No abstract

Solid oxide fuel cells (SOFC) are of great interest for cleaner energy production. As the technology moves to more widespread commercial adoption, balance of plant (BOP) components such as heat exchangers and other structural components are becoming increasingly important [1]. These components operate at ~600-950°C in the presence of aggressive species such as water vapor. Currently used chromia-forming Fe-based stainless steels and Ni-based alloys suffer from accelerated oxidation under these operation conditions due to enhanced internal oxidation and Cr oxy-hydroxide volatility [1,2]. Further, the volatilization of Cr species from BOP alloys results in Cr poisoning of the SOFC stack and a concomitant reduction in performance [1]. The use of alumina-forming alloys in SOFC BOP applications is a promising approach to reducing Cr contamination of the SOFC stack [3]. Alumina scales offer 1 to 2 orders of magnitude lower oxide growth rate than chromia and are far more stable in water vapor. Most alumina-forming alloys also make use of Cr alloying additions to promote the formation of the protective alumina; however, studies suggest > 10x reduction in Cr release can be achieved despite the presence of Cr in these alumina-forming alloys [3]. Alumina-forming austenitic (AFA) alloys are a new class of stainless steels that offer the potential for improved oxidation resistance via an alumina scale combined with good creep resistance due to their austenitic structure, in a lower cost Fe-based alloy formulation [4]. The alloy design compromises to achieve both creep resistance and alumina formation result in a loss of protective oxidation and a transition to internal oxidation of Al with increasing temperature, typically in the range of ~750-900°C for wrought 20-25Ni mass % based AFA alloys depending on composition [5]. (A 35Ni-based cast grade of AFA alloy capable protective alumina formation to 1100-1200°C is available but is not suitable for SOFC BOP components) [6]. The goal of the present work was to develop a lower cost, 25 Ni mass % grade AFA alloy capable of long-term SOFC BOP use at ~800-950°C in water vapor containing environments. The composition range of interest was based on Fe-25Ni-(13-18)Cr-4Al-(1-2.5)Nb-(0.5-2)Mn-(0.15-0.5)Si-(0-2)W-(0-2)Mo-(0-1)Cu-(0-0.1)Ti-(0.1)V-(0.05-0.2)Zr-(0.05-0.2)Hf-(0-0.1)Y-(0.02-0.2)C-(0.005-0.015)B mass %. The effort focused on systematic variation of C, Cr, and Nb levels ± Hf, Y, and Zr, which impact oxidation resistance, manufacturability, and cost. Key findings to achieve oxidation resistance at ≥850-900°C in air with 10% H2O without loss of creep resistance included: Protective alumina formation was enhanced by use of Zr additions, rather than previously demonstrated, more costly additions of Hf and Y. Levels of Nb as low as 1.5 mass % were sufficient to achieve protective alumina formation, as compared to previously used 2.5 mass % Nb levels, which reduces cost and aids manufacturability. The Cr addition levels were critical to pushing protective alumina formation past 900°C, with the critical transition occurring between ~14.5 and 16 mass % Cr. Protective alumina formation was achieved with lower than expected C addition levels, as low as 0.02 to 0.03 mass % C, which aids manufacturability to thin product forms by reducing primary coarse NbC formation. Details of the alloy design strategy and oxidation behavior out to 10,000 h of exposure at 900-1000°C in air with 10% H2O will be presented. Insights for the oxidation mechanism relative to Cr evaporation behavior will also be discussed. References L. Zhou, J. H. Mason, W. Li, and X. Liu, Renewable and Sustainable Energy Reviews Volume 134, December 2020, 110320 S.R.J. Saunders, M. Monteiro, F. Rizzo, Progress in Materials Science, 53 (2008) 775-837. M. Stanislowski, E. Wessel, T. Markus, L. Singheiser, W.J. Quadakkers, Solid State Ionics 179 (2008) 2406–2415. Y. Yamamoto, M.P. Brady, M.L. Santella, H. Bei, P.J. Maziasz, B.A. Pint, Metallurgical and Materials Transactions A, 42 (2011) 922-931. M.P. Brady, K.A. Unocic, M.J. Lance, M.L. Santella, Y. Yamamoto, L.R. Walker, Oxidation of Metals, 75 (2011) 337-357 M. P. Brady, G. Muralidharan, Y. Yamamoto, B.A. Pint, , Oxidation of Metals, 87 (1), pp. 1-10 (2017) Notice: This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The US government retains and the publisher, by accepting the article for publication, acknowledges that the US government retains a nonexclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for US government purposes. DOE will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan).

