It is now well established that the stored energy (SE) during plastic deformation due to storage and organization of dislocations during plastic deformation plays a significant role in both nucleation and growth steps of the classical discontinuous or continuous recrystallization process observed in metal forming, including Strain Induced Boundary Migration (SIBM) which can play a significant role in the formation of specific textures like those developed for the electrical steels. However, it is also recognized that one single parameter is not enough to account for the various nucleation and growth mechanisms which can simultaneously occur within a single sample and to clearly explain the link between deformed, recrystallized states and final mechanical properties. Although SE is still often related to one single dislocation density parameter, it has also been shown that it may also depend on details of the dislocation structure that forms, with any long-range dislocation stress field playing a significant role. Furthermore, even if SE is a driving force for grain boundary migration, this migration also depends on the mobility of the moving interfaces as well as the grain size and shape distributions, which are in turn affected by the distribution of orientations and misorientations within the deformed stat.

Some previous works mainly performed on Cu and Fe have established a clear link between SE and recrystallization and have led to the conclusions that (i) the zones within the deformed sample associated with a high value of this stored energy always disappear first and (ii) a strong SE gradient (DE) is necessary for the growth of a nucleus and the achievement of the recrystallization process. This is generally also assumed for the simulations of texture evolution during annealing, for which a DE threshold value for the initiation of boundary migration is arbitrarily set in order to reproduce the experimental evolutions. However, even if it is possible to reproduce annealing textures based on these observations, it is still impossible to be predictive, since the evolution of texture and microstructure during annealing may depend on very local details within the microstructure, such as local heterogeneities of hardening state or grain orientations, which are hard to completely characterize.

The present paper will first briefly recall the various methods used for evaluating stored energy and illustrate them by data obtained in various metallic alloys. It will then present some more recent data obtained mainly in copper using a laboratory X-Ray diffraction (XRD) equipment which will demonstrate, through a comparison with simulations performed with dislocation –based hardening laws, that XRD allows to get data which are complementary from those obtained through scanning electron microscopy, often preferred because of their greater facility of use.