JP5023103B2 - Permanent magnet material - Google Patents

Description

  The present invention relates to the structure of a transition metal compound permanent magnet.

  In the case of a permanent magnet material having a low Curie temperature, the coercive force must be improved to prevent demagnetization in a high temperature environment. The coercive force is probably influenced by the structure of the fine structure. Therefore, as described in Patent Document 1, the coercive force is reduced by reducing the grain size and unevenly distributing elements having a large magnetic anisotropic magnetic field near the grain boundary. Attempts have been made to improve.

  Further, as disclosed in Patent Document 2, for example, by introducing an appropriate element, the symmetry of the crystal lattice of the master alloy is lowered, the distance between transition metals is increased, the magnetic interaction is improved, and the magnetic moment Attempts have been made to increase the temperature and increase the Curie temperature. Further, as disclosed in Patent Document 3, by forming a core-shell structure of FePt and Fe, high magnetic properties are realized by forming a nanocomposite of a hard magnetic phase and a soft magnetic phase.

WO2006043348A1 JP 2008-78610 A JP 2008-248364 A

  In particle size control, the mother alloy is easily affected by oxidation, and there is a limit to miniaturization. The structure in which an element having a large magnetic anisotropy magnetic field is arranged in the vicinity of the grain boundary described in Patent Document 1 has a problem in resource security because it uses the rare resource elements Dy and Tb. In addition, since the structure in which an appropriate element has intruded into the crystal lattice described in Patent Document 2 is a metastable phase, there is a problem that the invading element is detached at a high temperature and cannot be sintered. The core-shell structure described in Patent Document 3 uses the rare resource element Pt, which still has a problem in resource security.

  Note the large magnetovolume effect of rare earth transition metal alloys. However, it relates to all material systems with a large magnetovolume effect. A core-shell structure in which fine crystal grains are mainly composed of two types of phases is formed, and the crystal lattice volume of the main phase or the distance between each atomic site is controlled by the large magnetic volume effect of the rare-earth-3d transition metal magnet, and exchange mutual The action is increased. The core-shell structure is composed of a main phase in which exchange interaction is increased, and a second phase in which the crystal lattice of the main phase is controlled by stress. When the main phase is Fe-based, tensile stress is applied from the second phase due to the difference in thermal expansion coefficient, thereby causing an increase in exchange interaction due to expansion of the crystal lattice. In order for the tensile stress to be properly applied, the thermal expansion coefficient of the second phase needs to be smaller than the thermal expansion coefficient of the main phase. In particular, the effect is large when the thermal expansion coefficient of the second phase is negative. As such a second phase material system, a substance group exhibiting a magnetovolume effect (invar effect) is suitable. On the other hand, when the main phase is a Co group or a Ni group, an increase in exchange interaction is caused by contraction of the crystal lattice due to the compressive stress applied from the second phase due to the difference in thermal expansion coefficient. In order for the compressive stress to be properly applied, the thermal expansion coefficient of the main phase needs to be smaller than the thermal expansion coefficient of the second phase. In particular, the effect is large when the thermal expansion coefficient of the main phase is negative. As such a main phase material system, a substance group exhibiting a magnetovolume effect (invar effect) is conceivable. In addition, domain wall pinning occurs from the structure by heterogeneous mixing, and the coercive force increases.

  By using the present invention, it is possible to provide a highly functional permanent magnet material that does not use rare resource elements such as Dy and Tb and is structurally stable. It is possible to provide a permanent magnet material characterized in that the exchange interaction of the main phase is augmented by the stress generated by the difference in volume change between the two phases, and the Curie temperature is increased.

The magnetic structure (1) using stress is shown. 2 shows a magnetic structure (2) using stress. 2 shows a magnetic structure (3) using stress. The magnetic structure (4) using stress is shown. The temperature and magnetic field at which crystal lattice distortion and magnetic order occur in each combination of the main phase and the second phase are shown. The element selective demagnetization curve obtained by XMCD measurement is shown.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  In magnetic materials based on transition metals, magnetism is manifested by band polarization. In many cases, the 3d transition metal group having relatively strong itinerary is described by the Hubbard model, and in the 4f rare earth metal group having strong localization, the Anderson model is often used. In the Hubbard model, the electronic state and magnetic structure are determined by the competition between the gain of kinetic energy reduction due to the spatial spread of electrons and the increase of Coulomb energy due to the proximity of electrons. In the Anderson model, the electronic state and magnetic structure are determined by further considering the interaction between conduction electrons and localized electrons in the Hubbard model. The focus of the present invention relates to the Kanamori theory predicted from a single Hubbard model of a 3d transition metal. The Kanamori condition removes the overestimation of Coulomb energy from the Stoner condition and gives an indication of ferromagnetism.

