CHARACTERS OF THE SALTS OF IRON. 617 soluble in excess of alkali. Sulphocyanide of potassium in neutral or acid solutions gives an intense blood-red solution; ferrocyanide of potassium, a bright blue precipitate of ordinary Prussian blue. Ferricyanide of potassium occasions no precipitate in solutions of the ferric salts, but the liquid acquires a greenish hue; the ferric salts may thus be distinguished from the ferrous salts. Tincture of galls, in neutral solutions, yields a bluish-black inky precipitate; it is the colouring matter of ordinary writing-ink; this test is rendered much more delicate in its indications by the addition of water holding a little carbonate of calcium in solution in carbonic acid. In neutral solutions the benzoates and the succinates of the alkali-metals give voluminous insoluble precipitates: benzoate or succinate of ammonium or potassium is sometimes employed to separate iron from nickel and cobalt, as the benzoates and the succinates of these metals are soluble. If a solution of a ferric salt to which an alkali has been added till it begins to occasion a permanent precipitate be raised to the boiling-point, it is completely decomposed, and an insoluble basic ferric salt is precipitated this property is often turned to account in the separation of iron from cobalt, nickel, and manganese, which are not precipitated under similar circumstances. When a ferric salt in solution is digested with a bar of zinc in a flask provided with a tube for the escape of the gas, the zinc becomes dissolved, hydrogen is evolved, and the whole of the iron is precipitated as peroxide, whilst a salt of zinc is formed in the liquid. Before the blowpipe both classes of the salts of iron act alike: with borax in the reducing flame they give a green glass, which becomes colourless, or yellowish (if the iron be in large quantity) when held in the oxidating flame. (774) Estimation of Iron.-In estimating the quantity of iron for the purposes of analysis, it should always be first converted into a ferric salt, by boiling with nitric acid or otherwise, after which it may be precipitated by excess of ammonia, and then well washed and ignited: pure sesquioxide of iron remains, consisting in 100 parts of 70 of iron and 30 of oxygen. Iron is thus readily separated from the alkalies and alkaline earths. If magnesia be present, it is apt to be partially precipitated with the oxide of iron, unless the solution contain a considerable quantity of chloride of ammonium. In the presence of tartaric acid, of sugar, and of various other forms of organic matter, ammonia precipitates the peroxide of iron very imperfectly from its solutions: in such a case sulphide of ammonium must be employed. The iron is then thrown down completely as sulphide; this precipitate must be 618 SEPARATION OF IRON FROM OTHER METALS. redissolved in nitric acid, and afterwards the iron may be obtained as peroxide by adding an excess of ammonia to the solution. (775) Separation of Iron from Aluminum and Glucinum.-If alumina and glucina are contained in the liquid, they accompany the peroxide of iron when precipitated by ammonia. When these earths are present the precipitation should be effected by an excess of caustic potash instead of by ammonia; the precipitate should be gently warmed with the liquid, for the purpose of dissolving out the earths. The solution is filtered from the peroxide of iron, which requires long washing with boiling water to remove the last traces of potash. The alumina and glucina are obtained from the alkaline filtrate by neutralizing it with hydrochloric acid, and then adding a slight excess of ammonia; the alumina and glucina are precipitated together, and must be separated in the manner described in (677). (776) Separation of Iron from Zinc, Cadmium, Cobalt, Nickel, and Manganese.-Having precipitated the cadmium by sulphuretted hydrogen, and reconverted the iron into peroxide by boiling the liquid with a small quantity of nitric acid, the solution is to be largely diluted with water, and carbonate of sodium added gradually to the acid liquid until a permanent precipitate is formed, though the liquid remains acid. The solution must be boiled, and the liquid filtered from the bulky precipitate of the basic ferric salt the clear solution must then be slightly supersaturated with carbonate of sodium, and afterwards feebly acidulated with acetic acid on again boiling the liquid, the last trace of iron is thrown down in the form of basic acetate, whilst the other metals are retained in the solution: the precipitated salt of iron must be redissolved in hydrochloric acid, and the iron thrown down as peroxide by the addition of ammonia. Sometimes ammonia in excess is made use of to separate iron from these metals, which all form soluble compounds with ammonia, and which, it is supposed, will retain them in solution ; but this method should never be resorted to in analysis, because the oxide of iron always retains a large quantity of the other oxides. (777) Separation of Iron from Uranium.