US2838454A - Electrolytic cell - Google Patents

Description

June 10, 1958 M. E. WASHBURN 2,833,454

ELECTROLYTIC CELL Filed Oct. 12, 1954 2 Sheets-Shet 1 mmvron. MALCOLM E WASHBUEN A TTORNEY June 10, 1958 M. E. WASHBURN 2,838,454

ELECTROLYTIC CELL 2 Sheets-Sheet 2 Filed Oct. 12, 1954 1 I VENTOR. MALCOLM E. VVASHBUEN United Sttes Patent ELECTROLYTIC CELL Malcolm E. Washburn, Northboro, Mass, assignor to Norton Company, Worcester, Mass, in corporation of Massachusetts Application October 12, 1954, Serial No. 461,740

1 Claim. (Cl. 204-246) The invention relates to electrolytic cells and particularly to cells for extraction of metals from metal carbides by methods in accordance with the disclosures of my colleagues Ervin and Ueltz, Serial Nos. 356,423 to 356,428, inclusive, filed May 21, 1953, Serial No. 313,171, filed October 4, 1952, and Serial No. 394,753, filed November 27, 1953, in which the metal carbide is the anode in the cell, which employs an electrolyte of Another object of the invention is to provide a cell of general application for the extraction of refractory metals of the subgroups of groups IV, V and VI of the periodic table as follows from their carbides.

Group IV V VI Atomic No 22 23 24 etal Ti V Cr 40 41 42 Zr Nb(Cb) Mo 72 73 74 Hf Ta W Another object of the invention is to provide an efficient electrolytic cell of the character indicated. Another object of the invention is to provide a cell having cell elements which are not too expensive and are easily made.

In the accompanying drawings illustrating several embodiments of the electrolytic cell according to the invention,

Figure l is a vertical axial sectional view of a cell,

Figures 2 to 5 inclusive are horizontal sectional views illustrating the metal cell, graphite container, the anode or anodes and the cathode or cathodes in further embodiments of the invention.

, The following description of the structure illustrated in Figure l and of the materials used to make such structure is exemplary only.

A cylindrically shaped cell 1 made of nickel or other suitable material has a bottom to which an integrally depending drain pipe 2 is attached. The cell 1 has a hollow head 3 to which is bolted by means of studs 4 a hollow head 5, each stud 4 being insulated from the head 5 by means of a sleeve 6. Nuts 7 on the studs 4 hold the parts together. The peripheries of-the heads 3 and 5am drilled and tapped and pipes 8 and 9 are screwed thereinto one of which conducts water into one head and the other of which is the drain from the other head. The heads are drilled and tapped on the periphery ice at another location and pipes 10 and 11 connected by a rubber tube 12 conduct the water from the inside of the one head to the inside of the other head. Thus the heads 3 and 5 are kept reasonably cool. To seal the cell a sealing ring 13 of rubber or the like is provided between grooved flanges 14 of the heads 3 and 5.

Welded to the inside of a central hole through the head 5 is an upwardly extending pipe 15 which is externally threaded at the top to receive "a cap 16 which has a central hole plugged by an insulating bushing 17 preferably made of refractory material such as asbestos cement through which extends a cathode 18. Where the metal being extracted is titanium this cathode 18 is preferably a titanium tube. In other cases the cathode 18 might be a nickel tube but preferably the cathode 18 is made/of the same metal which 'is being extracted. Zirconium with its almost inevitable content of hafnium can be drawn into tubes and so can vanadium, chromium, niobium (columbium) and tantalum although the last two are rare at present. Molybdenum and tungsten cannot at present be had in the form of tubes but solid rods can be had andused in this invention. Titanium can readily be drawn into tubes and I have used a titanium tube cathode 18. The advantage of using a tube is that a thermocouple rod 19 having thermocouple wires 20 can be inserted in the cathode 18 to keep careful check 011 the temperature deep in the cell. The bushing 17 is sealed to the pipe 15 by means of a sealing ring 21 and the bushing 17 is also sealed to the cathode 18 by means ofa short section of rubber tubing 22 so therefore, since the heads 3 and 5 are sealed by the ring 13 and the head 3 is integral with the cell 1, the inside of the cell 1 is hermetically sealed from the atmosphere except for pipes 24 which extend through both walls of the head 5 and are welded thereto. These pipes are connected to rubber hoses, not shown; one hose leads to a bottle of gas and the other hose leads to a needle valve to throttle the flow to keep a low pressure of gas within the cell, for example of the order of one or two inches of mercury which is sufficient to keep out the air and to avoid wasting the gas. Preferably the gas is an inert gas of which the common examples are helium, neon, argon and krypton and from many practical points of view I prefer argon. It is not too expensive and it is heavier than air so it has less tendency to diffuse then helium which is much cheaper but is very much lighter than air. The molecules of helium are of course very small and they diffuse readily, that is to say they escape through minute cracks or pores. Nevertheless helium in some cases might be preferred.

