Article pubs.acs.org/jced
Isobaric Vapor−Liquid Equilibrium for Acetone + Methanol + Phosphate Ionic Liquids Xiaochun Chen,† Bin Yang,† Ahmed A. Abdeltawab,‡ Salem S. Al-Deyab,‡ Guangren Yu,*,† and Xingyue Yong*,† †
Beijing Key Laboratory of Membrane Science and Technology & College of Chemical Engineering, Beijing University of Chemical Technology, Beijing 100029, P. R. China ‡ Petrochemicals Research Chair, College of Science, King Saud University, Riyadh 11451, Saudi Arabia S Supporting Information *
ABSTRACT: Isobaric vapor−liquid equilibrium (VLE) at atmospheric pressure (101.3 kPa) for the binary systems of acetone + methanol, acetone + phosphate ionic liquids (ILs), and methanol + phosphate ILs, and for the ternary system of acetone + methanol + phosphate ILs are measured using a circulation VLE still. The phosphate ILs include 1,3-dimethylimidazolium 1,3-dimethylimidazolium dimethylphosphate ([MMIM][DMP]), 1-ethyl-3-methylimidazolium diethylphosphate ([EMIM][DEP]), and 1-butyl-3-methylimidazolium dibutylphosphate ([BMIM][DBP]). The addition of these phosphate ILs to the azeotropic acetone + methanol system results in a salting-out effect on acetone and makes the azeotropic point disappear. The relative volatility α12 of acetone over methanol increases with increasing molar fraction of ILs. The equilibrium data were well fitted by the electrolyte nonrandom two-liquid model (e-NRTL). Compared with some reported ILs previously such as pyridinium hexafluorophosphate, imidazolium trifluoromethane sulfonate and imidazolium dicyanamide, such imidazolium phosphate ILs are fluorine-free, are prepared more simply with lower cost, and make the azeotropic point disappear at less added amount of ILs. This work shows that such phosphate ILs are a class of potential solvents to separate azeotropic acetone + methanol system.
■
INTRODUCTION Acetone and methanol are a typical azeotropic system in chemical industries such as furfural production and Fischer−Tropsch process.1,2 For such an azeotropic system, simple distillation usually does not work, and some special distillation technologies such as azeotropic distillation,3 pressure swing distillation4,5 and extractive distillation6,7 are needed. Among these methods, extractive distillation is preferred. Typically, ethylamine was used as entrainer for the separation of the azeotropic system of acetone + methanol.8 However, the volatility and corrosion of ethylamine limits its industrial application; therefore an alternative to such an unfavorable entrainer is desired. Recently, ionic liquids (ILs) have received much more attention as entrainers used in extractive distillation after they were first studied by Arlt et al. for their unique properties.9 As “designer green solvents”, ILs have three advantages compared with some traditional entrainers: (i) they can be easily added to the reflux stream because of low melting points (typically below 373 K); (ii) their great solubility for inorganic, organic, and polymeric materials makes them have a higher salting-out effect with higher concentration of electrolyte; (iii) ILs can be regenerated after removing some volatile compounds from ILs by distillations such as flash distillation due to the negligible vapor pressure of ILs. Also, they have some other advantages such as higher thermal stability and less susceptibility to corrosion.10 Phase equilibrium data are necessary to study using ILs as entrainers in extractive distillation separation for azeotropic © XXXX American Chemical Society
systems. Isothermal phase equilibrium data were determined for acetone + methanol + ILs (e.g., N-butyl-pyridinium hexafluorophosphate, [bpy][PF6];11 1-ethyl-3-methylimidazolium hydrogen sulfate, [EMIM][HSO4];12 1-ethyl-3-methylimidazolium
Figure 1. Structure of phosphate-based ILs. Received: August 7, 2014 Accepted: December 30, 2014
A
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
Table 1. Specifications of Chemical Samples chemical name
CAS
mass fraction purity
source
purification method
final water mass fraction
analysis method
0.99
none
not detected
GCd
0.999
none
not detected
GCd
654058-04-5
Tianjin Guangfu Technology Development Co. Ltd. Tianjin Guangfu Technology Development Co. Ltd. prepared in this work
0.98
0.0005
KFe
[EMIM][DEP]b
848641-69-0
prepared in this work
0.98
0.0005
KFe
[BMIM][DBP]c
663199-28-8
prepared in this work
0.98
vacuum desiccation vacuum desiccation vacuum desiccation
0.0005
KFe
acetone
67-64-1
methanol
67-56-1
[MMIM][DMP]a
a c
[MMIM][DMP] = 1, 3-dimethylimidazolium dimethylphosphate. b[EMIM][DEP] = 1-ethyl-3-methylimidazolium diethylphosphate. [BMIM][DBP] = 1-butyl-3-methylimidazolium dibutylphosphate. dGC = gas chromatography. eKF = Karl Fischer titration.
