Supported zeolite films and methods for preparing

Abstract

A method for producing a crystalline film comprising zeolite and/or zeolite-like crystals on a porous substrate is described. The method has the steps of: providing a porous support; modifying at least a surface of the top-layer of said porous support by treatment with a composition having one or more cationic polymer(s); rendering at least the outer surface of said porous support hydrophobic by treatment with a composition having one or more hydrophobic agent(s); subjecting said treated porous support to a composition having zeolite and/or zeolite-like crystals thereby depositing and attaching zeolite and/or zeolite-like crystals on said treated porous support, and growing a crystalline film of zeolite and/or zeolite-like crystals on said treated porous support and calcination. Crystalline films find use in a variety of fields such as in the production of membranes, catalysts etc.

Claims

1. A method for producing a crystalline film comprising zeolite and/or zeolite-like crystals on a substrate or porous support, said method comprising the steps of: a) charge modifying a substrate or porous support, b) then rendering the charge modified substrate or porous support hydrophobic, c) depositing particulate zeolite and/or zeolite-like crystals onto the charge modified and hydrophobic substrate or porous support, and d) growing a crystalline film comprising zeolite and/or zeolite-like crystals from said deposited particulate zeolite and/or zeolite-like crystals to provide a crystalline film comprising zeolite and/or zeolite-like crystals on a substrate or porous support.

2. A method as claimed in claim 1, wherein the substrate or pororus support is impregnated with a solvent, a solution, or a combination thereof before charge modification of the support, wherein the solvent and/or solution comprises water, an alcohol, an aqueous ammonia solution, or a combination thereof.

3. A method as claimed in claim 1, wherein the charge modification is performed using one or more cationic materials.

4. A method as claimed in claim 3, wherein at least one cationic material of the one or more cationic materials is a cationic polymer.

5. A method as claimed in claim 4, wherein the substrate or porous support is immersed in a solution of cationic polymers.

6. A method according to claim 5, wherein the solution is an aqueous ammonia solution.

7. A method as claimed in claim 6, wherein the pH of the aqueous ammonia solution comprising cationic polymer(s) is from 7 to 12.

8. A method according to claim 1, wherein said substrate or porous support is made hydrophobic using hydrophobic agent(s) comprising one or more hydroxamate(s) and/or one or more silane(s).

9. A method according to claim 1, wherein said substrate or porous support is made hydrophobic using hydrophobic agent(s), wherein said hydrophobic agent(s) is/are selected from one or more of the following compound(s): octylhydroxamate, decylhydroxamate,dodecylhydroxamate, octadecyltrihydrosilane, phenyltrimethoxy silane, triethoxypropylsilane, 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane, 1H, 1H, 2H, 2H-perfluorodecyltrichlorosilane, trimethyloctylsilane, octyltrimethoxysilane, 1H, 1H, 2H,2H-perfluorocetylmethyldichlorosilane, 1H, 1H, 2H, 2H-perfluorooctyltrimethoxysilane, or a combination thereof.

10. A method according to claim 8, wherein the zeolite material is selected from the group consisting of silicalite-1, zeolite A, zeolite Beta, the zeolites L, Y, X, ZSM-22, ZSM-11, ZSM-5, ZSM-2, LTA, SAPO-34, DDR, mordenite, chabazite, faujasite, sodalite, ferrierite, MFI, and phillipsite.

11. A method according to claim 1, wherein in step a) the porous support is asymmetric comprising a top layer and the top layer is charge modified.

12. A method according to claim 1, wherein a top surface of the substrate or porous support is charge modified.

13. A method according to claim 1, wherein the product of steps (a) to (c) is used in subsequent steps of the method without calcination.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1. A CCD (charge coupled device) camera image of a water droplet on a top layer of a porous α-alumina support surface modified according to a new and improved method described in the present invention (Example 1). The measured contact angle is 116°.

(2) FIG. 2a. Scanning Electron Microscopy image of silicalite-1 crystals in a densely packed monolayer on the top layer of a porous α-alumina support surface modified according to Example 2.

(3) FIG. 2b. Scanning Electron Microscopy image of densely packed silicalite-1 crystals on the top layer of a porous α-alumina support surface modified according to Example 2.

(4) FIG. 3a. CCD camera image of a water droplet on the top-layer of a porous α-alumina support surface modified according to the prior art described in Example 3.

(5) FIG. 3b. CCD camera image of a water droplet on the top layer of a porous α-alumina support surface modified according to the prior art described in Example 3, but rinsed only once after treatment with cationic polymer solution.

(6) FIG. 4a. CCD camera image of a water droplet on the top layer of a porous α-alumina support surface modified according to the prior art as described in Example 4.

(7) FIG. 4b. Scanning Electron Microscopy image of silicalite-1 crystals on the top layer of a porous α-alumina support surface modified according to the prior art as described in Example 4.

(8) FIG. 5a. Scanning Electron Microscopy image of a surface of a calcined zeolite MFI film synthesized according to the prior art as described in Example 5.

