Magnet production

Abstract

A process is provided for the production of rare earth magnets comprising the steps of exposing a rare earth alloy to hydrogen gas at an elevated temperature so as to effect hydrogenation and disproportionation of the alloy, mechanically processing the disproportionated alloy, and degassing the processed alloy so as to effect hydrogen desorption and recombination of the alloy. The process of the invention finds use in the production and shaping of rare earth magnets, and may be particularly applicable to the production of thin magnetic sheets.

Claims

1. A process for the production of rare earth magnets via a non-powder route, the process comprising the steps of: exposing a rare earth alloy to hydrogen gas at an elevated temperature so as to effect hydrogenation and disproportionation without decrepitation of the rare earth alloy; mechanically processing the disproportionated rare earth alloy; and degassing the mechanically processed rare earth alloy so as to effect hydrogen desorption and recombination of the mechanically processed rare earth alloy, wherein the rare earth alloy is a solid material and does not break apart into a powder.

2. The process according to claim 1, wherein the rare earth alloy is selected from NdFeB, SmCo.sub.5, Sm.sub.2(Co,Fe,Cu,Zr).sub.17 and SrFe.sub.12O.sub.19.

3. The process according to claim 1, wherein the rare earth alloy is NdFeB.

4. The process according to claim 1, wherein the pressure of the hydrogen gas is from 1 mbar to 20 bar.

5. The process according to claim 1, wherein the rare earth alloy is exposed to hydrogen gas for a period of time from 30 minutes to 48 hours.

6. The process according to claim 1, the rare earth alloy is constrained during the step of exposing the rare earth alloy to hydrogen gas.

7. The process according to claim 6, wherein the rare earth alloy is constrained within a constraining element selected from a mould, a tube, a sleeve, or a ring.

8. The process according to claim 1, further comprising the steps of casting a molten rare earth alloy into a mould and solidifying the molten rare earth alloy, prior to exposing the rare earth alloy to hydrogen gas.

9. The process according to claim 8, wherein the cast rare earth alloy remains within the mould during the step of exposing the rare earth alloy to hydrogen gas.

10. The process according to claim 6, wherein the elevated temperature is at least 400° C.

11. The process according to claim 1, wherein the elevated temperature is at least 600° C.

12. The process according to claim 1, wherein the elevated temperature is no more than 1000° C.

13. The process according to claim 1, wherein mechanically processing the disproportionated rare earth alloy comprises pressing, rolling, compacting, shaping, and/or extruding the disproportionated rare earth alloy.

14. The process according to claim 1, wherein mechanically processing the disproportionated rare earth alloy comprises forming the disproportionated rare earth alloy into sheets.

15. The process according to claim 1, wherein the processed disproportionated rare earth alloy is degassed at a pressure of no more than 100 mBar.

16. The process according to claim 1, wherein the processed disproportionated rare earth alloy is degassed at a temperature of from 600 to 700° C.

17. The process according to claim 1, wherein the process further comprises homogenising the disproportionated rare earth alloy by exposing the disproportionated rare earth alloy to hydrogen gas at a temperature of at least 900° C. for at least 6 hours.

18. The process according to claim 1, wherein the rare earth alloy is a cast ingot, solid sintered magnet, or strip cast flakes.

19. A process for the production of rare earth magnets via a non-powder route, the process comprising the steps of: exposing a rare earth alloy to hydrogen gas at an elevated temperature so as to effect hydrogenation and disproportionation of the rare earth alloy; mechanically processing the disproportionated rare earth alloy; degassing the mechanically processed rare earth alloy so as to effect hydrogen desorption and recombination of the mechanically processed rare earth alloy; and wherein the rare earth alloy is a non-powder bulk solid material and wherein the non-powder bulk solid material is selected from the group consisting of a cast ingot and a solid sintered magnet.

