Background of the Invention Based on the new class of superconductors, henceforth referred to as high-Tc superconductors, which were discovered by Bednorz and Müller and disclosed in their article"Possible High-Tc Superconductivity in the Ba-La-Cu-O System", Zeitschrift fur Physik B, Condensed Matter, Vol. B64,1986, pp. 189-193, a vari- ety of superconducting wires, cables and tapes have been developed for the transport of electrical current. A key parameter defining the performance and thus the economic benefit of these conductors is given by their so-called critical current density, which is the maximum current these conductors can carry as so-called supercurrents in the superconducting state divided by the cross- sectional area A of the superconductor. The critical current density is a specific property for a given superconductor, and, for the practical use of a supercon- ductor, one aims to maximize the critical current density.

Chaudhari et al. have taught in their article"Direct Measurement of the Super- conducting Properties of Single Grain Boundaries in YBa2Cu307", Physical Review Letters, Vol. 60,1988, pp. 1653-1655, that the limiting factor for the critical current density of polycrystalline high-Tc superconductors is the elec- tronic behavior of the boundaries formed by the crystalline grains of these ma- terials. It was shown that the critical current densities of these grain boundaries are smaller by one to two orders of magnitude than the critical current densities of the grains abutting the grain boundaries.

Further, Dimos et al. have taught in their publication"Superconducting Trans- port Properties of Grain Boundaries in YBa2Cu307 Bicrystals", Physical Re- view B, Vol. 41,1990, pp. 4038-4049, that superconducting grains with a small misorientation (below typically 8'to 10') behave as strongly coupled supercon- ductors whereas larger misorientations (also called large-angle grain bounda- ries) are weakly coupled, showing Josephson junction-like properties. The teaching by Dimos et al. is the finding that the detrimental effect of the grain boundaries can be reduced by aligning the superconducting grains with respect to their crystalline main axes.

Following this proposal, wires and tapes of high-Tc superconductors have been fabricated, the critical currents of which are enhanced by aligning the super- conducting grains by a variety of means, such as rolling processes or ion beam assisted techniques. Although these technologies have lead to the fabrication of high-T, superconductors with current densities of the order of 100 000 A/cm2 at temperatures of 4.2 K, it remains desirable to fabricate high-Tc supercon- ductors with still higher critical current densities or with processes which are less costly and faster than the known ones.

Mannhart and Tsuei have revealed in their publication"Limits of the Critical Current Density of Polycrystalline High-Temperature Superconductors Based on the Current Transport Properties of Single Grain Boundaries", Zeitschrift fur

Physik B, Vol. 77,1989, pp 53-59, that the critical current density of a three- dimensional conductor can exceed by an order of magnitude the given critical current density of the grain boundaries. This approach, illustrated in Fig. 1, is based on the fact that the critical current of conductor is a function of the grain boundary critical current density as well as of the effective area of the grain boundaries, which may be much larger than the cross-sectional area A of the conductor. The effective grain boundary area A'may be enhanced, e. g. , by adjusting the microstructure of the superconductor such that the grains have a large aspect ratio, the long sides of the grains being oriented predominantly parallel to the supercurrent flow. In their publication, Mannhart and Tsuei also revealed a procedure to calculate the critical current as a function of the grain aspect ratio. These calculations show that the critical current strongly increases with the aspect ratio of the grains, being ultimately limited only by the intragrain critical current densities.

It was pointed out by Mannhart and Tsuei that large critical current densities may be attainable by using superconducting films with aligned needle-shaped grains. Although this proposal shows the right way to fabricate tapes with large critical current, despite more than 10 years of intense R&D efforts on this prob- lem, no way was found to fabricate such conductors. The resulting problematic situation of the field of high-Tc superconducting cables is presented in"R. F.

Service, YBCO confronts life in the slow lane"Science, Vol. 295, page 787 (February 1 sot, 2002).

