Experimental Measurement and Modeling of Phase Diagrams of

Article pubs.acs.org/jced

Experimental Measurement and Modeling of Phase Diagrams of Binary Systems Encountered in the Gasoline Desulfurization Process Using Ionic Liquids Yushu Chen, Fabrice Mutelet,* and Jean-Noel̈ Jaubert Université de Lorraine, Ecole Nationale Supérieure des Industries Chimiques, Laboratoire Réactions et Génie des Procédés (UMR CNRS 7274), 1 rue Grandville, 54000 Nancy, France ABSTRACT: The aim of this work is to investigate phase equilibrium of binary systems encountered in the extractive desulfurization of gasoline and diesel using ionic liquids. This work is focused on two ionic liquids: 1,3-dimethylimidazolium methylphosponate [DMIM][Ph] and 1-ethyl-3-methylimidazolium thiocynate [EMIM][SCN]. Vapor−liquid equilibrium (VLE) measurements on binary systems of ionic liquids with various solutes including thiophene, pyridine, toluene, and water at pressures close to the atmospheric pressure were performed and correlated by the PC-SAFT equation of state. The molecular parameters of pure ILs were determined from experimental density data, whereas binary interaction parameters kij were optimized on experimental VLE data. The eight binary mixtures: {thiophene+ILs}, {pyridine+ILs}, {toluene+ILs}, and {water +ILs} studied in this paper were accurately described by the PC-SAFT equation of state.

■

INTRODUCTION Recently, the European Community proposed strict restrictions on sulfur contents in produced fuels as a result of the environmental protection strategies. Sulfur contents combustion gives birth to sulfur oxides which are believed to cause most environmental problems such as global warming effects, acid rain, and air pollution. Therefore, the desulfurization of fuels has attracted much considerable attention in Europe and Japan where attempts are being made to diminish the sulfur contents in gasoline to less than 10 ppm by 2010.1 Until now, the most widely used desulfurization process in industry has been hydrodesulfurization (HDS), by which most of the sulfur compounds such as thiols, sulfides, and disulfides can be converted to H2S using the catalysis of Co/Mo/Al2O3 or Ni/ Mo/Al2O3 at high temperature and high pressure. Such a method however suffers several drawbacks: (i) it is operated at very high temperatures and pressures, (ii) octane and cetane numbers are reduced due to hydrogenation side reactions, (iii) it is efficient for the removal of sulfides, thiols, and thiophenes; however, it is more difficult to remove refractory sulfur compounds like dibenzothiophene, benzothiophene, and their alkyl derivatives2,3 because of their very low hydrogenation activity.4 For these reasons, it is still a challenge for deep desulfurization using the traditional HDS process, and alternative approaches should be developed in order to produce gasoline with low or even ultralow sulfur content. An alternative process, extractive desulfurization using ionic liquids (ILs), is regarded as a promising process. It has a high sulfur removal ratio and great selectivity under mild operating conditions; moreover, it is safe and simple but also reproducible. The use of such solvents in the field of extractive desulfurization presents a great potential because of thermodynamic properties of these solvents: negligible vapor pressure © 2014 American Chemical Society

and great thermal−chemical stabilities. Furthermore, ILs may be used for the extraction of various organic but also inorganic compounds. Their physical−chemical properties like polarity or miscibility can be modified according to certain system requirements. For all these reasons, ILs are being considered suitable and promising solvents for a large number of applications such as liquid−liquid extraction, catalysis, synthesis, and gas separations.5−7 The first study presenting the application of ILs for the extractive desulfurization of diesel was published by Bosmann et al.8 in 2001. Then, extractive desulfuration9−18 and denitrogenation19 processes of gasoline oil using different ILs were extensively studied and reported. Most studies demonstrated that the ILs present greater extraction ratios and selectivity in extractive desulfuration than conventional organic solvents. Among other studies, [PF6] ¯or [BF4]¯ based imidazolium ILs presented great selectivity for organic sulfur and nitrogen compounds extraction.20,21 However, the main drawback observed during the extractive desulfurization is the appearance of aromatic hydrocarbons extraction and the decrease of the fuel octane number.10 Because of corrosion and stability problems and also the total expenses of extractive desulfurization of fuel, it is still difficult and a challenging task to choose the most suitable ionic liquid for the extractive desulfurization process. As an example, it was found that the imidazolium thiocyanate ILs, [Cn-mim][SCN] could selectivity separate22−24 hydrocarbons from sulfur compounds. Since sulfur compounds are partially miscible with ILs, phase equilibrium measurements were mainly focused on liquid− Received: April 17, 2013 Accepted: February 11, 2014 Published: February 25, 2014 603

dx.doi.org/10.1021/je400510j | J. Chem. Eng. Data 2014, 59, 603−612

Journal of Chemical & Engineering Data

Article

(ãassoc) to the Helmholtz free energy is absolutely necessary to model mixtures containing associating molecules and is defined by27,28

liquid equilibrium (LLE). It is however obvious that binary mixtures encountered in the extractive desulfurization process present not only LLE but also VLE regions.25 Furthermore, the experimental VLE data for IL-containing binary mixtures is essential for the calculation of phase diagrams in multicomponent systems using a thermodynamic model, whereas the VLE data is extremely scarce until now. For these reasons, it was decided to measure the vapor−liquid equilibrium of various systems involved in the extractive desulfurization of fuels from ILs. VLE data of binary systems containing either 1,3dimethylimidazolium methylphosphonate [DMIM][Ph] or 1ethyl-3-methylimidazolium thiocynate [EMIM][SCN] and one of the following solutes, thiophene, pyridine, toluene, or water, are determined in this work. A total of eight binary systems: {thiophene + ILs}, {pyridine + ILs}, {toluene + ILs}, and {water+ ILs} were thus studied. The PC-SAFT26 (perturbedchain SAFT) EoS was chosen to correlate the acquired data. The pure ILs parameters were determined from density measurement also performed in this paper.

