THE ANALYST. 123 INORGANIC ANALYSIS. The Estimation of Opalescent Silver Chloride Precipitates. Roger Clark Wells. (Amer. Chem. Joz~rn., 1906, vol. 35, 99.)-A number of small pre- cautions are described, which must be used in connection with the nephelometer (ANALYST, 1905, xxix., 203). It is found that ground-glass standards are to be preferred, and that for every concentration a suitable solution and excess of pre- cipitant are required. It is also shown that the presence of electrolytes both augments the maximum opalescence and hastens its dissppearance through coagulation. The best procedures in a number of special cases are also described. A. G. L. The Volumetric Determination of Copper. G. Denigbs. ( A m . de C h h . anal., 1906, vol. 11, pp. 10-12.)-Ammoniacal solutions of cupric salts form double colourless compounds with alkali cyanides, thus : CUSO, + 4NH4CN = (NHJ,SO, + Cu(CN),.2(NH),CN.The double cyanide thus formed is very stable, but gives up a, portion of its cyanogen on treatment with silver ritrate, the amount varying, e.g., with the degree of alkalinity of the liquid. The author has found that when the quantity of the copper salts taken contains from 1 to 50 mgms. of the metal, whilst 20 C.C. of & potassium cyanide solution and 2.6 grams of ammonia are used, the total amount124 THE ANALYST. of liquid being 120 c.c., each cgm. of copper gives up to the silver an amount of cyanogen equal to 2.5 C.C. of the potassium cyanide solution, or in other words each C.C. of cyanide solution is equivalent to 4 mgms.of copper. In making a determina- tion, a quantity v of the solution of the cupric salt containing not more than 50 mgms. of copper is treated with a volume w’ of ammonia solution containing 2.6 grams of ammonia, and with 20 C.C. of -& potassium cyanide solution (standardized on the silver nitrate solution). Then, after the addition of 1 C.C. of a 10 per cent. solution of potassium iodide, and 99 - (v + w’) C.C. of water, the liquid is titrated with & silver nitrate until there is a faint but persistent turbidity. If n C.C. be used, the value (20 - n) C.C. represents the cyanide taken up by the silver, and (20 - 72) x 0.004 gram gives the amount of metallic copper in the amount, v, of solution taken. I n order to obtain concordant results it is necessary that the volume of w + v’ should not exceed 100.In testing cupric sulphate by this method the presence of zinc does not inter- fere with the results unless it amounts to 40 to 50 per cent, of the quantity of the copper salt. Volumetric Determination of Copper and Mercu~y in Admixture with Each Other.- A quantity v (not exceeding 100 c.c.) of the solution, which should not contain more than 0.5 gram of mercury and 0.1 gram of copper, is treated with 5 C.C. of hydro- chloric acid (specific gravity 1-18>, and 5 C.C. of a solution (1 : 5) of cystallized sodium phosphite, and boiled for five minutes with continual agitation, after which it is filtered, and the filtrate and wastings made up to 150 C.C. The copper is then deter- mined in 75 C.C. of this liquid in the manner described above, with the difference that it is necessary t o add a constant of 0.8 C.C.to the volume of silver nitrate, n, con- sumed, to make up for the modification in the reaction caused by the added salts. The precipitate (containing the mercury) left on the filter is detached with the aid of a fine jet, and treated with 5 C.C. of water, 5 C.C. of hydrochloric acid, and 0.25 gram of potassium chlorate, the whole being heated until solution is complete, and the liquid then cooled and made up to 100 C.C. An aliquot portion (a c.c.) of the solution is now mixed with 12 C.C. of ammonia solution (25’ BQ., S.G. 09), 10 C.C. of TG potassium cyanide solution, 100-n C.C. of water, and 1 C.C. of a 10 per cent. solution of potassium iodide, and titrated with TG silver nitrate solution. If q C.C.be used, (10 - q ) = n represents the amount of cyanogen in terms of TG potassium cyanide solution given up by the mercury to the silver. This figure must be corrected by multiplying by the factor 0.96 if it lies between 0 and 5.5, or by the factor 1.04 and subtraction of 0.45 of the product if it lies between 5.5 and 9.5. The corrected value, c, multiplied by 0.02 gram (one ten-thousandth of the atomic weight of mercury) gives the quantity of that metal in the amount of solution, a, taken. C. A. M. The Separation of Metals of the Arsenic Group. 