Predictive Thermodynamic Modeling of Poorly Specified Mixtures and Applications in Conceptual Fluid Separation Process Design

Abstract

In many chemical engineering processes, mixtures occur whose composition is not well known. Simulations of processes with such poorly specified mixtures are basically unfeasible, as thermodynamic models require knowledge of the speciation. As a
workaround, pseudo-components can be introduced, which are generally defined using ad-hoc assumptions. In a previous series of works, we have developed and tested a method for quantifying the group composition of poorly specified mixtures based on
simple NMR experiments, without having to solve the much more complicated task of quantifying all species. This method has been extended recently to a rational method for defining pseudo-components in poorly specified mixtures. In the present work, this method is combined with several thermodynamic group-contribution methods. The resulting thermodynamic models were applied to various poorly specified mixtures and used for solving two typical tasks from conceptual fluid separation process design: a) solvent screening for liquid-liquid extraction processes and b) simulation of open evaporation processes. The results show excellent agreement with those obtained using the complete information on the speciation, which was known for the studied test mixtures but disregarded in applying the new approach.