Energy-adaptive resistive switching with controllable thresholds in insulator-metal transition.

Resistive switching has provided a significant avenue for electronic neural networks and neuromorphic systems. Inspired by the active regulation of neurotransmitter secretion, realizing electronic elements with self-adaptive characteristics is vital for matching Joule heating or sophisticated thermal environments in energy-efficient integrated circuits. Here we present energy-adaptive resistive switching via a controllable insulator-metal transition. Memory-related switching is designed and implemented by manipulating conductance transitions in vanadium dioxide. The switching power decreases dynamically by about 58% during the heating process. Furthermore, the thresholds can be controlled by adjusting the insulator-metal transition processes in such nanowire-based resistive switching, and then preformed in a wide range of operating temperatures. We believe that such power-adaptive switching is of benefit for intelligent memory devices and neuromorphic electronics with low energy consumption.