神经形态工程学
计算机数据存储
材料科学
计算机科学
数码产品
纳米技术
人工智能
计算机硬件
电气工程
人工神经网络
工程类
作者
Ruizhi Yuan,Yingjie Cao,Xiyu Zhu,Xiuming Shan,Bo Wang,Shouxin Zhang,Sen Chen,Jing Liu
标识
DOI:10.1002/adma.202309182
摘要
Abstract Storage systems are vital components of electronic devices, while significant challenges persist in achieving flexible memory due to the limitations of existing storage methodologies. Inspired by the polarization and depolarization mechanisms in the human brain, here a novel class of storage principles is proposed and achieve a fully flexible memory through introducing the oxidation and deoxidation behaviors of liquid metals. Specifically, reversible electrochemical oxidation is utilized to modulate the overall conductivity of the target liquid metals, creating a substantial 11‐order resistance difference for binary data storage. To obtain the best storage performance, systematic optimizations of multiple parameters are conducted. Conceptual experiments demonstrate the memory's stability under extreme deformations (100% stretching, 180° bending, 360° twisting). Further tests reveal that the memory performs better when its unit size gets smaller, warranting superior integrability. Finally, a complete storage system achieves remarkable performance metrics, including rapid storage speed (>33 Hz), long data retention capacity (>43200 s), and stable repeatable operation (>3500 cycles). This groundbreaking method not only overcomes the inherent rigidity limitations of existing electronic storage units but also opens new possibilities for innovating neuromorphic devices, offering fundamental and practical avenues for future applications in soft robotics, wearable electronics, and bio‐inspired artificial intelligence systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI