Vorträge und Posterpräsentationen (mit Tagungsband-Eintrag):
A. Saleh, H. Zahedmanesh, H. Ceric, K. Croes, I. De Wolf:
"Dynamics of Electromigration Voids in Cu Interconnects: Investigation Using a Physics-Based Model Augmented by Neural Networks";
Vortrag: IEEE International Interconnect Technology Conference (IITC),
San Jose, USA;
27.06.2022
- 30.06.2022; in: "2022 IEEE International Interconnect Technology Conference (IITC)",
(2022),
ISBN: 978-1-6654-8646-0;
S. 22
- 27.
Kurzfassung englisch:
Physics-based numerical simulations have been widely employed for understanding electromigration (EM) induced voiding in Cu interconnects. Yet, their application has remained limited to exploratory studies given their computational cost. To achieve fast, yet accurate, void dynamics simulations, in this study, a neural network (NN) is trained to determine local current density distributions around void surfaces. Up to 85% reduction of computational time was achieved by replacing the finite-element (FE) solver with the NN. The model was used to investigate the impact of interconnect linewidth, line and via aspect ratio and microstructure on void dynamics.
"Offizielle" elektronische Version der Publikation (entsprechend ihrem Digital Object Identifier - DOI)
http://dx.doi.org/10.1109/IITC52079.2022.9881303
Erstellt aus der Publikationsdatenbank der Technischen Universität Wien.