The rapidly growing energy demand associated with artificial intelligence and data-intensive technologies is accelerating the search for novel hardware paradigms capable of overcoming the intrinsic limitations of conventional von Neumann computing architectures. In this context, neuromorphic devices, which emulate key features of biological information processing, are emerging as promising candidates for next-generation, energy-efficient computing systems.

A central challenge in the field is the development of materials and device architectures capable of emulating synaptic functionalities such as memory, plasticity, and stimulus-dependent conductance modulation. While oxide-based memristive systems have a relatively high maturity, increasing attention is now being directed toward materials that can couple electrical and optical stimuli, enabling multimodal operation.

In this presentation, we discuss recent progress in the exploration of neuromorphic behavior and switching mechanisms in halide perovskites and low dimensional semiconductors. These material platforms offer a promising route toward adaptive, low-power, and optoelectronically tunable neuromorphic devices.

Halide perovskites are mixed ionic electronic semiconductors in which mobile ions strongly affect charge transport, hysteresis, and resistive switching. This ion-coupled response is attractive for memory and neuromorphic devices, but poorly controlled ion migration can also cause stochastic switching and limited reproducibility. A central challenge is to control ionic motion while preserving its functional role in device operation.

 Here, we present two strategies to manage ionic processes in halide perovskite memristors. At the material level, compositional engineering of lead-free bismuth halide perovskites enables low voltage resistive switching with composition dependent hysteresis. Analysis of time-dependent responses reveals the ionic origin of switching and allows key physical parameters governing the dynamic hysteresis to be extracted.[1] At the device level, interfacial engineering suppresses stochastic filamentary pathways and redirects ion accumulation toward barrier modulation, resulting in forming free and programmable switching with dual inductive characteristics.[2]

 These results show that mobile ions in halide perovskite memristors can be treated as controllable transport variables rather than only as sources of instability. By combining compositional control, interfacial engineering, and physical dynamic interpretation of hysteresis, this work provides a route toward reliable and programmable ion coupled electronic memory and neuromorphic hardware.

Organic electrochemical transistors (OECTs) combine ionic charging of a mixed ionic–electronic channel with electronic current transport, giving rise to transient switching, transfer hysteresis, and synaptic-like responses. However, these phenomena are often studied separately, obscuring their common physical origin and their distinction from memristive behaviour. Here, we establish a unified interpretation of OECT dynamics based on delayed ionic channel charging and its electronic readout.

Using PEDOT OECTs under gate-voltage steps, we show that gate-current transients exhibit a nearly drain-bias-independent decay, enabling extraction of an ionic diffusion time of approximately 30–40 ms. In contrast, the drain-current transient changes strongly with drain-bias magnitude and polarity, reflecting the contribution of lateral electronic transport to the readout of the same ionic charging process. Despite their different temporal morphologies, the characteristic decay times extracted from gate and drain currents remain consistent, confirming vertical ionic diffusion as the intrinsic timescale governing channel charging. [1]

Building on this separation between ionic relaxation and electronic readout, we relate transfer-curve hysteresis to pulse-driven synaptic plasticity within a reduced diffusion–transport model. For a fixed device polarity, operating window, and pulse protocol, the orientation of the transfer loop provides an operational indicator of whether repeated pulses produce potentiation or depression. Both observables arise from the delayed evolution of the same ionic state, while their measured sign is determined by the competition between ionic relaxation and electronic readout timescales. [2]

Finally, comparison with a conductance-activated memristor clarifies that similar pulse-to-pulse current accumulation does not imply the same memory variable. In memristors, the internal state directly controls conductance; in OECTs, the evolving state is reversible channel charging. Consequently, conventional OECT potentiation is intrinsically volatile in the long-time limit unless an additional retained redox, ionic, trapping, or material state is introduced. This framework connects transient response, hysteresis, and neuromorphic function while defining the physical boundary between charging-based transistor dynamics and conductance-state memory.

Electrostatic gating provides a powerful route to interrogate and engineer biological interfaces. In this talk, I will discuss biofunctionalized solid/liquid interfaces as iontronic systems in which ionic screening, interfacial polarization, molecular dipoles, and field-driven rearrangements jointly determine device response.

The focus will be on physisorbed antibody monolayers on gold and their coupling to electronic and optical transducers. Although physisorption is often regarded as less controlled than covalent immobilization, our recent results show that antibody adlayers can form robust, function-preserving biointerfaces whose electrostatic properties are highly informative. Kelvin probe force microscopy, polarization-modulation infrared reflection–absorption spectroscopy, and Raman spectroscopy provide complementary access to the electrostatic and structural state of the biolayer, and to its modulation by ionic, molecular, and electric-field-driven processes.

I will first discuss how antibody adsorption is governed by a spreading–packing competition. Using PM-IRRAS and two-dimensional correlation spectroscopy, the structural evolution of anti-IgM monolayers can collapse onto a surface-density coordinate, linking adsorption conditions to secondary-structure organization. I will then show how electric-field cycling, applied within a non-faradaic window, acts as an annealing-like process for physisorbed antibody layers: it does not simply align molecular dipoles, but reduces the dispersion of biolayer polarization, thereby improving reproducibility across devices. Finally, I will connect these interfacial effects to transduction, including graphene-based optoelectronic platforms in which biomolecular recognition at a remote gold biointerface is capacitively coupled to graphene and read optically through Raman phonon shifts.

