US2762691A - Method of preparing halides - Google Patents

Description

2,762,691 Patented Sept. 11, 1956 NEETHQD (1 F PREPARKNG HALIDES Eugene Wainer, Cleveland Heights, Ohio, assignor to Horizons Incorporated, Princeton, N. 3., a corporatton of New Jersey No Drawing. Application November 10, 1952, Serial No. 319,791

2 Claims. or. 2s 2ss This invention relates to the production of halides other than the fluorides of metals other than the alkali metals, the alkaline earth metals and the rare earth metals.

The halides of metals are of particular importance because they are frequently the starting compounds used for the production of other salts of the metals. The most common method of producing these halides is by halogenation, often at elevated temperature, of the oxide or other compound of the metal. The halogenating agents for this purpose include not only the halogen itself but also the hydrogen halides. Such halogenation procedures, particularly at the elevated temperatures required, dictate the use of corrosion-resistant apparatus and the maintenance of numerous safety requirements for protection of operating personnel.

I have now devised a method of producinghalides, other than the fluorides, of metals other than the alkali metals, the alkaline earth metals and the rare earth metals without using a corrosive or hazardous halogen or halogen compound. Thus, the method of my invention makes it possible to produce the desired halides with conventional equipment and without hazard to operating personnel. Moreover, the method of my invention is amenable to a recycling or recovery of the by-products of the reaction with a resulting over-all economy which makes this method particularly attractive for the commercial scale production of the halides.

My novel method of producing halides other than the fluorides of metals other than the alkali metals, the alkaline earth metals and the rare earth metals, which metal halides are volatile at temperatures below 1200 C., com prises heating to a temperature of at least 500 C. an anhydrous mixture of an alkali metal double fluoride of the metal and a chloride, bromide or iodide of magnesium, calcium or lithium, and withdrawing from the heated mixture the resulting evolved vapors of the desired metal chloride, bromide or iodide. In the course of this reaction, the initial magnesium, calcium or lithium halide is converted to the corresponding fluoride and this fluoride in turn may either be used as a source of fluorine for the production of the metal fluoride used as the starting material or it may be transformed into another halide and used as the reagent which is reacted with the metal fluoride.

The metals whose alkali metal double fluorides may be converted to other halides pursuant to the method of my invention comprise all metals other than the alkali metals, the alkaline earth metals and the rare earth metals. For example, the method of my invention may be used to produce substantially anhydrous halides (other than the fluoride) of such diverse and representative metals as titanium, silicon, germanium, zirconium, hafnium, thorium, aluminum, boron, niobium (columbium), tantalum, iron, tin, molybdenum, tungsten, arsenic, anti mony and gallium.

The metal fluorides used as starting materials in the method of my invention comprises complex fluorides, to

wit, the alkali metal double fluorides. Thus, both the sodium and potassium double fluorides of the aforementioned metals may be used with particular advantage in the practice of the method of my invention. Accordingly, the term metal shall be used hereinafter to embrace in eachinstance allmetals other than the alkali metals, the alkaline earth metals and the rare earth metals, and the term metal fluoride shall be used to embrace the double fluorides of these metals.

The metal fluoride is converted to the desired other halide pursuant to my invention by reaction with a halide other than the fluoride of either magnesium, calcium or lithium. The choice of the halide of these latter metals depends entirely upon the halide which it is desired to produce. For example, magnesium chloride will, by the method of the invention, yield the desired metal chloride. The choice in the use of either magnesium, calcium or lithium as the metal component of this halide is primarily one of economics and may difler under varying circumstances in the availability of these metal halides. Thus, I have found that both magnesium and lithium chlorides yield the metal chlorides with at least efficiency whereas calcium chloride produces the same result at about 85% efliciency. The bromides of magnesium, calcium and lithium are nearly as eflicient as their corresponding chlorides, Whereas the iodides give considerably lower yields. Of the two more efficient halides, i. e. the magnesium and lithium chlorides, bromides and iodides, the magnesium halides are presently preferred because they are more readily available and are considerably less expensive than the corresponding lithium halides.