Because of the insulation of the cathode 18 from the pipe 15 it is really not necessary to insulate the head 5 from the head 3 so therefore the sleeves 6 might be dispensed with but they are added as a matter of precaution. The cathode is connected to negative electricity by means of a clamp 26 while the cell 1 is connected to positive electricity by means of a clamp 27. The pipes 8 and 9 can readily be insulated by connecting them to rubber hoses while the lower head 3 rests upon an annular plate of asbestos fibre material 30; thus the cell 1 and conveniently shaped into cylindrical form. n the outside thereof there is preferably a helical groove in which is a winding of wire 38 having suitable electrical resistance and being refractory enough to withstand the temperature involved. Many suitable metals and alloys are known which are satisfactory for this purpose and it is sufiicient to mention Nichrome (U. S. Patent 811,859). The grooving of the cylinder 36 can be dispensed with and is a matter of choice. Outside the winding 38 is a coating of refractory cement 40 which can be applied after the Wire is wound. Alumina cement is satisfactory for this purpose. Between the cement 40 and the inside of the can 32 is a filling 41 of any suitable insulating material which is refractory enough and there are many such materials such as fullers earth and diatomaceous earth. This material 41 can be simply loose material poured into the space.

One end 42 of the Winding 38 extends to a binding post 43 connected to a wire 44 and the other end 45 of the winding 38 extends to a binding post 46 connected to a wire 47. The binding posts 43 and 46 extend through the fibreboard 30 as shown. The wires 44 and 47 are connected to a suitable source of electrical energy, either direct current or alternating current although the latter is usually more readily available, and suitable controls are provided for heating the apparatus to the required temperature. Such electrical apparatus is well known and need not be further described.

Inside of the cell 1 is a cylindrically shaped container 50 closed off at one end, herein designated as a cup, made out of carbide of the metal which is being extracted. The cup can be formed by taking grains or lumps of the carbide and bonding with a carbonaceous binder such as coal tar pitch. This operation is conveniently carried out in an inert atmosphere, such as argon, at 900 C. to 1000 C. Other binders could be used, and it is even possible to produce satisfactory anodes by pressing, sintering or hot-pressing without the presence of added residual binding material, so long as the particles are in good electrical contact to constitute an electrically-conducting structure. In the case of titanium carbide there should be neither oxygen nor nitrogen in the atmosphere. Titanium and its carbide react quickly with nitrogen.

The carbide cup 50 will in most cases be somewhat porous so I provide an outer cup 51 made of graphite which fits snugly between the cup 50 and the cell 1. Graphite is inert to the hot salt.

The heads 3 and can be made of nickel but it is cheaper to have them made of steel with a heavy nickel plate. All inside surfaces which the salt could possibly contact even assuming the cell were turned upside down should be heavily plated with nickel. Nickel is the most practical metal although theoretically there are others available. Welding solves the problem of attaching the head 3 to the cell 1. It is not necessary to nickel plate the water chambers of the heads 3 and 5.

For convenience in moving the cathode 18 up and down, I have provided a clamp 55 held by an insulating bushing 56 on a screw stud 57 which is supported by a thumb nut 58. The lower end of the screw stud 57 is screwed into and is supported by a clamp 62 secured to the pipe 15.

As examples of the use of the apparatus of the invention, titanium metal was made therein in accordance with the following examples.

