHO will be able to operate at least more than 4 to 5 times longer than the LTA zeolites in a typical commercial rapid swing adsorption process under wet CO.sub.2 conditions. The degradation has correlation with cycle temperature and water content in feed condition. At higher regeneration temperatures and higher water loadings the samples show more degradation for the same number of cycles (i.e., the samples lost more capacity). To accelerate degradation in minimum cycles, the operating test conditions reported here used a high regeneration temperature at 400° C. and sample almost fully loaded with water (>10 mol/kg) when exposed to 0.7 bar CO.sub.2. For conventional temperature swing cycle with LTA zeolites, the adsorbents bed are normally changed after approximately 2000 cycles. Giving the RHO material shows no degradation after 300 cycles at severe conditions, it should easily last more than 2000 cycle with a corresponding capacity drop less than 10% over 2000+ cycles.

(40) In another TSA cycle experiment, the material stability for LTA zeolite structures were compared under the wet CO.sub.2 atmosphere with temperature cycling from 35 to 400° C. The desorption process used wet CO.sub.2 feed including 20 sccm CO.sub.2 and 2 sccm N.sub.2 balance and 5 sccm N.sub.2 bubbling through water at room temperature. This is different compared to previous treatment using dry CO.sub.2 conditions. The adsorption condition is same as previous treatment. The results of this testing are shown graphically in FIG. 7. Compared to the results in FIG. 6, all of the LTA zeolite structures showed more degradation over 60 adsorption cycles under wet CO.sub.2 atmosphere conditions for both adsorption and desorption process. As can be seen in FIG. 7, with similar Si/Al ratios of 1, and increasing pore opening from 3A, to 4A, and to 5A the LTA zeolites show more degradation under wet CO.sub.2 condition for temperature swing adsorption/desorption cycles under wet CO.sub.2 atmosphere conditions. As can be seen in FIG. 7, the LTA zeolites are not be stable over continued adsorption/desorption cycles under the wet CO.sub.2 atmosphere conditions. In FIG. 7 it can be seen that the Zeolite 3A lost 5.2% of its initial water capacity over 220 cycles, while Zeolite 4A lost 6.2% of its initial water capacity, and Zeolite 5A lost 15.2% of its initial water capacity over only 60 cycles. As can be seen in FIG. 7, all of the LTA zeolites showed a clear trend to a near linear decrease in capacity as a function of the # of cycles. As such, the water adsorption of the LTA zeolites are expected to significantly deteriorate over a short life span of adsorption cycles conditions under wet CO.sub.2 cycle conditions. Accordingly, the higher initial capacity LTA zeolite structures show a high rate of degradation (water capacity loss) over multiple adsorption/desorption cycles under wet CO.sub.2 atmosphere conditions.

EXAMPLE 2

Water Kinetics Testing

(41) Under rapid cycle temperature swing and/or pressure swing conditions, the adsorption step may be on the order of just a few seconds. As such, the selected adsorbent generally operates under a “kinetic” separation regime. That is, that in conventional adsorption swing processes, the adsorption time is sufficient to allow the adsorption of the various feed stream components to come to an equilibrium state. These conventional adsorption swing processes generally have an adsorption time on the order of a few hours to a few days. However, under rapid cycle temperature swing and/or pressure swing conditions, the kinetics of the separation process are utilized as an advantage in the separation by selecting an adsorbent and adsorption conditions so that the kinetic adsorption (or diffusivity) selectively favors the desired component to be adsorbed (in this case, H.sub.2O) over the component or components that are not desired for adsorption.

(42) Kinetics of zeolite samples can be measured in the laboratory using ballistic chromatography. For quantification of fast diffusivity measurements, a variation of the chromatographic breakthrough technique may be utilized. In this technique, a small amount of sample (e.g., zeolite crystals) is placed in a packed bed of about 1 centimeter (cm) in lengths, and about 0.1 cm in diameter. The weight of the dry sample in the packed bed is accurately measured and depending on how the packed bed is loaded, the sample can range from between about 2 milligrams (mg) to about 20 mg. The sample placed into the packed bed can be composed of individual zeolite crystals or small aggregates of the crystals. For water vapor delivery helium gas stream is passed through a bubbler, which is maintained at a temperature lower than the temperature of the adsorption bed to avoid condensation. A mass spectrometer with a fast data acquisition is utilized to monitor the effluent concentration of the water vapors.

(43) Typically, the gas velocity is on the order of about 30 centimeters per second (cm/s), and the corresponding gas residence time is very short). The response of the column is indicative of the equilibrium and kinetics of the adsorption process. If the kinetics of the sample are fast, a sharp breakthrough front appears at a time that is more than 30 seconds later than the time at which a front appears with no sample in the cell. The swing adsorption capacity of the sample at the point of breakthrough can be calculated from the time of breakthrough and can be directly calculated from the rate at which molecules are being fed into the bed.

(44) To establish the intrinsic kinetics of the NaCs—RHO zeolite using ballistic chromatography, a small (e.g., 3 milligram (mg) to 10 mg) packed adsorbent bed of zeolite crystals was used to measure breakthrough in a short residence time. The tests were performed at ambient conditions, with the pressure drop in the column typically not exceeding ˜1-4 bar, depending on the crystal size. The testing was performed by flowing helium gas through water.

