Distributed Bimodular Sensors for Onboard Diagnostics of Automotive Aftertreatment Devices

The monitoring of gaseous exhaust (e,g., NOx, NH3) and particulate matters (PMs) has been critical in the automotive diesel engine combustion emission control, for which various catalytic converters need be operated efficiently and environment-friendly, including diesel oxidation catalysts (DOC), diesel particulate filters (DPF), and selective catalytic reduction (SCR). For instance, in the DPF, the device regeneration usually requires the monitoring and control of NOx to help reduce theregeneration cycle number and fuel consumption for a robust operation of the particulate filtration. While the SCR, as the most reliable and successful approach to remove NOx from diesel engineexhaust, requires an injection of NH3 in the form of urea to catalytically convert NOx into nontoxic N2 and water. However, the amount of injected NH3 needs be monitored on the basis of NOx composition (NO/NO2 ratio) presented in exhaust to prevent over injection and secondary pollution. Therefore, the on-board analysis of NOx has been an urgent issue needed to solve for emission control. At the University of Connecticut (UConn), nanostructure array (nanoarray) based bimodular sensors have been demonstrated to accurately differentiate NOx, SO2, and NH3 in ppm level with excellent sensitivity and selectivity. In this STTR phase I project, 3D Array Technology and UConn propose to develop an innovative bimodular sensing strategy to additively deploy and distribute locally on the commercial monolithic catalysts. By combining the resistor mode and impedance mode, the single sensor device can in-situ and real-time measure the gas composition in NOx mixture, as well as monitoring the PM and catalyst surface composition locally. Meanwhile, the simplicity and ease of fabrication/deployment of nanoarray-based sensors brought by the bimodular sensing strategy and in-situ growth on honeycomb monoliths could further optimize the cost and reliability. Specifically in this Phase I project, metal oxide nanoarray based bimodular sensors will be in-situ grown on commercial monolith substrates. The chemical and physical structure and properties of as– prepared sensors will be well characterized and evaluated and their initial sensing performance towards mixture analysis will be evaluated and calibrated. In addition to the prototype bimodular NOx sensor tested in simulated gas atmosphere and low concentration detection window (<10 ppm), the reliability test will also be conducted in simulated exhaust atmosphere. The corresponding data processing on the basis of data-driven machine learning will be developed to analyze the gas mixture with heavy NOx loading and severe competitive adsorption to accommodate an enlarged detection window. A rich set of relevant parameters that can be interrogated in-situ and real-time include NOx composition, solid particulate size, mass and flow rate, substrate/catalyst surface composition, etc.. The distributed nanoarray architecture sensing technology proposed here is expected to enablea truly real-time and in situ on-board NOx/PM/catalyst diagnostics with good location precision, therefore help interpret and understand the dynamic aftertreatment device operation process. Further, remarkable spatial and temporal resolutions may be achieved as the built-in nanoarray sensors also directly contribute to the catalytic functions locally via real-time responses to the gaseous and solid species involved. Upon implementation, the proposed research will provide a unique in-situ and real-time interrogation approach for quantifying and understanding the commercial catalytic converters in operation. Further, the distributed monolithic sensor architectures will be the first of its kind as efficient sensing and control solutions with potentially remarkable spatial and temporal resolutions.