CORROSION OF IRON AND STEEL Gl 



It appears from recent work^ that 80° C. is the temperature of 

 maximum corrosion. 



(6) Surface Effect. — ^It has been shown'^ that a rough surface 

 tends to dissolve more than a smooth surface. This will, of course, 

 only have an effect on the initial rate of corrosion. 



(7) Physical Condition of the Metal. — A metal if slowly cooled 

 is to a large extent in an annealed condition. The annealed con- 

 dition corresponds to the crystalline state of the metal. The effect 

 of working the metal is to break down the crystalUne structure to 

 some extent and to produce the ammorphous form. The author 

 has already shown that the tendency of a hammered metal to 

 •dissolve is greater than that of the annealed metal by 10 milli- 

 volts in certain cases. 



Iron is subjected to a good deal of straining in rolling or 

 hammering processes, and differs in hardness between one place 

 and another in an especial degree. This circumstance must, 

 therefore, assist greatly the corrosion of iron. 



(8) As influencing corrosion exceptionally there is to be con- 

 sidered the chemical condition of the metal. 



It is well known that metals which are impure usually dissolve 

 more easily than pure metals. Pure iron will not rust^- Iron in its 

 commercial forms, however, may be described as being the most 

 impure of all metals in their commercial varieties, and oxidises 

 with great readiness. 



Carbon. — -Iron always contains carbon. This is present in all 

 ordinary steels in the combined form cementite (FeaC), which 

 contains about 7 per cent, of carbon. The cementite is often inti- 

 mately mixed with the pure iron to form other constituents such 

 as the eutectics pearlite and sorbite with .9 per cent, of carbon ; in 

 these eutectics the iron particles are, of course, quite distinct from 

 the cementite particles, and at present, from the corrosion stand- 

 point it is probably more correct to look on steels as a mixture of 

 iron and cementite. A number of steels all in the annealed 

 ■condition, containing varying amounts of carbon up to 0.9 per 

 cent, were tested for their solution pressure by the E.M.F. method,* 

 and were found all to give the same value. This showed the 

 presence of some pure iron in all cases, which was giving the 

 value of the potential obtained. 



Actual corrosion experiments on steels with varying per- 

 centages of carbon do not in general show that the presence of carbide 

 has a pronouncedly bad or good effect.^ 



At my suggestion Mr. G. J. Burrows, B.Sc, tested some steels 

 in a corrosive water containing sodium chloride, sodium bicarbo- 

 nate, and magnesium sulphate. There were seven steels tested 

 with the following percentages of carbon : 0.18, 0.23, 0.40, 0.62, 



1 Hehn & Bauer : Mitt. Kg/., Matrial-Prufungsamt, 1910. 



2 Fawsitt ; J ourn. Soc. Chem. hid., 1906, 25,11;}:!. 



3 Lambert and Thomson ; Proc. Chem. Sor., 26, 290. 



4 See /ourn. ^oc. Chem. Ind.. 1906, vol. 25, ll:i3. 



a See Cushman and Ciardner. " The Corrosion of Iron and Steel," p. 76. 