(45) FIG. 8 is a diagram of the water breakthrough on a NaCs—RHO zeolite packed adsorbent bed. The breakthrough curves of RHO samples at water concentration of about 3 percent (%) in a packed bed configuration along with the associated blank configuration. The response curve 801, which scale has been normalized to a value of 1.0, is for an empty bed containing no adsorbents (or “blank”) to determine measurement response time, while the response curve 805 is for NaCs—RHO packed bed (shown along with shifted blank curve 801a—which a superimposed shifted version of response curve 801). As can be seen in FIG. 8, the NaCs—RHO zeolite data is relatively flat before the breakthrough front (starting at approximately 20 minutes), with no initial bleed through of water. Thus, by comparing response curves 801 and 805, it can be seen that the uptake of water is significantly faster than the residence time of the water in the sample beds, thereby confirming that the NaCs—RHO zeolite possess a high kinetic adsorption rate for water. Additionally, the slope of breakthrough curves of the experimental blank response curve (see shifted/overlaid curve 801a) and sample NaCs—RHO experimental response curve 805 are very comparable, which confirms that there is no appreciable slowdown of adsorption kinetics for water in the NaCs—RHO zeolite sample.

EXAMPLE 3

Long-Term Stability Testing

(46) To test the stability of multiple adsorbent samples for many cycles of heating and cooling, a cyclic stability test apparatus was utilized. The test apparatus had the capability to test up to 8 samples each is in its own channel within a heating/cooling block. Samples are typically run in duplicates within each test as per results shown in FIGS. 9 and 10 for 3A and NaCs—RHO samples. For the testing examples below, the test apparatus was typically run as 4 samples in duplicate under controlled gas environments while cycling from cool to hot.

(47) The test examples provided below, i.e., the “3A (LTA) Testing” and “NaCs—RHO Testing” of Example 3 below, were tested under cyclic conditions cycling between 30° C. to 350° C., with a typical full adsorption/desorption cycle lasting approximately 20 minutes. During the cool portion of the cycle, the zeolite samples (loaded as a small, approximately 25 mg sample contained in a 1 mm tall by 6 mm wide (pancake) bed with flow through gas) were each exposed to a 300 sccm flow of water saturated gas composed of 70% nitrogen and 30% carbon dioxide (plus the water content). Due to the bed and sample holder pressure drops under this moderately high gas flow, the saturator and the bed entry side was typically at approximately 2.5 bara pressure, with the bed exit side held at approximately 1 bara pressure. During the cool adsorption period (or stage), both water and CO.sub.2 are adsorbed onto the bed. The adsorption period was held long enough to ensure that complete breakthrough of water had occurred by monitoring gas hydrometers in line with the exit side of each bed.

(48) Following the cool adsorption portion of the experiment, the beds, contained in a single aluminum temperature control block were all rapidly heated to approximately 350° C. using resistive heater rods embedded in the aluminum block. The heating ramp-up period normally required less than 2 minutes to attain 350° C. To allow desorption of the adsorbed water and carbon dioxide, the samples were then held at 350° C. for an additional 60 seconds, followed by cooling of the temperature control block to 30° C. utilizing cold water flow through channels in the block. During both the heating and high temperature hold (desorption) periods, the gas flow through the samples was switched to 300 sccm of dry carbon dioxide. The cooling period (or stage) typically required approximately 4-5 minutes to reduce the temperature to stabilize at approximately 30° C. As the sample temperatures decreased below 100° C., the gas flow was switched again to the water saturated gas composed of 70% nitrogen and 30% carbon dioxide (plus the water content).

(49) For the data disclosed below for the “3A Testing” and “NaCs—RHO Testing” of Example 3 according to the described test methods, approximately 3000 total cycles were performed with intermediate testing of the water adsorption capacity using an off-line TGA was performed approximately every 500 to 1000 cycles to determine the sample's water adsorption capacity relative the fresh adsorbents' initial water adsorption capacities. The TGA testing procedure was to heat the samples to 400° C. for a 2 hour drying period, followed by adsorbing water from a gas steam composed of 50% water saturated helium blended with 50% dry helium. The measured capacities were then reported in both absolute and “relative residual capacity.”

(50) 3A (LTA) Testing of Example 3:

(51) As per the test methods described in Example 3 above, two samples of 3A zeolite were tested as described above to establish a base case for comparative testing. The results from these tests are shown in FIG. 9. The water capacity determined for the fresh 3A samples under the standardized TGA measurement conditions (above) was 23.47 wt. % based on the dry sample weight with a two sigma standard deviation of 0.16% based on 4 measurements. FIG. 9 shows the residual percent initial capacity over the course of approximately 3000 heating and cooling cycles for the two (2) 3A samples from the test apparatus (samples in Channels 1 and 5, respectively). It may be observed that the 3A samples lost between 44 and 58% of their initial water capacity.

(52) As can be seen, such a deterioration in adsorption capacity for an LTA zeolite would require significant oversizing of the adsorption bed size to allow for the later life capacity loss. This would result in oversizing of associated equipment and additional volume costs for the zeolite adsorbent and related adsorbent bed construction materials, or would require significant downtime for frequent changing out of the adsorbent materials for adsorption swing processes under wet CO.sub.2 conditions.

(53) NaCs—RHO Testing of Example 3:

(54) As per the test methods described in Example 3 above, two (2) samples of NaCs—RHO zeolite were tested under the same conditions as the zeolite 3A testing above. The results from these tests are shown in FIG. 10. The water capacity determined for the fresh NaCs—RHO samples under the standardized TGA measurement conditions (above) was 15.75 wt. % based on the dry sample weight with a two sigma standard deviation of 0.34% based on 4 measurements. FIG. 10 shows the residual percent initial capacity over the course of approximately 3000 heating and cooling cycles for the two (2) NaCs—RHO samples from the test apparatus (samples in Channels 3 and 7, respectively). It may be observed that the NaCs—RHO samples lost only between 5 and 6% of their initial water capacity, an approximate 10× lower loss rate as compared to 3A zeolite under the same conditions.