Description

(1) Embodiments of the invention will now be described by way of example only, with reference to the accompanying drawings in which:

(2) FIG. 1 shows a schematic flow diagram of a conventional manufacturing route for producing sintered NdFeB magnets;

(3) FIG. 2 shows a schematic flow diagram of a process for producing NdFeB magnets according to an embodiment of the present invention;

(4) FIG. 3 shows a schematic flow diagram of a process for producing NdFeB magnets according to another embodiment of the present invention;

(5) FIG. 4 shows a schematic flow diagram of a process for producing NdFeB magnets according to a further embodiment of the present invention;

(6) FIG. 5a shows a SEM micrograph of partially disproportionated material following exposure of a NdFeB type alloy to hydrogen gas under conventional hydrogenation and disproportionation conditions;

(7) FIG. 5b shows a SEM micrograph of partially disproportionated material following exposure of a constrained NdFeB type alloy to hydrogen gas under hydrogenation and disproportionation conditions according to an embodiment of the present invention;

(8) FIG. 6a shows a cylinder of hydrogen-treated NdFeB material;

(9) FIG. 6b shows a cylinder of hydrogen-treated NdFeB material after compression at 20 tonnes;

(10) FIG. 6c shows a cylinder of untreated NdFeB material after compression at 20 tonnes;

(11) FIG. 7 is a back-scattered SEM image of a region of a treated Nd.sub.12.2Fe.sub.81.3B.sub.5 alloy after disproportionation and compression;

(12) FIG. 8 is a back-scattered SEM image of a region of a treated Nd.sub.12.2Fe.sub.81.3B.sub.6.5 alloy after compression, where the compression axis is indicated by arrows;

(13) FIG. 9 is a back-scattered SEM image of a region of a treated Nd.sub.15Fe.sub.77B.sub.8 alloy after compression;

(14) FIG. 10a is a stress-strain curve of a treated Nd.sub.12.2Fe.sub.81.3B.sub.6.5 alloy compressed at a rate of 0.5 mm/min;

(15) FIG. 10b is a stress-strain curve of a treated Nd.sub.15Fe.sub.77B.sub.8 alloy and an untreated alloy compressed at a rate of 0.5 mm/min;

(16) FIG. 11a is a magnetic hysteresis loop for a treated Nd.sub.15Fe.sub.77B.sub.8 alloy after compression and recombination; and

(17) FIG. 11b is a magnetic hysteresis loop for a treated Nd.sub.15Fe.sub.77B.sub.8 alloy after recombination only.

COMPARATIVE EXAMPLE 1: CONVENTIONAL MANUFACTURING ROUTE

(18) FIG. 1 shows a schematic flow diagram of a conventional manufacturing route for producing fully dense sintered NdFeB magnets. The molten NdFeB type alloy may be cast, using standard casting procedures such as book moulding or strip casting. In book moulding, the molten alloy is poured into a suitable mould and cooled to form an ingot. Free iron (α-Fe) may form on the surface of the casting and which reduces the ease of processing of the ingot. Heat treatment of the alloy, for a period of up to 24 hours, may therefore be required to remove the free iron. Alternatively, in strip casting, the molten NdFeB type alloy is poured onto a cooled copper wheel and the NdFeB type alloy solidifies into flakes. Strip casting suppresses the formation of free iron since the free iron does not have time to form.

(19) The cast NdFeB type alloy is then reacted with hydrogen gas at room temperature to effect decrepitation of the alloy into a friable powder. Since the friable powder is air sensitive, the powder has to be stored and transported under an inert atmosphere (e.g. argon) and it is preferable to carry out all subsequent steps of the process in an inert atmosphere. The friable powder is then jet milled to reduce the size of the powder particles.

(20) The particles of the milled powder are then aligned in a magnetic field and subsequently pressed to provide a green compact. Green compacts produced in this way will typically have a density of approximately 69% of the theoretical density of the finished magnet.

(21) The pressed green compact is then sintered at a temperature of approximately 1000° C. The sintering process is required to further increase the density of the green compact and provide the fully dense NdFeB type magnet.