Where U is the Coulomb energy, G (0,0) is a parameter between two electrons having a wave vector of 0, and the magnitude D (E _F ) of the reciprocal of the 3d bandwidth is the density of states of electrons in the Fermi level. Respectively. The Kanamori condition indicates that in order for ferromagnetism to occur, the bandwidth needs to be quite large and at the same time the density of states on the Fermi surface must be locally increased. Since the state density increases as the crystal lattice increases, the electronic state density changes sharply due to the volume change of the unit cell. Therefore, when a volume change is introduced into the unit cell by forced or spontaneous force, it is expected that a large change in magnetism occurs due to a change in the density of states in the vicinity of the Fermi level.

For example, a Bethe-Slater curve showing the interatomic distance and the magnitude of the exchange interaction of a 3d transition metal alloy is a curve showing that the itinerant electron magnetic interaction oscillates according to the interatomic distance. In order to maximize the magnitude of the exchange interaction in the Bethe-Slater curve, it is known that αFe has an interatomic distance that is too short and Co and Ni have an interatomic distance that is too wide. This is because αFe has too much itinerant nature of electrons and few localized electrons, so exchange interaction is small, and Co and Ni have too strong localization and small overlap of wave functions, so exchange interaction is small. Means. In other words, it means that the exchange interaction can be increased by increasing the interatomic distance in Fe and by reducing the interatomic distance in Co and Ni. Actually, the Curie temperature correlated with the magnitude of the exchange interaction is 0 ± 0.1 for αFe, 0.35 ± 0.02 for Ni as pressure dependence (∂Tc / ∂P) (° C./kbar), Co indicates 0 ± 1. In particular, the Ni alloy has a large pressure dependency. In addition, in the RKKY interaction that can be predicted from the Anderson model and is actually observed, the surrounding conduction electrons are subjected to spin polarization by the electric field of the electrons localized in the atoms, thereby causing the local atoms to be localized. Interacts with electrons. In the RKKY interaction, the exchange interaction oscillates according to the interatomic distance.

In the present invention, based on the above idea, the volume change is applied to the rare earth-transition metal magnet, and the exchange interaction is increased by controlling the crystal lattice volume or the distance between each atomic site. In the 4f rare earth element-3d transition metal magnet, for example, R ₂ T ₁₄ B, R ₂ T ₁₄ C, R ₂ Fe ₁₄ BH, R ₂ T ₁₇ , RT ₅ , RT ₇ A _x , RT ₉ A _x , RT ₁₂ A _x , R ₃ T _(29-x) A _x or the like (where R is a rare earth element, T is a 3d transition metal, A is a structural stabilizing element, x is a suitable positive number) or a mixture thereof. In all compositions, the atomic weight of T is larger than R.
In particular, a promising composition as a permanent magnet material has an atomic weight of T with respect to R of 5 or more. Therefore, in the above, the Bethe-Slater curve based on the Hubbard model is used as a reference for consideration because it is strongly influenced by the band polarization of T. However, the essence of the present invention is to control the interaction of itinerant electron magnetism by the difference in thermal expansion coefficient (also referred to as the thermal expansion coefficient), particularly by the magnetic volume effect. It is not limited.

  The structure for realizing the present invention requires a fine structure composed of at least two components. A core-shell structure is particularly desirable. One is a main phase in which exchange interaction is increased, and one is at least a second phase in which the crystal lattice of the main phase is controlled by stress. However, the stress includes a normal stress in which the normal direction of the minute surface and the direction of the force coincide, and a shear stress that does not coincide. Normal stress is called tensile stress when the force acting surface is perpendicular to the force acting direction, acting in the direction of pulling the acting surface, and compressive stress when acting in the direction of pushing the acting surface. Are used according to this definition. In FIG. 1 (magnetic structure (1) using stress), the main phase is a) Fe group, b) Co group or Ni group as the magnetic structure in which exchange interaction increases using stress. Indicates. In the case of Fe group, the tensile stress is applied from the second phase due to the difference in thermal expansion coefficient, thereby causing an increase in exchange interaction due to expansion of the crystal lattice. In order for the tensile stress to be properly applied, the thermal expansion coefficient of the second phase needs to be smaller than the thermal expansion coefficient of the main phase. In particular, the effect is large when the thermal expansion coefficient of the second phase is negative. As such a second phase material system, a substance group exhibiting a magnetovolume effect (invar effect) is suitable. On the other hand, in the case of Co group and Ni group, the compressive stress is applied from the second phase due to the difference in thermal expansion coefficient, which causes an increase in exchange interaction due to the shrinkage of the crystal lattice. In order for the compressive stress to be properly applied, the thermal expansion coefficient of the main phase needs to be smaller than the thermal expansion coefficient of the second phase. In particular, the effect is large when the thermal expansion coefficient of the main phase is negative. As such a main phase material system, a substance group exhibiting a magnetovolume effect (invar effect) is conceivable.

