The diffusion of hydrogen through hydrogen‐purified carbonyl iron shows breaks in the log (rate) vs. 1/T curve at 834, 909 (body‐centered→face‐centered phase change), 945, 1009, 1056, 1082 and 1103 deg. centigrade. These are average values for three samples of ``Mehl'' iron. The log (rate) vs. 1/T curves at these temperatures are composed of arcs, the above points being ends of these arcs. These points fall on a line whose slope is 4085 deg. Kelvin. Thus b in the expression exp (—b/RT) is given as 18,680 cal./g atom for these points. The above temperatures belonging to the ends of the arcs can be fitted to a Balmer type formula, i.e., T (°K) = C(1/32—1/n2), where n = 6, 7, 8, ..., and C = 13,430±30 for the breaks starting with n = 7, while C = 13,301 for the break with n = 6. This difference is presumably caused by the n = 6 break being in the body‐centered phase, while all breaks starting with n = 7 are in the face‐centered phase.
[1]
R. Fowler,et al.
A Theoretical Formula for the Solubility of Hydrogen in Metals
,
1937
.
[2]
J. C. Slater.
The Ferromagnetism of Nickel. II. Temperature Effects
,
1936
.
[3]
C. J. Smithells,et al.
Diffusion of Gases through Metals
,
1934,
Nature.
[4]
J. B. Austin.
Heat Capacity of Iron - A Review
,
1932
.
[5]
G. Borelius,et al.
Durchgang von Wasserstoff durch Metalle
,
1927
.
[6]
O. Richardson,et al.
I. The diffusion of hydrogen through hot platinum
,
1904
.