Table 2. Vapor−Liquid Equilibrium Data for Acetone (1) + Methanol (2) at 101.3 kPaa
Figure 2. Schematic diagram of vapor−liquid equilibrium measurement: 1, heating rod; 2, riser; 3, liquid phase sampling point; 4, vapor liquid boiling chamber; 5, thermometer; 6, condenser; 7, vapor phase sampling point; 8, desiccant; 9, manometer; 10, N2 vessel; 11, valve; 12, buffer.
methyl sulfate, [EMIM][MeSO4]12), where P−x−y data were obtained. Isobaric phase equilibrium data were also determined for acetone + methanol + ILs (e.g., 1-butyl-3-ethylimidazolium trifluoromethane sulfonate, [beim][triflate];10 1-butyl-1-methylpyrrolidinium trifluoromethane sulfonate, [bmpyr][triflate];10 1-ethyl-3-methylimidazolium trifluoromethane sulfonate, [emim][triflate];131-ethyl-3-methylimidazolium dicyanamide, [emim][dca]14), where T−x−y data were obtained. Some of these ILs that contain fluorine in anions easily decompose and release corrosive hydrogen fluoride, and some other ILs are expensive with expensive raw materials and complex synthesis routes. In this work, we studied the ability of three phosphate-based ILs to separate the azeotropic system of acetone + methanol, namely, 1, 3-dimethylimidazolium dimethylphosphate ([MMIM][DMP]), 1-ethyl-3-methylimidazolium diethylphosphate ([EMIM][DEP]), and 1-butyl-3-methylimidazolium dibutylphosphate ([BMIM][DBP]),15,16 which are fluorine-free, inexpensive, and easily prepared. The structures are illustrated in Figure 1. We determined the phase equilibrium data for the binary systems of acetone + methanol, acetone + phosphate ILs and methanol + phosphate ILs, and for the ternary system of acetone + methanol + phosphate ILs, and correlated equilibrium data using Mock’s electrolyte nonrandom two-liquid model.17,18 This work shows such phosphate ILs have better ability to separate acetone + methanol than some ILs reported previously.
T/K
x1
y1
T/K
x1
y1
337.25 337.05 336.05 335.30 334.35 333.60 333.05 332.75 332.15 331.80 331.30 330.95 330.70 330.25 329.85 329.55 329.50
0.000 0.030 0.060 0.093 0.125 0.152 0.177 0.209 0.238 0.267 0.293 0.317 0.341 0.387 0.414 0.444 0.457
0.000 0.068 0.125 0.181 0.232 0.272 0.301 0.334 0.369 0.394 0.428 0.446 0.462 0.503 0.526 0.546 0.556
328.90 328.55 328.40 328.20 328.35 328.20 328.20 328.15 328.05 328.0 327.95 328.20 328.15 328.20 328.20 328.65 329.00
0.487 0.535 0.561 0.590 0.612 0.639 0.676 0.693 0.732 0.746 0.793 0.819 0.842 0.884 0.916 0.948 1.000
0.579 0.608 0.628 0.644 0.663 0.679 0.697 0.715 0.745 0.754 0.789 0.813 0.832 0.870 0.903 0.938 1.000
a Standard uncertainties u are u(x1) = 0.001, u(y1) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
synthesized following our previous work.15,16 ILs were dried for 48 h at 393 K under 0.2 kPa in order to remove volatile byproducts and water before use. The specifications of chemicals used are summarized in Table 1. Apparatus and Procedures. The vapor−liquid equilibrium (VLE) for both binary and ternary systems were measured in a circulation VLE still (a modified Othmer still) at 101.3 kPa, which was described in detail in our previous work.15,16 Schematic diagram of vapor−liquid equilibrium measurement is shown in Figure 2. In a typical experiment, 60 mL of mixtures of liquids were added to the VLE still and heated using a heating rod. The equilibrium temperature was measured by standard thermometer with a standard uncertainty of 0.05 K. The thermometer was calibrated against the ice and steam points of distilled water. The pressure was measured by a manometer with standard uncertainty of 0.1 kPa. The apparatus pressure was kept at 101.3 kPa by controlling the pressure using the vacuum pump connected with the still. For the binary solvent + IL system, every experiment point was obtained from an initial sample of solvent + IL with the highest IL concentration at which different quantities of solvent were added. For the ternary systems, several acetone + IL mixtures of known composition were prepared, and different quantities of methanol + IL mixtures were added in order to keep the
■
EXPERIMENTAL SECTION Materials. The solvents used in the experiment were acetone (AR grade, 100 w ≥ 99.0), methanol (GC grade, 100 w ≥ 99.9). [MMIM][DMP], [EMIM][DEP], and [BMIM][DBP] were B
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
Figure 3. Isobaric VLE diagram for the system of acetone (1) + methanol (2) at 101.3 kPa: ●, this work; ○, Toshihiko Hiaki;20 △, Nurbay Gultekin;21 □, Maria C. Iliuta;21 ◇, Weidong Yan;22 solid lines, correlated using e-NRTL model.
Table 3. Vapor−Liquid Equilibrium Data for Acetone (1) + [MMIM][DMP] (3) at 101.3 kPaa x3
T/K
x3
T/K
0.259 0.247 0.230 0.211 0.187 0.169 0.157 0.150 0.142
330.95 330.70 330.55 330.30 330.10 330.00 329.95 329.85 329.70
0.131 0.112 0.091 0.083 0.069 0.051 0.032 0.021
329.65 329.60 329.55 329.50 329.50 329.45 329.35 329.20
a
Standard uncertainties u are u(x3) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
Table 4. Vapor−Liquid Equilibrium Data for Methanol (2) + [MMIM][DMP] (3) at 101.3 kPaa x3
T/K
x3
T/K
0.259 0.235 0.227 0.22 0.217 0.203 0.19 0.182 0.168 0.162 0.148 0.126 0.119 0.109
371.35 366.85 364.95 363.70 362.80 360.70 358.20 357.55 354.75 354.20 352.15 349.50 348.65 347.45
0.100 0.093 0.089 0.085 0.081 0.077 0.071 0.066 0.060 0.058 0.054 0.050 0.049
346.65 345.35 344.30 343.90 343.55 342.90 342.45 342.25 341.90 341.40 340.95 340.75 340.65
Figure 4. Variation of the activity coefficient of solvent γi and boiling temperature with the mole fraction of IL x3 in the solvent + [MMIM][DMP] systems at 101.3 kPa: △, acetone (1) + [MMIM][DMP] (3); □, methanol (2) + [MMIM][DMP] (3); solid lines, correlated using the e-NRTL model.
