Is there any proof that the tilt improves the signal, or does it only affects the pattern center positioning since the working distance can be increased so that the gnomonic distortion are not that dramatic? If they are, the band detection is not that precise so that the orientation determination is less accurate.
Well, first thing coming to my mind is that this should be following Lambert's cosine law, and the signal intensity should therefore increase when tilting. Why would the band detection be less accurate when tilting? Due to increased path length through sample thickness or due to increase of the interaction volume caused by tilt?
Actually the interaction volume of electrons which hit the detector can become smaller for a slightly tilted sample.
To my opinion, the reason is just in the design and geometry of the sample holder versus elevation angle of the EBSD detector: varius SEMs have different available z-movement, different dimensions of the stage craddle and different elevation angles of the EBSD detector. There is then the need to optimize the geometry of the sample vs. detector to be able to insert the detector nose as close as possible beneath the sample.
I think that tilting the sample will only affect the pattern center position and, of course, the particular segment of the orientation space you see on the detector screen.
I guesst the effect of tilt on the signal is barely measureable since the tilt angles are generally small and the samples thin, so the beam path is more or less the same. If there is any detectable effect, it would be improvement in spatial resolution as can be anticipated from Nathnanael's answer.
If this is of general interest, I could give a talk about it on the next EBSD Arbeitskreistreffen in germany. Currently I'm doing some experiments and simulations on the topic of sampletilt and detectorsignal (still in the process, so I can't say something definitive now, but in a few months it should be ok).
PC should definitely not be affected. But I'm curious about the reduced 'interaction volume' due to tilt, any publications on that you might want to share? By interaction volume, which in case of EBSD/TKD might be an arguable term, I'm referring to the volume from which the EBSP stems. Maybe I'm thinking too much EBSD and not TKD here. Is the difference in path length due to tilt working as sort of energy filter resulting in higher spatial resolution, or what is that based on?
The problem is, that people normally consider all electrons for interaction volume. But if you only look at the electrons, that actually hit the detector, the results change. For those electrons the pathlength can decrease if the sample is tilted (depending on the setup and detector position relative to the sample).
- Badges
- Science topic
- Similar topics
- Physical Sciences
- Materials Science