0. Materne. (Bz~ll. Xoc. Chirn. BeZg., 1905, xix., 241.)-The mixed sulphides of arsenic, antimony, and tin are boiled with a 2 per cent, solution of borax containing some ammonium chloride (amount not stated), and the liquid allowed to cool and filtered.The filtrate contains the arsenic, which is precipitated by means of tartaric acid. The residue of the two other sulphides is treated with a boiling 5 per cent. solution of sodium carbonate (containing some ammonium chloride), which dissolves the antimony sulphide andTHE ANALYST. 125 leaves the stannous sulphide. On treating the filtrate with tartaric acid the antimony sulphide is re-precipitated, whilst the residue of stannous sulphide can be brought into solution by means of a 10 per cent. solution of sodium hydroxide. C. A. M. Determination of Carbon in Steel by Direct Ignition with Red Lead. Charles Morris Johnson. (Proc. Eng . SOC. West. Pen?zsyEvania, 1906, xxi., 586.)- The author has found that for molybdenum and chromium steels the ordinary potas- sium cupric chloride method of determining carbon gives very discordant results, as the carbide obtained appears to oxidize rapidly if exposed to air, or, in some cases a t least, to acid.He therefore burns such steels directly with red lead in a current of oxygen, the resulting carbon dioxide being absorbed in barium hydrate, and the barium carbonate formed filtered off and weighed. The baryta is contained in two large test-tubes, through which the carbon dioxide bubbles. The steel used should pass a 20-mesh sieve and be retained by a 30-mesh, or else it may be in the form of fine drillings. For high carbon steels, 0.5 to 1 gram is taken ; for low carbon steels, 2 or 3 grams may be used.The sample is placed in a weighing-bottle with 4 grams of red lead, and well mixed by shaking; the mixture is then transferred to a porcelain boat and covered over with asbestos, which prevents creeping. The highest heat obtainable should be used for the combustion. Old boats may be cleaned with nitric acid. A combustion by this method takes in all only seventy-five minutes from start to reporting, and results obtained on ordinary steels check exceedingly well with those given by the usual method. A. G. L. Determination of Sulphur in Roasted Zinciferous Ores and Similar Materials. G. Lunge and R. Stierlin. (Zeits. ungew. Chem., 1906, xix., 21.)- The authors show that the Watson-Lunge method is not applicable to materials containing zinc, and recommend the following : Exactly 2 grams of sodium bicar- bonate, the alkali content of which is accurately determined, is mixed in a small nickel crucible with 3.206 grams of the finel? powdered sample and 2 grams of potassium chlorate. The crucible is covered and heated for one hour, for the greater part of the time very gently, but finally to dull redness, without, however, allowing the mixture to fuse.The mixture is then extracted with water to which 25 C.C. of sodium chloride solution, absolutely neutral and free from magnesium salts, have been added. The whole is boiled until sodium chloride commences to separate out, when tihe insoluble residue is filtered off and washed with sodium chloride solution (to prevent iron oxide from passing through the filter). The filtrate is then titrated with hydrochloric acid, using methyl orange as indicator. The difference between the alkali value of the sodium bicarbonate taken and that found corresponds to the sulphur.If more than 6 per cent. of sulphur is present, only 1.603 grams of substance should be taken, and 2 grams of ferric oxide added, to prevent fusion of the mixture. A. G. L. Titanium Chloride in Volumetric Analysis. E. Knecht and E. Hibbert. (Clzem. Zeit. Rep., 1905, xxix., 380.)