New Fitting Method for Vapour-Liquid Equilibrium Data and its Application for Ethanol-Water Distillation Process Modeling

Csaba Dezso Andras,Eva Molnos,Alexandru Szep,Laszlo Matyas

doi:10.37358/rc.20.7.8219

Abstract

For modeling and simulation of distillation processes, the main information required is the vapour-liquid equilibrium (VLE) data. These are available as discrete values, but for modeling purpose, an analytical function would be more suitable. We developed a simple calculation protocol to obtain a single continuous function for the VLE curve on the entire concentration domain, without using thermodynamic functions. The fitting parameters of binary ethanol-water mixture VLE curve were determined. The methodology is suitable for fitting continuous functions on VLE data of any binary or multicomponent, non-ideal (even azeotropic) mixtures.

Highlights

The modeling of separation processes in chemical engineering is indispensable for the modern design of chemical equipment
To find an appropriate fitting function for the relative volatility (RV), firstly, we obtained their values from the vapour-liquid equilibrium (VLE) data, using the defining equation and making their graphical representation in function of ethanol molar concentration in the liquid phase (x)
The modeling of innovative distillation methods with general nonlinear VLE behavior requires more accurate non-iterative design algorithms [16], and the presented fast, and accurate method could be a part of the modeling arsenal

Summary

Introduction

The modeling of separation processes in chemical engineering is indispensable for the modern design of chemical equipment. Poor thermophysical property or equilibrium data lead to significant calculation errors and conduct to erroneous equipment design [1]. Any improvements in this direction increase the accuracy of the predictions. A fitting method is needed, but in many cases (mainly for highly non-ideal, azeotropic mixtures), which presents a concave region, this could be a challenging task For this purpose, equations of state [2] or Gibbs excess energy method with different activity coefficient (NRTL, Wilson, UNIQUAC) or thermodynamic models (Margules or van Laar equations) are commonly used [3]

Methods

Results

Conclusion