TY - JOUR
T1 - Simulation of Phonon Transport in Semiconductors Using a Population-Dependent Many-Body Cellular Monte Carlo Approach
AU - Sabatti, Flavio F.M.
AU - Goodnick, Stephen
AU - Saraniti, Marco
N1 - Funding Information:
This work has been supported by the Defense Advanced Research Projects Agency Grant Nos. FA8650-10-1-7045 and FA8650-15-1-7525. We are grateful to Dr. John Albrecht, Dr. Avram Bar-Cohen, and Dr. Daniel Green for their contribution in managing the research program from DARPA at different times.
Publisher Copyright:
Copyright © 2017 by ASME.
Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 2017/3/1
Y1 - 2017/3/1
N2 - A Monte Carlo rejection technique for numerically solving the complete, nonlinear phonon Boltzmann transport equation (BTE) is presented in this work, including three particles interactions. The technique has been developed to explicitly model populationdependent scattering within a full-band cellular Monte Carlo (CMC) framework, to simulate phonon transport in semiconductors, while ensuring conservation of energy and momentum for each scattering event within gridding error. The scattering algorithm directly solves the many-body problem accounting for the instantaneous distribution of the phonons. Our general approach is capable of simulating any nonequilibrium phase space distribution of phonons using the full phonon dispersion without the need of approximations used in previous Monte Carlo simulations. In particular, no assumptions are made on the dominant modes responsible for anharmonic decay, while normal and umklapp scattering are treated on the same footing. In this work, we discuss details of the algorithmic implementation of both the three-particle scattering for the treatment of the anharmonic interactions between phonons, as well as treating isotope and impurity scattering within the same framework. The simulation code was validated by comparison with both analytical and experimental results; in particular, the simulation results show close agreement with a wide range of experimental data such as thermal conductivity as function of the isotopic composition, the temperature, and the thin-film thickness.
AB - A Monte Carlo rejection technique for numerically solving the complete, nonlinear phonon Boltzmann transport equation (BTE) is presented in this work, including three particles interactions. The technique has been developed to explicitly model populationdependent scattering within a full-band cellular Monte Carlo (CMC) framework, to simulate phonon transport in semiconductors, while ensuring conservation of energy and momentum for each scattering event within gridding error. The scattering algorithm directly solves the many-body problem accounting for the instantaneous distribution of the phonons. Our general approach is capable of simulating any nonequilibrium phase space distribution of phonons using the full phonon dispersion without the need of approximations used in previous Monte Carlo simulations. In particular, no assumptions are made on the dominant modes responsible for anharmonic decay, while normal and umklapp scattering are treated on the same footing. In this work, we discuss details of the algorithmic implementation of both the three-particle scattering for the treatment of the anharmonic interactions between phonons, as well as treating isotope and impurity scattering within the same framework. The simulation code was validated by comparison with both analytical and experimental results; in particular, the simulation results show close agreement with a wide range of experimental data such as thermal conductivity as function of the isotopic composition, the temperature, and the thin-film thickness.
UR - http://www.scopus.com/inward/record.url?scp=85026890981&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85026890981&partnerID=8YFLogxK
U2 - 10.1115/1.4035042
DO - 10.1115/1.4035042
M3 - Article
AN - SCOPUS:85026890981
SN - 0022-1481
VL - 139
JO - Journal of Heat Transfer
JF - Journal of Heat Transfer
IS - 3
M1 - 032002
ER -