This perspective frames bioelectronic sensing as a condensed-matter and interfacial-physics problem, discussing how soft, polar, nanoscale biological matter organizes under ionic, electrostatic, and molecular constraints to generate measurable electronic and optical signals.

Electricity can be generated from ambient heat via the hydrovoltaic effect, which refers to the generation of electrical energy from the interaction of water or aqueous electrolytes with nanostructured surfaces.[1], [2] The transpiration of aqueous electrolytes through nanoporous materials can sustain this voltage generation reliably. The use of a hermetic hydrovoltaic cell (HHC) provides insulation from meteorological variations and results in negligible consumption of the electrolyte. Previously, a sustained voltage generation in HHC has only been achieved by fabricating a heterogeneous wicking bilayer structure consisting of a carbon black film (deposited onto a substrate) layered on top with a tissue paper.[3] In our HHC, comprising of only a homogeneous film of spherical carbon nanoparticles, we sustained voltage output of ~450 mV  for 50 hours. Herein, the effect of electrolyte (ion-size and concentration) and carbon nanoparticle surface chemistry on the performance of hermetic hydrovoltaic cell will be discussed. The concentration of electrolyte and nanoparticle surface chemistry provide means to control the voltage from ~100 mV to ~450 mV in a single HHC. The obtained results can be explained within the framework conventional streaming potential and pseudo-streaming mechanisms. Our findings provide an avenue for exploiting low-grade ambient heat for electricity generation under high humidity conditions.

Ion transport is increasingly recognized as a key physical mechanism for unconventional computation, enabling nonlinear dynamics, memory, and adaptive behavior beyond conventional electronics. Across diverse iontronic platforms—including mixed ionic–electronic conductors, ionic nanopores, and semiconductor devices with ionic or capacitive dynamics—information can be processed through self-sustained oscillations, spiking, synchronization, and phase relationships.

In this talk, I present a unified dynamical framework showing that these seemingly different systems share a common architecture: a fast electronic or transport instability coupled to a slow ionic recovery process arising from ion migration, charge accumulation, or configurational relaxation. This slow–fast interplay drives self-sustained oscillations through a Hopf bifurcation and provides universal design rules for tuning frequency, waveform, synchronization, and responsiveness.

Recent work illustrates this approach across three representative iontronic platforms: ultrasmooth silicon thyristor oscillators, where capacitive charge storage acts as the slow variable; single-transistor organic electrochemical oscillators, where ionic transport modulates electronic conduction; and rectifying nanopore oscillators, where ion accumulation and deactivation generate autonomous liquid-phase dynamics. Despite their different physical implementations, all exhibit the same bifurcation structure and can be designed within a common nonlinear dynamical framework.

These results establish ionic dynamics as a unifying physical principle for oscillator-based computation, bridging solid-state, organic, and fluidic technologies and providing a general strategy for developing scalable, energy-efficient neuromorphic and unconventional computing systems.

Ions are among the most familiar and fundamental components of matter. All living organisms rely on ions to sustain essential biological functions. In humans, sodium and potassium ions regulate body-fluid balance and support the circulatory system, while calcium ions play key roles in sensing information such as pain. Ions are also involved in the repair and maintenance of biological functions, including the formation and regeneration of bones and teeth. In this sense, ions are deeply involved in almost all complex life processes.

Living organisms also possess sophisticated receptors, such as mechanoreceptors and photoreceptors. These receptors convert physical stimuli applied to the body surface into chemical signals, which are then transmitted to the central nervous system through sensory nerves. In other words, biological systems can convert mechanical and optical energy into chemical signals through highly controlled ionic processes. This demonstrates that living organisms possess remarkably advanced ion-control mechanisms, suggesting that the potential of ions is far more diverse than currently imagined.

However, despite this biological sophistication, artificial ion-control technologies remain limited. Practical applications of ions have so far been largely confined to energy-storage functions, such as batteries and capacitors, where electrochemical reactions and ion accumulation/release are utilized. Recently, we demonstrated a new function of ions by fixing ionic distributions to form an electric-double-layer electret, in which electric charge can be retained semi-permanently.

More recently, we have discovered that ions can be controlled by cross-correlated energies, such as light, mechanical force, and magnetic fields, which are conventionally considered unrelated to ion control. In these phenomena, the application of mechanical, optical, or magnetic energy to the electric double layer of ionic liquids induces rapid and soft changes in the molecular structures of the constituent cations and anions, as well as in their aggregates. These changes also modify the distribution of positive and negative ions within the electric double layer, resulting in variations in ionic density and, consequently, changes in the electric-double-layer capacitance.

In this talk, I will present emerging functionalities of ions enabled by cross-correlated energies and discuss their potential applications in future ion-based devices.

