for equilibrium t o be established R hen running a sample which differs materially in its concentration from the previous samples. Before the 10% samples shown in Figure 3 were run, the instrument was equilibrated to samples having a 1 atom % concentration of deuterium. The zero time on Figure 3 corresponds t o the time a t which the instrument was equilibrated t o the 1 atom % samples. An accurate determination of the difference between the two 10% samples could thus have been determined samples were run. It in about 1.5 hours after the 1 atom is possible that by further improving the procedure this equilibrium time could be reduced.
A series of experiments was perforined in order to determine the accuracy that could be obtained in measuring differences in hydrogen concentrations in samples having very high deuterium concentration. As no synthetic samples of a known small difD), two ference were available in the 99%+ range of D / ( H samples of slightly different concentration were prepared from a sample of nominally 99.57, D / ( H D ) by distilling approximately half of the sample into an ampoule which had some normal water adsorbed upon the walls. Runs were then made to determine the reproducibility of the measured H / ( H D) difference between these two samples. I n the first attempt to introduce these samples, the procedure of using a microburet for introducing the sample into a mercurycovered sintered disk was employed. This procedure was found to be entirely inadequate, giving very erratic results. This was probably due t o the fact that the small samples, during introduction, were contaminated by x a t e r contained in the mercury covering the sintered disk, and by atmospheric water adsorbed on the microburet. A second method of introduction was therefore devised, wherein each sample was placed in a small ampoule and connected to one arm of a Y connection through a stopcock, the Y connection being connected into the system in place of the sintered disk. Each sample was then introduced by opening its stopcock and allowing the vapor to pass into the inlet sample bottle and through the instrument until the mass 20 ion current had reached a standard intensity. (When this procedure of sampling vapor is used, it should be remembered that the isotopic composition of the vapor is not identical to that of the liquid. As the standard and unknown samples have very nearly the same composition, however, any error due to this effect cancels out when t h e difference in ratios is obtained.) T h e procedure following this was similar to that used for the introduction of the low concentration samples. I n other words, the system \vas flushed three times with one sample, and then the 19 and 20 peaks were recorded upon the fourth and fifth introductions. The second sample was then used to flush the system three times and its 19/20 ratio n-as determined upon the fourth and fifth introductions.
These operations were repeated on a definite time schedule, and the results obtained are shom-n in Figure 4. The difference in concentration of H/(H D ) obtained, after the samples had approached equilibrium, was 0.017 atom %. The values plotted on
.\Iass Spectrometer Deterniiriatiori of Difference of Concentration of Hydrogeii
T w o sumplrc of upproximate compoJitinn of 99.5 a t o m 7c D ( I 1 T D). i t zero t i m e appuratu+ u a - cquilihruted to sample of unknown rumpobition
these curves were calculated on an absolute basis, as there was no standard sample of known concentration available. Data similar t o those shown in Figure 4 Lvere taken on two other days. T h e D) values obtained on those days for the difference in H/(H concentration were 0.014 and 0.013 atom %, respectively. Khile relatively few results were obtained a t this high deuterium concentration, those given above indicate that a reproducibility of perhaps f 0 . 0 0 2 atom % may be eypected.
(1) Evans, M.W., Bauer, N., and Beach, J. Y., J . Chern. Pkys., 14, 701 (1946). (2) Gifford, 4.P., Rock, S. A I , , and Comaford, D. J., A s . 4 ~ .CHEN., 21, 9 (1949). (3) Kirshenbaum, I., “Physical Properties and Analysis of Heavy Water,” National Nuclear Energy Series, Div. 111, Vol. 4a, p. 54, New York, McGraw-Hill Book Co., 1951. (4) Sier, A. 0. C., Stevens, C. M., and Rustad, B., “Mass Spectrometer for Routine Hydrogen Isotope Analysis,” Atomic Energy Commission AECD-2767. (5) Orchin, M., Wender, J., and Friedel, R. .L,ANAL.CHEW,21, 1072 (1949). (6) Purdy, K. M.,and Harris, R. J., Ibid., 22, 1337 (1950). (7) Thomas, B. W., Ibid., 22, 1476 (1950). RECEIVED December 31, 19.51. dccepted September 2 9 , 1952. Presented before Section 2 , Analytical Chemistry, a t the XIIth International Congress of Pure and Applied Chemistry, S e w York, S. Y., September 10 to 13, 1951.
Analytical Methods for Germanium HORATIO H. KRAUSE
OTTO H. JOHNSON
School of Chemistry, L’nicersity of Minnesota, Minneapolis, J f i n n .
