US2960451A

US2960451A - Electrolytic production of refractory multivalent metals

Info

Publication number: US2960451A
Application number: US795340A
Authority: US
Inventors: Harvey L Slatin
Original assignee: Individual
Current assignee: Individual
Priority date: 1959-02-25
Filing date: 1959-02-25
Publication date: 1960-11-15
Anticipated expiration: 1977-11-15
Also published as: FR1249133A

Description

ELECTROLYTIC PRODUCTION OF REFRACTORY MULTIVALENT METALS No Drawing.

This invention relates to the production of high purity and ductile refractory and multivalent metals tungsten and molybdenum in the form of large crystals from their oxides and oxidic ores and compounds by fusion electrolysis.

The primary object of this invention is to provide a process for the production of high purity multivalent refractory metals tungsten and molybdenum, which is adaptable to commercial production because (a) crystalline metal of large size is formed by the process and thereby is resistant to attack by moisture or the atmosphere;

(b) the metal is substantially free of bronzes, lower oxides, carbides, borides, phosphides, sulfides, selenides, arsenides, silicides, or the like, and consequently is ductile;

(c) the electrodeposited metal adheres to a depended cathode and can be Withdrawn from the electrolyte bath with little loss;

(d) the metal product is recovered from the bath components by simple water washing without difliculty;

(e) the process operates at high current efliciency and high recovery efiiciency;

(f) the metal, because of its high purity and ductility and crystal size, is easier to consolidate and work than hydrogen reduced metal; and

(g) the electrolyte is cheap and the drag-out compounds need not be recovered to render the process economic.

These and other advantages are realized by the invention by electrolyzing an oxide of tungsten or molybdenum dissolved in a fused bath consisting of at least one calcium halide and containing calcium oxide, electrolyzing said bath between an insoluble anode and a cathode, said electrolyte being substantially free of alkali metal ions, borates, phosphates, pyrophosphates, metaphosphates, silicates, carbonates, and the like, thereby depositing ductile metal of said oxide of high purity in large crystalline form at said cathode.

The term high purity metal as used herein refers to metal whose purity lies in the range of at least 99.8+% to over 99.99%.

In the reported preparations of tungsten and molybdenum by electrolysis, the processes result in the production of a non-adherent finely divided powder that falls to the bottom of the electrolytic cell Where it forms a sludge with the electrolyte. The metal powder so formed is contaminated with lower oxides of tungsten, tungsten bronzes, borides, carbides, silicides, phosphides or the like so that the separation and recovery of pure tungsten is either impossible or costly. This situation has been clearly recognized by Li and Wang (Tungsten. Third edition 1956. Reinhold) and confirmed by the views of Smithells (Tungsten. 1952 edition. Chapman and Hall) that no practical commercial electrolytic process for producing tungsten has been found. The presently used process is that of hydrogen reduction of purified oxide. This process produces a grey metal tates Paten M 2,960,451 Patented Nov. 15, 1960 powder which because of its finely divided condition is not of the highest purity and increases the cost and difliculty of further working of the metal.

Contrary to published data, I do not find that tungstic oxide reacts with NaCl with the liberation of chlorine or HCl. In fact, tungstic anhydride, W0 is violently expelled by fused NaCl so that little, if any, W0 is retained in the solution. With CaCl alone or in combination with CaF and/or alkali halide, the same phenomenon of ejection of the tungstic oxide is observed so that no useful or suitable concentration of tungsten oxide is built up in the bath and as a result, on electrolysis, no practical or recoverable production of tungsten is possible. Similarly, with molybdic anhydride, M00 no solution in NaCl is observed.

