technical report
Sealing Fuel Cells for Peak Performance
- Senior Product Engineer
- Garlock Sealing Technologies
- P.E. Manager Business Development & Integration Pikotek-PSI
- Garlock family of companies
There are a number of different types of fuel cells, all of which are based on electrochemical reactions to convert chemical energy from a fuel into electricity. These technologies are frequently defined by the material used for the electrolyte separating the anode and cathode. Two of the most common types are solid-oxide (SOFC) and polymer electrolyte membrane (PEM) fuel cells.
SOFCs are designed primarily for stationary power applications which include distributed generation, providing primary or redundant power for office buildings, hospitals and other large facilities, giving them emergency backup and a degree of independence from the commercial power grid. They also are used as auxiliary power units in truck cabs, RVs and military vehicles and equipment, as well as for residential heat and power. PEMs, by contrast, are designed to provide portable power for automobiles, material handling and other mobile equipment.
The principal advantages of SOFCs are their high efficiency and ability to use a variety of fuels. Their biggest disadvantage is high operating temperatures (500̊C-1,000̊C; 1,200̊oF-1,800̊F), which require long startup times and can cause corrosion and degradation of cell components. Operating at much lower temperatures (50̊C-100̊C; 120̊F-212̊F), PEM fuel cells provide quicker startup, but require costly catalysts and are sensitive to fuel impurities. In both types, individual power-generating modules are stacked to deliver the requisite power for their intended application.
Critical components in these devices are the gaskets used to seal these modules to prevent mixture or leakage of air and fuel, which can adversely affect efficiency. Connections to ancillary components such as burners, pre-heaters and condensers also must be sealed.
Critical Factors
Among the factors to be addressed in gasket selection are temperature, the effects of water vapor, gasket compression, deflection and sealing characteristics, expected service life and maintenance. Unfortunately there are no standards for sealing fuel cells.
Fig. 1 Typical compressible gaskets (from left) mica-, talc- and vermiculite-based materials
Temperature is the first and perhaps most important consideration when evaluating gasket materials. As noted SOFCs have dramatically higher temperature requirements than PEM fuel cells. In addition some materials accommodate thermal cycling better than others.
Non-compressible glass and ceramic materials are commonly used to seal SOFCs. Glass offers the added advantage of being able to have its coefficient of thermal expansion controlled through crystallization. Both of these materials, however, are brittle, rigid and prone to cracking during movements induced by thermal cycling. Compressible seals are a better choice for SOFCs since they remain resilient even at extremely high temperatures (see Figure 1).
Sealing conditions are not particularly challenging with PEM fuel cells as their lower operating temperatures open a wide range of potential options. An adequate level of sealing can be achieved using ethylene propylene diene monomer (EPDM), polyurethane or silicone seals.
However special attention must be paid to material formulation to assure the gasket will not out-gas, leach into or extract from the cell fluids. These reactions will degrade the seal and compromise cell performance.
The media to be sealed is air and a combination of hydrogen, nitrogen, steam, carbon monoxide and carbon dioxide. Particularly critical gaskets are those that separate the air and fuel intakes, which are subject to long-term exposure to chemical oxidizing and reducing environments. In addition these gaskets must be able to withstand the deflective stresses from the expansion and contraction of cell components with different coefficients of thermal expansion.
Process Compatibility
When evaluating gasket materials for fuel cells, sensitivities to the electrochemical process and corrosion also must be taken into account. The gaskets must not only be compatible with the process, but provide electrical insulation to prevent the drain of current. The constituents of all the materials used in the construction of a fuel cell need to be assessed to determine if they are potentially corrosive to the metallurgy or could alter the fuel or electrolyte membranes. Most elastomeric gaskets, for example, contain sulfur which will form corrosive by-products such as dilute sulfuric acid when it combines with condensed water.
The required life expectancy for fuel cells is generally 40,000 hours, representing 4½ years of continuous service. This is acceptable for the present, but to be economically viable they will have last 8 to 10 years, which means the gasketing will have to last this long as well. In the case of SOFCs this poses a technological challenge due to extremely high operating temperatures.
The effectiveness of a seal is largely dependent on the design of the gasketed joint in which it is installed. Compared with industrial connections that exert stresses of 5,000 psi to 10,000 psi on a gasket, compressive loads on fuel cell joints can be as low as 5 psi. These low loads can make it difficult to seal SOFC joints. Because the gaskets usually contain very little elastomeric content, they require more compressive load to create a seal. Compressive load can be increased by modifying gasket geometry or using an alternate bolting scheme.
