Method to Regenerate Ammonia for the Capture of Carbon Dioxide

to separate ammonia from CO2 after thermal decomposition of ammonium bicarbonate also hinders the ammonia scrubbing technique from being applied in ...
0 downloads 0 Views 105KB Size
Download PDF
904

Energy & Fuels 2002, 16, 904-910

Method to Regenerate Ammonia for the Capture of Carbon Dioxide Houping Huang and Shih-Ger Chang* Environmental Energy Technology Division, Lawrence Berkeley National Laboratory, University of California, Berkeley, California 94720

Thomas Dorchak National Energy Technology Laboratory, Morgantown, West Virginia 26507 Received November 15, 2001

Compared to the conventional MEA process for the capture of carbon dioxide from flue gas, the ammonia scrubbing technique provides advantages of lower material costs and less corrosion to absorber, as well as the potential for saving energy, which will eventually make the process less expensive. However, in addition to the highly volatile nature of ammonia, the lack of a method to separate ammonia from CO2 after thermal decomposition of ammonium bicarbonate also hinders the ammonia scrubbing technique from being applied in practice. This paper investigates a method to regenerate ammonia so as to allow for the ammonia scrubbing technique to be practical in the capture of CO2. In the new method, a weakly basic ion-exchange resin containing amine functional groups is used to regenerate ammonia through absorbing carbonic acid at ambient temperatures from ammonium bicarbonate, the main product formed after the absorption of CO2 by ammonia. The resin can then be regenerated when it is heated by water at temperatures equal to or greater than 50 °C. A 50% energy saving can be expected of the new method when compared to the energy requirement of the MEA process. Also, through several runs of a resin absorption-desorption experiment, the sustainability of the resin for the regeneration of ammonia from ammonium bicarbonate was evidently demonstrated. The regeneration capability of the resin can be maintained consistently at 87% of the original regeneration capability of the resin. The issue of the volatility of ammonia is also addressed.

Introduction Although being the most widely used for the capture of CO2 from flue gas, the processes based on chemical absorption with monoethanolamine (MEA) or diethanolamine (DEA) as the absorbents are still too expensive for large-scale application such as the sequestration of CO2 from power plants.1-6 To reduce the cost of capture of CO2, one method is to find a low cost solvent that can minimize energy requirements, the equipment size, and equipment corrosion. * Corresponding author. E-mail: [email protected]. (1) Buzek, J.; Podkanski, J.; Warmuzinski, K. The Enhancement of the Rate of Absorption of CO2 in Amine Solutions Due to the Marangoni Effect. Energy Convers. Manage. 1997, 38, S69-S74. (2) Chakma, A. CO2 Capture ProcessessOpportunities for Improved Energy Efficiencies. Energy Convers. Manage. 1997, 38, S51-S56. (3) Yeh, J. T.; Pennline, H. W.; Resnik, K. P. Study of CO2 Absorption and Desorption in a Packed Column. Energy Fuel 2001, 15, 274-278. (4) Herzog, H.; Drake, E.; Adams, E. CO2 Capture, Reuse, and Storage Technologies for Mitigating Gobal Climate Change, DOE Final Report DE-AF-22-96PC01257, January 1997. (5) Erga, O.; Juliussen, O.; Lidal, H. Carbon Dioxide Recovery by Means of Aqueous Amines. Energy Convers. Manage. 1995, 36, 387392. (6) Mimura, T.; Simayoshi, H.; Suda, T.; Iijima, M.; Mituoka, S. Development of Energy Saving Technology for Flue Gas Carbon Dioxide Recovery in Power Plant by Chemical Absorption Method and Steam System. Energy Convers. Manage. 1997, 38, S57-S62.

With its lower costs, higher capacity, and efficiency7 for the absorption of carbon dioxide, and the lower decomposition temperatures of ammonium bicarbonate, as well as its being less corrosive to absorber material, the ammonia scrubbing technique possesses many advantages for the capture of carbon dioxide over the conventional MEA process. Currently though, besides concerns of the highly volatile nature of ammonia, a process to regenerate ammonia from its carbonate salts is still lacking, thus rendering the ammonia scrubbing technique impractical. This paper studied the capability of anion-exchange resins to regenerate ammonia from ammonium bicarbonate as well as the feasibility for the regeneration of resin by heated water and a collection of CO2. In essence, the combination of the use of ammonia for the absorption of CO2 and the use of a basic anion-exchange resin for the regeneration of ammonia is another dual alkali approach8 we are researching to lower the cost for the capture and separation of CO2 from flue gas. The (7) Yeh, A. C.; Bai, H. Comparison of Ammonia and Monoethanolamine Solvents to Reduce CO2 Greenhouse Gas Emissions, The Science of the Total Environment. 1999, 228, 121-133. (8) Huang, H. P.; Shi, Y.; Li, W.; Chang, S. G. Dual Alkali Approaches for the Capture and Separation of CO2. Energy Fuel 2001, 15, 263-268.