With specific reference to titanium, the following equations illustrate the nature of the basic chemistry;

(and analogously with the applicable bromides and iodides and Ti, Ge or Si double fluorides).

The reaction between the metal fluoride and the magnesium, calcium or lithium halide other than thefluoride will produce the desired metal halide provided that the reactants are substantially completely anhydrous at the time of reaction. Toward this end, the individual reactants may be rendered anhydrous before being introduced into the reaction zone, or the admixed reactants may be subjected to a dehydrating operation, advantageously in situ in the reaction zone. The individual double (complex) fluorides may be rendered completely anhydrous by heating them to a temperature between and 200 C. in a vacuum of about 1 millimeter of mercury or less. The lithium, magnesium and calcium halides used in practicing the invention may be rendered anhydrous by any of a variety of techniques. One technique, in the case of the chlorides for example, comprises heating the chloride to about 350-400 C. while passing dry hydrogen chloride gas through the salt until the weight of the salt becomes constant. An alternative technique is to admix the magnesium, calcium or lithium halide with the corresponding ammonium halide and then heat the mixture to a temperature of about 500 C. under subatmosphen'c pressure. A third effective technique is to heat the magnesium, calcium or lithium halide to a temperature of about 350 400 C. in a vacuum of about 1 millimeter of mercury or less. Where the reactants are admixed prior to introduction. into the reaction zone,

anhydrous condition resulting from any of the aforementioned procedures, the metal halides produced by the method of my invention will be virtually anhydrous and substantially free of products of hydrolysis.

Inasmuch as impurities introduced with the reactants may be carried over into the metal halide product of the method of the invention, it is preferable to use substantially pure reactants. The double fluorides of the metals may be obtained in a state of suitable purity by at least one recrystallization of the double fluoride from an aqueous medium. The halides of magnesium, calcium and lithium, on the other hand, are available in commercial quantities in a state of purity wholly suitable for use directly in the practice of the method of my invention without intermediate purification.

The amounts of the two reagents used in the practice of the invention may vary considerably without significant effect upon the efliciency of the reaction. I have found, however, that substantially stoichiometric quantities of the reactants lead to as good results as an excess of either reactant and that furthermore the use of stoichiometric quantities of the reactants results in the formation of a reaction residue containing a minimum of unconsumed reactants. Inasmuch as it is my presently preferred practice to recycle the reaction residue, it is advantageous to use substantially stoichiometric (equimolar) amounts of the reactants and thereby minify the burden of recycled unconsumed charge components.

Temperatures of at least about 500 C. are required for effecting reaction between the aforementioned charge components, although the minimum effective temperature varies for the various metals. Higher temperatures promote more rapid and more complete reaction, and at temperatures 50 to 100 C. above this minimum temperature the reaction is substantially complete and the resulting metal halide is completely volatilized from the reaction mass. Still higher temperatures may be used in practicing the invention although they tend to transform the solid reagents or residues, or both, to the molten state with resulting decrease in the rate of evolution of the vapors of the desired metal halide. Thus, for each metal there is a maximum satisfactory temperature which is marked either by fusion taking place in the reaction zone or by decomposition of one or more of the reagents or residual compounds with resulting contamination of the evolved metal halide with a volatile product of such decomposition. Accordingly, useful temperatures for practicing the method of my invention range between that which causes the reaction to proceed (and generally at least 500 C.) and that at which decomposition of either a reactant or a reaction product occurs with liberation of a volatile substance which would contaminate the evolved metal halide.

The range of minimum and maximum temperatures for producing a variety of metal halides pursuant to the Table Maximum Reaction 'Iernp.,

Minimum Reaction Temp., C.

Source of Yield, Metal Halide Formed Metal Percent In each of the reactions represented in the table, the double fluoride was purified before use by recrystalliza- .tion from an aqueous medium and was dried under vacuum at a temperature of C. The yield figures in the table are those obtained at a reaction temperature 50 C. above the stated minimum temperature with a reaction period of one hour.

