arrow
Return

All-In-One Sensing-Memory-Computing System Based on a Self-Powered MoS2 P-N Homojunction Transistor

delete2026-05-15
delete0
PRE
AI
W
Wen Deng
X
Xin Yan
H
Han Xu
H
Hao Luo
H
Hao Zhang
R
Rui Su
L
Lisheng Wang
N
Niannian Yu
B
Bo Wen Jia
J
Jinsong Wu
罗雯 (Wen Luo) *
DOI:10.1002/adfm.75864delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Transition metal chalcogenides are promising for integrated sensing-memory-computing optoelectronics, owing to their atomic-scale thickness and outstanding electrostatic control. However, the fast charge dynamics required for sensing conflict with the long charge retention needed for storage and synapses, hindering their functional integration in a single device. This work demonstrates a multifunctional MoS2 P-N homojunction ambipolar transistor, formed by vertically stacking intrinsic MoS2 with uniform P-type MoS2-xOδ prepared by ultraviolet-ozone photochemical doping. This type-II homojunction exhibits gate-tunable ambipolarity, enabled by oxygen-doping-modulated band structure and interfacial charge transfer. Kelvin probe force microscope (KPFM) measurement directly corroborate the built-in electric field and efficient photogenerated charge separation in the junction region, confirming its type-II band alignment and self-powered operation. Simply by switching the gate polarity, the device can be reversibly reconfigured. Under negative gate bias, it acts as a fast, sensitive self-powered photodetector. Under positive gate bias, it exhibits self-powered persistent photoconductivity, nonvolatile multilevel memory, and synaptic plasticity. Based on these reconfigurable functions, the platform integrates on-chip high-contrast single-pixel imaging, long-term optical storage, and an optical front-end bionic visual neural network. This work provides a high-performance compact, low-power platform for brain-inspired perception and computing from 2D materials defect engineering perspective.
Keywords:
ambipolar transistor
oxygen-doped MoS2
P-N homojunction
self-powered
sensing-memory-computing integration

Journal

Advanced Functional Materials cover
Advanced Functional Materials
IF:
19
Papers:
3.4W
Citations:
32.1W

Organization

W
wuhan university of technology
Scholars:
7.2K
Papers: 2.2K
Citations: 0