- Email: [email protected]
Partial miscibility behavior of the ternary mixture carbon dioxide + 1-methylnaphthalene + acetone
Fluid Phase Equilibria 198 (2002) 29–36
Partial miscibility behavior of the ternary mixture carbon dioxide + 1-methylnaphthalene + acetone Lydia E. Gutiérrez M, K.D. Luks∗ Department of Chemical Engineering, University of Tulsa, Tulsa, OK 741043189, USA Received 3 January 2001; accepted 12 October 2001
Abstract The liquid–liquid–vapor (llg) partial miscibility behavior of the mixture carbon dioxide +1-methylnaphthalene + acetone was experimentally studied by the use of a visual cell (stoichiometric) technique. Phase compositions and molar volumes of the two liquid phases in equilibrium are reported as functions of temperature and pressure within the llg region. The addition of the co-solvent acetone extends the three-phase llg region in pressure–temperature space from the binary carbon dioxide + n-1-methylnaphthalene llg locus downward in pressure, towards the vapor pressure curve of the co-solvent acetone. The three-phase region is bounded from above by an upper critical end point (UCEP) (l-l=g) locus. The presence of the co-solvent acetone significantly enhances the amount of the solute 1-methylnaphthalene taken up by the second (carbon-dioxide-rich) liquid phase. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Data; Mixture; Liquid–liquid–vapor equilibria; Extraction; Co-solvent
1. Introduction Our group has been experimentally studying liquid–liquid–vapor (llg) phase equilibria in well-defined binary and ternary systems. These systems are typically composed of one or more gas solvents (e.g. carbon dioxide, ethane, propane, nitrous oxide and/or xenon) and one or more relatively non-volatile solutes chosen from homologous series of n-alkanes, n-alkylbenzenes and/or 1-alkanols. The primary goals of these studies are to map out the patterns of the multiphase equilibria of these prototype mixtures in thermodynamic phase space and to generate data that support the prediction of phase equilibria within and near the regions of llg partial immiscibility. Successful supercritical extraction at temperatures near and above the critical temperature of a gas solvent is often accompanied by the presence of llg phase equilibria at lower temperatures [1,2], be it Abbreviations: AAD, average absolute deviation; LCEP, lower critical end point; UCEP, upper critical end point ∗ Corresponding author. Tel.: +1-918-631-2226; fax: +1-918-631-3268. E-mail address: [email protected] (K.D. Luks). 0378-3812/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 3 8 1 2 ( 0 1 ) 0 0 7 3 5 - X
30
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
actual or metastable (masked by the formation of a solid solute phase). Therefore, models purporting to describe supercritical extraction should be capable of describing llg phase equilibria as well. Additionally, there is the attractive prospect of carrying out subcritical extraction in an llg configuration, wherein a solute is selectively removed from the solute-rich liquid phase by the gas-solvent-rich liquid phase [the extract phase] in the presence of a gas-solvent-rich vapor phase. Compared to supercritical extraction, subcritical extraction would operate at more modest pressures and temperatures. Investigators of supercritical extraction processes have addressed the problem of enriching the extract phase in the solute, and the addition of a modifier (co-solvent) to the extraction system has been carried out with some success. In the case of the gas solvent carbon dioxide, the co-solvent is generally a lighter molecular species of considerable dipole moment such as methanol (1.7 Debye) or acetone (2.88 Debye). In two recent studies, we examined the effect of the co-solvents methanol and acetone, respectively on the llg phase equilibria of the partially miscible binary mixture carbon dioxide + n-tetradecane [3,4]. In this present study, we report the llg phase equilibria of the ternary mixture carbon dioxide + 1-methylnaphthalene + acetone, initiating an inquiry whether common polar co-solvents can enhance the concentration of a heavy aromatic hydrocarbon in the carbon dioxide-rich-liquid [l2 , or extract] phase. The data were obtained using a visual cell (“stoichiometric”) experimental technique. The boundaries of the llg region were mapped out in pressure–temperature space and then the compositions and the molar volumes of the two liquid phases were found along the 288.15 and 298.15 K isotherms. The effectiveness of acetone as a co-solvent is discussed in the context of the binary llg system carbon dioxide +1-methylnaphthalene.
