Patent Application
20110094380
1. Field of the Invention
This invention relates to silicoaluminophosphate (SAPO) membranes. More particularly, this invention relates to SAPO membranes supported on porous supports. This invention further relates to SAPO membranes for selective separation of gases in a gas mixture. This invention further relates to supported SAPO membranes and methods for producing such membranes.
2. Description of Related Art
One of the more significant contributors to global warming is the emission of greenhouse gases, particularly carbon dioxide (CO2), into the atmosphere. The primary sources of CO2 emissions are fossil fuel combustion, natural gas sweetening, synthesis gas production and certain chemical plants. The United States is committed to reducing the greenhouse gas intensity of the American economy by 18% over the 10-year period from 2002 to 2012.
Low-temperature distillation is a widely used commercial process for purification and liquefaction of CO2 from streams containing CO2 fractions larger than 90%. However, with atmospheric pressure flue gases, CO2 cannot be effectively condensed. Alkaline sorbents and scrubbing solutions are also employed to remove CO2 from various gas mixtures. Compared to these methods, membrane separation processes are far less expensive, require less energy to operate, and do not need chemicals or regenerating absorbents to maintain. Additionally, membranes are compact and can be retrofitted onto the tail end of power-plant flue gas streams without complicated integration. Permeance and selectivity are two of the basic characteristics or properties of membranes which are useful for determining the potential of a membrane for gas separations.
For the separation of CO2 from the flue gas, it has been reported that a CO2/N2 selectivity of >70 and a minimum CO2 permeance of 3.3×10−7 mol/(m2·s·Pa) or 1,000 GPU (GPU is an industrial unit equivalent to 10−6 cm3(STP)/(cm2·s·cmHg)) are required for the economic operation. The driving force across a gas-separation membrane is the pressure differential between the feed side and the permeate side. Creating this driving force accounts for most of the cost for membrane separation since flue gases are at or slightly above atmospheric pressure. The majority of previous studies have compressed the feed gas to a higher pressure (15 to 20 bar) and set the permeate stream at atmospheric pressure (designated as pressurized feed/atmospheric permeate mode). Under this mode, the feed-gas and the post-separation compressors account for over 50% of the capital and operating costs. To reduce the cost of compressing, another approach is to leave the feed gas close to atmospheric pressure and use vacuum to draw the permeate (designated as atmospheric feed/vacuum permeate mode). Using this mode, it has been estimated that the cost of capturing CO2 using gas-separation membranes is only about 65% of the cost using a pressurized feed.
Polymeric membranes have been successfully applied for the separation of CO2 from natural gas streams. However, they have limitations for flue gas application because of their poor performance, stability at high temperature, and their intolerance to harsh chemicals. Although flue gases can be cooled prior to a separation, the associated energy consumption increases the cost. Therefore, the CO2 permeances and CO2/N2 selectivities of polymeric membranes need to be significantly improved to lower the total cost. An alternative approach is to develop membrane materials that are inherently stable at higher temperatures and harsh chemicals. Molecular sieve materials (such as zeolite) are one such class of materials for highly selective membranes that overcome problems associated with existing polymer materials, and that offer an opportunity to expand membrane technology.
Zeolite membranes are multi-crystalline materials synthesized as a dense layer on the surface of a porous support (α-Al2O3, γAl2O3, or stainless steel) and/or within the pores of the support. The porous support can be thick but with large pores (0.1-5 μm). They provide mechanical strength without introducing additional mass transfer resistance. Because the zeolite membrane is an inorganic oxide and the underlying support is a ceramic or metal, these membranes are far more robust than conventional polymeric membranes and they are usable in high-pressure environments. In addition, these membranes are stable to at least 400° C. as well as in chemically corrosive conditions. In addition to their robustness, zeolite membranes are of interest because they are able to separate gas mixtures with high selectivity. Depending upon the type of zeolite, the mixture system, and the operating conditions, mixtures are separated in accordance with at least the following three principles or mechanisms: 1) molecular sieving, where larger molecules are unable to fit into the pores, and thus the smaller molecules preferentially permeate; 2) differences in diffusivity, where the smaller, less hindered type of molecule in a mixture diffuses faster than the larger ones; and 3) competitive adsorption, where one type of molecule is more strongly adsorbed on the zeolite and thus can dramatically inhibit permeation of another type of molecule. Separation selectivity depends on the particular zeolite used for the membrane, its chemical composition (e.g., Si/Al ratio), the crystal orientation, the identity of the charge neutralization ion, and the quality of the membrane. The kinetic diameters of CO2 and N2 are 0.33 nm and 0.364 nm, respectively. Thus, to obtain a high CO2 flux, by way of differences in diffusivities responsible for the CO2/N2 selectivity, zeolite membranes should have pore sizes (diameters) in the range of about 0.35-0.55 nm. SAPO-34 membranes, which have pore sizes of 0.38 nm, have been shown to be effective for removal of CO2 from natural gas with CO2/CH4 separation selectivities higher than 170, CO2 permeances as high as about 2×10−6 mol/(m2·s·Pa) at 22° C., and a feed pressure of 224 kPa.
