INORGANIC ANALYSIS 511 INORGANIC ANALYSIS. Two Methods of Separation of the Metals of the Alkaline-Earth Group. A. G. Paterson. (J. *4mer. Chsm. Soc., 1915, 37, 2346-2352.bThe methods described for separations in qualitative analysis are based on the fact that, in the preeipitation of a mixture, the separation of the least soluble component is always completed first, and that if a relatively insoluble precipitate be treated with a reagent capable of forming a still less soluble compound by double decomposition, such exchange takes place completely. Method I.-A solution containing a mixture of salts of the alkaline-earth methods is treated with an excess of a mixture of ammonium carbonate and sulphate.Tho barium is precipitated as sulphate, and the strontium and calcium as carbonates.The liquid is brought to the boil and allowed to stand for twenty to thirty minutes with occasional stirring, in order to complete the reaction. Some magnesium may be precipitated. The mixture is filtered and the precipitate washed; the latter is removed from the filter with a minimum quantity of water, and boiled with 1.5 vols. of 1 per cent. acetic acid. The undissolved barium sulphate is filtered off, and the filtrate made slightly alkaline with ammonia, then treated with an excess of ammonium carbonate and oxalate together. The liquid is again brought to the boil, and allowed to stand for twenty to thirty minutes with stirring.This precipitate consists of calcium oxalate and strontium carbonate. It is filtered off, washed, and agitated vigorously with 1 per cent.cold acetic acid; the calcium oxalate remains, and thO strontium earbonate is dissolved. A small amount of magnesium may pass into solution as magnesium oxalate. The essential points for a, clean separation are the complete washing of the precipitates, and the digestion for about thirty minutes before filtering. Method I1.-This separation depends on the fact that the solubility of lead sulphate stands between that of barium sulphate and strontium sulphate. A portion of the mixture is first tested for barium by potassium chromate and the flame.The remainder is treated with ammonium chloride and sulphate, the mixture being brought to the boil. The precipitate consists of barium and strontium sulphates, with, perhaps, a little calcium sulphate ; most of the calcium remains in the filtrate.The precipitate is washed, transferred to a beaker, and digested with an excess of a normal solution of lead acetate ; the mixture is warmed for a few minutes, but must not be boiled ; it is then filtered. The whole of the barium sulphate remains on the filter with lead sulphate ; the strontium sulphate is completely decomposed.The sohtion contains strontium, calcium, and lead acetates. The lead is removed by sulphilretted hydrogen, and the strontium and calcium separated by Method I. It is noted that a separation of cadmium sulphide from copper sulphide may be effected on the same principle by digestion with lead nitrate, the lead displacing the cadmium. J. 2’. 13. Reagent for the Detection and Colorimetric Estimation of Aluminium.F. W. Atack. (J. SOC. Chem. Ind., 1915,34,936-937.)-The reagent used is a 0.1 per cent. aqueous solution of alizarin (red) S, the sodium salt of alizarinmonosulphonic512 ABSTRACTS OF CHEMICAL PAPERS acid. The solution is filtered before use, and is pale yellow in the presence of acids, but gives a purple coloration with alkalies.Detection of AZurrtinium.-To about 5 C.C. of the solution under examination is added 1 C.C. of the reagent and then ammonia until the solution is alkaline, as shown by the purple colour. The solution is boiled, cooled, and acidified with acetic acid, when a red coloration or precipitate remaining is conclusive evidence of the presence of aluminium. The red calcium, strontium, barium, zinc, and magnesium salts, and salts of other metals later than Group II., are readily soluble in acetic acid, and do not interfere with the aluminium reaction.The precipitate obtained with cobalt is not dissolved by acetic acid if a large quantity -of ammonia has been used; should cobalt be present (for detection of cobalt, cf. Atack, ANALYST, 1915, 414), an excess of ammonia must be avoided.The reaction with aluminium is not affected by the presence of phosphates or of chromium, and only by comparatively large amounts of iron. The test will detect the presence of 1 part of aluminium in 10 million parts of water. Colorimetric Estimation of AZu- milzizm.