Thermal properties of single-layer MoS₂-WS₂ alloys enabled by machine-learned interatomic potentials

Area of Science:

• Materials Science
• Condensed Matter Physics
• Computational Chemistry

Background:

• Two-dimensional (2D) quantum materials, such as transition metal dichalcogenides (TMDs), offer transformative potential for electronics and chemical sciences.
• Understanding thermal transport in these materials is crucial for applications in heat management, energy storage, and conversion.

Purpose of the Study:

• To develop and validate a computationally efficient method for modeling thermal transport in single-layer (1L) and multi-layer TMDs, including MoS2, WS2, and their alloys.
• To demonstrate the synergy between density functional theory (DFT) and algorithmic training for generating accurate inter-atomic potentials.

Main Methods:

• Utilized density functional theory (DFT) calculations to generate training data.
• Employed algorithmic training to create a moment tensor potential (MTP) for modeling 1L-MoS2, 1L-WS2, and their alloys.
• Validated the MTP by comparing its predictions of vibrational properties and thermal conductivities against DFT calculations and the Stillinger-Weber (SW) potential.

Main Results:

• The trained MTP accurately describes the vibrational properties and thermal conductivities of 1L-MoS2 and 1L-WS2.
• The MTP shows consistent agreement with established DFT calculations and the SW potential.
• Investigated the impact of sulfur vacancies on thermal conductivity in 2D TMD alloys, finding minimal effect.

Conclusions:

• The combined DFT and MTP approach provides a powerful tool for studying thermal transport in 2D quantum materials.
• The developed inter-atomic potentials are suitable for high-performance computing simulations of quantum materials under thermal stress.
• The insensitivity of thermal conductivity to sulfur vacancies in 2D alloys suggests potential for fine-tuning material properties for specific applications.