The physical state of the reactants may, as indicated hereinbefore, be that of either a solid or a liquid. Because of the ease of handling a solid charge and a solid residue, the solid state reaction is presently preferred, and with this prescription in mind the reaction temperature should be limited to that below which fusion of the reactants and of the residue of the reaction remain solid. I have found it particularly advantageous, when carrying out the reaction in the solid state, to charge the reactants in the form of briquettes in a batch operation. The water of hydration of the components of the charge mixture is sufficient to promote agglomeration of the charge components under compressive pressure, and the resulting briquettes are then subjected to dehydration in the reaction zone prior to carrying out the reaction. The size of such briquettes is not critical, the most advantageous size depending largely upon the size and shape of the reaction vessel. For example, in a reaction chamber 10 inches in diameter, I have found that briquettes in the form of about 1 inch cubes are particularly satisfactory. On the other hand, where it is desired to carry out the reaction in the liquid phase, the charge components and their ultimate residue may be maintained in a suitable fluid condition by the use of a reaction temperature of 1000 C. or higher, the specific temperature depending upon the specific reactants. If desired, the fluid state reaction may be carried out in a diluent bath composed of one or more alkali metal halides. In any such fused-state operation, the reactants may be continuously or at leastincrementally charged to the reaction crucible, or the crucible may be initially charged with a relatively large amount of the magnesium, calcium or lithium halide and subsequently the metal fluoride may be charged incrementally to the fused body of the other halide during evolution of the desired metal halide.

Regardless of the physical state of the reactants, the atmosphere in which the reaction is carried out should be substantially inert so as to avoid contamination of the evolved metal halide. For this purpose either vacuum pumping or sweeping with an inert gas such as argon or helium may be used efiectively. In either procedure, the pumping or sweep circuit should be such as to remove the evolved metal halide vapors as they are formed and deliver them to a cooling zone in which the vapors may [condense and thence be recovered. In practice, I have found it advantageous to use a combination of these expedients. For example, after a substantial vacuum is established while maintaining the charge 'at an elevated temperature below the reaction temperature to insure dehydration of the charge, argon may be admitted to the reaction vessel, and then the argon and evolved metal halide vapors are withdrawn during the subsequent reaction by means of active vacuum pumping. On the other hand, if the charge is completely dehydrated prior to introduction into the reaction vessel, it may be charged continuously or incrementally into a vessel such as a rotary kiln wherein the reaction is carried out while maintaining an inert atmosphere by sweeping the interior of the vessel with an appropriate inert gas. In the latter case, the operation can be made continuous by charging the reactants to one portion of the kiln,.withdrawing the residue from another portion of the kiln, and recovering the metal halide from the eflluent diluent sweep gas stream.

Recovery of the evolved metal halide may be readily efiected by withdrawing th evolved vapors from the reaction zone and by permitting them to cool with resulting condensation either to the liquid or solid state. In practice, I have found it to be particularly satisfactory to interpose a condenser in a vapor draw-off line beyond the reaction zone, advantageously between the reaction zone and .a vacuum pump where the pump serves not only to draw off the evolved metal halide vapors but also to maintain a suitably inert reaction atmosphere. The metal halide vapors thus removed from the reaction zone are readily condensed to the liquid or solid form by the degree of cooling effected by a water cooled condenser, and virtually none of the metal halide is lost by carryover into the vacuum pump if the condenser capacity is adequate.

Other than the evolved vapors of the desired metal halide, the products of the reaction are normally solid. For example, in the production of titanium tetrachloride by reaction between the alkali metal-titanium fluoride and magnesium chloride, the residual reaction product consists of magnesium fluoride and the corresponding alkali metal fluoride obtained pursuant to the reaction equation: K2TiF6+2MgCl2- TiCls-I-ZMgFz-I-ZKF. This residue is particularly amenable to recovery as a source of fluorine in the production of the aforesaid double fluoride of the metal from other sources of the metal component such as the oxides or hydrates of the metals. It will be seen, accordingly, that all of the fluorine component and all of the alkali metal component of the residue are recovered by this procedure with resulting economy of the source materials.

