Study On Memristor Characteristics And Neurosynaptic Biomimetic Function Based On Zinc Oxide Thin Film

Memristors have been intensively studied for their potential applications in next-generation high-density and non-volatile memories.Among them,as an important application of the memristor in the memory field,resistive random access memory(RRAM)has attracted much attention due to its reliable performance.However,in order to achieve the goal of practical non-volatile memory applications,one major obstacle must be overcome,that is,namely the dispersion unevenness of the memristive set/reset switching voltage(V_S and V_R).Therefore,the development of new structural devices with stable performance will play a driving role in the future development of the storage field.Other than that,memristors can be work as artificial neuron synapses implementing bio-synaptic plasticity,which can be applied in key research fields including neural networks and artificial intelligence(AI).In view of the existing problems,this thesis mainly conducts the following three research works:Firstly,memristor with W/ZnO/FTO structure is fabricated,and we developed obtained typical memristive electrical properties and reliable simulated biological synaptic functions.The device current response and attenuation obtained by applying the voltage pulse simulates three phases of the refractory action potential.By comparing the effects of different electrodes on ZnO-based device,the physical mechanism of W/ZnO/FTO device functions was further explored.Therefore,the proposed W/ZnO/FTO device based on the realization of refractory function promotes the progress of artificial nerve synaptic bionic function.Secondly,in order to better optimize the resistive switching performance and switching characteristics of the device,we have adopted another effective method,that is,the double-layered mixed structure of oxide and two-dimensional(2D)material as the functional layer of the device.By adopting the ZnO/BNNSs hybrid double-layer structure,we obtained the excellent memristive device characteristics.Moreover,a pulse train can be applied to memristor device to tune resistance gradually,simulating external stimuli of a biosensing system.This research not only provides an approach to improve its performance in terms of stability and uniformity in oxide-based RRAM,but also the ZnO/BNNSs hybrid double-layer structure device can be implemented as reliable building units for the realization of artificial neuromorphic functional units.Thirdly,antimony quantum dots have higher stability and conductivity tunability in the field of electronic devices.Here,by embedding antimonene quantum dots(AQDs),we prepared Ag/ZnO/AQDs/Pt devices to improve the repeatability and parameter uniformity of ZnO-based devices.The embedded quantum dots enhance the surrounding local electric field,guide the growth direction of the conductive filaments,and improve the dispersion of the switching characteristics of the memristor.This research provides a simple technical method for improving the performance of memristors.This work not only shows the huge application prospects of quantum dots in the field of neuromorphic device and neural network,but also has important guiding significance for the design of future artificial intelligence and neural bionic hardware units.