2. Experimental Fall and Luks [5] and Jangkamolkulchai and Luks [6] give a detailed description of the experimental apparatus. The procedures for performing the llg studies are described in Fall et al. [7] and Jangkamolkulchai and Luks [6]. The experimental apparatus employs a stoichiometric approach, wherein known amounts of 1-methylnaphthalene and acetone are added to a volumetrically calibrated visual (glass) equilibrium cell. The total volume of the cell is typically 7–9 cm3 . The cell vapor space is thoroughly flushed with carbon dioxide gas, after which measured amounts of the carbon dioxide gas are added to the cell from a high-pressure bomb. The cell contents are brought to equilibrium by a magnetically actuated steel ball stirrer mechanism. With the use of mass balances, the compositions and molar volumes of the phases in a ternary llg system at a given temperature and pressure can be determined from “conjugate” measurements. Our experience [8] indicates that the presence of 1-methylnaphthalene in the vapor phase is negligible. It was assumed that the presence of acetone was also masswise negligible in the vapor phase. Therefore, the mass balance calculations required a set of conjugate runs in which the l1 and l2 phases were respectively dominant in terms of volumetric fraction. l1 - and l2 -dominant run data at a specific temperature and pressure enabled quantitative determination of the compositions and molar volumes of these phases. Since the vapor phase was assumed to be pure carbon dioxide, its stoichiometry was calculated from tables [9] rather than measured here. The visual cell temperature was measured with a Pt-resistance thermometer to an estimated accuracy of ±0.02 K. Pressures were measured to ±0.07 bar with pressure transducers that were frequently calibrated with a dead-weight gauge. Phase volumes in the calibrated visual cell were determined with the aid of a cathetometer to an accuracy of at least 0.01 cm3 .
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
31
The acetone and 1-methylnaphthalene were purchased from Aldrich. The acetone had a stated purity of >99.9%. The 1-methylnaphthalene had a stated purity of 98%. Chromatographic analysis indicated purities consistent with these stated purities. These chemicals were used without further purification. The carbon dioxide was purchased from Air Products and Chemicals Inc as “Coleman grade” with a stated purity of 99.99 mol%. The carbon dioxide was first transferred to an evacuated storage bomb immersed in an ice bath. The vapor phase was vented to remove any light impurities. The purity was verified by liquefying the carbon dioxide at 298.15 K. The vapor pressure at this temperature was within 0.04 bar of that reported by Vargaftik [10]. In addition, the critical temperature and pressure were within 0.06 K and 0.07 bar of those reported by Vargaftik [10].
3. Results In Fig. 1, portions of the boundaries of the llg surface of the ternary system carbon dioxide + 1-methylnaphthalene + acetone are shown. The surface extends from the binary llg locus of carbon dioxide+1-methylnaphthalene downward in pressure with the addition of the third component acetone (towards the acetone vapor pressure curve). The surface is terminated from above by an upper critical end point locus (UCEP, or l-l=g), given in Table 1. There appears to be, based on a cursory scanning exercise, no llg partial immiscibility for the binary system carbon dioxide + acetone. This suggests that, the continued addition of acetone to the binary mixture carbon dioxide + 1-methylnaphthalene will eventually lead to a termination of llg partial miscibility, its terminus being a critical end point locus of the type l=l-g. That locus would eventually intersect the critical end point locus reported in Table 1 at a tricritical point. In contrast to the earlier study of the system carbon dioxide + n-tetradecane + acetone, the partial immiscibility in the system carbon dioxide + 1-methylnaphthalene + acetone is more pronounced. The l=l-g locus would occur at much higher concentrations of acetone relative to 1-methylnaphthalene. We did not study mixtures with such high acetone concentrations, since the role of the acetone from a practical standpoint, is that of an additive co-solvent. The boundary points were determinable to ±0.1 K and ±0.07 bar and are considered to be good to ±0.1 K and ±0.10 bar.
Fig. 1. Deviation of the binary llg and ternary l-l=g boundaries of the ternary llg surface from the vapor pressure curve for pure 0 carbon dioxide in pressure–temperature space for the system carbon dioxide + 1-methylnaphthalene + acetone. PCO is the vapor 2 pressure of carbon dioxide as represented by the correlation ln(P r ) = 6.576183 (l-l/Tr ). (䊉) l-l=g critical end points.
32
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
Table 1 Temperature T and pressure P raw data for the l-l=g boundary loci for the ternary mixture carbon dioxide+1-methylnaphthalene+ acetone UCEP (l-l=g) T (K)
P (bar)
308.82 309.76 309.93 310.05 309.98 310.07 311.47 311.64 312.20 312.17 312.67 313.38 313.68 314.36 314.42 315.36
79.30 80.23 80.35 80.42 80.43 80.46 81.44 81.85 82.35 82.36 82.77 83.47 83.76 84.38 84.44 85.39
A regressed straight line through the data in Table 1 yields an average absolute deviation (AAD) of about 0.05 bar. The llg data for the ternary system carbon dioxide+1-methylnaphthalene+acetone along the isotherms 288.15 and 298.15 K are given in Figs. 2–6 and Tables 2 and 3. The ternary mole fractions are estimated to be accurate to ±0.004 and the molar volumes to ±0.6 cm3 /mol. These are conservative estimates, based on AAD of the raw data from simple smoothed curves and our experience at being able to generate
Fig. 2. Mole fraction of carbon dioxide as a function of pressure P for the l1 and l2 phases for the llg region of carbon dioxide + 1-methylnaphthalene + acetone at 288.15 K.
