Revisiting the Gas Diffusion Layer Water Inventory - from Micro to Macro Scale for Modern GDLs

Abstract

Listen

Translate article

Significant progress has been made over the last decade in improving Polymer Electrolyte Membrane fuel cells (PEMFC), resulting in usable current densities of up to 3.0 A/cm². Specifically for high current density values, it is important to maintain a well-balanced water inventory inside the fuel cell, particularly within the GDL, to prevent flooding of the porous material by excess product water and maintain high permeability for reactant gases. Twenty years ago, several authors conducted research [1,2,3,4] to enhance the modelling of permeability and wetting behavior of gas diffusion layers (GDLs) for PEMFC models. However, the standard Leverett J-function model [5], which was originally developed for porous rocks, remains the default description of saturation in many macro scale fuel cell models. Additionally, the latest GDL materials are much thinner and technologically advanced compared to that time. Thus, this work aims to model the water saturation and permeability of gases in modern GDLs using various simulation methods starting from the micro scale (GeoDict, Pore-Network-Modelling (PNM)). The micro-scale simulations are based on high-resolution X-ray scans of the GDL structure. These scans are analyzed and reconstructed with GeoDict to create digital twins that perform statistically identically and allow for virtual material modifications. Parametric studies for various local operating conditions are then performed using these twins, including e.g. different compression rates, temperatures, water production rates, and gas velocities. These simulations provide insight into the behavior of water saturation within the GDL structure and its influence on the diffusion of oxygen and hydrogen towards the catalyst layer. PNM modelling is employed to ensure the representativeness of the results regarding the macro-scale behavior of the porous material. Using the knowledge gained from the microstructure simulations, the simple Leverett J-function will be benchmarked. Depending on the results, ways to improve the J-function parameters or devising a new macro-scale functional description of the GDL water saturation and the resulting gas diffusion properties of the GDL will be explored.This work has received financial support by the German Federal Ministry for Digital and Transport within the GALLIA project (NIP II, 03B10114D).Citation:[1] Gostick, Jeffrey T., et al. "Capillary pressure and hydrophilic porosity in gas diffusion layers for polymer electrolyte fuel cells." Journal of power sources 156.2 (2006): 375-387.[2] Kumbur, E. C., K. V. Sharp, and M. M. Mench. "On the effectiveness of Leverett approach for describing the water transport in fuel cell diffusion media." Journal of Power Sources 168.2 (2007): 356-368.[3] Kumbur EC, Sharp KV, Mench MM. Validated Leverett approach for multiphase flow in PEFC diffusion media. J Electrochem Soc 2007;154. https://doi.org/10.1149/1.2784283.[4] Sarkezi-Selsky, Patrick, et al. "Lattice Boltzmann simulation of liquid water transport in gas diffusion layers of proton exchange membrane fuel cells: Parametric studies on capillary hysteresis." Journal of Power Sources 535 (2022): 231381. [5] Leverett, MoC. "Capillary behavior in porous solids." Transactions of the AIME 142.01 (1941): 152-169.