This study demonstrates that in addition to association, polarity, and other interactions, phase equilibria of the ionic liquid (IL) systems are also influenced by the global phase diagram factors, namely, the differences between values of the pure compound critical points. As these differences increase, increasing thus the system asymmetry, the extent of phase splits also increases, and the molar solubilities correspondingly decrease. It is shown that the pressure dependence of the IL densities indicate the higher experimentally inaccessible T c and P c values, making their systems more asymmetric, which reduces the solvent capacities and vice versa. In addition to this, phase equilibria are also influenced by the critical constants of the solvents. Their higher T c and P c values typically increase the symmetry, and therefore, the solubilities. The CP-PC-SAFT equation of state rigorously obeys the T c and P c of solvents and accurately represents the densities of ILs under a wide range of conditions. Such features endow this model with a remarkable predictive potential even while neglecting the association and polar interactions. In most of the cases, CP-PC-SAFT with k 12 = 0 correctly estimates the differences between solubilities in the experimentally investigated systems comprising fluoro- and chloromethanes, ethanes, ethenes, propanes, and propenes with [C2mim]­[Ntf2], [C4mim]­[Ntf2], [C6mim]­[Ntf2], [C8mim]­[Ntf2], [C4mim]­[PF6], [C4mim]­[BF4], and [C8mim]­[BF4]. The quantitative accuracy of this model is usually reasonably good as well. Although SAFT-VR-Mie obeys the literature T c values of solutes, it overestimates their P c. In addition to this, it yields higher imaginary critical constants of the considered ILs than CP-PC-SAFT. These factors may explain its tendency to underestimate the solubility data with k 12 = 0. Nevertheless, in most of the cases, SAFT-VR-Mie correctly describes the solubility tendencies in the considered systems.

This study compares the estimations of two predictive approaches, namely, CP-PC-SAFT and CS-SAFT-VR-Mie for vapor pressures, saturated and high-pressure densities, sound velocities, isobaric and isochoric heat capacities of 17 representative substituted fluoro- and chloromethanes, ethanes, ethenes, propanes, and propenes, whose data are available under a wide range of conditions. Predictions of vapor–liquid equilibria (VLE) in their binary systems comprising inter alia carbon dioxide, hydrocarbons, and light gas systems with k 12 = 0 are considered as well. These models do not incorporate associative and polar interactions, which allows parametrization based on the particularly limited amount of literature data. It is demonstrated that although the overall accuracies of the considered approaches are reasonably good, each of them has their advantages and disadvantages. For example, SAFT-VR-Mie typically predicts more accurately some VLE in the symmetric systems, while CP-PC-SAFT is advantageous in estimating phase behavior of the asymmetric ones. Both models can be applied for preliminary estimation of unavailable halocarbon data for which each of them exhibits a relative superiority. In addition to this, it is demonstrated that SAFT-VR-Mie attached by square gradient theory with predictive correlation for influence parameters accurately estimates surface tensions of the considered compounds.

This study compares accuracies of PC-SAFT and its critical point-based revision (CP-PC-SAFT) with k 12 = 0 along with the predictive 1978 Peng−Robinson EoS (PPR78) in estimating the available high-pressure phase equilibria data in binary systems of methane, ethylene, ethane, propane, and n-hexane with heavy branched, alkylbenzoic, bicyclic aliphatic and aromatic substances represented by squalane, n-dodecylbenzene, cis-decalin, tetralin, and naphthalene. Predictions of the pure compound properties of the considered heavy substances are evaluated as well. It is found that some data were predicted more accurately by CP-PC-SAFT and othersby Peng− Robinson EoS and PPR78. At the same time, the overall performance of PC-SAFT in predicting these data is found inferior in comparison with two other considered approaches.

This study compares performances of the Critical Point-based revision of Perturbed-Chain SAFT (CP-PC-SAFT) and the SAFT of Variable Range and Mie Potential (SAFT-VR-Mie) in predicting the available data on VLE, LLVE, critical loci and saturated phase densities of systems comprising CO, O2, CH4, H2S, SO2, propane, the refrigerants R22, R23, R114, R124, R125, R125, R134a, and R1234ze(E) and ionic liquids (ILs) with 1-alkyl-3-methylimidazolium ([Cnmim]+) cations and bis(trifluoromethanesulfonyl)imide ([NTf2]−), tetrafluoroborate ([BF4]−) and hexafluorophosphate ([PF6]−) anions. Both models were implemented in the entirely predictive manner with k12 = 0. The fundamental Global Phase Diagram considerations of the IL systems are discussed. It is demonstrated that despite a number of quantitative inaccuracies, both models are capable of reproducing the regularities characteristic for the considered systems, which makes them suitable for preliminary estimation of selectivity of the ILs in separating various gases.

No abstract