where ρj = xjρ and ρ is the molar density, Δ represents the association strength which depends on εAiBj, association energy, and kAiBj, association volume, between associating substances i and j.

PC-SAFT MODELING A thermodynamic model based on statistical associating fluid theory (SAFT) has been developed by Huang and Radosz.27,28 In 2001, a modified SAFT equation of state (EoS) named PCSAFT was developed by Gross and Sadowski26 applying the perturbation theory to a hard-chain reference fluid, to compare with the original SAFT which used a hard-sphere reference fluid. In terms of the residual Helmholtz energy, the PC-SAFT EoS is usually presented as follows:

EXPERIMENTAL SECTION Materials. 1,3-dimethylimidazolium methylphosponate [DMIM][Ph] (purity > 98 %) was obtained from Solvionic and 1-ethyl-3-methylimidazolium thiocynate [EMIM][SCN] was purchased from Fluka with a purity > 95 %. Pyridine, thiophene, and toluene were provided by Sigma-Aldrich with a purity of 99.8 %, 99.8 %, and 99 %, respectively. Tetrachloroethylene (purity > 99 %) and n-dodecane (purity > 99 %) were used as reference fluids for measuring the density of ILs. All the materials employed in this work are presented in Table 1.

nc

a assoc = ̃

i=1

res

⎝

−

⎤ X Ai ⎞ 1 ⎥ ⎟ + Mi 2 ⎠ 2 ⎥⎦

(4)

Bj

(5) AiBj

■

(1)

Table 1. Materials Used in Experiments along with Suppliers and Purities

nc

i=1

Ai

j

hc

∑ xi(mi − 1)ln gijhs

⎣

i

X A i = [1 + NAV ∑ ∑ ρj X BjΔA iBj ]−1

where ã is the residual Helmholtz free energy and ã (hardchain contribution), ãdisp(dispersion contribution), and ãassoc(assoicating contribution) refer to different microscopic contributions to the free energy of the system. The choice of these different terms is according to certain system requirements. The hard chain term comes from Wertheim’s theory. In this work, the hard chain expression is provided by Gross and Sadowski: hs a ̃hc = ma ̅ ̃ −

⎛

where Mi presents the number of bonding sites on each molecule, ∑Ai represents a sum over all associating sites (starting with A) on each molecule and XAi is the mole fraction of molecules i not associated at site A and expressed as

■

a ̃res = a ̃hc + a disp ̃ + a assoc ̃

⎡

∑ xi⎢⎢∑ ⎜ln X A

(2)

materials

suppliers

[DMIM][Ph] [EMIM][SCN] pyridine thiophene toluene n-dodecane tetrachloroethylene

Solvionic Fluka Sigma-Aldrich Sigma-Aldrich Sigma-Aldrich Sigma-Aldrich Acros Organics

purity/% > 98 % > 95 % 99.8 % 99.8 % 99 % > 99 % > 99 %

Apparatus and Procedure. Density measurement. The density measurements of pure ILs were performed on an Anton Paar DMA 60 densimeter at atmospheric pressure. The densimeter consists of a digital vibrating-tube with a DMA 512P measuring cell. A platinum resistance thermometer Pt100 which has an accuracy of ± 0.1 K was used to measure the internal temperature in the vibrating-tube cell and a thermostatic oil bath was employed as circulating fluid in the measuring cell in order to hold the internal temperature in the cell constant to ± 0.1 K. n-Dodecane and tetrachloroethylene were employed as reference fluids for the calibration of the densimeter for the purpose of determining the pure IL densities. VLE Measurement. The detailed procedure for the experimental measurement of VLE has been reported in the previous paper of Revelli et al.29 and only a short description will be provided here. The vapor−liquid equilibrium measurements of mixtures are carried out in a glass cell employing a static method. The experimental instrument is displayed in

ghs ij

where is the hard-sphere radial pair distribution function, m̅ is the mean segment number, and ãhs is the hard-sphere contribution to the free energy of system. ãdisp is the dispersion contribution to the total Helmholtz free energy of system, which accounts for van der Waals forces. In this study, the dispersion expression was defined according to the paper of Gross and Sadowski:26 2 2 3 a disp ̃ = −2πρ ̃I1m2εσ 3 − πρ ̃mC (3) ̅ 1I2m ε σ where ρ̃ is the number density, the coefficient C1 can be calculated by m̅ and η, I1 and I2 are power series which depend on η, and m, σ, and ε are the pure component PC-SAFT parameters. The evaluation of the vapor−liquid equilibrium of nonassociative fluids requires only the knowledge of the terms ãhc and ãdisp in the Helmholtz free energy equation. In this case, the three parameters m, σ, and ε/kB are necessary to characterize each pure compound. However, the association contribution

604



Article

Figure 1 and used to the VLE measurement of isothermal P−x data up to 40 kPa and 313 K.

