Alkali-promoted activated alumina adsorbent

Abstract

An adsorbent for removing CO.sub.2 from a gas mixture, the adsorbent comprising alumina and a carbonate compound where the carbonate to alumina IR absorbance intensity ratio is reduced by washing the adsorbent with water. The disclosure also describes a method of making adsorbent particles, process for removing CO.sub.2 from a gas mixture using the adsorbent, and an adsorption unit using the adsorbent.

Claims

1. A process for removing CO.sub.2 from a gas mixture containing CO.sub.2, the process comprising: passing the gas mixture containing CO.sub.2 into a bed containing an adsorbent, the adsorbent comprising: alumina; a carbonate compound; and one or more alkali metals; wherein the total amount of alkali metals in the adsorbent is from 0.9 weight % to 10 weight %; and wherein the adsorbent has a carbonate to alumina intensity ratio, R, the carbonate to alumina intensity ratio, R, having a value less than or equal to 0.0150, where the carbonate to alumina intensity ratio is as determined by Fourier Transform infrared (FTIR) spectroscopy of a crushed sample of the adsorbent, wherein the carbonate to alumina intensity ratio is a ratio of a peak absorbance intensity for carbonate, AI.sub.carbonate, to a peak absorbance intensity for alumina, AI.sub.alumina, (i.e. R=AI.sub.carbonate/AI.sub.alumina), each peak absorbance intensity obtained after subtracting a baseline signal intensity, where the peak absorbance intensity for alumina, AI.sub.alumina, is observed at an FTIR wavenumber in a range from 420 cm.sup.1 to 520 cm.sup.1, and the peak absorbance intensity for carbonate, AI.sub.carbonate, is observed at an FTIR wavenumber in a range from 1300 cm.sup.1 to 1400 cm.sup.1; and withdrawing a CO.sub.2-depleted gas from the bed.

2. The process as claimed in claim 1 wherein the total amount of alkali metals in the adsorbent is from 1.0 weight % to 8 weight %.

3. The process as claimed in claim 1 wherein the surface area of the adsorbent ranges from 220 m.sup.2/g to 400 m.sup.2/g.

4. The process as claimed in claim 1 wherein the amount of alumina in the adsorbent is from 90 to 99 weight %.

5. The process as claimed in claim 1 wherein the alkali metals are Na and/or K.

6. The process as claimed in claim 1 wherein in the adsorbent the amount of carbonate is lower than the amount of carbonate arithmetically necessary for stoichiometrically compensating the charge of the total amount of alkali metals.

7. The process as claimed in claim 1 wherein the value of the carbonate to alumina intensity ratio ranges from 0.003 to 0.0150.

8. The process as claimed in claim 1 wherein the gas mixture containing CO.sub.2 has a concentration of CO.sub.2 that ranges from 5 ppmv CO.sub.2 to 1 mole % CO.sub.2.

9. The process according to claim 1 wherein the gas mixture contains oxygen, nitrogen, and water.

10. The process as claimed in claim 1 wherein the gas mixture is a feed to a cryogenic air separation unit.

11. An adsorption unit comprising a bed containing an adsorbent, the adsorbent comprising: alumina; a carbonate compound; and one or more alkali metals; wherein the total amount of alkali metals in the adsorbent is from 0.9 weight % to 10 weight %; and wherein the adsorbent has a carbonate to alumina intensity ratio, R, the carbonate to alumina intensity ratio, R, having a value less than or equal to 0.0150, where the carbonate to alumina intensity ratio is as determined by Fourier Transform infrared (FTIR) spectroscopy of a crushed sample of the adsorbent, wherein the carbonate to alumina intensity ratio is a ratio of a peak absorbance intensity for carbonate, AI.sub.carbonate, to a peak absorbance intensity for alumina, AI.sub.alumina, (i.e. R=AI.sub.carbonate/AI.sub.alumina), each peak absorbance intensity obtained after subtracting a baseline signal intensity, where the peak absorbance intensity for alumina, AI.sub.alumina, is observed at an FTIR wavenumber in a range from 420 cm.sup.1 to 520 cm.sup.1, and the peak absorbance intensity for carbonate, AI.sub.carbonate, is observed at an FTIR wavenumber in a range from 1300 cm.sup.1 to 1400 cm.sup.1.

12. The adsorption unit as claimed in claim 11 wherein the total amount of alkali metals in the adsorbent is from 1.0 weight % to 8 weight %.

