A predictive model for the entropies and heat capacities of zeolites

The classical predictive models for thermodynamic entities are based on summation techniques. At the time these predictive methods were designed, no or few calorimetric data on zeolites were available. The number of calorimetric measurements (third-law entropy and heat capacities) performed since 1990 on various zeolitic minerals allows a quick verification of all algorithms. It shows that the predicted values display a discrepancy above 4% relative to the measured values. Recent predictive methods for entropy and for heat capacity also give very high differences between predicted and experimental values. With a limited number of thermodynamic values of zeolites and their dehydrated forms, the entropy and heat capacity predictions have been improved by using the thermodynamic properties of zeolite-like silica polymorphs and of zeolitic water obtained from the difference between the thermodynamic parameters of anhydrous and hydrated zeolites. The improvements introduced in this work reduce the prediction errors to about 2.95% for entropy and 2.86% for heat capacities.