J. MATER. CHEM., 1994, 4(8), 1353-1355 Superconductivity up to 95 K in Mercury-substituted 121 2 Thallium Cuprates (TI,Hg)lSr2+yNdl $u207 F. Letouze, S. Peluau, C. Michel, A. Maignan, C. Martin, M. Hervieu and B. Raveau Laboratoire CRISMAT, CNRS URA 1318-ISMRA, Universite de Caen, Boulevard du Marechal Juin, 14050 Caen Cedex, France The solid solution TI, -xHgxSr,+yNd, -yCu,O,-d has been characterized; the homogeneity range extends up to x =0.2 and y=0.5. Electron diffraction and XRD studies confirm the statistical distribution of the cations in the different sites of the 121 2-type structure. The best properties are observed for x =0.1 and y =0.5, with a T, of 95 K and a diamagnetic volume fraction of 70%. The influence of mercury substitution for thallium in the phase is discussed.Numerous superconducting thallium cuprates have been isolated to date, besides the classical superconductors TlBa,CaC~,0,l,~ and TlSr,CaCu,O, which are rarely pre- pared as a pure pha~e.~,~ In these compounds, whose structure (Fig. 1) consists of pyramidal copper layers intergrown with distorted rock-salt-type layers, thallium can be associated with different cations such as bismuth5 and lead6 without changing dramatically the critical temperature of these phases. In contrast, the nature of the cations interleaved between the pyramidal copper layers plays a determining role in supercon- ductivity. In most of these compounds, the presence of calcium is necessary for the existence of superconductivity; its partial replacement by a trivalent lanthanide may enhance the super- conducting properties, but beyond 20% substitution T, is affected; the superconductivity disappears completely for a total substitution, e.g.for TlBa,NdCu,O, .7,8 In this respect, the cuprate TlSr, _,L~,CU~O~~-~~ is very interesting since it is the only 1212 superconductor without calcium, with a T, onset of 95 K. Divalent mercury, owing to its 5d1' electronic configuration similar to that of Tl"', is a good candidate for partial replacement of this cation. The stabilization of mixed layers [Tll-,Hg,O], was indeed proved in the 2223 cuprate T1, ~,Hg,Ba,Ca,Cu3010'2 for a substitution rate x limited to 0.4. Such a substitution did not modify the critical temperature of this phase, which remained close to 130K.The present study is devoted to the substitution of mercury for thallium in calcium-free 1212 cuprates belonging to the system T1Sr,NdCu20, -,-T1Sr3Cu,07 -Fig. 1 Idealized 1212 structure Mixtures of the oxides T1203, HgO, SrO,, Nd203 and CuO according to the nominal compositions Tll -xHg,Sr2+yNdl -yC~Z09-xx/2+y/2 were heated in cvacuated silica tubes. The temperature was raised slowly o~er 8 h to 900 "C, maintained for 8 h and then slowly decreased to room temperature. The nominal oxygen content is much higher than that required for the ideal composition '0,' of the 1212 oxides, so that an oxygen pressure of several bar exists in the tube in order to favour a partial oxidization of Cu" into Cu"' and to avoid dissociation of mercury oxide into metallic mercury.Different thermal treatments were performed in order to improve the superconducting properties via the optimization of the oxygen content: at 350 "C under an oxygen pressure of 100 bar or at 280 "C under an Ar-H2 (90 : 10) flow. The samples were systematically analysed by X-ray diffrac- tion (XRD), electron diffraction (ED) and energy-ciispersive microanalysis (EDS). The XRD patterns were recorded with a Seifert vertical diffractometer equipped with a primary monochromator (Cu-Ka, radiation). Data were collected by step scanning in the range 10<28/degrees<110 with an increment of 0.02" (28). Lattice constants were determined using the Rietveld method (computer program DBW 3.213).Electron diffraction was performed with a JEO I, 200CX electron microscope, operating at 200 kV and equipped with a eucentric goniometer (k60").Electrical resistivity measure- ments are carried out by the four-probe technique. 