(9) FIG. 5b. A plot of helium permeance vs. relative vapor pressure of n-hexane as measured in permporometry experiments, illustrating the effect of large defects such as pinholes on the permeance of zeolite films (upper trace) and the permeance behavior for a membrane without large defects such as pinholes (lower trace).

(10) FIG. 6. Scanning Electron Microscopy image of a silicalite-1 membrane synthesized on the top layer of a porous α-alumina support after a masking according to the prior art as described in Example 6.

(11) FIG. 7. Scanning Electron Microscopy image of a silicalite-1 membrane synthesized on the top layer of a porous 10 cm α-alumina tubular support prepared according to the present invention as described in Example 8.

(12) FIG. 8. Scanning Electron Microscopy image of a silicalite-1 membrane synthesized on the top layer of a porous 50 cm α-alumina tubular support, prepared according to the present invention as described in Example 9.

(13) FIG. 9. Scanning Electron Microscopy image of a silicalite-1 membrane synthesized on the top layers of a porous α-alumina support with 19 channels, prepared according to the present invention as described in Example 10.

(14) FIG. 10a. Scanning Electron Microscopy image of a ZSM-5 membrane synthesized on the top layer of a porous α-alumina disc support, prepared according to the present invention as described in Example 11.

(15) FIG. 10b. XRD pattern of the ZSM-5 membrane prepared according to the present invention as described in Example 11.

(16) FIG. 11. Scanning Electron Microscopy image of a ZSM-5 membrane synthesized on the top layer of a porous α-alumina disc support, prepared according to the present invention as described in Example 12;

(17) FIG. 12a. Scanning Electron Microscopy image of a cross section of a CHA membrane synthesized on the top layer of a porous α-alumina disc support, prepared according to the present invention as described in Example 13.

(18) FIG. 12b. XRD pattern of the CHA membrane prepared according to the present invention as described in Example 13.

(19) FIG. 13a. Scanning Electron Microscopy image of a cross section of a FAU membrane synthesized on the top layer of a porous α-alumina disc support, prepared according to the present invention as described in Example 14.

(20) FIG. 13b. XRD pattern of the FAU membrane synthesized according to the present invention as described in Example 14. Reflection marked with an asterisk originate from the support.

(21) FIG. 14. Shows the best literature data reported for CO.sub.2/CH.sub.4 separation (filled triangles) and data for membranes prepared according to the present invention Example 15 indicated by squares, and

(22) FIG. 15. A plot of helium permeance vs. relative vapor pressure of n-hexane as measured in permporometry experiments, measured for the tubular MFI membrane described in Example 16.

EXPERIMENTAL SECTION

Example 1

(23) A circular disc support of porous α-alumina (25 mm in diameter, 3 mm thick), provided with a 30 μm thick top layer of 100 nm pores and a 3 mm thick base layer with 3 μm pores was calcined in air at 500° C. for 5 hours to remove any organic substances that might affect the interaction between α-alumina and chemicals used in subsequent steps. The porous support was filled with filtered distilled water by immersion and then the support was treated with filtered aqueous cationic polymer (poly(dimethylamine-co-epichlorohydrin-co-ethylenediamine), 1 wt.-% in water, pH 8) solution for 30 minutes. After 30 minutes in the aqueous polymer solution, the support was rinsed 4 times with a 0.1 M aqueous ammonia solution and subsequently, after drying in oven for 2 hours at 105° C., the support was immersed into a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane for 1 hour. Subsequently, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying in oven for 2 hours at 105° C., the contact angle between a droplet (4 μL) of milliQ water and the disc support was measured with a FibroDat 1121/1122 system equipped with a CCD camera. According to the FibroDat software, the measured contact angle (A) was 116° (FIG. 1). Before treatment with 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane, the contact angle was too low to be measured by the FibroDat system)(<10° and the droplet was quickly disapering into the disc support.

Example 2

(24) A porous α-alumina disc support, as described in Example 1, was calcined in air at 500° C. for 5 hours. After filling the support with filtered distilled water by immersion, the support was immersed in an aqueous solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich). The pH of the aqueous polymer solution had been adjusted to 8 using 25 wt.-% aqueous ammonia solution. After 30 minutes in the aqueous polymer solution, the support was rinsed 4 times with a 0.1 M aqueous ammonia solution and subsequently, after drying in oven for 2 hours at 105° C., the support was immersed into a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying in oven for 2 hours at 105° C., the hydrophobic support was immersed in an aqueous dispersion containing 1 wt.-% silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with 0.1 M aqueous ammonia, the support was calcined in air at 500° C. for 5 hours.

(25) The deposition of the molecular sieve crystals onto the support was examined with Scanning Electron Microscopy (SEM) using a FEI Magellan 400 field emission instrument. The sample was not coated with any coating before the SEM image was recorded. As shown in FIGS. 2a and 2b, the support surface was covered with a densely packed layer of silicalite-1 crystals suitable to grow and intergrow a zeolite film. In FIG. 2a the magnification was 50 000 times and in FIG. 2b the magnification was 100 000 times. All parts of the support were covered with an equally dense layer of silicalite-1 seeds.

