INORGANIC ANALYSIS 103 INORGANIC ANALYSIS. Analysis of Mixed and Spent Acids. L. Wuyts. ( J . SOC. Chem. Ind., 1916, 35, 149-151.)-The proportion of nitrous and nitric acids in the mixed acids used for nitrating toluene may be rapidly estimated by a modification of Schloesing’s method. The apparatus required consists of a pear-shaped Kjeldahl flask with a capacity of about 200 c.c., provided with a 40 C.C.tap-funnel and a delivery tube. The stern of the tap-funnel is of capillary tubing, and is bent twice at right angles, with the object of preventing its being immediately above the Bunsen burner, whilst the delivery tube has a tap, and is bent so as to deliver the gas beneath a eudiometer, graduated in 0.1 C.C. I n making an estimation 40 C.C. of ferrous chloride solution (450 grms.per litre with a few drops of HCI), and 30 C.C. of hydrochloric acid (1 vol. of concentrated acid with 2 vols. of water, boiled to expel air), are placed in the Kjeldahl flask, and the stopper with funnel and delivery tube inserted. I n the case of the mixed acids for preparing mononitro-toluene, about 15 C.C. of water are first introduced into the funnel, and suficient run into the flask to expel the air from the tube.The contents of the flask are boiled until all air is removed, and a eudiometer tube filled with water is placed over the end of the delivery tube. To prevent water being drawn back into the flask through back pressure, the end of the delivery tube is formed into a T-piece, one end of which passes up into the eudiometer while the other dips into mercury below the water in the trough.About 0.7 C.C. of the acid is introduced into the tap-funnel by means of the bulb pipette shown in the diagram, and, together with the water already there, is allowed to fall drop by drop into the boiling ferrous chloride solution in the flask. A further 10 C.C. of water is subsequently added to rinse the funnel, and the rinsing is twice repeated after the evolution of gas has nearly stopped.Meanwhile the amount of acid L’ used in the estimation is weighed. After all the gas has been collected, the eudioineter is transferred to a cylinder containing water saturated with nitric oxide, while a fresh estimation may be made without interrupting the boiling. In the case of the mixed acids for preparing trinitro-toluene, about 0.9 C.C.is used, while with (‘ spent acid T.N.T.” a weight corresponding to about 4 C.C. is taken, and the liquid is diluted with 25 C.C. of water. For ‘‘ spent acid M.N.T.” a weight corresponding to about 10 C.C. is required, and 25 C.C. of water should be added. Blank estimations with 20 C . C . of a standard solution of nitrate (16.5 grms. of dry sodium nitrate or 20 grms.of dry potassium nitrate per litre) are made before and after a series of estimations, and the results are conipared with the volumes of the gas yielded by the mixed acids. If V, represent the volume of gas from the standard nitrate solution, and To1 the104 ABSTRACTS OF CHEMICAL PAPERS volume from the saniple of acid of weight W , the percentage of nitric acid corre.: sponds to 24*4629V01+Vy,TV. For preserving the tubes of gas until convenient to read, the best results were obtained by the use of water saturated with nitric oxide, This was prepared by leaving two or more burettes containing the gas in a gas cylinder filled with water, until the volume showed no further sign of reduction when reduced to standard temperature and pressure4 C.A. M. Analysis of a Mixture of Alkali Sulphides, Thiosulphates, and DithionatesL (Bull. Soe. C'him., 1916, 19, 8-9.)--In a mixture of alkali salts of this J. A. Muller. nature the pre-existing sulphates are estimated by precipitation as barium sulphate, in highly dilute solution, slightly acidulated with acetic acid and maintained in an atmosphere of carbon dioxide.I n another portion of the solution the sulphides and thiosulphates are estimated together by iodinietric titration in highly dilute solution slightly acidified by acetic acid. Another portion is placed in a small flask, acidified with acetic acid, and the flask connected with a water suction-pump. The contents of the flask are vigorously shaken for about ten minutes under the vacuum of th0 pump, until all the sulphuretted hydrogen is expelled and the thiosulphates are then titrated iodimetrically.The difference between the two titrations gives the sulphides. I n these titrations the presence of compounds of the thionic series does not interfere, provided the titration be carried out rapidly, as the action of iodine on these com- pounds in the cold is practically negligible.The thionates art! estimated