However, in accordance with the proposal of Mannhart and Tsuei, cables based on high-Tc superconductors. such as Bi2Sr2Ca1Cu208+ or Bi2Sr2Ca2Cu301o+s (BSCCO) so-called first generation cables have been fabricated, which use superconductors which contain platelet-like grains arranged in such a manner that huge effective grain boundary areas are obtained (see Fig. 2 for an illustra- tion). This is described by Mannhart in"Critical Currents in High-Tc Supercon- ductors"in"Physics of High Temperature Superconductors", Proceedings of

the Toshiba International School of Superconductivity, Kyoto, Japan, July 15- 20,1991, Springer Series in Solid State Sciences, Vol. 106,1992, 367-393 (1991). This method, also called powder in tube method, uses large grain boundary areas in bulk BSCCO with success. In this method, Ag-tubes are mechanically filled with BSCCO-powder. The tubes are then drawn and rolled into a final, tube-or tape-like shape and then fired for reaction and annealing.

The large grain boundary areas in the bulk BSCCO-filling of the tubes originate from the platelet-like microstructure of the very anisotropic BSCCO-compounds.

Unfortunately the material cost involved in the powder in tube process are so high that this technology cannot be commercially competitive. Furthermore, due to the inferior flux pinning properties of the BSCCO compounds, at the pre- ferred operation temperature of 77 K, the critical current density of such wires is low in applied magnetic fields.

Another method to fabricate conductors made of high-Tc superconductors avoids the use of bulk materials and silver (as done in the powder in tube proc- ess), but instead enhances the grain boundary critical current densities by aligning the grains by epitaxially depositing superconducting films. It is this technology, called coated conductor technology, about which the invention is primarily concerned. Tapes fabricated by the coated conductor technology are also called conductors of the second generation, as this process has the poten- tial to solve the cost-problem of the first-generation conductors. The grain alignment (texturing) required at present is achieved, for example, by depositing the superconductor on a template that has a textured surface (an overview of coated conductors is provided by D. Larbalestier et al.,"High-Tc Superconduct- ing Materials for Electric Power Applications", Nature, Vol. 414,2001, pp 368- 377 and references therein). Deposition is usually done using standard vapor phase deposition techniques such as sputtering, laser deposition, or thermal evaporation. Recently, non-vacuum techniques like sol-gel methods or dip coating have also been used for this purpose.

At present, coated conductors are fabricated predominantly by using three different processes. In all of them the conductor typically consists of a sub- strate, for example a metallic tape, a buffer layer system, which usually is based on a series of oxide films, a layer of a high-Tc superconductor such as YBa2Cu307, and possibly several doping and capping layers. In contrast to the polycrystalline superconductor of the powder-in tube conductors, the grains in the superconductor layer of the coated conductors generally form a two- dimensional network, because the high-Tc superconductor is in most cases epitaxially deposited as a polycrystalline film.

The first technique to produce coated conductors to be described here is known as the rolling assisted biaxially textured technique (RABiTS). When tapes of nickel based alloys or similar materials are rolled and suitably heat treated, the Ni-grains become textured along two of their main crystal-axes, so that the grains are aligned in all directions. This makes the metallic tape a useful sub- strate for the fabrication of a coated conductor. On the surface of the tape a buffer layer, usually composed of CeO2 and Y-stabilized ZrO2, is grown. On top of this buffer layer a high-Tc material, typically YBa2Cu307s, is deposited as film. These epitaxial films reproduce the microstructure of the buffer layer, which in turn has replicated the microstructure of the nickel alloy substrate. The thickness of the superconducting films is in the range of a few microns, the entire tapes being 25-50 micrometers thick. This process, known as the rolling assisted biaxially textured technique (RABiTS) is capable to produce low-angle boundaries (e. g. 5°-8°), consequently, the critical current density is relatively high, reaching values above 105 A/cm2 at 77 K in a field of one Testa.

