Identifying relevant machine learning features for multi-sensing platforms is both an applicative limitation to recognize environments and a necessity to interpret the physical relevance of transducers’ complementarity in their information processing. Particularly for long acquisitions, feature extraction must be fully automatized without human intervention and resilient to perturbations without significantly increasing the computational cost of a classifier. In this study, we investigate the relative resistance and current modulation of a 24-dimensional conductimetric electronic nose, which uses the exponential moving average as a floating reference in a low-cost information descriptor for environment recognition. In particular, we identified that depending on the structure of a linear classifier, the ‘modema’ descriptor is optimized for different material sensing elements’ contributions to classify information patterns. The low-pass filtering optimization leads to opposite behaviors between unsupervised and supervised learning: the latter favors longer integration of the reference, allowing the recognition of five different classes over 90%, while the first one prefers using the latest events as its reference to cluster patterns by environment nature. Its electronic implementation shall greatly diminish the computational requirements of conductimetric electronic noses for on-board environment recognition without human supervision.

Conducting Polymer Dendrites (CPD) can engrave sophisticated patterns of electrical interconnects in their morphology with low-voltage spikes and few resources: they may unlock in operando manufacturing functionalities for electronics using metamorphism conjointly with electron transport as part of the information processing. The relationship between structure and information transport remains unclear and hinders the exploitation of the versatility of their morphologies to store and process electrodynamic information. This study details the evolution of CPD's circuit parameters with their growth and shape. Through electrochemical impedance spectroscopy, multiple distributions of relaxation times are evidenced and evolve specifically upon growth. Correlations are established between dispersive capacitances of dendritic morphologies and growth duration, independently from exogenous physical variables: distance, evaporation or aging. Deviation of the anomalous capacitance from the conventional Debye dielectric relaxation can be programmed, as the growth controls the dispersion coefficient of the dendrite's constant-phase elements relaxation. These results suggest that the fading-memory time window of pseudo-capacitive interconnects can practically be conditioned using CPD morphogenesis as an in materio learning mechanism. This study confirms the perspective of using electrochemistry for unconventional electronics, engraving information in the physics of conducting polymer objects, and storing information in their morphology, accessible by impedance spectral analysis.

Conventional electronics is founded on a paradigm where shaping perfect electrical elements is done at the fabrication plant, so as to make devices and systems identical, “eternally immutable.” In nature, morphogenic evolutions are observed in most living organisms and exploit topological plasticity as a low-resource mechanism for in operando manufacturing and computation. Often fractal, the resulting topologies feature an inherent disorder: a property that is never exploited in conventional electronics manufacturing, while necessary for data generation and security in software. In this study, we present how such properties can be exploited to implement long-term and evolvable synaptic plasticity in an electronic hardware. The rich topology of conducting polymer dendrites (CPDs) is exploited to program the non-ideality of their electrochemical capacitances containing constant-phase-elements. Their evolution through structural changes alters the characteristic time constants for them to charge and discharge with the applied voltage stimuli. Under a train of voltage spikes, the evolvable current relaxation of the electrochemical systems promotes short-term plasticity, with timescales ranging from milliseconds to seconds. This large window depends not only on the temporality of the voltage pulses used for reading but also on the structure of a pair of CPDs on two electrodes, grown by voltage pulses. This study demonstrates how relevant physically transient and non-ideal electrochemical components can be exploited for unconventional electronics, with the aim to mimic a universal property of living organisms, which could barely be replicated in a silicon monocrystal.

Conducting polymer dendrite (CPD) morphogenesis is an electrochemical process that enables in materio evolving intelligence in wetware devices. During CPD morphogenesis, voltage transients drive the physical evolution of electrically conductive structures, thereby programming their filtering properties as nonlinear analog devices. Whether studied in an electrochemical experiment or in neuromorphic devices, the dependence of the electrical properties of the electrogenerated structures on the chemical composition of their growth environment is still unreported. In this study, we report the existing interconnection between the nature and concentration of the electrolytes, electroactive compounds and co-solvents and the electrical and electrochemical properties of CPDs in an aqueous electrolyte. CPDs exhibit various chemical sensitivities in water: their morphology is highly dependent on the nature of the chemical resources available in their environment. The selection of these resources therefore critically influences morphogenesis. In addition, the concentrations of the different electrochemical species have varying impacts on growth dynamics, modulating the balance between thermodynamic and kinetic control over polymer electrosynthesis. By correlating the dependencies of these evolving objects with the availability of the chemical resources in an aqueous environment, this study offers guidelines to tune the degree of evolution of electronic materials in water and highlights potential avenues for their application. Such evolving hardware is envisioned to exploit the chemical complexity of real-world environments as part of information processing technologies.

Croissance de dendrites de polymère conducteur influencée par l’électroconvection et étude de leurs relaxations électrochimiques: En 2022, près de 62 millions de tonnes de déchets électroniques ont été produites à l'échelle mondiale, un chiffre en constante augmentation. Cette situation critique invite à repenser la manière dont les circuits électroniques sont conçus, la rigidité des dispositifs actuels les rendant rapidement obsolètes et difficilement recyclables.Dans la nature, les organismes sessiles tels que les plantes ou les champignons poussent en fonction de leur environnement. Le cerveau, performant et économe en énergie, forme des connexions en permanence. Cette flexibilité structurelle apparaît comme une ressource avantageuse pour faire des circuits plus intelligents. Dans cette optique, la croissance dendritique est un mécanisme original et peu coûteux conduisant à des structures ramifiées analogues à celles des plantes. Bien connue chez les métaux, elle peut être mise en oeuvre de manière contrôlée avec des polymères conducteurs comme le PEDOT.Ce travail étudie le mécanisme de croissance des dendrites de polymère conducteur (CPDs), en particulier l'influence de facteurs tels que la température, la viscosité du milieu et la nature du contre-ion sur leur morphologie. Des parallèles sont faits avec les dendrites métalliques, offrant une perspective nouvelle sur celle des CPDs. Jusqu'alors exclusivement observées en régime alternatif, il est montré que les CPDs croissent aussi sous tension constante. De plus, des tourbillons électroconvectifs sont observés pendant la croissance. Ces tourbillons prennent une importance particulière dans un montage à deux plaques parallèles, où les signaux de tension donnant habituellement des dendrites donnent des motifs plats. Ces motifs sont explorés afin d'en comprendre l'origine, rapidement reliée à l'électroconvection. Il apparaît plus difficile, dans cette configuration à deux plaques, de faire croître des dendrites en raison des forts flux convectifs agitant la solution. Une modification de la géométrie des plaques a cependant permis de débloquer leur croissance et d'étudier la transition entre motifs et dendrites.Pour finir, les CPDs sont étudiées en tant que composants capacitifs en régime petits signaux au moyen de la spectroscopie d'impédance électrochimique. Un bilan des différentes relaxations électrochimiques existant dans le système est établi. À l'état non connecté, les dendrites se comportent comme un circuit à éléments à phase constante, dont les coefficients de dispersion α semblent corrélés à la morphologie. Ce résultat est confirmé dans le domaine temporel. Enfin, la croissance de dendrite activée par un capteur polymère est étudiée, démontrant la possibilité d'utiliser ces structures pour encoder dans le matériau un historique d'expositions à des espèces volatiles.