JP7727280B2 - Magnetic refrigeration material and its manufacturing method - Google Patents

Description

The present invention relates to a magnetic refrigeration material and a method for producing the same.

To realize a future hydrogen society that promotes energy conservation and a low-carbon society, it is essential to utilize liquid hydrogen, which has 1/800 the volume of gaseous hydrogen and features characteristics such as mass transportation, mass supply, mass storage, space-saving, and ultra-high purity. However, current hydrogen liquefaction technology using compressors has problems such as low liquefaction efficiency during production and loss due to evaporation. Magnetic refrigeration is a cooling technology that uses a magnetic material that exhibits the magnetocaloric effect as a refrigerant. A ferromagnetic-paramagnetic phase transition occurs through cycles of increasing and decreasing the magnetic field, and the resulting endothermic and exothermic reactions are used to refrigerate. Magnetic refrigerators are being developed as refrigerators that are more energy efficient and capable of liquefying hydrogen at low cost compared to refrigerators that use gas compression and expansion (see, for example, https://www.jst.go.jp/mirai/jp/program/large-scale-type/theme06.html).

Various magnetic refrigeration materials have been proposed in the past (for example, Patent Documents 1 to 3). Furthermore, in developing magnetic refrigeration materials, it is important to bring about a large magnetic entropy change when the magnetic field of the magnetic material is turned on and off near the target temperature (20 Kelvin for hydrogen liquefaction). To date, room-temperature magnetic refrigeration materials, such as La(FeSi) ₁₃ , have been developed. Furthermore, it is known that materials such as Laves phases (for example, ErCo ₂ ) have a large magnetic entropy change even near the hydrogen liquefaction temperature.

However, magnetic refrigeration materials can form hydrides depending on the environment in which they are used. Excessive hydriding typically embrittles the base metal, resulting in a deterioration of mechanical properties.

On the other hand, materials that have a slower hydrogen diffusion rate than metals, such as metal oxides such as Al ₂ O ₃ and Cr ₂ O ₃ , metal nitrides such as TiN and SiN, and metal carbides such as SiC, are known (for example, Non-Patent Document 1).

Japanese Patent Application Laid-Open No. 2008-214733 Japanese Patent Application Laid-Open No. 2007-262457 Japanese Patent Application Laid-Open No. 2006-265631

Vincenc Nemanic, “Hydrogen permeation barriers: Basic requirements, materials selection, position methods, and "quality evaluation", Nuclear Materials and Energy 19 (2019) 451-457

As mentioned above, various magnetic refrigeration materials have been proposed, but the behavior of these materials in their manufacturing and usage environments has not been thoroughly investigated. In particular, the chemical reactions between hydrogen and magnetic refrigeration materials, which are thought to come into frequent contact with hydrogen, are worthy of attention, but it cannot be said that sufficient knowledge has been gained.

The inventors conducted extensive research against this background and discovered that when hydrogenation progresses in magnetic refrigeration materials made of metals, many of them may shatter due to the volume expansion that occurs after hydride formation. They also discovered that hydrogenation significantly changes the magnetic properties, reducing their functionality as magnetic refrigeration materials.

Hydrogenation of metallic magnetic refrigeration materials leads to the formation of hydrides, which affect their mechanical and magnetic properties. To prevent this, a hydrogen barrier layer can be used. Possible hydrogen barrier materials include metals such as Mo and W, oxides such as Al ₂ O ₃ and Cr ₂ O ₃ , nitrides such as TiN and SiN, and carbides such as SiC.

We therefore discovered that forming a hydrogen barrier layer on the surface of the magnetic refrigeration material can effectively prevent hydrogenation. For example, by performing heat treatment in an oxygen atmosphere to form an oxide layer on the surface, it is possible to achieve a hydrogen barrier coating that has good adhesion to the underlying magnetic refrigeration material and is inexpensive and easy to process. This makes it easier to use Laves phase compounds, which have a large magnetocaloric effect but are easily hydrogenated, as magnetic refrigeration materials in a hydrogen atmosphere.

When magnetic refrigeration materials are used in magnetic refrigerators, they are often formed into particles (spheres) with a diameter of approximately 300 microns to maximize heat exchange with gases or liquids. Magnetic refrigeration materials with a large magnetocaloric effect, such as _ErCo2 , a Laves phase compound, are preferred. To improve their magnetic properties, they are preferably subjected to a long-term homogenization heat treatment, approximately one week, at approximately 800°C in an inert atmosphere or vacuum. If a sample after homogenization heat treatment is exposed to a hydrogen atmosphere, it may shatter within a short time due to hydrogenation, significantly changing its magnetic properties and potentially failing to function as a magnetic refrigeration material. Furthermore, uniformly depositing conventional hydrogen barrier materials onto particles (spheres) with a diameter of 300 microns and having gaps and irregularities on the surface is not only technically difficult, but also expensive, even if feasible.

