Proton Conductor Gated Synaptic Transistor Based on Transparent IGZO for Realizing Electrical and UV Light Stimulus

Abstract

Synaptic transistors mimicking the biological synapse's short term plasticity and short-term memory property were demonstrated using the amorphous indium-gallium-zinc oxide channel in combination with the nanogranular SiO 2 as the gate oxide. The lowest energy consumption was ∼1.08 pJ per pulse activity and the operating voltage was within 100 mV. The device's plasticity and memory characteristics can be explained by the movement of protons in the insulating layer. The proton relaxation was revealed by two ways of dual sweeping: continuous and discontinuous sweepings. We observed that the excitatory postsynaptic current (EPSC) rose as the voltage decreased anomaly during the backward sweeping process. In the electrical stimulus, both the short-term potentiation and depression were observed for this proposed device. The amplitude of the EPSC changed with the pulse number following a saturating exponential function. For the electrical stimulus under constant illumination, the UV light wavelength, intensity and duration time were found to have little effect on the paired pulse facilitation. While in the light stimulus, the light frequency promoted the paired pulse facilitation and had more effect on the synapse's plasticity than the other light pulse parameters including intensity, numbers and width. INDEX TERMS Synaptic transistor, transparent oxide, IGZO, nanogranular SiO 2 , UV light.

Bio‐Inspired Photoelectric Artificial Synapse based on Two‐Dimensional Ti3C2Tx MXenes Floating Gate

“…c) The minimum energy consumption of the synaptic devices reported by previous studies and this work for the past 6 years. [13,21,43,53,[56][57][58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75] d) Emulation of classical conditioning learning for MXFGT, for which the UV and the electrical pulses stimuli are used as food and bell signals for conditional learning, respectively. e) The process of the supervised learning task completed through five MXFGT devices, which include training, forgetting and recognition periods.…”

“…The EPSC results of the MXFGT under (a) electrical pulses stimulus and (b)UV pulses stimulus at low-energy-consumption working condition. c) The minimum energy consumption of the synaptic devices reported by previous studies and this work for the past 6 years [13,21,43,53,[56][57][58][59][60][61][62][63][64][65][66][67][68][69][70][71][72][73][74][75]. d) Emulation of classical conditioning learning for MXFGT, for which the UV and the electrical pulses stimuli are used as food and bell signals for conditional learning, respectively.…”

Recent progress on optoelectronic synaptic devices

“…氧化物半导体是制备光电神经突触器件的重要选择之一, 这类器件的突触特性主要来源于氧化物 半导体中普遍存在的持续光电导 (persistent photoconductivity, PPC) 及其相关弛豫现象 [45,93,102,103] . 尽管目前关于氧化物半导体中光生载流子的产生及复合机制没有统一的认识, 但是主流的看法是由氧 空位 (V O ) 在光刺激下离化而带正电, 光刺激结束后发生缓慢的去离化过程所致 [104,105] .…”

Red‐Light‐Stimulated Photonic Synapses Based on Nonvolatile Perovskite‐Based Photomemory

“…[3,18,19] More importantly, limited by the large bandgap of photoactive material in the artificial photonic synapse, ultraviolet light, or blue light is typically the main light source to trigger the synaptic functions. [1,11,20] Therefore, exploration of red-light recordable device to emulate the permanent memory behavior in brain is essential to further reduce the energy consumption in neuromorphic computing. However, the photonic synapses stimulated by red-light or even near-infrared (NIR) region are still limited now.…”