K. Scott Weil

Selection and Evaluation of Heat-Resistant Alloys for SOFC Interconnect Applications

Over the past several years, the steady reduction in SOFC operating temperatures to the intermediate range of 700~850oC [1] has made it feasible for lanthanum chromite to be supplanted by metals or alloys as the interconnect materials. Compared to doped lanthanum chromite, metals or alloys offer significantly lower raw material and fabrication costs. However, to be a durable and reliable, a metal or alloy has to satisfy several functional requirements specific to the interconnect under SOFC operating conditions. Specifically, the interconnect metal or alloy should possess the following properties: (i) Good surface stability (resistance to oxidation, hot corrosion, and carburization) in both cathodic (air) and anodic (fuel) atmospheres; (ii) Thermal expansion matching to the ceramic PEN (positive cathode-electrolyte-negative anode) and seal materials (as least for a rigid seal design); (iii) High electrical conductivity through both the bulk material and in-situ formed oxide scales; (iv) Bulk and interfacial thermal mechanical reliability and durability at the operating temperature; (v) Compatibility with other materials in contact with interconnects such as seals and electrical contact materials.

Chemical Stability of Glass Seal Interfaces in Intermediate Temperature Solid Oxide Fuel Cells

In intermediate temperature planar solid oxide fuel cell (SOFC) stacks, the interconnect, which is typically made from cost-effective, oxidation-resistant, high-temperature alloys, is typically sealed to the ceramic positive electrode-electrolyte-negative electrode (PEN) by a sealing glass. To maintain the structural stability and minimize the degradation of stack performance, the sealing glass has to be chemically compatible with the PEN and alloy interconnects. In the present study, the chemical compatibility of a barium-calcium-aluminosilicate (BCAS) based glass-ceramic (specifically developed as a sealant in SOFC stacks) with a number of selected oxidation resistant high temperature alloys (and the yttria-stabilized zirconia electrolyte) was evaluated. This paper reports the results of that study, with a particular focus on Crofer22 APU, a new ferritic stainless steel that was developed specifically for SOFC interconnect applications.

Reactive Air Brazing: A Novel Method of Sealing SOFCs and Other Solid-State Electrochemical Devices

High-temperature electrochemical devices operate via an ion gradient that develops across a solid electrolyte. Consequently, hermeticity across this membrane is paramount. Not only must the electrolyte contain no interconnected porosity, but it must be connected to device chassis with a gas-tight seal. Here we report a new method of brazing developed specifically for solid-state electrochemical applications. We demonstrate that the seal is hermetic and resistant to thermal aging, can be thermally cycled under rapid heating rates with no measurable loss in seal strength, and has shown promise in sealing full-size planar solid oxide fuel cell ~pSOFC! components.

Rupture Testing as a Tool for Developing Planar Solid Oxide Fuel Cell Seals

One of the critical issues in designing and fabricating high-performance planar solid oxide fuel cell (pSOFC) stacks is the ability to hermetically seal adjacent metal and ceramic components. In our pSOFC development program, we have designed a testing technique that allows us to screen through the numerous variables involved in developing glass seals. Using this test for example, we have found that the composition of the metal component plays an important role in the strength of the seal. Microstructural analysis of as-sealed specimens revealed that an interfacial reaction zone forms during joining, and it appears that the thickness and composition of this layer are the dominant parameters that control joint strength. In this paper the details of the seal test are reported. The results have proven particularly significant in the development of the next-generation stack design. Supporting microstructural and chemical analyses collected on the test specimens are also presented and used to interpret the seal test results in an effort to identify the necessary steps toward improving glass pSOFC seals.

Joining Mixed Conducting Oxides Using an Air-Fired Electrically Conductive Braze

Pacific Northwest National Laboratory, Richland, Washington 99352, USAThe lanthanum strontium cobalt ferrites ~LSCF! are a well-known family of mixed oxygen-ionic/electronic conductors that are ofinterest for use in high-temperature electrochemical devices, such as solid oxide fuel cells and oxygen separation membranes. Oneof the challenges however in developing these types of devices is joining the constituent metallic and ceramic components,particularly with a technique that is compatible with the electrochemically active membrane materials. A new joining process hasbeen developed which employs a silver-copper oxide braze material. In the present study, we found that braze compositionscontaining between 1.4 and 16 mol % CuO offer the best combination of wettability, joint strength, and electrical conductivity. Ourresults indicate that the wettability of the Ag-CuO brazes on LSCF substrates increases with CuO content and that electricalconductivity through the joint remains high as long as the silver content is greater than 66 mol %. Results from long-term,high-temperature resistance experiments confirmed the latter finding, as minimal degradation in conductivity was observed in thevarious test joints over 500 h of testing at 750°C. Data obtained from three-point bend testing showed that joints formed with abraze composition of 1.4 and 16 mol % CuO displayed maximal average flexural strength.© 2004 The Electrochemical Society. @DOI: 10.1149/1.1764782# All rights reserved.Manuscript submitted September 26, 2003; revised manuscript received January 22, 2004. Available electronically June 17, 2004.

Alternative planar SOFC sealing concepts

Abstract One of the challenges in manufacturing planar solid oxide fuel cells (pSOFCs) is in hermetically sealing the ceramic and metallic components such that the resulting joint remains rugged and stable over the lifetime of the stack. Traditionally, glass joining or compressive sealing has been used. While short-term success has been achieved with these techniques, it is apparent that to meet the long-term operational needs of stack designers, alternative sealing concepts will need to be conceived. At Pacific Northwest National Laboratory we have been developing two such alternatives, air brazing and bonded compliant sealing, the details of which are outlined here.