EXAMPLE 2: MANUFACTURING ROUTE USING HDDR PROCESS

(22) FIG. 2 shows a schematic flow diagram of a manufacturing route for producing fully dense NdFeB magnets according to an embodiment of the invention. A molten NdFeB type alloy is cast, using standard casting procedures, into a mould and solidified. The cast NdFeB type alloy is then cut into coarse blocks being exposed to pure hydrogen gas (1 bar) at a temperature of over 650° C. to effect hydrogenation and disproportionation of the alloy into NdH.sub.2, Fe.sub.2B and predominantly α-Fe.

(23) The disproportionated material is homogenised under hydrogen gas (1 bar) at ˜950° C. for up to 12 hours, such as 3-5 hours, to optimise the microstructure of the material.

(24) The material is then mechanically processed by, for example, hot pressing or cold compaction to form a green compact. The green compacts produced in this way will typically have a density of approximately 94% of the theoretical density of the finished magnet.

(25) In alternative embodiments, the disproportionated material could be extruded or hot rolled into thin sheets, followed by punching of the thin sheets to provide discrete pieces of material that will eventually form individual magnets.

(26) Following hot pressing, the processed disproportionated material is degassed under vacuum at a temperature of around 650° C. to effect hydrogen desorption and recombination of the NdFeB type alloy. The resulting magnet can then be placed into a device, such as a motor.

(27) With reference to FIGS. 3 and 4, a process in accordance with an embodiment of the invention can similarly be applied using recycled magnetic powder, melt spun or strip cast ribbon or flakes, solid sintered magnets, or powder obtained by the hydrogen decrepitation (HD) of a cast ingot. As with the cast alloy, these materials are first disproportionated by exposure to hydrogen at a temperature of over 650° C. Optionally, the disproportionated material is homogenised (FIG. 4). The disproportionated material is then compressed, for example by hot or cold pressing, to produce a compact, which is then shaped. The shaped material is then degassed under vacuum at a temperature of around 650° C.

(28) These processes results in the production of a fully dense aligned rare earth magnet without the need to produce an air-sensitive powder. Processes in according with the invention enable the production of rare earth magnets with a significant reduction in the number if process steps and materials wastage. The increased ductility of the intermediate disproportionated material allows the shaping of the alloy as desired.

EXAMPLE 3: DISPROPORTIONATION STUDIES

(29) The formation of the disproportionated constituents can be observed by carrying out SEM studies on the disproportionated material. FIG. 5a shows a SEM micrograph of a partially disproportionated material following exposure of a NdFeB type alloy to hydrogen gas at a temperature of 880° C., i.e. under conventional hydrogenation and disproportionation conditions. The grey regions are where very fine mixtures of NdH.sub.2, Fe.sub.2B and α-Fe have formed.

(30) FIG. 5b shows a SEM micrograph of a partially disproportionated material following exposure of a constrained NdFeB type alloy to hydrogen gas. A sample of NdFeB was placed within a copper sleeve at exposed to hydrogen gas (1 bar) at a temperature of 400° C. for 6 hours. The presence of the grey regions in the SEM image indicates that the initiation of the disproportionation reaction at the original grain boundaries has occurred at a much lower temperature than that anticipated from normal kinetic arguments. This may be a result of local increases in temperature due to the constrained nature of the NdFeB type alloy and to the associated exothermicity of the hydrogenation and disproportionation reactions.

EXAMPLE 4: DUCTILITY STUDIES

(31) The ductility of the solid bulk disproportionated material obtained from hydrogenation and disproportionation of NdFeB was assessed by measuring the density of green compacts obtained by pressing the disproportionated material.

(32) Powdered NdFeB was exposed to hydrogen at a rate of 10 mbar/min up to 1200 mbar, at a temperature of 875° C., and held for 1 hour to effect hydrogenation and disproportionation. SEM was used to determine that disproportionation was complete and that the NdFeB had fully converted to the constituents NdH.sub.2, Fe.sub.2B and α-Fe.

