First-Principles Calculation of Thermal Transport in the Metal/Graphene System

TL;DR

This study uses first-principles phonon transport modeling to analyze how different metal/graphene interfaces affect thermal transfer, revealing that bonding strength and interfacial microstructure critically influence thermal resistance.

Contribution

It provides a detailed atomistic analysis of thermal transport in various metal/graphene interfaces using first-principles calculations and Landauer formalism, highlighting the role of bonding nature.

Findings

Chemisorbed interfaces have lower thermal resistance than physisorbed.

Pd/Gr interface exhibits unusually high thermal resistance due to mixed bonding.

Good agreement with experimental data supports the model's validity.

Abstract

Thermal properties in the metal/graphene (Gr) systems are analyzed by using an atomistic phonon transport model based on Landauer formalism and first-principles calculations. The specific structures under investigation include chemisorbed Ni(111)/Gr, physisorbed Cu(111)/Gr and Au(111)/Gr, as well as Pd(111)/Gr with intermediate characteristics. Calculated results illustrate a strong dependence of thermal transfer on the details of interfacial microstructures. In particular, it is shown that the chemisorbed case provides a generally smaller interfacial thermal resistance than the physisorbed due to the stronger bonding. However, our calculation also indicates that the weakly chemisorbed interface of Pd/Gr may be an exception, with the largest thermal resistance among the considered. Further examination of the electrostatic potential and interatomic force constants reveal that the mixed bonding force between the Pd and C atoms results in incomplete hybridization of Pd and graphene orbital states at the junction, leading effectively to two phonon interfaces and a larger than expected thermal resistance. Comparison with available experimental data shows good agreement. The result clearly suggests the feasibility of phonon engineering for thermal property optimization at the interface.