JP7287220B2 - Rare earth magnet and manufacturing method thereof - Google Patents

Description

The present invention relates to a rare earth magnet having main phase crystal grains of a ThMn ₁₂ -type compound containing yttrium (Y) and a method for producing the same.

In recent years, there has been a demand for the development of magnets with a reduced content of rare earth elements. As used herein, rare earth elements refer to at least one element selected from the group consisting of scandium (Sc), yttrium (Y), and lanthanides. Here, the lanthanoid is a general term for 15 elements from lanthanum to lutetium. RT ₁₂ (R is at least one rare earth element, T is Fe, Co or Ni) having a body-centered tetragonal ThMn ₁₂ type crystal structure is a ferromagnetic alloy containing a relatively small composition ratio of rare earth elements. Are known. Although RT ₁₂ has high magnetization, it suffers from thermal instability of the crystal structure.

Patent Document 1 discloses a rare earth permanent magnet in which a portion of Fe, which is a T element, is partially replaced with Ti, which is a structure stabilizing element, to improve thermal stability in exchange for high magnetization. there is

In Patent Document 2, by partially substituting the R element of the RFe ₁₂ -based compound with an element such as Zr or Hf, the amount of the structure stabilizing element Ti or the like substituting the transition metal element is reduced to increase the saturation magnetization. Rare earth permanent magnets are disclosed which are stabilized while preserving the _ThMn12 structure.

Further, Patent Document 3 discloses an R'--Fe--Co system ferromagnetic alloy in which Y or Gd is selected as part of the R element of RFe ₁₂ , and this R'--Fe--Co system ferromagnetic It is described that the alloy exhibits high magnetic properties due to having a ThMn ₁₂ -type crystal structure produced by an ultraquenching method.

Further, Patent Document 4 discloses a rare earth permanent magnet in which the stability of the _ThMn12 structure is improved by adding Ti or the like to Patent Document 3.

Further, Patent Document 5 describes that the addition of Cu generates a non-magnetic, low-melting-point 1-4 composition phase, which enables sintering and high coercive force.

Further, in Patent Document 6, Sm ₅ Fe _17- based phase, SmCo _5- based phase, Sm ₂ O _3- based phase, and Sm ₇ Cu _3- based phase, which are different from the ThMn ₁₂ type as secondary phases for the main phase ThMn ₁₂ It is described that high coercivity can be achieved by including at least one of the phases. In addition, Non-Patent Document 1 describes that a stable ThMn type ₁₂ crystal structure can be obtained by structure stabilizing elements (Ti, V, Nb, Ta, Mo, W).

JP-A-64-76703 JP-A-4-322406 JP 2015-156436 A JP 2018-125512 A Japanese Patent Application Laid-Open No. 2001-189206 Japanese Unexamined Patent Application Publication No. 2017-112300

K. H. J. Buschow, Rep. Prog. Phys. 54 (1991) 1123-1213.

The rare earth permanent magnet described in Patent Document 1 has improved thermal stability due to element substitution of Fe with Ti, but since the amount of Fe substituted with Ti is large, the magnetization is correspondingly reduced, and sufficient magnetic properties are obtained. I can't get it.

On the other hand, in the rare earth permanent magnet described in Patent Document 2, although the ThMn ₁₂ structure is stabilized by substituting the transition metal element with Ti or the like, the (Fe, Co, Ti) phase of the soft magnetic bcc and Laves phase are likely to form, and sufficient coercive force cannot be obtained.

In the R'--Fe--Co ferromagnetic alloy described in Patent Document 3, since the Fe element is not substituted with a structure stabilizing element, high magnetization, large magnetic anisotropy, and a high Curie temperature are obtained. Since it is a non-equilibrium phase, the main phase compound may decompose in a high temperature densification process such as sintering.

Unlike Patent Document 2, the rare earth magnet described in Patent Document 4 can suppress the generation of the soft magnetic phase, but the coercive force is not sufficient.

The rare earth magnet described in Patent Document 5 may not have high magnetic physical properties due to the large amount of Ti added.

In the rare earth magnet described in Patent Document 6, when the rare earth-rich secondary phase Sm ₇ Cu ₃ is used, there is concern that the equilibrium state will shift to a composition richer in rare earth than the main phase during heat treatment, and the main phase ratio will decrease. be done. In addition, the coercive force of the secondary phase determines the coercive force of the entire bulk, and it is difficult to say that the main phase ThMn ₁₂ itself contributes to the coercive force.

An embodiment of the present disclosure provides a rare earth magnet having main phase crystal grains of a ThMn ₁₂ -type compound, in which coercive force is improved while suppressing a decrease in magnetization.

A rare earth magnet of the present disclosure, in an exemplary embodiment, is a rare earth magnet having main phase grains of an Fe-based ThMn ₁₂ -type compound containing Sm and Y and a grain boundary phase containing Cu, wherein the main The phase grain has a ThMn _12- type core phase and a ThMn ₁₂ -type shell phase, and the Sm concentration in the shell phase is higher than the Sm concentration in the core phase.

In one embodiment, the main phase grains contain at least one M element selected from the group consisting of Ti, Si, Al, and Ga, and the concentration of the M element in the shell phase is higher than the concentration of the M element in

In one embodiment, Si is contained as the M element, and the Si concentration gradient at least within the core phase is represented by a position-dependent linear function.

In an exemplary embodiment of the method for producing a rare earth magnet of the present disclosure, in the composition formula R1 _1-x R2 _x (Fe _1-y Co _y ) _w Ti _z Cu _α , R1 has at least Y, and Gd and R2 has at least Sm and is at least one selected from the group consisting of La, Ce, Nd and Pr, and x, y, z and w are each 0. Bulk of alloy for rare earth magnet, wherein 5≤x≤1.0, 0≤y≤0.4, 8≤w≤12, 0.42≤z<0.70, 0.40≤α≤0.70 preparing a body; preparing a diffusion alloy having a composition formula R1 _1-χ R2 _χ Cu _ω -based diffusion alloy where χ>x and 2≦ω≦4; Main phase crystal of ThMn ₁₂ type compound in which the Sm concentration in the shell phase is higher than the Sm concentration in the core phase by heating in contact with the alloy and diffusing the component elements of the diffusion alloy into the bulk body. forming a rare earth magnet with grains.

