Investigations of crystalline phases in the melting of Bi2Sr2CaCu2Ox

Appried Suprconducriuify Printed in Great Britain

Vol. 1, No.

l/2, pp. 5MO.

0964-1807/93

1993

$5.00 + 0.00

Pergamon Press Ltd

INVESTIGATIONS OF CRYSTALLINE PHASES IN THE MELTING OF Bi, Sr,CaCu, 0, MINGXu, J. POLONKA, A. I. GGLDMAN and D. K. FINNEMORE Ames Laboratory, U.S. Department of Energy and Iowa State University, Ames, IA 50011, U.S.A. Abstract-The melting of Bi,Sr,CaCu,O, and Bi,Sr,CaCu,O,/Ag materials have been investigated by high-temperature X-ray diffraction, scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) in order to study the phases that formed during melting-processing. At least three distinct phases of (Sr, _xCa,)CuO,, (Sr, _,Ca,),CuO, and (Sr, _,Ca,)O have been observed in the Bi-rich matrix depending upon operating temperatures. Either the addition of Ag or processing in a pure N, atmosphere decreases the melting point by about 3WC. Crystallization from the melt by fast cooling usually results in the coexistence of B&compounds and these Sr-CaCu-O phases.

1. INTRODUCTION

Since Maeda et al. first discovered superconductivity in the Bi-Sr-Ca-Cu-0 compounds [l], the richness of both the chemistry and crystallography of these systems has been explored. It is well known that there exist three superconducting Bi-based phases: the 10 K phase, Bi,Sr,CuO, (2001) with a single Cu-0 layer; the 80 K phase, Bi,Sr,CaCu,O, (2212) with a double Ct.4 layer; and the 110 K phase Bi,Sr,Ca,Cu,O, (2223) with a triple Cua layer [2]. Bismuth based superconductors are some of the most promising materials for practical applications, especially for wire and coil fabrications. The partial melting and solidification technique for Bi-based compounds is known as a useful method to make c-axis oriented materials with a good value of critical current density (J,) [3-6]. During fabrication, since the Bi compound decomposes into liquid and solid phases at the partially-molten state, the liquid phase is very important to promote textured grain growth for high-quality materials in the temperature range 830-880°C. On the other hand, the viscosity of the B&based liquid is very high, on the order of 60 P at these temperatures, so it is not always apparent that a liquid has formed [7]. The high viscosity is also important for controlling the migration of the liquid. Thus, considerable interest has been raised in the many phases that form at these high temperatures in the fabrication of the high-quality Bi-(2212) or Bi-(2223) superconductors. In recent high-temperature X-ray diffraction studies [8-121 it has been shown that upon melting Bi-(2212) material several different insulating phases coexist with in the liquid. Oka et al. first reported that two phases of (Sr,_.Ca,)CuO, or (11) and (Sr,_.Ca,),CuO, or (21) were observed in the melted state depending upon the starting compositions of Bi-(2212) and Bi-(4336) [8]. It is unclear, however, how the addition of Ag and variations in an atmosphere effect the formation of these phases. The purpose of this work is to investigate the crystallization behavior, phase formation and the factors controlling the phase growth during melting at high temperatures in the Bi-(2212) system. The phases present in the melted state were investigated by high-temperature X-ray diffraction, SEM and energy dispersive spectroscopy for Bi-(2212) and Bi-(2212)/Ag in air, and B&(2212) in an atmosphere of pure N2 gas. 2. EXPERIMENTAL

DETAILS

The BL(2212) materials used in the high-temperature X-ray diffraction experiments were prepared using a solid-state reaction technique reported elsewhere [6]. The sample crucible is a Pt strip about 0.15 mm thick, 10 mm wide and 100 mm long with an indentation about 8 x 8 mm in the center to hold the B&(2212) powder. The sample region of the hot stage usually was coated with Au to inhibit the formation of possible Pt compounds. A chromel-alumel thermocouple was spot welded to the Pt boat in the center for temperature determination. High-temperature X-ray data were taken with a rotating anode Cu Ka source and a linear position-sensitive detector. 53

MING

54

Xv et al.

