REFERENCES.

1 J. J. Berzelius, Svenska Akad. Handl., 1, 1829; Pogg. Ann., 16. 385, 1829; J. J. Chydenius, ib., 119. 43, 1863; Kemisk Undersökning af Thorjord och Thorsalter, Helsingfors, 1861; T. Sollmann and E. D. Brown, Amer. Journ. Physiol., 18. 427, 1907; P. T. Cleve, Bull. Soc. Chim., (2), 21. 115, 1874; C. R. Böhm, Chem. Ind., 29. 460, 1906; A. C. Neish and J. W. Burns, Journ. Canadian Inst. Chem., 5. 69, 1921; J. O. Perrine, Phys. Rev., (2), 22. 48, 1923; H. Freundlich and M. Wreschner, Zeit. phys. Chem., 106. 366, 1923; V. Cuttica and L. Bonamici, Gazz. Chim. Ital., 53. ii, 761, 1923; V. Cuttica and A. Tocchi, ib., 54. i, 628, 1924; A. C. Neish, Journ. Amer. Chem. Soc., 26. 781, 1904; Chem. News, 90. 196, 1904; B. Brauner, Journ. Chem. Soc., 73. 984, 1898; A. T. Cameron and W. Ramsay, ib., 91. 1606, 1907; O. Fuhse, Zeit. angew. Chem., 10. 116, 1897; J. F. Bahr, Pogg. Ann., 119. 578, 1863; I. Koppel and H. Holtkamp, Zeit. anorg. Chem., 67. 290, 1910; A. Kolb, ib., 60. 124, 1908; G. Krüss and W. Palmaer, ib., 14. 366, 1897; R. J. Meyer and R. Jacoby, ib., 27. 359, 1901; Ber., 33. 2135, 1900; A. C. Brown, Journ. Chem. Soc., 121. 1736, 1922; R. Jacoby, Die Doppelnitrate des vierwertigen Ceriums und des Thoriums, Berlin, 1901; R. J. Meyer and A. Anschütz, Ber., 40. 2639, 1907; R. J. Meyer and A. Gumperz, ib., 38. 818, 1905; W. Biltz and F. Zimmermann, ib., 40. 4983, 1907; A. Sachs, Zeit. Kryst., 34. 163, 1901; A. Kolb, G. Melzer, A. Merckle, and C. Teufel, Zeit. anorg. Chem., 60. 123, 1908; A. Kolb, ib., 83. 143, 1913; F. Halla, ib., 79. 260, 1912; W. Crookes, Proc. Roy. Soc., 66. 409, 1900; F. Soddy, Phil. Mag., (6), 16. 513, 1908; G. P. Drossbach, Journ. Gasbeleucht., 38. 482, 1895; 0. Angelucci, Atti Accad. Lincei, (5), 16. ii, 196, 1907; B. Szilard, Journ. Chim. Phys., 5. 636, 1907; G. Wyrouboff and A. Verneuil, Ann. Chim. Phys., (8), 6. 492, 1905; Compt. Rend., 127. 865, 1898; H. A. McTaggart, Phil. Mag., (6), 44. 386, 1922; H. R. Robinson, ib., (6), 50. 241, 1925; W. Ramsay and F. L. Usher, Ber., 42. 2930, 1909; W. Ramsay, Journ. Chem. Soc., 95. 624, 1909.

