Separation of Carbon Dioxide and Sulfur Dioxide Using Room


Nov 19, 2009 - Environmental Science & Technology 2013 47 (3), 1598-1605. Abstract | Full Text .... Hou , Weize Wu. Journal of Hazardous Materials 201...
0 downloads 0 Views 2MB Size


Energy Fuels 2010, 24, 1001–1008 Published on Web 11/19/2009

: DOI:10.1021/ef900997b

Separation of Carbon Dioxide and Sulfur Dioxide Using Room-Temperature Ionic Liquid [bmim][MeSO4] Mark B. Shiflett*,† and A. Yokozeki‡ †

DuPont Central Research and Development, Experimental Station, Wilmington, Delaware 19880, and ‡ 109-C Congressional Drive, Wilmington, Delaware 19807 Received September 8, 2009. Revised Manuscript Received October 30, 2009

A ternary equation of state (EOS) model for the CO2/SO2/1-butyl-3-methylimidazolium methyl sulfate ([bmim][MeSO4]) system has been developed in order to gain further our understanding of capturing and enhanced gaseous selectivity of industrial flue gases containing CO2 and SO2 using room-temperature ionic liquids. The present model is based on a generic Redlich-Kwong (RK) EOS. The empirical binary interaction parameters have been determined using our measured vapor-liquid-equilibrium (VLE) data for SO2/[bmim][MeSO4] and literature data for CO2/[bmim][MeSO4] and CO2/SO2. The validity of the present EOS has been checked by conducting ternary VLE experiments for the present system. With this EOS, an isothermal ternary phase diagram and solubility (VLE) behaviors have been calculated for various (T, P, and feed compositions) conditions. The addition of the [bmim][MeSO4] for small and equimolar CO2/SO2 mole ratios significantly increased the selectivity. For large CO2/SO2 mole ratios, the selectivity was high for even a small addition of ionic liquid and in certain cases showed a maximum selectivity due to preferential chemical absorption of SO2. The enhancement in CO2/SO2 selectivity using [bmim][MeSO4] was significantly higher than using 1-hexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([hmim][Tf2N]) from our previous work and may make the simultaneous capture and separation of these acid gases practical.

gases5-16 and may make it possible for the simultaneous capture and separation of CO2 and SO2. The solubility of CO2 in many RTILs has been studied;17-29 however, in the case of SO2, only a few RTILs have been examined for the binary phase PTx (pressure-temperature-composition) behavior.14,30,31 In our first report,30 we examined the physical absorption of

1. Introduction Due to the adverse environmental effect of releasing CO2 and SO2 from the combustion of fossil fuel, new separation processes are becoming an important area of research. Conventional removal techniques include using limestone1,2 and organic solvents;3,4 however, certain types of limestone and solvents may be poisoned by the continuous removal of SO2. Room-temperature ionic liquids (RTILs) are a new type of solvent being considered for capturing these acidic flue

(14) Anderson, J. L.; Dixon, J. K.; Maginn, E. J.; Brennecke, J. F. J. Phys. Chem. B 2006, 110, 15059–15062. (15) Jiang, Y.-Y.; Zhou, Z.; Jiao, Z.; Li, L.; Wu, Y.-T.; Zhang, Z.-B. J. Phys. Chem. B 2007, 111, 5057–5061. (16) Jiang, X.; Nie, Y.; Li, C.; Wang, Z. Fuel 2008, 87, 79–84. (17) Shiflett, M. B.; Yokozeki, A. J. Phys. Chem. B 2007, 111, 2070– 2074. (18) Shiflett, M. B.; Kasprzak, D. J.; Junk, C. P.; Yokozeki, A. J. Chem. Thermodyn. 2008, 40, 25–31. (19) Yokozeki, A.; Shiflett, M. B.; Junk, C. P.; Grieco, L. M.; Foo, T. J. Phys. Chem. B 2008, 112, 16654–16663. (20) Shiflett, M. B.; Yokozeki, A. Ind. Eng. Chem. Res. 2005, 44, 4453–4464.  Tuma, D.; Xia, J.; Maurer, G. J. Chem. (21) Perez-Salado Kamps, A; Eng. Data 2003, 48, 746–749. (22) Husson-Borg, P.; Majer, V.; Costa Gomes, M. F. J. Chem. Eng. Data 2003, 48, 480–485. (23) Shariati, A.; Peters, C. J. J. Supercrit. Fluids 2004, 30, 139–144. (24) Scovazzo, P.; Camper, D.; Hieft, J.; Poshusta, J.; Koval, C.; Noble, R. Ind. Eng. Chem. Res. 2004, 43, 6855–6860. (25) Liu, Z.; Wu, W.; Han, B.; Dong, Z.; Zhao, G.; Wang, J.; Jiang, T.; Yang, G. Chem. Eur. J. 2003, 9, 3897–3903. (26) Aki, S. N. V. K.; Mellein, B. R.; Saurer, E. M.; Brennecke, J. F. J. Phys. Chem. B 2004, 108, 20355–20365. (27) Blanchard, L. A.; Gu, G.; Brennecke, J. F. J. Phys. Chem. B 2001, 105, 2437–2444. (28) Shiflett, M. B.; Yokozeki, A. J. Chem. Eng. Data 2009, 54, 108– 114.  (29) Carvalho, P. J.; Alvarez, V. H.; Schr€ oder, B.; Gil, A. M.; Marrucho, I. M.; Aznar, M.; Santos, L. M. N. B. F.; Coutinho, J. A. P. J. Phys. Chem. B 2009, 113, 6803–6812. (30) Yokozeki, A.; Shiflett, M. B. Energy Fuels 2009, 23, 4701– 4708. (31) Shiflett, M. B.; Yokozeki, A. Ind. Eng. Chem. Res. 2009, accepted.