-The iron having been converted into peroxide is precipitated by a large excess of sesquicarbonate of ammonium, which retains most of the uranium. This process, however, although usually adopted, is imperfect: for if the quantity of iron be at all large, a considerable proportion of uranium is precipitated along with it. (778) Estimation of Ferrous Salt in a Mixture of Ferric and ESTIMATION OF IRON IN ITS MIXED OXIDES. 619 Ferrous Salts.-It often happens that the chemist has to ascertain the relative proportions of protoxide and sesquioxide of iron which a compound contains. If the compound of iron for examination be soluble in hydrochloric acid, the following process by Penny will be found both easy in execution and accurate in its results. It is based upon the power which a solution of the anhydrochromate (bichromate) of potassium in excess of hydrochloric acid possesses of converting ferrous chloride into ferric chloride, while the chromic acid is reduced to the state of chromic chloride ::Ker2,+6 FeCl2 + 14 HCl=3 Fe, Cl+r,Cl +2 KCl+7 H2Ð. In order to carry this process into effect, 44°4grains of pure anhydrochromate of potassium are introduced into an alkalimeter burette (577), which is to be filled up to o° with tepid water; the mixture is to be agitated until the salt is dissolved. Each division of the instrument contains sufficient of the anhydro-chromate to convert half a grain of metallic iron, present in the form of ferrous chloride, into sesquichloride. The ore for experiment having been reduced to an extremely fine powder, 100 grains of it are boiled in a flask for ten or fifteen minutes with about 2 ounces of hydrochloric acid of sp. gr. 1'100: about 6 ounces of boiling distilled water are added, and the mixture immediately transferred to an evaporating basin, taking care to rinse out the flask thoroughly. A white plate is then spotted over with a few drops of a weak solution of ferricyanide of potassium, and the anhydro-chromate is cautiously added from the alkalimeter to the solution of iron (which is kept in continual agitation), until it assumes a dark-greenish shade; as soon as this begins to appear, it must be tested after each addition of the anhydro-chromate, by taking out a drop of the solution on a glass rod, and adding it to one of the drops of the ferricyanide. When the last drop no longer occasions a blue precipitate, the operation is ended, and the number of divisions of the liquid which has been added, when divided by two, indicates the amount of metallic iron which exists in the form of a ferrous compound in 100 parts of the ore. The total quantity of iron present in the solution may be ascertained by making a second experiment on a fresh portion of the ore, and reducing the metal whilst still in the flask with the hydrochloric acid, to the state of a ferrous salt: this is readily effected, either by transmitting a current of sulphuretted hydrogen and then expelling the excess of that gas by ebullition; or by boiling the concentrated solution with metallic zinc; or by nearly neutralizing the liquid with carbonate of sodium, and adding a solution of sulphite of sodium until a drop of the liquid 620 ESTIMATION OF IRON IN ITS MIXED OXIDES. ceases to give a red colour when mixed with a drop of a solution of sulphocyanide of potassium,* placed upon a white plate: the liquid is then boiled, to expel the excess of sulphurous acid. When the iron has thus been reduced to the state of ferrous salt, the whole quantity of the metal present may be ascertained by means of the solution of anhydro-chromate of potassium, as before; the difference between the two results will give the per-centage of metallic iron present in the form of ferric salt. Another excellent process for determining the amount of a ferrous salt present was contrived by Margueritte (Ann. de Chimie, III. xviii. 