§ 15. Thorium Phosphates

J. J. Berzelius,1 P. T. Cleve, and T. Sollmann and E. D. Brown prepared tetrahydrated thorium orthophosphate, Th3(PO4) 4.4H2O, as a white gelatinous mass by adding a soln. of sodium hydrophosphate to one of a thorium salt. The product, said J. J. Berzelius, is insoluble in water and in aq. phosphoric acid; and, according to P. T. Cleve, in acetic and mineral acids. T. Sollmann and E. D. Brown added that it is insoluble in dil. hydrochloric, acetic, citric, and tartaric acids and in soln. of sodium citrate and tartrate. Carbonates, citrates, and oxalates, but not the tartrates, hinder the precipitation of the phosphate. The citrate mixture is again precipitated by boiling, long standing, or adding hydrochloric acid, but not by citric, or tartaric acid, or sodium hydroxide. The salt dissolves in 30 per cent. hydrochloric acid, and the soln. is not precipitated by dilution. The salt is also soluble in a soln. of ammonium citrate. According to A. Colani, if 2-3 grms. of anhydrous thorium metaphosphate, or an alkali phosphate, be heated with 40 grms. of anhydrous thorium chloride, in a tube through which dry carbon dioxide is passed, some of the thorium chloride is volatilized; and the white crystalline powder which remains after washing with water, is a thorium chlorophosphate, ThCl4. Th3(PO4) 4, or ThCl(PO4). It is insoluble in water, and, with the exception of boiling sulphuric acid, it is scarcely attacked by dil. or conc. acids, but it is easily broken down by fused alkali carbonates. The corresponding thorium bromophosphate, Th(PO4)Br, is obtained in a similar manner. Where the thorium metaphosphate was similarly treated with anhydrous calcium chloride, small, optically inactive crystals of calcium thorium orthophosphate, CaTh(PO4)2, were formed. With strontium chloride, optically active crystals of strontium thorium orthophosphate, SrTh(PO4)2, were obtained. The corresponding barium thorium orthophosphate was an amorphous powder.

K. A. Wallroth made sodium thorium orthophosphate, NaTh2(PO4)3, by saturating fused microcosmic salt with thoria, and treating the cold mass with hydrochloric acid. L. Troost and L. Ouvrard have stated that normal sodium phosphate does not form a double salt with thorium phosphate. The prismatic crystals are insoluble in all acids. L. Troost and L. Ouvrard made potassium thorium orthophosphate, KTh2(PO4)3, by melting potassium metaphosphate, KPO3, and saturating it with thorium oxide, phosphate, or anhydrous chloride. The slowly cooled mass was washed with acidulated water. The microcrystalline

prisms act on polarized light; their sp. gr. is 5·75 at 12°; they are insoluble in hydrochloric or nitric acid, and in aqua regia. If the potassium pyrophosphate be used in place of the metaphosphate in the preceding process, dipotassium thorium orthophosphate, K,Th(PO4)2, is formed in microcrystalline octahedra. The sp. gr. is 4-688 at 7°; and the salt is insoluble in water but soluble in nitric acid. The corresponding disodium thorium orthophosphate, Na, Th(PO4)2, was made in a similar manner; the crystals probably belong to the triclinic system. L. Troost and L. Ouvrard reported hexagonal plates of the potassium thorium phosphate, K12Th3(PO4)8, 6K2O.3Th02.4P2O5, of sp. gr. 3.98 at 12°; and a white powder of NagTh(PO4)3, or 5Na2O.2ThO2.3P2O5, of sp. gr. 3.843 at 7°. G. Saring prepared potassium calcium phosphatothorate, Ca3(PO4)2(CaO)2(K2O)2ThO2, analogous with the corresponding phosphatosilicate.

P. T. Cleve, and C. Volck made monohydrated thorium hydrophosphate, Th(HPO4)2.H2O, as a white gelatinous mass, by mixing soln. of orthophosphoric acid, and thorium chloride. The washed precipitate was dried to constant weight at 60°-100°. P. T. Cleve said that the salt loses water when calcined; and C. Volck found it to be soluble in boiling water. P. T. Cleve made thorium pyrophosphate, ThP2O7.2H2O, by adding a soln. of pyrophosphoric acid to one of thorium chloride; or a soln. of normal sodium pyrophosphate to one of thorium nitrate. The washed precipitate was dried at 100°. It is soluble in soln. of pyrophosphoric acid or sodium pyrophosphate. The last-named soln. is precipitated neither by ammonia nor by oxalic acid. It will be observed that the salt is isomeric with monohydrated thorium hydrophosphate. P. T. Cleve made sodium thorium pyrophosphate, Na4Th(P2O7)2, as a white crystalline powder by dissolving thorium pyrophosphate in a boiling sat. soln. of sodium pyrophosphate, and allowing the mixture to stand some days. The soln. is precipitated neither by ammonia nor by oxalic acid.

