Topic of Research Seminar: Recent Advances in Computational Modelling: From Molecular Simulations to Machine Learning

Abstract: Computational modelling is an effective technique that uses computer resources to investigate the structural information, transformations, and electronic behaviour of bulk periodic systems such as MOF and COF. Due to the recent advancements in high-performance computer systems and the development of more effective algorithms, three computational methods, namely quantum chemistry (QC), molecular dynamics (MD), and Monte Carlo (MC) methods, have been widely employed in modelling large systems containing 5000 atoms in a unit cell. Quantum chemical methods, e.g., DFT, are typically based on the accurate solution of the Schrödinger wave equation, therefore, these methods are highly accurate in computing. However, these methods are computationally costly for bulk periodic systems. The latter two are empirical force field methods, and are consequently up to five to ten orders of magnitude faster, and therefore applicable for larger MOF(s) and COF(s) (>1000 atoms). However, these empirical methods are less accurate as well as limited to certain properties because these do not account for the electronic behaviour. Due to such limitations, a novel computational methodology for large-scale screening is applied by machine learning technologies. This approach is a promising trade-off between the accuracy of ab initio methods and the speed of classical approaches.

Subject Field of Topic: Computational Chemistry

Name of Speaker: Dr. Hasnain Sajid

Professorial Rank of Speaker: Post-Doc

Email of Speaker: [email protected]

Affiliation of Speaker: University of Ottawa

Date and Venue: (Tuesday) 08 May 2025, 02:00 pm, Room # 205, New Building, School of Natural Sciences (SNS),  NUST Islamabad