US3809147A - Method for making products suitable for use in forming composite superconductors - Google Patents

Abstract

A plurality of rods are assembled in a predetermined configuration to form a core which is surrounded by a molten matrix metal within a heated crucible. The temperature of the thusly charged crucible''s upper portion is maintained above the matrix metal''s melting point. In this respect, as the heat is applied to the top of the melt, the crucible is maintained in a hot environment while the bottom of the crucible is centrally chilled. In this manner the charge is solidified from the bottom toward the top so that the solidification progresses upwardly and outwardly in a conical pattern. After the controlled solidification is completed the casting is separated from the crucible to form a cored extrusion billet. In one embodiment the rods are separated from the casting and the resulting open holes are filled with superconductive material to form a composite superconductor extrusion billet. In another embodiment the rods themselves are made of a superconductive material so as to eliminate the step of separating the rods from the casting.

Description

United States Patent [191 Raymond et al.

11] 3,809,147 [451 May7, 1974 METHOD FOR MAKING PRODUCTS SUITABLE FOR USE IN FORMING COMPOSITE SUPERCONDUCTORS [76] Inventors: Jan Wayne Raymond, 250 S.

t Oneida, Apt. 4018; Clay N.

Whetstone, 840 S. Oneida, Apt. 403A, both of Denver, Colo. 80222 22 Filed: Sept.24, 1971 21 App1.No.:183,594

Related US. Application Data [62] Division of Ser. No. 47,390, June 18, 1970.

[52] US. Cl 164/'l22, 29/599, 164/4, 164/132, l74/DIG. 6 [51] Int. Cl B22 d 27/04 [58] Field of Search 164/79, 98, 100, 103, 105, 164/108, 110,112, 4,122, 60,132; 174/126 CP, DIG. 6; 335/216; 29/194, 599, 527.5

[56] References Cited UNITED STATES PATENTS FOREIGN PATENTS OR APPLICATIONS 860,126 2/1961 Great Britain OTHER PUBLICATIONS A. Hare et al., A Method of Casting Radiator-Type Fuel Elements for a Nuclear Reactor, Transactions of the American Foundrymens Society, Vol. 66 (1958) pp. 210-212. 1 V

Primary Examiner-Charles W. Lanham Assistant Examiner-D. Reiley, III

Attorney, Agent, or Firm-Griffin, Branigan and Kindness 1 [57] ABSTRACT A plurality of rods are assembled in a predetermined configuration to form a core which is surrounded by a molten matrix metal within a heated crucible. The temperature of the thusly charged crucible s upper portion is maintained above the matrix metals melting point. In this respect, as the heat is applied to the top of the melt, the crucible is maintained in a hot environment while the bottom ofthe crucible is centrally chilled. In this manner the charge is solidified from the bottom toward the top so that-the solidification progresses upwardly and outwardly in a conical pattern. After, the controlled solidification is completed the casting is separated from the crucible to form a cored extrusion billet. In one embodiment the rods are separated from the casting and the resulting open holes are filled with superconductive material to form a composite superconductor extrusion billet. In another embodiment therods themselves are made of a superconductive material so as to eliminate the step of separating the rods from the casting. a

6 Claims, 4 Drawing Figures m nnow 71914 3809.147

SHEU 2 OF 2 HEAT FURNACE 46 4 HEAT TEMP CONTROL d@ CHILL WATER FIG. 5

This is a division, of application Ser. No. 47,390,filed June 18, 1970.

BACKGROUND OF THE INVENTION The present invention relates to a method and apparatus for producing pattern-cored extrusion billets for composite superconductors; and to the production of particularly high quality superconductorwire.

Composite superconducting wire is comprised of strands of superconductive material imbeded in a matrix of a metal such as copper and is frequently extruded from a billet. One form of superconductive wire includes upward to 1000 filaments of superconductor alloy distributed in an array in a normal metal matrix such as copper. These filaments should be continuous fromone end of the wire or strip to the other and separated from one another over their entire length by the matrix material. Such composite structures are currently fabricated by techniques involving extrusion of composite billets that are subsequently swaged, drawn, or rolled into superconductive wire, rod or strips which form the final product. Each billet used in the extrusion V process is conventionally in the form of a right circular cylindrical matrix of normal metal such as copper in which rods of a superconducting material are arranged with their long axes parallel to the longitudinal axis of the matrix cylinder. In this respect, it is customary to form such a billet by drilling holes in a copper slug and then loading the holes with a superconductive material.