13. The adsorption unit as claimed in claim 11 wherein the surface area of the adsorbent ranges from 220 m.sup.2/g to 400 m.sup.2/g.

14. The adsorption unit as claimed in claim 11 wherein the amount of alumina in the adsorbent is from 90 to 99 weight %.

15. The adsorption unit as claimed in claim 11 wherein the alkali metals are Na and/or K.

16. The adsorption unit as claimed in claim 11 wherein in the adsorbent the amount of carbonate is lower than the amount of carbonate arithmetically necessary for stoichiometrically compensating the charge of the total amount of alkali metals.

17. The adsorption unit as claimed in claim 11 wherein the value of the carbonate to alumina intensity ratio ranges from 0.003 to 0.0150.

Description

BRIEF DESCRIPTION OF SEVERAL VIEWS OF THE DRAWINGS

(1) FIG. 1 is a plot of end of feed CO.sub.2 concentration versus feed contact time for as-received (unwashed) and washed adsorbent AA4.

(2) FIG. 2 is a plot of FTIR absorbance spectra for as-received (unwashed) and washed AA1 adsorbent.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

(3) The ensuing detailed description provides preferred exemplary embodiments only, and is not intended to limit the scope, applicability, or configuration of the disclosure. Rather, the ensuing detailed description of the preferred exemplary embodiments will provide those skilled in the art with an enabling description for implementing the preferred exemplary embodiments, it being understood that various changes may be made in the function and arrangement of elements without departing from the scope as defined by the claims.

(4) The articles a and an as used herein mean one or more when applied to any feature in embodiments described in the specification and claims. The use of a and an does not limit the meaning to a single feature unless such a limit is specifically stated. The article the preceding singular or plural nouns or noun phrases denotes a particular specified feature or particular specified features and may have a singular or plural connotation depending upon the context in which it is used.

(5) The adjective any means one, some, or all indiscriminately of whatever quantity.

(6) The term and/or placed between a first entity and a second entity includes any of the meanings of (1) only the first entity, (2) only the second entity, and (3) the first entity and the second entity. The term and/or placed between the last two entities of a list of 3 or more entities means at least one of the entities in the list including any specific combination of entities in this list. For example, A, B and/or C has the same meaning as A and/or B and/or C and comprises the following combinations of A, B and C: (1) only A, (2) only B, (3) only C, (4) A and B and not C, (5) A and C and not B, (6) B and C and not A, and (7) A and B and C.

(7) The phrase at least one of preceding a list of features or entities means one or more of the features or entities in the list of entities, but not necessarily including at least one of each and every entity specifically listed within the list of entities and not excluding any combinations of entities in the list of entities. For example, at least one of A, B, or C (or equivalently at least one of A, B, and C or equivalently at least one of A, B, and/or C) has the same meaning as A and/or B and/or C and comprises the following combinations of A, B and C: (1) only A, (2) only B, (3) only C, (4) A and B and not C, (5) A and C and not B, (6) B and C and not A, and (7) A and B and C.

(8) The term depleted means having a lesser mole % concentration of the indicated component than the original stream from which it was formed. Depleted does not mean that the stream is completely lacking the indicated component.

(9) The terms rich or enriched means having a greater mole % concentration of the indicated component than the original stream from which it was formed.

(10) The present disclosure relates to an adsorbent comprising alumina, one or more alkali metals, and a carbonate compound for use in a process to remove CO.sub.2 from a gas mixture containing CO.sub.2, a process and adsorption unit for removing CO.sub.2 from a gas mixture containing CO.sub.2 using adsorbent particles, and a method for making the adsorbent particles.

(11) The adsorbent may be in any known particle form, for example, pellets, beads, powder, monoliths, laminates, or any other form known in the art.

(12) The gas mixture may contain oxygen and nitrogen and may be a feed to a cryogenic air separation unit. The gas mixture may have a concentration of CO.sub.2 that is 1 mole % or less than 1 mole %. The gas mixture may have a concentration of CO.sub.2 that is greater than 5 ppmv CO.sub.2. The gas mixture may also contain water and the adsorbent may also remove water from the gas mixture. The adsorbent may demonstrate simultaneous water and CO.sub.2 removal.

