Development of Synthesis Process of Magnetocaloric Material of La(Fe, Co, Si)13

(1)

Development of Synthesis Process of Magnetocaloric

Material of La(Fe, Co, Si)

13 Akiko T. Saito

1,2

*, Takayoshi Hirakawa

1

, Shiori Kaji

2 1

Department of Urban Environment Systems, Chiba University, 1-33 Yayoi, Inage, Chiba 263-8522, Japan

2

Corporate Research & Development Center, Toshiba Corporation, 1, Komukai-Toshiba, Saiwai, Kawasaki, 212-8582, Japan *Corresponding Author's Email: [email protected]; [email protected]

Abstract – La(Fe,Si)13 – based compound has been

considered as one of the excellent magnetocaloric working substance for an environmentally-friendly cooling technique of magnetic refrigeration. A new synthesis process of La(Fe,Co,Si)13 compounds was developed and compared to

the conventional fabrication process. Melting the starting materials of La, Fe, Co and Si elements is the conventional method and it requires homogenized heat treatment at above 1000°Cfor over a week after melting, in order to form the La(Fe,Co,Si)13 phase by atomic inter-diffusion between

La-rich and Fe-La-rich phases. Rotating electrode process enables to form the spherical particles of near-net-shape of magnetic refrigerant and also shorten the heat treatment time to the formation of the La(Fe,Co,Si)13-phase to 3 days, because the

spherical particles have fine metallographic structure consisted of La-rich and Fe-rich phases. Moreover, applying the spark plasma sintering process to this material was investigated. The phase formation behavior depends on both the starting materials and the process conditions of SPS.As a results, a thin plate, which is another suitable shape for magnetic refrigeration, formed of almost the La(Fe,Si)13

-phase was successfully obtained only by the use of the SPS process with starting material of Fe, Co, LaFeSi and LaFe2Si2

powder. This process enables to achieve a substantial reduction in time and energy to fabricate the near-net-shape magnetic refrigerant of La(Fe,Si)13–based compounds,

because the post annealing is not necessary for this process.

Keywords – Spark Plasma Sintering (SPS), Rotating Electrode Process (REP), Plasma-Arc Melting, Homogenized Heat Treatment.1

I. I

NTRODUCTION

In the last few decades, global environmental problems such as the ozone layer destruction or the global warming have been emerged. Then, the conventional fluids of vapor compression refrigeration of chlorofluorocarbons, hydro-chlorofluorocarbons, and hydro fluorocarbons have been phased out under the Montreal and Kyoto Protocols, and replaced with natural refrigerants. However, these alternative refrigerant fluids are not ideal, for example, isobutane is flammable and ammonia is harmful. Moreover, refrigeration and cooling technology have a large impact on the power consumption in total electrical power use.

Magnetic refrigeration, in which magnetocaloric material is used as the working substance, has received much attention as a highly-efficient and

This research was partially supported by the Cabinet Office, Government of Japan and the Japan Society for the Promotion of Science (JSPS) through the NEXT Program.

friendly cooling technique. In 1997, continuous operation of active magnetic regenerative (AMR) refrigeration cycle near room-temperature had been demonstrated for 18 months [1], [2]. This work showed the higher potential coefficient of performance (COP) of AMR refrigeration than that of the household refrigerator and shed light on the alternative to conventional vapor compression method using working gas fluid having a large impact on the environment. However, it was realized by using large magnetocaloric effect of gadolinium (Gd) obtained only by applying a high magnetic field generated by a super-conducting magnet. Since then, AMR refrigeration device using permanent magnets instead of superconducting magnet [3]-[15] and magnetocaloric materials [16]-[23] have been actively investigated toward practical applications. La(Fe,Si)13–based compounds were proposed

as an excellent candidate in terms of both magnetocaloric performance and green technology, although Gd is a rare-earth element. However, there have been a few reports of attempt to apply the La(Fe,Si)13– based compounds to the

AMR refrigeration[6], [12], [15], [23]. This is owing to the fact that long-term heat treatment over a week is necessary to obtain the La(Fe,Si)13 compounds. Further-more,

magnetocaloric material is required to be formed into the suitable shape, such as a small sphere or thin plate, for practical use in the AMR refrigeration[1], [2], [9], [22].In the case of the La(Fe,Si)13–based compounds, material

fabrication process[24], [25] is a bottleneck in the advancement of application to the AMR refrigeration.

