Probing Lattice Anharmonicity and Thermal Transport in Ultralow-κ Materials Using Machine Learning Interatomic...

Area of Science:

• Materials Science
• Condensed Matter Physics
• Computational Chemistry

Background:

• Ultralow lattice thermal conductivity (κ) is critical for thermoelectric energy conversion and thermal barrier coatings.
• Conventional theoretical models struggle with heat transport in strongly anharmonic materials where perturbation theory fails.
• Understanding these materials requires advanced computational methods beyond traditional frameworks.

Purpose of the Study:

• To develop and apply machine learning interatomic potentials (MLIP) for investigating thermal transport in anharmonic materials.
• To explore heat transport mechanisms in TlAgSe and Cs2PbI2Cl2, focusing on their ultralow κ properties.
• To provide a robust theoretical framework for designing materials with tailored thermal conductivity.

Main Methods:

• Development of machine learning interatomic potentials (MLIP) for accurate interatomic force calculations.
• MLIP-driven molecular dynamics (MD) simulations to analyze anharmonic lattice dynamics and structural properties.
• Application of the Green-Kubo (GK) framework via equilibrium MD to compute lattice thermal conductivity (κ).

Main Results:

• MLIP successfully modeled anharmonic lattice dynamics and finite-temperature distortions in TlAgSe and Cs2PbI2Cl2.
• Calculated κ values using the non-perturbative GK framework closely matched experimental data.
• Evidence of phonon scattering beyond the Ioffe-Regel limit and high anharmonicity (σA > 0.5) confirmed the materials' nature.

Conclusions:

• The MLIP-integrated approach provides a powerful, non-perturbative method for studying heat transport in strongly anharmonic materials.
• This framework enhances the physical understanding of thermal conductivity in materials like TlAgSe and Cs2PbI2Cl2.
• The study offers guidance for the rational design of novel materials with ultralow thermal conductivity for advanced applications.