Organic Solid-State Electrolyte Synaptic Transistors with Photoinduced Thiol-Ene Cross-linked Polymer Electrolytes...

Area of Science:

• Materials Science
• Organic Electronics
• Neuromorphic Computing

Background:

• Solid-state polymer electrolytes (SPEs) are crucial for developing advanced electronic devices.
• Electrolyte-gated organic field-effect transistors (EGOFETs) offer potential for low-power, flexible electronics.
• Synaptic behavior in artificial devices is key for efficient neuromorphic computing.

Purpose of the Study:

• To develop a novel SPE-based EGOFET using a photocurable nitrile butadiene rubber (NBR) network.
• To investigate the electronic and synaptic properties of the NBR/LiTFSI EGOFET.
• To demonstrate the device's capability for handwritten digit recognition in a deep neural network (DNN).

Main Methods:

• Fabrication of photocurable SPE films using thiol-ene-assisted photo-cross-linking of NBR with LiTFSI.
• Patterning of SPE films using photolithography.
• Characterization of EGOFET electronic properties, including transconductance and on/off ratio.
• Evaluation of synaptic behavior and performance in a DNN for handwritten digit recognition.

Main Results:

• The NBR/LiTFSI EGOFET exhibited excellent electronic properties with high transconductance (11.9 mS) and on/off ratio (10^5).
• Significant current hysteresis was observed, enabling crucial synaptic learning and memory functions.
• The device achieved a high handwritten digit recognition accuracy of 91.9% in a DNN.

Conclusions:

• The developed solid-state NBR/LiTFSI EGOFET demonstrates promising potential for neuromorphic applications.
• The photocurable nature and embedded electrolyte offer a pathway for creating efficient, low-energy artificial intelligence hardware.
• This research highlights the viability of NBR-based SPEs for next-generation neuromorphic devices.