(13) The adsorbent according to the present disclosure comprises alumina, a carbonate compound, and one or more alkali metals. The adsorbent may contain at least 0.9 weight %, or 1.0 weight %, or 1.5 weight %, or 2.0 weight %, or 2.5 weight %, or 3 weight % of the one or more alkali metals. The adsorbent may contain at most 10 weight %, or 9 weight %, or 8 weight %, or 7 weight %, or 6 weight %, or 5 weight % of the one or more alkali metals. The weight % of the one or more alkali metals is the weight % of the respective weight of the alkali metal, not the weight % of the alkali compound. The one or more alkali metals are present in ionic form. The adsorbent may be 90 to 99 weight % alumina. The carbonate compound may be an alkali carbonate, for example, K.sub.2CO.sub.3. The weight % alkali metal may be determined by X-ray fluorescence (XRF) spectroscopy.

(14) The alumina and carbonate compound may be co-formed or spray-formed to form the adsorbent.

(15) The surface area of the adsorbent may range from 220 m.sup.2/g to 400 m.sup.2/g. Adsorbents with surface areas in this range are suitable for air pre-purification. Alumina adsorbents used in air pre-purification typically have a surface area greater than 220 m.sup.2/g because this surface area is required to maintain high water capacity, and the alumina adsorbent must also be capable of removing water in addition to CO.sub.2 to prevent solidification of both CO.sub.2 and water in downstream cryogenic processes. Adsorbents having a surface area less than 220 m.sup.2/g are generally not suitable for removing water from the feed gas mixture for air pre-purification for a cryogenic air separation plant.

(16) The present inventors have discovered that washing an alkali salt promoted activated alumina adsorbent with water significantly improves the performance of the adsorbent for removing CO.sub.2 from air in pressure swing and temperature swing adsorption processes. The adsorbent retains some alkali after washing so that enhanced CO.sub.2 capacity remains, but water soluble carbonate species that appear to hinder the adsorbent's cyclic ability to sorb and desorb CO.sub.2 are removed from the adsorbent.

(17) In the adsorbent of the invention the amount of carbonate may lower than the amount of carbonate arithmetically necessary for stoichiometrically compensating the charge of the total amount of alkali metal, or in the adsorbent the amount of carbonate may be at least by a factor of 1.1 lower than the amount of carbonate arithmetically necessary for stoichiometrically compensating the charge of the total amount of alkali metals. In order to fully compensate the charge of the alkali metals in the adsorbent arithmetically an amount of two alkali metal ions per carbonate (ion) is necessary, as e.g. in K.sub.2CO.sub.3. Accordingly, in this embodiment of the adsorbent, the molar ratio of alkali metal ions to carbonate ions is higher than 2, or is 2.2 or higher, respectively. The amount of alkali metals in the adsorbent can be determined using X-ray fluorescence spectroscopy data. The amount of carbonate ions can be determined using the FT IR method performed to determine the carbonate/alumina peak ratios (see below), by using standards of known carbonate contents and correlating the carbonate IR peak areas to a carbonate quantity.

(18) The adsorbent may be characterized by a carbonate to alumina intensity ratio, R, having a value less than or equal to 0.0150 as determined by Fourier Transform infrared (FTIR) spectroscopy of a crushed sample of the adsorbent. The value of the carbonate to alumina intensity ratio may be less than or equal to 0.014, or less than or equal to 0.013. The value of the carbonate to alumina intensity ratio may be equal to or more than 0.003, or equal to or more than 0.0035, or equal to or more than 0.005. This carbonate to alumina intensity ratio is lower than the carbonate to alumina intensity ratios found in prior art adsorbents.

(19) The carbonate to alumina intensity ratio correlates to the ratio of the mass fraction of carbonate to alumina.

(20) The Fourier Transform infrared (FTIR) spectroscopy may be done, for example, using a Nicolet Nexus 670 FTIR interferometer.

(21) The crushed sample is formed by crushing a sample of the adsorbent, for example, using a mortar and pestle. The crushed sample may have a mean particle size ranging from 10 microns to 300 microns. The particle size may be determined by the method described by Eshel et al. in Soil Sci. Soc. Am. J. 68:736-743 (2004), using a Horiba LA-950 laser particle size analyzer.

(22) As part of the measurement method, the crushed sample of adsorbent is pressed on a diamond crystal in a SmartORBIT Attenuated Total Reflectance accessory.

