When a proton exchange membrane fuel cell (PEMFC) is operated under sub-cooled conditions (stack/ambient temperature below 0°C, often with warmer reactant inlets), product water can persist as supercooled liquid and accumulate in the porous catalyst layer (CL) and gas diffusion layer (GDL) before freezing. The source study of sub-cooled PEMFC performance (Syed, Ahmed and Sung, 2011) predicted substantial liquid-water saturation in this regime with a steady-state two-phase model that contains no ice phase, and identified the formation of ice from this accumulated water as an unresolved risk to cell performance. Later literature resolved neighboring pieces — transient startup models with ice formation, ex situ ice-crystallization kinetics in GDL and CL, and neutron-radiography observation of where ice accumulates during subfreezing operation — but a coupled, experimentally validated prediction of ice nucleation, growth, spatial distribution, and the resulting performance loss under sustained sub-cooled operation has not been established. The problem asks for exactly such a coupled multiphysics model and its validation against experimental measurements under sustained sub-cooled operation.

Background

A proton exchange membrane fuel cell (PEMFC) generates electricity from hydrogen and air; water is produced at the cathode. The cell is a layered porous structure: a catalyst layer (CL, a thin porous film of Pt catalyst on carbon mixed with ionomer, a proton-conducting polymer phase, where the reactions occur) and a gas diffusion layer (GDL, a porous carbon cloth or paper) on each side of a proton-conducting polymer membrane; together these form the membrane-electrode assembly (MEA). 'Cold start' denotes startup from a sub-freezing environment, a requirement for automotive deployment (United States Department of Energy (DOE) targets have included unassisted startup from −30°C or below). 'Sub-cooled operation' in the source paper denotes operation with the ambient/stack temperature held below the freezing point of water, typically with reactant inlet streams warmer than the surroundings; in this regime product water can remain as supercooled liquid (liquid water below 0°C, observed in operating cells by Ishikawa et al.) and accumulate in CL and GDL pores before nucleating ice. 'Liquid saturation' is the fraction of pore volume filled with liquid water, predicted by two-phase porous-media models. The source study (Syed, Ahmed and Sung, 2011, Int. J. Energy Res. 35:365) built a three-dimensional steady-state, two-phase model of a cell with a serpentine flow field (a single snaking gas channel machined in the flow plate) and found that at sub-cooled ambient temperature with warm inlets, liquid saturation is high in the 'shoulder' region under the land (the solid rib of the flow plate between channel segments) near the inlet; their abstract states there is a high probability of ice formation throughout the system, which may further degrade performance, yet the model includes no ice phase. Prior and parallel literature covers parts of the physics: transient cold-start models coupling heat, water transport, and ice in the CL (Mao and Wang 2007; Meng 2008; Balliet and Newman 2011) address the startup transient rather than sustained sub-cooled operation; isothermal ice-crystallization kinetics in GDL and CL materials were measured ex situ by Dursch et al. (2012, 2014); in-situ imaging (Ge and Wang 2006) and high-resolution neutron radiography (imaging with a neutron beam, which is strongly attenuated by hydrogen-rich water and ice but passes through the metal hardware, so water and ice inside an operating cell can be mapped; Mishler et al. 2012) located ice in the cathode CL region during subfreezing operation and quantified electrochemically active surface area (ECSA, the catalyst area available for reaction) loss over cold-start/freeze-thaw cycles; and freeze/thaw damage to MEAs has been characterized since Cho et al. (2003) and in later durability work (Srouji and Mench 2012; Zhang et al. 2024). What has not been achieved is a single validated framework connecting, for sustained sub-freezing operation, supercooling-dependent ice nucleation and growth inside the porous layers, the spatial distribution of ice, and the resulting loss of performance through blocked reactant transport and covered active area; the further mechanical consequences of pore ice (stress, cracking, delamination of CL and membrane) form a related durability thread that motivates the problem but lies outside its completion criterion. Cold start is reviewed as an ongoing challenge in recent literature (Tajiri and Wang 2009; Yu et al. 2025).