The re-use of the solid residue of the reaction stage in the practice of my invention does not in general entail any intermediate treatment of this residue. As pointed out hereinoefore, the residue resulting from the reaction between a metal fluoride and a magnesium halide other than the fluoride comprises the relatively Water-insoluble magnesium fluoride accompanied by an alkali metal fluoride. When a lithium halid is used as a reagent in the production of the metal halide, the residue comprises the more water insoluble lithium fluoride, and the residue obtained with a calcium halide comprises the still more water-insoluble calcium fluoride. The magnesium fluoride-containing residue may be used directly in an aqueous phase process for the production of the metal fluoride, and the lithium or calcium fluoride-containing residue may be used directly in a fusion process for the production of the metal fluoride starting material. For example, aluminum sulfate, produced by sulfuric acid digestion of aluminum oxide raw material or of an alkaline earth metal aluminate resulting from sintering' of alumina with an alkaline earth metal oxide, will react readily in the hot aqueous phase with magnesium fluoride and an alkali metal fluoride to form the corresponding alkali metal-aluminum double fluoride. As pointed out hereinbefore, when such a double fluoride is reacted with a magnesium halide other than the fluoride in practicing the method of my invention, the solid residue of that reaction will consist essentially of precisely such a mixture of magnesium fluoride and alkali metal fluoride as that required for conversion of aluminum sulfate to the corresponding double fluoride with an alkali metal. The lithium fluoride in a residue containing this salt can be readily converted to lithium chloride and recycled directly in the method of my invention. This conversion may be readily effected by reacting the lithium fluoride with calcium chloride in the aqueous phase with the resulting formation by metathesis of lithium chloride and the still more water-insoluble calcium fluoride. The lithium chloride is recovered from the separated aqueous phase by evaporation and is then returned directly as the lithium halide reactant in the method of the invention, and any alkali metal fluoride in the calcium fluoride precipitate and in the remaining aqueous liquor can be used as the source both of the alkali metal and of the fluorine required for the production of the double fluoride of the metal and "an alkali metal. Thus, the method of my invention is characterized by the possibility of recycling the by-products of the main reaction with resulting economy of all components of the reactants other than those which are combined in the desired main product of the reaction.

The following examples are illustrative of the method of the invention:

Example 1.-A reaction chamber is provided which is in the form of a crucible consisting of an inner liner of graphite and an outside container made of impervious quartz. The crucible is covered tightly and the cover is fitted with a hopper feeding-device through which solid material can be fed into the crucible by the action of a screw drive. An inlet is available in the top of the hopper through which dry argon is fed. A condenser outlet is also attached to the cover and the condenser outlet is lagged so that .its temperature up to the condenser proper does not drop below 400 C. The condenser is water-cooled and air-tight. Means for heating the crucible and for making temperature measurements are available.

170 grams of anhydrous lithium chloride (approximately 4 mols) is thoroughly mixed with 240 grams of KzTiFs (approximately 1 mol) and the reaction mass is briquetted under pressure. The briquettes are put in the crucible. All the air is swept out with the dry argon, and the temperature of the crucible is raised rapidly to a range between 400 and 450 C. and a rapid evolution of gaseous titanium tetrachloride takes place which is condensed to a liquid in the water cooled condenser. After baking for a few minutes at 450 C., the mass is allowed to cool. The contents of the crucible consist of a graywhite porous sinter which has shrunk away from the side of'the crucible, and is found to weigh 225 gnams or substantially equivalent to two mols of potassium fluoride plus 4 mols of lithium fluoride, indicating the quantitative nature of the reaction.

This porous sinter is digested in 1000 cc. of warm water for about an hour, after which the precipitate obtained is separated from the solution by filtration and washing. The filtrate is retained for further use. The precipitate consisting substantially of pure lithium fluoride is dispersed in 500 cc. of water and 300 grams of calcium chloride dihydrate is added. The bath is digested at C.

for about an hour with continuous stirring, and the precipitate which is obtained is separated by washing and iiitration. Analysis indicates that the precipitate is substantially pure calcium fluoride. The filtrate is evaporated to dryness and then baked at 200 C. for approximately an hour and is found to weigh grams, indicating that the recovery of the lithium chloride is substantially quantitat-ive. The filtrate from the first digestion of the lithium fluoride is combined with the precipitate from the calcium chloride reaction to serve as reagents for the decomposition of fresh titanium raw material.