E R X 4 S I L X has suddenly become of industrial signifi-
cance. Because the available information on the determination of this element is scattered and hard to find, this review of work done in this field has been compiled. QUALITATIVE ANALYSIS
The separation of germanium from other elements for purposes of identification is accomplished by distillation as the germanium(IV) halide or by precipitation with inorganic or organic precipitants. Distillation Procedures. The distillate obtained from a hydrobromic acid solution of a comple.: mixture contains arsenic and selenium as well as traces of tin and antimony (73’). Sele-
nium present in the distillate is removed by precipitation with hydroxylamine hydrochloride ( 1 2 , 7 3 ) or with sulfur dioxide (107). Separation of the germanium from the arsenic is effected by distillation of germanium(1V) chloride from concentrated hydrochloric acid solution containing an oxidant, preferably a stream of chlorine, t o keep the arsenic in the nonvolatile quinquevalent stat,e ( 1 7 , 18, 32, 107). This procedure will detect. 0.5 mg. of germanium dioxide in 100 grams of arsenic trioxide. Precipitation Procedures. Slinerals containing germanium and arsenic are fused with sodium carbonate and sulfur. Arsenic is t,hen precipitated from dilute acid solution as the sulfide ( 1 , 38) or magnesium ammonium arsenate (12)arid the germanium is recovered by precipitation from concentrated hydrochloric
V O L U M E 25, N O . 1, J A N U A R Y 1 9 5 3
This review summarizes the qualitative and quantitative methods for determining germanium. The majority of the methods require a preliminary separation of the germanium either by distillation as the halide or by precipitation. Quantitative procedures include gravimetric, volumetric, colorimetric, polarographic, and spectrographic. Most procedures are applicable only in specific cases and the selection of the method is determined by the nature of the interfering substances present.
groups in the ortho position and having, in a position para to one hydroxyl, another group capable of producing, in acid medium, a prototropic change which favors the ionization of the opposed phenol group. Included in this type of aromatic nuclei are the hydroxyazobenzenes, triphenylmethanes, oxazines, Schiff bases, and numerous vegetable dyes (98). Polyhydroxy alcohols such as glycerol and mannitol form complex acids with metagermanic acid, HZGeO3. These complex acids are more highly ionized than the original germanic acid and the presence of the germanic acid is detected by the decrease in the p H of the resulting solution (82). Molybdic Acids. Germanates combine with molybdic acid t o form molybdogermanic acid, HsGe( h'fozO7)e. \Then this compound is treated with benzidine, the color of the reduction moduct. molvbdenuni blue. is Eeinforced by that of the oxidation product, benzidine blue, Table I. Factors for Heteropoly Acid Complexes the combination of the two Pyridine Salt Cinchonine Salt Oxine Salt . colors making the test very Useful range, Useful range, Cseful range, mg. Ge Factor mg. Ge Factor mg. Ge Factor sensitive (56). Ions such as BIolybdogernianium Complexes tin(II), iron(II), and arsenic(III), capable of reducTheoretical factors (BaseIrH4 [Ge ( M o ~ 0 r ) e l 0.0326 0.02359 0.02927 ing molybdates, must be absent (Base14(GeMo~zOro] 0.0333 0.02387 0.03009 as well as substances such as Experimental factors Geilmann and Brunger (40) 0 . 0 3 5 3 a r s e n a t e s , phosphates, and 0.5-4.0 0,02385 Davies and Morgan ( 2 9 ) 1.0 -3.5 0,0317 Alimarin and Alekseeva (4) 0.0311 silicates which cause interfer0.125 0.01901 0.05-0.25 0.02807 Herht and Barthelmus ( 4 6 ) 0.05-2.0 0.03487 2.0-4.0 0,02374 0,s-1.0 0.03003 ence by forming heteropoly 3.5-3.5 0.0311 acids with molybdenum,
acid solution as germanium(1V) sulfide. The sulfide method will detect as little as 0.005 mg. of germanium in a 1-mg. sample (12, 17, 67). Germanium(1V) compounds are reduced by sodium hypophosphite in acid solution to the soluble bivalent state, while arsenic compounds are reduced to insoluble elementary arsenic which is then removed by filtration (61). Germanium frequently occurs with zinc compounds and the two metals may be separated by reduction with zinc dust, which yields insoluble elementary germanium and leaves the zinc compounds in solution (65). Reduction of monogermane with hydrogen deposits a germanium mirror similar to the Marsh test for arsenic. Various adaptations of this test have been tried, the chief variations being in the hydrogen source ( 2 , 26, 27, 70).
