Diffusion layer costs

Chemical polishing, yielding a surface of high specular reflectivity, exploits fully optimised bright dip solutions often achieved by the further addition of phosphoric acid at the expense of the residual water. Because phosphoric acid is relatively viscous at lower temperatures (e.g. less than 40°C) it can act as diffusion layer promoter (C), but its presence increases the chemical costs considerably. 

Another important parameter that has to be taken into account when choosing the appropriate diffusion layer is the overall cost of the material. In the last few years, a number of cost analysis studies have been performed in order to determine fuel cell system costs now and in the future, depending on the power output, size of the system, and number of xmits. Carlson et al. [1] reported that in 2005 the manufacturing costs of diffusion layers (for both anode and cathode sides) corresponded to 5% of the total cost for an 80 kW direct hydrogen fuel cell stack (assuming 500,000 units) used in the automotive sector. The total value for the DLs was US 18.40 m-, which included two carbon cloths (E-TEK GDL LT 1200-W) with 27 wt% P ILE, an MPL with PTFE, and Cabot carbon black. Capital, manufacturing, tooling, and labor costs were included in the total. 

In 2007, the same consulting company published another report in which the cost of the DLs had increased slightly to 6% of the overall cost of the stack, compared to the 5% previously estimated [2]. One issue with these analyses and predictions was that they were based on carbon cloth as the diffusion layer, but this material does not reflect what most of the fuel cell companies use (i.e., carbon fiber paper) [3]. 

In another report, James and Kalinoski [4] performed an estimation of the costs for a direct hydrogen fuel cell system used in automotive applications. The assumed system consisted of an 80 kW system with four fuel cell stacks, each with 93 active cells this represents around 400 MEAs (i.e., 800 DLs) per system. The study was performed assuming that the DL material used for both the anode and cathode sides would be carbon fiber paper with an MPL. In fact, the cost estimate was based on SGL Carbon prices for its DLs with an approximate CEP value of around US 12 m for 500,000 systems per year. Based on this report, the overall value of the DLs (with MPL) is around US 42.98 per kilowatt (for current technology and 1,000 systems per year) and 3.27 per kilowatt (for 2015 technology and 500,000 systems per year). Figure 4.2 shows the cost component distribution for this 80 kW fuel cell system. In conclusion, the diffusion layer materials used for fuel cells not only have to comply with all the technical requirements that different fuel cell systems require, but also have to be cost effective. 

Proton exchange membrane (PEM) fuel cells are the primary choice for transportation systems, but they can also be useful for stationary power production or local hydrogen production. Most of the challenges of PEM fuel cell commercialization center around cost and materials performance in an integrated system. Some specific issues are the cost of catalyst materials, electrolyte performance, i.e., transport rates, and water collection in the gas diffusion layer (GDL). 

In PCR, the forward to reverse period ratio is typically between 20/1 and 30/1. The PCR has two main effects. The current density may be increased without anode passivation. The reversal current depletes built up metal concentration within the anodic boundary layer. This helps in avoiding the precipitation of metal salts, which is one of the causes of anode passivation. On the cathode, thinning of the diffusion layer and selective removal of nodules during the reverse current phase result in smoother deposits. The major disadvantage of PCR is higher energy costs. 

A number of different methods exist for the production of catalyst layers [97-102]. They use variations in composition (contents of carbon, Pt, PFSI, PTFE), particle sizes and pds of highly porous carbon, material properties (e.g., the equivalent weight of the PFSI) as well as production techniques (sintering, hot pressing, application of the catalyst layer to the membrane or to the gas-diffusion layer, GDL) in order to improve the performance. The major goal of electrode development is the reduction of Pt and PFSI contents, which account for substantial contributions to the overall costs of a PEFC system. Remarkable progress in this direction has been achieved during the last decade [99, 100], At least on a laboratory scale, the reduction of the Pt content from 4.0 to 0.1 mg cm-2 has been successfully demonstrated. 

The electrodes represent, in stack manufacture, the main outlay (77%), while membranes (6%), gas diffusion layers (5%), and bipolar plats (5%) follow spaced out. Electrode costs are strictly related to the very expensive Pt content. Thus,... 

The cost of non-active materials (gas diffusion layer, membrane, and bipolar plates) dominate stack cost at very low platinum loadings, while ohmic losses limit the benefit of increasing platinum loading beyond some point. 

In addition the Siemens employees can use the know-how gathered in 40 years of developing and producing electrodes. Electrodes are the core element of a PEM electrolysis system. After they are coated on a membrane it is called membrane electrode assembly (MEA). Packed with two bipolar plates (BIP) and one gas diffusion layer (GDL), one electrolysis cell is ready. In this context a rough rule of thumb says that the MEA is responsible for the lifetime, the BIP causes the costs. 

The analysis of carbon materials used as catalyst support, gas diffusion layer, and current collector and bipolar plates is performed in Chap. 7. A number of carbon materials including carbon blacks, nanotubes, nanofibers, and structured porous carbon materials are analyzed and compared as catalyst support in direct methanol fuel cells. Commercial and non-commercial gas diffusion layers are described along with the role of the mesoporous layer on the fuel cell performance. Finally, synthetic graphite and carbon composites used as current collector and bipolar plates are discussed, focusing on their mechanical and electrical properties and production costs. 

Probe-beam deflection is a technique in which a monochromatic source, typically a He-Ne laser beam, passing parallel to the electrode surface, is used to monitor refractive index changes in the diffusion layer. It is a simple and cost-effective way of profiling the diffusion layer, and is used to monitor the diffusion layer ingress and egress of ions, particularly protons. It is often used in conjunction with electrochemical quartz crystal microbalance (EQCM) measurements that monitor mass changes within or on the electrode layer itself. 

Dow Chemical Co., Asahi Chemical Co. and Chloride Engineers Ltd. make a similar product. The polymer membrane only conducts H" " when fully hydrated and one solution to this problem has been to use carbon fiber wicks (Figure 23.9). SGL Carbon is making fiber based gas diffusion layers in a roll form. The cell is sensitive to low levels of CO, which can be removed by a Pt/Ru catalyst, but the high cost restricts use. However, a later technique to increase the surface area has reduced the cost by a factor of 75. 

From Fig. 9.4, it becomes clear that the stack is the most expensive single component of the fuel cell system, and that on stack level, the catalyst is the most expensive single component. It is important to notice that all costs are expressed per kW system output. It can therefore be important to improve a compmient that has a big impact on fuel cell performance without being expensive itself. When a better performing membrane or gas diffusion layer leads to a doubling of the power density while the catalysts cost per square meter stays the same, its cost... 

---