神经形态工程学
晶体管
计算机科学
感知
材料科学
计算机体系结构
薄膜晶体管
光电子学
神经科学
人工智能
纳米技术
电气工程
人工神经网络
工程类
心理学
电压
图层(电子)
作者
Xiaoqian Li,Lin Yi,Xuemei Yin,Jiafeng Cheng,Qian Xin,Aimin Song
标识
DOI:10.1038/s41528-025-00425-4
摘要
Abstract Large-area, paper-based ZnO synaptic transistor arrays for visual perception and neuromorphic computing have been fabricated for the first time entirely by screen printing. The channel ink was formulated by dispersing ZnO nanoparticles with a small amount of hydroxyl-rich ethyl cellulose in terpineol, which converted into a semiconducting film at a low temperature of 90 °C. The paper-based transistor arrays exhibited desirable electrical properties, large-area uniformity, environmental stability and biodegradable, making them particularly promising as disposable devices. The printed ZnO synaptic transistors demonstrated exceptional photoelectric synaptic behaviors, including paired-pulse facilitation and depression, high-pass and low-pass filtering, learning, forgetting, relearning, Morse code recognition, and short-term/long-term plasticity, all at a low energy consumption of about 3.7 pJ per synaptic event. Artificial visual learning and information storage capabilities were achieved owing to the persistent photoconductance effect of the printed ZnO films, achieving an accuracy of 91.4% in neuromorphic computing through optoelectronic co-modulation.
科研通智能强力驱动
Strongly Powered by AbleSci AI