370 M. Deville on the various forms of Silicon. 



which have been attacked by contact with the sodium, together 

 with their contents, are exposed in a carbon crucible to the heat 

 of a wind-furnace. The common salt volatilizes, the porcelain 

 melts, and globules of silicon presenting remarkable traces of 

 crystallization are obtained. 



It is also obtained by the action of sodium on a glass free from 

 iron, and specially prepared for the purpose by melting together 

 silica, Iceland spar, and carbonate of potash. This glass, coarsely 

 powdered, is mixed with sodium and heated to redness in a glass 

 tube. The black matter which is obtained is heated in a wind- 

 fui-nace in a carbon crucible placed inside another crucible. 

 The silicon is found dispersed in the vitreous mass, which is 

 smoked like obsidian, in the form of small, very brittle, steel- 

 gray globules. 



Silicon may also be extracted from silica by means of the gal- 

 vanic current, by fusing a mixture of fluorides of sodium and 

 potassium, adding calcined silica, which is very readily dissolved, 

 and then inserting the platinum and carbon poles of a Bunsen's 

 battery of four elements. Oxygen is disengaged at the negative 

 pole, and the silicon set free at the platinum pole unites with it 

 to form a siliciuret of platinum, which fuses with extreme ease. 



Deville obtained simultaneously with, but independently of 

 Wohler*, the graphitoidal form of silicon, but by a different 

 method. In preparing aluminium by the battery he obtained 

 an alloy of aluminium and silicon, which by treatment with acids 

 yields the graphitoidal form of silicon. But the best method of 

 preparing it is that of Wohler, and Deville confirms the descrip- 

 tion of its properties given by that chemist. 



The octahedral form of silicon is obtained by a process sus- 

 ceptible of application to the preparation of fiH the fixed simple 

 bodies which form compounds that are volatile and decomposable 

 by some substance capable in itself of dissolving them. They 

 are then obtained crystalline. Deville terms this, crystallization 

 by igneous solution. 



On passing chloride of carbon over aluminium, or over so- 

 dium, amorphous carbon is obtained ; neither of these metals can 

 dissolve carbon. But by treating melted iron (or better, cast 

 iron), which has the property of dissolving carbon, with chlo- 

 ride of carbon, a crystalline substance different, however, to iron- 

 graphite is obtained. 



Deville prepares this form of silicon as follows : — Aluminium 

 placed in a small tray is introduced into a porcelain tube, through 

 which a current of hydrogen saturated with the vapours of chlo- 

 ride of silicon is passed. The tube is kept at a clear cherry-red 



* Phil. Mag. vol. xii. p. 64. 



