Quaternary (liquid + liquid) equilibria for systems of imidazolium based ionic liquid + thiophene + pyridine + cyclohexane at 298.15 K: Experiments and quantum chemical predictions

doi:10.1016/j.jct.2013.02.020

The Journal of Chemical Thermodynamics

Volume 62, July 2013, Pages 142-150

https://doi.org/10.1016/j.jct.2013.02.020 Get rights and content

Abstract

1-Ethyl 3-methylimidazolium acetate [EMIM][OAc], 1-ethyl 3-methylimidazolium ethylsulfate [EMIM][EtSO₄] and 1-ethyl 3-methylimidazolium methylsulfonate [EMIM][MeSO₃] are found to be effective for the selective removal of a five member ring sulfur compound, viz. thiophene, and a six member ring nitrogen compound, viz. pyridine from a hydrocarbon, viz. cyclohexane at room temperature and atmospheric pressure. The liquid liquid equilibrium (LLE) data for the quaternary mixtures of ionic liquid (1) + thiophene (2) + pyridine (3) + cyclohexane (4) are experimentally determined. The experimental tie line data are successfully correlated with the Non Random Two Liquid (NRTL) and UNIversal QUAasi-Chemical (UNIQUAC) models, which provide a good correlation of the experimental data with root mean square deviation (RMSD) values less than unity for all the studied systems. The results suggest that the structure and size of the anion greatly affects the extractive performance of ionic liquids. The reliability of experimental data is ascertained by applying the quantum chemical based COnductor like Screening Model-Segment Activity Coefficient (COSMO-SAC) model. The goodness of the fit is determined by calculating the RMSD values. The RMSD values obtained for [EMIM][OAc], [EMIM][EtSO₄] and [EMIM][MeSO₃] are 10.4%, 8.1% and 12.2%, respectively.

Graphical abstract

Highlights

► Quaternary (liquid + liquid) equilibria are measured for ionic liquids, cyclohexane, pyridine and thiophene. ► Extraction of thiophene and pyridine from cyclohexane using ionic liquids is investigated. ► Three different ionic liquids are investigated. ► The experimental data can be successfully correlated using NRTL and UNIQUAC models. ► The experimental data are in moderate agreement with a priori predictions made using the COSMO-SAC model.

Introduction

In the past decade, much attention has been paid to the production and use of more environmentally friendly transportation fuels with lower concentration of aromatic nitrogen and sulfur [1]. Hydrodenitrification (HDN) and hydrodesulfurization (HDS) are the conventional methods to remove these compounds in industrial processes. The main disadvantages of HDN and HDS are high operating temperatures (>300 °C), high pressures (>4 MPa), high energy costs, and large reactor size. These limitations can lead to an increase in the associated costs such as capital and operating costs. Moreover it is very difficult to remove both kinds of compounds in a single step at ambient conditions because the aromatic nitrogen compounds poison the active catalyst leading to coke formation. Nitrogen and sulfur in diesel oil also remain a major source of NO_x and SO_x which contribute to air pollution, global warming, greenhouse effect and acid rain [2], [3]. Thus the threshold limits for nitrogen and sulfur in diesel oil are expected to be regulated on a global level to less than 1 ppm of nitrogen and 10 ppm of sulfur [4], [5], [6], [7].

Recently ionic liquids have been used as extracting solvents for the removal of nitrogen and sulfur from liquid fuel at ambient temperature and pressure. Usually, ionic liquids are salts which are in liquid state at a temperature below 100 °C. Ionic liquids have a negligible vapor pressure. Hence, there is no emission of volatile organic compounds to the environment. They are nonvolatile and exhibit a high ionic conductivity. Ionic liquids are also highly solvating and non-coordinating mediums [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30]. Moreover, ionic liquids consist of cation and anion combinations. A large number of cations and anions that can form ionic liquids are available. Hence, the number of ionic liquids that can be synthesized from their combinations is large. As a result, their physiochemical properties (i.e. melting point, viscosity, density, solubility etc.) can be tuned by varying the cations and anions.