Self-sustained oscillatory devices are emerging as key building blocks for iontronic, neuromorphic, and unconventional computing architectures, where information can be processed directly through the nonlinear dynamics of materials and devices. In these systems, variables such as phase, frequency, synchronization, entrainment, and spiking activity can encode and transform information in real time, enabling new approaches to tasks such as optimization, pattern classification, and temporal inference. [1] A common route to generate self-oscillations in electronic and iontronic devices is the presence of negative differential resistance (NDR), which provides the active feedback required to destabilize a stationary state and sustain a limit-cycle oscillation. However, the observation of NDR alone is not sufficient to predict oscillatory behavior, since the onset of instability depends on the coupling between the active material dynamics, internal relaxation processes, and external circuit elements. [2,3]

Here, we show that impedance spectroscopy provides a powerful experimental framework to identify, interpret, and predict the emergence of self-sustained oscillations in S-type NDR devices. Using a thyristor-based oscillator as a model system, we first demonstrate that frequency-domain measurements reveal direct signatures of nonlinear dynamical phenomena such as stability loss, Hopf bifurcation, autonomous oscillations, entrainment, phase locking, and divergence of the impedance response near the oscillatory threshold. By mapping impedance spectra onto an equivalent-circuit representation, we connect experimentally accessible circuit parameters with the local linearized dynamics of the system. This establishes a bridge between impedance spectroscopy, nonlinear bifurcation theory, and classical control concepts such as transfer-function stability criteria. [4]

Building on this framework, we further show that impedance spectroscopy can be used not only to detect dynamical instabilities, but also to reconstruct the experimental stability map of a self-oscillatory device. From the equivalent-circuit parameters, we define a stability time constant τ_s obtained from the parallel coupling of the dynamic relaxation branches of the system. This time constant is directly related to the trace of the Jacobian of the underlying dynamical equations, providing an experimental criterion for whether small perturbations around a stationary state are damped or amplified. Its sign therefore determines the transition between stable operation and self-sustained oscillations, while its zero-crossing identifies the bifurcation boundary.

By varying the external capacitance coupled to the device, we experimentally construct the current–capacitance (I₀, C₀) bifurcation diagram and validate the predicted stable and oscillatory regimes with time-domain voltage measurements. This approach allows the stability landscape of the device to be mapped without requiring a complete microscopic nonlinear model. More broadly, it enables the identification of the physical or physico-chemical branch responsible for the destabilizing feedback, including ionic motion, delayed conductance activation, interfacial charge storage, or other emergent material processes.

These results establish impedance spectroscopy as a stability-resolved diagnostic tool for iontronic and mixed ionic-electronic devices. The method provides a practical route to discover hidden instabilities, design oscillatory regimes, and control nonlinear dynamics in emerging material-based hardware for neuromorphic and physical computing. [3]

Neuromorphic systems often emulate neuronal firing by combining memory elements with dedicated reset mechanisms [1] or strongly nonlinear switching devices [2]. In this work, we show that repetitive spike generation can emerge from a considerably simpler framework [3]. We investigate a threshold-responsive memristive circuit composed of a passive RC network and a dynamic conductance element described by a single internal variable [4]. When driven by periodic voltage pulses, the system displays stable sequences of current spikes despite the absence of reset operations, oscillatory feedback loops, or negative differential resistance.

Through numerical and analytical analysis, we find that spike generation is governed by the competition between memory evolution and charge accumulation processes. The firing regime is not determined by static device properties alone, but by the synchronization of several characteristic timescales associated with relaxation, charging, leakage and external stimulation. Within a specific dynamical window, these processes self-organize into repetitive cycles of energy storage and rapid release, producing well-defined spiking activity.

A notable feature of the mechanism is the appearance of an effective separation between fast and slow dynamics, even though the memristive element contains only a single intrinsic memory timescale. This emergent behavior results from the nonlinear interaction between the internal state and its voltage-dependent operating point. We derive simple criteria that predict the onset and stability of the firing regime and identify the conditions leading to spike suppression or irregular activity.

These results suggest that neuronal-like firing can be achieved in a broader class of adaptive electronic materials than previously assumed. By emphasizing dynamical operation rather than static switching characteristics, this work provides new guidelines for the design of compact neuromorphic hardware based on memristive technologies.

I will introduce an electric-field-sensitive solid polymer electrolyte for which the electric-double layer (EDL) dynamics (i.e., formation and dissipation) can be tuned by an applied voltage. The electrolyte is designed to both facilitate ion motion and undergo the Menshutkin reaction, triggered by the large electric fields generated by the EDL itself. For programming voltages larger than 2 V, the reaction crosslinks a copolymer, trapping ions at the interface between the electrolyte and the channel of a 2D-dimensional crystal transistor, giving rise to persistent channel doping that remains after grounding the gate terminal.  The EDL dynamics can be tuned by tuning the magnitude of the programming voltage. Low-temperature, dual-gated measurements confirm that ion trapping leads to non-volatility by distinguishing mobile, bulk ions from those trapped at the interface. The programmed states show an average retention of the induced sheet-carrier density of 80% over seven weeks at ambient conditions. Such functionality could be applied to temporal matching for neuromorphic computing.