In one embodiment, the diffusion alloy is represented by a composition formula R1 _1-χ R2 _χ (Cu _1-ν-τ X _ν Si _{τ Znσ} ) _ω , where X is at least one of Al or Ga, and χ >x, 0<ν≤0.3, 0<τ<0.5, 0<σ<0.2, 2≤ω≤4.

According to an embodiment of the present invention, it is possible to provide a rare earth magnet having main phase crystal grains of a ThMn ₁₂ -type compound, in which a decrease in magnetization is suppressed and coercive force is improved.

FIG. 10 is a diagram showing a tissue in Experimental Example 9 of the present disclosure; FIG. 11 is a diagram showing a tissue in Experimental Example 11 of the present disclosure; FIG. 10 is a diagram showing a composition line analysis result in Experimental Example 9 of the present disclosure; FIG. 11 is a diagram showing line analysis results of composition in Experimental Example 11 of the present disclosure; It is a figure which shows the nuclear density distribution of 3D, (100) plane, and (110) plane derived|led-out by MEM. FIG. 2 is a diagram showing the correspondence of each site during the structural change from the TbCu ₇ type to the ThMn ₁₂ type seen in the Y—Fe-based compound.

<Rare earth magnet>
The rare earth magnet of the present disclosure is a rare earth magnet having main phase grains of an Fe-based ThMn ₁₂ -type compound containing Sm and Y and a grain boundary phase containing Cu. Further, the main phase grains have a ThMn ₁₂ -type core phase and a ThMn ₁₂ -type shell phase, and the Sm concentration in the shell phase is higher than the Sm concentration in the core phase. That is, in the present disclosure, the shell phase refers to a phase having a higher Sm concentration than the central portion (core phase) of the main phase crystal grains. In addition, in the present disclosure, a plurality of main phase crystal grains as a whole is simply referred to as a "main phase", and a phase present at the grain boundary of the main phase crystal grains is generally referred to as a "grain boundary phase".

In an exemplary embodiment, the main phase grains of the rare earth magnet of the present disclosure contain at least one M element selected from the group consisting of Ti, Si, Al, and Ga, and the M element in the shell phase is higher than the concentration of M element in the core phase.

Sm is expected to increase the magnetic anisotropic energy, and M elements such as Ti and Si are expected to have the effect of increasing the magnetic anisotropic magnetic field and reducing the stress and strain associated with the decrease in magnetization. These elements entering the main phase contribute to an increase in coercive force. On the other hand, the introduction of Sm means a decrease in the amount of Y and impairs the structural stability, so it is necessary to increase the amount of M element to compensate for the destabilization of the structure. It significantly reduces the magnetization of the grains. That is, introduction of Sm and M elements creates a trade-off relationship between magnetization and coercive force. Therefore, by forming a shell phase in which Sm or both Sm and M elements are concentrated in the outer periphery of the main phase crystal grain, which is the starting point of magnetization reversal, the decrease in magnetization of the magnet as a whole is suppressed and the coercive force is increased. improvement is possible.

The main phase crystal grains preferably have a form in which the shell phase thinly covers the core phase. Especially when magnetization reversal is likely to occur from a specific crystal face, it is desirable to form a thin shell phase only on that crystal face. By forming the thickness of these shell phases to be equal to or greater than the thickness of the region serving as the starting point of magnetization reversal, the effect of sufficiently improving the coercive force can be obtained. The scale on which the stress/strain at the edge of the grain is relaxed varies depending on the crystal system.

The Sm concentration in the shell phase is higher than that in the core phase. Such a configuration is realized, for example, by diffusing a metal element from the outside into the bulk body of the rare earth magnet alloy. Naturally, this is also possible for samples that are not sufficiently densified, such as magnetic powder and temporary compacts. However, in order to obtain a rare earth magnet constructed from main phase crystal grains of thin shell phases, it is important to suppress the diffusion of elements into the core phase and efficiently form only the shell phase. For this purpose, it is sufficient that dissolution and reprecipitation occur at a temperature above which the grain boundary phase has sufficient fluidity. It becomes difficult to realize. Therefore, by preparing a bulk body such as a densified compact or sintered body of a rare earth magnet alloy and a diffusion alloy and diffusing the diffusion alloy along the grain boundaries in the bulk body of the rare earth magnet alloy, Rare earth magnets are preferably made. It is explained below. On the other hand, there is also a method of constructing a Sm-rich shell by utilizing the fact that Y contained in the main phase is very easily oxidized. In the heat treatment process for producing a bulk body such as a densified molded body or a sintered body, Y is removed from the main phase by oxidation at the outer periphery of the magnetic powder, thereby forming an Sm-rich shell. In this case, it is necessary to make the main phase a Y-rich composition in advance.

<Method for producing rare earth magnet>
Before explaining the method for producing a rare earth magnet, the compositions of the rare earth magnet alloy and the diffusion alloy will be explained.

[Composition of alloy for rare earth magnet]
The rare earth magnet alloy of the present disclosure is represented by the composition formula RT _w Ti _z Cu _α in a non-limiting exemplary embodiment.

In the rare earth magnet alloy having the composition formula RT _w Ti _z Cu _α , R is at least one selected from the group consisting of scandium (Sc), yttrium (Y), and lanthanides, and T is Fe or Fe and Co. The ranges of the composition ratios w, z, and α are respectively 8≦w≦12, 0.42≦z<0.70, and 0.40≦α≦0.70. Hereinafter, this alloy may be referred to as "R--Fe--Co--Ti--Cu rare earth magnet alloy". A structure obtained by densifying the powder of this alloy is sometimes called a "bulk body".

More specifically, the R—Fe—Co—Ti—Cu based rare earth magnet alloy in this embodiment is described by the composition formula R1 _1-x R2 _x (Fe _1-y Co _y ) _w Ti _z Cu _α . R1 has at least Y and may also have Gd. R2 has at least Sm and may contain at least one selected from the group consisting of La, Ce, Nd and Pr excluding Gd. x, y, z, and w are respectively 0.5≦x≦1.0, 0≦y≦0.4, 8≦w≦12, 0.42≦z<0.70, 0.40≦α It is a composition ratio that satisfies ≦0.70.

