Session: 11-05-01: Thermophysical Properties: Characterization and Modeling Across Scales

Paper Number: 150663

150663 - Universality of Interfacial Thermal Conductance via Interfacial Phonon Localization 

For over 80 years and ever since its discovery by Kapitza, the interfacial thermal resistance has been explained by phonon scattering at the plane in which they encounter a change in the medium. The well adopted acoustic mismatch and diffuse mismatch models (AMM and DMM) succeed only partially for a particular set of circumstances (interface type, temperature range, etc..) and fail otherwise. The failure is exemplified when inelastic phonon scattering starts to contribute to thermal transport in highly mismatched surfaces. Several modifications have been attempted in the literature to expand on these benchmark models, all of which maintained the picture of phonon reflection at a given plane as the origin for the interfacial thermal resistance. The above treatments, however, dismiss the fact that in most cases for solid-solid interfaces, the change in the medium takes place over a finite thickness region rather than a sharp transition. Even for atomically sharp ones, as in exfoliated 2D materials supported on a substrate, the atomic interactions at the interface are completely different from those of both adjoining sides. This suggests the possibility that thermal transport is mediated via a new type of phonons that are extremely localized within the interface region which takes place over a finite domain. A great analogy to distinguish both arguments can be drawn from surface chemistry when comparing the Gibbs dividing surface and the Guggenheim interfacial layer. Both models are meant to treat interfacial phenomena in multiphase systems. The former treats the interface as a mathematical dividing plane with zero thickness that separates the two mediums. The latter, however, assumes that there exists an interfacial layer of finite thickness with unique and distinct properties best known as the Guggenheim interfacial layer. Here, we propose a universal physical model to treat the interfacial thermal conductance (G) that takes into account a finite-thickness region termed as the equivalent interfacial medium (EIM) by which the heat is mediated. The interfacial thermal resistance that is conventionally explained by phonon scattering is now attributed to the conduction resistance within the EIM over a finite thickness. The EIM is shown to exhibit unique and distinct properties that are governed by what we term the interface characteristic temperature (Θ_int), which is found to be distinct from the Debye temperature of the adjoining materials. Due to the anticipated irregularities within the EIM, since it is a mixture of two different materials, the heat conduction is treated as in disordered solids where the main mechanism for thermal transport is via localized oscillators as first envisioned by Einstein over a 100 years ago. The proposed model fits the reported literature data for G with remarkable accuracy across all temperature range and regardless of the interface type by making use of only two fitting parameters, each of which carries significant physical meaning: the interface characteristic temperature and the energy carrier transfer time which indicates how fast the energy can transmit across the EIM. The work further provides a highly confident prediction of G at different temperatures from extremely limited data with as few as two experimental data points. We demonstrate that this prediction is within the uncertainty of any experimental technique, which will help facilitating and validating future measurements. Moreover, the model provides a quantitative prediction for the maximum G an interface can sustain termed as G_max. Very strikingly, under normalized temperature (T/ Θ_int) and interfacial thermal conductance (G/G_max), all literature data for the evolution of G with temperature can be universally grouped in a single curve. This temperature dependency resembles that of the heat capacity of solids under the Debye model, which further fortifies the proposed concepts (Θ_int - G_max) and their relevance to interfacial thermal transport. This work sheds new light on a century-old open challenge and is expected to attract broad interests for scientific understanding and engineering of interfaces for material design, process control, and micro/nanoelectronics development.

Presenting Author: Ibrahim Al Keyyam Iowa State University

Presenting Author Biography: A second year PhD student majoring in Mechanical Engineering with dual minors in Physics and Materials Science at Iowa State University.

Authors:

Ibrahim Al Keyyam Iowa State University
Xinwei Wang Iowa State University