Carbon Dioxide Capture Using Ionic Liquid 1-Butyl-3

Sep 7, 2010 - The need for affordable post-combustion (PC) CO2 capture processes for existing coal-fired power plants is of particular interest in the...
2 downloads 13 Views 1MB Size
Download PDF
Energy Fuels 2010, 24, 5781–5789 Published on Web 09/07/2010

: DOI:10.1021/ef100868a

Carbon Dioxide Capture Using Ionic Liquid 1-Butyl-3-methylimidazolium Acetate Mark B. Shiflett,*,† David W. Drew,‡ Robert A. Cantini,‡ and A. Yokozeki§ †

DuPont Central Research and Development, Experimental Station, Wilmington, Delaware 19880, ‡DuPont Engineering, 1007 Market Street, Wilmington, Delaware 19898, and §109-C Congressional Drive, Wilmington, Delaware 19807 Received July 8, 2010. Revised Manuscript Received August 23, 2010

Carbon dioxide (CO2) capture using aqueous amine scrubbing is currently considered the most feasible option for separating CO2 from post-combustion flue gas. Using simple absorption and stripping configurations, monoethanolamine has been commercially demonstrated to effectively scrub CO2 from post-combustion flue gas. However, the current capital and operating costs are high and do not meet the target of the Department of Energy to remove 90% of CO2 from post-combustion flue gas with no more than a 35% increase in the cost of electricity. The evaluation of advanced absorbents, adsorbents, and membranes is under way to find the most energy-efficient CO2-capture technology. We have modeled an ionic liquid that can reduce the energy losses by 16% compared to a commercial monoethanolamine process. The choice of the ionic liquid, 1-butyl-3-methylimidazolium acetate, has not been optimized but was chosen based on chemical absorption behavior and the desire to understand performance. Engineering design estimates indicate that the investment for the ionic liquid process will be 11% lower than the aminebased process and provide a 12% reduction in equipment footprint. A parametric study examined four improvements in the ionic liquid technology, which may reduce even further the energy and cost required for CO2 capture. capture technologies are critical to ensure that the DOE targets and timing can be met to reduce future CO2 emissions while minimizing the impact of increasing electricity prices on consumers. Several of the approaches that have been proposed to remove CO2 from PC flue gases on a large scale include cryogenic distillation, purification with membranes, absorption with liquids, and adsorption with solids.5-8 While the concentration of CO2 in PC flue gas varies with fuel source, boiler age and design, and generating load, most modern coalfired boilers produce a flue gas that contains approximately 12-14% CO2 by volume at about atmospheric pressure. Efficient capture of CO2 at this low concentration and pressure will require chemical sorption. A variety of amines have been studied, which chemically react with CO2, such as monoethanolamine (MEA), proprietary amines (KS-1), and mixed amines.1 Additives, such as piperazine, have also been studied to improve the reaction kinetics.9-11 Absorptionbased processes are often used as the benchmark, and more

1. Introduction Development of economically viable carbon dioxide (CO2) capture processes is becoming increasingly important as concerns over greenhouse gas emissions continue to receive worldwide attention. The need for affordable post-combustion (PC) CO2 capture processes for existing coal-fired power plants is of particular interest in the United States because these plants generate approximately 50% of the electricity for the nation and produce about 30% of the CO2 emissions.1 Currently, amine-based scrubbing is the most feasible technology for PC CO2 capture that is commercially deployable at required scales.2 However, the current energy requirement of an amine-based scrubbing system does not meet the Department of Energy (DOE) PC targets defined by the National Energy Technology Laboratory (NETL).3 The DOE states that PC CO2 capture processes should be capable of removing 90% of the CO2 from the flue gas with no more than a 35% increase in the cost of electricity (COE).3 Current projections of the best state-of-the-art amine-based scrubbing systems indicate parasitic power requirements of 22-30% of the power plant output1 and a projected increase in COE of 81% for a supercritical pulverized coal plant.4 Thus, improvements in the energy efficiency and economics of PC CO2

