D. Dong
Rowan University,
United States
Keywords: membrane degradation, fluoride emission rate, diagnostics, sensors
Summary:
Non-Proprietary Summary: Initially funded as a seed discretionary project through Million Mile Fuel Cell Truck (M2FCT) consortium, the project provides a complementary diagnostics tool for providing key insights for mitigating proton exchange membrane fuel cell (PEMFC) efficiency and durability concerns. On the topic, we have an issued patent of systems and methods for monitoring fuel cell membrane degradation (US Patent 11,811,116, Nov. 2023), and several publications (Dong et al., 2022 ECS Sens. Plus 1 (3), 035601; Dong et al., 2022 ECS Meet. Abstr. 242, 2256). It is highly demanding to date is to develop innovative material or integration as the basis for PEMFC towards commercial EV applications where the efficiency and durability become more critical. The team of the sensors and systems in Dong's laboratory at Rowan University has been working on sensor metric-based degradation prediction and predictive maintenance (PM) using deep learning algorithms. The team employs ion-sensitive-field-effect-transistor (ISFET) microsensor to real-time continuously monitor ionomer degradation, and to use deep learning for sensor-guided PM of fuel cells. The intellectual merits are: (i) complementary diagnostics by combining the ISFET sensor technology and advanced data-driven analysis for PM, (ii) the bridging of the research expertise of the interdisciplinary including PEMFC membrane, integrated micro-electronics and computing analysis, (iii) the consideration of the fuel cell EV dynamic operating space in real driving. Vision and Goals: In Nafion®-based PEMFCs, radical attack causes polymer chain scission and irreversible reaction, and results in the global and local thinning of the ionomer, followed by producing fluorinated and sulfated degradation materials into reactant outlet streams. Synchronous fluorinated and sulfated byproducts will be accumulated into reactant outlet streams. The concentrations are enhanced at elevated temperatures and lower humidity conditions. The fluoride and sulfate anions emission rates can be drawn as the signature of the PEMFC degradations. ISFETs have been promising microsensors due to low cost, small size, robustness, and their applications for real-time continuous monitoring. Leveraging on the evolution of ISFET technology, we dedicate efforts to the development of inline sensors integrated at the cathode and anode exhaust for monitoring the PEMFC degrading status. We use fluoride emission as a sensing model in practice. Highly fluoride sensitive membranes (LaF3/CaF2) for microsensors will be introduced into a thin layer of insulator in the ISFETs. The functionalization of the insulator layer varies the selectivity/sensitivity of ISFET. Moreover, deep learning (DL) methods can be employed for ISFET sensor based predictive maintenance (PM) of PEMFC. PM is one of the most important diagnostics tools in smart manufacturing and Industry 4.0. Accurate prediction of part degradation of the cell with the employment of Internet of Thing (IoT) could be achieved. It gives a complementary approach to existing PEMFC characterization and diagnostics techniques. The focus is to develop inline sensors for real-time continuous monitoring of the byproducts and thus membrane loss, and to realize the PM of PEMFC based on DL, towards commercial fuel cell EV applications.