DOI:10.30919/esmm5f105

ES Materials & Manufacturing, 2018, 1, 35-40

Published online: 25 Sep 2018

Received 13 Aug 2018, Accepted 24 Sep 2018

In-situ X-ray Observation of Synthesizing Process for Rare-earth Transition-metal Pnictides under High Temperature and High Pressure

 Chihiro Sekine,1*Hidetoshi Osanai,1Keisuke Ikemori,1Ryosuke Nakajima,1Shingo Deminami,1JirattaganSirimart1and Hirotada Gotou2

 

1Graduate School of Engineering, Muroran Institute of Technology, Muroran, Hokkaido 050-8585, Japan

2Institute for Solid State Physics, University of Tokyo, Kashiwa, Chiba 277-8581, Japan

Abstract:

We have performed in-situ x-ray diffraction (XRD) measurements in order toacquire the information about synthesis conditions of several rare-earth transition-metal pnictides. By considering the results for a battery of the in-situ XRD measurements, we have obtained conditions optimal for synthesizing layered rare-earth zinc phosphides RZn3P3 (R = Y and Dy) and a new filled skutterudite compoundTbxCo4Sb12. According to the optimum condition acquired by the in-situ XRD experiments, we have succeeded in synthesizing high-quality samples.

Table of Content

By in-situ XRD measurements, we have obtained the optimum conditions for synthesizing TbxCo4Sb12, DyZn3P3 and YZn3P3 under high temperature and high pressure

 

Keywords:High-temperature and high-pressure synthesis; Rare-earth transition-metal pnictides; In-situ x-ray diffraction


1. Introduction

    Transition metal pnictides have attracted much attention for their various anomalous physical properties and application as functional materialssuch as thermoelectric (TE) materials, superconductors and catalysts.1-7However, asynthesis of good quality samplesfree from secondary phases or impuritiesfor transition-metal pnictides is extremely difficult at ambient pressure because the vapor pressure of pnictogen is considerably high. The high-temperature and high-pressure (HTHP) synthesis is one of the most effective methods to prepare transition-metal pnictides.8 Nevertheless, HTHP synthesispractically takes long time and much effort to decide the optimum conditions to synthesize high-quality samples or new compounds. Therefore, it is extremelyefficient to search the processes for synthesis of target materials at HTHP by in-situ x-ray diffraction (XRD) measurementsin advance.8-10

     In this study, we focus on two systems of rare-earth transition-metal pnictides because both systems are important for future application as functional materials. First system is skutteruditecompounds, which are expected as high-performance TE materials, and the other system is layered rare-earth zinc phosphides RZn3P3 (R=rare-earth element), which are magnetic frustrated materials and have a possibility of multiferroic. However, a synthesis of good quality samples for both systems is extremely difficult at ambient pressure. Therefore, we have performed in-situ XRD measurements for the systems in order to synthesize good quality samples free from secondary phases or impurities under high pressure.

     The skutterudite family includes binary, unfilled and ternary, filled types withthe same space group. The binary compounds have the general expressionMX3 or ☐M4X12, where Mis transition metal such as Co, Rh and Ir (site 8c), Xispnictogen (site 24g), and the symbol ☐stands for a vacancy (site 2a) in the large cages formed by the M and X ions. The crystal structure of unfilled skutterudite compounds is body centered cubic (CoAs3 type), space group is Im3  (T5h,No. 204).11This system is most promising high-performance TE materials. The TE performance is estimated by the figure of merit Z= sS2/k. Here, s, k and S are the electrical conductivity, the thermal conductivity and the Seebeck coefficient, respectively. Furthermore,k is given by kE+kL, wherekE and kL are electronic and lattice contribution, respectively.12The binary compound CoSb3 (or ☐Co4Sb12) exhibitssuperior TE properties (large S and high hole mobility).13In contrast tothe large power factor s S2 of CoSb3, kL is high. However, the weak point can be improved by partially filling guest ions into the vacancy☐of ☐Co4Sb12. It is conceived that the guest ions located inside the vacancyexhibita random motion (rattling) around the equilibrium positions.2Then, the rattling causesa significant reduction of kL by a marked phonon scattering. Actually, it was reported that the vacancy of CoSb3 could be partially filledby rare-earth (R) ionsand a significantreduction ofkL were observed for RxCo4Sb12.13,14In order to cause a significant rattling effect, smaller andheavier ions such as heavy rare-earth ionscould bemore efficient. Therefore, heavier R ions could be desirable for guest ions filled the vacancy of CoSb3. However, synthesis of heavy rare earth (except for Yb) filled CoSb3 has not been reported so far whileRions such as La, Ce, Nd and Eu havebeen successfully filled into the voids of CoSb3by means of conventional methods at ambient pressure.13-16Mei et al.have carried out a systematic theoreticalresearch of the filling fraction limit (FFL) of Ratoms in CoSb3at ambient pressure.17Based on their calculations, the FFLdecreases rapidly from La to Sm, and becomes zero for heavy rare earths from Gdto Lu, with the exception of Eu and Yb. However, high pressurecouldbenefit the entrance of heavier rareearths into the vacancy of CoSb3 than ambientpressure. Therefore, we have currentlytriedto synthesizepartially heavy rare earth (Gd-Lu) filled CoSb3, systematically. Among them, we introduce the results of partially Tb (which is one of heavy rare earth elements) filled CoSb3 (TbxCo4Sb12) by the HTHP synthesis as an example.As far our knowledge, TbxCo4Sb12 is a new compound.

