New atomistic framework for exploring phonon-structure interactions
Date
2020
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
Understanding the interaction of phonons with nanostructures is critical to engineering materials with favorable thermal properties. In the same way that more familiar waves, such as acoustic waves and light (photons), interact with imperfections in their environment, phonons are scattered as they transport throughout solids with imperfections. Common imperfections include material interfaces, alloy impurities, or nanoparticles. By carefully engineering these nanostructures we can control the flow of phonons inside materials, thereby controlling their thermal conductivity. The ability to understand and manipulate thermal transport inside solids could, for example, enable brighter quantum cascade lasers, higher-power/higher-frequency switches, thermoelectric materials that efficiently convert heat to electricity without moving parts, faster and lower-power phase-change memory alloys, and ultra-high capacity hard disks relying on heat-assisted magnetic recording heads. However, there remain several outstanding questions regarding thermal transport in nanostructured materials, in part because the methods for modelling phonon interaction with intentionally placed nanostructures are computationally expensive. ☐ To resolve this, I've developed a new computational method: the Frequency Domain - Perfectly Matched Layer (FD-PML) method. The FD-PML method approaches the phonon-nanostructure scattering problem using a frequency-domain decomposition of the atomistic equations of motion and the use of perfectly matched layer (PML) boundaries. The use of PML boundaries enables rapid absorption of scattered wave energies at the boundaries and provides a simple interpretation of the scattered phonon energy flux as the energy dissipation rate in the PML, which can be inexpensively computed. We describe the optimal design of the PML boundary parameters and then use the approach to explore new physical effects associated with phonon scattering from embedded nanoparticles and across interfaces. ☐ For example, by investigating phonon interaction with interdiffused interfaces we have been able to show that the Diffuse Mismatch Model (DMM) is fundamentally inconsistent with its own assumptions for extremely high levels of interdiffusion. However, this study also suggest that some of the basic physics is correct for sub-monolayer levels of interdiffusion: interdiffused interface do open additional transmission pathways that are otherwise forbidden by scattering selection rules, a phenomenon accurately captured by the DMM but not by specular models like AMM. ☐ With regards to nanoparticles, we compare the atomistic predictions of FDPML against continuum mechanics approach, where we show that the simulation results are in reasonable agreement with with analytical continuum formulations that consider Mie regime scattering accurately. However, in-case of scattering of optical phonons we observe that the long wavelength optical phonons depict a geometric scattering behaviour. A behaviour we attribute to optical phonons being analytical continuation of acoustic phonons. Finally, we show that the an interdiffused nanoparticle is a more effective scatterer for mid-high frequency phonons compared to a solid nanoparticle. ☐ In chapter 5, we extend the capabilities of FDPML to simulate large domain sizes by using massively parallel processing techniques. The approach here is to efficiently parallelize the FDPML solution procedure to achieve a linear speed-up with the number of processors used. In reality, using our approach we see a slightly less (~15%) than the ideal behaviour. Lastly in chapter 6, we present a theoretical framework for obtaining nanoparticle scattering rates from directionally dependent scattering properties, i.e. transition probabilities, which to our surprise had never been addressed from the point of view of phonon transport.
Description
Keywords
Atomistic Simulations, Interfacial Thermal Conductance, Microscale Thermal Trasport, Phonon Nanoparticle Scattering