J. MATER. CHEM., 1994,4(11), 1745-1748 Formation and Decomposition of LaBa,Cu,O, -& J. M. S. Skakle and A. R. West Department of Chemistry, University of Aberdeen, Meston Walk, Aberdeen, UK AB9 2UE The formation and decomposition of LaBa,Cu,O, -,has been studied by X-ray diffraction and thermogravimetric analysis. The attempted preparation of LaBa,Cu,O,-, in air or oxygen gives a mixture of a solid solution of composition La, +xBa,-,Cu,07-d (xz0.2),and BaCuO, (J. M. S. Skakle and A. R. West, Physica C, 1994, 220, 187). However, heating this mixture in a less oxidising atmosphere such as flowing argon at temperatures above 850°C for a short time, ca. 16 h, yields a single-phase tetragonal product, which, when annealed in oxygen at lower temperature, gives an orthorhombic superconductor with T, (onset)z94 K.The orthorhombic and tetragonal structures have been confirmed by Rietveld refinement of X-ray powder diffraction data; an orthorhombic-tetragonal transition occurs at 6 =0.2. The stability of the single-phase material under air and oxygen has been studied by thermogravimetry. It is stable up to 800 "C in air; at higher temperatures, decomposition to La, +xBa2--xC~307-,(x %0.2), and BaCuO, occurs. On prolonged heating of single-phase tetragonal LaBa,Cu,O,-, at high temperatures, e.g. 875 "C for 64 h under argon, decomposition to give La,-,Ba,+,Cu,-,O,,-, (xz0.2),BaCuO, and BaCu,O, occurs. This is similar to the behaviour of YBCO under low oxygen pressure. Thus, under all conditions studied so far, LaBa,Cu,O,-, may be considered to be a non-equilibrium phase.In orthorhombic samples of LaBa,Cu,O,-, prepared by quenching, T, varies linearly with 6; samples cooled slowly show evidence of a plateau at ca. 94 K for 6%-0.05 to -0.15. The 93 K superconductor YBa,Cu,O, may be readily pre- pared by sintering in air or oxygen at ca. 950°C followed by a post-sinter anneal at 400 "C in oxygen. Rare-earth-metal analogues, REBa,Cu,O, (RE=Lu, Yb, Tm, Er, Ho, Dy, Gd, Eu and Sm) may easily be synthesized under similar con- ditions, but for the larger rare-earth metals, e.g. La, prep- aration of a single-phase 123 material in air or oxygen is not possible. This difficulty in preparation has been attributed to the ready formation of the tetragonal solid solutions La, +,Ba2-,Cu30, (x#O),l,, in which the similarity in the sizes of La and Ba permits cross-substitution on the cation sites.Accurate structure determination on La-123 has also proved difficult, since La and Ba have similar X-ray scattering factors,, and several refinements of powder diffraction data have apparently failed to account for the presence of BaCuO, as a second Some studies have reported the La-123 phase as tetragonal with T,<90 K (e.g.ref. 3 and 6), consistent with the formation of a solid solution member, x>O, rather than 123, whilst others have shown that preparation in air gives a mixture of a tetragonal solid solution composition, La, +xBa2-xCu30z and BaCuO, .7-9 More recently, it has been shown that under less oxidising conditions, such as heating under pure nitrogen or argon, a phase-pure 123 material can be synthesized, which, after low-temperature oxygenation, is orthorhombic with a T,of 93 K.lo," Another study has shown that in oxygen atmospheres richer than 50/002/N2, La-123 does not form.', In single-phase orthorhombic La-123, struc- tural studies have shown that the oxygen sites in the basal plane exhibit more disorder than in YBCO.' In particular, some oxygen is present in the (!LOO) sites while the (O%O) sites may not be fully occupied.Several studies have shown that excess oxygen (above 7) may be accommodated