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
33
Fig. 3. Molar volume v as a function of pressure P for the l1 and l2 phases for the llg region of carbon dioxide + 1-methylnaphthalene + acetone at 288.15 K.
Fig. 4. Mole fraction of carbon dioxide as a function of pressure P for the l1 and l2 phases for the llg region of CO2 + 1methylnaphthalene + acetone at 298.15 K.
Fig. 5. Molar volume v as a function of pressure P for the l1 and l2 phases for the llg region of carbon dioxide + 1-methylnaphthalene + acetone at 298.15 K.
34
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
Fig. 6. Concentration c of 1-methylnaphthalene in liquid phase 2 as a function of pressure P for the llg region of the system carbon dioxide + 1-methylnaphthalene + acetone at 288.15 and 298.15 K. Quadratic curve representations are offered. Table 2 Liquid phase l1 and l2 compositions (mole fractions x) and molar volumes v as a function of pressure P for the llg region of carbon dioxide + 1 methylnaphthalene + acetone at 288.15 K l1 Phase
l2 Phase
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
49.61 49.18 49.07 48.69 48.54 48.36 48.21 47.92 47.72 47.43
0.4537 0.4600 0.4729 0.4721 0.4724 0.4743 0.4833 0.4890 0.4917 0.4987
0.0000 0.0091 0.0105 0.0132 0.0163 0.0204 0.0238 0.0357 0.0385 0.0440
95.5 95.8 93.1 92.2 92.4 91.7 90.6 89.2 89.1 88.4
49.61 49.16 48.53 48.43 48.30 47.84 47.60
0.9793 0.9693 0.9644 0.9628 0.9598 0.9527 0.9455
0.0000 0.0047 0.0137 0.0156 0.0171 0.0218 0.0259
53.5 53.9 53.6 53.6 54.1 53.9 53.8
Table 3 Liquid phase l1 and l2 compositions (mole fractions x) and molar volumes v as a function of pressure P for the llg region of carbon dioxide + 1 methylnaphthalene + acetone at 298.15 K l1 Phase
l2 Phase
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
62.46 62.03 61.97 61.20 61.08 60.83 59.61 59.39 59.38
0.4736 0.4831 0.4823 0.4949 0.4934 0.5009 0.5244 0.5224 0.5307
0.0000 0.0099 0.0104 0.0214 0.0227 0.0249 0.0409 0.0556 0.0434
93.6 92.6 93.0 90.3 91.0 90.1 86.3 86.0 85.7
62.46 61.95 61.07 61.04 60.88 60.87 60.00 59.93 59.43 59.29
0.9851 0.9785 0.9716 0.9695 0.9695 0.9670 0.9555 0.9492 0.9512 0.9422
0.0000 0.0042 0.0091 0.0118 0.0104 0.0133 0.0224 0.0258 0.0276 0.0286
58.3 57.8 57.4 57.7 57.1 57.4 56.5 56.2 55.7 55.9
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
35
Table 4 Concentration c of 1-methylnaphtalene, mole ratio of 1-methylnaphtalene to CO2 in liquid phase 2, and mole fraction of acetone in liquid phase 2 as a function of pressure at temperatures 288.15 and 298.15 K for the llg region of carbon dioxide + 1-methylnaphtalene + acetone Liquid phase 2 at 288.15 K
Liquid phase 2 at 298.15 K
P (bar)
c(1-MN) mol cm−3
r(1-MN/CO2 )
x (C3 H6 O)
P (bar)
c(1-MN) mol cm−3
r(1-MN/CO2 )
x (C3 H6 O)
49.61 49.16 48.53 48.51 48.43 48.30 47.84 47.60
3.875E-04 4.818E-04 4.081E-04 4.475E-04 4.023E-04 4.283E-04 4.727E-04 5.323E-04
2.118E-02 2.677E-02 2.270E-02 2.539E-02 2.241E-02 2.414E-02 2.673E-02 3.029E-02
0.0000 0.0047 0.0137 0.0133 0.0156 0.0171 0.0218 0.0259
62.46 61.95 61.07 61.04 60.99 60.88 60.87 60.00 59.93 59.43 59.29
2.555E-04 3.000E-04 3.368E-04 3.252E-04 3.624E-04 3.529E-04 3.425E-04 3.898E-04 4.456E-04 3.812E-04 5.210E-04
1.511E-02 1.780E-02 1.999E-02 1.930E-02 2.151E-02 2.095E-02 2.033E-02 2.314E-02 2.645E-02 2.263E-02 3.092E-02
0.0000 0.0042 0.0091 0.0118 0.0133 0.0104 0.0133 0.0224 0.0258 0.0276 0.0286
