Limits on the simultaneous correlation of gE and hE data by the NRTL, LEMF and Wilson's equations

Abstract

Three currently popular excess free energy models (Wilson's equation, the NRTL equation, and the LEMF equation) were subjected to a theoretical parametric analysis to determine limits to their ability to correlate experimental g E and h E data simultaneously. The LEMF equation was found to be distinctly superior in its ability to predict VLE data from h E data. Both Wilson's equation and the NRTL equation were shown to break down to ideal solution models in the limit of large intermolecular interactions (| h E| max. > 200 cal gmol −1) whereas the LEMF equation does not. For mixtures whose h E data exhibit maxima less than 100 cal gmol −1 and which have positive s E the LEMF equation coupled with the method of Hanks, Gupta, and Christensen can predict reliable VLE data from h E data. For | h E| max. > 200 cal gmol −1, the LEMF equation/Hanks—Gupta—Christensen method is accurate to within 10–15% where the other two equations generate errors in excess of 40%.