Phase transitions perception in nonreciprocal mechanical metamaterials through electromagnetic resonance.

Phase transitions of metamaterials are critical in advancing energy conversion efficiency and controlling mechanical performance. However, the design method and localized perception of phase transitions remain challenging. Inspired by the passive coupling mechanisms in ostrich locomotion, this work proposes nonreciprocal metamaterials that can perceive real-time phase transition. These architectures enable the topological solitons to propagate unidirectionally and overcome dispersive and dissipative effects through bistable-to-monostable state switching between adjacent units. The integration of electromagnetic resonators within the metamaterial units enables real-time detection of dynamic phase transitions, as soliton propagation or external loads induce resonance frequency shifts between distinct stable states. By arraying these mechanoreceptive units and the combination of the machine learning, it can encode information and compute programmatically. Furthermore, the mechanoreceptors hold promising applications in robotics. This work provides an approach for integrating phase transition perception and nonlinear wave manipulation and offers insights into dynamic material intelligence and energy management systems.