Example 3. Prior Art Comparison

(26) A circular disc support of porous α-alumina (25 mm in diameter, 3 mm thick), provided with a 30 μm thick top layer of 100 nm pores and a 3 mm thick base layer with 3 μm pores was calcined in air at 500° C. for 5 hours. The support was then treated by a masking procedure known in the art (WO 00/53298) involving two steps; the first step implies that the top surface of the 100 nm pores was coated with a layer of poly (methyl methacrylate), in the second step the interior of the support was filled with molten hydrocarbon wax at elevated temperature. The hydrocarbon wax solidified after cooling. Subsequently, the poly (methyl methacrylate) was dissolved in acetone and the surface rinsed with 0.1 M aqueous NH.sub.3 before the contact angle between a droplet (4 μL) of milliQ water and the dry disc support was measured with a FibroDat 1121/1122 system equipped with a CCD camera. The contact angle was 92°. This value was expected to be influenced by the measured hydrophobicity of the hydrocarbon wax (contact angle θ=107°). It shows that the probed surface consists of hydrophobic parts (wax) and hydrophilic parts (α-alumina), i.e. the surface is heterogenous from a surface energy point of view. Then the support was treated with an aqueous solution adjusted to pH 8 containing 1 wt.-% cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) in order to reverse the charge of the support surface. Then the support was rinsed in 4 different 0.1 M aqueous ammonia solutions and dried on a laminar flow bench for 12 hours at ambient temperature.

(27) The contact angle measured after treatment with cationic polymer was 87° (FIG. 3a). This value shows that the support surface became hydrophilic after treatment with the cationic polymer solution. To further elucidate the effect of the cationic polymer, the treatment of the support with cationic polymer was repeated and excess polymer on the support surface rinsed away by dipping the support once into 0.1 M aqueous NH.sub.3 before drying it on a laminar flow bench for 12 hours at ambient temperature. It is evident that less rinsing resulted in a contact angle of 64° and accordingly a more hydrophilic support surface due to higher concentration of cationic polymer on the support surface (FIG. 3b).

Example 4. Prior Art Comparison

(28) A circular disc support of porous α-alumina (25 mm in diameter, 3 mm thick), provided with a 30 μm thick top layer of 100 nm pores and a 3 mm thick base layer with 3 μm pores was calcined in air at 500° C. for 5 hours. The calcined support was first immersed in 2.5 wt.-% of 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane in ethanol as solvent (WO 2014140291). The immersion time was 1 hour. After this period of time, the support was rinsed with ethanol according to the procedure described in Example 1 and dried in oven at 105° C. for 2 hours. The hydrophobic support was then immersed for 30 minutes in aqueous ammonia solution (pH=8) containing one wt.-% of the cationic polymer used in Examples 1-3 and rinsed in 4 different 0.1 M aqueous ammonia solutions. After drying in oven for 2 hours at 105° C., the contact angle was measured using the FibroDat 1121/1122 system equipped with a CCD camera. According to the FibroDat software, the measured contact angle was 130° (FIG. 4a). The contact angle of the hydrophobic surface before treatment with cationic polymer was 145°, indicating that the attached cationic polymer made the surface less hydrophobic. Subsequently, the support was immersed in a 1 wt.-% aqueous dispersion containing silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with 0.1 M aqueous ammonia, the support was calcined in air at 500° C. for 5 hours.

(29) The deposition of the molecular sieve crystals onto the support surface was examined with Scanning Electron Microscopy using a FEI Magellan 400 field emission instrument. The sample was not coated with any coating before the SEM image was recorded. As shown in FIG. 4b, the support surface was covered with a sub-monolayer of silicalite-1 crystals, a coverage not suitable to grow and intergrow a zeolite film. Some spots of the support were not covered with silicalite-1 crystals. These empty spots without seed crystals may result in pinholes in the synthesized membrane and/or prevent intergrowth of the crystals in the film.

Example 5. Prior Art Comparison

(30) A tubular support of porous α-alumina with geometry; L=100 mm, D.sub.outer=10 mm, and D.sub.inner=7 mm, was calcined in air at 500° C. The tube was provided with a 30 μm thick inner layer with 100 nm pores and an outer layer with 3 μm pores. The calcined support was first immersed in 2.5 wt.-% of 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane in ethanol as solvent. The immersion time was 1 hour. After this period of time, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. Subsequently, the hydrophobic tubular support was allowed to dry in oven at 105° C. for 2 hours and then the support was immersed in an aqueous solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich). The aqueous polymer solution was adjusted to pH 8 using 25 wt.-% aqueous ammonia solution. After 30 minutes in the aqueous polymer solution, the support was rinsed with 0.1 M aqueous ammonia solution by dipping the support into the ammonia solution four times after which the solution was replaced by a fresh solution. This procedure was repeated four times. Subsequently, the tubular support was treated with wt.-% aqueous dispersion containing silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with 0.1 M aqueous ammonia, the support was immersed in a synthesis solution with the composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1450H.sub.2O and hydrothermally treated in an oil bath at 88° C. for 71 hours. After cooling, the thin-film composite anisotropic membrane was rinsed with 0.1 M aqueous ammonia to remove loosely bound crystals and oligomers from the surface of the membrane. The single gas helium permeance of the membrane was measured to 0.17 before calcination, which proves the presence of defects such as pinholes in the membrane. A perfect membrane would have zero permeance before calcination since the zeolite pores are blocked with template molecules. The surface of the membrane was investigated carefully by Scanning Electron Microscopy. Most of the surface appeared defect free, but pinholes in the film can be observed at certain locations. A representative image of a pinhole in the membrane is shown in FIG. 5a. The pinholes are a result of the empty spots in the seed layer (Example 4). FIG. 5b shows a plot of helium permeance vs. relative vapor pressure of n-hexane, as measured in a permporometry experiment of the membrane prepared according to example 5 (upper trace) and for a membrane prepared according to Example 8 in the present invention (lower trace). As described in the art [Permporometry analysis of zeolite membranes, Journal of Membrane Science. 345 (2009) 276], the helium permeance measured at a relative vapor pressure of n-hexane of (p/p.sub.0)=1 in the permporometry experiment, indicates the amount of defects such as pinholes in the membranes. The membrane prepared according to example 5 has a high helium permeance of 1.4 at this relative vapor pressure (FIG. 5b, upper trace), indicating the existence of defects such as pinholes.

