# Thermal boundary conductance and phonon transmission in hexagonal boron nitride/graphene heterostructures

## TL;DR

This study measures and models the thermal boundary conductance at the h-BN/graphene interface, revealing the dominant phonon transmission mechanisms and improving predictive models for 2D material heterostructures.

## Contribution

It provides experimental TBC measurements and compares anisotropic diffuse mismatch models, enhancing understanding of phonon transmission in 2D heterostructures.

## Key findings

- TBC at h-BN/graphene interface is 35.1 MW/m²-K.
- Piecewise anisotropic DMM predicts phonon transmission better.
- Flexural and longitudinal acoustic phonons dominate heat transfer.

## Abstract

Increased power density in modern microelectronics has led to thermal management challenges which can cause degradation in performance and reliability. In many high-power electronic devices, the power consumption and heat removal are limited by the thermal boundary conductance (TBC) at the interfaces of dissimilar materials. Two-dimensional (2D) materials such as graphene and hexagonal boron nitride (h-BN) have attracted interest as a conductor/insulator pair in next-generation devices because of their unique physical properties; however, the thermal transport at the interfaces must be understood to accurately predict the performance of heterostructures composed of these materials. We use time-domain thermoreflectance (TDTR) to estimate the TBC at the interface of h-BN and graphene to be 35.1 MW/m2-K. We compare the phonon transmission and TBC at the h-BN/graphene interface predicted by two different formulations of the diffuse mismatch model (DMM) for anisotropic materials. The piecewise anisotropic DMM model, which uses two different phonon velocities near the center and at edge of the first Brillouin zone, results in better prediction of phonon transmission rates. The phonon transmission and temperature dependence of TBC confirms the flexural branch in ab-plane and c-plane longitudinal acoustic branch of graphene and h-BN are the dominant contributor when implementing both the A-DMM and PWA-DMM models. The methodology used here can be employed to heterostructures of other 2D materials.

---
Source: https://tomesphere.com/paper/1903.06385