Method for prediction of liquid-vapor critical points in binary mixtures: geometrical-EOS model.

A new method for predicting the Liquid-Vapor critical point of binary mixture, is presented, which is based in geometrical distances. Actually, the method is based on the minimization of the distance between the experimental and calculated values of the critical temperatures and critical pressures. The SRK and PR equations of state along with classical mixing rules of van der Waals were used as thermodynamic models to calculate the critical point of a given mixture. The proposed method requires that the mixture parameters a, b, and the covolume e = b/v of each equation of state be determined at each iteration by solving the resulting cubic equation. For nine binary mixtures containing: hydrocarbon derivatives, carbon dioxide and alcohols are studied. The AARE of the calculated values is about 0.86% for critical temperature and 2.07% for critical pressure. Good agreements are found between the calculated results and experimental data. The technique is a general purpose one and can be applied in connection with other thermodynamic models. [ABSTRACT FROM AUTHOR]