Magnetocaloric particles of the Laves phase compound HoAl₂ prepared by electrode induction melting gas atomization

Takafumi D. Yamamoto [email protected] Hiroyuki Takeya Akiko T. Saito Kensei Terashima Takenori Numazawa Yoshihiko Takano National Institute for Materials Science, Tsukuba, Ibaraki 305-0047, Japan University of Tsukuba, Tsukuba, Ibaraki 305-8577, Japan

Abstract

Processing magnetocaloric materials into magnetic refrigerant particles is an essential issue in developing high-performance magnetic refrigerators. Here, we succeed in stably producing magnetocaloric particles of the promising material HoAl₂ by a newly devised method based on electrode induction melting gas atomization process. The particle size range is on the order of submillimeters, which is suitable for practical refrigeration systems. The resulting particles with less contamination have good morphological, magnetic, and magnetocaloric properties: (i) almost spherical shapes with few internal pores, (ii) a sharp ferromagnetic transition around 30 K, and (iii) a large magnetocaloric effect comparable to the bulk counterpart. These features suggest the HoAl₂ gas-atomized particles have the potential of use as a magnetic refrigerant. The presented method can be applied not only to HoAl₂ but also to other brittle magnetocaloric materials with high melting points, facilitating the production of various magnetic refrigerants needed to develop magnetic refrigerators for hydrogen liquefaction.

keywords:

Magnetocaloric effect, Intermetallic compounds, Gas-atomization, Thermodynamic properties

^†^†journal: Journal of Magnetism and Magnetic Materials

1 Introduction

Magnetocaloric effect (MCE) is a magneto-thermodynamic phenomenon in which a magnetic material absorbs/generates heat when exposed to a varying magnetic field. The magnetic refrigeration based on the phenomenon has attracted attention as a promising alternative to conventional gas compression/expansion refrigerations, owing to the advantages of being highly efficient, environmentally friendly, and feasible over a wide temperature range from extremely low temperatures to room temperature [1, 2, 3, 4]. Recently, the need for lowering any costs related with liquid hydrogen, a potential medium for economical transportation and storage [5, 6], has increased toward realizing a future society using hydrogen energy. In this context, great interest has been focused on the development of high-performance magnetic refrigeration systems for hydrogen liquefaction [7, 8], which has driven extensive research on exploring for new magnetocaloric materials working at cryogenic temperatures [8, 9, 10, 11, 12] and developing highly efficient refrigerators based on active magnetic regenerator (AMR) cycles [13, 14, 15].

The Laves phase compounds RT₂ (R = Gd-Yb, T = Al, Ni, Co) exhibit large MCEs associated with a ferromagnetic transition occurring in the temperature range between the two liquefaction temperatures of hydrogen (20 K) and nitrogen (77 K) [16, 17, 18]. Moreover, their Curie temperature ( $T_{\rm C}$ ) is easy to tune by element substitutions. Such features make these compounds to be one of the candidate materials for magnetic refrigerants used in magnetic liquefiers for hydrogen. On the other hand, practical use of promising magnetocaloric materials in AMR systems requires them to be processed into spherical particles with a size of 200-400 $\mu$ m for optimal system performance [19, 20].

The material properties of the RT₂ compounds may hinder producing spheres by some conventional spheroidization processes. For example, they have high melting points above 1000 ^∘C and high reactivity. In the processes that melt and/or superheat a target material in a crucible, such as the centrifugal atomization method [21, 22, 23] and the gas atomization process [24, 25], these properties may cause the reaction of the target material and the crucible, preventing the mass production of high-quality particles. The plasma rotating electrode process [26, 27] can produce high-purity spherical particles by non-contact melting of a material’s electrode rod that rotates at high speed. Nevertheless, the brittle nature of RT₂ compounds can make it challenging to fabricate electrode rods that can withstand for high-speed rotations.

As a viable route for producing magnetic refrigerant particles, we have focused on the electrode induction melting gas atomization (EIGA) technique [28], in which an electrode rod of a target material is inductively melted in a non-contact manner. Very recently, we actually succeeded in fabricating spherical particles of a giant magnetocaloric material, HoB₂, with a melting point of 2350 ^∘C [29]. In the EIGA process, ingots are typically used as electrode rods. However, since the RT₂ compounds are pretty sensitive to thermal shock, the breakdown of the ingot can occur during induction heating. In this study, we devise a new method to use a sintered body as the electrode rod in the EIGA process and realize stable production of spherical particles of HoAl₂, one of the RT₂ compounds with the $T_{\rm C}$ of about 30 K [16] and a melting point of 1530 ^∘C. The resulting particles are evaluated in terms of morphological, crystallographic, and physical properties.