  The above structure causes a steep change in the density of states of electrons near the Fermi level of the main phase, causing an increase in exchange interaction, resulting in a change in magnetism in the form of an increase in magnetic moment or an increase in Curie temperature. .

  When the volume change due to a magnetic field and the volume change due to pressure occur by the same mechanism, the following thermodynamic relational expression generally holds.

Where V is the volume of the system, G is the free energy of the cast, P is the pressure, H is the magnetic field, T is the temperature, M is the magnetization, and ω is the spontaneous volume magnetostriction (ΔV / V). The equality relationship between the terms 2 and 3 in this equation means that the magnetization is increased by pressurization and the volume contraction by applying a magnetic field is equivalent. Hereinafter, the magnetic field change (∂ω / ∂H) _{T, P} of the spontaneous volume strain is defined as ν. Based on this, when the exchange interaction of the main phase is increased by the pressure and the magnetic moment is improved, the structure can be defined by ν. ν is a physical quantity obtained by measuring the magnetic field dependence of crystal lattice strain and can be measured by the optical lever method, the capacitor capacitance method, the inductance method, and the strain gauge method.

FIG. 2 (a magnetic structure using stress (2)) shows a magnetic structure in which exchange interaction is increased using stress. The main phase is a) when ν> 0, and b) when ν <0. Show. When ν> 0, tensile stress is applied from the second phase due to the difference in thermal expansion coefficient, thereby causing an increase in exchange interaction due to the expansion of the crystal lattice. In order for the tensile stress to be properly applied, the thermal expansion coefficient of the second phase needs to be smaller than the thermal expansion coefficient of the main phase. In particular, the effect is large when the thermal expansion coefficient of the second phase is negative. As such a second phase material system, a substance group exhibiting a magnetovolume effect (invar effect) is suitable. On the other hand, when ν <0, a compressive stress is applied from the second phase due to the difference in thermal expansion coefficient, thereby causing an increase in exchange interaction due to shrinkage of the crystal lattice.
In order for the compressive stress to be properly applied, the thermal expansion coefficient of the main phase needs to be smaller than the thermal expansion coefficient of the second phase. In particular, the effect is large when the thermal expansion coefficient of the main phase is negative. As such a main phase material system, a substance group exhibiting a magnetovolume effect (invar effect) is conceivable.

The structure described in Example 1 or Example 2 must be such as to give strain to the crystal lattice. As shown in FIG. 3 (magnetic structure using stress (3)), the structure composed of the main phase and the second phase is preferably superparamagnetic critical particle size or more and 100 nm or less, preferably about 50 nm. It is as follows. In that case, the above-described tensile stress and compressive stress are appropriately applied to the crystal lattice. However, both the main phase and the second phase must be larger than the superparamagnetic critical particle size. This is because, in the superparamagnetic state, the increase in entropy due to thermal fluctuations lowers the free energy rather than the gain in energy decrease due to the uniform magnetic moment, so that spontaneous magnetization and magnetic anisotropy do not appear. For example, in a system with large magnetic anisotropy energy, the superparamagnetic critical particle size is small, about 2 nm for Sm ₂ Fe ₁₇ and about 3 nm for PtFe.

  When the crystal grain size is on the order of submicrons, the microstructure is more deformed than the crystal lattice is deformed by stress, so that cracks occur and the stress does not contribute to the crystal lattice effectively. From the above, the effect of the present invention is most remarkable in the core-shell structure of 20 nm to 40 nm. Further, due to the matching of the crystal lattice in the boundary region of the core-shell structure, the stress between the main phase and the second phase is appropriately applied to the crystal lattice, and the effect of the present invention becomes more remarkable. Such a structure can be constructed by a nanoparticle process or a thin film process. For example, the vapor phase method includes a thermal CVD method, a plasma CVD method, a molecular beam epitaxy method, a sputtering method, an EB vapor deposition method, a reactive vapor deposition method, a laser ablation method, a resistance heating vapor deposition method and the like. The liquid phase method includes coprecipitation method, microwave heating method, micelle method, reverse micelle method, hydrothermal synthesis method, sol-gel method and the like.

