Some aspects of the thermodynamic behavior of high-Tc oxide systems via EMF and related measurements

Some aspects of the thermodynamic behavior of high-Tc oxide systems via EMF and related measurements

PhysicaC 198 (1992) 109-117 North-Holland Some aspects of the thermodynamic behavior of high-T, oxide systems via EMF and related measurements M. Tet...

775KB Sizes 0 Downloads 56 Views

PhysicaC 198 (1992) 109-117 North-Holland

Some aspects of the thermodynamic behavior of high-T, oxide systems via EMF and related measurements M. Tetenbaum,

P. Tumidajski,

D.L. Brown

and M. Blander

Chemical Technology Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439, USA Received 4 February I992 Revised manuscript received 16 March I992

EMF measurements of oxygen fugacities as a function of stoichiometry have been made in the YBazCusO,, NdBa,CujO,, and Nd,.s,Ce,,,9CuOx superconducting systems in the temperature range 400-750°C by means of an oxygen titration technique with a yttria-stabilized zirconia electrolyte. The object of our current measurements is to investigate the effect of ionic size of Y, and Nd on the immiscibility and other structural transitions in the LnBazCu30X system. The shape of the 400°C isotherm for NdBazCu90, indicates that, if a miscibility gap is present, it exists at lower temperatures and at values of x that are higher than those for the miscibility gap we deduce in the YBazCu,O, system. The relative locations of the miscibility gaps are consistent with the effects of ionic radii on the composition dependence of T, for these systems. Our results explain the two plateaus in measured values of T, as a function of composition for the YBa,CuXO, system an appear to be consistent with the less pronounced T, plateau found for the NdBazCu30, system at higher values of stoichiometry. For a given oxygen stoichiometry, partial pressures of oxygen above NdBa2Cu30, are higher than for the YBa#&O, system. Results of limited measurements on the n-type (electron-doped) superconducting Nd,.8,Ce0.,&uOx system will be presented, as well as a thermodynamic assessment and intercomparison of our oxygen partial pressure measurements with the results of other measurements.

1. Introduction The objective of our studies is to investigate the effect of ionic size of yttrium and neodymium on the structural transformations, nonstoichiometric and thermodynamic behavior of the high-T, superconducting oxide systems YBa2Cu30, and NdBa2Cu,0, as a function of oxygen partial pressure, oxygen stoichiometry, and temperature by means of a coulometric titration technique where the oxygen content can be varied by well-defined small amounts. The objective of lower temperature measurements is to determine whether the structural changes indicated by the models and calculations of Curtiss et al. [ 11, Khachaturyan et al. [ 21 and Wille et al. [ 3 1, as well as the experimental studies described below, could be confirmed via EMF measurements carried out at a relatively low temperature. We will review and update the results of our previously reported measurements on the YBa2Cu30, system and report on our recent results on the NdBazCu30, system. The results of limited mea-

surements on the n-type (electron-doped) superconducting Nd,.*,Ce,,. ,&uO, system will also be presented. It should be emphasized that the advantage of the EMF technique compared to thermogravimetric analyses (TGA) is that very small samples (-2530 mg) can be studied, and the composition of the condensed phase varied precisely by small incremental amounts. In addition measurements can be made down the much lower oxygen pressures than are possible by TGA methods. One gram samples are generally used in TGA; problems due to convectional gas flow and thermal drift can reduce the sensitivity of the TGA technique.

2. Experimental

The electrochemical resented by

0921-4534/92/%05.00 0 1992 Elsevier Science Publishers B.V. All rights reserved.

cell can be schematically

rep-

M. Tetenbaum et al. / Thermodynamic behavior ofhigh-T, oxide systems

110

Pt, MM@))

IZr02(Y&)

lAir(~O~),,~ ~0.21 atm), Pt

,

and where M = YBa2Cu30,, NdBazCu30,, Nd, .8,Ce,,. , JuO,. The equilibrium oxygen partial pressures, p(02), above the sample in the cell can be calculated by means of the Nemst equation

P(Oz)=P(02)dexp

( > - g

,

where E is the EMF (open circuit potential) of the cell, F is the Faraday constant, T is the absolute temperature and R is the gas constant. The oxygen content of the sample under investigation can be varied by means of coulometric titration and calculated from the relationship An(0,)

showed only the orthorhombic structure; oxygen analysis via iodometric titration gave 6.9 10 + 0.005 for x. The composition of the as-received n-type samples (coarse powder ) was Nd, .81Ce,,. , &u04. Xray examination showed essentially the tetragonal structure and a minor cubic phase estimated to be The transition temperature was reCeo.sNdo.0.,,. ported to be -24 K [7].