With the scattering plane vertical, X-rays were directed onto a graphite monochromator and impinged on the sample at an angle of 26.5” to the horizontal. The linear position detector covered a range of 28 from 30 to 58”. It was sometimes moved to a higher angle range to observe high-angle diffraetion peaks to help in the identification of phases. In SEM and EDS studies, the samples were quenched from high t~m~ratures by switching off the power supply, while (Sr,_,Ca,)CuO, phase, (Sr,_,Ca,),CuO, or (Sr,_,Ca,)O phase diffraction peaks were observed in the high-temperature X-ray detector. Microstructure and phase

895*C,

3 min

8 8O”C, 6 min

25

30

35

40

45

50

55

60

28 Fig. f. Hid-tem~rat~~ X-ray detraction patterns for the melting B&(2212) at various tem~ratures in air: three different phases of (Sr,_,Ca,)CuO, or (1 I), (Sr,_,CaJzCu03 or (21), and (Sr, _,Ca.JO or (10) formed as indicated. lottam pattern was taken at room temperature indicating a pure Si-(2212) phase.

55

investigations of crystalline phases

composition were examined by a Cambridge S-200 SEM equipped with a Tracer-Northern energy dispersive spectrometer (EDS) operated at 15 kV. Chemical analysis was carried out on the EDS spectrum with a standardless semi-quantitiative analysis (SSQ) program. 3. RESULTS

AND

DISCUSSIONS

3.1. In situ X-ray dijfiiaction

As shown in Fig. 1, a typical melting sequence for Bi-(2212) powder shows that the different phases appeared at high temperature in air. The scans are 3 min long and were taken while the boat was held at the constant temperature marked on each pattern. Starting at 800°C two scans (6 min) were taken at 10°C intervals up to 860°C. At 865°C the spectrum is essentially the same as at room temperature. In the first 3-min scan at 870°C the Bi-(2212) diffraction peaks collapse and two strong lines grow at 44.0 and 51.3”, indicating Bi-(2212) phase melting and a new phase developing at 870’C. These lines belong to the (0,8,0) and (2.0.0) lines of a Sr rich orthorhombic

87O”C,

3 mln (10)

(10)

860°C,

3 min

85O”C,

3 min

84O”C,

3 min /\

-

-

-

A

Gw

A

Ag

Ag

50

55

60

Fig. 2. High-tempera&ure X-ray diffraction pastterns for the melting Bi-(2212)jAg at various tem~ratures in air: various phases formed as indicated.

56

MING Xu et al.

(Sr, _.Ca,)CuO, or (11) phase determined by SSQ chemical analysis, where the longest lattice constant has been chosen to be the &axis (the space group is CmCm) [13,14]. A line at 54.0” also starts to grow at 870°C as shown in Fig. 1. After two scans or 6 min at 870°C the peaks of the (11) phase diminished and the lines at 42.8 and 54.0” grew very rapidly at 880°C. These two line are (6,0,0) and (0,0,2) lines of the orthorhombic (Sr, _,Ca,),CuO, or (21) phase, where the longest lattice constant is chosen as the u-axis (the space group here is Immm) [14-161. These identifications can be confirmed by detailed SEM studies as shown below. After two scans or 6 min at 880°C the temperature was stepped to 890, 895, and 900°C for the 3-min scans. The (21) phase diminished and two new peaks at 31.9 and 36.7” started to grow. Peaks well also found at 62.8 and 65.9” by shifting the detector to the high angle range. Thus, in combination with the 52.9” peak, these peaks can be identified as the rock-salt structure of (Sr,_,Ca,)O with (Sr/Ca) ratio of about l/3. The effect of Ag addition to Bi-(2212) on the melting point was tested in air by in situ X-ray scans as shown in Fig. 2. A fine dispersion of metallic Ag with a particle size of about 1 ,um was micromilled with Bi-(2212) powder. Three-minute scans were taken at 10°C intervals in the range 800-870°C. As the sample was heated in a Au-plated Pt boat to 600°C the powder settled in the boat, causing a shift in the lines by about 0.4”. The Bi-(2212) peaks disappeared between 840 and 850°C in successive 3-min scans at 10°C intervals from 820 to 870°C indicating that the melting point of the Bi-(2212)/Ag composites was about 840°C. Between 850 and 870°C the (21) peak at 54.0” rises and falls. At 870°C the (10) peaks at 31.8 and 36.7” are clearly visible. Melting of Bi-(2212) in an atmosphere of a pure N, gas was also investigated by in situ X-ray diffraction. The melting started at 830°C which is about 40°C lower than in air. After melting, several phases were observed including Sr-Ca-Cu-0 phases plus unknown phases. Melting is a more complicated process in an N, atmosphere. After the sample had been taken through a similar sequence to that in air, then quenched from 750°C SEM studies indicated that the (Sr, _ xCa,)O or (10) phase remained during this process in N, . 3.2. SEA4 and EDS studies Extensive SEM and EDS studies have been carried out on the samples quenched from high temperatures either in air or N, gas. The first sample was quenched in air from a temperature of about 870°C as (Sr, _.Ca,)CuO, or (11) phase peaks appeared in the X-ray spectrum. As can be seen in Fig. 3, rectangular crystalline grains are embedded in Bi-rich matrix. EDS analysis reveals that this rectangular needle is the (11) phase and the Bi-rich matrix is the Bi-(2201) phase. From But, it should chemical analysis, the ratio of Sr/Ca is found to be x N 0.33 in (Sr, _,Ca,)CuO,.

