Search

• Quasi in-situ observation of catalyst-coated membrane cracks was achieved. • A freeze–thaw method under saturated humidity condition was established. • Cracks tend to extend along the larger pores formed between agglomerates. • Cracks in the catalyst layer led to the appearance of cracks in the membrane. • The membrane cracks can even traverse the entire membrane. The evolution of catalyst-coated membrane (CCM) under saturated humidity conditions during freeze–thaw cycling was investigated, with a particular focus on the evolution of initial cracks within the catalyst layer (CL) and their impact on the proton exchange membrane (PEM). A quasi in-situ fuel cell fixture was designed to enable direct observation of the CCM via scanning electron microscopy (SEM). As the number of freeze–thaw cycles increases, the initial cracks in the CL gradually propagated, with new cracks appearing around them. These CL cracks tend to extend along the larger pores, which essentially represent gaps between agglomerates. The overall trajectory of formed cracks appears to be influenced by the flow channel direction. Additionally, under freeze–thaw cycling, cracks in the CL led to the formation of cracks in the PEM. A strong correlation is found between the direction of membrane cracks and CL cracks, suggesting a direct impact of the CL defects on the structural integrity of the PEM. Furthermore, some cracks at the tips of CL cracks even penetrate through the entire membrane. This study provides valuable insights into the degradation mechanisms of CCMs under low-temperature environmental conditions and could inform future design improvements for enhanced durability. [ABSTRACT FROM AUTHOR]

Assembly error, misalignment of components, is inevitable in the fabrication of proton exchange membrane (PEM) fuel cells. It will induce complex and contradictory effects, such as the higher contact resistance and better gas permeability, making it difficult to evaluate the error tolerance. These coupled impact on the performance of fuel cell needs to be clarified. This study establishes a performance prediction model considering the contact resistance between bipolar plate (BPP) and gas diffusion layer (GDL), deformation of membrane electrode assembly (MEA), and distribution of porosity and permeability in GDL to analyze the effect of assembly error. Assembly error brings better performance due to the higher porosity, permeability, and a more uniform distribution of oxygen. However, contact resistance also increases. Cases with narrow channels, thinner GDL, and lower compression ratios are more susceptible to assembly error, mainly manifested in the higher contact resistance, lower porosity and permeability growth, and less oxygen distribution improvement. Cases with wider channels, thicker GDL, and higher compression ratios have a higher tolerance for assembly error. Considering that contact resistance has a greater impact in most cases, assembly error needs to be avoided. The tolerance of assembly error is proposed to guide the fuel cell fabrication. • A method for analyzing the effects of assembly error on performance is proposed. • Results of the assembly model are input into the performance prediction model. • MEA deformation, the distribution of GDL porosity and permeability are considered. • Effects of channel width, GDL thickness, and compression ratio are analyzed. • Tolerance of assembly error under different conditions is proposed. [ABSTRACT FROM AUTHOR]

Contact behavior between the gas diffusion layer (GDL) and bipolar plate (BPP) is of significant importance for proton exchange membrane fuel cells. Most current studies on contact behavior utilize experiments and finite element modelling and focus on fuel cells with graphite BPPs, which lead to high costs and huge computational requirements. The objective of this work is to build a more effective analytical method for contact behavior in fuel cells and investigate the size effect resulting from configuration alteration of channel and rib (channel/rib). Firstly, a mathematical description of channel/rib geometry is outlined in accordance with the fabrication of metallic BPP. Based on the interface deformation characteristic and Winkler surface model, contact pressure between BPP and GDL is then calculated to predict contact resistance and GDL porosity as evaluative parameters of contact behavior. Then, experiments on BPP fabrication and contact resistance measurement are conducted to validate the model. The measured results demonstrate an obvious dependence on channel/rib size. Feasibility of the model used in graphite fuel cells is also discussed. Finally, size factor is proposed for evaluating the rule of size effect. Significant increase occurs in contact resistance and porosity for higher size factor, in which channel/rib width decrease. [ABSTRACT FROM AUTHOR]

Fuel cell assembly plays a dominant role in performance and lifetime of proton exchange membrane (PEM) fuel cell. Most current methods for assembly design are based on experiment and finite element (FE) model, which need high cost and huge computation for the whole fuel cell. Unfortunately, there are only few theoretical methods which contribute to the whole stack assembly, especially those in 3D model. This study develops a comprehensive methodology to simulate the stack assembly and to design the clamping displacement/pressure. At first, contact pressure field on the GDL is predicted with the use of the continuous equivalent model. The total contact resistance and porosity of fuel cell are proposed as the evaluation indexes combined by the desirability of function method. Then, in order to validate the contact pressure prediction, experiments with dimensional-error metallic bipolar plate are carried out and the numerical results show good agreements with experimental results. At last, clamping pressure of the endplate is calculated to optimize the assembly process. The methodology in this study is beneficial to the understanding of the internal contact behavior of the whole stack and helpful to guide the assembling of PEM fuel cell. [ABSTRACT FROM AUTHOR]

• A type of gradient wettability cathode CL with three sub-layers is designed. • Gradient CL's performance (895 mW/cm2@0.6 V) is higher than normal (721 mW/cm2@0.6 V). • This CL can adapt high current density, cathode humidity and air stoichiometry. • Gradient CL with best performance is found through comparison of 18 structures. The design of cathode catalyst layer (CL) is essential to improve the mass transfer capacity of proton exchange membrane (PEM) fuel cell and increase power density. In this work, cathode CL is divided into three sub-layers, and each sub-layer is added nano-particles with different wettability. The gradient CL with hydrophilic SiO 2 particles at inner layer and hydrophobic polytetrafluoroethylene (PTFE) particles at outer layer significantly enhances the performance of membrane exchange assembly (MEA). Its performance is 895 mW/cm2@0.6 V, which is 24.1 %higher than CL without any particles (721 mW/cm2@ 0.6 V). Under operating conditions of high current density, high cathode humidity and high air stoichiometry, the gradient CL has only a little voltage loss. Through Electrochemical Impedance Spectroscopies (EIS) impedance analysis under high current density (1.8A/cm2), mass transfer resistance of gradient CL is 25.4 Ω, and is much smaller than the mass transfer resistance of the homogeneous CL of 35.1 Ω, which reflects the significant enhancement in mass transfer capacity of gradient CL. The gradient catalyst layer is suitable for a wider range of current density, humidity, and stoichiometry, but excessive cathode gas stoichiometry causes a decrease in performance, which is caused by excessive drainage capacity. In addition, 18 different gradient CLs are designed and manufactured, and the gradient CL with catalyst coated membrane (CCM) structure has the best performance. In gradient CL, increasing the capillary pressure difference between sublayers is the key to performance improvement. It is confirmed that the property of MEA with appropriate wettability gradient design can be significantly improved. [ABSTRACT FROM AUTHOR]