Development of Thermodynamic Models with a View to the Synthesis and Design of Separation Processes

Abstract

The design of the different processes in chemical industry can be carried out with the help of computers by solving the resulting equations for the material and enthalpy balance. As input besides kinetic data (transport properties, reaction rate information) information about the real behavior of fluid mixtures as function of concentration, temperature and pressure is required.

For separation processes especially the reliable knowledge of the different phase equilibria (vapor-liquid equilibria (VLE), liquid-liquid equilibria (LLE), ..) in multicomponent mixtures is of great importance.

These phase equilibria can be described by using activity coefficients (g^E-models) or fugacity coefficients (equations of state) to account for the deviation from ideal behavior. Both methods do require experimental binary information. When this binary information is missing reliable group contribution methods can be used to supplement the experimental information.

In this paper different predictive methods are presented which can be applied to predict the real behavior of fluid mixtures. Furthermore the importance of the Dortmund Data Bank for the development of these methods is discussed and different important applications of these models are shown.