First-Principles Studies of the Surface and Interface between/Among Metal Oxides Used in SOFCs

On the surface and interface, all materials change the property such as stability and electronic states from them on bulk. Human being continue to apply the phenomena to the key technologies from early times. For example, in solid oxide fuel cells (SOFCs), important phenomena such as fuel decomposition, charge transfer etc., occur at surface or triple phase boundary (TPB) that consists of catalyst, electrolyte, and gas phase. To solve remaining issues of SOFCs such as degradation or efficiency improvement, important is to know the detail of property changes of materials used in SOFCs on the surface and interface. Theoretical study is efficient approach to investigate material property, especially electronic states. Therefore, in this study, we have focused on surface and interface of metal oxides used as electrolytes in SOFCs and investigated theoretically their property using two calculation models with first-principles calculation. In concrete, one is density functional theory (DFT) calculations with the TPB model that consists of Ni catalyst cluster, YSZ electrolyte, and gas phase. In this case, we have tried to decide the reliable structure to discuss properties on TPB and evaluated adsorption reactivity etc. The other is DFT calculations with the hetero-structure interface model that consists of zirconia (ZrO 2 ) and ceria (CeO 2 ). In this case, we have tried to evaluated electronic states on that interface and discussed in terms of conductivity control. Here, only the results of optimized YSZ (111) surface model and density of state (DOS) change in ZrO 2 /CeO 2 hetero-structure are shown as examples for word limitation. The CASTEP software was used for all calculations. Geometry optimization of YSZ (111) surface model have been performed with some arrangements of yttrium (Y) atoms and oxygen vacancy in order to identify the most stable structure. Our model consists of (111) three layers includes 2×2 or 2×4 ZrO 2 unit cells, 15 Å vacuum layer, and 9.1 % substituted Y atoms. In the most stable YSZ 2×2 model, one Y atom is on the top Zr atoms layer and another Y atom is on the middle Zr atoms layer nearest to the top Y atom, and oxygen vacancy is on the 2nd O atoms layer nearest to the top Y atom. We have found the position of oxygen vacancy against Y atoms and surface is sensitive to the YSZ surface stability. Oxygen vacancy does not prefer to stick around two Y atoms. It indicates that improvement for the geometry unstability of cubic ZrO 2 is more important than keeping local electroneutrality even near on the surface. By applying this finding, we have performed geometry optimization of YSZ 2×4 unit cell model with changing Y atom arrangements in the same layer and keeping the stable structure in 2×2 unit cell model. The most stable YSZ 2×4 surface model finally-obtained is shown in attached figure (A). Our 2nd topic is for the electronic states change on the interface of metal oxides. We have calculated DOS with pure ZrO 2 , pure CeO 2 , and several ZrO 2 /CeO 2 models hetero-connected with (111) surfaces. DOS changes by hetero-connection are shown in attached figure (B). A new small peak not appeared in the cases of pure ZrO 2 and pure CeO 2 is observed in 01 eV near Fermi level. It is considered that the peak is caused by the interaction between 4f orbital shown in 13 eV of Ce atoms and 2p orbital shown in -50 eV of O atoms in ZrO 2 on the interface. In addition, the peak height is influenced only by the thickness of CeO 2 layers, not that of ZrO 2 layers. Our results suggest that we can control electronic conductivity of these metal oxides with hetero-structuration in atomic level especially at low temperature. Other results and detailed discussion will be reported at our meeting publications (ECS Transactions) and presentation in this conference. Figure 1