2 Material and methods

Master alloy of HoAl₂ was prepared by high-frequency vacuum melting from a stoichiometric mixture of Ho (3N) and Al (4N) elements. The alloy was crushed into powder, filled into a quartz tube, and sintered in a vacuum at 1000-1200 ^∘C for 20-24 h. The resulting sintered rods were 400-500 g in weight for each. Figure 1(a) depicts a schematic image of the present EIGA process. In this process, one end of the sintered rod was fixed with a chuck and hung, and the other end was inductively melted by an induction coil in an Ar-atmosphere. The molten metal freely fell into the lower sample chamber, atomized into droplets by jetting Ar-gas, and then solidified into particles while flying in the chamber.

After the EIGA process, the collected atomized powder was sieved into six fractions, $<$ 100 $\mu$ m, 100-212 $\mu$ m, 212-355 $\mu$ m, 355-500 $\mu$ m, 500-710 $\mu$ m, and $>$ 710 $\mu$ m, through a series of JIS Z 8801 standard sieves with a FRITSCH ANALYSETTE 3 vibratory sieve shaker. Since the sieved powder was composed of round particles and irregular ones, the former were manually separated from the latter by rolling them on a sloping board, though the particles less than 100 $\mu$ m in diameter could not be separated well. No heat treatment was not applied to both the master alloy and the atomized particles.

The particle morphology was observed using a Hitachi SU-70 scanning electron microscope (SEM) operated at 20 kV. The energy-dispersive X-ray spectroscopy (EDS) analysis was performed to observe element distribution of the particles using a Hitachi TM4000 operated at 10 kV. The particles were embedded in bakelite resin with carbon filler for polishing to a flat surface prior to cross-section observation. Powder X-ray diffraction (XRD) measurements were carried out at room temperature by a Rigaku MiniFlex600 diffractometer (Cu K $\alpha$ radiation).

The two magnetocaloric properties, namely, the magnetic entropy change $\Delta S_{\rm M}$ and the adiabatic temperature change $\Delta T_{\rm ad}$ were evaluated from the magnetization and specific heat data for the round particles and the master alloy. The details of our evaluation method for $\Delta S_{\rm M}$ and $\Delta T_{\rm ad}$ were described elsewhere [30, 31]. Temperature dependence of magnetization ( $M$ - $T$ curve) was measured by a Quantum Design SQUID magnetometer from 60 to 2 K under various magnetic fields ( $\mu_{0}H$ ) ranging from 0.01 to 5 T. Field dependence of magnetization ( $M$ - $H$ curve) was taken at 2 K between 0 T and 5 T. Specific heat ( $C$ ) was measured at 0 T between 2 K and 65 K by a thermal relaxation method using a Quantum Design PPMS. In the magnetization measurements, the master alloy had a rectangular shape with dimensions of 1.5 $\times$ 0.3 $\times$ 0.3 mm³, and dozens of the round particles were arranged vertically long in contact with each other with an aspect ratio of 2-4. The magnetic field was applied along the longitudinal direction of both the shaped master alloy and the arranged particles. The magnetization data in this study were not corrected for the demagnetization effect because it is difficult to determine the exact demagnetization factor for the arranged particles.

3 Results and discussion

3.1 Particle size distribution

Figure 1(b) shows the yield of HoAl₂ gas-atomized particles as a function of particle size. Here the yield was calculated as the ratio of the weight of the round particles for each diameter to the total weight of the collected atomized powder. The present EIGA process was able to run many times, so the average values and the corresponding standard deviations are presented together. The particle size is distributed in the range of 100-710 $\mu$ m, roughly centered around 212-355 $\mu$ m, for which the yield of 14-22 wt% is obtained. This result contrasts with that the particle size is typically 100 $\mu$ m or less in conventional EIGA experiments [32, 33, 34, 35, 36]. The atomizing gas pressure values in this study, 1.5-3.5 MPa, are lower than those used in these previous studies. The low gas pressure means that the kinetic energy of the atomizing gas is low, which gives the molten metal a moderate breakup effect [34]. This should have provided the particle size range on the order of submillimeters that meets the requirements of AMR systems.