Alternatively, a plurality of wafers of the matrix metalhave holes drilled therein in a predetermined pattern. The wafers are then stacked up so that their holes are aligned for receipt of superconductive rods. This wafer-rod structure is then encased to form an extrusion billet. Still another method of forming such billets is to form a pattern-arranged bundle of rods comprised uf the matrix material and superconductive materials. This pattern arranged bundle of rods is then encased to form either an extrusion billet or a smaller bundle that can be swaged or drawn into a final product without employing the extrusion step. A method of this type is described in (1.8. Pat. Nos. 3,465,429 and 3,465,430 to Barber et al.

All of the above methods of producing composite superconductor materials have certain drawbacks. For example, there is considerable waste and expense involved in drilling holes inthe matrix metal; and it is also expensive to properly encase the matrix and rods. Also, particularly in connection with the wafer-rod structure, there is a tendency for the superconductive rods to break whilethey are being reduced to final form. Similarly, where it is necessary to encase the billet elements, the casing must be removed 'after extrusion or it frequently causes undesirable burrs or jaggedprotrusions on the finished wire. Hence, objects of this invention are to provide not only a more economical and reliable method and apparatus for producing superconductive wire, but to provide a superconductive wire which itself is of a considerably higher quality than that obtained by prior-art methods.

An intermediate object of the invention is to provide a superconductor extrusion billet that is substantially 2 defect free. In this regard it is a principle of the invention to form such a billet by means of a unique casting technique. Barber et al. have suggested that extrusion billets can be cast, but such techniqueshave nothere- I tofore proven practical. One reason for this is the generally accepted belief that extremely expensive and sophisticated casting equipment would be necessary to make defect free castings of the required geometry and size because of the large shrinkage tendencies and other defect producing mechanisms encountered during solidification of normal matrix metals.

In accordince with principles of the invention a cored billet matrix is formed in a manner'similar to that some times used in connection with fuel elements for nuclear reactors. One such technique is described in an article by A W. Hare and R. F. Dickerson appearing at p. 210 et seq., Vol 66 of the Transactions of American Foundrymens Society 1958). In this regard a plurality of rods are assembled in a predetermined configuration to form a core which is surrounded by a controlled purity and composition molten matrix metal within a heated crucible. The top of the crucible andits charge are maintained above the matrix metals melting point; and, at the same time, a chill block is brought into contact with the bottom center of the crucible while the sides of the crucible are maintained in a hot 'environment. In this manner, the charge solidifies from the bottom up and the center out so that it solidifies in the pattern of an upwardly progressing cone. In this manner shrinkage, porosity and other defects are eliminated so that the resulting matrix metal is nonporous and uniform throughout. Moreover, the cast billet is formed without sophisticated andcomplex zone refining equipment, vibrational casting apparatus, centrifugal casting devices or the like. Consequently, the method of the invention is relatively inexpensive.

A major advantage of the invention is the provision of a cored superconductor billet having a unitary matrix metal structure. In this regard, where a plurality of matrix metal rods are assembledin a container there is considerable difficulty in bonding the similar-metal rods together. Some writers hold that it is not necessary to obtain complete bonding between the matrix metal elements, but we have found thatcomplete bonding is quite important; and one of the reasons our method results in a superior product is that the unitary matrix metal portion of our extrusion billet has no unbounded portions as will nowbe briefly discussed,

When pattern arranged bundles of matrix metal rods and superconductor rods are placed in a container and co-reduced as in the method of Barber et al., the ductile but abrasive superconductor rods are almost instantaneously bonded to their adjacent rods of normal matrix. metal. Contiguous matrix metal-rods, on the other hand, shift, slide, and readjust under the stresses of'coreduction. They then transmit uneven stresses to the ductile superconductor filaments so that the resulting filaments in the compositebillet do not have uniform cross-sections throughout theirlengths. Moreover, if the rods of normal metal are not substantially absolutely clean they are even less adequately bonded and the resulting filamentsare even less uniform. This, in turn, places a restriction uponthe critical current density of the superconductor wire that is'drawn or otherwise formed from the composite billet. The products resulting from the method of the instant invention, on the other hand, have uniform superconductor filaments whereupon they exhibit substantially higher useful critical current densities than corresponding composite superconductors made by the above described techniques of the prior art.

