Mn 3x Zn x O 4 spinel phases in the Zn–Mn–O system M. Peiteado a, * , S. Sturm b , A.C. Caballero c , D. Makovec a a Department of Materials Synthesis, Jozef Stefan Institute, 1000 Ljubljana, Slovenia b Department for Nanostructured Materials, Jozef Stefan Institute, 1000 Ljubljana, Slovenia c Department of Electroceramics, Instituto de Cera ´ mica y Vidrio, CSIC, 28049 Madrid, Spain Received 24 December 2007; received in revised form 17 March 2008; accepted 15 April 2008 Available online 22 May 2008 Abstract The thermal evolution of the Zn–Mn–O system in air was studied by X-ray diffraction, transmission electron microscopy–energy dis- persive spectroscopy and electron energy loss spectroscopy. The obtained results suggest that this evolution involves the formation of different Mn 3x Zn x O 4 spinel-type phases. With increasing temperature these spinels experience phase transformations which are found to be induced by the Mn(IV) to Mn(III) reduction process. This last cation is an active Jahn–Teller ion which leads to an appreciable distortion of the Mn 3x Zn x O 4 spinel structure, from a cubic symmetry at low temperatures to highly distorted tetragonal symmetries at high temperatures. Ó 2008 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. Keywords: Manganese-doped ZnO; Binary oxides; Spinels; Phase transformations; EELS 1. Introduction In recent years there has been a growing interest in ZnO systems doped with manganese because of their potential for applications in diluted magnetic semiconductors for spintronic devices [1–4]. The manufacturability of the materials depends on the thermal equilibrium solubility of the dopant in the host ZnO; however, the specialized lit- erature of the past decade comprises a nearly 50–50 split in the papers evidencing symptomatic room-temperature fer- romagnetic behavior for these materials and the papers in which such a possibility is completely rejected. On the basis of this lack of reproducibility some authors claim that the inconsistency is somehow precipitated by an inadequate characterization of the materials, with the issues of struc- ture, composition and secondary-phase formation not being addressed with sufficient care [5]. In short this sug- gests a worrisome lack of knowledge about the Zn–Mn– O system itself, which can probably explain the origin of some of these misinterpretations. The first studies on the reactivity between ZnO and manganese oxides were carried out in the early 1960s by Tuissant [6], who reported the for- mation of ZnMn 2 O 4 and ZnMnO 3 compounds, the latter having a cubic symmetry. The first report on a possible phase diagram, ZnO–MnO, dates back to 1966 when Bates et al. [7] studied the solubility of different transition-metal oxides in zinc oxide over the temperature range 500– 800 °C. In this work the measurements were made using metallic manganese as a starting material to keep the man- ganese in a divalent state, and as a consequence no second- ary phases were detected. Simultaneously, but taking into account the different redox processes of manganese oxides, the phase equilibrium in the system Zn–Mn–O in air was analyzed by Driessens and Rieck [8]. Working with samples prepared by conventional ceramic methods, these authors suggest the existence of different unidentified spinel-like phases, mixed oxides of the formula Zn x Mn 3x O 4+y . How- ever, since these early works few efforts have been made to describe more clearly the evolution of this pseudo-binary system. Only a few experiments have been conducted in relation to the structure of the ZnMn 2 O 4 and ZnMnO 3 1359-6454/$34.00 Ó 2008 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.actamat.2008.04.024 * Corresponding author. Tel.: +34 91 7355840; fax: +34 91 7355843. E-mail address: peitead@icv.csic.es (M. Peiteado). www.elsevier.com/locate/actamat Available online at www.sciencedirect.com Acta Materialia 56 (2008) 4028–4035