Monitoring soil health is very important in agriculture. Soil health can be monitored by the sensing the metabolomic activities of microbiota as increased concentrations in CO2, CH4, N2O, vapors in the soil. A vapor sensor based on microfabricated cantilever array is an ideal sensor platform for field deployable miniature sensors. Many such platforms, such as cantilever arrays, have been demonstrated as sensitive sensors for chemical and biological detection. In fact, the array-based sensor platform satisfies almost all of the requirements of an ideal high-performance sensor, such as its very high sensitivity, miniature size, array based multi-analyte detection, and low power consumption. However, poor molecular selectivity, especially for small molecules at trace level, as well as the lack of reproducibility, pose formidable challenges in translating this highly sensitive sensor platform into a practical reality. Traditionally, selectivity in chemical and biological sensing is achieved by using immobilized receptors or chemical interfaces on sensor surfaces. While poor selectivity in small molecule detection using reversible receptors is due to the lack of uniqueness of receptor-analyte interaction, nonuniformity in the immobilized receptor graft density is the reason for unacceptable rates of reproducibility in these sensors. We have used a multi-modal, multi-physics approach in overcoming these challenges. Multi-modal data when analyzed using deep learning techniques, show enhanced selectivity, sensitivity, and reliability. In order to operate these sensors in a field-scale study, a long range through-the-soil (TTS) wireless power transfer method is developed. Energy from an adjacent solar panel array is distributed wirelessly via conduction currents that propagate outward along and through the soil. Power transfer within a 0.25-ha (0.6acre) area is demonstrated using an average input power of 6W.This research is supported by NSF-SitS Award Nos. 2226614 and 2226612
Read full abstract