Generally aromatic nitrogen and sulfur based compounds have low activity coefficient at infinite dilution in ionic liquids, whereas diesel compounds shows high activity coefficients [3], [31], [32], [33] in ionic liquids. Smaller the values of the activity coefficient, stronger are the interactions between the solute and ionic liquid. Since, there is a large difference in the activity coefficients of the nitrogen and sulfur compound present in diesel in the ionic liquid, and the activity coefficient of the hydrocarbons in the ionic liquid, a high selectivity is obtained. In addition to the high selectivity (S), high capacity (C) is also essential. This can be achieved by selecting an ionic liquid with a large solubility of the nitrogen and sulfur compounds. The selectivity and capacity for different combinations of cation and anion can be measured. Alternatively, it can be calculated using predictive models such as the COnductor like Screening MOdel for Real solvent (COSMO-RS) [34] and the COnductor like Screening Model-Segment Activity Coefficient (COSMO-SAC) [35]. Previous work [36], [37] shows the feasibility and ability of ionic liquids to remove various aromatic sulfur compounds viz. thiophene, benzothiophene, and dibenzothiophene, and nitrogen compounds (five member ring: pyrrole, indole, indoline, carbazole, benzocarbazole, and dibenzocarbazole, six member ring: pyridine, quinoline and benzoquinoline) from diesel oil. Simultaneous extraction of aromatic nitrogen and sulfur from diesel oil compounds with ionic liquids is expected to require less number of stages and less energy consumption than extraction with conventional solvents because ionic liquids have a negligible vapor pressure and hence can be recycled. Shah and Yadav [38] had provided a criterion for screening of ionic liquids for aromatic/aliphatic separations using the Cosmo-SAC model and the “Cosmo” files for eight cations and sixteen anions corresponding to 128 potential ionic liquids have been provided for the use of the general scientific community to predict any thermodynamic equilibria involving ionic liquids without the use of any molecular modeling software. They also used the model for prediction of liquid–liquid equilibria for biofuel applications [39] and prediction of sorption in polymeric membranes [40].

The aim of this work is to investigate the feasibility of the liquid–liquid extraction process for the separation of thiophene and pyridine from diesel compounds at ambient temperature using 1-ethyl-3-methylimidazolium acetate [EMIM][OAc], 1-ethyl-3-methylimidazolium ethylsulfate [EMIM][EtSO₄] and 1-ethyl-3-methylimidazolium methylsulfonate [EMIM][MeSO₃] ionic liquids. The cation and anions have been selected on the basis of the selectivities predicted by the COSMO-RS model [3], [32], [33]. The ability of the selected ionic liquids to separate the nitrogen and sulfur compounds from diesel has been evaluated by measurement of quaternary (liquid + liquid) equilibrium (LLE) data and the results have been shown in terms of a ternary diagram with the ionic liquid, (thiophene + pyridine) and cyclohexane as the three components in the ternary diagram. These data can be helpful in understanding the thermodynamic limit in separation, and the design of the liquid–liquid extraction column. The (liquid + liquid) equilibria for the quaternary mixtures of [EMIM][OAc] + thiophene + pyridine + cyclohexane, [EMIM][EtSO₄] + thiophene + pyridine + cyclohexane and [EMIM][MeSO₃] + thiophene + pyridine + cyclohexane have been determined at 298.15 K. The selectivity and the distribution coefficient have also been determined from the experimental data. Further, the obtained experimental data are benchmarked with the predictions of the COSMO-SAC model and correlated with the NRTL and UNIQUAC models.

Section snippets

Experimental section

The ionic liquids: 1-ethyl-3-methylimidazolium acetate [EMIM][OAc](C₈H₁₄N₂O₂) of purity >99% and 1-ethyl-3-methylimidazolium ethyl sulfate [EMIM][EtSO₄] (C₈H₁₆N₂O₄S) of purity >99% were supplied by Sigma Aldrich, Germany. 1-ethyl-3-methylimidazolium methanesulfonate [EMIM][MeSO₃] (C₈H₁₆N₂O₃S) ionic liquid was purchased from Fluka with a purity of >99%. Pyridine (C₅H₅N) and thiophene (C₄H₄S) were received from Sigma Aldrich with mass fractions greater than 98%. Cyclohexane (C₆H₁₂) was obtained

NRTL and UNIQUAC model

The non-ideal liquid phase activity coefficient of a component i in a liquid mixture can be correlated using the NRTL and UNIQUAC models. The mathematical expressions for these two models have been presented elsewhere [41], [42], [43], [44], [45] and are therefore not discussed here. From the experimental LLE data, the binary interaction parameters have been calculated by minimizing the sum square of errors. The following objective function (equation (1)) has been maximized: $Maximize : F_{(\begin{matrix} with \end{matrix})}$

