Investigation of convective transport in the gas diffusion layer used in polymer electrolyte fuel cells

Abstract

Recent experimental data on a fuel cell-like system revealed insights on the fluid flow in both free and porous media. A computational model is used to investigate the momentum and species transport in such system, solved using the finite elements method. The model consists of a stationary, isothermal, diluted species transport in free and porous media flow. The momentum transport is treated using different formulations, namely Stokes-Darcy, Darcy-Brinkman, and a hybrid StokesBrinkman formulations. The species transport is given by the advection equation for a reactant diluted in air. The formulations are compared to each other and to the available experimental data, where it is concluded that the Darcy-Brinkman formulation reproduces the data appropriately. The validated model is used to investigate the contribution of convection in reactant transport in porous media of fuel cells. Convective transport provides a major contribution to reactant distribution in the so-called diffusion media. For a serpentine channel and flow with Re = 260 to 590, convection accounts for 29 to 58% of total reactant transport to the catalyst

Spatially resolved oxygen reaction, water, and temperature distribution: Experimental results as a function of flow field and implications for polymer electrolyte fuel cell operation

“…manifold and seal) to the flow path in a serpentine configuration causes a substantial improvement in the performance of a PEFC [58] which also suggests the contribution of secondary flow to more effective mass transport of reactant species, in agreement with the present findings. Similar effect was observed in a modelling study [59] Figure 2 and on our recent modelling results [18], vortices arising at U turns might probably be responsible for the accumulation of water at these regions by pushing species away from the gas pathway, thus hindering its transport down the channel. As pointed out earlier, these vortex regions are seen as regions of high reactant partial pressure (e.g.…”

“…Data presented in Figure 2 also illustrates that the reactant partial pressure at the catalyst layer is increased upon U turns in the serpentine-type flow fieldsthere is a decrease in reaction upstream of the U-bend but increased reaction at the downstream side of the U-Bend. This phenomenon was also seen in the computational model [18], being interpreted as a consequence of secondary flow, in the form of vortices, which strongly shape the partial pressure and distribution of reactant gas throughout the catalyst layer and positively impacts PEFC performance. The existence of such vortices has also been measured through laser Doppler anemometry analysis of an in operando PEFC, [57].…”

“…It is also noteworthy that this second section (1st peak) highlights the previously assigned convective transport effects [22], such as the spread of reactant after each turn of the channel, and the reduced partial pressure before each turn. Finally, the third section (2nd peak, between 60−95 Pa) illustrates similar features to the second section (1st peak), while strongly emphasizing the most prominent feature assigned to the convective transport [18,22], that is secondary flows leading to local maxima of reactant partial pressure at the second corner of each turn. It should be noted that there is some "mixing" between the assigned sections, of course, as the reactant partial pressure drops along the channel.…”

“…The analysis will first focus on the SS flow field, in order to correlate with the other imaging techniques used. We shall return later (section 4) to compare between flow fields, using the species distribution surfaces presented in Figure 2 Recently, we have shown that the validated computational model strongly suggests that the ozone transport closely reproduces the transport of oxygen in an actual fuel cell [18]. Therefore, the response in Figure 2 closely represent the distribution of oxygen gas along the catalyst layer of an actual PEFC.…”

“…[17] These findings suggest that, species distribution through fuel cell layers is an important step towards understanding and improving fuel cell performance, durability and reliability, since it enables us to understand and optimize the flux of reactants and heat distribution through each cell component with experimentally validated advanced computational models. [18] Therefore, in situ techniques enabling a high-resolution detection of species and heat distribution through fuel cell layers are of paramount importance towards practical applications of such devices. Some in situ techniques have been developed over the years to characterize polymer electrolyte fuel cells (PEFCs), as can be seen in references [5,19,20].…”