IES20020538A2 - A process for producing soft magnetic composites - Google Patents

Abstract

A process for producing polycrystalline soft magnetic composites from alloys which consists of at least two elements. One is a readily oxidisable non-ferromagnetic element such as a rare earth element and the other is a ferromagnetic element such as iron. Under specified conditions the alloy can be internally oxidised by the inward diffusion of oxygen into the bulk to produce a mixture of oxide of the non-ferromagnetic element and the ferromagnetic element in a largely non-oxidised condition. The oxide provides electrical insulation and the ferromagnetic element provides the soft magnetic component. Similar effects can be achieved if the non-ferromagnetic element exhibits a high affinity for other gaseous agents such as nitrogen.

Description

A Process for producing Soft Magnetic Composites Inventors: I.R. Harris and J.M.D. Coey Abstract A process for producing polycrystalline soft magnetic composites from alloys which consists of at least two elements. One is a readily oxidisable non-ferromagnetic element such as a rare earth element and the other is a ferromagnetic element such as iron. Under specified conditions the alloy can be internally oxidised by the inward diffusion of oxygen into the bulk to produce a mixture of oxide of the nonferromagnetic element and the ferromagnetic element in a largely non-oxidised condition. The oxide provides electrical insulation and the ferromagnetic element provides the soft magnetic component. Similar effects can be achieved if the nonferromagnetic element exhibits a high affinity for other gaseous agents such as nitrogen.

I. Background to the invention There is a well-developed trend towards smaller and more energy efficient electrical devices. Essential components of electrical energy converters and circuits are inductors with a soft ferromagnetic core. Laminated stacks of sheets of silicon-iron about 300 pm thick insulated with a layer of electrical varnish are often used as cores for low-frequency applications at frequencies below several hundred hertz. At frequencies in the kilohertz range, thinner laminate may be used to reduce eddycurrent losses. For higher frequencies up to one hundred kilohertz or more it is customary to use soft ferrites such as nickel ferrite or zinc ferrite. Although these materials have relatively low permeabilities and a limited saturation induction (Bs) of about 0.3-0.4 tesla, because of the presence of the oxygen in their structure they have the advantage of being electrically insulating.

The development in the magnetic properties of ferrites has not kept pace with the development in electrical electronic circuitry. The ferrite core is now the bulkiest component of a typical device power converter. Further progress depends on the development of new materials which can operate effectively at frequencies above 100 KHz with acceptable losses and at greater flux densities than ferrites.

The phenomenon of internal oxidation is well established and published work [1] has shown that, in the case of Nd-Fe-B magnets, it is possible to produce internal oxidation of the Nd component both at the grain boundaries and within the Nd2Fei4B ferromagnetic matrix phase. This results in a nanoscale mixture of Nd2C>3 and Fe and the volume change is such that, unlike the reaction with hydrogen, the bulk magnet remains fully dense and there is no evidence of cracking.

The present invention is a process by which a new type of soft composite magnetic material can be produced by internal oxidation resulting from heating in oxygen or in other gases such as nitrogen. f ί j OPEN TO PUBLIC INSPECTION ’ UNDER SECTION 28 AND RULE 2c Ϊ JNLNo~..11.££-----II. Description The magnetic material of the invention is a composite where fine grains of the soft magnetic component are separated by insulating regions. The orientation of the magnetic component can be controlled by the processing conditions and by the application of a magnetic field. In the case of particulate material, the shape can be in the form of platelets or needles depending on the initial state of the alloy. The advantages of employing such a structure include the following: (a) The metallic soft magnet component can have a saturation polarization Bs, much greater than that of a soft ferrite. Bs of a soft ferrite is limited to no more than about 0.5 T whereas metallic soft ferromagnets are known to have Bs values in the range 0.55 - 2.4 T. These materials include ferrous alloys based on iron, cobalt and nickel. (b) The structure of aligned platelets/needles is advantageous because there is a low demagnetization factor and hence a high permeability for a field applied in any direction within the plane of the platelets or along the axis of the needles. This permits the realization of magnetic circuits from blocks of material with the flux path following particular directions. (c) The metallic regions may be smaller than the electrical laminations prepared by the methods such as hot rolling planar flow casting or melt extraction. This facilitates flux penetration at high frequencies with reduced eddy current losses. (d) The particles are insulating because the metallic regions are insulated by nonconducting regions. (e) The bonded material has further insulation provided by the resin. High packing densities with volume fractions greater than 0.6 can be achieved with the platelet particles. Preferred alignment can be achieved by pressing, rolling or extrusion possibly at elevated temperatures. (f) Fully dense materials which maximizes Bs can be produced in particular shapes by hot pressing or liquid phase sintering the metallic powder prior to gaseous treatment.