(55) As can be seen in FIG. 10 that, especially as compared to 3A (LTA) zeolite in FIG. 9, the NaCs—RHO is quite stable over a significant number of swing adsorption cycles modeling the desired duration of a commercial application (approximately 3000 heating and cooling cycles) between adsorbent change outs/replacements. As such, the NaCs—RHO zeolite enables the adsorption bed not to have to be oversized to accommodate for significant later life capacity losses. This improved stability leads to lower costs for the zeolite adsorbent and related adsorbent bed construction materials as it allows the adsorbent bed to run for a significantly larger number of adsorption/desorption cycles than the LTA zeolites. This also results in significantly reduced downtime for changing out of the adsorbent materials for adsorption swing processes under wet CO.sub.2 conditions.

EXAMPLE 4

Water Isotherms for NaCs—RHO Zeolite

(56) FIG. 11 shows water isotherms on NaCs RHO NaCs—RHO zeolite crystal (synthesized per examples herein) over temperature ranges (30° C. to 200° C.) and water pressure ranges (0.00001 to 0.05 bar). The wide range of temperature measurement provides data for design basis for a TSA cycle, and the wide range of pressure measurements additionally provides data for the design basis for deep dehydration cycles to sub ppm level additionally utilizing pressure swing. As water removal for LNG or NGL specifications typically has to meet 0.1 ppmv, water isotherms have been measured down to this range for design purposes.

(57) Even though water capacity on the NaCs—RHO is less than conventional LTA molecular sieves (3A, 4A, and 5A) at room temperature, however, the working capacity, which is defined as the difference of loading between water partial pressure at 0.01 bar and 0.00001 bar at room temperature, is ˜7.5 mol/kg for NaCs—RHO compared to 8.6 mol/kg for 3A zeolite. As illustrated in FIG. 12, in this range, NaCs—RHO has a slightly less working capacity than the 3A zeolite; about 15% less observed for this pressure swing range. However, when comparing the working capacity for the water loading between 0.03 bar and 0.00001 bar at room temperature, both 3A and NaCs—RHO have a similar working capacity of about 9.5 mol/kg. FIG. 13 shows comparison of water isotherms of 3A and NaCs—RHO at high temperature for regenerating samples. For regeneration, NaCs—RHO shows an advantage compared to a conventional LTA molecular sieve. As an example, the water capacity for NaCs—RHO at 150° C. and 0.001 bar is less than that for 3A at 200° C. and 0.001 bar, which means NaCs—RHO is more easily regenerated at lower temperatures. This can save energy cost for regeneration, and more importantly, use of the NaCs—RHO greatly minimizes material degradation which occurs increasingly at higher temperature cycles.

EXAMPLE 5

Stability Comparison of NaCs—RHO to Typical Proton Form of RHO

(58) Counter examples of NH.sub.4—RHO and CaNH.sub.4—RHO have been synthesized using similar steps. The H.sup.+ could be generated from protons in the exchange solution or the use of NH.sub.4 followed by the removal of NH.sub.3 upon heat treatment or desorption in the process. With the same testing conditions used in Example 1 for NaCs—RHO, these proton forms of RHO show significant degradation under similar wet CO.sub.2 cycle conditions. This example shows that not all RHO zeolites maintain high stability for dehydration process, especially when acidic gas is present. FIG. 14 illustrates that the testing of comparative sample of NH.sub.4—RHO (proton form of RHO) shows a loss of water capacity of approximately 6% after only 100 cycles, and approximately 12% after only 200 cycles, and approximately 18% after only 300 cycles. The comparative sample of CaNH.sub.4—RHO (partially exchanged proton form of RHO) shows an approximate 19% water capacity loss after only 100 cycles. In comparison, the NaCs—RHO (i.e., a cationic zeolite RHO of the present disclosure) shows almost no loss of water capacity after 300 cycles.

(59) Since RHO can be ion-exchanged with other alkali metals, an example with Li form RHO (LiCs—RHO) has been synthesized and has been tested under the same conditions. As seen in Na form RHO, the LiCs—RHO shows great stability as well as compared to the proton forms of RHO. No loss of water capacity has been observed in LiCs—RHO after 100 cycles for the same testing conditions in Example 1. Si/Al=6, Li/(Li+Cs)=0.94, (Li+Cs)/Al=0.84 as measured via ICP. FIG. 14 shows a comparison of water adsorption capacity stability among these different forms of RHO under the same wet CO.sub.2 cycle conditions.

EXAMPLE 6

Preparation #2 of NaCs—RHO Zeolite

(60) This example provides another scalable synthesis route for to obtain zeolite RHO materials that are stable for swing adsorption dehydration processes. In particular, it is stable for processes used to dehydrate feeds that contain acid gases.

(61) 159.2 g of 50 wt % NaOH was added to 236.0 g of distilled water. 49.6 g of CsOH monohydrate was added to 38.9 g of distilled water. 58.2 g of dried aluminum hydroxide gel was dissolved in the NaOH solution. A cesium hydroxide solution was added to the sodium aluminate solution. While stirring the mixture, 517.8 g of LUDOX HS-40 was added to the aluminate solution. The mixture was stirred for 30 minutes. The mixture was aged for 18 hours while stirring at room temperature. The mixture was crystallized at 90° C. for 24 hours in a 2 L stainless steel reactor while stirring at 250 RPM. Upon completion of the reaction, the slurry was filtered warm and washed with 2 volumes of water on the filter. After washing, the crystal was dried at 120° C. for 16 hrs. The resulting product was determined to be zeolite RHO via X-ray powder diffraction with crystal sizes of approximately 1 micron or less. Si/Al=2.8, Cs/(Cs+Na)=0.61, (Cs+Na)/Al=1.8 as measured via XRF.