In this study, the fabrication process of La(Fe,Si)13–

basedcompounds having near-net-shape were developed, and compared to the conventional fabrication process. Finally, we show advanced synthesis process of the magnetocaloric material of La(Fe,Si)13–based compounds

without post annealing.

II. E

XPERIMENTAL

A.

Material preparation and fabrication process

In a conventional fabrication method, raw material of La, Fe, Co and Si was prepared respectively in the appropriate amount in the ratio of the nominal composition of La(Fe,Co,Si)13, and melting them together in a plasma-arc

furnace under argon atmosphere. Button samples were obtained after repeat a melting several times. After that, button samples were cut into several pieces and subjected to homogenized heat treatment under specific conditions.

In order to obtain spherical particles of La(Fe,Co,Si)13

(2)

casting from raw materials of La, Fe, Co and Si in the appropriate amounts in the ratio of the nominal composition of La(Fe,Co,Si)13. Secondly, the cylindrical

rod alloy was mounted on a rotating shaft in a large chamber, and the rotating electrode of the cylindrical alloy was melted into drops by arc jet and solidified while fling in helium gas atmosphere. The diameter of the spherical particles was controlled by changing the rotation speed of the mother alloy electrode. After that, the spherical particles were subjected to homogenizing anneal under several conditions.

In contrast to above-mentioned alloying procedure, powder metallurgy technique is free from the melting process. Additionally, we were focused on the different generation pass to form the La(Fe,Co,Si)13compound, and

employed the starting materials of Fe, Co, LaFeSi and LaFe2Si2 powder. The mixture of specific amounts of Fe,

Co, LaFeSi and LaFe2Si2 powder was filled into the mold

which consists of a carbon-die and-punch. The die was set in a chamber, and the powder mixture was sintered by spark plasma sintering (SPS) process, where the filled sample is heated by a pulsed electric current under high pressure. The SPS apparatus using this work is SPS-511S (Sumitomo Coal Mining Co. Ltd.). In the SPS process, it is considered that an electrical discharge between the powder fine particles plays important role for sintering as well as heating by pulsed electric current. Therefore, the electric pulse pattern was also varied as a parameter in addition to the temperature and current value. The sample shape can be changed by mold shape. Thin plate sample was obtained using rolled out green body of powder mixture with solvent additive of polyvinylpyrrolidone.

B. Material characterization

The La(Fe,Si)13-based compounds have a NaZn13-type

crystal structure, and could not be obtained after just only melting the elements of the nominal composition. There is no solid solution and inter metallic compound between Fe and La. Although the NaZn13-type crystal structure is

stabilized by partial replacement of Zn-site with Si or Co elements, La(Fe,Si)13or La(Fe,Co,Si)13 phases are

metastable states in liquid phase. Therefore, the materials undergo phase separation into Fe-rich and La-rich phases trough the melting process. The characterization of materials fabricated by different method was performed mainly by X-ray powder diffraction (XRD) and scanning electron microscopy (SEM).The micro structure composed of multiple phases was specified by backscattered electron (BSE) imaging. Finally, the magnetocaloric properties were confirmed by magnetization measurement for the well-homogenized sample using a superconducting quantum interference device (SQUID) magnetometer.

III. R

ESULTS AND

D

ISCUSSION

First of all, the results of material fabrication in a conventional method are shown. As mentioned before, in the La(Fe,Si)13-based compounds, single phase having a

NaZn13-type crystal structure can hardly be obtained by

only melting process. Therefore, we executed homoge-nized heat treatment after arc-melting. Figure 1 shows X-ray diffraction patterns of the samples as arc-melting and after annealing at 1060°C for 3 and 7 days which

Fig.1. X-ray diffraction patterns of as arc-meltingand after annealing samples at 1060°C for 3 and 7 days.

