INORGANIC ANALYSIS 451 INORGANIC ANALYSIS. Estimation of Traces of Carbon Monoxide in Air. A. Gautier. (Bull. SOC. Chim., 1915, 17, 256-260.)-1n view of a recent paper by Graham and Winmill (ANALYST, 1914, 39,445) on the estimation of carbon monoxide, the author mentions that he used the iodine pentoxide method for the purpose in the year 1890, and that the reaction had been described by Ditte (BUZZ.SOC. Chim. 1870, 13, [ii.], 318). The author found that the reaction between carbon monoxide and iodine pentoxide commences at 35" C., and is rapid and complete at 75' C.; further, ethylene, acetylene, and the vrtpours of benzene and alcohol, w.hen present in air to the extent of 1 part per 1,000 of air, are oxidised partially by iodine pentoxide, and at the same time interfere with the oxidation of the carbon monoxide.When the gases men-452 ABSTRACTS OF CHEMICAL PAPERS tioned are present in qusntity not exceeding 1 ptrt per 100,000 parts of air, they do not affect the oxidation of the carbon monoxide, nor are they thernselves oxidised to any appreciable ex tent . w. P. s. Standardised Coloured Fluids. H. V. Arny and C. H. Ring. (J. Franklin Inst., 1915, 180, 200-213; through J.SOC. Chem. Ind., 1915, 34, 925.)-To produce solutions of definite tint and colour intensity which can be easily prepared and are more readily accessible than the Lovibond tintometer glasses, the authors suggest Tc solutions of simple metallic salts in acid and ammoniacal solvents respectively. (The cmcentration & in all cases refers to the metal present.) The standard red solution is prepared by dissolving 13.5 grms.of roseo-cobaltic chloride (CoCI,,SN H40 H) in sufficient 2.8 per cent. ammonia solution to make 1 litre. The yellc~zu solution is prepared by dissolving 2.1 grms. of ammonium bichromate in 50 C.C. of 1 per cmt. hydrochloric acid, then adding 50 C.C. of 2.8 per cent. ammonia, solution, and diluting with water to 1 litre. The standard blue solution is made by dissolving 12.486 grms.of copper sulphate crystals in 200 C.C. of water, adding 100 C.C. ol 2.8 per cent. amnionia solution, and diluting with water to 1 litre. Other similar series of blue, yellow, and green solutions are made by dissolving cobalt, iron, and copper salts in dilute hydrochloric acid. Pink tints are obtained from a potassium permanganate solution.I t is stated that practically every tint of colour manifested in fluids can be duplicated by the proper blending of eight standard fluids. In use, the various fluids are mixed in such proportions that the total volume employed is exactly 12 C.C. The volume of any solution in the mixture is always a whole number of C.C. If fractions are required, the standard solutions are diluted to one-fifth of their strength.I t is seldom that more than three solutions are required to match a tint. Although most of the solutions are permanent in colour when blended, it is prefer- able to blend when required. The colour of the separate solutions is stable for more than a year. Estimation of Ferrous Iron in Silicates by Titration with Dichromate.0. L. Barnebey. (J. Anzer. Chem. SOC., 1915, 37,1829-1835.)-Ferrous iron cannot be titrated accurately with dichromate in a hydrofluoric acid solution. Moreover, ferrous iron in fluoride solution oxidises so rapidly that a slow titration, such as accompanies the use of an outside indicator, may introduce a considerable error. If, however, the mineral be dissolved in hydrofluoric and sulphuric (or hydrochloric) acid, with due precaution against oxidation of ferrous fluoride, and excess of boric acid then added so as to convert all fluoride into fluoborate (cf.ANALYST, 1915, 334), the tendency to oxidation by the air is entirely overcome, and the resulting solution can be titrated as exactly with dichromate as if fluorine compounds were absent.Attention is recalled to the fact that the sharpness of the end-point in a titration of ferrous salt with dichromate is seriously interfered wit,h by high concentrations of hydrogen chloride. The author uses a 0.05 per cent. solution of potassium ferri- cyanide as an outside indicator. With this indicator, one drop withdrawn from 200 c . ~ . of a solution, as regards hydrogen chloride, and containing 0.1 C.C.of & ferrousINORGANIC ANALYSIS 453 solution, gives a distinct reaction within a minute. With ten times as high a con- centration of hydrogen chloride, about ten times as high a concentration of ferrous salt is required to give an equally quick development of colour. The absolute error due to this cause is much less when comparatively large quantities of ferrous iron are being titrated, and in such cases the percentage error becomes negligibly small.On the other hand, with small quantities of ferrous iron the error is as stated above -so large relatively that the results are quite worthless. Fortunately the observation that large quantities could be estimated exactly suggested a means of estimating smaller ones with equal accuracy.I t is found that in presence of sufficient ferric chloride (5 C.C. or more of an approximately FG solution in 200 C.C. of the ferrous solution to be titrated) even small quantities of iron can be estimated in presence of $ hydrochloric acid with no greater error than is the ca.se when the acid concen- tration is kept quite low-namely, such an error as corresponds to 0.05 C.C.of & dichromate. G. C. J. Estimation of Lead as Sulphite. G. S. Jamieson. (Amer. J . Sci., 1915, 40,157-160; through J. SOC. Chem. Ind., 1915, 34, 925.)-Lead can be pre- cipitated quantitatively from slightly acid solutions by adding an excess of sodium or ammonium bisulphite or of sulphurous acid, provided that excessive acidity be avoided by addition of ammonia.As an example, 22 C.C. of a solution of lead acetate containing 7.832 grms. P b and 10 grms. of free acetic acid per litre were diluted to about 100 c.c., and treated with an excess of 2 per cent. sodium bisulphite solution, the mixture being well stirred, and allowed to settle for not more than an hour. The precipitate was collected in a Gooch crucible and washed with cold water, without allowing the whole of the liquid to drain until the final washing.After drying at 150" C, the lead sulphite weighed 0.239 grrn., corresponding to 0.1733 grm. P b ; error, - 0.0001 grm. The lead was also precipitated with sulphurous acid containing 36.8 grms. SO, per litre. The addition of sodium acetate is recommended in this case, and a large excess of sulphurous acid must be avoided.The method was successfully applied to mixtures of lead with copper and zinc; it is inapplic- able in presence of the alkaline-earth metals, or of metals which are reduced by sulphites. Separation of Manganese as Manganous Sulphide from Alkali and Alkaline-Earth Metals. V. M. Fischer. (J. Russ. Phys. Chenz. SOC., 1914, 46, 1519-1526 ; through J. Chem. Soc., 1915, 108, ii., 487-489.) - The separation of manganese as sulphide by precipitation with an alkali sulphide in presence of ammonia and ammonium chloride is attended with difficulty, the precipitate being usually the rose-coloured, amorphous form, troublesome to filter and wash, and readily oxidisable in the air.The author, having determined the conditions under which the green sulphide may invariably be obtained in the cold, gives the following method for separating manganese as sulphide from solutions containing metals of the alkalis and alkaline earths.To 100 to 200 C.C. of the liquid are added 5 to 15 grms. of ammonium chloride and 50 to 60 C.C. of ammonia solution (sp. gr. at454 A BSTRACTY OF CHEMICAL PAPERS 16' C. 0,895) ; the solution should remain perfectly clear.From a dropping funnel with an orifice 0-75 mm. in diameter 50 to 100 C.C. of ammonium hydrogen sulphide solution freshly prepared from 2.5 per cent. ammonia solution are slowly added, the addition occupying ten to fifteen miniites, and the vessel being meanwhile constantly and vigorously shaken. Conversion of the flesh-coloured manganese sulphide into the green variety proceeds during the addition of the precipitant, and, when it is not irninediately completed, the Erlenmeyer flask should be corked ; the change occupies from a quarter to one