The titanium raw material is obtained through the medium of reaction of sulfuric acid on either TiOz or ilmenite. In the case of TiOz, the reaction takes place through treatment of one mol of TiOz with two mols of sulfuric acid at elevated temperature. In the case of ilmenite, the ore is decomposed with H2804 and the desired titanium compound is hydrolytically separated from the iron through the medium of precipitation of a water-insoluble basic sulfate of titanium. The basic sulfate may be transformed to a water-soluble sulfate by adding the requisite amount of sulfuric acid, or a mixture of basic sulfate and the requisite amount of sulfuric acid may be used directly in the fluorinating reaction.

Then a reaction equivalent approximately to the following is carried out. Two mols of CaFz+-2 mols of KF+1 mol of titanium sulfate+1 mol of sulfuric acid is digested for extended periods in the region of the boiling point in the presence of an adequate amount of water. One or two per cent excess KCl is advantageously added at this stage. The products of the reaction are KzTiFs and CaSO4, the titanium compound being in solution and the calcium sulfate compound being separated therefrom by filtration of the hot solution. The potassium titanium fluoride is recovered by concentrating and cooling the filtrate. After drying, it is then ready for fresh admixture with the lithium chloride for repetition of the reaction.

perature treatment of the fluorides of calcium and potassium on the titanium sulfate followed by subsequent digestion and the like.

Example 2.Same as Example 1 except that approximately 4 mols of lithium bromide are substituted for the lithium chloride, 350 grams of lithium bromide being used. The same amount of combined potassium lithium fluoride residue is obtained as in Example 1, indicating a quantitative recovery of titanium tetrabromide. In the recycling operation based on lithium fluoride, 620 grams of CaBmGHzO are used to reconstitute the lithium bromide, this calcium compound being used in place of the calcium chloride compound indicated in Example 1. Again as pointed out, the reaction is substantially quantitative.

Example 3.-Same as Example 1 except that 545 grams (approximately 4 mols) of lithium iodide are substituted for the chloride used in Example 1. Again, 810 grams of CaI2.6H2O are used in place of the calcium chloride indicated in Example 1 for the reconstitution of the lithium iodide from the lithium fluoride. Again, a high yield is obtained throughout. In the case of the titanium iodide, it is necessary to run the reaction and the temperature of the condenser cross-over in a slightly higher range than indicated for the chloride and bromide. The reaction vessel is heated to approximately 500 C. and the condenser cross-over is maintained at 450 C. rather than 400 C. The purpose of maintenance of this latter temperature is to prevent the condensation of the iodide in the crossover arm, with the possibility of choking up the equipment.

Example 4.Same as in Example 1 except that 210 grams of sodium titanium fluoride (approximately 1 mol) are substituted for the potassium titanium fluoride used in Example 1. The fluoride residue obtained as a result of the reaction of the formation of titanium tetrachloride weighs 190 grams, substantially equivalent to a mixture of 2 mols of sodium fluoride plus 4 mols of lithium fluoride. The same recycling operation is complete-d as in the case of Example 1 except that the product obtained as a result of decomposition of the ilmenite or TiOz is sodium titanium fluoride rather than potassium titanium fluoride.

Example 5.-Same as Example 1 except that 225 grams (approximately 1 mol) of potassium fluosilicate is substituted for the fluotitanate used in Example 1. The yields of anhydrous fluoride left as a residue in the container are substantially the same as in Example 1, and

the reconstitution of the lithium chloride is carried out identically as indicated in the original example. The formation of fresh potassium fluosilicate is accomplished through the medium of a fairly specific series of reactions. The potassium fluoride obtained as a residue from the lithium fluoride digestion (in which case the potassium fluoride remains in the solution) is transformed to calcium fluoride in exactly the manner by which the lithium chloride is reconstituted, that is by treatment of the potassium fluoride with calcium chloride. This makes available approximately 3 mols of calcium fluoride or roughly 240 grams. These three mols of calcium fluoride are mixed with one mol of silica and three mols of sulfuric acid, in which case the sulfuric acid .is maintained at a concentration not exceeding per cent. The slurry is then digested at about 60 C. for two or three hours, after which the insoluble precipitate consisting chiefly of calcium sulfate is removed by filtration. The potassium chloride obtained from the metathetical reaction between potassium fluoride and calcium chloride is then added in solution form to the solution of silica rapidly and with stirring, and the precipitation of KzSiFs takes place substantial-ly quantitatively. After filtration, washing, and

drying, the material isthen ready for recirculating in the initialreaction for the formation of silicon tetrachloride.

