Dynamic Memory Effects in Aqueous Ion Transport through 2D Nanopore Arrays

Nanofluidic ion-conducting devices with built-in memory are gaining attention as potential components for brain-mimetic computing systems. Here we present examples of the dynamic memory effects in the aqueous ion transport through 2D nanopore membranes under alternating bias voltages. In the first example, we explore the transport of KCl and NaCl salt mixtures through arrays of graphene-embedded crown-like pores, known to selectively trap aqueous K⁺ ions. In this system, trapped K⁺ ions act as pore cloggers, while Na⁺ ions are the main charge carriers. We demonstrate that the dynamically changing state of the system defined as the ratio of pores <i>unblocked</i> by K⁺ (and thus the number of conductive paths available to Na⁺) is marked by a basic time delay. This delay is shown to result in distinct memristive ion transport under alternating voltage bias.<br/>In the second example, we investigate capacitive transport of water-dissociated RbCl in the presence of sub-nm pore arrays in hexagonal boron nitride (hBN). In contrast to the first example, the pores are impermeable to Rb⁺and such a system behaves as a capacitor with a built-in chemical barrier, causing current spiking. The external bias voltage is shown to induce ordered adsorption of Rb⁺on one side of the membrane, representing the charging phase. Alternating the voltage direction triggers rapid ion discharge, leading to current spikes with current polarity dependent on the prior input.<br/>The physical processes described in this work are remarkably illustrative, potentially extending beyond the presented examples. By focusing on the mechanisms, we provide a clear insight into the design of nanofluidic systems capable of memristive and spike ion transport within realistic timescales.