General Embodiments

(62) As discussed, swing adsorption processes may be used to remove water vapor (or simply “water” herein) from a feed stream (such as a gas mixture) because water selectively may adsorb into the micropore of the adsorbent material, and may fill the micropores in certain situations with a greater selectivity than other components of the gas mixture. The swing adsorption processes (e.g., PSA and TSA) may be used to separate gases of a gas mixture because different gases tend to fill the micropore of the adsorbent material to different extents.

(63) As shown in the examples herein, the cationic zeolite RHO materials as disclosed are particularly stable under “wet CO.sub.2 environments” or (or equivalently “acid gas environments”). By the term “wet CO.sub.2 environments” herein, it is meant gas mixtures containing at least 100 ppmv water and 50 ppmv of CO.sub.2 and/or H.sub.2S. A preferred form of the cationic zeolite RHO for use in this process is the aluminosilicate form with Na and Cs cations.

(64) The cationic RHO zeolite used in the swing adsorption dehydration process can be in a non-dehydrated form, a dehydrated form, or a calcined form.

(65) For an example related to the dehydration processes herein, if a feeds stream, such as natural gas containing water (or water vapor), is passed under pressure through a vessel containing an adsorbent material that is more selective towards water vapor than it is for methane, at least a portion of the water vapor is selectively adsorbed by the adsorbent material, and the gas exiting the vessel is enriched in methane. As such, the adsorbent would be considered to have a “selectivity” (or “greater selectivity”) for water over methane. Before the adsorbent material reaches the end of its capacity to adsorb water vapor it is switched from an adsorption step (or “cycle”) to a desorption step. Desorption can be accomplished by raising the temperature of the adsorbent (TSA), purging the adsorbent with a dry stream (PPSA), reducing the pressure of the adsorbent (PSA) or by combinations of these methods. Once the adsorbent has gone through a desorption step it is ready for another adsorption step. Other additional steps such as depressurization, purging, repressurization, or reheating, may alternatively be included in the overall process steps. The combination of the overall steps from the beginning of one adsorption step to the next adsorption step may be referred to as the “total cycle” or the “swing adsorption process cycle”.

(66) Rigorous dehydration is the removal of water so that the concentration of water in the product gas or stream (e.g., the gas exiting the adsorbent bed during the adsorption step) to typically less than 10 ppm on a mole basis, preferably less than 1 ppm on a mole basis or even more preferably less than 0.1 ppm on a mole basis.

(67) In performing rapid cycle swing adsorption system, the adsorbent bed (e.g., in one embodiment a substantially parallel channel contactor) is regenerated before the adsorbent material reaches the end of its capacity to adsorb water vapor. PSA processes can be used to regenerate the adsorbent used for dehydration, but sufficient regeneration involves low pressures (e.g., vacuum pressures) and long periods of time for regeneration. For rapid cycle dehydration processes, after the adsorption step, the adsorbent bed will undergo a desorption step wherein a desorption step product is produced (enriched in water) using rapid cycle PSA, rapid cycle TSA and/or rapid cycle PPSA processes, or a combination thereof (e.g., the desorption step may include both a “pressure swing” in combination with a “temperature swing”). After the desorption step, the adsorbent material may be optionally purged, repressurized, and/or cooled prior to the next adsorption step. The adsorbent material is thus prepared for another adsorption cycle.

(68) In particular, is disclosed the cationic zeolite RHO herein may be utilized in pressure swing adsorption processes (PSA), temperature swing adsorption processes (TSA), partial pressure purge swing adsorption (PPSA), rapid temperature swing adsorption (RTSA), rapid cycle pressure absorption (RCPSA), rapid cycle partial pressure swing adsorption (RCPPSA), or a combination thereof, which may be collectively referred to herein as “swing adsorption processes” unless further defined. The term “rapid cycle swing adsorption processes” refer to rapid temperature swing adsorption (RTSA), rapid cycle pressure absorption (RCPSA), rapid cycle partial pressure swing adsorption (RCPPSA), or a combination thereof. The term “rapid cycle swing adsorption processes” will include such processes as mentioned wherein the total cycle time or period for the rapid cycle swing adsorption processes to go through a full cycle, such as feed/product step(s), desorption step(s), purge step(s) and repressurization step(s) and back to the next feed/product step(s), is a period greater than 1 second and less than 600 seconds. In preferred embodiments, total cycle time or period for the rapid cycle swing adsorption processes is greater than 2 seconds and less than 300 seconds. For example, the total cycle times may be less than 600 seconds, less than 300 seconds, preferably less than 200 seconds, more preferably less than 90 seconds, and even more preferably less than 60 seconds. In rapid cycle processes the residence time of the gas contacting the adsorbent material in the adsorbent bed during the adsorption step is typically short.