10.1021/ef010270i CCC: $22.00 © 2002 American Chemical Society Published on Web 05/17/2002

Method to Regenerate NH3 for the Capture of CO2

Energy & Fuels, Vol. 16, No. 4, 2002 905

effectiveness of various types of resin to regenerate ammonia from ammonium bicarbonate will be demonstrated. This paper will also test the sustainability of the resin through a series of adsorption-desorption experiments. Experimental Section Materials. The following materials were used: CO2 (8.5%, with N2 as balance, Baygas Co.), ammonia (NH4OH, 28-30%, J. T. Baker Inc.), ammonium bicarbonate (99+%), Purolite A-830 (Polyamine, moisture: ∼50%, weakly basic anion exchange, Purolite Co.), Amberlite IRA-96 (Polyamine, moisture: ∼60%, Sigma Co.), Amberlite IRA-67 (Polyamine, moisture: ∼60%, Sigma Co.), Supelco WA-30 (Alkylamine, moisture: 49%, Supelco), Amberlyst A-7 (Amine, moisture: 56%, Janssen Chimica), Amberlyst A-21 (Tertiary amine, -N(CH)3, moisture: ∼57%, Janssen Chimica). The following reagents were prepared for titration: 1% phenolphthalein was prepared by dissolving 1.0 g of phenolphthalein, 90 mL of ethanol, and 10 mL of deionized water; 0.1% methyl orange solution was prepared by dissolving 0.1 g of methyl orange into 100 mL of deionized water; 0.077 M hydrochloric acid standard solution (standardized by sodium tetraborate decahydrate. Absorption of CO2. At a flow rate of 500 mL/min, a CO2 (8.5%) gas was passed through a glass reactor with a sintered bottom containing 25 mL of an absorbent (15% NH4OH or the regenerated ammonia solution) to determine CO2 absorption capacity. The temperature was held at 25 °C with a constanttemperature bath (PolyTemp). A cold trap with ice was employed to condense the moisture in the CO2 gas stream that came out of the reactor. The change in the concentration of CO2 with time was monitored by a FT-IR spectrometer (model 1605, Perkin-Elmer Co.) with a 10-cm optical pass gas cell, while the amount of CO2 absorbed was calculated using a concentration vs time curve. Regeneration of Ammonia from Ammonium Bicarbonate by Resin. A volume of 200 mL of 1 M ammonium bicarbonate solution was added to a water-jacketed reactor. Before the addition of a resin, the initial pH of the solution was measure using a digital pH meter with a combination electrode. A 50 g sample of resin (containing 40-60% moisture) was then washed by 200 mL of deionized water. The resin was filtered and the water was discarded. The washed resin was later transferred to the reactor to initiate the adsorption reaction for the regeneration of ammonia. The mixture was stirred at a constant rate of 800 rpm. The temperature was held constant at 25 °C. To trace the kinetic process of the reaction between a resin and ammonium bicarbonate, five samples were taken for analysis, each one lasting 15 min longer than the previous one (0, 15, 30, 45, and 60 min after reaction). Before each sample was taken, the stirring was stopped. A volume of 2 mL of the liquid was then obtained by a pipet and transferred to a 250 mL Erlenmeyer flask. The stirring was resumed to continue the reaction after each sample was taken. While the reaction in the reactor was proceeding, each sample was analyzed through titration.9 For titration, a certain volume of deionized water was added to the 250 mL Erlenmeyer flask with the sample to make up 25 mL of a solution, and 1 drop of phenolphthalein was also added (9) Hou, T. P. Manufacture of Soda with special reference to the ammonia process, a practical treatise, 2nd ed.; American Chemical Society: Hafner Publishing Company, New York and London; 1969; pp 485. (10) Chang, S. G.; Toossi, R.; Novakov, T. The Importance of Soot Particles and Nitrous Acid in Oxidizing SO2 in Atmospheric Aqueous Droplets. Atm. Environ. 1981, 15, 1287-1292. (11) Weast, R. C.; Lide, D. R.; Astle, M. J.; Beyer, W. H. CRC Handbook of Chemistry and Physics, 70th ed.; CRC Press: Boca Raton, FL, 1989-1990.