(5) Aaron, D.; Tsouris, C. Separation of CO2 from flue gas: A review. Sep. Sci. Technol. 2005, 40, 321–348. (6) Choi, S.; Drese, J. H.; Jones, C. W. Absorbent materials for carbon dioxide capture from large anthropogenic point sources. ChemSusChem 2009, 2, 796–854. (7) Ho, M. T.; Allinson, G. W.; Wiley, D. E. Reducing the cost of CO2 capture from flue gases using membrane technology. Ind. Eng. Chem. Res. 2008, 47, 1562–1568. (8) Britt, D.; Furukawa, H.; Wang, B.; Glover, T. G.; Yaghi, O. M. Highly efficient separation of carbon dioxide by a metal-organic framework replete with open metal sites. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 20637–20640. (9) Dugas, R.; Rochelle, G. Absorption and desorption rates of carbon dioxide with monoethanolamine and piperazine. Energy Procedia 2009, 1, 1163–1169. (10) Closmann, F.; Nguyen, T.; Rochelle, G. T. MDEA/Piperazine as a solvent for CO2 capture. Energy Procedia 2009, 1, 1351–1357. (11) Freeman, S. A.; Dugas, R.; Van Wagener, D.; Nguyen, T.; Rochelle, G. T. Carbon dioxide capture with concentrated, aqueous piperazine. Energy Procedia 2009, 1, 1489–1496.

*To whom correspondence should be addressed. E-mail: mark.b. [email protected]. (1) Rochelle, G. T. Amine scrubbing for CO2 capture. Science 2009, 325, 1652–1654. (2) Plaza, J. M.; Wagner, D. W.; Rochelle, G. T. Modeling CO2 capture with aqueous monoethanolamine. Energy Procedia 2009, 1, 1171–1178. (3) National Energy Technology Laboratory (NETL). Bench-Scale and Slipstream Development and Testing of Post-combustion Carbon Dioxide Capture and Separation Technology for Application to Existing Coal-Fired Power Plants; Funding Opportunity Number DE-FOA0000131. (4) National Energy Technology Laboratory (NETL). Cost and Performance Baseline For Fossil Energy Plants, 2007; http://www.netl. doe.gov/energy-analyses/baseline_studies.html. r 2010 American Chemical Society

5781

pubs.acs.org/EF

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al. 16-37

than 30 power plants currently remove CO2 (at small scale) from flue gas using aqueous amines, such as MEA.1,12 The primary thermodynamic limitation with amine-based scrubbers is the energy required to decompose the carbamate at high temperatures (383-403 K) during regeneration.13-15 The focus of this paper is to compare the energy requirement and economic investment of a commercial MEA PC CO2 capture facility with a new process designed to use ionic liquids (ILs). Research on ILs for CO2 capture has been

government underway for several years in academia, laboratories,38-43 and private industry;44-46 however, no process simulations of an IL-based scrubber compared to a MEA-based process have appeared in the literature. The need for such a comparison motivated us to publish the first study. In our previous work, we measured and modeled the global phase behavior [vapor-liquid equilibria (VLE) and vaporliquid-liquid equilibria (VLLE)] of CO2 in ILs using a modified Redlich-Kwong (RK) equation of state (EOS).47-52 Knowledge of solvent phase behavior is important in determining the attractiveness of using an IL in new applications, such as CO2 capture. In this report, the choice of the IL, 1-butyl-3-methylimidazolium acetate [bmim][Ac], was not optimized but was selected on the basis of chemical absorption behavior.50,51 Figure 1 provides the chemical structure for [bmim][Ac]. The binary system (CO2 þ [bmim][Ac]) has a highly unusual phase behavior, as shown in Figure 2. At low CO2 concentrations (less than ca. 20 mol %), the binary mixtures have hardly any vapor pressure, reflecting a strong attractive interaction (chemical absorption) between CO2 and [bmim][Ac], while at high CO2 concentrations (above ca. 70 mol %),