*To whom correspondence should be addressed. Telephone: 302-6952572. Fax: 302-695-4414. E-mail: [email protected] (1) Ryu, H. J.; Grace, J. R.; Lim, C. J. Energy Fuels 2006, 20, 1621– 1628. (2) Wu, S.; Uddin, Md. A.; Su, C.; Nagamine, S.; Sasaoka, E. Ind. Eng. Chem. Res. 2002, 41, 5455–5458. (3) Boniface, J.; Shi, Q.; Li, Y. Q.; Cheung, J. L.; Rattigan, O. V.; Davidovits, P.; Worsnop, D. R.; Jayne, J. T.; Kolb, C. E. J. Phys. Chem. A 2000, 104, 7502–7510. (4) DuPont BELCO Clean Air Technologies sells Labsorb regenerative scrubbing systems to remove sulfur dioxide from gas streams while minimizing solid and liquid discharges. http//:www.belcotech.com. (5) Brennecke, J. F.; Maginn, E. J. Purification of Gas with Liquid Ionic Compounds; U.S. Patent 6579343, 2003. (6) Wang, Y.; Pan, H.; Li, H.; Wang, C. J. Phys. Chem. B 2007, 111, 10461–10467. (7) Shi, W.; Maginn, E. J. J. Phys. Chem. B 2008, 112, 16710–16720. (8) Wu, W.; Han, B.; Gao, H.; Liu, Z.; Jiang, T.; Huang, J. Angew. Chem., Int. Ed. 2004, 43, 2415–2417. (9) Yuan, X. L.; Zhang, S. J.; Lu, X. M. J. Chem. Eng. Data 2007, 52, 596–599. (10) Mochizuki, Y.; Sugawara, K. Energy Fuels 2008, 22, 3303–3307. (11) Ando, R. A.; Siqueira, L. J. A.; Bazito, F. C.; Torrest, R. M.; Santos, P. S. J. Phys. Chem. 2007, 111, 8717–8719. (12) Huang, J.; Riisager, A.; Wasserscheid, P.; Fehrmann, R. Chem. Commun. 2006, 4027–4029. (13) Huang, J.; Riisager, A.; Berg, R. W.; Fehrmann, R. J. Mol. Catal., A 2008, 279, 170–176. r 2009 American Chemical Society

1001

pubs.acs.org/EF

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

Figure 1. Chemical structure of [bmim][MeSO4].

SO2 in a common RTIL, 1-hexyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide ([hmim][Tf2N]) and found that the capture and separation of CO2 and SO2 was possible, but [hmim][Tf2N] was not necessarily the best choice of ionic liquid for this application. In our continuation to find a more optimum RTIL, we measured the chemical absorption of SO2 in 1-butyl-3-methylimidazolium methyl acetate ([bmim][Ac]) and 1-butyl-3-methylimidazolium methyl sulfate ([bmim][MeSO4]).31 In that work, we showed two types of reversible chemical associations (AB and AB2 where A is RTIL and B is SO2) and concluded that effective capture of SO2 from flue gas requires strong chemical absorption because of the low partial pressures (e.g., 0.2 vol % at atmospheric pressure) of SO2 in this stream. In the present study, we continue our previous work to understand capturing and enhanced gaseous selectivity of CO2 and SO2 using [bmim][MeSO4]. In this report, we construct the ternary phase diagram of CO2 þ SO2 þ [bmim][MeSO4] for the first time, using our ternary equation of state (EOS) model.17,30,32 The model is based on a generic Redlich-Kwong (RK) EOS, with empirical binary interaction parameters determined from our previous measurements for SO2/[bmim][MeSO4]31 and literature data for CO2/[bmim][MeSO4]33 and CO2/SO2.34 In order to check the validity of the present EOS model, VLE experiments for the present ternary system were performed under various T, P, and feed compositions, and the EOS validity was satisfactorily confirmed. Then, the capturing and selectivity of these acid gases with RTIL [bmim][MeSO4] are calculated at several feed, T, and P conditions. The selectivity advantage using this RTIL and the simultaneous CO2/SO2 capture characteristics are discussed based on the present ternary phase calculations.

Figure 2. Schematic diagram of a sample cell.

