Experimental Characterization and Two-Phase Flow Modelling of Water Transport Phenomena in PEM Fuel Cells

Proton Exchange Membrane (PEM) fuel cells are a promising alternative to conventional power sources like combustion engines in automotive applications. A key property of these fuel cells is their operating temperature below 100°C, since, among other advantages, heat-up times are significantly lower when compared to high temperature fuel cells. Since the capabilities of one fuel cell are limited, many individual cells are combined to fuel cell stacks to reach the power output required to move a vehicle.

Maximizing the power density and longevity of fuel cell stacks while maintaining reasonable development and production costs requires a high level of understanding of all contributing physical phenomena as advancements in modelling can amplify the reliability of simulations in the development cycle.

On the fuel cell’s cathode side, water is produced in the fuel cell reaction. Due to the low operating temperature, this product water can condense into its liquid state, blocking up the reactive zone as well as the gas supply. Therefore, water removal and consequentially the design of the gas supply channels are a key optimization factor in fuel cell development.

Current multi-physical models of a complete fuel cell, modelling electro-chemistry, heat transfer, electric and ionic charge, gas species transport and fluid flow, are, as of now, not sufficiently representing the formation and transport of liquid water in the cell as well as the consequences of its presence for its performance.

The goal of this project is to advance the modelling of liquid water in PEM fuel cells by providing reference data via optical in-situ experiments in the first phase. In the second phase of the project, the reference data is used to develop and implement an improved liquid water model that can reliably predict flow regimes in the gas channels and the resulting influence on the fuel cell performance based on an OpenFOAM fuel cell solver.