A Carbon‐Introduced Single Layer GaN Artificial Optoelectronic Synapse for Human Action Recognition and Spatio‐Temporal Encoding
Authors:
Open Access
- 6 July 2026
- journal article
- research article
- Published by Wiley in Advanced Optical Materials
- Vol. 14 (27) , e71353
Abstract
To realize real‐time, device‐based cognitive tasks, there is an urgent need to develop high‐performance optoelectronic synapse devices that integrate perception and storage functions. While optoelectronic synapse devices with various structures such as hybrid‐dimensional or heterostructure expand the functional design space, applicable neuromorphic vision hardware requires material platforms that are scalable, manufacturable and reproducible. This study demonstrates an array‐level single layer GaN synapse fabricated by conventional semiconductor processes. Although the direct bandgap of GaN is unfavorable for the persistent photoconductivity (PPC) effect, it can be obtained through intentionally introduced carbon atoms in GaN film by adjusting the growth parameters and optimizing the Ti/Al–GaN interface. The optoelectronic synapse device exhibits repeatable excitatory postsynaptic current (EPSC), paired‐pulse facilitation (PPF), tunable short‐term plasticity (STP) and long‐term potentiation (LTP). The carrier capture and release at carbon‐related deep‐level defects enable exceptional device uniformity and stability. The STP characteristics of this device are employed for human action recognition, enabling efficient processing of dynamic temporal signals. Moreover, array‐level LTP is used for spatio‐temporal encoding, demonstrating stable signal storage. This work establishes a concise, stable, and CMOS‐compatible fabrication strategy for optoelectronic synapse arrays, presenting a reliable and scalable hardware platform for future brain‐inspired perception and intelligent recognition systems.Keywords
Funding Information
- National Natural Science Foundation of China (62374062)
- National Natural Science Foundation of China (62375090)
- National Natural Science Foundation of China (52002135)
This publication has 39 references indexed in Scilit:
- Memristors Based on 2D Materials: Bridging Device Theory and Potential ApplicationsAdvanced Functional Materials, 2025
- Neuromorphic Transistors Integrating Photo‐Sensor, Optical Memory and Visual Synapses for Artificial Vision ApplicationAdvanced Materials, 2025
- Ultraviolet Optoelectronic Synapse Based on AlScN/p‐i‐n GaN Heterojunction for Advanced Artificial Vision SystemsAdvanced Materials, 2025
- Retina‐Inspired Artificial Synapses with UV Modulated and Immediate Switchable PlasticityAdvanced Functional Materials, 2025
- Realizing Bidirectional Photocurrent in Monolithic Dual‐Mode Device for Neuromorphic Vision and Logically‐Encrypted TransmissionAdvanced Functional Materials, 2024
- Superior AlGaN/GaN‐Based Phototransistors and Arrays with Reconfigurable Triple‐Mode Functionalities Enabled by Voltage‐Programmed Two‐Dimensional Electron Gas for High‐Quality ImagingAdvanced Materials, 2024
- Fully UV Modulated Artificial Synapses with Integrated Sensing, Storage and ComputationAdvanced Functional Materials, 2024
- Broadband Artificial Tetrachromatic Synaptic Devices Composed of 2D/3D Integrated WSe2‐GaN‐based Dual‐Channel Floating Gate TransistorsAdvanced Functional Materials, 2024
- Memristor‐Based Neuromorphic ChipsAdvanced Materials, 2024
- Dual‐Mode Conversion of Photodetector and Neuromorphic Vision Sensor via Bias Voltage Regulation on a Single DeviceAdvanced Materials, 2023