Example I Titanium carbide of grit size from 12 to 90 was mixed with ten percent of pitch. The mixing was accomplished by heating the pitch in a metal container on a hot plate until molten and then adding the titanium carbide to the molten pitch, heating continuously while mixing with a spatula until the mass was homogeneous. A total of 421 grams of carbide was used. This mixture was used to line the inside of the graphite cup 51, which was 3 inches outside diameter by 2 /2 inches inside diameter and 7 /2 inches deep. The graphite cup 51 was lined by tamping the hot mixture in it around a slightly tapered graphite plug that had a A inch clearance. The A inch lining over the bottom of the cell was tamped in before the plug was inserted. All components had to be kept sufficiently hot to maintain the pitch in a plastic condition, heating them when required during the tamping operation. The plug which extended beyond the cylinder was then removed by simultaneous twisting and pulling motions after allowing slight cooling. The lining was A inch thick. The graphite cup 51 so lined was then buried in carbon black in a refractory container made of silicon carbide and was fired for four hours at 1000 C. The volatiles of the pitch were driven off, leaving a fairly hard bonded structure titanium carbide cup 50.

This titanium-carbide-lined cup was then placed in the cell 1 and was filled with a eutectic mixture of mols of potassium chloride and 60 mols of lithium chloride in accordance with the disclosure in the application of my colleagues Dr. Ervin and Dr. Ueltz, Serial Number 394,753 filed November 27, 1953. The quantity of this eutectic mixture was 466 grams, which melted to form molten salt about 6 /8 inches deep.

The apparatus was then completely assembled, heated to the operating temperature, the water was turned on and the argon caused to flow from one pipe 24 to the other pipe 24, thus driving the air out of the cup and the space above it. The titanium tube 18 was adjusted so that it extended four inches into the molten salt. Then, the apparatus was energized with electricity at 3.4 volts E. M. F., causing electricity to flow from the cup 1 through the cup 51 and anode cup 50, through the salt bath to the cathode rod 18, and so out of the cell. At the same time, the temperature of the molten salt was raised to 900 C. by control of the electric current from the wire 44 to the wire 47.

The exposed surface area of the cathode rod 18 in the salt bath was 30.4 square centimeters. The electric current was amperes, giving a current density at the cathode of 164 amperes per square decimeter. After minutes the electric current was turned off, both between the clamps 27 and 26 and between the wires 44 and 47, and the apparatus andcontents were allowed to cool. The cathode 18 was raised out of the salt bath soon after the apparatus was de-energized. After allowing to cool to the vicinity of room temperature, the apparatus was disassembled. Up to this time, argon had been continuously flowing from one pipe 24 to the other one.

Frequently after one run is completed, only the oathode tube 18 and its supports, such as the cap 16, are removed and after scraping 01f the sponge metal, these parts are replaced, the current from wire 44 to wire 47 is turned on, and when the salt is molten, the cathode 18 is again lowered and the process is run again. But, sometimes the carbide lining (cup) 50 needs renewing, in which case a sealing plug 63 at the bottom of the pipe 2 is removed and the graphite cup 51 is forced out by inserting a tool through the bore of the pipe 2. This cup 51 sticks in the cell 1 because of the salt which penetrates every crevice and freezes therein, but I am able easily to loosen the graphite cup 51 with a steel rod through the pipe 2 and a hammer. Of course, the head 5 has to be taken off before the cup 51 can be removed.

The deposit of metal on the cathode tube 18 was then scraped off and was placed in dilute sulphuric acid. The sulphuric acid was then poured off and the salt was leached from the metal with distilled water. The weight of the sponge metal titanium produced was 7.6 grams, giving an efficiency of 34 percent, based on the electric current value and total time involved,

g Example II Using the same apparatus as in Example I, the cup 50 was prepared in the same manner out of 374 grams of titanium carbide, grit size from 12 to 90, with ten percent of pitch. Four hundred and eighty-six grams of the eutectic mixture of lithium chloride and potassium chloride were placed in the cup and melted. The molten salt was 6 inches deep.

The cathode tube 18, being the same one previously used, was immersed 4 inches deep in the molten salt and the cell was operated for 60 minutes at 1,000 C. at 50 amperes under an electromotive force of 3.3 volts. The current density was 164 amperes per square decimeter.

The same procedure was used to remove the tube and to allow the apparatus to cool sufiiciently so as not to risk contamination of the metal sponge with oxygen or nitrogen. The same procedure as described in connection with Example I was used to wash the salt out of the metal sponge. Eight and six-tenths grams of titanium metal was obtained, giving an efiiciency of thirty-eight percent.