Figure 1. The VLE measurement apparatus: VP, vacuum pump; VT, vacuum trap; A, magnetic stirrer; C, equilibrium cell; PT-100, platinum resistance thermometer; T, temperature indicator, M, calibrated pressure sensor, P, digital pressure indicator; and TB, thermostatic bath.

■

Figure 2. Temperature-density diagrams for ILs: ◇, [EMIM][SCN] and ○, [DMIM][Ph]. Symbols, experimental densities data; solid lines, PC-SAFT estimations.

RESULTS AND DISCUSSION Pure components. First, the PC-SAFT EoS was used to represent the thermodynamic properties of pure ILs like

temperature-dependent density. Density measurements for 1,3dimethylimidazolium methylphosponate [DMIM][Ph] were performed at different temperatures (from 298.15 K to 318.75 K) at atmospheric pressure. The experimental densities of [DMIM][Ph] measured in this work are reported in Table 2, while the densities of [EMIM][SCN] were taken from literature.30 The parameters of PC-SAFT for pure ILs were determined using experimental densities by minimizing the following objective function (OF):

Table 2. Experimental Densities of Pure [DMIM][Ph] in the Range of the Temperature from 298.15 K to 318.75 Ka

a

ILs

temp [K]

density [g/cm3]

[DMIM][Ph]

298.15 299.45 300.75 302.15 303.45 304.75 305.85 307.15 308.65 310.35 311.75 313.05 314.45 315.85 317.25 318.75

1.2306 1.2298 1.2293 1.2286 1.2279 1.2273 1.2266 1.2258 1.2249 1.2239 1.2231 1.2224 1.2215 1.2207 1.2199 1.2192

⎛ ρ sat,exp − ρ sat,cal ⎞2 i ⎟⎟ OF = ∑ ⎜⎜ i sat,exp ρ ⎠ i i=1 ⎝ n pts

(6)

where the difference between experimental liquid densities and calculated values can be evaluated. In this work, thiophene and toluene were modeled as nonassociating substances, and they were thus presented by three nonassociating molecular parameters: m, σ, and ε/kB. On the other hand water and pyridine were considered as self-associating compounds and were presented by three nonassoicating parameters (m, σ, ε/kB) but also two self-associating parameters (association energy, −εAiBj, and association volume, −kAiBj). The PC-SAFT parameters for pyridine, thiophene, toluene, and water were taken from literature26,31−33 and presented in Table 3.

The standard uncertainty u is u(ρ) = 0.0001 g/cm3.

Table 3. Optimized PC-SAFT Parameters of Pure ILs: [EMIS][SCN], [DMIM][Ph], and PC-SAFT Parameters of Pyridine, Thiophene, Toluene, and Water σ

Mw ‑1

ε/kB

εAiBj

ILs

g·mol

Å

K

m

[EMIM][SCN] [DMIM][Ph] H2O thiophene pyridine toluene

169.2500 192.1500 18.0150 84.1420 79.1014 92.1410

4.2200 4.1300 3.0007 3.5655 3.8066 3.7169

383.80 411.40 366.51 301.73 250.65 285.69

3.0500 3.4200 1.0656 2.3644 2.0352 2.8149 605

AiBj

k

density AAD

K

%

0.002250 0.002250 0.034868

3450.00 3450.00 2500.70

0.32 0.12

0.189332

1890.30



Article

Table 4. Experimental VLE and VLLE Data for Various Molar Fractions of Different Compounds in Binary Systems Containing ILsa T/K 283.15

293.15

288.15

298.15

288.15

298.15

x1

P/kPa

1.0000 0.8971 0.8016 0.7061 0.6183 0.5110 0.4113 0.3253 0.2642 0.1229 1.0000 0.8971 0.8016 0.7061 0.6183 0.5110 0.4113 0.3253 0.2642 0.1229

1.27 1.20 0.97 0.79 0.63 0.45 0.32 0.23 0.20 0.15 2.33 2.19 1.78 1.48 1.19 0.85 0.62 0.46 0.37 0.25

1.0000 0.8879 0.7900 0.6647 0.5968 0.4984 0.3834 0.3008 0.1872 0.0957 0.0438 1.0000 0.8879 0.7900 0.6647 0.5968 0.4984 0.3834 0.3008 0.1872 0.0957 0.0438

2.42 2.38 2.33 2.33 2.33 2.33 2.33 2.32 1.98 1.28 0.63 3.81 3.80 3.76 3.76 3.76 3.76 3.76 3.75 3.20 1.96 1.15

1.0000 0.8953 0.7889 0.6954 0.6009 0.4941 0.3985 0.2999 0.2028 0.0980 0.0526 1.0000 0.8953 0.7889 0.6954

6.78 6.67 6.68 6.64 5.83 4.94 4.01 3.11 2.28 1.15 0.68 10.47 10.37 10.36 10.25

T/K

state

[EMIM][SCN] + H2O (1) VLE 288.15

VLE

298.15

[EMIM][SCN] + Toluene (1) VLE 293.15 VLLE

VLE

x1

P/kPa

state

1.0000 0.8971 0.8016 0.7061 0.6183 0.5110 0.4113 0.3253 0.2642 0.1229 1.0000 0.8971 0.8016 0.7061 0.6183 0.5110 0.4113 0.3253 0.2642 0.1229