K. R. Johnsson made thorium metaphosphate, Th(PO3)4, by R. Maddrell's process in which phosphoric acid at a red-heat is treated with thorium sulphate, and heated until the sulphur trioxide is expelled. The excess of phosphoric acid is removed by decantation with water. L. Troost and L. Ouvrard also made it by the action of an excess of metaphosphoric acid on anhydrous thorium chloride or bromide; and they gave 4.08 for the sp. gr. of the rhombic plates at 16.4°; K. R. Johnsson gave 3.922 for the sp. gr. of the microscopic plates. A. Colani found the salt to be isomorphous with uranium metaphosphate, U(PO3)4, with which it forms mixed crystals. L. Troost and L. Ouvrard said that the salt is insoluble in water; they also made sodium thorium metaphosphate in triclinic prisms of sp. gr. 5-62 at 16°; the crystals act on polarized light; and are soluble in nitric and hydrochloric acid.

The Lindsay Light Co. prepared thorium sulphatometaphosphate, Th(PO3)2SO4, in the form of white, acicular crystals, by dissolving 120 grms. of octohydrated thorium sulphate, in 50 c.c. of 80 per cent. phosphoric acid, and heating for 10 hrs. at 280°. Water and sulphuric acid fumes are evolved and the product forms a nearly solid, crystalline mass, insoluble in water or dil. acids. The same substance is also obtained by heating thorium phosphate with an equal weight of sulphuric acid, preferably in presence of 20-40 per cent. of phosphoric acid, for 10 hrs. at a temp. above 260°.

REFERENCES.

1 J. J. Berzelius, Svenska Akad. Handl., 1, 1829; Pogg. Ann., 16. 385, 1829; P. T. Cleve, Bull. Soc. Chim., (2), 21. 115, 1874; T. Sollmann and E. D. Brown, Amer. Journ. Physiol., 18. 427, 1907; C. Volck, Zeit. anorg. Chem., 6. 163, 1894; Lindsay Light Co., U.S. Pat. No. 1323735, 1920; Brit. Pat. No. 156892, 1920; A. Colani, Compt. Rend., 149. 207, 1909; Ann. Chim. Phys., (8), 12. 59, 1907; L. Troost and L. Ouvrard, ib., (6), 17. 229, 1889; Compt. Rend., 101. 211, 1885; 102. 1423, 1886; 105. 30, 1887; K. R. Johnsson, Ber., 22. 979, 1889; K. A. Wallroth, Efvers. Akad. Förh., 40. 3, 1883; Bull. Soc. Chim., (2), 39. 321, 1883; R. Maddrell, Mem. Chem. Soc., 3. 273, 1848; Phil. Mag., (3), 30. 322, 1847; G. Saring, Versuche über den Aufschluss von Phosphaten durch Kieselsäure bei hohen Temperaturen, Dresden, 1906.

CHAPTER XLV

GERMANIUM

§ 1. The Discovery and Occurrence of Germanium

K. G. A. VON WEISSENBACH,1 and A. Breithaupt described a mineral, obtained from St. Michaelis, near Freiberg (Saxony), which was called Plusinglanz-from λoúσios, rich-in allusion to the lustre. An earth from Himmelsfürst in the same district was described by E. W. Neubert, and A. Weisbach, and named argyrodite from ȧpyvpádns, argentiferous-in allusion to its composition. A. Weisbach reported an analysis indicating the presence of a small proportion of mercury. C. Winkler obtained 74-72 per cent. Ag; 17.13, S; 0·66, FeO; 0-22, ZnO; and 0-31, Hg. The result was about 7 per cent. too low. He attributed the deficiency to the presence of an unknown element, precipitated as sulphide in the hydrogen sulphide group, and dem Antimon in mancher Beziehung ähnlichen, but yet quite different. The new element was called germanium from the Latin name Germania.

On account of the likeness of argyrodite to antimony glance (stibnite), C. Winkler at first assumed the new element must be D. I. Mendeléeff's unknown eka-antimony, occupying a place between antimony and bismuth in the periodic table. When C. Winkler had worked over the properties of germanium it was found that it did not fit well into the places assigned in the periodic table to ekaantimony and eka-cadmium, but it did fit well into the place reserved for ekasilicon. This is illustrated by Table IV, 1. 6, 4. G. A. Quesneville proposed to ́use D. I. Mendeléeff's term eka-silicon in place of germanium, but this was not adopted.