(23) The carbonate to alumina intensity ratio is a ratio of a peak absorbance intensity for carbonate, carbonate, to AI.sub.carbonate, a peak absorbance intensity for alumina, A/alumina, (i.e. R=AI.sub.carbonate/AI.sub.alumina), each peak absorbance intensity obtained after subtracting a baseline signal intensity. The peak absorbance intensity for alumina and the peak absorbance intensity for carbonate are observed by Fourier Transform infrared (FTIR) spectroscopy of the crushed sample of the adsorbent. The peak absorbance intensity for alumina, AI.sub.alumina, is observed at an FTIR wavenumber in a range from 420 cm.sup.1 to 520 cm.sup.1, and the peak absorbance intensity for carbonate, AI.sub.carbonate, is observed at an FTIR wavenumber in a range from 1300 cm.sup.1 to 1400 cm.sup.1. The peak absorbance intensity is the maximum value in the specified range.

(24) The baseline signal intensity is a linear baseline function determined from two local minimum values of absorbance intensity between frequencies ranging from 1100 cm.sup.1 to 1800 cm.sup.1. Determining baseline signal intensities and subtracting baseline signal intensities from absorbance intensity spectra is routine and standard in the field of FTIR spectroscopy.

(25) The peak absorbance intensity for alumina and the peak absorbance intensity for carbonate may be as determined from an FTIR spectrum as obtained by co-adding multiple scans. In the present case, it has been found beneficial to co-add 128 scans at 4 cm.sup.1 resolution using a Nicolet Nexus 670 FTIR interferometer.

(26) The present disclosure also relates to methods for making adsorbent particles, e.g. particles of adsorbents as described in any of the above embodiments, comprising alumina, one or more alkali metals, and a carbonate compound for adsorbing CO.sub.2 having improved sorbing and desorbing characteristics.

(27) The methods for making adsorbent particles comprise washing alumina materials comprising alumina, one or more alkali metals, and a carbonate compound, which are often referred to as alkali-promoted activated alumina materials comprising a carbonate compound, with water, and drying the washed alumina materials to form the adsorbent particles. The materials are preferably washed with water containing less than 100 ppm total dissolved solids, and more preferably deionized water. Total dissolved solids may be measured, for example, by the method described in American Society of Testing and Materials (ASTM) D5907-13.

(28) Alkali-promoted activated alumina materials comprising a carbonate compound are available commercially, for example, from Axens, BASF, Porocel and/or UOP.

(29) Alkali-promoted activated alumina materials may be formed by incorporating an alkali metal carbonate into activated alumina materials to form alkali-promoted alumina materials. The alkali-promoted alumina materials may be calcined in an atmosphere, for example an air atmosphere, to form the alkali-promoted activated alumina materials, which are subsequently cooled. Calcining temperatures typically range from about 300 C. to 350 C. Activated alumina production is well-known and described, for example, in U.S. Pat. Nos. 3,226,191, 4,568,664, and 5,935,894.

(30) The alkali-promoted activated alumina materials may be made by incorporating alkali metal carbonate into an activated alumina structure by spray coating an alumina support structure. The alkali-promoted activated alumina materials may be made by incorporating alkali metal carbonate into an activated alumina structure by co-forming the alkali metal carbonate with alumina to form the alkali-promoted activated alumina materials. Any known process for making alkali-promoted activated alumina materials may be used.

(31) The washing may be done in a batch or continuous process.

(32) Washing does not remove all of the alkalisome remains either as water insoluble K.sub.2Al(CO.sub.3).sub.2OH dawsonite phase, or potassium cations intercalated into gibbsite phases on the activated alumina surface. This remaining alkali enhances the surface basicity enough to increase equilibrium CO.sub.2 capacity compared to unpromoted alumina, but does not hinder CO.sub.2 kinetics. Pure K.sub.2CO.sub.3 can limit rates of CO.sub.2 sorption by its relatively slow reaction with CO.sub.2 as described by Rahimpour et al., in Chemical Engineering and Processing 43 (2004) pp. 857-865.

(33) The carbonate to alumina intensity ratio of a crushed sample of the adsorbent particles as determined by FTIR spectroscopy is decreased by washing the adsorbent with water.

(34) The alkali-promoted activated alumina materials may be washed sufficiently to produce adsorbent particles having a carbonate to alumina intensity ratio, R.sub.2, which is smaller than the carbonate to alumina intensity ratio, R.sub.1, of the alkali-promoted activated alumina materials before washing.

(35) The alkali-promoted activated alumina materials may be washed sufficiently to produce adsorbent particles having a carbonate to alumina intensity ratio, R, having a value less than or equal to 0.0150 as determined by Fourier Transform infrared (FTIR) spectroscopy of a crushed sample of the adsorbent particles. The value of the carbonate to alumina intensity ratio may range from 0.003 to 0.0150 or may range from 0.0035 to 0.014 or from 0.005 to 0.013 as a result of washing with water.

