[ 18 ] 



II. On the Deportment of Bismuth during Solidification. 

 By R. Schneider*. 



DURING the change of bismuth from the fluid to the solid 

 state of aggregation, it often happens that the surface of 

 the apparently sohdified mass is broken through in one or naore 

 places by the liquid metal underneath, which soon soUdifies 

 in globular forms. This phsenomenon is generally accepted as a 

 sure proof that bismuth at its moment of solidification expe- 

 riences a considerable expansion. Marx, indeed, has deter- 

 mined the expansion from the magnitude of the globules thus 

 forced through the metallic crust, and found that they amount 

 to j^rd of the entire mass. The above proof is not correct; 

 pure bismuth, whatever may be its expansion at the moment of 

 sohdification, does not exhibit the squeezing out of the globules 

 of metal. This phsenomenon is observed with impure bismuth 

 only, and it is a remarkable fact that the bismuth thus squeezed 

 out is foimd to possess a high degree of purity, even when the 

 metal made use of contains a considerable quantity of foreign 

 matter. The following experiments will serve as a confirmation 

 to this statement. 



1. In a mass of bismuth sold to me as pure, which, however, 

 during solidification showed the protrusion of the metallic spheres, 

 2*5 per cent, of impurity was found upon analysis ; chiefly con- 

 sisting of sulphur, a little copper, and a trace of iron. From 

 64 grammes of this metal, by repeated melting and pouring out 

 upon a clean plate of porcelain, 32 grammes or 50 per cent, of 

 metallic globules were gradually collected. These globules were 

 melted altogether, and the mass again poured out ; during solidi- 

 fication no globules of bismuth appeared in this case. The ana- 

 lysis of the mass showed that out of 100 parts 99-92 were bis- 

 muth. From sulphur, copper, and iron the mass was totally free. 



2. 100 grammes of almost pure bismuth were melted with 

 3 grammes of sulphur, 1 of copper, 0-25 of a gramme of silver, 

 1 gramme of nickel, and 1*25 gramme of arsenic; that is, the 

 substances with which the commercial bismuth is most usually 

 associated. Although at the high temperature necessary to melt 

 the mass a portion of the sulphur and arsenic was lost, the foreign 

 constituents could nevertheless be rated at 5 per cent, at least. 

 From this metal, heated as before, 25 per cent, of spherical bis- 

 muth was separated. The latter being melted and sufi"ered to 

 solidify, yielded no globules. The analysis showed that this mass 

 contained 99-78 per cent, of bismuth, 0*11 per cent, of silver, 

 and a trace of sulphur. Copper, nickel, and arsenic were not to 

 be found. It is worthy of remark that when silver is contained 



* From PoggendorfiP's Annalen, No. 11. 1855, p. 494. 