in another portion of the liquid, after acidulating as before and removing the sulphuretted hydrogen in waeuo. The solution is filtered, then made distinctly alkaline with potassium hydroxide, and evaporated to dryness in a silver capsule, with the addition of a little nitre. The dried residue is carefully roasted and the total sulphates are estimated as barium sulphate, from which the equivalent of the thiosulphates and pre-existing sulphates is subtracted.The result may be expressed in terms of dithionate, since in presence of excess of alkali sullphides the tri- and tetra-thionates are decomposed. If the original solution contained disulphides and polysulphides, sulphur would be precipitated on acidification with acetic acid.This sulphur, at first colloidal, coagulates when the last traces of sulphuretted hydrogen have been expelled. I t is then easily collected on st small filter which, after washing, is heated in a silver capsule with potassium hydroxide and nitrate, and the sulphur estimated as barium sulphate. J. F. B. Estimation of Bromine and Iodine in the Presence of Chlorides, L.W. Winkler. (Zeitsch. angew. Chem., 1915, 28, 477-480, 494-496.)-The per- manganate method is recommended for the estimation of bromine in bromides. I n the case of sodium or potassium bromide containing but traces of chloride, about 0.1 grm, of the salt dissolved in 100 C.C. of water is acidified with 25 c.c, of 50 per cent. sulphuric acid, the mixture is boiled, and permanganate solution is run in until a permanent pink coloration is obtained.The permanganate solution should be standardised against pure potassium bromide. When the bromide contains its own weight of chloride, the titration is carried out in a distillation flask, the jet of the burette being fitted into the neck of the flask. After each addition of permanganateINORGANIC ANALYSIS 105 2 C.C.of the liquid are distilled, and this is continued until the distillate is colourless -z.e., until bromine no longer distils over. If iodide is present, iodine distils before the bromine, and can be collected separately; nitrates and carbonates may be present in the solution, but reducing substances and ammonia must be removed previously. In the case of sea-water, the bromine is concentrated by acidifying 100 C.C.of the water with hydrochloric acid, adding sulphuric acid, and distilling with the addition of permanganate, the distillate being collected in dilute sulphurous acid. Traces of iodine do not interfere, but if a large quantity is present, the greater part should be removed by first distilling the water with the addition of sulphuric acid and ferric chloride solution. The distillate is then placed in a flask, sulphuric acid is added, and, when all the air has been expelled by boiling, permanganate solution is run in under the conditions described above.The liberated bromine is collected in hydrogen peroxide solution, and the hydrobromic acid thus formed is estimated either by titration with lirneLwater, by titration with silver nitrate solution after neutralisation with lime-water, or gravimetrically as silver bromide.The following method is recommended for the estimation of bromine in mineral waters, and is similar to the above, except that the distillate containing the bromine is collected in 10 C.C. of A few C.C. of carbon tetrachloride are then added to the distillate, the mixture is acidified with 2 or 3 drops of concentrated hydrochloric acid, and shaken.After fifteen minutes i3a arsenious acid solution ig added in small quantities at a time, and the mixture shaken until the carbon tetrachloride layer is colourless, Iodine, if present, is also titrated ; in this case the carbon tetrachloride changes in colour from brownishLyellow to colourless and then rose-red, at whichpoint the titration is ended.For the estimation of iodine in the presence of bromides and chlorides modifica- tions of the nitrite method are described: The sample is acidified, and shaken with sodium nitrite solution in the presence of carbon tetrachloride ; after several of these treatments, the separated carbon tetrachloride solutions are treated with a small quantity of urea to decompose traces of nitrite, and then titrated with thiosulphate solution, If bromides are present, the bromine may be estimated in the same portion of the sample; the solution remaining after the carbon tetrachloride has been separated is acidified with sulphuric acid, boiled until all nitrous acid and dissolved carbon tetrachloride have been expelled, and then titrated with perman- sodium hydroxide solution.ganate solution as described. w. P. s. Estimation of Carbon Dioxide in Air by Haldane’s Apparatus. R. C. Frederick. (J. SOC, Chem. Ind., 1916, 35, 96-99.)