Modelling of PEM Fuel Cell Degradation – From Accelerated Stress Tests to Real-Life Stress Conditions

  • Keržič, Rok (AVL-AST d.o.o.)
  • Eisenberger, Lukas (AVL Deutschland GmbH)
  • Prah, Ivo (AVL-AST d.o.o.)

Please login to view abstract download link

The durability of proton exchange membrane (PEM) fuel cells is of major importance for their successful implementation in real-world applications. While accelerated stress tests (ASTs) are widely used to characterize and quantify material degradation under controlled laboratory conditions, translating these results to actual fuel cell stack operation remains challenging. This paper presents a comprehensive methodology for modelling PEM fuel cell degradation by linking mechanistic models, calibrated using AST protocols, with load profiles and stressors representative of real-life operating conditions, such as standardized drive cycles (e.g. WLTC). The modelling framework follows a mechanistic, physics‑based approach that enables the isolation and quantification of individual degradation pathways. These include catalyst degradation, covering carbon support oxidation, platinum support oxidation and dissolution, catalyst redistribution, and particle growth leading to electrochemical surface area (ECSA) loss, as well as membrane degradation, which accounts for ionomer chemical attack through peroxide and radical formation, and the development of platinum bands. The mechanistic degradation models are implemented within a multi‑physics system simulation platform, allowing seamless transfer across different scales, from material development to full system integration. A degradation model calibrated using single‑cell ASTs targeting specific mechanisms can be upscaled to a discretized stack model and subsequently embedded in a multi-domain digital twin. This integrated system can couple the PEM fuel cell stack with the balance‑of‑plant (BoP), thermal management, or driveline subsystems, enabling the simulation of real-life operational stress conditions in a versatile, fully modular approach. Overall, the presented framework provides a robust connection between rapid experimental AST methods and the performance and durability targets of the vehicle industry, thereby supporting the development of reliable and long‑lasting fuel cell powertrains through highly accelerated degradation simulation.