Results and discussion

The composition of experimental (liquid + liquid) equilibrium data (tie lines) for [EMIM][OAc] (1) + Thiophene (2) + Pyridine (3) + Cyclohexane (4), [EMIM][ EtSO₄] (1) + Thiophene (2) + Pyridine (3) + Cyclohexane (4) and [EMIM][MeSO₃] (1) + Thiophene (2) + Pyridine (3) + Cyclohexane (4) at 298.15 K are shown in TABLE 3, TABLE 4, TABLE 5. Values of selectivity and distribution coefficient are also shown in these tables. These parameters are defined as follows: $Selectivity (S) = \frac{(x_{thiophene + pyridine}^{E}) \cdot (x_{cyclohexane}^{R})}{(x_{}}$

Conclusion

In this work, the LLE for the quaternary system of [EMIM][OAc] (1) + Thiophene (2) + Pyridine (3) + Cyclohexane (4), [EMIM][EtSO₄] (1) + Thiophene (2) + Pyridine (3) + Cyclohexane (4) and [EMIM][MeSO₃] (1) + Thiophene (2) + Pyridine (3) + Cyclohexane (4) have been experimentally determined at 298.15 K and the corresponding selectivities and distribution coefficients have been calculated. The selectivity values are higher than unity for all of the studied systems, whereas the distribution coefficient values are

Acknowledgments

The authors are grateful to the Department of Science and Technology (DST), Government of India for the financial support through project SR/FTP/08-08 under the Fast Track Scheme to T. Banerjee. M.R. Shah acknowledges support from University Grants Commission, as D.S. Kothari post doctoral fellow. G.D. Yadav received support from the R.T. Mody Distinguished Professor Endowment and J.C. Bose National Fellowship from DST.

References (53)

C. Song
Catal. Today
(2003)
C. Song et al.
Appl. Catal. B: Environ.
(2003)
H. Yang et al.
Catal. Today
(2005)
E. Ito et al.
Catal. Today
(2006)
A. Marciniak et al.
Fluid Phase Equilib.
(2012)
A. Marciniak et al.
J. Chem. Thermodyn.
(2012)
B. Rodriguez-Cabo et al.
Fluid Phase Equilib.
(2012)
H. Rodriguez et al.
Fluid Phase Equilib.
(2010)
M. Francisco et al.
Fluid Phase Equilib.
(2010)
A. Arce et al.
J. Chem. Thermodyn.
(2010)

R. Shafia et al.

Catal. Today

(2000)

Cited by (48)