3.2 Morphological and microstructural properties

Figure 2(a) shows an SEM image of the surface for HoAl₂ particles with 212-355 $\mu$ m diameter. The majority of the particles are almost spherical, but some are fractured or have small hollows on the surface. The latter is probably due to particle/particle and particle/chamber collisions. It is worth noting here that the particles have few internal pores, as seen from the cross-sectional image in Fig. 2(b). The internal pore is a well-known significant issue that causes degradation of gas-atomized particles and results from the atomizing gas trapped inside the droplet. These pores are more often found in particles larger than 100 $\mu$ m [35], whereas this is not the case in the HoAl₂ particles. According to the consideration based on fluid mechanics [36, 37], the higher the atomizing gas pressure, the higher the probability of gas-trapping. Therefore, the low atomizing gas pressure is also likely to help prevent the formation of internal pores.

Figures 2(c) and 2(d) show a magnified cross-sectional backscattering electron image of the particle and EDS spectrum taken at several positions. One can find three characteristic areas: the major light gray area (such as position 1 and 2), the dark gray area (such as position 3 and 4), and the white spot (such as position 5). Compared to the light gray area, the dark gray area has a stronger relative peak intensity of Al to Ho in the EDS spectrum, implying that this area corresponds to HoAl₃. In the white spot area, the intensity of Al is weak and that of O is strong, indicating the presence of Ho₂O₃. Note that the microcracks observed in the dark gray areas may have occurred when the particles were polished.

3.3 Crystallographic properties

The upper pannel of Fig. 3 shows XRD patterns of HoAl₂ spherical particles with various diameters between 100 $\mu$ m and 710 $\mu$ m. The data for the master alloy is shown at the bottom for comparison. The calculated patterns of some materials including HoAl₂ are shown in the lower panel. The primary phase is indexed as HoAl₂ [38] in both the master alloy and the atomized particles. Moreover, there are no differences in position and sharpness of diffraction peaks, suggesting that the atomization process does not affect the crystallographic properties of HoAl₂. The intensity of impurity peaks does not increase much before and after the atomization process, which implies less contamination during a series of processes. However, the type of impurities is different between the two samples; in addition to Ho₂O₃ [39], which is common to both, the master alloy has trigonal HoAl₃ [40], and the atomized particles has cubic HoAl₃ [41].

The intensity of Ho₂O₃ peaks tends to increase with increasing particle size. In general, oxidization is facilitated at high temperatures. Given that the larger particles need longer cooling times, this result implies that the gas-atomized particles take up oxygen during solidification. Note that the HoAl₂ particles do not oxidize under ambient conditions. It is interesting that HoAl₃ changes its crystal structure through the atomization process. Cannon and Hall [41] have reported that such transformation occurs at high temperatures ( $\sim$ 1000 ^∘C) and high pressures (6.4 GPa), but this is far from the present experimental conditions. One possibility is that the rapid solidification of the atomized particles causes the transformation. Indeed, the cubic HoAl₃ peaks are prominent in smaller particles, in which the solidification time may be shorter. The experiments are underway to verify the quenching effect on trigonal HoAl₃.

3.4 Effects of atomization process on the magnetism

Figure 4(a) shows the temperature dependence of specific heat at 0 T in HoAl₂ particles with 212-355 $\mu$ m diameter and the master alloy. Both samples exhibit a specific heat peak around 30 K, ascribed to the paramagnetic-ferromagnetic transition in HoAl₂. The peak in the particles is as clear as that in the master alloy, suggesting that the ferromagnetic transition in the former is sharp as well as the latter. In addition, as shown in Figs. 4(b) and 4(c), the $M$ - $H$ curves at 2 K and the $M$ - $T$ curves under 0.01 T are similar for both the atomized particles and the mater alloy. These results suggest that the HoAl₂ gas-atomized particles have good sample quality comparable to the bulk counterpart. Note that the difference in magnitude of $M$ at low magnetic fields is due to that in demagnetization factor between the master alloy and the particles.

