The SOFC

Solid oxide fuel cells operate at temperatures at which certain oxidic electrolytes become oxygen ion, O2-, conducting. It is the same effect that takes place in the lambda sensor supplied with three-way catalytic converters in spark ignition cars2, and lambda sensors can be used as convenient lab models for SOFCs. The oxides employed are mixtures of yttria and zirconia, and their use goes back as far as early work by Nernst — see Chapter 2.

The two electrode reactions are expressed as:

2 A lambda sensor can be viewed as a SOFC operated as an oxygen concentration cell.

Overall cell reaction and standard reversible potential are the same as for the other fuel cells and, as was the case for the MCFC, water is generated at the anode.

The discussion of the advantages and disadvantages of the high-temperature concept runs entirely parallel to the case of the MCFC, and what was said in Section 8.1.2 applies similarly to the SOFC. The SOFC benefits from excellent kinetics at anode and cathode. However, for thermodynamic reasons, the reversible potential at the operating temperature is somewhat lower than for low-temperature fuel cells (Chapter 3).

Inherent advantages of the SOFC are the entirely solid-state design and, in contrast with the PEMFC, the absence of water management problems. Yet materials problems, particularly related to sealing and thermal cycling, are even more severe than with MCFC technology. In fact, the search for the right stack design has been a focus of active research and development for decades — and still is.

Some information on the tolerance of SOFC technology with respect to fuel impurities is summarized in Table 8.1.

SOFC technology uses two basic designs. The planar design follows the same principle as discussed with other fuel cells, while currently by far the most advanced design is that of tube bundles, originally developed by Westinghouse since the mid-1960s and now commercialized by Siemens-Westinghouse.

Tubular SOFC Designs

The tubular design originates from sealing problems with planar SOFC stacks. Its principle is shown in Fig. 8.1. Fuel and air are supplied to the outside and the inside, respectively, of extended solid oxide tubes closed on one end. In the so-called air electrode supported (AES) technology, the tube itself forms the cell cathode or air electrode — compare Fig. 8.3(a). The tubes are sealed only at the open end, where

Air Cathode

FIGURE 8.1 (a) Siemens-Westinghouse air electrode supported (AES) tube bundle. (b) Schematic arrangement of tubes in a power plant. (Photograph and drawing courtesy of Siemens-Westinghouse.)

FIGURE 8.1 (a) Siemens-Westinghouse air electrode supported (AES) tube bundle. (b) Schematic arrangement of tubes in a power plant. (Photograph and drawing courtesy of Siemens-Westinghouse.)

Exhaust 850°C

Process Air 630°C

Exhaust 850°C

Process Air 630°C

Mcfc Brennstoffzelle

Desulfurized Natural Gas

FIGURE 8.2 Seal-less mounting of an SOFC tube in a fuel cell stack. (Courtesy of Siemens-Westinghouse.)

Desulfurized Natural Gas

. Combustion Plenum Depleted Fuel Recirculation Plenum Cell Stack At 1000°C

- Fuel Ejector • Stack Reformer Prereformer

FIGURE 8.2 Seal-less mounting of an SOFC tube in a fuel cell stack. (Courtesy of Siemens-Westinghouse.)

some stress relief may be allowed by appropriate sealing or by an entirely seal-less construction — see Fig. 8.2. The gases are supplied via manifolds and, in the case of air, through alumina dip tubes protruding deep into the fuel cell tubes.

At operating temperatures of 1000°C, the choice of materials and manufacturing techniques is particularly critical in order to avoid thermal stress, unwanted sintering, and corrosion. SOFC cathodes are usually made of doped lanthanum manganite. The porous cathode tubes in Siemens-Westinghouse fuel cells have a diameter of 22 mm and a total length of 1810 mm with an active length of 1500 mm, equivalent to 834 cm2 active area. Onto these tubes, a stripe of lanthanum strontium chromite is plasma sprayed as the interconnect, as shown in Fig. 8.3(a). Currently, the deposition of the 40-micron yttria (Y2O3) stabilized zirconia (ZrO2) or YSZ electrolyte layer is done by an expensive electrochemical vapor deposition (EVD) process. The final coat of the nickel intermixed YSZ anode or Ni/ZrO2 cermet is applied by dip coating followed by simultaneous firing in two different atmospheres for cathode and anode (Blum et al., 2001). The tubes deliver approximately 150 W electric power at 950°C each (Blum et al., 2001), which corresponds to an area power density of 0.180 W cm-2.

