260 THE ANALYST. INORGANIC ANALYSIS. S. W. Parr. (Journ. Amer. Chenz. SOC., xxiv., 580.)-In the method depending on the oxidation by perman- ganate of cuprous thiocyanate (Jourlz. Amer. Chem. SOC., xxii., 685) the end-reaction is sometimes unsatisfactory. This can be obviated by the following modification : The washed precipitate of cuprous thiocyanate and the asbestos pulp are returned to the original beaker, 10 C.C. of a 10 per cent. potassium hydroxide solution and 10 C.C. of ammonia (specific gravity, 0.96) are added, and potassium permanganate solution is immediately added until, upon warming to 50°, a green colour remains. About one-third or one-fourth of the quantity of permanganate necessary for this is then run in, and the whole allowed to stand for five minutes, acidified with 25 C.C.of sulphuric acid (1 : 1 or Z), and the titration finished as usual. The results obtained with this method are very good. To calculate results, the iron value of the perman- ganate is multiplied by the factor 0.1602. A Note on the Volumetric Determination of Copper. A. G. L. The Volumetric Determination of Copper by means of Potassium Iodide. F. M. Litterscheid. (Zezt. anal. Chem., 1902, xli., 219-227.)-0n treating a solution of a copper salt with sulphur dioxide, and then adding potassium iodide in slight excess, the copper is precipitated quantitatively as represented in the equation The solution of the copper salt, which should be neutral or only slight acid, and should contain not less than 0.1 gramme of copper, is mixed with a few C.C.of sulphur dioxide solution, and then with a slight excess of decinormal potassium iodide solution. After standing for an hour for the cuprous iodide to become crystalline, the liquid is made up to definite volume and filtered, and an aliquot portion of the filtrate mixed with an excess of decinormal silver nitrate solution, and then with 10 C.C. of dilute nitric acid, and titrated with decinormal thiocyanate solution, with iron ammonium sulphate as indicator. Lead, mercury, and other metals precipitated under these conditions must be absent, and also substances that liberate iodine from potassium iodide. Copper and Silver.-About 1 gramme of the alloy is dissolved in nitric acid, and the solution mixed with about 5 C.C. of dilute sulphuric acid, and evaporated nearly to dryness. It is then diluted to about 50 c.c., and any separated silver sulphate brought into solution by warming.When cold, 1 drop of a 5 per cent. solution of potassihm nitrite and some starch paste are introduced, and the silver titrated with decinormal potassium iodide. The liquid is then filtered, and the copper determined in the filtrate as described above. ~CUSO, + 2KI + SO, + 2H20 = CU& + SKHSO, + H2S0,.THE ANALYST. 261 Copper a d Bismuth.-The solution of the alloy is concentrated to a small volume, the bismuth separated as basic carbonate by the addition of ammonium carbonate and ammonia, and the copper determined in the filtrate. The method is also applicable to the determination of nickel and zinc in the presence of copper. When lead is present, it is first determined in the form of sulphate.C. A. M. Preparation of Arsenic-Free Zinc, Otto Hehner. (Jozw. SOC. Chem. 1 9 4 1902, xxi., 675.)-Zinc suitable for use in the Marsh-Berzdius test is prepared as follows : Melt about a pound or two of ordinary block-zinc, no matter how impure, in a clay crucible over a good gas fire. When quite fluid, throw into the metal a piece of sodium, taking for each pound of zinc about 1 gramme of sodium at a time. Stir the molten metal with a piece of hard-glass tubing bent at right angles. A black scum forms immediately, which is removed with a china spoon or a crucible lid held in the tongs. Stir vigorously, and remove the scum from time to time till the sodium appears to have oxidized out; then add another piece of sodium, stir and skim as before, and oxidize well.This takes altogether about ten minutes. Finally, pour out the metal into a second clean clay crucible, and repeat the sodium treatment once more. It is best to keep two crucibles with file-cuts for the two stages. Allow the molten metal to cool considerably before granulating, as when the zinc is very hot and fluid heavy drops like shot are obtained, whereas when the zinc is near its solidifying point, thin flakes, presenting a large surface to the acid, are produced, A. M. ___ The Electrolytic Determination of Bismuth. 