Macroscale nanofluidic memristors enable iontronic memory, sensing and programmable separation

Memristors—electronic and ionic elements with memory—have transformed neuromorphic computing, but their extension to practical fluidic systems has been limited to individual or few nanochannels. Here, we demonstrate a macroscale nanofluidic memristor, combining experiments and simulations in a scalable membrane platform. We reveal that asymmetric ion polarization dynamics in nanochannels under AC fields generate memristive response—a prediction confirmed experimentally in a scalable membrane architecture. The macroscopic design exhibits tunable ion transport with programmable on/off states, enabling spike-timing-dependent control of ion fluxes that mimics synaptic plasticity. Beyond neuromorphic behavior, we demonstrate ion-specific chemical sensing and programmable, on demand ion separation. This work bridges nanofluidic phenomena with macroscale functionality, opening avenues for large-area iontronic memory, adaptive filtration, bio-inspired computing, and selective ion detection. Memristors have transformed neuromorphic computing, yet fluidic implementations remain limited. Here, the authors demonstrate a scalable nanofluidic memristor for memory, neuromorphic ion transport, sensing, and programmable ion separation.