= $;,

where An (0,) = moles of oxygen transported, I is the controlled current flowing through the cell for a time t(s) , and F is the Faraday constant; it should be noted that yttria-stabilized zirconia has a high mobility of oxygen ions in the lattice of the electrolyte. The cell used in this study was based on the cell designs described by Tretyakov and Rapp [4], except that soft glass was used as a sealant instead of Pyrex. Pure analyzed air circulation over the cell served as the reference electrode. The cell EMF was measured with a high-impedance electrometer and a Keithley digital voltmeter. Coulometric titrations were performed with a Princeton Applied Research Model 173 Potentiosat/Galvanostat. measurements. The samples used for the EMF measurements were prepared by the Materials Science Division at the Argonne National Laboratory. The YBazCusO, sample was received in the form of a thick sintered tape ( - 95 theoretical density). The transition temperature was reported to be - 90 K [ 5 1. X-ray examination showed only the orthorhombic structure; oxtitration) gave analysis (iodometric ygen x= 6.89 1 + 0.005. The tape was broken up into small granules (sample size, - 30 mg) for use in the EMF measurements. The NdBa2Cu30, sample was received in the form of a coarse powder. The transition temperature was reported to be -96 K [ 61. X-ray examination

3. Results and discussion 3.1. Y-Ba-Cu-0

systems

The essential highlights of our previously published results [ 8-101 on the YBa2Cu30, system are summarized below. The results of our EMF measurements are shown in fig.1 where equilibrium oxygen pressures calculated for isotherms in the temperature range 400-

0

1

2

E s 2: ON3 n E 0 T 4

Af? *

5

0 5oooc

A

A

v 476OC

6 6.1

62

6.3

64 x

65

66

67

66

6.9

70

in YBa,Cu,O,

Fig. 1. Partial pressure (fugacity) of oxygen as a function of temperature and oxygen stoichiometry in YBa,C&O,

M. Tetenbaum et al. / Thermodynamic behavior of high-T, oxide systems

750°C are plotted as a function of x in YBazCu30,. There is no apparent sign of a change in curvature at an oxygen stoichiometry of x- 6.5 where the formal valence of copper is two, and where the O-T transition had been assumed to occur by various investigators during the early stages of research on this material. Of particular importance is that the change in curvature of the lower temperature isotherms above x=6.5 shown in fig. 1 and an inflection around x=6.65 are consistent with the presence of a miscibility gap at lower temperatures. Based on the excellent linear fit, of RT logp(0,) versus T, we extrapolated values of logp(Oz) versus x, below 400°C. Results indicate the presence of an orthorhombicorthorhombic miscibility gap with a consolute temperature of _ 200 + 50°C. For the terminal compositions extending between x=6.55 to x=6.75 we estimated a temperature of N 35 “C. In addition, a relatively rapid increase of the partial molar entropy of solution of oxygen in YBa,Cu,O, in the composition range x z 6.75-6.55 appears to reflect the properties of an orthorhombit-orthorhombic miscibility gap since the extrapolated measurements appear to be completely within the orthorhombic phase field. The increase in ti( 02) with oxygen deficiency in YBa2Cu30x at ~~6.5 is related to the O-T transition and reflects the differences between the distributions of oxygen atoms and vacancies in the relatively disordered tetragonal structure and the more ordered orthorhombic structure. The presence of a miscibility gap at low temperatures due to two orthorhombic structures has been suggested by a number of theoretical studies. Curtiss et al. [ 1 ] postulated a phase diagram based, in part, on semiempirical molecular orbital theory, which suggested two orthorhombic structures in the region x>6.5. They speculated that one of the structures would have fully occupied oxygen chain sites (0 I), while the other would have partially occupied chain sites. Khachaturyan et al. [ 21 estimated a phase diagram based on a mean field model and experimental data for the T+O transition. Their calculated diagram predicts a peritectoid decomposition into two phase fields below -190°C namely, 0’ (where x=6.5) +O,, or G’+T. Wille et al. [ 31, have calculated a phase diagram by application of the cluster