III. Examples Nd-Fe-B alloys provide an excellent example to illustrate the principles of the process of the invention. For instance, the alloy Ndi5Fe77Bg where the composition is expressed in atomic percents, is the basis of hard magnets based on the ferromagnetic phase NdiFe^B. This alloy consists of a Nd2Fe4B matrix phase with Nd-rich material at the grain boundaries. The bulk alloy can be readily reduced to a fine powder by the hydrogen decrepitation (HD) process and after subsequent jet milling the powder can be aligned in a magnetic field and then liquid phase sintered by melting the Nd-rich phase to form fully dense magnets with a grain size of 1-30 pm). The grain size of the bulk alloy (~ 200 pm) can be reduced to ~ 0.3 pm by the hydrogenation, IE 02 Os 38 disproportionation, desorption and recombination (HDDR) process and the Nd-rich phase can be redistributed around the fine grains by subsequent annealing at around 650°C. Fine grain powder particles can be produced by atomisation and ribbon-like material with a thickness of ~30 pm and a grain size of ~ 50 nm can be produced by melt spinning the bulk alloy and such materials are available commercially.

All the above materials can be internally oxidized by heating in air in the range 3001000°C. In this process there is a rapid inward diffusion of oxygen which results in the selective oxidation of the Nd, first at the grain boundaries and then within the NdiFenB matrix phase. This results in the production of nanoscale grains of NdiCh and free iron, without any cracking within the bulk material despite the very brittle nature of the matrix phase (see Fig.l). There is an increase in the permeability and in the electrical resistance and a fall in the coercivity.

Thus internal oxidation can be employed to fully oxidize the Nd component to produce a NdiCh/Fe composite for soft magnetic applications. Scrap Nd-Fe-B magnets are not recycled and hence could be a source of cheap raw material. The process would work even better with cheap cerium misch-metal material instead of Nd. The process can be carried out in a magnetic field to produce preferred alignment of the soft magnetic component. The powder material can be hot pressed or bonded.

The soft magnetic characteristics of all these materials can be modified by subsequent high temperature (600-1200°C) annealing in vacuum to prevent further oxidation. Such a treatment modifies the microstructure compared with that characteristic of the as-oxidised material.

Reference [1] D.S. Edgley, J.M. Le Breton, S Steyaert, F.M. Ahmed, I.R. Harris and J. Teillet, J. Magn. Magn. Materl, 173,29 (1997).

Claims (1)

1. 4. Claims (i) A process for producing a soft magnetic composite material by internal reaction (with gaseous agent) of a single or multiphase alloy composed at least of a reactive non-ferromagnetic element and a less reactive ferromagnetic element, where the reaction may optionally be conducted in a magnetic field. (ii) A R-Fe or R-Fe-X alloy where R represents a rare earth metal or a mixture of rare earths and X is another element or a mixture of elements such as phosphorus, silicon, carbon or boron. R is internally oxidised by heating in air at 300-1000°C for a period of time in excess of one second to produce a fine scale mixture of Fe and R 2 O 3 which exhibits electrical resistance in excess of 10 pQm, a high spontaneous polarisation in excess of 0.5 T, a low coercivity of less than 1000 A/m and a relative permeability in excess of 5. The soft magnetic characteristics may optionally be modified by high temperature (600-1200°C) annealing in vacuum. (iii) A fully dense soft magnetic composite produced by the internal oxidation of a fully dense cast alloy, a sintered or hot pressed body or a deposited film. The soft magnetic properties may optionally be modified by high temperature annealing in vacuum at a temperature of 600-1100°C. (iv) A soft magnetic resin-bonded composite produced from internally oxidized powder. The soft magnetic properties of the powder may optionally be modified by high temperature annealing in vacuum prior to bonding at a temperature of 600-1200°C. (v) Magnetic cores manufactured from the internally oxidised composite magnetic material of claims (i), (ii), (iii) or (iv).