Fig.2. BSE images of the cross-sectional area of the samples(a) as arc-meltingand (b) after annealing at 1060°C

for 7 days.

correspond to the nominal composition of La(Fe0.85Co

0.07-Si0.08)13. For the sample as arc-melting, the main-phase of

-Fe having body center cubic (bcc) structure and small amount of the sub-phase corresponds to La(Fe,Co)Si were observed in the XRD pattern, in which the closed and open triangles come from bcc and Cu2Sb-type crystalline

structure phases respectively. These phases correspond to the dark gray and bright gray parts in a BSE imageas shown in Fig. 2(a), those are, the main-phase of -Fe appears dendritic structures in the sub-phase of La(Fe,Co)Si as a result of the peritectic reaction. After annealing at 1060°C for 3days, indication of growth of La(Fe,Co,Si)13 phase having a NaZn13-type crystal

structure was observed as shown by the closed circle in Fig. 1. The La(Fe,Co,Si)13 phase grows to most part in the

sample after annealing for 7 days as shown in Fig. 1 and Fig. 2(b) in which middle gray part corresponds to the La(Fe,Co,Si)13 phase. The BSE image shows clear contrast

(3)

each phase.

Fig.3. (a) Schematic view of rotating electrode process, and photographs of (b) a cylindrical electrode of mother

alloy, and (c)spherical particles.

Fig.4. BSE images of the cross-sectional area of the particle samples (a) as REPed and (b) after annealing at

1050°C for 3days.

One of the suitable shapes of magnetocaloric materials for applying to AMR refrigeration near room temperature is a small sphere ranging from 0.2 to 1 mm. Therefore, spherical particles were made by means of the REP using a cylindrical mother alloy having the nominal composition of La(Fe0.85Co0.07Si0.08)13asshown in Fig. 3.The

cross-sectional image of micro-structure observed by SEM are shown in Fig. 4 for the sample (a) as-REPed and (b) after annealed the RERed particles at 1050°C for 3 days. Though the dendritic structures appear in the REPed sample in analogy with the sample as arc-melting shown in Fig. 2 (a), the microstructure in the REPed sample is much finer than that in the arc-melting sample. The fine structure was due to the rapid solidification because of the small particle size and the effective heat-transfer by use of helium gas. As a result, the La(Fe,Co,Si)13 phase have

formed for the most part in the sample after annealing within 3daysas shown in Fig. 4 (b).These results are consistent with a prediction of La(Fe,Co,Si)13 phase

formation based on the atomic inter diffusion reaction between -Fe phase and La(Fe,Co)Si phase by reference to a phase diagram of La-Fe-Si [26]. Figure 5 shows a schematic view of the ternary phase diagram of La-Fe-Siin an isothermal section at T = 900°C. This phase diagram teaches that La(Fe,Si)13 phase having NaZn13-type crystal

structure could never be obtained by inter diffusion reaction directly between the Fe2Si, FeSi, FeSi2, LaSi2,

LaSi, La5Si4, La3Si2, La5Si3, or La2FeSi3phases at 900°C.

The phases next to the La(Fe,Si)13 phase are Fe (including

Fe-Si solid-solution), LaFe2Si2, and LaFeSi phases.

Therefore, the formation of the La(Fe,Si)13 phase need to

atomic interdiffusion between Fe (including Fe-Si) phase

and LaFeSior LaFe2Si2 phases. In both samples as

arc-melting and as REPed, shown in the Fig. 2(a) and Fig. 4(a) respectively, -Fe phase and La(Fe,Co)Si phase next to each other. In comparison with the arc-melting sample, the interdiffusion length between -Fe and La(Fe,Co)Si phases, necessary for the formation of the La(Fe,Co,Si)13

phase, in the sample as REPed is so smaller, that annealing time required to form La(Fe,Co,Si)13 phase becomes

shorter.