by increasing the occupancy of the (%OO) In an attempt to clarify the conditions of synthesis, we have studied the effects of different atmospheres on the formation and decomposition of La-123, together with the structure, stability, and superconducting properties of the single-phase material, and we report the results here.Experimental The starting materials were La,O, (Aldrich 99.99%) BaC0, (Aldrich 99.9%) and CuO (BDH 99.5%); the La,O, was dried at 1000°C prior to use, the CuO at 700°C and the BaCO, at 300°C. These were weighed out to give ca. 5 g rcaction mixtures and mixed together with acetone in an agate mortar and pestle. The mixture was pressed into 13 mm pellets, placed on pre-seasoned partially yttria-stabilized zirconia discs, heated in a muffle furnace at 900°C for 12 h to decarbonate it, then heated at 950°C for 24 h. Heat treatments under different atmospheres were carried out in a horizontd tube furnace.For quenching experiments, ca. 0.2g of the sample was placed in a Pt foil envelope and suspended by Pt thread in a vertical tube furnace; after 30min the sample was dropped into mercury, and measurements were carried out immediately. Phase identification was carried out using a Hagg (hinier camera with Cu-Ka, radiation, and data for unit-cell determi- nation and Rietveld refinements were collected using a Stoe STADI/P automated powder diffractometer in transmission mode with linear position-sensitive detector. Ac susceptibility measurements were carried out on powder samples itsing a Lakeshore AC7000 susceptometer. A Stanton Redcroft STA 1500 simultaneous TG-DTA instrument was used for t hermo- gravimetric measurements under different atmospheres, using a heating rate of 10 "C min-'.Results and Discussion Formation of LaBa,Cu,O, -The initial reaction in air or oxygen at temperatures ranging from 900 to 1000°C always gave a mixture of BaCuO, and a 123-related solid solution La, +,Ba2 -,Cu3OZ [Fig. 1(a)]whose composition was determined as x z0.2 from its lattice param- eter~.~The effect of heating this mixture under flowing argon was studied; heat treatment at 875 "C for 4-16 h gave single- phase La-123, as determined by X-ray diffraction. At 850 "C it was necessary to heat the same mixture for 25-40 h to give a phase-pure material, but at 825 "C stoichiometric La-123 did not begin to form even after 7 days.Structural Studies The oxygen content of a phase-pure sample heated at 875 "C for 16 h was determined as 6.49f0.03 by reduction in 10% H,-N, on the thermobalance. The structure was determined I A I, 20 30 40 50 60 2fldegrees Fig. 1 X-Ray diffraction patterns of LaBa,Cu,O,-d under different preparative conditions. (a) Air at 950 "C for 24 h, (6) Ar at 875 "C for 16 h, (c) Ar at 875 "C for 16 h followed by 0, at 400 "C for 4 h, (d) Ar at 875 "C for 32 h, (P) Ar at 875 "C for 48 h and (f)Ar at 875 "C for 64 h. 0,BaCuO,; V,422; 3,BaCu,O,; V,Cu,O. as the tetragonal YBCO structure by Rietveld refinement [Fig. 1(b)];starting parameters were taken from David et a1.,I5 and a squared Lorentzian function was used to model the peak shape.The final structural parameters are shqwn in Table 1 (a), with unit-cell parameters a =3.9250( 2) A, c = Table 1 (a) Results of Rietveld refinement on composition LaBa,Cu,O,.,, position ~ 0 X Y Z uiso La Id 1.O 0.5 0.5 0.5 0.015(3) Ba 2h 1.O 0.5 0.5 0.1796(5) 0.020(2) Cu(1) Cu(2) O(1) O(2) O(3) la 2g 4i 2g 2f 1.o 1.0 1.o 1.0 0.245 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.5 0.0 0.5 0.0 0.350(1) 0.375(2) 0.157(4) 0.0 0.025( 6) 0.041(4) 0.03(1) 0.03(1) 0.03 (2) space group P4/mmm, u =3.9250( 2) A,c = 11.9263( 