1.73 1.62 1.37 1.09 0.84 0.65 0.46 0.36 0.30 0.21 3.20 2.93 2.45 1.96 1.53 1.13 0.82 0.58 0.50 0.30

VLE

1.0000 0.8879 0.7900 0.6647 0.5968 0.4984 0.3834 0.3008 0.1872 0.0957 0.0438

3.18 3.10 3.04 3.04 3.04 3.04 3.04 3.04 2.68 1.62 0.86

VLE VLLE

1.0000 0.8953 0.7889 0.6954 0.6009 0.4941 0.3985 0.2999 0.2028 0.0980 0.0526

8.70 8.54 8.54 8.38 7.64 6.33 5.20 3.82 2.80 1.52 0.92

VLE

VLE

VLE VLLE

VLE

[EMIM][SCN] + Thiophene (1) VLE 293.15 VLLE

VLE

VLE VLLE

VLE

VLE VLLE

606



Article

Table 4. continued T/K

288.15

298.15

283.15

293.15

288.15

x1

P/kPa

0.6009 0.4941 0.3985 0.2999 0.2028 0.0980 0.0526

9.73 8.13 6.70 5.14 3.63 1.89 1.18

1.0000 0.9022 0.8001 0.7185 0.5964 0.4948 0.3971 0.3050 0.1956 0.1100 1.0000 0.9022 0.8001 0.7185 0.5964 0.4948 0.3971 0.3050 0.1956 0.1100

1.65 1.59 1.45 1.36 1.14 0.96 0.80 0.59 0.44 0.28 2.83 2.71 2.54 2.38 2.03 1.70 1.42 1.12 0.82 0.57

1.0000 0.8981 0.8003 0.7072 0.6093 0.5232 0.4160 0.3274 0.2154 1.0000 0.8981 0.8003 0.7072 0.6093 0.5232 0.4160 0.3274 0.2154

1.27 0.96 0.72 0.47 0.35 0.26 0.17 0.13 0.10 2.33 1.86 1.28 0.89 0.62 0.49 0.35 0.26 0.20

1.0000 0.8916 0.7910 0.6859 0.6014 0.4946 0.4007 0.2999 0.2037 0.1004 0.0654 0.0434

2.35 2.30 2.27 2.21 2.25 2.27 2.27 2.27 2.27 2.17 1.45 0.90

T/K

state

x1

P/kPa

state

1.0000 0.9022 0.8001 0.7185 0.5964 0.4948 0.3971 0.3050 0.1956 0.1100

2.20 2.13 1.95 1.84 1.53 1.30 1.08 0.80 0.63 0.38

VLE

1.0000 0.8981 0.8003 0.7072 0.6093 0.5232 0.4160 0.3274 0.2154 1.0000 0.8981 0.8003 0.7072 0.6093 0.5232 0.4160 0.3274 0.2154

1.73 1.37 0.93 0.67 0.50 0.36 0.25 0.20 0.15 3.20 2.43 1.92 1.16 0.85 0.70 0.47 0.35 0.25

VLE

1.0000 0.9235 0.8532 0.7762 0.6903 0.7567 0.4788 0.3568 0.2352 0.1139 0.0651 0.0414

3.05 2.96 2.88 2.88 2.88 2.88 2.88 2.88 2.89 2.79 1.86 1.23

VLE VLLE

[EMIM][SCN] + Thiophene (1) VLE

[EMIM][SCN] + Pyridine (1) VLE 293.15

VLE

[DMIM][Ph] + H2O (1) VLE 288.15

VLE

298.15

[DMIM][Ph] + Toluene (1) VLE 293.15 VLLE

VLE

607

VLE

VLE



Article

Table 4. continued T/K 298.15

288.15

298.15

288.15

298.15

a

x1

P/kPa

1.0000 0.9235 0.8532 0.7762 0.6903 0.7567 0.4788 0.3568 0.2352 0.1139 0.0651 0.0414

3.84 3.82 3.77 3.76 3.76 3.76 3.76 3.76 3.77 3.68 2.20 1.48

1.0000 0.9507 0.9155 0.8939 0.8917 0.8502 0.6953 0.6556 0.4985 0.3724 0.3001 0.2353 0.1071 1.0000 0.8945 0.6953 0.6556 0.4985 0.3646 0.2308 0.1187

6.75 6.69 6.60 6.57 6.55 6.55 6.54 6.54 6.54 5.90 5.18 4.16 2.54 10.58 10.38 10.36 10.38 10.24 9.29 5.93 3.70

1.0000 0.9013 0.7794 0.7066 0.5934 0.5061 0.4364 0.3010 0.2201 0.1242 1.0000 0.9013 0.7794 0.7066 0.5934 0.5061 0.4364 0.3010 0.2201 0.1242

1.72 1.66 1.58 1.58 1.57 1.48 1.38 1.08 0.84 0.53 2.85 2.76 2.57 2.56 2.43 2.32 2.26 1.86 1.40 0.91

