DESCRIPTION
CERAMIC SUPERCONDUCTING DEVICES AND FABRICATION METHODS
Field of the invention
This invention concerns superconducting materials and their manufacture and is directed to improvements in the design and fabrication of high critical temperature superconducting composite conductors. Typically such conductors depend for their superconducting properties on high field, high critical temperature superconducting oxide or nitride phases, and possess superconducting properties above 77°K.
Background to the invention
Although superconducting composite conductors have been available for some years they have hitherto depended on metallic superconducting materials for their properties. Examples of existing commercial composites are Nb-Ti alloy composites and composites based on A15 intermetallic compounds such as Nb.Sn. A new class of higher critical temperature, high field, superconducting oxide phases presents new opportunities for applications of superconductivity but has created new problems for the design of high performance conductors.
Some features of the new materials make the design of a practical conductor or wire particularly difficult.
- 2 -
In the first place the materials are inherently brittle, and this raises acute problems for the support a conductor under the extreme conditions of mechanical stress encountered in many high field applications.
Secondly in their optimum superconducting state these materials are chemically extremely active, in particular exhibiting extreme sensitivity to the presence of oxygen and moisture, and in view of their extreme chemical activity, the components are also susceptible to corrosion and environmental degradation.
In certain respects the design of a composite conductor for operation at 77°K and above presents fewer problems than at lower temperatures. Thus for example the specific heat of materials is an order of magnitude, or more, greater, at elevated temperatures and this makes the design of a thermally stable conductor simpler.
Also, for operation under steady state conditions it is likely that conductors with superconductor cross-sections, larger than commonly used in currently available- composites, will be acceptable.
The invention seeks to provide a method of construction and manufacture which enables brittle reactive superconducting oxide or nitride phases to be incorporated into a robust, prestressed, environmentally protected high critical current composite conductor, and to provide a form of construction in which the conductor may be reactivated into an optimum state should its performance deteriorate with time.
Summary of the invention
According to one aspect of the invention thre is provided a method of forming a superconducting composite from an outer supporting casing of a non-superconducting inert material having active high temperature superconducting material in the form of a thick walled tube located therewithin, which involves the step of passing oxygen through the hollow interior of the tube to produce oxidation of the active material, to create a superconducting oxide phase after fabrication.
Fabrication may involve for example laying up, or winding into a coil, incorporation into a machine or device.
The composite may be cooled or heated to obtain the devised oxidation.
The internal oxidation process may be carried out in more than one step, with oxygen being reintroduced into the hollow interior at intervals after previously introduced oxygen has been partially consumed in the reaction process.
The oxygen may be introduced under high pressure.
Where appropriate, oxidation is controlled so as to produce a superconducting phase having the properties for the particular application in mind.
In particular dimensional changes such as contraction on cooling of the superconductor after reaction, and differential thermal contraction as between the conductor and support tube on cooling from the reaction temperature to the working temperature, can be used to place the
- 4 - superconducting phase under compressive stress. Prestressing in this way avoids brittle in service failure of the superconductor.
Preferably the hollow interior is defined by a tube of oxygen permeable metal such as a silver or alloy. This inner tube gives support to the superconducting material during fabrication and reaction but is penetrable by oxygen to enable the oxidation step to be performed.
The final composite conductor will normally be required to be of relatively small cross-section and relatively much greater length. To this end the composite may be formed by extrusion over a mandrel so as to provide the hollow interior and if initial extrusion does not provide the desired final cposs-sectional size, the latter may be achieved by drawing in known manner. Thus during the initial stages of composite fabrication the region of the cross-section which is to comprise the hollow core may be filled with a suitable material which can be removed readily after fabrication. This material may be left in place during any subsequent drawing to further reduce the cross-sectional area of the composite. However at the end, -it must be capable of being removed so as to enable oxygen to be pumped along the hollow interior to produce the oxidation of the material coating the inside of the supporting casing to form the superconducting material previously referred to.
