THE ANALYST. 345 INORGANIC ANALYSIS. On the Analysis of the Metals of the Platinum Group. N. A. Orlow. Chem. Ztg., 1906, xxx., 714.)-Platinum, iridium, ruthenium, and rhodium, act catalytically on hydrogen peroxide ; osmium, on the other hand, is dissolved by hydrogen peroxide solutions, as is also osmium hydroxide, 0 s (OH),, which is oxidized to the tetroxide. Histological preparations blackened by osmic acid are decolourized by hydrogen peroxide, the osmium going into solution. The reaction may be used to separate osmium from the other platinum metals. Palladium chloride, PdCl,, acts on silver iodide with formation of the black insoluble palladium iodide. The chlorides of the other platinum metals do not show this reaction, which may be used to separate palladium from the other metals.The palladium may be recovered from the precipitate by treatment with potassium iodide or thiocyanate, or else by igniting it and treating the residue with aqua regia. A. G. L. Apparatus for the Determination of Arsenie. A. Kleine. (Chem. ztg., 1906, XXX., 585.)-The apparatus consists of it flask for the determination of arsenic by distillation. The flask has an upwardly-inclined side exit-tube, bent down sharply near its end, and ground into the funnel-shaped top of tt vertical spiral condenser ; the joint is made absolutely tight by keeping a little water in the funnel, The flask itself has a funnel-shaped top, with a central tube reaching to the bottom of the flask,346 THE ANALYST. which is closed during the distillation by means of a ground-in glass rod, this joint also being made quite tight by keeping some liquid in the funnel, The apparatus is made by Strohlein and Go., of Dusseldorf.A. G. L. A New Method of Separating Antimony and Tin. A. Czerwek. (Zeit. anal. Chem., 1906, xlv., 505-512.)-The method is based upon the formation of a double compound of stannic acid and phosphoric acid which is insoluble in water and nitric acid, but is partially soluble in dilute hydrochloric acid. I n order, therefore, to separate tin from antimony by converting the former into this compound it is essential that the nitric acid solution shall be completely free from chlorine ions. I n analysing an alloy, 0.5 gram of the sample is treated at 40° to 50" C. with a solution consisting of 15 C.C. of nitric acid (specific gravity 1*42), 15 C.C.of water, and about 6 grams of tartaric acid, and allowed to stand in a warm place, or at the ordinary temperature, for two or three hours, when the metals will have completely dissolved. The solution is now rapidly heated to incipient boiling and treated with 5 to 30 drops (according to the amount of tin) of a 45 per cent, solution of phosphoric acid (specific gravity 1*3), being meanwhile continually shaken. I t is next diluted with about 300 C.C. of boiling water and the precipitate allowed to subside on the water-bath, which will take about fifteen minutes, The clear supernatant liquid is decanted through a filter, the precipitate washed with hot water containing ammonium nitrate, and while still moist dissolved in warm ammonium sulphide ; the tin is then reprecipitated from the coZd and diluted solution by the addition of sulphuric acid.The beaker is left on the water-bath for some time, and the greenish-grey precipitate separated and washed as before. Finally it is dried, oxidized in a weighed crucible with nitric acid (specific gravity 1-42), dried on the water-bath, ignited, and weighed as tin dioxide. The filtrate from the tin is neutralized with ammonia heated with a sufficient quantity of ammonium sulphide, and acidified with acetic acid. The precipitate of antimony sulphide is left to settle on the water-bath and filtered off while still hot. It is then washed with water containing ammonium nitrate, and dissolved jn ammonium sulphide, the solution evaporated to dryness on the water-bath, and the residue oxidized in the usual way with fuming nitric acid, and weighed as antimony tetroxide.The accuracy of the method is shown by test analyses of alloys containing known quantities of antimony and tin. I n the case of alloys containing other metals in addition the tin is precipitated as described above, and the filtrate neutralized and treated with sodium or ammonium sulphide solution, which precipitates the other metals, whilst the antimony is separated from the new filtrate by means of acetic acid. Electrolytic precipitation from an ammonium sulphide solution has been found a suitable method of throwing down antimony and tin prior to their separation from each other. C. A. &I. A Modification of Schlagdenhaufen's Reaetion for Magnesium.J. Bellier. (Ann. de Chim. And., 1906, vol. 11, pp. 283-285.)-The following method of appIying the colour reaction of magnesium salts with alkali hypoioditesTHE ANALYST. 