111

variation method to an asymmetric two-dimensional Ising model with interaction parameters selected to guarantee the stability at x=6.5 of the experimentally observed double cell structure. However, their calculated phase-diagram is not consistent with our measurements. Goodenough and Manthiram [ 111 have also suggested that interchain ordering of the oxygens can give rise to several discrete Magneli-type orthorhombic structures. From the more recent results of Khachaturyan et al. [ 2 1, de Fontaine et al. [ 31 and Hyland et al. [ 12 1, we conclude that at a stoichiometry of x= 6.5, a coulombically ordered structure exists at low temperature with oxygens in the basal planes forming alternating planes of chains of divalent oxygen and vacancies. At the idealized stoichiometry of x= 7.0, monovalent oxygens occupy all of the vacancies. it appears that the valence of copper remains at 2 at x> 6.5 and that beyond this stoichiometry; some of the oxygens in the basal plane are monovalent. Therefore, compositions between 6.5 and 7.0 can be thought of as being formed by mixing ordered O*-vacancy pairs (which have net dipoles) with O(which have no net charge in the solid). Mixing of material containing dipole “ion” pairs with a “neutral” (uncharged) material will generally be accompanied by a positive energy change and by large positive deviations from ideal solution behavior if the effective dielectric constant is not very large. We discuss this model below. 3.2. Nd-Ba-Cu-0

systems

In agreement with the results obtained with the YBalCujO, system, no apparent change in curvature or discontinuity was observed at an oxygen stoichiomen-y of x= 6.5 where copper is divalent and where the orthorhombic-tetragonal transition had been assumed to occur. The results of our EMF measurements are shown in fig. 2, where equilibrium oxygen pressures calculated for isotherms in the temperature range 400-600°C are plotted as a function of x in NdBa&u,O,. It should be emphasized that for a given isotherm the EMF was reversible for oxygen absorption and desorption over the composition range investigated. Our earlier measurements [ 13 ] reported on the NdBa&u,O, system showed some slight hysteresis at 400°C in the composition range

M. Tetenbaum et al. / Thermodynamic behavior of high-T, oxide systems

112

-

0

ometry where superconductivity is lost increases with increasing ionic radius. in addition, their results show that the lower plateau is nearly absent for the case of NdBa,Cu,O,. Ionic radii of trivalent Y, Gd and Nd (reported in their paper) are 0.99, 1.02 and 1.10 A respectively. According to Shaked et al. [ 16 1, the repulsion energy of oxygen atoms in 0( 1) and 0( 5) lattice sites in NdBazCuSO, is smaller than in YBa2Cu30,. This lower repulsion energy apparently stabilizes the orthorhomic structure at higher values of x in NdBazCu30, compared to YBa2Cu30r Our results obtained with the NdBa,Cu30, system show that at a given oxygen stoichiometry, partial pressures of oxygen are higher than for the YBa2Cu30X system, in accord with calculated partial thermodynamic quantities. Typical results at 400°C are shown in fig. 3, and intercomparison of partial thermodynamic quantities calculated from our EMF measurements on these systems are given in figs. 4 and 5. A comparison of partial molar entropy and

ND123 -1

-2

-3

E F -, a0 n _0

-1 -5,

-6

-7

-8

-Y

- 10 Ii 1

62

63

64 x

Ii 11

6 Ii

Ii 7

Ii 8

6 ‘I

: 0 I

In NdBa,Cu,OY

Fig. 2. Partial pressure (fugacity) of oxygen as a function perature and oxygen stoichiometry in NdBa&u,O,.