The specific heat peak is shifted to higher temperatures in the atomized particles, which is indicative of the increase of $T_{\rm C}$ . Here we define $T_{\rm C}$ as a peak temperature of the temperature derivative of magnetization (d $M$ /d $T$ ) at 0.01 T. As can be seen from d $M$ /d $T$ in Fig. 4(c), the $T_{\rm C}$ is evaluated to be 28 K for the master alloy and 30 K for all the particles with various diameters. Considering that the X-ray diffraction peaks of HoAl₂ does not change before and after the atomization and the $T_{\rm C}$ is independent of particle diameter, or equivalently, solidification time, the structural properties are unlikely to be responsible for the $T_{\rm C}$ shift. Rather, the difference in $T_{\rm C}$ between the atomized particles and the master alloy seems to be attributed to whether the crystal structure of HoAl₃ is trigonal or cubic. Trigonal HoAl₃ has been reported as an antiferromagnet with a N $\acute{\rm e}$ el temperature of 9 K [43], while the magnetism of cubic HoAl₃ has not been clarified. The difference in their magnetic properties might affect the physical properties of HoAl₂, but the further investigations are needed.

One can find another anomaly around 20 K in both $C$ and $M$ . This is ascribed to a spin-reorientation transition [44, 45], in which the easy axis of magnetization changes from <100> direction above 20 K to <110> direction below 20 K. The anomaly looks similar in all the samples, implying that this transition is largely unaffected by the atomization process. This result seems to be consistent with the absence of changes in the crystallographic properties of HoA₂ by the atomization process, as strong magneto-elastic coupling has been observed associated with the spin-reorientation transition [46].

3.5 Magnetocaloric properties

Here we will evaluate the magnetocaloric properties for HoAl₂ spherical particles with 212-355 $\mu$ m diameter and the master alloy. Figures 5(a) and 5(b) show the magnetic entropy change $\Delta S_{\rm M}$ and the adiabatic temperature change $\Delta T_{\rm ad}$ for the magnetic field change $\mu_{0}\Delta H=$ 1, 3, and 5 T, where the initial magnetic field was set to zero. Both the $\Delta S_{\rm M}$ and the $\Delta T_{\rm ad}$ for each $\mu_{0}\Delta H$ take a maximum value at around $T_{\rm C}$ . For the spherical particles, the maximum values for $\mu_{0}\Delta H=$ 5 T are 23.9 J kg^-1 K^-1 and 9.1 K, respectively, which are close to those for the master alloy (25.7 J kg^-1 K^-1 and 9.7 K). The values of $\Delta S_{\rm M}$ are consistent with those reported in the literatures [47, 48]. Surprisingly, to our best knowledge, this is the first report on $\Delta T_{\rm ad}$ for $\mu_{0}\Delta H=$ 5 T in HoAl₂. Its maximum value is comparable to the reported value of 10 K for ErCo₂ [49], a well-known giant magnetocaloric material with a similar $T_{\rm C}$ of 32 K. For further comparisons, see the table in Ref. [12] summarizing magnetocaloric properties of various materials with $T_{\rm C}$ below 30 K.

It would be noticed that the shape of $\Delta S_{\rm M}$ and $\Delta T_{\rm ad}$ curves below $T_{\rm C}$ is a bit different between the spherical particles and the master alloy. Anisotropic MCE has been reported in HoAl₂ single crystal in association with the spin-reorientation transition [50]. In this context, Gil et al. [51] have pointed out that the texture effects in polycrystalline samples can affect the shape of $\Delta S_{\rm M}$ especially for $T\leq$ 20 K. Accordingly, sample dependence possibly appears, but it will not be a serious problem for the performance of the HoAl₂ gas-atomized particles as a magnetic refrigerant.

4 Conclusions

The electrode induction melting gas atomization process was successfully applied for fabricating magnetocaloric particles of the Laves phase compound HoAl₂ by using the sintered rod as the electrode rod. The particle size range of the resulting particles is suitable for AMR systems, with the highest yield of 14-22 wt% of the total atomized powder for 212-355 $\mu$ m diameter. These gas-atomized particles have mostly spherical shapes and few internal pores and are also less contaminated during processing. Furthermore, they exhibit the sharp ferromagnetic transition and large magnetocaloric properties, similar to the bulk counterpart.

All these results indicate that the HoAl₂ gas-atomized particles are of potential use as a magnetic refrigerant. Nevertheless, we notice that other important properties should be guaranteed for these particles before implemented in AMR systems. In particular, mechanical stability is a significant factor in determining the usefulness of the produced particles, and we have a future plan to assess the mechanical properties of the gas-atomized particles. Meanwhile, the technique devised in this study would facilitate the stable production of spherical particles of various brittle magnetocaloric materials with high melting points, including the RT₂ compounds, helping develop high-performance magnetic liquefiers for hydrogen.

