Activity coefficients at infinite dilution of alkanes, alkenes, alkyl benzenes in dimethylphosphate based ionic liquids using gas–liquid chromatography

J. Chem. Thermodynamics 91 (2015) 279–285

Contents lists available at ScienceDirect

J. Chem. Thermodynamics journal homepage: www.elsevier.com/locate/jct

Activity coefficients at infinite dilution of alkanes, alkenes, alkyl benzenes in dimethylphosphate based ionic liquids using gas–liquid chromatography Ming-Lan Ge a,⇑, Chun-Yang Lu a, Xiao-Yan Liu a, Xiang-Bo Li b,a, Jin-Yuan Chen a, Jie-Ming Xiong a a b

Department of Chemical Engineering, Beijing Institute of Petrochemical Technology, Beijing 102617, People’s Republic of China College of Chemical Engineering, Beijing University of Chemical Technology, Beijing 100029, People’s Republic of China

a r t i c l e

i n f o

Article history: Received 8 April 2015 Received in revised form 9 July 2015 Accepted 30 July 2015 Available online 8 August 2015 Keywords: Activity coefficient at infinite dilution Gas–liquid chromatographic Dimethylphosphate based ionic liquids Organic solutes Selectivity

a b s t r a c t Activity coefficients at infinite dilution ðc1 i Þ for organic solutes: alkanes, alkenes, and alkyl benzenes in two ionic liquids (ILs) 1,3-dimethylimidazolium dimethylphosphate ([MMIM][DMP]) and 1-ethyl-3methylimidazolium dimethylphosphate ([EMIM][DMP]) have been measured by the gas–liquid chromatographic method (GLC). The measurements were carried out in the temperature range of (313.15 Þ were derived to 363.15) K. The values of the partial molar excess enthalpies at infinite dilution ðHE;1 i E;1 Þ and Gibbs energies ðGiE;1 Þ from the temperature dependence of the c1 i values. The entropies ðT ref Si of organic solutes at a reference temperature Tref = 298.15 K were also calculated from the c1 i values. 1 Selectivity ðS1 ij Þ and capacity ðkj Þ at infinite dilution at T = 323.15 K have been determined for hexane (i)/benzene (j), cyclohexane (i)/benzene (j). The results were analyzed in comparison to literature data for other ILs with the [MMIM] and [EMIM] cations. For three isomeric xylenes separation problems, selectivity at T = 323.15 K were also obtained from the c1 i values. Ó 2015 Published by Elsevier Ltd.

1. Introduction Ionic liquids (ILs) have been considered to be environmental friendly solvents in recent years due to their unique physical and chemical properties, such as negligible vapor pressures, a high solvating capacity for both polar and non-polar compounds, high thermal stability, high ionic conductivity, nonflammability, and large liquid-state temperature range [1–4]. For ILs to be used effectively as solvents, it is essential to know their interaction with different solutes. The activity coefficient at infinite dilution ðc1 i Þ describes the degree of nonideality for species i in a mixture, gives a quantitative measure of interactions between unlike molecules in the absence of solute–solute interactions. Values of c1 also provide information on the intermolecular i energy between ILs and organic solutes and can be used to quantify the selectivity and solvent power of ILs [5–8]. Activity coefficients at infinite dilution have a wide range of applications in the field of chemical engineering and can be used for the pre-screening of solvents to be used in unit operations such as extractive distillation and liquid–liquid extraction. The use of a steady-state gas–liquid ⇑ Corresponding author. E-mail addresses: [email protected], [email protected] (M.-L. Ge). http://dx.doi.org/10.1016/j.jct.2015.07.046 0021-9614/Ó 2015 Published by Elsevier Ltd.

chromatographic (GLC) technique for the determination of c1 i has been employed by many researchers and has proved quite reliable [9]. ILs are in particular reliable for the determination of c1 i by this method because their negligible vapor pressures and high thermal stability make themselves to be ideal stationary phases [2]. Until now, many research groups have measured c1 i of various solutes in a number of ILs using the gas–liquid chromatograph [10–16]. Our group has focused attention on the determination of c1 i of various solutes (i) in ILs by the GLC method [17–20]. In this paper, c1 have been measured for 16 organic solutes: alkanes i (pentane, hexane, heptane, octane, nonane, decane, cyclohexane, methylcyclohexane, and 2,2,4-trimethylpentane), alkenes (cyclohexene and styrene), and alkyl benzenes (benzene, toluene, o-xylene, m-xylene, and p-xylene) in 1,3-dimethylimidazolium dimethylphosphate ([MMIM][DMP], CAS Registry No. 654058-045) and 1-ethyl-3-methylimidazolium dimethylphosphate ([EMIM] [DMP] CAS Registry No. 945611-27-8) by the GLC method in the temperature range of (313.15 to 363.15) K. The values of the partial molar excess enthalpies at infinite dilution ðHE;1 Þ were derived i from the temperature dependence of the c1 values. The i Þ and Gibbs energies ðGE;1 Þ of organic solutes at entropies ðT ref SE;1 i i a reference temperature Tref = 298.15 K were also determined from the c1 i values.