Therefore, rather than attaching a hydrogen barrier to the surface, the inventors succeeded in forming a hydrogen barrier layer by treating the surface of the magnetic refrigeration material.

Such treatment can be carried out by heating in an oxidizing atmosphere. The oxidizing atmosphere may be, for example, an oxygen-containing atmosphere or an atmosphere containing a compound that releases oxygen. Heat treatment in such an oxidizing atmosphere is preferably performed after homogenization heat treatment. If homogenization heat treatment is performed after heat treatment in an oxidizing atmosphere containing oxygen, oxygen may diffuse further inward, potentially impairing the hydrogen barrier effect. It is preferable that the formed oxide layer be at least 10 nm thicker than the native oxide film. If it is too thin, a sufficient hydrogen barrier effect may not be achieved. On the other hand, if it is too thick, the magnetic properties as a magnetic refrigeration material may deteriorate. The volume ratio of the oxide layer to the entire material is preferably 3/4 or less.

After the oxide layer is formed, it is preferable to perform a heat treatment in an inert atmosphere or under vacuum at a temperature equal to or lower than the homogenization heat treatment temperature. This may further enhance the hydrogen barrier effect. The magnetic refrigeration material may be in the form of bulk, thin film, particles, fine particles, ultrafine particles, etc. For example, in a magnetic refrigeration material consisting of particles or spheres measuring 30 nm to 3000 μm, heat treatment in an oxygen atmosphere to form an oxide layer can form a uniform oxide layer all the way to the surface of the cavities inside the particles or spheres connected by gaps. When forming the particles or spheres, gaps or irregularities can form on the surface, or cavities can form inside. Therefore, it is not always easy to form a uniform oxide layer using dry or wet film formation methods such as sputtering or plating, and these methods are relatively expensive.

Specifically, this may include the following:
A magnetic refrigeration material comprising a Laves phase compound represented by _RT2 , wherein R is at least one rare earth element, and T is at least one element selected from the group consisting of Co, Ni, Al, and Fe, the thickness of a surface layer containing an oxide is 10 nm or more, and the ratio of the volume of the formed oxide layer to the total volume is 3/4 or less.
In any of the above-described magnetic refrigeration materials, R is at least one element selected from the group consisting of scandium, yttrium, lanthanum, cerium, praseodymium, neodymium, promethium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, and lutetium.
Any of the above magnetic refrigeration materials, wherein R is erbium.
Any of the above magnetic refrigeration materials, wherein T is cobalt.
The oxide layer may be thicker than a so-called native oxide film.
The oxide layer is _a magnetic refrigeration material containing a compound represented by _ErxCoyOz ( _x +y+z=1), where 0≦x≦0.6, 0<y≦0.6, and 0<z≦0.6. Here, 0≦x≦0.7 or 0≦x≦0.65 may be satisfied. 0<y≦0.7 or 0<y≦0.65 may be satisfied. 0<z≦0.7 or 0<z≦0.65 may be satisfied.
Any of the above magnetic refrigeration materials, wherein the shape of the magnetic refrigeration material is a sphere, an ellipsoid, or a shape that can be approximated to an oblate ellipsoid.
Any of the above magnetic refrigeration materials, wherein the diameter of the magnetic refrigeration material when its shape is approximated as a sphere is 60 nm or more and 6000 μm or less.
In any of the above magnetic refrigeration materials, the average particle size of each constituent particle, when approximated as a sphere as described above, is 60 nm or more and 6000 μm or less. The average particle size may be 100 nm or more, 200 nm or more, or 300 nm or more. The average particle size may be 6000 μm or less, 1000 μm or less, or 300 μm or less.
An AMR bed comprising any of the above magnetic refrigeration materials.
A magnetic refrigerator equipped with any of the above-described AMR beds.
Any of the above-described magnetic refrigeration devices further comprising a magnetic field application means for applying a magnetic field to the AMR bed, a refrigeration stage for cooling an object to be cooled by cold, and a heat exchanger for dissipating heat generated by the AMR bed.
A method for improving the hydrogen resistance of a magnetic refrigeration material containing a Laves phase compound represented by _RT2 , wherein R is at least one rare earth element and T is at least one element selected from the group consisting of Co, Ni, Al, and Fe, the method comprising the steps of: molding the magnetic refrigeration material; and forming a hydrogen barrier layer on a surface of the magnetic refrigeration material.
A method for improving the hydrogen resistance of a magnetic refrigeration material containing a Laves phase compound represented by _RT2 , wherein R is at least one rare earth element and T is at least one element selected from the group consisting of Co, Ni, Al, and Fe, the method comprising the steps of: molding the magnetic refrigeration material; homogenizing the molded magnetic refrigeration material; and forming a hydrogen barrier layer on the surface of the magnetic refrigeration material. The step of forming the hydrogen barrier layer may include a step of heat-treating the magnetic refrigeration material in a predetermined atmosphere. The predetermined atmosphere may include an oxidizing atmosphere. The oxidizing atmosphere may include an atmosphere containing oxygen or a compound capable of releasing active oxygen. In the oxidizing atmosphere containing oxygen, the partial pressure of oxygen is preferably 100 Pa or more, more preferably 1,000 Pa or more, and even more preferably 10,000 Pa or more. There is no particular upper limit to the partial pressure of oxygen, but a pressure that allows the formation of a desirable hydrogen barrier layer (which may include an oxide layer) is preferred, and may be, for example, approximately the partial pressure of oxygen at atmospheric pressure.
Any of the methods described above, further comprising the step of heat-treating the magnetic refrigeration material in an inert atmosphere after the step of forming the hydrogen barrier layer, wherein the step of forming the hydrogen barrier layer may include the step of forming an oxide layer.
The step of forming the hydrogen barrier layer may include a step of holding the molded magnetic refrigeration material in an oxidizing atmosphere within a predetermined temperature range. The predetermined temperature range may be 250°C or higher and 850°C or lower. The predetermined temperature range may be 150°C or higher, 200°C or higher, or 250°C or higher. The predetermined temperature range may be 1000°C or lower, 900°C or lower, or 850°C or lower.
A method for producing a magnetic refrigeration material, the method including a method for improving the hydrogen resistance of any of the magnetic refrigeration materials described above.

