Search

Abstract The RKPR EoS was developed in the past decade as a three-parameter model, with the intention to overcome some of the limitations of classic two-parameter cubic equations of state. More recently, parameter correlations were presented for alkanes and their mixtures, showing excellent capacity to predict the fluid phase behavior of binary mixtures up to the more asymmetric combinations and in wide ranges of temperature and pressure. In this work, different types of phase equilibrium data for multicomponent mixtures are considered, in order to see whether the good performance previously demonstrated for binary systems is also transferable to multicomponent mixtures like the ones typically encountered in synthetic fluids designed in the lab as model reservoir fluids. In the case of ternary systems with good availability of constant temperature and pressure data sets, the type and behavior of these Gibbs triangle diagrams are analyzed with reference to the general or global behavior of the system, based on pure compound saturation and binary critical lines, something we had not found in the literature preceding this work. The results and comparisons presented in this article allow concluding that RKPR2015 can be a useful tool with very good predictive capacity to describe the fluid phase equilibria of hydrocarbon mixtures. Advantages over the Peng-Robinson equation appear more clearly for the more asymmetric fluids, including synthetic gas condensates. [ABSTRACT FROM AUTHOR]

The solid–liquid equilibrium phase diagrams of binary and ternary mixtures formed by n-decane and the fatty acid ethyl esters ethyl laurate and ethyl palmitate were studied through differential scanning calorimetry (DSC). Phase change properties of polymorphs were obtained by resolving the thermograms into the corresponding overlapping peaks through a fitting analysis with a Gaussian function. The binary systems show immiscibility in solid phase. The system formed by ethyl laurate and n-decane presents two peritectic transformations, which were also observed in the ternary phase diagram. Equilibrium data were thermodynamically modeled using different models for liquid phase non-ideality and considering different polymorphs in solid phase. The results of modeling are in good agreement with the eutectic behavior of the system formed by ethyl palmitate and n-decane. However, higher discrepancies were observed if no peritectic behavior for the binary system formed by ethyl laurate and n-decane is considered. The results obtained in this work constitute another step forward to enhance the understanding and description of the complex behavior of biodiesel/diesel mixtures at low temperatures. [ABSTRACT FROM AUTHOR]

The solid–liquid equilibrium diagrams of binary and ternary mixtures formed by dodecylcyclohexane and the fatty acid esters ethyl laurate and ethyl palmitate was determined herein. The liquidus and solidus lines were obtained through differential scanning calorimetry. The thermograms of pure compounds were resolved into the corresponding overlapping peak components by a fitting analysis, allowing a better estimation of phase-change properties of polymorphs. All binary systems show immiscibility in the solid phase, although those containing dodecylcyclohexane present small regions of partial miscibility. The system formed by ethyl palmitate and dodecylcyclohexane presents a peritectic transition, which was also observed in the ternary system. The thermodynamic modeling was conducted using different models for liquid-phase non-ideality and considering different polymorphs in solid phase. The modeling shows that a probable inversion of polymorphs relative stability occurs for the binary system formed by ethyl laurate and dodecylcyclohexane. The results from this work may contribute to enhancing the understanding and description of the complex behavior of biodiesel/diesel mixtures at low temperatures. [ABSTRACT FROM AUTHOR]

Multiphase, multicomponent equilibrium systems are encountered commonly in the chemical industry. The non-random two-liquid (NRTL) and universal quasi-chemical (UNIQUAC) models are applied widely to correlate activity coefficients in phase equilibria calculations. We demonstrated previously the predictive capability of a quantitative structure–property relationship (QSPR)-generalized NRTL model (NRTL-QSPR) for vapor–liquid equilibrium (VLE) property predictions of diverse binary systems. In this study, we further evaluate the prediction quality of the NRTL and UNIQUAC models as they apply to multicomponent systems. Specifically, we evaluate both the representation and prediction of vapor–liquid equilibrium properties of multicomponent systems using regressed and QSPR-generalized parameters of the constituent binaries. To complete this evaluation, a database comprised of 57 ternary VLE systems and their 75 constituent binary systems was assembled. In addition, 29 ternary systems were employed to examine the efficacy of the models when at least one constituent binary is missing. For ternary systems where all three constituent binary systems were available, the NRTL-QSPR model provided overall percent absolute average deviations (%AADs) of 3.7, 0.3, 9.0, and 8.7 for pressure ( P ), temperature ( T ), mole fraction ( y 1 ), and K-values , respectively. In comparison, the generalized UNIQUAC-QSPR model exhibited errors of 3.5, 0.3, 8.2, and 7.9, respectively. Thus, for the systems considered, both models exhibit comparable accuracy. Further, for these systems the group-contribution method Modified UNIFAC (mUNIFAC) produced slightly better results. For systems where at least one constituent binary is missing in our database, errors for the NRTL-QSPR are 6.3, 0.5, 11.2, and 11.4 %AADs for P , T , y 1 , and K-values , respectively. The comparable errors for the UNIQUAC-QSPR for this case are 8.4, 0.7, 12.9, and 12.5, respectively. For the same set of ternary systems, the mUNIFAC model produced %AADs of 20.7, 2.1, 25.7, and 26.7 for P , T , y 1 , and K-values , respectively. As such, the mUNIFAC model results in three times higher errors than our generalized models. Although additional evaluation using a comprehensive database is required, the results presented in this study demonstrate the efficacy of the QSPR binary parameter generalizations, as applied to the NRTL and UNIQAC models, for VLE multicomponent property predictions. [ABSTRACT FROM AUTHOR]

