Nanofluidic Volatile Threshold Switching Ionic Memristor: A Perspective
Miliang Zhang, Guoheng Xu, Hongjie Zhang, Kai Xiao
Abstract
The fast development of artificial intelligence and big data drives the exploration of low-power computing hardware. Neuromorphic devices represented by memristors may provide a possible computing paradigm beyond von Neumann's architecture because they enable the integration of processing and storage units by mimicking how the brain processes complex information in parallel. In the brain, information is processed via multilevel spiking coding and event-driven mechanisms, whose simplified neural circuit is represented by the leaky-integration-and-fire model combining volatile threshold switching memristors and capacitors. As a computing unit to emulate the working environment and explore the unique functions of ions and molecules of biological systems, nanofluidic volatile threshold switching ionic memristors become essential but are still missing. This Perspective will review the mechanism and role of threshold switching memristors as a building block for neuromorphic computing and list three possible routes for nanofluidic ones.