首页 抖音快讯文章正文

Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉

抖音快讯 2025年08月22日 18:54 1 admin

近日,暨南大学大学关贺元,卢惠辉团队以「Tunable optoelectronic memristor based on MoS2/BaTiO3 for neuromorphic vision」¹为题在Chip上发表研究论文,文章展示了一种基于MoS2/BaTiO3异质结的光电器件,成功实现了紧凑结构下的多级电导调控和类生物突触行为模拟,为低功耗人工智能芯片的发展提供了新思路。第一作者为方子良、陈炳羽,通讯作者为杨铁锋副教授和关贺元教授。


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉

图1 | 受脑启发的神经形态计算。


视觉神经形态器件通过将感知、存储和计算功能集成于单一器件中,模拟生物视觉系统的高效信息处理能力²,为解决冯·诺依曼架构在数据传输过程中存在的高能耗和高延迟问题提供了重要方向(图1)。


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉

图2 | MoS2/BaTiO3器件的光电特性。


本研究提出了一种基于MoS2和BaTiO3的异质结光电器件。利用BaTiO3中的铁电极化弛豫滞后于电和光激励特性³ ⁴,实现了电场与光脉冲双重调控下的电导状态切换。实验中,该器件展现出清晰的电阻切换行为,以及通过光脉冲实现的多电导状态(图2)。


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉

图3 | 基于MoS2/BaTiO3器件的感知器网络。


同时,利用光电响应和神经形态行为的双重特性,构建了一个集视觉感知和图像识别于一体的多层感知器网络。基于该器件的多层感知网络在MNIST手写数字识别任务中实现了97.6%的识别准确率,充分展示了该器件在低功耗人工智能芯片和神经形态计算领域的广阔应用前景


Tunable optoelectronic memristor based on MoS2/BaTiO3 for neuromorphic vision¹


Inspired by the human visual system, researchers from Jinan University developed a heterostructured optoelectronic memristor based on MoS2 and BaTiO3. This device integrates sensing, computing, and memory functions, providing a compact and cost-effective solution for neuromorphic vision applications².


With a simple planar structure, the MoS2/BaTiO3 neuromorphic devices shows 16 distinct conductive states due to the lag of ferroelectric polarization decay behind electrical and optical excitation³ ⁴, and successfully mimics biological synaptic behaviors such as short-term memory (STM) and long-term memory (LTM).


Furthermore, the device was used to construct a multi-layer perceptron network integrating visual perception and image recognition. In the MNIST handwritten digit recognition task, the network achieved an impressive accuracy of 97.6%. This study demonstrates the potential of MoS2/BaTiO3-based neuromorphic devices in advancing compact, low-power artificial intelligence systems.


参考文献


1. Fang, Z. et al. Tunable optoelectronic memristor based on MoS2/BaTiO3 for neuromorphic vision. Chip 4, 100136 (2025).

2. Li T, Miao J, Fu X, et al. Reconfigurable, non-volatile neuromorphic photovoltaics. Nat. Nanotechnol. 18, 1303-1310 (2023).

3. Guan H, Liu F, He Z, et al. Efficient electrical extraction of nonlinear response and large linear dynamic range implementation in MoS2/BaTiO3 hetero-Integrated photodetector. Laser Photonics Rev. 18, 2400445 (2024).

4. Liang X, Guan H, Luo K, et al. Van der waals integrated LiNbO3/WS2 for high-performance UV-Vis-NIR photodetection. Laser Photonics Rev. 17, 2300286 (2023).

5. Hu M, Graves C E, Li C, et al. Memristor‐based analog computation and neural network classification with a dot product engine. Adv. Mater. 30,1705914 (2018).


论文链接:

https://www.sciencedirect.com/science/article/pii/S2709472325000103


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉

扫码阅读原文


作者简介


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉


方子良,暨南大学博士研究生,研究方向主要为微结构光电子器件。


Ziliang Fang, a Ph.D. student at Jinan University, focuses on microstructured optoelectronic devices.


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉


陈炳羽,暨南大学博士研究生,研究方向主要为光学鉴相器。


Bingyu Chen, a Ph.D. student at Jinan University, focuses on optical phase demodulators.


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉


杨铁锋,暨南大学副教授,主要从事基于铌酸锂/层状材料异质集成高性能光电子器件研究。发表学术论文50余篇,含2篇高被引,2篇封面论文,总引用次数超过2300余次。


Tiefeng Yang, primarily focusing on high-performance optoelectronic devices based on lithium niobate/layered material heterogeneous integration, has published over 50 academic papers, including 2 highly cited papers and 2 cover articles, with a total citation count exceeding 2,300.


Chip发表:基于MoS₂/BaTiO₃的可调谐光电子忆阻器用于类脑视觉


关贺元暨南大学物理与光电工程学院副院长,光电工程研究所所长,国家级高层次青年人才、广东省杰青,广东省科协青年科技人才。主要面向国家重大工程需求,研究薄膜界面增强的光栅集成器件、铌酸锂集成器件、太赫兹集成器件和光电集成芯片等。在Laser & Photonics Reviews(6篇)、ResearchPhotonics Research等高水平期刊上发表论文100余篇,其中封面论文4篇,邀请论文3篇,获授权发明专利5项。研究成果被LPRPRL等权威学术期刊以及中国科学报和人民号等权威媒体报道。


Heyuan Guan, Deputy Dean of the college of Physics & Optoelectronic Engineering at Jinan University, is a recipient of national high-level youth talent awards, the Guangdong Outstanding Youth Fund, and a member of the Guangdong Association for Science and Technology Youth Talent Program. His research focuses on the development of grating-integrated devices with enhanced thin-film interfaces, lithium niobate integrated devices, terahertz integrated devices, and optoelectronic integrated chips to address the needs of major national engineering projects. He has published over 100 papers in high-impact journals such as Laser & Photonics Reviews (5 papers), Research, and Photonics Research, including 4 cover articles and 3 invited papers, and holds 5 authorized invention patents. His research findings have been featured in prestigious academic journals such as LPR and PRL, as well as in authoritative media outlets like China Science Daily and People's Daily.

发表评论

泰日号Copyright Your WebSite.Some Rights Reserved. 网站地图 备案号:川ICP备66666666号 Z-BlogPHP强力驱动