id_768. FROM RIGID TO FLEXIBLE: REDOX-ACTIVE MACROMOLECULAR SYSTEMS FOR EMULATING NEURAL PLASTICITY IN ORGANIC ELECTRONICS
Anna Kostecka, Wojciech Wieczorek, Krystian Sokołowski, Tomasz Mazur, Konrad Szaciłowski, Michał Szuwarzyński
AGH University of Krakow, Academic Center for Materials and Nanotechnology, Krakow, Poland
INTRODUCTION: Bio-inspired computing requires hardware mimicking the processing power and physical nature of the biological brain. Unlike rigid silicon electronics, organic memristors offer a unique combination of analog switching (plasticity) and mechanical flexibility, making them ideal "artificial synapses" for bio-integrated electronics. However, the computational fidelity of such devices depends critically on the nanoscale engineering of the polymer-metal interface.
AIM(S): This study evaluates poly(glycidyl methacrylate) (PGMA) brushes as a platform for emulating neural functions. We investigate how the top electrode deposition technique (Thermal Evaporation vs. Magnetron Sputtering) and synthesis method influence resistive switching and the ability to mimic synaptic plasticity in both rigid and flexible systems.
METHOD(S): PGMA brushes were synthesized via surface-initiated polymerization on rigid (ITO/glass) and photopolymerization on flexible (PET/ITO) substrates to suit soft electronics. Functionalization with redox-active aminoferrocene introduced charge-trapping states, alongside control amines (hexylamine, fluorobenzylamine). Top gold electrodes were deposited via low-energy Thermal Evaporation or high-energy Magnetron Sputtering. Morphology and composition were characterized by AFM, XPS, and IR spectroscopy, followed by electrical measurements to assess neuromorphic behavior.
RESULTS: Analysis reveals that the interface formation technique governs the symmetry and linearity of switching dynamics (learning rules). We compare deposition routes to determine optimal stability conditions. Furthermore, the performance of photopolymerized flexible systems is assessed under mechanical bending to verify their potential for wearable neuromorphic applications.
CONCLUSIONS: This work establishes design rules for organic artificial synapses. By analyzing fabrication parameters, we provide essential insight into developing thin-film macromolecular systems tailored for efficient and flexible neural function emulation.
FINANCIAL SUPPORT: This study was funded by National Science Centre (Poland) [Sonata Bis Grant no. 2021/42/E/ST4/00290].