reproducible data using the stoichiometric technique described herein. The AADs for the l1 data are 0.003 for the carbon dioxide mole fractions and 0.4 cm3 /mol for the liquid phase molar volume. For the l2 data, these AADs are 0.001 and 0.2 cm3 /mol, respectively. 4. Discussion A successful co-solvent should enhance the capacity of the l2 phase for the solute of interest. Table 4 suggests that acetone is an effective co-solvent here, in the context of 1-methylnaphthalene in a carbon dioxide medium. The absolute molar concentration c1-MN of the solute 1-methylnaphthalene is significantly increased as the ternary system enriches with acetone, at both 288.15 and 298.15 K. The measure r, which is a ratio of 1-methylnaphthalene to carbon dioxide also increases significantly. These enhancements occur with relatively modest additions (x < 0.03) of the co-solvent acetone to the overall phase mixture. List of symbols c molar concentration g gas or vapor phase l, l1 , l2 liquid phases l-l=g a three-phase critical end point where the less dense liquid phase (l2 ) and the vapor phase (g) are critically identical; also referred to as the UCEP l=l-g a three-phase critical end point where the two liquid phases (l1 and l2 ) are critically identical; also referred to as the LCEP P pressure r mole ratio of n-1-methylnaphthalene to carbon dioxide T temperature
36
v x
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
molar volume mole fraction
Acknowledgements The support of this research was provided by the National Science Foundation (Grant No. CTS-9730853). The apparatus used is a part of the PVTx Laboratory at the University of Tulsa and was purchased with funds provided by several industries, the University of Tulsa and a National Science Foundation specialized equipment (Grant No. CPE-8104650). References [1] S.C. Hwang, Y.N. Lin, S.W. Hopke, R. Kobayashi (Eds.), Relation of liquid–Liquid Equilibrium Behavior at Low Temperatures to Vapor–Liquid Equilibria Behavior at High Temperatures and Elevated Pressures, in: Proceedings of the 57th Annual GPA Convention, 1978, pp. 7–12. [2] R.M. Lansangan, A. Jangkamolkulchai, K.D. Luks, Binary vapor–liquid equilibria behavior in the vicinity of liquid–liquid–vapor loci, Fluid Phase Equilb. 36 (1987) 49–66. [3] C.M. Foreman, K.D. Luks, Partial miscibility behavior of the ternary mixture carbon dioxide + n-tetradecane + methanol, J. Chem. Eng. Data 45 (2000) 334–337. [4] C.M. Foreman, K.D. Luks, Partial miscibility behavior of the ternary mixture carbon dioxide + n-tetradecane + acetone, Fluid Phase Equilb. 169 (2000) 65–73. [5] D.J. Fall, K.D. Luks, Phase equilibria behavior of the systems carbon dioxide + n-dotriacontane and carbon dioxide + n-docosane, J. Chem. Eng. Data 29 (1984) 413–417. [6] A. Jangkamolkulchai, K.D. Luks, Partial miscibility behavior of the methane + ethane + n-docosane and the methane + ethane + n-tetradecylbenzene ternary mixtures, J. Chem. Eng. Data 34 (1989) 92–99. [7] D.J. Fall, J.L. Fall, K.D. Luks, Liquid–liquid–vapor immiscibility limits in carbon dioxide + n-paraffin mixtures, J. Chem. Eng. Data 30 (1985) 82–88. [8] L.E. Gutiérrez M, K.D. Luks, Three-phase liquid–liquid–vapor equilibria of the binary mixture carbon dioxide + 1-methylnaphthalene, J. Chem. Eng. Data, in press. [9] S. Angus, B. Armstrong, B.K.M. de Reuck (Eds.), International Thermodynamic Tables of the Fluid State Carbon Dioxide, Pergamon Press, Oxford, 1976. [10] N.B. Vargaftik, Tables on the Thermophysical Properties of Liquids and Gases, 2nd Edition, Wiley, New York, 1975, pp. 167–168.
