US2789897A - Magnesium reduction process for production of uranium - Google Patents

Description

quantity.

Un t 5W8 P w-" MAGNESIUM REDUCTION PROCESS FOR PRODUCTION OF URANIUM Charles B. Sawyer and Bengt R. F. Kjellgren, Cleveland,

Ohio, assignors to the United States of America as reprgesented by the United States Atomic Energy Commiss on No Drawing. Application July 19, 1946, Serial No. 684,794

1 Claim. (Cl. 7 5-841) The present invention relates to the production of uranium metal in massive and substantially pure form.

Heretofore uranium has been produced on an experimental basis by electrolysis of a fused salt, such as KUFs.

The cost of metal so produced was expensive, amounting suggested by the prior art are several attempts to obtain the metal by reacting a salt thereof with an active metal. For example, in the British patent to Westinghouse No. 230,865 (1925) KUFs is reduced by means of magnesium metal, the process being carried out under a vacuum. Obviously such a method does not solve the problem of a substitute for KUFs, nor is it convenient to operate under a vacuum if a method can be evolved for carrying on the reaction in the open air.

Prior to our invention it was also known that uranium in hard brittle form could be obtained by reducing uranium chloride with calcium in a high vacuum. However, the metal produced in this manner was not of suflicient purity for certain uses of uranium, and metallic calcium is expensive and not readily obtainable in It is the object of this invention to provide a new, economical and easy method for producing metallic uranium in massive and substantially pure form.

The object of this invention is obtained by reacting uranium tetrafluoride and magnesium metal and supplying sufficient heat in addition to the heat of reaction to permit the uranium metal to coalesce into a coherent mass. By use of our method, it is not necessary to carry on the reaction under a vacuum, and the employment of KUF5 and calcium is entirely eliminated.

We have found that magnesium metalwill reduce uranium tetrafiuoride according to the following equation:

By substituting the atomic weights for the several items in Equation 1, we obtain:

Patented Apr. 1957 2 Dividing Equation 2 by 48.6, we obtain the relative weights of materials required for the reaction:

From 3 it is apparent that in a theoretically perfect reaction, 1 gram of metallic magnesium and 6.45 grams of uranium tetrafluoride should yield 4.9 grams of metallic uranium and 2.56 grams of magnesium fluoride.

Upon the assumption that about 25% of the uranium tetrafiuoride will form a low melting flux and will remain in the slag, we prefer to use more than the theoretical amount of uranium tetrafiuoride required. In practice we have found that a ratio of 7.30 grams of uranium tetrafluoride to 1 gram of magnesium gives satisfactory results.

'When dealing with materials in amounts in the neighborhood of several kilograms, we have found that the heat of reaction between the uranium tetrafluoride and the magnesium is insufficient to make the resulting uranium metal fluid enough to separate from the slag and form a pool in the bottom of the crucible used. We therefore heat the reaction crucible before charging it with the material for the reaction. We continue to heat themixture while the reaction is in progress and for a sufiiciently long time thereafter to assure separation of molten uranium from the slag.

In practicing our invention we have found that magnesium vapor formed in the reaction effectively protects the materials from the atmosphere, and that the use of a vacuum or a protective atmosphere may be eliminated. It is to be understood, however, that our method may be performed under a vacuum or in aninert atmosphere, such as argon, if desired.

An electrically heated graphite crucible has been found satisfactory for our purposes. This is heated while empty to approximately 1000 vC. to 1200" C. A small charge of magnesium is then introduced, after which small charges of uranium tetrafluoiide and magnesium are alternately introduced into the crucible until the total charge has been placed therein. After the charging operation is complete, a cover is placed over the charging hole. The initial efiect of charging the crucible is a drop in temperature, since the materials absorb heat until the reaction temperature is attained. Once the reaction be tween the uranium tetrafiuoride and magnesium begins, however, the temperature rises sharply. We continue heating the crucible for about an hour after the reaction has taken place to permit the metal formed in the reaction to coalesce into a pool in the bottom of the crucible.

Since the melting point of uranium isabout 1125 C., the charge should be heated to at least this temperature,

and preferably should be heated higher in order to assure temperatures above 1400 C.

The crucible was electrically heated to about 1220 C., v

The following data Time, KW UF Mg, Temp, Remarks p. m. gr. gr. C.

5. 5 1,050 Empty crucible. 6. 5 1, 220 Do. 4. 5 190 about 800 4. 5 1, 390 about 800 No reaction observed. 4. 5 180 about 800 4. 5 1, 310 about 800 Mg burned some. and

U sparks observed. 4. 5 150 about 800 4. 5 1,090 about 800 Quiet or no reaction. 4. 5 130 about 800 Do. '4. 5 050 about 800: Do. 4. 5 120 about 800 Do. 4. 5 880 about-800 Do. 4. 5 120 about'800 Do. 4. 5 about 800 Do. 4. 5 about 800 Do. 4. 5 about 800 Do. 4. 5 Mgburncd violently.

.' Added 110 gr. U1", powder to protect metal.

Metal feels sandy. Surface fumes. Metal si-ill sandy. Added another 110 gr. of UF Shut down.

In the described experiment a total of 1000 grams of magnesium'and 7300 grams ofuranium tetrafiuoride were used in the charge. An additional 220 grams of uranium tetrafiuoride were added for fiuxing.

When opened, the crucible was found to contain a regulus of uranium weighing 2070 grams and less than .05 percent of magnesium. This amounted to 42.3 percent of the theoretical yield of'4900 grams of metal.

Example 2.An experiment was conducted in a manner similar to the'first example, except for the addition of an argon atmosphere. A graphite tube in the crucible cover adjacent the charging hole was used for introducing the gas into the crucible. The argon atmosphere was maintained both while the reaction was taking place and during the cooling period. By this method 6055 grams of uranium tetrafiuoride yielded about 1600 grams of metal.

It will be understood that we intend to include variations and modifications of the invention and that the preceding examples are illustrations only and are not to be construed as limitations upon our invention, the scope of which is defined in the appended claim.

We claim:

The method of producing substantially pure massive metallic uranium that comprises the heating of a charge of uranium tetrafiuoride and magnesium to cause reaction between them with reduc tion of uranium fluoride to uranium metal and formation of magnesium fluoride, continuing heating afterreaction and :maintaining the reacted charge at a temperature above at least 1125 C.

- for sufficient time for sep aration of 1 liquid uranium and slag into" two layers, said operations being carried out at substantially "atmospheric pressure.

References=Cited=-inthc file of this, patent UNITED *STATES 'PATENTS 1,415,516 Bridge ee May 9, 1922 1,437,984 Marden Dec. 5, 1922 1,602,542 Marden Oct. 12, 1926 1,659,209 'Marden Feb. 14, 1928 2,446,780 Newton Aug. 10, 1948 FOREIGN PATENTS 230,865 Great'Britain Dec. 10, 1925 OTHER REFERENCES Spcdding'et aL: USAfE/C. Publication No. CT-2712 date June 26,1945, Declas. Jan. 4, 1956 pages 1-12, entire publication 67 pages.

Goggin et at; Industrial and Engineering'Chemistry, vol. 18 (Feb. 1926),'pp. 11 4-116. i