Imágenes de páginas
PDF
EPUB

IV.

THE ANALYSIS OF URANIUM ORES

AND ALLOYS.

3

Pitchblende.-Pitchblende is the best-known ore of uranium; it is an impure U2O, containing from 40 to 95 per cent. of that oxide, but not always exactly in the proportion of UO,: 2UO3. The following elements have been found associated with it-sulphur, selenium, phosphorus, calcium, magnesium, aluminium, silicon, vanadium, manganese, arsenic, bismuth, antimony, tin, zinc, lead, iron, cobalt, nickel, copper, and silver. Many of these occur in such small amounts, that their presence is negligible so far as commercial analyses are concerned. Nitrogen and a number of rare earths are also to be found in uraninite from some localities (vide "On the Occurrence of Nitrogen in Uraninite, and on the Composition of Uraninite in General," Hillebrand, U.S. Geological Survey Bulletin, No. 78, and Chemical News, vol. lxiv., p. 221, etc.). The following outline method of analysis meets most

industrial demands. The particular form in which the results are expressed is a matter of expediency.

Digest 4 grams of the finely powdered ore with 20 c.c. (1°42) nitric acid, boil down to low bulk, and filter off the siliceous residue. This residue very rarely contains uranium; besides silica, alumina, and ferric oxide, it may contain small amounts of vanadium, tin, and antimony. A further analysis is made, or not, according to circumstances.

The soluble portion is evaporated with hydrochloric acid, diluted, any silica or silver chloride filtered off, and a current of sulphuretted hydrogen passed. The precipitated sulphides of antimony, arsenic, tin, lead, copper, and bismuth, are separated in the usual manner into two groups with sodium sulphide, and the separate determinations made as usual.1 A separate determination of the arsenic may be desirable, as a portion is probably volatilized from the main portion during the evaporation with hydrochloric acid.

The main filtrate is boiled to expel sulphuretted hydrogen, and nitric acid added to oxidize the

1 It is hardly feasible to give details of these separate determinations, but the general procedure followed by the author is dealt with, in relation to the analysis of white-metal alloys, in "The Analysis of Steelworks Materials" (Longmans, Green & Co.).

iron. It is then made up to 250 c.c.; 100 c.C., or some other volume well suited to the supposed amount of the impurities to be determined, is measured into a flask containing a few crystals of ammonium phosphate, then nearly neutralized, and poured into an excess of aqueous sodium carbonate. The mixture is raised to boiling, and at least as much ammonium chloride added as will destroy the free soda. After standing a few hours the precipitated phosphates are filtered off. Uranium is determined in a portion of the filtrate after acidifying and boiling free from CO2. The residue is dissolved in sulphuric acid. A portion of this acid solution is reduced with sulphur dioxide, and the iron titrated with permanganate, care being taken to make allowance for any vanadium which may be present. The vanadium may also be estimated, colorimetrically, on this portion if its amount is small compared with the alumina and ferric oxide. The remaining portion of the sulphuric acid solution is boiled with sodium phosphate and thiosulphate to obtain a precipitate from which alumina and titantic oxide may be determined; any uranium not having passed into the ammonic carbonate solution would be found here.

To another fraction (100 c.c.) of the main

solution, an excess of microcosmic salt is added, then an excess of ammonium acetate to ensure the complete precipitation of uranium, iron, aluminium, etc., from a strongly acetic acid solution. A fraction of this solution is filtered, most of its free acetic acid neutralized, and the lime precipitated as oxalate, and the magnesia subsequently as phosphate. There is generally not sufficient manganese in these ores to interfere with this procedure.

The remaining portion (50 c.c.) of the main filtrate is used for the determination of phosphoric acid via the molybdate precipitation; or a separate portion may be prepared by digesting the ore with nitric acid in case no appreciable amount of tin is present.

Separate determinations may be made of manganese by digesting 1 gram of the ore with nitric acid, oxidizing the solution with sodium bismuthate, and titrating the permanganate formed; of vanadium, by fusing 1 gram with sodium carbonate, extracting with water, and determining colorimetrically with hydrogen peroxide, or otherwise; of nickel, cobalt, and zinc, by precipitating uranium, aluminium, iron, etc, as phosphates from strongly acetic acid solutions, and applying the usual processes to the filtrate.

The sulphur and moisture are determined as in samples of pyrites.

Carnotite. This ore, which has attained commercial importance, is found chiefly in Western Colorado. It is a canary yellow compound associated with a fragile sandstone bearing some resemblance to Roscoelite. The analysis is comparatively easy.

Digest 3 grams of the ore with nitric acid, and filter. The ignited siliceous residue may be reported as such, or fused with sodium carbonate, evaporated with hydrochloric acid, etc. Uranium is usually entirely absent from the residue, but vanadium (V2O1) and barium sulphate are frequently present along with the usual constituents of silicates.

The filtrate is evaporated with hydrochloric acid, the trifling amount of silica collected, and a few drops of sulphuric acid added to precipitate any barium. If no barium is indicated, sulphuretted hydrogen is passed to separate copper, and, in rare cases, other metals of this group. The filtrate is boiled, oxidized with nitric acid, and made up to 300 c.c. From this point the analysis proceeds on similar lines to that of pitchblende. Vanadium is estimated on a separate portion by titration with ferrous sulphate or otherwise,

« AnteriorContinuar »