     Then, we focus on layered rare-earth zinc phosphides RZn3P3 (R = rare-earth element). The crystal structure of layered rare-earth zinc phosphides RZn3P3 (R = rare-earth element) is hexagonal (ScAl3C3 type), space group is P63/mmc(D46h,No. 194).18The Zn and P atom form hexagonal net, wherein the Zn and P atoms alternate. The R atoms form layers separating the Zn-P nets and a two-dimensional triangular lattice. Therefore, the compounds are expected as a frustrated system. Since the discovery of the giant magnetoelectric effects in a perovskite manganite TbMnO3,19multiferroicsstudies have become active. Multiferroics have potential for applications as next generation devices such as magnetic field sensors, actuators, switching device and new types of memory.20Many maltiferroics have been discovered in magnetic materials with a geometrical frustration in the crystal structure.19,21,22 There is a possibility that RZn3P3 (R = rare-earth element)isalso a magnetic frustrated system.However, the physical properties for most of RZn3P3except CeZn3P3have not been studied so far, while the synthesis of the phosphides by a flux method was reported.18It was reported that CeZn3P3 exhibits an antiferromagnetic ordering at TN = 0.8 K. The entropy of the magnetic contribution is ∼0.4Rln2 at TN, and reaches Rln2 around 10 K.23 This suggests that CeZn3P3 has the short-range magnetic interaction above TN and the compound could be a frustrated system.In order to study RZn3P3 systematically, we have tried to synthesize DyZn3P3 with a large magnetic moment of DyandYZn3P3 as a nonmagnetic reference under high pressure at the beginning. The high-pressure synthesis of RZn3P3 (R = rare-earth element) has not been reported so far. Therefore, we have performed in-situ XRD measurements to search the processes for synthesis of DyZn3P3 and YZn3P3 at HTHP.

     As a pressurization method for high-pressure experiments using quantum beams such as synchrotron X-ray, neutron, and muon, multi-anvil assembly 6–6 (MA6–6) with a DIA-type cubic-anvil high-pressure apparatusis becoming to be a standard.24 MA6–6 consists of six small second-stage anvils with an anvil guide. Recently, we have developed a new anvil guide, which can make it much easier to perform the experiments using MA6-6. In this paper, we also present a detailed design of the new anvil guide for MA6-6 and report results of pressure generation tests.

