Accurate estimation of interfacial thermal conductance between silicon and diamond enabled by a machine learning interatomic potential

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autorschaft

  • Ali Rajabpour
  • Bohayra Mortazavi
  • Pedram Mirchi
  • Julien El Hajj
  • Yangyu Guo
  • Xiaoying Zhuang
  • Samy Merabia

Organisationseinheiten

Externe Organisationen

  • Imam Khomeini International University
  • Université Claude Bernard Lyon 1
  • Harbin Institute of Technology
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Aufsatznummer109876
Seitenumfang10
FachzeitschriftInternational Journal of Thermal Sciences
Jahrgang214
Frühes Online-Datum18 März 2025
PublikationsstatusVeröffentlicht - Aug. 2025

Abstract

Thermal management at silicon-diamond interface is critical for advancing high-performance electronic and optoelectronic devices. In this study, we calculate the interfacial thermal conductance between silicon and diamond using a computationally efficient machine learning (ML) interatomic potential trained on density functional theory (DFT) data. Using non-equilibrium molecular dynamics (NEMD) simulations, we compute the interfacial thermal conductance (ITC) for various system sizes. Our results reveal an extremely close agreement with experimental data than those obtained using traditional semi-empirical potentials such as Tersoff and Brenner which overestimate ITC. In addition, we analyze the frequency-dependent heat transfer spectrum, providing insights into the contributions of different phonon modes to the interfacial thermal conductance. The ML potential accurately captures the phonon dispersion relations and lifetimes, in good agreement with DFT calculations and experimental observations. It is shown that the Tersoff potential predicts higher phonon group velocities and phonon lifetimes compared to the DFT results. Furthermore, it predicts higher interfacial bonding strength, which is consistent with higher interfacial thermal conductance as compared to the ML potential. This study highlights the use of ML interatomic potentials to improve the accuracy and computational efficiency of thermal transport simulations of complex material interface systems.

ASJC Scopus Sachgebiete

Zitieren

Accurate estimation of interfacial thermal conductance between silicon and diamond enabled by a machine learning interatomic potential. / Rajabpour, Ali; Mortazavi, Bohayra; Mirchi, Pedram et al.
in: International Journal of Thermal Sciences, Jahrgang 214, 109876, 08.2025.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Rajabpour A, Mortazavi B, Mirchi P, El Hajj J, Guo Y, Zhuang X et al. Accurate estimation of interfacial thermal conductance between silicon and diamond enabled by a machine learning interatomic potential. International Journal of Thermal Sciences. 2025 Aug;214:109876. Epub 2025 Mär 18. doi: 10.1016/j.ijthermalsci.2025.109876, 10.48550/arXiv.2407.15404
Download
@article{2d41001b4e464701bc90154a82460fa6,
title = "Accurate estimation of interfacial thermal conductance between silicon and diamond enabled by a machine learning interatomic potential",
abstract = "Thermal management at silicon-diamond interface is critical for advancing high-performance electronic and optoelectronic devices. In this study, we calculate the interfacial thermal conductance between silicon and diamond using a computationally efficient machine learning (ML) interatomic potential trained on density functional theory (DFT) data. Using non-equilibrium molecular dynamics (NEMD) simulations, we compute the interfacial thermal conductance (ITC) for various system sizes. Our results reveal an extremely close agreement with experimental data than those obtained using traditional semi-empirical potentials such as Tersoff and Brenner which overestimate ITC. In addition, we analyze the frequency-dependent heat transfer spectrum, providing insights into the contributions of different phonon modes to the interfacial thermal conductance. The ML potential accurately captures the phonon dispersion relations and lifetimes, in good agreement with DFT calculations and experimental observations. It is shown that the Tersoff potential predicts higher phonon group velocities and phonon lifetimes compared to the DFT results. Furthermore, it predicts higher interfacial bonding strength, which is consistent with higher interfacial thermal conductance as compared to the ML potential. This study highlights the use of ML interatomic potentials to improve the accuracy and computational efficiency of thermal transport simulations of complex material interface systems.",
author = "Ali Rajabpour and Bohayra Mortazavi and Pedram Mirchi and {El Hajj}, Julien and Yangyu Guo and Xiaoying Zhuang and Samy Merabia",
note = "Publisher Copyright: {\textcopyright} 2025 The Author(s)",
year = "2025",
month = aug,
doi = "10.1016/j.ijthermalsci.2025.109876",
language = "English",
volume = "214",
journal = "International Journal of Thermal Sciences",
issn = "1290-0729",
publisher = "Elsevier Masson",