Texturing can also be induced by ion beam assisted deposition (IBAD) or depo- sition under a glancing angle, which is the so called inclined substrate deposi-

tion method (ISD). In these techniques the buffer layer is textured during growth. This is done in the ISD-process by using a shallow angle between the incoming beam of adatoms and the substrate surface, and in the IBAD tech- nique by irradiating the growing film with additional ions. The critical current densities of the superconducting films, having again a typical thickness of a few micrometers, exceed 106A/cm2 at 77 K and zero external magnetic field. A limiting factor for applications of these processes is their low speed, caused by the cumbersome alignment processes.

Immense efforts are devoted in Asia, the US and in Europe to improve the coated conductor processes. Despite these efforts, possible market applica- <BR> <BR> tions are at best several years away (see, e. g. ,"R. F. Service, YBCO confronts life in the slow lane"Science, Vol. 295, page 787, February 1St, 2002). The reason is that the texturing of the tapes is a tedious and costly process. Due to this, the maximum length of the coated conductors produced today is approxi- mately ten meters only, and no practical way has been found to produce larger lengths at competitive costs. It is clear that the commercial breakthrough of conductor conductors could be obtained if the current density of the cable could be enhanced significantly for a given grain alignment. Therefore such methods are sought since many years with great intensity as described by P. Grant in "Currents without Borders"Nature Vol. 407,2000, pp 139-141. If such a method was found, one could benefit for given production costs from an en- hanced critical current, or, if the grain alignment was relaxed, from standard critical currents at much lower costs.

The present invention provides the solution to his problem.

Summary of the Invention A general objective of the present invention is to provide a solution for an in- creased use of high-Tc and other superconducting materials by improving the current transport mechanisms within such materials, in particular for increasing the upper limit of the achievable supercurrent density within such materials.

A specific objective, as mentioned above, is to provide an approach by which the current carrying capabilities of superconductors, e. g. superconducting wires or tapes, can be significantly improved in a wide temperature range.

A further objective of the invention is to provide a technique for a simpler manufacturing process, resulting in cheaper mass production of polycrystalline superconductors with large critical current densities compared with presently used techniques. At present, the fabrication of such superconductors requires cumbersome and costly processes to optimize the grain boundary alignment.

The novel solution taught by the invention is based on the understanding that the critical current density of coated conductors is a monotonously increasing function of the grain boundary critical current density and of the effective grain boundary area. Therefore, by enhancing the effective grain boundary area, the critical current of such a conductor will considerably exceed the product of the grain boundary critical current density and the geometrical cross-section of the conductor.

It is the gist of the invention, that for polycrystalline superconductors consisting of grains which are significantly thinner than long or wide, and which are ori- ented predominantly with their flat side parallel to the superconductor's surface, the effective grain boundary area can be readily enhanced. Intriguingly, such superconductors are the typical superconductors fabricated with the coated conductor technologies. A major enhancement of the grain boundary area is

achieved by joining two such superconductors with their superconducting sides facing each other, such that a good superconducting contact is established between the two superconducting layers, as sketched in Fig. 3 and Fig. 4. In this case the supercurrent can meander along the tapes, in part bypassing the grain boundaries within one layer by shifting into grains of the other layer as illustrated in Fig. 3. Hereby the supercurrent can readily pass across the grain boundaries associated with the transition from one layer to the others, as these grain boundaries have a very large area.

In the result a novel superconductor is obtained, the critical current of which exceeds the sum of the critical currents of its constituents.

It is the core of the invention that this enhancement of the critical current den- sity is achieved particularly easily, as various straightforward and inexpensive technologies can be used to establish the required superconducting contacts between the layers.

