264 M. Berzelius on the Prepayation 



the silicate may be sepai'ated by means of water, but the re- 

 mainder can only be i-emoved by fluoric acid. The quantity of 

 silicium obtained is extremely small ; and this manner of ob- 

 taining it merits attention only because it gives the same re- 

 sult as the double salt when treated with potassium ; which 

 proves that the evolution of hydrogen that takes place when 

 the mass is treated by water is owing to the potassium, and 

 not to the combustible radical of fluoric acid. It is by the 

 process in question ih^t Davy endeavoured to reduce silica; 

 and he obtained with the mixed silicate of .potash only a brown 

 pulverulent matter, which dissolved in the water, giving it a 

 greenish-gray colour. I observed the same colour in the so- 

 lution ; but it disappears when this becomes clear. 



1 passed silicated fluoric gas through a red-hot iron tube filled 

 with turnings of the same metal, and it was not sensibly ab- 

 sorbed. The turnings when taken out were of a nut-brown co- 

 lour, similar to that of silicium, and had the taste of fluate of 

 iron, in the places where the heat had had the most intensity. 

 This last having been dissolved by water, a perceptible pelli- 

 cle of silicium was left on the surface of the iron, but so thin 

 that it was not possible to separate it. We see by this ex- 

 periment that iron at a suflSciently elevated temperature has 

 an affinity strong enough to decompose the gas ; but that the 

 decomposition soon ceases, because the silicium which is de- 

 posited at the surface of the metal prevents its action. By 

 fusing in a covered iron vessel a mixture of very fine iron 

 filings with fluate of silica and potash, the salt was decom- 

 posed and converted into double proto-fluate of iron and pot- 

 ash : after having dissolved it in warm water, a combination 

 of iron and of silicium was left. I hoped to be able to sepa- 

 rate the iron by means of an acid ; but the silicium became 

 oxidated at the same time, although I employed the liquid 

 sihcated fluoric acid. In drying this alloy it oxidated more 

 quickly in the air than it dried, and was changed into an ochre 

 of a rusty yellow colour. 



Composition of Silica. 

 Since silicium can be obtained in a direct manner, it was na- 

 tural to endeavour to ascertain its oxidation in a direct manner. 

 For this purpose I burnt ] 00 parts of pure silicium, dried in 

 a vacuum, by heating it with carbonate of soda. The mass, 

 treated by muriatic acid, evaporated to dryness, and strongly 

 heated, was dissolved in water, and left some silica coloured 

 gray by carbon, which, well washed and ignited, became snow- 

 white, and weighed 203'75. The solution obtained and the 

 waters of the washing were evaporated afresh, and the saline 



residuum 