Example 6. grams of lithium chloride are mixed with 170 grams of sodium chloride and the batch melted in the reaction chamber described in Example 1, quiet fusion being obtained in the range of 600 to 650 C. After the atmosphere has been completely purged with argon, powederd potassium titanium fluoride in the same 9 amount as used in Example 1 is fed into the molten salt at a uniform rate, the molten salt being stirred while this addition takes place. Titanium tetrachloride is evolved substantially quantitatively from the reaction. In this case, the residue consists of a mixture of potassium fluoride, lithium fluoride, and sodium chloride, and after the initial digestion reaction for the recovery of lithium fluoride, the filtrate contains then a mixture of potassium fluoride and sodium chloride. The presence of the sodium chloride does not appear to make any major difference in the precipitation reaction leading to the formation of the potassium titanium fluoride.

Example 7 .Same as Example 6 except that the carrying salt holding the lithium chloride consists of 85 parts of potassium chloride and 85 grams of sodium chloride, and these pass through the reaction zone substantially unchanged. The same general recovery and reconstitution reactions as described in Example 1 are effective.

Example 8.-'In a specially constructed chamber, 225 grams of anhydrous CaClz (approximately 2 mols) are mixed and briquetted with 240 grams (approximately 1 mol) of KzTiFs. The mixture is heated as in Example 1 and the residue is found to weigh 275 grams approximately equivalent to 2 mols of CaFz and 2 mols of RF, indicating substantially quantitative recovery of TiCl4.

The fluoride residue is digested in water with 1 mol of Ti(SO4)2.9H2O to form KzTiFe and CaSOr separated by filtration, and the KzTiFs recovered by crystallization from the solution. The CaSO4 is discarded, and the KzTiFe is retained for reuse in the original reaction.

Example 9.--Same as in Example 8 except that grams (approximately 2 mols) of anhydrous MgClz is used in place of the CaClz. After the chlorination reaction, 245 grams of residue is obtained, which is approximately equivalent to 2 mols of MgFz and 2 mols of KF. These are used as reagents for decomposition of to form KzTiFs and MgSO4 which are separated from each other by crystallization of the KzTiFs.

\It must be understood that the method of my invention is equally applicable to the treatment of either the individual metal fluorides, or mixtures thereof, for the production of other halides of either the individual metal or mixtures of such metals. It must also be understood that the method of my invention is equally applicable to the production of the metal chlorides, bromides and iodides which are stable within the recited reaction temperature range and that these metal halides are generally the halides of the highest stable valence of each metal where more than one valence state exists. The use of alkali metal double fluorides as the metal fluorides referred to herein makes possible the production of these metal halides in a state of purity much higher than if the simple fluorides of the metals were used. The double fluorides used in the practice of my invention can be readily purified by one or more simple recrystallizations, are not significantly unstable at reaction temperatures of 500 C. and higher, and therefore make possible the production of the desired metal halides of greater purity and in more nearly quantitative .yields than can be achieved by the use of the simple fluorides.

The use of the hereindescribed method for the production of thorium tetrahalides other than the fluorides is described and claimed in my application Serial No. 319,793, and the use of this method for the production of zirconium and hafnium tetrahalides other than the fluonides is described and claimed in my application Serial No. 319,792, both filed concurrently herewith.