Materials. Sulfur dioxide (mole fraction purity >0.9998, CAS No. 7446-09-5) and carbon dioxide (mole fraction purity >0.9999, CAS No. 124-38-9) were purchased from MG Industries (Philadelphia, Pennsylvania). The [bmim][MeSO4] (ECOENG 411, purity >99%, C9H18N2O4S, Lot number 99/ 851, CAS No. 401788-98-5) was obtained from Solvent Innovations (Cologne, Germany). Figure 1 provides the chemical structure for [bmim][MeSO4]. The [bmim][MeSO4] ionic liquid sample was dried and degassed by first placing the sample in a borosilicate glass tube and pulling a coarse vacuum on the sample with a diaphragm pump (Pfeiffer, model MVP055-3) for about 3 h. Next, the sample was fully evacuated using a turbopump (Pfeiffer, model TSH-071) to a pressure of about 4  10-7 kPa while simultaneously heating and stirring the ionic liquid at a temperature of about 348 K for 5 days. The final water content was measured by Karl Fischer titration

(Aqua-Star C3000, solutions AquaStar Coulomat C and A) and the ionic liquid contained less than 1  10-3 mass fraction of water. Binary VLE Measurement. In our previous report, we measured the gas solubility of SO2 and [bmim]MeSO4] using a gravimetric microbalance20 (Hiden Isochema Ltd., IGA 003). Detailed descriptions of experimental equipment and procedures for the VLE are given in our previous reports.20,30,31 Safety precautions must be considered when handling SO2 gas, which is highly toxic with an 8 to 12 h allowable exposure limit (AEL) of 1 ppm. Detailed SO2 safety and handling procedures can be found in our previous reports.30,31 Solubility of CO2 and [bmim][MeSO4] data were taken from Kumezan et al.,33 and CO2/SO2 data were selected from the DDB database.34 Binary VLLE Measurement. Two high-pressure sample containers were filled with dried [bmim][MeSO4] following the procedures outlined in our previous publications.18,35,36 The two samples contained about 67.6 and 87.8 mol % CO2. VLLE experiments have been made with these samples at constant temperatures of about 275 and 295 K using the volumetric method.18,35,36 The VLLE determined by this method required only mass and volume measurements without any analytical method for molar composition or volume analysis. Special attention must be given to ensure no leaks occur from the sample containers after being filled with the high-pressure CO2. Weights of sample containers were checked several times before starting and after completing the VLLE experiments to quantify whether any CO2 had escaped from the sample container. The samples were placed inside a constant temperature bath (Tamson Instruments, TV4000LT, Zoetermeer, Netherlands) which was initially set at about 295 ( 0.2 K. The sample cells were vigorously shaken to assist with mixing prior to being immersed in the tank. The bath temperature was calibrated using a standard platinum resistance thermometer (SPRT model 5699, Hart Scientific, range 73-933 K) and readout

(32) Yokozeki, A. Int. J. Thermophys. 2001, 22, 1057–1071.  Tuma, D.; Maurer, G. (33) Kumezan, J.; Perez-Salado Kamps, A; J. Chem. Eng. Data 2006, 51, 1802–1807. (34) Dortmund Data Bank. 2008, http://www.dechema.de.

(35) Shiflett, M. B.; Yokozeki, A. J. Phys. Chem. B 2006, 110, 14436– 14443. (36) Shiflett, M. B.; Yokozeki, A. J. Chem. Eng. Data 2006, 51, 1931– 1939.

2. Experimental Section

1002

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

(Blackstack model 1560 with SPRT module 2560). The Blackstack instrument and SPRT are a certified secondary temperature standard with a NIST traceable accuracy to (0.005 K. The uncertainty in the bath temperature was 0.2 K. It is important to mention that the vapor phase density that contains CO2 with a negligible contribution of [bmim][MeSO4] must be properly accounted for in the mass balance equations described in refs 35 and 36. Ternary VLE Measurement. We have also conducted VLE experiments for the present ternary system (CO2/SO2/[bmim][MeSO4]) at several thermodynamic conditions similar to our previous experiments with CO2/SO2/[hmim][Tf2N]30 in order to verify the present EOS model. Five sample cells have been constructed as shown in Figure 2. Each cell was made using Swagelok fittings, two Swagelok ball valves (SS-426S4), stainless steel cylinder, and a pressure gauge (Parker Instruments, 0 to 0.75 MPa). The internal volume of each cell was calculated by measuring the mass of methanol required to completely fill the cell and knowing the density of methanol at the fill temperature. The internal volume of each cell (VT) was 90 ( 2.5 cm3. Ionic liquid was loaded by mass (0.75-5.72 g) and weighed on an analytical balance with a resolution of 0.01 g (Mettler Toledo PG-4002-S) inside a nitrogen-purged drybox. In order to load ionic liquid in the cell, the pressure gauge, valves, and fittings were removed and a glass pipet (10 mL) that fit through the cylinder opening was used for filling. The pressure gauge, valves, and fittings were assembled as shown in Figure 2 and the sample cell was removed from the drybox. After filling with the ionic liquid, the sample cell was always maintained in a vertical upright position when valve 1 was open to prevent contact with the valves and pressure gauge. If the sample cell had to be mixed or weighed in a horizontal position, valve 1 would be momentarily closed and then opened again in the vertical position. The cell was connected to the diaphragm pump, with both valves open, to remove residual nitrogen. Once evacuated, the ball valve (valve 2) was closed and the cell was weighed to obtain the initial ionic liquid mass. The CO2/SO2 gas mixtures were also loaded by mass (0.45-1.59 g) from a high pressure gas cylinder. Three CO2/ SO2 gas mixtures (8.1/91.9, 48.3/51.7, and 86.8/13.2 mol % CO2/ SO2) were prepared by weight and analyzed by gas chromatography (GC) (Hewlett-Packard HP6890) using an isothermal (353.15 K) method (GS-GASPRO capillary column, 60 m length, 0.32 mm I.D., model 113-4362, Agilent Technologies, inlet injector temperature 473.15 K, thermal conductivity detector temperature 523.15 K, helium carrier gas, flow rate 55 cm3 min-1 with a 20:1 split ratio, injection volume 25 μL). Special care must be taken when preparing the CO2/SO2 gas mixtures to prevent the SO2 from condensing. The saturation vapor pressure for CO2 at 293 K is 6.0 MPa; however, the saturation pressure for SO2 at 293 K is much lower at 0.35 MPa. Therefore, the total pressure for the three gas mixtures (8.1/91.9, 48.3/51.7, 86.8/13.2 mol % CO2/SO2) were 0.31, 0.48, and 0.60 MPa, respectively. The five sample cells were placed inside a Plexiglas tank and the temperature was controlled with an external temperature bath (VWR International, Model 1160S) that circulated water through a copper coil inside the tank. The bath was stirred with an agitator (Arrow Engineering Co., Inc. model 1750) and the temperature was measured with a thermocouple (Fluke 52II thermometer). The temperature was initially set at about 297 K. The sample cells were vigorously shaken to assist with mixing prior to being immersed in the tank. The water level in the tank was adjusted such that the entire cell was under water, including the bottom 2 cm of the pressure gauge. The pressure in each cell was recorded until no change in pressure was measured. To ensure the samples were at equilibrium and properly mixed, the cells were momentarily removed from the water bath and vigorously shaken. The cells were

