神经形态工程学
逻辑门
晶体管
可编程逻辑器件
材料科学
逻辑族
非易失性存储器
双极扩散
极性(国际关系)
逆变器
计算机科学
电子线路
灵活性(工程)
通流晶体管逻辑
可编程逻辑阵列
电子工程
与非门
和或反转
电阻器–晶体管逻辑
多发射极晶体管
NOR门
现场可编程门阵列
和大门
电压
电气工程
或门
逻辑综合
计算机体系结构
计算机硬件
宏单元阵列
编码(内存)
光电子学
能量(信号处理)
可重组计算
数字电子学
双极结晶体管
电阻随机存取存储器
嵌入式系统
阈值电压
CMOS芯片
作者
Zhenyu Feng,Wei Li,Yan Xu,Cuijie Qin,Xiaoyue Huang,Shenglan Hao,Jie Yang,Guangdi Feng,Chun‐Gang Duan,Junhao 君浩 Chu 褚,Bobo Tian
摘要
ABSTRACT Reconfigurable transistors integrating memory and logic functions are promising for beyond‐von‐Neumann computing, yet existing architectures often suffer from limited operational flexibility and coupled programming/readout mechanisms. Here, a ferroelectric‐enabled dual‐gate WSe 2 ambipolar transistor is demonstrated using a ferroelectric P(VDF‐TrFE) top gate and a conventional back gate to achieve decoupled non‐volatile polarity programming and volatile logic operation. The ambipolar WSe 2 channel can be reversibly configured into stable n‐type or p‐type states, enabling symmetric dual‐mode multi‐state memory with an ON/OFF ratio up to 10 7 , switching speed down to 500 ns, energy consumption of 0.2 fJ, and endurance exceeding 10 7 cycles. Leveraging the intrinsic polarity reconfigurability, multiple Boolean logic gates and a compact 1‐bit content‐addressable memory (CAM) cell are implemented within a single transistor. In addition, a programmable complementary inverter with tunable switching thresholds enables multi‐state voltage encoding for enhanced neuromorphic vision processing. This work provides a versatile platform for multifunctional, energy‐efficient, and adaptive in‐memory computing systems.
科研通智能强力驱动
Strongly Powered by AbleSci AI