Modelling the mechanical response of structural steel requires the knowledge of the evolution of the density of dislocations with strain. The heat treatment applied after quenching in the case of martensitic steel plays a significant role onto the evolution of the microstructure and mechanical properties. Recovery results in the rearrangement of dislocations, known to play a key role onto the mechanical properties of the martensite [1], leading to dislocation annihilation and/or sub-grain boundaries formation and polygonization. The dislocation density thus needs to be integrated as a key parameter in predictive models to fully describe the mechanical properties. However, dislocation density is not straightforward to estimate. Transmission Electron Microscopy (TEM) seems to be the obvious choice to reach an accurate measurement. However, this technique lacks easiness of use due to its difficulties in the sample preparation, observation, and due to its substantial cost. An alternative is the x-ray diffraction (XRD). Studies brought improvement of the Williamson and Hall model [2] based on width of diffraction peak to estimate dislocation density. A modified and improved model emerged [3] and established a parameter M, attributed to the arrangement of dislocations, necessary to be estimated for an accurate measurement [4] but seems to be neglected [5,6].

 The present study aims at comparing measurements based on XRD and TEM. Synchrotron XRD measurements were carried out on tempered martensitic steels (as shown on Figure 1). Dislocation densities were estimated following the modified Williamson and Hall model. These ones were then compared to TEM observations.


References

[1] F. Christien; M.T.F. Telling; K.S. Knight Scripta Materialia, 2013, 68, 506-509.
[2] G.K. Williamson; W.H. Hall Journal of Acta Metallurgica, 1953, 1, 22-31.
[3] T. Ungar; A. Borbely Applied Physics Letters, 1996, 69, 3173-3175.
[4] A. Borbély Scripta Materialia, 2022, 217, 114768.
[5] C. Couchet; S. Allain et al. Materials Characterization, 2021, 179, 111378.
[6] B.B. He; B. Hu et al. Science, 2017, 357, 1029-1032.