 

2. Experimental

     In-situ XRD measurements at HTHP have been performed atPhoton Factory (PF) in High Energy Accelerator Research Organization (KEK) (Tsukuba, Japan).XRD patterns were taken by an energy-dispersive method using synchrotron radiation (white X-ray) and a solid-state detector at the beam line AR-NE5C.8-10 High pressure was applied using multi-anvil assembly 6–6 (MA6–6) with a DIA-type cubic-anvil high-pressure apparatus, the MAX80 system, installed at the beam line. MA6–6 consists of six small second-stage anvils with an anvil guide and can be compressed by a DIA-type cubic-anvil apparatus.24 We have to set the second stage anvils at the accurate positions of the DIA geometry, carefully. Therefore, an anvil guide is really needed. The anvil guide (frame) is generally made of tool steel (SUS304) andthe holes are opened at the frame along one of the diagonal direction for access to the incident and diffracted X-rays.8 Fig. 1(a) shows a schematic illustration of MA6-6

Fig. 1 Schematic illustration of MA6-6 using a regular steel frame (a) and MA6-6using aplastic frame (b) 1, sample cell; 2, anvil guide (steel frame); 3, spacer glued to the anvil guide (balsa); 4, spacer glued to the anvil (Teflon); 5, anvil guide (plastic frame); 6, second stage anvil.

using a conventional steel frame. The spacers (Teflon) are necessary in setting the second stage anvils at their initial positions in the DIA geometry. Furthermore, we glue four spacers (balsa) on each outer surface of the anvil guide. When pressurizing is carried out, these spacers make it possible thatsecond stage anvils contactwith the first stage anvils tenderly becausethe first stage anvils compress the spacers first and then contact the second stage anvils.

     Although the use of MA6-6 simplifies the anvil replacement process, we have to still consume time and labor for setting, namely, we have to attach an electric insulating tapes on the surfaces of the steel frame and build up many parts (spacers). This process takes long time (several hours) because the anvil positions should be adjusted carefully. Therefor, we developed an integrated-type plastic frame (Fig. 1(b)) using 3D printer (KEYENCE, AGILISTA-3200). Fig. 2(a) shows a design of a plastic anvil guide integrated with spacers, which are also made of the same plastic. It is our original idea that plastic spacers are integrally molded on the plastic anvil guide using 3D printer. Fig. 2(b) and (c) show photographs of the plastic frame made by 3D printer and MA6-6 with the plastic frame set on the first stage anvils of a DIA-type cubic-anvil high-pressure apparatus, the MAX80 system, respectively. The plastic anvil guide with spacers can make it much easier to assemble MA6-6 because we do not have to attach electric insulating tapes on the surfaces of the anvil guide and do not have to glue spacers on the anvils and the anvil guide. We can just insert the anvils to the frame andthe anvils locate at accurate positions automatically.

Fig. 2. Design of a plastic frame integrated with spacers (a) a photo of the plastic frame made by 3D printer (b), and a photo of MA6-6using theplastic frame set on the first stage

Pressure generation of the MA6-6 with the plastic frame was tested using the MAX80 system installed at PF, KEK. Pressure was determined by the lattice constant of the NaCl. Pyrophyllite cube with edge length of 7 mm was used as the pressure medium. The truncated edge length (TEL) of the second-stage anvil made of tungsten carbide (WC) is 4 mm. The anvil guidemade of plasticwith an outer edge length of 28 mm was used. The anvil guide made of tool steel (SUS304) was also used for comparison. The TEL of the first-stage anvil is 27 mm.Fig. 3 shows the determined pressure calibration curves. The plastic frame can be applied without problems while the curve for the plastic frame is slightly less efficient than that for a regular steel frame.

Fig. 3. Generated pressure plotted as a function of applied load for a combination of MA6-6 with the plastic frame and DIA-type cubic-anvil high-pressure apparatus. The TELof the second stage anvil is 4 mm. The anvil guide made of tool steel (SUS304) was also used for comparison.

     The sample cell assembly of the in-situ XRD measurements for rare-earth transition-metal pnictides is almost the same as that in previous report.8 Thestarting materials, which are mixture of each element chips or powders, are put in a boron nitride (BN) crucible. The crucible with a graphite electric heater is inserted in a cubic solid pressure medium made of pyrophyllite. Pressure is determined by the lattice constant of the NaCl internal pressure marker.

      In order to prepare large bulk samples at HTHP, adouble-stage multi-anvil high-pressure apparatus (Kawai type system) was used (quench experiments).8Pressurizing method of the Kawai-type system is as follows.First,a hydraulic ram drives six first-stage anvils. Then, the six first-stage ones compress eight second-stage cubic anvils. Finally, the second-stage ones compress an octahedral solid pressure medium made of magnesia (MgO + 5% Cr2O3). For the second-stage anvils, WC cubes with a TEL of 11 mm wereused. The sample cell assembly is similar to that used for in-situ experiments. The pressure medium was transformed into an octahedron witha length of each side of 18 mm. The starting substances, which are mixture of each element chips or powders, were put in a crucible made of BN. The crucible, with a graphite heater surrounded with a zirconia (ZrO2) thermal insulator, was inserted into themagnesia octahedron. The prepared samples were characterized by powder x-ray diffraction using Co Kα1 radiation and silicon as a standard.
 