244). It consists in ascertaining the quantity of a measured solution of permanganate of potassium of known strength, which the cold acidulated and largely diluted solution of iron in hydrochloric acid, can deoxidize and deprive of colour, owing to the reaction expressed in the following equation :— 2 KMnO4 + 10Fe”Cl2 + 16 HCl = 2 MnCl2 + 2 KCl+ 2 The strength of the solution of permanganate is ascertained by dissolving 5 grains of clean iron wire in boiling hydrochloric acid, diluting the solution largely, and ascertaining the number of divisions of permanganate measured from a burette, which it is capable of decolorizing. The total quantity of iron present in an ore or other compound may be ascertained by a second experiment upon a fresh portion of the ore, reducing the iron to the state of a ferrous salt by means of zinc, or otherwise, as described when treating of Penny's process. (779) Analysis of Cast Iron, Steel, and Bar Iron. For this purpose the metal must be reduced to a fine state of subdivision by means of a new file, previously freed from oil by the action of a solution of potash; the fine particles detached are to be sifted through a lawn sieve. Some kinds of cast iron are too hard to admit of being filed; they must be crushed in a small mortar made of hard steel. 1. The proportion of carbon is ascertained by mixing from 50 to 100 grains of finely-divided iron with about 10 times its weight of chromate of lead or of oxide of copper; then placing it in an *The reducing effect of sulphite of sodium on the perchloride of iron may be explained by the following equation, from which it will be seen that the sulphite is converted into sulphate of sodium during the operation; Fe""1⁄2Cl+ H2O+Na,S0=2 Fe"Cl2+2 HCl+Na,S04. ANALYSIS OF CAST IRON, STEEL, AND BAR IRON. 621 apparatus similar to that shown in fig. 288, p. 79, and burning the iron in a very gentle current of oxygen; the carbonic anhydride which is formed is collected in a solution of potash placed in Liebig's bulbs. The tube which contains the iron is gradually heated with charcoal, commencing at the extremity nearest the potash bulbs, and the fire is slowly advanced towards the other end, until, when the operation is completed, the whole length of the tube is red hot. From the quantity of carbonic anhydride thus obtained, the proportion of carbon in the iron may be calculated.* But it has been already explained that cast iron contains carbon partly in chemical combination, partly in a state of mechanical mixture, and it is important to determine the relative proportions of the carbon which exist in these different conditions. This may be effected by dissolving the iron in hydrochloric acid. In this operation all the carbon which was chemically combined with the iron is separated in the form either of a gaseous compound of carbon and hydrogen, or as a liquid hydrocarbon; whilst the scales of graphite mechanically diffused through the metal are not acted upon by the acid, and are left in a solid form mixed with silica. In order to ascertain the proportion of graphite in this residue, it is collected on a small weighed filter, and washed with ether, to remove any adhering liquid hydrocarbon; the filter and its contents are dried at 212°, and weighed in a covered crucible. The residue is then burned, and the silica which remains is deducted from the weight of the precipitate collected on the filter. 2. Nitrogen. Two methods were employed by Boussingault for ascertaining the amount of nitrogen: the first consisted in oxidizing a known weight of iron (by heating it to redness) in a current of steam, condensing the water after it had passed over the iron, and determining the amount of ammonia that it contained, by a method previously contrived by him for ascertaining its amount in rainwater: the second consisted in converting a given weight of iron into sulphide, by heating it with cinnabar, and measuring the amount of nitrogen in the gaseous state by a method exactly analogous to that invented by Dumas for determining the amount of nitrogen in an * A less rapid but accurate plan consists in digesting the iron in filings or in fragments with an excess of normal chloride of copper (EuCl,) dissolved in water; the iron is slowly dissolved, and copper precipitated in its place, whilst the carbon is left in a finely divided condition. The precipitate must be collected on a filter, dried, and transferred to a tube in which it is burned, as above directed. |