(12) Ramezan, M.; Skone, T. J.; Nsakala, N.; Liljedahl, G. N. Carbon Dioxide Capture from Existing Coal-Fired Power Plants; DOE/NETL401/110907, Nov 2007. (13) Fisher, K. S.; Searcy, K.; Rochelle, G. T.; Schubert, C. Advanced Amine Solvent Formulations and Process Integration for Near-Term CO2 Capture Success; DE-FG02-06ER84625, June 2007. (14) House, K. Z.; Harvey, C. F.; Aziz, M. J.; Schrag, D. P. The energy penalty of post-combustion CO2 capture and storage and its implications for retrofitting the U.S. installed base. Energy Environ. Sci. 2009, 2, 193–205. (15) Schach, M.-O.; Schneider, R.; Schramm, H.; Repke, J.-U. Technoeconomic analysis of postcombustion processes for the capture of carbon dioxide from power plant flue gas. Ind. Eng. Chem. Res. 2010, 49, 2363– 2370. (16) Brennecke, J. F.; Maginn, E. J. Purification of gas with liquid ionic compounds. U.S. Patent 6,579,343, 2003. (17) Maginn, E. J. Design and Evaluation of Ionic Liquids as Novel CO2 Absorbents, Quarterly Technical Report to Department of Energy (DOE); Sept 30, 2006. (18) Brennecke, J. F.; Anthony, J. L.; Maginn, E. J. Gas Solubilities in Ionic Liquids, Ionic Liquids in Synthesis; Wiley-VCH: Weinheim, Germany, 2003; pp 81-92. (19) Anderson, J. L.; Anthony, J. L.; Brennecke, J. F.; Maginn, E. J. Gas Solubilities in Ionic Liquids, Ionic Liquids in Synthesis; Wiley-VCH: Weinheim, Germany, 2008; pp 103-129. (20) Aki, S. N. V. K.; Mellein, B. R.; Saurer, E. M.; Brennecke, J. F. High-pressure phase behavior of carbon dioxide with imidazoliumbased ionic liquids. J. Phys. Chem. B 2004, 108, 20355–20365. (21) Muldoon, M. J.; Aki, S. N. V. K.; Anderson, J. L.; Dixon, J. K.; Brennecke, J. F. Improving carbon dioxide solubility in ionic liquids. J. Phys. Chem. B 2007, 111, 9001–9009.  Tuma, D.; Maurer, G. (22) Kumezan, J.; Perez-Salado Kamps, A.; Solubility of CO2 in the ionic liquid [hmim][Tf2N]. J. Chem. Thermodyn. 2006, 38, 1396–1401.  Tuma, D.; Maurer, G. (23) Kumezan, J.; Perez-Salado Kamps, A.; Solubility of CO2 in the ionic liquids [bmim][CH3SO4] and [bmim][PF6]. J. Chem. Eng. Data 2006, 51, 1802–1807.  Maurer, G. (24) Kumezan, J.; Tuma, D.; Perez-Salado Kamps, A.; Solubility of the single gases carbon dioxide and hydrogen in the ionic liquid [bmpy][Tf2N]. J. Chem. Eng. Data 2010, 55, 165–172. (25) Costa Gomes, M. F. Low-pressure solubility and thermodynamics of solvation of carbon dioxide, ethane, and hydrogen in 1-hexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)amide between temperatures of 283 and 343 K. J. Chem. Eng. Data 2007, 52, 472–475. (26) Kim, Y. S.; Choi, W. Y.; Jang, J. H.; Yoo, K.-P.; Lee, C. S. Solubility measurement and prediction of carbon dioxide in ionic liquids. Fluid Phase Equilib. 2005, 228-229, 439–445. (27) Kim, Y. S.; Jang, J. H.; Lim, B. D.; Kang, J. W.; Lee, C. S. Solubility of mixed gases containing carbon dioxide in ionic liquids: Measurements and predictions. Fluid Phase Equilib. 2007, 256, 70–74. (28) Lee, B.-C.; Outcalt, S. L. Solubilities of gases in ionic liquid 1-nbutyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide. J. Chem. Eng. Data 2006, 51, 892–897. (29) Chinn, D.; Vu, D. Q.; Driver, M. S.; Boudreau, L. C. CO2 removal from gas using ionic liquid absorbents, U.S. Patents 20060251558A1 and 20050129598A1. (30) Zhang, X.; Liu, Z.; Wang, W. Screening of ionic liquids to capture CO2 by COSMO-RS and experiments. AIChE J. 2008, 54, 2717–2728. (31) Carvalho, P. J.; Coutinho, J. A. P. On the non-ideality of CO2 solutions in ionic liquids and other low volatile solvents. J. Phys. Chem. Lett. 2010, 1, 774–780. (32) Energy Systems Research Unit (ESRU). The Capture and Sequestration of Carbon Dioxide, 2008; http://www.esru.strath.ac.uk. (33) Blasucci, V.; Dilek, C.; Huttenhower, H.; John, E.; LlopisMestre, V.; Pollet, P.; Eckert, C. A.; Liotta, C. L. One component, switchable, neutral to ionic liquid solvents derived from siloxylated amines. Chem. Commun. 2009, 116–119. (34) Eckert, C. A.; Liotta, C. L. Reversible Ionic Liquids as DoubleAction Solvents for Efficient CO2 Capture, Quarterly Progress Report to Department of Energy (DOE); Jan 27, 2009.