3. Results and discussion

    In the following sections, results of the in-situ XRD of synthesizing process under HTHP for skutterudites and layered rare-earth phosphides RZn3P3 will be reported.

3.1 Skutterudite

    We have performed in-situ XRD measurements of synthesizing process for CoSb3 as a test run of MA6-6 with the plastic frame. Fig. 4 shows XRD patterns of synthesizing process of CoSb3 at 2 GPa. Fig. 4(a) depicts XRD pattern of the starting materials, which are mixture of Co and Sb powders in the atomic ratio of Co : Sb = 1 : 3, at room temperature (RT) and ambient pressure. Solid triangles and solid

Figure4. XRD patterns of synthesizing process of CoSb3 at 2 GPa. Solid triangles and circles indicate the Bragg peaks of Co and Sb, respectively. Solid squares designate the characteristic x-ray of Sb. Numbers in the spectra are the Miller index of the Bragg peaks for CoSb3. Crosses show the peaks of a secondary phase. The starting materials at room temperature (RT) and ambient pressure (a), RT and 2GPa (b), 530 °C (c), 630 °C (d), 915 °C (e) and 1050 °C (f).

   circles indicate the Bragg peaks of Co and Sb, respectively. Solid squares designate the characteristicx-ray for Sb. Fig. 4(b) shows XRD pattern at RT after applying pressure up to 2 GPa. The peak positions of the starting materials shifted with increasing pressure and any additional diffraction with pressure was not observed. Accompanying the temperature rise, the Bragg peaks of the starting materials faded out, and then the peaks of the skutterudite structure was observed above 500 °C. Fig. 4(c) shows XRD pattern at 2GPa and 530°C. The diffraction peaks of CoSb3 were observed in addition to those of the starting materials. The numbers indicate the plane index of skutterudite structure. All diffraction lines were indexable using the skutterudite structure at 630 °C (Fig. 4(d)). Then, the peaks of a secondary phase (CoSb2), which are shown by crosses, appeared at 915°C (Fig. 4(e)). Further, most of the diffraction peaks disappeared and only the peaks of the characteristic x-ray of Sb were observed at 1050 °C (Fig. 4(f)). This indicates that the sample meltsat this temperature. By considering the results, we obtained an appropriate temperature range of 650-850 °C for synthesizing CoSb3 at 2GPa.This result is consistent with the previous report 9). Moreover, it could be confirmed that the MA6-6 system with the plastic frame is available for experiments at temperature up to 1050 °C.

     Then, we have performed in-situ XRD experimentsof synthesizing process for partially filled skutterudite compound TbxCo4Sb12 at 4 GPa. Starting substances are mixture of Tb chip, Co and Sb powders in the atomic ratio of Tb : Co : Sb = 0.5 : 4 : 12. By considering the results for a battery of the in-situ XRD measurements, we obtained an appropriate temperature range of 500-700 °C for synthesizing TbxCo4Sb12 at 4 GPa. Then, we have actually synthesized TbxCo4Sb12 at 4 GPa and 600 °C using a Kawai-type multi-anvil high-pressure apparatus according to the optimum condition acquired by the in-situ XRD experiments. Fig. 5 shows the XRD pattern of TbxCo4Sb12 prepared under high pressure using mixture of Tb chip, Co and Sb powder in the atomic ratio of Tb : Co : Sb = 0.5 : 4 : 12 as the starting materials. Themost of the observed diffraction lines were indexable using the skutterudite structure. The lattice constant determined by a least-squares fit to the data was 9.049 Å. The lattice constant of TbxCo4Sb12 is larger than that of CoSb3. This suggests that the filling of Tb to CoSb3 was succeeded. To determine the actual filling rate of Tb, the chemical compositional analysis was conducted using a scanning electron microscope (SEM) with an energy dispersive X-ray spectrometer (EDX) (JEOL 6510). For the point analysis, 8 to 10 different points were carefully chosen to reduce the errors. The actual filling ratioestimated by SEM-EDX was found to be x = 0.11. Although the theoretical calculations indicate that Tb with relatively smaller ionic radii cannot be inserted into the lattice voids of CoSb3 (FFL=0) at ambient pressure,17 we have succeeded in synthesizing 10% Tb filled Co4Sb12 under high pressure. TbxCo4Sb12 is a new compound.