a Standard uncertainties u are u(x3) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
scheduled mole fraction of IL in each series. And the equilibrium was reached in about 40 min indicated by the constant temperature. Sample Analysis. For solvent + IL, the IL mole fraction in the liquid phase was gravimetrically determined by drying a known mass of samples until constant weight to remove solvent using a vacuum drying oven (393 K, 0.2 kPa). An analytical balance with a standard uncertainty of 0.001 g was used to weigh the sample. The vapor phase was made up of pure solvent as the effective vapor pressure of ILs is very small.
Figure 5. Boiling temperature with the mole fraction of IL x3 in the methanol (2) + [MMIM][DMP] (3) systems at 101.3 kPa: ○, Cai, F.;16 ●, this work.
For the ternary system, IL mole fraction in the liquid phase was gravimetrically determined in the same way. The compositions of the condensed vapor and the liquid phase were analyzed by gas chromatography (GC112A-1, China) which was equipped with a FID detector and a packed column (3 m × 3 mm). The carrier C
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
Table 5. Vapor−Liquid Equilibrium Data for Acetone (1) + [EMIM][DEP] (3) at 101.3 kPaa
Table 8. Vapor−Liquid Equilibrium Data for Methanol (2) + [BMIM][DBP] (3) at 101.3 kPaa
x3
T/K
x3
T/K
x3
T/K
x3
T/K
0.313 0.308 0.266 0.235 0.229 0.207 0.202 0.175 0.155 0.145
334.20 334.05 333.10 332.70 332.55 331.85 331.75 330.90 330.85 330.55
0.133 0.126 0.124 0.114 0.106 0.103 0.089 0.077 0.068 0.059
330.30 330.25 330.25 330.20 330.10 330.15 330.00 329.85 329.75 329.40
0.218 0.191 0.180 0.155 0.144 0.138 0.133 0.126 0.119 0.105 0.095 0.085 0.077
364.75 360.20 358.95 354.45 352.70 351.90 351.45 350.05 349.20 347.20 345.75 344.35 343.65
0.071 0.066 0.060 0.055 0.050 0.044 0.041 0.037 0.035 0.030 0.028 0.025
343.20 342.25 341.70 341.55 340.85 340.25 339.95 339.80 339.60 339.45 339.30 339.25
a
Standard uncertainties u are u(x3) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
a Standard uncertainties u are u(x1) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
Table 6. Vapor−Liquid Equilibrium Data for Methanol (2) + [EMIM][DEP] (3) at 101.3 kPaa x3
T/K
x3
T/K
0.252 0.214 0.196 0.189 0.168 0.161 0.157 0.147 0.140 0.121 0.116 0.113 0.107 0.091
371.25 364.50 361.40 359.25 356.95 356.10 354.65 353.45 352.10 349.10 348.45 348.15 346.90 345.35
0.086 0.082 0.076 0.064 0.057 0.053 0.049 0.046 0.040 0.035 0.033 0.031 0.027 0.024
344.55 343.80 343.25 341.85 341.85 340.50 340.15 339.80 339.60 339.15 339.00 338.65 338.45 338.25
Figure 6. x′1−y1 diagram for the system of acetone (1) + methanol (2) + [MMIM][DMP] (3) at 101.3 kPa: ◇, x3 = 0; △, x3 ≈ 0.05; ○, x3 ≈ 015, □, x3 ≈ 0.20; solid lines, correlated using e-NRTL model.
■
RESULTS AND DISCUSSION Experimental Data. All VLE data are determined at atmospheric pressure (101.3 kPa). The results for acetone (1) + methanol (2) are shown in Table 2 and Figure 3; for acetone (1) + [MMIM][DMP] (3) in Table 3 and Figure 4; for methanol (2) + [MMIM][DMP] (3) in Table 4 and Figures 4 and 5; for acetone (1) + [EMIM][DEP] (3) in Table 5 and Supporting Information, Figure S1; for methanol (2) + [EMIM][DEP] (3) in Table 6 and Figures S1 and S2; for acetone (1) + [BMIM][DBP] (3) in Table 7 and Figure S3; for methanol (2) + [BMIM][DBP] (3) in Table 8 and Figures S3 and S4; for acetone (1) + methanol (2) + [MMIM][DMP] (3) in Table 9 and Figures 6 and 7; for acetone (1) + methanol (2) + [EMIM][DEP] in Table 10 and Figures S5 and S6 ; for acetone (1) + methanol (2) + [BMIM][DBP] (3) in Table 11 and Figures S7 and S8. To test whether the equilibrium apparatus and the procedure used in this work are precise enough to perform such measurements, we compare the VLE data for binary systems of acetone + methanol with those reported by Hiaki,19 Gultekin,20 Iliuta,21 and Yan22 at the same pressure; the compared results are shown in Figure 3. In addition, the consistency of thermodynamic data was studied using the Herington test23. Equations 1 to 4 are used for this test.