The analyses of C. Winkler, S. L. Penfield, V. M. Goldschmidt, F. Kolbeck, A. Frenzel, and G. T. Prior and L. J. Spencer show that different samples of argyrodite contain from 4.99 to 7.05 per cent. of germanium; and that the mineral is best regarded as 4Ag2S.GeS2-C. Winkler first gave 3Ag2S.GeS2-i.e. silver sulphogermanate, AggGeS. T. Kolbeck, and A. Frenzel showed that the mineral Plusinglanz is a variety of argyrodite. A. Weisbach first described argyrodite as belonging to the monoclinic system. S. L. Penfield obtained a similar mineral from Potosi, Bolivia, but furnishing cubic crystals, thence he inferred that argyrodite is dimorphous, and he assigned the name canfieldite-after F. A. Canfield-to the cubic form. A. Weisbach subsequently showed that the mineral from Freiberg is cubic, and tetrahedral. S. L. Penfield described a mineral from La Paz, Bolivia, which was previously thought to be argyrodite, but which was shown to be a kind of stanniferous argyrodite with Ge: Sn=5: 12, and he transferred the name canfieldite from the Bolivian argyrodite to the isomorphous stanniferous form. As emphasized by A. W. Stelzner, tin and germanium belong to the same chemical group, and are isomorphous with one another, and that silver sulphostannate isomorphous with argyrodite is to be anticipated. Canfieldite, said S. L. Penfield, may therefore be regarded as silver sulphostannate, AggSnSe, which occurs in Bolivia admixed with the isomorphous sulphogermanate, furnishing Agg(Ge, Sn)Sg. The Bolivian mineral has about 1·83 per cent. Ge, and 6-10 per cent. Sn. The deposits of argyrodite have been described by A. W. Stelzner, A. Frenzel, R. Peele, and

H. Reck. V. M. Goldschmidt emphasized the analogy between the formula of fahlerz, 3R,S.RR'S, of sp. gr. 4.921, and mol. vol. 165, and argyrodite 3R2S.R11⁄2RTMS, of sp. gr. 6-266 and mol. vol. 180. The sp. gr. is 6·1-6-3; and the hardness 2-3. J. L. C. Schröder van der Kalk discussed the colour of argyrodite; and A. de Gramont, the spark spectrum. G. T. Prior and L. J. Spencer showed that A. Damour's brongniardite is nothing but a variety of argentiferous argyrodite. The dark reddish-grey mineral germanite obtained by O. Pufahl from Tsumeb, S.W. Africa, has a composition approximating Cu5 (Cu,Fe) AsGeS12. Analyses were reported by H. Schneiderhöhn, F. W. Kriesel, W. Keil, and E. Thomson. The latter represented it by the formula 10Cu2S.4GeS2. As2S3. J. Lunt made a spectroscopic analysis of germanite. The black mineral, ultrabasite, reported by V. Rosicky and J. Sterba-Böhm from the Himmelsfürst mine, Freiberg, has the composition of a silver germanium lead sulphantimonite, 11Ag2S.28PbS.3GeS2.2Sb2S3, and P. Groth and K. Mieleitner suggest that it is a mixture of lead germanium sulphide, 3GePb2S4, with 2Pb3(SbS3)2, 16PbS, and 11Ag2S. The axial ratios of the rhombic or pseudotetragonal crystals are a:b: c=0.988: 1:0.462, when those of teatlite are a:b: c =0.93: 1:1·31. The hardness of ultrabasite is 5, and sp. gr. 6-026. It decrepitates and decomposes when heated; and is slowly attacked by hydrochloric and nitric acids.

Germanium is a scarce element. J. H. L. Vogt 2 estimated that the igneous rocks in the earth's crust contain n×10-12 per cent. of this element; and F. W. Clarke and H. S. Washington made a similar estimate. According to H. A. Rowland, lines in the solar spectrum correspond with those of germanium, and indicate the presence of that element in the sun. C. Winkler 3 said that germanium although sparse may be really more widely distributed than it is at present assumed to be because, owing to the want of a delicate and sharp distinguishing test, its presence may easily be overlooked. G. Urbain examined the ultra-violet spectrum of 64 samples of zinc blende, and he found germanium to be present in 38 samples, and appreciable quantities were found in five samples from Webb City, Missouri; Stolberg, Aix-la-Chapelle; Turkey in Europe; Raible, Corinth; and Mexico. A. del Campo y Cerdan also examined 68 zinc blendes and found germanium to be present in 50 of the samples-vide indium, 5. 35, 1. G. H. Buchanan also found germanium present in many zinc blendes, and J. H. Müller, in the smithsonite ore of Kentucky, and in the mine-water and ore washings. W. F. Hillebrand and J. A. Scheerer found it in zinc blendes and carbonate ores from Missouri, Idaho, Colorado, Nevada, and Utah.