SAPO-34 is a silicoaluminophosphate having the composition SixAlyPzO2 where x=0.01-0.98, y=0.01-0.60, and z=0.01-0.52. The SAPO-34 structure is formed by substituting silicon for phosphorous in the AlPO4 which has a neutral framework and exhibits no ion exchange capacity. SAPO-34 has been found to be highly stable in humid atmospheres at temperatures over 100° C. Below this temperature, 2 days of hydration reduces the crystallinity and porosity, but they are completely recovered by calcination in a dry environment.
However, for other CO2-containing mixtures, there remains a need for improved methods for making SAPO membranes, in particular, SAPO membranes having improved separation selectivities for CO2, in particular over N2 in flue gas treatment.
This invention provides methods for making crystalline silicoaluminophosphate (SAPO) membranes on a porous support. Inorganic membranes such as SAPOs can have superior thermal, mechanical and chemical stability, good erosion resistance, and high pressure stability as compared with conventional polymeric membranes. The methods of this invention can produce SAPO membranes and, in particular, SAPO-34 membranes, having improved CO2/N2 selectivities as compared with conventional membranes and which are capable of separating CO2 from post-combustion flue gas.
This invention describes a method for preparing CO2 selective zeolite membranes having thicknesses up to about 5 microns supported on mechanically strong substrates, such as ceramic, metal or carbons, to obtain adequate mechanical strength. The membrane is useful for CO2 capture from post-combustion flue gas, in particular, CO2/N2 separations. The major steps used to make the membrane thin and highly CO2 selective include:
1) Selection of zeolite—The selected zeolites have pores that can discriminate between molecules approx. 0.35-0.5 nm in size. The zeolites also have higher adsorption capacity for CO2 than N2, which is useful because adsorbed CO2 would narrow down membrane pores and further block N2 through.
2) Seeding the substrates—Homogenous zeolite crystals with sizes smaller than 200 nm are used as seeds in the membrane fabrication. A seeding technique, for example, electrophoretic deposition, is applied to attach nano-sized seed crystals to the substrates.
3) Formation of continuous zeolite layer—The seeded support is placed in a synthesis gel followed by hydrothermal synthesis to obtain the desired zeolite layer and structure. The layer has a low fraction of large non-zeolitic pores (grain boundaries) and is about 1 micron thick.
4) Post-synthesis treatment—To tailor pore structure, membrane is post-treated (for example, by using chemical layer deposition) to systematically reduce the zeolite and possible non-zeolite pore sizes, thereby further decreasing the diffusivity of N2 and, thus, increasing CO2/N2 selectivity.
