Ivanlh20 / multem

MULTEM is a powerful and advanced collection of C++ routines with CUDA support, designed to perform efficient and accurate multislice simulations for various TEM experiments such as HRTEM, STEM, ISTEM, ED, PED, CBED, ADF-TEM, ABF-HC, EFTEM, and EELS.
GNU General Public License v3.0
65 stars 26 forks source link

Potential slicing #30

Closed emichr closed 5 years ago

emichr commented 5 years ago

I have some questions about slicing when using ePS_dz_Proj potential slicing in MULTEM. That I hope is relatively straight forward to answer.

First of all, is the first slice defined to be from z=0 to z=dz? or from z=-dz/2 to z=dz/2?. I.e. will an atom positioned at z=0 be situated in the middle of a slice, or at the boundary of a slice? If the latter case, should one offset atoms in order to make a slice belong to one atomic plane alone?

Secondly, is the slice thickness spec_dz defined as the complete slice thickness, or the half-thickness of the slice? This will of course be connected to the first question.

I've made a quick and dirty figure below to try to illustrate what I mean.

slicing

If I've misunderstood something about how MULTEM computes or uses the slices, please let me know.

Ivanlh20 commented 5 years ago

Check out the following examples.

https://github.com/Ivanlh20/MULTEM/blob/master/mex_examples_multem/example_specimen_planes.m

https://github.com/Ivanlh20/MULTEM/blob/master/mex_examples_multem/example_specimen_slicing_show.m

https://github.com/Ivanlh20/MULTEM/blob/master/mex_examples_multem/example_specimen_slicing_zone_axis.m

El mar., 7 may. 2019 a las 16:20, Emil Christiansen (< notifications@github.com>) escribió:

I have some questions about slicing when using ePS_dz_Proj potential slicing in MULTEM. That I hope is relatively straight forward to answer.

First of all, is the first slice defined to be from z=0 to z=dz? or from z=-dz/2 to z=dz/2?. I.e. will an atom positioned at z=0 be situated in the middle of a slice, or at the boundary of a slice? If the latter case, should one offset atoms in order to make a slice belong to one atomic plane alone?

Secondly, is the slice thickness spec_dz defined as the complete slice thickness, or the half-thickness of the slice? This will of course be connected to the first question.

I've made a quick and dirty figure below to try to illustrate what I mean.

[image: slicing] https://user-images.githubusercontent.com/16645501/57306829-ca928d80-70e3-11e9-9d60-8d77b463f8e9.png

If I've misunderstood something about how MULTEM computes or uses the slices, please let me know.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/Ivanlh20/MULTEM/issues/30, or mute the thread https://github.com/notifications/unsubscribe-auth/ACBRRN6TX3QLYQG6E2CKSXLPUGF4FANCNFSM4HLJKQWQ .

emichr commented 5 years ago

Thanks for your help!

By playing around with parameters in the examples, it seems that the first slice is centered on z=0, and spec_dz is the full slice thickness (although some of the slices are sometimes thicker when needed, e.g. the first slice has thickness 5.78 Å in example_specimen_slicing_zone_axis when modified with input_multislice.potential_slicing = 2 and spec_dz=5).