A Radial Flow Characterization
Technique to Measure In Plane Permeability of Gas Diffusion Layers
Summary:
In-plane permeability of Gas Diffusion Layer (GDL) plays an important role in convective flow of reactants and products in polymer electrolyte fuel cells. A technique to measure in-plane permeability is developed which should assist with the ability to design fuel cells with increased convective performance. Our characterization technique uses radial permeability experiments to measure in-plane permeability of gas diffusion layers. These experiments can use either a wetting liquid or a gas of known viscosity as the host fluid and reach identical conclusions. However, flow rates dependence on pressure is different for gases and liquids and must be recognized when large pressure differentials are present. Results on three commonly used gas diffusion layer materials show that the non-woven material SGL31BA and the carbon fiber cloth material Avcarb 1071-HCB have in-plane permeabilities substantially higher than those reported for other materials throughout the literature.
Measurement Technique:
A radial flow apparatus shown here was fabricated to test samples of GDL for in-plane permeability at various levels of compressive strain. The samples consisted of annuli of material 15 cm OD x 9 cm ID stacked to a height ofapproximately 1 mm with each layer of material separated by thin layer of brass shim. Shim stock was used to control the total thickness of the compressed stack and avoid nesting effects. The characterization method was developed to work with a liquid or gas flow through the GDL. For gas permeability experiments, compressed air (0-550 kPa) was forced through the sample, passed from the outlet to a variable area rotameter for flowrate measurements. In liquid permeability experiments, a pressurized tank (0-200 kPa) forced water though the sample and was collected in a graduated cylinder at the outlet. In the case of gas permeability, pressure was measured using gauges at both the inlet and the outlet as rotometers were found to account for a significant pressure loss. For the case of liquid permeability, pressure was measured by a gauge on the inlet only and assumed to be atmospheric pressure at the outlet.
Results:
Permeability was measured on three types of gas diffusion layers: woven carbon fiber (cloth), non-woven carbon fibers, and carbon fiber paper. Measurements were taken at multiple levels of compression thought to correspond with typical levels in a fuel cell. The woven carbon fiber sample was Avcarb 1071-HCB (Ballard). The non-woven carbon fiber and paper based samples were SGL31BA (SGL Carbon) and TGP-60-H (Toray).
Summary:
In-plane permeability of Gas Diffusion Layer (GDL) plays an important role in convective flow of reactants and products in polymer electrolyte fuel cells. A technique to measure in-plane permeability is developed which should assist with the ability to design fuel cells with increased convective performance. Our characterization technique uses radial permeability experiments to measure in-plane permeability of gas diffusion layers. These experiments can use either a wetting liquid or a gas of known viscosity as the host fluid and reach identical conclusions. However, flow rates dependence on pressure is different for gases and liquids and must be recognized when large pressure differentials are present. Results on three commonly used gas diffusion layer materials show that the non-woven material SGL31BA and the carbon fiber cloth material Avcarb 1071-HCB have in-plane permeabilities substantially higher than those reported for other materials throughout the literature.
Measurement Technique:
A radial flow apparatus shown here was fabricated to test samples of GDL for in-plane permeability at various levels of compressive strain. The samples consisted of annuli of material 15 cm OD x 9 cm ID stacked to a height ofapproximately 1 mm with each layer of material separated by thin layer of brass shim. Shim stock was used to control the total thickness of the compressed stack and avoid nesting effects. The characterization method was developed to work with a liquid or gas flow through the GDL. For gas permeability experiments, compressed air (0-550 kPa) was forced through the sample, passed from the outlet to a variable area rotameter for flowrate measurements. In liquid permeability experiments, a pressurized tank (0-200 kPa) forced water though the sample and was collected in a graduated cylinder at the outlet. In the case of gas permeability, pressure was measured using gauges at both the inlet and the outlet as rotometers were found to account for a significant pressure loss. For the case of liquid permeability, pressure was measured by a gauge on the inlet only and assumed to be atmospheric pressure at the outlet.
Results:
Permeability was measured on three types of gas diffusion layers: woven carbon fiber (cloth), non-woven carbon fibers, and carbon fiber paper. Measurements were taken at multiple levels of compression thought to correspond with typical levels in a fuel cell. The woven carbon fiber sample was Avcarb 1071-HCB (Ballard). The non-woven carbon fiber and paper based samples were SGL31BA (SGL Carbon) and TGP-60-H (Toray).
