<p>Rapid advances in artificial intelligence, machine learning, and neural computing propel the demand for multifunctional optoelectronic devices that can integrate sensing, memory, and processing capabilities. Although two-dimensional (2D) van der Waals heterostructures (vdWHs) offer unique advantages for multifunctional integration, their practical applications remain limited by complex device architectures and inefficient mode-switching. Herein, we propose a MoTe<sub>2</sub>/SnS<sub>2</sub> anti-ambipolar heterojunction device enabling single-gate reconfiguration among frequency doubling, optoelectronic detection, and neuromorphic computing. The device exhibits a peak-to-valley ratio (PVR) of 465 for efficient frequency doubling and outstanding optoelectronic performance across the broad wavelength range of 520–2200 nm. Additionally, it can simulate complete synaptic behaviors, including short-term plasticity (STP), long-term plasticity (LTP), and paired-pulse facilitation (PPF). When integrated into a reservoir computing (RC) system trained on a vehicle motion dataset, it achieves a directional recognition accuracy of 98.7%. This work opens a new path for multifunctional integration and low-power neuromorphic computing, advancing the development of next-generation intelligent optoelectronic systems.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Single-gate reconfigurable multifunctional devices based on anti-ambipolar van der Waals heterojunctions

  • Tingting Guo,
  • Zixu Sa,
  • Xiaoyong Jiang,
  • Zhidong Pan,
  • Jing Li,
  • Yehui Shen,
  • Jialin Yang,
  • Chuyao Chen,
  • Hengze Qu,
  • Nengjie Huo,
  • Gangyi Zhu,
  • Xiang Chen,
  • Jinshui Miao,
  • Zai-Xing Yang,
  • Shengli Zhang,
  • Xiufeng Song,
  • Haibo Zeng

摘要

Rapid advances in artificial intelligence, machine learning, and neural computing propel the demand for multifunctional optoelectronic devices that can integrate sensing, memory, and processing capabilities. Although two-dimensional (2D) van der Waals heterostructures (vdWHs) offer unique advantages for multifunctional integration, their practical applications remain limited by complex device architectures and inefficient mode-switching. Herein, we propose a MoTe2/SnS2 anti-ambipolar heterojunction device enabling single-gate reconfiguration among frequency doubling, optoelectronic detection, and neuromorphic computing. The device exhibits a peak-to-valley ratio (PVR) of 465 for efficient frequency doubling and outstanding optoelectronic performance across the broad wavelength range of 520–2200 nm. Additionally, it can simulate complete synaptic behaviors, including short-term plasticity (STP), long-term plasticity (LTP), and paired-pulse facilitation (PPF). When integrated into a reservoir computing (RC) system trained on a vehicle motion dataset, it achieves a directional recognition accuracy of 98.7%. This work opens a new path for multifunctional integration and low-power neuromorphic computing, advancing the development of next-generation intelligent optoelectronic systems.