The alloy in such an embodiment contains Sm as R2, and further contains Sm from the outside by grain boundary diffusion. Sm can improve the magnetic anisotropy that is important for increasing the coercive force.

Addition of Cu produces a grain boundary phase that coexists with the main phase and is richer in rare earth elements than the main phase. The generation of the grain boundary phase rich in rare earth elements provides the following effects that are effective in obtaining anisotropic magnetic powder.
・Crystal growth of the main phase is facilitated by heat treatment.
・The generation of heterogeneous phases during melting and solidification can be reduced.
・Since the rare earth-rich grain boundary phase absorbs and releases hydrogen, cracks are generated between the main phase and the grain boundary phase, which can be efficiently pulverized into single crystal grains.

The composition ratio of R1, R2 and Ti determines the magnetic property value and high temperature stability of the main phase. From the viewpoint of magnetic anisotropy, it is desirable that the composition ratio (x) of R2 is greater than or equal to the composition ratio (1-x) of R1. Therefore, it is desirable that 0.5≦x≦1.0. From the viewpoint of saturation magnetization, Ti should preferably be as small as possible, but from the viewpoint of high-temperature stability, it should be large. As a result of studies by the inventors, the range of 0.42≦z<0.70 is desirable. Note that 50 mol % or less of Ti may be replaced with V, W, Nb, Mo, or the like.

Part of Fe is preferably replaced with Co from the viewpoint of improving magnetization and magnetic anisotropy at practical temperatures associated with increasing the magnetic moment and Curie temperature. However, if the amount of substitution is too large, the magnetization and magnetic anisotropy will rather be lowered, which is not desirable. Specifically, the Co substitution amount y is desirably 0≤y≤0.4, and more desirably 0.1≤y≤0.3. Magnetic properties are maximized when y is about 0.2. Since grain boundary diffusion tends to increase the Co concentration in the shell phase slightly, by designing the composition of the rare earth magnet alloy of the present invention so that y < 0.2, the function of Co can be efficiently performed in the shell phase. It is possible to express

The amount of Cu determines the amount of grain boundary phase that forms in the rare earth magnet alloy. If the amount of grain boundary phase is small, bcc (Fe, Co, Ti) or Nd ₃ (Fe, Ti) ₂₉ -type crystal structure phase (hereinafter referred to as 3-29 phase) or Th ₂ Zn ₁₇ during melting and solidification Not only can heterogeneous phases such as type, Th ₂ Ni ₁₇ type, and a phase with a crystal structure having a partial irregular substitution site inside it (hereinafter referred to as 2-17 phase) not disappear, an anisotropic sintered magnetic powder can be obtained. It is not easy to grow the crystal to a size large enough for Grain boundary diffusion into the inside of the magnet is also not easy. Also, when the amount of the grain boundary phase is large, the ratio of the main phase is lowered, so that the magnetization of the magnet body is lowered. As a result of studies by the present inventors, the range of 0.40≦α≦0.70 is desirable.

The resulting grain boundary phase is Cu-based. This grain boundary phase is mainly a KHg ₂ type crystal structure phase (hereinafter referred to as 1-2 phase), and also has a composition in which the ratio of R to Cu, Fe, Co is 1:4 (hereinafter referred to as 1-4 composition). Regarding the R element constituting the grain boundary phase, both the 1-2 phase and the 1-4 composition phase have a higher molar ratio of R2/(R1+R2) than that of the main phase. In addition, Fe and Co are dissolved slightly, and the solid solution amount is larger in the 1-4 composition phase than in the 1-2 phase. Ti hardly dissolves in both phases.

The appropriate amount of w varies depending on the amount of Cu added, but is 8≤w≤12. This is because a soft magnetic bcc (Fe, Co, Ti) phase is generated when w is too large, and a 2-17 phase or a 3-29 phase is generated when w is too small. Especially when constructing a Sm-rich shell by Y oxidation, w is preferably a relatively low value, and 8≤w≤10 is appropriate.

The R--Fe--Co--Ti--Cu based rare earth magnet alloy thus obtained contains a main phase having a ThMn _12- type crystal structure. The TnMn ₁₂ -type compound phase in the alloy in the embodiment of the present invention can typically exist stably even at 1000° C. or higher, and is suitable for adopting high-performance magnet fabrication processes such as sintering. be able to. In general, the “ThMn type ₁₂ crystal structure” is a tetragonal crystal, but in the embodiment of the present invention, the crystal lattice of the tetragonal crystal is slightly distorted to have orthorhombic symmetry, or Even if the atomic periodicity is slightly disturbed, it is regarded as a "ThMn type ₁₂ crystal structure". Moreover, when the Ti element concentration is less than 7.7 at % regardless of the presence or absence of Cu, the crystal may include a portion in which the rare earth element and two Fe elements (Fe dumbbells) are irregularly substituted. Specifically, the axial direction of the Fe dumbbell is parallel to the c-axis, and the random substitution rate is 5% at maximum. This also applies to structure stabilizing elements other than the Ti element, and by using Y, a ThMn type ₁₂ crystal stabilized with a stabilizing element in an amount smaller than the substitution composition range of the stabilizing element described in Non-Patent Document 1 It is commonly observed in structures.

Zn or Sn may be further added to the rare earth magnet alloy. It brings about a lower melting point of the grain boundary phase and the formation of a uniform two-particle grain boundary, making it possible to sufficiently supply the diffused alloy to the inside of the bulk body by grain boundary diffusion. There is also an effect that can result from a decrease in the amount of Cu solid solution in the main phase grains.

[Composition of diffusion alloy]
As a result of diligent research by the present inventors, it is preferable that the composition of the diffusion alloy according to the embodiment of the present invention is richer in Sm than the composition of the grain boundary phase of the rare earth magnet alloy. By introducing the diffusion alloy along the dissolved grain boundary phase, the 1-12 phase, which is in equilibrium with the Sm-rich grain boundary phase, is constructed at the outer periphery of the main phase by dissolution and reprecipitation. Therefore, when the diffusion alloy composition is expressed as an R1 _1-χ R2 _χ Cu _ω system diffusion alloy, it is desired that χ>x. ω satisfies 2≦ω≦4.

