Faculty Sponsor: Francis W. Starr
Casey Gao
Casey Gao is a rising junior (’27) studying Mathematics with a minor in Physics, doing research at Wesleyan in a Computational Physics Group. Over the past year, Casey has been studying the phase change material Ge2Sb2Te5 (GST), and modelling its unique behaviours. At Wesleyan, he also works as a Math Course Assistant, a Teaching Fellow for the National Education Equity Lab, and for Sound Co-op. Casey is from Toronto, Ontario, and enjoys to spend as much time doing anything outdoors.
Abstract: Ge₂Sb₂Te₅ (GST) is a phase change material known for its ability to rapidly switch between its amorphous and crystalline states. Both states are stable and exhibit distinct optical and electronic properties, making them key candidates for non-volatile, high-speed data memory applications. GST typically forms a polycrystalline structure, made of multiple grains with varying orientations and separated with grain boundaries. These factors can influence the properties of the material, making it crucial to investigate.
Recent advancements in new machine-learned (ML) potentials have enabled computer simulations of large-scale polycrystalline systems and studies of its structure. The ML potentials can accurately model these materials, while being at a fraction of the computational cost of previous methods. In this work, we employ a recently developed ML Atomic Cluster Expansion (ACE) potential to model 10 080-atom GST polycrystalline systems. We investigate the effects of the annealing temperature while preparing the crystal in the resulting crystal grain sizes.
Furthermore, existing literature has shown pressure-dependant phase transitions. To successfully model GST at these high pressures, a new ML potential was trained and successfully models the high-density amorphous and body-centered cubic crystal states. These transitions were quantified using its crystallinity and structure factor. This development unlocks further studies in the crystallization kinetics which govern these phase transitions.