-The authors have successfully used the strong126 THE ANALYST. reducing action of titanium chloride for the volumetric analysis of other substances, especially coloured organic compounds which form colourless (‘ leuco-” bodies. This is usually effected by the addition of 2 atoms of hydrogen to the molecule.The colouring matter acts as its own indicator in the titration, as the end of the reaction is marked by the solution becoming colourless. The experiments must be conducted in absence of air, owing to the ease with which the “leuco-” bodies are oxidized. The authors also succeeded in accurately estimating inorganic substances by the. same reagent. E. K. H. Analysis of Incandescence Mantles. T. B. Stillman. (Chem. Zeit., 1906, XXX., 60.)-The following scheme of procedure is given : Five mantles are burnt off, broken into small pieces, weighed, and evaporated with sulphuric acid in a platinum basin to dryness. This digestion is repeated four times. When cold, the mass is extracted with cold water, and the solution filtered, if necessary. The residue, if any, is fused with potassium hydrogen sulphate, diluted with water, and added to the main solution.Ammonium sulphide is now added in slight excess, and the precipitate collected on a filter and washed with water containing ammonium sul- phide. The filtrate contains the calcium and magnesium, which are separated in the usual manner, after boiling the solution to drive off hydrogen sulphide, and filtering to remove separated sulphur. The precipitate contains the hydroxides of cerium, thorium, lanthanum, zirconium, aluminium, didymium, and yttrium. I t is dissolved in hydrochloric acid, diluted with water, boiled; a dilute solution of sodium thio- sulphate is added, the whole boiled for one hour, then allowed to stand for twelve hours and filtered, giving precipitate ‘‘ A ” and filtrate ‘ 6 B.” Precipitate ‘( A ” contains the oxides of thorium, zirconium, and aluminium.These are dissolved in hydrochloric acid filtered from precipitated sulphates ; the latter are dried, ignited, fused with potassium hydrogen sulphate ; the melt dissolved in cold water, precipitated with ammonia, filtered, and the precipitate dissolved in hydrochloric acid and added to the main hydrochloric acid solution. The latter is treated with an excess of sodium thiosulphate, boiled, filtered, treated with ammonia, in slight excess, Itgain filtered, and washed with water. The precipitate is dissolved in hydrochloric acid, supersaturated with oxalic acid, boiled for five minutes, allowed to stand for twelve hours, and filtered. Precipitate. Wash with cold water, dry, ignite, and weigh as Tho,.Gives no ab- sorption spectrum. Tho,. Filtrate. Treat with excess of sodium hydroxide, boil for ten minutes, filter, and wash with hot water. Precipitate. Dry, ignite, and weigh as Zr0,. Gives no absorption spectrum. ZrO,. Filtrate. Acidify with hydro- chloric acid, render feebly alkaline with ammonia; boil, filter, wash, and weigh as A1,0,. No absorp- tion spectrum. A1,0,. Filtrate ‘( B ” contains the oxides of cerium, yttrium, lanthanum, didymium, and thorium. The solution is rendered alkaline with ammonia, filtered, washed withTHE ANALYST. 127 cold water, the precipitate dissolved in hydrochloric acid, and the solution treated with dilute sodium thiosulphate solution to precipitate any .thoria, present, Any precipitate obtained is dissolved in hydrochloric acid and added to the hydrochloric acid solution of precipitate ‘‘ A.” The filtrate is precipitated with ammonia, filtered, washed, the precipitate dissolved in sulphuric acid; the excess of the latter is ignited, and the residue dissolved in cold water.The solution is now treated with a saturated solution of potassium sulphate, and the precipitate collected on a filter after standing for four hours, and washed with a, solution of potassium sulphate. Precipitate. Dissolve in dilute hydrochloric acid, add oxalic acid, filter, dry, and ignite. Dissolve the residue in hydro- chloric acid, precipitate with sodium hydroxide, saturate with