The cost and durability of polymer electrolyte fuel cells (PEFCs) are critical to their massive commercialization. To further reduce the cost of PEFCs, optimized catalyst layers (CLs) with low Pt loading and high performance are necessary. For the CL with low Pt loading, mass transport loss is a critical factor and limits the performance at high current density conditions. The micro-structure of a CL needs to be optimized to reduce the transport loss through heterogeneous porous structures and make reactants more accessible to the Pt sites. Pore-scale simulation directly computes detailed multiphysics phenomena occurring inside the heterogeneous nanoscale structure of the CL [1], thereby improving our understanding of micro-structural effects and guiding the design of an optimized CL. However, for the wide-spread use of pore-scale simulation in the design process, the computational cost needs to be reduced.

In this work, the multiscale method developed in Zheng and Kim [2] is used to accelerate multiphase pore-scale simulation, where water generation and liquid water transport processes in the CL micro-structure are explicitly computed. In the multiscale method, a physical quantity is decomposed into macroscopic and local variations. The macroscopic variations occur at the scale of the CL thickness, and are described by the volume-averaged equation as in the macroscopic method [3]. The local variations are due to the complex micro-structure of CL, and are captured by pore-scale simulation. The multiscale method is applied to the governing equations of electrolyte phase potential and gas phase oxygen concentration in the multiphase pore-scale simulation. The calculation of the electrolyte phase potential shows that the multiscale method substantially increases the convergence rate without sacrificing the accuracy. The application of the multiscale method to the oxygen concentration will be presented. The performance and computational cost of the multiscale method in simulating multiphase transport-electrochemistry coupling in the cathode CLs will be discussed.

Reference:

[1] J. Kang, K. Moriyama, S. H. Kim, An extended stochastic reconstruction method for catalyst layers in proton exchange membrane fuel cells, Journal of Power Sources 325 (2016) 752-761.

[2] W. Zheng, S. H. Kim, A multiscale approach to accelerate pore-scale simulation of porous electrodes, Journal of Power Sources 348 (2017) 21 - 29.

[3] D. M. Bernardi, M. W. Verbrugge, A mathematical model of the solid-polymer-electrolyte fuel cell, Journal of the Electrochemical Society 139 (9) (1992) 2477-2491.