-For tlie estimation of small quantities of carbon dioxide in air, Haldane’s apparatus is the most convenient and, in skilled hands, it gives sufficiently accurate results.The carbon dioxide of the air sample is absorbed by caustic potash, and the consequent diminution in volume, measured on a graduated scale, gives a direct reading of the quantity of carbon dioxide in 10,000 parts of air. The author employs a potash solution of 10 per cent. strength coloured with methyl orange. The apparatus is fully described and illustrated, and complete details are given for its correct manipulation.A more definite reading may be obtained by placing a sheet of black material behind the106 ABSTRACTS OF CHEMICAL PAPERS tubes on which the adjustment marks are situated. Instead of agitating the water- jacket by blowing air through the water by mouth, a rubber bulb is preferably supplied for the purpose. Thin pipe cleaners which can pass right through the stop- cocks serve admirably for dislodging dirt from the glass tubes and air burette, which must be perfectly clean.The simplest operation is when the air to be analysed can be taken directly into the instrument. For analyses at a distance small sample bottles of special pattern are required. The method of collecting tha samples and discharging the contents of the bottles into the Haldane apparatus, with the assistance of a mercury bath, is described.Certain modifications have been introduced in the details of the Haldane apparatus, notably a device for extending the range of the instrument designed to register up to 100 parts of carbon dioxide per 10,000, so that it can be used to record up to 500 parts by the temporary attachment of a special scale.Another adjunct described by the author relates to a counterbalancing device to reduce the physical effort required for raising and lowering the mercury reservoir of the apparatus when several analyses have to be performed. Stops are also provided which render it impossible for the mercury to run over into the potash vessel, or vice versa. J. F. B. Analysis of Copper-Aluminium-Zinc Alloys.H. Graefe, (Chem. Zeit., 1916, 40, 102.)-Alloys consisting of copper, 25 to 27 per cent., aluminium, 14 to 18 per cent., and zinc, 54 to 57 per cent., together with small quantities of silicon, cadmium, tin, lead, and iron, may be analysed as follows: A weighed quantity of about 0.5 grm. of the turnings is dissolved in 15 C.C. of nitric acid (sp.gr. 14), and any insoluble metastannic acid is separated byfiitration. The filtrate is diluted to 150 c.c., heated to 60' to 70" C., and submitted to electrolysis; using a current of 0.2 to 0.4 ampere and 2 to 3 volts, the copper and lead are deposited within forty-five minutes. Solid potassium hydroxide is now added to the solution until the acidity of the latter is nearly neutralised, and the aluminium, iron, and zinc are precipitated together by the addition of sodium carbonate.The precipitate is collected, washed with hot water, dissolved in the minimum quantity of dilute sulphuric acid (1 : 4), and the solution treated with an excess of potassium hydroxide. The solution, containing a small quantity of undissolved ferric hydroxide, is diluted to 150 C.C.and electrolysed ; a copper-coated cathode is employed, and this is rotated at 600 to 800 revolutions per minute, the current being increased from 1 to 5 amperes in steps of 1 ampike every ten minutes. The zinc is thus deposited in forty minutes, and is washed without interrupting the current, then dried with alcohol, and weighed, The iron and aluminium remaining in the electrolyte are then estimated in the usual way.w. P. s. Differential Iodimetry. Estimation of Periodates, Iodates, Bromates, and Chlorates in Presence of Each Other, 0. L. Barnebey. (J. Amer. Chenz. SOC., 1916, 38, 330-341.)-Periodates, iodates, bromates, and chlorates can be esti- mated iodimetrically in presence of each other, and of perchlorate, by regulation of the concentration of reagents, especially the acidity, the temperature, and time of reaction.Periodate reacts completely with iodide in saturated boric acid solutionINORGANIC ANALYSIS 107 containing sufficient borax to diminish the acidity to a slight extent, forming iodate and free iodine ; under these conditions the other salts are unaffected. Iodate as well as periodate is actedon by & iodide in acetic acid solution, and the free iodine can be titrated; bromate, if present, is not unaffected under these conditions, but e simple correction makes the method