DR BROWN ON THE PRODUCTION OF SILICON FROM PARACYANOGEN. 237 



they must be regarded as mixtures of two or more definite compounds. The for- 

 mation of these mixed metallic products explains the necessity of care in the per- 

 formance of the process for paracyanogen. 



2. An analogous mixed compound of iron is procured by a similar procedure 

 with the iron high-pressure tube. It requires a higher temperature, and contains 

 a larger proportion of silicon. 



3. A new platinum crucible was half filled with paracyanogen, and ignited 

 for three hours. On being opened, it was empty and clean, the metallic lustre 

 having been only slightly dimmed. This was repeated, till the metal would ab- 

 sorb no more of what was yielded to it by the paracyanogen. It was now grey 

 and brittle ; and, on analysis of 20 grs. of the broken crucible, was found to 

 have been composed of platinum and silicon, containing nearly 4 per cent, of the 

 latter. This analysis was effected by the reaction of nitro-hydrochloric acid ; and 

 the undissolved residue, a mixture of silicic acid and a black powder, having been 

 fused with carbonate of potassa, the silicic acid was separated by the ordinary 

 process, and the constituent silicon was inferred by calculation. 



The singularly powerful attraction of platinum for silicon has been often ob- 

 served. DESCOTILS ignited it in contact with incandescent charcoal, and procured 

 a frangible substance, which he represented as an indefinite combination of plati- 

 num and carbon. M. BOUSSINGAULT examined the same product, and found it to 

 be a true siliciuret. BERZELIUS* accounts for its formation by supposing that 

 the metal decomposes the silicic acid which is known to exist in charcoal, and 

 appropriates the base, while, it is to be presumed, the oxygen disappears in the 

 form of carbonic acid. This rationale is certainly incongruous with the general 

 plan of chemical reaction as now understood, and appears to be a mere evasion 

 of an apparently insurmountable difficulty ; for silicic acid, whether nascent or 

 produced, may be ignited in platinum crucibles with true carbon, in any excess, 

 without being decomposed, the carbon being taken in, and the silicic acid left un- 

 touched. But for this the common analysis for silica would in many cases be 

 nugatory. It must be remembered that charred wood is not carbon, but contains 

 a large proportion of some exorganic compound of that element, especially if it 

 have been produced at low temperatures ; it forms artificial tannin with nitric 

 acid, another exorganic proximate which I have tried in vain to produce with true 

 carbon, however finely divided, such as is prepared by the decomposition of the 

 solid iodide of carbon. Grant that common charcoal contains a compound radi- 

 cal analogous in constitution to paracyanogen, and BOUSSINGAULT'S siliciuret of 

 platinum is explained ; it may have been produced by transformative decomposition. 

 This postulate is rendered probable by Mr JOHNSTONE'S observations on the coals, 



* Traite de Chimie BERZELIUS, par VALERIUS, 1838, p. 426. 

 VOL. XV. PART I. 38 