of tem-

during desorption of oxygen from NdBazCuJO, after x had reached a value of _ 6.90. It should be noted that coulometric titrations yield the change in stoichiometry, 6x, from a given composition, x, in NdBa$u,O, The reference composition after our cell was sealed and the EMF measured prior to coulometric titrations was based on the data of Kishio et al. [ 141. The minimum of the slope of the 400°C isotherm shown in fig. 2 NdBa$&O,, although not as pronounced as with the YBa2Cu30x system, suggests the possible presence of a miscibility gap at lower temperatures, at values of x that are higher than those in the YBa2Cuj0, system. It should be noted that the location of the miscibility gap is consistent with the observed compositional dependence of T, for the YBazCu30, systems. The results of Veal et al. [ 15 1, show that with increasing ionic radius in the sequence Y, Gd, Nd, the rate of fall-off of T,increases with oxygen deficiency. Also, the oxygen stoichix=6.65-6.85

400

c

0

-1

-7 62

63

65

64 x

in

66

67

68

69

7 0

LnBa,Cu30x

Fig. 3. Comparison of oxygen partial pressures Y 123 systems at 400°C.

above Nd I23 and

M. Tetenbaum et al. / Thermodynamic behavior of high-T, oxide systems

113

10

y

20

g 0 o ,‘

0

l

30

.

a0

9

0

8 0

lboo

SW ‘2 I

40 0 Nd123 0 Y123

50 6.1

6.2

6.4

6.3

6.5

6.6

6.7

6.8

6.9

7.0

x in LnBa,Cu,O, Fig. 4. Comparison ofAS

20

values in Nd123 and Y123 systems.

/

0

0

:* O. l

o

o.

.

0*a .f.**.h.

o&%~***”

99

0

0”

8%QP 0"0"

+

61

62

63

64

65 x

66

67

68

69

70

in LnBa,Cu,O,

Fig. 5. Comparison of AH( 0,) values in Nd 123 and Y 123 systems.

enthalpy quantities derived from TGA measurements of Hasegawa et al. [ 171 and our EMF measurements of the NdBa2Cu30, system are shown in figs. 6 and 7.

We are currently collecting EMF data on the NdBa2Cu30, system at values of x approaching 6.0, and greater than 6.9. In addition, measurements on the GdBa2Cu30, have been initiated. The results of

M. Tetenbaum et al. / Thermodynamic behavior of high-T, oxide systems

114

60’

1

61

62

63

64

65

66

67

68

69

70

x in NdBa,Cu,O, Fig. 6. Comparison of AS( 0,) values in NdBa2Cu,0, derived from EMF measurements and TGA measurements of Hasegawa [ 13 1.

61

62

63

64 x

65

66

67

68

69

7 0

In Nd~a,Cu,Ox

Fig. 7. Comparison of Lw( 0,) values in NdBa,Cu30, derived from EMF measurements and TGA measurements of Hasegawa [ 131.

these measurements will be reported in a later paper. 3.3. Nd-Ce-G-0

(electron-doped)

superconductor NdI.slCeo.,9Cu0, [7,18]. X-ray examination of the starting material showed the expected tetragonal cubic phase and also a minor structure Reducing the oxygen partial presCco.sNdo.sOo1.75. (T,=24

system

Preliminary EMF measurements as a function of oxygen stoichiometry were made on the n-type

K, x=4)

M. Tetenbaum et al. / Thermodynamic behavior of high-T, oxide systems

sure via coulometric titration at 750°C resulted in an oxygen partial pressure plateau, p( 0,) = 3.2 x 1OF4 atm, over a narrow composition range, x= 3.97,-3.966, indicating a diphasic region. Reversibility was not achieved during oxygen uptake. The results are shown in fig. 8. An X-ray spectrum of the residue was identical to that of the starting material even though the transition temperature was found to be 13 K [ 71. Additional runs are planned over a wide range of temperatures and composition. It is apparent from the result shown in fig. 8 that the oxygen nonstoichiometry obtained by Suzuki et al. [ 191, via TGA measurements with Nd,.&e0.15Cu04_s at 750°C is much smaller, namely, 6=0.002, than that obtained via our EMF measurements. Extensive studies of Lightfoot et al. [20], on the Nd,_Ce,CuO, system by pulsed neutron and high revolution X-ray diffraction show that a single phase was formed only with the undoped

0

750

c

x

1

.

