SLE Thermodynamic Consistency Tests

In Equations 10.2-10.5, is the measured or extrapolated melting point of the mixture in the limit X 1, x, is the mole fraction of the component i, is the pure melting point temperature of component i, and d. is the absolute uncertainty in tf. If the absolute uncertainty (0) for the experimental data (t ) is higher than the values found for the variable (0°), then the quality factor can be considered equal to 1, once it is not possible to determine the quality of the mixture data end points. 

Another method to analyze the consistency of the pure component data-points has been proposed by Kang et al. [43]. It analyzes the quality of the data-points close to the end points (molar fraction between Xj = 0 and Xj = 0.2, and between Xj=0.8 and Xj = l). In this methodology, the increase of the component 1 in the SEE can be calculated by considering the following equation  

This requirement is based on the Gibbs energy and SEE condition  

0/ j is the enthalpy of transfer of component 1 from solid to liquid state (enthalpy of fusion if Xj=0). The activity coefficient can be considered the unit for x, (X = 1), using the following approximation  

The quality factor is calculated using the following equation  

---

The quality of NRTL-SAC model predictions depends on the quality of the experimental data used to identify the solute parameters, as observed previously in the SLE thermodynamic consistency tests subsection of this chapter. Figure 10.11 shows the comparison with experimental data, the predictions made with the parameters obtained through UNISAC model and through regression considering only NRTL-SAC. The prediction accuracy depends on the choice of the selected data. However, when the parameters or experimental data for the new compound is not available, UNISAC model becomes very useful. 

---