280

M.-L. Ge et al. / J. Chem. Thermodynamics 91 (2015) 279–285 1

The selectivity ðS1 ij Þ and the capacity ðkj Þ at infinite dilution directly calculated from c1 i offer an important means to evaluate the performance of ILs as solvents in various separation problems. 1 S1 ij and kj at T = 323.15 K for two ILs have been also calculated for hexane (i)/benzene (j), cyclohexane (i)/benzene (j). The results were analyzed in comparison to literature data for other ILs with the [MMIM] and [EMIM] cations. For three isomeric xylenes separation problems, selectivity at T = 323.15 K were also obtained from the c1 i values. 2. Experimental 2.1. Chemicals and materials The ILs [MMIM][DMP] and [EMIM][DMP] were purchased from Shanghai Chengjie Chemical Co., Ltd. and had a purity of >0.99 mass fraction according to manufacturer’s specifications, with the following certified mass fraction of impurities: w (Cl) < 5  104, water < 103. Before use, the ILs were subjected to vacuum evaporation at T = (323 to 333) K over 24 h to remove possible traces of solvents and moisture. The water mass fraction analyzed by Karl Fischer analysis was less than 4  104. The chemical structures of [MMIM][DMP] and [EMIM][DMP] are given in figure 1. The organic solutes were purchased from Beijing Chemical Reagents Company. Their mass fraction purities were greater than 0.99. The solutes were used without further purification. The sources and mass fraction purities of materials used are listed in table 1S in Supplementary Material.

work we always use the large column packing, which prevents the residual adsorption of solute onto the column packing. The measurements for organic solutes were carried out in the temperature range of (313.15 to 363.15) K. The columns were filled uniformly with the help of an ultrasound vibrator and finally heated under nitrogen for 8 h at the column temperature of 160 ° C. Dry nitrogen was used as the carrier gas. The flow rate of carrier gas was determined using a calibrated soap bubble flowmeter which was placed at the outlet after the detector. The flow rate was set for a series of runs and was allowed to stabilize for at least 15 min before any c1 i determinations were made. The volume of the samples injected into the GC probes was about (0.05 to 0.5) lL, and the peaks were found to be symmetrical, independent of the carrier gas flow rate. The temperature of the GC column was maintained constant within ±0.05 K. At a given temperature, each operation was repeated at least three times to check the reproducibility. The deviation of the retention time of the three measurements was within ±0.05 min. The value of the dead time tG was determined with methane as the nonretainable pure component under the assumption that the effect of the solubility of methane in the ILs was negligible. The measured dead time in the temperature range has a deviation of ±0.01 min. To check the stability of the experimental conditions, such as the possible elution of the stationary phase by the carrier gas stream, the measurements of retention time were repeated systematically every (6 to 8) h for hexane and benzene. No change of the retention time was observed during 80 h of continuous operation.

2.2. Apparatus and procedure

3. Theory

The experiments were performed on a SP-3420A gas chromatograph equipped with a thermal conductivity detector. The columns preparation and the packing method used in this work have been described previously [16,20]. Chromosorb WAW DMCS 80/100 mesh was used as the solid support and was supplied by SUPELCO. Coating the solid support with ILs was performed by dispersing a known mass amount of the Chromosorb in a solution of IL in ethanol followed by evaporation of the solvent in a rotating evaporator. The Chromosorb was weighed on an electronic balance of precision of ±0.0001 g before and after the coating process. The column packing in this work were 35.98% (3.374 mmol) mass percent of [MMIM][DMP] and 44.11% (2.077 mmol) mass percent of [EMIM][DMP] with a precision of ±0.0001 g. In our experimental

In (gas–liquid) chromatography, the activity coefficient at infinite dilution c1 were obtained by the equation proposed by i Cruickshank et al. [21] and Everett [22].

