STUDIES IN THE METALLURGY OF BERYLLIUM BY OTTO ERSKINE HUNTLEY A. B. Hope College, 1918 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of MASTER OF ARTS IN CHEMISTRY IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS 1921 UNIVERSITY OF ILLINOIS THE GRADUATE SCHOOL may 3 1 1 92_Jr I HEREBY RECOMMEND THAT THE THESIS PREPARED UNDER MY SUPERVISION BY QITQ-JSRSKINE HUNTLEY ENTITLED STUDIES IH THE METALLURGY O ff BERYLLIU M BE ACCEPTED AS FULFILLING THIS PART OF THE REQUIREMENTS FOR Recommendation concurred in* Committee on Final Examination* ^Required for doctor’s degree but not for master’s 2 > T-- . 2 , Digitized by the Internet Archive in 2016 https://archive.org/details/studiesinmetalluOOhunt \^l\ sg TABLE OF CONTENTS ~ Page INTRODUCTION '0 I. Object 1 II. Occurrence 1 III. Historical 3 EXPERIMENTAL IV, Crude Material 5 V, Parsons’ Method of Extraction 5 VI. Preparation of Beryllium Chloride 7 VII, Electrolysis of Beryllium Chloride 7 VIII, Treatment of Beryllium- Amalgam 9 CONCLUSION IX, Summary 11 X . B ibl io graphy 1 3 V ACKNOWLEDGEMENT The writer takes this opportunity to thank Professor B. S. Hopkins, who suggested thi3 research, and directed the work, for the invaluable suggestions, advice, aid and encouragement. Thanks are also tendered to Dr. H. C. Kreniers for placing apparatus at the writer’s disposal, and for several valuable suggestions. . . . . . . INTRODUCTION - 1 - I. OBJECT « The object of this investigation was to study the metal- lurgy of beryllium with a view to the discovery of a method that would prove successful on the large scale. The plan suggested was to carry out the investigations on a small scale and if a successful method was discovered, to try it out on a large scale . The difficulty which is encountered in the extraction of beryllium from beryl has kept beryllium in the category of «rare« elements altho beryllium bearing minerals occur exten- sively. When a method is proposed by which beryllium can be prepared cheaply an adequate supply of the mineral will be (1,3) available , Beryllium possesses properties which make it useful for the manufacture of special electrical and other scientific instruments and apparatus) its alloys with copper are known to possess sonorous properties; its salts are useful in the synthetic preparation of pure gems. II. OCCURRENCE. Beryllium occurs chiefly as the silicate, beryl, Be 3 Alg(Si0 2 ) 6 , which contains 11-15$ BeO. Emerald and aqua marine are of this formula. Gadolinite, FeBe 2 Y 2 Si 2 Q| 0 or Be 2 F 3 (Y0) 2 (Si0 4 ) 2 , 5-11$ BeO: Leucophanite NaBeFCa(Si0 3 ) 2 , 10-13$ BeO; chrysoberyl, Be(A10 2 ) 2 , 19-30$ BeO; euclase, Be (AlQH) S10 a . 17-16$ BeO:, Bertrandite, Be 2 (3e0H) 2 Si 2 0 7 , - 2 - 40 -43$> BeO; and eudidymite, HHaBeSl 3 0 Q , 10-11$ BeO , are other (2,3) important minerals. Many other minerals contain beryllium. Beryl was the source of material for this investigation. III. HISTORICAL. In 1798, L, N. Vauquelin announced the discovery of a new earth, »la terre du Beril" . He made the discovery in attempt- ing to prove the identity of beryl and emerald, previously reported by Hauy, when he found that on boiling a solution of a precipitate previously supposed to be aluminium hydroxide in potassium hydroxide, that a portion of the precipitate was (4) thrown out of solution. The name proposed by Vauquelin was translated into the German as "Beryllerde" from whence came "Beryllium" . The editors of Annales de Chimie suggested the name "glucine" because of the sweet taste of the fulfate, Hut later investigators have objected to the name because of (5) the similarity to "glycine," and others because the sweet- ness, is not peculiar to beryllium compounds alone, but is ( 6 ) possessed by the yttrium earths. Following the discovery of the element, Vauquelin studied and