Reconfigurable High‐Performance Memristors Based on Few‐Layer High‐κ Dielectric Bi2SeO5 for Neuromorphic Computing

IF 18.5 1区 材料科学 Q1 CHEMISTRY, MULTIDISCIPLINARY
Fang Yang, Yuwei Xiong, Zhaofu Chen, Shizheng Wang, Yinan Wang, Zhihao Qu, Weiwei Zhao, Jiayi Li, Kuibo Yin, Zhenhua Ni, Jing Wu, Diing shenp Ang, Dongzhi Chi, Xin Ju, Junpeng Lu, Hongwei Liu
{"title":"Reconfigurable High‐Performance Memristors Based on Few‐Layer High‐κ Dielectric Bi2SeO5 for Neuromorphic Computing","authors":"Fang Yang, Yuwei Xiong, Zhaofu Chen, Shizheng Wang, Yinan Wang, Zhihao Qu, Weiwei Zhao, Jiayi Li, Kuibo Yin, Zhenhua Ni, Jing Wu, Diing shenp Ang, Dongzhi Chi, Xin Ju, Junpeng Lu, Hongwei Liu","doi":"10.1002/adfm.202514338","DOIUrl":null,"url":null,"abstract":"Memristors are pivotal for energy‐efficient artificial intelligence (AI) hardware, potentially eliminating the von Neumann bottleneck by in‐memory realizations of synaptic operations. However, the dynamic requirements of neuromorphic computing on specific electronic devices pose reliability and universality challenges, limiting progress toward more widely applicable computing platforms. Here, a 2D high‐κ dielectric‐based memristor with the desired reconfigurable resistive switching behavior is successfully demonstrated. Utilizing a few layered Bi<jats:sub>2</jats:sub>SeO<jats:sub>5</jats:sub> possessing excellent electrical insulation properties as the switching medium, the device features a low operating voltage (≈0.5 V), low operation current (10 pA), long memory retention (&gt;10<jats:sup>3</jats:sup> s), large switching window (≈10<jats:sup>8</jats:sup>), steep slope (&lt;1 mV dec<jats:sup>−1</jats:sup>), fast switching speed (40 ns), and low energy dissipation (≈1 pJ). The switching characteristics between volatile and non‐volatile memory can be achieved on demand by regulating compliance currents, offering the possibility of implementing multiple neural computational primitives. A simulated convolutional neural network (CNN) based on long‐term potentiation/depression (LTP/D) achieves 85% accuracy in complex image recognition. Furthermore, MNIST and fashion‐MNIST recognition with built reservoir computing (RC) utilizing volatile behaviors reach 97% and 85% accuracy, respectively. This work opens new opportunities for 2D high‐κ dielectrics in next‐generation AI hardware with enhanced energy efficiency and computational versatility.","PeriodicalId":112,"journal":{"name":"Advanced Functional Materials","volume":"108 1","pages":""},"PeriodicalIF":18.5000,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Advanced Functional Materials","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1002/adfm.202514338","RegionNum":1,"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

Memristors are pivotal for energy‐efficient artificial intelligence (AI) hardware, potentially eliminating the von Neumann bottleneck by in‐memory realizations of synaptic operations. However, the dynamic requirements of neuromorphic computing on specific electronic devices pose reliability and universality challenges, limiting progress toward more widely applicable computing platforms. Here, a 2D high‐κ dielectric‐based memristor with the desired reconfigurable resistive switching behavior is successfully demonstrated. Utilizing a few layered Bi2SeO5 possessing excellent electrical insulation properties as the switching medium, the device features a low operating voltage (≈0.5 V), low operation current (10 pA), long memory retention (>103 s), large switching window (≈108), steep slope (<1 mV dec−1), fast switching speed (40 ns), and low energy dissipation (≈1 pJ). The switching characteristics between volatile and non‐volatile memory can be achieved on demand by regulating compliance currents, offering the possibility of implementing multiple neural computational primitives. A simulated convolutional neural network (CNN) based on long‐term potentiation/depression (LTP/D) achieves 85% accuracy in complex image recognition. Furthermore, MNIST and fashion‐MNIST recognition with built reservoir computing (RC) utilizing volatile behaviors reach 97% and 85% accuracy, respectively. This work opens new opportunities for 2D high‐κ dielectrics in next‐generation AI hardware with enhanced energy efficiency and computational versatility.
基于少层高κ介电Bi2SeO5的可重构高性能忆阻器用于神经形态计算
忆阻器是节能人工智能(AI)硬件的关键,有可能通过在内存中实现突触操作来消除冯·诺伊曼瓶颈。然而,神经形态计算对特定电子设备的动态要求对可靠性和通用性提出了挑战,限制了向更广泛应用的计算平台的发展。本文成功地展示了一种2D高κ介电基忆阻器,该忆阻器具有所需的可重构电阻开关行为。该器件采用具有优良电绝缘性能的多层Bi2SeO5作为开关介质,具有低工作电压(≈0.5 V)、低工作电流(10 pA)、长记忆保持(>103 s)、大开关窗(≈108)、陡斜率(<1 mV dec−1)、快开关速度(40 ns)和低能量损耗(≈1 pJ)等特点。易失性和非易失性存储器之间的切换特性可以根据需要通过调节顺应电流来实现,从而提供了实现多个神经计算原语的可能性。基于长期增强/抑制(LTP/D)的模拟卷积神经网络(CNN)在复杂图像识别中达到85%的准确率。此外,利用挥发性行为的内置油藏计算(RC)的MNIST和fashion - MNIST识别准确率分别达到97%和85%。这项工作为下一代人工智能硬件中的2D高κ电介质提供了新的机会,具有更高的能源效率和计算通用性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
Advanced Functional Materials
Advanced Functional Materials 工程技术-材料科学:综合
CiteScore
29.50
自引率
4.20%
发文量
2086
审稿时长
2.1 months
期刊介绍: Firmly established as a top-tier materials science journal, Advanced Functional Materials reports breakthrough research in all aspects of materials science, including nanotechnology, chemistry, physics, and biology every week. Advanced Functional Materials is known for its rapid and fair peer review, quality content, and high impact, making it the first choice of the international materials science community.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信