0. Brunck. (Berichte, 1902, xxxv., 1871-1873.)-Winkler’s method (Berichte, xxxii., 2192), io which wire gauze is employed as the cathode, has been found by the author to yield good results in the determination of bismuth. The metal or its salt is dissolved in nitric acid, the amount of which may be from twenty to twenty-five times that of the bismuth, though its strength must not exceed about 2 per cent, Another essential condition is that the tension of the current must not exceed 2 volts.The permissible strength of the current varies with the concentration of the solution. Thus, if more than 0.1 gramme of bismuth in 100 C.C. be present, a current of 0.5 ampere or more may be employed, whilst with less than 0.05 per 100 c.c, a current not exceeding 0.1 ampere is best. The electrolyte is first heated to 7Oo-8O0C., so that its resistance is lessened, and it is possible to transmit a fairly large current, notwithstanding the low tension. As the solution cools and becomes poorer in metallic ions the strength of the current diminishes, until at length it falls to only a few hundredths of an arnpbre.The deposition of even large quantities of bismuth is complete in two to three hours, and when the above conditions are observed the gray deposit is compact and adherent. After completion of the electrolysis both electrodes are transferred to a beaker of water, and the current continued for a short time, in order to effect the deposition of any bismuth dissolved by the adherent nitric acid. Finally, the cathode is washed successively with water, alcohol, and ether, and weighed. The results of the test experiments described agree well with the theoretical amounts.262 THE ANALYST. Sulphuric acid alone is not a suitable solvent for the metal, owing to its being necessary to use such large quantities to prevent the formation of basic salts.All the attempts to Reparate lead from bismuth by this means were unsuccessful, the lead peroxide invariably containing more or less bismuth. C. A. M. On the Compounda of Tellurium with Bismuth and the Quantitative Separation of the Two Elements. A. Gutbier. (Zeits. Anorg. Chem., xxxi., 331.)- I n the analysis of various mixtures of bismuth and tellurium prepared by himself, the author has examined three methods of separating these two elements. He condemns the first, in which tellurium is precipitated by sulphurous acid, bismuth being thrown down by hydrogen sulphide in the filtrate ; and the second, in which bismuth is first precipitated as oxychloride, as being inexact.He obtained good results, however, with the third method, which depends on the solubility of tellurium sulphide in alkaline sulphides. The method was carried out as follows: The powdered sub- stance was dissolved in aqua regia, and the solution repeatedly evaporated on a water-bath with hydrochloric acid until the smell of chlorine had disappeared. After diluting with water as far as possible without causing bismuth oxychloride to separate, and heating to 60" to 70°, the solution was saturated with hydrogen sulphide, the precipitate allowed to settle, and then filtered through a weighed filter-paper placed in a funnel with ground edges, and having a stopcock fitted to its stem. The filtrate and washings were concentrated and again treated with hydrogen sulphide, but one precipitation was generally found to be sufficient to throw down the whole of the elements. The precipitated sulphides were then treated on the filter with warm potassium sulphide solution until all the tellurium was extracted, the residue washed thoroughly with water, finally with alcohol and carbon bisulphide, dried at 1000, and weighed as Bi,S,.The tellurium was precipitated in the filtrate by cautiously adding hydro- chloric acid and gently heating. After filtering off, the precipitate was dissolved in fuming nitric acid, the solution evaporated repeatedly on the water-bath with hydro- chloric acid, the residue dissolved in water and a little hydrochloric acid, and boiled with an excess of barium bicarbonate solution to remove sulphuric acid.In the filtrate from the barium sulphate and carbonate, tellurium was precipitated by any one of the usual methods and weighed as Te. The author finds it best t b use a Neubauer's (platinum) crucible for the filtration and drying of the tellurium, on account of the ease with which it oxidizes. A. G. L. A New Qualitative Teat for Cobalt. J. L. Daneiger. (Journ. Arner. Chem. SOL, xxiv., 578.)