Four-phonon scattering significantly reduces the predicted lattice thermal conductivity in penta-graphene: A machine learning-assisted investigation

Yifan Wang,Wenjie Huang,Junwei Che,Xuezhi Wang
DOI: https://doi.org/10.1016/j.commatsci.2023.112435
IF: 3.572
2023-08-15
Computational Materials Science
Abstract:Embellish the following academic abstract: Penta-graphene (PG) has recently been characterized as a novel two-dimensional carbon allotrope, exhibiting many unique optoelectronic properties. In this paper, with the assistance of machine learning potential, we meticulously study the phonon thermal conductivity ( κ ) by solving the state-of-the-art Boltzmann transport equation including both three- and four-phonon scattering. The intrinsic κ of PG is found to be about 298–45 Wm −1 K −1 at 300–900 K, which is much lower than that of graphene. A pivotal revelation arises from our work, indicating the profound influence exerted by four-phonon scattering in mitigating the intrinsic κ of PG. Consequently, our findings underscore the necessity of including the four-phonon scattering mechanism to ensure accurate predictions of κ for PG. Moreover, it is found that the contribution of diffusive phonon modes to the total κ can be neglected despite a strong anharmonicity of PG. Overall, this study provides new insight into the phonon thermal transport in PG, which could be useful for the regulation of thermal properties in PG.
materials science, multidisciplinary
What problem does this paper attempt to address?