Amorphous and nanocrystalline soft magnetic materials: from design to application

Advanced soft magnetic materials are required to match high-power density and switching frequencies made possible by advances in wide band-gap semiconductors. Magnetic materials capable of operating at higher operating frequencies have the potential to reduce the size of power converters significantly. Amorphous magnetic alloys lack long-range atomic order and consequently exhibit high electrical resistivity, no macroscopic magnetocrystalline anisotropy and no microstructural discontinuities, e.g., grain boundaries or precipitates, on which magnetic domain walls can be pinned. Consequently, they show excellent performance in DC and AC magnetic fields due to their low hysteresis and eddy current losses. Much work on this class of materials has been carried out, yet there are some critical issues in design, fabrication and optimisation that need to be addressed. First, one of the main challenges is producing amorphous alloys with a lower content of non-magnetic elements. The alloys comprising more abundant magnetic elements show low amorphisation capability, so non-magnetic metals have been added to these alloys to tackle this issue. However, adding these elements can significantly reduce the saturation magnetisation and increase the cost of alloy. To overcome these hurdles, kinetic, thermodynamic and topological parameters have been exploited to predict and tune the amorphisation capability of Co-Fe-B alloys. Based on this, seven alloys with very attractive magnetic properties have been designed and fabricated by melt spinning. Five out of seven alloys are amorphous, highlighting the efficiency of the design procedure, among which there is an alloy with ultra-low coercivity, 2.9 A/m. This substantiates the efficiency of the amorphous alloy design policy. Additionally, the crystallisation behaviour of alloys correlates significantly with their amorphisation ability. In this regard, the eutectic mode of crystallisation can be a sign of higher amorphisation ability, whereas alloys that crystallise through other mechanisms can offer lower amorphisation ability. Two in-situ crystallised alloys offer low power loss and high saturation magnetisation. The surface crystallised alloy exhibits not only very low coercivity but also surprisingly 20% lower power loss compared to the best amorphous sample. The mechanism behind this observation is investigated in more detail. The surface crystallisation can be used to induce anisotropy and decrease the power loss of melt-spun ribbons, eliminating the need for magnetic annealing. In addition, it has been shown that the amorphisation capability of an alloy can be tuned, based on the as-mentioned parameters in order to produce an in-situ nanocrystallised alloy with remarkable Bs = 1.57 T. The precipitation of the kinetically-favoured metastable Co7Fe3 nanocrystalline phase in the amorphous matrix is responsible for the high Bs of this alloy. The atomic structure of the amorphous matrix is evaluated based on Mössbauer spectroscopy, and the distribution of the hyperfine magnetic field implies that cobalt atoms form clusters and boron atoms undergo only short-range ordering. Therefore, designing alloy compositions using the as-mentioned predictive parameters can give one the opportunity to manipulate the microstructure of alloys to greatly benefit from their optimised properties. To further lower the cost of amorphous alloys, ultra-thin soft magnetic amorphous ribbons are fabricated via a single-step rapid-quenching process. The ribbons, with the thicknesses of 5.5 µm, fabricated by a high speed melt spinner offer ultra-soft magnetic properties. It is shown that in-situ thinning process results in a substantial reduction in the cost of the material. It also eliminates the need for post-processing steps, providing the opportunity for mass production of high-performance soft magnetic amorphous ribbons at a relatively low cost. Despite the low hysteresis and eddy current loss of amorphous metals, the measured losses, especially at high frequencies, are often in excess of these classical loss contributions. To gain a better understanding of this observation, the measured total power loss is resolved into hysteresis, eddy current, and anomalous losses. It is found that the main contributor to the power loss is the anomalous loss that can be reduced substantially through annealing in a transverse magnetic field at temperatures lower than the ribbons crystallisation temperature. It has been shown that transverse magnetic annealing not only alters the mechanism of magnetisation, from domain wall motion to magnetisation rotation, but it results in domain pinning due to the increased number of domains by decreasing their width. The latter promotes the magnetisation rotation mechanism. These effects account for a 75% decline in the total power loss in the melt-spun ribbons, making them desirable candidates for power converters working at mid- and high frequency. In addition, soft magnetic nanocomposites can be produced through annealing the amorphous ribbons, with a specific composition, at temperatures higher than crystallisation temperature. The nanocrystallisation of Co23B6, with network-like FCC structure, in the amorphous matrix, leads to substantial progress in DC and AC magnetic properties of ribbons. These metastable complex phases with 116 atoms per unit cell are not well understood. Therefore, it is essential to investigate the kinetics and thermodynamics of this kind of phase transformation in order to identify the substantial effects of the resulting phase on magnetic properties. This has been done using CALPHAD and nucleation theory. As a result of transverse magnetic annealing leading to the combined effects of nanocrystallisation and coherent magnetisation rotation, the anomalous loss declines by 70% in the ribbons annealed at 525 °C. Therefore, this soft magnetic nanocomposite can be utilised as the core of efficient power convertors