(35) Camper, D.; Bara, J. E.; Gin, D. L.; Noble, R. D. Roomtemperature ionic liquid-amine solutions: Tunable solvents for efficient and reversible capture of CO2. Ind. Eng. Chem. Res. 2008, 47, 8496–8498. (36) Bara, J. E.; Carlisle, T. K.; Gabriel, C. J.; Camper, D.; Finotello, A.; Gin, D. L.; Noble, R. D. Guide to CO2 separations in imidazoliumbased room-temperature ionic liquids. Ind. Eng. Chem. Res. 2009, 48, 2739–2751. (37) Bara, J. E.; Camper, D. E.; Gin, D. L.; Noble, R. D. Roomtemperature ionic liquids and composite materials: Platform technologies for CO2 capture. Acc. Chem. Res. 2010, 43, 152–159. (38) Magee, J. W.; Frenkel, M. National Institute of Standards and Technology (NIST) Ionic Liquids Database; International Union of Pure and Applied Chemistry (IUPAC) Project 2003-020-2-100, 2005; http:// ilthermo.boulder.nist.gov/ILThermo/mainmenu.uix. (39) Luebke, D. R.; Ilconich, J. B.; Myers, C. R.; Pennline, H. W. Carbon Dioxide Selective Supported Ionic Liquid Membranes: The Effect of Contaminants; DOE/NETL-IR-2008-115, 2008. (40) Luebke, D. R.; Ilconich, J. B.; Myers, C. R.; Pennline, H. W. Carbon Dioxide Seperation with Supported Ionic Liquid Membranes; DOE/NETL-IR-2007-124, 2007. (41) Luebke, D. R.; Ilconich, J. B.; Myers, C. R.; Pennline, H. W. Carbon Dioxide Separation with Supported Ionic Liquid Membranes. DOE/NETL-IR-2006-108, 2006. (42) Pennline, H. W.; Luebke, D. R.; Jones, K. L.; Myers, C. R.; Morsi, B. I.; Heintz, Y. J.; Ilconich, J. B. Progress in carbon dioxide capture and separation research for gasification-based power generation point sources. Fuel Process. Technol. 2008, 89, 897–907. (43) Heintz, Y. J.; Sehabiague, L.; Morsi, B. I.; Jones, K. L.; Luebke, D. R.; Pennline, H. W. Hydrogen sulfide and carbon dioxide removal from dry fuel gas streams using an ionic liquid as a physical solvent. Energy Fuels 2009, 23, 4822–4830. (44) BASF Chemical Company. http://www.basionics.com/en/ ionic-liquids/. (45) Evonik Chemical Company. http://www.evonik.com. (46) Merck Chemical Company. http://www.merck-chemicals.de/. (47) Shiflett, M. B.; Yokozeki, A. Solubilities and diffusivities of carbon dioxide in ionic liquids: [bmim][PF6] and [bmim][BF4]. Ind. Eng. Chem. Res. 2005, 44, 4453–4464. (48) Shiflett, M. B.; Yokozeki, A. Solubility of CO2 in room-temperature ionic liquid [hmim][Tf2N]. J. Phys. Chem. B 2007, 111, 2070– 2074. (49) Yokozeki, A.; Shiflett, M. B. Hydrogen purification using roomtemperature ionic liquids. Appl. Energy 2007, 84, 351–361. (50) Shiflett, M. B.; Kasprzak, D. J.; Junk, C. P.; Yokozeki, A. Phase behavior of carbon dioxide þ [bmim][Ac] mixtures. J. Chem. Thermodyn. 2008, 40, 25–31. (51) Yokozeki, A.; Shiflett, M. B.; Junk, C. P.; Grieco, L. M.; Foo, T. Physical and chemical absorptions of carbon dioxide in room-temperature ionic liquids. J. Phys. Chem. B 2008, 112, 16654–16663. (52) Shiflett, M. B.; Yokozeki, A. Phase behavior of carbon dioxide in ionic liquids: [emim][acetate], [emim][trifluoroacetate], and [emim][acetate] þ [emim][trifluoroacetate] mixtures. J. Chem. Eng. Data 2009, 54, 108–114.

5782

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al.

Figure 1. Chemical structure of [bmim][Ac].

Figure 3. IR spectra of [bmim][Ac] (A, - - -) and (CO2 þ [bmim][Ac]) (B, ;).50.

