Liquid—liquid phase equilibria for dearomatisation of ATF fraction

The present work describes a method to develop a simulation model for dearomatisation of the straight-run fraction that boils in the range 140–240°C (also called Aviation Turbine Fuel fraction) of Bombay high crude (India—offshore). Sulpholane was used as a solvent to extract aromatics from this fraction. A mass spectrometric analysis of the ATF fraction identified six hydrocarbon types; each of these was represented by one or more components. The group contribution model, UNIFAC, has been investigated for the prediction of the multicomponent liquid—liquid equilibria. The compositions of the extract and raffinate phases, thus predicted, compare closely with single stage experimental data. Multistage countercurrent extraction calculations have also been carried out using the above model and checked satisfactorily against experimental data that were available on a 27 mm-diameter packed laboratory extraction column. It has been shown that the representation of ATF fraction by model hydrocarbons, together with the equilibria predictions by UNIFAC, form a sufficiently accurate data base for the simulation model.     

Liquid–liquid equilibrium calculations for methanol–gasoline blends using continuous thermodynamics

This work presents the application of continuous thermodynamics to investigate the limited miscibility of methanol–gasoline blends. To predict the liquid–liquid equilibrium of these systems, the Gaussian distribution function was used to represent the composition of paraffins in the gasoline. The naphthenes and aromatics were represented by model compounds. A model has been developed using three different continuous versions of the UNIFAC model. Methanol is an associating component, and association affects phase equilibria. Therefore, the CONTAS (continuous thermodynamics of associating systems) model based on the Flory–Huggins equation, for multicomponent methanol–gasoline blends has also been investigated. The predicted results including the cloud point curve, shadow curve and phase separation data have been compared with experimental data and good agreement was found for the two UNIFAC and CONTAS models.     

Prediction of infinite dilution activity coefficients for systems including water based on the group contribution model with mixture-type groups: II. Extension and application

Group contribution models such as ASOG or UNIFAC were known to be inaccurate in the prediction of infinite dilution activity coefficients (γ^∞) for most of the systems containing water. To overcome the weakness inherent with the UNIFAC models, Zhang et al. (Fluid Phase Equil. 149 (1998) 27) have recently proposed a group-contribution-based model with mixture-type groups, where the mixture-type group is a hypothetical concept for taking into account the particular hydrophobic effects in aqueous organic systems. The proposed methodology has been proven to be applicable to alkane/water and alcohol/water mixtures in our previous study. In this work, the proposed method was further extended to the other classes of compounds, e.g. aromatics, ketones, acids, aldehydes, esters, ethers, nitriles and halogenated compounds. Compared to the conventional UNIFAC models, the proposed method demonstrates significant improvements in accuracy for various organic compounds in water mixtures.     

Solubility of pyrene in simple and mixed solvent systems

The solubility of pyrene was experimentally determined in simple and complex solvent systems (single, binary, ternary, quaternary and pentinary solvent systems) composed of benzene, ethylbenzene, hexane, hexanol and methylcyclohexane over a temperature range from 293 to 318 K. In addition, six models were used in this study to represent pyrene solubility in the different solvent systems. The interaction parameters for modified Wilson, NIBS/Redlich-Kister, UNIQUAC and NRTL models were estimated using the solubility data generated for pyrene in single, binary and ternary solvent systems. By re-adjusting the interaction parameters reported for Dortmund UNIFAC and ASOG models, a better representation of the solubility of pyrene was obtained compared to using reported values. Furthermore, a correction term is introduced for the ASOG model in this study to better improve pyrene solubility prediction in simple and mixed solvent systems. These estimated or re-adjusted interaction parameters for the different models, along with the reported parameters for Dortmund UNIFAC and ASOG models, were tested on complex solvent systems (quaternary and pentinary solvent mixtures), in order to check their validity and accuracy for such predictions.     

Liquid phase equilibria of (water + propionic acid + oleyl alcohol) ternary system at several temperatures

(Liquid–liquid) equilibrium (LLE) data are investigated for mixtures of (water + propionic acid + oleyl alcohol) at 298.15, 308.15 and 318.15 K and atmospheric pressure. The solubility curves and the tie-line end compositions of liquid phases at equilibrium were determined, and the tie-line results were compared with the data predicted by the UNIFAC method. The phase diagrams for the ternary mixtures including both the experimental and correlated tie-lines are presented. The distribution coefficients and the selectivity factors for the immiscibility region are calculated to evaluate the effect of temperature change. The reliability of the experimental tie-lines was confirmed by using Othmer–Tobias correlation. It is concluded that oleyl alcohol may serve as an adequate solvent to extract propionic acid from its dilute aqueous solutions. The UNIFAC model correlates the LLE data for 298.15, 308.15 and 318.15 K with a root mean square deviation of 5.89, 6.46, and 6.69%, respectively, between the observed and calculated mole concentrations.     