O

N

H 3C

N

O

-

P

O

CH3

O

C2 H 5

CH3

H 3C

O

1,3-dimethylimidazolium dimethylphosphate O

N H3 C

C2H5

N

O

P

C2 H5 O

1-Ethyl-3-methylimidazolium dimethylphosphate FIGURE 1. The chemical structures of 1,3-dimethylimidazolium dimethylphosphate and 1-ethyl-3-methylimidazolium dimethylphosphate.

ln c1 i ¼ ln



 n3 RT Bii  v i 0 2Bi2  v 1 i  pi þ J 32 po ; 0 RT RT V N pi

ð1Þ

where c1 i is the activity coefficient of solute i at infinite dilution in the stationary phase (3), p0i is the vapor pressure of the pure liquid solute i, n3 is the number of moles of the stationary phase component on the column, and VN is the standardized retention volume obtained by equation (2), 1

V N ¼ ðJ 32 Þ U 0 ðtr  tG Þ

  T col p0 1 w ; Tf po

ð2Þ

where tr denotes the retention time, tG the dead time, U0 the flow rate of the carrier gas, Tcol the column temperature, Tf the flowmeter temperature, p0w the saturation vapor pressure of water at Tf, and po the pressure at the column outlet. The second and third terms in equation (1) are correction terms arising from the nonideality of the mobile gaseous phase and the effect of pressure. Bii is the second virial coefficient of the solute, Bi2 is the cross second virial coefficient of the solute (i) with the carrier gas (2), vi is the liquid molar volume of pure solute, and v1 i is the partial molar volume of the solute in the stationary phase (3) at infinite dilution. For all solutes, values of p0i were calculated from the Antoine equation, with Antoine coefficients given by Boublik et al. [23]. Molar volumes of solutes vi were estimated using their experimental densities [24]; partial molar volumes of solutes at infinite dilution v 1 i have been assumed to be equal to vi. Bii and Bi2 have been estimated according to the equations suitable for nonpolar liquids by Tsonopolous’s method [25] with an uncertainty of <±10 cm3  mol1. The critical parameters and acentric factor x needed for the calculations were available from the literature

281

M.-L. Ge et al. / J. Chem. Thermodynamics 91 (2015) 279–285

[24,25]. The cross critical properties pcij, Tcij, vcij, Zcij, and acentric factor xij were calculated by using equations given in the literatures [25,26]. The vapor pressure of the solutes (i) at temperatures of (313.15 to 363.15) K, the critical constants Tc, Pc, Zc, Vc, and acentric factors x of the solutes and the carrier gas used in calculation of the virial coefficients were presented in the Supplementary Materials of our previous paper [20].

TABLE 3 in the ionic liquid 1,3-dimethylimidazolium Comparison of values of c1 i dimethylphosphate ([MMIM][DMP]) obtained in this work with c1 measured by i Gmehling et al. [28] at T = 323.15 K.

The pressure correction term J 32 is given by [27]

J 32 ¼

2 ðpi =po Þ3  1 ; 3 ðpi =po Þ2  1

ð3Þ

where pi and po are the inlet and outlet pressures of the GC column, respectively. The inlet column pressure pi was determined by inner manometer. Outlet pressure po was kept equal to atmospheric pressure. 4. Results and discussion Experimental results of c1 i for the solutes in [MMIM][DMP] and [EMIM][DMP] in the temperature range of (313.15 to 363.15) K are presented in tables 1 and 2, respectively. For the two ILs, the c1 i values for the linear n-alkanes increase with increasing chain

a

Solutes (i)

c1 i this work

a c1 i

Octane Cyclohexane Cyclohexene Benzene Toluene

325.1 55.98 31.64 3.622 7.538

335 44.5 25.2 3.61 7.30

Reference [28].

length. The branching of the alkane skeleton (e.g., cyclohexane, methylcyclohexane, or 2,2,4-trimethylpentane) reduces the values of c1 i in comparison with the corresponding linear alkanes: hexane, heptane, and octane. The introduction of the double bond in the six-membered ring (cyclohexene) causes a reduction of c1 i . For the aromatic compounds, the values of c1 i are distinctly lower in comparison with those of the alkanes, and the values of c1 i increase with increasing size of the alkyl group. The high c1 i values of alkanes indicate low solubility and weak solute-IL interactions. Cyclic alkanes, alkenes, and aromatic molecules interact more strongly with the ILs, as indicated by the lower c1 values. The i

TABLE 1 Experimental activity coefficients at infinite dilution c1 i for various solutes in the ionic liquid 1,3-dimethylimidazolium dimethylphosphate ([MMIM][DMP]) at po = 101.3 kPa and different temperatures.a Solutes (i)

Pentane Hexane Heptane Octane Nonane Decane Cyclohexane Methylcyclohexane 2,2,4-trimethylpentane Cyclohexene Styrene Benzene Toluene o-Xylene m-Xylene p-Xylene a

T (K)