323 K, these new peaks at 1666, 1508, 1323, and 791 cm-1 were significantly diminished, and at 352 K, the new peaks were virtually absent. This is a significantly lower temperature range than required for amine-based scrubbers (383-403 K) and provided the first clue that IL-based CO2 capture processes may be more energy-efficient. Figure 2. (P, T, x) phase diagram of the (CO2 þ [bmim][Ac]) system. (; and 3 3 3 ) Calculated using the modified RK EOS model. (b) Measured vapor-liquid equilibrium data.50 (9 and - - -) Measured vapor-liquid-liquid equilibrium data and tie lines.50

2. Process Simulation Simulations have been performed to compare the MEA- and IL-based processes. The MEA simulation is based on a CO2 capture process at a coal-burning power plant (180 MWe) in the northeastern United States. The plant burns bituminous coal, and the combustion technology is based on a circulating fluidized bed. To control SO2 emissions, limestone is co-fed into the fluidized bed of the boiler. A selective non-catalytic reduction system provides supplemental control of nitrogen oxides (NOx). A bag house controls particulate emissions in the boiler flue gas. A slip stream of the flue gas exiting the bag house is diverted from the stack and pressurized using two blower fans (1.22 MPa) to feed the CO2 capture plant. The flue gas enters a desulfurization system (two-stage NaOH scrubber) to remove additional SO2 and is cooled with a heat exchanger before entering the absorption column. A schematic of the simplified CO2 capture process flow diagram is shown in Figure 4. Not included in the figure or the simulation were the blowers, which are needed to overcome the pressure drop in the capture process, the desulfurization system, the flue gas heat exchanger, or the compressor, which is required for pressurizing and liquefying CO2 for storage. These were assumed to be similar for both the MEA and IL processes. The MEA-based process was sized to capture approximately 47 000 tons (metric tons) of CO2 per year or about 3-4% of the annual CO2 output from the power plant. The process consists of an absorption column for separating CO2 from the flue gas using an aqueous MEA solution (15-16 mass % MEA). MEA is regenerated by stripping with steam to produce CO2. CO2 is compressed, cooled, purified, and stored on site, where it can be shipped by truck and purchased for sale. The stripping column includes a reboiler, condenser, and reflux tank. Absorber and stripper pumps circulate the rich and lean MEA solution between the columns, and a process heat exchanger reduces energy load. Two additional heat exchangers heat and cool the rich and lean MEA solution at the inlet of the stripping and absorption columns, respectively. Table 1 provides the size of the absorber and stripper. Aspen Plus simulator (version 13.1) of Aspen Technology, Inc.55 was used to model the MEA process, and the input file is provided in the Supporting Information. The absorption and

the binary solutions show liquid-liquid separations (or immiscible VLLE).50 Such a highly asymmetric phase behavior with respect to concentrations is extremely rare53 and can provide a unique means of chemically capturing CO2. The modified RK EOS model successfully correlated the unusual VLE data and even predicted the VLLE behavior at high CO2 concentrations.50 The excess thermodynamic functions calculated with the EOS are large and negative, except for at high CO2 concentrations, and indicate the possibility of intermolecular complex formation or chemical reaction.50 In our previous report, we also measured the IR spectra using Fourier transform infrared (FTIR) and attenuated total reflectance infrared (ATR-IR) for samples of [bmim][Ac] and CO2 þ [bmim][Ac] to gain a better understanding of the sorption mechanism.50,54 In the 900-2000 cm-1 region, FTIR identified new peaks at 1668, 1509, and 1324 cm-1 for the CO2-containing IL. The new peaks suggest the presence of a carboxylate salt (COO-). In the 600-2000 cm-1 region, the ATR-IR identified new peaks at 1666, 1508, 1323, and 791 cm-1, as shown in Figure 3. These peaks are consistent with those measured by FTIR, except for the additional peak at 791 cm-1. The 1173 cm-1 peak belongs to the [bmim] cation and can serve as a reference peak because it has appeared in many imidazolium-based ILs with different anions from our previous work.51 The acetate carboxylate (CdO and C-O combined) peaks are at 1578 and 1379 cm-1. An interesting similarity exists with the identification of peaks at 1666, 1323, and 791 cm-1 in [bmim][Ac] containing dissolved CO2 and with oxalate salts (O2C-CO2)22-, which have peaks near 1620, 1320, and 770 cm-1.50 Upon heating the sample to (53) Rowlinson, J. S.; Swinton, F. L. Liquids and Liquid Mixtures; Butterworth: London, U.K., 1982. (54) Yokozeki, A.; Kasprzak, D. J.; Shiflett, M. B. Thermal effect on C-H stretching vibrations of the imidazolium ring in ionic liquids. Phys. Chem. Chem. Phys. 2007, 9, 5108–5026.