I find, however, if a suitable quantity of calcium oxide is dissolved in calcium chloride prior to the addition of the multivalent metal solute, W0 or M00 then a non-turbulent, clear melt is obtained as the tungsten and molybdenum oxides are accepted and readily dispersed and dissolved in the bath. On electrolysis, the oxides are not now rejected and pure metal may be recovered from the electrolyte. In this way, about a 30-40% solution of W0 in a CaCl -CaO bath may be made, or about a 10% M00 solution in CaCl CaO is obtained. Also, instead of W0 or M00 the lower oxides of these metals may be used as feed for the bath and so may such ores as scheelite (CaWO powellite (CaMoO and the like he used.

A typical electrolyte has the following composition:

Weight percent It is possible to add varying amounts of CaF to the bath, and a eutectic mix of CaCl -CaF with up to 50 weight percent CaO serves as an excellent solvent for W0 However, in addition to the difficulty of subsequently removing CaF- from the deposited metal, I find that for some reason the presence of fluoride in the electrolyte causes a pronounced reducing in the particle size of the deposited metal and an increase in the metal hardness. For these reasons, an electrolyte composed of CaCl CaO as solvent and W0 as solute is preferred. Such electrolytes are liquid at about 700 C. and are stable at temperatures up to 1400 C. or higher.

The concentration of calcium oxide has been varied over a wide range from below 5 weight percent to 40 and over with effective results, but concentrations of CaO of the order of 1020% by weight are preferred.

The presence in the bath of alkali compounds, particularly sodium and potassium (lithium being the least offensive of the group), is especially conducive to the production of tungsten bronzes and impure powdery metal. Further, I find that the presence of such compounds as borates, phosphates, carbonates, silicates, and the like in the electrolyte not only prevents the production of high purity tungsten in excess of 99.9+% metal, but also markedly contributes to the reduction of particle size so that the powder material that is thereby produced is more difiicult to work due to poor ductility and for other reasons. Accordingly, in a CaCl CaOWO calcium halide-calcium oxide-tungstic oxide, electrolyte which is substantially bereft of alkali metal compounds such as the halides, borates, phosphates and the like, I am able to produce by electrolysis large bright shiny silvery crystals of tungsten metal of high purity and ductility, instead of the dull gray powder of the hydrogen reduction process or the usual dark slimy powder associated with prior electrolytic processes. Other alkaline earth halides and oxides may be used such as strontium or barium and the like, but for many reasons, the calcium chloride and oxide are preferred.

The metal produced is not only silvery bright in appearance and macroscopic in size, but it is also free of any discolorations such as bronzes. It lends itself readily and easily to water washing and no complex or arduous operations are necessary to separate the metal from the electrolyte. The metal does not form a slime or sludge nor does it react with the electrolyte in any way so that the metal is quickly separated from the salts by decantation or elutn'ation without loss and without metal contamination. Many of the crystals are over of an inch in length and the mass usually exceeds a sixteenth inch in length. The largest crystals are over 99.9'+% pure and the mass is over 99.8+% pure. A screen classification of metal showed, for example, the following retention:

Percent On'a No. 10 screen approx 70 On a 20 mesh screen approx 20 On a 60 mesh screen approx 8 Through a 150 mesh screen less than 1 The concentration of the W solute in the electrolyte may be varied over a considerable range without impairing the high purity of the metal deposited or adversely affecting its physical size and appearance. Baths have been run wherein the electrolyte was saturated with W0 and electrolyzed until stripped of W0 i.e., from about 43 weight percent W0 to zero weight percent W0 In a continuously operating cell wherein the cathode and its ahering metal deposit is gradually and automatically withdrawn the electrolyte, a solute concentration below about 15 weight percent and above 5 weight percent is preferred. On the other hand, where the bath is operated discontinuously as in the case of the use of a cathode basket which is periodically removed with its adherent cathode deposit after stripping the bath, it is preferable to start (and run) at higher W0 concentrations, i.e., above 20 i i weight percent. However, the precise concentration of oxide is not critical and a preferred range lies between 10 and 30 weight percent. The electrolysis takes place at the expense of the W0 alone and no chlorine gas has been detected at the anode.