Tungstogermanium Complexes Theoretical factors (BaseIjH4 [Ge(W1z04z)I Experimental factors ( 4 6 )
Other precipitation methods include the addition of potassium ferrocyanide, which forms a white precipitate tentatively assigned t h e formula, (Ge0)2Fe(CN)6.2Hn0 (80). Treatment with tannin or quinine tannate under specified conditions precipitates t h e germanium and by varying the conditions of the precipitation, separation from many other metals is possible (29, 48, 49, 104,105). Hydrogen selenide in aqueous formaldehyde forms a yellox precipitate of unknown composition with germanium, the test having a sensitivity of 0.2 p.p.m. (59). A sensitive test suitable for micro work is the formation of crystals of sodium hexafluogermanate(1V) or rubidium molybdogermanate (21). Organic Reagents. The simplest and most specific test for the germanium(1V) ion is phenylfluorone (9-pheny1-2,3,;trihydroxy-6-fluorone). Test paper is prepared by treating spot paper with a 0.0570 solution of phenylfluorone in alcohol. The solution to be tested is acidified to a p H of l with hydrochloric acid, and a drop is placed on the test paper and treated wit,h concent,rated nitric acid. An intense rose color indicates germanium. The only interfering ions are strongly oxidizing ones such as dichromate, permanganate, or cerium(1V) (44, 57). Other organic reagents are hydroxynaphthacene quinonesulfonic acid (81), alizarin S , quinalizarin, purpurin, and p-nitrobenzeneazochromotropic acid (36,43,82,86). In general, germanium is precipitated by aromatic organic reagents having two hydroxy
T h e best method for separating germanium from complex mixtures is by distillation of the tetrachloride from a concentrated hydrochloric acid solution (99). Germanium is then determined in the distillate. Gravimetric Procedures. Germanium dioxide is the most common weighing form, but there are numerous variations in the methods used t o obtain it. I-sually the germaniuni is precipitated from solution as the sulfide. Germanium(1V) sulfide is unusual in that precipitation is complete only when a rather high concentration of acid is present. Germanium is complesed by fluoride or oxalate in acid solution and hence may be separated from arsenic and antimony and many other metals of the hydrogen sulfide group by a preliminary precipitation of these elements with hydrogen sulfide. Germanium( IV) sulfide i p converted to the dioxide by direct ignition, but this proredwe is unsatisfactory, because the formation of volatile germanium(I1) oxide causes indeterminate losses ( 6 1 ) . Direct oxidation with nitric acid is generally unsatisfactory, being subject to mechanical losses by spattering and the formation of volatile germaniurn(I1) oxide (SI, 110, 111). Solution of the disulfide in ammonia followed by oxidation with 3y0 hydrogen peroxide has bem used (29, 5 3 ) , as has hydrolysis with steam (69). The germanium(1V) ion is quantitativt4)- precipitated in a 5y0 tannin solution, giving a white flocculent precipitate that is
converted to germanium(IV) oxide by direct ignition (69). This procedure was used in the Manhattan Project to determine Table 11. Effect of Temperaturq on Gravimetric Forms germanium in alloys (99). Germanium forms the complex, Temp. Ranae of H,Gr( C Z O ~ )with ~ , oxalic acid and the addition of 5,6-benzoColist. quinoline oxalate to this solution gives a nonstoichiometric Gravimetric Cpmp., Reagent Precipitate Form C. precipitate which is ignited to germanium(1V) oxide for n-eighHzS GeSz GeOz 410-946 ing (33, 95, 109). MgzGeO, MgzGeO4 280-814 GeOa 900-950 F t n L Ge tannate Regardless of the method used to obtain the germanium(IV) 8-Naphthoquinoline oxide, drying of the precipitate a t 100' C. will not remove all (benso[f]quinoline) Bz[Ge(CzOdzI GeOz >800 of the n-ater and ignition a t 900" to 1000° C. is customary. When B4 Ge\lolnOo Ge02.12MoOs 440-813 tetramine GeOz.12MoOs 429-813 Bd[Geko120o] Pyfidine filter paper is ignited, reduction to volatile germanium(I1) 50-115 Ba[Geb~oi~Ora] GeOt.lZMoOa.B, Oxine oxide must be avoided (29). Nonstoichiometric Indeterminate 450-900 Cinchonine Magnesium orthogermanate, MgzGe04, is precipitated from aqueous solutions of germanium(1V) ions by a mixture of magmesium sulfate, ammonium sulfate, and concentrated ammonia. Table 111. Comparison of Germanium Lines with Copper, This procedure is subject to serious error because of coprecipitaBismuth, and Platinum tion mith magnesium hydroxide (29,68). Ge c u Bi Pt Ref. The precipitation of salts of molybdogermanic or tungstogermanic acid by organic bases such as pyridine, cinchonine, 82709.6 2768 (16) ?Si;: 93 2651.15 ... 