Example 6. Prior Art Comparison

(31) A porous α-alumina disc support, as described in Example 1, was thoroughly rinsed with acetone, ethanol and 0.1 M aqueous ammonia to remove dust and any organic molecules attached to the surface of the substrate. The support was then treated by a masking procedure known in the art (WO 00/53298) involving two steps; the first step implies that the top surface of the 100 nm pores was coated with a thin layer of poly (methyl methacrylate), in the second step the remainder of the support was filled with hydrocarbon wax. Subsequently, the all of the poly(methyl methacrylate) was dissolved in acetone and the surface rinsed with 0.1 M aqueous NH.sub.3 before the support was treated with an aqueous solution adjusted to pH 8 (using 25 wt.-% NH.sub.3(aq)) containing 1 wt.-% cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) in order to reverse the charge of the support surface. After rinsing the support with a 0.1 M aqueous ammonia solution, seeding of the support surface was accomplished by immersion in 1 wt.-% silicalite-1 dispersion for 10 minutes. After seeding, the support was rinsed four times with 0.1 M aqueous solution NH.sub.3 to remove excess seed crystals. The seeded support was immersed in a synthesis solution with the composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1500H.sub.2O and hydrothermally treated in an oil bath at 100° C. for 36 hours. After cooling, the sample was rinsed with 0.1 M aqueous NH.sub.3 solution to remove loosely bound crystals and oligomers from the surface of the membrane.

(32) A representative Scanning Electron Microscopy image of the cross section of the calcined film is shown in FIG. 6. Invasion from the synthesis solution could be detected; the pores between the alumina grains in the support are partially or completely filled with zeolite or silicates from the synthesis solution used for film growth. This is because poly (methyl methacrylate) penetrated into the 100 nm pores or the hydrocarbon wax did not reach completely up to the poly (methyl methacrylate) coating. As the porosity of the alumina support typically is only about 40%, invasion will reduce the permeability of the membrane substantially. Also, invasion makes it difficult to optimize the thickness of the synthesized membrane and may also result in leaching of the support.

(33) The Si/AI ratio of the entire cross section of the film was measured using energy dispersive spectroscopy (EDS) with a method described in the art [J. Membr. Sci. 360 (2010) 265-275]. Depending on the location, the Si/AI ratio varied in the range 45-133. This shows that the masking of the support was uneven. Although the synthesis mixture was free from aluminium, much aluminum was leached from the support and incorporated in the grown film due to the incomplete masking of the alumina support. Consequently, it is not possible to prepare MFI films with high Si/AI ratio on alumina supports by this prior art method.

Example 7

(34) A porous α-alumina disc support, as described in Example 1, was calcined in air at 500° C. for 5 hours. After filling the support with distilled water, the support was immersed in a solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) dissolved in an aqueous ammonia (pH=8.5) solution. After 10 minutes in the aqueous polymer solution, the support was rinsed in 0.1 M aqueous ammonia solution and subsequently, after drying for 2 hours on a hot plate (110° C.) placed in a laminar flow bench, the support was immersed completely in a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying again for 2 hours on a hot plate (110° C.) placed in a laminar flow bench, the support was immersed completely in aqueous dispersion (pH 10) containing 1 wt.-% silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with a 0.1 M aqueous ammonia solution, the support was immersed completely in a synthesis solution with the molar composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1500H.sub.2O and hydrothermally treated in an oil bath at 100° C. for 36 hours. After cooling, the sample was rinsed with 0.1 M aqueous NH.sub.3 solution to remove loosely bound crystals and oligomers from the surface of the membrane.

(35) The Si/AI ratio of the entire cross section of the film was measured at various locations of the support using (EDS), a method described in the art [J. Membr. Sci. 360 (2010) 265-275]. Depending on the location, the Si/AI ratio varied between 97 and 152. The Si/AI ratios measured with EDS in Example 6 (prior art) are smaller than the Si/AI ratios measured with EDS in Example 7 and illustrates the possibility to prepare MFI films with higher Si/AI ratio (less hydrophilic characteristic) with the method disclosed herein as compared to prior art (Example 6).