Furthermore, by including at least Al or Ga in the diffusion alloy, it becomes possible to construct a Sm- and Ti-rich shell phase deep in the bulk body. More specifically, Al and Ga diffuse along the grain boundary phase, but replace part of the 3d element of the main phase and diffuse into the grains. , Co, Ti, and Cu are ejected into the grain boundary phase. In particular, since Al and Ga are selectively coordinated to the 8i site of the main phase (see Non-Patent Document 1), Ti, which is strongly oriented to the 8i site, also remains and concentrates in the main phase, and Fe is ejected. In other words, the amount of Al or Ga has a correlation with the amount of Fe element discharged into the grain boundary phase, and it is appropriate that the amount of Fe element can be absorbed in the grain boundary phase. Furthermore, increasing the amount of Al or Ga tends to increase the thickness of the generated shell phase. From the above, when the diffusion source composition is represented by R1 _1-χ R2 _χ (Cu _1-ν X _ν ) _ω (X=Al, Ga), the range of 0<ν≦0.3 is suitable. More preferably, 0.1≤ν≤0.2. In addition, depending on the presence or absence of Si or Zn, the composition of ω may be different between Al addition and Ga addition. <2.8 may be more desirable.

In addition, further increasing the concentration of the M element in the shell phase not only further enhances the thermal stability of the _ThMn12 type, but also enhances the magnetic anisotropic magnetic field, thereby improving the coercive force. is important. According to Non-Patent Document 1, V can replace up to 4/12 of the T element, and Si can replace up to 2/12 of the T element. However, when the grain boundary is narrow, V tends to form a compound with Fe element, and it is difficult to introduce a sufficient amount deep into the bulk body by grain boundary diffusion. This also applies to Ti, Cr, Nb, Mo, W, etc. On the other hand, it was found that a sufficient amount of Si can be introduced deep into the bulk body along grain boundaries. Si can be suitably used to construct a shell in which the M element is more concentrated in the deep part of the bulk body. If the amount of Si introduced is too large, the amount of the R--Cu--Si-based compound produced becomes too large, hindering sufficient diffusion deep into the bulk body. When the diffusion alloy composition is represented by R1 _1-χ R2 _χ (Cu _1-ν-τ X _ν Si _τ ) _ω , the range of 0<τ<0.5 is appropriate. In addition, Si is characterized by having a higher affinity with Cu than Fe and by having a smaller atomic diameter. A diffusion process may preferably be used.

Therefore, by combining the above three, the main phase crystal grains have a ThMn12 type core phase and a ThMn12 type shell phase, and the shell phase richer in Sm, Ti, and Si than the core phase is constructed. . In that case, the diffusion source composition is R1 _1-χ R2 _χ (Cu _1-ν-τ X _ν Si _τ ) _ω 0.5, 2≤ω≤4) is desirable. Furthermore, the addition of Zn to the diffusion source composition improves the wettability between the main phase and the grain boundary phase, and makes it possible to introduce a larger amount of the metal element contained in the diffusion source deep into the bulk body. desirable. Zn tends to be unevenly distributed in the grain boundary phase and hardly diffuses into the main phase. In this case, the diffusion source composition is R1 _1-χ R2 _χ (Cu _1-ν-τ X _ν Si _{τ Znσ} ) _ω (X=Al, Ga, χ>x, 0<ν≦0.3, 0<τ<0.5,0<σ<0.2, 2≦ω≦4) is desirable.

<Step A> Step of Preparing Bulk Body of Rare Earth Magnet Alloy A bulk body of a rare earth magnet alloy (R—Fe—Co—Ti—Cu based rare earth magnet alloy) having the composition described above is prepared. Known techniques such as die casting, centrifugal casting, strip casting, and liquid ultra-quenching can be employed as methods for producing the R--Fe--Co--Ti--Cu based rare earth magnet alloy. In order to minimize the formation of phases such as the bcc (Fe, Co, Ti) phase, which are particularly unfavorable as raw material alloys for magnets, during the solidification of the molten alloy, the strip casting method or liquid ultra-quenching method, which has a relatively high cooling rate, is used. It is appropriate to adopt If the cooling rate during solidification is slow, the grain size of the precipitated heterogeneous phase becomes large, making it difficult to eliminate the heterophase in the subsequent heat treatment step. Even nanocrystals produced by the liquid ultra-quenching method can easily grow into crystal grains of 10 μm or more suitable for obtaining anisotropic magnetic powder by going through the following heat treatment process. It is better to solidify at speed. It is preferable that the solidified structure has little macrosegregation and is as uniform as possible. For example, when it is produced by a liquid ultraquenching method, a roll peripheral speed of 1 to 40 m/s is preferable in an inert atmosphere.

By applying heat treatment to the rare earth magnet alloy in the embodiment of the present invention, the heterogeneous phase generated in the solidification process is reduced, and the powder composed of single crystal-like particles useful as a raw material for anisotropic sintered magnets is pulverized. The crystal grains can be coarsened for easy obtaining by the method. Although it varies depending on the composition, the eutectic temperature between the 1-2 phase and the 1-4 composition phase is around 820°C, the melting point of the 1-2 phase is around 860°C, and the melting point of the 1-4 phase is around 880°C. be. Therefore, the heat treatment temperature is preferably 900° C. or higher and 1250° C. or lower, more preferably 1000° C. or higher and 1100° C. or lower. The heat treatment time depends on the temperature, but is preferably 5 minutes or more and 50 hours or less. If the time is too short, sufficient reaction to eliminate the heterogeneous phase will not occur or grain growth will be insufficient, and if the time is too long, the rare earth element will evaporate or oxidize, and operational efficiency will be poor. In this temperature range, the grain boundary phase becomes a liquid phase and part of the main phase dissolves and reprecipitates. When the grain size is not large, the heterogeneous phase can also be easily eliminated. The amount of the grain boundary phase is desirably 3 wt % or more and 10 wt % or less. If the amount of the grain boundary phase is small, not only will the heterogeneous phase not disappear during melting and solidification, but it will also be difficult to grow the crystals to a size sufficient to obtain an anisotropic sintered magnetic powder. This is because, if there is a large amount of , the ratio of the main phase is lowered, so that the magnetization of the magnet body is lowered. Incidentally, the (Fe, Co, Ti) phase of bcc, which was not completely eliminated in this step, can be eliminated in the step B if the amount is small.