This process, often referred to as a substitution process, may for example involve the use of a ductile metal or metal alloy, wax or the use of water or other liquid such as castor oil which may be frozen into the cross-section region of the composite prior to the fabrication stages
— _> - and then removed by heating or other convenient means typically under pressure so as to enable the cross-section of the final fabrication to be heated and then oxidised in the manner required by the invention.
It is to be understood that the invention is not limited to a composite with only a single superconducting element within the cross-sectional area of the composite, but envisages composites having two or more such superconducting elements within the cross-section of the composite, extending the length of the composite.
The starting materials from which the superconducting layer is to be formed may for example consist of mixed powdered oxides, carbonates, or other compounds such as co-precipitated nitrates or metals in any combination, and typically the starting materials are pre-treated, reacted, graded and preferably magnetically separated prior to insertion into the composite.
Most of the starting materials referred to are in the form of a clinker when in the solid and the first' step will typically involve grinding the clinker or other solid form into a powder. If this is then cooled to a temperature at which the material becomes superconducting (typically the temperature of liquid nitrogen) a magnetic field will cause any superconductive powder to move due to the interaction between the magnetic field and the induced currents in the conductive material and this can be used to separate superconductive powder particles from non- superconducting particles. Thus for example the superconductive particles can be caused to levitate or simply be deflected depending on the direction of the magnetic field and the design "of the separator. The non-
separated material may typically be reprocessed and subjected to a further magnetic separation, to further refine the material.
Carbonates may be used as the raw material if an oxide is not available or is not suitable. Mention has also been made of the use of co— recipitated nitrates as possible starting materials and examples are typically co- precipitates from aqueous solution of yttrium nitrate, barium nitrate and copper nitrate.
Where a substitution process is- employed and the hollow" core is filled with a material which is to be removed after final fabrication and before the oxidation stage, the composite will normally have to be cooled before it is subjected to extrusion or swaging or wire drawing techniques since all of which normally increase the temperature of the composite and could cause a failure of the cavity forming characteristic of the substitute material within the core of the composite.
In a preferred embodiment of the invention, the outer supporting casing is formed from copper or stainless steel and the wall of the hollow core is formed from silver or silver alloy with the high temperature superconductor material sandwiched therebetween. Preferably the inner surface of the outer supporting casing is nickel or silver or tantalum lined.
Typically a composite has an initial diameter of 75mm with an outer support casing thickness-of 12.5mm, a powder annulus having a radial thickness of some 15mm, an inner copper liner of 1mm thickness and a central core of substitute material of diameter 18mm. After appropriate
- 7 - processing as by drawing, extrusion etc., the fina diameter of the conductor would be of the order of 1 to 2mm.
Preferred materials for the substitute core material are low melting point lead alloys, wax, and inorganic salts.
Where materials such as aluminium etc. have been employed as the core substitute, chemical means may be employed for removing the material.
Where the material can be removed by application of gas pressure, the removal of the material and the oxidation of the superconductive material, may be achieved simultaneously by using oxygen under pressure to remove the unwanted substitute material.-
Where the final form of the composite requires that different points along the length of the composite must be electrically separated from adjoining sections thereof, as when wound in the form of a coil, the composite preferably includes an outer electrical insulating layer.
Where the environment of the final composite is such that the brittle nature of the superconducting material may cause problems, appropriate non-reactive strengthening rods, fibres or mesh material may be located within the powder sleeve of the initial composite and worked as by extrusion or otherwise with the remaining materials of the composite so as to extend along the length of the final composite within the superconducting region. Since this region is conductive, it is probably irrelevant whether the strengthening material is itself conductive or non- conductive. However where it is conductive, this will.
- 8 - further assist in stabilising the superconductive material which, as is well known, can be damaged if .local hot-spots occur.
The nickel, silver or tantalum coating of the interior of the support casing reduces the reaction which may otherwise occur between the outer casing and the alkali metals or important rare earth materials which will be present in the superconducting layer. In the alternative, gold may be used in place of nickel, silver or tantalum.
It will be appreciated that gold may in fact be substituted for the copper or silver previously referred, where circumstances and applications.-are appropriate.