The temperature of the electrolytic bath may be from about 700 C. to over 1400 C. However, at the lower temperatures, below about 800 C., the metal is more white gray in color and less of the silvery bright species. Also, the crystal size is not the largest obtainable, although the purity is still high. Consequently, I prefer to operate the electrolyte above about 750 C., i.e., between 750 and 100 C., and excellent results are consistently obtained at about 1050 C. to 1200 C.

The anode is preferably insoluble and made of graphite or carbon. Where the electrolysis is to be used for refining impure metal or where it is desired to maintain the composition of the bath constant, or where it is desired to grow large single crystals, the metal to be refined may be used as an anode. During the course of electrolysis, the anodes are consumed taking the shape of a carrot or an inverted cone, and must be progressively lowered into the bath to maintain the proper current density. Due to the elevated temperature of electrolysis, it is advisable to protect the anodes from useless corrosion and atmospheric oxidation as by encasing them in high fired ceramic A1 0 or MgO sleeves, or by cooling the anodes. In the course of my research, I have found that what small amount of A1 0 or MgO may enter into the electrolyte does not have any noticeable deleterious effect on the purity or character of the metal produced. The reaction at the anode is one of the discharge of oxygen and the union of most of the oxygen with the carbon to give CO and C0 The heat of electrolysis has proven sufiicient to maintain the bath in a liquid state and no auxiliary external heat source may be required. The anode current density may be varied widely without atfecting the course of electrolysis adversely, i.e., from a fraction of an ampere per square inch to over 25 amperes per square inch. However, I. prefer to maintain the anode current density below about 10 amperes per square inch, viz., about 5-7 amperes per square inch.

The cathode may be made of such materials as titanium, tungsten, molybdenum or iron, and although it is advisable to use a cathode made of the metal to be deposited, nevertheless I prefer to collect my electrodeposit on an iron cathode. The cathode current density may likewise be varied widely without affecting the workability of the process, and cathode densities of a fraction of an ampere to over 600 amperes per square inch have been used. In speaking of cathode current densities, it is difiicult to estimate what the actual current density is inasmuch as when the electrolysis is started, the crystals immediately begin to grow and the true area of the electrode is rapidly increased. In a cell where the cathode and its adherent deposit is being gradually withdrawn from the bath during the course of electrolysis, thereby maintaining a fairly constant current density and increasing the crystal growth, I prefer a cathode current density between about 10 and 70 amperes per square inch (calculated as initial). At higher cathode current densities, electrode cooling may be required. Excellent results are obtained at 55 to 65 amperes per square inch for instance. However, in a cell where the bath is periodically stripped of its solute, I prefer initial current densities below about 75 amperes per square inch and excellent results have been obtained at 23 amperes, for example.

In either case, whether the cell is run continuously or intermittently, the deposited metal is found to cling to the cathode and is readily separated from its film of electrolyte by washing with water. Although it is not mandatory, the first wash water should have a small amount of hydrochloric acid added in order to facilitate the solution of any residual CaO that was dragged out of the bath. The only feed material is the oxide of tungsten, and occasionally the bath level should be restored by some addition of CaCl or CaO to make up for the dragout. In a properly run cell practically no tungstic oxide will be found in the metal and the metal and its salts will be devoid of bronzes, insoluble compounds, lower oxides, nitrides, carbides, silicides, borides, phosphides, etc. The salts will be absolutely white and with no discolorations whatever. The metal will be silvery bright and similarly free of any foreign non-metallic bodies or discolorations, or gray powdery material. Also, the cell should be covered to prevent contamination of the electrolyte and possible downgrading of the deposited metal. An inert gas atmosphere such as argon or helium may be used above the electrolyte or in the cathode compartment, if desired.

The cathode current efficiency is always about and the recovery effici'ency may be greater than 88% depending on the cell design and operating conditions. In a basket" cathode and lined cell, recoveries approaching 100% are realizable. Finally, the metal crystals are easier to consolidate into rods than the common hydrogen reduced powder due to their larger crystal size, purity, and ductility and greater resistance to oxidation, as their physical characteristics lend them to consolidation techniques that cannot be applied to powder or amorphous metal.