On the basis of above-mentioned considerations, we also examined by powder metallurgical method to form La(Fe,Si)13 phase using the stating powder materials

selected from the group of Fe, Fe-Si, Co, La(Fe,Co)Si and La(Fe,Co)2Si2. In order to compare with the results

indifferent fabrication process, one targetmaterial was set to La(Fe0.85Co0.07Si0.08)13in terms of the magnetic transition

temperature. The process conditions of the spark plasma sintering; maximum current value, hydraulically-operated applying pressure, isothermal holding time, rate of temperature rise, electric pulse pattern, were varied to find the optimal conditions for generation of the La(Fe,Co,Si)13

compound. At last, we succeeded in obtaining the almost single phase of the La(Fe,Co,Si)13by only spark plasma

sintering.

Fig.5. Schematic view of the ternary phase diagram of La-Fe-Si in an isothermal section at T = 900°C.

Table I: The Process Conditions

Sample

Parameters of spark plasma sintering process Maximum

current (A)

Applying pressure

(MPa)

Holding time (minute)

Electric pulse pattern Non-Noff

(a) 400 40 30 20-2

(b) 400 40 60 20-2

(c) 400 40 90 20-2

(d) 390 40 90 20-2

(e) 380 40 90 20-2

(4)

BSE for several typical samples fabricated around optimal conditions are shown in Fig. 6 (a) ~ (f), and their process conditions are given in Table I. The starting powder materials in these samples were the same, the mixture of Fe, Co, and LaFeSi where just La was slightly added the amount by 4% to the nominal composition of La(Fe0.85Co0.07Si0.08)13 considering a loss by vaporization

or oxidization while sintering.

The electric pulse pattern of Non-Noff is illustrated in the

inset of Fig. 6. As regarding process temperature, it is difficult to specify the accurate powder sintering temperature. As mentioned before, in the SPS process, the powder mixture filled in a carbon-diesintered by the electrical discharge between powder fine particles and the heating induced by a pulsed electric current. Thus, we could not know the accurate temperature of the powder while sintered in different process condition even by monitoring the die-temperature. For these reasons, we controlled the input energy not by the temperature but by the input current. To compare with the different fabrication process, relation between the input current and die-temperature was shown in Fig. 7as a guide.

In the case of the sample sintered by SPS with a holding current of 400A as seen in Fig.6 (a) ~ (c), the holding time of 60 minutes shows better result than those for 30 or 90 minutes. The middle gray part corresponding to the La(Fe,Co,Si)13 phase is spread to most part in the area

while dark gray and bright gray parts, correspond to the  -Fe and La(-Fe,Co)Si, exist in small amounts. As for the case of varying the holding current with the same holding time of 90 minutes as seen in Fig.6 (c) ~ (e), the holding current of 390Ashows better result than 380 or 400A. These results indicate that there are optimum conditions neither too much energy nor too less energy to encourage the interdiffusion between Fe, Co, and LaFeSi phases. Finally, almost single phase of the La(Fe,Co,Si)13

compound was obtained by only spark plasma sintering under the condition of the holding current of 390A for 90 minutes, and the electric pulse pattern of 20-4. Considering the relation between the input current and die-temperature as shown in Fig. 7, this optimal condition could correspond to the temperature about 1000 ~ 1100°C. It is consistent with the former results on the different fabrication process such as arc-melting, REP and post-anneal.

Fig.6. BSE images of the cross-sectional area of thesamples as sintered by SPS with a holding current of (a) 400Afor 30 min., (b) 400Afor 60 min., (c) 400Afor 90 min., (d) 390Afor 90 min., (e) 380Afor 90 min. and (f) 390Afor 90 min.,

respectively, where the nominal compositions correspond to La(Fe0.85Co0.07Si0.08)13. The lowerleft and middle

illustrationsare schematic views of the main part of SPS apparatus and electric pulse pattern.

(a)

(b)

(c)

(d)

(e)

electrode (－)

electrode (＋)

press

spacer punch

die thermocouple

mixed powder

spacer punch

Electric pulse pattern

Non Noff

Time

Curr

en

t

(f)

20mm

20mm 20mm

20mm

(5)

die-temperaturewhile sintering.

