A Symmetry-Based Approach to Minimum Energy Pathways

Non-technical description: Symmetry is a powerful tool in all of physical sciences. Distortions, often a collection of atomic trajectories, are important for understanding a range of kinetic materials processes where atoms move under external stimuli such as fields, temperature, and stress. Predicting which path a material would take in undergoing a distortion is important for any application which involves such atomic motion. However, predicting such paths, called minimum energy pathways (MEPs), is currently stochastic in nature, and one can never be sure of the final results. This project aims at developing a new symmetry-based framework and computational tools through the introduction of new ideas such as distortion-reversal symmetry and distortion groups; it significantly improves the speed, numerical accuracy, and the comprehensive nature of search for such paths. Experimental verification of the theory predictions is the other aspect of this research. The growing need for educational training in computation materials research is being addressed with hands-on workshops in density-functional theory, active mentoring of undergraduates in research including women and underrepresented groups, and through active participation in STEM Academy of the Upward Bound Math and Science Program, as well as Penn State Millennium Scholars Program that attracts and guides talented STEM students to the undergraduate program.

Technical description: This is an integrated theory-experimental proposal that develops new symmetry ideas and computational tools to explore minimum energy pathways (MEPs) for distortions, and more specifically, for ferroelectric domain switching and domain wall motion. The MEPs are computed by the Nudged Elastic Band method (NEB); however, it currently relies on a combination of stochastic process and user intuition to construct initial paths. The principal investigators are focused on the concept of the symmetry group of a path and are develop a code (to be made publicly available when ready) to implement these ideas in Quantum Espresso, an open-source density functional theory (DFT) software. Three model system classes are picked, namely, in simple perovskites, in layered oxides, and in multiferroics, to make predictions. The predicted MEPs with this approach are tested using the experimental techniques that push in-situ dynamics with 2-5 picometers metrology in electron microscopy, and ultrafast electron diffraction microscopy techniques.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.