New sealing concept for planar solid oxide fuel cells

A key element in developing high-performance planar solid oxide fuel-cell stacks is the hermetic seal between the metal and ceramic components. Two methods of sealing are commonly used: (a) rigid joining and (b) compressive sealing. Each method has its own set of advantages and design constraints. An alternative approach is currently under development that appears to combine some of the advantages of the other two techniques, including hermeticity, mechanical integrity, and minimization of interfacial stresses in the joint substrate materials, particularly the ceramic cell. The new sealing concept relies on a plastically deformable metal seal; one that offers a quasi-dynamic mechanical response in that it is adherent to both sealing surfaces, i.e., non-sliding, but readily yields or deforms under thermally generated stresses. In this way, it mitigates the development of stresses in the adjacent ceramic and metal components even when a significant difference in thermal expansion exists between the two materials. The pre-experimental design of the seal, initial proof-of-principle results on small test specimens, and finite-element analyses aimed at scaling the seal to prototypical sizes and geometries are described herein.

Development of Ti-6Al-4V and Ti-1Al-8V-5Fe Alloys Using Low-Cost TiH2 Powder Feedstock

Thermo-mechanical processing was performed on two titanium alloy billets, a beta-titanium alloy (Ti1Al8V5Fe) and an alpha-beta titanium alloy (Ti6Al4V), which had been produced using a novel low-cost powder metallurgy process that relies on the use of TiH2 powder as a feedstock material. The thermomechanical processing was performed in the beta region of the respective alloys to form 16-mm diameter bars. The hot working followed by the heat treatment processes not only eliminated the porosity within the materials but also developed the preferred microstructures. Tensile testing and rotating beam fatigue tests were conducted on the as-rolled and heat-treated materials to evaluate their mechanical properties. The mechanical properties of these alloys matched well with those produced by the conventional ingot processing route.

Materials Properties Database for Selection of High-Temperature Alloys and Concepts of Alloy Design for SOFC Applications

To serve as an interconnect / gas separator in an SOFC stack, an alloy should demonstrate the ability to provide (i) bulk and surface stability against oxidation and corrosion during prolonged exposure to the fuel cell environment, (ii) thermal expansion compatibility with the other stack components, (iii) chemical compatibility with adjacent stack components, (iv) high electrical conductivity of the surface reaction products, (v) mechanical reliability and durability at cell exposure conditions, (vii) good manufacturability, processability and fabricability, and (viii) cost effectiveness. As the first step of this approach, a composition and property database was compiled for high temperature alloys in order to assist in determining which alloys offer the most promise for SOFC interconnect applications in terms of oxidation and corrosion resistance. The high temperature alloys of interest included Ni-, Fe-, Co-base superalloys, Cr-base alloys, and stainless steels. In the US alone, there are hundreds of commercial compositions produced, over 250 of which are listed in Appendix A. Two initial criteria (oxidation resistance and oxide scale electrical conductivity) were used to reduce the list of alloys to manageable proportions. Thermal expansion and fabrication characteristics were then considered to further reduce the list of stainless steels. Due to their outstanding oxidation resistance and their potential to be used in SOFC components that can exclude alumina scales from the stack electrical path, alloys with a sufficient amount of aluminum were classified into a separate alumina-forming alloy category. The down-selected compositions (approx. 130 in number) and their characteristics and/or applications are listed in the Selected Alloy Compositions tables (Appendix B). Following the down-selection of alloy compositions, materials properties of interest corresponding to the their functional requirements in SOFC stacks were compiled in a tabular form (Appendix C). For comparison, the properties of selected noble metals and intermetallics were also collected and compiled and are listed in a separate table in Appendix C. Analysis of the pertinent literature indicated that, for a wide variety of alloys, there remains a lack of information on specific materials properties. Also, we have observed a large scatter in the reported database. For those cases, we employed general alloying principles as a tool of choice to approximate the unavailable data and to evaluate the reliability and consistency of collected data. Though numerous high temperature alloys look promising, it is anticipated that there will be few, if any, “off the shelf” alloy compositions which could completely satisfy the materials requirements as an interconnect, especially for a long term in a specific SOFC design. Therefore, some concepts of alloy design, including composition, constitution, and structure, as well as their effects on properties relevant to SOFC applications, are elaborated in an attempt to provide guidance for modification of current compositions and development of new alloys. Acknowledgement: This work was funded by the U.S. Department of Energy’s National Energy Technology Laboratory (NETL) under the Core Technology Program (CTP) of the Solid-State Energy Conversion Alliance (SECA).

Effect of Pd additions on the invariant reactions in the Ag-CuOx system

Palladium was added as a ternary component to a series of copper oxide-silver alloys in an effort to increase the use temperature of these materials for potential ceramic brazing applications. Phase equilibria at low palladium and copper oxide concentrations in the Pd−CuOx−Ag system were determined experimentally using differential scanning calorimetry, microstructural analysis, and x-ray diffraction. Small additions of palladium were generally found to increase the temperature of the eutectic reaction present in the pseudobinary system but have little effect on a higher temperature monotectic reaction. However once enough palladium was added (≈5 mol%) to increase the new eutectic temperature to that of the original pseudobinary monotectic reaction, the pseu doternary monotectic temperature correspondingly began to move upward as well. The addition of palladium also forced the eutectic point to slightly lower silver concentrations, again causing a convergence with the former monotectic line.