JOURNAL OF SOLID STATE CHEMISTRY ARTICLE NO.
133, 198—200 (1997)
Single Crystal Growth and Properties of Incongruently Melting TbB6 , DyB6 , HoB6 , and YB6 K. Takahashi* and S. Kunii Department of Physics, Tohoku University, Aramaki, Aoba-ku, Sendai 980-77, Japan; and *Institute for Material Research, Tohoku University, Katahira, Aoba-ku, Sendai 980-77, Japan Received April 7, 1997; accepted April 10, 1997
Details of the single crystal growth of incongruently melting TbB6 , DyB6 , HoB6 , and YB6 using a crucible-free vertical floating zone method are reported. Magnetic susceptibility is reported and discussed. It is suggested that antiferro-quadrupolar ordering occurs in DyB6 between 30 and 25.6 K. ( 1997 Academic Press
Rare-earth hexaborides with a simple cubic CaB -type 6 structure have been attracting much attention because of their variety of electronic and magnetic properties. Recent development of the study is due to a single crystal growth technique that overcomes the very high melting temperature (&2500°C). Heavy rare-earth hexaborides melt incongruently in a condition which makes it difficult to grow big single crystals. We have, however, succeeded in growing big single crystals of them by a crucible-free vertical floating zone method. The details of sample preparation are similar to those described in Ref. (1) except for the starting composition of materials and the growing speed. The purpose of the present paper is to show (a) the details of the single crystal growth procedure and (b) the results of magnetic susceptibility measurements. SAMPLE PREPARATION AND CRYSTAL GROWTH
Incongruently melting rare-earth hexaborides were first obtained in powder form by the following reactions. Tb O #31B P4TbB #7BO and RE O #15B P 4 7 6 2 3 2REB #3BO (RE"Dy, Ho, and Y). Rare-earth oxide 6 powders (99.99%) were purchased from Japan Yttrium Company. For the borothermal reduction, we used an amorphous submicron boron powder of 99.999% purity, purchased from Aldrich Chemical Company (USA). In the case of 11-boron-enriched samples for neutron measurements, 99.5 at.% enriched 11B was purchased from Isotec (USA). Mixed powders of oxide and boron were pressed into a square rod (10]10]200 mm3) at 2.5 kg/mm2. The
reactions were performed at about 1650°C for 2 h, removing the vaporized BO in vacuum. An induction furnace was used with a graphite susceptor surrounded outside by a carbon felt for temperature insulation. To avoid graphite contamination we used a boron-nitride pipe surrounding the powder rod. By the above procedure, however, we always got a mixture of three compounds, REB , REB , and REB , 4 6 12 due to the incongruently melting property. Based on diffusion and reaction kinetics, the intensity ratio among the three types of diffraction lines is changeable depending on the reaction temperature. Therefore we can arrange the overall composition of the mixed compounds at P on the liquidus curve of Fig. 1. The powder of the mixed compounds was pressed hydrostatically into a rod of 10 mm £ (diameter)]100 mm at 2.5 kg/mm2 using a special rubber scak and then sintered at 1700°C for 2 h in the same surface. Using the sintered rod obtained above, the incongruently melting rare-earth hexaborides have been grown by a crucible-free vertical floating zone method under 1 MPa pressurized high-purity (99.9995%) argon. The single crystal growing speed was 4 mm/h. Figure 2 shows the cross section of the growing crystal; the total length of it is about 10 cm. There is a surface skin (&1 mm thick) indicated in Fig. 2 that is always formed when growing REB crystals. Only in6 side is the single crystal of REB . REB and REB com6 4 12 pletely disappeared and only REB remains. In Fig. 1, REB 6 6 is not a line compound; that is, REB has some width of 6 composition. However, Fig. 1 represents the situation under normal pressure (1 atm), whereas the rare-earth element is always defective (see B/REB "85.7% ("6/7)&89% in the 6 DyB case). In our floating zone procedure, we use 1 MPa 6 of pressurized argon, preventing rare-earth defects. This is confirmed by lattice parameter (a ) measurements in which 0 (a ) is consistent with stoichiometric REB (3). 0 6 RESULTS AND DISCUSSIONS
To know the magnetic ground state, magnetic susceptibility and specific heat measurements have been carried out.
198 0022-4596/97 $25.00 Copyright ( 1997 by Academic Press All rights of reproduction in any form reserved.
PROPERTIES OF TbB , DyB , HoB , AND YB 6 6 6 6
FIG. 1. The phase diagram of the Dy—B system after (2).