T/K

state

[DMIM][Ph] + Toluene (1) VLE 303.15 VLLE

VLE

[DMIM][Ph] + Thiophene (1) VLE 293.15 VLLE

VLE

VLE VLLE

303.15

VLE

[DMIM][Ph] + Pyridine (1) VLE 293.15

VLE

303.15

x1

P/kPa

state

1.0000 0.9235 0.8532 0.7762 0.6903 0.7567 0.4788 0.3568 0.2352 0.1139 0.0651 0.0414

4.73 4.68 4.63 4.58 4.52 4.58 4.55 4.61 4.60 4.50 2.90 1.88

VLE VLLE

1.0000 0.9497 0.9115 0.8891 0.8635 0.8945 0.6953 0.6556 0.4985 0.3646 0.2308 0.1187

8.74 8.69 8.52 8.49 8.53 8.47 8.40 8.43 8.48 7.58 4.84 3.16

1.0000 0.8945 0.6953 0.6556 0.4985 0.3646 0.2308 0.1187

12.98 12.44 12.40 12.36 11.24 10.57 7.20 4.29

VLE VLLE

1.0000 0.9013 0.7794 0.7066 0.5934 0.5061 0.4364 0.3010 0.2201 0.1242 1.0000 0.9013 0.7794 0.7066 0.5934 0.5061 0.4364 0.3010 0.2201 0.1242

2.25 2.13 2.05 2.06 1.98 1.91 1.79 1.46 1.09 0.69 3.45 3.37 3.12 3.08 2.90 2.82 2.75 2.41 1.78 1.16

VLE

VLE

VLE VLLE

VLE

VLE

VLE

Standard uncertainties u: u(P) = 0.05 kPa, u(T) = 0.1 K, and u(x1) = 0.0002.

associating compounds. m, σ, and ε/kB as three nonassociating parameters can be determined by a fitting procedure on pure-

In the case of pure ILs, the set of PC-SAFT molecular parameters were characterized considering them as self608



Article

Table 5. kij Interaction Parameters of Binary Mixtures at Different Temperatures compound H2O

pyridine

toluene

thiophene

H2O

pyridine

toluene

thiophene

T/K 283.15 288.15 293.15 298.15 288.15 293.15 298.15 288.15 293.15 298.15 288.15 293.15 298.15 283.15 288.15 293.15 298.15 288.15 293.15 298.15 303.15 288.15 293.15 298.15 303.15 288.15 293.15 298.15 303.15

kij [EMIM][SCN] −0.069 −0.070 −0.074 −0.075 0.059 0.062 0.061 0.070 0.071 0.070 0.040 0.043 0.045 [DMIM][Ph] −0.154 −0.153 −0.154 −0.156 0.071 0.068 0.063 0.060 0.045 0.045 0.044 0.045 0.036 0.037 0.034 0.028

AAD % on mole fraction 5.89 4.94 3.59 2.05 7.86 8.48 8.83 3.43 5.81 4.37 11.31 14.10 14.01

Figure 4. Experimental phase equilibrium data of the binary mixtures {H2O + [DMIM][Ph]} at different temperatures: T = ○, 283.15 K; □, 288.15 K; ∗, 293.15 K; ◇, 298.15. Solid lines are the PC-SAFT calculations.

7.35 6.28 4.94 4.80 13.52 9.04 4.09 5.46 16.90 13.34 14.81 14.48 10.16 9.30 8.61 5.70

Figure 5. Experimental phase equilibrium data of the binary mixtures {pyridine + [EMIM][SCN]} at different temperatures: T = ○, 288.15 K; □, 293.15 K; ∗, 298.15. Solid lines are the PC-SAFT calculations.

parameters to be adjusted. These associating values were chosen and kept constant because the length of alkyl chain in the cation has a negligible effect on the strength of the associating bonds. Molecular parameters of the ILs presented in this paper and the absolute average deviation (AAD %) on density obtained after fitting procedures are presented in Table 3. Moreover, Figure 2 depicts the temperature−density diagrams. These results highlight that the density of pure ILs is well correlated by PC-SAFT EoS. Binary Systems. Results of experimental VLE and VLLE data measured in this work with standard uncertainties are provided in Table 4. The values were correlated with the PCSAFT EoS. In this study, binary mixtures are classically depicted with Berthelot−Lorentz combining rule, and a binary interaction parameter (kij) is thus introduced to present the dispersive-cross energy parameter by the equation εij = (εiεj)1/2(1 − kij). The interaction parameters kij of various

Figure 3. Experimental phase equilibrium data of the binary mixtures {H2O + [EMIM][SCN]} at different temperatures: T = ○, 283.15 K; □, 288.15 K; ∗, 293.15 K; ◇, 298.15. Solid lines are the PC-SAFT calculations.

component data with eq 6. Furthermore, the two selfassociating parameters (εAiBj and kAiBj) were assumed constant and taken from those values of 1-alkanols34 (i.e., εAiBj = 3450 K and kAiBj = 0.00225) in order to reduce the number of molecular 609



Article

Figure 6. Experimental phase equilibrium data of the binary mixtures {Pyridine + [DMIM][Ph]} at different temperatures: T = ○, 288.15 K; □, 293.15 K; ∗, 298.15; ◇, 303.15 K. Solid lines are the PC-SAFT calculations.

Figure 8. Experimental phase equilibrium data of the binary mixtures {Toluene + [DMIM][Ph]} at different temperatures: T = ○, 288.15 K; □, 293.15 K; ∗, 298.15; ◇, 303.15 K. Solid lines are the PC-SAFT calculations.