2 2 m I; 0” 3 a G 0 T

\ *\

=

.

.

a**

4

. . 5

.

0

Thus work.

X Th!s 3 Suzuki.

.

decreasing

work.increaslng

x

TGA

6 3.95

3.96

3.97

3 98

3 99

4 00

x in Nd,,8,Ceo,,,Cu0, Fig. 8. Partial pressure of oxygen as a function of oxygen stoichiometry in Ndl.sLCeO.&uOx at 750°C (Suzuki data for Ndls5Ce0.&uOx-

[lb].

115

compound NdzCu04 and for the optimum superconducting composition x= 0.165. Clearly, further characterization of these intriguing n-type (electrondoped) class of superconducting materials are required. 3.4. Miscibility gaps The most significant result is our determination of a miscibility gap for the YBa&u,O, system. If one considers the compositions between x=6.5 and x=7.0 to be a mixture of these end members, then one mixes the coulombically ordered ortho-11 structure with an ortho-I structure in which all of the 0 ( 1) sites are occupied by O- ions which have a zero net charge. Because of the strength of the coulomb interaction, the closest vacancy-02pair on adjacent chains in the basal plane is the smallest stoichiometric unit which can be considered as a high temperature mixing unit. Because of the relatively low dielectric constant (relative to water) of oxides, such pairs could not dissociate in even relatively dilute solutions in the ortho-I structure. On the other hand, O-, if it tends to form pairs is likely to do so much more weakly and one thus considers O- as the fundamental mixing unit. Thus, in the a direction in the basal plane because of Coulomb forces, one could not have close vacancies on nearest neighbor chains without a significant repulsion. Oxygens can more readily occupy 0( 1) sites on chains nearest to those containing vacancies. At low temperatures, this pairing is between planes perpendicular to the basal planes of oxygen and vacancies in relatively large domains. At higher temperatures, the vacancies and oxygens can randomize in these planes and the domains become smaller. For high temperatures, in the simplest solution model, the entropy of mixing is largest at the 6.66 composition. In addition, the effective mixing must have a dimensionality greater than unity to have a miscibility gap. This implies randomization and mixing of vacancies, 02- and Oin more than one dimension. In either case, some of the attractive forces (and the Coulomb energy) are decreased by forming a mixture which leads to a positive energy of mixing analogous to what would happen hypothetically if one could mix molten NaCl with molten argon. Such a positive interaction should lead to a miscibility gap.

116

M. Tetenbaum et al. / Thermodynamic

For the Y123 system, a simple version of this model leads to a consolute composition close to the measured value of x=6.65 which arises because of Coulomb interaction. Below the consolute temperature the miscibility gap can be characterized by the following five regions. (a) A stable region between x=6.5 and the edge of the gap where the ortho-II structure is stable. (b) A metastable region between the edge of the gap and the spinodal branch at the lower value of x. In this region there are thermodynamic barriers to phase separation which can preserve the metastable orthoII structure. (c) An unstable region where there are not thermodynamic barriers to phase separation (spinodal decomposition) and the major barrier is the kinetics of atomic rearrangement which controls the size of domains if there is a phase separation. (d) A second metastable region at relatively high values of x between the high x side of the spinodal and the high x side of the miscibility gap where the ortho-1 phase is metastable and there is a thermodynamic barrier to decomposition into the stable ortho-I and ortho-11 phases which might preserve the metastable ortho-I phase. (e) A region between the high x side of the miscibility gap and x=7 where the ortho-I structure is stable. Veal et al. [ 151 have quenched samples of YBa,Cu30, and measured the superconducting transition temperatures as a function of x. If there is some degree of phase separation due to this immiscibility during quenching, one might detect the influence of this in the composition dependence of T,.Their results between x=6.5 and 7 contain two plateaus at 60 K and 90 K at low and high values of x, respectively, with a line joining these two plateaus in a span of compositions (-6.63-6.80)which is narrower than our estimate of the width of the miscibility gap (even at values at 35 “C states above). If the 60 K value of T, characterizes the ortho-II structure and the 90 K value characterizes the ortho-I structure, then this indicates that the metastable ortho-11 structures are preserved, at least in part of the metastable regions outside the spinodal (i.e., x from 6.55 to 6.63). The sloped line for T,is probably largely inside the spinodal and, apparently the spinodal de-