-From 5 to 20 C.C. of the solution (containing from 0-005 to 0.05 mgrm. of aluminium) are acidified with hydrochloric acid, 10 C.C.of glycerol and 5 C.C. of the reagent are added, the solution is diluted with water to 40 c.c., and then rendered slightly ammoniacal. After standing for five minutes, the solution is acidified with acetic acid, the latter being added until no further change in the coloration occurs. The mixture is then diluted to 50 C.C. and compared with a standard. Iron and chromium may be prevented from interfering by the addition of a citrate to the acidified solution before adding the ammonia.w. P. s. Detection of Free Carbon Dioxide in Water. L. W. Winkler. (Zeitsch. angew. Ckem., 1915, 28, 376.)-The presence of free carbon dioxide in moderately hard waters may be detected by adding two drops of 10 per cent. copper sulphate solution to 100 C.C. of the water; if free carbon dioxide is absent, the solution becomes turbid, but if the gas is present the solution remains.clear.If the clear solution is shaken with air, the carbon dioxide is expelled and a turbidity is produced ; if a turbidity does not develop under these conditions, the water contains a non- volatile acid (humic acid) or a large quantity of organic matter.The copper sulphate test fails, however, in the case of soft waters, but free carbon dioxide may be detected in these waters by adding ten drops of 1 per cent. alizarin solution (in alcohol) to 100 C.C. of the water. A blue-red coloration indicates the absence of free carbon dioxide ; a copper-red coloration, a small quantity ; a, reddish-yellow coloration, a moderate quantity; and a, yellow coloration, a large quantity of the gas.w. P. s. Estimation of Hydrogen in Gas Mixtures by Catalytic Absorption. E. Bosshard and E. Fischli. (Zeitsch. angew. Chem., 1915,28,365-366).-The hydro- gen is absorbed by a mixture of sodium oleate and metallic nickel ; the nickel is pre- pared by reducing nickel oxide in hydrogen at 340' C. The gas under examination should be freed previously from carbon dioxide, heavy hydrocarbons, oxygen (by ammoniacal cuprous chloride solution, and not by means of phosphorus), and carbon monoxide ; it is then passed into the absorption pipette, which contains concentratedINORGANIC ANALYSIS 513 sodium oleate solution mixed with 3 per cent.of its weight of reduced nickel, shaken with this mixture for three minutes, then passed into a second pipette containing a similar mixture, and, after the addition of 1 C.C. of alcohol to break down the foam, the residual gas is passed back in the measuring burette.The hydrogen is removed com- pletely from the gas by this treatment, whilst any nitrogen and methane which may be present are not absorbed. The results obtained for hydrogen agree well with those found by the ordinary explosion method.w. P. s. Composition and Analysis of Lime-Sulphur Liquors. J. Bodnar (Chem. .&it., 1915, 39, 715-716.)-The lime sulphur solutions employed as insecti- cides in agriculture, and prepared by boiling together water, lime, and sulphur, contain mainly calcium polysulphides and calcium thiosulphate with a small quantity of calcium sulphate; it is very improbable that calcium sulphite is present, since in aqueous solution this salt reacts with polysulphides and yields thiosulphate.Analyses of different lime-sulphur liquors showed that they contain per 100 C.C. : Sulphur as thiosulphate, 2.84 to 3-50 grms. ; as polysulphides, 10.20 to 16.29 grms. ; as sulphide, 2.84 to 3.61 grms. ; as sulphate, 0.1 grm.Concentrated lime-sulphur liquors are also on the market; these contain sulphur as polysulphides, 31-26 to 33-80 grms., and as thiosulphate, 1.85 to 1.92 grms. per 100 C.C. The author proposes a method for the estimation of the polysulphide, sulphide, and thiosulphate sulphur in lime-sulphur liquors which depends on the reactions shown in the following equations : (1) CaS, + 2AgN0, = Ag2S + S,.