substantially exact even in presence of bromate.I n hydrochloric acid containing Tv iodide, bromate as well as periodate end iodate is completely decomposed, and the free iodine can be titrated. Finally 6 N hydro- chloric acid acting in presence of TG to iodide decomposes chlorate completely, as well as periodate, iodate, and bromate, whilst perchlorate is unaffected.After rendering the solution alkaline and then acidifying, the iodine can be titrated with thiosulphate. Of a mixture of the above salts, 2 grms. are dissolved and the solution is diluted to 1,000 C.C. Periodate is estimated in one 100 C.C. portion by adding 2 grms.borax, an excess of boric acid, and 20 C.C. of iodide, allowing to stand three minutes, and then titrating the liberated iodine with TG thiosulphate. In another 50 C.C. portion, periodate and iodate are estimated by cooling to 3" or 4 O C. (to limit the tendency of bromate to react), adding 20 C.C. water, 10 C.C. acetic acid, also cooled, and commencing to titrate the liberated iodine with thiosulphafe after one or two minutes.When the titration i s complete, which should not be less than three minutes after adding the iodide and acid, the mixture is allowed to stand for a further equal interval of time, after which any iodine which appears is titrated with thio- sulphate. This is a fairly exact measure of the error due to decomposition of bromate, and is deducted from the consumption of thiosulpbate in the estimation proper.Periodate, iodate, and bromate are estimated in another 50 C.C. portion by adding 40 C.C. water, 10 C.C. iodide, and 5 C.C. 6 N hydrochloric acid and titration of the liberated iodine with thiosulphate. Finally, periodate, iodate, bromate, and chlorate are estimated in a 50 C.C.portion which is placed in 500 C.C. stoppered flask. The air in the flask is replaced by carbon dioxide, 10 to 20 C.C. iodide and 50 C.C. concen- trated hydrochloric acid are added, and the mixture is allowed to stand twenty minutes. The flask is then placed in running water, 50 C.C. of sodium hydroxide solution at least equivalent to the acid used (520 grms. NaOH per 1,000 c.c.) are added all at once and the mixture cooled to room temperature.Hydrochloric acid (1 : 1) is then added slowly until iodine is liberated in quantity, followed by 3 to 5 C.C. more of the acid. iodide, and 20 C.C. The free iodine is then titrated with thiosulphate. G. C. J. Detection of Nitrates in the Presence of Organic Matter. A. Tingle. (J. SOC. Chern. Ind., 1916, 35, 77-78.)-The reagent consists of 3 grms.of salicylic acid in 100 C.C. of concentrated sulphuric acid. A control test should show that the reagent, after being warmed, contains no unsulphonated salicylic acid ; in the absence of nitrates an ethereal extract of the reagent should give no red coloration with ferric chloride solution. The solution to be tested-e.y., 2 C.C. of a saturated solution of cane-sugar, containing 0.1 per cent. of potassium nitrate-is treated with an equal volume of the reagent, and warmed until a charred mass is formed.The char is extracted with 10 C.C. of boiling water, and the solution, after cooling, shaken with an equal volume of ether. One half of the ethereal extract is shaken with aI08 ABSTRACTS OF CHEMICAL PAPERS little aqueous ammonia, which produces an orange colour with the nitrosalicylic acid ; t h e other half is shaken with 2 C.C.oE a 1 per cent, aqueous solution of ferric chloride, which in presence of nitrosalicylic acid gradually forms a red coloration in the aqueous layer. J. F. B. Electrometrie Titration of Vanadium. G. L. Kelley and J. B. Conant. ( J . Amer. Chem Soc., 1916, 38, ;j41-351.)-Vanadates may be titrated with ferrous sulphate, using a change in the electromotive force of a suitable cell as the indicator.It is not necessary to measure the potential involved, and an ordinary suspension galvanometer serves, with a resistance box containing two sliding contacts. The resistance box used by the authors consists of two coils and sliding contacts with resistances of 195 and 5 ohms, and is wired so that one resistance serves to vary the v E.M.F.of the dry cells used as a source of current, and the other serves as the potentiometer box. A represents a calomel electrode, the dotted area representing the mercury, and the bottle being filled with a solution of potassium ni- trate, and fitted with a plunger, slight pressure on which causes liquid through a capillary tube.B is the beaker in which the titration is made, and contains the platinum electrode, E, and the stirrer, 8. The galvanometer is shown a t G, and the resistance box at R, with its two sliding