Abstract
| Thermochemical databases have demonstrated to be extremely important due to their usefulness in the obtention of critical materials and process design. CALPHAD (CALculation of PHase Diagram) method is used to develop thermochemical databases for multicomponent system, which requires the knowledge of thermodynamic properties such as enthalpy of formation, standard entropy and heat capacity of all the present phases (compounds and solutions) of a chemical system. Nevertheless, information of many phases is still scarce or inexistent. A reliable and precise method to estimate such properties for a wide variety of inorganic compounds would be a key of the advancement in thermodynamic modeling without demanding a large quantity of resources and time. The objective of this project is to develop a hybrid model based on Polyhedron method and Neural Network for predicting thermodynamic properties of several oxides. |