Example 8

(36) A porous α-alumina tubular support with geometry; L=100 mm, D.sub.outer=10 mm, and D.sub.inner=7 mm, was calcined in air at 500° C. The support was provided with a 30 μm thick top-layer with 100 nm pores and an 1.5 mm thick base layer with 3 μm pores. In-between these layers, there is also a 30-40 μm thick layer of intermediate pore size. The porous support was filled with filtered distilled water and then immersed in an aqueous solution containing one wt.-% of cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) for 30 minutes. The aqueous polymer solution was adjusted to pH 8 using 25 wt.-% aqueous ammonia solution. Subsequently, the tubular support was rinsed with filtered 0.1 M NH.sub.3 solution, by dipping the ceramic tube 10 times in each of 6 Teflon tubes filled with filtered 0.1 M NH.sub.3 solution, and then dried on a laminar flow bench for 12 hours. Thereafter the whole support was treated with filtered 2.5 wt.-% 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane in ethanol as solvent for 1 hour. After this treatment, the whole support was rinsed with filtered ethanol (99.7%) for 1 hour (6 Teflon tubes with filtered ethanol×10 min. in each tube) and dried on a laminar flow bench for 12 hours. Subsequently, the tubular support was treated with filtered one wt.-% aqueous dispersion containing silicalite-1 crystals (pH 10) with an average diameter of about 50 nm. Deposition of the crystals onto the substrate surface was allowed to proceed for 15 minutes. After rinsing with 0.1 M aqueous ammonia solution, the support was immersed in synthesis solution with the molar composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1450H.sub.2O and hydrothermally treated in an oil bath at 88° C. for 71 hours. After cooling, the sample was rinsed with filtered 0.1 M aqueous NH.sub.3 to remove loosely bound crystals and oligomers from the surface of the membrane. The single gas helium permeance of the membrane before calcination was lower than 0.02 verifying that the membrane was essentially free from cracks and pinholes.

(37) A representative Scanning Electron Microscopy image of the cross section of the calcined membrane is shown in FIG. 7. As illustrated in FIG. 7, the pores of the support are fully open. No zeolite invasion of the support could be detected by recording numerous SEM images at various locations of the cross section, implying that synthesis solution could not penetrate through the hydrophobic barrier of the support.

(38) FIG. 5b shows a plot of helium permeance vs. relative vapor pressure of n-hexane, as measured in a permporometry experiment, of the membrane prepared according to Example 8 in the present invention (lower trace). The membrane prepared according to Example 8 in the present invention has a much lower helium permeance of only 0.17 at a relative pressure p/p.sub.0=1, which confirms a much smaller number of defects such as pinholes in comparison to the membrane prepared according to prior art in example 5.

Example 9

(39) A porous α-alumina tube with geometry; L=500 mm, D.sub.outer=10 mm, and D.sub.inner=7 mm, was calcined in air at 500° C. The tube was provided with a 30 μm thick inner layer with 100 nm pores and an 1.5 mm thick outer layer with 3 μm pores. In between these layers there is a 30-40 μm thick layer of intermediate pore size. The porous support was filled with filtered distilled water and then immersed in an aqueous solution containing one wt.-% of cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) for 30 minutes. The aqueous polymer solution was adjusted to pH 8 using 25 wt.-% aqueous ammonia solution. Subsequently, the tubular support was rinsed with filtered 0.1 M NH.sub.3 solution, by dipping the ceramic tube 10 times in each of 6 Teflon tubes filled with filtered 0.1 M NH.sub.3 solution, and then dried on a laminar flow bench for 12 hours. Thereafter, the whole tube was treated with filtered 2.5 wt.-% 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane in ethanol as solvent for 1 hour. After this treatment, the whole support was rinsed with filtered ethanol (99.7%) for 1 hour (6 Teflon tubes with filtered ethanol×10 minutes in each tube) and dried on a laminar flow bench for 12 hours. Subsequently, the tubular support was treated with filtered one wt.-% aqueous dispersion containing silicalite-1 crystals (pH 10) with an average diameter of about 50 nm. Deposition of the crystals onto the substrate surface was allowed to proceed for 15 minutes. After rinsing with 0.1 M aqueous ammonia solution, the support was immersed completely in synthesis solution with the molar composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1450H.sub.2O and hydrothermally treated in an autoclave at 88° C. for 56 hours. After cooling, the sample was rinsed with filtered 0.1 M aqueous NH.sub.3 to remove loosely bound crystals and oligomers from the surface of the membrane. The single gas helium permeance of the membrane before calcination was lower than 0.02, verifying that the membrane was free from large defects such as pinholes.

(40) A representative Scanning Electron Microscopy image of the cross section of the calcined membrane is shown in FIG. 8. As illustrated in FIG. 8, the pores of the support are fully open. No zeolite invasion of the support could be detected by recording numerous SEM images at various locations of the cross section, implying that synthesis solution could not penetrate through the hydrophobic barrier of the support. This illustrates that the method disclosed herein is suitable also for the preparation of large single channel membranes.

