Abstract
This study proposes a method to optimize the charge transfer mechanism in synaptic devices, a practical core component of neuromorphic systems, by employing bandgap engineering (BE) with high-k materials. The conventional Al2O3 blocking oxide was substituted with a single HfO2 layer and a stacked HfO2/Al2O3 structure to enhance the program and erase characteristics. Simulation results indicate that the structures utilizing high- k materials demonstrated a larger threshold voltage (Vth) shift during program operations compared with the conventional structure. This improvement is attributed to the increased electron acceleration and the reduction in equivalent oxide thickness (EOT) due to the high permittivity of high-k materials. Moreover, a greater Vth shift was documented during erase operations, which is explained by the band offset between the blocking oxide and the nitride trap layer. Consequently, the BE charge-trap flash device demonstrated an enhancement of 2.22 V in the memory window compared with device in the conventional structure.
| Original language | English |
|---|---|
| Pages (from-to) | 1098-1102 |
| Number of pages | 5 |
| Journal | IEEE Journal of the Electron Devices Society |
| Volume | 13 |
| DOIs | |
| State | Published - 2025 |
Bibliographical note
Publisher Copyright:© 2013 IEEE.
Keywords
- Bandgap engineering
- high-k materials
- low power operation
- neuromorphic computing
Fingerprint
Dive into the research topics of 'Bandgap-Engineered Side-Path Synaptic Device Utilizing High-κ Materials for Low-Power Operation'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver