FIELD OF THE INVENTION

The invention comprises a method for preparing HTSC materials having high critical current.

BACKGROUND

Many hιgh-T _Λ superconducting cuprates (HTSC) are known to have superconducting transition temperatures, T _r exceeding the temperature at which liquid nitrogen boils, 77 K. As such they have a potentially large number of applications ranging from power generation, distribution, transformation and control, to high-field magnets, motors, body scanners, telecommunications and electronics. T _r values may be of the order of 93 K for example for YBa^Cu ₃ 0 _{7 i} , 95K for example for Bi ₂ SR ₂ CaCu ₂ 0 ₈ , 109 K for example for Bi ₂ Sr ₂ Ca,,Cu.,0| _ϋ , 120K for example for TIBa ₂ Ca ₂ Cu ₃ O ₁₀ or as high as 134 K for HgBa ₂ Ca ₂ Cu,O _]0 . For many of these applications such T _r values alone do not guarantee the utility of these HTSC at 77K or high temperatures Often these applications require large critical currents in the HTSC in the presence of a magnetic field. Even if the grains of the HTSC are crystallographically aligned, otherwise known as textured, and well sintered together, as is commonly achieved in thin-films, such that weak links between the grains are removed, a high critical current in the presence a magnetic field is not guaranteed. Such high currents are only achieved if there is strong flux pinning within the individual grains.

A convenient measure of the intrinsic ability of HTSC materials to support high critical currents in the presence of a magnetic field at a particular temperature is the temperature-dependent irreversibility field, H*(T). For magnetic fields exceeding H* the magnetisation is reversible and hence dissipative while it is irreversible for fields less than H* and hence substantially non-dissipative. If, for a given temperature. H* is large then, at that temperature, the critical currently density maybe high provided that the field H<H*.

As the superconducting Cu0 ₂ layers are weakly Josephson coupled to each other H* is known to be dependent upon the spacing between these Cu0 ₂ layers. If this spacing is small H* is large. In particular, the Josephson coupling energy is dependent upon the c-axis conductivity and within a tunnelling model for c-axis transport this varies as exp(- d/Q) where d, is the interlayer spacing between Cu0 ₂ layers and 0 is a correlation length. Thus H* is proportional to exp (d/C) the proportionality being dependent upon the hole concentration and upon a universal function f(T/T _c ). A convenient hole concentration for comparing large number of HTSC materials is that at optimal doping when T _c =T _{( maΛ} . A convenient temperature for comparing a large number of HTSC materials is T=0.75T _t . Evidently, if the spacing d _< can be significantly reduced then H* can be increased.

SUMMARY OF THE INVENTION

In broad terms in one aspect the invention comprises a method of preparing a high temperature superconducting cuprate material (HTSC) having a crystalline structure comprising intrinsically superconducting Cu0 ₂ layers and non-superconducting intermediate layers between the superconducting layers in the crystalline structure of the material, comprising preparing the material such that at least one intermediate layer

between the CuO _a layers is conductive whereby superconductivity can be induced in the intermediate layer(s) between the superconducting layers by proximity effect to thereby increase the critical current of the material.

Preferably the method includes causing the intermediate layer(s) to become conductive by introducing sufficient oxygen into the intermediate layer(s) during preparation of the HTSC material.

The HTSC may be a Bi-Sr-Ca-Cu-O based HTSC and said intermediate layers comprise Bi-O layers. The HTSC may be of nominal composition Bι ₂ Sr ₂ CaCu ₂ O _s where Bi may be partially substituted by Pb, Hg, Re, Os, Ru. Tl. V. Cr. Zr, Nb. Mo, Hf, Ta. W, Co, or Sm, Sr may be partially substituted by Ba or a larger lanthanide rare earth element, and Ca may be partially substituted by Y or a lanthanide rare earth element. The HTSC may be of nominal composition Bι ₂ Sr ₂ Ca ₂ Cu ₃ O[ ₀ , where Bi may be partially substituted by Pb, Hg, Re, Os, Ru, Tl. V, Cr, Zr, Nb, Mo, Hf, Ta, W, Co, or Sm, and Sr may be partially substituted by Ba or a larger lanthanide rare earth element, and Ca may be partially substituted by Y or a lanthanide rare earth element.

The HTSC may be a Tl-Sr-Ca-Cu-O based HTSC and said intermediate layers comprise Tl-O layers.