Figure 7. Experimental phase equilibrium data of the binary mixtures {toluene + [EMIM][SCN]} at different temperatures: T = ○, 288.15 K; □, 293.15 K; ∗, 298.15;. Solid lines are the PC-SAFT calculations.

Figure 9. Experimental phase equilibrium data of the binary mixtures {thiophene + [DMIM][Ph]} at different temperatures: T = ○, 288.15 K; □, 293.15 K; ∗, 298.15; ◇, 303.15 K. Solid lines are the PC-SAFT calculations.

binary systems were obtained by minimizing the deviations on mole fraction between experimental VLE data and the calculated values. The corresponding values with absolute average deviations on molar fraction are presented in Table 5. First, the experimental VLE data of two binary mixtures {H2O + [EMIM][SCN]} and {H2O + [DMIM][Ph]} at different temperatures: T = 283.15 K, 288.15 K, 293.15 K, and 298.15 K were correlated, and the corresponding isothermal phase diagrams are depicted in Figure 3 and Figure 4. In the experimental range of temperature, vapor−liquid equilibrium is observed in these systems for the overall range of composition. The high accuracy of the PC-SAFT EoS to correlate these data is obvious. Moreover, [EMIM][SCN] presents a greater selectivity for water than [DMIM][Ph]. Second, binary systems {pyridine + ILs} were also studied at four temperatures: T = 283.15 K, 288.15 K, 293.15 K, and 298.15 K. All experimental data along with their correlation by

the PC-SAFT model are presented in Figures 5 and 6. For the system {pyridine + [EMIM][SCN]} shown in Figure 5, a good agreement between the calculated values and experimental VLE data is observed. For the binary system {pyridine + [DMIM][Ph]} presented in Figure 6, the PC-SAFT EoS predicts a vapor−liquid−liquid line which is not observed experimentally at 288.15 K. The flatness of the experimentally determined bubble curve however highlights that T = 288.15 K is very close to the upper critical solution temperature, and it is thus not surprising that a 3-phase equilibrium was calculated by the PCSAFT EoS. Futhermore, the VLE data evidence that for a given temperature, pyridine is less soluble in [DMIM][Ph] than in [EMIM][SCN]. Third, phase diagrams of binary mixtures {toluene + [EMIM][SCN]} and {toluene + [DMIM][Ph]} are presented in Figures 7 and 8. Vapor−liquid and vapor−liquid−liquid 610



■

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel.: +33 (0) 3.83.17.51.31. Fax: +33 (0)3.83.17.51.52. Notes

The authors declare no competing financial interest.

■

LIST OF SYMBOLS ILs ionic liquids PC-SAFT pertubed chain-statistical associating fluid theory EoS equation of state LLE liquid−liquid equilibrium VLE vapor−liquid equilibrium ãres residual Helmholtz free energy (J·mol−1) ãhc hard chain contribution to Helmholtz free energy (J·mol−1) disp ã dispersion contribution to Helmholtz free energy (J· mol−1) ãassoc associating contribution to Helmholtz free energy (J·mol−1) hs ã hard-sphere contribution to Helmholtz free energy (J·mol−1) m̅ mean segment number xi mole fraction of component i ghs radial pair distribution function of hard-sphere ij ρ̃ number density (number of molecules in unit volume) (Å−3) m segment number (-) ε segment energy (K) σ segment diameter (Å) C1 coefficient number η pure-component reduced density XAi mole fraction of molecules i not bonded at site A Mi number of association sites in component i NAV Avogadro constant ≈ 6.022 × 10−23 (mol−1) ρj molar density of component j (mol·m−3) AiBj Δ association strength εAiBj association energy (K) kAiBj association volume (-) OF objective function AAD % absolute average deviation kij interaction parameter

Figure 10. Experimental phase equilibrium data of the binary mixtures {thiophene + [EMIM][SCN]} at different temperatures: T = ○, 288.15 K; □, 293.15 K; ∗, 298.15. Solid lines are the PC-SAFT calculations.

equilibrium are observed in both binary systems. For both systems, along the three-phase line, temperature has a very small influence on the composition of the liquid-phase rich in IL. Figures 7 and 8 show that this model may be used to predict the phase behaviors of a binary system containing these ILs and toluene with high accuracy. Marciniak35 concluded that hydrocarbons present the highest solubility in the IL with trifluoromethanesulfonate anion and the lowest in [EMIM][SCN]. In this work, we found that [DMIM][Ph] has a poor efficiency for the absorption of toluene as compared to [EMIM][SCN]. Finally, systems {thiophene + ILs} have been studied and reported. Figure 9 depicts the phase diagrams obtained for {thiophene + [DMIM][Ph]} as compared to experimental data. VLE and VLLE are observed in this binary system. Along the three-phase line, the mole fraction of thiophene in the liquid phase rich in IL changes from 0.52 to 0.66 with temperature increases from 298.15 K to 303.15 K. A good agreement between calculated values and experimental vapor−liquid data is demonstrated. However, this model has difficulty representing the binary mixture {thiophene + [EMIM][SCN]} with high accuracy, which is presented in Figure 10. Indeed the PC-SAFT EoS predicts VLE, whereas LLE is experimentally determined. For a given temperature, thiophene is more soluble in [EMIM][SCN] than in [DMIM][Ph].

