PROCESSES FOR THE CONVERSION OF CARBON INTO SILICON. 549 



in water, and added in atomic proportions to solutions of cyanide of potassium. 

 Where the iodide was used, the yellow colour of the precipitate shewed that it was 

 a compound of the iodide and cyanide of lead. The precipitate varied in appear- 

 ance and properties, where the chloride was employed, according as it or the cya- 

 nide of potassium was made the precipitant ; but in every case it seemed to be a 

 mixture of cyanide and chloride of lead. It retained moisture with great obsti- 

 nacy, and could scarcely be deprived of it by heat without decomposition. It was 

 accordingly placed when moist in vacuo over sulphuric acid, and afterwards dried 

 completely in a current of heated air passed through chloride of calcium. But this 

 tedious desiccation introduced a new element of impurity ; for although the cyanide 

 of potassium employed was originally quite free from carbonate of potass, the pre- 

 cipitate, when dry, gave off carbonic acid when treated with dilute acids. The 

 proportion of chloride present in the precipitates varied much, and the physical 

 characters of the body along with it. Where the cyanide of lead preponderated, it 

 was, when dry, of a primrose yellow colour, and was changed by heat into a bluish 

 crumbling powder. When the chloride of lead was in excess it was cream-coloured ; 

 and when heat was applied, fused and adhered to the glass. 



At this stage of the inquiry, we abandoned the cyanide of lead as a raw ma- 

 terial for transmutation, having exhausted for the time all our resources for its pre- 

 paration. 



We subjected, however, most of the lead precipitates to the treatment pre- 

 scribed by Dr BROWN for the conversion of the carbon of the cyanide into silicon, 

 in the expectation of procuring sufficient quantities of the latter substance to esta- 

 blish its nature unequivocally. 



When the precipitate prepared by his own methods was heated with the pre- 

 cautions indicated in his " Two Processes," we obtained a powder of a bluish-grey 

 colour, closely corresponding in physical characters to the " leaden " product he 

 describes. When this was treated with dilute nitric acid, all the lead of the com- 

 pound dissolved, and an insoluble brown powder remained, which we trusted was 

 silicon. It was quite soluble in oil of vitriol, and was only partially destroyed by 

 fusion with chlorate of potass. When it was fused with carbonate of potass, and 

 the product of fusion treated with muriatic acid, evaporated to dryness, and redis- 

 solved in water, according to the ordinary method for the separation of silica from 

 silicate of potass, an insoluble yellowish-white residue remained, amounting in 

 weight to a fraction of that of the original brown powder. Had the latter been 

 silicon, it should have yielded by this treatment twice its weight of silica.* In 

 one recorded experiment, we find that 4'5 grs. of the brown powder gave, after the 

 treatment described, 0-4 grs. of insoluble residue; in another, 2-1 grs. gave 0-3 

 grs ; and the proportion generally obtained was, as nearly as possible, a tenth of 

 the original powder. 



* The atomic weight of silicon, according to Berzelius, is 22-22 ; that of silica 46'22. 