ZMGTM232G10 is Fe-Cr ferritic alloy, which has better oxidation resistance for SOFC interconnects. Oxidation behavior of ZMG232G10 was investigated in air at 750 and 850°C for 40,000 hours. Bulk specimens of ZMG232G10 showed stable oxidation behavior according to parabolic law. However, thinner specimens showed anomalous oxidation behavior for 40,000 hours at 850°C. Furthermore, ferrite to austenite phase transformation was observed in a part of specimens which was oxidized at 850°C. This unstable oxidation behavior resulted from Cr consumption in alloy matrix by oxidation.

Solid Oxide Fuel Cell (SOFC) is known as clean and high efficiency co-generation systems. SOFC operates at high temperatures (600 – 900°C), therefore, heat resistant materials are used for SOFC components. For SOFC interconnects that connect between ceramic cells electrically, the required properties are as follows; good oxidation resistance at operating temperature, high electrical conductivity at operating temperature and thermal expansion close to ceramic cells. There are many research works about long term degradation behavior of anode, cathode and electrolyte materials, because the lifetime of SOFC is expected from 40,000 hours (about 5 years) to 90,000 hours (about 10 years). On the other hand, degradation behavior of metal interconnects, that means oxidation behavior of metal interconnects, has not been evaluated sufficiently despite relevance to ohmic loss of SOFC stack. Fe-Cr ferritic alloy for SOFC interconnects “ZMGTM232G10” was developed and its long term oxidation resistance has been evaluated. At this time, we will report 40,000 hours’ oxidation test result about ZMG232G10, for example the changes of oxidation weight gain and microstructure. The alloy was melted with 10kg lab-scale vacuum melting furnace, and then hot forged to bar shape. After that, forged material was cut and cold rolled to sheets with some target thicknesses. Sample sizes cut from sheets are 10mm square with 3mm thickness and 15mm square with 0.3 and 0.5mm thicknesses. The surfaces of samples were ground with #1,000 dry paper. Oxidation test was performed at 850°C in air, which is an accelerated condition compared to 750°C. The test pieces were put in the ceramic containers, and put into electric furnaces. The oxidation weight gain of each test piece was measured after each 500 hours’ oxidation test period. After the measurement, some specimens were cut, plated with Ni, mounted in the resins, ground and polished, and then the cross-sectional microstructures were observed with optical microscope and EPMA and analyzed with EPMA. Figure1 shows the oxidation weight gain of ZMG232G10 vs. exposure time at 850°C in air. In general, oxidation behavior of the alloy with dense oxidation layer is explained with parabolic law. ZMG232G10 with 3mm thickness (bulk sample) showed stable oxidation behavior for 40,000 hours according to parabolic law. However, oxidation behavior of thinner samples deviated from parabolic law, and anomalous oxidation was observed at the edge of the samples. Microstructure of oxide layer of ZMG232G10 consisted of (Mn, Cr, Cu) spinel layer and Cr2O3 layer located beneath spinel layer. Ferrite phase and oxide layer were observed in almost all samples, however martensitic transformation under the oxide layer was observed in thinner sample after long term oxidation test. It indicated that ferrite phase transformed to austenite phase during long-term oxidation test and then austenite phase transformed to martensite during cooling to room temperature. This is the reason why martensite was observed at room temperature. Because the protective oxide layer of ZMG232G10 is mainly Cr2O3, Cr are consumed gradually from the alloy matrix during high temperature oxidation. Then, Cr contents in the alloy matrix were analyzed with EPMA. ZMG232G10 with 3mm thickness contained Cr amount enough to keep protective oxide layer after 40,000 hours oxidation test. However Cr amount in thin samples decreased faster with exposure time than thick one. It was difficult to keep protective oxide layer dense because of less total Cr amount and decreasing Cr in the alloy matrix. As a result, faster oxidation and earlier anomalous oxidation occurred in thin samples than thick one. Furthermore, low Cr content was not able to keep ferrite phase stable and then caused transformation from ferrite phase to austenite phase at high temperature. Through long term oxidation test on ZMG232G10, we obtained the knowledge as follows ; (1) although bulk sample of ZMG232G10 shows stable oxidation behavior according to parabolic law, thin ones show faster oxidation and earlier anomalous oxidation than thick one. (2) In thin samples, ferrite phase may transform to austenite phase in long term oxidation. Both phenomena were caused by less total Cr amount in thin samples and rapid decrease of Cr at high temperature for long term oxidation than thick one. Figure 1