Page images

dissolved in water, and the silicic acid filtered off and determined in the ordinary way. This analysis is completed in one and a half hours, whilst the error does not exceed 0.1 per cent. D. B.

Speedy Volumetric Determination of Carbonic Anhydride. By W. MARCET (Proc. Roy. Soc., 41, 181-195). The method, described at length, is based on the absorption of carbonic anhydride in a closed receiver by potassium hydroxide, and the accurate measurement of the volume of dry atmospheric air required to re-establish the atmospheric pressure after complete absorption. The volume of air will correspond with the volume of the carbonic anhydride, and the weight of the latter is deduced by the ordinary method of calculation. The process consists in passing the air to be analysed from one of two small gasometers through an absorption apparatus into the other, the amount of absorption being recorded by a pressure-gauge. Air is then added from a small receiver enclosed over mercury until the atmospheric pressure in the whole apparatus is restored, and the volume of air thus added is read off.

Certain mechanical difficulties were at first experienced, and the methods by which they were successfully overcome are described in full. Comparative analyses made by this method and by Pettenkofer's are given; the latter, as a rule, gives slightly lower results than the former, the average difference being about 0.4 per cent.

V. H. V.

Detection of Normal Carbonate in Hydrogen Carbonates of the Alkali Metals. By E. KUHLMANN (Arch. Pharm. [3], 25, 72— 73). A concentrated solution of pure hydrogen sodium carbonate to which a fragment of rosolic acid has been added remains perfectly colourless, even after standing a quarter of an hour. If 1-4 per cent. of normal carbonate is present a rose tint very shortly appears. If over 4 per cent. is present, the colour rapidly becomes purple. With potassium carbonate, the reaction is more sensitive. Previous writers have noted the sensitiveness of phenolphthalein as a test, in fact, it is too sensitive for commercial purposes in this application.

J. T.

Separation of Sodium and Potassium from Lithium, Magnesium, and Calcium by the Action of Amyl Alcohol on the Chlorides. By F. A. GoоCH (Chem. News, 55, 18-19, 29-30, 40-41, 56-57, 78-79).-After criticising the methods previously adopted for the estimation of lithium in the presence of potassium and sodium, and exposing their weak points, the author proceeds to compare the relative solubilities of lithium, potassium, and sodium chlorides in amyl alcohol, and he finds that 1 part of lithium chloride requires 15 parts of this solvent, 1 of potassium chloride 24,000 parts, and one of sodium chloride 30,000; and when used to wash the solid chlorides the quantity of potassium or sodium chloride dissolved is quite insignificant. It is upon this basis that the following accurate method of separating and estimating lithium in the presence of potassium and sodium is founded.

Amyl alcohol is added to the concentrated solution of the chlorides, the mixture is heated gently at first until the water is driven off, it is

then raised to boiling temperature and boiled; potassium and sodium chlorides are precipitated, lithium chloride passes into solution; the liquid is now allowed to cool, a few drops of hydrochloric acid added to convert any lithium hydroxide into chloride, then it is boiled again

to drive off water.

If great accuracy is required, and the quantity of lithium exceeds 10-20 mgrms., the liquid is decanted off, the residual chlorides washed with amyl alcohol, then redissolved in water and re-treated. The combined extracts and washings are evaporated to dryness, treated with sulphuric acid, ignited, fused, and weighed. For details of the precantions required, the original paper should be consulted. Magnesium and calcium chlorides are also soluble in amyl alcohol, they may therefore be separated from sodium and potassium chlorides in the manner described above: in the case of calcium chloride, a second treatment is always necessary. Preliminary experiments indicate the probable extention of the method to nitrates. D. A. L.

Separation of Metals by Oxalic Acid. By C. LUCKOW (Chem. Zeit., 11, 5-6). (See Abstr., 1886, 922.)-Of the metals of group 6, oxalic acid precipitates tin and antimony when occurring as stannous and antimonious salts, but not when in the state of stannic and antimonic compounds. Arsenic is not precipitated, and solutions of arsenious chloride mixed with oxalic acid may be evaporated without loss of more than traces of the metal. From a solution containing the three metals and oxalic acid, ammonia precipitates the tin only. All the metals of group 5 are precipitated by oxalic acid. The oxalates of lead, bismuth, silver, copper, and mercury are almost absolutely insoluble; that of cadmium is very sparingly soluble. All are the denser if thrown down by boiling oxalic acid. Of the fourth group, oxalic acid precipitates nickel, cobalt, manganous, ferrous, and uranous oxides (but not the corresponding higher oxides), also zinc from neutral or moderately acid solutions. They require a longer time for complete separation than those of the fifth group; addition of ammonium chloride or nitrate favours their precipitation, but the best way, especially with zinc, is to evaporate to dryness. Dilute sulphuric and nitric acids do not greatly increase the solubility of the precipitates; hydrochloric and strong oxalic acid have a much greater solvent action. Many of the precipitates dissolve in alkaline oxalates. The exceptions are the strontium, barium, calcium, silver, lead, and mercuric salts. The barium, magnesium, and mercuric oxalates dissolve in ammonium chloride. M. J. S.