Fig. 3. SEM (backscattering

electron)

photograph for quenched samples embedded in Bi-rich matrix.

in air: (Sr, _,Ca,)CuO,

phase

Investigations

Fig. 4. SEM (backscattering

electron)

of crystalline

phases

photograph for quenched samples embedded in a Bi-rich matrix.

in air: (Sr,

,Ca,),CuO,

phase

be pointed out that this semi-quatitative chemical analysis is not quite accurate since the sample surface is not smooth nor well-polished. While the temperature is raised to about 870°C in air, most X-ray peaks of the (11) phase disappear and peaks belonging to (Sr, _.xCa,r),CuO, or the (21) phase start to develop. The second sample was quenched from these conditions. As shown in Fig. 4, there are still rectangular crystalline needles embedded in the Bi-rich matrix. However, the shapes of these needles are different from the (11) phase. These needles seem to be narrower and sharper than the (11) phase, as shown in Fig. 3. EDS studies indicate that the narrower and sharper rectangular needles belong the (21) phase and the Bi-rich matrix is still the Bi-(2201) phase. It should be pointed out that although the dominant phase is the (21) phase, the (11) phase is still maintained as a minor phase. EDS analysis indicates that the (21) phase has a composition of .Y - 0.55 in (Sr, ,Ca,),CuO,. As shown in Fig. 5, the sample quenched in pure Nz gas from 750°C contains the cubic (Sr, rCa,)O or (10) phase, which are dark clusters in this SEM picture. Besides (10) phase particles, (21) phase needles are also observed, which are embedded in the Bi-rich matrix. Thus, the pure N? gas not only lowers the melting point of Bi-(2212) but also helps the formation of the (10) phase.

Fig. 5. SEM (backscattering electron) photograph for quenched samples in N2 gas: (Sr, _., Ca,)O is a major phase in the dark cluster shape; and (Sr, _.qCa,),CuO, phase is also observed in a dark needle-like shape, both of them embedded in the Bi-rich matrix.

MING Xu et al.

