Abstract
Argyrodite-type compounds are renowned for their exceptional thermoelectric performance and ultralow thermal conductivity. While the latter is commonly attributed to the superionic behavior of cations, there has been limited research into how cations' static or dynamic behavior affects the thermal transport properties of argyrodites. To address and bridge this research gap, we employ a wide range of measurements and develop ab-initio based machine-learning interatomic potentials to perform large-scale molecular dynamics simulations on Ag8SiTe6 under different temperatures. We highlight the symmetry breaking and lattice-distortion scattering caused by chilled ions at low temperatures and the enhanced ionic diffusion behavior at elevated temperatures endowing argyrodites with superior superionicity and liquid-like thermal conductivity. Our findings also provide valuable insights into the ionic diffusion kinetics and the exotic lattice dynamics of liquid-like thermoelectrics.
Original language | English |
---|---|
Article number | 109324 |
Journal | Nano Energy |
Volume | 122 |
DOIs | |
State | Published - 04 2024 |
Bibliographical note
Publisher Copyright:© 2024
Keywords
- AI machine learning
- Argyrodites
- Liquid-like thermoelectric materials
- Molecular dynamics simulation
- Superionicity
- Ultralow thermal conductivity