The invention thus provides a support sheath for very brittle and readily damaged superconducting materials which not only protects the latter against degradation and attack from elements such as water, but also provides mechanical structural support. Preferably the outer support sheath, if not also any inner support sheath, must be capable of conducting electricity to stabilise the" superconductor and where reinforcing elements are employed throughout the cross-section of the superconducting material, these are also, advantageously, electrically conductive.
The invention also enables an insulating outer surface to be provided which will enable separation between windings in a coil and the like.
The invention also lies in a method of fabricating a superconducting composite involving the steps of forming as an annular sandwich between two conductive tubular
- 9 - members, a layer of a material which on subsequent processing can be rendered superconductive; extruding, swaging, drawing or otherwise forming the composite into a conductor of desired overall cross-section whilst ensuring that the internal tube does not collapse so that a hollow passage extends throughout the entire length of the finally fabricated conductor; forming the latter into a coil or other electrical component by shaping, forming or otherwise; passing a reactive gas or vapour or liquid through the hollow interior of the formed composite whilst maintaining the temperature of the formed composite at a temperature at which appropriate reaction will occur to form the sandwiched material into a superconductive material.
In a development of the method of the invention, the internal tubular member may be filled with a removable substance such as a metal having a melting point lower than that of the other materials which the composite is formed, a metal alloy, castor oil, wax or even water so as to prevent the collapse of the inner tubular member during the working of the initial composite to achieve the cross- section size reduction and this substitute material is advantageously removed before the final reaction phase at which the gas or vapour or other material which is to be reacted with the sandwiched material is passed through the central core, whilst maintaining the composite at the appropriate temperature so as to achieve the desired reaction and render the elongated annulus of material superconductive.
Whatever the material that is chosen for the inner tubular support, it is of course essential that the material is
either removable before the reaction stage or is permeable to the gas vapour or liquid which is to react with the sandwiched material.
According to another aspect of the invention, a layer of nickel or tantalum is applied to the inner surface of the outer conductive sheath before the space between it and the inner core is filled with a material which after reaction is to be the superconducting medium.
According to another development of the method, a layer of gold or silver may advantageously be applied to the inner surface of the nickel or tantalum layer before the material which is to be reacted to from the superconductive material is added.
It is of course not necessary for the inner support tube to be of metal such as silver or copper since other materials may be used such as a plastic materials., the only requirement being that the material concerned is permeable to the reactive agent which is to be added to the composite at the time when the material sandwiched between the inner and outer supports is to be rendered superconducti e.
The invention will now be described by way of example with reference to the accompanying drawings in which;
Figure 1 is a cross-section of a composite conductor constructed in accordance with the present invention,
Figure 2 is a cross-section through a silver yttrium barium copper oxide reference wire having an outer diameter of 1.5mm,
Figure 3 is a cross-section through a multicore hollow core conductor,
Figure 4 is a cross-section through a single hollow core conductor,
Figure 5 illustrates the critical current transition for the reference wire at (a) and for silicone at (b) , the reference wire with a silica addition
Figure 6.is an axial section through a circular section hollow conductor constructed in accordance with the invention/
Figure 7 is a cross-section on the line AA through the . hollow single core conductor of Figure 6,
Figure 8 is a cross-section through a twin conductive tape having a twin core and stainless steel outer,
Figure 9 illustrates graphically the resistive transition at 77K in zero magnetic field for external diffusion (ED) conductors sintered for 24 hours at 940°C under oxygen, having an outer diameter of yttrium barium copper oxide of 0.7mm,
Figure 10 illustrates graphically the critical current value as a function of heat treatment time for ED conductor under oxygen at 940βC (starting material: powder after calcination (SSR)) with the outer diameter of the yttrium barium copper oxide of 0.7mm,
Figure 11 illustrates graphically the reduced critical
current i as a function of preform particle size. The applied field Ha perpendicular to the current direction. •
Figure 12 illustrates graphically resistive transitions at (a) internal diffusion twin filament composite and at (b) external diffusion (ED) single filaments.
