Multiphysics Modeling of Proton Exchange Membrane Fuel Cells: A Systematic Parametric Assessment of Operating and Electrochemical Parameters

Proton exchange membrane fuel cells (PEMFCs) are promising electrochemical energy conversion technologies owing to their high efficiency, rapid dynamic response, and low-emission operation. Their performance is governed by strongly coupled electrochemical, protonic, mass-transport, and species-transport phenomena. In this study, a two-dimensional five-layer membrane electrode assembly (MEA) model was developed using the Hydrogen Fuel Cell interface in COMSOL Multiphysics. The model incorporates electronic and ionic charge transport, multicomponent gas diffusion, Darcy flow, Butler–Volmer kinetics, oxygen transport limitations, and hydrogen crossover. The model was calibrated against experimental polarization data obtained under humidified air and oxygen conditions at 100 °C and subsequently validated against experimental data at 100 °C and at 80 °C and 70% relative humidity. The calibrated parameter values obtained at 100 °C were directly applied to the 80 °C condition without further parameter fitting. Four electrochemical parameters—membrane electrolyte conductivity, ORR reference exchange current density, cathodic charge transfer coefficient, and limiting current density were calibrated to improve agreement with the experimental data. Subsequently, a one-factor-at-a-time (OFAT) analysis was performed by varying temperature (60–100 °C), relative humidity (40–70%), membrane thickness (5–15 µm), electrolyte conductivity (5–15 S m−1), ORR exchange current density (10−4–10−2 A m−2), and cell voltage (0.40–0.80 V). Electrode potential, electrolyte potential, pressure drop, and local O2, H2O, and N2 mole fractions were evaluated. ORR kinetics and cell voltage exhibited the strongest effects on the electrochemical responses, whereas temperature and relative humidity primarily influenced protonic, pressure, and species-transport behavior. Membrane thickness and electrolyte conductivity had comparatively limited effects within the investigated ranges. The agreement between the model predictions and experimental data at 80 °C and 70% relative humidity further demonstrates the model’s predictive capability across the investigated operating conditions. The results provide a physically interpretable framework for PEMFC model calibration, parametric assessment, and subsequent optimization studies.