Reflecting the rare earth-rich composition, the grain boundary phase contained in the present alloy causes hydrogen absorption and desorption phenomena, which occur remarkably particularly by heat treatment in hydrogen. For example, hydrogen absorption occurs at temperatures between 250°C and 400°C, and hydrogen release occurs between 540°C and 660°C. Therefore, after heating the rare earth magnet alloy to 400° C. or higher in hydrogen to absorb hydrogen, the atmosphere is switched to a vacuum atmosphere to sufficiently release hydrogen. In that case, the temperature for switching to the vacuum atmosphere is 700° C. or less. By absorbing and desorbing hydrogen, the rare earth-rich phase undergoes volumetric expansion and contraction, and cracks occur between the main phase grains and the subphase. This increases the probability of the magnetic powder cracking at the cracks during the pulverization process using a jet mill, stamp mill, ball mill, etc., and makes it possible to obtain highly oriented anisotropic magnetic powder containing a large amount of single-crystal unit fine powder. Become.

The anisotropic magnetic powder thus produced is compacted in a magnetic field and heat-treated at 900° C. to 1200° C. for 5 minutes to 50 hours to obtain a densified bulk body. At that time, a method of putting Sm-based alloy fine powder into an ampoule at the same time is known in order to suppress the evaporation of Sm, and it may be employed. Also, since Y is easily oxidized, a metal or alloy fine powder that is more easily oxidized may be added as an oxygen getter at the same time. In order to build a Sm-rich shell by positively utilizing the fact that Y is easily oxidized, the composition of the entire alloy may be designed to be on the Y-rich side. Alternatively, hot forming or the like may be used to obtain a densified bulk body. Embodiments of the present invention are preferably used in a densified bulk body composed of a main phase and a grain boundary phase, and the effect is irrespective of whether or not the magnetic particles are oriented. Although the effect of improving the coercive force is found even in magnetic powder that is not densified, it is not suitable because it is incompatible with volume magnetization as a bulk body. In that sense, it is more appropriate for the bulk body to be oriented. In addition, it may contain a small amount of oxide phase or a phase derived from unavoidable impurities.

If there is foreign matter on the surface of the densified bulk body, grain boundary diffusion in step C may be hindered. Therefore, by grinding the surface of the bulk body before the grain boundary diffusion heat treatment, a sufficient effect of improving the coercive force can be obtained.

<Step B> Step of Preparing Diffusion Alloy A diffusion alloy having the composition described above is prepared. The diffusion alloy may be amorphous or composed of a plurality of phases, but it is preferable that it is not extremely segregated. If there is extreme segregation, diffusion into the bulk body may be inhibited. Therefore, it is appropriate to adopt the strip casting method, the liquid ultra-quenching method, the gas atomizing method, the disk atomizing method, or the arc melting method, which have a relatively high cooling rate. Alternatively, a homogeneous alloy may be prepared using a ball mill, planetary ball mill, or the like. In addition, a nanoparticle synthesis method using a chemical liquid phase method such as a chemical reaction precipitation method, a reverse micelle method, a hydrothermal synthesis method, or a sol-gel method may be used.

<Step C> A step of forming a rare earth magnet having main phase crystal grains of a ThMn ₁₂ -type compound in which the Sm concentration in the shell phase is higher than that in the core phase. A heat treatment is applied for introduction into the bulk body. Since the diffusion alloy has good wettability with the main phase, if a bulk body with a thickness of about 1 mm is placed in a state in which at least a sufficient amount is in contact with the bulk body, the bulk body can spread along the grain boundary phase. It can be sufficiently diffused inside. If the bulk body has a thickness exceeding 1 mm, a diffusion source is laid on the bottom surface of the bulk body. Furthermore, if it is intended to increase the coercive force limited to the edge of the bulk body, the location where the constituent elements of the diffusion alloy are diffused will have a high coercive force, so the diffusion alloy should be arranged accordingly or the mask of the bulk body should be used. Processing or the like is required and may be performed as appropriate.

Suitably, the diffusion alloy is placed between 1 wt% and 80 wt% of the bulk weight. If the amount of the diffusion alloy is too small, the effect of improving the coercive force cannot be obtained sufficiently. It is not a good trend, such as being untargetable.

The heat treatment temperature is preferably 500° C. or higher and 1000° C. or lower. More desirably, the temperature is 700°C or higher and 1000°C or lower. If the temperature is too low, the diffusion alloy and the grain boundary phase will not dissolve, and a sufficient amount of the diffusion alloy cannot be introduced into the bulk body. As a result, the decrease in magnetization of the entire bulk becomes remarkable. Depending on the thickness of the bulk body, the heat treatment time can be set between 5 minutes and 50 hours. If it is too short, sufficient diffusion to the depth of the bulk body will not occur, and if it is too long, the coercive force improving effect will tend to saturate, resulting in poor operational efficiency.

Further, heat treatment again at a temperature lower than the diffusion heat treatment temperature may be effective because the coercive force is improved.

<Step D> Step of Grinding Bulk Body Surface After diffusion, the bulk body may be ground by a known processing method. The surface of the bulk body that has undergone the diffusion heat treatment has a residue of the diffusion alloy, and there are many phases such as the 2-17 phase and the 3-29 phase that are undesirable in terms of magnetic properties at the grain boundaries near the diffusion alloy. By grinding the outer peripheral portion of the bulk body by about 10 μm to 500 μm, such a low magnetic property region can be removed, and the magnetic properties of the bulk as a whole can be improved. Grinding powder can be reused by subjecting it to an appropriate treatment, so it does not greatly impair economy. A plating process or the like may be added after this depending on the intended use, and the present invention is not limited by the subsequent processes.

Experimental examples of the present invention will be specifically described below, but the present invention is not limited to these experimental examples.