As shown in FIG. 4 (magnetic structure using stress (4)), in the structures described in Example 1 and Example 2, when the main phase is Fe group, the main phase is Co. In the case of a group and a Ni group, a substance group exhibiting an Invar effect is appropriate for the main phase. For example, as a material system showing the Invar effect, Ni ₃₆ Fe ₆₄ , PtFe ₃ , Laves phase intermetallic compounds (for example, RCo ₂ , Zr _1-x Nb _x Fe ₂ , YMn _2, etc.), alloys containing 4f rare earth elements, Compound, reverse perovskite type manganese nitride (for example, Mn ₃ CuN etc.) and the like. The Invar effect is defined as an Invar effect if there is a temperature range where the coefficient of thermal expansion is 0 or less, but in a broad sense, the Invar effect occurs when the behavior deviates along with the magnetic order due to the temperature dependence of the volume due to the phonon thermal oscillation. Or it is called an effect like Invar. Here, the two are not distinguished and referred to as the Invar effect or the magnetic volume effect. Below the Curie temperature of these substance groups, stress is applied to the main phase, and the exchange interaction of the main phase is improved. Moreover, by applying stress, domain wall pinning occurs, and an increase in coercive force can be expected.

As the main phase, a larger amount of T than R is suitable as a permanent magnet from the viewpoint of magnetic flux density. For _{example, R 2 T 14 B, R} 2 T 14 C, R 2 Fe 14 BH, R 2 T 17, RT 5, RT 7 A x, RT 9 A x, RT 12 A x, R 3 T (29-x ₎ A _x or the like (where R is a rare earth element, T is a 3d transition metal, A is a structural stabilizing element, and x is an appropriate positive number) or a mixture thereof. These compositions may contain a rare earth element R having a large spin orbit interaction, some have a negative thermal expansion coefficient, and have a larger magnetovolume effect (invar effect) than other magnetic materials. Yes.
However, the generation mechanism of the magnetovolume effect is different for each substance, and the details are often unknown.

In the case of rare earth transition metal alloys and compounds, the distance between transition metal elements varies depending on the crystal lattice and the site. Depending on the distance between the sites, the magnitude of the exchange interaction is different. By selectively changing the volume of antiferromagnetic interaction or geometrically unstable spin sites, the exchange interaction can be improved. A stable spin configuration can be realized, and the magnetic moment can be increased or the Curie temperature can be increased. For example, the low Curie temperature of the R ₂ Fe ₁₇ crystal structure is due to the fact that the value of exchange interaction between Fe at the 4f site shows a large negative value. It is known that the Curie temperature is drastically improved by increasing the distance between them and blocking the direct exchange interaction. Thus, by improving a specific site, it is possible to dramatically improve the characteristics of the entire crystal system. This example details a method for changing the magnetism of a rare earth transition metal alloy or compound by improving the site characteristics by using the magnetic volume effect. The magnetovolume effect is used in the sense of strong spin / lattice coupling.

  Specifically, the overall exchange interaction can be improved by expanding the sites that are antiferromagnetic interactions or blocking the exchange interaction. In order to perform site-selective adjustment, elements that cause spin / lattice coupling with one element (elements with large orbital angular momentum) and elements with local strain due to the Jan-Teller effect are candidates. Examples of elements having large spin / lattice coupling include rare earth elements excluding zero orbital angular momentum. As for the Yang-Teller effect, an element having an open shell orbit (magnetic element) is selected for each system due to the symmetry of the crystal, while local anion O, F, Cl, S, Br, I is introduced locally. Crystal field adjustment can be performed by strain. In addition, when Pd, Rh, and Pt, which are observed to be paramagnons and are said to be on the verge of ferromagnetism, are expected to undergo a significant change in magnetism, there are many unexplained cases.

  The magnetovolume effect of the present invention does not indicate a phenomenon including a change in material structure such as magnetostriction, but indicates a physical property phenomenon caused by coupling between a crystal lattice and magnetism. Generally, it is a phase transition phenomenon, and there is a correlation between magnetic order and crystal lattice. For this reason, phase transitions with temperature and external magnetic field as parameters are defined, and magnetic order and crystal lattice distortion occur at the same temperature and the same magnetic field.

