Organic Neuromorphic Electronics relies on neuron-like transient properties of organic (semi-)conductive devices, such as Electrolyte Gated Organic Transistors (EGOTs). These devices, operated under pulsed or alternate biasing conditions, constitute novel electronic building blocks for applications ranging from biosensing to on-board signal processing. Conversely to conventional inorganic semiconductors, organic materials exhibit unique properties which allow them to interact seamlessly with biological systems through their mixed ionic-electronic conduction. These characteristics make organic neuromorphic devices particularly suitable for bio-interfacing applications, where they can sense, process, and modulate signals interrogating ions and molecules dynamics, which constitute the main information transmission paradigm in biological systems. This PhD thesis builds on these concepts, to extend the understanding of the transient response of electrolyte-gated organic transistor (EGOT) and of two- and one-terminal architectures and to devise novel translational applications of the neuromorphic behavior/response of organic electronic devices. In particular, this work investigates neuromorphic organic devices with the dual purpose of gaining understanding and control of their tunability and of developing organic platforms for on-board signal processing, classification and bio-sensing. This lays the foundations for envisioning the integration of organic neuromorphic electronics in novel smart neural interfaces. The versatility of organic-mixed ionic-electronic conductor-based EGOTs for neuromorphic applications is emphasized, demonstrating gate-driven reversible fine-tuning of EGOT plasticity, arising from reversible modulation of its response timescales. Besides, the discussion is extended to ambipolar materials, demonstrating multilevel memory and signal rectification in individual circuit components. The role of neuromorphic responses in determining the capability to efficiently sort physiological signals, separating them in their relevant spectral components and independently routing them, is discussed as a foundational step towards on-board signal processing in neuroelectronic devices and benchmarked in electrophysiological experiments. Additionally, such sensitivity to time-encoded information is leveraged in a neuro-inspired circuit performing pattern recognition by mimicking the dendritic integration function of biological somas is presented. Finally, neuromorphic response is successfully translated to intracortical organic interfaces capable of mapping physiological level of dopamine in vivo, paving the way for the implementation of neuromorphic responses as novel sensing observables.
L'elettronica neuromorfica organica si basa su proprietà transitorie simili a quelle dei neuroni emulabili dai dispositivi organici (semi-)conduttivi, come i transistor organici a “gating” elettrolitico (EGOT). Questi dispositivi, operanti sotto condizioni di polarizzazione pulsata o alternata, danno origine a nuovi elementi costitutivi dell’elettronica per applicazioni che spaziano dal bio-sensing all'elaborazione del segnali a livello dell’interfaccia stessa. A differenza dei semiconduttori inorganici convenzionali, i materiali organici presentano proprietà uniche che permettono loro di interagire perfettamente con i sistemi biologici grazie alla loro conduzione “ionico-elettronica”. Queste caratteristiche rendono i dispositivi neuromorfici organici particolarmente adatti alle applicazioni di “Bio-interazione”, dove possono rilevare, elaborare e modulare segnali interrogando la dinamica di ioni e molecole. Tale dinamica costituisce il principale paradigma di trasmissione delle informazioni nei sistemi biologici. Questa tesi di dottorato si basa su questi concetti per ampliare la comprensione della risposta transitoria dei transistor EGOT e di architetture a uno e due terminali, nonché per sviluppare nuove applicazioni traslazionali dei fenomeni neuromorfici nei dispositivi elettronici organici. In particolare, questo lavoro indaga i dispositivi neuromorfici organici con il duplice obiettivo di comprendere e controllare la loro modulabilità e di sviluppare piattaforme organiche per l'elaborazione del segnali nel dispositivo, la classificazione e il “bio-sensing”. Ciò getta le basi per l'integrazione dell'elettronica neuromorfica organica in nuove interfacce neurali intelligenti. Viene enfatizzata la versatilità degli EGOT basati su conduttori organici a conduzione mista “ionico-elettronica” per applicazioni neuromorfiche, dimostrando la possibilità di una fine regolazione reversibile della plasticità degli EGOT attraverso il controllo della scala temporale della loro risposta tramite il gate. Inoltre, la discussione si estende ai materiali ambipolari, dimostrando la capacità di memoria multilivello e la rettificazione del segnale in singoli componenti circuitali. Il ruolo delle risposte neuromorfiche nella capacità di selezionare efficientemente i segnali fisiologici, separandoli nelle loro componenti spettrali rilevanti e instradandoli in modo indipendente, è discusso come passo fondamentale verso l'elaborazione “in-situ” del segnale nei dispositivi neuro-elettronici e viene convalidato tramite esperimenti elettrofisiologici. Inoltre, questa sensibilità alle informazioni codificate temporalmente viene sfruttata in un circuito neuro-ispirato che esegue il riconoscimento di pattern imitando la funzione di integrazione dendritica dei soma biologici. Infine, la risposta neuromorfica viene tradotta con successo in interfacce organiche intra-corticali capaci di mappare i livelli fisiologici di dopamina in vivo, aprendo la strada all’implementazione delle risposte neuromorfiche come nuovi osservabili sensoriali.
Trransient phenomena in organic neuroelectronics: foundamentals and applications
RONDELLI, FEDERICO
2025
Abstract
Organic Neuromorphic Electronics relies on neuron-like transient properties of organic (semi-)conductive devices, such as Electrolyte Gated Organic Transistors (EGOTs). These devices, operated under pulsed or alternate biasing conditions, constitute novel electronic building blocks for applications ranging from biosensing to on-board signal processing. Conversely to conventional inorganic semiconductors, organic materials exhibit unique properties which allow them to interact seamlessly with biological systems through their mixed ionic-electronic conduction. These characteristics make organic neuromorphic devices particularly suitable for bio-interfacing applications, where they can sense, process, and modulate signals interrogating ions and molecules dynamics, which constitute the main information transmission paradigm in biological systems. This PhD thesis builds on these concepts, to extend the understanding of the transient response of electrolyte-gated organic transistor (EGOT) and of two- and one-terminal architectures and to devise novel translational applications of the neuromorphic behavior/response of organic electronic devices. In particular, this work investigates neuromorphic organic devices with the dual purpose of gaining understanding and control of their tunability and of developing organic platforms for on-board signal processing, classification and bio-sensing. This lays the foundations for envisioning the integration of organic neuromorphic electronics in novel smart neural interfaces. The versatility of organic-mixed ionic-electronic conductor-based EGOTs for neuromorphic applications is emphasized, demonstrating gate-driven reversible fine-tuning of EGOT plasticity, arising from reversible modulation of its response timescales. Besides, the discussion is extended to ambipolar materials, demonstrating multilevel memory and signal rectification in individual circuit components. The role of neuromorphic responses in determining the capability to efficiently sort physiological signals, separating them in their relevant spectral components and independently routing them, is discussed as a foundational step towards on-board signal processing in neuroelectronic devices and benchmarked in electrophysiological experiments. Additionally, such sensitivity to time-encoded information is leveraged in a neuro-inspired circuit performing pattern recognition by mimicking the dendritic integration function of biological somas is presented. Finally, neuromorphic response is successfully translated to intracortical organic interfaces capable of mapping physiological level of dopamine in vivo, paving the way for the implementation of neuromorphic responses as novel sensing observables.File | Dimensione | Formato | |
---|---|---|---|
TESIfinale_DopoRevisioni_FedericoRondelli.pdf
accesso aperto
Dimensione
7.04 MB
Formato
Adobe PDF
|
7.04 MB | Adobe PDF | Visualizza/Apri |
I documenti in UNITESI sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/20.500.14242/218632
URN:NBN:IT:UNIFE-218632