A Comprehensive Compact Model for Multilevel Switching in TaOx-Based Memristive 1T-1R Cells
Filamentary switching memristive devices based on the valence change mechanism (VCM) are promising for non-volatile memory applications due to their ability to store multiple resistance states within a single device. To facilitate the integration into circuits, this study presents an advanced compact model designed for multilevel switching in the VCM devices serially connected with transistors in a 1T-1R configuration. This model is an extension of the existing JART (Jülich Aachen Resistive Switching Tools) VCM v1b model, by incorporating state-dependent effective thermal resistance (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$R_{\mathrm {th,eff}}$ </tex-math></inline-formula>) based on an electro-thermal continuum model. This enables precise modeling of multilevel behavior and includes the variability in switching cycles to reflect experimental conditions. The validation with TaOx-based VCM devices co-integrated with 180 nm n-MOS transistors demonstrates the model’s accuracy, achieving consistent multilevel programming across 7-states and capturing cycle-to-cycle variability effectively. This model offers a robust tool for designing reliable, high-density multilevel ReRAM memory system.
Authors
- Daniel Schön (ORCID: https://orcid.org/0000-0002-6967-0032)
- Ankit Bende (ORCID: https://orcid.org/0009-0008-6434-7667)
- Vikas Rana (ORCID: https://orcid.org/0000-0001-5432-0286)
- Rana Walied Ahmad (ORCID: https://orcid.org/0000-0002-0445-8110)
- Stephan Menzel (ORCID: https://orcid.org/0000-0002-4258-2673)
- Dennis Nielinger (ORCID: https://orcid.org/0000-0002-7540-7062)
- Seokki Son (ORCID: https://orcid.org/0000-0003-4478-3775)
Institutions
- X-Fab (Germany) (DE)
Publication Details
- Journal
- IEEE Access
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1109/access.2025.3590140
- Citations
- 1
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.73