Doctor of Philosophy
Faculty of Engineering
Reid, Mark H., Metal components of ceramic oxide fuel cells, Doctor of Philosophy thesis, Faculty of Engineering, University of Wollongong, 2012. http://ro.uow.edu.au/theses/3647
The subject of this thesis is the metallurgical stability of an alloy selected for use in a ceramic oxide fuel cell. Fuel cells are a new and developing power generation technology with numerous designs and materials being considered for the construction of fuel cells. One particular design, a multilayer, laminated fuel cell stack utilising a stabilised zirconia wafer as an electrolyte is under development by Ceramic Fuel Cells Ltd. The functional elements of the fuel cell also include the anode (a nickel/zirconia cermet), the cathode (a Sr doped LaMnO3), and an electrically conductive interconnect to allow multiple wafer stacks for the generation of usable voltages. In addition, the interconnects provide a heat sink (due to the exothermic nature of the fuel oxygen reaction) and a high thermal conductivity pathway for heat to be extracted from the fuel cell (for secondary power generation). The oxygen conductivity of the zirconia wafer requires that the fuel cell operates in the temperature range of 950-1000ºC.
The material requirements of the interconnect components are challenging. They must be resistant to oxidising and reducing gases at the operating temperature, have a reasonable thermal expansion match with the electrolyte/electrode wafer, be microstructurally stable over long periods at the fuel cell operating temperature, and be re-usable (for economic reasons) at the end of the fuel stack operating life. A number of metallic alloys have been considered for the role of interconnect for the fuel stack, these include: nickel-based alloys, high chromium alloys, iron-aluminium alloys and iron-chromium-aluminium alloys. After assessment of these alloys, an iron 20wt% chromium 5wt% aluminium alloy, with trace amounts of rare earth elements was deemed by Ceramic Fuel Cells Ltd to be the best potential as a candidate for the metallic components of the fuel cell. This alloy forms an aluminium oxide scale, resistant to both oxidising and reducing gases and has a reasonable thermal expansion match with the zirconia wafer over the operating temperature range of the fuel stack. Two forms of the alloy were examined, a 0.5mm thick sheet called Alfa IV and a 1mm sheet called NCA-S.
This thesis is concerned with the microstructural stability of the Alfa IV / NCA-S alloy as this was identified as being potentially detrimental to the economic and technical performance of the fuel cell stack. In particular, the low temperature (400-550ºC) decomposition of the alloy due to the miscibility gap in the iron chromium system was assessed as potentially damaging to the recyclability of the metallic components.
The decomposition of iron chromium alloys, due to the miscibility gap in the iron chromium system, has been the subject of considerable research. This is due, industrially, to the importance of these materials in numerous industries, such as the chemical and nuclear industries, and scientifically, as an ideal example of a miscibility gap in a binary metallic system. While the decomposition of iron chromium alloys at low temperature has been well described in high purity alloys, an effect, termed the high temperature effect, has been observed in alloys of commercial purity whereby the rate of decomposition at low temperatures is affected by prior high temperature treatment (> 550ºC). This effect is poorly understood.
An experimental program was conducted to assess the effects of high temperature exposure and cooling rate on the low temperature decomposition of the Alfa IV / NCA-S alloy. The decomposition reaction was primarily followed through the change in hardness of the material during aging at 475ºC. The high temperature effect was clearly evident in the material when solution treated and quenched from 1100ºC and above, with a substantial reduction in the rate of hardening observed during aging at 475ºC when compared to lower temperature solution treatments or to slow cooling from 1100ºC and above. The differing decomposition processes were confirmed in samples quenched and slow cooled from 1300ºC and aged at 475ºC using atom probe microscopy. Mechanical testing of the Alfa IV / NCA-S demonstrated that in addition to the effects of low temperature decomposition, grain growth at high temperature has a significant effect on the mechanical properties of the Alfa IV / NCA-S material.
Two theories have been proposed to account the high temperature effect. The first is that the chromium distribution at high temperature is inhomogeneous in iron chromium alloys and the second is that the distribution of interstitial elements, dissolved into solution at high temperature, effects the low temperature decomposition. The results of this thesis strongly support the second proposed theory. Thermodynamic modelling was used to demonstrate that the dissolution of chromium carbides above approximately 950ºC can account for all of the experimental results collected regarding the behaviour of the Alfa IV / NCA-S material after the various solution treatments and during ageing at 475ºC.
The results of this thesis demonstrate that the Alfa IV / NCA-S material is not suitable for use as a metallic component of a fuel cell. While the expected shutdown period of 24 hours will not provide sufficient time for decomposition to inhibit the recycling of the interconnect component, grain growth at the operating temperature of the fuel stack will result in significant embrittlement of the metallic components. It is recommended that the Alfa IV/ NCA-S material be substituted with an oxide dispersion strengthened material of the same matrix composition to avoid the detrimental effects of grain growth.