The following additional examples are given to further distinguish and illustrate my invention and are not intended to limit the scope thereof.

Example I In a graphite crucible, fused 100 parts of anhydrous CaCl and 16.5 parts of CaF (eutectic mix melting at about 660 C.). and carefully added a small amount of W0 pure anhydrous oxide. On addition, dense white fumes were observed which were odorless. With further addition of W0 (predried in an oven at about 500 C.

for 6 hours), the same result followed. Using the walls of the graphite crucible as anode and an iron rod dipping into the center of the bath as a cathode, the bath was electrolyzed. For a brief few moments, more dense white smoke was discharged from the cell and then a quiet smooth electrolysis ensued. The statistics are shown in Table I below. As far as could be determined, only calcium metal was deposited and no recoverable satisfactory deposit of tungsten was obtained.

The same negative results were obtained with a fused NaCl bath.

TABLE I Electrolyte Solvent Eutectic CaCl -CaF Concentration of W0 Unknown, if any. Temperature 725-750 C. Cathode Current Density 40+ amperes/inF. Anode Current Density 1.5 to 2 amperes/inF. Deposit No acceptable W, if any.

Example II In a covered graphite crucible identical in size to the one used in Example I above, there were fused 150 parts of CaCl and 20 parts of CaO obtaining a clear melt. About 30 parts of W0 were added. This time the tungstic anhydride was quickly accepted and dissolved in the bath without fuming, smoking, or losses of any kind. Using the graphite crucible as anode and an iron rod as cathode as above the bath was electrolyzed. During the course of electrolysis, the cathode rod and its adhering electrodeposit was gradually withdrawn from the bath. The electrolysis was discontinued when the bath was stripped of W0 as indicated by a sudden rise in the back The cathode and its deposit was allowed to cool to room temperature and washed with cold water to remove the salts. The results are summarized in Table II. The salts were white and easily dissolved away leaving no discolored residues. The metal was seen to be in the form of crystals, silver in color, brilliant, with individual crystal sizes in excess of inch long. There were no discolored particles among the crystals nor any gray powder.

TABLE II Electrolyte Composition CaCl -CaOWO Temperature 875-975 C. Initial Cathode Current Density 120 amperes/in. Anode Current Density 2.5-4 amperes/in. Deposit Tungsten 99.87%.

Example 111 Fused 1100 parts of purified anhydrous OaCl in a covered metal pot which was lined internally with a high fired A1 0 sleeve. After attempting to dissolve dried anhydrous tungstic oxide in the bath and finding no success whatever, added about 200 parts of CaO and stirred. Obtained a clear fluid melt. Passed a low voltage A.C. current through the bath to raise the temperature from 700 C. to about 1050 C. Electrolyzed with low voltage DC. in order to rid the salts of impurities, moisture, etc. Added 226 parts of W0 to the melt and obtained a quite thin melt. Compartmentalized the cell by using the exposed bottom of the metal pot as the cathode and inserting a graphite rod a short way into the bath from above as anode. The anode was protected above the solution level by a ceramic A1 0 sleeve. Commenced the electrolysis and passed only the theoretical number of ampere hours needed to consume all the W0 in the electrolyte through the cell. The back was measured periodically and when it was seen to jump, indicating the depletion of W0 in the bath, the event was recorded and noted that it coincided substantially with the equivalent ampere hours required to consume all the W0 The electrolysis was discontinued thereafter. The bath was allowed to cool to room temperature. The salts were washed away from the deposit by soaking in dilute 10% HCl. After the metal was rinsed in cold water, it was given a caustic soda solution wash, rinsed again in water and air dried. The metal was meticulously examined under a 40 and power microscope. All the metal was found to be bright silver in luster and in well-defined crystals. No discolorations could be seen. The salt had been white and no discolorations were seen therein either. Most of the crystals seemed to be at least /s inch in length and many were A of an inch and longer. The conditions are shown in Table III. The recovery efficiency was 97% TABLE III Electrolyte CaCl CaO-WO Temperature 1050 to 1250 C. Initial cathode current density 7.2 :arnperes/inF. Anode current density 4.5 amperes/in Deposit, mass 99.92% tungsten. Single crystals, large 99.96% tungsten.

