2D (NH<sub>4</sub>)BiI<sub>3</sub> enables non-volatile optoelectronic memories for machine learning

doi:10.1038/s41467-025-56819-5

IMR OpenIR

	2D (NH₄)BiI₃ enables non-volatile optoelectronic memories for machine learning
	Tong, Bo 1,2; Xu, Jiajun 1,2; Du, Jinhong 1,2; Liu, Peitao 1,2; Du, Tianming 3; Wang, Qiang 1,2; Li, Langjun 4; Wei, Yuning 1,2; Li, Jiangxu 1; Liang, Jinhua 1,2; Liu, Chi 1,2; Liu, Zhibo 1,2; Li, Chen 3; Ma, Lai-Peng 1,2; Chai, Yang 5,6; Ren, Wencai 1,2
通讯作者	Ren, Wencai(wcren@imr.ac.cn)
	2025-02-13
发表期刊	NATURE COMMUNICATIONS
卷号	16 期号:1 页码:11
摘要	Machine learning is the core of artificial intelligence. Using optical signals for training and converting them into electrical signals for inference, combines the strengths of both, and thus can greatly improve machine learning efficiency. Optoelectronic memories are the hardware foundation for this strategy. However, the existing optoelectronic memories cannot modulate a large number of non-volatile resistive states using ultra-short and ultra-dim light pulses, leading to low training accuracy, slow computing speed and high energy consumption. Here, we synthesized a van der Waals layered photoconductive material, (NH4)BiI3, with excellent photoconductivity and strong dielectric screening effect. We further employed it as the photosensitive control gate in a floating-gate transistor, replacing the commonly used metal control gate, to construct an optical floating gate transistor which achieves adjustable synaptic weights under ultra-dim light without gate voltage assistance. Moreover, it shows ultra-low training energy consumption to generate a non-volatile state and the largest resistive state numbers among the known non-volatile optoelectronic memories. These exceptional performances enable the construction of one-transistor-one-memory device arrays to achieve similar to 99% accuracy in Artificial Neural Networks. Moreover, the device arrays can match the performance of GPU in YOLOv8 while greatly reducing energy consumption.
资助者	CAS \| Institute of Metal Research, Chinese Academy of Sciences (IMR, CAS) ; National Natural Science Foundation of China ; Chinese Academy of Sciences ; Ministry of Science and Technology of China ; LiaoNing Revitalization Talents Program ; Special Projects of the Central Government in Guidance of Local Science and Technology Development ; Guangdong Basic and Applied Basic Research Foundation
DOI	10.1038/s41467-025-56819-5
收录类别	SCI
语种	英语
资助项目	CAS \| Institute of Metal Research, Chinese Academy of Sciences (IMR, CAS)[52188101] ; CAS \| Institute of Metal Research, Chinese Academy of Sciences (IMR, CAS)[52172057] ; CAS \| Institute of Metal Research, Chinese Academy of Sciences (IMR, CAS)[52002375] ; CAS \| Institute of Metal Research, Chinese Academy of Sciences (IMR, CAS)[52422112] ; National Natural Science Foundation of China[ZDBS-LY-JSC027 [WCR]] ; Chinese Academy of Sciences[2021YFA1200804] ; Ministry of Science and Technology of China[XLYC2201003] ; LiaoNing Revitalization Talents Program[2024010859-JH6/1006] ; Special Projects of the Central Government in Guidance of Local Science and Technology Development[2020B0301030002] ; Guangdong Basic and Applied Basic Research Foundation
WOS研究方向	Science & Technology - Other Topics
WOS类目	Multidisciplinary Sciences
WOS记录号	WOS:001421591600024
出版者	NATURE PORTFOLIO
引用统计	被引频次：1[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.imr.ac.cn/handle/321006/179906
专题	中国科学院金属研究所
通讯作者	Ren, Wencai
作者单位	1.Chinese Acad Sci, Inst Met Res, Shenyang Natl Lab Mat Sci, Shenyang, Peoples R China 2.Univ Sci & Technol China, Sch Mat Sci & Engn, Shenyang, Peoples R China 3.Northeastern Univ, Microscop Image & Med Image Anal Grp, Shenyang, Peoples R China 4.China Med Univ, Clin Coll 1, Shenyang, Peoples R China 5.Hong Kong Polytech Univ Kowloon, Dept Appl Phys, Hong Kong, Peoples R China 6.Hong Kong Polytech Univ Kowloon, Joint Res Ctr Microelect, Hong Kong, Peoples R China
推荐引用方式 GB/T 7714	Tong, Bo,Xu, Jiajun,Du, Jinhong,et al. 2D (NH₄)BiI₃ enables non-volatile optoelectronic memories for machine learning[J]. NATURE COMMUNICATIONS,2025,16(1):11.
APA	Tong, Bo.,Xu, Jiajun.,Du, Jinhong.,Liu, Peitao.,Du, Tianming.,...&Ren, Wencai.(2025).2D (NH₄)BiI₃ enables non-volatile optoelectronic memories for machine learning.NATURE COMMUNICATIONS,16(1),11.
MLA	Tong, Bo,et al."2D (NH₄)BiI₃ enables non-volatile optoelectronic memories for machine learning".NATURE COMMUNICATIONS 16.1(2025):11.