Long‐ and Short‐Term Memory Characteristics Controlled by Electrical and Optical Stimulations in InZnO‐Based Synaptic Device for Reservoir Computing

Hyogeun Park,Dongyeol Ju,Chandreswar Mahata,Andrey Emelyanov,Minsuk Koo,Sungjun Kim
DOI: https://doi.org/10.1002/aelm.202300911
IF: 6.2
2024-03-29
Advanced Electronic Materials
Abstract:In this study, synaptic characteristics of ITO/ IZO/Al2O3/TaN device are emulated by light stimulations and electric pulses. Moreover, reservoir computing is implemented by optical stimulations for temporal learning. The insertion of a thin Al2O3 layer improves the resistive switching characteristics such as cycle‐to‐cycle and device‐to‐device uniformity and reduces the power consumption compared to the ITO/IZO/TaN device. In this study, the resistive switching phenomenon and synaptic mimicry characteristics of an indium tin oxide (ITO)/indium zinc oxide (IZO)/Al2O3/TaN device are characterized. The insertion of a thin Al2O3 layer via atomic layer deposition improves the resistive switching characteristics such as cycle‐to‐cycle and device‐to‐device uniformity and reduces the power consumption of the proposed device with respect to a single‐layer ITO/IZO/TaN device. The proposed device exhibits the coexistence of volatile and nonvolatile characteristics under optical and electrical measurement conditions. Nonvolatile memory characteristics with stable retention results are used for synaptic applications by emulating potentiation, depression, and spike‐timing‐dependent plasticity. Furthermore, the device shows volatile characteristics under ultraviolet‐light illumination, emulating paired‐pulse facilitation and excitatory post‐synaptic current responses. Finally, optical‐enhanced reservoir computing is implemented based on the nonlinear and volatile nature of the IZO‐based resistive random‐access memory device.
materials science, multidisciplinary,physics, applied,nanoscience & nanotechnology
What problem does this paper attempt to address?