Ternary miscibility data, usually reported in the Roozeboom triangular diagram, have been analyzed in Cartesian coordinates defined by X = ( x i − x j )/3 and Y = x k where x i , x j , and x k are the mole fractions of the two solvents and of the solute, respectively. This has allowed for an easy analytical representation of all the elements which are characteristic of the system: binodal curve, tie lines, plait point. Three novel methods for determining the plait point composition have been proposed based only on its fundamental properties. The results have been presented by applying the methods to some systems described in the literature. Differences between the values obtained with the proposed methods and those obtainable with the NRTL and UNIQUAC models have been found and are discussed. [ABSTRACT FROM AUTHOR]

Combined chemical and vapor–liquid equilibrium (ChVLE) data for the ternary system ethylene + water + ethanol are required for the conceptual design of a reactive separation process to obtain ethanol by the hydration of ethylene. Due to the absence of experimental data for the combined ChVLE of the reacting system, molecular simulation looks appealing to predict such data. In this work, the reaction-ensemble Monte Carlo (RxMC) method was used to calculate the chemical equilibrium of the ternary system in the vapor phase, and the reactive Gibbs-ensemble Monte Carlo (RxGEMC) method was used to calculate its combined ChVLE. A set of previously validated Lennard-Jones plus point-charge potential models were employed for ethylene, water, and ethanol. The RxMC predictions for the vapor-phase chemical equilibrium composition of the ternary system and the equilibrium conversion of ethylene to ethanol, at 200 °C and pressures of 30, 40, 50, and 60 atm, were found to be in good agreement with predictions made by use of a previously proposed thermodynamic model that combines the Peng–Robinson–Stryjek–Vera equation of state, the Wong–Sandler mixing rules, and the UNIQUAC activity coefficient model. The RxGEMC simulations were used to predict the reactive phase diagram (two-dimensional graph of pressure versus transformed liquid and vapor-phase ethylene mole fractions) at 200 °C. In contrast to the thermodynamic model, molecular simulation predicts a wider reactive phase diagram (due to a reactive dew-point line much richer in ethylene). However, these two independent approaches were found to be in very good agreement with regard to the predicted bubble-point line of the reactive phase diagram and the approximate location of the reactive critical point. [ABSTRACT FROM AUTHOR]

In the present work, the ionic liquids: N -hexylisoquinolinium thiocyanate, [HiQuin][SCN], N -octylisoquinolinium thiocyanate, [OiQuin][SCN], N -hexylisoquinolinium bis{trifuoro-methylsulfonyl}imide, [HiQuin][NTf 2 ] and N -octylquinolinium bis{trifuoromethyl-sulfonyl}imide, [OQuin][NTf 2 ] have been investigated as a solvent for the separation of aromatic nitrogen compounds from heptane. Ternary (liquid + liquid) phase equilibria have been determined for the following systems: {[HiQuin][SCN], or [[OiQuin][SCN], or [HiQuin][NTf 2 ], or [OQuin][NTf 2 ] + pyridine + heptane} and for {[OiQuin][SCN}] + 3-methylpyridine + heptane} at temperature T = 298.15 K at atmospheric pressure. The influence of ionic liquid structure on the separation of aromatic nitrogen compounds from heptane was determined. High solubility of aromatic nitrogen compounds and particular complete immiscibility of heptane with the tested ionic liquids were observed. The experimental data were correlated using the NRTL equation, and the binary interaction parameters have been reported. The phase equilibria diagrams for the ternary mixtures including the experimental and calculated tie-lines have been presented. The values of selectivity and solute distribution ratios have been calculated based on experimental results and were compared to the literature data for other ionic liquids. [ABSTRACT FROM AUTHOR]