Table 1. EOS Constants for Pure Compounds Used in the Present Study

-1

molar mass (g 3 mol ) TC (K) PC (MPa) β0 β1 β2 β3 a

sulfur dioxide

carbon dioxide

64.07 430.64 7.884 1.002 22 0.450 287 -0.048 279 0.001 2512

44.01 304.13 7.377 1.000 49 0.438 660 -0.104 98 0.062 500

[bmim][MeSO4]a 250.32 920.1 2.806 1.0 0.611 705

Taken from our previous work, see ref 31.

Table 2. Optimal Binary Interaction Parameters in eqs 6-8 system (1)/(2)

l12

l21

m12 = m21 τ12 = τ21/K

0.142 37 0.167 42 -0.188 14 CO2/ [bmim][MeSO4] -0.082 38 -0.198 49 0.170 43 SO2/[bmim][SO4]a 0.031 95 0.905 96 0.0 CO2/SO2 a

33.925 0.0 -2.853

Taken from our previous work, see ref 31.

Figure 3. Isothermal Px (pressure-liquid composition) phase diagram, or solubility chart. (a) SO2 þ [bmim][MeSO4] binary system, lines; the present EOS model calculations, symbols; experimental data taken from ref 31; (b) CO2 þ [bmim][MeSO4] binary system, solid lines; the present EOS calculations, circle symbols; VLE data taken from reference 33, square symbols and broken lines; the present experimental VLLE data and the LLE tie lines.

placed back in the bath and the process was repeated until no change in pressure was measured. In most cases, the cells reached equilibrium in 12-24 h. The process was repeated at a higher temperature of about 319-322 K. The pressure gauges were calibrated using the Paroscientific Model 765-1K pressure transducer. The Fluke thermometer was calibrated 1003

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

Table 3. Experimental VLLE for CO2 (1) þ [bmim][MeSO4] (2) T (K)

x01

(mol %)

x1 (mol %)

294.5 275.2

50.5 ( 1.0 52.9 ( 1.0

99.6 ( 0.4 99.6 ( 0.4

a

V 0 a (cm3 3 mol-1) 122.5 ( 1.0 116.1 ( 1.0

0

V a (cm3 3 mol-1)

V ex b (cm3 3 mol-1)

V exb (cm3 3 mol-1)

60.0 ( 1.0 49.0 ( 1.0

-9.0 ( 1.0 -5.6 ( 1.0

0.5 ( 1.0 0.3 ( 1.0

Observed molar volume. b Excess molar volume.

using the standard platinum resistance thermometer described previously. The vapor space of each sample was analyzed by placing a rubber septum over the end of the valve and inserting a gastight syringe (Hamilton Company, Reno, Nevada) to remove a small sample (25 μL) that was analyzed by GC. To prevent condensation of SO2 when taking samples at higher temperature (319-322 K), the syringe was preheated in an oven to 343 K. The CO2 and SO2 peaks appear at about 3.5 and 12.2 min with a total method run time of 15 min.

3. Thermodynamic Model To study the phase behavior of a ternary system of CO2/ SO2/[bmim][MeSO4], we have developed thermodynamic models based on EOS, which have been successfully applied for refrigerant/lubricant oil mixtures32 and various CO2,17-20 SO2,30,31 hydrofluorocarbon,37 NH3,38,39 and H240 mixtures with ionic liquids. It is based on a generic RK type of cubic EOS: P ¼

RT aðTÞ V -b VðV þ bÞ

aðTÞ ¼ 0:427480

R2 Tc2 RðTÞ Pc

b ¼ 0:08664

RTc Pc

Figure 4. Isothermal ternary CO2/SO2/[bmim][MeSO4] phase diagram calculated by the present EOS model at T = 298.15 K.