(36) The description above relating to the adsorbent regarding the carbonate to alumina intensity ratio, FTIR spectroscopy, crushed sample, etc. applies also to this embodiment of the method for making adsorbent particles.

(37) The pH of a washing solution in equilibrium with a quantity of adsorbent after washing is also decreased by washing the adsorbent with water. The method of making the adsorbent particles can therefore also be characterized by the pH of a washing solution after washing the adsorbent.

(38) The alkali-promoted activated alumina materials may be washed with water until the alkali-promoted activated alumina materials have a pH in solution of 9.5 or less than 9.5 or less than 9 thereby forming washed alumina materials. The pH in solution is determined by measuring the pH of an equilibrated 2 liter solution of deionized water containing 100 g of the washed alumina materials. The solution may be considered equilibrated if the pH does not change after repeated measurements over the period of several hours. The pH may be measured using a digital Fisher Science Education pH/ion 510 meter. The pH meter may be calibrated with standard buffer solutions at 4.0, 7.0, and 10.0.

(39) Alkali-promoted activated alumina materials as received from a supplier have a pH in solution of greater than 9 where the pH in solution is as determined by measuring the pH of an equilibrated 2 liter solution of deionized water containing 100 g of the as-received alkali-promoted activated alumina materials.

(40) The washed alumina materials may be dried in an oven at a temperature ranging from 25 C. to 100 C.

(41) The washed alumina materials may be dried in air, a vacuum, or an atmosphere containing greater than 79 mole % N.sub.2 to 100 mole % N.sub.2, or an atmosphere containing Ar or He. Generally, the drying atmosphere should preferably contain less than 50 ppmv CO.sub.2 and contain water with a dew point ranging from 90 C. to 40 C.

(42) The present disclosure also relates to a process for removing CO.sub.2 from a gas mixture containing CO.sub.2. The gas mixture may contain oxygen and nitrogen, and may be a feed to a cryogenic air separation unit. The gas mixture may have a concentration of CO.sub.2 that is 1 mole % CO.sub.2 or less than 1 mole % CO.sub.2. The gas mixture may have a concentration of CO.sub.2 that is greater than 5 ppmv. The gas mixture may also contain water and the adsorbent may also remove water from the gas mixture.

(43) The process comprises passing the gas mixture containing CO.sub.2 into a bed containing the adsorbent having the desired characteristics as described above or made by the methods described above, and withdrawing a CO.sub.2-depleted gas from the bed.

(44) The process may be a pressure swing adsorption (PSA) process. Pressure swing adsorption is well-known. PSA cycles suitable for use with the present adsorbent include U.S. Pat. Nos. 5,656,065, 5,919,286, 5,232,474, 4,512,780, 5,203,888, 6,454,838, and 5,156,657, and U.S. Pat. Appl. No. 2014/0373713.

(45) The process may be a temperature swing adsorption (TSA) process. Temperature swing adsorption is well-known. TSA cycles suitable for use with the present adsorbent include U.S. Pat. Nos. 5,614,000, 5,855,650, 7,022,159, 5,846,295, 5,137,548, 4,541,851, 4,233,038, and 3,710,547.

(46) The present disclosure also relates to an adsorption unit. The adsorption unit comprises a bed containing the adsorbent having the desired characteristics as described above and/or made by the methods described above.

Examples

(47) Several samples of commercially-supplied activated alumina beads having a bead diameter of 2 mm, were washed with deionized water. The specifications for each of the samples are shown in Table 1. Alkali promoter loadings are nominal values, as an average weight percent from a typical commercial lot manufacture. The measured alkali metal content is from X-ray fluorescence spectroscopy data.

(48) Adsorbent AA1 is promoted with 5 wt % potassium carbonate. The adsorbent is a spray-formed adsorbent, and falls within the scope of the adsorbent described in U.S. Pat. No. 5,656,064. 102 grams of the adsorbent was added to 2 liters of deionized water and manually stirred for several minutes. After equilibration, the pH of this 2 liter water solution was 11. The pH was measured using a digital Fisher Science Education pH/ion 510 meter, calibrated with standard buffer solutions at 4.0, 7.0, and 10.0. The solution was decanted. This soaking/washing process was repeated 14 times, after which the beads were filtered from the washing solution with a Buchner funnel. The solution over the alumina beads on the final wash measured a pH of 9.