A preferable way to fabricate superconductors containing flat grains with large aspect ratios is to use the RABiTS technology. The growth is started by using a polycrystalline, textured substrate. Such a template layer is routinely produced by conventional metallurgical processes. In case it is desired, but in many cases this will not even be necessary, these grains may also be oriented such that low angle grain boundaries are formed, as is done in the standard RABiTS process. Onto this layer a buffer layer and the high-Tc superconductor are de- posited as films, such that the shape and the orientation of the grains of the template layer is replicated by the grains of the oxide layer and by the grains of the superconductors.

Coated conductors containing flat, superconducting grains with large aspect ratios may also be obtained using the IBAD or ISD technologies

A superconducting layer fabricated by one of these techniques, consists of a sheet of flat superconducting grains. The critical current density of such a grain- network is given by the critical current density of the so-called limiting path. The limiting path is an area crossing the sheet which is composed by those grain boundaries, of which the sum has the smallest critical current of all such areas.

The effective area of the grain boundaries in the limiting path is approximately given by the product of the width and the thickness of the superconductor. This area is small, as the superconductors are usually thin (about 0.5 pm to 1.5 lit).

However, if two of such sheets are brought into intimate superconducting con- tact, the supercurrent can bypass the grain boundaries in a given sheet, by meandering into the other sheet, as shown in Fig. 3. Thereby the critical current density will be greatly enhanced. This is because the widths and lengths of the grains considerably exceed the film thickness. The effective grain boundary area available for the meandering is therefore much larger than the grain boundary area limiting the current flow within one isolated sheet. This is obvi- ous, because for a given grain, the area available for the current transfer into the other sheet is proportional to the product of the width and length of the grain, whereas the grain boundary area for the current flow within a sheet is approximately the product of the width and the thickness of the grains. For standard RABiTS tapes, for example, the grains have a size of the order of 100 m*100 m*1 jm (fength * width * thickness), so that both areas differ by two orders of magnitude.

Due to the large grain boundary area the current can therefore transfer easily from one sheet to the next, even if the critical current density of an intersheet grain boundary happened to be small. Likewise, due to the large areas in- volved, the relatively small critical current density of high-Tc superconductors for c-axis conduction has no limiting effect. Having passed into the second sheet, the supercurrent can readily flow within a'grain, due to the large value of the intragrain critical current density. For a large fraction of the grains in the second

sheet, the current can bypass a grain boundary in the first sheet and then transfer again into the first sheet, in many cases without the need to cross a grain boundary with a low critical current density in the second layer (see Fig. 3). The same arguments apply vice versa for a the current flowing in the second sheet. Therefore such a double tape will have a critical current density which will exceed significantly the sum of the critical current densities of the individual sheets, the ultimate limit being given by the intragranular critical current density.

In an example, we consider two sheets, each of which consists of grains of a size of approximately of 100 m*100 m*0. 5 m (tength * width * thickness) with an intragrain critical current density of 5*106 A/cm2. Each sheet is sup- posed to have a critical current density of 105 A/cm2 and both sheets are as- sumed to be coupled by a superconducting contact with a critical current den- sity of 1*104 A/cm2. From this we estimate that on average for each grain an area of more than 2 500 lim2 is available for current transfer into the next layer.

This area can carry a supercurrent of 250 mA, which, in principle may bypass the grain boundary within the first sheet as it corresponds to a intragrain current density of 5*105A/cm2. The boundary in the first sheet carries by itself only a supercurrent of 50 mA. Although this estimation does not consider in any detail complicating percolation effects, it is obvious that a large gain in enhancement of the critical current (from 50 mA to 250 mA in the example) can be achieved by joining such two sheets.

The increase of the critical current of the conductor will take place also in ap- plied magnetic fields and for all temperatures below the superconducting tran- sition temperature, which presents significant advantages for applications.

Another advantage of the invention is that it simplifies the manufacturing proc- ess of superconductors for technical applications by allowing cheaper mass production of polycrystalline high-Tc and other superconductors with large criti-

cal current densities. This is the case because the large current densities achieved by the invention relax the requirements on the costly and tedious grain alignment.

In the following, implementations of the invention shall be described, illustrated by the appended drawings.