Connections should be designed to provide as much compressive load possible. There should be no metal-to-metal contact of the flanges, which can result in leaks if the gasket material creeps due to temperature and thermal cycling. Gasketing for SOFCs should be 20% to 35% compressible with 15% to 20% recovery. These values will facilitate a seal between thin flanges that may deflect when bolted together.
Fig. 2 Typical gasket for auxiliary connection to a SOFC
Fuel stack elements and joints to connect ancillary components may not be able to develop high compressive loads, since they do not have the same rigidity as standard pipe or equipment flanges. However internal pressures rarely exceed atmospheric conditions (see Figure 2).
Predicting Performance
Fig. 3 Compressive load vs. thickness for a talc-based gasket
The success of a gasket material for fuel cell applications is linked to the relationship of the compressive stress and the resulting seal tightness (Figure 3a). Stack height can be critical when planning for 30, 60 or more cells. If a gasket will not compress (deflect) within a defined range and maintain a seal, it will compromise cell performance.
Fig. 3a xCompressive load vs. leakage for a talc-based gasket
For each gasket thickness being evaluated, compression vs. deflection data should be compared. By combining compressive stress vs. seal tightness and deflection, expected performance can be assessed prior to validation in prototype or production units (see Figures 3 and 3a). Maintenance of key joint components also needs to be considered. Ideally gaskets should not leave a residue on the connection faces, due to the difficulty of removing it in the confined spaces for which many fuel cell assemblies are designed. Even if a gasket is removed intact, it cannot be used again due to plastic deformation which decreases compressibility.
Gasket extracts can affect electrochemical processes or corrode the materials of which the cell is constructed. The constituents of all materials used in fuel cell construction need to be assessed to determine if they are potentially corrosive, or will alter fuel quality, electrolyte or membrane. This type of information is beyond the scope of a material safety data sheet. Rather it calls for the sealing manufacturer to disclose the components and elemental composition of the gasket. This may require a non-disclosure agreement to protect both parties’ intellectual property.
A good starting point is for the cell manufacturer to identify the materials and their concentrations that are not desired in the system. For example, there are various grades of natural gas in terms of calorific value and purity. Lower grades and liquefied natural gas can contain high amounts of sulfur, which at elevated temperatures has detrimental long-term effects on metals. This concern is intensified by the fact that most fuel cells are designed to work with all the common grades of natural and liquefied gas.
As noted gasket materials with elastomeric binders may contain leachable sulfur and sulfates up to thousands of parts per million. Trace amounts of other potentially reactive materials in high-temperature fibers and fillers may be incompatible with the exotic custom alloys used in SOFCs or molten carbonate fuel cells (MCFC). In some cases these minute constituents will be published in a specification sheet, but are not the norm for a Material Safety Data Sheet (MSDS).
Moisture Effects
One of the by-products of fuel cell reactions is water vapor. Although seemingly benign, it can degrade gaskets and impair cell performance. That means using a hydrophobic gasket material that will not absorb moisture instead of a hydrophilic one that can even absorb humidity in the air stream. Steam vapor content of less than 5%, typical in high-temperature fuel cells, can be enough to degrade gaskets. If a condenser component fails, moisture levels could increase three to four times, magnifying the problem.
Gaskets can fail for a number of reasons, but the leading cause is insufficient compressive load due to the joint design or improper installation. Below is a summary of application variables and usual failure modes.
The demand for greater fuel cell efficiency and longevity will no doubt continue to increase as the demand for more environmentally friendly, alternative energy sources increases. Sealing these devices for peak performance will play a small but vital role in meeting these demands. Unfortunately gasket selection for fuel cell applications is sometimes an afterthought in the larger development process.
Yet the multiple variables summarized above, and their complex interactions with one another as well as various sealing materials call for a holistic approach to gasket selection.
This is best achieved through close collaboration with sealing manufacturers, even if that means sharing proprietary information that can impact both gasket and fuel cell performance.
References
Article: Chemical interaction between Crofer 22 APU and mica-based gaskets under simulated SOFC conditions, F. Wiener, M. Bram, H.-P. Buchkremer & D. Sebold
Article: Effect of sulfur-containing compounds on fuel cell performance, D. Imamura, E. Yamaguchi, Y. Hashimasa
Article: Sealing Glass-Ceramics for Planar Solid Oxide Fuel Cells, Alexander Fluegel
Article: Filled Glass Composites for Sealing of Solid Oxide Fuel Cells, Sandia Labs