Specifically, some of the structures described in Example 4 were prepared and evaluated. FIG. 5 (temperature and magnetic field at which crystal lattice distortion and magnetic order occur in each combination of the main phase and the second phase) shows the main phase and its average grain size, the second phase and its average thickness, and the phase transition in each composition. Temperature and phase transition magnetic field are shown. The average particle diameter and average thickness of each sample were determined by scanning transmission electron microscopy (STEM), and the phase transition temperature and phase transition magnetic field were evaluated by X-ray diffraction (X-ray diffrac).
iton, XRD), and superconducting quantum interference device (SQUID). Magnetic order temperature of the second phase (Curie temperature or Neel temperature)
Is higher than the magnetic ordering temperature of the main phase, the measured phase transition temperature and phase transition magnetic field are almost equal to those of the second phase. As can be seen from Table 1, the phase transition temperature and the phase transition magnetic field were measured by XRD and SQUID and had almost the same value, indicating that this was a phenomenon in which magnetism and crystal lattice were coupled. However, a clear phase transition magnetic field was not measured with the combination of Nd ₂ Ni ₁₇ and Fe.

In the fine crystal grains of the present invention, the main phase is an alloy containing a rare earth element having a large magnetic anisotropy (rare earth transition metal alloy) or a structure-stabilizing element added to the alloy, and in that case Often has a large magnetovolume effect. For example, the main phase is RT ₅ , RT ₇ , RT ₉ , R ₂ T ₁₇ , R ₃ T _(29-x) A _x , RT ₁₂ A, etc., and the second phase contains at least part of the main phase. , Carbonized, nitrided, oxidized, and fluorinated structures. The combination of the main phase and the second phase is arbitrary as long as it satisfies the condition of the coefficient of thermal expansion in the operating temperature range, and is not limited to the described composition and structure. In the case of Fe group, such a composition and structure is obtained by coating at least one of boride, carbonization, nitridation, oxidation, and fluorination with a pre-treatment alloy, In the case of a Co group, the main phase can be easily obtained by performing at least one of boring, carbonization, nitriding, oxidation, and fluorination later, the extent of which depends on the heat treatment temperature and heat treatment time. , Easily achieved by atmosphere control. The obtained structure has a concentration gradient in at least one of B, C, N, O, and F. In addition, it is possible to cause exchange interaction between two phases by nanocompositing, but it is difficult to control stress.

For example, Sm ₂ Fe ₁₇ in the main phase, is described a method for manufacturing of the second phase when you select the Sm ₂ Fe ₁₇ N _3. As the main phase Sm ₂ Fe ₁₇ , a magnetic powder obtained by pulverizing an Sm—Fe-based ribbon produced by rapidly cooling a main phase alloy having an adjusted composition was used. Sm-Fe-based main phase alloys take into account the evaporation of rare earths, mix Sm with Fe more than the stoichiometric ratio, and dissolve in vacuum, inert gas or reducing gas atmosphere to make the composition uniform ing. If necessary, the main phase alloy, which is cut on the surface of the rotating roll, is dissolved in an inert gas or reducing gas atmosphere such as argon gas. After the jet quenching and thinning, heat treatment is performed in an inert gas or a reducing gas atmosphere. The heat treatment temperature is 200 ° C. or more and 700 ° C. or less, and Sm ₂ Fe ₁₇ crystal grain growth is generated by this heat treatment, so adjustment is necessary as appropriate. A part of Sm ₂ Fe ₁₇ is subjected to a heat treatment for 2 hours to 10 hours (4 hours this time) in a nitrogen atmosphere of 200 ° C. or higher and 500 ° C. or lower (this time 350 ° C.), so that the nitrogen content becomes Sm ₂ Fe ₁₇ composition. Completely invaded. The Sm ₂ Fe ₁₇ N ₃ compound is transferred to an environment of 400 ° C. in a vacuum of 5.0 × 10 ⁻⁵ Pa or less, and a method using a roll such as a single roll or a twin roll method is applied to the surface of the rotating roll. The dissolved Sm ₂ Fe ₁₇ main phase alloy was sprayed onto the Sm ₂ Fe ₁₇ N ₃ compound. Core-shell type magnetic powder was produced by controlling the atmosphere for an appropriate time. However, because of the influence of oxidation, the area around the magnetic powder was carbonized and hydrocarbonized.

