A statistically-based thermal conductivity model for fuel cell Gas Diffusion Layers

Many of the characterization and theoretical analyses of Gas Diffusion Layers (GDL) transport properties have to date been done as a function of porosity with little attention paid to other geometrical properties. In this paper, a statistical unit cell approach is presented for predicting the thermal conductivity of GDLs by introducing geometrical properties such as the intersecting fiber angles and characteristic distances/aspect ratios. These geometrical properties are deconvoluted analytically using optical and porometry data. The dependency of the thermal conductivity on the geometrical parameters is analyzed, and the GDL structure for optimal heat conduction is identified. It is shown that aspect ratio is as important as porosity in determining conductivity, and that the traditional notion that a porous medium with higher porosity has a lower thermal conductivity is not always true. The maximum thermal conductivity depends primarily on porosity and fiber angle, occurs at an aspect ratio of around unity, and shows negligible dependence on fiber diameter. The geometrical concepts and the measured data presented in this paper help unravel some unexplained trends reported in the literature and provide novel insights on avenues to optimize GDLs. The methodology can be extended to estimate other transport properties such as permeability.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► A statistical model is developed for estimating GDL thermal conductivity.► The model considers the statistical distribution of GDL geometrical parameters. ► Angle and aspect ratio distributions are measured for some GDLs. ► The dependency of thermal conductivity on geometrical parameters is considered. ► New techniques are introduced for GDL geometrical property measurements.