One prior art technique for obtaining better bonding between the matrix metal elements of a patterned-rod billet has been to co-reduce the billet at a relatively high temperature. Such high temperatures, however, cause the normal metal surfaces to react with surrounding active gases to severly reduce the bonding characteristics of the normal metal elements and thus degrade the quality of the product. Hence, when this technique has been employed it has been necessary to initially coreduce such composite structures in a vacuum or an inert-gas atmosphere. This, however, is a cumbersome and expensive procedure which is not required when the method of the invention is employed.

It is often desirable to use long length of composite superconductor wire rather than two or more shorter pieces; and, longer pieces of such composite wire are more susceptible to undesirably low critical current densities because there is a greater probability that the longer wire will have a filament defect somewhere along its length. Hence, it is another object of this invention to provide a method of making long lengths of composite superconductor wire having high critical current density ratings. In this regard, it has become conventional practice to twist composite superconductor material in order to increase its useful critical current density in magnet applications. When composite superconducting materials are thusly twisted, however, normal metal bonding defects are magnified and thus degrade the full benefitof the twisting step. One of the advantages of the instant invention, therefore, lies in the ability of its resulting wire to be twisted so as not to destroy the filament integrity and thus obtain the full benefits of the critical current density increases due to the twist.

Another object of the invention is to provide a method of easily, accurately, and economically con trolling the resistivity ratio of composite superconductive wire; and in accordance with the principles of the invention dealing with this particular object, an alloying or contaminant material is added to the molten matrix metal in an amount corresponding to a predetermined resistivity ratio of the corresponding superconductor.

And finally, in accordance with further aspects of the invention, after the casting has been solidified as described above the cored casting is separated from the crucible so that the rods can be removed if desired. In this event the resulting holes can be filled with a superconductive material to form an extrusion billet which is drawn or otherwise suitably reduced to form superconductive wire, rod, or strip. In this manner the resulting product not only has operating superconducting characteristics that are far superior to those produced by prior art methods, but the wire is free of burrs or jagged protrusions resulting from outer containers such as those described in U.S. Pat. No. 3,465,430. Hence the smooth surface obtained by this invention is easier to both fabricate and insulate.

BRIEF DESCRIPTION OF THE DRAWINGS The foregoing and other objects, features, and advantages of this invention will be apparent by the following more particular description of preferred embodiments thereof as illustrated in the accompanying drawings wherein the same reference numerals refer to the same parts throughout the various views. The drawings are not necessarily intended to be to scale, but rather are presented so as to illustrate the principles of the invention in clear form.

In the drawings:

FIG. 1 is an end view of a billet core used in an embodiment of the inventions method;

FIG. 2 is a sectional view of the FIG. 1 billet core taken along the lines 2-2 of FIG. 1;

FIG. 3 is a schematic view of a portion of a furnace and crucible adapted to practice a preferred embodiment of the inventions method;

FIG. 4 is a schematic illustration of the manner in which a billet matrix is solidified during a directionally controlled freezing step of the methods preferred embodiment; and, 7

FIG. 5 is a schematic illustration of apparatus for controlling the solidification of the billet matrix illustrated in FIG. 4.

DETAILED DESCRIPTION In the embodiment of the core array illustrated in FIGS. 1 and 2, a retainer plate 10 is affixed to both a core support rod 12 and a lower pattern plate 14. These elements are made of a high purity graphite.

The lower pattern plate 14 and a corresponding upper pattern plate 16 have holes 18 drilled therein to accommodate the ends of a plurality of rods 20 preferably hollow quartz which have at least one end sealed as at 22 to prevent the escape of air and the entrance of copper during an immersion step to be described shortly.

The core array of FIG. 1 is assembled by screwing the graphite core support rod 12 into the lower rod retainer plate 10, sliding the lower pattern plate 14 down the core support rod, and sliding the upper pattern plate 16 into its proper position on the core support rod. Both the upper and lower pattern plates are then locked in position with small tungsten or graphite pins such as 26 and 28. Next, a suitable number of quartz tubes 20 are loaded in the partially assembled core by inserting them, closed end 22up (to the right in FIG. 2) through the holes 18 in the top pattern plate 16 so that they terminate in corresponding holes in the bottom pattern plate 14. Subsequent to completely installing the appropriate rod pattern, the top retainer plate 24 is slid down the core support rod 12 and securely pinned as described above.