The conceptual design of a reactive separation process for the hydration of ethylene to ethanol requires reliable vapor–liquid equilibrium (VLE) data for the ternary system ethylene + water + ethanol. Due to the paucity of experimental data points in the VLE phase diagrams that have been reported for that system, molecular simulation looks appealing in order to predict such data. In this work, the Gibbs-ensemble Monte Carlo (GEMC) method was used to calculate the VLE of the pure components (ethylene, water, and ethanol), the binary subsystems (ethylene + water, ethylene + ethanol, and ethanol + water), and the ternary system (ethylene + water + ethanol). A set of previously validated Lennard-Jones plus point-charge potential models were chosen for the pure components, and the validity of these models was corroborated from the good agreement of the GEMC simulation results for the vapor pressure and the VLE phase diagrams of those components with respect to calculations carried out by means of the most accurate (reference) multiparameter equations of state currently available for ethylene, water, and ethanol. These potential models were found to be capable of predicting the available VLE phase diagrams of the binary subsystems: ethylene + water at 200 and 250 °C, ethylene + ethanol at 150, 170, 190, 200, and 220 °C, and ethanol + water at 200, 250, 275, and 300 °C. Molecular simulation predictions for the VLE phase diagrams of the ternary system at 200 °C and pressures of 30, 40, 50, 60, 80, and 100 atm, were found to be in very good agreement with predictions previously made by use of a thermodynamic model that combines the Peng–Robinson–Stryjek–Vera equation of state, the Wong–Sandler mixing rules, and the UNIQUAC activity coefficient model. The agreement between the predictions of these two independent approaches gives confidence for the subsequent use of molecular simulation to predict the combined phase and chemical equilibrium of the ternary system and check the validity of predictions previously made by means of the thermodynamic model. [ABSTRACT FROM AUTHOR]

Models of the equation of state (EOS) type, used for describing fluid phase equilibria, typically assume that the system behavior can be described from binary contributions only. This is not enough for obtaining good predictions for the high-pressure phase equilibria of ternary highly non ideal systems. On the other hand, cubic mixing rules (CMRs) provide both, binary and ternary interaction parameters. Such ternary parameters have the potential of improving the reproduction of ternary experimental data while leaving invariant the description of the three constituent binary subsystems. In this work, we evaluate the possibilities of the CMRs approach for ternary systems by studying the phase equilibria of two of them, i.e., CO 2 + n -hexadecane + 1,8-octanediol and CO 2 + H 2 O + 2-propanol. For the system CO 2 + n -hexadecane + 1,8-octanediol, we have available both, experimental isothermal binary and ternary phase equilibrium data. For such system, we investigate the influence of the ternary parameters on the size of ternary two-phase and three-phase equilibrium regions, and compare it to that of the experimental data. We also consider predicted values of the ternary parameters, which we need when ternary experimental data are not available. For the system CO 2 + H 2 O + 2-propanol, we examine the effect of the ternary interaction parameters on some calculated ternary univariant lines, ternary invariant points and ternary three-phase equilibria, over wide ranges of conditions. In this study, we use the Peng–Robinson (PR) EOS and Soave–Redlich–Kwong (SRK) EOS, both coupled to CMRs. Our results make evident the flexibility and improvement that can be gained in the description of ternary phase equilibria by resorting to ternary interaction parameters. [ABSTRACT FROM AUTHOR]

Abstract: The representation of liquid–liquid equilibria (LLE) in ternary systems composed by water, 1,4-dioxane and different grades of poly(lactic acid) (PDLLA and PLLA), has been addressed through the PC-SAFT equation of state (EoS), in which the scheme of induced association is used to represent the interaction between solvent (dioxane) and non-solvent (water). The model parameters devoted to the description of pure component properties, as well as those pertinent to the representation of thermodynamic behaviour of solvent/non-solvent mixtures, were tuned on the basis of specific pressure–volume–temperature (PVT) data for the corresponding systems. Only the binary parameters for polymer–solvent and polymer/non-solvent pairs were adjusted to obtain a useful representation of experimental LLE data for the ternary systems. A suitable description of the thermodynamic properties of ternary mixtures was obtained using temperature-independent binary interaction parameters in the range 25–80°C, and the consistency of the approach in the entire composition range was verified against experimental solubility data specifically measured for the polymer/non-solvent pair. The model shows good ability in the description of the thermodynamic properties of the system and it represents a reliable tool for the prediction of LLE also at conditions different from those considered for its set-up. This approach thus represents a useful designing tool for processes, such as thermally induced phase separation (TIPS), used in the preparation of microporous polymeric scaffolds. [Copyright &y& Elsevier]

