May_EDFA_Digital

edfas.org 31 ELECTRONIC DEV ICE FA I LURE ANALYSIS | VOLUME 23 NO . 2 condition. Therefore, the LSM phase in the oxygen-tested cell can still maintain a stable A-site deficient stoichiom- etry level during the cell testing, which stabilizes the La 2 O 3 phase, restrains the diffusion of the La and Zr ions, and prevents the formation of the LZO phase. It is also likely that the Mn 3 O 4 phase is present as a barrier layer at the LSM/YSZ interface, which could effectively limit cation inter-diffusion across the LSM/YSZ interface as shown in Fig. 8. Despite the Mn 3 O 4 phases that form along the LSM/YSZ interface, it is noted that they mainly exist in the areas slightly distant from the pores (see Fig. 7) and may therefore not affect the activity of the TPB sites. To date, despite numerous studies on the LZO forma- tion mechanisms, information and understanding of where the zirconate phase actually nucleates and forms have been rather limited. Most studies suggest that the formation of the LZO phase is La diffusion controlled. [8,25] The TEM results on the air-tested cell demonstrate that the Zr-rich phase should form and grow inside the LSM grains following the depletion of Mn. The only explanation for the reason that these Zr-rich phases can form inside the LSM phase is that the Zr ions diffuse from the YSZ into the LSM. Consequently, the zirconate phase formation is suggested to be primarily associated with Zr diffusion. This is in agreement withMitterdorfer andGauckler, [6] who concluded that the growth of LZO islands is controlled by the surface diffusion of Zr +4 and these LZO islands grow in a direction toward the LSM grain. Finally, it should be noted that all the major LSM-YSZ interactions are mainly observed in the vicinity of the cathode/electrolyte interface. This can be explained as follows: when the cathode is polarized under a current load, a gradient of the oxygen partial pressure is built up across the cathode. In the case of the air-tested cell, the oxygen partial pressure increases from approximately 2.5 Pa at the interface to 2.1x10 4 Pa away from the inter- face in the cathode. [5] The increase of oxygen partial pres- sure results in a significant increase of excessMn solubility in LSM as shown in Fig. 11. Consequently, the LSM phase away from the interface could still maintain its A-site deficient chemistry and therefore, the LZO formation is hindered. CONCLUSION The effect of the oxygen partial pressure on the cathode/electrolyte interfacedegradation in technological SOFCs after long-term durability testing was investigated through chemical and electronic structure analyses by employing various analytical TEM techniques. It has been revealed that the LSM-YSZ reaction mechanisms vary substantially with respect to the oxygen partial pres- sure at the cathode/electrolyte interface. Being tested at 750°Cwith air as the cathode gas, formation of nano-sized Zr-rich particles on the surface of LSM phase, which is in contact with YSZ, is observed. These Zr-rich particles are considered as precursors of the insulating LZO phase that often occur at the TPB active sites. These nano-sized zirconate particles would suppress reactions and conse- quently cause the cell to degrade. While tested in oxygen, formation of zirconate phase is largely suppressed as the LSM can maintain the A-site deficient stoichiometry and its stability. As a result, the cell tested in oxygen exhibits a much lower degradation rate in comparison to the cell tested in air. The experimental observations are further understood by theoretical calculations of the Mn solubility in the LSM andYSZ. It is the variations of theMn solubilitywith respect to the oxygen partial pressure and temperature that play important roles in affecting the Mn transportation, LSM stability, and the resulting LSM-YSZ reactions. It is elu- cidated that the changes in the interfacial structure and chemistry even on a nanoscale can have strong impact on the overall loss of cathode performance in SOFC. In view of this, analytical TEM techniques are further proven to be of powerful tools in studies of cell degradation. ACKNOWLEDGMENT This work was supported financially by Energinet.dk through the PSO project SOFC R&D II 2008-1-0065. REFERENCES 1. T. Iwata: J. Electrochem. Soc., 143, 1996, p. 1521-1525. 2. S.P. Jiang and W. Wang: Solid State Ionics, 176, 2005, p. 1185-1191. 3. R. Barfod, A. Hagen, S. Ramousse, and P.V. Hendriksen: Solid State Electrochemistry, Proc. 26th Risø International Symposium on Materials Science, S. Linderoth, A. Smith, N. Bonanos, A. Hagen, L. Mikkelsen, K. Kammer Hansen, D. Lybye, P. V. Hendriksen, F.W. Poulsen,M.Mogensen,W. G.Wangeds, RisøNational Laboratory, Roskilde, 2005, p. 121. 4. A. Hagen, Y.L. Liu, R. Barfod, and P.V. Hendriksen: J. Electrochem. Soc . , 155, 2008, B1047. 5. Y.L. Liu, et al.: Solid State Ionics, 179, 2009, p. 891-895. 6. A. Mitterdorfer, L.J. Gauckler: Solid State Ionics, 111, 1998, p. 185-218. 7. A. Chen, et al.: J. Am. Ceram. Soc., 91 (8), 2008, p. 2670-2675. 8. C. Brugnoni, U. Ducati, and M. Scagliotti: Solid State Ionics, 76, 1995, p. 177-182. 9. H. Taimatsu, K. Wada, H. Kaneko, and H. Yamamura: J. Am. Ceram. Soc., 75 (2), 1992, p. 401-405. 10. A. Hagen, et al.: Fuel Cells, 6, 2006, p. 146. 11. R.F. Egerton: “Electron Energy-Loss Spectroscopy in the Electron Microscope,” third edition, Springer, 2011. 12. J.-O. Andersson, et al.: “Thermo-Calc &DICTRA, Computational Tools for Materials Science,” Calphad, 26 (2), 2002, p. 273-312. 13. A.N. Grundy, B. Hallstedt, and L.J. Gauckler: “Assessment of the