A new correlation method for vapor-liquid equilibria and excess enthalpies for nonideal solutions using a genetic algorithm. Application to ethanol + an n-alkane mixtures

Abstract

This study has used a new polynomial expression with temperature-dependent coefficients as a function based on the so-called active fraction of the mixture components, z(i), which enabled the thermodynamic vapor-liquid equilibrium quantities, the Gibbs excess energy function and corresponding activity coefficients, and the excess enthalpies all to be correlated simultaneously. Correlation is carried out by using a least-squares procedure implemented in a genetic algorithm that optimized an objective function selected for that purpose. The model is applied to a set of thermodynamic data for binary systems composed of ethanol + one of five n-alkanes (hexane to decane) collected as part of the study. Good results were obtained using the model, which yielded good estimates of the mixing properties, gE and hE, which in turn were used for comparison with the results obtained using certain more conventional models. The expression that yielded the best correlation of the excess Gibbs energy function for the binary systems considered as a function of the concentration of one of the components and temperature was (gE/RT)(T,x1) z 1z2[(A01T + A02/T + A03) + (A21T + A22/T + A23)z1 2].