The HTSC may be an Hg-Ba-Ca-Cu-0 based HTSC and said intermediate layers comprise Hg layers.

In accordance with the invention the effective interlayer spacing d, referred to above us reduced and H* is thereby increased. The method of the invention may be described as

metallising one or more of the non-Cu0 ₂ layers between groups of Cu0 ₂ layers. The Cu0 ₂ layers, which intrinsically superconduct, induce superconductivity in the metallic intermediate layer(s) by proximity effect and then the d, spacing is not the spacing between the Cu0 ₂ layers but the smaller spacing between a Cu0 ₂ layer and the intermediate layer. If the intermediate layer lies halfway between the Cu0 ₂ layers then the d, spacing is halved and H* may increase 10 fold.

BRIEF DESCRIPTION OF THE FIGURES

The invention is further described with reference to the accompanying figures wherein:

Figure IA is a diagram of the structure of the bismuth oxide layer of Bι ₂ Sr ₂ CaCu ₂ O _s , _A showing the pseudo-perovskite strip of structure running through the general background rocksalt structure, Figure IB shows additional oxygen loaded into the pseudo-perovskite strip to provide a direction connectivity of the strip, and Figure 1C shows the ideal structure for Bι ₂ Sr ₂ CaCu ₂ O _β+4 where a single additional oxygen atom is added to each BiO layer to convert the layer from rocksalt to perovskite and thus reduce the Bi-O bondlength;

Figure 2 is a diagram showing T _c plotted against the muon spin depolarisation rate of σ(o) for several HTSC materials showing the typical loop followed by most HTSC and the plateau for Y-123 and Hg, Re- 1212 due to the enhanced superfluid density occurring within metallised interlayers;

Figure 3 is a diagram of the irreversibility field H* plotted against T/T _r for a range of HTSC materials; and

Figure 4 is a diagram of H* (evaluated at T=0.75T _C ) plotted against the d, spacing separating the superconducting Cu0 ₂ planes - the effective d, spacing for Y-123, Y-247 and Hg, Re-1212 can be seen to be halved due to the metallisation of any mterlayer, the CuO chains in the former two and the (Hg.Re)O layer for the last.

DETAILED DESCRIPTION

The structure of Bi-Sr-Ca-Cu-O (BSCCO) based HTSC is characterised by perovskite packing of the Ca-Cu0 ₂ -SrO layers and rocksalt packing of the B1O-B1O layers. In these compounds the stacking of interlayers between CuO,. layers follows the sequence: CuO _z - SrO-BιO-BιO-SrO-Cu0 ₂ so that adding additional oxygen to or metallising the BiO layer substantially reduces the d, spacing. The in-plane separation of a pair of Bi atoms is 0.54nm and if the linking oxygen atom were to reside in the mid-pomt position the in- plane BiO bond length would be 0.27nm. This is too long for metallic bonds and too long for a typical BiO bondlength. To satisfy the latter requirement the oxygen atom typically sits closer to one Bi atom than to the next thus resulting in the bondlength of about 0.24nm which is, however, still too long to metallic bonds. The structural modulation known to exist in these compounds is provided by an m-plane strip of pseudo-perovskite packing as shown in Figure IA. Here the bond length is 0.193nm but oxygen vacancies are distributed along the strip so that there is no connectivity of the short bond length sufficient to metallise this strip as shown in Figure IB. Preierably substantially the entire bismuth oxide layer is converted to perovskite packing so that the Bi-O bondlength is between 0.19 and 0.22nm, and typically between 0.19 to 0.20nm as shown in Figure 1 C. This is accomplished by adding nearly one oxygen atom per formula unit to the BiO layer such that it becomes a Bi0 ₂ layer and the formulae become Bi ₂ Sr ₂ CaCu ₂ 0 _{1(W A} or Bι ₂ Sr ₂ Ca ₂ Cu ₃ 0 _12tA , in both cases where -0.2< δ < +0.2.

Typically additional oxygen is introduced by annealing the material in an oxygen containing atmosphere. This may take place when forming long-length flexible wires or tapes by the technique known m the art as powder-in-tube processing. Powders of these materials or precursors to these materials are packed into a metallic tube, often made of silver metal or silver alloy, and then by a process of deformation and heat treatment the tube is drawn out into a long wire and oxide reacted to form a highly-textured HTSC core. Oxygen may also be added to electrochemically loading oxygen. Oxygen may be introduced by substituting a higher valence cation for a lower valance cation. A combination of any of the above may be used.