Fig.8. BSE images of the samples obtained by SPS with a holding current of (a)375A, and (b) 400Afor 120 minutes,

from the starting materials of Fe and LaFe2Si2.

Figure 8 (a) and (b) show the BSE images of the cross-sectional area of the samples as sintered by SPS from the starting materials of Fe and LaFe2Si2, where the nominal

compositions correspond to La(Fe0.88Si0.12)13.The holding

currents were 375A and 400A respectively, and holding time for both samples is 120 minutes. As seen above, interdiffusion between Fe and LaFe2Si2phasesis proved to

be an another possible pass to form the La(Fe,Si)13 phase.

For applying the La(Fe,Co,Si)13-based material to AMR

refrigeration, thin plate is a suitable shape besides the small sphere. Then, the plate sample was also prepared by SPS using rolled out green body of powder mixture with solvent additive of polyvinylpyrrolidone. The nominal composition set to La(Fe0.85Co0.07Si0.08)13. Figure 9 (a) and

(b) show the rolled out green body before SPS and sintered thin plate with the size of 60 mm in diameter and of 0.4 mm in the thickness. As described above, applying

Fig.9. photographs of (a) rolled out green body, and (b) thin plate fabricated by SPS.

Fig.10. Temperature dependences of magnetic entropy change of the La(Fe0.85Co0.07Si0.08)13 compound fabricated

by SPS and Gd-metal.

the SPS process enable us to form the La(Fe,Co,Si)13

-based material easily without post-annealing, and shape into a near-net-shape at the same time.

At the end of this work, magnetocaloric properties were confirmed. Figure 10 shows the temperature dependence of magnetic entropy change, S of the La(Fe0.85Co0.07

-Si0.08)13 compound fabricated by SPS process in

comparison to a conventional magnetocaloric material of Gd-metal, where an external magnetic field µ0Hext is 5

Tesla. The magnetic entropy changes were estimated from magnetization curves using the Maxwell relation:

H T T

H T M H

H T S

   

 

      

 



 ( , ) ( , )

,

where T, S, Hand M are temperature, entropy, magnetic field and magnetization, respectively. The S of La(Fe0.85

-Co0.07Si0.08)13 show a peak at around it’s Curie temperature

(TC), which is almost the same value as the TC of Gd.

(Because the composition of La(Fe0.85Co0.07Si0.08)13 was

designed to conform with Gd in the Curie temperature). The S peak values in La(Fe0.85Co0.07Si0.08)13 is 1.5 times

larger than that in Gd. As seen above, La(Fe0.85Co0.07-

Si0.08)13 compound fabricated by SPS process shows a

better performance in magnetocaloric property.

(6)

actively investigated in terms of both physical properties and metallographic aspect [27]-[32]. In the pseudo binary La(Fe1-xSix)13 system, a first order transition was observed

which is accompanied by a large magnetic entropy change in the vicinity of its TC due to meta-magnetic transition for

the composition range of 0.10 ≤ x ≤ 0.14, and a second

order transition of typical ferromagnetic to paramagnetic transition was observed for 0.14 <x[20], [33]. Therefore, the lower Si content is favorable for obtaining the larger magnetocaloric effect. However, obtaining a single phase of La(Fe,Si)13 with the NaZn13-type crystal structure

becomes difficult with lowering the Si content, and is hard for the range of Si content x < 0.10. A systematic study of annealing temperature dependence on the microstructure and physical properties of the La(Fe1-xSix)13 fabricated by

induction melting method was executed in the critical Si content region of x = 1.08[31]. They demonstrated that when annealing temperature is too high, particular grain growth of Fe was observed. When the temperature is too low, many small grains of Fe still remained because atomic diffusion could not be fully advanced. Our results are fairly consistent with this systematic study [31]. On the other hand, rapid cooling, such as melt spinning method, is one effective way to form the NaZn13-type crystal

structure phase[32], because molten alloy is almost solidified before segregating to La-rich and Fe-rich phases. In the melt spinning method, relatively homogeneous micro structure is obtained compared to the casting method. Powder metallurgical method is further advanced way in the same concept, and was employed in our study. By the way, in order to increase TC to room temperature