Despite its successes — solid oxide fuel cells are currently the fuel cells with the longest operating record, some 69,000 h for single tubes (Williams, 2001) — the tubular design in its present form suffers from two major problems. First, manufacturing of the tubes and assembly of the tube bundles are process and labor-intensive production steps, and it is hard to envisage major breakthroughs in cost-effective mass production. Second, the tubular design optimizes the gas flow reactor design at the expense of the electronic features. The tubes are connected in series or in parallel using the lanthanum-stron-tium-chromite interconnects and flexible Ni felts, which can only make good contact at one line along the circumference — see Fig. 8.3(a). Most current flows along long segments of the tubes and incurs significant ohmic losses. Therefore, tubular SOFC technology usually exhibits inferior power densities compared to planar fuel cells.

Siemens-Westinghouse is seeking to address the cost issue by producing more key components in-house rather than sourcing them3. Part of the production is going to be automated, which will increase output, and two costly electrochemical vapor deposition steps have already been cut out of the production leaving just the electrolyte EVD step (Williams, 2001). Siemens-Westinghouse is also considering a novel tube design, shown in Fig. 8.3(b). The flattened tube design or high power density (HPD) SOFC stack features electronically conducting ribs inside the air inlet, which make the dip tubes formerly

3Previously, cathode support tubes were purchased at an approximate cost of $1600 each (Stover et al., 2001).

Siemens Westinghouse Tubular
Nickel Felt
Flat Tube Sofc Design

FIGURE 8.3 (a) Cross-section of a Siemens-Westinghouse air electrode supported (AES) tube showing attached interconnect with Ni felt. (b) Cross-section of the flattened tube according to Siemens-Westinghouse's novel high power density (HPD) SOFC design. The ribs in the AES tube improve electronic conductivity, allow higher packing densities, and enable air guidance without the need for additional tubing. (Drawings courtesy of Siemens-Westinghouse.)

FIGURE 8.3 (a) Cross-section of a Siemens-Westinghouse air electrode supported (AES) tube showing attached interconnect with Ni felt. (b) Cross-section of the flattened tube according to Siemens-Westinghouse's novel high power density (HPD) SOFC design. The ribs in the AES tube improve electronic conductivity, allow higher packing densities, and enable air guidance without the need for additional tubing. (Drawings courtesy of Siemens-Westinghouse.)

employed redundant and help to improve electronic conductivity. The overall tube length will be reduced to one-third. This design also gives a better space filling and, hence, higher volumetric power densities, which are projected to rise from currently 136 to 388 kW m-3 (Stover et al., 2001).

Other developers of tubular fuel cells include TOTO Ltd., Kyushu Electric Power Co., and Nippon Steel Corporation, co-funded by the NEDO New Sunshine Project, and Mitsubishi Heavy Industries (MHI), in cooperation with Electric Power Development Co. (EPDC) and Chubu EPC. Using 2.2-cm-diameter, 90-cm-long tubes manufactured by sintering technology, the former consortium has developed 3-kW modules (36 cells) operating under atmospheric conditions and on reformed natural gas at an area-based power density of up to 0.192 W/cm2 (Nakayama and Suzuki, 2001), very similar to the Siemens-Westinghouse technology. The MHI group uses the concept of segmented, staggered tubes of 72 cm length which consist of 22 cells connected in series4. The tubes are currently produced by plasma spraying. Fuel is supplied through the inside of the tubes. Pressurized operation on coal gas at initially 10-kW module size is envisaged (Iritani et al., 2001).

4 This concept is also known as "bell-and-spigot" design (EG&G, 2000).

Planar SOFC Designs

Despite Siemens' decision to discontinue its in-house development of planar fuel cell technology in 1998 and to focus on the tubular concept within the newly founded Siemens-Westinghouse Power Corporation, there is widespread belief that, in the future, cost-effective SOFCs will be based on planar designs. With this in mind, the U.S. Department of Energy has launched the Solid State Energy Conversion Alliance (SECA), with a vision of producing a modular, 5-kW, low-cost planar fuel cell stack that will be used in a wide range of applications (Williams, 2001). But what are planar SOFC developers actually trying to achieve?