Partial miscibility behavior of the ternary mixture carbon dioxide + 1-methylnaphthalene + acetone Lydia E. Gutiérrez M, K.D. Luks∗ Department of Chemical Engineering, University of Tulsa, Tulsa, OK 741043189, USA Received 3 January 2001; accepted 12 October 2001
Abstract The liquid–liquid–vapor (llg) partial miscibility behavior of the mixture carbon dioxide +1-methylnaphthalene + acetone was experimentally studied by the use of a visual cell (stoichiometric) technique. Phase compositions and molar volumes of the two liquid phases in equilibrium are reported as functions of temperature and pressure within the llg region. The addition of the co-solvent acetone extends the three-phase llg region in pressure–temperature space from the binary carbon dioxide + n-1-methylnaphthalene llg locus downward in pressure, towards the vapor pressure curve of the co-solvent acetone. The three-phase region is bounded from above by an upper critical end point (UCEP) (l-l=g) locus. The presence of the co-solvent acetone significantly enhances the amount of the solute 1-methylnaphthalene taken up by the second (carbon-dioxide-rich) liquid phase. © 2002 Elsevier Science B.V. All rights reserved. Keywords: Data; Mixture; Liquid–liquid–vapor equilibria; Extraction; Co-solvent
1. Introduction Our group has been experimentally studying liquid–liquid–vapor (llg) phase equilibria in well-defined binary and ternary systems. These systems are typically composed of one or more gas solvents (e.g. carbon dioxide, ethane, propane, nitrous oxide and/or xenon) and one or more relatively non-volatile solutes chosen from homologous series of n-alkanes, n-alkylbenzenes and/or 1-alkanols. The primary goals of these studies are to map out the patterns of the multiphase equilibria of these prototype mixtures in thermodynamic phase space and to generate data that support the prediction of phase equilibria within and near the regions of llg partial immiscibility. Successful supercritical extraction at temperatures near and above the critical temperature of a gas solvent is often accompanied by the presence of llg phase equilibria at lower temperatures [1,2], be it Abbreviations: AAD, average absolute deviation; LCEP, lower critical end point; UCEP, upper critical end point ∗ Corresponding author. Tel.: +1-918-631-2226; fax: +1-918-631-3268. E-mail address: [email protected] (K.D. Luks). 0378-3812/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 3 8 1 2 ( 0 1 ) 0 0 7 3 5 - X
30
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
actual or metastable (masked by the formation of a solid solute phase). Therefore, models purporting to describe supercritical extraction should be capable of describing llg phase equilibria as well. Additionally, there is the attractive prospect of carrying out subcritical extraction in an llg configuration, wherein a solute is selectively removed from the solute-rich liquid phase by the gas-solvent-rich liquid phase [the extract phase] in the presence of a gas-solvent-rich vapor phase. Compared to supercritical extraction, subcritical extraction would operate at more modest pressures and temperatures. Investigators of supercritical extraction processes have addressed the problem of enriching the extract phase in the solute, and the addition of a modifier (co-solvent) to the extraction system has been carried out with some success. In the case of the gas solvent carbon dioxide, the co-solvent is generally a lighter molecular species of considerable dipole moment such as methanol (1.7 Debye) or acetone (2.88 Debye). In two recent studies, we examined the effect of the co-solvents methanol and acetone, respectively on the llg phase equilibria of the partially miscible binary mixture carbon dioxide + n-tetradecane [3,4]. In this present study, we report the llg phase equilibria of the ternary mixture carbon dioxide + 1-methylnaphthalene + acetone, initiating an inquiry whether common polar co-solvents can enhance the concentration of a heavy aromatic hydrocarbon in the carbon dioxide-rich-liquid [l2 , or extract] phase. The data were obtained using a visual cell (“stoichiometric”) experimental technique. The boundaries of the llg region were mapped out in pressure–temperature space and then the compositions and the molar volumes of the two liquid phases were found along the 288.15 and 298.15 K isotherms. The effectiveness of acetone as a co-solvent is discussed in the context of the binary llg system carbon dioxide +1-methylnaphthalene.
