IROSr 1029 



partially decomposed solution of indigo in sulphuric acid, which is said to keep its 

 characteristic colour without alteration for a considerable time. Another recommended 

 by Hetman consists of a mixture of bichromate of potash and nitrate of cobalt. In 

 most cases, however, the direct system of comparison with solutions obtained of steel 

 of known composition is preferred. 



Silicon. The determination of silicon in iron and steel is a matter of some difficulty, 

 as the mere weighing of the ignited insoluble residue from the hydrochloric acid 

 solution and calculation of the silica found, as silicon, depends upon the assumption 

 that the iron is free from silicates, which is often contrary to the fact, especially in 

 malleable iron, which may contain a notable amount of intermingled slag that has 

 not been perfectly expelled in the welding, and exists as a mechanical impurity, having 

 no relation whatever to the real composition of the metal. This defective interpreta- 

 tion of the ultimate analyses is probably the cause of the contradictory statements 

 current as to the amount of silicon that may be present beneficially in metallic iron. 

 Eggertz has introduced a method of determining silicon in the presence of slags, 

 which is based upon the fact that when iron is slowly acted on by bromine, or a solution 

 of iodine, it dissolves, and the silicon set free is converted into silica, which is com- 

 pletely soluble in a boiling solution of carbonate of soda, while that in combination in 

 the slag, if any be present, is not acted upon. The same method may be employed 

 with cast iron, as blast-furnace slag, which sometimes occurs in it as a mechanical 

 impurity is not sensibly acted upon either by iodine, bromine, or carbonate of soda. 



The process is conducted as follows : Three grammes or '45 grain of iron in the 

 state of filings or borings sufficiently small to pass through a sieve with meshes ^th 

 of an inch in diameter, is treated with five times its weight of iodine, in a volume of 

 fifteen cubic centimeters of water contained in a beaker of about six or seven times that 

 capacity. Water that has been boiled to free it from air is used, and the operation 

 must be performed at as low a temperature as possible, in order to permit oxidation 

 of the iron by the air ; it is usual, therefore, to keep the beaker cool by the use of ice. 

 When the iron is completely dissolved, the solution is increased to three times the 

 original volume by the addition of very cold water, in order to prevent the separation 

 of basic salts of iron, and is well stirred and left to settle. The lighter scales of 

 graphitic carbon remain in suspension, but are poured off with the bulk of the liquid 

 into a filter 2 inches in diameter, with only about one-tenth of the original quantity of 

 the insoluble residue which forms a heavy powder at the bottom of the beaker. A few 

 drops of hydrochloric acid are added, and the liquid stirred with a glass rod ; if this 

 produces a disengagement of gas, the whole of the iron is not dissolved, and a little 

 iodine and carbonate of soda are added in order to complete the solution. The acidified 

 water must be poured quickly on the filter in order to prevent the decomposition of 

 the slag. When the whole of the residue is transferred to the filter it is washed with 

 cold water until the whole of the iron is found to be completely removed when tested 

 by a solution of ferrocyanide of potassium. The filtrate is evaporated to dryness with 

 the addition of hydrochloric acid in order to recover any small amount of silica that 

 may have gone into solution with the iron. The original residue, which may contain 

 graphite silica and slag, is transferred without drying into a large platinum crucible, 

 and after concentrating the waste water to 6 cubic centimeters, a saturated solution of 

 carbonate of soda is added, and the crucible is heated for one hour in a water-bath, the 

 liquid being stirred from time to time with a platinum spatula in order to dissolve any 

 lumps in the insoluble residue. The solution is then carefully poured from the in- 

 soluble mass in the crucible on to a small filter, and a fresh quantity of carbonate-of- 

 soda solution is added and boiled for another hour when the whole contents of the 

 crucible are thrown upon the filter and washed. The alkaline solution of silica is 

 acidified with hydrochloric acid added to that containing the iron, and evaporated to 

 dryness in a water-bath. This operation is repeated with the addition of fresh acid 

 until the silica is perfectly freed from iron, when it is filtered, dried, ignited, and 

 weighed, as often as may be necessary to ensure its complete purity which must be 

 further tested by heating it with hydrochloric acid. When 3 grammes of iron are 

 treated 0'016 gramme of silica corresponds to O'OOl gramme of silicon. The insoluble 

 residue from the carbonate-of-soda solution may contain graphite, slag, oxide of iron, 

 and titanic acid. There is no way of distinguishing the amount of oxide of iron present 

 as such, in addition to that in combination with silica, except by assuming the com- 

 position of the slag to be constant, which is not the case. 



Eggertz found by the use of this process that the amount of silicon in good bar 

 iron may vary from 0*01 to 0*10 per cent. ; but in two samples of Krupp's cast steel it 

 was as high as 0'30 per cent. Slag has been found usually in mere traces in cast steel, 

 but in one instance it mounted to 0'2 per cent. Malleable iron, however, contained 

 much more. In wire iron from a charcoal hearth 0'38 p. c. was found, in armour-plates 

 from 0-75 to 3*00 p. c,, and in a rail as. much as from 4'00 to 5'00 p. c. H.B. 



