Fast predictions of ice melting during cold start of PEMFCs: Insights from pore-scale simulations

The existence of ice in the gas diffusion layer (GDL) hinders the transport of reactant gas. Accurate prediction of the ice-melting process in the is crucial for optimizing the cold start performance of proton exchange membrane fuel cells (PEMFCs). In this work, a pore scale model is developed to predict the ice melting process during the cold start of PEMFCs. A general conjugate boundary treatment is proposed for explicitly addressing fiber-level thermal contact resistance, which is validated by the experiments and benchmarks, quantitatively resolving the effects of microstructure and showing great advantages for highly heterogenous ice distribution situations. Additionally, a mean field model with microstructure-informed effective thermal conductivity is developed to rapid predict the ice melting in GDLs. Leveraging pore-scale insights, the mean field model achieves high prediction accuracy for homogeneous, hierarchical ice distributions and for scenarios considering fiber-level thermal contact resistances. The proposed mean field model can be used for the fast predictions of ice melting in GDLs, and thus enables the modeling and optimization of cold start performances of PEMFCs.