Carbon-based memristors for neuromorphic computing

IF 11.6 1区物理与天体物理 Q1 PHYSICS, APPLIED

Applied physics reviews Pub Date : 2025-10-13 DOI:10.1063/5.0260582

Zheng Wang, Kangli Xu, Jialin Meng, Bo Feng, Tianyu Wang

{"title":"Carbon-based memristors for neuromorphic computing","authors":"Zheng Wang, Kangli Xu, Jialin Meng, Bo Feng, Tianyu Wang","doi":"10.1063/5.0260582","DOIUrl":null,"url":null,"abstract":"Driven by the rapid advancement of the Internet of Things and artificial intelligence, computational power demands have experienced an exponential surge, thereby accentuating the inherent limitations of the conventional von Neumann architecture. Neuromorphic computing memristors are emerging as a promising solution to overcome this bottleneck. Among various material-based memristors, carbon-based memristors (CBMs) are particularly attractive due to their biocompatibility, flexibility, and stability, which make them well suited for next-generation neuromorphic applications. This review summarizes the recent advancements in CBMs and proposes potential application scenarios in neuromorphic computing. Representative CBMs and preparation methods of carbon-based materials in different dimensions (0D, 1D, 2D, and 3D) are presented, followed by structural, storage, and synaptic plasticity testing and switching mechanisms. The neural network architecture built by CBMs is summarized for image processing, wearable electronics, and three-dimensional integration. Finally, the future challenges and application prospects of CBMs are reviewed and summarized.","PeriodicalId":8200,"journal":{"name":"Applied physics reviews","volume":"37 1","pages":""},"PeriodicalIF":11.6000,"publicationDate":"2025-10-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied physics reviews","FirstCategoryId":"101","ListUrlMain":"https://doi.org/10.1063/5.0260582","RegionNum":1,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"PHYSICS, APPLIED","Score":null,"Total":0}

引用次数: 0

Abstract

Driven by the rapid advancement of the Internet of Things and artificial intelligence, computational power demands have experienced an exponential surge, thereby accentuating the inherent limitations of the conventional von Neumann architecture. Neuromorphic computing memristors are emerging as a promising solution to overcome this bottleneck. Among various material-based memristors, carbon-based memristors (CBMs) are particularly attractive due to their biocompatibility, flexibility, and stability, which make them well suited for next-generation neuromorphic applications. This review summarizes the recent advancements in CBMs and proposes potential application scenarios in neuromorphic computing. Representative CBMs and preparation methods of carbon-based materials in different dimensions (0D, 1D, 2D, and 3D) are presented, followed by structural, storage, and synaptic plasticity testing and switching mechanisms. The neural network architecture built by CBMs is summarized for image processing, wearable electronics, and three-dimensional integration. Finally, the future challenges and application prospects of CBMs are reviewed and summarized.

查看原文本刊更多论文

用于神经形态计算的碳基忆阻器

在物联网和人工智能快速发展的推动下，计算能力需求呈指数级增长，从而凸显了传统冯·诺伊曼架构的固有局限性。神经形态计算记忆电阻器正在成为克服这一瓶颈的一个有希望的解决方案。在各种基于材料的记忆电阻器中，碳基记忆电阻器（CBMs）由于其生物相容性，灵活性和稳定性而特别具有吸引力，这使得它们非常适合下一代神经形态应用。本文综述了CBMs的最新进展，并提出了CBMs在神经形态计算中的潜在应用前景。介绍了不同维度（0D、1D、2D和3D）碳基材料的代表性cbm和制备方法，以及结构、存储和突触可塑性测试和开关机制。总结了基于CBMs构建的神经网络结构在图像处理、可穿戴电子和三维集成等方面的应用。最后，对信任措施的未来挑战和应用前景进行了回顾和总结。

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

Applied physics reviews PHYSICS, APPLIED-

CiteScore

22.50

自引率

2.00%

发文量

113

审稿时长

2 months

期刊介绍： Applied Physics Reviews (APR) is a journal featuring articles on critical topics in experimental or theoretical research in applied physics and applications of physics to other scientific and engineering branches. The publication includes two main types of articles: Original Research: These articles report on high-quality, novel research studies that are of significant interest to the applied physics community. Reviews: Review articles in APR can either be authoritative and comprehensive assessments of established areas of applied physics or short, timely reviews of recent advances in established fields or emerging areas of applied physics.