(55) Aspen Technology, Inc. ASPEN Plus Simulator, Version 13.1; Aspen Technology, Inc.: Cambridge, MA, 2005.

5783

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al.

Figure 4. Simplified MEA process flow diagram. Table 1. Equipment Size for MEA and IL

absorber diameter height stripper diameter height flash tank diameter height

Table 2. Operating Conditions for MEA and IL

MEA (m)

IL (m)

2.9 31.7

1.8 23.2

MEA

IL

T (K) P (MPa) m (kg/s) T (K) P (MPa) m (kg/s) absorber flue gas inlet vent gas outlet absorbent inlet absorbent outlet stripper CO2 outlet absorbent inlet absorbent outlet flash tank CO2 outlet absorbent inlet absorbent outlet

2.6 25.5 1.0 0.9

stripping columns were simulated using Aspen Plus Radfrac. The CO2 amine solution physical properties were estimated using the Kent-Eisenberg electrolyte method as implemented by Aspen Technology.56 This method provides VLE and enthalpy values. The equipment and operating conditions were chosen to match the commercial MEA process. The flue gas enters the absorption column at T = 312 K and P = 0.112 MPa, as shown in Table 2, and contains approximately 78% N2, 13% CO2, and 9% H2O, by volume. The CO2 concentration is reduced to less than 1.0 vol % at the exit of the absorber. The aqueous amine solution (CO2-lean solvent) enters the top of the column and flows countercurrent to the flue gas, separating CO2 from the other gases. The CO2-rich solution exits the bottom of the absorber and is pumped through a process heat exchanger to the top of the stripping column.

312 313 308 328

0.112 0.101 0.136 0.112

8.7 7.1 43.8 45.4

329 361 389

0.143 0.239 0.164

1.6 45.4 43.8

312 280 273 296

0.791 0.618 0.791 0.618

8.7 7.1 48.2 49.8

344 344 344

0.108 0.446 0.108

1.6 49.8 48.2

The inlet and outlet temperatures for the CO2-rich solution in the process heat exchanger are 328 and 361 K, respectively. The MEA-absorption column was modeled using 25 theoretical stages. The treated gas is vented to the atmosphere from the top of the column after passing through a water scrubber to remove traces of MEA (not shown in Figure 4). The MEA portion of the absorption column is packed with ∼19 mm polypropylene packing, and the wash-water section at the top of the column is packed with ∼12.7 mm polypropylene packing. In the stripping column, the CO2-rich solvent is regenerated by heating the reboiler with steam. The CO2-lean solvent exits the bottom of the stripper and is pumped back through the process

(56) Kent, R. L.; Eisenberg, B. Better data for amine treating. Hydrocarbon Process. 1976, 55, 87–92.

5784

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al.

Figure 5. Simplified [bmim][Ac] process flow diagram.

The regeneration process for the IL-based system is very different from the MEA-based scrubber. In this case, the IL is regenerated using a flash technique, where the pressure is reduced and the solvent is heated. The flash process was simulated using the Aspen Plus reactor model RCSTR. Because of the difference in pressure between the absorption column and flash tank, a pump is no longer required, as shown in Figure 5. The IL solvent exits the absorption column and enters the process heat exchanger. Next, the IL passes through a flash preheater and enters the flash tank. The flash tank is essentially a simple single-stage stripper, where the rich solvent is regenerated by heating with steam. The CO2-lean IL exits the bottom of the flash tank and is pumped back through the process heat exchanger and cooled before entering the absorption column. Because of the very low vapor pressure of the IL, the flash tank vapor is assumed to contain only CO2 and a condenser is not required. Table 2 and Figure 5 provide additional details for the IL-based process, and the ASPEN Plus input file is provided in the Supporting Information.