Analysis of Copper. By W. STAHL (Dingl. polyt. J., 262, 277— 278). In estimating copper by electrolysis, it is recommended to remove all traces of lead and silver, as the latter is deposited at the cathode together with the copper whilst the lead is retained by the anode, thus causing too much resistance to the passage of the current. The addition of nitric acid to the sulphated electrolyte is said to exert a beneficial effect on the passage of the current, and prevent the volatilisation of arsenic and antimony at the cathode. In most cases, the electrolytic action may be continued until the colour of the

electrolyte has been entirely removed without the risk of depositing antimony or arsenic. Bismuth, if present, is separated with the copper and is removed by dissolving the total deposit in dilute nitric acid, treating the solution with hydrochloric acid, evaporating to dryness, and precipitating the bismuth as oxychloride. This treatment should be repeated for the purpose of removing basic cupric chloride which is carried down with the oxychloride. The precipitate is then dissolved in dilute nitric acid, the solution neutralised with ammonia, precipitated with ammonium carbonate, ignited, and weighed as bismuthic oxide. D. B.

Employment of Nitroso-ẞ-naphthol in Quantitative Analysis. By G. v. KNORRE (Ber., 20, 283-290; compare Ilinski and v. Knorre, Abstr., 1885, 840; 1886, 100).-Copper can be separated from lead, cadmium, magnesium, manganese, mercury, zinc, and other metals, by precipitation as copper nitroso-B-naphthol, Cu(O°C10H6 NO)2. This is obtained as a coffee-brown precipitate of metallic lustre. The separation is effected by neutralising with ammonia the solution, which must contain the metals as sulphates or chlorides, acidifying with a few drops of hydrochloric acid, heating nearly to boiling, and adding excess of nitroso-8-naphthol dissolved in 50 per cent. acetic acid; after remaining for some hours in the cold, the precipitate is collected, dried, ignited, and weighed as cupric oxide. The separation of iron from chromium, manganese, nickel, zinc, &c., can be effected by the method previously given for its separation from aluminium (loc. cit.). Full details of the methods and of the results of test analyses are given in the original. A. J. G.

Detection of Alum in Flour. By J. HERZ (Dingl. polyt. J., 262, 96).—A glass cylinder is filled from one-fourth to one-third full with the flour to be examined and the flour is moistened with water. It is then treated with a few c.c. of alcohol and two drops of a freshlyprepared solution of logwood (5 grams logwood in 100 c.c. of alcohol). After agitating the mixture, the cylinder is filled up with a saturated solution of sodium chloride. The depth of colour of the salt solution is then compared with that obtained by treating pure flour and flour containing 0.01 per cent., 0.05 per cent., and 0.1 per cent. of alum in a similar manner. The colour is permanent for days, being blue in presence of 0.05-01 per cent. of alum, and violet-red with 0.01 per D. B.


Detection of Free Sulphuric Acid and of Aluminium Hydroxide in Aluminium Sulphate. By K. J. BAYER (Chem. Zeit., 11, 53).—Referring to Hager's note (this vol., p. 182), the author remarks that free sulphuric acid is detected more readily and with greater certainty by the use of tropaolin as previously described (Abstr., 1886, 281; 651). If aluminium sulphate contains aluminium hydroxide which passes into solution as basic salt, it can be quantitatively determined by titration with normal sulphuric acid, using tropaolin as indicator. The turbidity noticed by Hager (loc. cit.) when dissolving aluminium sulphate in 2 parts of water, is due either

to silicic hydroxide or to the basic sulphate, 3Al2O3,2SO3 + 9H2O (this vol., p. 448). D. A. L.

Volumetric Estimation of Manganese. By C. MEINECKE (Chem. Zeit., 11, 137-138).-The author, in his earlier determinations of manganese by the chlorate precipitation method (compare Hampe, Abstr., 1885, 101), generally found that when iron was present, some of the manganese passed into the filtrate. This is now traced to the use of warm water for diluting. It appears that in presence of iron, although not otherwise, warm dilute nitric acid dissolves some of the precipitated manganese peroxide. When, after precipitation, the liquid was cooled and diluted with cold water, the whole of the manganese was found in the precipitate. M. J. S.

