The alloys Cr5Fe1Y 2 O 3 and the ferritic steel Crofer22APU are typical alloys used as solid oxide fuel cell ͑SOFC͒ interconnect materials. Alloy Cr5Fe1Y 2 O 3 is an oxide dispersion strengthened ͑ODS͒ alloy developed by Plansee, Reutte, Austria, for use at high temperature. A typical material for medium-temperature SOFC, is the high chromium ferritic steel Crofer22APU supplied by Thyssen Krupp VDM, Germany. The two alloys form different oxide scales which affect chromium poisoning. Chromium vaporization as source term and electrochemical degradation of La 1−x Sr x MnO 3 ͑LSM͒ and La 0.58 Sr 0.4 Co 0.2 Fe 0.8 O 3 ͑LSCF͒ describing the poisoning were studied for the two alloys. The dynamics of the chromium deposition in porous perovskite cathodes was studied by a dc method and impedance spectroscopy. Electrical degradation of the LSM cathode by alloy Cr5Fe1Y 2 O 3 was significantly higher than for Crofer22APU. The microstructure of the cells was studied after measurements by scanning and energy filtering transmission electron microscopy. Significant amounts of chromium were observed at the TPB in the functional layer of cells, with the LSM cathode giving insight into the degradation mechanism. Cells tested with the LSCF cathode clearly show Cr poisoning. Formation of large SrCrO 4 crystals was observed on the surface of the LSCF cathode.

The vaporization of chromium species from chromia scales limits the applicability of chromia-forming steels at high temperatures and is one of the major reasons for degradation in the development of planar solid oxide fuel cells ͑SOFCs͒. Cr͑VI͒ vaporized from the interconnector is reduced at the cathode and deposits in the form of solid Cr͑III͒-oxide, thereby inhibiting the electrochemical processes. This work presents the first systematic study on the Cr vaporization of Cr-, Fe-, Ni-, and Co-based alloys in air and in H 2 atmospheres at high temperatures. The influence of outer oxide layers of ͑Cr,Mn͒ 3 O 4 , ͑Fe,Cr͒ 3 O 4 , Co 3 O 4 , TiO 2 , and Al 2 O 3 on the Cr vaporization is investigated. It is shown that the Cr vaporization of chromia-forming steels can be reduced by more than 90% by alloying. An estimate of the expected degradation effects on planar SOFC designs for the use of uncoated interconnector materials is used to show that in order to achieve the desired lifetimes for SOFC systems, additional Cr-retention coatings are necessary. Additionally, equilibrium vaporization measurements are carried out for pure Cr 2 O 3 ͑s͒ in humid air in order to elucidate controversies in the literature concerning the thermodynamic data of CrO 2 ͑OH͒ 2 ͑g͒.

No abstract

No abstract

The microstructural development of an alumina scale formed on a model FeCrAlY alloy during oxidation at 1200°C was characterized for up to 2000 hours of growth. Quantitative scanning electron microscopy (SEM) studies revealed that the scale had a columnar microstructure, with the grain size being a linear function of the distance from the scale/gas interface. For a given fixed distance from the scale/gas interface, there was found to be no change in the oxide grain size for exposure times ranging from 24 to 2000 hours at 1200°C, up to 100 hours at 1250°C. Thus, there was no significant coarsening of existing grains in the scale. Through oxygen tracer experiments, the scale-growth mechanism was shown to be predominated by inward oxygen diffusion along the oxide grain boundaries. Electron backscatter diffraction (EBSD) analysis further revealed that a competitive oxide-grain growth mechanism operates at the scale/alloy interface, which is manifested by a preferential crystallographic grain orientation. The scale-thickening kinetics were modeled using the experimentally-derived, microstructural parameters and were found to be in excellent agreement with converted thermogravimetric (TG) measurements. The model predicted a subparabolic oxidation rate, with the time exponent decreasing with increasing exposure time. The values of the time exponent were shown to be approximately 0.35 to 0.37, at oxidation times commonly reached in the TG experiments, i.e., a few tens of hours. At longer oxidation times of a few thousand hours and with a constant rate of average oxide-grain size increase, the time exponent was predicted to approach 0.33, corresponding to an ideal cubic oxidation rate.