Figure 1. Absorption of CO2 by original and regenerated ammonia. as the first equivalent point indicator. The solution was titrated using hydrochloric acid until it became colorless, thus indicating the solution had reached the first equivalent point of the titration. The volume of the HCl solution consumed was recorded as V1 (mL). After 1 h, the reaction was terminated. Besides 2 mL of liquid as a sample was taken and analyzed to obtain the concentration of the released ammonia and the total concentration of ammonium, the final pH of the bulk solution in the reactor was also measured by the digital pH meter. The concentration of ammonia (M) regenerated in the solution by the resin was calculated on the basis of eq 1:

[NH4OH] ) (NHClV1)/Vs

(M)

(1)

where NHCl is the concentration of the standard HCl (M), Vs is the volume of each sample (mL), [NH4OH] is the concentration (M) of ammonia released by a resin in the reaction with ammonium bicarbonate, including not only the concentration of free ammonia but also the concentration of ammonia that had been regenerated by the resin and then reacted with ammonium bicarbonate to form ammonium carbonate. [NH4OH] is equal to the concentration of H2CO3 adsorbed by the resin from bicarbonate. The ammonia regeneration efficiency (ARE) of a resin was defined as the amount (mg) of ammonia regenerated by 1 g (on the dry weight basis) of a resin. The value was estimated according to eq 3:

ARE ) ([NH4OH] V MNH3)/[Wresin(1% moisture)]

(mg/g) (3)

where [NH4OH], V, MNH3, and Wresin are the concentration of ammonia regenerated by the resin (M) after reaction for 1 h, the volume of ammonium bicarbonate solution (mL), ammonia molecular formular weight (g/mol), and the weight (g) of the resin with moisture. Following the same procedure stated as above, six kinds of resin, including A-830, IRA-67, IRA-96, WA-30, A-7, and A-21 were investigated. The kinetic curves for the reactions between each of these resins were obtained by plotting the concentration of released ammonia with reaction time, as shown in Figure 4. The ammonia regeneration efficiencies of various resins are summarized in Table 1, as will be discussed later. To examine the effect of ammonium bicarbonate on ammonia regeneration efficiency of resin, 50 mL of 0.1, 0.5, 1.0, and 1.5 M ammonium bicarbonate solutions were employed to react with 50 g of IRA-67 with moisture, respectively. The mixture was stirred at a speed of 800 rpm. The temperature was controlled at 25 °C. After 15 min, the reaction reached its equilibrium and was terminated. Two milliliters of sample from the reactor was taken to determine the concentration of the released ammonia by the titration method. The ARE was

906

Energy & Fuels, Vol. 16, No. 4, 2002

Figure 2. The dependence of ammonia regeneration efficiency on ammonium bicarbonate concentration.

Figure 3. The dependence of ammonia regeneration efficiency on the ratio of resin to ammonium bicarbonate volume.

Huang et al. IRA-67 was employed to react with 70, 120, 170, 220, and 300 mL of 1 M ammonium bicarbonate, respectively. For each run, the reaction was ended as it reached its equilibrium after 15 min, and then 2 mL of liquid was taken for analysis to obtain released ammonia data by titration. The ARE was finally calculated based on eq 3. The relationship between ammonia regeneration efficiency and the ratio is depicted in Figure 3. Recovery of Resin and Collection of Carbon Dioxide. Deionized water at various temperatures was studied for its effectiveness to regenerate the resin, which was believed to have adsorbed carbonic acid in the regeneration of ammonia, the CO2-bearing absorbent. The resin and deionized water were mixed in a 500 mL Erlenmeyer flask. The solution was stirred constantly at a speed of 800 rpm and heated to 50, 80, and 100 °C, respectively. A condenser connected to the flask prevented the water in the mixture from evaporating and separated gaseous CO2 from moisture. The CO2 was released by bubbling N2 gas through the reactor at a flow rate of 100 mL/min. The CO2 coming out of the reactor was then passed through a container with 100 mL of NaOH (1 M) solution to convert the CO2 into carbonate for the determination of the total amount of CO2. The concentration of carbonate in the NaOH (1 M) solution was determined by the titration of 1 mL of the sample solution. The titration was performed as described above for the first equivalent point. After the first equivalent point achieved, HCl standard solution volume used was recorded as V1 mL. Then, one drop of 0.1% methyl orange solution was added to the remaining solution as an indicator for the second equivalent point, and the solution was titrated from green yellow to slight orange red with the addition of hydrochloric acid. The orangered was an indication that the solution had reached its second equivalent point. The reading of the HCl solution consumed was recorded as V2 (mL). As the presence of excessive amount of NaOH, the original sample contained both NaOH and Na2CO3. When the sample was titrated by the addition of HCl to the first equivalent point, NaOH was converted into NaCl and H2O, while Na2CO3 was transformed into NaCl and NaHCO3. From the first equivalent point to the second equivalent point, the amount of HCl was consumed to titrate NaHCO3 to NaCl and H2CO3. As the concentration of Na2CO3 is equal to that of NaHCO3, which was calculated as eq 4:

[HCO3-] ) NHCl(V2 - V1)

(M)

(4)

The amount of CO2 absorbed in the NaOH solution, i.e., the emitted CO2 amount from the reaction, can be calculated as follows: Figure 4. Kinetic curves for the regeneration of ammonia by resin.