I claim:

1. The method of producing a halide of boron other than the fluoride of boron which comprises: forming an anydrous charge mixture of (1) an alkali metal fluoborate of the group consisting of sodium fluoborate and potassium fluoborate and (2) a halide other than the fluoride of a metal from the group consisting of magnesium, calcium and lithium in which the ingredients are stoichiometrically proportioned to produce as reaction products the desired boron halide and a mixture of the fluorides of the alkali metal and the metal of the group consisting of magnesium, calcium and lithium; heating the charge mixture to a temperature between about 500 C. and

1000 C.; recovering the desired boron halide in the form of a vapor evolved from the heated mixture; recovering the alkali metal and fluoride values in the residue remaining after evolution of the desired halide by reacting said residue with a boron compound whereby the alkali metal fluoborate is reconstituted from the values in the residue; and introducing said alkali metal fluoborate into the process for further reaction with a stoichio metric proportion of a halide other than the fluoride of a metal of the group consisting of magnesium, calcium and lithium.

2. The method of producing a halide of boron other than the fluoride of boron which comprises: forming an anhydrous charge mixture of (1) an alkali metal fluoborate of the group consisting of sodium fluoborate and potassium fluoborate and (2) a halide other than the fluoride of magnesium in which the ingredients are stoichiometrically proportioned to produce as reaction products the desired boron halide and a mixture of the flue rides or" the alkali metal and magnesium; heating the charge mixture to a temperature between about 500 C. and 1000 C.; recovering the alkali metal and fluoride values in the residue remaining after evolution of the desired halide by reacting said residue With a boron compound in the hot aqueous phase whereby the alkali metal fluoborate is reconstituted from the values in the residue; and introducing said alkali metal fluoborate into the process for further reaction with a stoichiometric proportion of a halide other than the fluoride of magnesium.

References Cited in the file of this patent UNITED STATES PATENTS 1,699,234 Gaus et al. Jan. 15, 1929 2,475,287 Kawecki July 5, 1949 2,594,370 Warburton Apr. 29, 1952 2,626,203 Blumenthal Jan. 20, 1953 2,694,616 Warner Nov. 16, 1954 FOREIGN PATENTS 141,908 Great Britain Apr. 29, 1920 OTHER REFERENCES High Temperature Experiments With Zirconium and Zirconium Compounds, by W. J. Kroll, W. R. Carmody, and A. W. Schlecten, page 6, Bureau of Mines Report of Investigation 4915, U. S. Dept. of the Interior, November 1952.

Inorganic Chemistry, W. Norton Jones, 1947 ed., page 567, The Blakiston Co., Philadelphia.

Claims (1)

1. THE METHOD OF PRODUCING A HALIDE OF BORON OTHER THAN THE FLUORIDE OF BORON WHICH COMPRISES: FORMING A ANYDROUS CHARGE MIXTURE OF (1) AB ALKALI METAL FLUOBORATE OF THE GROUP CONSISTING OF SODIUM FLUOBORATE AND POTASSIUM FLUOBORATE AND (2) A HALIDE OTHER THAN THE FLUORIDE OF A METAL FROM THE GROUP CONSISTING OF MAGNESIUM, CALCIUM AND LITHIUM IN WHICH THE INGREDIENTS ARE STOICHIOMETRICALLY PROPORTIONED TO PRODUCE AS REACTION PRODUCTS THE DESIRED BORON HALIDE AND A MIXTURE OF THE FLUORIDES OF THE ALKALI METAL AND THE METAL OF THE GROUP CONSISTING OF MAGNESIUM, CALCIUM AND LITHIUM; HEATING THE CHARGE MIXTURE TO A TEMPERATURE BETWEEN ABOUT 500* C. AND 1000* C., RECOVERING THE DESIRED BORON HALIDE IN THE FROM OF A VAPOR EVOLVED FROM THE HEATED MIXTURE; RECOVERING THE ALKALI METAL AND FLUORIDE VALUES IN THE RESIDUE REMAINING AFTER EVOLUTION OF THE DESIRED HALIDE BY REACTING SAID RESIDUE WITH A BORON COMPOUND WHEREBY THE ALKALI METAL FLUOBORATE IS RECO NSTITUTED FROM THE VALUES IN THE RESIDUE; AND INTRODUCING SAID ALKALI METAL FLUOBORATE INTO THE PROCESS FOR FURTHER REACTION WITH A STOICHIOMETRIC PROPORTION OF A HALIDE OTHER THAN THE FLUORIDE OF A METAL OF THE GROUP CONSISTING OF MAGNESIUM, CALCIUM AND LITHIUM.