2. Experimental Fall and Luks [5] and Jangkamolkulchai and Luks [6] give a detailed description of the experimental apparatus. The procedures for performing the llg studies are described in Fall et al. [7] and Jangkamolkulchai and Luks [6]. The experimental apparatus employs a stoichiometric approach, wherein known amounts of 1-methylnaphthalene and acetone are added to a volumetrically calibrated visual (glass) equilibrium cell. The total volume of the cell is typically 7–9 cm3 . The cell vapor space is thoroughly flushed with carbon dioxide gas, after which measured amounts of the carbon dioxide gas are added to the cell from a high-pressure bomb. The cell contents are brought to equilibrium by a magnetically actuated steel ball stirrer mechanism. With the use of mass balances, the compositions and molar volumes of the phases in a ternary llg system at a given temperature and pressure can be determined from “conjugate” measurements. Our experience [8] indicates that the presence of 1-methylnaphthalene in the vapor phase is negligible. It was assumed that the presence of acetone was also masswise negligible in the vapor phase. Therefore, the mass balance calculations required a set of conjugate runs in which the l1 and l2 phases were respectively dominant in terms of volumetric fraction. l1 - and l2 -dominant run data at a specific temperature and pressure enabled quantitative determination of the compositions and molar volumes of these phases. Since the vapor phase was assumed to be pure carbon dioxide, its stoichiometry was calculated from tables [9] rather than measured here. The visual cell temperature was measured with a Pt-resistance thermometer to an estimated accuracy of ±0.02 K. Pressures were measured to ±0.07 bar with pressure transducers that were frequently calibrated with a dead-weight gauge. Phase volumes in the calibrated visual cell were determined with the aid of a cathetometer to an accuracy of at least 0.01 cm3 .
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
31
The acetone and 1-methylnaphthalene were purchased from Aldrich. The acetone had a stated purity of >99.9%. The 1-methylnaphthalene had a stated purity of 98%. Chromatographic analysis indicated purities consistent with these stated purities. These chemicals were used without further purification. The carbon dioxide was purchased from Air Products and Chemicals Inc as “Coleman grade” with a stated purity of 99.99 mol%. The carbon dioxide was first transferred to an evacuated storage bomb immersed in an ice bath. The vapor phase was vented to remove any light impurities. The purity was verified by liquefying the carbon dioxide at 298.15 K. The vapor pressure at this temperature was within 0.04 bar of that reported by Vargaftik [10]. In addition, the critical temperature and pressure were within 0.06 K and 0.07 bar of those reported by Vargaftik [10].
3. Results In Fig. 1, portions of the boundaries of the llg surface of the ternary system carbon dioxide + 1-methylnaphthalene + acetone are shown. The surface extends from the binary llg locus of carbon dioxide+1-methylnaphthalene downward in pressure with the addition of the third component acetone (towards the acetone vapor pressure curve). The surface is terminated from above by an upper critical end point locus (UCEP, or l-l=g), given in Table 1. There appears to be, based on a cursory scanning exercise, no llg partial immiscibility for the binary system carbon dioxide + acetone. This suggests that, the continued addition of acetone to the binary mixture carbon dioxide + 1-methylnaphthalene will eventually lead to a termination of llg partial miscibility, its terminus being a critical end point locus of the type l=l-g. That locus would eventually intersect the critical end point locus reported in Table 1 at a tricritical point. In contrast to the earlier study of the system carbon dioxide + n-tetradecane + acetone, the partial immiscibility in the system carbon dioxide + 1-methylnaphthalene + acetone is more pronounced. The l=l-g locus would occur at much higher concentrations of acetone relative to 1-methylnaphthalene. We did not study mixtures with such high acetone concentrations, since the role of the acetone from a practical standpoint, is that of an additive co-solvent. The boundary points were determinable to ±0.1 K and ±0.07 bar and are considered to be good to ±0.1 K and ±0.10 bar.
Fig. 1. Deviation of the binary llg and ternary l-l=g boundaries of the ternary llg surface from the vapor pressure curve for pure 0 carbon dioxide in pressure–temperature space for the system carbon dioxide + 1-methylnaphthalene + acetone. PCO is the vapor 2 pressure of carbon dioxide as represented by the correlation ln(P r ) = 6.576183 (l-l/Tr ). (䊉) l-l=g critical end points.
32
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
Table 1 Temperature T and pressure P raw data for the l-l=g boundary loci for the ternary mixture carbon dioxide+1-methylnaphthalene+ acetone UCEP (l-l=g) T (K)
P (bar)
308.82 309.76 309.93 310.05 309.98 310.07 311.47 311.64 312.20 312.17 312.67 313.38 313.68 314.36 314.42 315.36
79.30 80.23 80.35 80.42 80.43 80.46 81.44 81.85 82.35 82.36 82.77 83.47 83.76 84.38 84.44 85.39
A regressed straight line through the data in Table 1 yields an average absolute deviation (AAD) of about 0.05 bar. The llg data for the ternary system carbon dioxide+1-methylnaphthalene+acetone along the isotherms 288.15 and 298.15 K are given in Figs. 2–6 and Tables 2 and 3. The ternary mole fractions are estimated to be accurate to ±0.004 and the molar volumes to ±0.6 cm3 /mol. These are conservative estimates, based on AAD of the raw data from simple smoothed curves and our experience at being able to generate
Fig. 2. Mole fraction of carbon dioxide as a function of pressure P for the l1 and l2 phases for the llg region of carbon dioxide + 1-methylnaphthalene + acetone at 288.15 K.