(69) In rapid cycle swing adsorption processes the residence time of the gas contacting the adsorbent material in the adsorbent bed during the adsorption step is typically short. For rapid cycle swing adsorption processes, the residence time for gas contacting the adsorbent material in the adsorbent bed during the adsorption step is preferably less than 5.0 seconds, more preferably less than 1.0 seconds and even more preferably less than 0.5 seconds. As such, the adsorbent bed unit may be configured to provide a residence time for gas in the gaseous feed stream contacting the adsorbent material during the adsorption step being less than 5.0 seconds, less than 1.0 seconds, and even more preferably less than 0.5 seconds. The rapid cycle swing adsorption system may be configured to provide a product stream (from the adsorption step) having a concentration of water in the product stream is less than 10 parts per million on a mole basis or more preferably less than 1 part per million on a mole basis. In order to meet these stringent requirements, in a preferred embodiment a parallel channel contactor may be utilized as the contactor's adsorbent bed. However it should be noted that the present invention is applicable to all types of swing adsorption contactor configurations. Non-limiting examples of other types of contactors are packed bed contactors.

(70) Substantially parallel channel contactors can be constructed by coating thin layers of the cationic zeolite RHO and a binder onto a monolith. Substantially parallel channel contactors, such as monoliths, provide very low pressure drop as compared to conventional pellet or other packed beds, which provides a mechanism for the economic use of significantly higher gas velocities and hence higher productivity. One of the primary factors to the performance of a substantially parallel channel contactor and its application for rapid cycle swing adsorption systems is to avoid or minimize mass transfer resistances, and thus allow the intrinsic speed of the primary adsorbent to operate in the kinetic adsorption regime. Avoidance of mass transfer resistances in rapid cycle contactors provide the conditions to facilitate the generation of sharp adsorption fronts, particularly for strong Type 1 isotherm adsorption systems, such as water, in the adsorbent material. Sharp fronts within the length of the contactor provide efficient adsorbate removal to very low concentrations.

(71) Minimization of mass transfer resistance may be accomplished in a substantially parallel channel contactor by several steps. Gas film transfer resistance is minimized by making the gas channels in the contactor of small diameter, such that the distance any adsorbate species has to diffuse in the gas phase is limited to one half the diameter of the gas channel Gas channel diameters, or heights, of less than 2 millimeters are preferred, less than 1 millimeter are more preferred, and less than 600 microns are most preferred. Secondly, limiting the thickness of adsorbate containing coatings reduces the distance that adsorbate species has to diffuse through the macropores and mesopores of the composited adsorbate coating. Preferably, the volume of the zeolite 3A or other molecular sieve is greater than that of the binder and thickness of the layer is less than 800 microns, preferably less than 200 microns and even more preferably less than 125 microns, most preferably less than 60 microns. Further, it is beneficial to minimize the amount of mesopores within the coating layer, with a predominance of macropores being preferred due to the faster diffusion speeds of gas species in macropores as compared to mesopores. It is preferred that at least 50% of the pore volume of the adsorbate coating layer is in macropores, i.e. pore diameters greater than 50 nanometers, more preferably at least 75%, and most preferably greater than 90%. Lastly, adsorbent coating layers with low intrinsic tortuosity are preferred.

(72) While not limiting, suitable contactors may be constructed of adsorbate coatings on ceramic monoliths, or spaced laminated support sheets of metal, metal mesh, polymer, or polymer mesh, or various screens when laminated and spaced with spacers or other means to provide a gas flow channel parallel to the coating layers. Corrugated metal sheets, either layered or spiral wound coated with an adsorbent layer are particularly useful and flexible in their possible designs and gas channel characteristics. Contactors constructed from multiple monoliths or other such structures stacked in series are also particularly useful, as spaces between the monoliths or such provide gas mixing and can minimize front dispersion caused by variations in adsorbate coating thicknesses or gas channel diameters.

(73) Included herein is a process for removing water from a gaseous feed stream, the process comprising performing a swing adsorption process by: a) performing an adsorption step, wherein the adsorption step comprises passing a feed stream comprising water through an adsorbent bed unit comprising at least one adsorbent bed wherein water is selectively separated from the feed stream to form a product stream which has a lower molar fraction of water than the feed stream, wherein the adsorbent bed comprises an adsorbent material which is a cationic zeolite RHO; b) interrupting the flow of the gaseous feed stream; c) performing a desorption step, wherein the desorption step comprises removing at least a portion of the water from the adsorbent material; and d) repeating the steps a) to c) for at least one additional cycle.

(74) In preferred embodiments, the gaseous feed stream comprises from 0.00001 to 0.3 bar of water (H.sub.2O) partial pressure, and from 0.005 to 3.0 bar of carbon dioxide (CO.sub.2) partial pressure. In preferred embodiments, the water uptake capacity of the cationic RHO zeolite after 3000 cycles is preferably at least 80%, more preferably at least 90%, and even more preferably at least 95% of the initial water uptake capacity of the cationic RHO zeolite. In preferred embodiments, this stability initial water uptake capacity of the cationic RHO zeolite after 3000 cycles is based on utilization in a rapid cycle temperature swing adsorption process wherein the temperature difference between the adsorption step and the desorption step is at least 100° C.; more preferably at least 150° C., and/or when the desorption temperature is at least 150° C., at least 200° C., or at least 250° C.