As described above, in the embodiments of the present invention, hydrogenation of the magnetic refrigeration material can be reduced, allowing the shape of the molded body of the magnetic refrigeration material to be maintained and its functionality as a magnetic refrigeration material to be maintained.

1 illustrates the crystal structure of _ErCo2, an example of a Laves phase compound that may comprise a magnetic refrigeration material usable in embodiments of the present invention. FIG. 1 is a diagram showing changes in hydrogen pressure when Sample 1 is sealed in hydrogen gas. FIG. 1 is a diagram showing the results of observation of grains of Sample 1 with an optical microscope before and after exposure to hydrogen. FIG. 10 is a diagram showing the temperature change of magnetization of Sample 1 before and after exposure to hydrogen. FIG. 10 is a diagram showing the change in magnetization of Sample 1 with respect to the magnetic field before and after exposure to hydrogen. FIG. 10 is a diagram showing changes in hydrogen pressure when sample 2 is sealed in hydrogen gas. FIG. 10 is a diagram showing the results of observation of grains of Sample 2 with an optical microscope before and after hydrogen exposure. FIG. 10 is a diagram showing changes in hydrogen pressure when sample 3 is sealed in hydrogen gas. FIG. 10 is a diagram showing the results of observation of grains of Sample 3 with an optical microscope before and after exposure to hydrogen. FIG. 1 shows EDS spectra of Samples 1 to 3. FIG. 10 is a diagram showing the change in magnetization of (1) to (4) with respect to a magnetic field in an embodiment of the present invention. FIG. 10 is a diagram showing the temperature change of magnetization of (1) to (4) in an example of the present invention. FIG. 10 is a diagram showing the temperature change of normalized magnetization of (1) to (4) in an example of the present invention. 1 shows a diagram of an analysis of a cross section of a particle of the magnetic refrigeration material of Sample 1. 1 shows a diagram of an analysis of a cross section of a particle of the magnetic refrigeration material of sample 2. 1 shows a diagram of an analysis of a cross section of a particle of the magnetic refrigeration material of sample 3. FIG. 1 is a diagram schematically illustrating a magnetic refrigeration device.

Embodiments of the present invention will be described below with reference to the drawings. Note that like elements are numbered likewise and their description will be omitted.

Figure 1 illustrates the crystal structure of _ErCo2 in an embodiment of the present invention. The structure is a Laves _phase MgCu2 type with lattice parameters a = b = c = 0.71536 nm and α = β = γ = 90°. In such a Laves phase, A and B metal elements with an atomic radius ratio of approximately 1.2:1 are bonded in an _AB2 composition ratio to form a compound. The crystal structure, consisting of large A atoms and small B atoms, can be thought of as a packing structure of large and small spheres, which occupy specific lattice positions, the A site and the B site. The A site has four A atoms and 12 B atoms as neighboring atoms, while the B site is surrounded by six A atoms and six B atoms. In a realistic Laves phase crystal, the atomic packing is such that A-A atoms and B-B atoms are in contact with each other, but there is no contact between A and B atoms. In such cases, the atomic radius ratio of the two atoms is RA/RB = √3/2 = 1.225. Generally, atoms located at the A site are arranged in a manner similar to that of a diamond structure, while atoms at the B site form tetrahedra around the A site. Because Laves phase compounds have a type of close-packed structure, they have three types of crystal structures: cubic MgCu _2- type (C15), hexagonal MgZn _2- type (C14), and MgNi _2- type (C36), due to differences in atomic stacking similar to the difference between a face-centered cubic lattice and a hexagonal close-packed lattice.