58

3.3. Phase formations

It should be noted that although both (11) and (21) phases are orthorhombic structures, they belong to different space groups. As temperature increases with liquid present, more Cu is rejected from the existing (11) phase to the Bi-rich liquid. Thus, the (21) phase can be formed. Eventually the (10) phase can be formed after completely losing Cu in the (21) phase. During cooling, the formation of the Bi-(2212) phase from the Bi-rich liquid is inhibited by the formation of (Sr,_,Ca,)CuO,, (Sr,_.Ca,),CuO, or (Sr,_.Ca,)O phases since Ca is not rich in the liquid. It will be helpful to illustrate the crystalline structures of these phases including Bi-(2212), (11) and (21) compounds. As can be seen in Fig. 6a, a B&(2212) unit cell is illustrated here, which is basically a layered structure. After melting, BiO and Ca layers may be immediately rejected to Bi-rich liquid. Figure 6b is not a physical process, but rather an illustration of the intermediate picture to help in the understanding of phase transformation. It is clear from Fig. 6 that CuO, layers lose 0 and are changed to CuO chains. There is also some rearrangement in Sr and 0 sites for SrO layers. After the unit cell is compressed, the (11) phase could be formed in this way, as shown in Fig. 6c. It can be noted that there exists a layered feature for SrO layers, in contrast to CuO chains. The transformation from the Bi-(2212) phase to (21) phase can be applied in the same way as shown in Fig. 7. After pulling away BiO and Ca layers and rejecting one CuO, layer in Bi-(2212), the Sr (21) phase could be formed once CuO, planes become CuO chains. The transition from the (2212) to (21) phase seems more straightforward than that from the Bi-(2212) to the (11) phase. But, since two more CuO, layers need to be rejected from Bi-(2212) to form the (21) phase, more energy should be required to do this. Therefore, we usually observed that the (11) phase is always formed first after the melting of Bi-(2212). In Fig. 8, we plot an illustration for the transformation from the (11) to (21) phase. It is apparent that the transformation happens if more CuO chains in planes are rejected from the (11) phase and the sites of Sr and 0 in SrO layers are rearranged. It is interesting to note that the longest lattice parameter for Bi-(2212) is about 30.5 A as the longest lattice parameters for the (11) and (21) phases are 16.3 and 12.8 A, respectively. There are 14 layers including BiO, SrO, Ca and CuO, layers in the unit cell of Bi-(2212). The inter-layer distance is about 2.3 A for Bi-(2212). Since there are nine layers and seven layers in one unit cell for the (11) and (21) phases, respectively, the inter-layer distances for both phases (-2.2-2.4 A) (a) Bi (2212)

(b) possible inter

(cl Sr (11) Q Bi 0 Ca 0 Sr

For Sr (11):

a = 3.575 A b = 16.34 A c - 3.918 A b

k

Fr

C a

B; (2;li):

- . c = 30.52

A

a Fig. 6. Illustration for the phase transformation from Bi-(2212) to Sr-(1 I): (a) unit cell of Bk(2212); (b) intermediate structure during the transition, which is not a real phase; and (c) unit cell of Sr-(1 1) phase.

59

Investigations of crystalline phases

(b) possibleintermediatepicture

For Sr (21) : a 7 12.77 A b = 3.926 A t = 3.511 A C

L

b

Fig. 7. Illustration for the phase transformation from Bi-(2212) to Sr42lf: (a) unit cefl of Bi-(2212); (b) intermediate structure during the transition, which is not a real phase; and (c) unit cell of Sr-(21) phase.

are basicafly the same as for the I%-(2212)phase.Tkxefore, the picture of rejecting the whole iayer during phase transition coutd be acceptable, at least in helping ~nd&rstanding the melting process of B&(2212). 4. CONCLUSIONS

We have shown that the melting of B&(2212) occurs in several steps and at least three phases form in sequence upon material composites, atmosphere and temperatures, first (Sr, _ .Ca,)CuQ, or the (11) phase, then (Sr, _,VCa,)2Cu0, or the (21) phase, finally (Sr, _,Ca,)O or the (10) phase, (a) Sr (11)

(b) Sr (21)

cu

0 00

For Sr (11) a = 3.575 A b = 16.34 A c = 3.918

A

Fur Sr (21) a = 12.77 b = 3.926

A

A c= 3.511 A a

Fig, 8. Illustration for the phase transformation from Sr-(I 1) to Sr-(21): (a) unit cell of Sr-(1 I) and (b) unit ceil of Sr-(21) phase.

MING Xv et al.