a
Standard uncertainties u are u(x1) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
Table 7. Vapor−Liquid Equilibrium Data for Acetone (1) + [BMIM][DBP] (3) at 101.3 kPaa x3
T/K
x3
T/K
0.343 0.310 0.302 0.276 0.245 0.208 0.194 0.159 0.149 0.136
341.45 338.90 338.45 336.90 335.85 333.55 332.75 331.60 331.45 331.00
0.131 0.117 0.109 0.103 0.097 0.085 0.075 0.067 0.058 0.048
330.95 330.60 330.55 330.35 330.25 330.20 330.15 329.90 329.75 329.65
a
Standard uncertainties u are u(x3) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
gas was nitrogen flowing at 24.4 cm3·min−1, the injection volume was 0.4 μL, and the operating conditions were as follows: injector temperature, 473 K; oven temperature, 403 K; and the detector temperature, 473 K. The calibration curve was carried out by injection of a set of acetone+ methanol mixtures of known composition prepared gravimetrically. The standard uncertainty of the components in the liquid and vapor phase is 0.001.
I=
∫0
1
ln(γ1,exp/γ2,exp) dx1
∑ =∫
0
D
(1)
1
|ln(γ1,exp/γ2,exp)| dx1
(2) DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
Table 9. Vapor−Liquid Equilibrium Data for Acetone (1) + Methanol (2) + [MMIM][DMP] (3) at 101.3 kPaa
a
x3
x1′
y1
T/K
x3
x1′
y1
T/K
0.051 0.051 0.051 0.051 0.051 0.051 0.051 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.050 0.151 0.151 0.147 0.147 0.152 0.152 0.152
0.838 0.817 0.759 0.703 0.640 0.572 0.520 0.460 0.399 0.341 0.268 0.215 0.154 0.097 0.051 0.891 0.801 0.751 0.683 0.614 0.567 0.491
0.929 0.916 0.878 0.838 0.795 0.747 0.705 0.657 0.611 0.558 0.483 0.417 0.327 0.232 0.132 0.987 0.968 0.951 0.930 0.901 0.877 0.835
330.05 330.10 330.35 330.55 330.80 331.25 331.55 332.15 332.50 333.05 334.20 335.25 336.40 338.15 339.35 330.75 331.15 331.50 332.15 333.35 334.25 334.85
0.151 0.151 0.151 0.152 0.152 0.152 0.152 0.152 0.207 0.207 0.207 0.203 0.203 0.203 0.203 0.197 0.197 0.197 0.203 0.203 0.203 0.230
0.442 0.373 0.301 0.260 0.185 0.140 0.090 0.033 0.794 0.724 0.664 0.563 0.485 0.451 0.408 0.334 0.260 0.220 0.175 0.121 0.063 0.030
0.803 0.743 0.685 0.633 0.525 0.439 0.316 0.140 0.976 0.961 0.946 0.913 0.880 0.861 0.834 0.780 0.705 0.648 0.589 0.466 0.291 0.159
335.85 337.10 338.70 339.65 342.05 343.75 346.00 349.55 332.55 333.35 333.20 335.20 336.35 337.15 338.70 340.45 342.65 344.30 345.75 349.15 353.65 356.60
Standard uncertainties u are u(x3) = 0.001, u(x1′ ) = 0.001, u(y1) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
Table 10. Vapor−Liquid Equilibrium Data for Acetone (1) + Methanol (2) + [EMIM][DEP] (3) at 101.3 kPaa
a
x3
x′1
y1
T/K
x3
x′1
y1
T/K
0.054 0.051 0.053 0.052 0.052 0.053 0.052 0.052 0.053 0.053 0.052 0.053 0.053 0.053 0.150 0.151 0.151 0.151 0.148 0.148 0.148 0.148
0.852 0.732 0.703 0.617 0.569 0.483 0.436 0.388 0.327 0.271 0.217 0.157 0.100 0.044 0.825 0.746 0.698 0.659 0.615 0.551 0.493 0.438
0.906 0.822 0.801 0.748 0.713 0.657 0.615 0.574 0.525 0.471 0.407 0.320 0.221 0.109 0.960 0.930 0.911 0.893 0.868 0.835 0.796 0.756
330.00 330.35 330.45 331.05 331.30 331.80 332.35 332.75 333.55 334.30 335.40 336.75 338.45 340.25 332.55 333.40 334.20 334.95 335.45 336.90 337.65 338.90
0.152 0.152 0.152 0.152 0.149 0.149 0.149 0.201 0.201 0.201 0.201 0.202 0.202 0.202 0.202 0.202 0.203 0.203 0.203 0.203 0.203
0.392 0.339 0.271 0.227 0.160 0.114 0.055 0.807 0.753 0.697 0.613 0.574 0.519 0.464 0.427 0.364 0.308 0.249 0.198 0.109 0.077
0.717 0.663 0.582 0.516 0.417 0.316 0.170 0.965 0.949 0.931 0.896 0.879 0.849 0.814 0.785 0.738 0.673 0.609 0.536 0.359 0.258
339.80 341.15 342.95 344.10 346.90 348.15 350.75 335.05 335.95 337.10 338.90 339.85 341.20 342.70 343.65 345.45 347.05 349.50 351.90 356.60 357.05
Standard uncertainties u are u(x3) = 0.001, u(x′1) = 0.001, u(y1) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
D=
J=
|I | × 100 ∑
150θ Tm
system. Thermodynamic consistency is good if the value D is less than J. In this experiment, the D is 0.628 and J is 4.524, so great thermodynamic consistency is shown for this binary system of acetone + methanol. Calculation of Phase Equilibrium. We use the electrolyte nonrandom two-liquid (e-NRTL) model to correlate the VLE data of solvent + IL systems because ILs are organic salts and exhibit ionic character. The e-NRTL model is an extension of the nonrandom two-liquid local composition proposed by Renon and Prausnitz24 from which Chen et al.25 derived a model for
(3)
(4)
In these eqs: γi,exp is the activity coefficient of solvent i calculated from experiment data; Tm is the lowest azeotropic temperature of this binary system; θ is the differential between the highest and lowest azeotropic temperature of this binary E
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
where γi is the activity coefficient of solvent i, the indices calcd and exptl denote the calculated and experimental values respectively, N is the number of experimental data points, and the summations are extended to the whole range of data points. Following this procedure, the 1−3 and 2−3 binary parameters were obtained by assuming ideal behavior of the vapor phase and iteratively solving the VLE conditions expressed in phase equilibrium eq 6 for the molecular solvent.