+ Ca(NO,), and CaS, + SAgNO, = Ag2S + s4 + Ca(NO& ; (2) CaS,O, + 2AgNO,= Ca(NO,), + Ag,S,O, and Ag2S,0, + H,O =Ag2S + H,so4. Ten C.C.of the sample are diluted to 100 c.c., and 10 C.C. of the solution are added to a, mixture of 50 C.C. of & silver nitrate solution and 20 C.C. of water ; after being shaken, the mixture is diluted to 100 c.c., and filtered. The filtrate is collected in 8 dry flask, which is then removed, and the precipitate washed with warm water, dried at 100" C., and weighed.Fifty C.C. of the filtrate are now mixed with 30 C.C. of ;v sodium chloride solution and 15 C.C. of FG sodium hydroxide solution, phenolphtba- lein is added, and the mixture titrated with TG sulphuric acid ; potassium chromate solution is then added, and the mixture titrated with Tv silver nitrate solution.The quantities of the various forms of sulphur present are then calculated from the following formulq the results expressing grms. per 100 C.C. : Thiosulphate sulphur, t , = 100 x 4(a - b) x 0.0016035 ; sulphide sulphur, s, = 100 x 2(25 + c - d ) x 0.0016035, where a = C.C. of =& sodium hydroxide solution, b= C.C.of & sulphuric acid, c = C.C. of & silver nitrate solution, and d = C.C. of & sodium chloride solution used ; poly- sulphide sulphur, p , = lOOe - 7*727(s + t/2), where e = weight of the precipitate. Sodium chloride is sometimes used in the preparation of the liquor, and in such cases its quantity must be estimated separately, and the corresponding quantity of silver chloride subtracted from the weight p .To estimate the sodium chloride, 20 C.C. of the 10 per cent. solution of the sample are rendered ammoniacal, hydrogen peroxide is added, the mixture boiled until the excess of hydrogen peroxide has been decom-514 ABSTRACTS OF CHEMICAL PAPERS posed and all free ammonia expelled, and the solution then titrated with =& silver nitrate solution.w. P. s. Nephelometric Estimation of Phosphorus. P. A. Kober and G. Egerer. (J. Amer. Chem. SOC., 1915, 37, 2373-2381.)-By a modification of the reagent of Pouget and Chouchak, a very sensitive test for phosphates, suitable for nephelo- metric determinations, has been obtained. The reaction is comparable in point of sensitiveness with Nessler's reaction, 1 mgrm. of phosphorus in 2 litres giving a very marked suspension, while the extreme limit of sensitiveness is 1 part in 20 millions.The reaction presumably depends on the formation of a phosphomolybdic acid com- plex of strychnine, and, by substituting hydrochloric acid for the nitric acid originally recommended, the authors have succeeded in preparing a stable and colourless reagent giving quantitative and constant results. The components of the reagent are sodium molybdate, strychnine sulphate, and hydrochloric acid.The sodium molybdate must be prepared from pure molybdic acid free from ammonia ; 35.5 grms. of this are boiled for about one hour with 50 C.C. of sodium hydroxide solution con- taining 400 grms. of 96 per cent. caustic soda per litre. The liquid is diluted to 84 c.c., and shaken until practically all the sodium molybdate is dissolved.A little talc powder is added, and the solution filtered and evaporated to dryness; the residue is extracted by grinding with 40 C.C. of 95 per cent. alcohol, and washed on a filter with several portions of 20 C.C. of alcohol, I t is dried at about 50" C. The strychnine solution is made by dissolving 2 grrns.of pure strychnine sulphate in hot water and making up to 100 C.C. when cold. Fifty C.C. of strong hydrochloric acid are diluted to 100 c.c., and 5 C.C. of this are again diluted to 100 C.C. and adjusted after titration; 24 C.C. of the acid should be equivalent to 30 C.C. of alkali. To prepare the reagent, 1.5 grms. of the pure sodium molybdate