A Model-Based Approach to Improved Understanding and Mitigation of Transport Losses in PEM Fuel Cells

Abstract

Platinum-based catalysts are used to activate hydrogen oxidation and oxygen reduction in conventional polymer-electrolyte membrane fuel cells (PEMFCs). Reducing the amount of Pt is an important step towards lowering cost [1]. Pt alloys and core-shell structures that are more active for oxygen reduction than pure Pt have been demonstrated and these materials can enable low Pt loadings [2]. In fact, membrane-electrode assemblies (MEAs) with low loadings of highly active Pt-group metal (PGM) catalysts have already been demonstrated to exceed the U.S. DOE’s high-efficiency target of > 0.3 A/cm2 at 0.8 V [3]. However, DOE’s rated-power target of 1 W/cm2 cannot be met with these MEAs due to voltage losses at higher current densities that increase as the amount of catalyst is reduced. This behavior needs to be understood and mitigated in order to maintain present levels of performance while achieving meaningful cost savings. United Technologies Research Center (UTRC) is leading a DOE-supported research project that is focused on understanding and mitigating the transport losses in MEAs with ultra-low loading of PGM catalysts (i.e., ≤ 0.125 Pt/cm2 on both anode and cathode). This project is part of DOE’s Fuel Cell Performance and Durability (FC PAD) Consortia. UTRC is utilizing an iterative approach to address this challenging problem, which is illustrated in the figure. UTRC has developed and validated a new cathode-catalyst layer (CCL) model that builds upon conventional agglomerate models by including localized spherical diffusion and slow adsorption at the scale of the catalyst particles in addition to the processes at the scale of agglomerates that are normally considered [4]. Including transport losses associated with each platinum particle leads to limiting currents that are proportional to the amount of platinum, which is not a characteristic of traditional agglomerate models. Predictions using morphological parameters and physical properties from the literature are commensurate with published investigations into the influence of platinum loading on oxygen transport. As depicted in figure, the team is using the insights into the key CCL transport phenomena provided by this new hierarchal model to predict how these losses may potentially be mitigated. After providing a brief summary of UTRC’s model, the major focus of this talk will on the team’s efforts to improve the performance of MEAs with ultra-low PGM-catalyst loadings. Acknowledgements This work is supported by the U. S. Department of Energy (DOE) under contract number DE-EE0007652. The authors would like to thank their FC-PAD Consortia colleagues, especially those at LANL and ORNL, who have helped the team characterize electrodes and MEAs, as well as our FC-PAD project partners, Ion Power and the University of Arkansas at Little Rock (UALR) for providing advanced materials. References H. Gasteiger, et.al., Appl. Cat. B: Environ., 56, 9 (2005).M. Shao, et.al., Chem. Rev., 116, 3594 (2016).D. Myers, et. al., DOE AMR, ID# FC106 (2015).A.Z. Weber, et. al., J. Electrochem. Soc., 161, F1254 (2014). Figure 1