yP = xiγiPi s i
(6)
In eq 6 yi is the vapor phase mole fraction of solvent i; P is the total pressure in the equilibrium system; Xi is the liquid phase mole fraction of solvent i based on the assumption of total dissociation of IL; γi is the activity coefficient of component i obtained from the electrolyte NRTL model; and Pis is the vapor pressure of pure component i at equilibrium temperature which was calculated using the Antoine equation. The coefficients of the Antoine equation were obtained from literature.12 The result of the optimized binary parameters is summarized in Table 12. The correlated results are shown in Figures 6 and 7, and Supporting Information, Figures S5 to S8. We also correlated the activity coefficients and boiling temperature of acetone (1) + IL (3) and methanol (2) + IL (3) which are shown in Figure 4, Supporting Information, Figures S1 and S3. The mean absolute deviation between the experimental and calculated VLE data is given in Table 13. According to the ternary systems VLE data of acetone (1) + methanol (2) + IL (3), we calculated the relative volatility of acetone (1) to methanol (2) α12 by eq 7. α12 =
γ2p2s
(7)
The experimental activity coefficients γi were obtained from experimental data using eq 6; the calculated data was obtained from the e-NRTL model. The calculated results are shown in Figure 8 and Supporting Information, Figures S9 and S10. Discussion. As shown in Figure 3, the VLE data for binary system of acetone + methanol in this work is similar to that in others’ work which shows the equilibrium apparatus and the procedure used in this work is precise enough to perform such measurements. In addition, the VLE data for the binary system of methanol + ILs ([MMIM][DMP], [EMIM][DEP], and [BMIM][DBP]) in this work are almost identical to that in another work at atmospheric pressure (101.3 kPa),16 as shown in Figure 5 and Supporting Information, Figures S2 and S4. From Figure 4 and Supporting Information, Figures S1 and S3, it is observed that γ1 decreases along with the addition of these ILs while the γ2 shows the reverse trend. Also the boiling temperature of both acetone (1) and methanol (2) increases along with the addition of these ILs, but the increasing amplitude for methanol is much more than that for acetone. As shown in Figures 6 and 7 and Supporting Information, Figures S5 to S8, the diagram of x1′ − y1 shifts up with the addition of ILs, and this means the volatility of acetone is increased. Figure 8 and Figures S9 and S10 indicate that the addition of ILs increases the relative volatility of acetone to methanol. The α12 reaches more than 1 in the whole range of solvent composition when the x3 reaches 0.05; this means that the azeotropic point can be totally eliminated at this situation. This is mainly because acetone is less polar than methanol as shown by the ETN values 0.355 and 0.762, respectively,27 and the ILs can be completely miscible in methanol but
Figure 7. Temperature−composition diagram for acetone (1) + methanol (2) + [MMIM][DMP] (3) at 101.3 kPa for three IL mole fractions: ▲, x1′ experimental for x3 ≈ 0.05; △, y1 experimental for x3 ≈ 0.05; ●, x′1 experimental for x3 ≈ 0.15; ○, y1 experimental for x3 ≈ 0.15; ■, x1′ experimental for x3 ≈ 0.20; □, y1 experimental for x3 ≈ 0.20; solid lines, calculated with the e-NRTL model; dotted lines, calculated IL-free system.
