Advanced Intelligent Systems (Nov 2023)

Enhanced Long‐Term Memory Properties of ZnO Nanoparticle‐Based Optical Synaptic Devices for Neuromorphic Computing

  • Jae-Hyeok Oh,
  • Jeong-Hyeon Kim,
  • Hee-Jin Kim,
  • Young‐Hyeun Kim,
  • Kyoung-Kook Kim,
  • Sung-Nam Lee

DOI
https://doi.org/10.1002/aisy.202300350
Journal volume & issue
Vol. 5, no. 11
pp. n/a – n/a

Abstract

Read online

The enhancement of long‐term memory properties in optical synaptic devices based on ZnO nanoparticles (NPs) is investigated. High‐temperature annealing improves crystal quality and carrier mobility, leading to efficient carrier generation and transport. The annealed ZnO NPs exhibit increased band edge luminescence and reduced deep‐level emission. Their larger surface grain size decreases oxygen adsorption, resulting in enhanced desorption by photoexcited carriers during UV exposure. The annealed devices show higher excitatory postsynaptic currents (EPSCs) and slower decay rates after UV termination, indicating better long‐term memory. They also demonstrate accelerated learning processes with fewer pulse cycles required to reach 100% EPSC. Overall, this research highlights the significance of high‐temperature annealing for improving long‐term memory in ZnO NP‐based optical synaptic devices, offering insights for advanced memory devices.

Keywords