Plasmonic and potentiometric platforms are well-established label-free approaches for immuno and molecular detections. However, when operated without amplification strategies, their analytical sensitivity has traditionally been limited to the nanomolar concentration range. In this lecture, I will discuss a comprehensive physical and statistical framework showing that biosensing responses over an exceptionally broad dynamic range, from nicromolar down to 10-zeptomolar concentrations, are governed by two distinct sensing regimes.

The study focuses on centimeter-scale sensing interfaces densely functionalized with pH-conditioned recognition layers, where electrostatic and ion-mediated effects play a central role in transducing molecular recognition events. At ultralow analyte concentrations, the sensor response is no longer described by ensemble-averaged affinity interactions. Instead, it is dictated by Poisson statistics associated with stochastic single- and few-molecule capture events occurring on large-area interfaces. In this regime, each binding event can trigger a local pH-conditioned mediated dielectric and electrostatic reorganization of the interfacial layer, producing a measurable plasmonic or potentiometric signal.

This mechanism enables direct, label-free detection at concentrations as low as 10–100 zM with 99% statistical confidence, while keeping false-positive and false-negative rates below 1%. As the analyte concentration increases, the sensing behavior progressively evolves toward a classical Maxwell–Boltzmann regime, characterized by conventional binding isotherms and ensemble-averaged equilibrium affinity.

By explicitly accounting for noise, uncertainty propagation, and statistically robust decision criteria, the full binding curves can be modeled across both regimes, allowing the extraction of equilibrium parameters, including an apparent equilibrium constant specific to the ultra-dilute limit. These results clarify some key aspects of the physical origin of extreme sensitivity in large-area biosensing platforms and highlight the role of electrostatic, ion-mediated interfacial phenomena in the design of next-generation single-molecule sensors operating without labels.

Mechanistically, the two regimes arise from the responsive nature of the pH-conditioned capturing layer. At ultralow antigen concentrations, a single antigen–antibody binding event is proposed to nucleate a localized domain where neighboring antibodies undergo collective dielectric and electrostatic rearrangement, likely involving the cooperative hydrogen-bond networks. This converts a discrete molecular event into a mesoscopic dielectric perturbation detectable by plasmonic or potentiometric transducers. At higher concentrations the formation of a continuous layer of target molecules that are affinity bound to the capturing layer is seen and the response is governed by ensemble-averaged affinity-bindings.

Since the discovery of graphene, two-dimensional (2D) materials have become a major research platform for exploring novel quantum phenomena and emerging device functionalities. While early studies focused on atomically thin crystals exfoliated from layered van der Waals (vdW) materials, recent research has increasingly emphasized artificial vdW heterostructures, where interlayer stacking and moiré superlattices provide powerful routes to engineering electronic states. These advances have established structural engineering as a central concept in the design of quantum materials.

In this talk, I will introduce an alternative strategy that exploits another unique structural degree of freedom inherent to layered materials: the van der Waals gap. By introducing and controlling ions within this gap, it becomes possible to tailor electronic structures and magnetic states beyond what can be achieved through stacking engineering alone. Intercalation has long been recognized as an effective means of modifying layered transition-metal dichalcogenides, while electrochemical ion gating offers a reversible and continuous approach for tuning carrier density and correlated electronic states. Together, these techniques provide a versatile platform for investigating and controlling magnetism in layered materials.

Our group has recently developed molecular beam epitaxy (MBE) techniques for the layer-by-layer growth of intercalated van der Waals materials, enabling precise control over crystal structure and intercalant concentration. Combined with post-growth thermal annealing under ultrahigh vacuum and electrochemical Li-ion gating, these epitaxial materials provide an ideal platform for systematically exploring gate-controlled magnetic phenomena.

The presentation will highlight two representative classes of layered magnetic van der Waals materials with distinct electronic structures. In one class, electrochemical ion gating is employed to tune carrier density in systems hosting semimetallic bands near the Fermi level, providing insight into the role of itinerant carriers in stabilizing ferromagnetism. In the other, ion gating is applied to magnetic materials exhibiting flat electronic bands, where carrier modulation offers a unique opportunity to investigate the interplay between electronic correlations and antiferromagnetic order. These studies demonstrate how electrochemical control of carrier density serves as a powerful tool for uncovering the microscopic mechanisms governing magnetism in layered van der Waals materials with diverse electronic structures.

By highlighting recent progress in epitaxial growth, ion engineering, and electrochemical control, this talk will illustrate how the van der Waals gap can serve as a new design space for quantum materials. The ability to manipulate ionic configurations and carrier density in a controlled and reversible manner opens promising opportunities for realizing emergent magnetic phases and developing next-generation spintronic functionalities.