reported the properties (7,8,9,10) (12), (11) of 3ome of its compounds. Berzelius, Gay-Lussac, and others also studied the known beryllium compounds, and prepared and reported others. (13) In 1826, Wohler obtained metallic beryllium as a dark gray powder by reducing sublimed beryllium chloride with potass- (14) ium in a platinum crucible. About the same time Bussy - 3 - prepared impure bervliium by the same method. ( 15 ) In 1842, Awdejew made the first determinations of the atomic weight of bervliium that have anv claim to accuracy. (16) Weeren (1854) published an extended and careful research on beryllium. He determined the atomic weight by the ratio Ee0:S0 3 in bervliium sulfate which he had carefully purified. ( 1 ?) In 1855, DeBray prepared metallic beryllium as a white metal, specific gravity 2.1, by the reduction of the chloride with sodium in a current of hydrogen. He describes the proper- (16) tie 3 of the metal and of various compounds. Charles A, Joy, published an extended bibliography of beryllium up to 1863. He also published results on various methods of decomposing beryl and methods of separating aluminium and bervliium. (19) Nilson and Petersson prepared beryllium by heating the chloride to bright redness with sodium in a sealed crucible. (30 T Kruss and Moraht reduced potassium beryllium fluoride with ( 21 ) sodium. Warren reports that he obtained beryllium by electrolyzing fused beryllium bromide. This has since been (3) declared a mistake by Parsons who states that the bromide is ( 22 ) not a conductor of electricity. Borchers electrolyzed (23) potassium ammonium beryllium fluoride. Lebeau obtained beryllium by electrolyzing sodium beryllium fluoride. A nickle crucible served as the cathode and a graphite anode was used. It ms only necessary to melt the mass with a Bunsen - 4 - burner and then the current (6-7 amps, 35 volts) was sufficient to keep it molten. This is a brief resume of the various methods used up to this time to prepare metallic beryllium. ■ . l EXPERIMENTAL ' r - 5 - IV. the crude material. The material used was beryl which is the only mineral of beryllium which contains enough beryllia to make its extract ion p ract ical . It is beryllium aluminium silicate, Be 3 Al 2 (Si0 3 ) 6 , and contains 11-15$ BeO . As impurities it contains iron, calcium, and magnesium. The beryl used came in lumps ranging in color from white to green with brown and blue streaks. It was prepared by crashing first with a sledge hammer, then thru a Elake jaw-crusher, where it was reduced to l/l 6 inch. The material was then rotated in a ball mill for several days until it passed thru a 100 -mesh sieve . (24) Analysis . Composition of Beryl : BeO 10 . $ A1 2 O 3 M CD • Si0 2 65.5 $ F e 2 0 3 1.7 io CaO,MgO ,H 2 0 3.8 $ (by difference) 100 . $ V. PARSON'S METHOD OF EXTRACTION . One hundred gram 3 of beryl were fused with an equal weight of sodium hydroxide in a nickel crucible, in a Case \ * . - 6 - furnace. The fused mass ms "broken up and just covered with water. Concentrated sulfuric acid was added in slight excess. The product of several fusions was treated in this my and then the whole mass was heated until fumes of sulfuric acid were given off. The residue now had the appearance of fine white powder. This residue was treated with hot water when the sulfates of "beryllium, aluminium, iron and sodium passed into solution and were separated from the 3ilica by filtration. The mother liquors were evaporated until the alums began to crystallize out, allowed to stand, and filtered from the alum. Most of the aluminium was separated in this way. The liquor from the alum wa3 treated with nitric acid to convert all the iron present to the ferric condition, neutralized with ammonium hydroxide, and sufficient sodium acid carbonate added, in crystals, to