}

Download

TY - JOUR

T1 - Accurate estimation of interfacial thermal conductance between silicon and diamond enabled by a machine learning interatomic potential

AU - Rajabpour, Ali

AU - Mortazavi, Bohayra

AU - Mirchi, Pedram

AU - El Hajj, Julien

AU - Guo, Yangyu

AU - Zhuang, Xiaoying

AU - Merabia, Samy

N1 - Publisher Copyright: © 2025 The Author(s)

PY - 2025/8

Y1 - 2025/8

N2 - Thermal management at silicon-diamond interface is critical for advancing high-performance electronic and optoelectronic devices. In this study, we calculate the interfacial thermal conductance between silicon and diamond using a computationally efficient machine learning (ML) interatomic potential trained on density functional theory (DFT) data. Using non-equilibrium molecular dynamics (NEMD) simulations, we compute the interfacial thermal conductance (ITC) for various system sizes. Our results reveal an extremely close agreement with experimental data than those obtained using traditional semi-empirical potentials such as Tersoff and Brenner which overestimate ITC. In addition, we analyze the frequency-dependent heat transfer spectrum, providing insights into the contributions of different phonon modes to the interfacial thermal conductance. The ML potential accurately captures the phonon dispersion relations and lifetimes, in good agreement with DFT calculations and experimental observations. It is shown that the Tersoff potential predicts higher phonon group velocities and phonon lifetimes compared to the DFT results. Furthermore, it predicts higher interfacial bonding strength, which is consistent with higher interfacial thermal conductance as compared to the ML potential. This study highlights the use of ML interatomic potentials to improve the accuracy and computational efficiency of thermal transport simulations of complex material interface systems.

AB - Thermal management at silicon-diamond interface is critical for advancing high-performance electronic and optoelectronic devices. In this study, we calculate the interfacial thermal conductance between silicon and diamond using a computationally efficient machine learning (ML) interatomic potential trained on density functional theory (DFT) data. Using non-equilibrium molecular dynamics (NEMD) simulations, we compute the interfacial thermal conductance (ITC) for various system sizes. Our results reveal an extremely close agreement with experimental data than those obtained using traditional semi-empirical potentials such as Tersoff and Brenner which overestimate ITC. In addition, we analyze the frequency-dependent heat transfer spectrum, providing insights into the contributions of different phonon modes to the interfacial thermal conductance. The ML potential accurately captures the phonon dispersion relations and lifetimes, in good agreement with DFT calculations and experimental observations. It is shown that the Tersoff potential predicts higher phonon group velocities and phonon lifetimes compared to the DFT results. Furthermore, it predicts higher interfacial bonding strength, which is consistent with higher interfacial thermal conductance as compared to the ML potential. This study highlights the use of ML interatomic potentials to improve the accuracy and computational efficiency of thermal transport simulations of complex material interface systems.

UR - http://www.scopus.com/inward/record.url?scp=105000050763&partnerID=8YFLogxK

U2 - 10.1016/j.ijthermalsci.2025.109876

DO - 10.1016/j.ijthermalsci.2025.109876

M3 - Article

AN - SCOPUS:105000050763

VL - 214

JO - International Journal of Thermal Sciences

JF - International Journal of Thermal Sciences

SN - 1290-0729

M1 - 109876

ER -

Von denselben Autoren