Acknowledgements

This work was supported by JST-Mirai Program Grant Number JPMJMI18A3, Japan.

References

[1] K. A. Gschneidner Jr., V. K. Pecharsky, A. O. Tsokol, Recent developments in magnetocaloric materials, Rep. Prog. Phys. 68 (2005) 1479-1539.
[2] A. Kitanovski, J. Tusek, U. Tomc, U. Plaznik, M. Ozbolt, A. Poredos, Magnetocaloric Energy Conversion, From Theory to Applications, Springer International Publishing, Cham, Switzerland, 2015.
[3] J. Lyubina, Magnetocaloric materials for energy efficient cooling, J. Phys. D: Appl. Phys. 50 (2017) 053002.
[4] V. Franco, J. S. Blázquez, J. J. Ipus, J. Y. Law L. M. Moreno-Ramírez, A. Conde, Magnetocaloric effect: From materials research to refrigeration devices, Prog. Mater. Sci. 93 (2018) 112.
[5] S. A. Sherif, N. Zeytinoglu, T. N. Veziro $\check{\rm g}$ lu, Liquid hydrogen: Potential, problems, and a proposed research program, Int. J. Hydrogen Energy 22 (1997) 683-688.
[6] A. T. Wijayanta, T. Oda, C. W. Purnomo, T. Kashiwagi, M. Aziz, Liquid hydrogen, methylcyclohexane, and ammmonia as potential hydrogen storage: Comparison review, Int. J. Hydrogen Energy 44 (2019) 15026-15044.
[7] T. Numazawa, K. Kamiya, T. Utaki, K. Matsumoto, Magnetic refrigerator for hydrogen liquefaction, Cryogenics 62 (2014) 185.
[8] H. Zhang, R. Gimaev, B. Kovalev, K. Kamilov, V. Zverev, A. Tishin, Review on the materials and devices for magnetic refrigeration in the temperature range of nitrogen and hydrogen liquefaction, Phys. B: Condens. Matter 558 (2019) 65-73.
[9] D. X. Li, T. Yamaura, S. Nimori, Y. Homma, F. Honda, Y. Haga, D. Aoki, Large reversible magnetocaloric effect in ferromagnetic semiconductor EuS, Solid State Commun. 193 (2014) 6-10.
[10] K. T. Matsumoto and K. Hiraoka, Magnetocaloric effect in Gd-based ferromagnet GdZn₂, J. Magn. Magn. Mater. 423 (2017) 318-320.
[11] H. Omote, S. Watanabe, K. Matsumoto, I. Gilmutdinov, A. Kiiamov, D. Tayurskii, Magnetocaloric effect in single crystal GdTiO₃, Cryogenics 101 (2019) 58-62.
[12] P. B. Castro, K. Terashima, T. D. Yamamoto, Z. Hou, S. Iwasaki, R. Matsumoto, S. Adachi, Y. Saito, P. Song, H. Takeya, Y. Takano, Machine-learning-guided discovery of the gigantic magnetocaloric effect in HoB₂ near the hydrogen liquefaction temperature, NPG Asia Materials 12 (2020) 35.
[13] T. Utaki, K. Kamiya, T. Nakagawa, T. A. Yamamoto, T. Numazawa, Research on a Magnetic Refrigeration Cycle for Hydrogen Liquefaction, Cryocooler 14 (2007) 645.
[14] K. Matsumoto, T. Kondo, S. Yoshioka, K. Kamiya, T. Numazawa, Magnetic refrigerator for hydrogen liquefaction, J. Phys.: Conf. Ser. 150 (2009) 012028.
[15] Y. Kim, I. Park, S. Jeong, Experimental investigation of two-stage active magnetic regenerative refrigerator operating between 77 K and 20 K, Cryogenics 57 (2013) 113-121.
[16] T. Hashimoto, K. Matsumoto, T. Kurihara, T. Numazawa, A. Tomokiyo, H. Yayama, T. Goto, S. Todo, M. Sahashi, Investigation on the possibility of the RAl₂ system as refrigerant in an ericsson type magnetic refrigerator, Adv. Cryog. Eng. Mater. 32 (1986) 279.
[17] A. Tomokiyo, H. Yayama, H. Wakabayashi, T. Kuzuhara, T. Hashimoto, M. Sahashi, and K. Inomata, Specific Heat and Entropy of RNi₂ (R: Rare Earth Heavy Metals) in Magnetic Field, Adv. Cryog. Eng. Mater. 32 (1986) 295.
