OF FLAMES RESULTING FROM BESSEMER PROCESSES. 
497 
greater portion of the oxygen of the air enters into combination with iron, forming 
magnetic oxide of iron which is probably reduced to a lower oxide. This oxide of 
iron will be reduced to the metal by the silicon, manganese, and cahoon present in the 
iron, and by the phosphorus or ferrous phosphide in presence of the lime in the order 
stated, and if the time of oxidation v/ere extended indefinitely each one would be 
oxidised almost completely in turn. In the converter the oxidation is so rapid that 
all the reactions go on together ; but the complete oxidation of each impurity is 
achieved in the order named. 
Any phosphorus oxidised in the early part of the “blow” in contact with and 
entering into comhination with the lime is finally removed from the iron, as the 
resulting compound is not reduced by the reducing agents ])resent. About one-half 
the phosphorus is oxidised in this way before the flame drops (‘ J. I. and S. Inst.,’ 
1896, No. 1, p. 465). 
The chief cause of the flame in the “acid” process has been shown to be carljon 
monoxide which, leaving the vessel at a high temperature, accompanied Ijy vapours of 
iron, manganese, &c., is burnt Ijy the oxj^gen of the air. The hottest 2 )art of this 
flame is the outer layer, where the temperature is increased by the comhusti(jn of 
the carbon monoxide and hydrogen. When the carbon is completely removed from 
the iron in the “ basic ” process, there is still the appearance of a flame which 
increases in length as the “ over-ldow ” proceeds ; the quantity of fume expelled 
increases rapidly at this period also. The short Tame is l.)right, and may consist 
principally of finely divided particles of oxide of iron heated to the temperature of 
the bath of metal and slag, and by the heat evolved by the combustion of metallic 
vapour inside the vessel. The particles of matter expelled from the vessel are of two 
kinds : one consists of larger particles of slag produced by tlie violent agitation of 
the fluid charge inside the converter; these quickly fall to the ground. The other 
kind is the fume proper, composed of the products of combustion of the metallic 
vapours. The particles forming this fume are undouljtedly very small; this is 
proved by the fact that they scatter the light falling on thiem, and cast'"' a brown 
sliadow, and also by the great height to which the cloud ascends. These very small 
particles will be kept at the temperature of the escaping gases. 
The spectrum during the “over-blow” is very feeble in the violet and does not 
extend l)eyond 4000. The spectrum of the lime-light extends into the ultra-violet, so 
also does the light emitted mainly from the furnace walls, through the woi'king doors 
of a Siemens-Martin steel-melting furnace. 
It appears, therefore, tlmt the temjDerature of the source of light in the flame of 
the “ over-blow ” is comparatively low, approaching that of a yellowish-white heat, 
and that the light of what is apparently a flame emanates from a torrent of very 
small particles, liquid or solid, heated to a temperature of such luminosity. 
* Mr. C. H. Ridsdale first remarked the colour of this shadow (see ‘ Jouni. Iron and Steel Institute,’ 
No. 2, 1895, p. 35). 
3 S 
VOL. CXCVI.-A. 
