Fuel-Cell Performance and Stability during Liquid-Water Removal Cycles

Water management is crucial for achieving high-performance proton-exchange-membrane fuel cells (PEMFCs), as it helps keep the electrolyte hydrated while avoiding performance degradation and cell shutdown due to liquid-water accumulation. According to ex-situ measurements [1-3], accumulation of liquid water in gas-diffusion layers (GDLs) of PEMFCs is a transient process that can be accompanied with oscillations in capillary pressure and saturation. To understand how liquid-water accumulation and drainage impact PEMFC performance hysteresis and stability, a transient cell-level model that accounts for electrode structure and composition and is computationally efficient is needed. A number of volume-averaged models that describe the electrode structure through pore-size distribution have been developed in the past [4-7], but they are steady-state and thus cannot predict dynamic PEMFC performance. Existing transient models have also not been used to analyze cyclic liquid-water accumulation [8-10]. In this work, a transient two-phase 2D PEMFC model is developed in the open-source fuel-cell modeling software OpenFCST [11] and applied to analyze PEMFC performance hysteresis and stability during liquid-water accumulation and drainage cycles. The model accounts for the electrode structure through a mixed-wettability pore-size-distribution framework and incorporates a novel dynamic boundary condition to describe the experimentally observed cyclic liquid-water accumulation. Results of the numerical simulations are compared to transient current-density and resistance data at two polarization scan rates and during voltage steps measured with an in-house single-channel cell at multiple operating conditions. This work demonstrates how high breakthrough pressure and rapid liquid-water removal from GDLs may cause highly unstable fuel-cell operation with strong hysteresis and oscillations in the polarization curve (such as those in the figure below) that closely resemble experimental reports [12]. Numerical simulations also reveal the existence of a scan rate that maximizes polarization hysteresis due to a match between the time scale of GDL flooding and of a complete voltage sweep. Fast-scan polarization sweeps are shown most suitable for detecting catalyst-layer flooding that depends on its wettability and occurs within single seconds in contrast to GDL flooding that takes hundreds of seconds. Overall, this work brings more attention to the transient analysis of fuel-cell performance under wet conditions. Figure: Polarization-curve oscillations caused by cyclic liquid-water removal from the cathode GDL. Similar fluctuations have been experimentally observed in [12]. Transient graphs show the dynamics of current density and cathode GDL saturation. Operating conditions are 60 °C, 90% RH, 1.5 atm; scan rate is 0.5 mV/s. References J. T. Gostick et al., J. Electrochem. Soc. 157.4 (2010) C. Quesnel et al., J. Phys. Chem. C 119.40 (2015) D. Ziegler, B.Sc. thesis, Hochschule Mannheim / University of Alberta (2020) A. Z. Weber et al., J. Electrochem. Soc. 151.10 (2004) M. Eikerling, J. Electrochem. Soc. 153.3 (2006) V. Mulone and K. Karan, Int. J. Hydrog. Energy 38.1 (2013) Zhou et al., J. Electrochem. Soc. 164.6 (2017) R. J. Balliet and J. Newman, J. Electrochem. Soc. 158.8 (2011) I. V. Zenyuk et al., J. Electrochem. Soc. 163.7 (2016) A. Goshtasbi et al., J. Electrochem. Soc. 166.7 (2019) M. Secanell et al. ECS Transactions 64.3 (2014) C. Ziegler and D. Gerteisen, J. Power Sources 188.1 (2009) Figure 1

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.