(75) As further enhancements, the process may include some additional variations to the process. For example, the rapid cycle swing adsorption process may comprise a rapid cycle pressure swing adsorption process, a rapid cycle temperature swing adsorption process, a rapid cycle partial pressure swing adsorption process, or any combination thereof; the desorption step may further comprise performing a purge step, wherein the purge step comprises passing a purge stream into the adsorbent bed unit to remove at least a portion of the water from the substantially parallel channel contactor to form a purge product stream; the rapid cycle swing adsorption process may comprise a rapid cycle pressure swing adsorption process; may include performing one or more depressurization steps after step b) and prior to step c), wherein the pressure within the adsorbent bed unit is reduced by a predetermined amount with each successive depressurization step; may include heating the substantially parallel channel contactor to promote the removal of at least a portion of the water from the substantially parallel channel contactor to form a purge product stream; and may include passing a heated purge stream through the substantially parallel channel contactor to promote the removal of at least a portion of the water from the substantially parallel channel contactor to form a purge product stream. The pressure of the feed stream may be in the range between 400 pounds per square inch absolute (psia) and 1500 psia, or in the range from 600 psia to 1200 psi.; wherein the gaseous feed stream may be a hydrocarbon containing stream having greater than one volume percent hydrocarbons based on the total volume of the feed stream; wherein the cycle duration is greater than 2 seconds and less than 300 seconds; wherein residence time for gas in the gaseous feed stream contacting the adsorbent material in the substantially parallel channel contactor during the adsorption step is less than 5.0 seconds, is less than 1.0 seconds or is less than 0.5 seconds; and/or wherein the concentration of water in the product stream is less than 50 parts per million on a mole basis, is less than 1 parts per million on a mole basis or is less than 0.1 parts per million on a mole basis.

(76) The present techniques involve one or more adsorbent bed units to perform a swing adsorption process or groups of adsorbent bed units configured to perform a series of swing adsorption processes. Each adsorbent bed unit may be configured to perform a specific cycle or cycles, which may include an adsorption step and a desorption step. As noted, additional steps may be further included.

(77) In certain configurations, the swing adsorption unit, which includes the adsorbent material, may process a feed stream that comprises hydrocarbons along with water and CO.sub.2. For example, the feed stream may be a hydrocarbon containing stream having greater than one volume percent hydrocarbons based on the total volume of the feed stream. By way of example, the stream may include H.sub.2O and CO.sub.2 as one or more contaminants and the gaseous feed stream may comprise H.sub.2O in the range of 10 parts per million (ppm) molar to 1,500 ppm molar; or in the range of 500 ppm to 1,500 ppm molar; and CO.sub.2 in the range of 50 parts per million (ppm) molar to 2 molar %; or in the range of 500 ppm to 2 molar %. Moreover, the feed stream may include hydrocarbons and H.sub.2O, wherein the H.sub.2O is one of the one or more contaminants and the feed stream comprises H.sub.2O in the range of two ppm molar to saturation levels in the feed stream.

(78) In certain configurations, the adsorbent material may be used in a rapid cycle swing adsorption process, such as a rapid cycle PSA process, to remove moisture from the feed stream. The specific level may be related to dew point of desired output product (e.g., the water content should be lower than the water content required to obtain a dew point below the lowest temperature of the stream in subsequent processing and is related to the feed pressure). As a first approximation, and not accounting for fugacity corrections as a function of pressure, the water concentration in ppm that yields a certain dew point varies inversely with the pressure. For example, the output stream from the adsorbent bed may be configured to be the cryogenic processing feed stream, which satisfies the cryogenic processing specifications (e.g., approximately −150° F. (−101.1° C.) dew point for NGL processes or approximately −60° F. (−51.1° C.) for Controlled Freeze Zone (CFZ) processes. The cryogenic processing feed stream specification may include a water content in the stream (e.g., output stream from the adsorbent bed or feed stream to the to be cryogenic processing) to be in the range between 0.0 ppm and 10 ppm, in the range between 0.0 ppm and 5.0 ppm, in the range between 0.0 ppm and 2.0 ppm, or in the range between 0.0 ppm and 1.0 ppm. The resulting output stream from the adsorbent beds during the purge step may include a water content in the stream to be in the range between 0.0 ppm and 7 pounds per standard cubic feet (lb/MSCF).

(79) In one or more embodiments, the present techniques can be used for any type of swing adsorption process. Non-limiting swing adsorption processes for which the present techniques may include pressure swing adsorption (PSA), vacuum pressure swing adsorption (VPSA), temperature swing adsorption (TSA), partial pressure purge swing adsorption (PPPSA), rapid cycle pressure swing adsorption (RCPSA), rapid cycle thermal swing adsorption (RCTSA), rapid cycle partial pressure swing adsorption (RCPPSA), as well as combinations of these processes, such as pressure and/or temperature swing adsorption. Exemplary swing adsorption processes are described in U.S. Patent Application Publication Nos. 2008/0282892, 2008/0282887, 2008/0282886, 2008/0282885, 2008/0282884 and 2014/0013955 and U.S. Ser. Nos. 15/233,617; 15/233,623; 15/233,631 and 16/233,640, which are each herein incorporated by reference in their entirety.

(80) Further, in certain configurations of the system, the present techniques may include a specific process flow to remove contaminants, such as water (H.sub.2O), in the swing adsorption system. For example, the process may include an adsorbent step and a desorption step, which form the cycle. The adsorbent step may include passing a feed stream at a feed pressure and feed temperature through an adsorbent bed unit having an adsorbent material (e.g., adsorbent bed or substantially parallel channel contactor) to separate one or more contaminants from the feed stream to form a product stream. The feed stream may be passed through the substantially parallel channel contactor in a forward direction (e.g., from the feed end of the substantially parallel channel contactor to the product end of the substantially parallel channel contactor). Then, the flow of the feed stream may be interrupted for a regeneration step. The regeneration step may include one or more depressurization steps, one or more purge steps and/or one or more re-pressurization steps. The depressurization steps may include reducing the pressure of the adsorbent bed unit by a predetermined amount for each successive depressurization step, which may be a single step and/or may be a blowdown step. The depressurization step may be provided in a forward direction or may preferably be provided in a countercurrent direction (e.g., from the product end of the substantially parallel channel contactor to the feed end of the substantially parallel channel contactor). The purge step may include passing a purge stream into the adsorbent bed unit, which may be a once through purge step and the purge stream may be provided in countercurrent flow relative to the feed stream. The purge product stream from the purge step may be conducted away and recycled to another system or in the system. Then, the one or more re-pressurization steps may be performed, wherein the pressure within the adsorbent bed unit is increased with each re-pressurization step by a predetermined amount with each successive re-pressurization step.