10) A,R, =5.05%, R,, =6.50Y0, R, =4.26%. (h)Results of Rietveld refinement on composition LaBa,Cu,O,,,, position 0 z uiso Ih 1.O 0.5 0.5 0.5 0.016(3) 2t 1.o 0.5 0.5 0.1799(5) 0.018(2) la 1.o 0.0 0.0 0.0 0.023( 6) 2q 1.o 0.0 0.0 0.350( 1) 0.040(3) 2s 1.O 0.5 0.0 0.387( 6) 0.03 (2) 2r 1.o 0.0 0.5 0.355 (6) 0.04( 1) 2q 1.o 0.0 0.0 0.157( 3) 0.03( 1) le 1.0 0.0 0.5 0.0 0.041 (7) Ib 0.14 0.5 0.0 0.0 0.039(6) space group Pmmm, a= 3.8779(4)A, b =3.9432(4) A, c = 11.7601(15)A, R,=4.71%, R,,=6.12%, R,=3.85%0.J. MATER. CHEM., 1994, VOL. 4 11.9263( 10) A R,, =6.50%. The difference plot from the Rietveld refinement is also shown in Fig. l(bl. Annealing tetragonal L~B~,CU,O~.,~ in oxygen at 400 "C for 4 h gave the orthorhombic YBCO structure with 6= -0.14+0.02. There are two possible explanations for an oxygen content of 7.14. First, as with YBCO, it may indicate the presence of the '124' phase, LaBa2Cu,0,, in the ratio 14% 124, 86% 123.16 This appears not to be the case in the present materials; X-ray data indicate the La- 123 to be phase- pure.In addition, magnetic susceptibility results show no evidence for a second, superconducting phase (Fig. 2), and there have been no reports on the La analogue of YBa,Cu,O, to date. Secondly, the excess oxygen may be present in the 123 structure in the O(5)sites of the basal plane, as indicated by previous ~tudies.'~,~~ Hence, attempts were made to refine the structure of L~B~,CU,O,~~,in both the orthorhombic space group Pmmm and tetragonal P4/mrnm; the excess oxygen in both cases was assumed to be in the 0(4)/0(5) sites. R,, for the tetragonal model was 11.87%, with the thermal parameters of O(4) diverging. In addition, it was not possible fully to index all of the diffraction lines adequately. Using the orthorhFmbic model all oparameters conve:ged, with u =3.8779(4) A, b = 3.9432(4) A, c= 11.7601(15)A, RW,=6.l2% [Fig.l(c)], with the difference plot for the Rietveld refinement shown beneath the X-ray diffraction pattern: final parameters are given in Table l(b). It is concluded, therefore, that LaBa,Cu,O,,,, is a single-phase, orthorhombic material, with oxygen occupying both sites in the basal plane. Oxygen Content of La-123 The stability of the single-phase material was studied by thermogravimetry and by annealing experiments followed by quenching into Hg. 80mg samples of the single-phase, 6= -0.14 material were heated in steps to 900 'C (a) in air and (b) in oxygen on the thermobalance; samples were held isothermally for 30 min at 50 "C intervals to achieve thermal equilibrium; typical TG traces are shown in Fig.3. These experiments showed that LaBa,Cu,O,,,, is stable in air to ca. 300°C and in oxygen up to ca. 400°C after which weight loss commences, corresponding to an increase in (j. The quenching experiments showed that the orthorhombic structure is retained up to 600°C in air, above which temperature the structure becomes tetragonal. The tetragonal structure is preserved up to ca. 800°C in air and 750°C in oxygen, above which decomposition occurs to give La,,,Bal,,Cu,O, and BaCuO,. This seems to suggest that decomposition is not associated with an absolute oxygen content.but rather that 0.5 94K 1 0 0 -IY0) E -0.5. 0 m :I I 't 0 02 -Q 0 0 -1.0. 