Machine-learning discovery of extreme coherent thermal transport governed by motif-level order in Si-Ge superlattices
Abstract
Controlling phonon-mediated heat transport in multilayer nanostructures is central to thermal management and energy-conversion technologies. Yet identifying architectures exhibiting extreme lattice thermal conductivity ( κ l ) remains challenging due to the combinatorial complexity of interface arrangements and the high computational cost of transport simulations. In superlattices, thermal transport arises from competing wave effects: coherent phonon propagation enhances transport, whereas interface scattering and localization suppress it. We study Si-Ge superlattices using an iterative machine-learning-guided non-equilibrium Green’s function (NEGF) approach, where a convolutional neural network trained on NEGF data efficiently explores an experimentally constrained design space to identify both low- and high- κ l limits. Dense interface clustering induces cumulative elastic scattering and reduces κ l by up to 35%, while sparse, locally ordered motifs preserve extended phonon modes and enhance κ l by up to 33% beyond periodic designs. These results are obtained at 200 K in a regime where phonon transport is predominantly coherent and elastic; thus, the identified structural trends are representative of the coherent transport regime. Our work reveals that local motif-level order, rather than global periodicity, governs extreme thermal transport in multilayer systems.
Related
- Coherent phonon transport and Anderson localization in WS2/MoS2 superlattices
- Interplay of apparent coherent-domain size, geometrical relative density and oxygen annealing in the electrical and thermal transport of Ta-doped SnO2 ceramics
- Mesoscale hydrogen-bond network engineering controls quantum-coherent proton transport to suppress aluminum corrosion
- The Essential Transition Metals Transport Systems in Mycobacterium tuberculosis: A Scoping Review of Emerging Targets for Antitubercular Drug Discovery
- Active Learning for Fragment-Based Kinase Inhibitor Discovery using Docking
BibTeX
@article{tyagi2026machinelearning,
title = {{Machine-learning discovery of extreme coherent thermal transport governed by motif-level order in Si-Ge superlattices}},
author = {Shubham Tyagi and Julian G Fernandez and Anass Sebbar and Natalia Seoane and Antonio Garcia-Loureiro and Marc Bescond},
journal = {npj Computational Materials},
year = {2026},
month = jun,
doi = {10.1038/s41524-026-02172-0},
url = {https://hal.science/hal-05732366v1},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Machine-learning discovery of extreme coherent thermal transport governed by motif-level order in Si-Ge superlattices
AU - Shubham Tyagi
AU - Julian G Fernandez
AU - Anass Sebbar
AU - Natalia Seoane
AU - Antonio Garcia-Loureiro
AU - Marc Bescond
PY - 2026
DA - 2026/06/04
JO - npj Computational Materials
DO - 10.1038/s41524-026-02172-0
UR - https://hal.science/hal-05732366v1
AB - Controlling phonon-mediated heat transport in multilayer nanostructures is central to thermal management and energy-conversion technologies. Yet identifying architectures exhibiting extreme lattice thermal conductivity ( κ l ) remains challenging due to the combinatorial complexity of interface arrangements and the high computational cost of transport simulations. In superlattices, thermal transport arises from competing wave effects: coherent phonon propagation enhances transport, whereas interface scattering and localization suppress it. We study Si-Ge superlattices using an iterative machine-learning-guided non-equilibrium Green’s function (NEGF) approach, where a convolutional neural network trained on NEGF data efficiently explores an experimentally constrained design space to identify both low- and high- κ l limits. Dense interface clustering induces cumulative elastic scattering and reduces κ l by up to 35%, while sparse, locally ordered motifs preserve extended phonon modes and enhance κ l by up to 33% beyond periodic designs. These results are obtained at 200 K in a regime where phonon transport is predominantly coherent and elastic; thus, the identified structural trends are representative of the coherent transport regime. Our work reveals that local motif-level order, rather than global periodicity, governs extreme thermal transport in multilayer systems.
SN - 2057-3960
ER -