I am not an expert in ceria but a lot of work in the field has been done in the field of zirconia. Li, Chen and Penner-Hahn "Effect of Dopants on Zirconia Stabilization-An X-ray Absorption I-III", J. Am. Ceram. Soc. Vol 77 , 1984) have investigated zirconia with different dopants. e.g. 3 vol-% Fe are stable as a tetragonal solid solution at 600°C while they de-mix forming hematite and monoclinic zirconia at 1300°C.

As the difference between the ionic radius is even larger in case of ceria with fluorite structure this de-mixing is not surprising.

Undersize dopants have lower solubility than oversized ones (e.g. Y3+) and they can only be maintained stable at low sintering temperatures. One reason maybe that most of the undesized dopants are located in the grain boundaries under different energetic conditions than in the bulk. At higher sintering temperatures the bulk volume expands at the expense of the grain boundaries so that the excess dopant is precipitated. Evidence for this can be found in: Ross et. al Scripta materialia 45, 2001.

alumina is found in the grain boundaries of 3Y-TZP only in the first few atomic layers from the grain boundary.

Thank you very much for your comments and sending the valuable references as the supporting evidence. They helped me a lot.

Frank Kern: I agree with you. The another possibility is the transition metals start sinters (crystallize) at lower temp where Ce, Zr (Big size) sinters at high temp. The mobility of transition metals also stats at low temp (ex. Vapour pressure is high at 1000C) rather than other rare earth. In my opinion Ionic radii is one factor, then phase change and vapour density at high temp is another imp factors. Plz say your opinion

We should not forget that if we consider stability we talk about a thermodynamic problem. Initially if the ceria and transition metal precursors are well distributed (e.g. by coprecipitation) we may result in a thermodynamically instable but kinetically stable composition which separates once thermal activation is sufficient and diffusion speed then get high enough. Considering this argument the question of mobility (vapor pressure, diffusion coefficients etc.) is important. From the thermodynamic point of view this does not matter, as it seems the mixture of Nickel oxide and ceria is instable at 1200°C, this is a question of lattice energies and ionic radii.

In XRD there is no indication of formation of a ternary compound or solid solution.

in the system CeO2-NiO, I found this in a thesis published in 2007 at University of Stuttgart by Dr. Nuri Solak : "Interface Stability in Solid Oxide Fuel Cells for Intermediate Temperature Applications." page 87 ff

Our university library has it under following link. However I am sure sure if someone from outside the campus has access.

http://elib.uni-stuttgart.de/opus/volltexte/2007/3104/pdf/solak.pdf

Komateedi Rao: sintering behavior is closely related to the chemical composition, particle and pore size of the green body. Please see this reference (Journal of Power Sources Volume 86, Issues 1–2, March 2000, Pages 383–389) where it is clearly shown that the initial sintering behavior of NiO and YSZ are very similar and starts sintering around 1000C. under the given conditions once a solid solution is observed in Ce-Ni system at lower calcination temperatures (600C) it is kinetically stable and with the increase of calcination temperature, Ce-Ni system becomes unstable in terms of both kinetic and thermodynamic. If the vapor pressure of transitional metal is a key for the mobility then a homogenously physically mixed NiO and CeO2 sample should have a non-homogeniously mixed NiO and CeO2 when the sample is heat treated at 1500C. But in solid oxide fuel cell anode preparation we always see almost initial state of homogeneity of NiO and CeO2 even if it is heat treated at 1500C. Zr is also a transition metal but it has different physical and chemical properties when compared to Ni, Cu, Fe, Co.

Thank you