Study Of Ion Gel-based Electric-Double-Layer Photoelectric Synaptic Transistor

Current computer system architecture based on the Von Neumann structure suffers severe challenges.Especially,its physical separation between the storage and the computing could cause to slow speed and high energy consumption,which make people seek other solutions to meet the massive data processing needs in the information field.Biologically,the development of neuroscience in the brain has inspired researchers to develop synaptic-based neuromorphic computing network systems for simulating the workings of the human brain to process and store complex information.It is used to meet the future development of the artificial intelligence and information field.In this paper,flexible synaptic devices with neuro-synaptic morphological functions had been investigated,which were composed of ion gel as a gate medium,and graphene and perovskite quantum dot/graphene heterojunction respectively as channel materials.Two optoelectronic synaptic devices had been fabricated,and their important synaptic plasticity had been mimicked.It provides a feasible scheme for flexible sensing and memory integrated neuromorphic computation system.The main tasks of this thesis are as follows:1、In this paper,we designed and prepared an electric-double-layer photoelectric synaptic transistor based on ion gel and graphene.Using the in-plane tunable carrier transport property of graphene and the high anioncation motion property of ion gel,various synaptic functions under electrical stimulation had been successfully mimicked under flat or curved(60°)conditions,including excitation/inhibition of postsynaptic currents,paired pulse facilitation,and long-time plasticity,etc.The effects of gate voltage amplitude,voltage polarity,pulse duration and frequency on postsynaptic currents had been discussed.Here,inspired by optogenetics,the long-term remembering or forgetting could be activated by the 450 nm light stimulus.A light-induced forgetting or remembering is demonstrated in the process of recovery of long-term synaptic plasticity under 20 s and 10 Hz positive or negative gate pulse stimulus,and 30 mW light illumination can promote or inhibit the recovery of current for 4 min.Furthermore,the light stimulus promotes the associative learning and remembering function,which has been demonstrated by Pavlov’s dog experiment with the cooperation of electrical pulses and light stimulus.Such light-assisted learning and memory function can be attributed to the trapping effects of the graphene surface on photo-generated electrons in our synaptic device.2.The application of above-mentioned graphene synaptic device on optical synaptic plasticity has been limited due to the small number of photogenerated carriers in graphene.Thus,a flexible electric-double-layer photoelectric synaptic transistor based on perovskite quantum dot/graphene heterojunctions was designed and prepared to enhance photogenerated carriers.The barrier potential formed by the perovskite quantum dot/graphene heterojunction promotes the separation of photocarrier pairs,and then inhibits their complexation,thus the lifetime of photogenerated carrier pairs can be effectively retarded.Based on this property,the optical synaptic plasticity had been successfully mimicked under the stimulation of 450 nm light pulses,such as excitation of postsynaptic currents,paired pulse facilitation,and spike rate-dependent plasticity,etc.Exploiting the tunability of the Fermi energy level of graphene and the quantum dot energy level pegging effect,the inhibition of light postsynaptic currents had been firstly achieved under a negative gate bias with an amplitude of-2.2 V in the visible light range.Similarly,electrical synaptic functions could also be mimicked by modulating carrier transport based on the double-electric-layer effect of the ion gel.Moreover,we simulated and discussed the practical applications of such optoelectronic synaptic device,such as logic computing,Morse code decoding,and storage learning.The ion-gel-based optoelectronic synaptic device proposed in the paper has been demonstrated to provide a new strategy for the implementation of flexible photoelectrically stimulated synaptic transistors and the development of future flexible sensing and memory integrated neuromorphic computation system.