Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
HL: Fachverband Halbleiterphysik
HL 51: Transport Properties (joint session HL/TT)
HL 51.2: Vortrag
Donnerstag, 20. März 2025, 15:15–15:30, H13
Influence of defects and shape of thin InAs nanowires on their thermal conductivity, assessed via machine-learning potentials — •Sandro Wieser1, YuJie Cen1, Georg K. H. Madsen1, and Jesús Carrete2 — 1Institute of Materials Chemistry, TU Wien, Wien, Austria — 2Instituto de Nanociencia y Materiales de Aragón (INMA), CSIC-Universidad de Zaragoza, Zaragoza, Spain
Nanowires (NWs) grown from the zincblende (ZB) phase of InAs in the (111) direction commonly contain twin boundary defects consisting of narrow wurtzite (WZ) (001) phase regions between ZB sections. To investigate the impact of these and other defects on heat transport, we employ Green-Kubo equilibrium molecular dynamics simulations utilizing cepstral analysis to efficiently process the noise, and an accurate MACE model trained via active learning strategies to achieve transferability for a wide range of surface conditions.
We show that these twin boundaries reduce the thermal conductivity with respect to that of defect-free WZ-phase (001) NWs by a factor of more than two and that surface conditions lead to lower thermal conductivity values for defect-free ultrathin InAs ZB NWs. Analysis of the shape of twinning NWs reveals that structures mimicking experimentally measured surface configurations can enhance heat transport compared to strictly hexagonal NWs. Additional insights are gained from an analysis of line-group symmetries and vibrational properties for various NW shapes. Furthermore, experimentally motivated symmetric and symmetry-breaking surface defects are studied to reveal more and less influential defect sites.
Keywords: machine learning; nanowires; semiconductors; thermal transport; defects