7 Example IV Fused in an Inconel cell about 6 pounds of CaCl by first dehydrating the salts under vacuum below about 200-250 C. and then rapidly fusing the melt. Obtained a clear water white melt. The temperature was 1380" F. Raised the temperature to 1800 F. and added 200 grams of precalcined CaO. Melt was crystal clear. Added about 200 grams of blue tungsten oxide after having electrolyzed the bath at low voltage to decompose residual water and to remove base impurities. Using a 1 /2 inch diameter graphite anode and the cell bottom as the cathode, electrolyzed the melt for about 925 ampere hours. The statistics for the run are summarized in Table IV below. During the course of the electrolysis, the bath was fed with additional oxide and at the end the bath was stripped of tungsten oxide solute. The salts were washed from the metal first with cold water. The metal was washed with 25% HCl and rinsed, and then with hot 30% Na CO solution followed by water rinsing and a rinse with HCl followed by water. The metal was air dried without heat. The metal weighed 2 /2 pounds and was silvery bright in lustre. There were 160 grams of graphite consumed in the electrolysis.

Molybdenum anhydride, M00 was found to be insoluble in either fused CaCl or NaCl-KCl eutectic.

Example VI Ina covered cell similar in size and shape to that used in Example II above, fused parts of anhydrous CaCl and added 30 parts of CaO. Obtained a clear melt at about 825 C. Added about 10 parts of M00 anhydride to the melt. M00 is not as compatible with this melt as the W0 The temperature was raised to 1000 C. by imposing a low voltage A.C. across the cell. Using the walls of the crucible as the anode and a nickel rod as a cathode, the DC. electrolysis was commenced. The electrolysis proceeded smoothly and was discontinued when the dissolved M00 was seen to be consumed as indicated by the sharp rise in the back The cathode and its adhering deposit was allowed to cool and washed free of salts in dilute HCl and water. The metal was air dried and on examination was found to consist solely of brilliant silver colored crystals. There was no evidence of carbide or lower oxide formation and the bath salts were white. The conditions are shown in Table VI.

7 TABLE VI Electrolyte CaCl CaO-MoO Temperature 975 to 1050 C. Initial Cathode Current Density 44 amperes/in Anode Current Density 3 amperes/inF. Deposit Silvery Mo Crystals.

The process hereinabove described is applicable to the direct production of other refractory multivalent metals from their oxides.

I have disclosed in great detail a process for the production of a refractory multivalent element of the group tungsten and molybdenum wherein by electrolyzing the respective oxides of these metals in an electrolyte consisting of calcium chloride and calcium oxide, one may (1) produce the respective metal of purity in excess of 99.8|%; (2) produce the metal in the form of large and ductile crystals free of powder or contaminating compounds; (3) utilize a bath which is quickly and easily separated from the pure metal by decantation, water washing, or elutriation; (4) produce a metal that is substantially free of borides, carbides, lower oxides, nitrides, phosphides, silicides and the like; (5) operate at high current and recovery efiiciencies; and (6) have an economical and commercially acceptable method for producing these metals. It is apparent that the invention is not limited to the details of procedure herein described but may be modified without departing from the spirit and scope of the disclosure as defined by the appended claims.

As indicating the effect obtained at lower percentage of CaO, the following example is further illustrative.

