Scientific Achievement
Enabled by the high accuracy and computing efficiency of machine learned interatomic potential trained on-the-fly, we performed dynamic simulation of multivalent ion diffusion in solid materials at nanosecond time scale. Our computations revealed diffusion mechanisms, where cation hopping coordinated by surrounding anions’ synchronous movements in the oxyhalide-based, novel magnesium anode coating materials.
Significance and Impact
Multivalent-ion batteries represents one of the next-generation battery designs for their high theoretical energy density, but its development is hindered by the sluggish intercalation of ions in the host lattice. Our simulation method and results contribute to atomistic understanding and rational SEI designs to the realization of high-performance multivalent batteries.
Research Details
- Molecular Dynamics at nanosecond-time scale were carried out at nearly ab initio accuracy using machine learned interatomic potential training on-the-fly that enables direct observation of Mg-ion diffusion process.
- Our simulation enhances the atomistic understanding of the diffusing species and mechanism in a Mg anode protection layers.
- The computational approach provides a viable approach to rapidly screen conducting medium for ions in multivalent-ion batteries.