Standard state

313.15

323.15

333.15

343.15

353.15

363.15

245.3 259.6 296.9 339.9 400.2 502.6 58.37 83.41 318.3 32.74 / 3.637 7.567 12.66 14.55 13.75

230.1 246.3 278.6 325.1 383.9 480.3 55.98 80.82 300.6 31.64 6.622 3.622 7.538 12.49 14.38 13.58

217.4 229.9 258.3 301.8 368.2 465.7 52.56 78.33 283.9 30.67 6.608 3.613 7.511 12.38 14.24 13.48

200.9 218.5 243.4 284.7 350.9 448.4 50.94 75.38 269.5 30.05 6.597 3.588 7.483 12.23 14.17 13.37

183.8 203.4 231.4 270.1 345.1 430.8 48.03 73.02 244.2 29.12 6.581 3.572 7.468 12.12 14.05 13.21

170.5 186.7 215.9 247.8 326.8 413.7 45.68 69.91 228.8 28.27 6.571 3.562 7.445 11.98 13.93 13.13

Gas Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid

T < 341.85 K

T < 353.85 K

T < 356.05 K T < 353.15 K

Standard uncertainties (u) are as follows: u(po) = ±1 kPa, uðc1 i Þ = ±5%, and u(T) = ±0.05 K.

TABLE 2 Experimental activity coefficients at infinite dilution c1 i for various solutes in the ionic liquid 1-ethyl-3-methylimidazolium dimethylphosphate ([EMIM][DMP]) at po = 101.3 kPa and different temperatures.a Solutes (i)

Pentane Hexane Heptane Octane Nonane Decane Cyclohexane Methylcyclohexane 2,2,4-trimethylpentane Cyclohexene Styrene Benzene Toluene o-Xylene m-Xylene p-Xylene a

T (K)

Standard state

313.15

323.15

333.15

343.15

353.15

363.15

124.8 138.7 162.5 201.9 271.8 429.9 53.90 77.65 182.8 25.37 5.45 3.51 6.45 10.20 12.42 12.07

106.5 122.1 150.6 187.2 242.9 388.1 48.92 68.34 170.3 22.40 5.29 3.33 6.14 9.80 11.76 11.46

88.5 107.1 138.4 173.3 224.8 354.4 44.00 61.54 157.5 20.03 5.16 3.20 5.94 9.59 11.24 10.92

69.8 87.6 123.5 158.2 206.8 316.0 40.57 55.93 143.3 18.70 5.01 3.12 5.73 9.30 10.85 10.52

60.5 72.2 110.6 143.7 196.8 294.3 37.70 50.45 131.9 16.87 4.86 3.02 5.56 9.14 10.50 9.99

52.9 61.5 97.7 132.1 182.3 245.4 32.88 45.48 119.1 15.52 4.71 2.87 5.40 9.00 9.99 9.53

Standard uncertainties (u) are as follows: u(po) = ±1 kPa, uðc1 i Þ = ±5%, and u(T) = ±0.05 K.

Gas Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid Liquid

T < 341.85 K

T < 353.85 K

T < 356.05 K T < 353.15 K

282

M.-L. Ge et al. / J. Chem. Thermodynamics 91 (2015) 279–285

toluene. For cyclohexane generally differences are about 20%. The experimental data and those reported in reference [28] for cyclohexane, cyclohexene, benzene, and toluene are compared graphically in figure 2. The temperature trends are the same for cyclohexane, but are opposite for cyclohexene, benzene, and toluene. For cyclohexene, the differences are even larger than 20%, if compared in the whole overlapping temperature range. The differences of experimental method (GLC and dilutor technique), and the thermophysical data of solutes taken to the calculations may result in the discrepancy in the results. From the temperature dependence of the activity coefficients at infinite dilution, ln c1 i can be split to its respective enthalpy and entropy terms

60 50 40

γi

30 20 10 0 300

310

320

330

340

350

360

370

HE;1 SE;1 i  i ; RT R

ln c1 i ¼

T/K

FIGURE 2. Plot of c1 i vs T for cyclohexane, cyclohexene, benzene, and toluene in [MMIM][DMP]: j, cyclohexane (*); s, cyclohexane (**); d, cyclohexene (*); ., cyclohexene (**); h, toluene (*); N, toluene (**); 4, benzene (*); 5, benzene (**). (*) this work; (**) reference [28].

smaller values indicate the stronger interactions between solvent and solute. Table 3 compares values of c1 i obtained in this work with those measured using dilutor technique by Gmehling et al. [28] The numerical differences are very small for octane, benzene, and

ð4Þ

where R is the gas constant. The temperature dependence of the activity coefficients can be calculated from equation (5)

ln c1 i ¼ aþ

b ; ðT=KÞ

ð5Þ

Thus the partial molar excess enthalpy HiE;1 ¼ Rb and entropy SiE;1 ¼ Ra at infinite dilution can be calculated from its slope and intercept, respectively.

