Publication
Journal of Applied Physics
Paper

Preparation and crystal chemistry of divalent europium compounds

View publication

Abstract

The discovery that the europium chalcogenides are a particularly simple and important group of magnetic materials has resulted in considerable work being done on the synthesis of these materials. Several methods of preparing pure EuO have been investigated and the results are reported. The method which resulted in the best EuO is when an excess of europium metal is reacted with Eu 2O3 at 800°C and then distilled off at 1150° in a high vacuum. Chemical analyses, microstructures, and magnetization measurements were used to determine the stoichiometry of EuO as a function of temperature. The results indicated that the rock salt phase can accommodate about 4 mole% trivalent europium in solid solution at 1300°C. Very little (<1%) europium metal was found to go into the EuO phase, but by the addition of 5% to 7% the paramagnetic Curie temperature θ was increased from 73° to 79°K. A number of new divalent europium compounds have been prepared and their magnetic properties investigated. Ferromagnetism was found in the silicates Eu2SiO4 and Eu3SiO5, the aluminates Eu3Al2O6 and Eu5Al 2O8, and the phosphate Eu3(PO4) 2. Other divalent europium compounds in which the europium concentration is lower but which had simple well-known crystal structures were also investigated. These included EuTiO3, EuZrO3, and CsEuF3 with the perovskite structure, EuF2 with the fluorite, EuWO4 with the scheelite, Eu2CaWO6 and Eu2SrWO6 with the cryolite, and Eu2ZrO 4 with the K2NiF4 structures. None of these compounds were shown to be ferromagnetic despite the fact that the nearest-neighbor Eu-Eu distances were less than 4.5 Å, the distance determined from the chalcogenide series to be the critical one for a positive ferromagnetic Eu-Eu interaction. The absence of ferromagnetism in these compounds is discussed in relation to the fact that, except in CsEuF 3, the co-ordination number of the Eu+2 ion is not six as it is in the chalcogenides. The assumption that the ferromagnetic interaction takes place via overlap of 5d orbitals is considered and discussed in relation to their structures. © 1965 The American Institute of Physics.

Date

Publication

Journal of Applied Physics

Authors

Share