US2081984A - Process for obtaining beryllium fluoride solutions - Google Patents

Description

Patented June 1, 1937 UNITED STATES PROCESS FOR OBTAINING BERYLLIUM FLUORIDE SOLUTIONS Harry Glaflin,

The Beryllium Corporation,

Marysville, Mich., assignor to New York, N. Y.,

a corporation of Delaware.

No Drawing. Application June 26, 1934, Serial N0.'732 ,43'7

I 1 Claim. 'I'hisinvention relatestoa convenient and inexpensive method for solubilizing and thereby making commercially available the beryllium content of beryl, beryllium ores, and other beryllium 5 compounds, at the same time leaving the other components of the ore in an insoluble state.

Since the discovery by Copaux that the treatment of beryl with twice its weight of sodium fiuosilicate results in the formation of a soluble double fluoride of beryllium, insoluble cryolite,

and insoluble silica, there have been notable advances in the technique of opening up beryllium ores, but all these have hinged upon the use of analkali metal fiuosilicate or its equivalent, and the consequent formation of a double alkali beryllium fluoride. For example, it was: found soon after Copauxs original discovery that twice the weight of reactant was not needed, and that limiting the ratio to approximately 1:1 gives resuits equally satisfactory from all viewpoints, the only difierence being that the aluminum content of the beryl remains as oxide, while with increase of fiuosilicate, transformation yields cryolite instead of aluminum oxide, as previously indicated. Similarly, I have shown, in Patent No. 1,929,014 that the amount of fiuosilicate per unit of ore can still further be curtailed in part if one mol. of sodium fluoride 42 parts by weight) be substituted for a mo]. of sodium fiuosilicate (approximately 188 parts by weight). This can be done, in theory, up to about half of the fiuosilicate in the 1:1 beryl-fiuosilicate mixture. Further, I have shown in a patent application now pending, Serial No. 570,756, filed October 23, 1931, now 5 Patent 1,991,272 issued Feb. 12, 1935, the fluosilicate use can be eliminated in entirety by the use of a soluble alkali fluoride in molecular proportion of two mols of the fluoride toevery mol. of beryllium present in the ore, and passing gaseous silicon fluoride over the intimately mixed ore and alkali fluoride.

It will be noted that whereas the original fluosilicate process, as outlined by Copaux, calls for a reaction following the formula (using beryl as the standard example, since it is most common among beryllium ores) the curtailment of sodium fiuosilicate down to the 1:1 ratio brings a change in final products in the nature of 2BesA12 (SiOs) s+6NazSiFe=6NazBeF4+ 2A1203+ 3SiF4+ 15SlO2 The substitution of sodium fluoride for part of the fiuosilicate is of advantage primarily as a matter of economy since it prevents the destru'c tion of fiuosilicate in situ for the purpose of making sodium fluoride. In this arrangement, operations follow the reaction:

The total elimination of fiuosilicate and the substitution of sodium fluoride plus silicon fluoride gas for it and the other reactants used in the past can readily be understood to follow the equivalent formula:

Careful comparison of these processes will show one factor common to all. All, irrespective of reactant, ratios, and other final products, yield the beryllium content in the form of the soluble alkali double beryllium fluoride-in the examples given above, as NazBeFi.

My present invention, While at first glance appearing quite analogous to the processes described above, in View of the reactant used, does actually differ sharply in both theory and in practice. No double fluoride is formed, and no alkali metal remains in the solution to interfere with beryllium reactions. Instead, the primary and only product is normal beryllium fluoride, uncontaminated. The value of such procedure is manifest; where it is desired to convert the beryllium to ammonium beryllium fluoride, for direct use in electrolysis, as described in my copending patent application Serial No. 432,436 there is. no need for removing the sodium fluoride equivalent, as has hitherto been so essential. Similarly, the beryllium fluoride solution can be taken down to dryness directly, to yield the oxyfluoride, which is directly usable as a source of beryllium in high temperature electrolysis.

To achieve my end in. this invention I pass gaseous silicon fluoride over the beryllium ore, in the total absence of any other reactant. The ore is best in a well-ground state, so as to insure intimate contact between it and the reactant SiF-1. gas. The reaction which occurs follows the formula:

2Be3Al2 (S103) e +3S iF4 6BeF2+2A1203 15SiO2 The optimum operating range for the reaction is between about 650 C. and about 850 C. Below 650 C. reaction is relatively slow; above 850 C, there is considerable danger of loss of BeFz by volatilization. I do not wish to be understood as limiting my invention to these temperature limitations; it is merely that the range indicated appears to represent the optimum condition.

Of course, this applies only to normal atmospheric pressures. Increase of pressure on the system may result in broadening the optimum band, since it tends to increase the, efficiency of operation as well as to speed up the reaction. Further, on the high side of the temperature range, pressure prevents volatilization of BeFz from the reaction mass. This increase in 'efficiency is to be expected; it is in perfect line with the law of mass action. However, the use of supernormal pressures is purely optional in my invention.

Following reaction, I normally leach the reaction mass with water, and thereby obtain a pure beryllium fluoride solution. An additional advantage of my invention here evinces itself. Whereas in all previous operations, using fluosilicate, and resulting in a double alkali beryllium fluoride solution, large amounts of water are required because of the comparatively low solubility of the double fluoride, my present invention yields an extremely soluble compound, BeFz, and much smaller volumes of water are needed for complete leaching out of the beryllium solubilized.

' From the BeFz solution, of course, further processing, of standard and known sort, will yield whatever compound may be desired. The advantage resulting from my invention here is that there is no alkali fluoride to interfere. No metal other than beryllium is to be found in the solution.

Having described my invention what I claim The process of solubilizing the beryllium content of beryllium containing ores including the step of contacting the ore with gaseous silicon fluoride in the presence of heat and the absence 20 of any other reactant, and leaching with water.

HARRY c. CLAFLIN.