The mineral franckeite, 5PbS.Sb2S5.2SnS2, described by A. Stelzner was found to contain about 0.10 per cent. of germanium. G. Krüss found up to about 0.1 per cent. of germanium in euxenite; K. von Chrustschoff, in harmony with D. I. Mendeléeff's suggestion, found this element to be present in many tantalum and columbium minerals; up to 1.5 per cent. in samarskite; 0.01 per cent. in tantalite; 0-03 per cent. in columbite and fergusonite; and traces in gadolinite. G. Krüss and L. F. Nilson reported germanium in fergusonite from Arendal. G. Lincio failed to find any trace of germanium in samarskite; and L. M. Dennis and J. Papish found none in American samarskite, and they suggested that K. von Chrustschoff's methods were unsatisfactory. A. Hadding found germanium in cassiterite from Finbo, and Mamacka, but none in cassiterite from Finland, and Bohemia. G. Neumann also reported germanium to be present in tin compounds. G. H. Buchanan found up to about 0-25 per cent. of germanium in zinc oxide, and spelter prepared from Missouri and Wisconsin zinc blendes. C. Winkler found no germanium in the flue-dust from the roasting chambers of the ores from the Freiberg mine where argyrodite occurs. J. Bardet reported germanium to be present in the mineral waters of Vichy to the extent of one part in 40,000,000 parts of water. E. Cornec detected traces of germanium in the ashes of the marine plant Laminaria.

REFERENCES.

1 A. Breithaupt, Vollständiges Charakteristik der Mineralsystems, Dresden, 277, 1832; K. G. A. von Weissenbach, Journ. tech. okon. Chem., 10. 205, 1831; Jahrb. Berg. Hütt. Sachsen, 226, 1831; E. W. Neubert, ib., 84, 1886; A. Weisbach, ib., 89, 163, 1886; Chem. News, 53. 257, 1886; Neues Jahrb. Min., ii, 67, 1886; i, 98, 1894; C. Winkler, Journ. prakt. Chem., (2), 34. 188, 1886; (2), 36. 177, 1887; Ber., 19. 210, 1886; 30. 15, 1897; 32. 307, 1899; 39. 4528, 1906; Chem. News, 53. 127, 1886; 54. 136, 1886; V. M. Goldschmidt, Zeit. Kryst., 45. 552, 1908; A. Frenzel, Jahrb. Berg. Hütt. Sachsen, 27, 61, 1900; Tschermak's Mitt., (2), 19. 244, 1909; Neues Jahrb. Min., ii, 125, 1893; S. L. Penfield, Amer. Journ. Science, (3), 46. 107, 1893; (3), 47. 451, 1894; F. Kolbeck, Centr. Min., 331, 1908; J. L. C. Schröder van der Kalk, ib., 79, 1901; A. de Gramont, Bull. Soc. Min., 18. 241, 1895; L. Moser, Ester. Chem. Ztg., 26. 67, 1923; F. W. Kriesel, Chem. Ztg., 48. 961, 1924; Metall. Erz, 20. 257, 1923; H. Schneiderhöhn, ib., 17. 364, 1920; O. Pufahl, ib., 19. 324, 1922; E. Thomson, Univ. Toronto Geol. Studies, 17, 1924; Amer. Min., 9. 66, 1924; J. Lunt, South African Journ. Science, 20. 157, 1923; D. I. Mendeléeff, Liebig's Ann. Suppl., 8. 133, 1871; A. W. Stelzner, Neues Jahrb. Min., ii, 114, 1893; Bol. Soc. Min. Santiago, 10. 62, 1893; Zeit. deut. geol. Ges., 49. 140, 1910; G. T. Prior and L. J. Spencer, Min. Mag., 12. 5, 1898; H. Reck, Petermann's Geogr. Mitt., 247, 1867; R. Peele, Eng. Min. Journ., 57. 78, 1894; G. A. Quesneville, Chem. News, 54. 49, 1886; A. Damour, Ann. Mines, (4), 16. 227, 1849; W. F. Hillebrand and J. A. Scheerer, Journ. Ind. Eng. Chem., 8. 225, 1916; V. Rosicky and J. Sterba-Böhm, Zeit. Kryst., 55. 430, 1916; P. Groth and K. Mieleitner, Mineralogische Tabellen, München, 29, 1921; W. Keil, Zeit. anorg. Chem., 152. 101, 1926.