(81) Additionally included herein is a swing adsorption system for removing water from a feed stream, the system comprising: a) at least one adsorbent contactor containing an adsorbent bed; b) a feed stream inlet fluidly connected to the at least one adsorbent contactor; and c) a product stream outlet fluidly connected to the at least one adsorbent contactor; wherein the adsorbent bed comprises an adsorbent material which is a cationic zeolite RHO.

(82) In preferred embodiments, the adsorbent bed is a parallel channel contactor. In other preferred embodiments, the swing adsorption system is configured to perform steps comprising: a) perform an adsorption step, wherein the adsorption step comprises passing a feed stream comprising water through the adsorbent bed to form a product stream; b) interrupting the flow of the feed stream; c) performing a desorption step, wherein the desorption step comprises removing at least a portion of the water from the adsorbent material; and d) repeating the steps a) to c) for at least one additional cycle.

(83) Also, the present techniques may be integrated into a various configurations, which may include a variety of compositions for the streams. Adsorptive separation processes, apparatus, and systems, as described above, are useful for development and production of hydrocarbons, such as gas and oil processing.

(84) These rapid cycle swing adsorption processes provide enhancements of using less adsorbent, reducing size of equipment to have less capital cost and foot print. In addition, the rapid cycle swing adsorption processes make possible a mobile system to be used in remote areas, offshore, and other hard to reach places.

Other Embodiments

(85) Herein listed are non-limiting embodiments of the invention as disclosed.

(86) Embodiment 1. A process for removing water from a feed stream, the process comprising performing a swing adsorption process by: a) performing an adsorption step, wherein the adsorption step comprises passing a feed stream comprising water through an adsorbent bed unit comprising at least one adsorbent bed wherein water is selectively separated from the feed stream to form a product stream which has a lower molar fraction of water than the feed stream, wherein the adsorbent bed comprises an adsorbent material which is a cationic zeolite RHO comprising at least one metal cation; b) interrupting the flow of the feed stream; c) performing a desorption step, wherein the desorption step comprises removing at least a portion of the water from the adsorbent material; and d) repeating the steps a) to c) for at least one additional cycle.

(87) Embodiment 2. The process of Embodiment 1, wherein the cationic zeolite RHO is a RHO framework zeolite comprising at least one metal cation selected from the Group 1 and Group 2 elements (new Group 1-18 IUPAC numbering), and the ratio of average atomic metal cation charge to the atomic Al plus Ga content in the crystal as measured by either XRF or AA/ICP is from 0.7 (atomic charges per atom) to 1.5 (atomic charges per atom).

(88) Embodiment 3. The process of any one of Embodiments 1-2, wherein the cationic zeolite RHO comprises an aluminosilicate, a gallosilicate, or a combination thereof.

(89) Embodiment 4. The process of any one of Embodiments 1-3, wherein the at least one metal cation is selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

(90) Embodiment 5. The process of Embodiment 4, wherein the at least one metal cation is selected from Li, Na, K, and Cs.

(91) Embodiment 6. The process of Embodiment 5, wherein the at least one metal cation is Cs.

(92) Embodiment 7. The process of any one of Embodiments 1-3, wherein the cationic zeolite RHO comprises at least two metal cations selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

(93) Embodiment 8. The process of Embodiment 7, wherein one of the at least two metal cations is Cs.

(94) Embodiment 9. The process of any one of Embodiments 1-3, wherein the cationic zeolite RHO comprises at least two metal cations selected from NaCs, KCs, and LiCs.

(95) Embodiment 10. The process of any one of Embodiments 1-9, wherein the cationic zeolite RHO further comprises a non-metallic cation selected from H.sup.+, NH4.sup.+, and a combination thereof.

(96) Embodiment 11. The process of Embodiment 10, wherein the cationic zeolite RHO comprises an aluminosilicate and the cationic zeolite RHO has a ratio of the average non-metallic cation charge to the atomic Al plus Ga content is less than 0.3 (atomic charges per atom).

(97) Embodiment 12. The process of any one of Embodiments 1-11, wherein the feed stream is air.

(98) Embodiment 13. The process of any one of Embodiments 1-11, wherein the feed stream comprises hydrocarbons.

(99) Embodiment 14. The process of Embodiment 13, wherein the feed stream further comprises an acid gas.

(100) Embodiment 15. The process of any one of Embodiments 1-14, wherein the water content of the feed stream is at least about 100 ppm mole fraction and the water content of the product stream is less than about 10 ppm mole fraction.

(101) Embodiment 16. The process of any one of Embodiments 13-15, wherein the product stream comprises at least 90 wt % of the hydrocarbons present in the feed stream.

(102) Embodiment 17. The process of any one of Embodiments 13-16, wherein the feed stream is natural gas.

