﻿effect of Temper, in case of Steel. 277 



cut more nearly clown into the heart of the inquiry and to 

 determine directly the carbon relations of steel as a function of 

 the temperature (0° to 400° ; 400° to 1,000°) and of the time of 

 annealing ; to do this with full reference to the physical occur- 

 rences observed in the first and second phases of the phenome- 

 non. So far as we know, M. Caron* alone has occupied him- 

 self with similar work ; but his researches, being largely 

 restricted to the extreme states hard and soft, are for the 

 present purposes incomplete. 



Glass- hard steel rods (Stubs) about 0\L cm in diameter and tem- 

 pered uniformly in the way described elsewhere,f were each 

 broken into four nearly equal parts and four samples of hard 

 steel identical in composition and temper thus obtained. These 

 samples were annealed at 20° (glass-hard), 100° 4 h , 200° l h , 

 360° l h , 1000° respectively. Having treated them with cold 

 HC1, we found that the rods annealed at 20° and at 100° dis- 

 solved apparently without residue; those annealed at 200° left 

 a trace of flocculent carbon. Eods annealed at 360° yielded 

 flocculent carbon in some amount ; rods annealed at 1000°, 

 finally, a comparatively copious and heavy carbon precipitate. 

 The residues were collected in a weighed Gooch crucible (asbes- 

 tos filter), thoroughly washed in the usual way,$ dried, weighed, 

 ignited in oxygen, again weighed, and the loss of weight on 

 ignition estimated as carbon. The results thus obtained are 

 sufficient for the present purposes : 



Annealed at : 20°, oo 100°, 4 h 200°, l h 360°, ]> 1000° 



TJncombined graphitic carbon \ m m Q . 00Q9 ^ 



per gram of steel, c = J N ^* 



In a second series of similar experiments we found : 



Annealed at: 20°, oo 100°, 10 h 200°, l h 360°, ]> 450°, 1* 1000°, 30 m Commer- 



Uncombined graphitic ) ciai (soft). 



carbon per gram of [• 0*0001 0-0005 0-0005 0-0014 0-0009 0-0033 0*0053 



steel, c — ) 



In general c increases at an accelerated rate with tempera- 

 ture. The large datum for the commercial state, as compared 

 with the smaller values of c for steel heated to redness in the 

 ordinary way is an interesting feature of these results. We 

 will refer to it again below. Temperatures even as low as 100° 

 when acting on hard steel for long intervals of time (10 h ) pro- 

 duce perceptible precipitation of the carbon in the steel. 



On closer inspection it appeared that steel annealed at 100° 

 is, cet par., more easily soluble than glass-hard steel ; steel 

 annealed at 200° more easily than steel annealed at 100° ; steel 



*Caron: Comptes Eend., lvi, pp. 43. 211, 325, 1863. 

 + Bull. XL S. G. S., No. 14, p. 29, 1885. 



% Using dilute HOI, hot water, solution KOH, alcohol, ether. Cf. Blair, "Re- 

 port of the Board of Testing Iron, etc.," Washington, Govt., I, 1881, p. 248. 