Problem Statement

Develop and experimentally validate a coupled multiphysics description of sustained sub-cooled PEMFC operation (stack/ambient temperature below 0°C, with the inlet-temperature and saturation conditions of the source study as the reference regime) in which liquid water that accumulates in the porous layers undergoes the liquid-to-ice phase transition. The framework must (i) couple two-phase water transport, heat generation and transport, electrochemistry, and ice nucleation and growth in the CL and GDL, using supercooling-dependent kinetics consistent with measured crystallization behavior rather than assuming immediate freezing at 0°C; (ii) predict where ice forms first and how the ice distribution evolves spatially within the cell (heterogeneous and anisotropic porous-layer properties may be included for added realism but are not a completion condition); (iii) quantify the resulting performance degradation caused by ice blocking reactant transport and covering electrochemically active surface area; and (iv) demonstrate quantitative validation of the predicted ice formation and the associated performance loss against experimental measurements under sustained sub-freezing operation (for example, neutron-radiography or cryo-electron-microscopy (electron microscopy of frozen samples) ice distributions and transient voltage evolution). The single scientific objective is a validated predictive link from operating conditions to ice formation and the resulting performance loss; the components above are coupled links in that chain, not independent deliverables.

References

  1. Syed, Anwar ul Hasson; Ahmed, Dewan Hasan; Sung, Hyung Jin (2011). Performance of sub-cooled PEMFCs. International Journal of Energy Research 35(5), 365–375. DOI: 10.1002/er.1691. https://doi.org/10.1002/er.1691
  2. Mao, Leng; Wang, Chao-Yang (2007). Analysis of Cold Start in Polymer Electrolyte Fuel Cells. Journal of The Electrochemical Society 154(2), B139–B146. DOI: 10.1149/1.2402123. https://doi.org/10.1149/1.2402123
  3. Meng, Hua (2008). Numerical studies of cold-start phenomenon in PEM fuel cells. Electrochimica Acta 53(22), 6521–6529. DOI: 10.1016/j.electacta.2008.04.044. https://doi.org/10.1016/j.electacta.2008.04.044
  4. Tajiri, Kazuya; Wang, Chao-Yang (2009). Cold Start of Polymer Electrolyte Fuel Cells. Modern Aspects of Electrochemistry 43, 89–128. DOI: 10.1007/978-0-387-98068-3_3. https://doi.org/10.1007/978-0-387-98068-3_3
  5. Sundaresan, M.; Moore, R. M. (2005). Polymer electrolyte fuel cell stack thermal model to evaluate sub-freezing startup. Journal of Power Sources 145(2), 534–545. DOI: 10.1016/j.jpowsour.2004.12.070. https://doi.org/10.1016/j.jpowsour.2004.12.070
  6. Balliet, Ryan J.; Newman, John (2011). Cold-Start Modeling of a Polymer-Electrolyte Fuel Cell Containing an Ultrathin Cathode. Journal of The Electrochemical Society 158(9), B1142–B1149. DOI: 10.1149/1.3607968. https://doi.org/10.1149/1.3607968
  7. Dursch, T. J.; Ciontea, M. Arianit; Radke, C. J.; Weber, A. Z. (2012). Isothermal Ice Crystallization Kinetics in the Gas-Diffusion Layer of a Proton-Exchange-Membrane Fuel Cell. Langmuir 28(2), 1222–1234. DOI: 10.1021/la2033737. https://doi.org/10.1021/la2033737