(103) Embodiment 18. The process of any one of Embodiments 1-17, wherein the cationic zeolite RHO is NaCs—RHO.

(104) Embodiment 19. The process of any one of Embodiments 1-18, wherein the swing adsorption process is selected from TSA, PSA, PPSA, VPSA, and combinations thereof.

(105) Embodiment 20. The process of any one of Embodiments 1-19, wherein the cationic zeolite RHO is in a non-dehydrated form.

(106) Embodiment 21. The process of any one of Embodiments 1-19, wherein the cationic zeolite RHO is in a dehydrated form.

(107) Embodiment 22. The process of any one of Embodiments 1-19, wherein the cationic zeolite RHO is in a calcined form.

(108) Embodiment 23. The process of any one of Embodiments 1-22, the feed stream is in a gaseous state.

(109) Embodiment 24. A swing adsorption system for removing water from a feed stream, the system comprising: a) at least one adsorbent contactor containing an adsorbent bed; b) a feed stream inlet fluidly connected to the at least one adsorbent contactor; and c) a product stream outlet fluidly connected to the at least one adsorbent contactor; wherein the adsorbent bed comprises an adsorbent material which is a cationic zeolite RHO comprising at least one metal cation.

(110) Embodiment 25. The system of Embodiment 24, wherein the adsorbent bed is a parallel channel contactor.

(111) Embodiment 26. The system of any one of Embodiments 24-25, wherein the mass of cationic zeolite RHO in the adsorbent bed is more than 20 wt % of the total materials making up the adsorbent bed.

(112) Embodiment 27. The system of any one of Embodiments 24-26, wherein the swing adsorption system is configured to perform steps comprising: a) perform an adsorption step, wherein the adsorption step comprises passing a feed stream comprising water through the adsorbent bed to form a product stream; b) interrupting the flow of the feed stream; c) performing a desorption step, wherein the desorption step comprises removing at least a portion of the water from the adsorbent material; and d) repeating the steps a) to c) for at least one additional cycle.

(113) Embodiment 28. The system of any one of Embodiments 24-27, wherein the cationic zeolite RHO is a RHO framework zeolite comprising at least one metal cation selected from the Group 1 and Group 2 elements (new Group 1-18 IUPAC numbering), and the ratio of average atomic metal cation charge to the atomic Al plus Ga content in the crystal as measured by either XRF or AA/ICP is from 0.7 (atomic charges per atom) to 1.5 (atomic charges per atom).

(114) Embodiment 29. The system of any one of Embodiments 24-28, wherein the cationic zeolite RHO comprises an aluminosilicate, a gallosilicate, or a combination thereof.

(115) Embodiment 30. The system of any one of Embodiments 24-29, wherein the at least one metal cation is selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

(116) Embodiment 31. The system of Embodiment 30, wherein the at least one metal cation is selected from Li, Na, K, and Cs.

(117) Embodiment 32. The system of Embodiment 31, wherein the at least one metal cation is Cs.

(118) Embodiment 33. The system of any one of Embodiments 24-29, wherein the cationic zeolite RHO comprises at least two metal cations selected from Li, Na, K, Rb, Cs, Mg, Ca, Sr, and Ba.

(119) Embodiment 34. The system of Embodiment 33, wherein one of the at least two metal cations is Cs.

(120) Embodiment 35. The system of any one of Embodiments 24-29, wherein the cationic zeolite RHO comprises at least two metal cations selected from NaCs, KCs, and LiCs.

(121) Embodiment 36. The system of any one of Embodiments 24-35, wherein the cationic zeolite RHO further comprises a non-metallic cation selected from H.sup.+, NH4.sup.+, and a combination thereof.

(122) Embodiment 37. The system of Embodiment 36, wherein the cationic zeolite RHO comprises an aluminosilicate and the cationic zeolite RHO has a ratio of the average non-metallic cation charge to the atomic Al plus Ga content is less than 0.3 (atomic charges per atom).

(123) Therefore, the present invention is well adapted to attain the ends and advantages mentioned as well as those that are inherent therein. The particular embodiments disclosed above are illustrative only, as the present invention may be modified and practiced in different but equivalent manners apparent to those skilled in the art having the benefit of the teachings herein. Furthermore, no limitations are intended to the details of construction or design herein shown, other than as described in the claims below. It is therefore evident that the particular illustrative embodiments disclosed above may be altered, combined, or modified and all such variations are considered within the scope and spirit of the present invention. The invention illustratively disclosed herein suitably may be practiced in the absence of any element that is not specifically disclosed herein and/or any optional element disclosed herein. While compositions and methods are described in terms of “comprising,” “containing,” or “including” various components or steps, the compositions and methods can also “consist essentially of” or “consist of” the various components and steps. All numbers and ranges disclosed above may vary by some amount. Whenever a numerical range with a lower limit and an upper limit is disclosed, any number and any included range falling within the range is specifically disclosed. In particular, every range of values (of the form, “from about a to about b,” or, equivalently, “from approximately a to b,” or, equivalently, “from approximately a-b”) disclosed herein is to be understood to set forth every number and range encompassed within the broader range of values. Also, the terms in the claims have their plain, ordinary meaning unless otherwise explicitly and clearly defined by the patentee. Moreover, the indefinite articles “a” or “an,” as used in the claims, are defined herein to mean one or more than one of the element that it introduces.

(124) All priority documents are herein fully incorporated by reference for all jurisdictions in which such incorporation is permitted. Further, all documents cited herein, including testing procedures, are herein fully incorporated by reference for all jurisdictions in which such incorporation is permitted.