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
33
Fig. 3. Molar volume v as a function of pressure P for the l1 and l2 phases for the llg region of carbon dioxide + 1-methylnaphthalene + acetone at 288.15 K.
Fig. 4. Mole fraction of carbon dioxide as a function of pressure P for the l1 and l2 phases for the llg region of CO2 + 1methylnaphthalene + acetone at 298.15 K.
Fig. 5. Molar volume v as a function of pressure P for the l1 and l2 phases for the llg region of carbon dioxide + 1-methylnaphthalene + acetone at 298.15 K.
34
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
Fig. 6. Concentration c of 1-methylnaphthalene in liquid phase 2 as a function of pressure P for the llg region of the system carbon dioxide + 1-methylnaphthalene + acetone at 288.15 and 298.15 K. Quadratic curve representations are offered. Table 2 Liquid phase l1 and l2 compositions (mole fractions x) and molar volumes v as a function of pressure P for the llg region of carbon dioxide + 1 methylnaphthalene + acetone at 288.15 K l1 Phase
l2 Phase
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
49.61 49.18 49.07 48.69 48.54 48.36 48.21 47.92 47.72 47.43
0.4537 0.4600 0.4729 0.4721 0.4724 0.4743 0.4833 0.4890 0.4917 0.4987
0.0000 0.0091 0.0105 0.0132 0.0163 0.0204 0.0238 0.0357 0.0385 0.0440
95.5 95.8 93.1 92.2 92.4 91.7 90.6 89.2 89.1 88.4
49.61 49.16 48.53 48.43 48.30 47.84 47.60
0.9793 0.9693 0.9644 0.9628 0.9598 0.9527 0.9455
0.0000 0.0047 0.0137 0.0156 0.0171 0.0218 0.0259
53.5 53.9 53.6 53.6 54.1 53.9 53.8
Table 3 Liquid phase l1 and l2 compositions (mole fractions x) and molar volumes v as a function of pressure P for the llg region of carbon dioxide + 1 methylnaphthalene + acetone at 298.15 K l1 Phase
l2 Phase
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
P (bar)
x (CO2 )
x (C3 H6 O)
v (cm3 /g mol)
62.46 62.03 61.97 61.20 61.08 60.83 59.61 59.39 59.38
0.4736 0.4831 0.4823 0.4949 0.4934 0.5009 0.5244 0.5224 0.5307
0.0000 0.0099 0.0104 0.0214 0.0227 0.0249 0.0409 0.0556 0.0434
93.6 92.6 93.0 90.3 91.0 90.1 86.3 86.0 85.7
62.46 61.95 61.07 61.04 60.88 60.87 60.00 59.93 59.43 59.29
0.9851 0.9785 0.9716 0.9695 0.9695 0.9670 0.9555 0.9492 0.9512 0.9422
0.0000 0.0042 0.0091 0.0118 0.0104 0.0133 0.0224 0.0258 0.0276 0.0286
58.3 57.8 57.4 57.7 57.1 57.4 56.5 56.2 55.7 55.9
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
35
Table 4 Concentration c of 1-methylnaphtalene, mole ratio of 1-methylnaphtalene to CO2 in liquid phase 2, and mole fraction of acetone in liquid phase 2 as a function of pressure at temperatures 288.15 and 298.15 K for the llg region of carbon dioxide + 1-methylnaphtalene + acetone Liquid phase 2 at 288.15 K
Liquid phase 2 at 298.15 K
P (bar)
c(1-MN) mol cm−3
r(1-MN/CO2 )
x (C3 H6 O)
P (bar)
c(1-MN) mol cm−3
r(1-MN/CO2 )
x (C3 H6 O)
49.61 49.16 48.53 48.51 48.43 48.30 47.84 47.60
3.875E-04 4.818E-04 4.081E-04 4.475E-04 4.023E-04 4.283E-04 4.727E-04 5.323E-04
2.118E-02 2.677E-02 2.270E-02 2.539E-02 2.241E-02 2.414E-02 2.673E-02 3.029E-02
0.0000 0.0047 0.0137 0.0133 0.0156 0.0171 0.0218 0.0259
62.46 61.95 61.07 61.04 60.99 60.88 60.87 60.00 59.93 59.43 59.29
2.555E-04 3.000E-04 3.368E-04 3.252E-04 3.624E-04 3.529E-04 3.425E-04 3.898E-04 4.456E-04 3.812E-04 5.210E-04
1.511E-02 1.780E-02 1.999E-02 1.930E-02 2.151E-02 2.095E-02 2.033E-02 2.314E-02 2.645E-02 2.263E-02 3.092E-02
0.0000 0.0042 0.0091 0.0118 0.0133 0.0104 0.0133 0.0224 0.0258 0.0276 0.0286
