(Summer 2022) Molecular Dynamics Simulations with Machine Learning Potential for Doped Lithium Garnet-type Oxides

Posted on: 04/13/2022

Contact: Hui Zheng (huizheng@lbl.gov)

Description: Solid state batteries have the potential to be safer and more energy dense than their traditional counterparts that use liquid electrolytes. Oxide-based solid electrolytes such as lithium garnet-type oxides have been popular to study due to their potentially high ionic conductivity and chemical stability with Li metal, but the conductivity has yet to be optimized using these materials. As a result, dopants have been considered to promote Li-ion conductivity. The student would use VASP to train machine learning potentials for lithium garnet-type oxides with different dopants to understand their effect on Li conductivity.

Qualifications: UC Berkeley undergraduate student with a background in Materials Science and comfortable with coding in Python.