Photoelectric Characteristics Study And Artificial Synapse Simulation Based On The Phase Change Mechanism Of Two-dimensional MoTe₂

The vigorous development of artificial intelligence,Internet of Things and machine learning places higher demands on the computing power of modern computing devices.Unfortunately,the normal electronic computer system based on the"von Neumann"architecture has an insurmountable efficiency bottleneck.In this context,the brain-inspired neuromorphic computing system has attracted great attention from researchers because of its natural"storage and computing integration"feature.This also makes the electronic device that can mimic the characteristics of biological synapses have been widely studied and reported.Due to the atomic-level thickness and low charge shielding ability,the physical properties of the two-dimensional material can be easily adjusted through various stimuli,which is very beneficial for synapse simulation.In addition,the novel physical properties and excellent photoelectric properties of two-dimensional materials have great potential for studying functional synaptic simulations,such as photosensitivity,hearing ability,and tactile ability.Among the branches of the two-dimensional material family,transition metal dichalcogenides（TMDCs）have received extensive attention due to their extremely excellent electrical properties.Especially their inherent polymorphism,which has shown a vigorous development trend in the field of phase change engineering,and MoTe₂ is a typical representative of them.Thus,this study is based on the excellent phase change characteristics and optoelectronic characteristics of MoTe₂ to carry out research work.The specific work content is as follows:Firstly,using the persistent photoconductivity（PPC）effect that is closely related to defects,the artificial synapse simulation based on the MoTe₂ field-effect transistor is realized.The influence of the gate voltage on the PPC effect in the device can be explained by the Fermi level shift and the change of the electron binding energy.The reconfigurable photo-synaptic transistor which can more flexible simulate the characteristics of biological synapses such as long-term enhancement,long-term depression,and paired-pulse facilitate was realized based on the regulation of the gate voltage.The device exhibits good adaptability to diverse neuromorphic computing applications.Secondly,after verifying the electric field can induce the MoTe₂ phase transition,the wide bandgap material GaSe was introduced as the dielectric layer to control the active area of the device,which successfully achieved the optimization of the device reliability and stability.On this basis,rich intermediate conductance states were modulated by appropriate pulse excitation,which can realize the simulation of long-term enhancement,long-term depression,and spike time-dependent characteristics in biological synapses.It is expected to be used in scalable and fast resistance switching synapses simulation devices.Finally,the changes in the effective light-absorbing area of the device channel during the MoTe₂ phase change were used to realize the regulation of the photovoltaic characteristics to obtain multiple adjustable and nonvolatile photovoltaic states,with a good light response,fast response speed and reconfiguration,etc.It shows the application prospects in the field of"electro-writing and optical-reading"non-volatile storage devices.