Example VII In a graphite lined crucible fused 80 parts of calcium chloride. The fused salt was transparent and water white. Added carefully to this melt l to 2 parts of yellow tungsten oxide. The oxide was rejected and expelled and little if any oxide was accepted by the melt. Added one part of calcium oxide and obtained a Water white transparent solution. Now when the tungsten oxide is added, it is readily accepted and dissolved in the bath leaving a clear and transparent solution. About 4 to 5 parts of tungsten oxide may be added before again some of the tungsten is found to be rejected by the solution. A bath of this composition on electrolysis at 1350 C.1400 C. using an iron cathode and the graphite crucible as anode gave a deposit of tungsten metal.

The experiment was continued and another part of calcium oxide was added to the melt. Four more parts of tungsten oxide on addition were readily accepted and dissolved therein. This mix was transparent and fluid and on electrolysis tungsten metal was produced. Finally, still another part of calcium oxide was added to the melt and 5 parts of tungsten oxide added thereafter. Again the bath was transparent and when subjected to electrolysis, tungsten metal was obtained.

From this work it may be seen that there must be some calcium oxide added to the melt above the normal oxide contaminant associated with calcium chloride before any appreciable amount of tungsten will enter the electrolyte. Concentrations of calcium oxide from about one percent up to 20 percent will work successfully. However, the lower calcium oxide baths containing below about 5 percent will dissolve proportionately less tungsten oxide. As stated, both the high concentrations and the low concentrations of tungsten oxide in the electrolyte have special advantages and determine the concentrations of calcium oxide needed therein.

This application is a continuation-impart of my copending patent application Serial No. 636,228 filed January 25, 1957 and embodies subject matter and claims therefrom and now abandoned.

I claim:

1. A process for the electrolytic production of a refractory metal of the group tungsten and molybdenum comprising dissolving at least one oxide of said refractory metal in a molten electrolyte composed of at least one halide of the group alkaline earth metal halides and at least one oxide of the group alkaline earth metal oxides in substantial amount in said electrolyte and sufficient to attain therein a non-turbulent, clear melt with the said refractory metal oxide, electrolyzing said molten electrolyte between an anode and a cathode, thereby depositing said refractory metal in pure ductile and large crystalline form on said cathode.

2. A process for the electrolytic production of a refractory metal as set forth in claim 1 wherein said alkaline earth metal halide is the chloride.

3. A process for the production of a pure refractory multivalent element of the group consisting of tungsten and molybdenum, comprising electrolyzing between an anode and a cathode a molten electrolyte comprising a solvent mixture of at least one halide of the group alkaline earth halides and at least one oxide of the group of alkaline earth oxides and also containing a substantial amount of at least one oxide of the said tungsten and molybdenum group, the amount of said alkaline earth oxide being substantial and sufficient to attain in said electrolyte a nonturbulent clear melt with the oxide of said tungsten and molybdenum group and thereby discharging oxygen at said anode and depositing said element of said group in pure, ductile and large crystalline form on said cathode.

4. A process for the production of a pure refractory multivalent element as set forth in claim 3 wherein said molten electrolyte comprises calcium chloride, calcium oxide and tungstic oxide.

5. A process for the production of a pure refractory multivalent element as set forth in claim 3 wherein the molecular ratio of CaO to W0 in said electrolyte is greater than 1:1.

6. A process for the production of a pure refractory multivalent element as set forth in claim 3 wherein said element is molybdenum and said molten electrolyte comprises calcium chloride, calcium oxide and molybdic oxide.

7. A process for the refining of an impure refractory multivalent metal of the group tungsten and molybdenum comprising electrolyzing a fused bath contacting a cathode and with an anode of said multivalent metal immersed in a molten electrolyte comprising a solvent mixture of at least one halide of the group of alkaline earth halides and at least one oxide of the group of alkaline earth oxides in substantial amount in said electrolyte and sufficient to attain therein a non-turbulent, clear melt with a substantial amount of at least one oxide of said tungsten and molybdenum group added to said electrolyte thereby depositing large refined crystals of said multivalent element on said cathode.