After the core array of FIG. 1 is constructed as de scribed above, it is placed in a furnace to be heated. At the same time, a crucible 30 (FIG. 3) is placed on a somewhat donutshaped stool 32 in a furnace'34. In this regard,.the inside walls of thecrucible are slightly tapered at a rate of about one quarter inch per foot so that the crucibles inside diameter is smaller at its bottom end than its top which is covered by an insulated plug 35. This graphite plug 35 is machined to both fit the top of the crucible and accommodate a tube 36 for delivering inert gas to the crucibles interior if desired.

The furnace has one or more primary heat inlets such as 38: one or more secondary heat inlets such as 40; and a hole 42 in the bottom thereof to accommodate a chill-block 44 which may be raised upwardly through the stool 32 to rest against the bottom of the crucible 30. A temperaturesensing element 46 enters the top of the crucible and passes along its side to sense the temperature of the crucible at various points along its length. A second temperature sensing element 48 extends up to the crucibles bottom adjacent to the chill block 44; and both of the temperature sensing elements are connected to a temperature indicator-controller 5 (see also FIG. 5).

The temperature indicator controller 50 provides outputs on lines 52 and 54 to control primary and secondary heaters 56 and 58 respectively which provide heat to the primary and secondary heat inputs 38 and 40 to the furnace as shown shown in FIG. 5. Similarly, the temperature control 50 provides an output on line 58 for controlling a chill water supply 60 which delivers chill water through conduit 62 to a chill water recess 64 in the furnaces chill-block 44. The heaters and chill water supply can also be controlled manually.

In practicing a preferred embodiment of theinventions method a charge of high purity oxygen-free, highconductivity (OFHC) copper is placed in the crucible 30 as illustrated by dotted line 65 in FIG, 3. In this regard, the crucible 30 is preferably of a high purity graphite in order to minimize melt contamination from this source. This is particularly important where the inventionss method is used to produce superconductive wire having a high resistivity ratio. That is, the ratio of its resistance at room temperature to its resistance at the superconductive temperature such as 4.2 K, for example. Typically desired resistivity ratios are about 150-200, but this ratio drops rapidly as impurities are introduced into the copper. To this end, the use of graphite is significant because it does not combine with copper, but the crucible can alsobe made out of other materials which would not contaminate the copper. Similarly, matrix material other than copper can also be used; and, in those cases where it is desired to provide a superconductor having a low resistivity ratio the copper or other matrix can be intentionally alloyed. For example, as little as 7 percent nickle drops the resulting structures resistivity ratio to about 5:1. In this regard, it has been found that the method of the invention is admirably suited forboth accurately and inexpensively controlling the resistivity ratio of composite superconductive wire. Forany given combination of normal metal and superconductor metal the composite wire s resistivity ratio can be controlled to within an accuracy that has not been previously obtainable in commercially available composite wire.

After the charge has been melted and superheated, it fills the crucible to about the level of line 66 in FIG. 3. At this point, the preheated core assembly is slowly lowered into the melt so as to be covered by molten copper. The primary heater is then turned off and the melt is subjected to a controlled unidirectional freezing step which will now be described.

The temperature controller 50 is adapted to direct ronment. In this manner, the matrix gradually solidifies from the bottom up and insideout in a cone-like fashion so asto eliminate the casting defects generally associated with uncontrolled solidification. For example, with the controlled solidification step a shrinkagecavity does not formin the center of the billet as occurs if the melt solidifies from the outside in.

In the above regard, it has-been found that by simultaneously controlling the chilling of the crucibles lower portion and heating its upper portion during the solidification process a pyramid or cone type solidification pattem results. That is, as illutrated in FIG. 4, the melt 68 solidifies first at its bottom center and then at its outer edges in the manner of an upwardly progressing pyramid or cone placed on top of the previously solidified matrix below. For example, dotted-line 70 might represent the extent of solidification at a first point in time; dotted line 72 might represent the extent of solidification at a subsequent point in time; and dotted-line 74 might represent the extent of solidification at a still later point in time. It is this solidification cone pattern of progressive solidification that provides a casting that is substantially free of undesired voids or conventional casting defects.