behavior of high-T, oxide systems

composed domains of ortho-I and ortho-II are small enough so that T,is an average value. The difference between the NdBazCu30, and YBa2Cu30x systems is probably related to the fact that near x=6.5, a significant number of oxygen atoms occupy 0( 5) sites in NdBa2Cu30r Since oxygens occupy more kinds of sites in NdBazCuJO, it has a higher entropy and is more disordered than YBa2Cus0,. Therefore, all other things being equal, the deviations from ideal mixing should be less positive and the consolute temperature should be lower than for the YBa2Cu30x system. In addition the composition of the consolute point should shift to higher values of x; since the analogue of the ordered yttrium compound where vacancies and oxygens each occupy one half of the I( 1) sites is at a value of x higher than 6.5 for the NdBa2Cu30x system, one can (crudely) consider this to be one member of the mixing pair (the other being at x= 7.0). This is consistent with the observations of Veal et al. [ 15 1, on the composition dependence of quenched compositions. Further work is needed to define this phenomenon in the NdBa&u,O, system.

4. Summary EMF measurements of oxygen fugacities as a function of oxygen stoichiometry in YBa2Cu30, and NdBazCuJO, have been made in the temperature range 400-750°C by means of an oxygen titration technique with a yttria stabilized zirconia electrolyte. From the plots of equilibrium oxygen pressures versus x, no sign of a change in curvature was obtained at an oxygen stoichiometry of x= 6.5, where copper is divalent, and where the orthorhombic-tetragonal structural transition had been thought to occur. Of importance, the shape of the 400 oC isotherm obtained with NdBazCusO, suggests the presence of a miscibility gap at lower temperatures, at values of x that are higher than those in the YBa2Cu30, system. The shifts in the location of the miscibility gaps are consistent with the changes in the composition dependence of T, with ionic radii for Y and Nd in YBa2Cu30, and NdBazCu30x. For a given oxygen stoichiometry, partial pressures above NdBa&u,O, were found to be higher than those for YBa$&O, as reflected in the calculated partial molar quantities

M. Tetenbaum et al. / Thermodynamic behavior of high-T, oxide systems

ti(02) and fl(O,). Our results to date appear to explain the two plateaus in measured values of T, as a function of oxygen stoichiometry in orthorhombic YBa2Cu30, and appear to be consistent with the less pronounced T, plateaus found for the NdBa2CujOx system at higher values of the oxygen stoichiometry. Our results indicate that the orthorhombic-orthorhombic structural transition is associated with order-disorder transitions in the basal plane of YBa2CuJ0, and perhaps also NdBa2Cu30, compounds. The irreversible behavior of the n-type Nd-CeCu-0 system requires further study. Thermodynamic and structural studies during various stages of reduction and oxidation of the Nd-Ce-Cu-0 system are needed to define the nonstoichiometric behavior of this system.

Acknowledgement This work is supported by the US Department of Energy, Division of Materials Sciences, Office of Basic Energy Science, under Contract No. 3 1- 109-ENG38. The authors are indebted to J.P. Singh, B.W. Veal and D.G. Hinks of the Materials Science Division of ANL for supplying us with samples for use in our EMF measurements. We would like to thank B. Tani for performing X-ray analyses and A. Essling for iodometric analyses of samples for oxygen content. The authors are indebted to R. Kumar and I.D. Bloom for their expert guidance in computer operations. The technical assistance of D. Diana and M. Hash is gratefully acknowledged. We acknowledge helpful discussions with L.A. Curtiss, D.G. Hinks and J.D. Jorgensen.