contacts, L and Lz. The batteries-ordinary dry cells-are shown at F, and K is a double knife switch, which throws in both the galvanometer and main circuits at one operation.The solution containing the vanadium as vanadate is placed in the beaker, and 25 C.C. of dilute (1 : 1) sulphuric acid are added, with water enough to bring the volume up to 200 C.C. The connections are then made, and resistances so adjusted that the needle comes to rest on the scale. Standard ferrous sulphate solution is then run in from the burette. At the first addition of ferrous sulphate the needle begins to move off the scale, showing a drop in potential.The sliders on the resistance box are then changed in such a manner as to keep the needle on the scale during the addition of further quantities of the reducing agent. This movement of the needle is slow, and continues until half to three-quarters of the equivalent quantity of ferrous sulphate has been added.During the addition of the last quarter the needle remains practically at rest until the end- point is reached, when the addition of a few tenths of a C.C. in excess causes a sudden movement. To obtain exact adjustment, dilute standard dichromate solution is added in small amount until the needle is returned to the position which it occupied before ferrous sulphate was added in excess.The addition of 2 or 3 4 A the ejection of a portion of theINORGANIC ANALYSIS 109 drops of ferrous sulphate solution is now usually enough to cause a sudden move- ment over several divisions. This is followed by the addition of exactly the quantity of dichromate necessary to return the needle to the original stationary position. The end-point of the reaction is thus indicated by a sharp throw of the needle, which is easily distinguished from the slow movement which occurs when the first portions of ferrous sulphate are added. Sharpness of end-point is promoted by low temperature (preferably 10" C.), high acid concentration, small rather than large amounts of vanadium, and low concentration of chromic and ferric saltg.The mean of the results obtained by this method is substantially identical with the mean of results obtained by reduction with sulphur dioxide, hydrogen peroxide, and oxalic acid, whilst duplicate estimations by the new method agree more closely than do duplicates obtained by the other methods. G. C. J. Rapid Method for Converting Scrap Platinum into Chloroplatinic Aeicl.J. B. Tingle and A. Tingle. ( J . SOC. Chem. Ind., 1916, 35, 77.)-The solution of scrap platinum in aqaa Tegia is a tedious process, necessitating the use of a dispro- portionate amount of acid. The authors employ a far more rapid process by fusing the platinum with fifteen times its weight of zinc under a layer of borax or other flux. The mass is fused for about ten minutes, and the temperature then raised until the zinc begins to boil and its vapour to burn above the layer of flux.Small quantities may be thus treated in a porcelain crucible over a Bunsen burner. The metallic mass is cooled and separated, then dropped into dilute commercial hydro- chloric acid. The zinc dissolves rapidly, leaving a black residue which is washed by decantation. The residue, still containing a little zinc, dissolves rapidly in aqua regia; the solution is evaporated to dryness and then redissolved in warm, very dilute hydrochloric acid. The solution may then be treated with pure zinc sheet or rod, and the precipitate of platinum washed with dilute hydrochloric acid until the washings are free from zinc. Alternatively, the platinum may be precipitated with sulphuretted hydrogen, and the sulphide ignited on the filter. The metallic platinum is re-dissolved in aqua regia, the solution evaporated almost to dryness, the residue dissolved in hydrochloric acid, and the solution again evaporated. The residue is dissolved in water, the solution of chlorplatinic acid filtered and diluted to the desired strength. J. F. B. Estimation of Free Sulphur in Antimony Sulphide Pigments. A. Hutin. (Ann. Chim. anal., 1916, 21, 32-33.)--Carbon bisulphide is unsuitable for the extrac- tion of sulphur from antimony sulphides, since when heated it tends to transform antimonic sulphide into a sulphide, Sb,Sp. Acetone is a more suitable solvent, not- withstanding the fact that the solubility of sulphur therein is very small (0.25 per cent. at 5 6 O C.). For the complete separation of free sulphur from 2 grms. of an antimony sulphide from six to eight hours' extraction beneath a reff ux condenser are required. The extract must be evaporated at a low temperature, preferably under diminished pressure, C. A. M.