TABLE 4 E;1 Coefficients a and b of equation (11), standard deviation r, c1 ), i at T = 298.15 K calculated using equation (5), values of the partial molar excess enthalpies at infinite dilution ðH i entropies ðT ref SE;1 ), and Gibbs energies ðGE;1 ) of organic solutes in [MMIM][DMP] at a reference temperature Tref = 298.15 K.a i i

a

Solute (i)

a

b (K)

c1 i 298:15 K

HE;1 =ðkJ  mol i

Pentane Hexane Heptane Octane Nonane Decane Cyclohexane Methylcyclohexane 2,2,4-trimethylpentane Cyclohexene Styrene Benzene Toluene o-Xylene m-Xylene p-Xylene

2.8612 3.2309 3.3792 3.5669 4.5605 4.8908 2.2797 3.1948 3.3755 2.4375 1.7843 0.3786 0.9034 0.3941 0.5785 0.7349

832.13 734.43 724.87 712.72 448.41 415.72 561.67 385.69 753.09 328.57 34.83 682.46 1140.05 1126.67 1274.99 1302.07

284.9 297.1 333.8 386.6 430.3 536.5 64.30 88.98 365.6 34.45 6.69 3.67 7.59 12.94 14.87 14.00

6.92 6.11 6.03 5.93 3.73 3.46 4.67 3.21 6.26 2.73 0.29 0.43 0.24 1.11 0.94 0.94

Standard uncertainties (u) are as follows: uðHE;1 Þ ¼ 0:5 kJ  mol i

1

, uðGE;1 Þ ¼ 0:5 kJ  mol i

1

Þ

T ref SE;1 =ðkJ  mol i

1

Þ

7.09 8.01 8.38 8.84 11.30 12.12 5.65 7.92 8.37 6.04 4.42 2.79 4.78 5.24 5.75 5.60 1

GE;1 =ðkJ  mol i

1

Þ

14.01 14.11 14.40 14.77 15.03 15.58 10.32 11.13 14.63 8.77 4.71 3.22 5.02 6.35 6.69 6.54

, and uðT ref SE;1 Þ ¼ 0:05 kJ  mol i

1

r 0.017 0.018 0.007 0.015 0.007 0.006 0.008 0.006 0.022 0.004 0.003 0.003 0.003 0.005 0.001 0.001

.

TABLE 5 E;1 Coefficients a and b of equation (11), standard deviation r, c1 Þ, i at T = 298.15 K calculated using equation (5), values of the partial molar excess enthalpies at infinite dilution ðH i E;1 a entropies ðT ref SE;1 Þ, and Gibbs energies ðG Þ of organic solutes in [EMIM][DMP] at a reference temperature T = 298.15 K . ref i i

a

Solute (i)

a

b (K)

c1 i 298:15 K

HiE;1 =ðkJ  mol

Pentane Hexane Heptane Octane Nonane Decane Cyclohexane Methylcyclohexane 2,2,4-trimethylpentane Cyclohexene Styrene Benzene Toluene o-Xylene M-Xylene p-Xylene

1.6230 1.0597 1.4216 2.2144 2.7670 2.2108 0.5410 0.5283 2.1225 0.2877 0.6490 0.1145 0.5830 1.4158 0.9988 0.8011

2025.54 1891.05 1158.55 973.84 884.27 1212.78 1081.64 1196.76 972.38 1099.67 328.81 427.87 399.74 281.77 474.90 529.74

176.0 196.9 201.8 240.0 308.8 533.0 64.64 93.90 217.9 29.98 5.765 3.745 6.846 10.60 13.35 13.17

16.84 15.72 9.63 8.10 7.35 10.08 8.99 9.95 8.08 9.14 2.73 3.56 3.32 2.34 3.95 4.40

Standard uncertainties (u) are as follows: uðHE;1 Þ ¼ 0:5 kJ  mol i

1

1

, uðGE;1 Þ ¼ 0:5 kJ  mol i

Þ

T ref SE;1 =ðkJ  mol i

1

Þ

4.02 2.63 3.52 5.49 6.86 5.48 1.34 1.31 5.26 0.71 1.61 0.28 1.45 3.51 2.48 1.99 1

, and uðT ref SE;1 Þ ¼ 0:05 kJ  mol i

GiE;1 =ðkJ  mol 12.82 13.10 13.16 13.59 14.21 15.56 10.33 11.26 13.35 8.43 4.34 3.27 4.77 5.85 6.42 6.39

1

.