Abstract: The unique physical properties of ionic liquids play a decisive part in many of their applications. Therefore, the ability to predict the physical properties of ionic liquids is extremely important for the rational design of proper ionic liquids with specific properties. In practice, the processes involving ionic liquids usually contain other components, in addition to the ionic liquids. Therefore, in addition to pure component properties, knowledge of the physical properties of mixtures are also crucial for various applications. In the present study, the feasibility of using a feed-forward multi-layer perceptron neural network (MLPNN) model was investigated to predict the electrical conductivity of the ternary mixtures of 1-butyl-3-methylimidazolium hexafluorophosphate ([bmim][PF6])+water+ethanol and [bmim][PF6]+water+acetone in the temperature range from 288.15K to 308.15K, consisting of 104 data points. Not only were different networks, namely the linear and the hyperbolic tangent sigmoid transfer functions, considered in this study, but also the effects of the number of hidden layers, hidden neurons and the training algorithm were investigated on the accuracy of the results using 78 data points as training data to minimize the average absolute relative deviation percent (AARD%), mean square error (MSE) and correlation coefficient (R 2). Among the various cases studies, statistical analyses indicated the best configuration of the network to include one hidden layer and seven neurons in the hidden layer. The optimum network was then validated using 26 data points (test data) not used in the training stage which indicated the good interpolative ability of the trained network with AARD%=1.44, MSE=2.87×10−8 and R 2 =0.9981. [Copyright &y& Elsevier]

Abstract: Knowledge of diffusion coefficients in liquids is of great importance for the calculation and simulation of mass transfer processes. In the literature, only a few models for multicomponent diffusion coefficients are found. Due to lack of experimental multicomponent diffusion data, these models have not been verified for real systems until now. To overcome this limitation, multicomponent diffusivities were measured within the whole concentration space of four ternary systems. Fick diffusivities were transformed to less concentration dependent Maxwell–Stefan diffusivities using the thermodynamic correction factor. Four prediction models were tested by comparing predicted values with experimental data. In systems with nearly ideal thermodynamic behavior, multicomponent diffusion prediction shows promising results. In strongly non-ideal systems, all tested prediction models show large deviations. Thus, the quality of predicted diffusion coefficients strongly depends on an accurate thermodynamic description of the system. [Copyright &y& Elsevier]

Abstract: The activity coefficients of sodium chloride in the NaCl+NaBF4 +H2O ternary system were experimentally determined at 298.15K, at ionic strengths of 0.3. 0.5, 1, 2 and 3molkg−1 from emf from the bi-ISE cell without liquid junction: The activity coefficients were fit to the Harned, Scatchard, Pitzer and Lim-HOLL models and compared to those of similar systems such as NaCl+NaClO4 +H2O and NaCl+NaNO3 +H2O. Other related magnitudes, such as the activity coefficients in trace quantities, osmotic coefficients, activity of water, and change in excess Gibbs energy of mixing, were evaluated and discussed in terms of ion–ion and ion–solvent interactions. [Copyright &y& Elsevier]

Abstract: Isothermal vapor–liquid equilibria measurements for binary and ternary mixtures containing carbon dioxide, 1-heptanol, and n-pentadecane at the temperatures of 298.15 and 316.15K are reported. Phase equilibria measurements were made in a high-pressure visual cell with variable volume based on the static-analytical method. Both binary systems exhibit type III phase behavior, according to the classification of van Konynenburg and Scott. The phase equilibrium in the binary systems was correlated with the Soave–Redlich–Kwong (SRK) equation of state coupled with the Huron–Vidal infinite dilution (HVID) mixing rules. The parameters for the binary mixtures are further used to predict the phase equilibria of the ternary system n-pentadecane+1-heptanol+carbon dioxide. Validation of the predictions was achieved by comparison with experimental results. The model has the capability to predict the cosolvency effect. [Copyright &y& Elsevier]