range, both Co substitution [21] and hydrogen absorption [20] are effective to the La(Fe1-xSix)13 compounds. Small

amount of Co substitution for Zn-site in La(Fe,Si)13

increases TC significantly, at the same time, magnetic

property is strongly affected. In some compositions, meta-magnetic property changes to ferrometa-magnetic and decreases the large magnetocaloric effect. To obtain relatively large entropy change, La(Fe,Co,Si)13 compound with low Si

content is favorable[21], although it is not easy to form the single phase having the NaZn13-type crystal structure by

ordinary casting method.

In this study, samples with nominal composition of La(Fe0.85Co0.07Si0.08)13, which is expected to exhibit larger

entropy change than Gd, were prepared by three different processes, they are, arc-melting, rotating electrode process, and spark plasma sintering with the post-anneal if needed. As a result, La(Fe,Co,Si)13 compound having NaZn13-type

crystal structure were obtained in all the process although appropriate post-anneal were necessary in the case of arc-melting and REP. The heat treatment time for these processes is summarized in Fig. 11. Time required to form the La(Fe,Co,Si)13 phase by innterdiffusion between -Fe

and La(Fe,Co)Si phases depend strongly on the metallo-graphic microstructure of the material. The REP cut the required time in half compared to the arc-melting process due to finer metallographic structure of dendrite morphology. Furthermore, the SPS process could reduce

the amount of time required to form the La(Fe,Co,Si)13

phase dramatically.

IV. C

ONCLUSION

Magnetic refrigeration is one of the alternative technologies to enable us to be free from the geo environ-mental impact of the refrigerant gas used in the conventional gas-compression cooling technique. Ever since La(Fe,Si)13-based compounds was proposed as an

excellent candidate of magnetocaloric material for the magnetic refrigeration, development of the manufacturing process for mass-production of these materials has been eagerly anticipated. In this work, we have succeeded in reduce the turnaround time to form La(Fe,Si)13-based

compounds from several days to just 3 hours at the same time as forming near-net-shape. This work will shed a ray of light on advances in the cooling system research of magnetic refrigeration.

R

EFERENCES

[1] C. B. Zimm, A. Jastrab, A. Sternberg, V. Pecharsky, K. A. Gschneidner, Jr., M. Osborne, and I. Anderson, “Description and performance of a near-room temperature magnetic refrigerator,”

Adv. Cryog. Engin., vol. 43, 1998, pp. 1759–1766.

[2] V. K. Pecharsky and K. A. Gschneidner, Jr., “Magnetocaloric effect and magnetic refrigeration,” J. Magn. Magn. Mater.,vol. 200, 1999, pp. 44-56.

[3] X. Bohigas, E. Molins, A. Roig, J. Tejada, and X. X. Zhang, “Room temperature magnetic refrigerator using permanent magnets,” IEEE Trans. Magn., vol.36, 2000, pp. 538-544. [4] P. Colt, D. Viallet, F. Allab, A. Kedous-Lebouc, J. M. Fournier,

and J. P. Yonnet, “A magnet-based device for active magnetic regenerative refrigeration,” IEEE Trans. Magn., vol. 39, 2003, pp. 3349-3351.

[5] T. Okamura, K. Yamada, N. Hirano, S. Nagaya, “Performance of a roomtemperaturerotary magnetic refrigerator,” International Journal of Refrigeration, vol. 29, 2002, pp. 1327-1331. [6] C. Zimm, A. Boeder, J. Chell, A. Sternberg, A. Fujita, S.Fujieda,

and K. Fukamichi, “Design and performance of a permanent magnet rotary refrigerator,” Proceedings of the First IIF-IIR International Conference on Magnetic Refrigeration at Room Temp., 2005, pp. 367-373.

[7] A.Tura, A.Rowe, “Design and testing of a permanent magnet magnetic refrigerator,”Proceedings of the Second IIF-IIR International Conference on Magnetic Refrigeration at RoomTemperature, April2007, pp. 363-371.

