Prakt. Met. Sonderband 38 (2006) 45 
averaged two times and the scan step size was chosen to 2.8 um (hexagonal pixels). In 
order to minimise the influence of noise in the results, a grain has at least to comprise six 
pixels, resulting in a minimum grain diameter of around 7 pm. Grain boundaries were 
characterised by a misorientation larger than 5° between neighbouring measurement 
points. 
3. RESULTS 
A useful possibility to characterize the low angle misorientations of the grains is the 
from hot determination of the overall spread in orientation within a grain. In this case the so called 
0gengous grain orientation spread was calculated by averaging first the orientation for each grain. 
C for 60 The spread is then obtained by averaging the deviation between the orientation of each 
and led to point in the grain and the average orientation for the grain. The results for both a fully and 
arried out a partly recrystallized specimen are shown in Fig. 2a. A clear bimodal distribution of the 
wn to the partly recrystallized specimen can be observed and in addition the first peak of this 
0.1 3 to specimen matches very well with the peak of the fully recrystallized specimen. The strong 
soak time scatter in the range of the deformed grains is caused by the small number of larger grains. 
perature. The coloured recrystallized grains of Figure 2b are achieved applying this approach and 
d Sou using the marked value of the orientation spread for the partitioning of the two fractions. 
=12 mm, 
sections. 0.30 
> element 
o half the n O28 
9 0.20 + os 
© 0.157 eons 
8 0.10 4 
© 
0.05 | 
0.00 ! oo 
8 1 2 5 4 56 6 7 
orientation spread in degrees 
— fully recrystallized specimen 
—. partly recrystallized specimen 
Fig. 2: a) Distribution of the grain orientation spread of a partly and a fully recrystallized 
m specimen with the vertical line marking the partitioning b) image quality map overlapped 
solution with the recrystallized grains 
As a consequence the discrimination between the deformed and the recrystallized fraction 
was performed with this approach for the following investigations. Generally, if 
interpretable diffraction patterns are obtainable also from the deformed structure of a 
material, the grain orientation spread can be recommended 
d ground 
ning step 3.1 DYNAMIC RECRYSTALLIZATION 
/) for 30 
2 Gemini The dynamic recrystallization was investigated on specimens that were immediately 
n EDAX- cooled down to room temperature after the loading. Therefore, these microstructures 
I primary represent the state at the end of the hot deformation. The evolution of the DRX structures 
rns were with strain is depicted in Fig. 3 a,b and clearly indicates that the DRX predominantly starts