| In the context of the artificial intelligence era,with the rapid accumulation and growth in the amount of information data to be stored and processed,the traditional von Neumann architecture is difficult to meet the demand due to its low transmission speed and high-power consumption.Memristors have attracted considerable attention for their efficient parallel processing mode,like the human brain,and are expected to disrupt traditional computer architectures.The memristor is the fourth basic component of passive circuits and can not only store information but also perform logical operations at the same time.The mechanism of memristors is similar to that of the synapses that connect neurons in the human brain and has been extensively studied in the simulation of brain-like synaptic functions.However,compared to conventional purely electrically regulated amnestic devices,optoelectronic amnestic synaptic devices with an optoelectronic response have the potential to integrate the sensing,processing,and memory of optical information into a single device with faster operation,ultra-low energy consumption,and more flexible regulation.As a result,optoelectronic memristors are considered a key device in the fields of sense,memory,process integration,and artificial synapses.At present,the memristor is still in the early stages of research,which involves some core issues such as how to select the appropriate functional dielectric layer material,how to understand and analyze the resistive mechanism of the memristor,and how to achieve the integration of the sense,memory,and processing functions.In addition,the simulation of synaptic plasticity,and the expansion of application scenarios need to be further explored and improved.Moreover,in recent years,the appearance of molybdenum disulfide as a novel material has sparked significant interest and garnered considerable attention from the scientific community.This is because molybdenum disulfide is controllable,reproducible,and has excellent optoelectronic properties,making it ideal for use in the modulation of optoelectronic signals and the simulation of neurosynaptic functions.It thus has a wide range of applications in many fields.In this dissertation,excellent MoS2 films were successfully prepared by the hydrothermal method,and a series of devices using molybdenum disulfide films as a resistive functional layer were prepared,as follows:1.The photoelectric synaptic characteristics of MoS2-based monolayer thin film memristors and WO3/MoS2 structured memristors were investigated:MoS2 thin films were created hydrothermally,and then analyzed the behavior of memristors made of Ag/MoS2/ITO in terms of their electrical and optical properties.By using magnetron sputtering technology,WO3 thin film was successfully prepared,and subsequently,an Ag/WO3/MoS2/ITO structured memristor was constructed,enabling basic behavioral and functional simulations of biological synapses at both the electrical and optical aspects.In the investigation of the enhancement of the UV signal,it was found that the enhancement of the 365 nm UV signal was time,intensity,and frequency dependent,which enhanced the tunability of the device to the light signal.2.The optoelectronic properties of a MoS2/CeO2 heterojunction optoelectrical memristor and its artificial vision system with nociceptive perception are investigated.The device’s electro-synaptic plastic behavior can be obtained by altering the electrical signal’s pulse amplitude,width,and interval time.Further,the STM/LTM transition of the synapses was simulated by constructing an arrayed pattern based on the function of the electrical synapses.It was found that the sustained photoconductive effect of the device can be clearly observed in the UV and visible light.Furthermore,under optics,the device exhibits synaptic plasticity behavior.Based on this,an array of optoelectrical memristors was constructed to simulate image perception and memory by simultaneously applying a mixed signal of UV light and voltage.The multi-wavelength photoconductive modulation of the devices under red,yellow,and green light was then explored to achieve the versatility of simulating the sense,memory and color recognition of color images by the human visual system.Finally,the prepared memristors successfully simulated the functions associated with biological visual nociceptors,achieving the four key features of nociceptors,namely“threshold”,“no adaptation”,“relaxation”and“nociceptive sensitization”,thus realizing an artificial visual bionic system with nociceptive perception capability. |
