Characterization and modeling of the electrical nociceptive properties of HfO2/TiOx/SiO2-based memristive devices

Abstract

This work explores the potential of Co/HfO2/TiOx/SiO2/Si-based memristive devices for mimicking sensory systems, leveraging the electrical nociceptive properties that arise from their resistive switching characteristics. We demonstrate tunable ON/OFF threshold voltages depending on the maximum current flowing through the device, enabling volatile and non-volatile memory states achieved using appropriate biasing conditions. To provide deeper insight into the observed phenomenology, a compact behavioral model driven by the memristor's internal state control variable is proposed. The model accurately reproduces the threshold voltage variation as a function of the applied current compliance. The versatility of the proposed approach is verified as it can handle both filamentary and electron-trapping-based devices. Remarkably, the volatile states exhibit frequency-dependent transient responses and sensitization to previous stimulus voltages. These findings pave the way for beyond-memory applications of memristors, including the development of artificial sensory systems.