After the casting is sufiiciently cooled, it is extracted from the crucible with the bore array cast inside. The portions of the casting containing the pattern and re tainerplates are then cut ofi and, if desired, the outer surface of the casting is turned to the desired dimensions. In this regard, however, one of the advantages of the invention is that only a small portion of the originally cast matrix material is wasted. For example, the above described steps of turning and removing the end and retainer plates involves removing only about 5 percent of the matrix material which, because of its high purity can be reused. Note, in this regard, that if holes are drilled in either a unitary structure or individual wafers it is quite difficult to maintain the purity of the thusly removed material, whereupon it is not satisfactory for subsequent use as a superconductive matrix.

If desired the quartz rods are next removed from the casting by a leaching process in which the quartz is dissolved under the action of molten sodium hydroxide. The hot billet is then water-quenched to ensure re: moval of any undisolved quartz and to minimize oxide tion of the billet surface. Alternatively, the, quartz rods can be removed by leaching in hydrofluoric acid solu-, tion or by mechanical means. After this the matrix casting is cleaned by brushing, leaching, and washing in suitable reagents to provide clean active surfaces. The matrix holes are then filled with rods of superconductive elements or alloys such as niobium, or some appro priate superconducting alloy. For example, satisfactory results can be obtained by using a niobiumtitanium alloy having up to 70 percent titanium (titanium 30 weight percent niobium); and in a preferred embodiment the composite billet consisted of titanium 45 weight percent niobium rods embedded in an OFI-IC copper matrix. Whichever the case the composite billet is then extruded, swaged, drawn or in some other way fabricated into composite superconductive wire, rod or strip.

It will be appreciated by those skilled in the art that the above described method and apparatus for producing a uniform matrix also provides a uniform, superior quality superconductive wire. In this: regard, not only is such wire more uniform, but there are far less incidents of filament damage than in wires made by prior art processes. Consequently, the resulting wire can be drawn into much longer lengths without suffering a reduction in useful critical current density. For example, a comparison was made between composite wire made by aconventional method and composite wire of the same diameter and composition, but made by the above described method. The maximum length of high quality composite 0.050 inch diameter wire made by the conventional method was 4,000 feet, while high quality 0.050 inch wire of the instant invention was drawn to 40,000 feet and it could have been longer if desired.

Also, the surface of the resulting wire is free of burrs as opposed to that of prior art processes because the composite extrusion billet is not segmented and thus it is not necessary that the billet be placed in a container prior to drawing. Hence, there are no undesirable protrusions or burrs in the final wire so that it is considerably easier to insulate. It should also be noted that when superconductive wires made in accordance with the invention are placed in structures such as superconductive magnets, they result in a magnet that is much easier to energize because it is both easier to insulate in a short-free manner; and capable of obtaining a higher flux density because of its filament integrity.

While the invention has been particularly shown and described with reference to preferred embodiments thereof, it will be understood .by those skilled in the art that various changes inform and details may be made therein without departing from the spirit and scope of the invention. For example, it will be appreciated that the above described central graphite supporting rod could be a supporting circumferential sleeve or longitu dinal straps, the quartz tubes could be drilled out; and the inventions method can also be applied to continuous casting. Also, other materials can be used than those specifically described. For example, rods of stainless steel or other suitable metals can be coated with Al Ti 0 or other refractory compounds, and such structures can be used in place of the quartz rods described above; and, if the rods are tapered, they can be removed mechanically.

The configuration of the crucible mold, core components and billet can. also be changed markedly without departing from the spirit of theinvention. For example, square or hexagonal cross sectioned billets can be produced; and optimum packing factors may make it desirable to use core rods having hexagonal, triangular or other geometrical shapes.

The quartz rods can also be replaced with superconductor rods whose surfaces have been treated with niobium, molybdenum, tungsten, or the like so as not to combine with the matrix metal-and/or form compounds detrimental to the fabrication and useful current density of the superconducting end product.