The membranes produced in accordance with the method of this invention can separate CO2 from other gases at elevated temperatures because they are thermally stable at temperatures up to 400° C. The transport mechanism for the membrane is based on an adsorption-diffusion mechanism having five steps: 1) adsorption onto the membrane surface; 2) migration into the zeolite micropores; 3) diffusion through the zeolite micropores; 4) migration out of the pores onto the membrane surface; and 5) desorption from the membrane surface. Competitive adsorption and difference in diffusivities are responsible for the high selectivity. The membrane is selective for CO2 over N2 because CO2 is smaller (diffuses faster) and has higher adsorption coverage than N2. More particularly, the kinetic diameters for CO2 and N2 are 0.33 nm and 0.364 nm, respectively. To obtain high CO2 flux while maintaining the difference in diffusivities responsible for CO2/N2 selectivity, the membranes have pore sizes of approximately 0.35 nm to about 0.5 nm in diameter. 8-member ring zeolites are good candidates for CO2/N2 separation and, thus, we selected a silicoaluminophosphate zeolite, SAPO-34, which has a composition (SixAlyPz)O2, where x=0.01-0.98, y=0.01-0.60, z=0.01-0.52, and x+z=y. It has a chabazite structure with a pore diameter of 0.38 nm (
In accordance with one embodiment, the method of this invention comprises the steps of a) providing a porous support; b) preparing a plurality of SAPO seed crystals; c) preparing an aqueous SAPO synthesis gel comprising a mixture of sources of aluminum, phosphorus, silicon, oxygen, water, and a templating agent; d) contacting the porous support with the SAPO seed crystals, forming a SAPO seeded porous support; e) filling the SAPO seeded porous support with the SAPO synthesis gel, forming a gel-filled porous structure; f) heating the gel-filled porous structure, forming a SAPO layer of SAPO crystals on the surface and/or within pores of the porous support; g) calcining the SAPO layer, thereby removing the templating agent and forming a supported porous SAPO membrane layer; and h) subjecting the supported porous SAPO membrane layer to a pore size reduction post-synthesis treatment process, producing a reduced pore size supported porous SAPO membrane layer having an average pore size of less than about 0.38 nm. As used herein, the term “porous” when used to describe the SAPO membranes, including the SAPO-34 membrane, refers to the porosity characteristics of the individual zeolite crystals of which the membrane is formed as opposed to inter-crystal voids that may undesirably exist in the membrane layer.
These and other objects and features of this invention will be better understood from the following detailed description taken in conjunction with the drawings, wherein:
SAPO-34 membranes, synthesized on porous α-Al2O3 supports by using multiple templates and reduced crystallization time in accordance with one embodiment of the method of this invention, show high CO2 permeability for separating CO2/N2 mixtures up to 230° C. At a trans-membrane pressure drop of 138 kPa and an atmospheric pressure on the permeate side, one such membrane had a CO2 permeance of 1.2×10−6 mol/(m2·s·Pa) (=3,500 GPU) with a CO2/N2 separation selectivity of 32 for a 50/50 feed at 22° C. At a feed pressure of 23 bar, the CO2 flux was as high as 75 kg/(m2·h). CO2/N2 separations were investigated in part by using vacuum permeate pumping, whereby the membrane showed a CO2 permeance of 7.7×10−7 mol/(m2·s·Pa) and a CO2 permeate concentration of 93% for an equimolar feed at 22° C. For a 10% CO2/90% N2 feed, to reach a CO2 permeate concentration of 99%, only three steps were required at 22° C. and 4 steps required at 110° C.
The membranes of this invention are formed by crystallization of an aqueous silicoaluminophosphate-forming gel containing an organic templating agent. The term “templating agent” or “template” is a term of art meaning a species added to the synthesis media to aid in and/or guide the polymerization and/or organization of the building blocks that form the crystal framework. Gels for forming SAPO crystals are known to those versed in the art, but preferred gel compositions for forming membranes may differ from preferred compositions for forming loose crystals. The preferred gel composition may vary depending upon the desired crystallization temperature.
SAPO-34 membranes were prepared by secondary growth onto a tubular porous α-Al2O3(0.2-μm pores) structure. The permeate area was approximately 5.5 cm2. Before synthesis, the supports were boiled in deionized water for 1 h and dried at 150° C. for 30 min.
In accordance with one exemplary embodiment of the method of this invention, 6.8 gm of Al(i-C3H7O)3 (>99%, Aldrich), 3.85 gm of H3PO4 (85 wt % aqueous solution, Aldrich) and 20 gm of deionized H2O were mixed together and stirred for 2 hrs to form an homogeneous solution. Then, 1.13 gm of Ludox AS-40 colloidal silica (40 wt % suspension in water, Sigma-Aldrich) was added to the stirred mixture and the resulting solution was stirred for 0.5 hrs. Next, 12.3 gm of tetraethylammonium hydroxide (20 wt % solution in water, Sigma-Aldrich) was added and the solution was stirred for another 0.5 hrs. Finally, 1.37 gm of dipropylamine (99%, Aldrich) and 1.34 gm of cyclohexylamine (99%, Sigma-Aldrich) were added and the solution stirred for 12 hrs at room temperature. The resulting solution was placed in an autoclave, and treated hydrothermally at 220° C. for 12 hrs, producing SAPO seeds. After cooling to room temperature, the seeds were centrifuged at 2,200 rpm for 20 minutes and washed with water. This procedure was repeated 4 times. The resultant precipitate was dried overnight and calcined at 500° C. for 5 hrs. The calcination heating and cooling rates were 1.0° C./min.