2 6 i 9 : 44 (87, 8 8 ) uuinolinol, or hexamethylenetetramine give weighing forms with 2651.60 ,.. 2627.93 2659.44 (87, 88) 3039.08 ... 2993.34 3064.71 (87, 8s) favorable gravimetric factors (4, 29, 33, 40, 46). The data are summarized in Table I. T h e temperature limits Table IV. Summary of Quantitative 3Iethods within which the different gravimetric forms have constant composition have been s u m m a r i z e d i n T a b l e I1 (33, 34). Volumetric Methods. Germanium(1V) ions form a complex Kith mannitol and this mannitol-germanium complex acts as a monoprotic acid when titrated with sodium hydroxide. This method is especially suitable when a preliminary separation of germanium as sulfide has been employed ( M , 96, 97). Treatment of the mannitol-germanium complex with a mixture of potassium iodide and potassium iodate in the presence of s t r o n g el e c t r 01 y t es liberates iodine q u a n t i t a t i v e l y , permitting titration with sodium thiosulfate (97). Other iodometric methods include the precipitation of the oxine salt of molybdogermanic acid. The precipitate is dissolved in a mixture of hydrochloric acid a n d ethanol, treated with an excess of b r o m i d e - b r o m a t e solution followed by potassium iodide, and the liberated iodine titrated with standard thiosulfate ( 4 ) . Germanium dioxide in aqueous solution is quantitatively converted into thiodiger manate by hydrogen sulfide or potassium sulfide in acetate-buffered solution:
+ + +
2Ge(OHh 5H8 2CH8 COOK +K2Ge2S5 2CH3COOH 8H2O
References (11, 89, $2, 65, 61, 69)
Pptn. a s GeSz
Pptn. with tannin Complex formation with oxalic acid Pptn. a s molybdogermanic or tungstogermanic salts of organic bases
P p t n . a s MgaGeOi Titration of mannitol-germanium complex with K a O H
GeOa also forms volatile GeO. Hyd. >y steam has been used. Oxidation by HzOz gives most consistent results P p t n . is quantitative in presence of ammonium ion. P p t . ignited t o GeOz for weighing Forms compound, HzGe(Cz0da. Addn. of 5,6benzoquinoline oxalate gives nonstoichiometric ppt.. which is ignited t o GeOz for weighing Pyridine, cinchonine, oxine, and hexamethylenetetramine give ppts. with the formulas (Base)&Hd[Ge (Moz07) a ] and, (Base) I [Gellloi~Oro]. These ppts. have empirical factors t h a t depend on treatment of ppt. and do not agree with theoretical gravimetric factors .4ddn. of MgSO4, (NHa)zSOd, and coned. NHs ives bulky white ppt.; coprecipitation of M g ( 8 H ) a may cause error Mannitol-germanium complex may be titrated a s monoprotic acid with NaOH. If CaClz or SrClz is added, complex IS titrated a s diprotic acid. Boron interferes. Preliminary pptn. of GeSz
(BS, 97, 96)
Iodometric titration of mannitolgermanium complex
Iodometric titration of ppt. of oxine salt of molybdogermanic acid
Iodometric titration of potassium thiodigermanate Iodometric titration of Ge(I1) salt KMnO4 or KBrOs titration ef GeO Colorimetric estn. of molybdogermanium complex Colorimetric estn. of molybdenum blue from molybdogermanium complex Colorimetric estn. of Ge(heniatoxylin)? complex
Colorimetric estn. of phenylfluorone complex Spectrographic procedures
Na,&Oa GeOz in aq. soln. is quant. converted t o KzGezSs b y HzS or KzS in CHICOOK buffered soln. KzGeaSs is oxidized with excess of standard 1 2 soln. and back-titrated with XazSzOs Ge(IV) reduced t o Ge(I1) by NaHzPOa and Ge(I1) determined by iodometric titration Ge(1V) salts reduced t o GeO by Zn in 25% His04 a n d titrated with KMnOd or KBrOs. Results are a n approximation only Permits estn. of 1 p.p.m. of Ge. Requires considerable preliminary prepn. of sample Molybdogermanium complex reduced by NazSOr. NazCOs and hydroquinone (NHdzFe(SO4 t in satd. C'HrCOONa, or (NHhzFe(S03z in $804. Preliminary pptn. as GeSz usually required. Usef u l for estn. of small amts. of Ge in silicate rocks Purple complex formed with oxidized hematoxylin is not interfered with b y other elements forming heteropoly acids. Considerable preliminary prepn. of sample required, Useful range 0.08 to 1.6 p.p.m. of Ge Very sensitive for small amounts of Ge Ge shows lines from 2198.7 to 4686.1; C u , Bi, and Pt may be used a s internal standards. Useful range very broad. Elements forming complex silicates must be removed G e + + + +gives two waves first reducing to G e + + , then t o metal, Ge + + c s n be determined cathodically in acid solution a t potential of about - 0 . 5 0 volt, dependent on both germanium and hydrogen ion concentration. Anodic wave can be used after reduction of Ge with sodium hypophosphite + + + +
(14-16, $5, 41. 63!64, rr, 78, 84, 87-90, OS, 99, 101, 106) (7, 88. $8.
