Search

Summary: The mechanical degradation of the proton exchange membrane (PEM) is one of the main aspects affecting the lifetime of proton exchange membrane fuel cells (PEMFCs). It was observed in our previous study that the stress/strain distribution in the PEM of fuel cells with metallic bipolar plates (BPPs) is more complex, owing to manufacturing and assembly errors of the BPPs. The present study further concentrates on the stress/strain evolution in the membrane of fuel cells throughout the assembly and running processes by a finite element model. In membranes at the joint area between the gasket and gas diffusion layers, a serious stress concentration aggravated as the misalignment displacement increases. As for the membrane in reaction area, the plastic strain reaches highest level at the center of the groove after hygrothermal loading. The maximum stress is mainly relevant to the temperature and humidity and has little concern with the misalignment. The model and results of this study offer guidance regarding the design of PEMFC. Owing to the stress concentration, an additional protection should be set in the joint area, and the assembly error should be limited within 0.05 mm. [ABSTRACT FROM AUTHOR]

Abstract Proton exchange membrane (PEM) fuel cells usually operate in dead-ended anode mode due to a comparatively simple system. Nevertheless, flooding in the anode channels in dead-ended mode is severe than that in flow-through configuration, which causes cell performance degradation and durability decrease. In this study, water droplet dynamics in the anode channel is investigated numerically using volume of fluid method to study water accumulation and drainage in the PEM fuel cell with the dead-ended anode. Simulations are divided into a dead-ended stage and a purge stage to study the two-phase flow behaviors. Impact of water accumulating volume is taken into consideration and, cases of different wettability of the gas diffusion layer surface and the bipolar plate surface are compared. The numerical results reveal that water droplets emerge from the water inlets and then accumulate and coalesce in the dead-ended stage. Most water droplets are drained out of the gas channel along the channel corners in the purge stage. It is found that larger water accumulating volume results in higher eliminating rate. The total purge time is mainly affected by the wettability of the bipolar plate surface. Highlights • Water droplet dynamics in a dead-ended anode channel is investigated numerically. • The majority of water droplets drain away along the channel corners. • Most water flows away from the gas channel in the middle period of the purge. • More accumulated water results in higher drainage rate. • The total purge time is highly affected by the wettability of the BPP surface. [ABSTRACT FROM AUTHOR]

Abstract Mechanical degradation, caused by local stress concentration and variation, significantly affects the lifetime of proton exchange membrane fuel cells. This study constitutes the first numerical investigation of stress evolution in the membrane between the frame of the membrane exchange assembly (MEA) and gas diffusion layer (GDL) throughout the processes of assembly, operation and gas filling in fuel cells. A finite element model is outlined to determine mechanical deformation of the membrane by exerting assembly displacement, hygrothermal conditions and gas pressure in turn. It is observed that severe stress concentration and bending deformations occur in the joint-area membrane. The results show that a plastic deformation occurs after the temperature and water content are increased, and would be substantially enhanced by the gas pressure difference between the anode and cathode. The in-plane stress may throw some light on the rapid degradation of the membrane between the frame and GDL. The gas pressure difference, which exceeds 10 kPa, leads to a rapid increase in the in-plane stress and plastic deformation. Decreasing the joint width may not be a good approach for reducing the stress/strain concentration. It is suggested that additional gasket seals or adhesive protection layers are helpful in joining frame and GDL. Highlights • Numerical model of fuel cells is developed to investigate membrane degradation. • Stress concentration and plastic deformation are caused in the joint-area membrane. • As much as 70% increase is found in the in-plane stress after gas filling. • Gas pressure difference exceeding 10 kPa leads to a rapid increase in deformation. • The maximum stress and plastic strain are obtained if joint width is 0.05–0.15 mm. [ABSTRACT FROM AUTHOR]

Highlights • Gas diffusion layer accounts for as much as 20% of the resistance in fuel cells. • Parallel fiber microstructure leads to lower electrical resistance. • Resistance of carbon paper and carbon cloth declines as the load cycles increases. • Fiber crack and loss in resistance occur when exceeding "break stress" of 2.0 MPa. • Carbon felt has the highest stability in electric resistance and microstructure. Abstract Electrical resistance accounts for a significant part of the performance loss in proton exchange membrane fuel cells. To the best of the authors' knowledge, this work represents the first direct experimental investigation and comparison of the bulk resistance and microstructure of commercially available gas diffusion layers, carbon paper, carbon cloth and carbon felt, under cyclic and steady loads, which are typical compression conditions in the fuel cell. It was found that with the improvement of contact conductivity between gas diffusion layer and bipolar plate, the bulk resistance of gas diffusion layer accounts for as much as 20% of the resistance in the fuel cell, especially when the assembly pressure is high enough. Experimental results indicate that three kinds of gas diffusion layers show various electrical behaviors under compression due to their different fiber structures. For carbon paper, the resistance displays a gradual decline as the load cycles increases. A reduction in the resistance and obvious fiber cracks are observed when the compression pressure exceeds the "break stress" of 2.0 MPa. For woven carbon cloth, more uniform decline of the resistance is caused by the increasing fiber cracks, which are pulled and bent in the middle of a weave. Although felt gas diffusion layer features the lowest electrical conductivity, its tortuous and thick fibers lead to higher stability in electric resistance and microstructure than bonded carbon paper and woven carbon cloth. This study is helpful for enhancing our understanding of the relationship between electrical resistance and compression loads in the fuel cell with various gas diffusion layers. [ABSTRACT FROM AUTHOR]

Metallic bipolar plates (BPPs) are promising candidates for replacing conventional carbon-based BPPs. However, manufacturing errors are unavoidable on metallic BPPs due to the characteristics of fabrication processes. This study is conducted to further develop tolerance design method for dimensional error and assembly error based on the contact behavior model proposed in our previous work. Contact pressure distribution, contact resistance and gas diffusion layer (GDL) porosity are used as the evaluation indexes combined by desirability function method. Based on the contact behavior model, effect of dimensional error is substituted into the model by Monte Carlo Simulation, and misalignment between the upper plate and lower plate is included in the model to obtain the effect of assembly error. It is found that dimensional error and assembly error contribute to uneven pressure distribution, larger contact resistance and higher GDL porosity. The acceptable maximum dimensional error of 0.007, 0.012, 0.010 mm and assembly error of 0.080, 0.10, and 0.060 mm are recommended for the fuel cells with the channel widths of 1.0, 2.0 and 4.0 mm, respectively. This methodology developed is beneficial to the understanding of the effect of dimensional error and assembly error, and effective supplement to fabrication management of metallic BPPs. Image 1 • Tolerance design method for fabrication error of metallic BPPs is introduced. • Response variables are almost in linear rise with increasing the dimensional error. • Assembly error smaller than 0.05 mm contributes little to the response variables. • Dimensional error of 7, 12, 10 μm considered acceptable for channels of 1, 2, 4 mm. • Assembly error of 80, 100, 60 μm considered acceptable for channels of 1, 2, 4 mm. [ABSTRACT FROM AUTHOR]