The ultimate goal of my research is to develop new simulation tools to study electrolyte degradation in condensed phases at interfaces. Unfortunately, current methods are either too expensive (ab initio MD) or inaccurate (semi empirical). Therefore, I have been working on developing machine learning potentials (MLP) for molecules which offer greater accuracy while also being able to simulate larger, more relevant systems. Furthermore, I am also working on improving the large training data bottleneck faced in developing MLP by using enhanced sampling methods to speed the generation of training data and improve the reconstruction accuracy of the potentials while also limiting the total amount of training data needed. Ultimately, I plan to facilitate the study of electrolytic degradation using MLP and develop a framework for clean energy methods by studying electrolyte degradation pathways and long-scale electrolyte dynamics.
Advisor: Jim Pfaendtner – Chemical Engineering