PROCESSUTVECKLING FÖR JÄRNBASERADE HÅRDMAGNETISKA PULVER UTAN SÄLLSYNTA JORDARTSMETALLER

Abstract: An experimental study was performed, aimed at finding a way to produce single crystalline powder of pure (Fe0.7Co0.3)2B with a particle size of 100-200nm. This was done with the ultimate goal of producing a hard magnetic ferrite powder for powder metallurgical production of permanent magnets. For this study, eight different compositions of iron, cobalt and boron were atomized using atomization by water and a gas-water combination atomization (GA/W). The atomization type was found to have a limited impact on the particle morphology and GA/W atomization also showed a slight reduction in the oxygen contents. X-Ray Diffraction (XRD) was used to characterize the phase composition in conjuncture with Simultaneous Thermal Analysis (STA) and dilatometry to find the temperature of any phase transitions. One of the alloys was also analyzed using XRD, STA and Light Optical Microscopy (LOM) at different stages of milling and heat treatment under different atmospheres so as to determine the best way to treat the powder for crystal growth. Representatives from all eight alloys was analyzed in a Vibrating Sample Magnetometer (VSM) to produce hysteresis loops and determine the magnetic characteristics. In these analyses the composition, heat treatment and milling was varied in an attempt to discover any dependencies. Two samples that were heat treated for this purpose, were also coated with SiO2 using tetra-ethyl-orto-silicate (TEOS) in an attempt to hinder sintering normally accompanied with long heat treatments. A SEM analysis indicated that the grain growth is a slow process but was unable to quantify a growth rate due to a large variance. The equivalent LOM analysis was inconclusive but suggested an estimation of the grain size. These analyses were also used to estimate the number of grains per particle but vary heavily between 10 and 1'000'000. All but two of the alloys were, with their stoichiometric composition, found to contain >80% Fe2B with the remainder being FeB. Trace amounts of other phases could exist but was not possible to resolve in the XRD analyses. The remaining two alloys, having a non-stoichiometric composition, were found to contain mostly (Fe,Co)3B2 (>50%) with the remainder being FeCo. Thus it's concluded that the stoichiometric composition is more beneficial to the project. Heat treatment of these materials was beneficial in the transformation of (Fe,Co)B into (Fe,Co)2B regardless if the atmosphere is N2 or H2. A phase transition between an (Fe,Co)2B type phase and another phase has been detected around 550OC using DSC. This is confirmed by phase diagram comparison, reference measurement of pure Fe in DSC and is consistent with DSC of etched samples. The best hard magnetic properties has been found to be in the least heat treated and least milled G336 sample. Heat treatments and milling both, consistently, showed a deleterious effect on the hard magnetic properties. The heat treatment having this effect remains unexplained but the milling introduced defect in the material that could have acted as nucleation points for magnetic sub-domains. The SiO2 coating worked well as a coating to prevent sintering but had a deleterious effect on the magnetic properties of the material. The reasons for this are discussed. Some of the alloys were found to contain a large fraction of particles containing bubbles, mainly from the water atomizing process. These have not been seen to have any effect on the magnetic properties.