Abstract: Liquid–liquid equilibrium (LLE) data for the ternary systems water+1-butanol+methyl methacrylate or butyl methacrylate or isobutyl methacrylate were determined at atmospheric pressure and at 288.15K and 318.15K. From these new LLE data, we can find that very small amounts of the organic compounds were found in the aqueous phase, while water dissolved appreciably in the organic-rich phase. The reliability of the experimental tie-line data has been tested through the Othmer–Tobias and Bachman plots. The experimental data were also compared with the values correlated by the NRTL and UNIQUAC models. Good agreement was obtained with these models. In general, the average deviations from the UNIQUAC model are slightly smaller than those from the NRTL model. [Copyright &y& Elsevier]

Abstract: Liquid–liquid equilibrium (LLE) data for the systems water+1-propanol+methyl methacrylate, butyl methacrylate, and isobutyl methacrylate were determined at atmospheric pressure over a temperature range of 288.15–318.15K. The areas of two-phase region decrease with increasing temperature for each ternary system and decrease in the mixture containing methyl methacrylate>butyl methacrylate or isobutyl methacrylate. The reliability of the experimental tie-line data has been tested with the Othmer–Tobias and Bachman correlations. The experimental data were also compared with the values calculated by the NRTL and UNIQUAC models. Good agreement was obtained with these models. In general, the average deviations from the NRTL model are slightly smaller than those from the UNIQUAC model. [Copyright &y& Elsevier]

Abstract: In the present paper, a study of temperature behaviour of the liquid–liquid equilibrium in ternary systems N,N-dimethylformamide+2-methylpentane+methanol and N,N-dimethylformamide+methylcyclohexane+methanol. The analysis of critical curves of the liquid–liquid equilibrium by means of the regular solution model was carried out. The acquired predictions were subsequently verified experimentally. [Copyright &y& Elsevier]

Abstract: The activity coefficients of sodium fluoride in the NaF+NaClO4 +H2O ternary system were experimentally determined at 298.15K, at ionic strengths of 0.1–0.9, in 0.1molkg−1 steps from emf from the bi-ISE cell without liquid junction: The activity coefficients were compared with those of the NaCl+NaClO4 +H2O and NaBr+NaClO4 +H2O systems by using the Harned, Scatchard, Pitzer, and Lim-HOLL models. Other related magnitudes, such as the activity coefficients in trace quantities, osmotic coefficients, activity of water, and change in excess Gibbs energy of mixing, were evaluated and discussed in terms of ion–ion and ion–solvent interactions. Finally, the solubility of the NaF in the mixtures with NaClO4 were predicted and compared with values collected from the literature, with good agreement obtained. [Copyright &y& Elsevier]

Abstract: For the first time vapor¿liquid equilibrium (VLE) data for ternary systems containing ionic liquids are reported. The data were measured by means of a computer-operated static VLE apparatus at 353.15K with the ionic liquids 1-ethyl-3-methyl-imidazolium bis(trifluoromethylsulfonyl)imide [EMIM]+[(CF3SO2)2N]- and 1-butyl-3-methyl-imidazolium bis(trifluoromethylsulfonyl)imide [BMIM]+[(CF3SO2)2N]- and acetone, 2-propanol and water. The experimental VLE data of the binary systems were correlated using the Wilson, NRTL and UNIQUAC models. The errors using Wilson, NRTL, and UNIQUAC are 3.92%, 1.45%, and 1.53%. The gE-model parameters of the binary systems were used to predict the VLE behavior of the ternary systems and the predictions were compared to the experimental datasets. The errors using Wilson-, NRTL-, and UNIQUAC-parameters are 5.61%, 7.22%, and 5.02%. [Copyright &y& Elsevier]

Abstract: Two different models based on Guggenheim''s renormalized canonical partition function proposed by Wang and Vera, the generalized quasi-chemical model (GEM-QC) and the generalized regular solution model (GEM-RS), have been used in molecular form to correlate the liquid–liquid equilibrium (LLE) of experimental data for ternary systems containing water, propionic acid (PA) and organic solvents. Both the GEM-QC and the GEM-RS models have nine adjustable parameters for a ternary system, which are obtained from a least square fit of the models to the LLE experimental data. In this work the same assumptions made to develop the simplified GEM-RS model (GEM-RS2) by Wang and Vera, have been applied to the GEM-QC model (GEM-QC2), in order to reduce the number of adjustable parameters from 9 to 6. The GEM-RS2 and the GEM-QC2 models have also been used for correlating the LLE experimental data for ternary systems. The results obtained from the studied models were compared to those of the NRTL and UNIQUAC models. The comparison shows that the GEM-QC model can correlate the LLE experimental data of ternary systems more accurately than the other studied models. Finally, the selectivity of five different organic solvents that are used to recover PA from its aqueous solution has been compared. [Copyright &y& Elsevier]

