Electron Backscatter Diffraction (EBSD) and Energy Dispersive X-Ray Spectroscopy (EDX)

Back to main page / Research

Electron backscatter diffraction patterns for materials with four different crystal structures are shown in fig. 1. The bands in the EBSD patterns correspond directly to the lattice planes of the crystal being examined and the points where several bands cross are the crystallographic zone axes. The angles between several zone axes can be used to identify different crystal structures, for example in multi-phase polycrystalline materials. In addition, one such EBSD pattern can be used to determine the orientation of the crystal in 3D. By scanning the electron beam over an area of the sample and recording EBSD patterns from each point enables a map of the crystal orientations to be recorded.

 

 

Fig. 1 EBSD patterns from (a) Nd2Fe14B – tetragonal, (b) Nd2O3 – hexagonal, (c) Nd1.1Fe4B4 – orthorhombic and (d) NdO – cubic.

                      

Combining EBSD with chemical analysis by energy dispersive x-ray spectroscopy (EDX) gives a powerful tool for identification of phases. For example, we have used this to map the distribution of all phases in the microstructure of a high performance Nd-Fe-B sintered magnet for the first time. Fig. 2 shows a backscattered electron image (bottom left), EBSD crystal structure map (top left) and EDX maps where the intensity shows the Fe and Nd distribution (bottom right) and oxygen distribution (top right). Direct correlation of the chemical and structural information allows unambiguous identification of the phases present.

 

 

Fig. 2 Backscattered electron image (bottom left), EBSD crystal structure map (top left, where blue = Nd2Fe14B, green = Nd2O3, purple = Nd1.1Fe4B4 and red = NdO/fcc Nd-rich) and EDX maps showing the Fe and Nd distribution (bottom right) and oxygen distribution (top right) [TGW Acta].

 

  The EBSD data can be used to map the orientation of the grains in the microstructure, as is shown for the Nd2Fe14B grains in fig. 3. In fig. 3a, the grains are coloured according to the angle between the c-axis and the out of plane reference direction. The grains are all blue which shows that c-axes are well aligned. This c-axis alignment is critical in order to develop high magnetic remanence. In fig. 3b, the grains are coloured according to the angle between the a-axis and the in plane reference direction. The grains have all different colours which shows that there is no preferred orientation in this direction. From these images we can show the well known <001> fibre texture present in the Nd2Fe14B grains in high performance Nd-Fe-B sintered magnets.

 

Fig. 3 EBSD orientation maps for the Nd2Fe24B grains. In (a) the grains are coloured according to the angle between the c-axis and the out of plane direction. In (b) the grains are coloured according to the angle between the a-axis and the in plane horizontal direction [TGW Acta].

 

The step size used in figs. 2 and 3 was 400 nm. There are therefore many measurement points within individual grains. We can qualitatively determine the defect density for each phase by calculating the misorientation angles between all possible pairs of points within individual grains. If a perfect single crystal is present, all of these misorientation angles should be zero. Fig. 4 shows this analysis presented as histograms for the various phases present in the Nd-Fe-B sintered magnet. The misorientation angles for the Nd2Fe14B phase are all within the experimental error of the technique (~ 1°) and therefore the defect density in this phase is shown to be very low. In contrast, the Nd-rich phases have much larger misorientation angles within individual grains which indicates the presence of defects. For further details, see [TGW Acta].

 

Fig. 4 Intragranular misorientation angles presented as histograms for the various phases in a Nd-Fe-B sintered magnet [TGW Acta].      

 

References:

[TGW Acta]   T.G. Woodcock and O. Gutfleisch, Multi-Phase EBSD Mapping and Local Texture Analysis in NdFeB Sintered Magnets, Acta Materialia, 59 (3), 2011, 1026-1036

 

↑Top