V O L U M E 2 5 , NO. 1, J A N U A R Y 1 9 5 3
78). When only traces of germanium are present, it must Source Process be concentrated (16, 61, 64, Composition: 4AgtS.GeSz Extraction from argyrodite 84, 89, 90, 106). Quantities Fusion with potassium nitrate and potassium hydroxide, of germanium as small as 1 followed by leaching with water. Precipitation as gerinanium(1V) sulfide from sulfuric acid solution m i c r o g r a m i n 1 g r a m of C o m ~ o s i t i o n : 7CuS.FeS.GeSz with v a r. r i n a- amounts of Extraction from germanite arsenic, zinc, lead, and gallium uranium (99) and as high as Heating in nitrogen followed b y reduction t o volatile ger5.07, in zinc sulfide ores (93) maniurn(I1) sulfide with ammonia Roasting followed b y oxidation t o germanium dioxide by have been determined specnitric-sulfuric acid mixture Direct chlorination, t h e n fractional distillation of chloride troscopically. I n minerals or a n d hydrolysis t o dioxide rocks containing extraneous Treatment with sodium hydroxide, followed by oxidation with nitric acid, precipitation with ammonia, and dematerials capable of forming hydration with concentrated sulfuric acid t o obtain dioxide , complex silicates, a preliminary Electrolysis of alkaline extract using copper or nickel separation of the germanium is cathode Extraction from zinc conDistillation from hydrochloric acid solution in a stream of necessary (36, 63). centrates chlorine, followed b y precipitation a s disulfide a n d conversion t o dioxide Copper, bismuth, and platiRoasting, followed b y sintering with salt t o obtain chloride, num are used as intcrnal standthen distillation in hydrochloric acid and hydrolysis to . . dioxide ards. The pairs of lines used Digest in concentrated sulfuric acid, convert t o sulfide with sodium sulfide, a n d oxidize t o dioxide with nitric acid for comparison are g i w n in Remove silicon as tetrafluoride, then distill germanium(I1') Extraction from materials Table 111. chloride containing silica Remove carbon by heating t o 600' C. or using bomb caloExtraction from coal X-ray spectroscopy has not rimeter Fuse ash with sodium carbonate, mixture of sodium carbonbeen adapted to quantitative ate and sulfur, mixture of sulfuric, perchloric, and hydrodeterminations, but some gerHuoric acids, or sodium peroxide Remove germanium b y usual distillation or precipitation manium lines have been obniet hods Smelting with soda, lime, cupric oxide, and iron yields Extraction from flue dusts served in qualitative studies copper-iron regulus containing germanium. Treatment (14,101). with chlorine converts t o chloride, followed by fractional distillation Polarographic Procedures. Dissolve in nitric-sulfuric acid mixture and distill from Extraction from steel hydrochloric acid Germanium(1V) chloride is reFuse with sodium carbonate or dissolve in hydrofluoricExtraction from glass duced a t thp dropping mercury sulfuric acid mixture a n d distill from hydrochloric acid Belgian Congo mineral reported t o have 6.4-7.8% gerExtraction from renierite electrode in two steps, indimaniiim. S o details of extraction given cating the formation of the G e + + ion before reduction to the metal ( 7 6 ) . I n the preaence of hydrofluoric and oxalic acid, reduction of germanium(1Vij The thiodigernianate is oxidized with an excess of standard iodine solution and back-titrated n-ith thiosulfate solution. High was not observed while germanium(I1) is reduced t o the metal. Concentrations of chloride ion and the presence of strong electroThe half-wave potential is 0.45 to 0.50 in 6 .V hydrochloric acid lytes cause appreciable error (109). J\-hen chlorides are present and lo-' M