Ion etching is assumed to be ideal for EBSD preparation which might be correct for phases which do not transform during bombardment, and when diffraction patterns are only applied for a rough orientation determination. However, I wonder how much the surface distortion by ion implantation affects misorientation interpretations (e.g. geometrically necessary dislocation (GND) applications) as well as the pattern quality as analytical information. Don't we affect our (intrinsic) property of the bulk material by an inacceptable (extrinsic) surface manipulation since only the first few nanometers are mainly forming the diffraction signal for EBSD? When ion etching is not useful despite a pattern formation?
It's a very interesting question, and ties into a large question of what is the importance and value of orientation precision measurements in EBSD. I looked at recent EBSD publications, and around 60% involved some sort of deformation analysis. How important are small changes in orientation in comparison to measuring something like grain size?
...and we are not yet discussion about the real information volume (very anisotropic in x, y and z) of the displayed EBSD pattern which is in general assigned as information for the respective pixel dimension. This is in fact only in very rare cases correct since either the covered area is either much smaller for low magnification measurements, or definitely bigger for high-resolution maps.
From my experience I may add that ion bombardment does not severely affect the precision of orientation measurement and give excellent patterns. But it is how you throw your ions. In cross sectional ion polishing (CSIP) the ions come at a low angle (typ. 5º or less) and remove the top damaged layers exposing the undestirbed ones. Ions are more likely to get reflected than penetrating the material. Similar ion-matter interactions apply for FIB and TEM ion preparation. When EBSD datasets are observed the variation in orientation wthin a grain is about 0.5º.
The problem regarding EBSD is only that the Information depth is in the scale of some nanometers. The top layer is clearly effected by ion etching. For TEM investigations this small layer might not be very important since4 the main information is built in the entire lamella. In so far strain in the top layer is not registered in TEM, but EBSD patterns are effected by this influence and might change the orientation locally so that any conclusion is doubtful.
The 0.5 degree approach for EBSD orientation measurements is also a very rough estimation and needs to be discussed from case to case. Recent developments considering more bands or improved refinement strategies enables clearly better values, but they have nothing to do with the intrinsic uncertainties generated by sample preparation.
- Badges
- Science topic
- Similar topics
- Engineering
- Materials Engineering
- Powders