Simulation of Proton Exchange Membrane Durability Under Fuel Cell Vehicle Operation – a Fundamental Study

Proton exchange membrane fuel cells (PEMFCs) have been proven to be a promising candidate to replace combustion engines due to their zero-carbon emission and high power densities. Despite the recent success in PEMFC commercialization, a number of challenges such as high cost and difficulty in lifetime estimation still hinder their further development. PEMFC durability tests require a long time to complete; therefore, durability predicting models are increasingly important as a supporting tool for further development and implementation. Fuel cell membranes undergo a variety of dynamic conditions during regular operation such as varying temperature, humidity, current density, and cell potential. The cyclic variations of humidity and temperature (hygrothermal variations) during dynamic operation lead to swelling and contraction of the membrane. The fluctuating stress caused by the continuous expansion and contraction of the membrane when confined within the cell leads to mechanical membrane degradation. The recurring swelling and contraction of the membrane which stem from water content changes in the membrane cause fatigue and creep and ultimately the formation of pinholes, cracks, and tears [1-2]. Chemical membrane degradation occurs when radicals such as hydroxyl and hydroperoxyl are formed and attack the membrane and is escalated by high operating cell potentials. Chemical degradation results in decay in the ionomer chemical structure, membrane thinning, increased gas crossover, and potential electrode shorting [3]. The outcomes of such degradation are aggravated in the presence of mechanical degradation [4]. Hence, the study of combined chemo-mechanical membrane degradation is crucial for overall fuel cell membrane durability [5]. The objective of the present work is to establish a predictive membrane lifetime simulation tool designed to represent chemo-mechanical membrane degradation under various operating conditions for automotive fuel cell applications. To this end, a statistical model is developed based on the membrane fibrillar morphology presented by El Hannach et al. [6-7]. A network of fibre bundles is generated to represent the membrane structure where any bundle is assumed as the primary building block of the membrane [8-10]. Each bundle contains an aggregate of backbone chains with typical mechanical and chemical properties. Mechanical and chemical degradation rates are separately studied and coupled in the model. The thermally activated breaking rate of each fibre is calculated to evaluate the mechanical degradation rate at any given time. In order to calibrate the mechanical degradation rate in the model, the mechanical degradation of a reinforced membrane under cross-pressure accelerated mechanical stress tests ( p-AMST) [11] is investigated to obtain the membrane fatigue lifetime for a variety of cross pressures and temperatures. Then, the genetic algorithm is successfully applied to optimize the experimental parameters considering the least squares method. Furthermore, accelerated membrane durability tests (AMDT) carried out by Macauley et al. [12] are considered to calibrate the chemical degradation rate parameter in the model. Finally, the fully calibrated model is used to estimate the membrane lifetime under realistic fuel cell vehicle operating conditions (e.g. drive cycles). Fuel cell components experience different degradation mechanisms. In this regard, this modelling framework on the membrane lifetime can be used in conjunction with other methodologies on the cathode catalyst lifetime such as [13-14] to scrutinize the key factors and their impacts on fuel cell durability under similar operating conditions. Acknowledgements This research was supported by the Natural Sciences and Engineering Research Council of Canada, Canada Research Chairs, and Simon Fraser University Community Trust Endowment Fund. References [1] A. Kusoglu, et al. Journal of power sources. 161 (2006) 987-996. [2] R. M. Khorasany, et al. Journal of Power Sources.252 (2014) 176-188. [3] V. A. Sethuraman, et al. Journal of The Electrochemical Society. 155 (2007) B50. [4] S. V. Venkatesan, et al. in Electrochemical Society Meeting Abstracts 226. (2014). 1298 [5] Y. Singh, et al. Journal of Power Sources. 412 (2019) 224-237. [6] M. El Hannach, et al. in Electrochemical Society Meeting Abstracts 230. (2016) 2836. [7] M. El Hannach, et al. in Electrochemical Society Meeting Abstracts 233. (2018) 1992. [8] L. Rubatat, et al. macromolecules. (2004) 7772-7783. [9] P. E. A. Melchy and M. Eikerling, Journal of Physics: Condensed Matter. 27 (2015) 325103 [10] N. S. Khattra, et al. Journal of The Electrochemical Society, 167 (2019) 013528. [11] A. S. Alavijeh, et al. Journal of Power Sources Advances. 2 (2020) 100010. [12] N. Macauley, et al. Journal of Power Sources. 336 (2016) 240-250. [13] M. Shojayian and E. Kjeang, in Electrochemical Society Meeting Abstracts 241. (2022) 2456. [14] M. Shojayian and E. Kjeang, in Electrochemical Society Meeting Abstracts 242. (2022) 1569.

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.