In _RAl2 (R = Ce, Pr, Nd, Sm, Gd, Tb, Dy, Ho, Er, Tm, Lu), there is clearly a ferromagnetic exchange interaction between the magnetic moments of the rare earth metals in _RAl2 , but as seen in _HoAl2 , the obtained magnetic moment is considered to be lower than the theoretical moment value gI of the trivalent ion of Ho.

For example, magnetic data have already been investigated for RNi ₂ (R = Pr, Nd, Sm, Gd, Tb, Dy, Ho, Er, Tm), RCo ₂ (R = Y, Ce, Pr, Nd, Sm, Gd, Tb, Dy, Ho, Er, Tm, Lu), and RFe ₂ (R = Ce, Sm, Gd, Tb, Dy, Ho, Er, Tm, Lu, Y) in RT ₂ (R = Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb).

Thus, it is clear that _ErCo2 , _ErNi2 , _ErAl2 , and _ErFe2 are not only _MgCu2- type Laves phase intermetallic compounds, but also function as magnetic refrigeration materials. Furthermore, the oxide layers on their surfaces can function as hydrogen barrier layers. In the following experimental examples, _ErCo2 will be described in particular, but those skilled in the art will understand that the same applies to any of these Laves phase intermetallic compounds.

[Experimental Example]
(Preparation of _ErCo2 )
As raw materials, lump erbium (3N, purity 99.9%) manufactured by Fujian Changting Golden Dragon Rare-Earth Co., Ltd. and lump cobalt (3N, purity 99.9%) manufactured by Sumitomo Metal Mining Co., Ltd. were weighed in a molar ratio to produce _ErCO2 and melted in an argon atmosphere in a high-frequency melting furnace to synthesize _ErCO2 . A cast ingot rod was produced from the synthesized _ErCO2 . Using a free-fall gas atomization device manufactured by Nisshin Giken Co., Ltd., this cast ingot rod was used as an electrode to high-frequency melt the tip to approximately 1400°C. Argon pressurized to 1-5 MPa was sprayed through a gas jet nozzle onto the resulting falling molten metal stream, stirring and atomizing the molten metal stream. This resulted in the formation of grains approximately 300 microns in diameter. The resulting particles, each approximately 300 microns in diameter, were wrapped in tantalum foil and placed in a stainless steel tube, which was then filled with argon gas at approximately 0.5 atmospheres. The tube was then placed in a muffle furnace and heated from room temperature at a rate of approximately 5°C/min. After reaching 850°C, the temperature was maintained for one week. The furnace was then turned off and cooled to room temperature. A portion of the sample was subjected to homogenization heat treatment at 850°C in an Ar atmosphere for one week and taken as Sample 1. The average diameter of the particles, approximated as a sphere, was approximately 300 μm. The average particle size can be determined using the method described below. The particle sizes ranged from approximately 212 μm to approximately 355 μm.

(Oxidation heat treatment)
The granular sample after the homogenization heat treatment was placed in an alumina crucible, the lid was closed with a small gap, and the sample was heated in an air atmosphere using a muffle furnace from room temperature to 500 ° C at a heating rate of about 25 ° C / min, and then held at 500 ° C for 30 minutes to perform an oxidation treatment. The furnace was then turned off, the temperature was cooled to 350 ° C at a rate of about 5 ° C / min, the muffle furnace door was opened, the crucible lid was closed, and the sample was cooled to 200 ° C or less at a rate of about 10 ° C / min. The covered crucible was removed from the furnace and left in room temperature air for 10 minutes, after which sample 2 was removed. The granular samples obtained in this way were bonded to each other at their contact points, but when lightly vibrated, they disintegrated and returned to their granular form.

(Vacuum heat treatment)
A portion of the granular sample (sample 2) after the oxidation heat treatment was wrapped in tantalum foil and placed in a one-sided sealed quartz tube, which was then evacuated and heated in an electric furnace at a rate of 6.7°C/min from room temperature to 530°C under a vacuum of approximately 3 x ^10-4 Pa, and maintained at this temperature for 5 hours (5-hour heat treatment at 500°C under vacuum). The power was then turned off, the sample was cooled to room temperature, the evacuation was stopped, air was introduced into the quartz tube, and the sample was removed. In this way, sample 3 was obtained.