Innovative nanowire-based prototypical device architectures with potential applications encompassing quantum technologies  [1] and nanoelectronics [2,3]  have been recently enabled by iontronics [4], combining semiconductor nanowires with (poly)electrolytes where ions are free to move upon electrical or thermal biases. These devices are multi-gate nanotransistors which exploit electrolytes as dielectrics to build up electric double layers at the nanowire-electrolyte interface, providing ultra-intense local electric fields that can be tailored to control the transport properties in the nanowire. Implementing this paradigm, ion-gated and ion-sensitive nanowire transistors were proven to enable alternative measures of electrical parameters as well as novel architectures for thermoelectrics, and more recently have been proposed for zero-dimensional density of states engineering in InAs nanowires, namely, for engineering the iontronic nanowire quantum dot. Notably, such quantum features observed at low temperature have been very recently identified as potential booster of neuromorphic fiunctionalities.  brief review of nanowire iontronics is discussed and the novel concept of quantum iontronics in nanowire-based devices is reported.

This study reports the first experimental demonstration of the confinement of gramicidin, a biological ion-channel peptide with a β-helical structure, inside a single-walled carbon nanotube (SWCNT). The main objective is to determine whether such a biological ion channel can be stably integrated into a solid nanochannel while preserving its function-like biological structure – an essential step toward the development of artificial conductive nanochannels with high ion permeability and selectivity for applications such as iontronics, desalination, and blue energy harvesting [1–3].

Two sets of SWCNTs, i.e., e-DIPS (Meijo Nano Carbon Co., Ltd., Japan), whose diameters range from 1.8 to 2.5nm, have been studied. The gramicidin used is provided by Aldrich (ref.G5002) and is made of 80-85% gramicidin A, 6-7% B, 5-14% C, and <1% gramicidin D. Its native form, gramicidin A (gA) is a “head-to-head” dimer of β-helices that selectively conducts monovalent cations while blocking divalent ions. [4] Its outer diameter, close to 2 nm, closely matches the inner diameter of the SWCNTs selected in this study.

The SWCNTs are impregnated with gA in pure ethanol. First, it is observed that gA acts as a surfactant, promoting the stable dispersion of SWCNTs. The resulting gA–SWCNT hybrid system is characterized using a combination of techniques: small-angle X-ray and neutron scattering, Raman spectroscopy, high-resolution Transmission Electron Microscopy, and water sorption isotherms. All results obtained with these techniques are consistent and demonstrate a strong interaction between gA and SWCNTs: gA adsorbs onto the surface and inserts into the tube’s channel.

In conclusion, this work provides the first experimental evidence that a β-helical biological ion channel can be confined within SWCNTs of compatible diameter. The results open promising perspectives for the design of bio-inspired hybrid materials that combine the exceptional mechanical and electronic properties of carbon nanotubes with the high selectivity and efficiency of biological ion channels.

 

References:

[1]. Xin, W., Fu, J., Qian, Y. et al. (2022) Biomimetic KcsA channels with ultra-selective K+ transport for monovalent ion sieving. Nat Communю 13, 1701.

[2]. Ye, T., Hou, G., Li, W. et al. (2021) Artificial sodim-selective ionic device based on crown-ether crystals with subnanometer pores. Nat Commun. 12, 5231.

[3]. Chen Q. (2025). Understanding Protein Adsorption on Carbon Nanotube Inner and Outer Surfaces by Molecular Dynamics Simulations; Langmuir 41, 6, 4318.

[4]. Kelkar D.A., Chattopadhyay A. (2007) The gramicidin ion channel: A model membrane protein. Biochimica et Biophysica Acta-Biomembranes. 1768(9), 2011.

The energy-efficient manipulation of magnetic textures such as domain walls and skyrmions is central to emerging spintronic memory technologies and is opening new opportunities for unconventional computing. Key challenges remain, however, including domain wall pinning, controlled skyrmion nucleation and propagation, and the high current densities typically required to manipulate these magnetic textures.

Magneto-ionics offers a powerful route to address these challenges by dynamically tuning magnetic anisotropy with voltage, thereby controlling the nucleation, propagation, and dynamics of magnetic textures. In this talk, I will show that using ionic liquid gating non-volatile magneto-ionic effects and volatile charge effects can be combined to facilitate current-driven skyrmion nucleation in magnetic tracks. In this device design, rather than occurring uniformly, skyrmion nucleation propagates as a front driven by an electric-field gradient coupled to the spin-orbit torque current pulse. This provides a means to control not only the nucleation but also the position of a skyrmion front along a track using an electric field and a single current pulse.

I will also present magneto-ionic devices in which voltage control of magnetic anisotropy enables reversible tuning of the dynamics of three-dimensional chiral magnetic textures in dipolarly coupled multilayers. Modifying the anisotropy of the top magnetic layer strongly influences the spin structure throughout the entire multilayer stack, providing a versatile approach to engineering 3D magnetic texture dynamics for voltage-programmable spintronic devices.

Finally, I will demonstrate that a volatile gate-induced reduction of magnetic anisotropy can create a time-dependent switching probability driven by spin-orbit-torque-induced domain wall motion. This enables a single device to combine long-term magnetic information storage with short-term ionic update eligibility, a functionality particularly attractive for reward-based learning schemes.

