GDM Intrusion Into Reactant Gas Channels and the Effect on Fuel Cell Performance

Author(s):  
Pinkhas Rapaport ◽  
Yeh-Hung Lai ◽  
Chunxin Ji

This paper reports on the study of gas diffusion media (GDM) intrusion into reactant gas channels and its effect on the performance of the proton exchange membrane (PEM) fuel cell. The PEM fuel cell under consideration consists of a membrane electrode assembly (MEA) sandwiched between two layers of gas diffusion media commonly made of carbon paper or cloth. The GDM/MEA/GDM assembly is then compressed between two adjacent bi-polar plates. In this configuration, the compression pressure is transmitted under the lands of the reactant gas flow-field onto GDMs on which the portion over the channels remain unsupported. Because of the relatively low bending and compressive stiffness, it is found that GDMs can easily intrude into the reactant gas channels. The direct consequence of GDM intrusion is the pressure drop increase in the reactant gases in the intruded channels. This is further compounded by cell-to-cell or channel-to-channel variation in GDM thickness and mechanical properties, which results in non-uniform reactant gas flow distribution and ultimately negatively impacts the fuel cell performance. In this study, we have developed a GDM intrusion model based on the finite element method (FEM. We have also devised an experimental setup to measure the GDM intrusion, in which we found good agreement between the model prediction and experimental measurement. Combining the FEM based intrusion model and a flow redistribution model we have investigated the effect of GDM channel intrusion on the reactant flow distribution and the impact on the fuel cell performance. It is found that a 20% reduction of reactant flow can be induced with a 5% additional blockage in channels by GDM intrusion. Based on the findings from the current study, we attribute the significant performance variation in a 30-cell fuel cell stack to the variation in reactant flow induced by the variation in GDM intrusion. The results from the analytical study and fuel cell testing both suggested that the product variations in GDM would need to be significantly reduced and the stiffness of the GDM would need to be increased if the PEM fuel cells of high power density were to be used reliably at a relatively low stoichiometry.

Author(s):  
J. P. Owejan ◽  
T. A. Trabold ◽  
D. L. Jacobson ◽  
M. Arif ◽  
S. G. Kandlikar

Water is the main product of the electrochemical reaction in a proton exchange membrane (PEM) fuel cell. Where the water is produced over the active area of the cell, and how it accumulates within the flow fields and gas diffusion layers, strongly affects the performance of the device and influences operational considerations such as freeze and durability. In this work, the neutron radiography method was used to obtain two-dimensional distributions of liquid water in operating 50 cm2 fuel cells. Variations were made of flow field channel and diffusion media properties, to assess the effects on the overall volume and spatial distribution of accumulated water. Flow field channels with hydrophobic coating retain more water, but the distribution of a greater number of smaller slugs in the channel area improves fuel cell performance at high current density. Channels with triangular geometry retain less water than rectangular channels of the same cross-sectional area, and the water is mostly trapped in the two corners adjacent to the diffusion media. Also, it was found that cells constructed using diffusion media with lower in-plane gas permeability tended to retain less water. In some cases, large differences in fuel cell performance were observed with very small changes in accumulated water volume, suggesting that flooding within the electrode layer or at the electrode-diffusion media interface is the primary cause of the significant mass transport voltage loss.


Author(s):  
Luis Breziner ◽  
Peter Strahs ◽  
Parsaoran Hutapea

The objective of this research is to analyze the effects of vibration on the performance of hydrogen PEM fuel cells. It has been reported that if the liquid water transport across the gas diffusion layer (GDL) changes, so does the overall cell performance. Since many fuel cells operate under a vibrating environment –as in the case of automotive applications, this may influence the liquid water concentration across the GDL at different current densities, affecting the overall fuel cell performance. The problem was developed in two main steps. First, the basis for an analytical model was established using current models for water transport in porous media. Then, a series of experiments were carried, monitoring the performance of the fuel cell for different parameters of oscillation. For sinusoidal vibration at 10, 20 and 50Hz (2 g of magnitude), a decrease in the fuel cell performance by 2.2%, 1.1% and 1.3% was recorded when compared to operation at no vibration respectively. For 5 g of magnitude, the fuel cell reported a drop of 5.8% at 50 Hz, whereas at 20 Hz the performance increased by 1.3%. Although more extensive experimentation is needed to identify a relationship between magnitude and frequency of vibration affecting the performance of the fuel cell as well as a throughout examination of the liquid water formation in the cathode, this study shows that sinusoidal vibration, overall, affects the performance of PEM fuel cells.