'Magnetic measurements were performed with a DC SQUID magnet- ometer; the samples are first zero-field-cooled to 5 K and a magnetic field of 10 G was then applied to register the temperature dependence of the magnetization. No deinagnetiz-ation corrections were performed, owing to the porous charac- ter of the bars. For the above thermal treatments, the formation of a 1212-type phase was observed over a large part of the 'pseudo quaternary' diagram T1, -,Hg,Sr, +,,Nd, -yCu207 -& (Fig. 2). The EDS analysis shows that in many crystals the substitution can be achieved up to x=0.5, i.e.'Hgo,5Tlo,5' and to y=O.8, i.e. 'sr2.8Ndo.i (the two discontinuous lines in Fig. 2). However, the XRD study shows that, in the selected experimental conditions, the monophasic domain i$ smaller and is limited by two lines x=O.2 and y=O.5, corre-sponding to the oxides Tl,~,Hg,~,Sr, +yNd, -,Cu,C), and T1, -xHgxSr2.5Nd0.5C~207 (continuous lines in Fig. 2). All the corresponding data are reported in Table 1. In this table, T,s correspond to the as-synthesized and optimized samples. An improvement in T, is observed only for two samples after oxygen pressure annealing. The reconstruction of the reciprocal space from the ED patterns (Fig. 3) shows a tetragonal cell with axap(up~3.8A, cell parameter of the ideal cubic perovskite), cz J 2 A and Fig.2 Pseudo quaternary diagram TISr,CuzO, -d-HgSr,Cu,O, -HgSr,NdCu,O, -,-TISr,NdCu,O, Table 1 Cell parameters and onset of T, of the as-synthesized samples and after optimization onset cation composition x, I’ (EDS analysis) a/A CIA as-synthesized optimized 0, 0 0, 0.5 3.8503( 1) 12.0724(5) 3.8403( 1 j 12.2073(4) ns 90 ns 95 0.1, 0 3.8506(2) 12.059(1) ns ns 0.1, 0.2 3.8421(2) 12.137(1) 80 85 0.1,0.35 3.8415(5) 12.191(2) 90 90 0.1, 0.5 3.8357(3) 12.208(1) 95 95 0.2, 0 3.8507(2) 12.058( 1) ns ns 0.2, 0.5 3.8339(5) 12.210(2) 95 95 ns, not superconducting. P-type symmetry, whatever x and y. No superstructure or streaks were observed, attesting to the statistical distribution of the various cations over their own sites. From the XRD data refinements, the cell parameters (Table 1) show that the substitution of mercury for thallium J.MATER. CHEM.. 1994, VOL. 4 does not influence the cell size, as observed for the 2223 cuprate.12 The evolution of these parameters us. y, and especi- ally the increase of c with y is consistent with Sr being larger than Nd. In order to understand the changes in superconduction induced by Sr/Nd substitution, the line x=O.1 (Fig. 2) has been especially studied (part 2 of Table 1). This corresponds to the formula Tlo,9Hgo,lSr2+,Nd, -,Cu207 y =0-0.5. As soon as trivalent Nd is replaced by divalent Sr the supercon- ductivity is induced.For the y =0.2 as-synthesized sample, the superconducting transition is broad (Fig. 4);such behaviour can be explained considering the EDS analysis performed on numerous microcrystals of the samples which shows a rather large dispersion of the y values with regard to the mean y= 0.2 value. The T, onset is close to 80 K, with a diamagnetic volume fraction of 25% at 5 K. The critical temperature increases with y (Table 1); the best properties are observed for the limiting composition x =0.1, y =0.5 (as-synthesized sample) with a T, onset of 95 K, a superconducting volume fraction of 70% [Fig. 5 (a)]and a rather sharp transition. Note that the EDS analysis performed on this sample attests to the good homogeneity of the sample with a very small dispersion of y values. Resistivity measurements confirm the T, onset value and the metallic behaviour of the material for r>T, in the normal state.For the annealings performed on this y=OS sample, a decrease in T, was observed [Fig. 5 (m, O)]indicating