Autonomous electrochemical monitoring with low-power sensors and application-specific energy harvesting

Abstract

Continuous electrochemical monitoring can provide information that is missed by intermittent sampling, but its practical use is often limited by the energy required for sensing, processing, and wireless communication. Greater autonomy therefore requires the sensing method, measurement schedule, and energy source to be matched to the analyte and the deployment environment. This approach was first examined for on-body sweat monitoring. A shoe-insole triboelectric nanogenerator was developed to harvest energy from gait and was characterized during stepping, walking, and jogging. The harvester was combined with a cotton-yarn potentiometric Na⁺ sensor, a silicone sweat-sampling patch, and Bluetooth Low Energy transmission. Potentiometric readout reduced the electrical demand by avoiding an applied bias and current measurement. The system demonstrated gait-powered sensing and wireless transmission, although the voltage-supervision circuit still required an auxiliary supply. The sensing-side power demand was further reduced for plant H₂O₂ monitoring. Conventional H₂O₂ sensors are commonly amperometric and require an applied potential and current readout. Here, a mixed-potential sensor was developed using Pt/Nafion electrodes modified with carboxyl-functionalized multi-walled carbon nanotubes. The sensor operated at open circuit, allowing H₂O₂ to be measured as a change in electrode potential without external excitation. The sensing chemistry was transferred to a 3D-printed microneedle platform for direct in-planta measurement. The response depended strongly on pH, which limits quantitative interpretation without compensation. The complete co-design approach was then implemented for plant NO₃⁻ monitoring. A potentiometric microneedle sensor was combined with a thermoelectric generator driven by the soil–air temperature gradient, a storage capacitor, a voltage-monitor unit, and intermittent Bluetooth transmission. The low-power readout and duty-cycled operation were matched to the small and time-varying harvested energy. Under reproduced soil–air thermal profiles, the system performed batteryless in-planta sensing and wireless communication. These results show that autonomy does not depend on the harvester or sensor alone. It depends on matching the analyte timescale, electrochemical transduction method, biological interface, communication schedule, and available ambient energy. The developed systems demonstrate how this co-design approach can reduce sensing power and enable self-powered or batteryless electrochemical monitoring across wearable and plant applications. Extending this approach will require fully integrated devices, long-term deployment, and adaptive multiparameter sensing that can compensate for changes in the local chemical environment.