2 J. H. L. Vogt, Zeit. prakt. Geol., 6. 225, 314, 337, 413, 1898; 7. 10, 274, 1899; F. W. Clarke and H. S. Washington, Proc. Nat. Acad. Sciences, 8. 108, 1922; The Composition of the Earth's Crust, Washington, 1924; W. Lindgren, Econ. Geol., 18. 419, 1923; H. A. Rowland, Johns Hopkins Univ. Cir., 85, 1891; Amer. Journ. Science, (3), 41. 243, 1891; Chem. News, 63. 133, 1891.

3 G. Urbain, Compt. Rend., 149. 602, 1909; G. Urbain, M. Blondel, and M. Obiedoff, ib., 150. 1758, 1910; J. Bardet, ib., 157. 224, 1913; 158. 1278, 1914; E. Cornec, ib., 168. 513, 1919; A. del Campo y Cerdan, Anal. Fis. Quim., 12. 80, 1914; K. von Chrustschoff, Journ. Russ. Phys. Chem. Soc., 24. 130, 1892; Neues Jahrb. Min., ii, 229, 1894; Proc. Russ. Min. Soc., 31. 412, 1894; G. Krüss, Ber., 21. 132, 2312, 1888; G. Krüss and L. F. Nilson, ib., 20. 1696, 1887; G. Lincio, Centr. Min., 142, 1904; A. Hadding, Zeit. anorg. Chem., 123. 171, 1922; L. M. Dennis and J. Papish, ib., 120. 1, 1921; Chem. News, 123. 190, 1921; Journ. Amer. Chem. Soc., 43. 2142, 1921; G. Neumann, Monatsh., 12. 515, 1891; A. Stelzner, Neues Jahrb. Min., ii, 114, 1893; D. I. Mendeléeff, Liebig's Ann. Suppl., 8. 200, 1871; G. H. Buchanan, Journ. Ind. Eng. Chem., 8. 585, 1916; 9. 661, 1917; J. H. Müller, ib., 16. 604, 1924; C. Winkler, Journ. prakt. Chem., (2), 34. 177, 1886; (2), 36. 177, 1887; W. F. Hillebrand and J. A. Scheerer, Journ. Ind. Eng. Chem., 8. 225, 1916.

§ 2. The Extraction and Preparation of Germanium

The chief associates of germanium are silver, copper, zinc, cadmium, gallium, indium, titanium, tin, lead, arsenic, antimony, tantalum, and columbium. Germanium sulphide is precipitated by hydrogen sulphide from highly acidified soln., and the sulphide is readily soluble in ammonium sulphide; consequently, germanium is easily separated from the majority of its associates. The main difficulty is the removal of arsenic and tin-which occur in the same analytical groupvide infra, reactions of analytical interest.

C. Winkler 1 extracted germanium from argyrodite by heating a mixture of the finely powdered mineral with sodium carbonate and sulphur at a moderate red heat. The product was extracted with water, and the soln. treated with the exact amount of sulphuric acid necessary to decompose the whole of the sodium sulphide present. After being left for a day, the soln. was separated from the precipitate consisting of sulphur, and arsenic and antimony sulphides, and treated with hydrochloric acid so long as a precipitate was formed. The whole was then sat. with hydrogen sulphide, filtered, and the white voluminous precipitate of germanium sulphide washed with 90 per cent. alcohol sat. with hydrogen sulphide. The sulphide was roasted at a low temp. warmed with conc. nitric acid, and the oxide so obtained ignited.

G. Krüss fused euxenite with potassium hydrosulphate and heated the product with hydrogen chloride to remove the iron. The washed product was treated with