germanium(II), but is dependent on both germanium and hydrogen ion concentration ( 7 ) . the germanium( IV) ion may be reduced by sodium hypophosphite to germanium(II), n.hich is then determined by iodometric The best procedure appears to be the reduction of germaniunitit.ration (51). (IV) to germanium( 11) with sodium hypophosphite, the gerJ\.hen :in approximation is sufficient, germanium(1T') salts manium(I1) then giving an anodic wave a t -0.130 volt (28). ma>-be r d u c e d t o germanium(I1) oxide by zinc in 25% sulfuric Chlorides interfere but may be complexed as CdCld-- and acid and then titrated n-ith potassium permanganat,e or potassium metals ordinarily present with germanium do not interfere n-ith bromatr (10). determinations down to 10-6 M germanium(1V) if conditions Colorimetric Methods. The yellow color of the niolybdogerare carefully regulated. Traces of germanium may be concenmanium complex follows Beer's l a x in concentrations up to 40 trated before polarographing by electrolyzing the solution with a mg. of germanium per liter and comparison with known color mercury cathode, Il-ith subsequent removal of the mercury by standarcls permits estimation of quantities of germanium as distillation in a current of nitrogen (39). small as 1 p.p.m. The method requires considerable preliminary Analysis of Organogermanium Compounds. In the deterpreparation of the sample, and phosphates and silicates interfere mination of germanium in organo-gprmanium compounds the with the determination (8, 55). most convenient method, when only a few determinations are to Reduction of the molybdogermariium complex to molybdenum be made, is that of direct oxidation to germanium(1V) oxide by blue increases the sensitivity (13, 40, 83). Beer's law is valid nitric and sulfuric acids (11, 50, 40,94). The method, however, for molybdenum blue solutions in concent'rations up to 1.5 is not very accurate; the percentage of germanium found in a micrograms of germanium per milliliter, the color intensity becompound such as chlorotricyclohe~ylgermane, for example, coming less than theoretical a t higher concentrations (60). varies from 19.6 to 20.5c0 (71). For volatile compounds, such This method is useful for silicate minerals cont.aining small as diphenylgermane, the method gives such erratic results as to amounts of germanium (37,50, 76). Oxidized hematoxylin forms be of no value (45). The use of the Parr bomb, in a manner a purple complex with germanium and has the advantage that similar to that developed for silicon (sa), followed by precipitaot,hrr elements forming heteropoly acids do not interfere. This tion with tannin gave inconclusive results in preliminary experimethod is suitable in the range 0.08 to 1.6 p.p.m. of germanium ments (45). ( 7 2 ) . T h e pink phenylfluoronegermanium complex is specific T h e presence of germanium causes low results for carbon following a preliminary separation by distillation from hydrowhen organogermanium compounds are analyzed by the chloric acid and follows Beer's Iav for amounts up to 1 p.p.m. standard microcombustion method. This error is probably of germanium ( 2 4 ) . due t o the formation of germnnium carbides which resist Spectrographic Procedures. The arc spectrum of germanium oxidation (71 ). has been studied in the range 2198.7 t o 4686 A. on a quantitaT h e various quantitative methods are summarized in Table IV. tive basis, the amount of the element present being estimated A complete description of present-day commercial methods for by the appearance or nonappearance of the various lines (77, the production of germanium has been given by O'Connor ( 7 4 ) .
Table Y. -4 Summary of Extraction Methods