The underlying goal is primarily the development of a cost-effective production method and the increase in area power density, which has the strongest impact on cost per kilowatt of power.

One aspect with a particularly strong potential for cost cutting is operation at lower temperatures in the 750 to 850°C (approximately 1400 to 1550°F) range, rather than the 950 to 1000°C (approximately 1750 to 1850°F) common in current technology5. This will require the deposition of very thin (5 to 10 micron) electrolyte layers because YSZ electrolyte conductivity drops dramatically with decreasing temperature. The benefits of low-temperature operation are the use of lower-cost stainless steel materials, for example for the cell separators, and probably largely improved stability towards thermal cycling.

Three strands of planar technology are currently being developed by a wide range of groups. One approach uses frames for one or more cells per plane. The best known example of this technology is the former planar Siemens SOFC, which now continues to be developed by a start-up, Entwicklungsgesellschaft Brennstoffzelle GmbH, and Fraunhofer-IKTS in Germany (Lequeux, 2001). It employs metal frames for up to 16 YSZ electrolyte-supported cells per plane. The Rolls Royce (U.K.) concept combines features of the planar with advantages of the tubular design in a ceramic multicell arrangement (Lequeux, 2001). DLR (Germany) is working on a framed single-cell arrangement, with the cell built up in layers by plasma spraying onto a porous metal substrate support (Stover et al., 2001).

The second approach is that of a Japanese group consisting of Mitsubishi Heavy Industries, Chubu Electric Power Company, and Electric Power Research and Development Center, who have jointly developed a monolithic fuel cell stack called MOLB (mono-block layer built). It is based on an embossed YSZ electrolyte layer and electrodes made of the standard materials LSM and Ni/YSZ cermet (Sakaki et al., 2001). The embossed cell structures, separated from each other by planar interconnects, also serve as flow distributors. In the advanced, t-MOLB version, ten cells are combined in one unit stack, and ten unit stacks are combined to form one train, as shown in Fig. 8.4. Performance is now at 0.35 W cm-2 (Sakaki et al., 2001), but the operating temperature is probably around 1000°C.

Stack Mcfc
FIGURE 8.4 t-MOLB type SOFC: several 10-kW class stack in 2000. (Photograph courtesy of A. Nakanishi of Electric Power Research and Development Center, Chubu Electric Power Company, Nagoya.)

5 In a similar fashion, this also applies to tubular SOFC technology.

Sulzer Hexis Sofc
FIGURE 8.5 Schematic of Sulzer Hexis planar cell concept. The drawing shows the gas flow (fuel inlet through the center of the stack) and cell interconnects. (Drawing courtesy of Sulzer Hexis.)

Finally, there are a vast number of cell designs in which the cells are built up layer by layer. These can be either electrolyte supported as in the case of the Sulzer Hexis (Switzerland) SOFC system or, now more commonly, anode supported. Sulzer, the most advanced developer in Europe with domestic CHP units in field trials, uses partially stabilized YSZ electrolytes and screen-printed anodes and cathodes (Batawi et al., 2001), as illustrated in Fig. 8.5. At least some components are or have been sourced from InDEC (see below). The rationales for using anode- rather than cathode-supported cells are "the higher thermal and electrical conductivity, superior mechanical strength, and minimal chemical interaction with the YSZ electrolyte at high temperatures encountered during cell fabrication" (Singhal, 2001). In such cells, electrolyte layers of 5-20 microns are considered feasible compared to 50-150 microns in electrolyte-supported cells (Singhal, 2001), where the electrolyte itself has to carry mechanical strength. Power densities of 1.8 W cm-2 at 800°C have been achieved with anode-supported cells (Kim et al., 1999). But with this type of information, it is always important to know what the exact operating parameters such as fuel and oxidant flow are, and what would be the lifetime at high power operation.

The layer-by-layer construction has the advantage of using low-cost volume production techniques known from ceramics processing in the electronics industry, i.e., tape casting for the base layer, as shown in Fig. 8.6, followed by screen printing of the subsequent layers (Larsen et al., 2001; Rietveld et al., 2001). Sintering steps take place after each step or, ideally, not until the final deposition step (Rietveld et al., 2001). Tape casting, in combination with screen printing and the use of belt furnaces, offers the potential of cheap mass production. In Denmark, Ris0 National Laboratory, three utilities, Haldor Tops0e, and IRD have embarked on a five-year, partially government-funded program to develop SOFC technology to the pilot-plant level. The SOFC is anode-supported, low-temperature (750-850°C) technology with Fe-Cr metal interconnects and deformable stubs for building up stacks, as shown in Fig. 8.6 (Larsen et al., 2001).