saturate. The solution was warmed(50° C . ) and shaken frequently for 24 hours. Most of the beryllium passed into solution almost free from aluminium and iron. Ey re- dissolving and treating the residue after filtration practi- cally all of the beryllium wa3 obtained in the bicarbonate solution. Ammonium sulfide was added to thi3 solution to remove any iron that might have dissolved, and the solution filtered. The solution was diluted to ten times its volume and 3team passed thru the liquid to the boil ing point , which caused the beryllium to be precipitated as a. fine, granular - 7 - basic carbonate, xBeC0 3 , yEe(OH) 2 . This is obtained quite (26) pure by filtering and washing. The yield was about 80% of the theoretical. / r 2 p cA VI. PREPARATION QP BERYLLIUM CHLORIDE * * The basic carbonate of beryllium as prepared above was dissolved in hydrochloric acid and evaporated to dryness. The residud was taken up in water and it was found that a few drops of concentrated hydrochloric acid were usually necessary (5) to bring it into solution, due to the formation, on evaporation (25) of basic mixtures of various degrees of hydration. VII. ELECTROLYSIS OF BERYLLIUM CHLORIDE. Several types of cell were tried but the cell finally used was of the nature of the accompanying drawing. The first ceil was a 600cc. beaker equipped with a mercury cathode and a graphite anode. Connection was made to the mercury through a platinum wire sealed in a short length of glass tubing. Direct currents of 110 volts, and of 12 volts were tried. With the 110 volt circuit a bank of 12 lamps and a carbon rheostat resistance were inserted and varying resistances were tried. The graphite anode continually wore awa 2 / under the action of the chlorine and settled on the amalgam from which it was separated by rolling in cloth. The crude cell did not give very satisfactory results and the cell as finally used was constructed *S*/u.t < on . - 8 - as follows:- The bottom of a 300 cc. bottle was removed and a rubber stopper substituted thru which were passed two lengths of glass tubing. One piece of tubing ended at the surface of the stopper and was used to draw off the amalgam and to introduce fresh mercury. This was accomplished by means of a Y connection (see figure N to which was attached two separatory funnels, one for the amalgam, the other for the mercury. The other piece of tubing terminated about 1/2 inch higher and was used to remove the electrolzed solution and to introduce fresh aqueous beryllium chloride. The process was made more or less continuous, although not automatic, by connecting this tube, likewise, to two separatory funnels. Connection was made to the mercury cathode by means of a Piece of glass tubing in which was sealed a short platinum wire . The tubing was filled with mercury to connect the wires from the source of current. The anode was a platinum wire similarly constructed , and distant from the cathode about 3 centimeters. Experiments proceeded with varying resistances on both the 110 and 10 volt circuit. As finally tried the 10 volt circuit w as used wi th a sufficient resistance to cause a currenl of 1-2 amperes to flow thru the cell. The electrolysis was not at all dependable. Concentration and resistance seemed to play an important part. At first the amalgam seemed to - 9 - form readily, but after a tire equilibrium was reached and no more amalgam formed. At other times the beryllium salt hydrolyzed and had to be removed and converted to the chloride. With careful efforts to have the conditions the same the most unexpected reactions occurred and had to be overcome. Both ( 3 ) aqueous and alcoholic solutions of beryllium chloride were tried in an effort to easily obtain the amalgam but with very little difference in the result. The electrolysis is still