Title: Threshold switching performance in Nickle Zinc-Oxide diffusive memristor for bio inspired artificial nociceptor

Abstract

In neuromorphic computing, the memristor's biomimetic features as an electronic synapse and neuron have motivated the emergence of new information technology. A nociceptor is the essential and unique sensory neuron receptor capable of detecting harmful signals and providing a quick warning to the central nervous system to initiate a motor response in the human body and humanoid robotics. We report that the memristor-based-nociceptor characteristics for electronic receptors are illuminated in P++-Si/Ni-ZnO/Au devices. The introduction of Nikle-doped Zinc oxide (NZO) layer between a P++-Si electrode and an Au electrode is used to eliminate the surface effects of the NZO layer, resulting in improved volatile threshold switching performance. The conduction mechanism analysis shows that the NZO layer acts as a barrier between the electrodes, in which the oxygen defects create localized trap sites for electron hopping in a low electric field as well as assisting electron tunneling in a high electric field. The P++- Si/NZO/Au memristor shows several keys nociceptive functions, including threshold, relaxation, allodynia, and hyperalgesia, depending on the strength, duration, and repetition rate of the external stimuli. Such nociceptive characteristics are attributed to the electron trapping/detrapping to/from the traps in the NZO layer. NZO based metal oxide devices of this type yield multifunctional nociceptor performance that is fundamental for artificial intelligence systems applications, representing an essential step in realizing neural integrated devices with nanometer-sized features.