Reconfigurable optoelectronic functionality implemented in a Ta₂NiSe₅/WS₂ heterostructure toward multifunctional...

Area of Science:

• Materials Science and Engineering
• Optoelectronics
• Artificial Intelligence Hardware

Background:

• Advancements in artificial intelligence (AI) drive demand for versatile optoelectronic devices.
• Multi-functional optoelectronic devices are crucial for streamlining AI system design and reducing costs.
• Fabricating devices for both photodetection and neuromorphic simulation presents structural challenges.

Purpose of the Study:

• To develop a single optoelectronic device capable of both photodetection and neuromorphic visual simulation.
• To overcome the conflicting structural requirements for dual-modal optoelectronic functionality.
• To present a novel heterostructure for integrated intelligent optoelectronic systems.

Main Methods:

• Fabrication of a tantalum selenide (Ta2NiSe5) and tungsten disulfide (WS2) heterostructure.
• Characterization of the device's performance in photovoltaic mode (self-powered and reverse bias).
• Evaluation of the device's performance in photoconductive mode (forward bias) for neuromorphic functions.

Main Results:

• Photovoltaic mode: High responsivity (6.58 A W⁻¹), detectivity (1.56 × 10¹² Jones), and fast response times (46.9/48.9 μs) for photodetection.
• Photoconductive mode: Demonstrated synaptic functions including short-term plasticity (STP), long-term plasticity (LTP), and learn-forget-relearn capabilities.
• Achieved a paired pulse facilitation (PPF) of 31.34% via photocarrier trapping/de-trapping mechanisms.

Conclusions:

• The Ta2NiSe5/WS2 heterostructure successfully integrates photodetection and neuromorphic visual simulation.
• This dual-modal device offers a novel approach for creating compact, intelligent optoelectronic systems.
• The findings pave the way for future multifunctional devices in AI applications.