Figure 13 illustrates graphically the reduced critical current i as .function of load P for an ED wire under
P load, and ED wire as before but after removal of the load and for an ID wire under load.
In the drawing the composite is shown as comprising an outer stainless steel sheath 10 and an inner silver sheath 12 between which is sandwiched an annulus of high " temperature superconducting material 14. Between" the latter and the outer sheath 10 is a thin film of nickel 16 and within the annulus are located elongated fibres 18 which reinforce the structure. Although shown in the drawing, it is to be understood that these reinforcing fibres are optional.
In accordance with the method of the invention, the hollow inte ior of the tube 12 is initially filled with a removable material, such as a low melting point alloy and this serves to "ensure that the inner tube 12 does not collapse during the reducing steps.
Also in accordance with the method of the invention, the core of material within the tube 12 is removed before the final stage of processing which involves passing a gas vapour or liquid through the tube which can permeate through the wall of the tube into the material 14 and react therewith to render the latter superconducting.
Typically oxygen is passed through the hollow tube 12 after the core material has been removed and the composit is maintained at an appropriate temperature at which oxidation of the material 14 will occur as the oxygen 5 permeates through the wall of the tube 12.
The method of the invention also involves the formation o the composite into a component such as a coil or the like before the core material is removed from the tube 12.
The invention also lies. in superconducting composite 10 conductors when formed by the method of the invention.
Four major considerations determine the design of a suitable wire:
1) The most critical problem is to ensure the supply of sufficient oxygen to the final product to obtain an
,15 optimum critical temperature. This can be done by exposing the superconductor directly to an oxygen source or by diffusing oxygen through a permeable barrier. Results using an "external diffusion" fabrication route have been reported by S Jin et al. The new "internal
20 diffusion" route of the invention is preferred.
2) Mechanical support is of importance for a brittle material. In commerical composites based on brittle unde compression, this can be done by incorporating, within the composite a metal with a high thermal expansion.
25 3) In order to transfer current to the superconductor an achieve adequate electrical stabilisation it is essential to clad the superconductor with a metallic coating that •
does riot develop an insulating oxide barrier at the superconductor interface.
4) Finally the composition, state of order, microstructure and defect structure of the superconductor must be optimised to support a high critical current density.
All the composites tested by the inventors use,d as their starting material finely powdered YBa2C -07 with a critical temperature of 91.5K and a transition breadth of IK. Details of preparation have been reported by R A Camps et al. An initial reference composite was fabricated using the standard method of swaging a. simple ' preform consisting .of a silver Ag tube 11 containing superconducting powder, (see Figure 2). The composite wire was sintered in oxygen at 950βC for 24 hours so that oxygen could permeate through the tube wall 11. It was then annealed at 600βC and 350βC for 8 and 12 hours respectively. The wire exhibited a critical current density of 525 A cm -2 at 77K xn zero field. X-ray diffraction analysis of the compound 13 inside the finished wire gave the same result as in the bulk compound.
It is clear that silver is an excellent cladding material both for oxygen transport by diffusion and for making good electrical contact to the superconductor. However the wires were mechanically extremely weak and an attempt was made to reinforce the composite by casing with an outer layer of stainless steel. This composite was not superconducting confirming that diffusion of oxygen through the silver during heat treatment is essential.
In order to facilitate reinforcement without cutting off the oxygen supply a composite was built as shown in Figure 3 or 4 which incorporate a hollow core 15 to allow oxygen penetration by "internal diffusion" in accordance with the invention.
In Figure 3 six cores of superconducting material 17, 19 etc are shown surrounded by silver 21 itself encased in a nitrate sheath 23.
In Figure 4 a single hollow core 15' is surrounded by a silver layer 25 separating 4 from the superconducting material 27 which is itself surrounded by a metal foil 29 separating it from a stainless steel outer 31.
In the example shown in Figure 3, the critical current of the multicore conductor was about 60 A cm -2 at 77K. At present this is limited by difficulty in making contact to the wire and by contamination of the internal surface of the rather irregular hollow core.