(Hydrogen exposure test)
Equal amounts of Samples 1 to 3 were weighed and placed in separate containers of equal volume. They were filled with hydrogen gas at 12 atmospheres and sealed. The hydrogenation process was monitored by measuring the subsequent pressure change at room temperature. The results are shown in Figures 2, 6, and 8. In these figures, the upper plot represents hydrogen pressure, and the lower plot represents temperature. After sealing, observations were continued for up to seven days. For Sample 1, the hydrogen pressure significantly decreased from 1.28 MPa to 1.08 MPa in approximately one hour and then remained nearly constant (Figure 2). This indicates that hydrogenation was complete. In other words, Sample 1, which was subjected only to homogenization treatment, did not exhibit the so-called hydrogen barrier effect. The temperature at this time was between 24°C and 23°C. Furthermore, for Sample 2, there was no sudden decrease in hydrogen pressure within one hour after sealing. Instead, the hydrogen pressure gradually decreased over time, from an initial value of 1.28 MPa to 1.25 MPa (a 2.2% decrease in hydrogen pressure) after seven days (Figure 6). As a result, hydrogenation of Sample 2 progressed gradually in proportion to time, and hydrogenation of approximately 14% of the amount of reacted hydrogen of Sample 1 was observed over seven days. The temperature at this time was 23.5°C to 22.5°C. In Sample 3, no decrease in hydrogen pressure was observed for two days after sealing, and then it gradually decreased, from 1.27 MPa to 1.25 MPa after seven days (a 2.0% decrease in hydrogen pressure) (Figure 8). Furthermore, the rate of decrease in hydrogen pressure became particularly slow from around the sixth day. Thus, although hydrogenation of Sample 3 progressed gradually over time, the reaction tended to saturate, and hydrogenation of approximately 13% of the amount of reacted hydrogen of Sample 1 was observed over seven days.

(Optical microscope observation)
The ErCo ₂ grains of Samples 1 to 3 were observed under an optical microscope before and after hydrogen exposure (Figures 3, 7, and 9). As can be seen from Figure 3, (A) grains were clearly visible before the test, but (B) almost all of the grains were broken into pieces after the test. This is thought to be due to the grains being broken into pieces due to volume expansion caused by hydrogenation after hydrogen exposure. Even without measuring the magnetic properties, it is clear that such broken grains are unsuitable for magnetic refrigeration materials. As can be seen from Figures 7 and 8, in Samples 2 and 3, (A) most of the grains remained intact before the test, while (B) only some of the grains were broken into pieces after the test. This indicates that volume expansion due to hydrogenation was suppressed after hydrogen exposure. This indicates that Samples 2 and 3 have a clear hydrogen barrier effect, and that the hydrogen barrier effect is greater in Sample 3 than in Sample 2.

(Evaluation of magnetic properties 1)
The magnetic properties of Sample 1 described above were evaluated before and after hydrogen exposure. The results are shown in Figures 4 and 5. In these figures, Sample 1 before hydrogen exposure is labeled "Ar-850°C-1w," and Sample 1 after hydrogen exposure is labeled "H ₂ -exposed." Figure 4 shows that Sample 1 before hydrogen exposure exhibited a significant change in magnetization at approximately 30 K, indicating excellent magnetic properties. On the other hand, Sample 1 after hydrogen exposure did not exhibit such properties, indicating that it is not suitable as a magnetic refrigeration material. Figure 5 also shows that Sample 1 before hydrogen exposure exhibited higher magnetization and superior magnetic properties from low to high magnetic fields compared to Sample 1 after hydrogen exposure. Furthermore, Sample 1's magnetization characteristics and their temperature dependence changed significantly before and after hydrogen exposure, indicating that hydrogenation makes it difficult for it to function as a magnetic refrigeration material. Samples 2 and 3 did not show any significant changes in their magnetic properties before and after hydrogen exposure, indicating that they function as magnetic refrigeration materials even after hydrogen exposure. The magnitude relationship of ΔS (magnetic entropy change) per unit volume, which indicates the performance as a magnetic refrigeration material, is estimated to be Sample 1 > Sample 2 > Sample 3, based on the temperature dependence of the magnetization characteristics. This is presumed to reflect the decrease in volume of _ErCo2 due to oxide formation, etc. Magnetic refrigeration materials are typically used in temperatures below liquid nitrogen. In magnetic refrigeration systems that undergo maintenance at room temperature for approximately one day once a year, exposure to hydrogen at room temperature during this period is thought to be the most severe environment for hydrogen barrier coating materials. However, magnetic refrigeration materials such as those in the examples are estimated to be able to withstand several years of maintenance.