聚合物溶液热力学模型的评述

预测聚合物溶液体系的气液平衡数据是热力学模型的一个重要内容.热力学模型大体可分3类,以过量吉布斯自由GF能表示的最为常用.本文选取基于GF的Flory-Huggins(FH)、UNIQUAC、UNIFAC和ENSIC 4种常用模型进行评述,认为FH模型出现较早,形式最简单;UNIQUAC模型发展较完善,应用范围较广,UNIFAC模型弥补了UNIQUAC模型参数缺乏的不足,应用最为;ENSIC模型具有较好的预测效果,但参数难以获得.     

An Economical Method for Isolation of Dioscin from Dioscorea nipponica Makino by HSCCC Coupled with ELSD, and a Computer-Aided UNIFAC Mathematical Model

An economical method for isolation of dioscin from Dioscorea nipponica Makino by high-speed counter-current chromatography (HSCCC) was successfully developed by using a UNIFAC mathematical model coupled with computer-aided counter-current chromatography solvent-selection software (CCC-SSS) for separate preparation of the components of the solvent system (i.e., the stationary and mobile phases). The solvent system n-hexane–ethyl acetate–ethanol–water 2:5:2:5 (v/v) was selected to demonstrate the feasibility of the approach. A comparative study was also carried out on different methods for preparation of the solvent system, namely conventional preparation of the mobile and stationary phases together in the same vessel and the method developed for separate preparation of the phases. The results indicated that purity and recovery of dioscin were no different when solvent systems prepared by the different methods were used for HSCCC separation. Much less n-hexane, ethyl acetate, and ethanol was used when the mobile and stationary phases were prepared separately, however. This was not only environmentally sensible, but also enabled conservation of resources. Use of the UNIFAC mathematical model combined with the CCC-SSS technique for separate preparation of the components of the solvent system in HSCCC is reported and explained. It is a simple and economical means of isolating pure dioscin from Dioscorea nipponica Makino.     

Solubility of polycyclic aromatics in binary solvent mixtures using activity coefficient models

Polycyclic aromatic hydrocarbons (PAHs) precipitation is one of the major problems in the hydrocracking units. In this investigation, pyrene and phenanthrene were selected because they were found to be in higher concentrations in the feed to hydrocracking units. Their solubilities were investigated in toluene solvent mixture of iso-octane and heptane over a temperature range from 293 to 323 K. The experimental solubility data were used to predict the interaction parameters for seven different solid–liquid equilibrium models. The following activity coefficient models were used; Wilson, NIBS/Redlich–Kister, UNIQUAC, modified UNIFAC, modified UNIFAC (Dortmund), Flory–Huggins and Sheng. The interaction parameters were expressed as a second-order polynomial function in temperature. In order to test the models, the average absolute deviation percentage (AADP) was used. The overall AADP was found to range from approximately 7 to 14%. The models can be arranged according to their accuracy in a descending order based on AADP as follows: NIBS/Redlich–Kister, Wilson, UNIQUAC, Sheng, Flory–Huggins, modified UNIFAC (Dortmund) and finally modified UNIFAC. All models used in this work gave reasonable results; however, the group contribution models can also be used as a predictive tool for the solubility measurement of pyrene and phenanthrene in other solvents containing the same groups of the solvents used in this study.     

柱液体色谱分析沥青类产品的烃族组成

以柱液体色谱联合红外光谱(IR)、元素分析等手段,建立了一项测定渣油和沥青中烃族的族组成分析方法.通过色谱参数的评选和红外光谱的核实所建立的此方法具有溶剂和担体用量少、色谱分辨率高的特点,是一安全、经济、先进的族组成分析方法.所建方法中,选择的模型化合物为试剂纯碳24烷烃、二苯并[ah]蒽、乙酰苯胺.冲洗所获模型物制备段分,其红外光谱特征和元素分析结果和原试剂相一致,且具有好的回收率.所建方法对渣油和沥青实样的测定,不仅有很好的重复性,还具有很高的分离、分辩能力.IR和核磁共振氢谱(1H-NMR)的测定充分地证实了各制备段的不同特征.     