single-solvent + electrolyte systems for liquid phase activity coefficients. Mock et al.17,18 extended it to mixed-solvent + electrolyte systems by neglecting the long-range interaction contribution term. The e-NRTL model has been proven suitable for IL-containing systems by the work of others.13−15 The e-NRTL model produces expressions for the liquid-phase activity coefficients of acetone (1) and methanol (2) in a binary or ternary system containing ionic liquids. These equations can be obtained in the work of Vercher et al.26 We should determine nine binary parameters for all of the solvent + solvent and solvent + IL in the systems to represent the phase equilibrium of mixedsolvent + electrolyte systems. Six of them are energy parameters (△g1,2, △g2,1, △g1,3, △g3,1, △g2,3, △g3,2) and three are nonrandomness factors (α1,2 = α2,1α1,3 = α3,1α2,3 = α3,2). The 1−2 binary acetone−methanol parameters were obtained from the literature.12 And the 1−3 and 2−3 binary parameters are obtained by adjusting the VLE ternary data in Tables 9−11 through the minimization of the objective function (F): 2 2 ⎡ γ1calcd ⎤ ⎡ γ2calcd ⎤ ⎢ ⎥ ⎢ ⎥ + 1− F=∑ 1− ⎢ γ1exptl ⎥⎦ ⎢⎣ γ2exptl ⎥⎦ N ⎣
γ1p1s
(5) F
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
Table 11. Vapor−Liquid Equilibrium Data for Acetone (1) + Methanol (2) + [BMIM][DBP] (3) at 101.3 kPaa
a
x3
x1′
y1
T/K
x3
x1′
y1
T/K
0.051 0.051 0.051 0.051 0.049 0.049 0.049 0.049 0.049 0.052 0.052 0.052 0.052 0.052 0.052 0.150 0.152 0.149 0.149 0.147 0.147 0.147
0.887 0.807 0.769 0.676 0.591 0.569 0.495 0.441 0.381 0.283 0.246 0.214 0.142 0.078 0.057 0.886 0.842 0.787 0.721 0.643 0.568 0.478
0.906 0.840 0.820 0.769 0.721 0.692 0.647 0.606 0.554 0.471 0.421 0.381 0.286 0.182 0.131 0.945 0.927 0.905 0.879 0.847 0.806 0.746
329.75 329.80 329.85 330.00 330.25 330.45 330.75 331.30 332.20 333.20 334.15 335.05 336.30 338.10 338.65 333.55 334.15 334.70 335.55 336.45 337.15 338.25
0.148 0.148 0.148 0.148 0.151 0.151 0.151 0.151 0.199 0.196 0.198 0.198 0.199 0.196 0.196 0.196 0.196 0.196 0.196 0.195 0.197 0.196
0.418 0.356 0.309 0.247 0.197 0.162 0.091 0.029 0.902 0.830 0.756 0.696 0.601 0.537 0.463 0.393 0.361 0.287 0.214 0.147 0.101 0.061
0.702 0.656 0.607 0.537 0.474 0.407 0.263 0.099 0.955 0.927 0.907 0.883 0.855 0.813 0.775 0.721 0.696 0.631 0.545 0.442 0.336 0.221
339.15 340.00 341.20 342.75 344.55 345.45 348.10 350.95 335.55 337.45 338.50 340.05 340.95 342.20 343.65 344.75 345.75 348.05 350.00 351.75 353.95 356.40
Standard uncertainties u are u(x3) = 0.001, u(x1′ ) = 0.001, u(y1) = 0.001, u(T) = 0.25 K, u(p) = 0.1 kPa.
Table 12. Estimated Values of Nonrandomness Factors, αi,j, and Energy Parameters, Δgi,j and Δgj,i, for the Electrolyte NRTL Model i component acetone acetone acetone acetone methanol methanol methanol a
j component methanol [MMIM][DMP] [EMIM][DEP] [BMIM][DBP] [MMIM][DMP] [EMIM][DEP] [BMIM][DBP]
Δgi,j/J·mol−1
αi,j a
0.300 0.225b 0.207b 0.210b 0.07b 0.019b 0.032b
a
924.2 19722.14b 20950.39b 23110.32b 49007.4b 7285.8b 78360.4b
observed that the salting-out effect of these three phosphate ILs follows [MMIM][DMP] > [EMIM][DEP] > [BMIM][DBP]. The mean absolute deviation between the experimental and calculated VLE data which is given in Table 13 and the e-NRTL model is able to properly correlate the experimental VLE data. According to the e-NRTL model, the azeotropic points for acetone + methanol system at 101.3 kPa disappear at x3 = 0.022, 0.025, and 0.029 for [MMIM][DMP], [EMIM][DEP], and [BMIM][DBP], respectively. So the ability of these three phosphate ILs as the entrainers for acetone + methanol system is [MMIM][DMP] > [EMIM][DEP] > [BMIM][DBP], determined by their different polarity. Compared with other ILs used for the acetone + methanol system, the data is x3 = 0.031 for [EMIM][DCA] at 100 kPa;14 x3 = 0.078 for [EMIM][triflate] at 100 kPa;13 x3 = 0.073 for [BEIM][triflate] at 100 kPa;10 x3 = 0.079 for [bmpyr][triflate] at 100 kPa.10 Furthermore, these three ILs produced no crossover effect over the acetone + methanol system, which appear for [EMIM][triflate],
Δgj,i/J·mol−1 863.1a −6484.11b −7527.98b −8520.73b −30703.7b −16799.0b −50805.3b
FromVercher et al.10 bFrom this work.
partially miscible in acetone. The “attractive interaction” between ILs and methanol is stronger than that between ILs and acetone; therefore, the addition of IL can lead to a great salting-out effect for acetone. From these figures it can be
Table 13. Mean Absolute Deviations, δy and δT, and Standard Deviations, σy and σT, between Experimental and Calculated Values of the Vapor-Phase Mole Fractions and the Equilibrium Temperatures
a
system
δya
σyb
δTc/K
σTd/K
acetone + methanol acetone + [MMIM][DMP] methanol + [MMIM][DMP] acetone + methanol + [MMIM][DMP] acetone + [EMIM][DEP] methanol + [EMIM][DEP] acetone + methanol + [EMIM][DEP] acetone + [BMIM][DBP] methanol + [BMIM][DBP] acetone + methanol + [BMIM][DBP]
0.007
0.009
0.005
0.007
0.006
0.008
0.009
0.012
0.25 0.40 0.30 0.25 0.30 0.30 0.45 0.20 0.30 0.30
0.35 0.55 0.45 0.35 0.45 0.45 0.65 0.35 0.45 0.45
δy = (1/N)∑|yexptl − ycalcd|. bσy = [1/(N − 1)][∑(yexptl − ycalcd)2]1/2. cδT = (1/N)∑|Texptl − Tcalcd|. dσT = [1/(N − 1)][∑(Texptl − Tcalcd)2]1/2.