ExampleIII In this example of the use of the apparatus, the anode cup was prepared out of 391 grams of titanium carbide of grit size from 12 to 90 with ten percent of pitch added. The baking was done in an inert atmosphere of argon at 900 C. Four hundred and seventy-four grams of the lithium chloride-potassium chloride eutectic mixture constituted the salt bath, which had a depth of .six inches. The voltage was 4.1 and the current 50 amperes, giving a current density of 164 amperes per square decimeter, the cathode tube 18 being immersed four inches in the salt bath, as in the other examples. The procedure was the same as previously described and 11.8 grams of titanium metal was collected, giving an efficiency of 52 percent.

Example IV current density was 246 amperes per square decimeter.

Other steps involving cooling with argon flowing and washing the metal to remove the salt were as described in the preceding examples. A total of 10.7 grams of titanium metal was collected, giving an efiiciency of 38 percent.

Example V Procedure, similar to Example I, was used for a series of consecutive runs using the same anode, and a sample of the product had the following analysis:

Percentage by weight Ti (including about 2% Zr) 1 99.3 0 0.1 C 0.4 Fe 0.2

1 This particular sample of TiC employed in these runs contained a little ZrC as impurity. By using TiC free of ZrC, titanium of better than 99% purity can be obtained.

By using as an anode a carbide cup, that is, having the metal carbide as the cylindrical part of the cell, surrounding the cathode, a larger weight of the metal carbide, which is the raw material for producing the metal, can be provided. The centrally located cathode rod or tube and its deposit of metal can be conveniently lifted out of the molten salt without opening the cell. Also because the cathode is centrally located, it can hold a larger deposit of metal sponge before bridging over to the anode. Another factor is the larger anode area in comparison to that of the cathode. This gives a lower anode current density which in turn gives improved electrical efficiency. With other electrode arrangements, current efliciencies were usually less than 10 percent; while with this arrangement, current efliciencies were from 30% to 50% and sometimes as high as Another advantage of this apparatus is that the use of a lining in a graphite cup also provides a practical way of making and replacing the anode.

While I have used an anode of cylindrical shape, many other shapes would be effective, the only requirement being that the cathode is partially or completely surrounded by the anode. For example, in some cases it might be more convenient to have a plane-sided anode, which might even be assembled from flat plates. In such a case the anode shape might be that of a hollow prism. The requirement is that at least 80% of the submerged area of cathode shall be surrounded by carbide anode. The cathode can be in several parts and so can the anode.

It isnt necessary to use pure carbide for the cup or lining 50. The term metal carbide is used here to include material that contains more than 50% by'weight of chemically combined metal and carbon, and which is electrically conducting.

In actual practice, the metal carbides have variable ratios of metal to carbon and are frequently contaminated with oxygen and nitrogen, but can be used effectively for raw materials for this process.

I have found if at least 80% of the total immersed unmasked area of the total cathode is surrounded by anode comprising electrically conductive carbide of metal of those listed, an operative apparatus results and the process can be effectively carried on. Figures 2 to 5 inclusive illustrate modifications illustrating this rule. In these figures numbers ending with the digit 5 designate the metal cells, numbers ending with the digit 1 designate the graphite containers, numbers ending with 0 designate the anode or anodes and numbers ending with 9 designate the cathode or cathodes.

Referring to Figure 2 if radiallines are drawn from the cylindrical cathode 79 it will readily be seen that at least 80% of the area thereof is surrounded by anode 70. If the cathode 79 is immersed for a length which is much greater than its diameter the fact that the bottom is not surrounded by anode is well accounted for.

In Figure 3 the structure is square in cross section except for the cathode 89 which is hollow and the anode completely covers the interior of the graphite container 81 so it can be in contact therewith as shown. This tion to the rule in that lines normal to the interior surface of the cathode 89 can reach anode material but to do so they have to pass through another part of the cathode so the inside area is masked and therefore doesnt count.

In Figure 4 the structure is octagonal except for a solid cylindrical cathode 99. An octagon approaches a cylinder and because of symmetry is preferable to a square. In Figure 5 four anodes 100 in the form of plates and four cathodes 109 in the form of rods are illustrated and it will readily be seen that the structure complies with the rule above given.