Prediction of thermodynamic properties of polymer solutions by a group-contribution method

The Flory–Huggins lattice-theory expression for solvent activity in a polymer-solution is commonly used to calculate the thermodynamic interaction parameter χ with the aid of experimental data from vapor pressure osmometry. This expression assumes that χ is independent of composition. However, experimental data for a variety of polymer-solvent mixtures indicate that χ exhibits an appreciable concentration dependence. A group contribution method, UNIFAC (UNIQUAC Functional-Group Activity Coefficients) incorporating the free-volume correction of Oishi and Prausnitz is used to predict the dependence of χ on solvent concentration. Agreement with previously reported experimental data is within 15%. Calculated values of χ obtained from the Flory–Huggins expression for solvent activity and from the corresponding Gibbs free energy of mixing (which does not assume that χ is independent of composition) are compared. Calculations based on the Gibbs free energy of mixing predict a somewhat larger value of χ relative to those based on solvent activity. The specific Gibbs free energy of mixing for polystyrene-solvent mixtures is calculated using the UNIFAC model, and is found to represent qualitatively the phase equilibrium behavior. Quantitative discrepancies are observed, however, for the polystyrene-acetone system in light of the actual experimental solubility reported by Suh and Clark (20). Most of the thermodynamic predictions for polymer-solvent systems investigated herein are correlated qualitatively with the relative mismatch between solubility parameters of both components.     

Solid-liquid equilibria and purity determination for binary n-alkane + naphthalene systems

Mixtures of heavy aromatics with high aliphatics are important in the formation of asphaltenes in the oil industry.This work reports binary solid-liquid equilibria for naphtalene + eicosane, +pentacosane, +hexatriacontane mixtures by differential scanning calorimetry. Results are compared with those from modified UNIFAC (Larsen and Gmehling versions) and ideal predictions. Finally, we determine the purity according to van’t Hoff equation. Results are in good agreement with values given by ultraviolet spectrophotometry.     

Revision of UNIFAC group interaction parameters of group contribution models to improve prediction results of vapor–liquid equilibria for solvent–polymer systems

The objective of this work was to improve the accuracy of group contribution models for prediction of solvent activities in polymer solutions by revising UNIFAC group interaction parameters using a wide range of vapor–liquid equilibrium (VLE) data of solvent–polymer systems. The group contribution models considered in this work were UNIFAC-FV, Entropic-FV, GK-FV and UNIFAC-ZM models. A total of 142 systems that consisted of 16 polymers and 36 solvents containing a large variety of solvent–polymer systems ranging from non-polar to polar substances were considered to optimize 46 pairs of group interaction parameters. Data considered were split up into systems containing alkane and cycloalkane, aromatic, and polar solvents. For athermal systems, the UNIFAC-FV model gave the best results. Therefore, the model was used in optimizing the group parameters. Revised group interaction parameters were found to improve the reliability of VLE predictions in solvent–polymer systems. A significant improvement of prediction results was achieved by UNIFAC-FV model from 20.0 to 10.8% absolute average deviation (AAD) in solvent activities for systems containing polar solvents and from 16.7 to 10.9% AAD for all systems. The prediction results of GK-FV and UNIFAC-ZM models were also improved.     

Calculation for liquid-liquid equilibria of quaternary alkane-ethyl acetate-methanol-water systems used in counter-current chromatography

The calculation of liquid-liquid equilibrium compositions of solvent systems is very important for the work on counter-current chromatography (CCC), especially the phase composition and volume ratio obtained from liquid-liquid equilibrium calculation. In this work, liquid-liquid equilibria of quaternary Arizona solvent systems, alkane-ethyl acetate-methanol-water, and related ternary systems are correlated and predicted using the non-random two-liquid model (NRTL). Hexane, heptane and isooctane are the used alkanes. The parameters in the model are regressed only with the special systems considered. Detailed comparison with experimental data shows that liquid-liquid equilibria of these systems can be predicted with greatly improved accuracy as compared to the group contribution method (UNIFAC).     

计算离子液体溶液汽液相平衡的分子热力学模型

用平均球近似理论、微扰理论和UNIFAC基团贡献方法分别考虑离子之间的长程静电作用、离子与溶剂之间的中程静电作用以及所有粒子之间的短程作用,本文提出了一种新的分子热力学模型,可用于离子液体溶液中溶剂活度系数的计算.通过对含烷基咪唑磷酸酯类离子液体与水、甲醇或乙醇组成的9个二元体系的饱和蒸汽压数据进行关联,获得了相关的模型参数,即溶剂的分子直径和基团之间的交互作用能参数.溶剂活度系数及饱和蒸汽压的计算结果与实验值的平均偏差为1.40%,符合良好,因此本模型可望用于含离子液体体系汽液相平衡的预测.     