heat exchanger and cooled before entering the absorption column. The inlet and outlet temperatures for the CO2-lean solution in the process heat exchanger are 389 and 355 K, respectively. The MEA-stripping column was modeled using 12 theoretical stages and contained 26 actual trays. The vapors from the column are condensed to recover water and produce gaseous CO2. CO2 is compressed, cooled, and liquefied using a multi-stage compressor and a NH3 refrigeration system (not shown in Figure 4 or included in this analysis). CO2 is purified using an additional scrubber to remove any traces of SO2 and passes through a series of driers and adsorbent beds before analysis, storage, and shipment. Cooling water was assumed to be available for the condenser and absorption precooler. The loading of the lean solvent is an important parameter concerning the energy required for the regeneration;12 therefore, this value was optimized by varying the MEA mass flow rate. The simplified flow diagram for the IL process is shown in Figure 5. In this case, the absorption column operates under pressure and the flue gas is compressed (P = 0.618 MPa) before entering the column. The absorption column was simulated using Radfrac, and the equilibrium calculations were based on our experimental measurements (CO2 þ [bmim][Ac]) and a simple VLE model. The absorber pressure and IL circulation rate were optimized to maximize CO2 recovery and minimize energy consumption. The inlet flue gas concentration was the same as the MEA base case, and the equipment size and operating conditions are provided in Tables 1 and 2. The solvent in this case is [bmim][Ac], which is cooled using refrigeration (T = 273 K) before entering the absorption column and flowed countercurrent to the flue gas. The treated gas is vented from the top of the column and, because of the very low vapor pressure for the IL, a secondary scrubber (i.e., wash-water section) to remove traces of IL from the vent is not required. The CO2 concentration in the vent stream is reduced to the same concentration as in the MEA scrubbing system. The IL-absorption column was modeled using 20 theoretical stages.

3. Results and Discussion The commercial MEA-based process separates about 90% of CO2 from the flue gas slip stream, with an annual capacity of approximately 47 000 metric tons. This result matches well with our simulated recovery of 91.4% and 47 100 metric tons, as shown in Table 3. CO2 recovered in the commercial process has a purity of about 95% (before purification), which is in good agreement with the simulation result (95.3%). The ILbased simulation was designed to match the CO2 scrubbing capacity of the MEA-based system. The IL-based process separates 46 900 metric tons/year of CO2, which results in a recovery of 91.3%. The IL process was able to achieve a higher CO2 purity of 98.7%. The mass flow rates (m) for the aqueous amine solution and IL absorbent were 43.8 and 48.2 kg/s, respectively. The higher 5785

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al.

Table 3. Scrubbing Performance for MEA and IL

CO2 capacity (metric tons/year)a recovery (%) purity (%) utilities steam (kg/s) cooling water (kg/s) energy steam (kW) electricity (kW) total (kW) a

Table 4. Cost Estimates for MEA and IL

MEA

IL

47100 91.4 95.3

46900 91.3 98.7

5.8 570

3.5 440

11627 13 11640

7145 2645 9790

equipment columns absorber stripper tanks stripper reflux flash heat exchangers process flash preheater absorber precooler absorber prechiller stripper preheater stripper condenser stripper reboiler compressor cooler pumps absorber stripper recycle flue gas compressor total equipment cost

1 ton (metric ton) = 1000 kg.

mass flow rate for the IL is a result of lower CO2 solubility. As mentioned previously, the selection of the IL ([bmim][Ac]) was not optimized but was selected on the basis of chemical absorption behavior. The absorption pressure was increased, and the temperature was decreased, to improve physical absorption. However, a trade-off exists between how much energy is required for compression and refrigeration to maximize the CO2-IL concentration in the absorber and how much energy is required for solvent regeneration. Contrary to our initial belief, the additional energy and equipment cost for compressing the flue gas and cooling the absorber in the IL process can be partially offset by equipment and energy savings in the regeneration process. Also, the higher pressure reduces the size of the absorption column and decreases the IL flow rate (i.e., reducing the size of the IL heat exchangers). The utility requirement (steam and cooling water) for both processes is shown in Table 3. The majority of the energy required for the MEA process is for heating the stripper reboiler to regenerate the solvent. Some energy is also required to preheat the CO2-rich MEA stream before entering the stripping column, and electricity is required to run the absorption and stripping pumps. The total amount of energy (steam and electricity) was 11 640 kW. The energy required to operate the IL process is also a combination of steam and electricity. Steam is required for heating the flash tank and flash preheater. Electricity is required to run the flue gas compressor, refrigeration machine, and IL recycle pump. The total amount of energy was 9790 kW, which is a 16% reduction in total energy compared to the MEA base case. To ascertain the potential economic benefits of an IL CO2 capture system, a financial analysis of the technology was performed and compared to the MEA-based system. To calculate the costs of the two processes, a cost model was developed using the results from the ASPEN simulations. ASPEN Icarus was used to estimate the engineered equipment costs based on the equipment size, materials of construction, and temperature and pressure ratings. The results for the major equipment are shown in Table 4. The total capital equipment costs for the MEA and IL processes were $1.623 and $1.192 million, respectively, which result in a 26% lower engineered equipment cost for the ILbased process. The IL process did require additional equipment, including a compressor and refrigeration machine. The cost of the refrigeration machine was not included with the engineered equipment because it could be purchased as a ready to install package. The total equipment costs were used as the basis for a factored estimate of the project investment. The project investments for the MEA and IL processes were $18.1 and 15.8 million, respectively. Typical and consistent factors were applied for both cases. Table 5 provides details