G
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
■
Article
AUTHOR INFORMATION
Corresponding Author
*Tel./Fax: +86 10 6443 3570. E-mail:
[email protected]. *E-mail:
[email protected]. Funding
This work is financially supported by National Natural Science Foundation of China (21176021, 21276020) and Basic Scientific Research Foundation for Chinese Universities (JD1301). We extend our appreciation to the Deanship of Scientific Research at King Saud University for funding the work through the research group project No. RGP-089.
Figure 8. Variation of the relative volatility α12 between acetone (1) and methanol (2) with the acetone molar fraction x1′ for different [MMIM][DMP] (3) molar fractions: △, x3 ≈ 0.05; ○, x3 ≈ 0.15; □, x3 ≈ 0.20; solid lines, correlated using e-NRTL model; dotted line, calculated IL-free system.
Notes
The authors declare no competing financial interest.
■
Table 14. Activity Coefficients at Infinite Dilution γ∞ i for ILs in Acetone and Methanol Calculated by the Electrolyte NRTL Model and COSMO-RSa e-NRTL
a
COSMO-RS
ILs
γ∞ i for acetone
γ∞ i for methanol
γ∞ i for acetone
γ∞ i for methanol
[MMIM][DMP] [EMIM][DEP] [BMIM][DBP]
0.129 0.126 0.121
1.864 1.832 1.383
0.144 0.123 0.125
1.912 1.800 1.425
Standard uncertainties u(γ∞ i ) = 0.001.
[BEIM][triflate], and [bmpyr][triflate].10,13 These three phosphate ILs show great potential as the entrainers for the acetone + methanol system. This is consistent with the activity coefficients at infinite dilution γ∞ i which was calculated by COSMO-RS;28 the calculated data are shown in Table 14.
■
CONCLUSIONS In this work, isobaric vapor−liquid equilibrium (VLE) at atmospheric pressure (101.3 kPa) for the binary systems of acetone + methanol, acetone + phosphate ionic liquids (ILs), and methanol + phosphate ILs, and for the ternary system of acetone + methanol + phosphate ILs are measured using a circulation VLE still, and the electrolyte nonrandom two-liquid (e-NRTL) model is used for correlating the data. The addition of these three phosphate ILs to the acetone + methanol system produces noticeable salting-out on acetone, and the azeotropic point is eliminated at x3 ≈ 0.05 for these three phosphate ILs. The least IL molar fractions for eliminating the azeotropic point are 0.022, 0.025, and 0.029 for [MMIM][DMP], [EMIM][DEP], and [BMIM][DBP], respectively. The least IL molar fraction for eliminating the azeotropic point is less than the molar fractions of other ILs that were tested from the e-NRTL model calculation. From the mean absolute deviation between the experimental and calculated VLE data, we can find the VLE data can be correlated by using the e-NRTL model. This study shows that the phosphate-based ILs can be used for separation of acetone and methanol as potential entrainer.
■
REFERENCES
(1) Zhong, W. H.; Wang, H. X. Disposal of industrial waste during the production of furfural. Chem. Eng. 2004, 3, 1002−1124. (2) Wu, L. F.; Cao, L. R.; Bai, L.; Li, Y.; Jia, R.; Li, Y. Study on the process of producing acetone and methanol from the water phase of Fischer−Tropsch synthesis. J. Fuel Chem. Technol. 2008, 36, 478−483. (3) Lei, L. A combined process of salt extraction with azeotropic distillation for separation of acetic acid−water mixture. Acetate Petrochem. Technol. 1990, 11, 749−752. (4) Luyben, L. W. Comparison of extractive distillation and pressureswing distillation for acetone−methanol separation. Ind. Eng. Chem. Res. 2008, 47, 2696−2707. (5) Modia, G.; Lang, P. Feasibility of separation of ternary mixtures by pressure swing batch distillation. Ind. Eng. Chem. Res. 2010, 49, 3785− 3793. (6) Li, Q.; Xing, F.; Lei, Z.; Wang, B.; Chang, Q. Isobaric vapor−liquid equilibrium for isopropanol + water + 1-ethyl-3-methylimidazolium tetrafluoroborate. J. Chem. Eng. Data 2007, 53, 275−279. (7) Li, Q.; Zhang, J.; Lei, Z.; Zhu, J.; Wang, B.; Huang, X. Isobaric vapor− liquid equilibrium for (propan-2-ol + water + 1-butyl-3methylimidazolium tetrafluoroborate). J. Chem. Eng. Data 2009, 54, 2785−2788. (8) Zhong, L. P.; Liu, J. Q.; Jia, Y. L. Separation of methanol−acetone binary azeotrope by extractive distillation. Chem. Ind. Eng. 2005, 2, 211− 215. (9) Arlt, W.; Seiler, M.; Jork, C.; Schneider, T. Ionic liquids as selective additive for the separation of close-boiling