In this regard the invention can be practiced by placing core rods composed of a superconducting material directly in the core assembly to produce a finished extrusion billet as the cast product rather than a cored extrusion matrix that must subsequently be loaded with a superconductive material to form the finished composite extrusion billet. For example Ti-Nb core rods can be directly cast in an aluminum matrix. Also, instead of inserting the core assembly into the molten matrix of open holes therein and suitable for use in forming a composite superconductor, said method comprising the steps of:

creating a melt of matrix metal in a form;

locating an array of preheated hole-shaping re'movable rods in said melt of matrix metal in a form such that the melt envelopes the array of rods;

sensing the temperature of said form at predetermined points running from the bottom to the top thereof;

controlling the solidification of said melt of matrix metal from the bottom of said billet toward the top thereof in accordance with saidsensed temperature in a manner such that solidification progresses upwardly in the pattern of a cone so as to eliminate undesired voids and casting defects, and obtain a nonporous uniform billet;

separating the solidified matrix and said array from said form to provide a cast billet suitable for manufacturing into a composite superconductor product; and,

separating said array of rods from said matrix metal to thereby leave open holes in said cast billet.

2. The method of claim 1 wherein said rods are formed of quartz and wherein said quartz rods are separated from said matrix metal by locating said billet in a bath comprised of constituents selected from the group consistingof molten sodium hydroxide and hydrofluoric acid solution.

3. The method of claim l wherein said controlled solidification step includes the step of:

cooling the melt at the bottom while:

i. maintaining said form in a hot environment above the melting point of said matrix metal; and,

ii. heating said form at the top of said melt to maintain the temperature thereof above the melting point of said matrix metal until the top of said melt solidifies.

4. The method of claim 2 wherein said quartz rods are hollow and have at least one end thereof sealed to prevent the escape of air and entrance of molten 'metal when said rods are located in the melt.

5. The method of claim 3 wherein said rods are comprised of hollow quartz having at least one end thereof sealed to prevent the escape of air and entrance of molten metal when said-rods are located in the melt.

6. The method of claim 5 including the step of removing said quartz rods from said casting by locating the casting in a-bath comprised of constituents selected from the group consisting of molten sodium hydroxide and a hydrofluoric acid solution.

l III 3 3 I UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Potent No. 3,809,147 Dated May 7. 197

Invennor(s)g g 1 W. Raymond and Clay N. Whetstone It is certified that error appears in the above-identified patent and that said Letter s Patent are hereby correctedas shown below:

r Add to the heading of the Patent as section (73) Assignees Cryomagnetics Corporation, Denver, Colorado Signed and sealedthis 8th day of October 1974."

(SEAL) Attest:

MCCOY M. GIBSON JR. c. MARSHALL DANN Attesting Officer Commissioner of Patents

Claims (6)

1. A method of making a cast billet having a plurality of open holes therein and suitable for use in forming a composite superconductor, said method comprising the steps of: creating a melt of matrix metal in a form; locating an array of preheated hole-shaping removable rods in said melt of matrix metal in a form such that the melt envelopes the array of rods; sensing the temperature of said form at predetermined points running from the bottom to the top thereof; controlling the solidification of said melt of matrix metal from the bottom of said billet toward the top thereof in accordance with said sensed temperature in a manner such that solidification progresses upwardly in the pattern of a cone so as to eliminate undesired voids and casting defects, and obtain a nonporous uniform billet; separating the solidified matrix and said array from said form to provide a cast billet suitable for manufacturing into a composite superconductor product; and, separating said array of rods from said matrix metal to thereby leave open holes in said cast billet.

2. The method of claim 1 wherein said rods are formed of quartz and wherein said quartz rods are Separated from said matrix metal by locating said billet in a bath comprised of constituents selected from the group consisting of molten sodium hydroxide and hydrofluoric acid solution.

3. The method of claim 1 wherein said controlled solidification step includes the step of: cooling the melt at the bottom while: i. maintaining said form in a hot environment above the melting point of said matrix metal; and, ii. heating said form at the top of said melt to maintain the temperature thereof above the melting point of said matrix metal until the top of said melt solidifies.

4. The method of claim 2 wherein said quartz rods are hollow and have at least one end thereof sealed to prevent the escape of air and entrance of molten metal when said rods are located in the melt.

5. The method of claim 3 wherein said rods are comprised of hollow quartz having at least one end thereof sealed to prevent the escape of air and entrance of molten metal when said rods are located in the melt.

6. The method of claim 5 including the step of removing said quartz rods from said casting by locating the casting in a bath comprised of constituents selected from the group consisting of molten sodium hydroxide and a hydrofluoric acid solution.

Images