1

Þ

r 0.027 0.042 0.028 0.015 0.013 0.034 0.020 0.007 0.018 0.012 0.005 0.008 0.004 0.006 0.006 0.005

283

6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0 3.8 3.6 2.7

lnγi

lnγi

M.-L. Ge et al. / J. Chem. Thermodynamics 91 (2015) 279–285

2.8

2.9

3.0

3.1

3.2

6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0 3.8 3.6 3.4 2.7

2.8

2.9

3.2 3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 1.4 2.8

2.9

3.0

3.1

1.2 2.7

3.2

2.8

2.9

2.8

2.9

3.0

3.1

3.2

1000K/T FIGURE 5. Plot of ln c1 i vs 1/T for alkyl benzenes in [MMIM][DMP] together with a linear correlation of the data: j, benzene; d, toluene; N, o-xylene; ., m-xylene; s, p-xylene; — linear correlation.

3.0

3.1

3.2

1000K/T FIGURE 7. Plot of ln c1 i vs 1/T for alkenes in [EMIM][DMP] together with a linear correlation of the data: j, cyclohexene; s, styrene. — linear correlation.

lnγi

FIGURE 4. Plot of ln c1 i vs 1/T for alkenes in [MMIM][DMP] together with a linear correlation of the data: j, cyclohexene; s, styrene. — linear correlation.

lnγi

3.2

3.4

1000K/T

2.8 2.7 2.6 2.5 2.4 2.3 2.2 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 2.7

3.1

FIGURE 6. Plot of ln c1 i vs 1/T for alkanes in [EMIM][DMP] together with a linear correlation of the data: j, pentane; d, hexane; N, heptane; ., octane; r, nonane; 4, decane; , cyclohexane; h, methylcyclohexane; s, 2,2,4-trimethylpentane; — linear fit.

lnγi

lnγi

FIGURE 3. Plot of ln c1 i vs 1/T for alkanes in [MMIM][DMP] together with a linear correlation of the data: j, pentane; d, hexane; N, heptane; ., octane; r, nonane; 4, decane; , cyclohexane; h, methylcyclohexane; s, 2,2,4-trimethylpentane; — linear fit.

4.0 3.8 3.6 3.4 3.2 3.0 2.8 2.6 2.4 2.2 2.0 1.8 1.6 2.7

3.0 1000K/T

1000K/T

2.6 2.5 2.4 2.3 2.2 2.1 2.0 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1.0 0.9 0.8 2.7

2.8

2.9

3.0

3.1

3.2

1000K/T FIGURE 8. Plot of ln c1 i vs 1/T for alkyl benzenes in [EMIM][DMP] together with a linear correlation of the data: j, benzene; d, toluene; N, o-xylene; ., m-xylene; s, p-xylene; — linear correlation.

284

M.-L. Ge et al. / J. Chem. Thermodynamics 91 (2015) 279–285 TABLE 6 1 Selectivity (S1 ij ) and capacity ðkj Þ at infinite dilution for different ionic liquids with 1,3-dimethylimidazolium cation or 1-ethyl-3-methylimidazolium cation at T = 323.15 K. S1 ij

Ionic liquids

[MMIM][CH3OC2H4SO4] [MMIM] [CH3SO4] [MMIM][DMP] [MMIM][N(CF3SO2)2] [MMIM][DMP] [EMIM][NO3] [EMIM][BF4] [EMIM][DEP] [EMIM][N(CF3SO2)2] [EMIM][EtSO4] [EMIM][N(CF3SO2)2] [EMIM][CF3SO3] [EMIM][SCN] [EMIM][FAP] [EMIM][DCA] [EMIM][MDEGSO4] [EMIM][TCB] [EMIM][DMP] a

1

kj

Hexane (i)/ benzene (j)

Cyclohexane (i)/ benzene (j)

(benzene)

39.6 13.3 – 23.1 68.0 – 43.0 27.9 20.0a 40.8a 20.6 28.5 70.4 7.48 43.4 39.0 11.7 36.6

24.7 16.2 12.5 13.2 15.4 23.9 23.0 13.5 11.2a 21.7a 11.4 14.1 25.7 7.32 17.9 17.2 8.36 14.7

0.23 0.16 0.28 0.74 0.28 0.21 0.48 0.41 0.83 0.36 0.85 0.45 0.29 0.92 / 0.41 0.75 0.30

References

[28] [28] [28] [30] This work [7] [17] [18] [30] [30] [31] [32] [33] [34] [35] [36] [37] This work

The values of c1 i from dilutor technique.