气液色谱法研究聚合物/溶剂体系气液平衡

 用气液色谱法测定了聚二甲基硅氧烷（PDMS）／溶剂、聚甲基丙烯酸甲酯（PMMA）／溶剂体系在不同温度下以质量分数表示的无限稀溶剂活度系数和Flory－Huggins相互作用参数。应用UNIFAC和UNIFAC－FV模型对PDMS／溶剂、PMMA／溶剂体系中以质量分数表示的无限稀溶剂活度系数进行了估算。结果表明,用这两个模型预测PDMS／溶剂、PMMA／溶剂体系中的无限稀溶剂活度系数有待修正或采用其它模型进行估算。     

Liquid–liquid equilibria of aromatics removal from middle distillate using NMP

We investigated the removal of aromatics from a typical Kuwait middle distillate using liquid–liquid extraction. A multicomponent system; dodecane + hexadecane + 1,3,5-trimethyl benzene (mesitylene) + butylbenzene; was used to represent the middle distillate mixture. N-methyl-2-pyrrolidone (NMP) was used as a solvent. The liquid–liquid equilibria (LLE) of the mixture was studied over a temperature range of 293–323 K and at different solvent to feed ratios (1.35–1.97). The selectivity and distribution coefficient of the solvent were correlated successfully using the universal quasi-chemical (UNIQUAC) model for the pseudo-ternary system (paraffins (1) + aromatics (2) + NMP (3)) at the studied temperature range. The UNIQUAC model fit the experimental data with an average percentage root mean square deviation (rmds) of 0.2980%.     

Liquid + liquid equilibria for the quaternary systems of (water + acetic acid + mixed solvent) at 298.2 K and atmospheric pressure

Liquid–liquid equilibrium (LLE) data for the quaternary systems of [water + acetic acid + mixed solvent (dipropyl ether + diisopropyl ether)] were measured at 298.2 K and atmospheric pressure, using various compositions of mixed solvent. Binodal curves and tie-lines for the quaternary systems have been determined in order to investigate the effect of solvent mixture, dipropyl ether (DPE) and diisopropyl ether (IPE), on extracting acetic acid from aqueous solution. A comparison of the extracting capabilities of the mixed solvents was made with respect to distribution coefficients, separation factors, and solvent free selectivity bases. Reliability of the data was confirmed by using the Othmer–Tobias and Hand plots. The tie-lines were also correlated using the UNIFAC model. The average root-mean-square deviations between the observed and calculated mass fractions for the studied systems were in the range of 10–14%.     

Prediction of phase equilibrium in mixtures containing ionic liquids using UNIFAC

The UNIFAC model is extended to mixtures of ionic liquids consisting of the imidazolium cation and the hexafluorophosphate anion with alkanes, cycloalkanes, alcohols and water. Two new main groups, the imidazolium and the hexafluorophosphate groups, are introduced in UNIFAC. The required group interaction parameters between these groups and the existing UNIFAC main groups, CH₂, OH and H₂O, are determined by fitting binary liquid–liquid equilibrium and infinite dilution activity coefficient experimental data. The predictive capability of the extended UNIFAC model is examined against experimental data for vapour–liquid equilibrium, liquid–liquid equilibrium and activity coefficients at infinite dilution of binary and ternary systems containing 1-alkyl-3-alkyl′-imidazolium hexafluorophosphate ionic liquids, alkanes, cycloalkanes, alcohols and water. The results indicate that UNIFAC is a reliable model for phase equilibrium predictions in mixtures containing this type of ionic liquids.     

逆流溶质出峰顺序的预测

 根据逆流保留方程可知,溶质的出峰顺序主要取决于其在互不混溶的两液相间分配系数的大小顺序。以饱和及不饱和脂肪酸乙酯、2,4 二硝基苯胺基脂肪醇和对硝基苯基葡萄糖甙等结构较为简单、极性差别较大的溶质系列为研究对象,应用修正的通用基团活度系数(UNIFAC)(Dortmund)模型,通过相平衡计算,预测了它们在含水或不含水的溶剂体系中分配系数的变化趋势,并与前人的工作进行了对比。结果表明,由该方法预测出的相似结构溶质分配系数的顺序基本上与实验值相同,其中对饱和脂肪酸乙酯在己烷 乙腈(体积比为1∶1)体系中分配系数的预测最好。     

Liquid—liquid equilibrium in methanol + gasoline blends

An investigation has been carried out to study the limited miscibility of methanol and gasoline blends over the temperature range −20 to 20°C. Two liquid phases in equilibrium were analysed by mass spectrometric methods and their composition reported, in addition to the methanol content, in terms of six principal classes of hydrocarbons. Liquid—liquid equilibrium was predicted using the UNIFAC group contribution model. In liquidliquid equilibrium calculations, gasoline was represented by a set of model compounds. The number of the different groups that comprise each model molecule was determined using the result of a distillation analysis and the paraffin—naphthene—aromatic composition. Estimation of conjugate phase composition using the UNIFAC model is reasonable at temperatures above 0°C. To describe correctly the limited miscibility of methanol+gasoline blends over the whole temperature range studied, we found that ‘specific’ UNIFAC interaction parameters were necessary.