Hydrogen fuel cells are currently an area of extensive research as the automobile industry strives to produce vehicles that rely less on fossil fuels. The design of a polymer electrolyte fuel cell (PEFC) incorporates many layers, all of which have important roles in the successful function of the cell (Figure 1). These consist of a polymer membrane, catalyst layers, and diffusion media layers. The diffusion layers and their physical properties will be the focus of this page. These layers are made of carbon fibers and play a crucial role in enabling the hydrogen gas to access the catalyst, in addition to ensuring that water produced as a by-product during cell operation is properly removed before it can pool and reduce performance. Thermal management is important in this layer, and as conduction is the primary heat transfer mechanism in this material, thermal conductivity must be properly understood in order to predict fuel cell performance. Many factors can affect the performance of this layer, including compression and saturation. Xu (2013) from the University of Tennessee studied the thermal conductivity of three different diffusion media products as a function of these two parameters.
Figure 1. Diagram outlining the different layers that compose a polymer electrolyte fuel cell.1
Three different diffusion media products were tested in this experiment, and were given the sample names of MRC 105, SGL 25 BC, and Auto-Competitive (AC). A specialized sample set up was used in order to test the thermal conductivity of the materials under a specific amount of compressive pressure. Identical pieces of the same diffusion media were sandwiched around a two sided Hot Disk sensor, which were then sandwiched themselves between two 25 mm thick stainless steel disks of a known conductivity (Figure 2). A load cell was used to measure the pressure being applied. Thermal conductivity was tested at 0.1, 0.5, 1, 1.5 and 2 MPa. Tests of 10 seconds with a power output of 2.2 W were performed at room temperature. The Hot Disk Transient Plane Source 2500 S was the thermal conductivity instrument used for this work. This system performs absolute measurements with no calibrations required from 0.005 to 1000 W/mK.
Due to the thin nature of the diffusion media samples, the Thin Film testing module was used during measurement. This module is part of the Hot Disk software and is designed to measure the thermal conductivity of thin samples such as coatings and films. Three specially designed thin film sensors, one of which was used for this testing, accompany this software. The thin film module can measure thermal conductivities between 0.005 and 10 W/mK and handles samples that are 0.1 to 10 mm thick. Xu (2013) complemented this module by optically measuring the stress-strain relationship of the sample using a microscope and a high powered camera. This enabled him to take exact measurements of the sample thickness as it was being compressed, which was used to help determine the thermal conductivity under compression.
Measurements of thermal conductivity under various saturation levels were also performed. The enviably short test times and small power input of the Hot Disk TPS 2500 S make it an excellent choice for performing measurements on saturated samples, as it prevents vapour loss or convection from influencing results. Samples were fully saturated with de-ionized water and then allowed to dry to specified saturation levels for testing. All saturation measurements were carried out a pressure of 2 MPa.
Figure 2. Test set-up used by Xu (2013) to measure the thermal conductivity of diffusion media samples with the Hot Disk TPS 2500 S.
Effect of Compression on the Thermal Conductivity of Diffusion Media
Xu (2013) determined that compression and thermal conductivity had a positive correlation in diffusion media materials. When compressive strength was lower, so was the thermal conductivity. When compressive strength was higher, thermal conductivity followed suit (Figure 3). The decrease in porosity and thermal contact resistance at higher compression levels can account for this increase. The increase in thermal conductivity under higher pressures was quite significant; the thermal conductivity of Auto-Competitive increased by 440% between 0.1 and 2 MPa.
Effect of Saturation on the Thermal Conductivity of Diffusion Media
Increased saturation levels had the same positive effect as compression on the thermal conductivity. The thermal conductivity of MRC 105 increased by 62% as the saturation level increased from 0-6%. SGL 25 BC had a 35% increase in thermal conductivity with a jump from 0 to 25% in its saturation level (Figure 3). Xu (2013) also predicted the maximum thermal conductivity that could be attained at a maximum saturation level. As is displayed in Figure 3, these maximum predictions were not reached during experimental testing due to the tortuosity that exists in the diffusion media. Xu concluded that saturation can have a large effect on the thermal conductivity and heat transfer potential of the diffusion media, and that it must be accounted for when using models to predict fuel cell performance.
Figure 3. Thermal conductivity results obtained by Xu (2013) using the Hot Disk TPS 2500 S when testing the effect of compression (left) and saturation (right) on the thermal conductivity of diffusion media used in fuel cells.
This work is an excellent example of the importance of being able to accurately measure the thermal properties of components in complex systems such as polymer electrolyte fuel cells. The ability of the Hot Disk TPS products to provide fast, reliable measurements on materials that may prove difficult for other systems gives both academia and industry an important tool for research and development.