ð1Þ ð2Þ ð3Þ

The temperature-dependent part of the a parameter in the EOS for pure compounds is modeled by the following empirical form:17-20 e3 X RðTÞ ¼ βk ð1=Tr -Tr Þk ðTr  T=TC Þ ð4Þ k ¼0

The coefficients, βk, are determined so as to reproduce the vapor pressure of each pure compound. The a and b parameters for general N-component mixtures are modeled in terms of their respective binary interaction parameters.17-20 a ¼

Figure 5. Comparison of experimental and calculated VLE data for the ternary CO2/SO2/[bmim][MeSO4] system. Calculated vaporphase compositions of SO2 (CO2 mole % = 100 - SO2 mol %) are compared with observed SO2 vapor compositions for various experimental conditions (see Table 4). Symbols: open symbols = about 297 K; solid symbols = about 319-322 K; circle symbols = 8.1/91.9 mol % CO2/SO2 feed; square symbols = 48.3/51.7 mol % CO2/SO2 feed; diamond symbols = 86.8/13.2 mol % CO2/SO2 feed.

N X pffiffiffiffiffiffiffiffi ai aj fij ðTÞð1 -kij Þxi xj , i, j ¼1

ai ¼ 0:427480 fij ðTÞ ¼ 1 þ τij =T,

R2 Tci2 Ri ðTÞ Pci

where τij ¼ τji , and τii ¼ 0:

lij lji ðxi þ xj Þ kij ¼ , lji xi þ lij xj

where kii ¼ 0

ð5Þ ð6Þ

b ¼

ð7Þ

N 1X ðbi þ bj Þð1 -kij Þð1 -mij Þxi xj , 2 i, j ¼1

bi ¼ 0:08664

(37) Yokozeki, A.; Shiflett, M. B. AIChE J. 2006, 52, 3952–3957. (38) Yokozeki, A.; Shiflett, M. B. Ind. Eng. Chem. Res. 2007, 46, 1605–1610. (39) Yokozeki, A.; Shiflett, M. B. Appl. Energy 2007, 84, 1258–1273. (40) Yokozeki, A.; Shiflett, M. B. Appl. Energy 2007, 84, 351–361.

RTci Pci

ð8Þ

where mij = mji, mii = 0; Tci is the critical temperature of the ith species; Pci is the critical pressure of the ith species; R is the 1004

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

Table 4. Experimental VLE Data for Ternary Mixtures feed CO2 (mol %)

feed SO2 (mol %)

feed [bmim] [MeSO4] (mol %)

6.9 ( 0.1 6.1 ( 0.1 5.3 ( 0.1 5.2 ( 0.1 4.3 ( 0.1 5.3 ( 0.1 5.2 ( 0.1 4.3 ( 0.1 42.1 ( 0.3 38.9 ( 0.3 33.6 ( 0.3 29.1 ( 0.3 19.9 ( 0.3 42.1 ( 0.3 38.9 ( 0.3 33.6 ( 0.3 19.9 ( 0.3 68.3 ( 1.0 61.4 ( 1.1 46.7 ( 1.1 35.6 ( 1.1 27.0 ( 0.9 68.3 ( 1.0 61.4 ( 1.1 46.7 ( 1.1 35.6 ( 1.1 27.0 ( 0.9

78.7 ( 0.5 69.7 ( 0.7 59.8 ( 0.6 58.7 ( 0.5 48.4 ( 0.4 59.8 ( 0.6 58.7 ( 0.5 48.4 ( 0.4 45.0 ( 0.3 41.6 ( 0.3 36.0 ( 0.3 31.2 ( 0.3 21.3 ( 0.3 45.0 ( 0.3 41.6 ( 0.3 36.0 ( 0.3 21.3 ( 0.3 10.4 ( 0.2 9.3 ( 0.2 7.1 ( 0.2 5.4 ( 0.2 4.1 ( 0.1 10.4 ( 0.2 9.3 ( 0.2 7.1 ( 0.2 5.4 ( 0.2 4.1 ( 0.1

14.4 ( 0.6 24.2 ( 0.7 34.9 ( 0.7 36.2 ( 0.6 47.3 ( 0.4 34.9 ( 0.7 36.2 ( 0.6 47.3 ( 0.4 12.9 ( 0.5 19.5 ( 0.6 30.4 ( 0.6 39.7 ( 0.6 58.8 ( 0.6 12.9 ( 0.5 19.5 ( 0.6 30.4 ( 0.6 58.8 ( 0.6 21.3 ( 1.2 29.3 ( 1.3 46.2 ( 1.3 59.0 ( 1.3 68.9 ( 1.0 21.3 ( 1.2 29.3 ( 1.3 46.2 ( 1.3 59.0 ( 1.3 68.9 ( 1.0

T (K)

P (MPa)

liquid SO2 calculated (mol %)

liquid [bmim][MeSO4] calculated (mol %)

vapor SO2 calculated (mol %)

vapor SO2 measured (mol %)

297.1 296.5 297.2 296.3 296.8 319.2 319.2 319.2 297.7 297.3 297.4 297.2 297.7 322.4 322.3 321.8 322.4 296.6 296.8 296.9 296.8 296.8 319.2 319.2 319.2 319.2 319.2

