Hydrogel Covered Solid-State Nanopores as Iontronic Memristors for Neuromorphic and Logic Functions

The human brain exhibits extraordinary computational capabilities, enabled by its intricate network of neuronal interconnections and parallel signal transmission pathways. Nanofluidic memristors, which dynamically modulate ionic conductance through controlled ion transport, have shown promise in emulating synaptic functions. Here, we report an integratable nanofluidic memristor based on a hydrogel covered SiN_x nanopore, which modulates its ionic conductance via voltage induced ion concentration polarization. By tuning the diameter of SiN_x nanopore, the ionic device functionalities can switch from memristor to capacitor. Moreover, this nanofluidic memristor are capable of mimicking diverse synaptic plasticity behaviors, such as paired-pulse facilitation/depression and synaptic weight potentiation/depression. Notably, it can serve as a neuromorphic synaptic element for information processing, storage, and encoding, achieving an accuracy of 93.8% in the MNIST handwritten digit classification task. Finally, fluidic memristors are integrated to construct fluidic ionic circuits, which are applied for logic functions. This architecture based on hydrogel covered SiN_x nanopore devices offers an innovative strategy for the design of integrated iontronic circuits and provides an experimental foundation for neuromorphic computing in liquid environments.