Further, Sm ₂ Ni ₁₇ in the main phase, is described a method for manufacturing of the second phase when you select the Sm ₂ Ni ₁₇ N _3. As in the case of the Fe group, the main phase Sm ₂ Ni ₁₇ was a magnetic powder obtained by pulverizing an Sm—Ni-based ribbon produced by quenching a main phase alloy whose composition was adjusted. In the Sm-Ni main phase alloy, Sm is mixed with Ni more than the stoichiometric ratio and dissolved in a vacuum, an inert gas, or a reducing gas atmosphere to make the composition uniform. If necessary, the main phase alloy, which is cut on the surface of the rotating roll, is dissolved in an inert gas or reducing gas atmosphere such as argon gas. After the jet quenching and thinning, heat treatment is performed in an inert gas or a reducing gas atmosphere. The heat treatment temperature is 200 ° C. or higher and 700 ° C. or lower, and Sm ₂ Ni ₁₇ crystal grain growth is generated by this heat treatment. A part of Sm ₂ Fe ₁₇ is subjected to heat treatment for 2 hours to 10 hours (2 hours this time) in a nitrogen atmosphere of 200 ° C. or higher and 500 ° C. or lower (this time 230 ° C.), thereby forming nitrogen into Sm ₂ Fe ₁₇ Only the outer peripheral part of was nitrogenated. However, because of the influence of oxidation, the area around the magnetic powder was carbonized and hydrocarbonized. The nitrogen substitution rate (Sm ₂ Fe ₁₇ N ₃ ) of the magnetic powder was judged from X-ray diffraction and Curie temperature measurement using magnetic powder produced under the same heat treatment conditions and atmosphere.

  The magnetic properties of the magnetic powder produced as described above were evaluated. In the case of Fe group, since the Curie temperature of the main phase was higher than that of the second phase, there was no significant increase, but the temperature increased from 480 ° C. to about 10 ° C. to about 490 ° C. In addition, a stepwise increase in magnetization was observed at the Curie temperature (120 ° C.) of the second layer. In the case of Ni, the Curie temperature was improved from about 480 ° C. to about 20 ° C. to about 500 ° C. At the Curie temperature of the second phase, a stepwise increase in magnetization was observed as before.

The main phase is an alloy containing a rare earth element having a large magnetic anisotropy, and in that case, it often has a large magnetovolume effect. For example, the main phase is RT ₅ , RT ₇ , RT ₉ , R ₂ T ₁₇ , R ₃ T _(29-x) A _x , RT ₁₂ A, etc., and the second phase contains at least part of the main phase. , Carbonized, nitrided, oxidized, and fluorinated structures. The combination of the main phase and the second phase is arbitrary as long as it satisfies the condition of the coefficient of thermal expansion in the operating temperature range, and is not limited to the described composition and structure. In the case of Ni group and Co group, such a composition and structure can be easily obtained by performing at least one of boring, carbonization, nitriding, oxidation, and fluorination on the main phase later. The degree can be easily achieved by heat treatment temperature, heat treatment time, and atmosphere control.

Sm ₂ Ni ₁₇ was selected as the main phase and Sm ₂ Ni ₁₇ N ₃ was selected as the second phase, and it was fabricated by sputtering.
Each target uses a purchased product, and the purity is 99.9%. The introduced argon gas (99.999%) was turned into plasma by applying a voltage with the substrate facing the sputtering target in an ultrahigh vacuum chamber (<1 × 10 ⁻⁷ Pa). At this time, the generated ions were accelerated by an electric field to strike the target, and the target component jumped out and adhered to the opposing substrate surface. The argon gas pressure was adjusted as appropriate. The used apparatus has three rooms, a metal film forming chamber, a reactive sputtering chamber, and a plasma nitriding chamber. Although Si was used for the substrate for convenience, it is not limited thereto. In the metal film forming chamber, Sm ₂ Ni ₁₇ was deposited on the Si substrate so that the film pressure was about 70 nm. This was transferred to a plasma nitriding chamber, and the outer peripheral portion of about 20 nm was nitrided. The structural evaluation was performed by X-ray diffraction and scanning transmission electron microscope energy dispersive X-ray analysis (STEM-EDS), and it was confirmed that a predetermined crystal phase was formed in both the main phase and the second phase. The magnetic properties of the fine crystal grains produced in this way were evaluated. As a result, the Curie temperature was 510 ° C., which was about 30 ° C. higher than that of Sm ₂ Ni ₁₇ of the main phase. Further, in the vicinity of 120 ° C., which is the Curie temperature of the second phase, a stepwise increase in magnetization was observed.

The grain boundary diffusion technology of rare earth fluoride was applied to the NdFeB sintered body. The techniques for thermally diffusing Dy inside the magnet body along the grain boundary include (1) a method of exposing to Dy vapor between 500 ° C. and 900 ° C., and (2) a method of applying a slurry of DyF _3. It has been. In this example, a sintered body diffused by grain boundaries was produced by the following method.

This magnet body is immersed in the DyFx solution. This DyFx solution is prepared by dissolving Dy (CH ₃ COO) ₃ as a raw material with H ₂ O and adding HF to gelatin-like DyF ₃ .XH ₂ O or DyF ₃ .X (CH ₃ COO) (x is About 3) was formed as a positive number, and the solvent was removed by centrifugation, and alcohol was added to obtain a DyF _x state. Specifically, a rare earth fluoride or alkaline earth metal fluoride coating film forming treatment solution was prepared as follows.