Tie-line data for ternary systems containing methanol (CH4O), ethanol (C2H6O), water (H2O), toluene (C7H8), and 2,2,4-trimethylpentane (C8H18) were investigated. Phase diagrams of {w1H2O+w2CH4O+(1−w1−w2) C7H8}, {w1CH4O+w2C7H8+(1−w1−w2)C8H18}, {w1 H2O+w2 C8H18+(1−w1−w2)C7H8}, and {w1H2O+w2C2H6O+(1−w1−w2)C7H8} ternary systems were obtained at 303.15 K, while data of the other partially miscible ternary systems: {w1H2O+w2C2H6O+(1−w1−w2)C8H18}, {w1H2O+w2C7H8+(1−w1−w2)C8H18}, and {w1H2O+w2CH4O+(1−w1−w2)C8H18} were taken from the literature. The quaternary systems {w1CH4O+w2C7H8+w3C8H18+(1−w1−w2−w3)H2O} and {w1C2H6O+w2C7H8+w3 C8H18+(1−w1−w2−w3)H2O} were also studied at the same temperature. From our experimental results we conclude that these quaternary systems present a very small water tolerance and that phase separation could produce a considerable loss of CH4O or C2H6O drawn into their respective aqueous phases. On the other hand, the aqueous phase generally contains a higher concentration of C7H8 than of C8H18 for both quaternaries. All the pairs of UNIQUAC interaction parameters obtained from the four ternary subsystems included in each quaternary system were averaged and then used to predict both quaternaries liquid–liquid equilibria (LLE) with this model. The UNIFAC group contribution method was used with the same purpose. The UNIQUAC equation appears to be more accurate than the UNIFAC method, particularly for {w1CH4O+w2C7H8+(1−w1−w2)C8H18} ternary system where UNIFAC predicts an immiscibility region much larger than experimentally observed, while for the {w1H2O+w2CH4O+(1−w1−w2)C7H8} ternary one it predicts a smaller region. Therefore, the UNIFAC method only provides qualitative results for these ternary systems. However, both the UNIFAC and UNIQUAC methods predict the quaternary systems with small and similar deviations. [Copyright &y& Elsevier]

In this work, the solubility of aspirin at 293.15–313.15 K was measured in several aqueous systems where one excipient out of PEG 6000, PEG 400, PVP K25, HPMC E3 and poloxamer 188 was present. It was found that PVP K25 had the strongest solubilizing ability and the solubility enhancement of aspirin by poloxamer 188 was greatly affected by temperature. In addition, the Wilson, NRTL and UNIQUAC models were employed to predict the solubility of aspirin in various organic solvents as well as in different mixtures of polymeric excipients and water through interaction parameters obtained from experimental data. The average ARDs of Wilson, NRTL, and UNIQUAC models were 0.0300, 0.0318, 0.0305 for the binary aspirin-solvent systems, respectively, and those were 0.4214, 0.1166, 0.0861 for the ternary aspirin-polymer-water systems. UNIQUAC model among these three models was found to have excellent performance for solubility prediction of aspirin in solvents and polymer-water solutions. Image 1 • The effects of five polymers on API solubility at various temperatures were compared. • Local composition model was applied from binary system to ternary system. • Local composition model was applied to predict aspirin-polymer-water systems. [ABSTRACT FROM AUTHOR]

The reliable description of a liquid-liquid equilibrium system is irrefragably important, given that liquid-liquid equilibrium has several important applications in industry, such as the extraction of propionic acid, reactant of esterification reaction for thermoplastics production, the removal of congeners of bioethanol through the equilibrium displacement by adding water, and the proper calculation of fuel blends equilibrium composition. The use of accurate models is one of the main necessities to achieve good predictions of a system. In this work, the use of the UNIFAC(LLE) and COSMO-SAC (2002) excess Gibbs energy models in liquid-liquid equilibrium is analyzed. Liquid-liquid equilibrium predictions of 55 ternary aqueous systems are performed, the models are compared and their accuracy evaluated. The systems may contain water, alkanes, alkenes, aromatic compounds, acids, alcohols and esters. The COSMO-SAC model yielded the most reliable results, although some systems were better described by the UNIFAC-LLE model. [ABSTRACT FROM AUTHOR]