[8] C. V. Muller,“Innovative design of a magnetocaloric system,”InternationalJournal of Refrigeration, vol. 29, 2006, pp.1318-1326.

[9] C. R. H. Bahl, K. Engelbrecht, D. Eriksen, J. A. Lozano, R. Bjork, J. Geyti, K. K. Nielsen, A. Smith, and N. Pryds, “Development and experimental results from a 1 kW prototype AMR,” Proceedings of the 5th IIF-IIR International Conference on Magnetic Refrigeration at Room Temperature, 2012, pp.53-60.

[10] J. M. Gatti, C. V. Muller, G. Brumpter, and P. Haegel, “Magnetic heat pumps – configurable hydraulic distribution for a magnetic cooling system,”Proceedings of the 5th IIF-IIR International Conference on Magnetic Refrigeration at Room Temperature, 2012, pp.341-348.

(7)

[12] S. Jacobs, J. Auringer, A. Boeder, J.Chell,L. Komorowski, J. Leonard, S.Russek, and C. Zimm, “The performance of a large-scale rotary magnetic refrigerator,” Proceedings of the 5th IIF-IIR International Conference on Magnetic Refrigeration at Room Temperature, 2012, pp.421-428.

[13] J. Tusek, A. Kitanovski, U. Flisar, S. Zupan, I. Prebil, and A. Ppredos, “Active magnetic regenerator (AMR) experimental test device,” Proceedings of the 5th IIF-IIR International Conference on Magnetic Refrigeration at Room Temperature, 2012, pp.429-436.

[14] A. T. Saito, T. Kobayashi, S.Kaji, J. Li, and H.Nakagome, “Environmentally friendly magnetic refrigeration technology using ferromagnetic Gd alloys,” International Journal of Environmental Science and Development , vol. 7, 2016, pp.316-320.

[15] A. T. Saito, T. Kobayashi, H.Fukuda, R. Arai, and H. Nakagome, “Cooling Properties of Gd alloys and La(Fe,Si)13-based compounds in active magnetic Refrigeration for environmentally-friendly coolingsystems,” International Journal of Environmental Science and Development , vol. 7, 2016, pp.335-339.

[16] H. Wada and Y. Tanabe, “Giant magnetocaloric effect of MnAs 1-xSbx,” Appl. Phys. Lett., vol. 79, 2001, pp.3302-3304.

[17] O. Tegus, E. Bruck, K. H. J. Buschow, and F. R.de Bore, “Transition-metal-based magnetic refrigerants for room-temperature applications,” Nature, vol. 415, 2002, pp.150-152. [18] X. X. Zhang, G. H. Wen, F. W. Wang, W. H. Wang C. H. Yu,

and G. H. Wu, “Magnetic entropy change in Fe-based compound LaFe10.6Si2.4,” Appl. Phys. Lett., vol. 77, 2000, pp. 3072-3074. [19] S. Fujieda, A. Fujita, and K. Fukamichi, “Large magneto-caloric

effect in La(FexSi1-x)13 itinerant-electron meta-magnetic compounds,” Appl. Phys. Lett., vol. 81, 2002, pp.1276-1278. [20] A. Fujita, S. Fujieda, Y. Hasegawa, and K. Fukamichi,

“Itinerant-electron metamagnetic transition and large magnetocaloric effects in La(FexSi1-x)13 compounds and their hydrides,” Phys. Rev. B., vol. 67, 2003, pp.104416:1-6. [21] A.T.Saito, H.Tsuji, and T.Kobayashi, “Magnetocaloric effects

and magnetic properties in intermetallic compounds La(Fe 1-x-yCoxSiy)13,” Proceedings of the 2005 International Magnetic

Conference, 2005, GU-15.

[22] A.T.Saito, T.Kobayashi, and H.Tsuji, “Magnetocaloric effect of new spherical magnetic refrigerant particles,” J. Magn. Magn. Mater, vol. 310, 2007, pp.2808-2810.