0 0 0 Fig. 2 Ac susceptibility (x)trace for LaBa,Cu,O,,,, J. MATER. CHEM., 1994, VOL. 4 1000 0 0 0 800 O oono [I 0 0 600 0 9i= 400 200 01 n]' ' 6.2 6.4 6.6 6.8 7.0 7.2 nominal oxygen content Fig. 3 TG trace for LaBa,Cu,O,-a heated in air (0)and oxygen (0). The oxygen contents refer to single phase 123 for the range 6.65 to 7.14 in air and 6.85 to 7.14 in oxygen; lower oxygen contents refer to multiphase mixtures. at a specific temperature and atmosphere, it is more favourable for the Lal.2Bal.8Cu30Z composition to form than the stoi- chiometric, tetragonal La-123 phase. The decomposition observed in the quenched samples also appears to correspond to a change of gradient in the weight loss traces shown in Fig.3; this implies that the Lal.2Bal&~30Z composition loses oxygen at a different rate than the La-123 phase. Unit-cell Data X-Ray diffraction data for a number of samples quenched from temperatures in the range 400-800°C were measured over the range 7-90" 28; the unit-cell results for 6 =0.49 and -0.14 were taken from the Rietveld refinements. It is essential when studying the orthorhombic-tetragonal phase transition in these triple perovskite materials to consider the ratio R= c// % (a +b). For the ideal triple perovskite, this ratio is equal to 3; the structure is pseudo-cubic. Deviations from the ideal value lead to line splitting, e.g.a pseudo-cubic (100) line splits to (100) and (001);however, the line-splitting does not imply that the structure is orthorhombic. For instance, in the XRD powder pattern of the tetragonal LaBa2Cu306,49 [Fig. 1(b)], R =3.038 and significant line splitting is observed. In the case of orthorhombic LaBa2Cu,07~,, [Fig. 1(c)], the strongest line shows no splitting, and R =3.007. For intermediate values of 6, attempts were made to index the data using both the orthorhombic and tetragonal unit cells. For values of 6 <0.2, the tetragonal model did not index all lines; samples with 6>0.2 could be accurately indexed on a tetragonal unit cell. The results shown in Fig. 4 indicate that an orthorhombic-tetragonal phase transition occurs at 6~0.2.This is in clear contrast with YBCO where the orthorhombic-tetragonal phase transition occurs at 6 z 0.4. The c parameter appears to show a slight change in gradient at the phase transition, again in contrast to YBCO where the change in c is more pron~unced.'~,~' T,Data Critical temperatures of the quenched samples were measured from ac susceptibility data; a plot of T,against oxygen content shows linear variation (Fig. 5), as is also seen in quenched samples of YBa2Cu307 Samples were also prepared by annealing in air at temperatures up to 750°C, followed by slow cooling to room temperature; this did not give as wide a variation in 6 as the quenching method, but clearly shows 11.951 3.9501 oxygen content Fig.4 Variation of unit-cell parameters with oxygen content for samples prepared by quenching loo r t 2o t t 01 \\-i 7.2 7.1 7.0 6.9 6.8 6.7 oxygen content Fig.5 Variation of T, with oxygen content for quenched (0)and slowly cooled (m) samples evidence for a plateau in T, us. oxygen content over the range ca. 7.05-7.15. A similar plateau is seen in YBCO, but at significantly lower oxygen contents of ca. 6.85-7 (e.g. ref. 18 and 20). Decomposition of LaBa,Cu,O, -The results for heating at 875 "C in flowing argon are summar- ised in Fig. l(d)-(f). The tetragonal phase remains phase- pure for heating times of 16 and 32 h, but after 48 h, lines corresponding to La,~2xBa2+2xCu2~,010~,,,x ~0.2,(422), BaCuO, and Cu20 are evident.After 64h, extra lines of La, -2xBa2+ 2xCu2-,Ole-2x, x M0.2, (422), BaCuOz and BaCu202 are present. This behaviour is similar to the gradual decomposition of YBCO under low oxygen partial press- ure~.