Dresden 2026 – wissenschaftliches Programm
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
FM: Fachverband Funktionsmaterialien
FM 11: Poster Session Functional Materials
FM 11.10: Poster
Dienstag, 10. März 2026, 18:00–20:30, P4
Structural Factors Governing Stability and Electronic Transport in Future Neuromorphic Memristor Technologies — •Wiktoria Szopa1,2,3, Marcin Krynski1, Attila Cangi2,3, and Piotr Wisniewski4 — 1Faculty of Physics, Warsaw University of Technology, 00-662 Warsaw, Poland — 2Center for Advanced Systems Understanding, 02826 Görlitz, Germany — 3Helmholtz-Zentrum Dresden-Rossendorf, 01328 Dresden, Germany — 4Centre for Advanced Materials and Technologies CEZAMAT, Warsaw University of Technology, 02-822 Warsaw, Poland
With the rapid rise of artificial intelligence, the demand for technologies that enable its development and application is high, as AI is now widely used across science, industry, and everyday life. Neuromorphic materials may offer a promising solution to the increasingly complex hardware requirements posed by next-generation, high-throughput AI.
Silicon-oxide-based memristors may operate as adjustable weights within hardware neural networks, enabling more efficient, neuromorphic computation. The electronic component contains two electrodes separated by a thin amorphous silicon oxide layer, within which charge transport occurs through filamentary silicon chains that form conductive pathways.
In this project, we investigate how the local atomic structure influences electronic transport. To this end, we generated hundreds of memristor models and identified their most stable configurations, with high structural robustness and defined conduction pathways by using a combination of DFT calculations and machine learning modeling.
Keywords: neuromorphic materials; memristors; electronic transport; density functional theory; machine learning modeling