  (1) A salt having high solubility in water, for example, in the case of Dy, 4 g of acetic acid Dy was introduced into 100 mL of water, and completely dissolved using a shaker or an ultrasonic stirrer.

(2) Hydrofluoric acid diluted to 10% was gradually added in an amount equivalent to the chemical reaction for producing DyF _x (x = 1-3).

(3) The solution in which the gel-like precipitate DyF _x (x = 1-3) was formed was stirred for 1 hour or more using an ultrasonic stirrer.

  (4) After centrifugation at 4000 to 6000 rpm, the supernatant was removed and approximately the same amount of methanol was added.

  (5) The methanol solution containing the gel-like DyF cluster was stirred to form a complete suspension, and then stirred for 1 hour or more using an ultrasonic stirrer.

  (6) The operations of (4) and (5) were repeated 10 times until no anion such as acetate ion or nitrate ion was detected.

In the case of the DyF system, it became a substantially transparent sol-like DyFx. A methanol solution containing 1 g / 5 mL of DyFx was used as the treatment liquid. This sintered body is dipped in the solution and vacuum degassed to remove the solvent. The operation of immersion and vacuum deaeration is appropriately adjusted according to the amount to be applied. This time, 5 times. Thereafter, DyF is thermally diffused inside the magnet body by heat treatment in a temperature range of 300 ° C. to 900 ° C. For example, heat treatment was performed at 800 ° C. this time. Although rare earth oxides containing Dy are already formed at the grain boundaries, Dy, C, and F constituting the fluorine compound diffuse along the grain boundaries and exchange with Nd constituting the crystal grains. Diffusion occurs. It is considered that such diffusion occurs in the diffusion along the grain boundary because the oxyfluoride is more stable than the rare earth oxide containing Dy. It has been found that an acid fluorine compound or a fluorine compound is formed at the grain boundary triple point and is composed of DyF ₃ , DyF ₂ , DyOF, NdOF, (Nd _0.5 Dy _0.5 ) OF, and the like. Furthermore, it was found that C is also contained in these acid fluorine compounds and fluorine compounds. Fluorine atoms are detected at the grain boundaries, and Dy is concentrated in an average range of 1 nm to 500 nm from the grain boundaries. At a distance of 100 nm from the center of the grain boundary, the concentration of Dy is 1/2 to 1/10 as a ratio (Dy / Nd) with Nd. Such a method for producing a high coercive force magnet utilizing thermal diffusion from the surface of the magnet body is particularly effective when applied to a magnet of 10 mm or less.

In order to evaluate the magnetization reversal state on the surface of the NdFeB grain boundary diffusion magnet produced as described above, the X-ray magnetic circular dichroism (XMCD) measurement was performed on the SPring-8 industrial beam. Conducted on line BL16XU. XMCD measurement is 5μ from the surface.
In order to detect magnetic signals from m to 10 μm, the magnet surface must be cleaned before measurement. Therefore, after polishing the magnet surface with a wrapping film for about 5 μm, the magnet surface was milled with Ar ions in a high vacuum of 4 × 10 ⁻⁵ Pa or less. In order to prevent the oxidation of the surface, the surface was covered with a diamond like carbon (DLC) film. Three such samples were prepared, magnetized at 4T, and then subjected to XMCD evaluation. In the XMCD experiment, a permanent magnet type magnetic field application mechanism was used, and a maximum magnetic field of about 0.7 was applied. Incident X-ray energy is Fe K edge (7.113 keV), Nd LII edge (6.720 keV), Fe X-ray fluorescence (Kα ray) intensity, Nd X-ray fluorescence (Lα ray, Lβ ray) The strength was measured respectively.