The internal diffusion method of invention enable's a good oxygen supply to the superconductor metal especially in t e single hollow core arrangement of Figure 4.
Some problems were encountered in making current and voltage contacts to sample wires. Current transfer effects can lead to negative-going voltages when the critical current is large and the superconductive cross- section is non-uniform of cracked. By reversing the direction of current flow we have established that these effects are thermoelectric in origin. Current and voltage contacts were made using In, Ag or Au pressed directly onto the surface of the wire. Conventional soldering
degraded the join between the outer cladding and the superconducting core, considerably reducing the value of •• the measured critical current.
Two different experimental composites were used to investigate the relation of the critical current to microstructure.
First in. a tape formed by rolling the reference wire, the critical current had a minimum value for a perpendicular applied field which points to the importance of powder compaction in the design of commercial conductors.
Secondly the .importance of trace impurities was investigated by adding 0.lwt% Si0~ to the reference wire. This reduced the critical current by a factor of 2 (see Figure 5) . This serious degradation has been correlated with the extensive occurence of silica rich inter-grain precipitates in the work done by R A Camps et al.
Example 1
The Fabrication of a Circular Section Hollow Conductor
As shown in Figure 6 a stainless steel tube (20) internal diameter 6mm, external diameter 8mm, was used as the external support cladding for the composite. A copper end stop (22) was screwed into one end of the tube to retain the inner portions of the composite during assembly and fabrication of the composite. The tube was then lined with nickel foil (24) fitted closely to its inner surface. Then a silver tube filled with low melting points Wood's metal was placed centrally in the stainless steel tube supported on the copper end stop.
The silver tube was prepared by filling a silver tube (26) of wall thickness 0.5mm and internal diameter 5mm with Wood's metal (28) and cold swaging to a final external diameter of 2.5mm prior to the insertion into the stainless steel tube (20) .
The resulting annular space was filled up completely with superconducting YBa.-.Cu-.O.-, powder (30) using a stainless steel rammer.
Finally the upper hole was closed with a second hollow copper end stop (32) screwed into place.
The assembled composite was cold swaged to a wire having 3mm outside diameter*
The copper containing end sections were cut-off and a plastics tube was sealed over one end and connected to a source of compressed air. The composite wire was then immersed in boiling water and the molten Wood's metal extruded under the . action of the compressed air. The inside of the silver tube was then cleaned using flowing acid, and the completed composite wire was finally sintered in a furnace at 920°C with a flow of oxygen passing through the hollow core to optimise the oxidation of the superconductor by diffusion of oxygen through the inner silver layer, after which the sample was slowly cooled in flowing oxygen.
A similar method was used to fabricate a hollow core composite containing six cores of silver clad superconductor as shown in Figure 3.
Example 2
The Fabrication of a twin Conductor Tape
In a further composite a twin core stainless steel clad tape was prepared which after fabrication had the cross- section of Figure 8 and contained a single hollow central core (34) and two silver clad superconducting filaments (36 and 38)'.
A stainless steel tube (40) was prepared by being cold rolled until it had two parallel sides and contained a flattened interior hole. Three silver clad wires of identical diameter surrounded by nickel foil (39) were inserted lengthwise into the tube (40). The"two outer wires contained yytrium barium copper oxide superconducting powder prepared by filling -a silver tube having a 0.5mm wall thickness and a _5mm internal diameter with powder, one end sealed with a copper end stop, and swaging the tube down to 2mm diameter. The third central wire contained Wood's metal. *
The whole composite was further rolled until the wires occupied the whole of the internal space in the tube.
The ends of the wire were then cut-off and the exposed ceramic ends were covered by silicone rubber. A tube was connected to one end of the composite and the other end of the tube was connected to a source of compressed air. The composite was immersed in boiling water and the Wood's alloy was removed by the air pressure.
The internal surface of the tube was cleaned by acid flow after which the now hollow composite was placed in an oven
for sintering. During sintering oxygen was blown through the central hollow core iri the wire.