that the optimal carrier density is achieved in the as-synthesized sample. This latter T, value of 95 K is indeed the same as that of the pure T1-optimized sample T1Sr2.5Ndo.5Cu207 the limit and of T1,-,,Hgo~2Sr,,,Ndo~,C~~207, compound (Table 1). In conclusion: A new solid solution has been characterized, Tl1-,Hg,Sr2+,Nd, -,,CU~O~-~.For the selected thermal pro- cess, the homogeneity range extends up to x=O.2 and y=O.5. Coupled EDS and XRD analyses suggest that these limits can be increased provided the conditions of synthesis (time- temperature-partial pressure) are adapted to each composi- tion.The EDS investigation and XRD refinements confirm the statistical distribution of the cations in the different sites of the 1212 structure. This study confirms the previous observations made for Fig. 3 [0101 ED pattern, x =0.2 and y =0.5 J. MATER. CHEM., 1994, VOL. 4 0.00 v, --0.05 -0.10 -f' c. x -0.15 --0.20--0.250 20 40 60 80 100 TIK Fig. 4 Temperature dependence of the susceptibility for the TI,,,HE,,~S~,,,N~,,,CU,O~-as-synthesized sample 0.0 -0.1 -0.2 4.3h Y? x -0.4 -0.5 -0.6 -0.7 0 20 40 60 80 100 TIK Fig. 5 Temperature dependence of the susceptibility for Tl,,,Hg,,,Sr,,,Nd,.,Cu,O,_, as synthesized (O), 280 "C, 40 min, Ar-H, annealed (D)and 350 "C -7 h -100 bar 0, annealed (0) the 2223 oxidel, concerning Hg for T1 substitution: a similar T, value of 95 K, after optimization, can be reached for the doped and undoped samples, i.e.T1,-,Hg,Sr,,,NdO~,Cu,O7 with x=O, 0.1, 0.2. It can also be compared to the oxides Tlo.8Pbo,2Sr,+,Nd, -,,CU,O-~ of ref. 10: for similar y values close to 0.5, the T,s are comparable, showing that the composition of the thallium layers, which may act as a reservoir of holes, is not of prime importance in the expected maximum T,; the optimal hole carrier density is the only important factor. Such an assertion was also made concerning the T1-deficient 2212 s~perconductor.'~ The role of the layer interleaved between the [CuO,], layers is very different.This fact was observed in numerous 2212 and 1212 oxides and is once more confirmed; a small variataon of the (Sr,Ca)/Ln ratio involves a dramatic change in the critical temperature. References 1 M. Hervieu, A. Maignan, C. Martin, C. Michel, J. Provost and B. Raveau, J. Solid State Chem., 1988,75,212. 2 B. Morosin, D. S. Ginley, P. F. Hlava, M. J. Carr, R. J. Haughman, J. E. Schirber, E. L. Venturini and J. F. Kwak, Physicsa C, 1988, 152,413. 3 W. L. Lechter, M. S. Osofsky, R. J. Soulen, V. M. Le Tourneau, E. F. Skelton, S. B. Qadri, W. T. Elam, H. A. Hoff, K.A. Hein, L. Humphreys, C. Skowronek, A.K. Singh, J. F. Gilfrich, L. E. Toth and S. A. Wolf, Solid State Commun., 1988,h8, 519. 4 F. Izumi, T. Kondo, Y. Shimakawa, T. Manako, Y. Kudo, H. Igarashi and H. Asano, Physica C, 1991,185-189,615. 5 S. Li and M. Greenblatt, Physica C, 1989, 157, 365. 6 M. A. Subramanian, C. C. Torardi, J. Gopalakrishnan, P. L. Gai, J. C. Calabrese, T. R. Askew, R. B. Flippen and A. W. Sleight, Science, 1988,242, 249. C. Martin, D. Bourgault, C. Michel, M. Hervieu and B. Raveau, Mod. Phys. Lett. B, 1989,3,93. C. Michel, E. Suard, V. Caignaert, C. Martin, A. Maignan, M. Hervieu and B. Raveau, Physica C, 1991,178,29. A. K. Ganguli, V. Manivannan, A. K. Sood and C. N. R. Rao, Appl. Phys. Lett., 1989,55,2664. V. Manivannan, N. Rangavittal, J. Gopalakrishnan and C. N. R. Rao, Physica C, 1993,208,253. Y. Xin, Y. F. Li, D. Ford, D. 0.Pederson and Z. Z. Sheng,J.J.A.P., 1991,30,1549. F. Goutenoire, A. Maignan, G. Van Tendeloo, <'. Martin, C. Michel, M. Hervieu and B. Raveau, Solid State Commun., 1994, 90,47. 13 D. B. Wiles and R. A. Young, J. Appl. Crystallogr., 198 1, 14, 149. 14 C. Michel, C. Martin, M. Hervieu, A. Maignan, J. Provost, M. Huve and B. Raveau, J. Solid State Chem., 1992,96,271. Communication 4/03564J; Received 13th June, 1994