2016 ◽  
Vol 41 (4) ◽  
pp. 3023-3037 ◽  
Author(s):  
M. Rahimi-Esbo ◽  
A.A. Ranjbar ◽  
A. Ramiar ◽  
E. Alizadeh ◽  
M. Aghaee

2016 ◽  
Vol 853 ◽  
pp. 410-415 ◽  
Author(s):  
Xiang Shen ◽  
Jin Zhu Tan ◽  
Yun Li

A proton exchange membrane (PEM) fuel cell is an electrochemical device that directly converts chemical energy of hydrogen into electric energy.The structure of the flow channel is critical to the PEM fuel cell performance. In this paper, the effect of the cathode flow channel baffles on PEM fuel cell performance was investigated numerically. A three-dimensional model was established for the PEM fuel cell which consisted of bipolar plates with three serpentine flow channels, gas diffusion layers, catalyst layers and PEM. Baffles were added in the cathode flow channels to study the effect of the cathode flow channel baffle on the PEM fuel cell performance. And then, numerical simulation for the PEM fuel cell with various cathode channel baffle heights ranging from 0.2 mm to 0.6 mm was conducted.The simulated results show that there existed an optimal cathode flow channel baffle height in terms of the electrochemical performance as all other parameters of the PEM fuel cell were kept constant. It is found that the PEM fuel cell had the good electrochemical performance as the flow channel baffle heights was 0.4mm in this work.


Author(s):  
Isaac Perez-Raya ◽  
Abel Hernandez-Guerrero ◽  
Daniel Juarez-Robles ◽  
M. Ernesto Gutierrez-Rivera ◽  
J. C. Rubio-Arana

This work presents the results of a study of a new radial configuration proposed for the gas flow field for a PEM fuel cell. The objective of this study is to understand the effects of this configuration on the fuel cell performance. The results are compared with the radial designs proposed in previous analysis. The proposed designs on this work show an improvement on the cell performance, with a better use of the reaction area compared with a flow free radial design. The results also show that the effect of channeling the flow inside these radial configurations helps to improve the fuel cell performance.


Author(s):  
Shan Jia ◽  
Hongtan Liu

In a PEM fuel cell, it has been shown that the compression under the land area is the main reason for the observed higher performance than that under channel areas. If the area under the channel can also benefit from such a compression the overall performance of the cell will increase. Since the areas under the channel are not directly compressed in an assembled fuel cell, it is the objective of this study to determine if a cold pre-compression treatment of the gas diffusion electrode (GDE) may have a significant positive effect on the overall performance of the cell. First, the GDE is cold pre-compressed to a level similar to the compression that would be experienced by the land areas in an assembled fuel cell. Then the pre-compressed GDE is assembled in a regular test fuel cell and the performances under various operating conditions are studied. Finally, the cell performance results are compared with the results obtained from a fuel cell with a regular GDE. The experimental results show that cold pre-compress of the GDE has significantly improved the overall performance of the fuel cell. Further experiments have also been conducted with five different levels of cold pre-compression to determine if there exists an optimal compression and its value if it exists. The experimental results show that the performance of the fuel cell first increases with the level of cold pre-compression, reaching a maximum and then decreases with the level of compression. These results clearly indicate that there indeed exists an optimal level of compression. Further studies using both cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) have further corroborated the cell performance findings as well as the underlying mechanism. The results of EIS indicate that the ohmic resistance is hardly affected by the cold pre-compression, while the charge transfer resistance is significantly affected, especially in high current density region. The CV results show that the electro-chemical area (ECA) is higher with the cold pre-compressed GDE and there is an optimal compression that results in the maximum ECA. Therefore, the experimental results have shown that (a) the cold pre-compression treatment of the GDE is an effective and simple technique to increase PEM fuel cell performances; (b) there exists an optimal compression level at which the cell reaches its maximum performance; and (c) the increased performance is due to the increase of ECA resulting from the cold pre-compression treatment.


2021 ◽  
Author(s):  
Peng Quan

Fuel cells are electrochemical devices that directly convert the chemical energy of reaction of a fuel and an oxidant (usually hydrogen and oxygen) into electricity. Detailed measurements of properties such as the fluid flow distribution, species concentration, temperature distribution and local current densities are difficult to obtain during the operation of a fuel cell. Nonetheless, information like that is critical for improving the fuel cell performance, reducing cost and identifying possible failure mechanisms. Therefore, numerical models that are capable of producing this information would be very useful for design and optimization as well as improved fundamental understanding of transport processes in fuel cells. In this project a two-dimensional non-isothermal and non-isobaric computational model of a PEM fuel cell is formulated. The model is coupled with a computational fluid dynamics model for diffusive transport in the electrodes and convective transport in the gas flow channel, which is built principally upon the conservation laws for mass, momentum and energy. Then the following four phenomenological equations are also involved: the Stefan-Maxwell equation for the description of multi-species diffusion; the Bulter-Volmer equation for the description of first-order reaction kinetics; the Nernst-Planker equation for the description of proton flux through membrane; and the Schlögl equation for the description of water velocity in membrane. The focus of the project is placed on the study of partial hydration in the membrane. For this propose [sic], we develop our membrane model partly based on empirical relationships to account explicitly for water diffusion, pressure distribution and electro-osmotic drag; meanwhile, thermal effects are introduced implicity via the variation of membrane transport parameters as functions of temperature. Although the solution of the present model is beyond the scope of the project, we are trying to embed the general solution ideas in our modeling work by giving, for all computation domains, detailed boundary conditions and corresponding properties and parameters that are critical for the prediction accuracy of the model.


Sign in / Sign up

Export Citation Format

Share Document