2D Materials Powering Neuromorphic Intelligence

IF 36.3 1区材料科学 Q1 Engineering

Nano-Micro Letters Pub Date : 2026-06-24 DOI:10.1007/s40820-026-02253-1

Jamal Kazmi, Waqas Ahmad, Muhammad Naqi, Yawar Abbas, Aumber Abbas, Peijian Wang, Mohd Ambri Mohamed, Federico Rosei, Zhiming M. Wang, Hongwei Song

引用次数: 0

Abstract

Highlights

Two-dimensional materials enable energy-efficient neuromorphic computing through gate tuneable bandgaps, fast switching kinetics, and compatibility with spiking neural networks for adaptive learning.
Emerging memory devices leveraging 2D materials—including resistive, ferroelectric, and phase-change systems—mimic synaptic plasticity to revolutionize neuromorphic architectures.
Future advances in thin-film synthesis, defect engineering, and quantum-inspired designs will unlock scalable, sustainable 2D neuromorphic systems for healthcare, edge AI, and quantum computing.

查看原文本刊更多论文

2D材料驱动神经形态智能。

对节能和自适应计算架构的指数级需求推动了人工智能（AI）和机器学习（ML）的发展。受生物神经网络的启发，神经形态计算克服了传统冯·诺伊曼架构的局限性，包括高能耗和有限的可扩展性。引入二维（2D）材料，如过渡金属二硫族化合物、六方氮化硼、黑磷和碲，使神经形态器件具有前所未有的电子和光电子特性控制。这些材料具有原子级的厚度、高载流子迁移率和可调带隙，促进了突触行为，如峰值时间依赖的可塑性和对脉冲促进。这篇综述描述了将二维材料集成到神经形态系统中，重点介绍了在可穿戴电子、脑机接口和量子神经形态平台中的应用。在可穿戴和边缘计算领域，基于2d的设备可以实现本地化、超低功耗的数据处理。在脑机接口中，它们增强了信号转导和神经接口。二维材料中的量子效应进一步使混合量子经典神经形态架构能够用于高维计算任务。尽管取得了重大进展，但在再现性、可伸缩性和稳定性方面仍然存在挑战。通过合成和缺陷钝化的创新来解决这些限制对于实际应用是必不可少的。这篇综述强调了基于2d材料的神经形态计算在节能人工智能方面的变革潜力。将二维材料集成到神经形态计算架构中，为构建生物学习机制与机器智能之间的桥梁，提供了一条通向节能和自适应系统的有希望的途径。

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

Nano-Micro Letters NANOSCIENCE & NANOTECHNOLOGY-MATERIALS SCIENCE, MULTIDISCIPLINARY

CiteScore

32.60

自引率

4.90%

发文量

981

审稿时长

1.1 months

期刊介绍： Nano-Micro Letters is a peer-reviewed, international, interdisciplinary, and open-access journal published under the SpringerOpen brand. Nano-Micro Letters focuses on the science, experiments, engineering, technologies, and applications of nano- or microscale structures and systems in various fields such as physics, chemistry, biology, material science, and pharmacy.It also explores the expanding interfaces between these fields. Nano-Micro Letters particularly emphasizes the bottom-up approach in the length scale from nano to micro. This approach is crucial for achieving industrial applications in nanotechnology, as it involves the assembly, modification, and control of nanostructures on a microscale.