reproducible data using the stoichiometric technique described herein. The AADs for the l1 data are 0.003 for the carbon dioxide mole fractions and 0.4 cm3 /mol for the liquid phase molar volume. For the l2 data, these AADs are 0.001 and 0.2 cm3 /mol, respectively. 4. Discussion A successful co-solvent should enhance the capacity of the l2 phase for the solute of interest. Table 4 suggests that acetone is an effective co-solvent here, in the context of 1-methylnaphthalene in a carbon dioxide medium. The absolute molar concentration c1-MN of the solute 1-methylnaphthalene is significantly increased as the ternary system enriches with acetone, at both 288.15 and 298.15 K. The measure r, which is a ratio of 1-methylnaphthalene to carbon dioxide also increases significantly. These enhancements occur with relatively modest additions (x < 0.03) of the co-solvent acetone to the overall phase mixture. List of symbols c molar concentration g gas or vapor phase l, l1 , l2 liquid phases l-l=g a three-phase critical end point where the less dense liquid phase (l2 ) and the vapor phase (g) are critically identical; also referred to as the UCEP l=l-g a three-phase critical end point where the two liquid phases (l1 and l2 ) are critically identical; also referred to as the LCEP P pressure r mole ratio of n-1-methylnaphthalene to carbon dioxide T temperature
36
v x
L.E. Guti´errez M, K.D. Luks / Fluid Phase Equilibria 198 (2002) 29–36
molar volume mole fraction
Acknowledgements The support of this research was provided by the National Science Foundation (Grant No. CTS-9730853). The apparatus used is a part of the PVTx Laboratory at the University of Tulsa and was purchased with funds provided by several industries, the University of Tulsa and a National Science Foundation specialized equipment (Grant No. CPE-8104650). References [1] S.C. Hwang, Y.N. Lin, S.W. Hopke, R. Kobayashi (Eds.), Relation of liquid–Liquid Equilibrium Behavior at Low Temperatures to Vapor–Liquid Equilibria Behavior at High Temperatures and Elevated Pressures, in: Proceedings of the 57th Annual GPA Convention, 1978, pp. 7–12. [2] R.M. Lansangan, A. Jangkamolkulchai, K.D. Luks, Binary vapor–liquid equilibria behavior in the vicinity of liquid–liquid–vapor loci, Fluid Phase Equilb. 36 (1987) 49–66. [3] C.M. Foreman, K.D. Luks, Partial miscibility behavior of the ternary mixture carbon dioxide + n-tetradecane + methanol, J. Chem. Eng. Data 45 (2000) 334–337. [4] C.M. Foreman, K.D. Luks, Partial miscibility behavior of the ternary mixture carbon dioxide + n-tetradecane + acetone, Fluid Phase Equilb. 169 (2000) 65–73. [5] D.J. Fall, K.D. Luks, Phase equilibria behavior of the systems carbon dioxide + n-dotriacontane and carbon dioxide + n-docosane, J. Chem. Eng. Data 29 (1984) 413–417. [6] A. Jangkamolkulchai, K.D. Luks, Partial miscibility behavior of the methane + ethane + n-docosane and the methane + ethane + n-tetradecylbenzene ternary mixtures, J. Chem. Eng. Data 34 (1989) 92–99. [7] D.J. Fall, J.L. Fall, K.D. Luks, Liquid–liquid–vapor immiscibility limits in carbon dioxide + n-paraffin mixtures, J. Chem. Eng. Data 30 (1985) 82–88. [8] L.E. Gutiérrez M, K.D. Luks, Three-phase liquid–liquid–vapor equilibria of the binary mixture carbon dioxide + 1-methylnaphthalene, J. Chem. Eng. Data, in press. [9] S. Angus, B. Armstrong, B.K.M. de Reuck (Eds.), International Thermodynamic Tables of the Fluid State Carbon Dioxide, Pergamon Press, Oxford, 1976. [10] N.B. Vargaftik, Tables on the Thermophysical Properties of Liquids and Gases, 2nd Edition, Wiley, New York, 1975, pp. 167–168.