Cooperative Biomemristor Based on PEDOT:PSS-Pectin: Realizing Artificial Synapses and High-Precision Reservoir...

Area of Science:

• Materials Science
• Neuroscience
• Computer Engineering

Background:

• Organic polymer-based memristors are promising for artificial synapses due to ease of fabrication, low cost, and biocompatibility.
• They are suitable for reliable reservoir computing (RC) applications.

Purpose of the Study:

• To develop an effective strategy for constructing organic artificial neural synapses.
• To utilize conductive polymer PEDOT:PSS and polysaccharide-pectin as a functional layer for memristor devices.
• To investigate the synaptic functions and reservoir computing capabilities of the fabricated devices.

Main Methods:

• Fabrication of a Cu/PEDOT:PSS-pectin/ITO (CPPI) memristor device.
• Investigation of conductive filament formation via metal ion migration (Cu2+).
• Characterization of synaptic plasticity functions (LTP, LTD, PPF, PPD, STDP) and short-term to long-term plasticity transition.
• Construction and testing of a physical reservoir computing system using the CPPI device.

Main Results:

• The CPPI device successfully simulated various neural synapse functions, including LTP, LTD, PPF, PPD, EPSC, and STDP.
• A high recognition rate of 99.0% was achieved for machine-written numbers (0-9) in a physical RC system.
• Integration of the CPPI device into an RC system yielded a 91.1% recognition rate for MNIST handwritten digits.

Conclusions:

• The study presents a novel organic artificial synapse based on PEDOT:PSS-pectin, demonstrating robust synaptic plasticity and high performance in reservoir computing.
• This work advances the development of organic polymers for artificial synapses and offers a new material platform for brain-inspired computing.
• The fabricated memristors show potential for next-generation computing applications.