WCO2 ) [HCO3-]VNaOHMCO2 (mg)

Table 1. Comparison of Ammonia Regeneration Efficiency of Various Resins

where WCO2, VNaOH, and MCO2 are the emitted CO2 amount (mg), the volume of NaOH absorption solution (mL), and molecular formular weight of CO2 (g/mol). By measuring the change of the emitted CO2 with the desorption reaction time, the dependence of the emitted CO2 on the reaction time at various temperatures was obtained (Figure 5). The Sustainability of Resin for Ammonia Regeneration and Its Recycling. Several cycles were run to examine the sustainability of resin for the regeneration of ammonia and itself with collection of CO2. Fifty grams (wet) of IRA-67 was employed to react with 200 mL 1 M of ammonium bicarbonate in a water-jacketed reactor at room temperature. The mixture was stirred as a stirring speed of 800 rpm. After reaction for 15 min, 2 mL of liquid was removed from the reactor for the analysis of released ammonia by titration as above, thus obtaining the ARE value the resin. Subsequently, the mixture was filtered, and the liquid was discarded. The resin remained

resin

regeneration capability (mg-ammonia/g-resin)

A-830 IRA-67 IRA-96 WA-30 A-7 A-21

38.8 47.3 18.4 12.4 8.5 6.8

also estimated according to eq 3. The results for the dependence of ammonia regeneration efficiency on ammonium bicarbonate concentrations are shown in Figure 2. The effect of the ratio of resin amount to ammonium bicarbonate volume on the ammonia regeneration efficiency of a resin was also studied by experiment. Under constant stirring at 800 rpm and a temperature of 25 °C, 50 g (wet) of

(5)

Method to Regenerate NH3 for the Capture of CO2

Energy & Fuels, Vol. 16, No. 4, 2002 907

or greater than 50 °C along with the collection of CO2 (eq 7)

2NH4HCO3(aq) + Ra(s) T Ra•H2CO3(s) + (NH4)2CO3(aq) (6) Ra•H2CO3(s) T Ra(s) + CO2(g) + H2O(aq)

Figure 5. Desorption curve for the regeneration of resin.

(7)

The enthalpy of the dissociation of the adsorbed CO2 from resin (eq 7) has not been reported in the literature. Judging from its low dissociation temperature, it is expected that the enthalpy of CO2 dissociation is small. However, a reasonable estimation of the enthalpy can be made. The enthalpy for the release of CO2 from the resin used would be slightly higher than that from ammonium bicarbonate, and less than that from ammonium carbonate. This is why the resin can extract CO2 from ammonium bicarbonate but not from ammonium carbonate. Therefore, the enthalpy of the dissociation of adsorbed CO2 from the resin should be between 6.5 kcal/mol (the enthalpy of eq 8) and 12.2 kcal/mol (the enthalpy of eq 9). A reasonable estimate of the enthalpy of the dissociation of the adsorbed CO2 from resin is 8 kcal/mol, which is about one-half of that of the MEA’s 16.5 kcal/mol. Therefore, a 50% energy saving is expected of the new method when compared to the MEA process.

2NH4HCO3(aq) T (NH4)2CO3(aq) + CO2(g) + H2O(aq) (8) Figure 6. Influence of ratio of resin to water volume on desorption of resin.

Figure 7. Sustainability of resin for regeneration of ammonia. in the reactor. To the reactor, 200 mL of water was added. The mixture was stirred constantly at 800 rpm, and heated to boiling. After boiling for 1 h, the resin was separated from the water and cooled to room temperature by using cold water. The recycled resin was used to repeat the operation 5 times as above, and to obtain the ARE data for each run. The results for each run are depicted in Figure 7.

Results and Discussion It was recently found that an anion-exchange resin (Ra) with a free base as a functional group, was able to adsorb H2CO3 from a NH4HCO3 solution resulting in the release of ammonia, which then combines with the remaining NH4HCO3 to form ammonium carbonate at room temperatures (eq 6). The resin could then be recycled by the use of water at temperatures equal to

(NH4)2CO3(aq) T NH3(aq) + CO2(g) + H2O(aq) (9) Regeneration of Ammonia by Resin. Due to the reaction between ammonia and CO2, ammonium bicarbonate is formed. If the ammonia scrubbing technique is to be applied, it is crucial to be able to regenerate ammonia from ammonium bicarbonate. Recently, it was discovered that a weakly basic anion-exchange resin was able to regenerate ammonia from ammonium bicarbonate. To demonstrate this capability of resin, an ammonia solution (pH ) 13.0) was first used to capture CO2 from simulated flue gas (8.5% CO2 with N2 as the balance) until it could not absorb any more CO2, i.e., the ammonia solution became saturated by CO2, as depicted in Figure 1. The saturated ammonia solution (pH ) 9.1), most of which should be ammonium bicarbonate, was then passed through a column containing a resin (A-830) at an ambient temperature of 25 °C. After the solution passed through the column with the resin, the pH of the solution was increased to 11.2, most of the solution should be ammonium carbonate. Subsequently, the resultant solution was recycled to absorb CO2 from the simulated flue gas. The absorption curve of CO2 by the recycled ammonia solution was also recorded and shown in Figure 1. It is evident that the recycled ammonia is able to absorb CO2, indicating the “dead” saturated ammonia solution became “alive” after passing through the column with the resin. The amount of CO2 absorbed by the solution was estimated by an integration method, and found to be 38% of that of CO2 absorbed by the original ammonia solution. The capability of the weakly basic anion-exchange of resins to regenerate ammonia from ammonium bicar-