(49) After filtration the beads were dried in air in an oven at 90 C.

(50) Other adsorbents were similarly washed and dried. These adsorbents included potassium carbonate co-formed adsorbent, AA2, which falls within the scope of the adsorbent described in U.S. Pat. No. 7,759,288, sodium oxide promoted activated alumina, AA3, which falls within the scope of U.S. Pat. No. 6,125,655, and AA4, another commercially available potassium carbonate co-formed adsorbent. An example of AA3 production is described in U.S. Pat. No. 6,125,655 col. 5 line 18. NaOH is used as the impregnating alkali compound, and calcination of NaOH on alumina is described to form an alumina that contains sodium oxide (Na.sub.2O). While no explicit mention of carbonate is made, it is well known in the art that sodium hydroxide will react with trace carbon dioxide in air to form sodium carbonate. Under the calcination temperature of >150 C. taught in U.S. Pat. No. 6,125,655, the formation of Na.sub.2CO.sub.3 is quite favorable (Harris, The Canadian Journal of Chemical Engineering, Volume 41, Issue 4, 1963). Furthermore, it is also highly thermodynamically favorable for Na.sub.2O to react with CO.sub.2 at ambient temperature to form Na.sub.2CO.sub.3 as well, with a heat of reaction of 77 kcal.

(51) TABLE-US-00001 TABLE 1 Adsorbent AA1 AA2 AA3 AA4 Surface area 230 266 230 261 (m.sup.2/g) Bulk density 873 913 766 791 (kg/m.sup.3) Particle size 95% 95% 95% 95% distribution between between between between 8 14 Tyler 8 14 Tyler 8 14 Tyler 8 14 mesh mesh mesh Tyler mesh Alkali promoter K.sub.2CO.sub.3 K.sub.2CO.sub.3 Na.sub.2O K.sub.2CO.sub.3 Promoter 5 8 13 8 loading (wt %) before washing Measured alkali 2.5 4.8 10.0 4.9 metal content before washing (wt. %) Measured alkali 0.7 2.5 4.0 3.1 metal content after washing (wt. %)

(52) Each of the adsorbents was tested in a pressure swing adsorption test rig. Unwashed and washed samples of the adsorbents were tested.

(53) For each adsorbent test, a 1.9 cm (0.75 inch) diameter by 45.7 cm (18 inch) tall vessel was filled with respective adsorbent particles. The single vessel was cycled under PSA conditions with a feed gas flowing at 5 liters per minute for 10 minutes at 308 kPa absolute (30 psig) pressure, followed by depressurization to 136 kPa absolute (5 psig), and a regeneration purge gas flowing at 3.5 liters per minute for 10 minutes at 136 kPa absolute (5 psig) pressure. The feed gas was a gas mixture containing air with 400 ppm CO.sub.2. The purge gas was N.sub.2. The vessel containing the respective adsorbents were cycled until a steady state was achieved.

(54) The concentration of CO.sub.2 at the exit end of the vessel was measured during the feed step. The concentration of CO.sub.2 at the end of the feed step was recorded and a mean value of the CO.sub.2 concentration at the end of feed step calculated for 10 cycles after steady state was achieved. The mean values of the CO.sub.2 concentration at the end of feed for each of the adsorbents in as-received and washed are shown in Table 2.

(55) It is shown that water washed versions of promoted alumina adsorbents reduce the amount of CO.sub.2 present at the end of PSA feed steps in cyclic operation, demonstrating that the washed alumina is removing more CO.sub.2 than the as-received promoted alumina under equivalent process conditions.

(56) In another series of tests, adsorbent AA4 was cycled with incrementally slower feed and purge gas flow rates, and longer feed and purge step times such that the total volume of feed gas (50 liters) and total volume of purge gas (35 liters) processed was the same as the PSA test conditions described above. The resulting effect was an increase to the empty bed contact time (gas flow rate divided by empty bed volume) in both feed and regeneration steps. As shown in FIG. 1, the unwashed AA4 requires a much longer contact time to achieve the same CO.sub.2 removal performance as washed AA4. The results show that washing the adsorbents improves the kinetics of CO.sub.2 adsorption and desorption such that equivalent PSA performance can be achieved with 45% less contact time vs. un-washed alumina adsorbent. The improved kinetics reduces bed size and associated capital costs.