347 is capable of detecting 1 part in 20,000: About 10 C.C. of the solution of the magnesium salt are mixed with 1 C.C. of a 1 per cent. solution of potassium iodide saturated with iodine, and the mixture shaken with 15 drops of & sodium hydroxide solution. I n the presence of 0.1 gram of magnesium the liquid becomes reddish-brown, and then gives a flocculent deposit of the same colour. With 0.05 gram there is only a brownish to yellowish-red coloration. Ammoniacal salts prevent the reaction altogether, as do also acids and alkalies, whilst calcium salts in large quantity render it less sensitive.The reaction cannot be used for the gravi- metric determination of magnesium, owing to the solubility of the precipitate in water. C. A. M. Iodometric Determination of Vanadic Acid. P. Hett and A. Gilbert. (Zeit. ofentl. Chem., 1906, xii., 265, 266.)-Although other chemists have found that the reaction between vanadic acid and potassium iodide in acid solution is untrust- worthy, the authors state that they have obtained good results by the method. The vanadate salt or ore is fused with sodium hydroxide, the melt dissolved in water, filtered, and the filtrate diluted to a known volume. A portion of the solution is then acidified with either hydrochloric or sulphuric acid, potassium iodide is added, and the liberated iodine titrated as usual. The reaction proceeds according to the equation, V,O, + 2HI = V,O, + H,O + 21.w. P. s. A Qualitative Test for Phosphorus. Mauricheau-Beaupre. (Comptes Rendas, 1906, cxlii., 1206, 1207.)-This is based on the fact that glass heated to red- ness and brought in contact with the vapours of phosphoric acid is rendered opaque. Hydrogen or acetylene burning in a Bunsen burner gives the best flame for the purpose, but either gas must be freed from any hydrogen phosphide by passing it through sulphuric and chromic acids. The test is so sensitive' that acetylene con- taining only 1 part in 10,000 of hydrogen phosphide makes the glass opaque. In applying the test, a piece of glass tubing, 5 t o 10 mm. in diameter, is supported on a platinum wire at the top of the oxidation zone of the flame.I n the case of metals, such as iron, in which the phosphorus is liberated by the action of an acid, the impure hydrogen is conducted into the air-holes of the Bunsen burner, or it can be burnt directly from a metallic jet. Organic compounds are burnt in the upper part of the blue zone of the flame, the glass tube being heated to redness above. The reaction does not take more than one to two minutes, whilst even ten minutes' heating is insufficient to devitrify the glass in the absence of phosphorus. Glass heated in a, pure flame loses in weight, whilst if phosphorus was present it gains in weight (e.g., 0.001 gram on 6.94 grams). I n one experiment the part of the tube rendered opaque was found to contain 0.718 per cent. of phosphorus, none being present in the unattacked glass.Hydrofluoric acid vapour when present in the gas in only small quantities does not attack glass at red heat, whilst compounds of arsenic, or antimony, or boric acid, do not give any deposits that could be confused with the phosphorus deposit. C. A. M.348 THE ANALYST. Determination of the Sodium Phosphates. C. C. Ahlum. (Journ. Amer. Chem. Xoc., 1906, xxviii., 533-537.)-Disodium hydrogen phosphate and trisodium phosphate are both alkaline to methyl orange, and may be titrated with hydrochloric acid, the reactions being : HCl + Na,HPO, = NaH,P04 + NaCl ; 2HC1+ Na,PO, = NaH,PO, + 2NaC1. Two methods are given for the separate estimation of the two salts in a mixture containing both, the first based on the conversion of Na,PO, into Na,HPO, with formation of Na,CO,, when CO, is passed through its aqueous solution, according to the equation GO, + 2Na,PO,+ H,O = 2Na,HPO, + Na,CO, ; and the second depending on the fact that, when a solution containing Na,HPO, and Na,PO, is acidified, and then neutralized with Na,CO,, adding an excess of the latter, the resulting solution contains Na2RP0, and Na,CO,.In both cases the total alkalinity of the original mixture is titrated with I n the first method, 2 grams of the mixed salts are dissolved in water, and CO, passed through the solution until reaction is complete (about ten to fifteen minutes). The solution is then evaporated to dryness, and the Na,CO, in the residue estimated in the Schrotter apparatus, the amount of CO, evolved multiplied by 7.4545 giving the amount of trisodium phosphate present.The number of C.C. & acid, equivalent to the trisodium phosphate, is readily calculated from the above equation, and this, deducted from the number of C.C. used in the titration of the original, gives the number required by the disodium hydrogen phosphate, and this, multiplied by 0.0142, gives its amount. For the alternative method, 1 gram is dissolved in 50 C.C. water containing a drop of methyl orange. A slight excess of HC1 is added, and the solution boiled for ten minutes, after which an excem of Na,CO, is added, and the solution concen- trated as far as possible by boiling, transferred to a tared platinum dish, evaporated on a steam-bath, dried, and weighed. The mass is then pulverized with a porcelain pestle, halved, and the CO, determined in one half by the Schrotter apparatus, the other half being titrated with HC1.The difference between the number of C.C. acid used in the titration of the original mixture and that used in the latter titration, minus the number of C.C. equivalent to the Na,CO, found by the Schrotter apparatus, represents the loss in alkalinity due to the conversion of the Na,PO, into Na,HPO,, and this, multiplied by 2 ~ 0 . 