[18] Y. Zhu, K. Asamoto, Y. Nishimura, T. Kouen, S. Abe, K. Matsumoto, T. Numazawa, Magnetocaloric effect of (Er_xR_1-x)Co₂ (R = Ho, Dy) for magnetic refrigeration between 20 and 80 K, Cryogenics 51 (2011) 494-498.
[19] J. A. Barclay and S. Sarangi, Selection of Regenerator Geometry for Magnetic Refrigerator Applications, Cryog. Proc. Equip. 51 (1984) 1.
[20] B. Yu, M. Liu, P. W. Egolf, A. Kitanovski, A review of magnetic refrigerator and heat pump prototypes, Int. J. Ref. 33 (2010) 1029.
[21] M. G. Osborne, I. E. Anderson, K. A. Gschneidner, Jr., Centrifugal atomization of neodymium and Er₃Ni regenerator particulate, Adv. Cryog. Eng. 40 (1994) 631.
[22] K. Matsumoto, K. Asamato, Y. Nishimura, Y. Zhu, S. Abe, T. Numazawa, Magnetocaloric effect of RM₂ (R $=$ rare earth, M $=$ Ni, Al) intermetallic compounds made by centrifugal atomization process for magnetic refrigerator, J. Phys.: Conf. Ser. 400 (2012) 052020.
[23] S. Wolf, T. M. Riedemann, J. Barclay, J. Holladay, I. E. Anderson, J. Cui, Synthesis and magnetic performance of gadolinium powder produced with rotating disk atomization, Powder Technol. 359 (2020) 331-336.
[24] I. E. Anderson, M. G. Osborne, H. Takeya, K. A. Gschneidner, Jr., Gas atomized Er₃Ni powder for cryocooler applications, Proc. 7th Int. Cryocooler Conf., Santa Fe, NM (1992) 1120.
[25] L. F. Aprigliano, G. Green, J. Chafe, Development of neodymium and Er₃Ni regenerator materials, Adv. Cryog. Eng. 37 (1992) 1003.
[26] S. A. Miller, J. D. Nicholson, K. A. Gschneider, Jr., A. O. Pecharsky, V. K. Pecharsky, Manufacturing Considerations for Rare Earth Powders Used in Cryocooler and Magnetic Refrigerator Applications, Cryocooler 11 (2001) 449.
[27] A. Fujita, S. Koiwai, S. Fujieda, K. Fukamichi, T. Kobayashi, H. Tsuji, S. Kaji, A. T. Saito, Active Magnetic Regeneration Behavior of Spherical Hydrogenerated La(Fe_0.86Si_0.14)₁₃ Fabricated by Rotating Electrode Process, J. J. Appl. Phys. 46 (2007) L154-L156.
[28] L. V. M. Antony, R. G. Reddy, Processes for production of high-purity metal powders, JOM 55 (2003) 14-18.
[29] T. D. Yamamoto, H. Takeya, A. T. Saito, K. Terashima, P. B. Castro, T. Numazawa, Y. Takano, Gas-atomized particles of giant magnetocaloric compound HoB₂ for magnetic hydrogen liquefiers, Appl. Phys. A 127 (2021) 301.
[30] T. D. Yamamoto, H. Takeya, S. Iwasaki, K. Terashima, P. B. Castro, T. Numazawa, Y. Takano, Magnetic entropy change of ErAl₂ magnetocaloric wires fabricated by a powder-in-tube method, J. Phys. D: Appl. Phys. 53 (2019) 095004.
[31] T. D. Yamamoto, H. Takeya, K. Terashima, S. Iwasaki, P. B. Castro, T. Numazawa, Y. Takano, Adiabatic temperature change in ErAl₂/metal PIT wires: A practical method for estimating the magnetocaloric response of magnetocaloric composites, J. Magn. Magn. Mater. 513 (2020) 167207.
[32] H. Franz, L. Plochl, F. P. Schimansky, Recent Advances of Titanium Alloy Powder Production by Ceramic-free Inert Gas Atomization, Titanium 2008 International Titanium Associtation (2008) 198.
