Ab-initio heat transport in defect-laden quasi-1D systems from a symmetry-adapted perspective

Abstract

Due to their aspect ratio and wide range of thermal conductivities, nanotubes hold significant promise as heat-management nanocomponents. Their practical use is, however, often limited by thermal resistance introduced by structural defects or material interfaces. An intriguing question is the role that structural symmetry plays in thermal transport through those defect-laden sections. To address this, we develop a framework that combines representation theory with the mode-resolved Green's function method, enabling a detailed, symmetry-resolved analysis of phonon transmission through defected segments of quasi-1D systems. To avoid artifacts inherent to formalisms developed for bulk 3D systems, we base our analysis on line groups, the appropriate description of the symmetries of quasi-1D structures. This categorization introduces additional quantum numbers that partition the phonon branches into smaller, symmetry-distinct subsets, enabling clearer mode classification. We employ an Allegro-based machine learning potential to obtain the force constants and phonons with near-ab-initio accuracy. We calculate detailed phonon transmission profiles for single- and multi-layer MoS$_\mathrm{2}$-WS$_\mathrm{2}$ nanotubes and connect the transmission probability of each mode to structural symmetry. Surprisingly, we find that pronounced symmetry breaking can suppress scattering by relaxing selection rules and opening additional transmission channels. Molecular dynamics shows that the behavior persists even when anharmonicity is considered. The fact that higher disorder introduced through defects can enhance thermal transport, and not just suppress it, demonstrates the critical role of symmetry in deciphering the nuances of nanoscale thermal transport.