(Evaluation of magnetic properties 2)
Table 1 summarizes the details of the samples whose magnetic properties were evaluated before and after hydrogen exposure.
(1) was a sample that was held at 850°C in an Ar atmosphere for one week (homogenization heat treatment) and then held at 500°C in an air atmosphere for 30 minutes (oxidation heat treatment). (2) was a sample that was held at 850°C in an Ar atmosphere for one week (homogenization heat treatment) and then held at 500°C in an air atmosphere for 30 minutes (oxidation heat treatment), followed by hydrogen exposure. (3) was a sample that was held at 850°C in an Ar atmosphere for one week (homogenization heat treatment), then held at 500°C in an air atmosphere for 30 minutes (oxidation heat treatment), followed by 530°C in a vacuum for 5 hours (vacuum heat treatment). (4) was a sample that was held at 850°C in an Ar atmosphere for one week (homogenization heat treatment), then held at 500°C in an air atmosphere for 30 minutes (oxidation heat treatment), followed by 530°C in a vacuum for 5 hours (vacuum heat treatment), followed by hydrogen exposure.

Figures 11 to 13 show the results of evaluating the magnetic properties of the above-mentioned (1) to (4). Figure 11 plots magnetization versus magnetic field, and it can be seen that there is no significant difference among (1) to (4). Figure 12 plots magnetization versus temperature, and Figure 13 normalizes the magnetization value at 60 K. Figure 12 shows that (1) has the best magnetic properties. Furthermore, comparing (1) and (2) and (3) and (4) before and after hydrogen exposure, it appears that the magnetic properties deteriorated due to hydrogen exposure. Next, in Figure 13, which is normalized, a similar comparison shows almost no difference between before and after hydrogen exposure. Furthermore, the magnitude relationship of ΔS predicted from the magnetization jump at the Curie temperature (up to 32 K) is thought to be as follows:
ΔS (Ar-850℃-1w) > ΔS ((1) & (2)) > ΔS ((3) & (4))
This suggests that the decrease in volume of _ErCo2 due to oxidation and the generation of _ErCo3 may have had an effect.

(Grain surface composition)
Figure 10 shows the EDS spectra of Samples 1 to 3. Sample 1 underwent homogenization heat treatment but not oxidation heat treatment, resulting in a small oxygen (O) signal (see Figure 10A). On the other hand, Sample 2 underwent oxidation heat treatment, resulting in a significant increase in the oxygen signal, which is believed to reflect the oxidation of the magnetic refrigeration material. At the same time, the erbium (Er) signal was significantly reduced. This suggests a change in the composition of the surface layer of the magnetic refrigeration material. Considering the blackening of the sample after oxidation, it is highly likely that Co oxides formed on the surface layer of the magnetic refrigeration material. Cobalt oxide is believed to function as a hydrogen barrier (see Figure 10B). Furthermore, in Sample 3, which underwent vacuum heat treatment, the intensity of the oxygen (O) signal decreased, but the intensity of the erbium (Er) signal did not recover (see Figure 10C). Since the sample surfaces turned grayish white, it is speculated that sample 2 may be Co ₃ O ₄ or Co ₂ O ₃ and sample 3 may be CoO.

(Cross-sectional structure of grains)
Figures 14 to 16 show cross-sectional analyses of each particle of the magnetic refrigeration materials of Samples 1 to 3. In Figure 14, no oxidative heat treatment was performed, suggesting the presence of a so-called native oxide film on the surface. In fact, when measuring the signal intensities of Er, Co, and O elements from the surface, as shown in the graph at the bottom right, a small O peak is observed on the outermost surface, while the Co intensity is relatively low, suggesting the formation of erbium oxide. Figure 15 shows a cross-sectional view of Sample 2, which underwent oxidative heat treatment. Elemental analysis revealed an oxide layer of less than several microns, primarily composed of Co oxide, on the outermost surface. Beneath this, a 10-15 micron-thick quasi-oxide layer, primarily composed of Er _x Co _y O _{z ,} was observed. Figure 16 shows a cross-sectional view of Sample 3, which underwent oxidative heat treatment followed by vacuum heat treatment. Elemental analysis revealed an oxide layer of less than several microns, primarily composed of Co oxide, on the outermost surface. Below this, a 20-30 micron thick sub-oxide layer consisting mainly of Er _x Co _y O ₂ was observed.