The coefficients a and b, the standard deviation r of the fitted equations, and the values of c1 i at T = 298.15 K are listed in tables 4 and 5, respectively. The plots of measured ln c1 i versus 1/T values and the linear fit of their data are given in figures 3 to 8, respectively, which show a fairly good fitting quality of equation (5). for the solutes studied are also listed in tables The values of HE;1 i all have positive values. This 4 and 5. For all the 16 solutes, HE;1 i is consistent with the fact that the c1 values of the solutes are i observed to decrease with an increase in temperature. The limiting ¼ RT ln c1 partial molar excess Gibbs energies GE;1 i of all the studi ied solutes in [MMIM][DMP] and [EMIM][DMP] at a reference temperature 298.15 K are also given in tables 4 and 5. As seen from tables 4 and 5, the GE;1 values of aliphatic hydrocarbons differs disi tinctly from that of other solutes. Obviously, the higher the cohesive energy density of the solute, the more energy has to be spent for breaking the solute–solvent interactions during the solution process. In case of aliphatic hydrocarbons this energy penalty is not compensated by formation of sufficiently strong solute–solvent interactions, which leads to poor miscibility of the aliphatics with [MMIM][DMP] and [EMIM][DMP]. 1 1 1 The selectivity, S1 ij , is defined as Sij ¼ ci;IL =cj;IL [29] (where i and j 1

refers to the solutes to be separated) and the capacity ðkj Þ is defined as follow:

1 kj

¼ 1=c1 j;IL . Table 6 lists

S1 ij

and

1 kj

through GLC measurements. The activity coefficients have been calculated for a series of organic solutes: alkanes, alkenes, and alkyl benzenes at the temperature range from (313.15 to 363.15) K. The values of the partial molar excess enthalpies at infinite dilution in two ILs were derived. The entropies and Gibbs energies of organic solutes at a reference temperature Tref = 298.15 K were also calculated from the c1 i values. Selectivity and the capacity at infinite dilution at T = 323.15 K have been determined for hexane (i)/benzene (j), cyclohexane (i)/benzene (j), and three isomeric xylenes separation problems. The results indicated that [MMIM][DMP] is a better extraction solvent for separation of hexane (i)/benzene (j) and is not the most ideal for cyclohexane (i)/benzene (j) binary systems. The studied ILs are not ideal extraction solvents for separation of xylenes. Acknowledgments This work was supported by Beijing Institute of Petrochemical Technology URT Program of National Level (Grant No. 2014X00009) and China National Science & Technology Pillar Program during the 12th Five-year Plan Period (2015BAK16B03). Appendix A. Supplementary data

for ILs based

on 1,3-dimethylimidazolium and 1-ethyl-3-methylimidazolium cations for hexane (i)/benzene (j) and cyclohexane (i)/benzene (j) separation problems at T = 323.15 K. The results indicated that [MMIM][DMP] is a better extraction solvent for separation of hexane (i)/benzene (j) and is not the most ideal for cyclohexane (i)/ benzene (j) binary systems, but as a solvent, many factors must be thought of, such as density, viscosity, toxicity, and cost. For [MMIM][DMP], the values of S1 ij for three isomeric xylenes at T = 323.15 K were 1.15 (m-xylene/o-xylene), 1.06 (m-xylene/pxylene), and 1.09 (p-xylene/o-xylene), and for [EMIM][DMP] were 1.20, 1.03, and 1.18, respectively. The results indicated that [MMIM][DMP] and [EMIM][DMP] were not ideal extraction solvents for separation of xylenes.

5. Conclusions In this work, we determined the activity coefficients at infinite dilution of organic solutes in ILs [MMIM][DMP] and [EMIM][DMP]

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jct.2015.07.046. References [1] R.D. Rogers, K.R. Seddon, Science 302 (2003) 792–793. [2] C. Yao, J.L. Anderson, J. Chromatogr. A 1216 (2009) 1658–1712. [3] J. Ranke, S. Stolte, R. Stormann, J. Arning, B. Jastorff, Chem. Rev. 107 (2007) 2183–2206. [4] N.V. Plechkova, K.R. Seddon, Chem. Soc. Rev. 37 (2008) 123–150. [5] M.-L. Ge, J.-M. Xiong, L.-S. Wang, Chin. Sci. Bull. 54 (2009) 2225–2229. [6] A. Marciniak, Fluid Phase Equilib. 294 (2010) 213–233. [7] W.E. Acree Jr., G.A. Baker, A.-L. Revelli, J.-C. Moise, F. Mutelet, J. Chem. Eng. Data 57 (2012) 3510–3518. [8] M. Sobota, V. Dohnal, P. Vrbka, J. Phys. Chem. B 113 (2009) 4323–4332. [9] C.F. Poole, J. Chromatogr. A 1037 (2004) 49–82. [10] A. Heintz, T.V. Vasiltsova, J. Safarov, E. Bich, S.P. Verevkin, J. Chem. Eng. Data 51 (2006) 648–655. [11] T.M. Letcher, A. Marciniak, M. Marciniak, U. Doman´ska, J. Chem. Thermodyn. 37 (2005) 1327–1331. [12] U. Doman´ska, P. Papis, J. Szydłowski, J. Chem. Thermodyn. 77 (2014) 63–70. [13] M. Królikowski, M. Królikowska, J. Chem. Thermodyn. 68 (2014) 138–144. [14] F. Mutelet, J.-N. Jaubert, J. Chromatogr. A 1102 (2006) 256–267.