Ionic liquid derived novel deep eutectic solvents as low viscous electrolytes for energy storage
2022, Journal of Molecular Liquids
This work investigates a newly emerging class of solvents to be used as alternatives to or complementarily with ionic liquid (IL) electrolytes. Three novel deep eutectic solvents (DESs) comprised of the hydrogen bond donors (HBDs), namely N-Methylacetamide (NMAc) and formamide, were formulated with three variants of the methyl imidazolium-based ionic liquids (ILs) as hydrogen bond acceptors (HBAs). ILs consisting of 1-butyl-3-methyl imidazolium ([BMIM]) cation with anions namely methanesulphonate ([MeSO₃]), bis(trifluoromethylsulfonyl)imide ([Tf₂N]), and hexafluorophosphate ([PF₆]) were used as hydrogen bond acceptors. Initially, the eutectic points of the DESs were estimated using a quantum chemical-based thermodynamic model, namely Conductor like Screening Model-Segment Activity Coefficient (COSMO-SAC). The calculated melting points of the DESs were determined to be markedly lower than those of both the constituent components. Further, physical and chemical characterization was done on the synthesized DESs. The viscosities (<20 cp) were found to be significantly lower than those of conventionally used pure room temperature ILs, and ionic conductivities (>5 mS/cm) were found to be appreciably higher than those of the corresponding constituent ILs. Thermogravimetric analysis revealed negligible mass loss up to 80–100 °C. Linear scan voltammetry (LSV) was employed to determine the electrochemical stable potential window (ESPW) of the solvents with a glassy carbon (GC) electrode. LSV provided electrochemical stabilities in the range of 3.8–4 V for all three studied deep eutectic solvents, comparable to pure ILs and far higher than the potential windows of mixtures of ILs and organic solvents, a blend commonly used as electrolytes.
Phase behavior of water-menthol based deep eutectic solvent-dodecane system
2021, Chemical Thermodynamics and Thermal Analysis
In the current work, two menthol-based hydrophobic deep eutectic solvents (DESs) have been synthesized to extract dodecane from water-dodecane emulsion at 298 K and atmospheric pressure. The equilibrium mixture produces two biphasic liquid-liquid equilibrium systems and one triphasic liquid-liquid-liquid equilibrium system. ¹H NMR has been used to calculate the molar composition of each phase. For both the lauric and decanoic acid-based DES, the extract phase of biphasic region-1 is rich in DES; whereas the raffinate phase is predominantly water. However, for the biphasic region-2, the extract phase is the dodecane rich phase and raffinate phase is the mixture of DES and dodecane having a dominant composition of DES. In case of triphasic region, i.e. top phase consists of dodecane and the bottom phase is only water. The middle phase becomes an emulsion where DES is the dominant component. The selectivity analysis in the triphasic region proves that both the DESs have high selectivity to extract dodecane from the water-dodecane microemulsion. The experimental mole fractions are correlated with genetic algorithm (GA) promoted NRTL and UNIQUAC models to predict the phase behavior. An average overall root means square deviation (RMSD, %) of 0.174% is observed for the biphasic region and 0.048% for the triphasic employing GA-NRTL model. GA-UNIQUAC predicts an average overall RMSD of 0.340% for biphasic regions and 0.129% for the triphasic region.
Value-added utilization of high-temperature coal tar: A review
2021, Fuel
Citation Excerpt :
Generally speaking, ILs are composed of organic cations and inorganic anions, and target compounds (TCs) with corresponding characteristics can be separated by designing ILs with different cations and anions [61]. At present, ILs are commonly used to separate VACs, such as alkanes and alkenes [62–64], arenes [65–69], NCAs [19,20,70–74], OCAs [75–81], and SCAs [82–89] (Tables 6-10). In addition, under the optimal extraction time, temperature, and IL: oil (w: w), imidazole ILs can separate more than 90% of indole from WO.
High-temperature coal tar (HTCT) is an important raw material for obtaining value-added aromatics. However, effective utilization of HTCT is not optimistic because of backward processing technology and higher energy consumption, resulting in fewer product types and lower purity. Accordingly, value-added utilization (VAU) of HTCT should be paid urgent attention. In order to realize VAU of HTCT, deep processing methods were systematically discussed. This paper mainly discusses the source, composition, research status, separation method, and upgrading of HTCT along with the utilization of coal tar pitch. Subsequently, the comprehensive processing route for VAU of HTCT was proposed. HTCT research ideas are put forward in combination with future research directions. We hope that this review will add to the ongoing discussions on the VAU of HTCT.
Tribological performance of silicon nitride and carbon black Ionanofluids based on 1-ethyl-3-methylimidazolium methanesulfonate
2020, Journal of Molecular Liquids
Citation Excerpt :
Fig. 10a shows that experimental density values for [EMIM][MS] (reported in Table 4) are in agreement with the corresponding literature data [37,40,56–68], with absolute average deviations within 0.4%. It should be noted that the content of water is slightly different among the reported literature, but always less than 1% [37,40,56–68]. The best agreements were found with the data from Domanska et al. [60], Rabari et al. [64], Krannich et al. [40], Marium et al. [66], and Chereches et al. [68], with absolute average deviations of 0.03%, 0.10%, 0.09%, 0.02% and 0.05%, respectively.