  8. Dursch, T. J.; Trigub, G. J.; Lujan, R.; Liu, J. F.; Mukundan, R.; Radke, C. J.; Weber, A. Z. (2014). Ice-Crystallization Kinetics in the Catalyst Layer of a Proton-Exchange-Membrane Fuel Cell. Journal of The Electrochemical Society 161(3), F199–F207. DOI: 10.1149/2.004403jes. https://doi.org/10.1149/2.004403jes
  9. Ishikawa, Y.; Hamada, H.; Uehara, M.; Shiozawa, M. (2008). Super-cooled water behavior inside polymer electrolyte fuel cell cross-section below freezing temperature. Journal of Power Sources 179(2), 547–552. DOI: 10.1016/j.jpowsour.2008.01.031. https://doi.org/10.1016/j.jpowsour.2008.01.031
  10. Ge, Shanhai; Wang, Chao-Yang (2006). In Situ Imaging of Liquid Water and Ice Formation in an Operating PEFC during Cold Start. Electrochemical and Solid-State Letters 9(11), A499–A503. DOI: 10.1149/1.2337860. https://doi.org/10.1149/1.2337860
  11. Mishler, Jeffrey; Wang, Yun; Mukherjee, Partha P.; Mukundan, Rangachary; Borup, Rodney L. (2012). Subfreezing operation of polymer electrolyte fuel cells: Ice formation and cell performance loss. Electrochimica Acta 65, 127–133. DOI: 10.1016/j.electacta.2012.01.020. https://doi.org/10.1016/j.electacta.2012.01.020
  12. Mishler, Jeffrey; Wang, Yun; Lujan, Roger; Mukundan, Rangachary; Borup, Rodney L. (2013). An Experimental Study of Polymer Electrolyte Fuel Cell Operation at Sub-Freezing Temperatures. Journal of The Electrochemical Society 160(6), F514–F521. DOI: 10.1149/2.051306jes. https://doi.org/10.1149/2.051306jes
  13. Cho, EunAe; Ko, Jae-Joon; Ha, Heung Yong; Hong, Seong-Ahn; Lee, Kwan-Young; Lim, Tae-Won; Oh, In-Hwan (2003). Characteristics of the PEMFC Repetitively Brought to Temperatures below 0°C. Journal of The Electrochemical Society 150(12), A1667–A1670. DOI: 10.1149/1.1621877. https://doi.org/10.1149/1.1621877
  14. Srouji, Abdul-Kader; Mench, Matthew M. (2012). Freeze Damage to Polymer Electrolyte Fuel Cells. In: Polymer Electrolyte Fuel Cell Degradation, pp. 293–333. Academic Press. DOI: 10.1016/b978-0-12-386936-4.10006-5. https://doi.org/10.1016/b978-0-12-386936-4.10006-5
  15. Ge, S.; Wang, Chaoyang (2014). Ice Formation and Mechanical Stress in the PEFC during Cold Start. ECS Meeting Abstracts MA2014-01(3), 344. DOI: 10.1149/ma2014-01/3/344. https://doi.org/10.1149/ma2014-01/3/344
  16. Zhang, Ruixuan; Chen, Tao; Zhang, Rufeng; Gan, Zhongyu (2024). Effect of freeze–thaw cycles on membrane electrode assembly of proton exchange membrane fuel cells and its fault diagnosis method. Fuel Cells 24(2), 78–89. DOI: 10.1002/fuce.202300134. https://doi.org/10.1002/fuce.202300134
  17. Kocher, Katharina; Kolar, Stefan; Ladreiter, Walter; Hacker, Viktor (2021). Cold start behavior and freeze characteristics of a polymer electrolyte membrane fuel cell. Fuel Cells 21(4), 363–372. DOI: 10.1002/fuce.202000106. https://doi.org/10.1002/fuce.202000106
  18. Yu, Yongsheng; Zheng, Weibo; Li, Bing; Zhang, Cunman; Ming, Pingwen (2025). A comprehensive review of cold start in proton-exchange membrane fuel cells: Challenges, strategies, and prospects. Applied Energy 390, 125846. DOI: 10.1016/j.apenergy.2025.125846. https://doi.org/10.1016/j.apenergy.2025.125846
  19. Liu, Guoqiu; Lv, Shuangyu; Chen, Lei; Tao, Wen-Quan (2026). Optimization of self-cold start for PEM fuel cell with serpentine flow field based on three-dimensional transient model: synergy of initial conditions and start-up strategies. Renewable Energy 256, 124165. DOI: 10.1016/j.renene.2025.124165. https://doi.org/10.1016/j.renene.2025.124165