Considerations on the sintering of manganese-zinc ferrite for power applications

Abstract

Manganese-zinc ferrites for power applications, such as switched-mode power supplies and transformers for television and video-recorders, can be defined by high saturation magnetization, high relative amplitude permeability and low losses with a negative temperature coefficient in the operating temperature range in order to avoid overheating [1]. The most suitable manganese-zinc ferrites for power applications are obtained with chemical compositions for which the compensation point of the magnetocrystalline anisotropy or the second permeability maximum are in the working temperature range of the transformer core (60-120 °C) [2]. A high saturation magnetization can be acquired for a composition of higher iron content (> 50 mol%), but this has an adverse effect on the permeability and losses. The hysteresis loss has been found to increase with the grain diameter and with the impurity content of the raw materials. To avoid the resultant eddy-current losses, it has been found necessary to form thin layers of high resistivity which separate low-resistivity granules by the addition of CaO [31. The search for optimum sintering conditions of a ferrite for power applications is complex because of the large number of parameters and the multitude of objectives to be realized. This letter presents the influence of the sintering parameters process on the microstructural characteristics of MnZn ferrites for power applications. We considered an MnZn ferrite with chemical composition 53 mol % Fe203, 34.5 tool % MnO and 12.5 tool % ZnO. Samples were obtained by conventional ceramic technology. The raw materials (0~-Fe203, Mn30 4 and ZnO) were mixed in suitable proportions in a steel ball-mill using water as the mixing medium. The mixture was dried and presintered at 1050 °C for 1 h in air. The X-ray diffractograms of the presintered powder contained only diffraction maxima specific for spinel phase (Fig. 1). The presintered powder was milled again to obtain a specific area of 2.5 m 2 g-1 (Brunauer-Emmett-Teller method) and kept ready for the preparation of samples. For the