The synthesis gel molar ratio was 1.0 Al2O3:1.0 P2O5:0.45 SiO2:1.2 TEAOH:1.6 dipropylamine:100 H2O. In accordance with one exemplary embodiment of the method of this invention, Al(i-C3H7O)3, H3PO4 and deionized H2O were mixed together and stirred for 0.5 hrs to form an homogeneous solution to which Ludox AS-40 colloidal silica was added and the resulting solution stirred for another 0.5 hrs. Then, tetraethylammonium hydroxide and dipropylamine were added, and the solution stirred for 12 hrs at room temperature. The membranes were prepared by rubbing the inside surface of a porous α-Al2O3 support with dry, calcined SAPO-34 seeds. The rubbed porous supports, with their outside wrapped with Teflon tape, were then placed in an autoclave and filled with synthesis gel. The hydrothermal treatment was carried out at 220° C. for 2-6 hrs, after which the membranes were washed with deionized water. The membranes were calcined in air at 390° C. for 10 hrs to remove the templates. The calcination heating and cooling rates were 0.6° C./min.
Powders were collected during the membrane synthesis, calcined at 550° C. for 8 hrs, and used for gas adsorption. Isotherms for CO2 and N2 were measured with pressure steps of approximately 10 bar. The samples were evacuated for 24 hrs or until no changes were observed in the mass anymore. One membrane was broken and analyzed by scanning electron microscopy (SEM).
Single-gas and mixture permeations were measured in a flow system. The membranes were mounted in a stainless steel module and sealed at each end with silicone o-rings. Mass flow controllers were used to mix pure CO2 and N2 gases. The pressure on each side of the membrane was independently controlled. The membrane module was placed in an oven so that separation could carry out at elevated temperatures (up to 250° C.). Fluxes were measured using a bubble flow meter. The compositions of the feed and permeate streams were measured by a CARLE Series 400 gas chromatograph equipped with a thermal conductivity detector and HAYESEP-A column. The oven was kept at 60° C. The permeance of the component i, Pi, is:
For the cross-flow configuration, because one component preferentially permeates through the membrane, the partial pressures in the feed and retentate are quite different. A Log-mean pressure drop was calculated by
where Ji is the flux through the membrane for component i; pf,i, pr,i and pp,i are partial pressures for component i, in feed, retentate, and permeate sides, respectively. The permeability is the permeance multiplied by membrane thickness. The ideal selectivity is the ratio of the single-gas permeances, and the separation selectivity, αi/jsep, is the ratio of the permeances for mixtures.
The SAPO-34 seeds used to synthesize membranes were cubic and rectangular crystals with sizes ranging from 0.5 to 1.2 μm (
Membrane M1, as shown in Table 1, prepared with a crystallization time of 6 hrs had a CO2/N2 separation selectivity of 32 with CO2 permeances of 1.2×10−6 mol/(m2·s·Pa) at 22° C. and under a feed pressure of 240 kPa and atmospheric permeate. Decreasing the crystallization time to 4 hrs (membrane M2) increased the concentration of non-zeolite pores and, thus, decreased the CO2/N2 selectivity. However, its permeance was high as 1.5×10−6 mol/(m2·s·Pa) under the same test conditions. The higher CO2 permeance for membrane M2 was because it was thinner than M1. Further decreasing of the crystallization time to 2 hrs (membrane M3) failed to produce a CO2 selective membrane. It appears that a continuous layer was not formed during such a short crystallization time. It should be noted that the permeances in GPU for M1 (3,500) and M2 (4,500) were much higher than that required for economic industrial operation (100).
Robeson, L. M., “The upper bound revisited”, J. Membr. Sci., 2008, 320, 390, describes recent revisions to the upper bound for CO2/N2 separation selectivities versus CO2 permeabilities (permeance×membrane thickness) of polymeric membranes at about 22° C. (
The CO2/N2 separation properties of the membrane were also evaluated with the feed gas at atmospheric pressure while drawing the permeate under a 5 kPa vacuum. As shown in
Large-pore, medium-pore, and small-pore zeolite membranes have been reported for CO2/N2 separation. Table 2 compares CO2 permeances and CO2/N2 selectivities with known membranes.