0.212 0.136 0.109 0.114 0.107 0.170 0.177 0.143 0.405 0.356 0.336 0.329 0.236 0.501 0.436 0.384 0.246 0.301 0.294 0.287 0.281 0.274 0.315 0.308 0.294 0.301 0.301

77.5 ( 0.3 68.3 ( 0.3 60.0 ( 0.3 59.3 ( 0.3 49.3 ( 0.4 57.1 ( 0.2 56.8 ( 0.1 48.1 ( 0.1 67.8 ( 0.5 60.7 ( 0.6 50.0 ( 0.6 41.4 ( 0.6 25.5 ( 0.5 61.2 ( 0.5 55.4 ( 0.6 46.8 ( 0.6 24.6 ( 0.6 27.7 ( 1.0 21.2 ( 0.9 12.2 ( 0.6 7.8 ( 0.4 5.3 ( 0.2 23.2 ( 0.6 18.5 ( 0.6 11.1 ( 0.4 7.4 ( 0.3 5.1 ( 0.2

21.3 ( 0.5 30.7 ( 0.3 39.0 ( 0.1 39.6 ( 0.2 49.5 ( 0.3 42.2 ( 0.1 42.3 ( 0.1 51.0 ( 0.1 29.3 ( 0.5 36.6 ( 0.6 47.4 ( 0.6 56.0 ( 0.6 72.3 ( 0.5 36.6 ( 0.5 42.5 ( 0.6 51.2 ( 0.6 73.9 ( 0.6 69.5 ( 1.0 75.9 ( 0.8 84.6 ( 0.4 88.8 ( 0.3 91.3 ( 0.1 74.7 ( 0.5 79.4 ( 0.5 86.6 ( 0.3 90.2 ( 0.2 92.4 ( 0.1

81.1 ( 2.2 74.9 ( 3.7 58.7 ( 4.8 52.3 ( 4.4 30.4 ( 2.9 73.0 ( 2.8 69.3 ( 3.2 52.1 ( 3.4 27.1 ( 1.4 19.8 ( 1.2 10.9 ( 0.8 6.3 ( 0.5 3.1 ( 0.3 36.2 ( 1.1 29.9 ( 1.1 20.2 ( 0.8 8.3 ( 0.3 2.7 ( 0.3 1.8 ( 0.2 1.0 ( 0.1 0.7 ( 0.1 0.5 ( 0.1 5.3 ( 0.4 4.0 ( 0.3 2.5 ( 0.2 1.7 ( 0.2 1.2 ( 0.1

80.3 ( 1.0a 74.6 ( 1.0a 58.9 ( 1.0a 53.3 ( 1.0a 32.5 ( 1.0a 74.2 ( 1.0a 69.2 ( 1.0a 51.0 ( 1.0a 26.3 ( 1.0b 18.1 ( 1.0b 9.5 ( 1.0b 5.4 ( 1.0b 2.7 ( 1.0b 36.1 ( 1.0b 29.6 ( 1.0b 19.5 ( 1.0b 11.2 ( 1.0b 2.8 ( 1.0c 1.7 ( 1.0c 0.8 ( 0.5c 0.4 ( 0.4c 0.2 ( 0.2c 5.0 ( 1.0c 3.5 ( 1.0c 1.7 ( 1.0c 1.0 ( 1.0c 0.5 ( 0.5c

CO2/SO2 gas mixtures: a (8.1/91.9) mol %; b (48.3/51.7) mol %; c (86.8/13.2) mol %. Liquid CO2 mol % = (100 - liquid SO2 mol % - liquid [bmim][MeSO4] mol %). Vapor CO2 mol % = (100 - vapor SO2 mol %, and vapor [bmim][MeSO4] = 0 mol %).

universal gas constant; and xi is the mole fraction of the ith species. In the above model, there are a maximum of four binary interaction parameters: lij, lji, mij, and τij for each binary pair. However, only two or three parameters are sufficient for most cases. Details for the fugacity coefficient φi of the ith species for the present EOS model, which is needed for the phase equilibrium calculation, can be found in our previous report.30 The equilibrium solubility for the ternary VLE system can be obtained by solving the following equilibrium conditions: xi φLi ¼ yi φV i , ði ¼ 1, 2, 3Þ

on data from refs 42 and 43. For the ionic liquid, the critical parameters (Tc and Pc) and βk in eqs 2-4 were taken from our previous work.31 Table 1 shows the EOS constants for the present compounds. Binary interaction parameters, lij, lji, mij, and τij in eqs 6-8, for each binary pair were obtained using nonlinear regression analyses of experimental PTx (pressuretemperature-composition) data for SO2 þ [bmim][MeSO4],31 CO2 þ [bmim][MeSO4],33 and CO2 þ SO234 systems. Details of the PTx data for CO2 þ SO2 mixtures selected from the Dortmund data bank (DDB) database can be found in our previous report.30 Table 2 presents optimal binary interaction parameters for each binary system. Figure 3a shows our model predictions of SO2 VLE in [bmim][MeSO4] taken from our previous work31 with an excellent standard deviation for the P versus x1 fit (dP = 0.0027 MPa). Figure 3b is a similar plot of our model predictions of CO2 VLE in [bmim][MeSO4] using the experimental data of Kumelan et al.33 The standard deviation for the P versus x1 fit is good (dP = 0.089 MPa). The present EOS has also predicted the VLLE (or liquid-liquid separation) in the CO2-rich side solution, as shown in Figure 3b. This VLLE prediction has been well confirmed by the present VLLE experiment. Table 3 provides the observed liquid phase compositions and molar volumes. VLE data of CO2 þ SO2 mixtures obtained from the DBB compared with the present EOS calculations can be found in our previous report.30 Although the solubility behavior of each binary system has been well correlated with the present EOS model as illustrated in Figure 3, the phase behavior prediction of the ternary system of CO2/SO2/[bmim][MeSO4] may not always be guaranteed based on the binary interaction parameters alone. Particularly for systems containing supercritical fluids and/or nonvolatile compounds such as the present case, the validity of a proposed EOS model for