138 Other methods for extracting germanium from native ran- materials have been summarized in Table V. LITERATURE CITED
( 5 5 ) Kitson, R. E., and hlellon, M. G., IXD.ENG.CHEM.,.IN.~L.
ED., 16, 128 (1944). ( 5 6 ) Komarovskii. -4.S.. and Poluektov. N. S.. M d w x h e m i e , 18,
66 (1935). Krause., H. H.. unuublished data. 1952. Kriesel, F. W.; Chem. Ztg., 48, 961 (1924). Kuenetsov, V. I., J . Gen. Chem. U.S.S.R., 9, 1049 (1939). Lloyd, D. J., and Pugh. W., J . Chem. Soc., 1943, 74. Lundin, H., Trans. Electrochem. Soc., 63, 149 (1933). McHard, J. A,, Servais, P. C., and Clark, H. -I., A F ~ LCHEM.. . 20, 325 (1948). Marks, G. W.,and Hall, H. T., U. S. Bur. Mines, Repts. ~~
(1) Abrahams, H. J., and Muller, J. H., J . Am. Chem. Soc., 54, 86
(1932). (2) Aitkenhead, W. C., and hliddleton, .I.R., IND.ENG.CHEM., ANAL.ED.. 10. 633 (19381. (3) Alimarin, I. P., T r u d y V s e s o y u z . Konferents. A n a l . Khim., 2, 371 (1943). (4) dlimarin, I. P., and Alekseeva, O. d., J . Applied Chem., U.S.S.R., 12, 1900 (1939). (5) Ibid., 13, 1393 (1940). (6) Alimarin, I. P., and Ivanov-Emin, B. S . ,Ibid., 13, 951 (1940). (7) Ibid., 17, 204 (1944). (8) Alimarin, I. P., and Ivanov-Emin, B. S . , Mikrochemie, 21, 1 (1926). (9) iilimarin, I. P., et al., Zarodskaya Lab., 9, 271 (1940). (10) Bardet, J., and Tchakirian, A , , Compt. rend., 186, 637 (1928). (11) Bauer, H., and Burschkies, K., Ber., 65B, 960 (1932). (12) Benedetti-Pichler, A. A,, and Rachele, J. R., IND.EXG.CHEM., ANAL.ED., 9, 589 (1937). (13) Bolts, D. F., and Mellon, .If. G., AXAL.CHEM.,19, 873 (1947). (14) Borovskii, I. B., T r u d y Vsesoyuz. Kon.ferents. A n a l . Khim. A k a d . Nauk S.S S.R., 1, 135 (1939). (15) Breckpot, R., Ann. soc. sci. Bruselles, 55B, 160 (1935). (16) Britske, hl. E.. and Varshavskaya, L. S . ,Izaest. A k a d . S a u l ; S.S.S.R., Ser. Fiz., 12, 455 (1948). (17) Browning, P. E., and Scott, S. E., A m . J . Sci., 44, 313 (1917). (18) Ibid., 46, 663 (1918). (19) Buchanan. G. H., J. IND. ENG.CHEW,8, 585 (1916). (20) Ibid., 9, 661 (1917). (21) Chamot, E. bf., and Cole, H. I., Mikrochemie, 4, 97 (1926). (22) Chapman, F. IT., Marvin, G. G., and Tyree, S. Y., ANAL. CHEM.,21, 700 (1949). (23) Cluley, H. J., Analyst, 76, 517 (1951). (24) Ibid., p. 523. (25) Ibid., p. 530. (26) Coase, 5. A., Ibid., 59, 462 (1934). (27) Ibid., p. 747. (28) Coezi, D., and Vivarelli, S., Mikrochemie oer. AZikrochim. Acta, 36/37, 594 (1951). (29) Davies, G. R., and hlorgan, G., Analyst, 63, 388 (1938). (30) Dennis, L. M., and Hance, F. E., J . A m . Chem. Soc., 47, 370 (1925). (31) Dennis, L. XI.,and Johnson, E. B., Ibid., 45, 1380 (1923). (32) Dennis, L. &I.,and Papish, J., Ibid., 43, 2131 (1921). (33) Dupius, T., and Duval, C., A n a l . Chim. Acta, 4, 186 (1950). (34) Duval, C., Ibid., 2, 432 (1948). (35) Efendiev, F. hI., Bull. acad. sei. C.R.S.S., Srr. phys.. 11, 313 (1947). (36) Feigl, F., “Qualitative Analysis by Spot Tests,” pp. 89-92, New York, Elsevier Publishing Co., 1946. (37) Fischer, W., and Keim, H., 2. anal. Chem., 128, 443 (1948). J . Am. Chem. Soc., 61, (38) Foster, L. S., and Thompson, R. P., 236 (1939). (39) Furman, N. H., Bricker, C. E., and McDuffie, B., J . Wash. Acad. Sci., 38, 159 (1948). (40) Gelimann, W., and Brunger, K., Biochem. Z.,275, 375 (1935). (41) Geilmann, W..and Brunger, K., 2. anorg. allgem. Chem., 196, 312 (1931). (42) Geilmann, W., and Steuer, E., Glastech. Ber., 18, 89 (1940). (43) Gillis, J., Bull. S O C . chim. France, 1946, 177. (44) Gillis, J., Hoste, J., and Claeys, A , Anal. Chim. Acta, 1, 302 (1947). (45) Harris, D. hf., Ph.D. thesis, University of Minnesota, p. 61, 1950. (46) Hecht, F., and Barthelmus, G., Mikrochemie