STRUCTURAL AND MORPHOLOGICAL CHARACTERIZATION OF IRON SULPHIDE (FES) TIN FILMS

ABSTRACT

This project successfully explored two methodologies for predicting the bulk modulus of Tantalum Iron Antimonide (TaFeSb) half-Heusler alloy. The first approach utilized first-principle calculations based on Density Functional Theory (DFT), which yielded a robust bulk modulus value of approximately 145 GPa, reflecting the material's impressive stiffness and suitability for applications in spintronics and thermoelectricity. Furthermore, the project employed a Machine Learning (ML) model, specifically a Support Vector Machine (SVM), to predict the bulk modulus of TaFeSb. This ML model was trained on a dataset of 2382 ternary compounds with elastic data from Aflow datasets, demonstrating the versatility of ML in materials science research. The model exhibited high accuracy and precision, with a Root Mean Squared Error (RMSE) of 0.0135 and a calibration uncertainty percentage error of 2.6%.