[23] A. Fujita, S. Koiwai, S. Fujieda, K. Fukamichi, T.Kobayashi, H.Tsuji, S.Kaji, and A.T.Saito, “Active magnetic regeneration behavior of spherical hydrogenated La(Fe0.86Si0.14)13 fabricated by rotating electrode process,”Jpn. J. Appl. Phys, vol. 46, 2007, pp. L154-156.

[24] L. F. Bao, F. X. Hu, L. Chen, J. Wang, J. R. Sun, and B. G. Shen,“Magnetocaloric properties of La(Fe,Si)13-based material and its hydrideprepared by industrial mischmetal,”Appl. Phys. Lett., vol. 101, 2012, pp.162406:1-5.

[25] S. Fujieda, K. Fukamichi, and S. Suzuki, “Microstructure and isothermal magnetic entropy change of La(Fe0.89Si0.11)13 in a single phase formation process by annealing,”J. Alloys Compd., vol.566, 2013, pp. 196-200.

[26] O. I. Bodak, and E. I. Gladyshevski, Visn. L’viv. Univ. Ser. Khim.

(Book style), no. 14, 1972, pp. 29-34.

[27] S. Fu, Y. Long, C. Wang, M. Zhang,S.Ohnuki, and F. Hu, “Formationof 1:13 phase in La(Fe,Si)13-based compounds by diffusion of LaFe,Si/-Fe(Si) couple,” IEEE Trans. Magn., vol. 48, 2012, pp. 3757-3759.

[28] C. Xiang, Y. Chen, and Y. Tang, “The effect of different temperature annealing on phase relation ofLaFe11.5Si1.5 and the magnetocaloric effects of La0.8Ce0.2Fe11.5-xCoxSi1.5alloys,” J.

Magn. Magn. Mater.,vol. 323, 2011, pp. 3177-3183.

[29] T. Liu, Y. Chen, Y. Tang, S. Xiao, E. Zhang, and J. Wang, “Structure and magnetic properties of shortly high temperature annealing LaFe11.6Si1.4 compounds,”J. Alloys Compd., vol.475, 2009, pp. 672-675.

[30] B. Bao,Y. Long, B. Fu, C. Wang, R. Ye, Y. Chang, J. Zhao, and J. Shen, “The study on the microstructure and the magnetocaloric effects in LaFe10.8Co0.7Si1.5C0.2compounds at different annealing times,” J. Appl. Phys, vol. 107, 2010,09A905.

[31] J. Liu, M. Krautz, K. Skokov, T. G. Woodcock, and O. Gutfleisch, “Systematicstudy of the microstructure, entropy change and adiabatic temperature change in optimized La-Fe-Si alloys,” ActaMater., vol.59, 2011, pp. 3602-3611.

[32] M. Jasinski,J. Liu, S. Jacobs, and C. Zimm, “La(Fe,Co,Si)13bulk alloys and ribbons with high temperaturemagnetocaloric effects,”

J. Appl. Phys, vol. 107, 2010, 09A953.

[33] A. Fujita, S. Fujieda, Y. Hasegawa, and K. Fukamichi, “Itinerant-electron metamagnetic transition and large magnetovolume effects in La(FexSi1-x)13 compounds,” Phys. Rev.

B., vol. 65, 2001, pp. 014410:1-6.

A

UTHOR

'

S

P

ROFILE

Akiko T. Saito

Doctor degree in energy engineering, Tokyo Institute of Technology, Tokyo, Japan, 1995. Senior Research Scientist, Corporate Research & Development center of Toshiba Corporation, Kawasaki, Japan. Researcher, Chiba University, Chiba, Japan. Her main research interest is the environmentally friendly techniques approached from an application of functional material. Email: [email protected]; [email protected].

Takayoshi Hirakawa

Doctor degree in energy and material engineering, Saga University, Kushu, Japan, 2005. Researcher, Chiba University, Chiba, Japan. His main research interest is the material engineering and nanotechnology.

Email: [email protected].

ShioriKaji

Doctor degree in science, Kushu University, Fukuoka, Japan, 2006. Research Scientist, Corporate Research & Development center of Toshiba Corporation, Kawasaki, Japan.