Brief Description of the Drawings Fig. 1 shows a sketch to illustrate that in a polycrystal grain boundaries provided by grains with large aspect ratios (on the right) can have an effective area A'which considerably exceeds the cross-sectional area A of the polycrys- tal. The figure has been taken from Mannhart and Tsuei"Limits of the Critical Current Density of Polycrystalline High-Temperature Superconductors Based on the Current Transport Properties of Single Grain Boundaries", Zeitschrift for Physik B, Vol. 77,1989, pp 53-59.

Fig. 2 shows a scanning electron microscopy image of a Ag-sheated Bi- based powder in tube conductor. Interconnected sheet-like BSCCO platelets are clearly visible. The figure has been taken from Y. Yamada et al."Properties of Ag-Sheated Bi-Pb-Sr-Ca-Cu-O Superconducting tapes Prepared by the Intermediate Pressing Process", Jpn. J. Appl. Phys. Vol. 29,1990, L 456-458.

The micrographs (a), (b), (c) present side-views, as shown in the sketch in the upper left.

Fig. 3 is an illustration showing how a current can meander between two sheets of polycrystalline superconductors, bypassing the critical current limiting grains boundaries in the sheets.

Fig. 4 is an illustration showing as an example how a superconductor with a large critical current density can be formed by joining two coated con- ductors.

Fig. 5 is an illustration of a cross-section of a superconductor containing two superconducting sheets on two substrates. The substrates are clamped by a mechanical joint, so that the resulting force establishes a superconducting contact between the two superconducting sheets.

Fig. 6 is a sketch of a superconductor containing six superconducting layers or sheets on four substrates fabricated with a coated conductor process.

The superconducting sheets are welded together by melting the intermediate layer, which has a lower melting temperature than the two superconducting sheets.

Fig. 7 is an illustration of a superconductor containing two supercon- ducting sheets deposited two-sided on a flexible substrate. By rolling the sam- ple, the superconducting sheets are brought into contact via an intermediate layer. The superconducting sheets are welded together by melting intermediate layers, which have a lower melting temperature than the two superconducting sheets.

Description of Examples and Embodiments We provide in the following examples for the fabrication of the superconductors according to the invention. As basis for the examples, we use standard coated conductors available at present.

Example 1 : In the first embodiment, two YBa2Cu307-8 tapes are used which are fabricated by a standard RABiTS Procedure. As substrate a Ni-alloy tape, e. g. , Ni-W with a thickness in the range of 20 gm to 100 ; j, m, is employed. According to the standard, published procedures, the tape is rolled and heated, so that a surface texture with aligned grains is produced, although the alignment of the grains is not a critical prerequisite for the invention. The surface orientation of the tape (for Ni, e. g. (111) ) has to be chosen to be appropriate for the later growth of a superconductor. On such tapes a buffer layer system is deposited. Such a buffer layer, which itself may consist of various sublayers such as CeO2NsZ/CeO2, is deposited on the carrier tape to prevent chemical reactions between the high-Tc film and the carrier tape material, or, e. g, the oxidation of the carrier tape during the growth of the superconductor.

A variety of materials have been found useful as buffer-layers. Beside the YSZ and Ce02 compounds already mentioned, NiO, LaTiO3, MgO or SrTiO3 have been used successfully, to name a few examples. These buffer-layers are usu- ally grown by standard technologies, such as sputtering, e-beam, or thermal evaporation, pulsed laser deposition or metal organic chemical vapor deposition (MOCVD). But also other non-vacuum techniques may be employed, such as dip-coating, spray-coating, or painting techniques.

