晶体管
数码产品
差速器(机械装置)
非线性系统
CMOS芯片
计算机科学
纳米技术
电子线路
逻辑门
材料科学
钥匙(锁)
物理
非常规计算
纳米电子学
电子工程
编码(集合论)
极限(数学)
半导体器件
可逆计算
神经形态工程学
输运现象
分子电子学
折叠(DSP实现)
拓扑(电路)
记忆电阻器
柔性电子器件
单调函数
软件
作者
Ying Luo,Weixiao Sun,Yongbiao Zhai,Sunyingyue Geng,俞庆森,Haitao Zhou,Ziyu Lv,Guanglong Ding,Ye Zhou,Su‐Ting Han
摘要
Negative differential resistance (NDR), a counterintuitive transport phenomenon in which current decreases with increasing voltage, challenges conventional transistor-centric computing paradigms based on monotonic electronic transport and opens new opportunities for beyond-Boolean computing. Here, we present a comprehensive and concept-driven review of NDR devices spanning memristor-based, diode-type, and transistor-based platforms. We establish a unified framework that links diverse NDR mechanisms, including resonant tunneling, electrothermal feedback, defect dynamics, and ferroelectric polarization, through their shared nonmonotonic transport characteristics. Beyond device-level classification, we further propose NDR as a physical foundation for functionally compressed computing, in which circuit functionalities traditionally implemented using multiple transistors and feedback networks can be partially embedded into the intrinsic nonlinear response of a single NDR device or compact device unit. We further compare representative NDR technologies using common performance metrics and analyze the key challenges that currently limit large-scale deployment, including variability, CMOS compatibility, compact modeling, and the distinction between intrinsic NDR behavior and measurement-induced artifacts. Finally, we discuss future opportunities in materials-by-design, heterogeneous and 3D integration, physics-informed modeling, and closed-loop intelligent systems. By connecting nonmonotonic transport physics with circuit and system-level functionality, NDR electronics offers a promising route toward compact and energy-efficient computing architectures in the post-Moore era.
科研通智能强力驱动
Strongly Powered by AbleSci AI