are dissolved in 2.5 C.C.of water, and 10 C.C. of the standard hydrochloric acid added while shaking ; the precipitate first formed should redissolve ; 1 C.C. of strychnine solution is then added, and the solution allowed to stand over night. I t must be filtered through an acid-extracted paper-e.g., C. S. and S., No. 575 or 589; ordinary filter- paper produces a cloudiness. A standard solution of potassium dihydrogen phos- phate is made up with 0-1 grm.of the salt in 2 litres of water, and 100 C.C. of this diluted to 1 litre serves as the standard for the nephelometer. For the precipita- tion, 5 C.C. of the reagent are added to a mixture of 30 C.C. of water with 5 C.C. of hydrochloric acid, and the solution shaken thoroughly; 10 C.C. of phosphate solution are then added slowly from a pipette, the solutions are mixed by gently rotating the flask, and the mixture is ready for observation after standing for three minutes. A number of readings have been made with the standard solution at various dilutions and curves plotted ; the use of the instrument and the relations exist- ing between the concentrations of turbidity and the readings are discussed.The completeness of a reaction may be tested by precipitating a standard solution and then diluting an aliquot portion of it, comparing it with the standard similarly diluted before precipitation.The divergence between the constants calculated from the two series of readings indicates the error due to incompleteness of reaction in dilute unknown solutions. In the present case a very fair agreement was found, indicating satisfactory conditions of precipitation for quantitative work.J. F. B.INORGANIC ANALYSIS 515 Estimation of Spelter Coating on Sheets and Wire. J. A. Aupperle. (Chem. Engineer, 1915, 22, 127-128.)-1t is suggested that the weight of the coating upon both wire and galvanised sheets should be expressed in ounces per square foot, and that the lengths of wire should be such that the area of the coated surface is equal to 5.079 Rquare inches.The samples of the sheet should also have the same area (2;t inches by 2$ inches), so as to obtain the final result without calculation. In dissolving the zinc, a, small amount of antimony chloride is added to the acid with the object of forming a thin film over the base of iron or steel and preventing its solution.Sheets.-Five samples (24 inches by 2a inches) are weighed together and immersed separatelyfor one minute in 100 C.C. of hydrochloric acid (sp. gr. 1-20), to which has been added 5 C.C. of antimony chloride solution, made by dissolving 20 grms. of antimony trioxide in 1,000 C.C. of hydrochloric acid (sp. gr. 1.20). The pieces are then washed and scrubbed under running water, dried with a cloth, and left for a few seconds in a warm place.The loss in weight of the five samples divided by 5 gives the weight of coating removed. Each grm. corresponds to 1 ounce of coating per square foot. Wire.-A small section of the sample is stripped with acid as described, and the diameter of the black wire measured to ascertain the length of wire required for the test. This may be found by reference to the subjoined table. The wire used in the estimation is cleaned with carbon tetrachloride, and treated in a tall glass cylinder with hydrochloric acid (sp. gr. 1-20> containing 2 to 3 C.C. of antimony chloride solution of the same strength as used on the sheets. After one minute’s immersion the wire is removed, scrubbed, dried, and weighed, as described above. For direct comparison with the weight of coating on galvanised sheets the loss in weight should be doubled, the sheets being coated on each side. Lengths of Wire to give Grms. of Coating Equivalent to Ouncesper Squa,re Foot. Gauge Number. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Diameter : Inch 0.340 0.300 0.284 0.259 0-238 0.220 0.203 0.1 80 0.165 0.148 0.134 0.120 0.1 09 0.095 0.083 0.072 0.065 0.058 0.049 Length for Test. Om’. 12.1 13.7 14-5 15.9 17.3 18.7 20.2 22.8 24.9 27.7 30.6 34.2 37.7 43.2 49.5 57.0 63.2 70.8 83.8 C . A. M.