[Experimental example]
<Step A> (Step of preparing bulk body of alloy for rare earth magnet)
Raw material metals of Y, Sm, Fe, Co, Ti, and Cu having a purity of 99.9% or more were weighed in increments, taking into consideration the evaporation of rare earth elements during melting. After sufficiently melting in the alumina crucible, the molten metal was tapped on a Cu roll rotating at a peripheral speed of 15 m/s. The super-quenched ribbon thus produced was wrapped in Nb foil and heat-treated at 1050° C. for 1 hour in an Ar stream. A magnetic powder having an average particle size of 10 μm or less was obtained by pulverizing with a jet mill. A pressed body molded in a 1 T magnetic field was placed in a furnace and subjected to heat treatment at 1150° C. for 4 hours. A bulk body of alloy for rare earth magnet with a thickness of 4 mm was produced by cutting the bulk body and subjecting it to surface polishing. SEM was used to assess material texture. _The composition _{of the main} phase was _Y0.37Sm0.63 ( _{Fe0.83Co0.17 ) 11.08Al0.07Si0.02Ti0.57Cu0.26 .} Al and Si are derived from the crucible used in this process.

<Step B> (Step of preparing diffusion alloy)
The diffusion alloy was produced using a super quenching device (NEV-A30023 manufactured by Nisshin Giken Co., Ltd.). A raw material with a purity of 99.9% or more weighed in consideration of the evaporation of rare earth elements is sufficiently melted in a quartz tapping tube, and then the molten metal is tapped on a Cu roll rotating at 20 m / s to form a ribbon shape. A diffusion alloy of Specifically, a diffusion alloy having a composition shown in Table 1 below was produced.

<Step C> (Step of forming a rare earth magnet having main phase crystal grains of a ThMn ₁₂ -type compound in which the Sm concentration in the shell phase is higher than the Sm concentration in the core phase)
The diffusion alloy as the lower layer, the bulk body of the alloy for rare earth magnets, and the diffusion alloy as the upper layer were placed in this order in a SUS container, and then covered with Nb foil. The weights of the diffusion alloys of the lower and upper layers were each 10% of the weight of the bulk body. That is, 20 wt% diffusion alloy was used for the bulk body. Rare earth magnets of Experimental Examples 1 to 13 were produced by heat treatment at 900° C. for 4 hours in an Ar stream.

A portion from the surface of the sample obtained in step C to a depth of 100 μm was removed with a surface grinder (step D). SEM was also used to evaluate the material structure. In addition, after magnetizing the bulk body and the rare earth magnets of Experimental Examples 1 to 13 with a pulsed magnetic field, the VSM was used to evaluate the magnetic properties at room temperature. Table 1 shows the difference in magnetization with respect to the bulk body (magnetization in the experimental example - magnetization in the bulk body), ΔMr (%), and the difference in coercive force with respect to the bulk body (coercive force in the experimental example - coercive force in the bulk body), ΔHc. (%). As shown in Table 1, it was possible to improve the coercive force while suppressing the decrease in magnetization in all the experimental examples.

For composition analysis, Experimental Examples 1 to 5 and Experimental Examples 9 to 13 were performed using a desktop scanning electron microscope (JCM manufactured by JEOL Ltd.) equipped with an energy dispersive X-ray spectrometer (EX-37001 manufactured by JEOL Ltd.) -6000Plus), and in Experimental Examples 6 to 8, a field emission scanning electron microscope (JSM-7001F, manufactured by JEOL Ltd.) equipped with an energy dispersive X-ray spectrometer (JED-2300F, manufactured by JEOL Ltd.). were used respectively. Table 1 shows the composition near the center of the shell phase in the main phase crystal grains of Experimental Examples 1 to 13. The composition near the center of the core phase in the main phase grains of Examples 1 to 13 was Y _0.37 Sm _{0.63 (Fe 0.63} (Fe _{0 .83} Co _0.17 ) _11.08 Al _0.07 Si _0.02 Ti _0.57 Cu _0.26 .

The magnetic properties were evaluated at room temperature with a vibrating sample magnetometer (VSM, manufactured by Toei Industry Co., Ltd., VSM-5-20) after 6T pulse magnetization. The absolute value of magnetization was corrected with Ni.

In Experimental Example 1, the Sm—Cu grain boundary phase was formed by diffusion. It was confirmed that the main phase crystal grains had a ThMn _12- type core phase and a ThMn ₁₂ -type shell phase, and that the shell phase was more Sm-rich than the core phase. However, since the formation of a shell could not be confirmed in the deep part of the bulk body, although the coercive force increased, ΔH _cJ was small compared to other experimental examples. This is due to the formation of Sm-rich shells and the disappearance of a small amount of the bcc (Fe, Co, Ti) phase that could not be completely removed in step A.

In Experimental Example 2, an Al-containing Sm—Cu grain boundary phase was formed by diffusion. It was confirmed that a Sm-rich shell phase was generated even in the deep part of the bulk body. _The average composition _{of the} shell _phase was _Y0.18Sm0.82 ( _{Fe0.83Co0.17 ) 10.58Al0.51Si0.02Ti0.62Cu0.27 .} The main phase grains have a ThMn ₁₂ -type core phase and a ThMn _12- type shell phase, and it was confirmed that a shell phase with a composition richer in Sm and slightly richer in Ti than the core phase was formed. Since the composition was richer in rare earth elements than the equilibrium composition, a 3-29 phase shell was also formed. Therefore, preferably, 2.8<ω≦4, which is poorer than the practical composition, is appropriate. Actually, in Experimental Example 3, when ω=3.4, no formation of the 3-29 phase was observed.

In Experimental Example 4, a Ga-containing Sm—Cu grain boundary phase was formed by diffusion. A shell phase with an average composition of Y _0.20 Sm _0.80 (Fe _0.82 Co _0.18 ) _10.60 Al _0.26 Si _0.13 Ti _0.71 Cu _0.30 was generated deep in the bulk body. . The main phase grains have a ThMn ₁₂ -type core phase and a ThMn ₁₂ -type shell phase, and it was confirmed that the shell phase had a composition richer in Sm, Si, Ga, and Ti than the core phase. On the other hand, since (Fe, Co, Ti) of bcc was generated at the grain boundary, the coercive force decreased. This is because the composition was poorer in the R element than the equilibrium composition. Therefore, 2≦ω<2.8, which is richer in rare earth elements than the actual composition, is more appropriate. Actually, in Experimental Example 5, when ω=2.3, formation of bcc (Fe, Co, Ti) was not observed. In addition, Al, which is an unavoidable impurity mixed in the process A, is characterized by being concentrated in the shell phase.