Example 10

(41) A porous α-alumina tubular support with 19 channels and the geometry; L=500 mm, D.sub.outer=25 mm, and a channel diameter of 3.5 mm was calcined in air at 500° C. Each channel was provided with a 30 μm thick inner/top layer with 100 nm pores and an outer/base layer with 3 μm pores. The porous support was filled with filtered distilled water and then immersed in an aqueous solution containing one wt.-% of cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) for 30 minutes. The aqueous polymer solution was adjusted to pH 8 using 25 wt.-% aqueous ammonia solution. Subsequently, the tubular support was rinsed with a filtered 0.1 M NH.sub.3 solution, by dipping the ceramic tube 10 times in each of 6 Teflon tubes filled with filtered 0.1 M NH.sub.3 solution, and then dried on a laminar flow bench for 12 hours. Thereafter, the support was immersed completely in filtered 2.5 wt.-% 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane in ethanol as solvent for 1 hour. After this treatment, the whole support was rinsed with filtered ethanol (99.7%) for 1 hour (6 Teflon tubes with filtered ethanol×10 minutes in each tube) and dried on a laminar flow bench for 12 hours. Subsequently, the tubular support was treated with filtered one wt.-% aqueous dispersion (pH 10) containing silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the substrate surface was allowed to proceed for 15 minutes. After rinsing with 0.1 M aqueous ammonia solution, the support was immersed completely in a synthesis solution with the molar composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1450H.sub.2O and hydrothermally treated in an autoclave at 88° C. for 56 hours. After cooling, the sample was rinsed with filtered 0.1 M aqueous NH.sub.3 solution to remove loosely bound crystals and oligomers from the surface of the membrane and calcined in air at 500° C.

(42) A representative Scanning Electron Microscopy image of the cross section of the calcined membrane is shown in FIG. 9. As illustrated in FIG. 9, the pores of the support are fully open. No zeolite invasion of the support could be detected by recording numerous SEM images at various locations of the cross section, implying that synthesis solution could not penetrate through the hydrophobic barrier of the support. This illustrates that the method disclosed herein is suitable also for the preparation of large multi-channel membranes.

Example 11

(43) A porous α-alumina disc support, as described in Example 1, was calcined in air at 500° C. for 5 hours. After filling the support with distilled water, the support was immersed in a solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) dissolved in an aqueous ammonia (pH=8.5) solution. After 10 minutes in the aqueous polymer solution, the support was rinsed in 0.1 M aqueous ammonia solution and subsequently, after drying for 2 hours on a hot plate (110° C.) placed in a laminar flow bench, the support was immersed completely in a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying again for 2 hours on a hot plate (110° C.) placed in a laminar flow bench, the support was immersed completely in aqueous dispersion (pH 10) containing 1 wt.-% silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with 0.1 M aqueous ammonia, the support was immersed completely in a synthesis solution with the molar composition 25 SiO.sub.2: 0.25 Al.sub.2O.sub.3: 3 TPAOH: 1 Na.sub.2O: 100 EtOH: 1600H.sub.2O: and hydrothermally treated in an oil bath at 100° C. for 22 hours. Aluminum isopropoxide (98.0%, Aldrich) was used as aluminum source to obtain a Si/AI ratio of 50 in the synthesis mixture and sodium hydroxide (NaOH, 99.0, Merck) was used to increase the basicity of the solution. After cooling, the membrane was rinsed with filtered 0.1 M aqueous NH.sub.3 to remove loosely bound crystals and oligomers from the surface of the membrane and calcined in air at 500° C.

(44) A representative Scanning Electron Microscopy image of the cross section of the resulting ZSM-5 membrane with an Si/AI ratio of 50 is shown in FIG. 10a. As illustrated in FIG. 10a, the pores of the support are fully open, and no invasion of the support could be detected by recording numerous SEM images at various locations of the cross section, implying that synthesis solution could not penetrate through the hydrophobic barrier of the support. This illustrates that it is possible to prepare MFI zeolite membranes with a higher aluminum content, i.e. ZSM-5 membranes, with the method of the present invasion.

(45) FIG. 10 b shows the XRD pattern of the ZSM-5 membrane, measured with a PANalytical Empyrean X-ray Diffractometer. The membrane was prepared according to the present invention as described in Example 11. The diffraction pattern verifies the MFI structure and reflections from the MFI structure are labeled with the appropriate Miller indices, while reflections from the support are labelled with “α-Al.sub.2O.sub.3”.

