The impressive impact of including enthalpy and heat capacity of mixing data when parameterising equations of state. Application to the development of the E-PPR78 (Enhanced-Predictive-Peng-Robinson-78) model.
Jean‐Noël Jaubert, Junwei Qian, Silvia Lasala, Romain Privat
Abstract
PPR78 (Predictive-Peng-Robinson78) is a well-established cubic equation of state in which, the temperature-dependent binary interaction parameters (bips) between two molecules i and j [ k ij (T) ] are predicted by a group-contribution method. In such a model, two group-binary-interaction parameters (G-bips), noted A kl and B kl , are required per ( k,l ) pair of groups. The purpose of this article is to demonstrate that the inclusion of enthalpy and heat capacity of mixing data in the process of optimisation of the G-bips leads to a remarkable overall reduction of the error on such properties, from 841.51% to 51.42%, while keeping constant the accuracy in the prediction of VLE data, 8.6% (without inclusion of mixing properties) to 8.8% (with inclusion of mixing thermal properties). This new version of the PPR78 model (with the new sets of A kl and B kl G-bips) has thus been called Enhanced -PPR78 ( E -PPR78) model. From 2013 to 2017, this model was continuously developed and it is currently based on 40 groups. The many G-bips were fitted over 131,207 vapour liquid equilibrium data and 33,629 mixing properties data ( h M and c P M ) belonging to 1301 binary systems. After almost 18 years from the initial development of the PPR78 model, this paper summarizes for the first time the overall accuracy of the E -PPR78 model.