Since the LII absorption edge excites Lβ rays, the XMCD intensity of Nd was derived from the Lβ rays. Since the intensity of the Lα ray has no significant correlation with the LII absorption edge, the elemental concentration of Nd was determined from the intensity of the Lα ray. Since the XMCD intensity depends on the intensity of the incident X-ray and the Nd element concentration, it was normalized by the intensity of the Lα ray. As a reference for the XMCD intensity of Fe and Nd, the XMCD intensity of Fe and Nd of a magnetized Nd—Fe—B sample (saturation magnetization is about 1.4 T) was measured. At this time, the background of the experimental apparatus was subtracted by changing the orientation of the magnetized sample and measuring the XMCD intensity when the magnetic moment of the sample was reversed by 180 degrees, and the net XMCD intensity of Fe and Nd was derived. As a result, the XMCD strength was 0.2% for Fe and 1.7% for Nd. FIG. 6 is an element selective demagnetization curve obtained by XMCD measurement. Here, the magnetic field dependence of the XMCD intensity of Nd and Fe at (a) room temperature and (b) 70 ° C. is shown. A demagnetization curve measured with a vibrating sample magnetometer (VSM) is also shown as bulk magnetization. The center position of the XMCD intensity is not adjusted because (a) it is difficult to discriminate at room temperature, and (b) the saturation magnetic field in the positive magnetic field direction (0T to 0.7T) and the negative magnetic field at 70 ° C. The average magnetization of each magnetic field (from −0.6 T to −0.7 T) in which the tendency of saturation was observed in the direction was taken, and the background was adjusted so that the center value became zero. In normalizing the values of the XMCD intensity and the magnetization, the saturation magnetization is considered to be the same, so that the average value between the magnetic field 0T and 0.7T is normalized to 1. A feature that does not depend on the measurement temperature is that the magnetization of the entire magnet and the magnetic field dependence of the XMCD intensity are clearly different. Since XMCD acquires magnetization information of about 4 μm to 10 μm from the magnet surface, it was found that the coercive force is different between the whole magnet and the magnet surface. It is considered that the coercive force is also lowered by the influence of the surface oxidation even in the region that enters the inside of about 4 to 5 μm from the magnet surface, that is, the region that enters the inside of about one crystal grain.

Next, as a feature depending on the measurement temperature, the coercive force of the entire magnet and the magnet surface tends to decrease with increasing temperature. The temperature coefficient of bulk coercivity can be estimated as -0.9% / ° C, and the temperature coefficient of surface coercivity can be estimated as -1.2 (± 0.1)% / ° C from the XMCD strength of Fe. It was found that the temperature coefficient of coercive force was larger. This indicates that the domain wall moves more easily on the magnet surface than on the inside. The reason why the magnetic field dependence of the XMCD intensity of Nd and Fe in (b) is slightly different is presumed to be due to a slight difference in the surface state of the measurement sample because different samples are measured in each measurement. From the above, it is found that the coercive force is reduced on the surface of the bulk magnet, and this appears remarkably in the shape of the demagnetization curve for a magnet of 1 mm or less. It is extremely important to improve the operating point to recover the coercive force drop on the surface of the bulk magnet having a small magnet thickness. Since this technique is the grain boundary diffusion of the DyFx solution from the magnet surface, it is a technique suitable for recovering the coercive force in the vicinity of the bulk magnet surface.

Claims (8)

Forming a core-shell structure in which fine crystal grains are mainly composed of two types of phases;
The main phase is composed of a rare earth transition metal alloy , and the second phase is composed by boriding, nitriding, oxidizing, or fluorinating the main phase,
The main phase is present on the shell side of the core-shell structure;
A permanent magnet material , wherein a thermal expansion coefficient of the main phase is larger than a thermal expansion coefficient of the second phase, and a magnetic field change of spontaneous volume magnetostriction of the main phase is positive . Forming a core-shell structure in which fine crystal grains are mainly composed of two types of phases;
The main phase is composed of a rare earth transition metal alloy, and the second phase is composed by boriding, nitriding, oxidizing, or fluorinating the main phase,
The main phase is present on the core side of the core-shell structure;
A permanent magnet material, wherein a thermal expansion coefficient of the main phase is smaller than a thermal expansion coefficient of the second phase, and a magnetic field change of the spontaneous volume magnetostriction of the main phase is negative. The permanent magnet material according to claim 1,
Permanent magnet material, wherein the pre-Symbol main phase is an Fe-based alloy or Fe-base compound. The permanent magnet material according to claim 2 ,
Permanent magnet material, characterized in that either before Symbol main phase is Ni base alloy, Co-based alloys, Ni-based compound, or a Co-based compound. The permanent magnet material according to any one of claims 1 to 4 ,
A permanent magnet material, wherein the fine crystal grains are not less than a superparamagnetic critical grain size and not more than 100 nm. The permanent magnet material according to any one of claims 1 to 5 ,
A permanent magnet material, wherein the phase transition of the magnetic order and crystal lattice strain occurs at the same temperature or the same magnetic field in the phase having the smaller thermal expansion coefficient of the main phase or the second phase. The permanent magnet material according to any one of claims 1 to 6 ,
A permanent magnet material, wherein at least one of the main phase and the second phase has a negative coefficient of thermal expansion. The permanent magnet material according to any one of claims 1 to 7 ,
A permanent magnet material, wherein at least one of B, C, N, O, or F elements has a concentration gradient in a boundary region between the main phase and the second phase.

Images