Example 12

(46) A porous α-alumina disc support, as described in Example 1, was calcined in air at 500° C. for 5 hours. After filling the support with distilled water, the support was immersed in a solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) dissolved in aqueous ammonia (pH=8.5). After 10 minutes in the aqueous polymer solution, the support was rinsed in 0.1 M aqueous ammonia solution and subsequently, after drying for 2 hours on a hot plate (110° C.) placed in a laminar flow bench, the support was immersed completely in a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying again for 2 hours on a hot plate (110° C.) placed in a laminar flow bench, the support was immersed completely in aqueous dispersion (pH 10) containing 1 wt.-% silicalite-1 crystals with an average diameter of about 50 nm. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with 0.1 M aqueous ammonia, the support was immersed completely in a synthesis solution with the molar composition 25 SiO.sub.2: 0.50 Al.sub.2O.sub.3: 3 TPAOH: 1 Na.sub.2O: 100 EtOH: 1600H.sub.2O: and hydrothermally treated in an oil bath at 150° C. for 14 hours. Aluminum isopropoxide (98.0%, Aldrich) was used as aluminum source to obtain a Si/Al ratio of 25 and sodium hydroxide (NaOH, 99.0, Merck) was used to increase the basicity of the solution. After cooling, the membrane was rinsed with filtered 0.1 M aqueous NH.sub.3 to remove loosely bound crystals and oligomers from the surface of the membrane and calcined in air at 500° C.

(47) A representative Scanning Electron Microscopy image of the cross section of the ZSM-5 membrane with an Si/Al ratio of 25 is shown in FIG. 11. As illustrated in FIG. 11, the pores of the support are fully open, and no invasion of the support could be detected by recording numerous SEM images at various locations of the cross section, implying that synthesis solution could not penetrate through the hydrophobic barrier of the support. This illustrates that it is possible to prepare MFI zeolite membranes with a higher aluminum content, i.e. ZSM-5 membranes, with the method of the present invasion.

Example 13

(48) A porous α-alumina disc support, as described in Example 1, was calcined in air at 500° C. for 5 hours. After filling the support with distilled water, the support was immersed in a solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) dissolved in aqueous ammonia (pH=10). After 10 minutes in the aqueous polymer solution, the support was rinsed in 0.1 M aqueous ammonia solution and subsequently, after drying in oven for 2 hours at 105° C., the support was immersed into a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying in oven for 2 hours at 105° C., the support was immersed in an aqueous dispersion containing 1 wt.-% chabazite crystals with an average diameter of about 178 nm according to dynamic light scattering. The chabazite crystals had been prepared from larger crystals by milling. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with a 0.1 M aqueous ammonia solution, the support was immersed in a synthesis solution with the molar composition of 1 SiO.sub.2: 1,4 N, N, N-trimethyl-1-adamantyl ammonium fluoride: 8H.sub.2O and hydrothermally treated in an oil bath at 160° C. for 12 hours. After cooling, the sample was rinsed with a filtered 0.1 M aqueous NH.sub.3 solution to remove loosely bound crystals and oligomers from the surface of the membrane.

(49) A representative Scanning Electron Microscopy image of the cross section of the calcined CHA membrane is shown in FIG. 12a. As illustrated by the representative SEM-image in FIG. 12a, the pores of the support are fully open, and no invasion of the support could be detected. Numerous SEM images were recorded at various locations of the cross section, with similar results. This shows that the synthesis solution could not penetrate through the hydrophobic barrier of the support, even though the synthesis of the CHA zeolite membrane was carried out at high temperature and pressure in an autoclave.

(50) FIG. 12b shows the XRD pattern of the CHA membrane, prepared according to the present invention as described in Example 13, verifying the CHA structure. The reflections from the CHA film are indexed by the appropriate Miller indices, while the reflection from the alumina support is indexed with “α-Al.sub.2O.sub.3”.

Example 14

(51) A porous α-alumina disc substrate, as described in Example 1, was calcined in air at 500° C. for 6 hours. After filling the support with distilled water, the support was immersed in a solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) dissolved in aqueous ammonia (pH=10). After 10 minutes in the aqueous polymer solution, the support was rinsed in 0.1 M aqueous ammonia solution and subsequently, after drying in an oven for 2 hours at 105° C., the support was immersed into a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluorodecyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying in oven for 2 hours at 105° C., the support was immersed in an filtered (0.45 μm) aqueous dispersion of 0.5 wt.-% FAU crystals. The immersion time was 10 minutes. After the seeded support had been rinsed with filtered 0.1 M ammonia solution, it was immersed in a synthesis solution with the composition: 80 Na.sub.2O: 1 Al.sub.2O.sub.3: 9 SiO.sub.2: 5000 H.sub.2O and treated hydrothermally in an oil bath at 100° C. for 3 hours. After cooling, the sample was rinsed with 0.1 M aqueous NH.sub.3 to remove loosely bound crystals from the synthesis solution attached on the surface of the membrane.

(52) A Scanning Electron Microscopy image of the cross section of the calcined FAU membrane showed that the pores of the support are fully open, and no zeolite invasion of the support could be detected by recording numerous SEM images at various locations of the cross section, implying that synthesis solution could not penetrate through the hydrophobic barrier of the support (FIG. 13a). FIG. 13b shows the XRD pattern of the FAU membrane, synthesized according to Example 14, verifying the FAU structure. Reflections marked with “#” emanates from the FAU structure, while the reflection marked with “*” originates from the porous α-alumina support.

