A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Bio-Nano Material Science and Engineering

Thesis (M. Eng.) - Asian Institute of Technology, 2024

Zeolitic imidazolate frameworks (ZIFs), which are porous crystalline materials composed of metal centers (mostly Zn (II) or Co (II)) and imidazole-based ligands, have garnered significant attention due to their versatile applications in gas separation and catalysis. In this regard, the mechanical properties of ZIFs are relevant to study, to obtain information on their flexibility and the effect on species adsorption and release. Traditional methods for predicting the properties of ZIFs are time-consuming and computationally intensive. This research explores the application of machine learning techniques to predict the mechanical properties of a set of ZIFs with sufficient accuracy and computational efficiency. By leveraging a dataset of ZIF structures and their corresponding mechanical properties, machine learning was trained and used to predict key mechanical attributes such as shear modulus (G) and bulk modulus (K). The results will be used not only to assess the potential of machine learning as a valuable tool for calculating and predicting the mechanical performance of ZIF materials but also to enable the design of novel ZIF-based materials with tailored mechanical characteristics for various applications. This research can offer valuable insights into the synergy between traditional computational chemistry and machine learning, opening new avenues for the efficient exploration and development of ZIFs with desired mechanical properties.