Commercial exploitation of high critical temperature superconductors will be severely limited unless mechanically robust, high critical current, fully stabilised wires and composite conductors can be fabricated with uniform. properties over lengths of the order a kilometre. The primary problem is the control of the critical current density, I . The production of ■' samples in wire and tape form enables a range of precisel controlled and reproducible experiments to be performed that would be much harder to achieve using sintered pelle samples. Four major considerations determine the design of a suitable wire, 1) the supply of sufficient oxygen to the final product to give a uniform high critical temperature, 2) mechanical support of the brittle ceramic 'phase, 3) the achievement of current transfer through normal leads or contacts and 4) the achievement of an optimum composition, state of order and microstructure fo the support of a high I-.
Initial work has concentrated on a composite formed by swaging or drawing a preform consisting of a silver tube containing superconducting powder. Preliminary work on '. steel clad composites has also been reported, although th critical current Ic was rather low. The invention provides for the fabrication of steel clad hollow conductors with Ic similar to that of silver clad conductors and with much improved mechanical properties. Measurements have been made and are reported of the dependence of I on details of the wire fabrication process.
Sample Preparation
The preform powders investigated were prepared from BaC0_, Y_03 and CuO powders (Johnson Matthey, - 99.999% purity) using two methods. Firstly powders mixed in stoichiometric proportion were prepared by solid state reaction solid state reaction (SSR) powders were investigated both after the initial calcination stage and when fully superconducting. Secondly the starting powders were dissolved in nitric acid and superconducting powders prepared by a citrate synthesis route (CS).
i) External diffusion (ED) composites - Wires for external diffusion were prepared using the standard method of swaging a simple preform consisting of a silver tube containing the preform powder being investigated as described below. Different powders could be placed in the same tube separated by solid silver spacer sections.
ii) Internal diffusion (ID) composites - Steel clad hollow core composites were constructed in a variety of • geometries. Single core, twin core and multicore '" composites have been tested. In each case the hollow core region was built using a metal core which was removed ,after fabrication to the final diameter, by chemical means or by melting. Both external and internal composites were reaction sintered after fabrication at about 940*C under oxygen and slowly cooled.
Superconducting Measurements
i) External diffusion - The critical current for identical wires prepared from different starting material is shown in Figure 9. The upper three traces used a
powder size range 25um to 38um and were compacted without evacuation. The lower trace is for a sample with the "as- ground" particle size spectrum and was evacuated under vacuum prior to compaction. The maximum value of I was obtained for superconducting powder swaged under air.
Problems were experienced with the vacuum compacted wire, after additional heat treatment for 60h under oxygen the . value of Ic only increased from 0.3A to 0.4A. In the case of wire made under air one can see that the value of the critical current increases strongly with the increase of
SSR time (see Figure 10). The effect of the preform particle size fraction on the magnetic field dependence of the critical current of standard wires is shown in Figure
11. The value of Ic in a mag Jnetic field increases with decreasing particle size.
ii) Internal diffusion - The values of Ic obtained in wires based on internal diffusion designs are close to the values obtained for external diffusion designs. Figure 12 shows I measurements for a twin filament tape form composite, compared with a single filament• treated by ED under the same conditions. The individual filaments are made from the same powder and have the same cross- section.
111) Stress - Silver clad wires are very weak and show irreversible degradation under load, in comparison the composites clad in steel are able to withstand high loads without degradation of I (Figure 13). The improvement in properties arises because the steel does not soften at the sintering temperature. The superconducting ceramic is therefore supported by a strong matrix, in addition, during cooling the differential thermal expansion places
the superconductor under compression.
Discussion and Conclusions
The maximum value of Ic obser_ve_d in YBa- 2.Cu- 3O-,-x in both I and ED conductors is 900Acm in zero field at 77K. Although I is still very low compared to commercial superconductors the results demonstrate clearly the feasibility of fabricating internal diffusion conductors with steel cladding. The ID designs are mechanically robust because of the rigidity of the cladding and becaus the brittle ceramic is under compression. Such ID conductors withstand high loads without degradation of I .