Robust multilevel storage characteristics of Al ₂ O ₃ /HfO ₂ /Al ₂ O ₃ trilayer-structured memristor fabricated by atomic layer deposition for neuromorphic computing

Memristors with multilevel storage capabilities have emerged as promising candidates for high-density memory and neuromorphic computing systems. In this study, a trilayer-structured memristor with an Al₂O₃/HfO₂/Al₂O₃(3/14/3 nm) dielectric stack was fabricated via atomic layer deposition, sandwiched between Ti and Pt electrodes. The analog switching characteristics of the memristor were systematically investigated through two strategies: adjusting the compliant current (<i>I</i>_cc) during the SET process and controlling the RESET-stop voltage (<i>V</i>_RESET-stop) in the RESET process. The experimental results indicate that<i>I</i>_ccprimarily modulates the values of low resistance states, whereas<i>V</i>_RESET-stopmainly influences the values of high resistance states. To validate multilevel storage feasibility,<i>I</i>_ccvalues of 0.5, 1, 2.5, and 5 mA and<i>V</i>_RESET-stopvoltages of 1.5, 1.7, 2, and 2.3 V were systematically applied. Statistical analysis demonstrated that<i>V</i>_RESET-stopmodulation yields more stable and repeatable resistance states compared to<i>I</i>_cctuning. Furthermore, the continuous resistance (or conductance) tuning capability of our fabricated memristor emulates neural network weight updates. This allows trained weights to be directly mapped to the memristor's conductance states, achieving 91.6% accuracy in handwritten digit recognition. This work underscores the significant potential of the Al₂O₃/HfO₂/Al₂O₃trilayer-structured memristor for high-performance multilevel storage and neuromorphic computing applications.