In Figure 2 the inside of the bottom of the metal cell 75 is covered with graphite which is to say that the graphite is in the form of a cup 71 as in the case of the cup 51 of Figure l. The anode plates 70 stand clear of contact with the graphite cup 71 so therefore the latter is not electrically energized.

In Figure 3-the metal cell contains a graphite cup 1 81 which contains a titanium carbide cup 50, so therefore the bottom of the cell is titanium carbide which is of course in electrical contact with the graphite and with the metal illustrating an arrangement similar to Figure 1 except that the cross section is square.

In Figure 4 I choose to illustrate the condition in which the metal cell 95 contains a graphite cup 91 whereas the carbide anode 90 has no bottom so therefore the bottom of the cell is graphite which is electrically energized. l find that this results in little loss of electric efficiency because the affected area of the bottom of the cathode rod 99 is relatively small. Nevertheless it would be better to have the bottom of the cell of Figure 54 covered with the carbide anode but the foregoing description of Figure 4 is given to illustrate what may be done and there may be some mechanical reasons in some cases for lining the graphite with carbide on the sides but not on the bottom.

In Figure the arrangement illustrated is similar to that of Figure 2, namely with a graphite bottom to the cell formed by lining the metal cell 105 with the graphite cup 101 and the anode plates 100 of course stand clear of the graphite bottom which is therefore not energized. In fact Figure 5 shows an arrangement almost the same as that of Figure 2 the difference being that the anode in Figure 5 takes the form of a plurality of rods 109 whereas in Figure 2 only a single rod 79 is shown. In Figure 5 there is some masking of some of the area of the cathode rods 199 as will readily be seen by drawing radial lines from the rods to the anode plates 100. This masked area does not count in the application of the rule of 80% as previously stated.

The embodiment of Figure l is preferred to all of the others but the embodiment of Figure 4 is quite efiicient. Next in order comes the embodiment of Figure 3. However there are some advantages from a mechanical and operational point of view to the embodiments of Figures 2 and 5 since the plates 70 and 100 can easily be replaced. In Figures 2, 4 and 5 it is assumed that the area of the side wall or walls of the graphite container below the level of the top of the salt bath is quite a lot greater than the area of the bottom of the container as in the case of the embodiment of Figure 1. It is the relatively small area of the bottom which makes it possible to have it coated with graphite electrically charged in the cases of Figures 2, 4 and 5 without being covered with carbide without substantial impairment of the efiiciency of the cell.

The electrolyte in this invention is composed of fused salt which is halide of metal selected from the group consisting of the alkali metals and alkaline earth metals including magnesium. As magnesium has for a long time been considered to be an alkaline earth metal it is not necessary to mention it if it is not excluded. The alkali metals are sodium, potassium, lithium, rubidium and cesium. The alkaline earth metals are magnesium, calcium, strontium and barium. Mixtures are definitely included and at present I prefer a eutectic mixture hereinbefore specified of potassium chloride and lithium chloride.

It will thus be seen that there has been provided by this invention an electrolytic cell in which the various objects hereinabove set forth together with many thoroughly practical advantages are successfully achieved. As many possible embodiments may be made of the above invention and as many changes might be made in the embodiments above set forth, it is to be understood that all matter hereinbefore set forth or shown on the accompanying drawings is to be interpreted as illustrative and not in a limiting sense.

I claim:

An electrolytic cell for the extraction of metal from metal carbide which comprises a metal container, a graphite container within said metal container, said graphite container being adapted to contain fused salt, means for maintaining a controlled inert atmosphere inside said metal container including a head secured to said metal container and an orifice for the introduction of said inert atmosphere, a cathode within said graphite container, a metal carbide anode within said graphite container surrounding at least of the submerged unmasked area of the cathode as determined by projecting the submerged area of the cathode radially in a horizontal direction from its surface to the submerged area of the anode, the area of said anode being considerably greater than the area of said cathode, and electrical leads for the anode and for the cathode whereby the cathode can be made negative and the anode positive.

References Cited in the file of this patent UNITED STATES PATENTS Haskell Sept. 17, 1935 Rosen Aug. 22, 1950 OTHER REFERENCES

Images