ohlorine gas, filter, and wash well. Precipitate. Dry, ignite, and weigh as CeO,. Gives no absorption spec- trum. Ce02.Filtrate. Acidify with hydrochloric acid, boil, precipitate with oxslic acid, filter, wash, dry, and ignite. The residue is dissolved in nitric acid, and divided into two parts : Test for Di with the spectroscope. A dark band shows that Di is present. Di,O,. Add ammonium acetate, then am- monia, and filter. Treat the precipi- tate with solid iodine. Lathanurn gives a blue colour, but no spectrum. La,O,. Filtrate. Add a dilute solution of sodium hydroxide ; filter off the pre- cipi t a t e, wash, dry, and weigh as Y,O,. Gives no absorp- tion spectrum. The didymium may consist of praseodymium and neodymium; the salts of the former are green and of the latter rose-red. w. P. s. The Separation of Iodine from Chlorine and Bromine by Means of Hydrogen Peroxide.J. Jannasch and F. Zimmermann. (Berichte, 1906, xxxix., 196, 197.)-The aqueous solution of the mixed halogens (120 to 150 c.c.) is treated with 15 c.c, of glacial acetic acid, and 3 C.C. of hydrogen peroxide (30 per cent.), which effects a quantitive separation of the iodine from the chlorine and bromine. The iodine thus liberated is distilled by means of a current of steam into three connected vessels containing respectively (a) 10 grams of hydrazine sulphate, and 10 C.C. of ammonium hydroxide in 80 to 100 C.C. of water, and (b) and ( c ) 0.5 gram of hydrazine sulphate, and 2 C.C. of ammonium hydroxide in 25 C.C. of water. After the whole of the iodine has passed over, the contents of the absorption vessels are cooled, mixed, and acidified with 30 to 40 C.C.of strong sulphuric acid, and the iodine determined as silver iodide. C . A. M,128 THE ANALYST. triangle or on a piece of platinum foil placed on the bottom of the furnace. The lid of the furnace is also covered with platinum foil. After passing a The Detection and Determination of Chlorate in Sodium Nitrate. L. Grimbert. (Jourrt. Pharm. Chim., 1906, xxiii., 98-100.)- A few C.C. of a, solution of the sodium nitrate are shaken with one drop of aniline, and a little strong sulphuric acid then poured down the side of the tube. In the presence of a chlorate an intense blue zone appears at the junction of the liquids. For the determination of the chlorate a given weight of the sodium nitrate is ignited with a, little cane sugar to effect reduction, the mass extracted with water, and the chloride determined by titration in the usual manner.In this way the author found a sample of sodium nitrate to contain 0.692 per cent. of sodium chlorate. Three other samples of sodium nitxate free from chloride also contained chlorate, but different samples of potassium nitrate examined were found to be quite pure. C. A. M. Use of the Electric Furnace for Burning Magnesium Phosphate Precipi- tates. F. Haussding. (Chzem. Zed., 1906, xxx., 60, 61.)-The construction of the furnace is shown in the illus- ‘I injurious action of ordinary gas-flames on The Determination of Carbon Nonoxide in Air. A. Levy and A. PBcoul. Conzptes Rendus, 1906, cxlii., 162.)-The method is based upon the fact, recorded by Gautier, that carbon monoxide, even when only present in traces in air, reduces iodic anhydride, and also on the colorimetric determination in a chloroform solution of the iodine thus liberated.Acetylene has a similar reducing action, but to a much smaller extent. Thus, air containing 4 parts per 10,000 of acetylene causes only a very slight reduction of the iodic anhydride, whereas with 1 part of carbon monoxide in 10,000 a large amount of iodine is liberated. Hence in testing ordinary air for carbon monoxide no special precautions are necessary, but in the case of industrial gas-products any acetylene, hydrogen sulphide, etc., must be absorbedTHE ANALYST. 129 before making a determination. a solution of potassium hydroxide for the absorption of the iodine. The most accurate results are obtained by using C.A. M. Determination of Water in Superphosphates and in Silicofluorides. Ludwig Schucht. (Zeits. angew. Chem., 1906, xix., 183.)