Researchers Develop Efficient Approach to Predict Transport Properties in Complex Materials

Researchers, including Zheyong Fan, Yang Xiao, Yanzhou Wang, Penghua Ying, Shunda Chen, and Haikuan Dong, have developed an efficient method for predicting thermal and electronic transport properties in complex materials.

The approach combines linear-scaling quantum transport and machine-learning molecular dynamics, using a machine-learned neuroevolution potential (NEP) trained with quantum-mechanical density-functional theory calculations. The method accounts for electron-phonon scattering and other disorders, providing a more accurate characterization of transport properties. This could have significant implications for the development of new materials and technologies. The research has been accepted for publication in the Journal of Physics Condensed Matter.

What is the New Approach to Predicting Thermal and Electronic Transport Properties in Complex Materials?

A team of researchers, including Zheyong Fan, Yang Xiao, Yanzhou Wang, Penghua Ying, Shunda Chen, and Haikuan Dong, has proposed an efficient approach for the simultaneous prediction of thermal and electronic transport properties in complex materials. This approach combines linear-scaling quantum transport and machine learning molecular dynamics. The team’s research has been accepted for publication in the Journal of Physics Condensed Matter.

The researchers’ approach begins with training a highly efficient machine-learned neuroevolution potential (NEP) using reference data from quantum-mechanical density-functional theory calculations. This trained potential is then applied in large-scale molecular dynamics simulations, enabling the generation of realistic structures and accurate characterization of thermal transport properties.

How Does the Approach Account for Electron-Phonon Scattering and Other Disorders?

In addition to the use of the NEP in molecular dynamics simulations, the researchers’ approach also involves coupling molecular dynamics simulations of atoms and linear-scaling quantum transport calculations of electrons. This coupling is done to account for the electron-phonon scattering and other disorders that affect the charge carriers governing the electronic transport properties.

The researchers demonstrated the usefulness of this unified approach by studying electronic transport in pristine graphene and thermoelectric transport properties of a graphene antidot lattice. They used a general-purpose NEP developed for carbon systems based on an extensive dataset.

What are the Limitations of Existing Computational Methods for Studying Transport Properties?

Thermal and electronic transports are two fundamental properties of a material. For simple solids, computational methods based on the electron and phonon Boltzmann transport equations have been widely used to compute the transport properties mediated by the heat and charge carriers. However, these methods can only efficiently deal with relatively simple systems and are generally not applicable to complex systems that cannot be properly represented by small periodic supercells.

How Does the New Approach Overcome These Limitations?

To efficiently compute transport properties in complex systems, one must resort to linear-scaling methods, i.e., methods with the computational cost that scales linearly with respect to the number of atoms in the periodic supercell. The researchers’ approach uses molecular dynamics (MD) simulation, a linear-scaling method at the atomistic level, provided that the interatomic potential used is a classical one and has a finite cutoff.

Machine-learned potentials (MLPs) have been routinely applied in MD simulations of thermal transport. Particularly, the neuroevolution potential (NEP) has been developed with a focus on thermal transport applications and has excellent computational efficiency.

What are the Implications of the New Approach?

The researchers’ approach, which combines linear-scaling quantum transport and machine learning molecular dynamics, offers a promising way to study thermal and electronic transports in complex materials. By accounting for electron-phonon scattering and other disorders, the approach provides a more accurate characterization of these transport properties. This could have significant implications for the development of new materials and technologies.