Effect of Data Type on Thermodynamic Model Parameter Estimation:  A Monte Carlo Approach

Abstract

Parameters for the UNIQUAC (universal quasi-chemical) model were regressed from VLE and LLE data for the systems:  diisopropyl ether + acetic acid + water; 1,1,2-trichloroethane + acetone + water; and chloroform + acetone + water. The results show a very significant effect of the data type used in the regressions (binary versus ternary) on the uncertainty of the predicted performance of sample liquid−liquid extraction units.

Publication
Industrial & Engineering Chemistry Research
Date