0 0 8 2 , gives the amount of Na,PO,, the difference between the number of C.C. of acid equivalent to the Na,PO, and that required by the original, multiplied by 0-0142, giving the amount of Na,HPO,. HC1 before treatment. W. H. S. Methods for the Determination of Carbonic Acid. W. Holtschmidt. (Chew%. Ztg., 1906, XXX., 621.)-The author uses acid potassium tartrate to decompose the carbonate, the evolved carbon dioxide being either weighed after absorption by soda-lime, or titrated after absorption by caustic alkali.The acid potassium tartrate acts on carbonates only at the boiling temperature, and hence vigorous and continued boiling is necessary to insure complete decomposition. In order to obviate theTHE ANALYST, 349 difficulties caused by the large amount of condensed water obtained, the author has devised special and complicated forms of apparatus, by the use of which very good results are obtained. A. G. L. Difficulties in the Determination of Carbon Monoxide in Gaseous Mixtures. A. Gautier and Clausmann. (BUZZ. SOC. Chim., 1906, XXXV., 513-519.) -In analysing a mixture of nitrogen or air with carbon monoxide, or of nitrogen and various combustible gases with carbon monoxide, it is not possible to recover the whole of the latter gas, either by explosion in the presence of oxygen, or by washing with cuprous chloride solution.Thus after two successive washings as much as 1 per cent. of carbon monoxide may be left in the residual gas, whilst from 0.3 to 0.5 per cent. may remain after explosion of the mixed gases in the eudiometer. If, how- ever, the gases left after the explosion or aft;er the washing (diluted with air or not according to circumstances) be made to circulate through a tube containing iodic anhydride heated to 70" C., the last traces of carbon monoxide are oxidized and an exact determination can be made. I n the ordinary method the small amount of carbon dioxide which may be found in the eudiometer after explosion of the gases left from the washings with potassium hydroxide, bromine, and cuprous chloride, might lead one t o infer the presence of saturated gases, such as methane, in the original mixture, whereas it might really be due to the oxidation of the carbon monoxide that had not been removed by the cuprous chloride. The apparent volume of the residual nitrogen is increased by the uncombined carbon monoxide and that formed from the pyrogallol, but is reduced by the slight oxidation of the nitrogen itself during the explosion in the eudiometer. C.A. M. Determination of Hydrogen Peroxide. W. E. Matthewson and J. W. Colvin. (Amer. Chem. Jozmz., 1906, vol. 36, pp.117-123.)-Hydrogen peroxide may be estimated by diluting with water until the solution is approximately q, and adding this solution from a burette to about 2 grams of ferrous ammonium sulphate dis- solved in a little water, together with a few grams of ammonium sulphate and some phosphoric acid, a solution of titanium, which gives a deep yellow colour with the slightest excess of H20,, being used as indicator. The titanium solution is prepared by fusing the dioxide with about ten times its weight of KHSO,, dissolving the fused mass in cold dilute H,SO,, and filtering. An attempt to determine sodium nitrite by titrating with a standard solution of H202 gave slightly too high results, and for accurate results the titre of the peroxide solution should be fixed with a known amount of nitrite. W. H. S. The Use of Sodium Hydrosulphite in Gas Analysis. H. Franzen. (Berichte, 1906, xxxix., 2069-2071.)-A solution of sodium hydrosulphite is recom- mended as an absorbent for oxygen in gas analysis. The absorption of the gas takes place in accordance with the equation-- Na,S,O,+ H,O + 0 = 2NaHSO,, according to which 64 C.C. of oxygen enter into combination with 1 gram of the salt. A suitable solution for gas analysis is obtained by mixing a solution of sodium hydro-350 THE ANALYST. sulphite (50 grams in 250 c.c.) with 40 C.C. of a solution of 500 grams of sodium hydroxide in 700 C.C. of water. One C.C. of this solution will absorb 10.7 C.C. of oxygen, the absorption being as rapid at low as at higher temperatures, The reagent, which is cheaper than pyrogallol, can be used in the anaIysis of mixtures of gases containing carbon monoxide and dioxide. C. A. M.