M.-L. Ge et al. / J. Chem. Thermodynamics 91 (2015) 279–285 [15] G. Foco, D. Bermejo, A.J. Kotlewska, F. Rantwijk, C.J. Peters, S.B. Bottini, J. Chem. Eng. Data 56 (2011) 517–520. [16] M.-L. Ge, L.-S. Wang, M.-Y. Li, J.-S. Wu, J. Chem. Eng. Data 52 (2007) 2257– 2260. [17] M.-L. Ge, L.-S. Wang, Q. Zhou, J. Chem. Eng. Data 53 (2008) 1970–1974. [18] M.-L. Ge, J.-B. Chen, J. Chem. Eng. Data 56 (2011) 3183–3187. [19] M.-L. Ge, X.-J. Song, G.-M. Li, Y.-H. Li, F.-Z. Liu, H.-L. Ma, J. Chem. Eng. Data 57 (2012) 2109–2113. [20] M.-L. Ge, X.-M. Deng, L.-H. Zhang, J.-Y. Chen, J.-M. Xiong, W.-H. Li, J. Chem. Thermodyn. 77 (2014) 7–13. [21] A.J.B. Cruickshank, M.L. Windsor, C.L. Young, Proc. R. Soc. London A295 (1966) 259–270. [22] D.H. Everett, Trans. Faraday Soc. 61 (1965) 1637–1645. [23] T. Boublik, V. Fried, E. Hala, The Vapor Pressure of Pure Substances, Physical Science Data 17, Elsevier, Amsterdam, Netherlands, 1984. [24] H.-Z. Lu, The Handbook of Petrolic Chemical Engineering Data, Chemical Industrial Press, Beijing, 1992. [25] B.E. Poling, J.M. Prausnitz, J.P. O’ Connell, The Properties of Gases and Liquids, McGraw-Hill Book Co, New York, 2000. [26] J.M. Prausnitz, R.N. Lichtenthaler, E.G. Azevedo, Molecular Thermodynamics of Fluid Phase Equilibria, Prentice Hall, New York, 1986.

285

[27] D.W. Grant, Gas–Liquid Chromatography, Van Nostrand Reinhold, London, 1971. [28] R. Kato, J. Gmehling, Fluid Phase Equilib. 226 (2004) 37–44. [29] D. Tiegs, J. Gmehling, A. Medina, M. Soares, J. Bastos, P. Alessi, I. Kikic, DECHEMA Chemistry Data Series IX, Part 1, DECHEMA, Frankfurt/Main, 1986. [30] M. Krummen, P. Wasserscheid, J. Gmehling, J. Chem. Eng. Data 47 (2002) 1411–1417. [31] A. Heintz, D.V. Kulikov, S.P. Verevkin, J. Chem. Eng. Data 47 (2002) 894–899. [32] E. Olivier, T.M. Letcher, P. Naidoo, D. Ramjugernath, J. Chem. Thermodyn. 42 (2010) 78–83. [33] U. Domanska, A. Marciniak, J. Chem. Thermodyn. 40 (2008) 860–866. [34] P.-F. Yan, M. Yang, X.-M. Liu, Q.-S. Liu, Z.-C. Tan, U. Welz-Biermann, J. Chem. Eng. Data 55 (2010) 2444–2450. [35] F. Mutelet, A.L. Revelli, J.N. Jaubert, L.M. Sprunger, W.E. Acree, G.A. Baker, J. Chem. Eng. Data 55 (2010) 234–242. [36] I. Bahadur, B. Govender, K. Osman, M.D. Williams-Wynn, W. Nelson, P. Naidoo, D. Ramjugernath, J. Chem. Thermodyn. 70 (2014) 245–252. [37] P.-F. Yan, M. Yang, X.-M. Liu, Ch. Wang, Z.-Ch. Tan, U. Welz-Biermann, J. Chem. Thermodyn. 42 (2010) 817–822.

JCT 15-243