(oxide layer)
In the examples of the present invention, as described with reference to FIGS. 15 and 16 , the surface and surface oxide layer can be easily observed by cross-sectional observation. For example, as described with reference to FIG. 14 , by scanning the cross-section with an electron gun or the like and continuously measuring the intensity of an element, the change in the concentration of that element can be determined. In this case, if the scanning speed is constant, the time is proportional to the scanning distance, and the change in the detected concentration over time can be determined as a change in distance. For example, elements such as Er, Co, and O can be investigated. For example, the detected concentration of oxygen (O) may be low and nearly constant inside the grain, but may suddenly increase toward the surface, reach a certain constant value, and then suddenly decrease at the surface. The midpoint between the point where the detected concentration suddenly increases and reaches a certain constant value can be determined as the boundary of the oxide layer. The distance from the surface to this boundary can then be used to determine the thickness of the oxide layer. Generally, multiple observation fields are set up, and the oxide layer thickness is measured at at least 10 locations in each field, and the arithmetic average is then used to determine the thickness of the oxide layer on the surface of the sample grain.

Here, if the radius of a particle when approximated as a sphere is R and the thickness of the oxide layer is L, the volume of the particle is 4/3πR ³ and the volume of the oxide layer is 4/3π{R ³ - (R - L) ³ }, and if the volume ratio is 3/4 or less, then L≦(1-0.25 ^(1/3) )R. R≧27 nm may satisfy L≧10 nm. Since the maximum value of L increases with increasing R, the substantial maximum value of L may be (1-0.25 ^(1/3) )R. In general, the penetration amount is considered to be inversely proportional to the penetration distance, which corresponds to the thickness of the so-called barrier layer. Therefore, if the volume of the oxide layer is 3/4 or less of the volume of the entire particle, its thickness may be 10 nm or more, 100 nm or more, or 1000 nm or more as R increases. On the other hand, in consideration of the deterioration of magnetic properties, it is preferable that the thickness of the oxide layer is as thin as possible within a range in which the desired hydrogen barrier effect can be obtained, but if it is too thin, the hydrogen barrier function decreases, and if it is 10 nm or less, the effect is low.

Figure 17 is a schematic diagram showing the main components of a magnetic refrigeration device. Magnetic refrigeration materials, including those described in the above-described embodiments, can be used. One form of this magnetic refrigeration material is particles with a particle diameter ranging from 50 μm to 1000 μm. For example, the particles may be spherically approximating particles with diameters of 50 μm to 100 μm, 200 μm to 2000 μm, 1000 μm to 500 μm. These upper and lower limits may also be appropriately combined to create a specified range. The particle form can increase the filling rate of the AMR bed, and the particle size can change the heat exchange cross-sectional area and pressure loss with the heat transport refrigerant. Smaller particle sizes increase the heat exchange cross-sectional area, which is effective in improving refrigeration performance. However, smaller particle sizes also increase pressure loss, resulting in reduced refrigeration performance. The actual pressure loss depends not only on the particle size but also on the type of heat transport refrigerant and operating conditions. Here, particle size refers to the volume-based median diameter (d50), and the volume-based average particle size can be measured, for example, using a microtrack or laser scattering method. More specifically, static image analysis and dynamic image analysis can be used. The former involves taking multiple particle images (e.g., SEM images) and using image analysis software to determine the particle diameter, converted into a circular shape, from the area of each particle. A magnetic refrigeration device 200 equipped with such a magnetic refrigeration material can be used to generate ultra-low temperatures, such as hydrogen liquefaction. The magnetic refrigeration device 200 further includes an AMR bed 220 filled with magnetic refrigeration material 210, a magnetic field application means 230 that applies a magnetic field to the bed, a cooling stage 290 that cools the object to be cooled by cold, and a heat exchanger 240 that dissipates heat generated by the magnetic refrigeration work in the AMR bed 220.

The magnetic field application means 230 can be any means for applying a magnetic field to the AMR bed 220; for example, it is practical to use a magnetic field with a strength of approximately 1 to 10 T (tesla). The magnetic field application means 230 can be a superconducting magnet, a permanent magnet, or the like. Furthermore, the relative positions of the magnetic field application means 230 and the AMR bed 220 can be changed using a drive mechanism (not shown), thereby changing the magnitude of the magnetic field applied to the AMR bed 220.

A pre-cooling stage 260 is provided on the high-temperature side of the AMR bed 220, with an 80K shield 270 connected to the low-temperature side of the pre-cooling stage 260 and a 300K shield 280 connected to the high-temperature side of the pre-cooling stage 260. Furthermore, a cooling stage 290 is provided on the low-temperature side of the AMR bed 220, with a liquefaction vessel 250 thermally connected to the cooling stage 290. In other words, gas to be cooled is supplied to the liquefaction vessel 250 and liquefied. The AMR bed 220 also has inlet and outlet ports for the heat transport refrigerant, allowing the heat transport refrigerant to flow back and forth within the AMR bed 220 through the gaps in the magnetic refrigeration material 210.

