Experimental measurement of dislocation density in metallic materials: A quantitative comparison between measurements techniques (XRD, R-ECCI, HR-EBSD, TEM)

The dislocation densities were measured on the same samples using transmission electron microscopy (TEM), scanning electron microscopy (electron channeling contrast imaging (ECCI) and high-angular-resolution-electron backscattered diffraction (HR-EBSD)), and X-ray diffraction (XRD). Notably, these different methods do not observe the same types of dislocations, i.e., statistically stored dislocations (SSDs) and/or geometrically neces-sary dislocations (GNDs). ECCI and TEM imaging are direct-measurement techniques, whereas HR-EBSD and XRD are indirect methods. Therefore, a quantitative comparison of the measurements obtained using these four techniques on undeformed and deformed duplex steels is proposed. For low deformation, where the dislocation density is quite small (1 -5 x 1013 m(-2)), imaging methods are rather performant, whereas XRD measurements suffer from high uncertainty levels. HR-EBSD measurements show results that are in good agreement with the other methods for these deformation levels. For higher deformation levels (with dislocation densities above 1 -3 x 1014 m-2), imaging methods are no longer relevant because of the increasing uncertainty arising from local contrast variation and overlapping of dislocations. The different results obtained highlight the necessity of taking a step back on each method used. Correctly defining what is to be measured (SSDs or GNDs), in which condition (solid material or thin plate) as well as the parameters (pixel size, area, etc.) and their bias is essential, especially if the objective is to use the measurement in a micromechanical model.