{"title":"Dual-Functional Optoelectronic Synaptic Device Based on MoTe2/h-BN Transistor Through UV Light Induced Doping","authors":"Budan Pei, Xuchen Han, Yan Wang, Jing Liu","doi":"10.1002/smll.202500184","DOIUrl":null,"url":null,"abstract":"Synaptic devices serve as the fundamental units of the brain-inspired neuromorphic computing architecture, which has been proposed to complement the drawback of von Neumann configuration in terms of computational efficiency. In this study, a dual-functional optoelectronic synaptic device is proposed based on the three-terminal MoTe<sub>2</sub>/h-BN transistor to seamlessly integrate both the synaptic and logic operation functions. The device can be switched between <i>n</i>- and <i>p</i>-type modes through ultraviolet (UV) light induced doping, allowing for versatile plasticity modulation strategies tailored to each operational mode. Comprehensive characterization of the synaptic behavior of the device reveals impressive stability and repeatability. The device is then explored to a virtual three-layered neural network array to classify the handwritten digit images from the Modified National Institute of Standards and Technology database, which achieves an accuracy of 95.4% and 94.2% for the <i>n</i>- and <i>p</i>-type modes, respectively, after 40 training cycles. The device also demonstrates its capability as optoelectronic logic gates, including “AND”, “OR” and “XOR” under different gate bias. This multifaceted operation signifies a substantial advancement in the development of hybrid systems that leverage both synaptic and traditional logic functionalities, thereby enhancing the overall efficiency of data processing tasks.","PeriodicalId":228,"journal":{"name":"Small","volume":"22 1","pages":""},"PeriodicalIF":13.0000,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Small","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1002/smll.202500184","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
Abstract
Synaptic devices serve as the fundamental units of the brain-inspired neuromorphic computing architecture, which has been proposed to complement the drawback of von Neumann configuration in terms of computational efficiency. In this study, a dual-functional optoelectronic synaptic device is proposed based on the three-terminal MoTe2/h-BN transistor to seamlessly integrate both the synaptic and logic operation functions. The device can be switched between n- and p-type modes through ultraviolet (UV) light induced doping, allowing for versatile plasticity modulation strategies tailored to each operational mode. Comprehensive characterization of the synaptic behavior of the device reveals impressive stability and repeatability. The device is then explored to a virtual three-layered neural network array to classify the handwritten digit images from the Modified National Institute of Standards and Technology database, which achieves an accuracy of 95.4% and 94.2% for the n- and p-type modes, respectively, after 40 training cycles. The device also demonstrates its capability as optoelectronic logic gates, including “AND”, “OR” and “XOR” under different gate bias. This multifaceted operation signifies a substantial advancement in the development of hybrid systems that leverage both synaptic and traditional logic functionalities, thereby enhancing the overall efficiency of data processing tasks.
期刊介绍:
Small serves as an exceptional platform for both experimental and theoretical studies in fundamental and applied interdisciplinary research at the nano- and microscale. The journal offers a compelling mix of peer-reviewed Research Articles, Reviews, Perspectives, and Comments.
With a remarkable 2022 Journal Impact Factor of 13.3 (Journal Citation Reports from Clarivate Analytics, 2023), Small remains among the top multidisciplinary journals, covering a wide range of topics at the interface of materials science, chemistry, physics, engineering, medicine, and biology.
Small's readership includes biochemists, biologists, biomedical scientists, chemists, engineers, information technologists, materials scientists, physicists, and theoreticians alike.