In Experimental Example 6, a Sm--Cu grain boundary phase containing Al and Si was formed by diffusion. Sm(Si _0.5 Cu _0.5 ) ₂ was newly generated in the diffusion alloy with the main composition. If the production amount of this R--Cu--Si-based compound becomes too large, sufficient diffusion to the deep portion of the bulk body is inhibited. Si dissolves 5 to 10 at % in the Cu portion in the 1-2 phase, and hardly dissolves in the 1-4 phase. Since Sm(Si _0.5 Cu _0.5 ) ₂ is produced, the alloy composition of the 1-2 phase and the 1-4 phase becomes the rare earth element poor side. Therefore, 2.8<ω≦4 shown in Experimental Example 2 was substantially satisfied, and the 3-29 phase was not confirmed. The average composition of the shell phase generated deep in the bulk body is Y _0.20 Sm _0.80 (Fe _0.82 Co _0.18 ) _10.60 Al _0.26 Si _0.13 Ti _0.71 Cu _0.30 became. The main phase crystal grains have a ThMn _12- type core phase and a ThMn ₁₂ -type shell phase, and it was confirmed that the shell phase had a composition richer in Sm, Si, and Ti than the core phase. A shell phase was clearly recognized even at a deeper portion of the bulk body than in Experimental Examples 2 and 3.

In Experimental Example 7, a Sm--Cu grain boundary phase containing Ga and Si was formed by diffusion. This is the same as in Experimental Example 6, although there is a difference in whether the constituent phases in the diffusion alloy are Ga or Al. Therefore, it is substantially 2.8<ω≦4, and bcc (Fe, Co, Ti) is generated at the grain boundary. Therefore, in Experimental Example 8, when ω was 2.3, formation of bcc (Fe, Co, Ti) was not observed at grain boundaries. The average composition of the shell phase was Y _0.20 Sm _0.80 (Fe _0.82 Co _0.18 ) _10.75 Al _0.09 Si _0.19 Ti _0.61 Cu _0.31 Ga _0.05 . . Construction of a shell phase having a composition that is richer in Sm and Si and slightly richer in Ga and Ti than the core phase was confirmed. A shell phase was clearly recognized even at a deeper portion of the bulk body than in Experimental Examples 4 and 5.

In Experimental Example 9, a Sm--Cu grain boundary phase containing Al, Si and Zn was formed by diffusion. _The average _{composition of the} shell _phase was _{Y0.20Sm0.80 ( Fe0.82Co0.18 ) 10.73Al0.26Si0.10Ti0.64Cu0.25Zn0.02} . However, Zn falls into the category of identification error. Construction of a shell phase with a composition richer in Sm, Si, Al, and Ti than the core phase was confirmed. A shell phase was clearly observed even at a deeper portion of the bulk body than in Experimental Example 6.

In Experimental Example 10, a Sm--Cu grain boundary phase containing Ga, Si, and Zn was formed by diffusion. This is the same as Experimental Example 7 except that the constituent phases in the diffusion alloy contain Zn or not. 2.8<ω≦4, and bcc (Fe, Co, Ti) was generated at the grain boundary. Therefore, in Experimental Example 11, the formation of bcc (Fe, Co, Ti) was not observed when the film was produced with ω=2.3. _The average composition _{of the shell phase} was _{Y0.20Sm0.80 ( Fe0.82Co0.18 ) 10.82Al0.03Si0.10Ti0.71Cu0.25Ga0.09} . Construction of a shell phase with a composition richer in Sm, Si, Ga, and Ti than the core phase was confirmed. A shell phase was clearly recognized even at a deeper portion of the bulk body than in Experimental Examples 7 and 8.

In Experimental Examples 12 and 13, diffusion alloys partially containing Y were used. The composition of the shell phase was not significantly inferior to the shell compositions of Experimental Examples 9 and 11, and the coercive force improvement rate was sufficient.

It was confirmed that the samples prepared in the above experimental examples had concentration gradients in the core phase and the shell phase corresponding to the element species. In particular, when Si is contained as the M element, the Si concentration gradient within the core phase is represented by a position-dependent linear function. Sm is often constant within the shell phase. This may be due to the fact that the concentration gradient in the core phase is due to solid-phase diffusion, and the concentration formation of the shell phase is due to dissolution and reprecipitation, which is mainly caused by applying the disclosed technology. It is one of the characteristic organizational forms.

For example, FIGS. 1A and 2A show the results of linear analysis of the structure and composition of Experimental Example 9, and FIGS. 1B and 2B show the results of linear analysis of the structure and composition of Experimental Example 11, respectively. In the measurement, since there is a part where the measurement is performed by interfering with both the core phase and the shell phase, that part is omitted (the part indicated by "//"). Si is shown to diffuse into the core phase as well. It was found that Al and Ga also diffuse into the core phase, but less so than Si. The Si concentration gradient in the core phase shows a position-dependent linear change. Although it is not clear, there is a possibility of a linear function in the range where Al and Ga are also diffused. We observed that Co tends to be slightly richer in the shell phase than in the core phase. This means that the amount of Co substituted in the core phase is slightly less than 0.2, which maximizes the magnetic property value, and it is possible to enrich Co in the shell phase to y ~ 0.2. do. By doing so, it is possible to reduce the amount of expensive Co used. Due to the concentration gradient of elements between the core phase and the shell phase, the lattice constant changes continuously and the lattice mismatch is eliminated.