In BSCCO compounds the stacking of interlayers between Cu0 ₂ , layers follows the sequence: Cu0 ₂ -SrO-BιO-BιO-SrO-Cu0 ₂ so that adding additional oxygen to, or metallising the BiO layer, substantially reduces the d, spacing. In BSCCO materials in particular additional oxygen may be introduced in the BiO layer by substituting a higher valence atom for Bi such as Pb, Hg, Re, Os, Ru, Tl, V, Cr, Zr, Nb, Mo, Hf, Ta. W, Co or Sm. The bond length may be lengthened to greater than 0.20 but less than 0.22 by further substituting sufficient Ba for Sr. In BSCCO materials in which Bi is partially substituted by Pb, the BiO bond length may also be reduced by increasing the substitution level of Pb because Pb ²⁺ is a larger ion than is Bι ^1τ . For the same reason the BiO bond length may be shortened by substituting Hg^ for Bi ^J+ .

In Tl-Sr-Ca-Cu-O compounds the stacking of interlayers between Cu0 ₂ , layers follows the sequence: Cu0 ₂ -BaO-T10-BaO-Cu0 ₂ so that adding additional oxygen to, or metallising the TIO layer, substantially reduces the d, spacing The HTSC may be of nominal composition TIBa _z Ca. ,Cu _n 0 _{2n f 1} or ,Cu_0 _{2n -} „ where Tl may be partially substituted by Bi, Pb or Hg, Ba may be partially or wholly substituted by Sr, and Ca may

be partially substituted by Y or any larger lanthanide rare earth element. A preferred compound is Tl ₀₅ Pb ₀₅ Sr ₂ Ca ₂ Cu _n O _{2-τ i} .

In Hg-Ba-Ca-Cu-O compounds the stacking of interlayers between Cu0 ₂ layers follows the sequence: Cu0 ₂ -BaO-Hg-BaO-Cu0 ₂ so that adding additional oxygen to or metallising the Hg layer substantially reduces the d, spacing. The HTSC may be of nominal composition HgBa ₂ Ca„ ,Cu _n 0 _2n , ₂ where n= 1 ,2 or 3, Hg may be partially substituted by Bi. Pb, Tl, Sr, Re, Os, Ru, Tl, V, Cr, Zr, Nb, Mo, Hf, Ta, W, Co, or Sn, and Ba may be partially or wholly substituted by Y or a larger lanthanide rare earth element, and Ca may be partially substituted by Y or any lanthanide rare earth element. Preferred compounds are those in which 25% of the Hg atoms are substituted by Re such that the Hg layer becomes an Hg _{o 75} Re ₀₂₅ 0 layer which is metallic.

The invention is further illustrated by the following examples:

Example 1

Polycrystalhne samples of YBa ₂ Cu ₃ 0 ₇₈ (Y-123) were synthesized using standard techniques from oxide precursors. These were annealed at different temperatures and oxygen partial pressures in order to provide a series of samples with different δ values which were then quenched in order to freeze m the oxygen content. The muon spin depolarisation rate, σ(O), was determined for each of the samples and is plotted in Figure 2 in the form of T _c plotted against σ(O). σ(O) is a measure of the superfluid density. In contrast to other HTSC materials which follow a loop as shown by the solid curve and squares in Figure 1 as they progress from the underdoped to the overdoped region, Y-123 follows a broad plateau shown by the open triangles. Here it can be seen that the