Elsewhere in Europe, ECN (the Netherlands), previously working with Siemens, has set up a small-scale production company, InDEC, in order to supply ceramic electrolyte-supported (Sulzer Hexis) and

Cathode current collection

Cathode

Electrolyte

Anode

Anode support

Metallic interconnect

FIGURE 8.6 Current stack design with metallic interconnect foils (Fe-Cr), anode-supported cells, and stub contact layers. (Drawing courtesy of P.H. Larsen, Ris0 National Laboratory.)

Cathode current collection

Cathode

Electrolyte

Anode

Anode support

Deformable stubs for current collection and gas distribution

Metallic interconnect

FIGURE 8.6 Current stack design with metallic interconnect foils (Fe-Cr), anode-supported cells, and stub contact layers. (Drawing courtesy of P.H. Larsen, Ris0 National Laboratory.)

anode-supported materials to stack developers using screen printing onto tape cast substrates. Partnerships exist with Alstom (U.K.), Forschungszentrum Jülich (Germany), and Prototech (Norway) as well as Fuel Cell Technologies Corporation (Canada) to develop technology for small-scale CHP systems.

The activities of the Jülich group focus on the development of an anode-supported cell system with a thin 5- to 10-micron electrolyte layer. Short five-cell stacks of 20 x 20 cm sizes have been produced (de Haart et al., 2001), and area power densities exceeding 0.5 W cm-2 have been achieved, albeit with smaller cells (Stöver et al., 2001). Jülich identified the stack joining and sealing technique used so far, solder glass, as one current technical problem (Stöver et al., 2001).

In North America, Pacific Northwest National Laboratory and Delphi Automotive Systems are working on the development of anode-supported SOFC stacks for 5-kW, 42-V systems for automotive auxiliary power supplies (APU), as discussed in Section 9.2.3 (Singhal, 2001). A first generation of SOFC stacks for APUs was supplied to Delphi by Global Thermoelectric (Canada) and gave a cell performance of 1 W cm-2 (hydrogen, 800°C) and a stack performance of 0.37 W cm-2 on reformate (750°C) (Mukerjee et al., 2001). Global Thermoelectric has developed low-temperature SOFC technology since 1997, when base technology was acquired from Forschungszentrum Jülich. Current cell sizes are limited to 10 x 10 cm, but Global appears to have found reliable high-temperature sealing and stack compression techniques (Ghosh et al., 2001).

Other North American developers include SOFCo (U.S.), a subsidiary of McDermott International Inc., and its partner Ceramatec Inc. This group wants to develop SOFC stacks based on tape casting, screen printing, and co-firing process technology (Elangovan et al., 2001). Honeywell (formerly Allied Signal) is also among SOFC systems developers. Stacks are of planar design (100 cm2) and give area power densities of up to 0.285 W cm-2 on syngas at 800°C (Minh et al., 2001). Again, a thin electrolyte anode-supported approach is taken, presumably with metallic interconnects (Minh et al., 1996).

As was said before, none of these developments has reached a state of maturity similar to that of the currently available tubular systems. When these fuel cell stacks are brought to the pilot system stage, a whole set of new materials problems may have to be faced. A good example is anode redox stability. So far, it has been acknowledged only for the Sulzer Hexis system that after shutdown at high temperatures, cells can be irreversibly damaged by oxidation of the anode due to air leakage and subsequent re-reduction. This problem has now been addressed by advanced anode formulations (Rietveld et al., 2001).

Guide to Alternative Fuels

Guide to Alternative Fuels

Your Alternative Fuel Solution for Saving Money, Reducing Oil Dependency, and Helping the Planet. Ethanol is an alternative to gasoline. The use of ethanol has been demonstrated to reduce greenhouse emissions slightly as compared to gasoline. Through this ebook, you are going to learn what you will need to know why choosing an alternative fuel may benefit you and your future.

Get My Free Ebook


Responses

  • petra
    How The Power Tube has been designed as a superior airflow management work?
    6 years ago

Post a comment