■

REFERENCES

(1) Zhu, W. S.; Li, H. M.; Hang, X.; Yan, Y. S.; Lu, H. D.; Xia, J. X. Oxidative Desulfurization of Fuels Catalyzed by Peroxotungsten and Peroxomolybdenum Complexes in Ionic Liquids. Energy Fuels 2007, 21, 2514−2516. (2) Baeza, P.; Aguila, G.; Gracia, F.; Araya, P. Desulfurization by Adsorption with Copper Supported on Zirconia. Catal. Commun. 2008, 9, 751−755. (3) Rang, H.; Kann, J.; Oja, V. Advances in Desulfurization Research of Liquid Fuel. Oil Shale 2006, 23, 164−176. (4) Shafi, R.; Hutchings, G. J. Hydrodesulfurization of Hindered Dibenzothiophenes: An Overview. Catal. Today 2000, 59, 423−442. (5) Revelli, A. L.; Mutelet, F.; Turmine, M.; Solimando, R.; Jaubert, J. N. Activity Coefficients at Infinite Dilution of Organic Compounds in 1-Butyl-3-methylimidazolium Tetrafluoroborate Using Inverse Gas Chromatography. J. Chem. Eng. Data 2009, 54, 90−101. (6) Revelli, A. L.; Mutelet, F.; Jaubert, J. N. Prediction of Partition Coefficients of Organic Compounds in Ionic Liquids: Use of a Linear Solvation Energy Relationship with Parameters Calculated through a Group Contribution Method. Ind. Eng. Chem. Res. 2010, 49, 3883− 3892.

■

CONCLUSIONS Eight binary systems encountered in the fuel desulfurization process using ILs were studied in this work. Attention was paid to a proper location of the three-phase line. Furthermore, we observed that [EMIM][SCN] presents a good capacity for the extractive desulfurization of fuels as compared to [DMIM][Ph]. A thermodynamic model based on the PC-SAFT equation of state was employed with success in the correlation of measured data, and the well-known pitfalls of such an EoS36,37 were not evidenced. 611



Article

(27) Huang, S. H.; Radosz, M. Equation of State for Small, Large, Polydisperse, and Associating Molecules. Ind. Eng. Chem. Res. 1990, 29, 2284−2294. (28) Huang, S. H.; Radosz, M. Equation of State for Small, Large, Polydisperse, and Associating Molecules: Extension to Fluid Mixtures. Ind. Eng. Chem. Res. 1991, 30, 1994−2005. (29) Revelli, A. L.; Mutelet, F.; Jaubert, J. N. (Vapor + Liquid) Equilibria of Binary Mixtures Containing Light Alcohols and Ionic Liquids. J. Chem. Thermodyn. 2010, 42, 177−181. (30) Freire, M. G.; Teles, A. R. R.; Rocha, M. A. A.; Schroder, B.; Neves, C.; Carvalho, P. J.; Evtuguin, D. V.; Santos, L.; Coutinho, J. A. P. Thermophysical Characterization of Ionic Liquids Able to Dissolve Biomass. J. Chem. Eng. Data 2011, 56, 4813−4822. (31) Van Niekerk, D.; Castro-Marcano, F.; Colina, C. M.; Mathews, J. P. Solvent Swelling Extent of Permian-Aged Vitrinite- and InertiniteRich Coals: Experiments and Modeling Using the Perturbed-Chain Statistical Associating Fluid Theory (PC-SAFT). Energy Fuels 2011, 25, 2559−2564. (32) Zuniga-Morena, A.; Galicia-Luna, L. A.; Betancourt-Cardenas, F. F. Compressed Liquid Densities and Excess Volumes of CO2+ Thiophene Binary Mixtures from 313 K to 363 K and Pressures up to 25 MPa. Fluid Phase Equilib. 2005, 236, 193−204. (33) Gross, J.; Sadowski, G. Application of the Perturbed-Chain SAFT Equation of State to Associating Systems. Ind. Eng. Chem. Res. 2002, 41, 5510−5515. (34) Pamies, J. C. Vapor-liquid equilibria and critical behavior of heavy nalkanes using transferable parameters from the softSAFT equation of state. Ph.D. Thesis, Universitat Rovira i Virgili, Tarragona, Spain 2003. (35) Marciniak, A. Influence of Anion Structure on the Liquid− Liquid Equilibria of 1-Ethyl-3-methyl-imidazolium Cation Based Ionic Liquid−Hydrocarbon Binary Systems. J. Chem. Eng. Data 2011, 56, 368−374. (36) Privat, R.; Gani, R.; Jaubert, J.-N. Are Safe Results Obtained when the PC-SAFT Equation of State Is Applied to Ordinary Pure Chemicals. Fluid Phase Equilib. 2010, 295, 76−92. (37) Privat, R.; Conte, E.; Jaubert, J. N.; Gani, R. Are Safe Results Obtained when SAFT Equations Are Applied to Ordinary Chemicals? Part 2: Study of Solid−Liquid Equilibria in Binary Systems. Fluid Phase Equilib. 2012, 318, 61−76.