(33) A uniaxial compacting pressure of 10 tonnes was applied to a 1 cm diameter die set containing the disproportionated material to form a green compact. The green compact formed from the solid bulk disproportionated material was found to have a density of 6.95 g/cc, and held its shape. The theoretical density of the final magnets produced is calculated to be 7.5 g/cc. Thus, the solid bulk disproportionated material was compacted to approximately 94% densification.

(34) In contrast, upon pressing the brittle, friable Nd.sub.2Fe.sub.14BH.sub.3 powder obtained from hydrogen decrepitation of NdFeB, the green compact was found to have a density of 5.13 g/cc. Thus, the brittle, friable powder was compacted to approximately 69% densification.

(35) In a further experiment, solid cast NdFeB was exposed to hydrogen at a rate of 10 mbar/min up to 980 mbar at 800° C. and held at temperature and pressure for 2 hours to effect solid hydrogenation and disproportionation. Again SEM was used to determine that disproportionation was complete and density was measured to be 6.87 g/cc.

(36) A uniaxial compacting pressure of 20 tonnes was applied to a 2 cm diameter die set containing the solid disproportionated material. The compact formed from the solid disproportionated material was found to have a density of 7.26 g/cc and a height change from 0.41 cm to 0.13 cm. Thus, the solid disproportionated material was compacted to approximately 97% densification.

(37) The much higher density of the disproportionated material compared to the decrepitated material and the large change in height of the solid disproportionated material indicates that the disproportionated material has a significantly improved ductility.

EXAMPLE 5: DISPROPORTIONATION STUDIES

(38) In this study, cast material of compositions Nd.sub.12.2Fe.sub.81.3B.sub.6.5 and Nd.sub.15Fe.sub.78B.sub.7 were employed. The materials were cut either into cylinders of ˜9.5 mm diameter and ˜5 mm in height, or cubes of ˜5×5×5 mm, using spark erosion, since this technique limits the chance of oxidation which could influence the disproportionation reaction.

(39) Disproportionation Technique

(40) To achieve disproportionation, the samples were heated under vacuum 915° C., and hydrogen was introduced to a pressure of 1200 mbar for varying periods of time of up to 6 hours. This technique avoids the hydrogen decrepitation process which occurs at lower temperatures, thus producing a completely solid material rather than a powder, and allowing compression, stress-strain measurements to be undertaken. The conditions were also adjusted to avoid formation of the more reactive NdH.sub.2.7 component, by cooling rapidly to room temperature under vacuum then heating to 350° C. with a 30-minute hold to remove H.sub.2. After a period of time sufficient to achieve 100% disproportionation (approximately 5 hours), the material was then cooled in hydrogen (1200 mbar) in order to maintain the disproportionated state.

(41) Compression Trials

(42) In order to assess whether there had been any radical change in mechanical behaviour resulting from disproportionation, both treated and untreated samples were compressed in 10 mm diameter specac die sets with an Atlas T25 press capable of a load of up to 20 tonnes.

(43) Microscopy

(44) A Joel 6060 and Joel 7000 scanning electron microscopes were employed in backscattered mode using 20 kV accelerating voltage in order to examine the structure of the disproportionated material both before and after deformation, in an attempt to relate the mechanical behaviour to any changes in the microstructure.

(45) Magnetic Measurements

(46) A Lakeshore vibrating sample magnetometer (VSM), capable of up to 1.5 T, was used to measure the magnetic properties of the material before and after compression.

(47) Results and Discussion

(48) The initial trials were carried out on the alloy Nd.sub.12.2Fe.sub.81.3B.sub.6.5 and specimens of this alloy were subject to a rapid compression test both in the initial condition and after the solid hydrogen disproportionation treatment by the method described above. The samples were compressed in a die set up to a maximum load of 15915 tonnes/m.sup.2. This provided a rapid means of assessing any effect of the hydrogen treatment on the mechanical behaviour prior to more detailed stress/strain measurements.