(57) TABLE-US-00002 TABLE 2 AA1 AA2 as-received washed as-received washed CO.sub.2 (ppm) 58.0 34.7 72.8 36.0 AA3 AA4 as-received washed as-received washed CO.sub.2 (ppm) >100 16.3 >100 39.5

(58) The impact of washing on the percent alkali utilization was also determined. The percent alkali utilization, or alkali utilization is defined as:
100*(C.sub.PAAC.sub.AA)/C.sub.CO2
where
C.sub.PAA=CO.sub.2 capacity of promoted activated alumina, in mmol/g;
C.sub.AA=CO.sub.2 capacity of unpromoted activated alumina, in mmol/g; and
C.sub.CO2=calculated capacity if all K.sub.2CO.sub.3 loaded on alumina reacted with CO.sub.2, in mmol/g.

(59) Unpromoted activated alumina, as-received (unwashed) and washed AA4 were tested using a thermogravimetric analyzer (TGA) and potassium content measured by XRF). The as-received AA4 had 4.9 weight % K, corresponding to 0.00063 moles K.sub.2CO.sub.3/g, and the washed AA4 had 3.1 weight % K, corresponding to 0.00040 moles K.sub.2CO.sub.3/g.

(60) For each of the samples, 50 mg of adsorbent were loaded into the TGA. An initial drying step under pure N.sub.2 to 100 C. was performed. While held isothermal at 30 C., 1% CO.sub.2 in N.sub.2 was flowed over the sample at 50 mL/min for 60 min. The sweep gas was then switched to pure N.sub.2, maintained at 30 C., and flowed for another 60 min. These latter two steps were repeated 5 times, and the weight change between the beginning and end of the 1% CO.sub.2 in N.sub.2 step recorded as CO.sub.2 uptake capacity.

(61) The results of the TGA are shown in Table 3.

(62) For as-received AA4 (K.sub.2CO.sub.3 promoted alumina), the utilization is shown to be about 40% in the first exposure to CO.sub.2, and drops to only 4% after 5 cycles of CO.sub.2 exposure and ambient regeneration in N.sub.2. To contrast, it is demonstrated that after water washing the AA4 adsorbent, the first cycle utilization of K.sub.2CO.sub.3 improves to 50%, while cycle 5 utilization improves to 12%, which is 3 times that of the sample without washing. In addition, the washed sample displays a slower rate of capacity loss than the unwashed sample. The unwashed retains only 22% (0.080/0.362) of its original capacity on the 5th cycle, while the washed sample retains 32% (0.104/0.323) of its original capacity.

(63) TABLE-US-00003 TABLE 3 CO.sub.2 capacity K.sub.2CO.sub.3 Utilization (mmol/g) (%) Cycle 1 Cycle 5 Cycle 1 Cycle 5 Unpromoted activated alumina 0.125 0.054 AA4 0.362 0.080 38 4 Washed AA4 0.323 0.104 50 12

(64) The change of the composition of each adsorbent as a result of washing each adsorbent was determined using Fourier Transform infrared (FTIR) spectroscopy.

(65) Each of the adsorbents, both as-received and washed samples, were each individually manually ground with a mortar and pestle until the mean particle size was between 10 and 300 microns as determined using a Horiba LA-950 laser particle analyzer. The resulting powder was dried in an oven at 90 C. for 12 hours in an air atmosphere. After drying, each powder sample was pressed onto a diamond crystal in a SmartORBIT ATR accessory and a spectrum was obtained by co-adding 128 scans at 4 cm.sup.1 resolution with a Nicolet Nexus 670 FT-IR interferometer.

(66) An adsorbance spectrum for AA1 is shown in FIG. 2. The peak absorbance intensity for alumina is observed at an FTIR wavenumber in a range from 420 cm.sup.1 to 520 cm.sup.1, and the peak absorbance intensity for carbonate is observed at an FTIR wavenumber in a range from 1300 cm.sup.1 to 1400 cm.sup.1.