Electro-dissolution and its Use in Analysis. By H. N. WARREN (Chem. News, 55, 62).-By placing iron boride at the positive pole with a platinum plate as the negative pole of a powerful battery, and using dilute sulphuric acid as a solvent, in 12 hours all the iron had dissolved, whilst the boron and other impurities were precipitated. A similar result was obtained with ferrous sulphide, whereas with silicon-iron and phosphide of iron, zinc, &c., the silicon and phosphorus respectively were only partially precipitated. hydrochloric acid, copper may be separated from arsenic, iron, zinc, &c., by electro-dissolution; and in ammonia, copper silver, zinc arsenic, and other alloys, may be resolved into their constituent metals. Electro-dissolution has also been applied to the preparation of unstable compounds, such as potassium ferrate, &c. D. A. L.


Detection of Adulteration in Metallic Nickel and other Metals, by the Magnet. By T. T. P. B. WARREN (Chem. News, 55, 37). In a mixed sample of nickel cubes, some were found to be nonmagnetic, and the sample was easily separated into two lots by means of a magnet. The non-magnetic cubes were a trifle whiter and had not the striated structure noticeable in the magnetic cubes, but there was no marked difference. On analysis, the following results were obtained::

Ni. Cu. C. Sio. Fe. As. Sn. Magnetic..... 96.67 0.08 0.07 0.41 2:46 0.12 0.75 Non-magnetic... 63.69 33.78 0.37 0.16 0.84 0.87 0.46 Other experiments show that the magnetic properties of both nickel and cobalt are affected by alloying with para-magnetic metals. Moreover, manganese and other metals are met with in commerce without magnetic properties.

When nickel apparatus is heated over a Bunsen flame, a black deposit forms, consisting mainly of graphitoidal carbon with traces of nickel, iron, and silica; therefore, the naked flame should not impinge on such apparatus. Other modes of heating are suggested. D. A. L.

Detection and Determination of Traces of Chromium. By E. DONATH and R. JELLER (Chem. Centr., 1887, 151).-If a solution

containing a chromic salt is boiled with sodium carbonate and potassium permanganate, the chromium is oxidised to chromic acid. On re-boiling with addition of alcohol, the permanganate is reduced and the manganese precipitated together with any iron, aluminium, manganese, or alkaline earths previously present. The chromate may be recognised by its colour or by adding sulphuric acid, potassium iodide, and starch. This reagent is more sensitive than hydrogen peroxide. For determination, the solution is re-boiled with alcohol after acidifying, and the chromium is precipitated by ammonium sulphide as usual. M. J. S.

Analysis of Chrome Iron Ore. By W. VENATOR and E. ETIENNE (Chem. Zeit., 11, 53).-The mineral is decomposed by fusion with caustic soda, and when cold is dissolved in hot water, treated with hydrochloric acid, evaporated to dryness, and dried for some time at 120°; the silica is then separated in the usual manner. Ammonium chloride and ammonia are added to the filtrate, and calcium and magnesium are determined in the solution; whilst the precipitate of aluminium, iron, and chromium hydroxides is weighed and fused with soda, the fused mass is boiled out with water and mixed with excess of ammonium carbonate. The solution is evaporated to dryness and treated with water, the sodium chromate then passes into solution, leaving the iron and aluminium oxides behind. To estimate the chromium, the chromate is reduced by means of sulphurous anhydride, the excess of the latter boiled off, and the chromnium precipitated by ammonia in a platinum vessel. Concordant results have been obtained. D. A. L.

Determination of Organic Matter in Air. By T. CARNELLEY and W. MACKIE (Proc. Roy. Soc., 41, 238-247).—Of the methods proposed for the estimation of organic matter in air, one consists in causing the air to bubble slowly through a dilute solution of potassium permanganate until the latter has become quite or nearly decolorised, any undecomposed permanganate being subsequently estimated by oxalic acid; the other consists in passing the air through distilled water in which the free and albuminoïd ammonia are subsequently determined by Wanklyn and Chapman's process. Both these methods, however, are tedious, and uncertain as regards the decomposition of the permanganate in the first, or the absorption of the organic matter by the water in the second. For the process described below, the authors claim rapidity and simplicity, a greater probability of exactness, and a more general applicability. The method proposed consists in a colorimetric estimation of the diminution of colour of a potassium permanganate of known strength, that is to say, the fractional bleaching of the solution effected by a given volume of air. The tint of the permanganate after passage of the air is compared with that of a standard solution of known tint. The method of procedure and the necessary precautions are described in full. Although the method gives concordant results, yet it is fully allowed that the organic matter is not directly estimated by it, that other gases likely to be present in air, such as hydrogen sulphide, nitrous and sulphurous acids, equally decolorise the permanganate, and that as the organic

« PreviousContinue »