w r . Mikrochim. Acta, 36/37, 466 (1951). (47) Hoffmann. J. I., and Lundell, G. E. F., J . Research S a t l . Bur. Standards, 22, 465 (1939). (48) Holness, H., A n a l . Chim. Acta, 2, 254 (1948). (49) Ibid., 3, 290 (1949). (50) Hybbinette, 9.G., and Sandell, E. B., IND. ESG. CHEM., ANAL.ED.. 14, 715 (1942). (51) Ivanov-Emin, B. N,, Zaaodskaya Lab., 13, 161 (1947). (52) James, C., and Fogg, H. C., J . Am. Chem. Soc., 51, 1459 (1929). (53) Johnson, E. B., and Dennis, L. M., Ibid., 47, 790 (1925). (54) Johnson, W. C., Foster, L. S., and Kraus, C. A,, Ibid., 57, 1828 (1935). ~
t . 3965 -T n w ..s.. - ..- (1946)
\ - - - - I
hfitchell, R. L., Bnalyst, 71, 361 (1946). Mitchell, W.E., U. 8.Patent 1,848,264 (March 8, 1932). hliiller. J. H.. J . Am. Chem. Soc., 43, 1088 (1921). Ibid.,p. 2549. Ibid.. 44. 2493 (1922). Miillkr, J. H., and Eisner, A, IND.ENG.CHEM.,. ~ N A L . ED., 4, 134 (1932). Muller, J. H., and Smith, N. H., J . Am. Chem. Soc., 44, 1909 (1922). Nebergall, K. H., Ph.D. thesis, University of Minnesota, p. 73, 1949. Newcombe, H., McBryde, IT. .I,, Bartlett, J.. and Beamish, F. E., h . 4 ~ CHEM.. . 23, 1023 (1951). Noyes, A. A , and Bray, W.C., “System of Qualitative Analysis for the Rare Elements,” pp. 1-70, New York, Jlacmillan Co., 1927. O’Connor, J. A , , Chcm. Eng., 59, S o . 4, 158 (1952). Orliac, M.. Compt. rend., 221, 500 (1945). gsterud, T., and Piyts, M.,Arch. Math. Naturaidenskab, 47, 73 (1943). Papish, J., Chem. S e u s , 124, 3 (1922). and Holt. D. A. J . A m . Chem. Soc., Pauish. J.. Brewer, F. M., 49, 3028 (1927). Patnode, W. I., and Work, R. W.,IND.ENG.C H n i . , 23, 204 (1931). Peisach, >I,, Pugh, IT., and Sebba, F., J . Chem. Soc., 1950, 949. Poluektov, N. S., J . Applied Chem. U.S.S.R., 9, 2302 (1936). Poluektov, K.S., Mikrochemie, 18, 48 (1935). Poluektov, 5 . S., 2. anal. Chem., 105, 23 (1936). Preuss, E., 2. angew. Mineral., 3, 8 (1940). Pugh, IT,, J. Chem. Soc., 1929, 2540. Ridder, hl. de, Chem. Zentr., 1943, 11, 444. Rusanov, A. K., and Bodunkov, B. J., Zaaodskaya Lab., 9, 183 (1940). Rusanov, A . K., and Kostrikin, T’. hI., J . Applied Chena. U.S.S.R., 9, 2305 (1936). Schleicher, A , , 2. anal. Chem., 101, 241 (1935). Scribner, B. F., and Mullin, H. R., J . Research 9 a t l . Bur. Standards, 37, 379 (1946). Sebba, F., and Pugh, W., J . Chem. Soc., 1937, 1371. Simek, B. G., Coufalik. F., and Stadler, A , , Zprciry cstavzb T’ddeckj ’C‘jzkum Uhli, 1948, 167. Strock. L. W,,Am. Inst. Mining Met. Engrs., Tech. Pz~b. 1866 (1945). Tabern, D. L., and Shelberg, E. F.. IND.E x . CHEM.,-IS(LL. ED., 4, 401 (1932). Tchakirian, A.. Ann. chim., 12, 415 (1939). Tchakirian, A., Bull. SOC. chim., 10, 98 (1943). Tchakirian, A,, Compt. rend., 187, 229 (1928). Tchakirian, A , and Bavillard, P., Compt. rend., 223, 256 (1951). Telford, R. E., and Furman, N.H., Natl. Nuclear Energy Ser., Div. VIII, 1, ilnal. Chem. Manhattan Project, 372 (1950). Thomas, J. S., and Pugh, IT., J . Chem. Soc., 125, 816 (1924). Turner, A. H., Chem. Eng. Mzning Rev., 20, 361 (1928). Urbain, G., Blondel, &I.,and Obiedoff, M., Compt. rend.. 150, 1758 (1910): Vaes, J. F., Ann. soc geol. Belg., Bull. 72, 19 (1948). Vanossi, R., Anales. asoc. qtcim. Argentina, 32, 164 (1944). Vanossi, R., Anales. soc. cient. Argentina, 139, 29 (1945). Veselovskii, B. K., Zaaodskaya Lab., 10, 372 (1941). Wada, I., and Kato, S., Sci. Papers Inst. P h y s . Chem. Research (Tokyo), 3, 243 (1925). Weissler, A,, IND. ESG. CHEM.,~ ~ N A ED., L . 16, 311 (1944). Willard, H. H., and Zuehlke. C. W., Ibid., 16, 322 (1944). Winkler, C., J . prakt. Chem., 142, (N. F. 3 4 ) , 188 (1886). Ibid., p. 228. RECEIVED for review M a y 2 2 1952. Accepted October, 1 1952.