908

Energy & Fuels, Vol. 16, No. 4, 2002

Huang et al.

bonate is attributed to the reaction of ammonium bicarbonate with the resin’s amine functional group. The basicity of the amine functional group induces the decomposition of ammonium bicarbonate into ammonia and carbonic acid. With the combination of carbonic acid and amine, the ammonia is released into the solution, resulting in an increase in pH (eq 10):

R-NH2 (or dNH, or tN) + NH4HCO3 T R-NH2 (or dNH, or tN)•H2CO3 + NH3 (10) Some of the released ammonia will react with the remaining ammonium bicarbonate to form ammonium carbonate, which results in the resin’s inability to completely regenerate ammonia. In other words, the resin cannot completely convert ammonium bicarbonate to ammonia due to the creation of ammonium carbonate as a product, as shown in eq 6. The inability of the resin to regenerate ammonia from ammonium carbonate was confirmed by the experimental facts that no reduction of carbonate ion concentration was observed when an ammonium carbonate solution was mixed with a resin. This also accounts for the fact that the amount of CO2 absorbed by the regenerated solution is only 38% of that by the original ammonia solution. Due to the nature of ammonium carbonate, the regenerated ammonium solution should contain ammonia and ammonium carbonate. Both ammonia and ammonium carbonate are ready to absorb CO2 (eqs 11 & 12):

CO2 + NH3 + H2O T NH4HCO3

(11)

CO2 + (NH4)2CO3 + H2O T 2NH4HCO3

(12)

Regarding the concern of the highly volatile nature of ammonia, a calculation was made to estimate the gas and aqueous compositions of ammonium carbonate under thermodynamic equilibrium conditions and based on the following reactions at 298 K. The results for the gas and aqueous components are shown in Table 2, along with calculation formulas and the parameters involved.

(NH4)2CO4(aq) T NH4+(aq) + NH4CO3-(aq)

(13)

NH4CO3-(aq) T NH4+(aq) + CO32-(aq)

(14)

NH4+(aq) + OH-(aq) T NH3(aq) + H2O(aq)

(15)

NH3(aq) T NH3(g)

(16)

CO32-(aq) + H+(aq) T HCO3-(aq)

(17)

HCO3-(aq) + H+(aq) T H2CO3(aq)

(18)

H2CO3(aq) T CO2(g)

(19)

The accuracy of the calculations was reflected by the agreement between the estimated values and the experimental results at ammonium carbonate concentrations of 4.41 M and temperatures at 298 K. According to the calculation, when pH ) 9.4, most of ammonia carbonate should have converted to ammonium bicarbonate and the equilibrium CO2 vapor pressure is