60

and the Bi-rich liquids (2201) and (2212). The SEM and EDS results are in very good agreement with the high-temperature X-ray data. These investigations reveal that the (11) and (21) phases are in the shape of rectangular needles floating on the surface of the Bi-rich liquid and that the Bi-rich liquid crystallizes into the Bi-(2201) and (2212) phases by quenching in air. The addition of Ag to Bi-(2212) lowers the melting point of the composites. The pure Nz gas results in both lowering the melting temperature and helping the formation of the (10) phase in Bk(2212). These phases are very important in controlling the formation and growth of the Bi-(2212) phase from the melting state. Acknowledgements-Ames Laboratory is operated for the U.S. Department of Energy by Iowa State University under Contract No. W-74O5-ENG-82. This work is supported by the Department of Energy Grant No. DE-FGO2-9OER45427 through the Midwest Superconductivity Consortium and the Office of Basic Energy Sciences.

REFERENCES H. Maeda, Y. Tanaka, M. Fukutomi and T. Asano, Jpn J. appf. Phys. 27, 209 (1988). J. M. Tarascon, W. R. Mckinnon, P. Barboux, D. M. Hwang, B. G. Bagley, L. H. Greene, G. W. Hull, Y. L. Page and M. Giroud, Phys. Rev. B-38, 8885 (1988). K. Sato, T. Hikata, H. Mukai, M. Ueyama, N. Shibuta, T. Kato, T. Masuda, M. Nagata, K. Iwata and T. Mitsui, IEEE Trans. Magn. MAG 27, 1231 (1991). J. Kase, K. Tagano, H. Kumakura, D. R. Dietderich, N. Irisawa, T. Morimoto and H. Maeda, Jpn. J. Appl. Phys. 29, L1096 (1990); also H. Kumakura, K. Tagano, H. Maeda, J. Kase and T. Morimoto, Jpn. J. appl. Phys. 29, L1096 (1990).

5. D. Shi, S. Salem-Sugui Jr, 2. Wang, L. F. Googrich, S. X. Dou, H. K. Liu, Y. C. Guo and C. C. Sorrel], Appl. Phys. Letr. 59,317l (1991); also S. X. Dou, H. K. Liu, M. H. Apperley, K. H. Song and C. C. Song, Supercond. Sci. Technol. 3, 138 (1990). 6. Q. Li, J. E. Ostenson and D. K. Finnemore, J. appl. Phys. 70, 4392 (1991). I. S. E. LeBeau, J. Righi, S. C. Sanders, J. E. Ostenson and D. K. Finnemore, Appl. Phys. Lett. 58, 292 (1989). 8. Y. Oka, N. Yamamoto, Y. Tomii, H. Kitaguchi, K. Oda and J. Takana, Jpn. J. appl. Phys. 28, L213 (1989); also Y. Oka, N. Yamamoto, H. Kitaguchi, K. Oda and J. Takana, Jpn. J. appl. Phys. 28, L801 (1989). 9. D. P. Matheis and R. L. Snyder and C. R. Hubbard, In Superconductivity and Its Applicarions (Edited by Y.-H. Kao, P. Coppcns and H.-S. Kwok), AIP Conf. Proc. 219, Buffalo, New York (1990). 10. M. F. Garbauskas, R. H. Arendt, J. D. Jorgenson and R. L. Hitterman, Appl. Phys. L&t. 58, 2987 (1991). 11. J. Polonka, M. Xu, A. I. Goldman, D. K. Finnemore and Q. Li, Supercond. Sci. Technol. 5, Sl57 (1992); also J. Polonka, M. Xu, Q. Li, D. K. Finnemore and A. I. Goldman, Appl. Phys. Lea. 59, 3640 (1991). 12. T. Hasegawa, T. Kitamura, H. Kobayashi, H. Kumakura, H. Kitagushi and K. Togano, Appl. Phys. L&t. 60, 2692 13. 14. 15. 16.

(1992). G. L. Teske and H. Muller-Buschbaum,

2. Anorg. A&. Chem. 371, 325 (1969); also 379, 234 (1970). K.-H. Breuer, W. Eysel and M. Behruzi, Z. Krisrallogr. 176, 219 (1986). M. Xu, E. T. Voiles, L. S. Chumbley, A. I. Goldman and D. K. Finnemore, J. Mater. Res. 7, 1283 (1992). A. M. M. Gadalla and J. White, Trans. Br. Ceram. Sot. 65, 181 (1966).