(492d) Predicting Fuel Cell Ink Aggregation and Drying

Polymer-electrolyte fuel cells (PEFCs) hold promise as a viable alternative to combustion engine for heavy duty vehicles and long-term energy storage. They have been sought-after in these fields for enabling minimal emissions technologies with high efficiencies and energy densities. Of the cost of the PEFC, the most expensive and complex part of the cell is the PEFC catalyst layer, which is key for enhancing the performance of the entire device. Numerous studies have been performed to investigate the correlation between the materials that comprise the catalyst layer and the performance of the cell, but the overarching relationship between the two remains undetermined. PEFC catalyst layers are typically constructed with a carbon supported catalyst particle (most commonly platinum) with an ionomer, often a perfluorosulfonic acid (PFSA) such as Nafion, which secures the carbon suspensions in the ink dispersions by preventing the particles from crashing out as well acting as a binder after the solvent has been evaporated off. To aid in the exploration and comparison of such complex systems, a few recent studies have looked at utilizing the role of the pH of the ink dispersion.¹

In this talk, we outline an approach to predicting the interactions between the particles through mathematical modeling using both macroscopic and microscopic ink properties. We utilize a kinetics-based procedure based on pairwise interaction spheres with partially uncovered side chain groups that simultaneously solves for the pH and sizes distributions of the particles under adjustable parameters, such as solvent and particle wt%. A novelty which has been included is the idea of backside repulsion of the particles, where the double layers on the opposite sides of interacting particles can further push them together, greatly impacting aggregation rates. Further, the outputs of this formulation will be used to ascertain fundamental properties of the catalyst layer such as the porosity using a space filling argument. To display the precision and robustness of this proposal, we juxtapose the predictions against experimental pH data for a range of concentrations and solvents. The intuition gained from this model will guide future design goals and process conditions.

Acknowledgements

This study was conducted under the Million Miles Fuel Cell Truck Consortium (M2FCT) funded by the Hydrogen and Fuel Cell Technologies Office in the Energy Efficiency and Renewable Energy Office of the U.S. Department of Energy under contract DE-AC02-05CH11231.

References

1. S. A. Berlinger, B. D. McCloskey, and A. Z. Weber, J. Phys. Chem. B, 122, 7790–7796 (2018).