References [ 1i L.A. Curtiss, T.O. Brun and D.M. Gruen, Inorg. Chem. 27 (1988) 1421. [ 2 ] A.G. Khachaturyan, S.V. Semenovskaya and J.W. Morris Jr., Phys. Rev. B 37 (1988) 2243; S.G. Khachaturyan and J.W. Morris Jr., Phys. Rev. Lett. 64 ( 1990) 76. A.G. Khachaturyan, S.V. Semenovskaya and J.W. Morris Jr., Phys. Rev. B, to be published.

117

[3] L.T. Wille, A. Berea and D. de Fontaine, Phys. Rev. Lett. 60 (1988) 1065; A. Berea and D. de fontaine, Phys. Rev. B 39 (1989) 6727; D. de Fontaine, M.E. Mann and G. Ceder, Phys. Rev. Lett. 63 (1989) 1300. 1Y.D. Tretyakov and R.A. Rapp, Trans. Met. Sot. AIME 245 (1969) 1235. J.P. Singh, private communication B.W. Veal, private communication. D.G. Hinks and J.D. Jorgensen, private communications M. Tetenbaum, A.P. Brown and M. Blander in: Ceramic Superconductors, vol. 2 Proc. Ceramic Superconductivity Symp. American Ceramic Society, ( 1988 ) 5 1. M. Tetenbaum, B. Tani, B. Czech and M. Blander, Physica C158(1989)377. ‘1M. Tetenbaum, L.A. Curtiss, B. Czech, B. Tani and M. Blander in: Physics and Material Science of High Temperature Superconductors, eds. R. Kossowsky, S. Methfessel and D. Wohlleben, NATO AS1 Series E, (Kluwer Academic The Netherlands. 18 1 1990) 279. [ 111 J.B. Goodenough and A. Manthiram, Int. J. Mod. Phys. (1988) 379. [ 121 G.J. Hyland in: Studies of High Temperature Superconductors, ed. A. Narlikar (Nova Science, Commack, NY 1989) 244. [ 131 M. Tetenbaum, P. Tumidajski, I.D. Bloom, D.L. Brown and M. Blander in: New Physical Problems in Electronic Materials ed. M. Borissov, N. Kirov, J.M. Marshall and A. Vavrek (World Scientific, Singapore 1991) 259. [ 141 K. Kishio, T. Hasegawa, K. Suzuki, K. Kitazawa and K. Fueki in: High Temperature Superconductors: Relationship Between Properties, Structure and Solid-State Chemistry, ed. J.D. Jorgensen, K. Kitazawa, M.J. Tarascon, M.S. Thompson and J.B. Torrance, Mat. Res. Sot. Symposium Proc. 156 (1989) 91. [ 151 B.W. Veal, A.P. Paulikas, J.W. Downey, H. Claus, K. Vandervoort, G. Tomlins, H. Shi, M. Jensen and L. Morss, Physica C 162-164 (1989) 97. [ 161 H. Shaked, B.W. veal, J. Faber Jr., R.L. Hitterman, U. Balachandran, G. Tomlins, H. Shi, L. Morss, and A.P. Paulikas, Phys. Rev. B 41 (1990) 4173. [ 171 T. Hasegawa, K. Kishio, K. Kitazawa and K. Fueki, Proc. Int. Conf. First Two Years of High Temperature Superconductivity, Tuscaloosa. Alabama, USA 1 l-l 3 April 1988. [ 181 Y. Tokura, H. Takagi and S. Uchida, Nature 377 (1989) 345. [ 19 ] K. Suzuki, K. Kishio, T. Hasegawa and K. Kitazawa, Physica c 166 (1990) 357. [20] P. Lightfoot, D.R. Richards, B. Dabrowski, D.G. Hinks, S. Pei, D.T. Marx, A.W. Mitchell, Y. Zhengand J.D. Jorgensen, Physics C 168 (1990) 627

1