(49) SEM Results

(50) SEM analysis of the Nd.sub.12.2Fe.sub.81.3B.sub.6.5 starting material revealed three phases in the material; several large dark areas, several light spots and a large grey area. Because the composition of the alloy was near that of stoichiometry and the 2/14/1 phase (large grey areas) area formed by a peritectic reaction, then some dendrites of free Fe were seen together (dark areas). A possible unseen phase of NdFe.sub.4B.sub.4 may also be present in the material.

(51) SEM analysis of the Nd.sub.15Fe.sub.77B.sub.8 starting material revealed that, unlike the Nd.sub.12.2Fe.sub.81.3B.sub.6.5 starting material, the material has no dark regions of Fe dendrites. Several larger areas of light Nd rich as well as a large area of the 2/14/1 phase were observed. Removing the Fe dendrites will considerably improve the magnetic properties of the recombined material.

(52) After hydrogen treatment of the Nd.sub.12.2Fe.sub.81.3B.sub.6.5 material, the large majority phase of 2/14/1 had transformed into a much finer disproportionated structure. The dark regions of Fe dendrites remained as they will not react with hydrogen but have a coarser disproportionated structure surrounding them. The small bright areas of Nd rich still remain after treatment. As well as this a new phase has appeared, confirmed by EDX to be NdFe.sub.4B.sub.4. Under the conditions employed in these experiments, there was no evidence of any reaction of this phase with hydrogen.

(53) The same hydrogen treatment was applied to the Nd.sub.15Fe.sub.77B.sub.8 material. The majority of the 2/14/1 material was transformed into the disproportionated phase, the lighter areas of Nd rich were still present and there was also a phase of NdFe.sub.4B.sub.4 material present along the Nd rich grain boundary which had become clearer after the formation of the disproportionated matrix.

(54) Initial Compression Trials

(55) Cylinders of NdFeB material were cut by a spark erosion technique to sizes of ˜9 mm diameter and varying heights from 4.1-5.4 mm (FIG. 6a). These samples were then compressed in a 20 mm diameter die set, in air, up to a load of 7 tonnes (˜1095 MPa), producing extensive cracking and disintegration of the untreated sample. In the disproportionated sample, only a minor change in height of 1.5% and no noticeable change in diameter was observed.

(56) The load was then increased to the maximum setting of 20 tonnes (˜3130 MPa). In the case of the treated samples, the compression dramatically changed the shape of the material which experienced a height change of up to 70%. The thin compacts could be handled without falling into a powder with little to no powder being left behind after the compression test (FIG. 6b). In contrast, untreated sample cracked and fell apart (FIG. 6c).

(57) These simple trials emphasise the dramatic change in mechanical behaviour after the hydrogen treatment with the untreated material exhibiting very little ductility. This dramatic change has been confirmed by the subsequent, more carefully controlled, compression trials.

(58) It can be surmised that the highly ordered NdFe.sub.4B.sub.4 will be of a similar brittle nature to that of Nd.sub.2Fe.sub.14B. This was confirmed by further SEM analysis of a region of a treated Nd.sub.12.2Fe.sub.81.3B.sub.6.5 alloy after compression, as shown in FIG. 7. A critical feature of this microstructure is that all of the cracking was confined to a phase which was identified by EDX (Energy Dispersive A-ray analysis) as NdFe.sub.4B.sub.4. The extensive ductility of this sample can therefore be ascribed completely to the behaviour of the disproportionated mixture.

(59) SEM analysis of a compressed sample revealed that where the disproportionated mixture had coarsened at the interface with the iron dendrites, it was possible to discern the elongated nature of the iron component such that the minor axis was perpendicular to the direction of compression (FIG. 8). This further confirmed the ductile nature of the disproportionated material.

(60) The density of the s-HD material was determined by weighing the sample in air and then in diethyl phthalate. The untreated cast material exhibited a density 7.548 gcm.sup.−3. After disproportionation the density of the material was measured to be 7.154 gcm.sup.−3, and once compressed by 20 tonnes this value was measured to be 7.067 gcm.sup.−3. The maximum possible density of stoichiometric disproportionated Nd.sub.2Fe.sub.14B is 7.18 gcm.sup.−3 The difference between this value and the value measured is due to the Fe dendrites and NdFeB.sub.4 phases present in the book mould material.