or azeotropic mixtures. WO2002/074718 A2, 2002-09-26. (10) Orchillés, A. V.; Miguel, P. J.; Llopis, F. J.; Vercher, E.; MartinezAndreu, A. Influence of some ionic liquids containing the trifluoromethanesulfonate anion on the vapor−liquid equilibria of the acetone + methanol system. J. Chem. Eng. Data 2011, 56, 4430−4435. (11) Kurzin, A. V.; Evdokimov, A. N.; Antipina, V. B.; Pavlova, O. S. Liquid−vapor equilibrium in the system acetone−methanol−Nbutylpyridinium hexafluorophosphate. Russ. J. Appl. Chem. 2007, 12, 2049−2050. (12) Matsuda, H.; Liebert, V.; Tochigi, K.; Gmehling, J. Influence of sulfate-based anion ionic liquids on the separation factor of the binary azeotropic system acetone + methanol. Fluid Phase Equilib. 2013, 340, 27−30. (13) Orchillés, A. V.; Miguel, P. J.; González-Alfaro, V.; MartínezAndreu, A. Ionic liquids as entrainers in extractive distillation: Isobaric vapor−liquid equilibria for acetone + methanol + 1-ethyl-3-methylimidazolium trifluoromethanesulfonate. J. Chem. Eng. Data 2007, 52, 141−147. (14) Orchillés, A. V.; Miguel, P. J.; González-Alfaro, V.; Vercher, E.; Martínez-Andreu, A. 1-Ethyl-3-methylimidazolium dicyanamide as a very efficient entrainer for the extractive distillation of the acetone+ methanol system. J. Chem. Eng. Data 2012, 57, 394−399. (15) Chen, X.; Cai, F.; Wu, X.; Asumana, C.; Yu, G. Isobaric vapor− liquid equilibrium for methanol + dimethyl carbonate + 1-butyl-3methylimidazolium dibutylphosphate. J. Chem. Eng. Data 2013, 58, 1186−1192.
ASSOCIATED CONTENT
S Supporting Information *
Additional figures as described in the text. This material is available free of charge via the Internet at http://pubs.acs.org. H
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
Journal of Chemical & Engineering Data
Article
(16) Cai, F.; Wu, X.; Chen, C.; Chen, X.; Asumana, C.; Haque, M. R.; Yu, G. Isobaric vapor−liquid equilibrium for methanol + dimethyl carbonate + phosphoric-based ionic liquids. Fluid Phase Equilib. 2013, 352, 47−53. (17) Mock, B.; Evans, L. B.; Chen, C. C. Phase equilibria in multiplesolvent electrolyte systems: A new thermodynamic model. Proc. Summer Comput. Simul. Conf. 1984, 558−562. (18) Mock, B.; Evans, L. B.; Chen, C. C. Thermodynamic representation of phase equilibria of mixed-solvent electrolyte systems. AIChE J. 1986, 32, 1655−1664. (19) Hiaki, T.; Kurihara, K.; Kojima, K. Vapor−liquid equilibria for acetone + chloroform + methanol and constituent binary systems at 101.3 kPa. J. Chem. Eng. Data 1994, 39, 714−719. (20) Gultekin, N. Vapor-liquid equilibria at 1 atm for ternary and quaternary systems composed of acetone, methanol, 2-propanol, and 1propanol. J. Chem. Eng. Data 1990, 35, 132−136. (21) Iliuta, M. C.; Thyrion, F. C. Vapour−liquid equilibrium for the acetone−methanol−inorganic salt system. Fluid Phase Equilib. 1995, 103, 257−284. (22) Yan, W.; Topphoff, M.; Zhu, M.; Gmehling, H. Measurement and correlation of isobaric vapor−liquid equilibrium data for the system acetone + methanol + zinc chloride. J. Chem. Eng. Data 1999, 44, 314− 318. (23) Peng, Y.; Liang, J.; Ping, L. J.; Mao, J. W.; HU, H. Thermodynamic consistency for the vapor−liquid equilibrium data of associating system. Acta Pet. Sin. (Petroleum Processing Section) 2009, 5, 019. (24) Renon, H.; Prausnitz, J. M. Local compositions in thermodynamic excess functions for liquid mixtures. AIChE J. 1968, 14, 135−144. (25) Chen, C. C.; Britt, H. I.; Boston, J. F.; Evans, L. B. Local composition model for excess Gibbs energy of electrolyte systems. Part I: Single solvent, single completely dissociated electrolyte systems. AIChE J. 1982, 28, 588−596. (26) Vercher, E.; Rojo, F. J.; Martínez-Andreu, A. Isobaric vapor− liquid equilibria for 1-propanol + water + calcium nitrate. J. Chem. Eng. Data 1999, 44, 1216−1221. (27) Iliuta, M. C.; Thyrion, F. C. Salt effects on vapor-liquid equilibrium of acetone-methanol system. Fluid Phase Equilib. 1996, 121, 235−252. (28) Kato, R.; Gmehling, J. Measurement and correlation of vapor− liquid equilibria of binary systems containing the ionic liquids [EMIM][(CF 3 SO 2 ) 2 N], [BMIM][(CF 3 SO 2 ) 2 N], [MMIM][(CH3)2PO4] and oxygenated organic compounds respectively water. Fluid Phase Equilib. 2005, 231, 38−43.
I
DOI: 10.1021/je5007373 J. Chem. Eng. Data XXXX, XXX, XXX−XXX