Example 15

(53) A porous α-alumina disc support, as described in Example 1, was calcined in air at 500° C. for 5 hours. After filling the support with distilled water, the support was immersed in a solution containing one wt.-% of a cationic polymer (Poly(dimethylamine-co-epichlorhydrin-co-ethylenediamine) with an average molecular weight of 75000, from Aldrich) dissolved in aqueous ammonia (pH=10). After 10 minutes in the aqueous polymer solution, the support was rinsed in 0.1 M aqueous ammonia solution and subsequently, after drying in oven for 2 hours at 105° C., the support was immersed into a 2.5 wt.-% ethanol solution of 1H, 1H, 2H, 2H-perfluoro decyltriethoxysilane for 1 hour. Thereafter, the support was dipped into ethanol (99.7%) four times after which the ethanol was replaced by new ethanol. This procedure was repeated six times. After drying in oven for 2 hours at 105° C., the support was immersed in an aqueous dispersion containing 1 wt.-% chabazite crystals with size between 20 and 300 nm according to dynamic light scattering. Deposition of the crystals onto the support surface was allowed to proceed for 10 minutes. After rinsing with a 0.1 M aqueous ammonia solution, the support was immersed in a synthesis solution with the molar composition of 1.0 SiO.sub.2: 0.35 N, N, N-trimethyl-1-adamantyl ammonium fluoride: 160.0 H.sub.2O and hydrothermally treated in an oil bath at 160° C. for 36 hours. After cooling, the samples were rinsed with a filtered 0.1 M aqueous NH.sub.3 solution to remove loosely bound crystals and oligomers from the surface of the membrane.

(54) An equimolar gas mixture of CO.sub.2/CH.sub.4 was separated by the CHA membrane at a feed pressure of 6-10 bar and a permeate pressure of 1-2 bar at a temperature from 253 to 320 K. Typical CO.sub.2/CH.sub.4 selectivity versus CO.sub.2 permeance data recorded for the membrane prepared in Example 15 are shown by squares in FIG. 14. In FIG. 14, filled squares indicate room temperature separation data and open squares indicate separation data recorded at temperature below room temperature (253-263 K). FIG. 14 also shows the best separation data for SSZ-13 and SAPO-34 membranes (filled triangles) published in literature viz. from S. Li, J. L. Falconer and R. D. Noble, Microporous Mesoporous Mater., 2008, 110, 310-317; H. Kalipcilar, T. C. Bowen, R. D. Nobel and J. L. Falconer, Chem. Mater., 2002, 14, 3458-3464; N. Kosinov, C. Auffret, C. Gücüyener, B. M. Szyja, J. Gascon, F. Kapteijn and E. J. M. Hensen, J. Mater. Chem. A, 2014, 2, 13083; E. Kim, W. Cai, H. Baik and J. Choi, Angew. Chem. Int. Ed., 2013, 52, 5280-5284; Y. Tian, L. Fan, Z. Wang, S. Qui and G. Zhu, J. Mater. Chem., 2009, 19, 7698; R. Zhou, E. W. Ping, H. H. Funke, J. L. Falconer and R. D. Noble, J. Membr. Sci., 2013, 444, 384.

(55) The combination of high permeance and high selectivity makes our membranes far more separation efficient than other inorganic membranes reported.

Example 16

(56) A porous α-alumina tubular support according to Example 8 was prepared and seeded according to the method described in Example 8. After rinsing with 0.1 M aqueous ammonia solution, the support was immersed in a synthesis solution with the molar composition: 25 SiO.sub.2: 3 TPAOH: 100 EtOH: 1450H.sub.2O and hydrothermally treated in an oil bath at 88° C. for 71 hours. After cooling, the MFI membrane was rinsed with filtered 0.1 M aqueous NH.sub.3 to remove loosely bound crystals and oligomers from the surface of the membrane. The single gas helium permeance of the MFI membrane before calcination was lower than 0.02 verifying that the membrane was essentially free from cracks and pinholes.

(57) FIG. 15 shows a plot of helium permeance vs. relative vapor pressure of n-hexane as measured in a permporometry experiment of the MFI membrane prepared according to Example 16 is about 0.01 after calcination and a relative vapor pressure of 1, which is much lower than for a MFI membrane prepared according to prior art in Example 5.

(58) An equimolar gas mixture of CO.sub.2/H.sub.2 was separated by the MFI membrane described in Example 16 at a feed pressure of 10 bar and a permeate pressure of 1 bar at room temperature. At a feed pressure of 10 bar, a CO.sub.2/H.sub.2 selectivity of 20 was measured and the CO.sub.2 permeance was as high as 99. The high selectivity shows that the tubular membrane is essentially free from defects and the very high permeance shows that the membrane is very thin and permeable.

(59) Throughout the description and claims of this specification, the words “comprise” and “contain” and variations of the words, for example “comprising” and “comprises”, means “including but not limited to”, and is not intended to (and does not) exclude other components, integers or steps.

(60) Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise. Features, integers, characteristics, compounds described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith.

(61) All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive. Each feature disclosed in this specification (including any accompanying claims, abstract and drawings), may be replaced by alternative features serving the same, equivalent or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.