Hydrogel materials have biocompatibility, flexibility, transparency, self-healing ability, adhesion with various substrates, anti-freeze ability, and high-temperature resistance. However, the existing hydrogel devices cannot continue to operate in the case of damage, and they cannot work during the repair period, which brings great challenges and threats to life safety. Herein, we have designed a bio-inspired combinable low-power device by imitating the generation of nerve signals whose components can be disassembled and can continue to operate during the period of reconstruction. And the mechanism and determinants of the above phenomena are revealed. The results indicate that this device can establish some information interaction relationships with the body or its surroundings to reflect and identify certain changes, implying that it will possess promising potential in feedback systems, power transformers, intelligence systems, soft robotics, wearable devices, implanted electronics with flexible characteristics matching biological tissues, etc.

Human skin, with beneficial elasticity and reparability, could sense the changes of the external environment through different receptors. Imitating these mechanical properties and perception of bionic skin with power supply function is an enormous challenge in the field of Internet of things and artificial intelligence. Herein, the neural signal transmission of human skin is imitated to create a smart self-powered bionic skin fragment integrating skin and power supply functions. Unlike the traditional bionic skin in essence, it can intelligently perceive the outside world by using anion-selective and cation-selective gels to control exchangeable anions and cations to realize the change of resting current and action current, and it can maintain the relatively stable self-powered current of 0.5 µA for nearly 2.2 h. Moreover, its mechanisms of current and voltage changes are systematically investigated. These results reveal that it can be applied to the synchronous transmission of signals for the next-generation neurologically integrated soft engineering systems such as bionic sensors, or prosthetic devices in hybrids of living and nonliving systems.