These findings highlight the versatility and promise of magneto-ionic devices for controlling the dynamics of magnetic textures and for the design of spin-dependent neuromorphic hardware

Bursting responses play a key role in spiking neural networks and neuromorphic computing, carrying richer temporal structure than isolated spikes and improving the reliability of neural communication. Since synaptic transmission is probabilistic, clustered spikes naturally increase the probability of successful downstream activation. Unlike conventional clock-driven computing, neuromorphic systems operate asynchronously in an event-driven manner,[1,2] enabling low-latency, energy-efficient processing while facilitating the integration of heterogeneous components without requiring clock synchronization.[3,4] These characteristics make them particularly well suited for real-time edge intelligence in dynamic environments.

Controlled bursting emerges in a negative differential resistance (NDR) system when it is biased near the folding point and driven by time-varying inputs. The nonlinear device response, together with external modulation, generates stable spike clusters whose temporal structure can be tuned through the input amplitude, frequency, and bias conditions. Consequently, both the burst rate and the number of spikes per burst are controllable.

This dependence on multiple control parameters enables flexible temporal encoding, making the system suitable for asynchronous signal classification based on dynamical patterns rather than fixed-rate sampling. Burst length and firing activity can be continuously adjusted through the operating conditions.

When the device is driven in the oscillatory regime near a Hopf bifurcation, it effectively suppresses high-frequency noise while amplifying weak coherent signals, enabling the detection of signals buried in noise at signal-to-noise ratios as low as 1/1000. This noise-filtering capability arises from relaxation toward stable attractors, which attenuate stochastic fluctuations while reinforcing coherent input-driven transitions.

Artificial neural network (ANN)-based computing can provide excellent learning, classification, and inference characteristics that are close to, and in some cases beyond, those found in natural intelligence (i.e., the human brain), whereas the enormous amounts of power required by ANN are far higher than that required by human beings. To overcome the low energy efficiency of ANN computing, physical reservoir computing (PRC) is particularly attractive because it can significantly reduce the computational resources required to process time-series data by leveraging the nonlinear responses of a ‘reservoir’ (a material or device acting as a dynamical system) to input signals. Recently, we have developed high-performance PRC devices based on iontronic phenomena. One example is an ion-gating reservoir (IGR), which utilizes ion-electron coupled dynamics in the vicinity of a solid electric double layer at the diamond/Li+ solid electrolyte interface [1]. The edge-of-chaos state of the IGR enabled the best computational capacity. Furthermore, the IGR, consisting of graphene and ion-gel, demonstrates an exceptionally broad responsive range, from 1 MHz to 20 Hz, while maintaining a high information processing capacity and adaptability across multiple time scales [2]. The IGR achieved deep learning (DL)-level accuracy in chaotic time series prediction tasks while reducing computational resource requirements to 1/100 of those needed by DL. Another example is a magnonic PRC device that utilizes the high-speed nonlinear dynamics of interfered spin waves in a ferrimagnetic Y3Fe5O12 single crystal [3].

The high PRC performance of the IGR is particularly advantageous for application to in-sensor computing, which is a system-level paradigm in which the sensor element itself not only converts an external stimulus into an electrical signal but also carries out local data processing (e.g., feature extraction, thresholding, logic, neuromorphic operations) on that signal [4]. One example is a real-time spoken-digit recognition system that directly processes throat vibrations by integrating a π–gel-electret mechano-electric generator (MEG) sensor with a multi-channel ion-gel/graphene IGR [5]. This hybrid MEG–IGR system achieves an impressive real-time spoken-digit recognition accuracy of 96.8%, demonstrating that IGR can efficiently extract discriminative spatiotemporal features from raw biomechanical signals. Another example is a high-speed gas classification with a membrane-type surface stress sensor. The IGR, with its broad responsive range and small volume, can serve as a versatile platform for high-performance in-sensor computing with various sensors.

Memristive devices have significantly developed in the last decade, expanding their application horizon much beyond memory applications. Especially important is their functionalities as artificial neurons and synapses, making them promising building units for the next generation bio-inspired neuromorphic hardware. However, other applications such as selectors, HF switches and sensors are also intensively studied and reported. Recently, we demonstrated a new application of memristors using the quantum conductance levels as standard for resistance in accordance with the new directions given by SI. Here the main challenge was to stabilize the quantum conductance levels over time for reliable comparison/verification.

In this contribution the main criteria for selection of materials systems for quantum memristors will be discussed. The advantages and disadvantages of different switching layers and active electrodes will be highlighted in terms of selection criteria for obtaining reliable QPC. Effects of impurities and protons in the processes and stability of the resistances will be critically overviewed.

The method of electrochemical polishing will be introduced, that allows to ensure stable, reproducible and adjustable quantum point contacts. The fundamentals of this procedure will be explained and its application to nanoscale electrochemical memristive devices will be critically discussed.

and the main principle of using the QPC as a standard for resistance will be provided. Different approaches for stabilization of the quantum steps with low number of participating atoms (5 <) will be shown. Fundamental materials principles and processes will be highlighted.

The THz spectral range is characterized by the low-energy excitations of condensed matter — phonons, polarons, excitons, electron-phonon coupling — while, from the technological point of view, it bridges the domains of electronics and photonics. In the first part of this talk, I will introduce terahertz time-domain spectroscopy (THz-TDS), in which the coherent detection of few-cycle THz pulses gives direct access to both the amplitude and phase of the transmitted field, and show how it can serve the field of iontronics, from fundamental studies of ion-gated materials to applicative aspects such as modulators and transistors [1].