ð9Þ

where xi is the liquid mole fraction of the ith species (x1 þ x2 þ x3 = 1); yi is the vapor mole fraction of the ith species (y1 þ y2 þ y3 = 1); φLi is the liquid-phase fugacity coefficient of the ith species; and φV i is the vapor-phase fugacity coefficient of the ith species. In the case of three phase equilibria (VLLE), equations corresponding to eq 9 become: L1 L2 V ¼ xL2 ¼ yV xL1 i φi i φi i φi , ði ¼ 1, :::, NÞ

ð10Þ

where superscripts, L1 and L2, denote one liquid phase (1) and another coexisting liquid phase (2) of VLLE, respectively. Numerical solutions of eq 9 or 10 (nonlinearly coupled equations) can be obtained by use of the TP-Flash (Rachford-Rice) method.41 EOS Model Parameters. Pure component EOS parameters for sulfur dioxide and carbon dioxide were determined based (41) Van Ness, H. C.; Abbott, M. M. Classical Thermodynamics of Nonelectrolyte Solutions; McGraw-Hill: New York, 1982; p.427. (42) Lemmon, E. W.; McLinden, M. O.; Huber, M. L. Standard Reference Data Program REFPROP, version 8.0; National Institute of Standards and Technology: Gaithersburg, MD, 2008. (43) Span, R.; Wagner, W. J. Phys. Chem. Ref. Data 1996, 25, 1509– 1596.

1005

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

ternary mixtures must be checked experimentally; see the next section. The isothermal ternary phase diagram predicted by the present EOS at 298.15 K is shown in Figure 4, where a large portion of the ternary composition exhibits the liquid-liquid separation (LLE) that reflects the immiscibility gap in the binary CO2 þ [bmim][MeSO4] system; see Figure 3b. Figure 5 presents the comparison of observed (i.e., present experimental measurements) and calculated values for SO2 mol % in the vapor phase (CO2 mol % = 100 - SO2 mol %, and [bmim][MeSO4] = 0 mol %) under various ternary feed compositions and T, P conditions; see Table 4. The model calculations and experimental data are in good agreement. 4. Results and Discussion Now that the present EOS model has been verified, we can predict the solubility behavior of the present ternary system with confidence. In order to assess the feasibility of the gas separation by the extractive distillation or selective absorption method, gaseous selectivity RA/B, ability to separate gases A and B in the gas phase, or gaseous absorption selectivity SB/A in the liquid phase is commonly defined as in refs 40, 44, and 45:     yA ð11Þ RA=B ¼ SB=A ¼ = xyB xA B where xA (or xB) and yA (or yB) are the mole fractions of A (or B) in the ionic liquid solution phase and vapor phase, respectively. Here we denote CO2 as A and SO2 as B. The CO2/ SO2 selectivity (RA/B) in the gas phase has been examined using the present EOS model at various T, P, and feed compositions, and results are shown in Figures 6 and 7. In Figure 6a, the CO2/SO2 selectivity (RA/B) is plotted as a function of the ionic liquid [bmim][MeSO4] concentration for ternary mixtures with three CO2/SO2 mole ratios (1/9, 1/1, and 9/1) at T = 298.15 K and P = 1 bar. The ternary mixture with the lowest CO2/SO2 mole ratio (CO2/SO2 = 1/9) or highest SO2 concentration shows a significant increase in selectivity (RA/B) from 42 to 346 with increasing ionic liquid addition. The equimolar CO2/SO2 case (CO2/SO2 = 1/1) shows a similar increase in selectivity from 90 to 310. For the highest CO2/SO2 mole ratio (CO2/SO2 = 9/1) or lowest SO2 concentration, the selectivity was initially high (258) for even a small addition (1 mol %) of [bmim][MeSO4] and increased to a maximum selectivity of about 348 at about 25 mol % ionic liquid addition. This characteristic behavior is due to the fact that SO2 has much stronger solubility in [bmim][MeSO4] than CO2, and the large SO2 concentration is absorbed in the liquid phase, leading to the high CO2/SO2 selectivity in the gas phase. The degree of selectivity is largely dependent on the CO2/SO2 feed composition and amount of ionic liquid as shown in Figure 6a. In order to provide clear insights into the change in selectivity due to the ionic liquid addition, Figure 6b shows the selectivity without [bmim][MeSO4] at 298.15 K plotted as a function of pressure for the same CO2/SO2 feed ratios (1/9, 1/ 1, and 9/1). The selectivity enhancement due to the ionic liquid addition can be well observed from the comparison between Figures 6a and 6b. For example, the feed ratio of 9/1

