Design of experiments for thermodynamic model discrimination applied to phase equilibria at high pressures

Abstract

Abstract Traditionally, the screening of thermodynamic models in phase equilibria is carried out based on the full data set and, therefore, after having obtained all the experimental points. The aim of this work is to show that sequential model discrimination may be used as a guide for obtaining experimental data for systems at high pressures. We have compared two discrimination procedures in order to select the next experimental point, one of them considering only the predicted responses (Hunter and Reiner [1]) and the other taking also into account the error limit inherent to each model (Ferraris and Forzatti [2]). In order to test these methodologies, literature data were used along with current thermodynamic models. It is shown that the sequential discrimination leads to the same conclusions as obtained when using the full data set, but with much less effort.