Like the buffer-layers, for the high-Tc superconductors a variety of materials, or multilayers may be used, again deposited by a multitude of techniques. Stan- dard materials are, for example, members of the so-called 123-family, com- prising the ReBa2Cu3Ox family, where Re is Y or a rare earth and x is a number of the order of seven. Into the superconductor additional layers, e. g. , doping layers may be embedded, as described by P. Grant in"Currents without Bor- ders", Nature Vol. 407,2000, pp 139-141. Here we will consider as example YBa2Cu307-s. Those skilled in the art will easily recognize many more varia-

tions of this scheme. These layers may be grown by using a variety of film growth techniques, such as such as sputtering, e-beam or thermal evaporation, pulsed laser deposition or MOCVD. But here again, other non-vacuum tech- niques, like, e. g. , dip-coating, spray-coating, or painting processes may be utilized.

The two tapes fabricated this way are then mounted together with their super- conducting layers facing each other. During the time between growth of the superconductor and joining of the tapes, the superconductors should not be exposed to any atmosphere causing degradation of the film's surfaces, includ- ing humidity. Therefore these process steps have best to be carried out in a nitrogen or oxygen atmosphere. A firm contact between the superconducting layers is established by mechanically fixing the tapes, for example by folding and pressing the two substrate tapes as shown in Fig. 5.

Example 2 : The second example is again based on coated conductor tapes, fabricated as described in example 1, the difference being that the buffer layers and super- conductors are deposited on both sides of the tapes. After the high-Tc super- conductor, e. g. YBa2Cu307 8 has been deposited, another layer, the intermedi- ate layer, is grown, which is also superconducting, but has a lower melting temperature than YBa2Cu307. s. As revealed by T. Puig et al. in"Self-seeded YBCO welding induced by Ag additives", Physica C Vol 363, pages 75-79, such a layer may consist, for example, of a 15 wt% Ag + 0.7 YBa2Cu307 0+0. 3Y2BaCu05 composite, the melting temperature of which is about 40°C lower than that of YBa2Cu307. The superconductors are again placed on top of each other and are heated under slight pressure to a temperature of about 990°C in an appropriate atmosphere (e. g. 0.5 bar of 02), as revealed by T. Puig et al. A superconductor fabricated this way is sketched as illustration in Fig. 6. It

is clear, that several of such tapes can be welded together. A similar procedure to establish a suitable contact between two superconductors is revealed in the European Patent Application 93102579.5, publication number: 0 556 837 A1.

Example 3 : The superconductor presented as third example is based on a double-sided superconducting tape including a layer with a lower melting temperature than described in example 2. This tape is then rolled up, as sketched in Fig. 7, and welded together like the superconductor in example 2.

In a modification of this embodiment, a tape covered with a superconducting layer, possibly even on one side only, is folded such that the superconducting layer is folded onto itself and a superconducting contact is established between various areas of one single superconducting layer. This folding step can be repeated several times to obtain a stack of superconducting double-layers.

It obvious to persons working in the field of superconductivity that, by using the invention, not only tapes for current transport can be fabricated, but that the invention also lends itself for the production of superconductors to be used for other purposes. By using the invention it is readily possible, for example, to fabricate superconducting foils or plates, which do not only have a substantial length, but also have a large width. Such foils or plates are highly suitable, for example, for magnetic shielding purposes.

Persons skilled in the art can apply this invention for cheaper mass production of superconducting tapes or wires, operated preferentially at 77 K or at any other temperature suitable for power applications or the design of strong mag- nets.

The main advantages offered by the invention, namely the enhanced current- carrying capabilities and the reduced sensitivity to magnetic fields, reduces the costly and cumbersome requirements to align the superconducting grains for wires as it is presently done for coated conductors. These properties make a superconductor according to the present invention perfectly suitable for many applications, e. g. as superconducting cables, wires or tapes, which have not been competitive and thus hardly marketable until now. With the invention, superconductors will become less expensive and thus broadly marketable and competitive for many applications.

Based on the above description and the given examples and applications, a person skilled in the art can easily vary the described invention, its values, and its materials given above, and adapt the invention to other implementations without departing from the inventive gist as defined in the appended claims.