Synergistic Approach of Interfacial Layer Engineering and READ-Voltage Optimization in HfO2-Based FeFETs for In-Memory-Computing Applications Article Swipe
YOU?
·
· 2022
· Open Access
·
· DOI: https://doi.org/10.1021/acsaelm.2c00771
· OA: W4307720929
This article reports an improvement in the performance of the hafnium oxide-based (HfO<sub>2</sub>) ferroelectric field-effect transistors (FeFET) achieved by a synergistic approach of interfacial layer (<i>IL</i>) engineering and <i>READ</i>-voltage optimization. FeFET devices with silicon dioxide (SiO<sub>2</sub>) and silicon oxynitride (SiON) as <i>IL</i> were fabricated and characterized. Although the FeFETs with SiO<sub>2</sub> interfaces demonstrated better low-frequency characteristics compared to the FeFETs with SiON interfaces, the latter demonstrated better <i>WRITE</i> endurance and retention. Finally, the neuromorphic simulation was conducted to evaluate the performance of FeFETs with SiO<sub>2</sub> and SiON <i>IL</i> as synaptic devices. We observed that the <i>WRITE</i> endurance in both types of FeFETs was insufficient to carry out online neural network training. Therefore, we consider an inference-only operation with offline neural network training. The system-level simulation reveals that the impact of systematic degradation via retention degradation is much more significant for inference-only operation than low-frequency noise. The neural network with FeFETs based on SiON <i>IL</i> in the synaptic core shows 96% accuracy for the inference operation on the handwritten digit from the Modified National Institute of Standards and Technology (<i>MNIST</i>) data set in the presence of flicker noise and retention degradation, which is only a 2.5% deviation from the software baseline.