-To determine moisture, as distinct from chemically combined water, in superphosphate, 2 grams of the sample are rubbed up in a glass dish with 20 C.C. of absolute alcohol. After one hour the liquid is decanted through a weighed filter-paper ; the insoluble material is brought on to the filter with more alcohol, washed with ether, dried at 40° C., and weighed. The filtrate is evaporated to dryness, and the residue dried at 120" C. to constant weight. The weighings must be made as quickly as possible. The difference between the original 2 grams and the sum of the two residues gives the moisture. For the determination of free acid, the author recommends the use of his oxalate method, and answers some objections which have been brought against it. TO determine the free acid in commercial siZico-fluoric acid, 50 C.C.are weighed Out, diluted with water to 1 litre, and 50 C.C. of this solution titrated hot with If free hydrofluoric acid is also present, the total acid is determined in this way, and the precipitate produced by potassium chloride and ethyl alcohol by Stolba's method is also weighed. The precipitate may also be titrated with sodium hydroxide. I n the analysis of commercial sodium silico-juoride, moisture is determined by heating 2 grams of the substance for three hours at 100" C. in a current of dry air which has to bubble through sodium hydroxide after passing over the substance, any volatilized acid being thus retained and allowed for.To determine free acid, 3 grams of the sample are moistened with methyl orange in a glass dish, and titrated with To determine the sodium silico-fluoride, 3 grams of the sample are dissolved in 450 C.C. of water, and the solution is titrated hot with 4 sodium hydroxide, using phenolphthalein as indicator. The free acid previously found must be deducted from the result, and the proportion of sample to water should always be that stated, otherwise errors due to hydrolysis will be made. sodium hydroxide, using phenolphthalein as indicator. sodium hydroxide. A. G. L. Determination of Available Plant Food in Soil by Use of Weak Acid Solvents. A. D. Hall and A. Amos. (Proc. Chem. XOC., 1906, xxii., 11.)-Repeated extraction, with water charged with carbon dioxide or with a 1 per cent.solution of citric acid, of soils of known history has shown that the first extraction does not remove the whole of the soluble phosphates, the reaction being a reversible one. With carbon dioxide and water the position of equilibrium is approximately constant for succegsive extractions, but with dilute acetic acid the amount of phosphoric acid dissolved falls for the first four or five extractions, then becoming nearly constant. No support is found for the theory that all soils establish in the soil-water a solution of phosphoric acid of approximately the same composition, and independent of the fertilizers the soil receives. W. H. S. Mechanical Analysis of Soils.J. A. Murray. (Chem. News, 1906, vol. 93, p. 40.)-The author has worked out the following scheme for the mechanical analysis130 THE ANALYST of soils : Five grams of the air-dried fine soil (passed through a 100-mesh sievej were disintegrated in weak ammonia solution and allowed to stand for fifteen minutes, after which the water containing the finer material in suspension was poured off. The coarser material was then introduced into a 200 C.C. Erlenmeyer flask, which was filled with water and attached by a rubber joint to a wide glass-tube of exactly the same diameter as the neck of the flask, from which the flange had previously been removed. The tube used was 147 cm. long and 2.3 cm. internal diameter. Water was next poured gently down the side of the tube so as to fill it, after which it was closed by a cork and inverted in a large glass basin full of water.The cork was then immediately removed and a small weighed porcelain dish placed under the open end of the tube under the water. The particles descending the tube were collected in a, series of fractions, the first porcelain dish being removed and another