-About 5 C.C. of the solution to be tested is acidified with hydro- chloric acid, solid ammonium thioacetate added, followed by a few drops of stannous chloride, and an equal volume of amyl alcohol; mixtures of acetone and ether, or alcohol and ether, may also be used, though less efficiently. The whole is well shaken and allowed to stand. If cobalt is present the upper layer will be coloured an intense blue from the formation of the double salt (CH&OS),Co, SCH,COSNH,.The test is as sensitive as Vogel's (ANALYST, xxvi., 218), and will serveTHE ANALYST. 263 to detect 1 part of cobalt in 500,000 parts of water. The object of adding stannous chloride is to reduce any ferric iron, which would give a deep red colour with the reagent. A. G. L. Estimation of Titanium in Steel and Iron. G. B. Waterhouse. (Chenz. News, lxxxv., 198.)-For the estimation of titanium in steels or pig-irons 5 grammes of the drillings are dissolved in 50 C.C. of strong hydrochloric acid, the solution is evaporated to dryness, the residue baked, heated for a few minutes with about 20 C.C. hydrochloric acid,and the solution boiled, after the addition of 70 C.C.water, until all the chlorides are dissolved. The solution is then filtered, the insoluble matter thoroughly washed with dilute hydrochloric acid, and the volume of the filtrate made up to about 150 C.C. Dilute ammQnia is then added till a small, permanent precipitate is obtained, which is dissolved with a few drops of hydrochloric acid, after which 50 C.C. of a, 20 per cent. solution of sodium sulphite are slowly added, any precipitate or dark colour being removed with a, little dilute sulphuric acid. The solution is then heated to boiling, a hot solution of 50 C.C. glacial acetic acid and 20 grammes sodium acetate added, and the whole boiled briskly for fifteen minutes. The precipitate formed is filtered off, thoroughly washed, and ignited together with the first insoluble residue.The oxides obtained are fused for several minutes with 10 grammes of sodium carbonate ; after cooling the melt is treated with 150 C.C. hot water, the precipitate well washed, transferred to a beaker, any adhering precipitate being dissolved in a little hydrochloric acid, and, after adding 10 C.C. dilute sulphuric acid, the solution is boiled down until sulphuric acid fumes come off, allowed to cool, diluted to 50 c.c., and filtered. The filtrate is made up to a volume of about 150 c.c., made slightly alkaline with ammonia, then just acid with hydrochloric acid, and, after adding 20 C.C. sodium sulphite solution, heated to boiling, when acetic acid and sodium acetate are added as before. After boiling for fifteen minutes, the precipitate is filtered off through a large double filter, well washed, ignited, and weighed as TiO,.The above method was tested on samples of iron to which known quantities of a standard titanium solution had been added, and gave good results, even in the presence of chromium, aluminium, nickel, tungsten, molybdenum, or vanadium, the largest error being found iu the presence of 1-1 per cent. of vanadium, in which case a percentage of 1.048 Ti, instead of 1.016, was found. A. G. L. The Electrolytic Determination of Vanadium. P. Truohot. (Ann. de (%Lirn. anal., 1902, vii., 165-167.)-This method is based upon the fact that a hot, slightly ammoniacal solution of sodium vanadate is decomposed by an electric current, with the deposition of hydroxide or a, mixture of hydroxides of vanadium on the cathode. The solution of the salt, which in the author's experiments contained from 0.012 to 0.05 gramme of VzO, in 200 C.C. was rendered slightly alkaline with 20 to 30 drops of ammonibm hydroxide, heated to 80" to 85" C., and electrolyzed with a current of 2 to 2.5 volts and 0.3 ampere for eight to ten hours, the liquid being meantime kept at the same level by the addition of amrnoniacal warm water. The platinum264 THE ANALYST. cone with the deposit was then washed with water and with alcohol, and ignited in a muffle for a few minutes to convert the hydroxides into vanadic anhydride (V,O,), allowed to cool slowly in a dry atmosphere, and weighed. The results thus obtained were within 0.0005 gramme of the theoretical amounts. C. A. If.