*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
THE SCHOOL OF MATERIALS SCIENCE AND ENGINEERING
GEORGIA INSTITUTE OF TECHNOLOGY
Under the provisions of the regulations for the degree
DOCTOR OF PHILOSOPHY
on Thursday, August 8, 2019
12:00 pm
in Love 295
will be held the
DISSERTATION PROPOSAL DEFENSE
for
James Chapman
"Accelerating Quantum-Accurate Atomic-level Materials Simulations with Machine Learning"
Committee Members:
Prof. Rampi Ramprasad, Advisor, MSE
Prof. Seung Soon Jang, MSE
Prof. Andrew Medford, ChBE
Prof. Chaitanya Deo, MSE/NRE
Prof. Le Song, CSE
Abstract:
Materials properties such as defect diffusion along surfaces, mechanical breakdown under dynamic conditions, and phase transformations under extreme temperatures and pressures, are governed by the subtle interactions at the atomic level under a plethora of unique environments. Computational tools have been instrumental in understanding the atomistic properties at these length scales. Over the past few decades, these tools have been dominated by two levels of theory: quantum mechanics (QM) based methods and semi-empirical/classical methods. The former are time-intensive, but accurate and versatile, while the latter methods are fast but are significantly limited in veracity, versatility and transferability. ML algorithms, in tandem with quantum mechanical methods such as density functional theory, have the potential to bridge the gap between these two chasms due to their (i) low cost, (ii) accuracy, (iii) transferability, and (iv) ability to be iteratively improved. In this work, we prescribe a new workflow for an emulation platform in which atomic forces, potential energy, stresses, and subsequently electronic structure, are rapidly predicted by independent machine learning models, all while retaining the accuracy of quantum mechanics. This platform has been used to study thermal, vibrational, and diffusive properties of a variety of elemental metals, highlighting the framework's ability to reliably predict materials properties under dynamic conditions. Further work is proposed to explore the capability of the ML framework to accurately model more complex phenomena such as crystal growth, mechanical failure, and the prediction of phase transformations under extreme conditions.