superfluid density increases dramatically as δ is reduced to zero, that is as the CuO chains are filled. This enhanced superfluid density arises from the fact that the CuO chains are metallic and superconductivity is induced on the chains by proximity to the Cu0 ₂ planes even though the chains may not be intrinsically superconducting. Figure 3 shows H*(T) plotted as a function of T/T _c for a variety of HTSC materials all carefully adjusted so as to be optimally doped with T _r =T _{c max} . Y-123 and the related compound Y ₂ Ba ₄ Cu ₇ 0 ₁₅ . ₄ (Y-247) can be seen to have very high values of H* in comparison with other materials. Like Y-123, Y-247 CuO chains and when these are filled, that is δ is reduced to zero, they also become metallic and superconduct by proximityeffect. The large values of H* are to be attributed to this induced superconductivity on the metallic chains. Figure 4 shows H* at T=0.75T _C plotted as a function of the d, spacing for each compound. The bismuth, thallium and mercury HTSC materials all follow the expected exponential dependence on d, but Y-123 and Y-247 (open circles) have values of H* which are too high for the correlation. If, however, the chains are superconducting by proximity effect then the relevant effective d, spacing is the plane-to-chain distance which is half of the plane-to plane distance. Using this shorter distance the data is plotted in Figure 4 for Y-124 and Y-247 using the solid circles. These points can be seen to match the correlation for the other HTSC materials. This was further checked by investigating a Ca-substituted Y-123 sample: Y ₀₈ Ca _{O 2} Ba ₂ Cu,O _{7 β} . The Ca substitution overdopes the sample so in order to achieve optimal doping some oxygen must be removed from the chains. For this sample δ=0.39 and the chains are so oxygen deficient that the metalhcity of the chains is destroyed and induced superconductivity on the chains is suppressed. As a consequence H* is reduced by nearly a factor of 10 and the data point is plotted in Figure 4 as Ca-123. Here the effective d, spacing can now be seen to be the full plane-to-plane distance as superconductivity on the chains is destroyed. Similar results are obtained when the chains are oxygen depleted then brominateri. The sample is optimally doped but the

chains are not metallic or superconducting and the data point annotated Br-123 shows also that the effective d, spacing is the plane-to-plane distance.

Example 2

Samples of Re-substituted mercury HTSC materials (Hg _o „Re ₀₂₅ Sr ₂ CaCu ₂ O ₇ ) were synthesized by high pressure techniques together with non-substituted samples (HgBa ₂ CaCu ₂ 0 ₆ and σ(O) and H* data for these samples, annotated Hg, Re- 1212 for the former and Hg- 1212 and Hg- 1223 for the latter are plotted in Figures 2,3 and 4 and the same behaviour can be seen as described in Example 1 for Y- 123. The Hg- O layer in Hg- 1212 and Hg-1223 which lacks any oxygen and is therefore insulating, on substitution of 25% Re for Hg takes in 1 oxygen atom per formula unit. This renders the layer metallic and therefore superconducting by proximity effect with considerable enhancement in H*. Figure 4 shows that from HgBa ₂ CaCu ₂ 0„ to Hg _o „Re ₀₂₅ Sr ₂ CaCu ₂ 0 ₇ there is a 50-fold increase in H* partly due to the replacement of Ba by the smaller Sr atom (which therefore shortens the d, spacing) and partly due to metallisation of the (Hg.Re)O layer. Other Re substituted mercury HTSC materials such as Hg _{ι y} Re,Ba,CaCu ₂ 0 _6^Λ ) shown in Figure 2 by the solid down triangles exhibit similar enhancements m superfluid density and irreversibility fields.

Example 3

Samples of Bι _{l s} Pb _o ,,Sr _{1 6} La ₀ .,CaCu ₂ O _8+A and Bι _{1 6} Pb ₀ ^Sr _{1 7} La ₀₂ CaCu _? O _{8 Λ} were synthesized by solid state reaction of precursor oxides at 840°C in air, then 860°C in air then at 910°C in air. Each reaction was carried out for 12 hours followed by grinding and die-pressmg pellets for further reaction. Samples were slow-cooled in oxygen flow from 750°C to

400°C. The samples took on a surprisingly large quantity of oxygen. Weight change showed the increase in δ to be 0.554 for the former and 0.623 for the latter. This is to be contrasted with changes in δ for the parent compounds of about 0.07. The substitution of La for Sr evidently has the desired effect of allowing a large increase in oxygen content in the BiO layers. As the same time the thermoelectric power reduced from 49.3μV/K (indicating strongly underdoped) to a 8.43μV/K (indicating weakly underdoped: S=2μV/K indicates optimal doping) for the former and from 38.5μ to 3.06μV/K for the latter. A subsequent anneal at 60 bar oxygen pressure in which the sample was cooled from 450°C to 400°C yielded total changes in δ of 0.614 and 0.675, respectively, and final thermoelectric power values of 5.33μV/K and 0.49μV/K, respectively, the last value indicating a slightly overdoped sample.

The foregoing describes the invention. Alterations and modifications as will be obvious to those skilled in the art are intended to be incorporated in the scope hereof as defined in the following claims.