3.2 Layered rare-earth zinc phosphide

      We have also performed in-situ experiments for layered rare-earth zinc phosphide DyZn3P3 at 4 GPa. Fig. 6(a) shows the XRD pattern of the starting substances, which are mixture of Dy chip, Zn and P powders in the atomic ratio of Dy : Co : Sb = 1 : 3 : 3, at room temperature (RT) and 4 GPa. Solid triangles and solid circles indicate the Bragg peaks of Zn and Dy, respectively. Solid squares designate the characteristic x-ray of Dy. Accompanying the temperature rise, the Bragg peaks of the starting substances faded out, and then the peaks of the ScAl3C3 type structure were detected above 900 °C. Fig. 6(b) shows the XRD pattern at 4 GPa and 990 °C. Most diffraction lines were indexable using the ScAl3C3 type structure while the peaks of secondary phases (ZnP2, Z3P2 and DyZnPO), which are indicated by crosses, were observed. We also performed the similar experiments for YZn3P3. By considering the results for a battery of the in-situ XRD measurements, we obtained an appropriate temperature around 980 amd950 °C for synthesizing DyZn3P3and YZn3P3at 4 GPa. Based on the results of in-situ experiments, we have succeeded in synthesizing large bulk samples ofYZn3P3 and DyZn3P3 under high pressure (quench experiment). Fig. 7 shows the powder XRD patterns of DyZn3P3 and YZn3P3synthesized under high pressure. Although a small quantity of impurity phases (Zn3Pand DyZnPO or YZnPO) was detected, most of the observed diffraction peaks were using the ScAl3C3 type structure.

Figure6. XRD patterns of synthesizing process of DyZn3P3 at 4 GPa. Solid triangles and circles indicate the Bragg peaks of Zn and Dy, respectively. Solid squares designate the characteristic x-ray of Dy. Numbers in the spectra are the Miller index of the Bragg peaks for DyZn3P3. Crosses indicate the peaks of secondary or impurity phases (ZnP2, Z3P2 and DyZnPO). The starting substances at room temperature (RT) and 990 °C (b).

Figure7. XRD pattern of DyZn3P3 (a) and Y Zn3P3 (b) synthesized under high pressure.Crosses indicate the peaks of secondary or impurity phases (Zn3P2 and DyZnPO or

      The lattice constants of the phosphides determined by a least-square fit to the data are summarized in Table 1. The obtained lattice constants are consistent with reported value.18

Tables 1 Crystal data and lattice parameters decided by XRD of layered rare-earth zinc phosphides RZn3P3 (R =Y and Dy) prepared under high pressure.

Compound

Crystal system

Space group

Lattice parameter

This work

Reference 18

a (Å)

c (Å)

a (Å)

c (Å)

YZn3P3

Hexagonal

P63/mmc

3.982

19.794

3.988

19.837

DyZn3P3

Hexagonal

P63/mmc

3.985

19.771

3.988

19.784


4. Conclusions

    In order to obtain the conditions optimal for synthesizing a new filled skutterudite compound TbxCo4Sb12 and layered rare-earth zinc phosphides RZn3P3 (R =Y and Dy) under high pressures, we have performed in-situ XRD experimentsat HTHP. We could obtain the optimum condition for synthesizing the compounds at 2~4 GPa.In this manner, in-situ XRDmeasurements at HTHP are splendid method for deciding the condition toacquire only target material free fromsecondary or impurity phases for solid-phase reaction synthesis under high pressure. Furthermore, we developed new integrated-type plastic frame for MA6-6 by using 3D printer. It was confirmed thatthe MA6-6 system with the plastic frame can be used without problem. This will make it much easier to assemble MA6-6.

 

Conflict of interest

There are no conflicts to declare.

