segregation profile along a distance. In fact, the initial 
Sg me segregation profile is constituted of several sine waves superimposed 
el with diffzrent wavelengths (15) Therefore, in any of the specimens 
analyzed, the maximum intensity recorded may not correspond to the 
same wave of the minimum intensity. For this reason, the average of 
ng 8 the ten greater and lower values of X-ray intensity was chosen. As 
tance and titanium and molybdenum form carbides, the theoretical model adopted 
could not be applied to these elements, as it does not take in 
account the dissolution of precipitates. Anyway, the results enable 
Co the assessment of homogenization kinectics as a whole. 
The average of the 300 distances measured between microsegregation 
oe striatons resulted 82pm. Therefore, using l=41um in equation (2) and 
tat, the the 8 values for nickel, the following values for diffusion 
coefficient at 1197°C are found : 
t (h) 1.5 J _5 ) , 
HS Dem? /S). 7. 3x0 14 7x0 BR TO O7 
These values are in the same order of magnitude as some values 
mentioned in literature (17,18) namely : D=8.4x10" em? /s which 
Geet corresponds to the diffusion coefficient for nickel in y-iron at 
1197°C and p=2.9x10" lemz/s in a 15.5% nickel-iron alloy at the same 
temperature. 
Co 4. CONCLUSIONS 
Fa The microsegregation degree can be qualitatively assessed through 
metallography, that is the aged state which gives the best 
RG indications. 
The white areas revealed by 4%Nital etchant are not constituted only 
of austenite. 
TEN It is possible to assess, semi-quantitatively, the microsegregation 
. degree and homogenization kinetics of nickel, molybdenum and titanium 
Cin through a WDS electron microprobe. 
Calculating the diffusion coefficient for nickel and, with the 
in integration of the second Fick's Law in the boundary conditions of 
Prakt. Met. Sonderbd. 21 (1990) 
50 
3E 
IC 2k 
149