and a, half hours, according to the proportion of ammonium chloride present and the rapidity of the precipitation.When all the precipitate is green, the liquid is diluted to 500 to 700 C.C.with cold, recently boiled water, and filtered. The precipitate is washed with dilute ammonium hydrogen sulphide solution containing a, little ammonium chloride, and is subsequently either heated with sulphur in a current of hydrogen and weighed as gulphide, or dissolved in dilute hydrochloric acid and converted into phosphate. This method yields very accurate results, and, as it is not affected by the presence of ammonium acetate, it is applicable after the removal of the iron by the acetate method.When sufficient ammonia is present, and the precipitation is effected slowly, it becomes unnecessary to convert the rose-coloured into the green sulphide, since the former is then obtained in a dense form, which can be readily filtered and washed, and does not undergo oxidation.I n this cme the following quantities should be used : 150 to 200 C.C. of solution, 5 to 10 grnm of ammonium chloride, 50 to 75 C.C. of ammonia solution, and 75 to 100 C.C. of the cold ammonium hydrogen sulphide solution. After diluting with 500 to 1,000 C.C. of cold, boiled water, the precipitate usually settles rapidly, leaving the supernatant liquid clear ; filtration, etc., may then be carried out imme- diately.If, however, the liquid remains turbid, the precipitate is allowed to change into the green form before filtration. Either procedure gives excellent results, which are not affected by the presence in the solution of hydroxylarnine hydrochloride (ZOC. cit.). Method for Controlling or Estimating the Quantity of Nickel Deposited during Nickel Plating.M. Pontio. (Comptes rend., 1915, 161, 175-177.)-The spot-test described depends on the action of a, mixture of mineral acids and hydrogen peroxide on the nickel-plated object. The reagent used is prepared by mixing 10 C.C. of nitric acid (sp. gr. 1.332), 20 c.cI of hydrochloric acid (sp. gr. 1*18), 20 C.C. of hydrogen peroxide (12 vol.), and 50 C.C. of water.The plated object to be tested is dipped for a few seconds in concentrated sulphuric acid, next washed with water, dried with a clean linen cioth, and a drop of the reagent is applied to the surface. After two minutes' contact 1 drop of ammonia is added, and, after the lapse of another minute, the drop is poured from the metal on to a porcelain slab. If it exhibits a blue coloration, the quantity of nickel deposited on the surface, in the case of a copper base, is less than 1 mgrm. per sq.cin. Iron previously coated with copper, and then nickel-plated, will give a brown coloration with the test if the nickel coating is less than 4 mgrrns. per sq. cm. I n cases of doubt, the blue copper coloration may be confirmed by the ferrocyanitle test. w. P. s,APPARATUS, ETC.455 Estimation of Sulphates in Soils. P. E. Brown and E. H. Kellogg. (J. 1w-L and Eng. Chem., 1915, 7, 686-687.)-Attempts to extract sulphates from soil by means of hydrochloric acid lead to inaccurate results, which may be high or low. Gemerally with low concentrations of acid the results are low, probably because of the interference of ferric oxide and organic matter (cf. Van Bemmelen, Landzl:. Vers. Sta., 37, 284). With 2 per cent. of acid the sulphates may be under-estimated by as much as 70 per cent. On the other hand, with 10 per cent. of acid they may be over-estimated by a8 much as 50 per cent., unless means be adopted to separate silica before proceeding to the precipitation of sulphate as barium sulphate. If this source of error be eliminated, the results will be as low as when weaker acid is used for the extraction. It is found that water is an excellent solvent. I t is recom- mended to shake the soil with twice its weight of water for six to eight hours. A shorter period may suffice with some soils, but not if much calcium sulphate is present. Hydrated calcium sulphate, when added to soils in proportions much greater than those commonly obtaining, is completely extracted in eight hours. G. C. J.