[33] M. Wei, S. Chen, J. Liang, C. Liu, Effect of atomization pressure on the breakup of TA15 titanium alloy powder prepared by EIGA method for laser 3D printing, Vaccum 143 (2017) 185-194.
[34] B. Xie, Y. Z. Fan, Q. D. Miao, Study on the process of TC4 powders prepared by electrode induction melting gas atomization for laser 3D printing, IOP Conf. Ser.: Mater. Sci. Eng. 668 (2019) 012008.
[35] R.-P. Guo, L. Xu, B. Y.-P. Zong, R. Yang, Characterization of Prealloyed Ti-6Al-4V Powders from EIGA and PREP Process and Mechanical Properties of HIPed Powder Compacts, Acta Metall. Sin. (Engl. Lett.) 30 (2017) 735-744.
[36] X. Sun, L. Yuan, L. Wang, L. Cheng, Study on the Formability of 3D Printed TC4 Alloy Powder by EIGA, IOP Conf. Ser.: Mater. Sci. Eng. 493 (2019) 012134.
[37] Y. F. Ternovoi, A. G. Tsipunov, S. B. Kuratchenko, O. M. Kuimova, K. V. Kondakova, Pore formation in atomized powders, Powder Metall. Met. Ceram. 24 (1985) 10-13.
[38] R. Gr $\rm{\ddot{o}}$ ssinger, W. Steiner, K. Krec, Magnetic investigations of pseudobinary SE(Fe,Al)₂ systems (SE = Y, Gd, Dy, Ho), J. Mang. Magn. Mater. 2 (1976) 196-202.
[39] E. N. Maslen, V. A. Streltsov, N. Ishizawa, A Synchrotron X-ray Study of the Electron Density in C-Type Rare Earth Oxides, Acta Crystallogr. B 52 (1996) 414-422.
[40] E. E. Havinga, Influence of repulsive energy on structural parameters of closed-packed metal structures, J. Less. Commn. Met. 41 (1975) 241-254.
[41] J. F. Cannon and H. T. Hall, Effect of high pressure on the crystal structures of lanthanide trialuminides, J. Less. Commn. Met. 40 (1975) 313-328.
[42] F. Izumi and K. Momma, Three-dimensional visualization in powder diffraction, Solid State Phenom. 130 (2007) 15.
[43] K. H. J. Buschow, in: E. P. Wohlfarth (Ed.), Handbook of Ferromagnetic Materials, vol. 1, Elsevier, 1980, pp. 297-414.
[44] T. W. Hill, W. E. Wallace, R. S. Craig, T. Inoue, Low temperature heat capacities and related thermal properties of TbAl₂ and HoAl₂, J. Solid State Chem. 8 (1973) 364-367.
[45] B. Barbara, M. F. Rossignol, J. X. Boucherle, Magnetic behavior of HoAl₂ single crystal, Phys. Lett. 55A (1975) 321.
[46] M. R. Ibarra, O. Moze, P. A. Algarabel, J. I. Arnaudas, J. S. Abell, A. del Moral, Magnetoelastic behaviour and the spin-reorientation transition in HoAl₂, J. Phys. C: Solid State Phys. 21 (1988) 2735-2748.
[47] J. C. P. Campoy, E. J. R. Plaza, A. A. Coelho, S. Gama, Magnetoresistivity as a probe to the field-induced change of magnetic entropy in RAl₂ compounds (R = Pr, Nd, Tb, Dy, Ho, Er), Phys. Rev. B 74 (2006) 134410.
[48] M. Khan, K. A. Gschneidner, Jr., V. K. Pecharsky, Spin reorientation transitions in Ho_1-xDy_xAl₂ alloys, J. Appl. Phys. 110 (2011) 103912.
[49] H. Wada, Y. Tanabe, M. Shiga, H. Sugawara, H. Sato, Magnetocaloric effects of Laves phase Er(Co_1-xNi_x)₂ compounds, J. Alloy. Compd. 316 (2001) 245-249.
[50] M. Patra, S. Majumdar, S. Giri, Y. Xiao, T. Chatterji, Magnetic, magnetocaloric and magnetoresistive properties of cubic Laves phase HoAl₂ single crystal, J. Phys.: Condens. Matter. 26 (2014) 046004.
[51] L. A. Gil, J. C. P. Campoy, E. J. R. Plaza, M. V. de Souza, Conventional and anisotropic magnetic entropy change in HoAl₂ ferromagnetic compound, J. Magn. Magn. Mater. 409 (2016) 45-49.