We have already reported on the electrical resistivity of REB (RE"Tb, Dy, and Ho) (4) and YB (5) and on the 6 6 specific heat (4, 6). Some investigations (7—9) of the magnetic properties of heavy rare-earth hexaborides have been carried out earlier except for HoB and at rather high tem6 peratrue. Our susceptibility measurements were performed using both a magnetic balance and a magnetic pendulum. Figures 3—5 show the temperature dependence of the inverse susceptibility for TbB , DyB , and HoB , respectively. 6 6 6 Curie—Weiss behavior is obeyed at higher temperature, and the effective Bohr magnetons are near the free-ion values (Table 1). This means the electric crystal field splitting of the ground state is within the energy of the corresponding linear-deviation temperature indicated by arrows in Figs. 3—5, which is consistent with the entropy analysis of the
FIG. 2. The schematic growing crystal.
FIG. 3. Temperature dependence of the inverse susceptibility of TbB . 6 (Inset) Magnetization M (in units of k /Tb) versus field H (tesla) at 4.2 K. B
specific heat measurements (10). A detailed specific heat analysis will be published in a forthcoming paper. For DyB there are two anomalies indicated by the ar6 rows in Fig. 4. From recent X ray and neutron diffraction measurements at various temperatures (11), the anomaly at the higher temperature (30 K) was found to correspond to a cubicPrhombohedral structure transition. The other anomaly at 25.6 K corresponds to the antiferromagnetic transition temperature revealed by our recent neutron diffraction measurement (11). At the higher transition temperature (30 K), a Jahn—Teller distortion is induced and the degeneracy of the magnetic ground state is removed. At this
FIG. 4. Temperature dependence of the inverse susceptibility of DyB . 6 (Inset) Magnetization M (in units of k /Dy) versus field H (tesla) at 4.2 K. B
TAKAHASHI AND KUNII
TABLE 1 Crystallographic and Magnetic Data on Rare-Earth Hexaborides Boride
a (As ) 0
¹ (K) N
TbB 6 DyB 6 HoB 6 YB 6
4.105 4.097 4.095 4.100
f8 f9 f 10 f0
19.5 25.6 5.6 (7.5)a
9.721 10.646 10.607
Exp. (k ) B
9.3 10.2 10.1
!22.9 !21.7 !16.3
a For YB , 7.5 is the superconducting transition temperature. The mag6 netization curve of YB shows a clear type II superconducting behavior, 6 that is, perfect diamagnetism M"!H/4 n due to Meissnher effect was observed below H . C1
FIG. 5. Temperature dependence of the inverse susceptibility of HoB . 6 (Inset) Magnetization M (in units of k /Ho) versus field H (tesla) at 4.2 K. B
temperature (30 K) the inverse magnetic susceptibility suddently increases with decreasing temperature in spite of no induced magnetic moment which is ascertained by the neutron diffraction measurement. Therefore susceptibility measurement suggests that the higher transition temperature corresponds to the antiferro-quadrupolar ordering temperature (12), in contrast to the ferro-quadrupolar moment of CeB in which the magnetic susceptibility increases 6 with decreasing temperature at the quadrupolar transition temperature of 3.2 K.
ACKNOWLEDGMENTS We are very grateful to Prof. M. Motokawa for useful and helpful discussions. This work was supported by the Grant-in-Aid from the Japanese Ministry of Education, Science, and Culture.
REFERENCES 1. S. Kunii, J. Phys. Soc. Jpn. 57, 361 (1988). 2. T. B. Masalski (Ed.), ‘‘Binary Alloy Phase Diagrams,’’ Vol. 1, p. 352. American Society for Metals, OH, 1986. 3. K. E. Spear, Boron, preparation, structure and properties. Ed. Sci. Moscow 207 (1974). 4. K. Segawa, A. Tomita, K. Iwashita, M. Kasaya, T. Suzuki, and S. Kunii, J. Magn. Magn. Mater. 104–107, 1233 (1992). 5. S. Kunii, T. Kasuya, K. Kadowaki, M. Date, and S. B. Woods, Solid State Commun. 52, 659 (1984). 6. K. Iwashita, T. Matsumura, K. Segawa, and S. Kunii, Physics B 186–188, 636 (1993). 7. Yu. B. Paderno, S. Pokrzywvicki, and B. Stalinski, Phys. Stat. Sol. 24, K73 (1967). 8. B. T. Mattias, T. H. Geballe, K. Andres, E. Corenzwit, G. W. Hull, and J. P. Maita, Science 159, 530 (1968). 9. Z. Fisk, R. H. Taylor, and B. R. Coles, J. Phys. C 4, L292 (1971). 10. S. Kunii and K. Iwashita. [in preparation] 11. K. Takahashi, H. Nojiri, M. Motokawa, K. Ohoyama, Y. Yamaguchi, and S. Kunii. [to be published] 12. P. Morin and D. Schmitt, ‘‘Ferromagnetic Materials’’ (K. H. J. Buschow and E. P. Wohlfarth, Eds.), Vol. 5, p. 34, Elsevier Sci., New York, 1990. 13. M. Kawakami, S. Kunii, T. Komatsubara, and T. Kasuya, Solid State Commun. 36, 435 (1980).