8. A process for the production of a pure refractory multivalent element of the group consisting of tungsten and molybdenum, comprising electrolyzing a fused electrolyte of at least one oxide of said multivalent element dissolved in a solvent consisting of about 10% or over calcium oxide and at least one calcium halide between an anode and a cathode, thereby depositing said multivalent element at said cathode in the form of large crystals.

9. A process for the production of a pure refractory multivalent element as set forth in claim 8 wherein said calcium halide is calcium chloride.

10. A process for the production of a pure refractory multivalent element as set forth in claim 8 wherein said fused electrolyte comprises and said metal oxide 10 to 40%.

11. A process for the production of a pure refractory multivalent element as set forth in claim 8 wherein the cathode current density is between 10 and 70 amperes per square inch initially, and wherein the anode current density is below 10 amperes per square inch.

12. A process for the production of a pure multivalent refractory element as set forth in claim 8 wherein said refractory element is tungsten and said metal oxide is tungstic oxide.

13. A process for the production of a pure multivalent refractory element as set forth in claim 8 wherein said refractory element is molybdenum and said metal oxide is M 14. A process for the refining of a refractory metal as set forth in claim 8 wherein the cathode is made of 15. A process for the refining of a refractory multivalent metal of the group tungsten and molybdenum, comprising electrolyzing a fused bath with an anode of said impure multivalent metal immersed in a fused salt bath composed of calcium chloride, about 10% or over calcium oxide, and at least one oxide of said multivalent metal, thereby depositing refined crystals of said multivalent element on said cathode and maintaining the concentration of said multivalent element solute in said fused bath substantially constant.

16. A process for the electrolytic production of tungsten in the form of bright silvery crystals in excess of 99.8% purity and substantially free from powder, lower oxides, tungsten bronzes, phosphides, carbides, borides, and silicides, comprising dissolving at least one oxide of tungsten in a fused salt electrolyte composed of calcium chloride and about 10% or over calcium oxide, electrolyzing said electrolyte between an insoluble anode and a cathode at a temperature in excess of 1050 C., thereby electrodepositing tungsten metal of high purity at said cathode.

17. A process for the electrolytic production of tungsten as set forth in claim 16 wherein said oxide of tungsten is W0 18. A process for the electrolytic production of tungsten as set forth in claim 16 wherein said oxide of tungsten is supplied from calcium tungstate, OaWO in the electrolyte.

19. A process as set forth in claim 1 wherein said refractory metal is tungsten and said tungsten oxide and said alkaline earth metal oxide are supplied by feeding sheelite to the electrolyte.

20. A process as set forth in claim 1 wherein said refractory metal is molybdenum and said molybdenum oxide and said alkaline earth metal oxide are supplied by feeding powerlite to the electrolyte.

References Cited in the file of this patent UNITED STATES PATENTS 1,821,176 Driggs Sept. 1, 1931

Claims

1. A PROCESS FOR THE ELECTROLYTIC PRODUCTION OF A REFRACTORY METAL OF THE GROUP TUNGSTEN AND MOLYBDENUM COMPRISING DISSOLVING AT LEAST ONE OXIDE OF SAID REFRACTORY METAL IN A MOLTEN ELECTROLYTE COMPOSED OF AT LEAST ONE HALIDE OF THE GROUP ALKALINE EARTH METAL HALIDES AND AT LEAST ONE OXIDE OF THE GROUP ALKALINE EARTH METAL OXIDES IN SUBSTANTIAL AMOUNT IN SAID ELECTROLYTE AND SUFFIICIENT TO ATTAIN THEREIN A NON-TURBULENT, CLEAR MELT WITH THE SAID REFRACTORY METAL OXIDE, ELECTROLYZING SAID MOLTEN ELECTROLYTE BETWEEN AN ANODE AND A CATHODE, THEREBY DEPOSITING SAID REFRACTORY METAL IN PURE DUCTILE AND LARGE CRYSTALLINE FROM ONE SAID CATHODE.