1813.] 



On the Changes of Colour on Steel, 



131 



Article VII. 



Oil the Cause of the Changes of Colour prodticed hy Meat, m 

 the Surface of Steel By Sir H. Davy, LL.D. F.R,S. 



(To Dr. Thomson.) 



DEAR SIR, Berkeley-square, Jan, 13, 1813. 



In the last edition of your elaborate and learned System of 

 Chemistry, vol. i. p. 224, you have stated that the changes of 

 colour produced by heat on the surface of polished steel takes 

 place under oil. In my Elements of Chemical Philosophy, page 

 390, I have said that these changes occur when the metal is 

 plunged beneath the surface of mercury, and we both conclude 

 that the elfect probably does not depend upon the oxidise- 

 ment of the metal. 



I was led to doubt of the perfect correctness of our statements-» 

 and the justness of our conclusions, by a letter from Mr. Stoddart, 

 who has made many accurate experiments on the tempering of 

 steel ; and that gentleman sent me two pieces of steel which had 

 been heated to the same degree, one in the atmosphere and the 

 other under the surface of pure mercury, where it had been 

 suffered to cool ; the first was blue, the second had suffered no 

 change of colour ; and both seemed to possess the same degree 

 of hardness. 



As 1 had formerly made but one experiment on this subject, 

 and as the mercury I used was impure and not cleaned with any 

 particular care, it appeared most likely that I had been deceived 

 by some metallic oxides, or saline matter adhering to the mer- 

 cury ; and I invited Mr. Stoddart to assist in some new trials on 

 the subject. 



A piece of polished steel was introduced into a retort, which 

 was exhausted and filled with hydrogen gas, and this hydrogen 

 gas was deprived of oxygen, a small quantity of which might 

 have entered with common air in the stop-cock, by melting 

 phosphorus in it ; the retort was then gradually heated. Where it 

 was in contact with the steel, a slight tint of yellow was soon ob- 

 served on the surface of the metal, but it did not increase as it 

 would have done, in the atmosphere during the increase of 

 temperature. 



A piece of polished steel was plunged in very pure olive oil, 

 which had been previously heated to deprive it of air ; the tem- 

 perature of the oil was increased until it began to boil, but no 

 change of colour took place on the surface of the steel, 



I had little doubt that the slight change of colour produced on 

 the metal in the hydrogen gas was owing to some aqueous 

 vapour in the gas, or to some action of the phosphorus, and I 

 have since proved the truth of this conjecture. 



I2 