Despite their scientific and technological appeal, THz waves have a major drawback coming from the absence of optoelectronic techniques and devices that can manipulate such light waves. To solve this issue, one of the most promising approaches relies on the use of metasurfaces, engineered composites whose optical properties can be specifically tailored to interact with THz waves. Research is now focused on designing reconfigurable metamaterials, also known as metadevices, able to modulate the amplitude, phase, and frequency of THz pulses. In our work, we explore for the first time the functioning and the effectiveness of an organic semiconductor-driven metadevice based on a matrix of metal Split-Ring Resonators, which can modulate the amplitude of a THz pulse passing through them. The modulation capabilities come from the electrically driven change of charge carrier density in the organic semiconductor, which enables the metadevice to act as an optical transistor, varying THz transmission around a specific frequency (~0.7 THz) with modulation depths of approximately 65%. These performances, which are comparable with current state-of-the-art technologies but with a way lower driving voltage (<1 V), result from the unique 3-dimensional charge modulation properties of a class of organic mixed ion-electron conductors based on conjugated polymers with glycolated sidechains, such as p(g2T-TT). We also show that it is possible to shift towards mass-scalable and cost-effective manufacturing techniques, exploiting high-throughput deposition methods for both the metamaterial matrix and the organic semiconductor, on either rigid or flexible plastic substrates, without losing modulation efficiency [2].

In the second part of the talk, I will show how iontronics can also control the ultrafast optical response of two-dimensional crystals, probed using mid-infrared pulses. Ionic liquid gating enables the accumulation of large charge carrier density upon applying moderate gating fields. Here, it is used to tune the Fermi level EF of single-layer graphene to hundreds of meV, demonstrating the strong tunability of its linear and nonlinear optical response in the mid-infrared range, where absorption is governed by the interplay between interband and intraband transitions. A femtosecond pump pulse photoexcites graphene, creating a charge carrier distribution at an elevated electronic temperature, and the cooling dynamics is tracked by monitoring the changes to graphene transmission with mid-infrared probe pulses of tunable wavelength between 3 and 7 microns. Ionic gating enables EF to cross the Pauli blocking threshold for interband absorption and switch the sign of the transient optical response - from a photobleaching to a photoinduced absorption - while higher doping levels suppress hot electron cooling via optical phonon emission. Interestingly, the femtosecond time resolution reveals a sub-picosecond, ultrafast sign reversal in transient transmission near the Pauli blocking threshold. While this phenomenon is of fundamental interest for understanding the properties of Dirac fermions, the electrostatic control of recovery dynamics and transmission opens exciting prospects for graphene-based ultrafast optical modulators and tunable saturable absorbers.

Monitoring the thermal history of perishable food products along the cold chain is critical to guarantee quality and safety for consumers. Time-temperature integrators (TTIs) offer solutions to guarantee cold chain efficiency, improve safety, and reduce food waste. Here, we present a sustainable TTI device based on the irreversible dehydration of a hydrogel, made of reduced graphene oxide and PEDOT:PSS, when exposed to varying environmental conditions. Dehydration kinetics of the hydrogel are investigated over a range of temperatures and relative humidity relevant to food-storage applications. The TTI response is quantified by the mass loss evolution which is interpreted by means of a kinetic model leveraging the initial hydrogel mass, the temperature and the relative humidity. The Arrhenius-like energy barrier is greater than the water binding enthalpy within the hydrogel, and the dehydration rate of the TTI device is finely tuned through the initial mass to mimic the microbial growth curve on fresh sea bream fillet. Deviations of the TTI from its evolution upon optimum conditions reveal the occurrence of a breach in the cold chain. We demonstrate that it is possible to translate the TTI assessment on a smartphone by imaging the evolution of shrinking of the projected area.

Electrolyte-gated organic transistors (EGOTs) emerged as a foundational platform for organic bioelectronics [1]. Their gating mechanism is governed by the electronic structure and the ion permeability of the organic semiconductor (OSC) [2]. In ion-permeable OSCs, electrolyte ions penetrate the semiconductor bulk and (de-)dope the active material, giving rise to the mixed ionic-electronic conductivity of organic mixed ionic–electronic conductors (OMIECs) [3]. EGOTs based on OMIECs are known as organic electrochemical transistors (OECTs) [4], in which the channel conductivity is modulated by gate-driven ionic diffusion.

The workhorse material for OECTs is poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS), a p-type OMIEC operating in depletion mode under positive gate–source biases (VGS). The current-modulation mechanism commonly invoked for PEDOT:PSS relies on an electrochemical (de-)doping process in which monovalent cations (M+) diffuse from the aqueous electrolyte into the OSC microstructure. There, cations compensate the negative charge of the PSS- dopant counterions, with PEDOT+ that is reduced to its neutral state PEDOT0 [5]. As a result, the number of holes in the channel decreases, leading to a drop in the drain current, and the device reaches the OFF state [4].