one substituted five minutes after the first particle had reached the bottom, the second dish being removed after another ten minutes, and the third after another twenty minutes (or else, after another twenty-five minutes, the times as given not corresponding exactly with each other). Nearly the whole of the remainder of the material was deposited in another hour. This last fraction was returned to the beaker containing the quantity of fine material originally poured off, well mixed with it, and the whole then placed in the flask and allowed to fall through the tube as before, the same water being used so as to avoid loss of material.In this way three more fractions were collected, after one, three, and seven hours respectively. The liquid remaining in the flask and tube was then measured and an aliquot part evaporated to dryness so as to give a, seventh and last fraction of finest material. The other fractions collected were also dried and weighed. The preliminary separation into finer and coarser material is necessary to prevent some of the fine particles being dragged down with the coarse material. The method enables a soil to be separated into a number of fractions con- taining particles of nearly uniform size; duplicate analyses made by it are in good agreement with each other.A. G. L. A Rapid Method for the Determination of Moisture in Fuels, especially Coal. E. Graefe. (Braunkohle, 1906, iv., 581; through Chem. Zeit. Rep,. 1906, xxx., 23.)-Ten grams of the coal are heated with 50 C.C. of solar oil or petroleum in EL 75 C.C. distilling flask until 25 C.C. of oil have been distilled off, which should require about five minutes. The distillate is caught in a 25 C.C. measuring cylinder, the amount of condensed water being read off directly; each & C.C. corresponds to 1 per cent. of moisture in the coal. The neck of the distilling flask should onlyreach a short distance above the side-tube, and a narrow condenser tube (above 5 mm.diameter) should be used. The method is stated to give serviceable results. A. G. L. Standardization of Chemieal Disinfectants. (Chem. Trade Joz~rrt., 1906, xxxviii., 4.)-The large variety of tbese now on the market renders desirable their classification according to their efficiency, and since they are sold for germicidal purposes, any comparison should be on this basis. The need for standardization is emphasized by the following table, taken from the Public HeaZth Eagiizeer.THE ANALYST. 131 - Sample. A. B. C. D. E. F. G. H. I. J. K. L. M. Carbolic Acid Coefficient. Price per Gallon, or 10% lbs. 0.02 0.30 8.00 1.00 2.50 1.40 0.30 0.10 0.10 0.90 2.50 0.03 11.00 6 8. d. 0 3 6 0 5 6 0 4 0 0 1 0 0 8 0 0 2 6 7 1 7 6 0 1 8 9 0 1 5 0 1 0 7 6 I 0 3 6 2 0 0 Cost of Disinfectant equivalent to 1 Gallon of Carbolic Acid. 6 s. d. 815 0 018 4 0 0 6 0 1 0 0 3 2 0 1 9 26 5 0 9 7 6 7 1 0 0 0 8 4 0 1 5 66 13 4 0 0 4 W. H. S. Determination of Turbidity of Water. J. F. Liverseege. (Journ. Sot. Chem. Id., 1906, xxv., 45.)-Test type as used by oculists, mounted on a board and attached to a cord running on a pulley, ig viewed through a 2-foot tube, filled with the water under examination, and placed with its end about 6 inches from the type, which is raised or lowered till a set is found that can be easily read. The turbidity figure is obtained by subtracting 2.5 feet from the normal range in feet for the particular type, and is recorded as ( ( turbidity expressed in feet of distance required to read standard type.” If a water is so turbid that 2 feet of it obscures the type entirely, a I-foot tube may be used, still reading at 2-5 feet from the type, but the results cannot be calculated from one tube to the other. W. H. S. Filtration of Finely Divided Precipitates. C. S. Palmer. (Chem. Zeit. Rep., 1905, xxix., 342.)-The author proposes to add 1 or 2 drops of albumen to the liquid to be filtered, and, after stirring, to quickly raise to boiling. The coagulated albumen enables the precipitate to be easily filtered, washed, and burnt. The trace of ash from the albumen is quite negligible. E. K. H.