On the other hand, in <Step A>, the alloy composition was designed such that w=9 and x=0.60 so that the composition of the entire alloy was Y-rich, and a densified bulk body was obtained in the same manner. It was found that inside the bulk body, a shell with a higher Sm concentration than inside the magnetic powder was formed on the outer periphery of the magnetic powder. The composition of the core is Y _0.40 Sm _0.60 (Fe _0.82 Co _0.18 ) _11.07 Al _0.04 Si _0.02 Ti _0.62 Cu _0.25 , whereas the composition of the shell is _{The composition was Y0.29Sm0.71 ( Fe0.83Co0.17 ) 11.08Al0.04Si0.02Ti0.62Cu0.24 .}

In order to increase the coercive force of a magnet having a ThMn ₁₂ -type crystal structure as the main phase, it is necessary to understand the crystal structure of the main phase crystal grains in detail. Therefore, nuclear density analysis was performed using the Maximum Entropy Method (MEM) by powder neutron Rietveld analysis. Since X-ray diffraction is caused by spatially spread electron density, resolution may be inferior to determine the positions of trace elements with high accuracy. On the other hand, since neutron beams are scattered from nuclei near mass points, they are suitable for determining the positions of trace elements with high accuracy. The measurement was performed by Ibaraki Prefecture material structure analysis installed at BL20 in the Material and Life Science Experimental Facility (MLF) of the Japan Proton Accelerator Research Complex (J-PARC). A device (iMATERIA) was used. Measured at room temperature in single frame mode. In order to correct the absorption of a sample containing Sm, which has a large neutron absorption, a hollow vanadium cylinder (φ5.8) was placed in a vanadium container (φ6), and a powder sample (<75 μm) was placed in the gap between them. was filled (sample geometry with cylindrical symmetry). Analytical data were analyzed in the range of 0.49 Å<d<2.57 Å using back bank data from which the background of the vanadium container was subtracted. Analysis codes used were Z-Rietveld 1.0.2 and Z-MEM 0.8.3.

The sample prepared in step A was pulverized to 75 μm or less and evaluated. This sample is a ThMn ₁₂ -type compound in which Y is partially substituted to reduce the substitution amount of the structure stabilizing element Ti. Moreover, it was confirmed that the following occurred regardless of the presence or absence of Cu addition.

FIG. 3 shows the 3D, (100) plane, and (110) plane nuclear density distributions derived by MEM. The non-isotropic nuclear density distribution corresponds to the anisotropic thermal vibration. It has been clarified that there are two types of places with high nucleus density distribution in the [001] direction of the 2a site, which is a normal rare earth site. In addition, the _4g1 site, which is a remnant of the pseudo-tetragonal system, was also observed, although it was extremely small. Crystallographically, it corresponds to the 2b site (the site where the interstitial element enters) and the 4e site. Since the 2a site is positioned between the 4e sites, the 4e site forms a dumbbell and can be viewed as a crystal structure in which the 2a site and the 2a site are irregularly substituted. It is inferred that the extruded rare earth element is located at the 2b site where the dumbbell at the 4e site replaces the rare earth element at the 2a site. When this fact was taken into the crystal structure model and the Rietveld analysis was performed again, the accuracy of the fitting analysis was improved. As a result, it was found that the random substitution rate is about 2%, which is about 1 per 25 unit cells. Considering the error and the change according to the amount of Ti, it may be included at a maximum of about 5%. This 2% to 5% random substitution may reduce the coercive force. Eliminating such disordered substitution inside the structure by concentrating the structure-stabilizing element in the outer periphery of the crystal may be equivalent to eliminating the origin of magnetization reversal. Actually, in the shells of Experimental Examples 2 to 11 in which Ti and Si are concentrated, there is a possibility that the coercive force is improved due to the reduction or disappearance of the random substitution ratio.

FIG. 4 shows the correspondence of each site during the structural change from the TbCu _7- type to the ThMn ₁₂ -type seen in the Y—Fe-based compound. The continuous lattice deformation of Y--Fe based compounds described in Patent Document 3 teaches the shortest path deformation from TbCu ₇ type to ThMn ₁₂ type. On the other hand, the structure near the Ti _0.5 composition, which is the transition region from the non-equilibrium phase to the equilibrium phase, which was newly discovered through extensive research, has an irregular substitution part in the direction of c _tetra , and the detour It became clear that it shows a structural change. In other words, it was shown that the _cortho -direction irregularities inherent in the TbCu _7- type disappear and the c- _tetra- direction irregularities appear and disappear, transforming into the ThMn _12- type. In addition, since the (100) in-plane direction is too narrow for the rare earth element to enter the 2b site, it is possible that an element other than the rare earth element may be contained.

An embodiment of the present disclosure provides a rare earth magnet having main phase crystal grains of a ThMn ₁₂ -type compound and having improved coercive force while suppressing a decrease in magnetization. Since such rare earth magnets can be suitably used in motors, actuators, etc., they have various industrial uses.

Claims (5)

A rare earth magnet having main phase crystal grains of an Fe-based ThMn ₁₂ -type compound containing Sm and Y and a grain boundary phase containing Cu,
The main phase crystal grains have a ThMn ₁₂ -type core phase and a ThMn ₁₂ -type shell phase,
A rare earth magnet, wherein the Sm concentration in the shell phase is higher than the Sm concentration in the core phase. The main phase crystal grains contain at least one M element selected from the group consisting of Ti, Si, Al, and Ga, and the concentration of the M element in the shell phase is equal to the concentration of the M element in the core phase. 2. The rare earth magnet of claim 1, wherein the rare earth magnet is higher than 3. The rare earth magnet according to claim 2, wherein Si is contained as the M element, and the Si concentration gradient in at least the core phase is represented by a position-dependent linear function. In the composition formula R1 _1-x R2 _x (Fe _1-y Co _y ) _w Ti _z Cu _α , R1 has at least Y and optionally Gd, and R2 has at least Sm and further at least one selected from the group consisting of La, Ce, Nd and Pr, and x, y, z and w are 0.5≦x≦1.0, 0≦y≦0.4 and 8, respectively; preparing a bulk body of rare earth magnet alloy, wherein ≤w≤12, 0.42≤z<0.70, and 0.40≤α≤0.70;
a step of preparing a diffusion alloy having a composition formula R1 _1-χ R2 _χ Cu _ω system, wherein χ>x and 2≦ω≦4;
By heating the bulk body and the diffusion alloy in contact with each other and diffusing the component elements of the diffusion alloy into the bulk body, the Sm concentration in the shell phase is higher than the Sm concentration in the core phase ThMn ₁₂ forming a rare earth magnet having main phase grains of a type compound;
A method for producing a rare earth magnet, comprising: The diffusion alloy is represented by a composition formula R1 _1-χ R2 _χ (Cu _1-ν-τ X _ν Si _τ Zn _σ ) _ω , where X is at least one of Al and Ga, χ>x, 0< 5. The method for producing a rare earth magnet according to claim 4, wherein ν≦0.3, 0<τ<0.5, 0<σ<0.2, and 2≦ω≦4.

Images