Development of nano-lubricants by the dispersion of nano-particles in current lubricants has contributed to improving energy efficiency reducing wear and friction. During the last two decades, ionic liquids have evolved as novel lubricants or lubricant additives, especially for high vacuum and high temperature. Nevertheless, the number of experimental studies regarding the tribological properties of Ionanofluids, defined as dispersions of nano-particles in ionic liquids, is limited. 1-Ethyl-3-methylimidazolium methanesulfonate, [EMIM][MS], is a promising candidate for lubrication applications due to its wide liquid range, high thermal conductivity, low friction coefficients and low compressibility in comparison with commercial mineral and synthetic hydrocarbon based lubricants. In this work, Ionanofluid lubricants based on dispersions of nano-additives (silicon nitride and carbon black) at mass concentration between 0.10 and 1.0 wt% in [EMIM][MS] were designed. The nearly spherical morphology of the nano-additives, silicon nitride and carbon black, was described by using scanning electron microscopy. The stabilities of the resulting nano-particle dispersions in the ionic liquid were analyzed by dynamic light scattering measurements of size for one month. Tribological characterization was performed by a rotational rheometer coupled with a tribology cell with ball-on-three-pins configuration (100Cr6 steel) in sliding conditions at 298.15 and 353.15 K. Afterwards, the wear track morphology of the worn pins was analyzed by a 3D optical profiler and Raman spectroscopy. The dispersions at the optimal nano-additive concentrations for lubrication reached friction coefficient decreases of up to 16% and wear tracks with a volume 28 times lower. Additionally, the density and dynamic viscosity of [EMIM][MS] and the optimal Ionanofluids for lubrication applications measured with a rotational Stabinger visco-densimeter in the 278.15 to 373.15 K temperature range showed small increases with almost no dependence on temperature.
Phase equilibria and energetics of binary mixtures of water with highly hydrophilic [EMIM]-based ionic liquids: Methanesulfonate, methylsulfate, and dimethylphosphate
2020, Fluid Phase Equilibria
Citation Excerpt :
The density data of Anantharaj and Banerjee [26], Shah et al. [27], and Singh et al. [28] for [EMIM][MeSO3] and those of Shekaari and Armanfar [29] and Wang et al. [30] for [EMIM][MeSO4] were found to deviate grossly from the measurements by numerous other authors, and consequently they are considered incorrect.
As a follow-up to our systematic studies on the aqueous solutions of 1-ethyl-3-methylimidazolium-based ([EMIM]) ILs, this work deals with phase equilibria and energetics of other three aqueous [EMIM] ILs whose anion is methanesulfonate ([MeSO₃]), methylsulfate ([MeSO₄]), and dimethylphosphate ([DMP]). We present here new and extensive experimental VLE, SLE and excess enthalpy data, our aim being to provide truly reliable information that would substantially upgrade that existing in the literature. After investigations of the phase behavior and properties of the pure ILs, we performed for the IL solutions systematic measurements of the water activity in the span from 0.05 to 1 and excess enthalpy in the highly dilute IL region (x_IL < 0.003) as functions of composition and temperature in the range of 30 K. The freezing temperatures of solutions were further measured in the water-rich region and, in case of [EMIM][DMP], also in the IL-rich region. The VLE and excess enthalpy data were then simultaneously correlated by a modified NRTL equation whose ability to reliably capture VLE, SLE and the energetic properties of the systems was subsequently demonstrated in detail. This also comprised due comparison and analysis of the available literature information which lead to the identification of some incorrect literature data. The calculated excess Gibbs energy and excess enthalpy of the systems revealed for all three systems large negative deviations from ideality that increase in the sequence [EMIM][MeSO₃] < [EMIM][MeSO₃] < [EMIM][DMP], the values exhibited by the (water + [EMIM][DMP]) belonging to the largest negative deviations from the ideal mixture behavior ever observed. In contrast to other aqueous [EMIM] ILs, the aqueous [EMIM][DMP] exhibits an enormously negative excess entropy which at x_w > 0.5 corresponds to an entropy loss upon mixing and indicates thus an unprecedented structuring effect of the system. The excess Gibbs energy and enthalpy values at extrema for aqueous solutions of five hydrophilic [EMIM] ILs studied here and in our previous work were further found to closely correlate with the solvatochromic H-bond basicity parameters of the ILs anions, which proved that the water – IL affinity is governed by the H-bonding of the water to the anions of the ILs. Finally, simple MD simulations gave insight into the way and the number of H-bonds between the water molecules and the anion of the IL in the water-rich region.
Liquid-liquid equilibria data and thermodynamic modeling for quaternary system of methanol, dimethyl carbonate, toluene and water at 298 to 318 K
2020, Journal of Molecular Liquids
Liquid-liquid equilibria (LLE) data for a quaternary mixture of methanol, Dimethyl carbonate, toluene and water were measured at 298, 308, 318 K and atmospheric pressure. The distribution of methanol and dimethyl carbonate in the organic phase and aqueous phase were searched. Experimental tie-line data reliability of the quaternary mixture was validated using the Othmer-Tobias and Bachman correlation method. This experimental data were correlated with the nonrandom tow-liquid (NRTL) and the universal quasichemical activity coefficient (UNIQUAC) model. The interactive binary parameters of these two models have been reported. Both models successfully correlated experimental data.

View all citing articles on Scopus

¹: Deemed University; Elite Status & Centre of Excellence, Govt. of Maharashtra; TEQIP PHASE II Funded.

View full text