In contrast to the known membranes, the SAPO-34 membranes of this invention have a high potential for CO2 capture in flue gas treatment since their CO2 permeances are about 1-2 orders of magnitude higher than other known small-pore zeolite membranes. Even at 110° C., the permeances for the membranes of this invention, either using the pressurized feed/atmospheric permeate mode (4.5×10−7 mol/(m2·s·Pa)) as shown in
The unique material properties and the progress in separation performance of SAPO-34 membranes offer several advantages over currently available membrane technologies for CO2 capture from flue gases. One advantage is the ability to operate continuously at higher temperatures with high flux. The second, but equally important, advantage is the improvement of CO2/N2 separation performance provided by post-synthesis treatment of the membranes to reduce the SAPO pores from their normal (natural) 0.38 nm size to less than 0.364 nm (the kinetic diameter of N2) so as to enhance the differences in diffusivity and molecular sieving in the separation process. The post-synthesis treatment is the key step to produce zeolite membranes with high CO2/N2 selectivity. In some cases, this treatment leads to blockage of non-zeolite pores (grain boundaries). In other cases, it models the zeolite pore size. By virtue of the reduction in pore size afforded by the post-synthesis treatment step, nitrogen may still fit into the pores of the treated membrane, but its permeation is expected to be inhibited more than CO2. Thus, CO2/N2 selectivity can be improved. Possible post-synthesis treatment methods are described herein below. However, any post-synthesis method which reduces the pore size of the membrane without affecting the integrity of the membrane may be employed. Methods may also be combined to take advantage of the features that each may offer.
In accordance with one embodiment of this invention, the post-synthesis treatment is an ion-exchange method in which the SAPO-34 structure is generally formed by substituting silicon for phosphorous in AlPO4, which has a neutral framework and exhibits no ion exchange capacity. Silica is tetravalent and, thus, the substitution creates acid sites that can be exchanged with alkali cations such as Li+, Na+, K+, NH4+, or Cu2+. The ion-exchange causes steric hindrance or pore narrowing by the adsorbed cations, and thus decreases the permeance of the bigger molecule more than that of the smaller molecule, thereby improving CO2/N2 selectivity.
In accordance with another embodiment of this invention, the post-synthesis treatment is a silylation method in which silane (SiH4) is used to treat the SAPO-34 membranes. Adsorption experiments have determined that n-C4H10 (0.43 nm kinetic diameter) fits into the SAPO-34 pores, but i-C4H10 (0.5 nm diameter) does not. SiH4 has a kinetic diameter of 0.41 nm, and would therefore fit into the SAPO-34 pores. In this process, the reactants SiH4 and O2 are respectively fed outside and inside of the membrane layer. The SiH4 and O2 counter diffuse through the pores, and upon reaction, deposit on the wall of the non-zeolite pores and on the mouth of the zeolite pores. Eventually, this self-limiting diffusion controlled process reduces the sizes of all pores.
In accordance with yet another embodiment of this invention, the post-synthesis treatment is a gas or liquid vapor chemisorption process in which the chemisorptions of gas or liquid on the SAPO pores decreases the fraction of non-SAPO pores. It could also decrease the size of zeolite pore opening.
In summary, the SAPO-34 membranes of this invention have high potential for CO2 capture in flue gas treatment. The membranes, synthesized on porous α-Al2O3 supports by using multiple templates and reduced crystallization time, showed high CO2 permeance for separating CO2/N2 mixtures up to 230° C. At a trans-membrane pressure drop of 138 kPa and an atmospheric pressure in the permeate side, the membranes had a CO2 permeances of 1.2×10−6 mol/m2×s·Pa (3,500 GPU) with CO2/N2 separation selectivity of 32 for a 50/50 feed at 22° C. At a feed pressure of 23 bar, the CO2 flux was as high as 75 kg/(m2·hr). CO2/N2 separations were also effective when using vacuum permeate pumping and leaving the feed at atmospheric pressure. For a 10% CO2/90% N2 feed, at 22° C., only three steps were required to reach a CO2 permeate concentration of 99%. For such a high-flux membrane, even at 110° C., the CO2 permeances generally meet the requirement for the economic industrial operation. The CO2/N2 separation selectivity, however, needs to be further improved by post-synthesis treatment.
While in the foregoing specification this invention has been described in relation to certain preferred embodiments thereof, and many details have been set forth for purpose of illustration, it will be apparent to those skilled in the art that the invention is susceptible to additional embodiments and that certain of the details described herein can be varied considerably without departing from the basic principles of the invention.
