Entropic-FV Models Research Articles

A new proposed thermodynamic model for aqueous polymer solutions

In this research a novel model was proposed for calculation of the phase behavior in polymer solutions. The proposed model consists of modified Freed-FV, non-randomness factor and the new expression derived volume-based group contribution local composition model activity coefficients. Firstly the proposed model was considered without pseudo chemical part to obtain the interaction parameters of functional groups (This is for the use of interaction parameters of functional groups in various systems). The binary interaction parameters of the functional groups of new local composition model have been calculated by correlation of the experimental vapor-liquid equilibria (VLE) data with different molecular weights (MW) and temperatures of several polymers. The interaction parameters of functional groups were reported and the deviations of new local composition model were compared with UNIFAC and Entropic-FV models and proved the proposed model has an acceptable advantage over the mentioned models. By obtaining the practical interaction parameters of functional groups, the proposed model was applied for modeling of liquid-liquid equilibria (LLE) of polymer-polymer aqueous two-phase systems. Also, it was compared with UNIQUAC and UNIQUAC-NRF models. The suggested model consists of three adjustable parameters. Moreover, the coordination number was developed based on molar fraction and molecular size of solution components. The results revealed that the proposed model can predict aqueous multicomponent polymer systems with great precision.

Mass transfer of hydroxyethyl cellulose membranes for desulfurization of FCC gasoline: Experimental and modeling

Hydroxyethyl cellulose (HEC) pervaporation membranes were employed for the desulfurization of fluid-catalytic-cracking (FCC) gasoline. Based on solution-diffusion mechanism, the solubility of sulfur during the pervaporation process was studied by carrying out experiments and modeling, and UNIFAC-FV, UNIFAC-ZM, and Entropic-FV models were compared. The results show that the prediction by UNIFAC-FV model is more accurate than the other two models, especially when the sulfur mass fraction in the feed is <4000μgg−1. The mass fraction of sulfur adsorbed onto the polymer surfaces increases with increasing temperature and sulfur content in the feed, and the latter is more effective than the former. The distribution coefficients of thiophene and n-heptane were calculated by using UNIFAC-FV model, and further compared to the experimental results.

Improvement of an Entropic-FV model based on solubility parameters for prediction of vapor–liquid equilibria of solvent–polymer systems

In this work, a modified Entropic-Free-Volume (FV) model was proposed for improved accuracy in the prediction of vapor–liquid equilibria (VLE) for solvent–polymer systems. The model was developed by combining an Entropic-FV combinatorial term with a residual term derived from the enthalpy of mixing determined from solubility parameters. Since solubility parameters can be predicted by group contribution models and extensive group parameters are available, the present model can be applied to a wide range of solvent–polymer systems. The accuracy of the present model was tested with finite-concentration VLE data for 16 polymers and 36 solvents composing a large variety of solvent–polymer systems including polar and nonpolar solvents and polymers. The data considered were divided into three categories: systems containing nonpolar, polar without hydrogen bonding and polar with hydrogen bonding solvents. The present model gave good accuracy for all solvent categories. In comparison with the original Entropic-FV model, the present model achieved significant improvements, reducing average absolute deviations in solvent activities from 7.3% to 2.9%, 8.8% to 2.2% and 18.5% to 3.0% for systems containing nonpolar, polar and polar-with-hydrogen bonding solvents, respectively.

A free-volume term based on the van der Waals partition function for the UNIFAC model

The objective of this work was to develop a model that takes into account the unique free-volume effect of water. The free-volume percentage of water is less than that of most polymers. This is opposite to the free-volume behavior of most solvents with polymers. As a result, the free-volume term in the UNIFAC-FV model introduced by Oishi and Prausnitz fails for aqueous systems. In the new model a free-volume term derived from the generalized van der Waals partition function is added to the combinatorial and residual terms of UNIFAC. In UNIFAC the geometric parameters were generally obtained directly from the van der Waals values. For water, however, these parameters were regressed. With the new free-volume term better predictions were obtained when the standard van der Waals geometric parameters for water were used. The new model was compared with the UNIFAC-FV model of Oishi and Prausnitz and the Entropic-FV model of Elbro et al. The new model was found to give better predictions for the aqueous polymer systems as well as for the non-aqueous polymer systems.

Prediction of mutual-diffusion coefficients in polymer solutions using a simple activity coefficient model

Using a simple activity coefficient model in conjunction with the Vrentas–Duda free-volume (FV) theory, the diffusion behavior of small molecular solvents between and amongst polymer chains is then predicted for several polymer/solvent systems over a wide range of concentrations. The estimations are comparable with experimental data for some systems. For diffusion over- and under-predicted systems, a pre-exponential factor D 0 can be adjusted to improve the predictions. It is found that the entropic-FV model is a suitable alternative to the Flory–Huggins equation for predicting mutual-diffusion coefficients in polymeric solutions. To understand the impact of the model on the prediction thoroughly a sensitivity analysis is done for free-volume. The influence of the free-volume of solvent is the opposite of that of the polymer.

Free-Volume Activity Coefficient Models for Dendrimer Solutions

The ability of the Unifac-FV and Entropic-FV models to predict phase equilibria for dendrimer systems is investigated in this work. Different approaches are considered for the estimation of the den...

Phase equilibria for complex polymer solutions

Many commercially important mixtures contain complex polymers, e.g. paints and coatings. If a good thermodynamic description can be given of these systems it is possible to develop paints, which possess a certain set of properties and at the same time meet some basic requirement as, e.g. regarding the content of organic solvents. This work presents an investigation of the three polymer models Entropic-FV (EFV), UNIFAC-FV (UFV) and GC-Flory (GCF) for their capability of predicting solvent activity coefficients in binary systems containing complex polymers. It is possible to obtain good predictions at finite concentrations and satisfactory predictions at infinite dilution, particularly with the EFV model. The investigation shows that EFV is the most robust and stable of the models, which indicates that it is the most well suited model for further development of methods for predicting the miscibility behavior of paints and related systems.

Prediction of Liquid-Liquid Equilibrium for Binary Polymer Solutions with Simple Activity Coefficient Models

Liquid-liquid equilibrium predictions for binary polymer solutions using four simple UNIFAC-based activity coefficient models are presented in this work. The four models are the original UNIFAC, the new UNIFAC recently developed in Lyngby, a modified Flory-Huggins model, and the recently developed entropic-FV model. All four models are purely predictive, since they are based on the group-contribution approach. They employ existing UNIFAC group interaction parameter tables, which have been estimated from vapor-liquid equilibrium data of mixtures with exclusively low molecular weight compounds. The investigated models are capable of predicting UCST qualitatively well, but only the modified Flory-Huggins and the entropic-FV models can predict LCST (near the critical temperature of the solvent) and are, thus, able to describe the combined UCST/LCST behavior often found in polymer solutions. In particular, the entropic-FV model can represent the five types of phase diagrams which are most often encountered in polymer solutions, including closed loops and hourglass type. Furthermore, the predictions with the entropic-FV model are in several cases (especially for nonpolar solutions) semiquantitatively correct and generally far more accurate than those provided by the classical UNIFAC models