Facile construction of electrochemical and self-powered wearable pressure sensors based on metallic corrosion effects

Flexible mechanical sensors are essential components for smart wearables. Conventional resistive, capacitive, or transistor-based mechanical sensors consume energy continuously, while piezoelectric and triboelectric sensors respond selectively to dynamic or transient mechanical stimulations. Developing mechanical sensors that do not necessitate external power supply but are able to monitor static mechanical stimuli can compensate for the deficiencies of existing sensing devices. Here, we present the facile construction of a new paradigm of electrochemical mechanical sensors based on ubiquitous metallic corrosion effects. The intrinsic differences in corrosion activities of diverse metals (e.g., zinc, aluminum, copper, etc.) are utilized to create potential differences between two electrodes, followed by encoding external mechanical stimulations into the potential difference variations via carefully selected solid electrolytes. The developed electrochemical mechanical sensors exhibit comparable performance (e.g., sensitivity, recovery/response speed, reproducibility, etc.) with that of conventional sensors, but possess significantly superior simplicity, cost-efficiency, and desirable capability in resolving static or slowly-varying mechanical stimulations in a self-powered manner. As proof-of-concept demonstrations, machine learning enabled speech recognition with high accuracy of 99.07% and monitoring of diverse human physiological activities are successfully demonstrated. These proposed unique electrochemical mechanical sensors based on the ubiquitous metallic corrosion phenomena provide a simple but effective approach for the burgeoning human-machine interfacing requirements with great benefit to the resource efficiency and sustainability of our society. • A new paradigm of electrochemical mechanical sensors based on ubiquitous metallic corrosion effects is presented. • The electrochemical sensors do not necessitate external power supply but are able to monitor static mechanical stimuli. • The sensors exhibit desirable sensing performance and possess superior simplicity and cost-efficiency. • Machine learning enabled speech recognition with high accuracy of 99.07% is demonstrated with such sensors. • Diverse human physiological activities can be monitored with the electrochemical sensors.