Substances containing oxide of iron yield the whole of the iron as metal when fused at a high temperature with charcoal and suitable fluxes. The metal, however, will contain varying proportions of carbon and other impurities, and its weight can only afford a rough knowledge of the proportion of the metal in the ore. There are two or three methods of dry assay for iron, but they are not only inexact, but more troublesome than the wet methods, and need not be further considered. Chalybite and the hydrated oxides dissolve very readily in hydrochloric acid; hæmatite and magnetite dissolve with rather more difficulty. Iron itself, when soft, is easily soluble in dilute hydrochloric, or sulphuric, acid. Pyrites, mispickel, &c., are insoluble in hydrochloric acid, but they are readily attacked by nitric acid. Certain minerals, such as chrome iron ore, titaniferous iron ore, and some silicates containing iron, remain in the residue insoluble in acids. Some of these yield their iron when attacked with strong sulphuric acid, or when fused with the acid sulphate of potash. Generally, however, it is better in such stubborn cases to fuse with carbonate of soda, and then attack the "melt" with hydrochloric acid.
When nitric acid, or the fusion method, has been used, the metal will be in solution in the ferric state, no matter in what condition it existed in the ore. But with dilute hydrochloric or sulphuric acid it will retain its former degree of oxidation. Hydrochloric acid, for example, with chalybite (ferrous carbonate) will give a solution of ferrous chloride; with hæmatite (ferric oxide) it will yield ferric chloride; and with magnetite (ferrous and ferric oxides) a mixture of ferrous and ferric chlorides. Metallic iron yields solutions of ferrous salts. It is convenient to speak of the iron in a ferrous salt as ferrous iron, and when in the ferric state as ferric iron. Frequently it is required to determine how much of the iron exists in an ore in each condition. In such cases it is necessary to keep off the air whilst dissolving; the operation should, therefore, be performed in an atmosphere of carbonic acid.
Separation.—The separation of the iron from the other substances is as follows:—Silica is removed by evaporating the acid solution, and taking up with acid, as described under Silica; the whole of the iron will be in solution. The metals of Groups I. and II. are removed by passing sulphuretted hydrogen, and at the same time the iron will be reduced to the ferrous state. The solution should be filtered into a 16 oz. flask, boiled to get rid of the gas, and treated (whilst boiling) with a few drops of nitric acid, in order to convert the whole of the iron into the ferric state. When this condition is arrived at, an additional drop of nitric acid causes no dark coloration. The boiling must be continued to remove nitrous fumes. Next add caustic soda solution until the colour of the solution changes from yellow to red. The solution must be free from a precipitate; if the soda be incautiously added a permanent precipitate will be formed, in which case it must be redissolved with hydrochloric acid, and soda again, but more cautiously, added. After cooling, a solution of sodium acetate is added until the colour of the solution is no longer darkened. The solution, diluted to two-thirds of the flaskful with water, is heated to boiling. Long-continued boiling must be avoided. The precipitate is filtered quickly through a large filter, and washed with hot water containing a little acetate of soda.
The precipitate will contain all the iron and may also contain alumina, chromium, titanium, as well as phosphoric, and, perhaps, arsenic acids.[64]
Dissolve the precipitate off the filter with dilute sulphuric acid, avoiding excess, add tartaric acid and then ammonia in excess. Pass sulphuretted hydrogen, warm, and allow the precipitate to settle. Filter and wash with water containing a little ammonic sulphide.
GRAVIMETRIC METHOD.
Dissolve the precipitate in dilute hydrochloric acid; peroxidise with a few drops of nitric acid and boil, dilute to about 200 c.c., add ammonia (with constant stirring) till the liquid smells of it, and heat to boiling. Wash as much as possible by decantation with hot water. Transfer to the filter, and wash till the filtrate gives no indication of soluble salts coming through. The filtrate must be colourless and clear. The wet precipitate is very bulky, of a dark-brown colour and readily soluble in dilute acids, but insoluble in ammonia and dilute alkalies. When thrown down from a solution containing other metals it is very apt to carry portions of these with it, even when they are by themselves very soluble in ammoniacal solutions. It must be dried and ignited, the filter paper being burnt separately and its ash added. When further ignition ceases to cause a loss of weight, the residue is ferric oxide (Fe2O3), which contains 70 per cent. of iron. The weight of iron therefore can be calculated by multiplying the weight of oxide obtained by 0.7.
The presence of ammonic chloride causes loss of iron during the ignition, and organic matter causes an apparent loss by reducing the iron to a lower state of oxidation. When the iron in the solution much exceeds 0.2 gram the volumetric determination is generally adopted, as the bulkiness of the precipitate of ferric hydrate makes the gravimetric method very inconvenient.
VOLUMETRIC METHODS.
As already explained these are based on the measurement of the volume of a reagent required to bring the whole of the iron from the ferrous to the ferric state (oxidation), or from the ferric to the ferrous (reduction). Ferrous compounds are converted into ferric by the action of an oxidising agent in the presence of an acid. Either permanganate or bichromate of potash is generally used for this purpose.[65]