10.96%, therefore the solution can hardly absorb CO2 from flue gas containing 12% CO2. However, as the pH increases to 11.0, the equilibrium CO2 vapor pressure is 0.05%, and ammonium carbonate is the major species in the solution, thus the CO2 removal efficiency from a 12% flue gas will reach nearly 99.6%. As stated above, these results basically agree with those from experiments. From the data in Table 2, it can be seen that ammonium carbonate concentration in the absorbent significantly impacts the escape of ammonia and CO2 removal efficiency. At a given pH, with the increase of the concentration, the escape of ammonia increases, while the CO2 removal efficiency decreases. At pH ) 9.0, when total ammonium carbonate concentration is 0.1 M, ammonia equilibrium vapor pressure is 0.0013 atm, CO2 removal efficiency is 94.4%, assuming an initial 12% CO2 in flue gas. However, when total ammonium carbonate increases to 1.0 M, the ammonia equilibrium vapor pressure increases to 0.0125 atm, and the CO2 removal efficiency decreases to 0%. With the increase of pH, the influence of total ammonium carbonate concentration on CO2 removal efficiency will decrease, whereas its influence on ammonia escape will increase. At pH ) 11.0, when total ammonium carbonate concentration is 0.1 M, ammonia equilibrium vapor pressure is 0.0034 atm, and CO2 removal efficiency is 100% from an initial 12% CO2 in flue gas. When total ammonium carbonate concentration is 1.0 M, ammonia equilibrium vapor pressure is 0.0345 atm, and CO2 removal efficiency is 99.7%. Therefore, taking account of the problem from the escape of ammonia and the CO2 removal efficiency, a diluted solution will be better than a concentrated solution for the capture of CO2 in flue gas. However, using a solution that is too diluted will increase the energy consumption due to the need of handling a larger amount of solution. On the basis of the results in Table 2, a reasonable total ammonium carbonate solution concentration to use is 0.25 M. At this concentration, when pH ) 9.0, the CO2 removal efficiency reaches 86.1% with an ammonia equilibrium vapor pressure of 0.0031 atm, and when pH ) 11.0, the CO2 removal efficiency will achieve 100% with the ammonia equilibrium vapor pressure of 0.0086 atm. In reality, when absorption and desorption mass transfer kinetics are considered, the CO2 removal efficiency and the amount of ammonia escaped will be less than the results indicated above. Further experimentation found that other sorts of resin, besides A-830, were also able to regenerate ammonia from ammonium bicarbonate. The experiments involved mixing a resin with an ammonium bicarbonate solution in a water-jacketed reactor by stirring the mixtures at 25 °C. The pH of each ammonium bicarbonate solution was 8.2. However, the pH values of the solutions were increased to 9.8, 9.7, 9.4, 9.1, 9.0, and 8.9 after each solution reacted for 1 h with A-830, IRA-67, IRA-96, WA-30, A-7, and A-21, respectively. The ammonia regeneration capability of a resin was estimated by the regenerated ammonia amount divided by the amount of the employed resin on the dry weight basis. The results for the ammonia regeneration capability of various resins are shown in Table 1. These

Method to Regenerate NH3 for the Capture of CO2

Energy & Fuels, Vol. 16, No. 4, 2002 909

Table 2. Concentrations of NH3(g) and CO2(g) above an Aqueous Solution of (NH4)2CO3 at Thermodynamic Equilibriuma,b

[(NH4)2CO3]hot

pH

[H+]

0.100

9.00 9.20 9.40 9.60 9.80 10.0 10.2 10.4 10.6 10.8 11.0 9.00 9.20 9.40 9.60 9.80 10.0 10.2 10.4 10.6 10.8 11.0 9.00 9.20 9.40 9.60 9.80 10.0 10.2 10.4 10.6 10.8 11.0 9.00 9.20 9.40 9.60 9.80 10.0 10.2 10.4 10.6 10.8 11.0 9.00 9.20 9.40 9.60 9.80 10.0 10.2 10.4 10.6 10.8 11.0

1.00E-09 6.31E-10 3.98E-10 2.51E-10 1.58E-10 1.00E-10 6.31E-11 3.98E-11 2.51E-11 1.58E-11 1.00E-11 1.00E-09 6.31E-10 3.98E-10 2.51E-10 1.58E-10 1.00E-10 6.31E-11 3.98E-11 2.51E-11 1.58E-11 1.00E-11 1.00E-09 6.31E-10 3.98E-10 2.51E-10 1.58E-10 1.00E-10 6.31E-11 3.98E-11 2.51E-11 1.58E-11 1.00E-11 1.00E-09 6.31E-10 3.98E-10 2.51E-10 1.58E-10 1.00E-10 6.31E-11 3.98E-11 2.51E-11 1.58E-11 1.00E-11 1.00E-09 6.31E-10 3.98E-10 2.51E-10 1.58E-10 1.00E-10 6.31E-11 3.98E-11 2.51E-11 1.58E-11 1.00E-11

0.250

0.500

1.00

4.41

CO2 removal efficiency (%)

pCO2(g) (atm)

pNH3(g) (atm)

[NH3]aq (M)

[NH3]aq (M)

[NH4+]aq (M)

[HCO32]aq (M)

[H2CO3]aq (M)

94.4 96.6 97.9 98.8 99.3 99.6 99.8 99.9 99.9 100 100 86.1 91.5 94.8 96.9 98.2 99.0 99.5 99.7 99.9 99.9 100 72.2 82.9 89.6 93.8 96.5 98.0 99.0 99.5 99.7 99.9 99.9 0.0 14.6 48.2 69.2 82.3 90.1 94.8 97.4 98.7 99.4 99.7 0.0 0.0 8.6 45.8 68.7 82.6 90.8 95.4 97.8 99.0 99.6

0.0067 0.0041 0.0025 0.0015 0.0009 0.0005 0.0003 0.0001 0.0001 0.0000 0.0000 0.0167 0.0103 0.0062 0.0037 0.0021 0.0012 0.0006 0.0003 0.0002 0.0001 0.0000 0.0333 0.0205 0.0124 0.0074 0.0043 0.0024 0.0013 0.0006 0.0003 0.0001 0.0001 0.1667 0.1025 0.0621 0.0369 0.0213 0.0118 0.0063 0.0032 0.0015 0.0007 0.0003 0.2940 0.1808 0.1096 0.0651 0.0376 0.0209 0.0111 0.0056 0.0027 0.0012 0.0005

