Abstract
An intriguing mechanical seed (MS) concept that modulates (in)homogeneous Li metal growth is proposed based on an in-depth understanding of its fundamental mechanism using unified atomistic computations. A large dataset of thermodynamic energies for Li disordered phase decouples the dual-body interactions into three components: i) crystal-like, ii) long, and iii) short bonds of Li─Li based on machine learning assisted by density function theory calculations. The contributions of these dual-body interactions offer a mechanical factor for controlling the disordered-ordered phase transition during electrochemical deposition. Macroscopic molecular dynamics simulations systematically construct the core–shell sphere and cross-sectional models to reinforce the MS premise. The former reveals that the lower energy level of disordered phase under the moderate compression causes a slow phase kinetics, whereas the strain-free mode exhibits a relatively fast transition. In addition, the cross-sectional model exhibits a smooth surface landscape for the strain-optimized case. These observations are attributed to the surface area evolutions depending on the MS conditions and elucidate the dynamic atomic displacements near the grain boundary from a local structural perspective. The proposed mechanical design concept facilitates uniform Li growth and is expected to be a global parameter in harnessing the full potential of Li metal batteries.
| Original language | English |
|---|---|
| Article number | 2300816 |
| Journal | Advanced Energy Materials |
| Volume | 13 |
| Issue number | 34 |
| DOIs | |
| State | Published - 8 Sep 2023 |
| Externally published | Yes |
Keywords
- dendrites
- density functional theory
- Li metal anodes
- mechanical seeds
- molecular dynamics