(7) Revelli, A. L.; Mutelet, F.; Jaubert, J. N. Partition Coefficients of Organic Compounds in New Imidazolium Based Ionic Liquids Using Inverse Gas Chromatography. J. Chromatogr. A 2009, 1216, 4775− 4786. (8) Bosmann, A.; Datsevich, L.; Jess, A.; Lauter, A.; Schmitz, C.; Wasserscheid, P. Deep Desulfurization of Diesel Fuel by Extraction with Ionic Liquids. Chem. Commun. 2001, 2494−2495. (9) Kedra-Krolik, K.; Fabrice, M.; Jaubert, J. N. Extraction of Thiophene or Pyridine from n-Heptane Using Ionic Liquids. Ind. Eng. Chem. Res. 2011, 50, 2296−2306. (10) Mochizuki, Y.; Sugawara, K. Removal of Organic Sulfur from Hydrocarbon Resources Using Ionic Liquids. Energy Fuels 2008, 22, 3303−3307. (11) Kulkarni, P. S.; Afonso, C. A. M. Deep Desulfurization of Diesel Fuel Using Ionic Liquids: Current Status and Future Challenges. Green Chem. 2010, 12, 1139−1149. (12) Hansmeier, A. R.; Meindersma, G. W.; de Haan, A. B. Desulfurization and Denitrogenation of Gasoline and Diesel Fuels by Means of Ionic Liquids. Green Chem. 2011, 13, 1907−1913. (13) Nefedieva, M.; Lebedeva, O.; Kultin, D.; Kustov, L.; Borisenkova, S.; Krasovskiy, V. Ionic Liquids Based on Imidazolium Tetrafluoroborate for the Removal of Aromatic Sulfur-Containing Compounds from Hydrocarbon Mixtures. Green Chem. 2010, 12, 346−349. (14) Francisco, M.; Arce, A.; Soto, A. Ionic Liquids on Desulfurization of Fuel Oils. Fluid Phase Equilib. 2010, 294, 39−48. (15) Zhang, S. G.; Zhang, Q. L.; Zhang, Z. C. Extractive Desulfurization and Denitrogenation of Fuels Using Ionic Liquids. Ind. Eng. Chem. Res. 2004, 43, 614−622. (16) Ko, N. H.; Lee, J. S.; Huh, E. S.; Lee, H.; Jung, K. D.; Kim, H. S.; Cheong, M. Extractive Desulfurization Using Fe-Containing Ionic Liquids. Energy Fuels 2008, 22, 1687−1690. (17) Huang, C. P.; Chen, B. H.; Zhang, J.; Liu, Z. C.; Li, Y. X. Desulfurization of Gasoline by Extraction with New Ionic Liquids. Energy Fuels 2004, 18, 1862−1864. (18) Kedra-Krolik, K.; Mutelet, F.; Moise, J. C.; Jaubert, J. N. Deep Fuels Desulfurization and Denitrogenation Using 1-Butyl-3-methylimidazolium Trifluoromethanesulfonate. Energy Fuels 2011, 25, 1559− 1565. (19) Alonso, L.; Arce, A.; Francisco, M.; Soto, A. Extraction Ability of Nitrogen-Containing Compounds Involved in the Desulfurization of Fuels by Using Ionic Liquids. J. Chem. Eng. Data 2010, 55, 3262− 3267. (20) Zhang, S. G.; Zhang, Z. C. Novel Properties of Ionic Liquids in Selective Sulfur Removal from Fuels at Room Temperature. Green Chem. 2002, 4, 376−379. (21) Alonso, L.; Arce, A.; Francisco, M.; Rodriguez, O.; Soto, A. Liquid−Liquid Equilibria for Systems Composed by 1-Methyl-3octylimidazolium Tetrafluoroborate Ionic Liquid, Thiophene, and nHexane or Cyclohexane. J. Chem. Eng. Data 2007, 52, 1729−1732. (22) Revelli, A. L.; Mutelet, F.; Jaubert, J. N. Extraction of Benzene or Thiophene from n-Heptane Using Ionic Liquids. NMR and Thermodynamic Study. J. Phys. Chem. B 2010, 114, 4600−4608. (23) Domanska, U.; Marciniak, A. Measurements of Activity Coefficients at Infinite Dilution of Aromatic and Aliphatic Hydrocarbons, Alcohols, and Water in the New Ionic Liquid EMIM SCN Using GLC. J. Chem. Thermodyn. 2008, 40, 860−866. (24) Domanska, U.; Laskowska, M. Measurements of Activity Coefficients at Infinite Dilution of Aliphatic and Aromatic Hydrocarbons, Alcohols, Thiophene, Tetrahydrofuran, MTBE, and Water in Ionic Liquid BMIM SCN Using GLC. J. Chem. Thermodyn. 2009, 41, 645−650. (25) Domanska, U.; Zawadzki, M.; Paduszynski, K.; Krolikowski, M. Perturbed-Chain SAFT as a Versatile Tool for Thermodynamic Modeling of Binary Mixtures Containing Isoquinolinium Ionic Liquids. J. Phy. Chem. B 2012, 116, 8191−8200. (26) Gross, J.; Sadowski, G. Perturbed-Chain SAFT: An Equation of State Based on a Perturbation Theory for Chain Molecules. Ind. Eng. Chem. Res. 2001, 40, 1244−1260. 612


Experimental Measurement and Modeling of Phase Diagrams of

Recommend Documents