(67) The measured carbonate and alumina peak intensities are corrected by subtracting the baseline signal intensity, measured as a linear baseline function determined from two local minimum values of absorbance intensity between frequencies ranging from 1100 cm.sup.1 to 1800 cm.sup.1. For example, the baseline function for the FTIR spectra measured for AA1, was calculated using values at wavelengths of 1292 cm.sup.1 and 1751 cm.sup.1, where the absorbance intensities (AI) were 0.0587 and 0.0574, respectively. The linear function:
AI=a*W+b
where
AI=absorbance intensity
W=wavelength in cm.sup.1
a=(0.05870.0574)/(12921751)=2.8310.sup.6
b=0.0587(2.8310.sup.6)*1292=0.0624
was determined. The absorbance intensity of the baseline at the carbonate peak wavenumber of 1359 cm1, was then calculated as:
AI=2.83106*1359+0.0624=0.0585.
The same baseline absorbance intensity is used for the alumina peak wavenumber, due to the relative closeness in frequency of the alumina and carbonate wavenumbers, and the insignificance of baseline variation versus the larger intensity of the alumina peak.

(68) The measured intensities, baseline intensity, and baseline subtracted corrected intensities taken from the adsorbance spectrum shown in FIG. 2 are summarized in Table 4 for AA1 and washed AA1. The baseline corrected carbonate peak intensity is the peak intensity minus the baseline intensity. For example, the baseline corrected carbonate peak intensity is 0.0724-0.0585=0.0139.

(69) TABLE-US-00004 TABLE 4 Baseline Baseline corrected corrected Carbonate Alumina carbonate alumina peak peak Baseline peak peak intensity intensity intensity intensity intensity AA1 as 0.0724 0.930 0.0585 0.0139 0.872 received AA1 0.0638 1.02 0.0604 0.00340 0.955 washed

(70) The baseline corrected absorbance intensity for the carbonate peak and the alumina peak, along with the carbonate to alumina intensity ratio, are shown in Table 5 for each of the adsorbents, as-received and washed. The carbonate to alumina intensity ratio is calculated by dividing the baseline corrected carbonate peak intensity by the baseline corrected alumina peak intensity, e.g. for the as-received AA1, the carbonate to alumina intensity ratio is 0.0139/0.872=0.0159.

(71) TABLE-US-00005 TABLE 5 absorbance intensity (counts/sec) AA1 AA2 as-received washed as-received washed carbonate 0.0139 0.00340 0.0151 0.00970 alumina 0.872 0.955 0.700 0.713 carbonate/alumina 0.0159 0.00356 0.0216 0.0136 absorbance intensity (counts/sec) AA3 AA4 as-received washed as-received washed carbonate 0.0267 0.0122 0.0121 0.00860 alumina 0.730 0.985 0.666 0.769 carbonate/alumina 0.0366 0.0124 0.0182 0.0112

(72) The results in Table 5 show that that the carbonate to alumina intensity ratio is significantly reduced as a result of washing the adsorbent, which correspondingly shows that the carbonate species are significantly reduced as a result of washing the adsorbent.

(73) As a result of washing the adsorbents the carbonate to alumina intensity ratio is reduced below 0.0150 for all of the adsorbents, a value less than any of the as-received adsorbents.

(74) It is unexpected that washing does not remove the majority of the potassium given the high water solubility of K.sub.2CO.sub.3. For the AA1 adsorbent, potassium content only drops from 4.9 wt % to 3.1 wt % after washing (63% remains on the surface). Table 1 in U.S. Pat. No. 7,759,288 patent shows CO.sub.2 capacity increases as you go from 0, 5, and 8 wt % K.sub.2CO.sub.3. The result in this disclosure, where removing carbonate from the alumina increases CO.sub.2 capacity in cyclic operation is highly unexpected.

(75) In view of the results shown in Table 2 above, it is seen that the low carbonate containing compositions allow higher utilization of the alkali, particularly for PSA cycles and low temperature TSA cycles.

(76) In this disclosure, it is shown that by removing the unreacted alkali carbonates via water washing, the kinetics of CO.sub.2 sorption/desorption is improved, cyclic CO.sub.2 capacity is increased, and utilization of the K.sub.2CO.sub.3 is increased. This leads to better overall performance in PSA cyclic conditions compared to as-received promoted activated alumina as shown in Table 2.

(77) The utility the present adsorbent provides is a new composition of alumina that provides better CO.sub.2 removal performance in a PSA system, particularly for CO.sub.2 removal prior to cryogenic distillation of air, and a method of manufacture where alkali carbonate or oxide promoted activated alumina can be modified, regardless of promotion method, for improved use in PSA cycles to remove CO.sub.2 from a gas composition.

(78) The washing step is shown to improve PSA performance in various promoted aluminas, including salt sprayed, co-formed, and Na or K impregnated species. IR spectroscopy confirms that alkali promoted aluminas can be modified to a unique composition after washing, where very little carbonate remains in the material.