(61) FIG. 9 shows the hydrogen-treated Nd.sub.15Fe.sub.77B.sub.8 material after compression. Much like the stoichiometric material, the NdFe.sub.4B.sub.4 material has begun to fracture whilst the disproportionated structure remains completely intact.

(62) Mechanical Testing

(63) Cylinders (˜9 mm diameter and ˜5 mm height) of the disproportionated cast materials were compressed in order to ascertain the detailed stress-strain behaviour of the various samples. FIG. 10a shows the curves for the hydrogen treated Nd.sub.12.2Fe.sub.81.3B.sub.6.5 cast material.

(64) FIG. 10b shows the stress-strain curve for treated and untreated Nd.sub.15Fe.sub.77B.sub.8 material. The apparent yield point for the hydrogen treated material and the unreacted material is dramatically reduced from 983 MPa to 446 MPa—almost a 50% reduction of the original stress. There is also a marked reduction in the elastic region for the Neomax alloy. There is still a stress relief after this point and this ends with a rapid increase in stress at around 67% change in thickness. The remarkable feature of FIG. 10b is the overall reduction in thickness of some 75% and, of this, up to 65% can be achieved at a very low stress level.

(65) Recombination Process

(66) After the compression trials, some of the samples were recombined by heating under vacuum to 900° C. at a rate of 10° C./minute and then cooled rapidly to room temperature. This treatment produced a solid sample with no powder break off and this resulted in a slight rise in density to 7.278 gcm.sup.−3. This increase can be attributed to the transformation back to Nd.sub.2Fe.sub.14B. The formation of cavitation, as shown by SEM, will lower the overall density as will the extensive cracking of the NdFe4B4 phase. Another distinctive feature of the microstructure is the ragged interface with the Fe dendrites which is indicative of the partial homogenisation process.

(67) Magnetic Measurements

(68) FIG. 11a shows the magnetic hysteresis loop for a treated Nd.sub.15Fe.sub.77B.sub.8 sample which has been compressed and recombined. The z direction is the direction of compression and these results would suggest that the compression has had an effect on the alignment of the material producing an easy axis.

(69) In FIG. 11b the magnetic hysteresis loop for a recombined Nd.sub.15Fe.sub.77B.sub.8 sample is shown. This sample has undergone no compression and shows no signs of magnetic alignment. Instead one finds that there is actually a decrease in the magnetic coercivity of the sample.

CONCLUSION

(70) The present investigations have demonstrated very clearly that the normally extremely brittle NdFeB-based alloys can be converted to a ductile form by the application of the solid disproportionation process. The present studies have shown that the intimate mixture of predominantly Fe and NdH2 exhibits substantial ductility and any brittleness originates from the presence of the NdFe4B4 which is fractured extensively after the compression treatment. Preliminary magnetic data has been obtained on the recombined material under present conditions has shown that it is possible to introduce anisotropy in the material through compression.

(71) Thus embodiments of the process of the present invention may provide one or more of the following advantages: The ability to provide magnets in a desired shape (e.g. a thin sheet) without the loss of material as caused by current shaping techniques. The invention makes use of the surprising finding that disproportionated material has increased ductility by pressing, rolling, extruding or otherwise forming the rare earth alloy while it is in the disproportionated state, prior to recombination. Deformation give alignment of grains, especially in the z direction, and improved magnetic properties; The provision of a process for producing fully dense and aligned rare earth magnets which avoids the use of an air-sensitive powder, in contrast to the known process based on hydrogen decrepitation; The provision of a process for producing fully dense and aligned rare earth magnets which involves fewer steps than known processes. In particular, the finding that exposing a constrained alloy to hydrogen reduces the temperature required for hydrogenation and disproportionation means that certain embodiments of the invention have reduced energy requirements.