The human brain performs remarkable feats of computation with minimal energy, an efficiency that modern artificial intelligence struggles to replicate using conventional, power-hungry hardware. This work explores a sustainable path toward brain-inspired computing by combining organic electrochemical transistors (OECTs) for synaptic emulation with a novel electrolyte derived from chitosan, an abundant and biodegradable biopolymer sourced from seafood waste. We first established a performance benchmark by rigorously characterizing OECTs, demonstrating their ability to replicate key synaptic functions like paired-pulse facilitation and extracting the fundamental ionic and electronic time constants that govern short-term plasticity. In parallel, we engineered and evaluated multiple chitosan formulations, culminating in an optimized eutectic gel that achieved a notable ionic conductivity of 3 mS/cm. Critically, this sustainable electrolyte successfully enabled fundamental synaptic learning rules in functional OECT devices. This work demonstrates that high-performance neuromorphic hardware need not rely on synthetic materials, effectively bridging the gap between computational efficiency and ecological responsibility for a new generation of green and bio-integrated electronics.

Synaptic Emulation with Chitosan‐Gated Organic Electrochemical Transistors

BRILLI, MATTEO
2024/2025

Abstract

The human brain performs remarkable feats of computation with minimal energy, an efficiency that modern artificial intelligence struggles to replicate using conventional, power-hungry hardware. This work explores a sustainable path toward brain-inspired computing by combining organic electrochemical transistors (OECTs) for synaptic emulation with a novel electrolyte derived from chitosan, an abundant and biodegradable biopolymer sourced from seafood waste. We first established a performance benchmark by rigorously characterizing OECTs, demonstrating their ability to replicate key synaptic functions like paired-pulse facilitation and extracting the fundamental ionic and electronic time constants that govern short-term plasticity. In parallel, we engineered and evaluated multiple chitosan formulations, culminating in an optimized eutectic gel that achieved a notable ionic conductivity of 3 mS/cm. Critically, this sustainable electrolyte successfully enabled fundamental synaptic learning rules in functional OECT devices. This work demonstrates that high-performance neuromorphic hardware need not rely on synthetic materials, effectively bridging the gap between computational efficiency and ecological responsibility for a new generation of green and bio-integrated electronics.
2024
File in questo prodotto:
File Dimensione Formato  
Brilli Matteo - Thesis .pdf

embargo fino al 04/11/2026

Dimensione 16.69 MB
Formato Adobe PDF
16.69 MB Adobe PDF

I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.14247/26902