The liquefaction vessel 250 is supplied with the gas 310 to be liquefied (e.g., hydrogen, helium (He), etc.) from a tank (not shown). The magnetic refrigeration device 200 may operate as follows: A magnetic field is applied to the AMR bed 220 filled with the magnetic refrigeration material 210 by the magnetic field application means 230, raising the temperature of the magnetic refrigeration material 210. Next, the heat transport refrigerant is caused to flow in a direction 300A from the low-temperature end to the high-temperature end of the AMR bed 220. The heat transport refrigerant exchanges heat with the magnetic refrigeration material 210 filled inside the AMR bed 220, receiving warmth, while flowing through the gaps in the magnetic refrigeration material 210 and flowing out from the high-temperature end of the AMR bed 220. The heat transport refrigerant flowing out from the high-temperature end of the AMR bed 220 flows through the pre-cooling stage 260 into the heat exchanger 240, which releases the warmth, and excess heat is released to the outside. Next, the magnetic field in which the magnetic refrigeration material 210 is filled is removed (reduced), causing the temperature of the magnetic refrigeration material 210 to drop.

The heat transport refrigerant is then caused to flow in direction 300B from the high-temperature end of the AMR bed 220 toward the low-temperature end. The heat transport refrigerant flows into the high-temperature end of the AMR bed 220 via the pre-cooling stage 260, and while being cooled by heat exchange with the magnetic refrigeration material 210 filled inside, flows through the gaps in the magnetic refrigeration material 210 and reaches the low-temperature end of the AMR bed 220. The flow of the heat transport refrigerant is driven by a refrigerant driving means (not shown). The refrigerant driving means is not particularly limited as long as it can drive an oscillating flow that causes the heat transport refrigerant to flow back and forth in synchronization with the AMR cycle, and examples include a system that combines a piston, a blower and a valve.

When the temperature of the low-temperature end of the AMR bed 220 drops below the boiling point of liquid hydrogen (20 K at atmospheric pressure), the hydrogen gas supplied to the liquefaction vessel 250 is cooled and concentrated/liquefied by heat exchange with the cooling stage 290 installed on the low-temperature end side of the AMR bed 220. This process is repeated, and the gas inside the liquefaction vessel 250 is periodically liquefied or cooled.

In the embodiments of the present invention, the magnetic refrigeration material is resistant to hydrogenation and can withstand long-term use. Such magnetic refrigeration materials are used in magnetic refrigeration devices and function effectively in hydrogen liquefaction, etc. This can contribute to the widespread use of hydrogen, which is a promising energy carrier.

200 Magnetic refrigeration device 220 AMR bed 230 Magnetic field application means 240 Heat exchanger 250 Liquefaction vessel 260 Pre-cooling stage 270 80K shield 280 300K shield 290 Cooling stage 300A Movement direction of heat transport refrigerant 300B Movement direction of heat transport refrigerant

Claims (9)

is a Laves phase compound represented by RT ₂ ,
R is at least one rare earth element,
T is at least one element selected from the group consisting of Co, Ni, Al, and Fe,
The thickness of the surface layer containing oxide is 10 nm or more,
A magnetic refrigeration material characterized in that the ratio of the volume of the formed oxide layer to the total volume is 3/4 or less. The magnetic refrigeration material of claim 1, wherein R is at least one element selected from the group consisting of scandium, yttrium, lanthanum, cerium, praseodymium, neodymium, promethium, samarium, europium, gadolinium, terbium, dysprosium, holmium, erbium, thulium, ytterbium, and lutetium. The magnetic refrigeration material described in claim 1 or 2, characterized in that, when approximated as a sphere, the diameter is 30 nm or more and 3000 μm or less. An AMR bed comprising the magnetic refrigeration material described in any one of claims 1 to 3. A magnetic refrigeration apparatus equipped with an AMR bed, wherein the AMR bed is the AMR bed described in claim 4. The magnetic refrigeration device of claim 5 further comprises a magnetic field application means for applying a magnetic field to the AMR bed, a refrigeration stage for cooling the object to be cooled by cold, and a heat exchanger for dissipating heat generated by the AMR bed. A method for improving the hydrogen resistance of a magnetic refrigeration material containing a Laves phase compound represented by RT ₂ , comprising:
R is at least one rare earth element,
T is at least one element selected from the group consisting of Co, Ni, Al, and Fe,
forming a magnetic refrigeration material;
a step of homogenizing the formed magnetic refrigeration material;
and a step of subjecting the magnetic refrigeration material after the homogenization treatment to an oxidation treatment. The method of claim 7, further comprising a step of heat-treating the magnetic refrigeration material in an inert atmosphere after the oxidation treatment step. A method for producing a magnetic refrigeration material, comprising a method for improving the hydrogen resistance of the magnetic refrigeration material described in claim 7 or 8.