Figure 6. (a) Plots of calculated selectivity defined by eq 11 vs [bmim][MeSO4] mole percent with three different CO2/SO2 feed ratios at T = 298.15 K and P = 1 bar. (b) Selectivity plots without ionic liquid [bmim][MeSO4] as a function of total pressure at T = 298.15 K; lines: dotted line = 1/9 CO2/SO2 feed mole ratio; broken line = 1/1 CO2/SO2 feed mole ratio; solid line = 9/1 CO2/SO2 feed mole ratio. (c) Comparison of calculated selectivity with [bmim][MeSO4] vs [hmim][Tf2N] mole percent at 9/1 CO2/SO2 feed ratio at T = 298.15 K and P = 1 bar, [hmim][Tf2N] data taken from ref 30.

(CO2/SO2) with the ionic liquid has a selectivity range of 226-348, whereas the corresponding case without the ionic liquid shows a selectivity range of 3-9. Similar trends are seen in the other feed ratio cases. Figure 6c shows the remarkable (i.e., order of magnitude) increase in CO2/SO2 selectivity for the feed ratio of 9/1 (CO2/SO2) using ionic liquid [bmim][MeSO4] versus our previous report30 using

(44) Yokozeki, A.; Shiflett, M. B. Ind. Eng. Chem. Res. 2008, 47, 8389–8395. (45) Peng, X.; Wang, W.; Xue, R.; Shen, Z. AIChE J. 2006, 52, 994– 1003.

1006

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

Figure 7. Plots of calculated selectivity defined by eq 11 vs [bmim][MeSO4] mole percent with three different CO2/SO2 feed ratios at (a) T = 333.15 K and P = 1 bar and (b) T = 333.15 K and P = 10 bar. (c) Selectivity plots without ionic liquid [bmim][MeSO4] as a function of total pressure at T = 333.15 K; lines: dotted line = 1/9 CO2/SO2 feed mole ratio; broken line = 1/1 CO2/SO2 feed mole ratio; solid line = 9/1 CO2/ SO2 feed mole ratio. Comparison of calculated selectivity with [bmim][MeSO4] vs [hmim][Tf2N] mole percent at 9/1 CO2/SO2 feed ratio (d) at T = 333.15 K and P = 1 bar and (e) at T = 333.15 K and P = 10 bar, [hmim][Tf2N] data taken from ref 30.

[hmim][Tf2N] (selectivity range of 29-31). The increase in CO2/SO2 selectivity is due primarily to the strong chemical absorption for SO2 in [bmim][MeSO4] versus SO2 in [hmim][Tf2N]. The selectivity characteristics at a high temperature (333.15 K) and pressures (1 and 10 bar) are shown in Figure 7, with and without the ionic liquid addition. The selectivity enhancement with the ionic liquid addition is clearly seen, and the general behaviors are similar to the cases shown in Figure 6. Figures 7d and 7e compare the order of magnitude increase in CO2/SO2

selectivity at T = 333.15 K for the feed ratio of 9/1 (CO2/SO2) using ionic liquid [bmim][MeSO4] versus [hmim][Tf2N] at P = 1 bar and P = 10 bar, respectively. In actual flue gas from the combustion of fossil fuels, the CO2/SO2 mole ratio is large (about 50/1) and larger than the present largest case (9/1). However, with the use of our EOS model the selectivity behavior observed with the present 9/1 mol ratio is very similar to the case of such large mole ratios; the selectivity does not depend strongly on the amount of the ionic liquid, but the ionic liquid addition significantly 1007

Energy Fuels 2010, 24, 1001–1008

: DOI:10.1021/ef900997b

Shiflett and Yokozeki

enhances the CO2/SO2 selectivity. Due to the stronger absorption of SO2 in the [bmim][MeSO4], the simultaneous capture and separation of CO2/SO2 in flue gas with the high mole ratios may be practical.

ionic liquid [hmim][Tf2N]. The present ionic liquid may still not be the best choice for the gaseous separation and/or capturing for CO2/SO2; however, the selection of a strong chemical absorbing ionic liquid such as [bmim][MeSO4] versus [hmim][Tf2N] had a significant effect on the gaseous selectivity.

5. Conclusions Acknowledgment. The authors thank Mr. Brian L. Wells and Mrs. Anne Marie S. Niehaus for their assistance with the vapor-liquid equilibrium and vapor-liquid-liquid equilibrium measurements and DuPont Central Research and Development for supporting the present work.

Although the separation concept of gaseous mixtures using room-temperature ionic liquids has been proposed in the past, no quantitative demonstrations have been reported in the literature for CO2/SO2, except for our previous report using [hmim][Tf2N]. In this report, we have developed a reliable EOS model for the ternary CO2/SO2/[bmim][MeSO4] system and have shown the significant improvement in gaseous selectivity and simultaneous capture of these acid gases, compared with not using an ionic liquid and with our previous work using

Note Added after ASAP Publication. There was an error in equation 5 in the version of this paper published ASAP November 19, 2009; the corrected version published ASAP November 24, 2009.

1008