0.0013 0.0016 0.0020 0.0024 0.0027 0.0030 0.0032 0.0033 0.0034 0.0034 0.0034 0.0031 0.0041 0.0051 0.0060 0.0068 0.0074 0.0079 0.0082 0.0084 0.0085 0.0086 0.0063 0.0082 0.0102 0.0121 0.0136 0.0149 0.0158 0.0164 0.0168 0.0171 0.0172 0.0125 0.0164 0.0204 0.0242 0.0273 0.0297 0.0315 0.0327 0.0336 0.0341 0.0345 0.0553 0.0725 0.0901 0.1066 0.1204 0.1311 0.1390 0.1444 0.1480 0.1504 0.1520

0.0714 0.0936 0.1165 0.1377 0.1556 0.1695 0.1796 0.1866 0.1913 0.1945 0.1965 0.1786 0.2341 0.2913 0.3443 0.3890 0.4237 0.4490 0.4666 0.4784 0.4861 0.4912 0.3571 0.4682 0.5825 0.6886 0.7780 0.8475 0.8980 0.9331 0.9567 0.9723 0.9823 0.7143 0.9365 1.1651 1.3773 1.5561 1.6949 1.7960 1.8663 1.9135 1.9445 1.9646 3.1500 4.1297 5.1381 6.0738 6.8623 7.4746 7.9205 8.2302 8.4385 8.5754 8.6640

0.1286 0.1064 0.0835 0.623 0.0444 0.00305 0.0204 0.0134 0.0087 0.0055 0.0035 0.3214 0.2659 0.2087 0.1557 0.1110 0.0763 0.0510 0.0334 0.0216 0.0139 0.0088 0.6429 0.5318 0.4175 0.3114 0.2220 0.1525 0.1020 0.0669 0.0433 0.0277 0.0177 1.2857 1.0635 0.8349 0.6227 0.4439 0.3051 0.2040 0.1337 0.0865 0.0555 0.0354 5.6700 4.6903 3.6819 2.7462 1.9577 1.3454 0.8995 0.5898 0.3815 0.2446 0.1560

0.0046 0.0071 0.0108 0.0160 0.0232 0.00324 0.0432 0.0547 0.0656 0.0752 0.0828 0.0114 0.0176 0.0269 0.0401 0.0581 0.0811 0.1080 0.1367 0.1641 0.1879 0.2069 0.0228 0.0353 0.0538 0.0802 0.1162 0.1621 0.2160 0.2733 0.3282 0.3759 0.4138 0.1142 0.1765 0.2688 0.4009 0.5810 0.8107 1.0801 1.3665 1.6411 1.8794 2.0690 0.2015 0.3113 0.4741 0.7072 1.0248 1.4300 1.9052 2.4105 2.8950 3.3153 3.6496

0.0952 0.0928 0.0892 0.0839 0.0767 0.0676 0.0568 0.0453 0.0344 0.0248 0.0172 0.2380 0.2320 0.2229 0.2098 0.1918 0.1689 0.1420 0.1133 0.0859 0.0621 0.0431 0.4760 0.4640 0.4458 0.4196 0.3837 0.3378 0.2839 0.2267 0.1718 0.1241 0.0862 2.3801 2.3200 2.2291 2.0979 1.9183 1.6889 1.4197 1.1334 0.8588 0.6206 0.4310 4.1985 4.0925 3.9322 3.7006 3.3839 2.9793 2.5044 1.9993 1.5150 1.0947 0.7603

0.0002 0.0001 0.0001 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0006 0.0003 0.0002 0.0001 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0011 0.0007 0.0004 0.0003 0.0001 0.0001 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0035 0.0021 0.0013 0.0007 0.0004 0.0002 0.0001 0.0001 0.0000 0.0000 0.0200 0.0061 0.0037 0.0022 0.0013 0.0007 0.0004 0.0002 0.0001 0.0000 0.0000

a Calculation formula: P + CO2(g) (atm) ) [H2CO3]aq/HCO2 PNH3(g) (atm) ) [H3]aq/HNH3 [NH3]aq (M) ) (2*[(NH4)2CO3]tot)/(1 + (Kb*(H ])/Kw) [NH4+]aq ) (2*[(NH4)2CO3]tot) - [NH3]aq [HCO3-]aq ) K1*[H2CO3]aq/[H+] [CO32-]aq ) K1*K1*[H2CO3]aq/([H+]*[H+]) b Parameters10,11: HCO2 ) 0.0340 M atm-1 HNH3 ) 57.0 M atm-1 Kb ) 1.80E-05 M K1 ) 4.20E-07 M K2 ) 4.80E-11 M Kw ) 1.00E-14 M2

results show that all types of weakly basic anionexchange resin have, to an extent, the ability to regenerate ammonia from ammonium bicarbonate, while A-830 and IRA-67 have capabilities in regenerating ammonia much greater than IRA-96, A-7, A-21, and WA-30. The reason for these results is probably due to the different composition of the amine functional group in various resins. For A-380 and IRA-67, the content of primary and secondary amines is greater than that for other resins, resulting in more ammonia produced from ammonium bicarbonate.

