Generalized Interaction Parameter for the Modified Nonrandom Two-Liquid (NRTL) Activity Coefficient Model

Activity
coefficients are used in phase equilibria calculations
to account for the nonideal behavior of liquids in a mixture. In our
previous work, we proposed a modification to the widely used nonrandom
two liquid (NRTL) activity coefficient model that reduced the number
of parameters from two to one. The current work is an extension to
the previous work, and it focuses on generalizing the parameter of
the modified NRTL (mNRTL1) model using a theory-framed quantitative
structure–property relationship (QSPR) modeling approach, where
the mNRTL1 model was used as a theoretical framework to develop the
behavior model, and QSPR was used to generalize the substance-specific
parameter of the model. In this work, a VLE database consisting of
916 binary systems with a wide range of functional-group interactions
was assembled. Data regression analyses were performed to determine
the parameter of the mNRTL1 model. The structural descriptors of the
molecules were used as inputs in the QSPR model to predict the regressed
parameter. The generalized QSPR model produced equilibrium property
predictions within two times the errors obtained through regression
of the experimental data. The newly developed QSPR model resulted
in comparable phase equilibria predictions to that of the group-contribution
method, UNIFAC. In comparison, the QSPR model has less than 10% of
the number of parameters used by UNIFAC and is applicable to a wider
range of systems.