MEA ($1000)

IL ($1000)

600 457

198

11 17 101

74 78 95 157

231 21 69 104

42 16 13 25 506 1192

1623

refrigeration system project investment working fluid total investment

850 18100 33 18133

15800 400 16200

Table 5. Investment Estimate Elements for IL Base Case investment component IL fluid process equipment insulation building and supports piping instrumentation electrical equipment refrigeration and utilities construction costs design costs contingency total

total estimate (%) 3.2 9.9 2.5 4.9 10.0 5.4 2.7 8.8 14.7 12.9 25.0 100.0

for the major project investment elements for the IL base case. All major investment areas were covered, including capital costs related to design and construction. The contingency of 25% is considered representative at this early stage of an evaluation, representing undeveloped scope, uncertainty in equipment cost, and other considerations. The cost of the absorbent was also considered because ILs are expensive solvents compared to aqueous amines. The MEA process requires an initial charge of about 68 m3 of aqueous MEA solution. On the basis of a MEA price of $2.25/kg, the cost of the MEA was about $33 000. The IL process required about 18.9 m3 of [bmim][Ac]. The prices of ILs are highly dependent upon the choice of cation, anion, purity, and manufacturer. Also, laboratory-quantity prices for ILs do not necessarily reflect the bulk chemical price.57 The price for [bmim][Ac] in multi-ton quantities with a purity of >95% was estimated to be $20/kg. This results in a total IL cost of about $400 000. The cost of the IL charge is more than 10 times the cost of the MEA charge, but equipment cost savings offset some of this difference, as shown in Table 4. Engineering design estimates indicate that the total project investment for the MEA-based process was $18.133 million. This (57) Wasserscheid, P.; Welton, T. Ionic Liquids in Synthesis; Wiley-VCH: Weinheim, Germany, 2004.

5786

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al.

Figure 6. MEA and IL equipment size and footprint.

agrees well (95%) and modeling results for the MEA process were in good agreement with a commercial process. Modeling indicates that the IL process can reduce the energy losses by 16% compared to the MEA process. Engineering design estimates indicate that the investment for the IL process will be 11% lower than the MEA-based process and provide a 12% reduction in equipment footprint. The calculated energy consumption and costs associated with capturing CO2 using both the MEA- and IL-based (60) Shiflett, M. B.; Yokozeki, A. Separation of carbon dioxide and sulfur dioxide using room-temperature ionic liquid [bmim][MeSO4]. Energy Fuels 2010, 24, 1001–1008. (61) Shiflett, M. B.; Yokozeki, A. Chemical absorption of sulfur dioxide in room-temperature ionic liquids. Ind. Eng. Chem. Res. 2010, 49, 1370–1377.

(59) Yokozeki, A.; Shiflett, M. B. Separation of carbon dioxide and sulfur dioxide gases using room-temperature ionic liquid [hmim][Tf2N]. Energy Fuels 2009, 23, 4701–4708.

5788

Energy Fuels 2010, 24, 5781–5789

: DOI:10.1021/ef100868a

Shiflett et al.

processes were high ($147/metric ton of CO2 and $140/metric ton of CO2, respectively). Use of available waste heat in the commercial plant reduces the energy penalty and cost of capturing to about $59/metric ton of CO2, which was in good agreement with the reported cost; however, this is the case with small-scale capture (3-4% CO2 capture). Large-scale (90% CO2 capture) plants will require more energy to operate, and efficient heat recovery will be necessary to minimize the operating costs. A parametric study examined four improvements in the IL technology that promise to reduce the energy

penalty and cost of CO2 capture. Future work will focus on modeling new ILs with improved economics. Acknowledgment. The authors thank Mrs. Anne Marie S. Niehaus for her assistance with the process schematics and DuPont Central Research and Development for supporting the present work. Supporting Information Available: ASPEN Plus input files for MEA and IL processes. This material is available free of charge via the Internet at http://pubs.acs.org.

5789