 

Acknowledgements

This work was carried out using the facilities of the Institute for Solid State Physics, the University of Tokyo. This work has been carried out under the approval of the Photon Factory Program Advisory Committee (Proposal Nos. 2017G164 and 2015G031).

 

References

1. H. Sato, H. Sugawara, Y. Aoki, H. Harima, Handbook of Magnetic Materials, 2009, 18, 1-110.

2. G. S. Nolas, D. T. Morelli and T. M. Tritt, Annu. Rev. Mater. Sci., 1999, 29, 89-116.

3. R. D. Blaugher and J. K. Hulm, J. Phys. Chem. Solids, 1965, 26, 2037-2039.

4. H. Barz, H. C. Ku, G. P. Meisner, Z. Fisk and B. T. Matthias, Proc. Natl. Acad. Sci. USA, 1980, 77, 3132-3134.

5. I. Shirotani, M. Takaya, I. Kaneko, C. Sekine and T. Yagi, Solid State Commun., 2000, 116, 683-686.

6. I. Shirotani, M. Takaya, I. Kaneko, C. Sekine and T. Yagi, Physica C, 2001, 357-360, 329-332.

7. Z. Yu, Y. Wang, Z. Sun, X. Li, A. Wang, D. M. Camaioni and J. Lercher, Green Chem., 2018, 20, 609-619.

8. C. Sekine and Y. Mori, Jpn. J. Appl. Phys., 2017, 56, 05FA09.

9. C. Sekine, T. Kachi, T. Yoshida, R. Abe, T. Namiki, K. Akahira and K. Ito, J. Phys.: Conf. Ser., 2010, 215, 012141.

10. C. Sekine, H. Kato, M. Kanazawa, Y. Kawamura, K. Takeda, M. Matsuda, K. Kihou, C. H. Lee and H. Gotou, J. Phys.: Conf. Ser., 2014, 502, 012017.

11. N. Mandel and J. Donohue, Acta Crystallogr., Sect. B, 1971, 27, 2288-2289.

12. G. J. Snyder and E. S. Toberer, Nat. Mater., 2008, 7, 105-114.

13. G. S. Nolas, J. L. Cohn and G. A. Slack, Phys. Rev. B, 1998, 58, 164-170.

14. J. Yang, Q. Hao, H. Wang, Y. Lan, Q. He, A. Minnich, D. Wang, J. Harriman, V. Varki, M. Dresselhaus, G. Chen and Z. Ren, Phys. Rev. B, 2009, 80, 115329.

15. V. L. Kuznetsov, L. A. Kuznetsova and D. M. Rowe, J. Phys.: Condens. Matter, 2003, 15, 5035-5048.

16. W. Jeitschko, A. J. Foecker, D. Paschke, M. V. Dewalsky, C. B. H. Evers, B. Kunnen, A. Lang, G. Kotzyba, U. C. Rodewald and M. H. Möller, Z. Anorg. Allg. Chem., 2000, 626, 1112-1120.

17. Z. G. Mei, W. Zhang, L. D. Chen and J. Yang, Phys. Rev. B, 2006, 74, 153202.

18. A. T. Nientiedt and W. Jeitschko, J. Solid State Chem., 1999, 146, 478-483.

19. T. Kimura, T. Goto, H. Shintani, K. Ishizaka, T. Arima and Y. Tokura, Nature, 2003, 426, 55-58.

20. R. Ramesh and N. A. Spaldin, Nat. Mater., 2007, 6, 21-29.

21. K. Taniguchi, N. Abe, T. Takenobu, Y. Iwasa and T. Arima, Phys. Rev. Lett., 2006, 97, 097203.

22. Y. Yamasaki, S. Miyasaka, Y. Kaneko, J. P. He, T. Arima and Y. Tokura, Phys. Rev. Lett., 2006, 96, 207204.

23. A. Yamada, N. Hara, K. Matsubayashi, K. Munakata, C. Ganguli, A. Ochiai, T. Matsumoto and Y. Uwatoko, J. Phys.: Conf. Ser., 2010, 215, 012031.

24. N. Nishiyama, Y. Wang, T. Sanehira, T. Irifune and M. L. Rivers, High Pressure Res., 2008, 28, 307-314.