Nanoscale Modeling of Surface Phenomena in Aluminum Using Machine Learning Force Fields

J Chapman and R Ramprasad, JOURNAL OF PHYSICAL CHEMISTRY C, 124, 22127-22136 (2020).

DOI: 10.1021/acs.jpcc.0c05512

The study of nanoscale surface phenomena is essential in understanding the physical processes that aid in technologically relevant applications, such as catalysis, material growth, and failure nucleation. While experimental observations, such as those based on various forms of microscopy, can be used to better understand surface diffusion mechanisms, the resulting information is often limited in both its spatial and temporal resolution. Therefore, computational methodologies have become critical in the study of the processes that occur in these domains. Until recently, these methodologies have fallen into two broad categories: quantum mechanics (QM) based methods and semiempirical/classical methods. The former are computationally demanding, but accurate and versatile, while the latter are computationally inexpensive, but are significantly limited in their versatility. Machine learning (ML) methods have shown the potential to bridge these two domains by combining the cost of classical methods with the accuracy of QM. In this work, we employ recently developed ML models to simulate a variety of surface phenomena of aluminum. Adatom diffusion barriers were predicted via nudged elastic band calculations. Surface dynamics were also considered by studying the melting temperature of Al slabs and nanoparticles, along with the epitaxial growth of the Al (110) surface.

Return to Publications page