The concentration of ammonium bicarbonate solution affects the regeneration efficiency of ammonia. As shown in Figure 2, when the ratio of resin (IRA-67) to the volume of ammonium bicarbonate is kept unchanged, the ammonia regeneration efficiency increases with the concentration of ammonium bicarbonate increasing. When the concentration of ammonium bicarbonate is 0.1 M, the ammonia regeneration efficiency is only 2.4 mg of ammonia per gram of resin. However, when the concentration of ammonium bicarbonate increases to 1.5 M, the ammonia regeneration efficiency

910

Energy & Fuels, Vol. 16, No. 4, 2002

increases to 26.1 mg of ammonia per gram of resin. Obviously, more CO2 will be captured by a higher concentration of regenerated ammonia solution than by a lower concentration of regenerated ammonia solution at the same solution volume. In addition to the concentration, the ratio of resin amount to the volume of ammonium bicarbonate solution also influences the regeneration efficiency of ammonia. When the concentration of ammonium bicarbonate solution is kept the same, the regeneration efficiency of ammonia decreases with an increase in the ratio of resin (IRA-67) amount to the volume of ammonium bicarbonate solution, as shown in Figure 3. When the ratio of resin amount to the volume of ammonium bicarbonate is 0.07 (g/mL), the ammonia regeneration efficiency is 60 mg/g. However, when the ratio of resin to the volume of ammonium bicarbonate solution increases to 0.28 (g/mL), the regeneration efficiency of ammonia becomes 30 mg/g, which is a half of that with the ratio of 0.07. Moreover, the reaction of resin with ammonium bicarbonate is fairly fast. Figure 4 shows the typical kinetic curves of ammonia regeneration by several different types of resin. It is apparent that each reaction between resin and ammonium reaches its equilibrium within 15 min. The kinetic characters with the reactions are to be investigated in more details. Recycling of Resin and Collection of Carbon Dioxide. It is essential to find a cost-effective way to recycle the resin, which had been used to adsorb carbonic acid in the regeneration of ammonia. It was found that water possesses the ability to regenerate the resin by heating the used resin to temperatures at or above 50 °C. However, the higher the temperature, the more effective the regeneration is. The higher the temperature, the better the regeneration efficiency will be. Along with regenerating the resin, CO2 can also be collected after the water vapor is separated by condensation. To demonstrate the feasibility of this process, we studied the release of CO2 from resin (IRA-67) at different temperatures. The results are shown in Figure 5. Only 13% of the total amount of CO2 absorbed was released at temperatures lower than 50 °C. As the temperature increased, the amount of CO2 released increased as well. When the temperature was kept

Huang et al.

constant at 80 °C, the desorption reached its equilibrium after 1 h. The emitted CO2 was 90% of the total amount of CO2 absorbed. When the temperature increased to 100 °C, the desorption reached its equilibrium in 15 min. The effect the ratio of resin to the volume of water had on the desorption of CO2 from resin was investigated. As shown in Figure 6, the emitted CO2 from each desorption experiment remained unchanged when the ratio of the amount of resin to the volume of water was raised from 0.08 to 0.25 g/mL. Thus, to save energy from being used to heat extra amounts of water, the ratio of resin to the volume of water should be as large as possible. The Sustainability of Resin. The sustainability of a resin for the regeneration of ammonia was also demonstrated by the continuous running of several adsorption-desorption cycles. Figure 7 depicts the ammonia regeneration capability of the resin at each operation. The results indicate that the ammonia regeneration capability of the recycled resin can be repeatedly kept at 87% of the original resin’s capability. Conclusion The current study demonstrates that it is feasible for a resin with a free amine functional group to regenerate ammonia from ammonium bicarbonate at ambient temperatures. It was also found that the regeneration of the resin by heating to 50 °C or higher with the collection of CO2 was possible. The resin proved to be sustainable as well, so as to be regenerated for repeated uses in the process of regenerating ammonia for the capture of CO2. The results of this study indicate that the thermal energy requirement is approximately 50% less in a dual alkali system using ammonia to absorb CO2 and anion-exchange resins to regenerate ammonia for reuse than using amine to absorb CO2 and steam stripping to dissociate carbamates. Acknowledgment. This work was supported by the Assistant Secretary for Fossil Energy, U.S. Department of Energy, under Contract DE-AC03-76SF0098 through the National Energy Technology Laboratory. EF010270I