The present work expands this picture by showing that the chemical identity of the monovalent alkali cation governs the gating effect. We describe the electrolyte/PEDOT:PSS interface as the concurrent interplay of three coupled processes: an acid-base equilibrium, a complexation reaction between the cations and the deprotonated sulfonate groups, and a cation-proton exchange at the protonated sulfonate groups. Solving this system, we derive a closed-form expression for ΔVT versus cation concentration, whose fit to the experimental data provides the cation-specific complexation constant (Kc). Across the alkali series, Kc follows a non-monotonic, V-shaped trend. When plotted against the hydrated mass of the cation, defined as the ion with its first hydration shell upon entering the polymer [6], this trend collapses onto a linear correlation.

Furthermore, we extended the study to the interaction between the PEDOT:PSS and the hydronium ion (viz., in acidic media). We demonstrated that the acid association constant between H3O+ and PSS- is consistent with the complexation-constant sequence established for the hydrated alkali cations. Notably, below pH 2, the hydrogen evolution reaction (HER) sets in, accompanied by the (over)oxidation of PEDOT:PSS channel. The resulting faradaic current markedly reshapes the transfer characteristics, confirming that redox-probe-triggered faradaic reactions at the PEDOT:PSS channel distort the transfer characteristics and transconductance (gm) profiles, as already demonstrated [7].

Taken together, these results broaden the purely faradaic description of OECT operation into a framework that rationalizes ion–polymer interactions, identifying ion hydration as the physicochemical descriptor that governs the complexation between alkali cations and the sulfonate groups of PEDOT:PSS.

Physical reservoir computing (PRC) is a neuromorphic computing framework that exploits the intrinsic nonlinear dynamics and fading memory of physical systems to process time-series information. Because only a simple output layer is trained digitally, PRC can reduce computational cost compared with conventional deep-learning approaches, making it attractive for edge applications where device volume, power consumption, and computing resources are limited. Ion-gating reservoirs (IGRs) are a promising class of PRC devices that use ion-gating-induced material property modulation in transistor structures.[1-5] Their large nonlinear transient responses enable efficient physical feature extraction from temporal signals.

Here, we report an ion-gel/graphene electric-double-layer transistor-based IGR for compact and computationally efficient processing of sensor data at the edge.[4,5] As distributed sensing systems generate continuous physical signals containing both fast and slow components, it is increasingly important to extract useful temporal features close to the sensor while reducing data-transfer volume, latency, power consumption, and digital computational load. The proposed IGR exploits the coexistence of ultrafast electric-double-layer charging at the ion-gel/graphene interface and slower surface-related memory processes on graphene. This combination provides nonlinear transformation and multi-timescale memory within a compact transistor structure, making the device suitable for broadband temporal signal processing.

In the fabricated six-channel ion-gel/graphene device, temporal input signals are applied to a common gate, and the resulting drain-current transients from multiple graphene channels are used as reservoir states. Owing to the ambipolar transport of graphene, channel-dependent geometries, and voltage-tunable ion dynamics, the device produces diverse transient responses over an ultrawide operating range from 1 MHz to 20 Hz, with a minimum relaxation time of 99 ns. These characteristics allow the IGR to respond to high-frequency signal components while retaining slower memory effects required for history-dependent information processing.

The reservoir performance was evaluated using benchmark time-series tasks, including nonlinear autoregressive moving-average tasks and chaotic time-series prediction based on the Mackey–Glass equation. The device achieved prediction accuracy comparable to deep-learning-based approaches, while reducing the computational load by approximately two orders of magnitude because the nonlinear feature expansion is performed physically by the device and only a simple readout layer is trained digitally. These results demonstrate that iontronic device dynamics can serve as a compact physical feature extractor for sensing data. The proposed IGR provides a promising platform for sensor-integrated edge AI hardware that combines ultrawideband operation, high temporal-processing performance, and lightweight computation for real-time processing of sensor signals.

We report a method for producing an array of fifty two ion-sensitive PEDOT:PSS organic electrochemical transistors on a glass coverslip, each featuring an integrated fluoropolymer microwell sealed with lipid bilayer into which membrane proteins can be inserted for simultaneous electrical and fluorescence microscopy studies. To demonstrate capability, we fill the microwells with an ‘inner’ phosphate assay buffer solution containing 10 µM Alexa-488 dye and 50 mM KCl, seal the microwells with lipid bilayer using an aqueous-organic-aqueous liquid exchange technique, and then fill the common flow-cell volume above the sealed microwells with a dye-free ‘outer’ phosphate assay buffer containing 100 mM KCl. We insert α-hemolysin, which embeds into the lipid bilayer forming a heptameric pore with diameter∼2.6 nm. The pore allows K+ ions to diffuse into the microwell and Alexa-488 dye molecules to diffuse out of the microwell producing a corresponding drop in transistor conductance and well fluorescence intensity, respectively. These two signals occur at different timescales, consistent with the known size difference between K+ ions and Alexa-488 molecules. Our approach to fabricating microwell arrays with PEDOT:PSS OECTs incorporated into the bottom of selected microwells distributed in the array is both scalable and versatile, opening a path to studies using larger arrays and with other membrane proteins embedded in the lipid bilayer sealing the microwells.