| Abstract Scope |
Ionic transport is of fundamental importance in batteries as both electrolytes and electrodes conduct ions. Atomistic simulations, complementary to experimental effort, provide atomic-scale visualization/quantification of ionic transport. Machine learning interatomic potentials (MLIPs), combining the accuracy of density-functional theory and the efficiency of classical interatomic potentials, have emerged as a new generation of potentials for materials research. In this talk, I will show our studies of applying MLIPs to study the ionic transport in lithium garnet oxides (lithium-ion solid electrolyte) and sodium nickel titanates (sodium-ion electrode). |