unsatisfactory and merits continued research. VIII, TREATMENT O 7 ? THE BERYLLIUM AMALGAM • An effort was made to obtain metallic beryllium from the amalgam. The first plan suggested was to heat the amalgam in a stream of nitrogen. The beryllium amalgam was placed in a combustion tube in a furnace and the tube connected with a vacuum pump and a source of nitrogen. The nitrogen was prepared by drawing air slowly through ammonium hydroxide and then through a combustion tube containing copper oxide to decompose the ammonia into nitrogen and water, then thru another combustion tube containing copper gauze to remove the oxygen from the air. The gas was then drawn thru a long series of drying towers containing concentrated sulfuric acid; stick potassium hydroxide; and phosphorus pentoxide respectively. - 10 r The product obtained was yellow to gray in color and was probably a mixture of the oxide and occluded mercury. A supposition that the nitride might have formed seemed unlikely since the material did not decompose in boiling water, but dissolved in hydrochloric acid. It was next decided to try to obtain beryllium from the amalgam in vacuum. The first trials were unsuccessful because the vacuum was not high enough. The metal oxidized and the product was white beryllium bxide. Other attemps were made using a mercury pump by which vacuum of 0.001 mm. of mercury could be obtained and a gray powder was the result. After several such runs, the product which contained a little mercury was dissolved in hydrochloric acid, separated from the mercury, and precipitated as the hydroxide with ammonium hydroxide. This was filtered, washed, and dissolved in potassium hydroxide. Upon boiling beryllium hydroxide was again precipitated. CONCLUSION? 7 11 > IX. SUMMARY • Eery Ilium basic carbonate was obtained from beryl by the method suggested by Parsons. No attempt was made to get large yields but rather to obtain purity of product. The basic carbonate was converted to the chloride and the aqueous solution electrolyzed with a mercury cathode. This gave a beryllium amalgam. The amalgam was heated in vacuum to obtain metallic beryllium. The investigation shows that beryllium amalgam may be formed by electrolyzing aqueous beryllium chloride. Metallic beryllium may be obtained by heating the amalgam in vacuum. A new method of obtaining metallic beryllium was discovered. * ' ■ • ' ■ ‘ ' :j . * cf l If ] - 13 X. BIBLIOGRAPHY 1. Negru, J. S. Chem. and Met. 21, 353 (1919) 2. Browning, P. E. Introduction to the Rarer Elements. 3. Parsons, C. L. The Chemistry and Literature of Beryllium. 4. Van quel in, L. N. Ann, de dhim. _26, 155-170 (1796) 5. Link, H. F. Allg. J.Chem. 3, 603 (1799) 6. Klaproth, M. H. Beitrage zur Kentneis der Mineralkorper 78 (1600) 7. Vauquelin, L. N. Ann.de chim. 26,170-177 (1798) 8. Vauquelin, L. N. Ann. de chim. 26,259 (1798) 9. Vauquelin, L. N. Allg.J. Chem. 1, 590-596 (1798) 10* Vauquelin, L. N. Ann. de chim. J30 , 82 (17 99) lit Gay Lussac, Ann. de chim. 7j3,273 (1811) 12. Berzelius, J.J. Ann. id Phys. 1, 22, 196 (1823) 13. Wohler, F. Berzelius J sb . 9, 96 (1628) 14. Eussy, Berzelius J sh . 9, 997, ’1838) 15. Awdejew, V, Berzelius J 3b. J33, 112, 185 (1842) 16. Weeren, Julius, Ann. Chim. 92, 262 (1854) 17. DeBray, Henri, J. Chem. Soc. 8, 242 (1855) 18. Joy, C. A. ahem. News 8, 183-19? (1863) 19. Nilson, L. F. and Petersson, Otto, Chem. News 42, 297 (1880) 20. Kruss, G. and Moraht, H. Berichte, _33, 730 (1890) 21. Warren, H. N. Chem. News 72, 310 (189 5) r 13 22. Borchlors, W. Z, Electrochem. .3, 40 (169 5) 23. Lebeau, Paul, Comp. rend. 136 . 744 (1696) 24. Cooper, I.E. Thesis. A Study of the Methods of Extracting Beryllia from Beryl. U. of I., (1920) 25. Parsons, C, L. J. Amer. Chem . Soc . 26, 721 (1904) 26. Parsons, C. L. J. Amer. Chem. Soc. _28, 1589 (1906)