Iontronic Neuromorphic Signaling with Conical Microfluidic Memristors

Synaptical Tunability of Multipore Nanofluidic Memristors

“…Remarkably, the currents follow the ionic mobility series Li + < Na + < K + (Figure d), suggesting that the validity of general ionic conduction principles should facilitate reliable operation. These experimental facts, together with the above frequency sensitivity and pinched current hysteresis, are typical traces of memristor systems and show reminiscences to other biomimetic pores and biological ion channels. ,,,,,, …”

“…Ion-based neuromorphic devices have attracted wide interest both in fundamental and applied chemistry audiences. − The basic building block of neuromorphic signal processing is the memristor. − Here, we describe a multipore nanofluidic memristor with conical pores on a polymeric substrate that shows a wide range of ionic conduction properties, including current rectification. This electrochemical memory resistor exhibits a robust history-dependent behavior based on the electrical interaction between the functionalized charges on the conical pore surface and the nanoconfined ionic solution. − We show that the memristor active response can be switched in current and polarity by pH control, which provides additional functionality for chemical computation and neuromorphic applications.…”

Iontronic Nanopore Model for Artificial Neurons: The Requisites of Spiking

“…Currently iontronic and nanofluidic devices are investigated for brain-like computation. , Some arrangements of microfluidic memristors have been suggested for functional iontronic neurons. , These previous works construct the neuron using the combination of different memristive channels in antiparallel polarity, following closely the structure of the HH model shown in Figure a, that represents the dynamics of biological neurons …”

“…22,23 Some arrangements of microfluidic memristors have been suggested for functional iontronic neurons. 24,25 These previous works construct the neuron using the combination of different memristive channels in antiparallel polarity, following closely the structure of the HH model shown in Figure 1a, that represents the dynamics of biological neurons. Here we show that the high biological plausibility of the HH model is not a necessary requirement to obtain biomimetic action potential oscillations useful to construct artificial neurons for computational purposes.…”

Concluding remarks: Iontronics, from fundamentals to ion-controlled devices – Random access memories

“…Going to salt and water, hence inspired by the brain, would be a great achievement. This begins to be possible, with the reports of various memristive behaviors in ionic systems 5–8 and the very first steps towards computations, like nanofluidic Hebbian learning, 6 or logic computations. 9 So there is a possible road towards nanofluidic computing, even though it involves considerable challenges: 10 in terms of a scale-up strategy for building interconnected nanofluidic circuits, reaching a sufficient complexity, but also in terms of signal processing and transport, in line with the voltage spiking occurring in neurons.…”