Nd₂NiO_4+δ with K₂NiF₄ type structure (T-phase) composed of rock salt layer and perovskite layer has attracted much interest as new cathode material for solid oxide fuel cell. It was reported that partial substitution of Cu for Ni-site in Nd₂NiO_4+δ improved ionic and hole conductivity [1]. However, crystal structure must change depending on Cu content, resulting in possible variation of electrical property, since crystal structure of Nd₂CuO₄ is T’-phase composed of fluorite layer and [CuO₂] sheet, which is completely different from T-phase. For optimization of Cu content in Nd₂Ni_1-xCu_xO_4+δ as cathode material, structural phase diagram between Nd₂NiO_4+δ and Nd₂CuO_4+δshould be established.

In this study, preparation of Nd₂Ni_1-xCu_xO_4+δ with various compositions was attempted and stability at high temperature was evaluated to establish structural phase diagram. It has been established that there exists miscibility gap in Nd₂Ni_1-xCu_xO_4+δ for 0.4≤x≤0.9 for the first time.

Nd₂Ni_1-xCu_xO_4+δ (0.0≤x≤1.0) were prepared by the Pechini method by which preparation of homogeneous specimens can be expected. Nd₂O₃, Ni(NO₃)₂·6H₂O and CuO powders were dissolved with mixture of dilute HNO₃ and H₂O₂, distilled H₂O and dilute HNO₃, respectively. They were mixed with nominal composition. After mixing ethylene glycol and citric acid, the mixture was heated at about 450 °C, resulting in precursor. The precursor obtained was calcined at 700 °C for 24 h in air, followed by pressing into pellet and heat-treatment at 1000 – 1200 °C in air for 10 h.

For identification of the crystal structure and evaluation of the lattice constants, X-ray diffraction measurement (XRD) was performed. Some XRD patterns were analyzed using the Rietveld analysis employing program RIETAN-FP [2]. Homogeneity of the specimens was evaluated by SEM-EDX analysis (JEOL: JCM-5700 equipped with JED-2300). Temperature dependence of lattice constants and molar volume and existence of the structural phase transition were evaluated using high-temperature X-ray diffraction measurements in the air.

XRD patterns at room temperature of Nd₂Ni_1-xCu_xO_4+δ prepared at 1200 °C could be indexed as single orthorhombic T-phase with space group of Fmmm (No. 69) and single tetragonal T-phase with I4/mmm (No.131) for 0.0≤x≤0.1 and 0.2≤x≤0.3, respectively. For diffraction patterns of 0.4≤x≤0.9, peaks assigned as T’-phase were observed in addition to ones indexed as T-phase. With increasing Cu content, intensity of the peaks indexed as T’-phase increased and ones assigned as T-phase decreased. All the peaks of XRD pattern for the specimen with x=1.0, i, e., Nd₂CuO_4+δ was successfully indexed as tetragonal T'-phase with space group of I4/mmm (No.131).

Rietveld analysis revealed that lattice constants and molar volume of both T-phase and T’-phase in the specimens with 0.4≤x≤0.9 were independent on Cu content, indicating existence of miscibility gap for 0.4≤x≤0.9. Also revealed by Rietveld analysis was linear relationship between Cu content and molar ratio of T’-phase which could be ascribed to satisfaction of lever rule in the miscibility gap. EDX image and spectra of Nd₂Ni_0.4Cu_0.6O_4+δ depicted in Fig. 1 indicate that two phases with Ni rich and Ni poor composition coexist in the specimen, showing agreement with proposed miscibility gap by Rietveld analysis.

High temperature XRD measurements revealed that the miscibility gap for 0.4≤x≤0.9 was maintained up to 1000 °C. For the specimens prepared at 1000 °C, the miscibility gap was observed for 0.3≤x≤0.9, suggesting expansion of the miscibility gap region by decreasing temperature.

References

[1] M. Yashima et al., J. Am. Chem. Soc.130 (2008) 2762.

[2] F. Izumi and K. Momma, Solid State Phenom., 130 (2007) 15.