~'-~~Subsequent reduction of these heat-treated samples in 10% H2-N, on the thermobalance gave the overall oxygen content at each of these stages. A decrease in the total oxygen content was observed for the increased heating times, corre- 1748 sponding to the processes La1.2Ba1.8C~20z+BaCuO, U L~B~,CU,O~,~ U Conclusions LaBa,Cu,O, -can be prepared as a single-phase tetragonal material by heating at 875 "C for 16 h under flowing argon which, on subsequent annealing in oxygen at 400°C gives an orthorhombic superconductor with T, (onset)=94 K.An orthorhombic-tetragonal phase transition occurs at 6 =0.2. Under the conditions described here, single-phase LaBa,Cu,O, -is not an equilibrium phase; when it is heated in air or oxygen at 800°C decomposition occurs to give La,.,Ba,,,Cu,O,~, and BaCuO,, and under flowing argon at temperatures of 850 "C and above, the material gradually decomposes to give La,.6Ba2,,Cul,80,,, +BaCuO, + BaCu,O,. On lowering the temperature to 825 "C, however, La-123 does not form in argon. It is possible that at some intermediate oxygen partial pressure and temperature, the stoichiometric single-phase material is thermodynamically stable. We thank SERC for a research grant (A.R.W.) and the International Centre for Diffraction Data for a scholarship (J.M.S.S.).References 1 M. Izumi, T. Yabe, T. Wada, A. Maeda, K. Uchinokura, S. Tanaka and H. Asano, Phys. Rev. B, 1989,40,6771. 2 A. Maeda, T. Noda, H. Matsumota, T. Wada, M. Izumi, T. Yabe, K. Uchinokura and S. Tanaka, J. Appl. Phys., 1988,64,4095. J. MATER. CHEM., 1994, VOL. 4 3 I. Nakai, K. Imai, T. Kawashima and R. Yoshizaki, Jpn. J. Appl. Phys., 1987,26, L1244. 4 M. Izumi, K. Uchinokura, A. Maeda and S. Tanaka, Jpn. J. Appf. Phys., 1987, 26, L1555. 5 R. Yoshizaki, H. Sawada, T. Iwazumi, Y. Saito. Y. Abe, H. Ikeda, K. Imai and I. Nakai, Jpn. J. Appl. Phys., 1987, 26, L1703. 6 M. Hikita, S. Tsurumi, K. Semba, T. Iwata and S. Kurihara, Jpn. J. Appl. Phys., 1987,26, L615. 7 J. M. S. Skakle and A. R. West, Physica C, 1994.220, 187.8 C. U. Segre, B. Dabrowski, D. G. Hinks, K. Zhang, J. D. Jorgensen, M. A. Beno and I. K. Schuller, Nature (London), 1987, 329,227. 9 E. Takayama-Muromachi, Y. Uchida, A. Fujimori and K. Kato, Jpn. J. Appl. Phys., 1987,26, L1546. 10 T. Wada, N. Suzuki, T. Maeda, A. Maeda, S. Uchida, K. Uchinokura and S. Tanaka, Appl. Phys. Letr., 1988,52,1989. 11 T. Wada, N. Suzuki, A. Maeda, T. Yabe, K. Uchinokura, S. Uchida and S. Tanaka, Phys. Rev. B, 1989,39.9126. 12 T-T. Fang, J-W. Huang and M-S. Wu, J. Muter. Res., 1994, 9, 1369. 13 L. Ganapathi, A. K. Ganguli, R. A. Mohan Ram and C. N. R. Rao, J. Solid State Chem., 1988,73, 593. 14 A. Sequeira, H. Rajagopal, L. Ganapathi and C. N. R. Rao, J. Solid State Chem., 1988,76,235.15 W. I. F. David, W. T. A. Harrison, J. M. F. Gunn, 0. Moze, A. K. Soper, P. Day, J. D. Jorgensen, D. G. Hinks, M. A. Beno, L. Soderholm, D. W. Capone 11, I. K. Schuller, C. U. Segre, K. Zhang and J. D. Grace, Nature (London), 1987,327,310. 16 D. M. Pooke, J. L. Tallon, R. G. Buckley and J. Loram, Physica C, 1991,185-189,563. 17 P. K. Gallagher, H. M. O'Bryan, S. A. Sunshine and D. W. Murphy, Muter. Res. Bull., 1987,22,995. 18 R. J. Cava, B. Batlogg, C. H. Chen, E. A. Rietman, S. M. Zahurak and D. Werder, Nature (London), 1987,329,423. 19 C. Namgung, J. T. S. Irvine and A. R. West, Physica C, 1990, 168, 346. 20 R. Beyers, B. T. Ahn, G. Gorman, V. Y. Lee. S. S. P. Parkin, M. L. Ramirez, K. P. Roche, J. E. Vazquez. T. M. Gur and R. A. Huggins, Nature (London), 1989,340,619. 21 J. S. Kim and D. R. Gaskell, Muter. Lett., 1993. 16, 327. 22 B. T. Ahn, T. M. Gur, R. A. Huggins, R. Beyers, E. M. Engler, P. M. Grant, S. S. P. Parkin, G. Lim, M. L. Ramirez, K. P. Roche, J. E. Vazquez, V. Y. Lee and R. D. Jacowitz. Physica C, 1988, 153-155,590. 23 T. Lada, A. Paszewin, R. Molinski, A. Morawski and W. Pachla, Appf.Supercon., 1993, 1, 591. Paper 4/04119D; Received 6th July, 1994