Forum: Open Discussion - Thread: Welcome to Open Discussion - Electron Holography Analysis Tools on SourceForge.JP

Parameters for the parabolic cluster

ragekeeper (ragekeeper) — Wed, 30 Nov 2011 09:09:13 GMT

Electrons cannot travel large distances in real solids without suffering inelastic losses, so that the region which actually contribute to the emission of elastically scattered electrons defines a finite cluster surrounding the emitter. Wor example we know maximum distance Rmax from the emitter up to which atoms we considered to belong to the cluster. Atoms further away are ignored. In other words, Rmax=|ri-r0|+zi-zsurf, where zsurf is the position of the surface and r0 is the the emitter. This criterion leads to a parabolic shape of cluster with the focus coinciding with the emitter. Parameteres for TMSP cluster can be defindes as follows: 1. Maximum depth of cluster Hmax=(Rmax+z0)/2 2. Cone angle alpha=arccos(z0/Rmax) 3. Back scattering depth Hbackscat=(Rmax-z0)/2 4. Minimum radius at (Z=0) rmin=sqrt(Rmax^2-z0^2) The rest is simple. Create cluster for each emitter and save. For the calculations we must stack together clusters into one file. For this copy and paste tegs .... There I found a problem. If we have a surface defect or large Rmax, then we must do this boring operation for each emitter of, for example, one hundred amount :(

RE: Welcome to Open Discussion

ragekeeper (ragekeeper) — Tue, 25 Oct 2011 09:42:11 GMT

I supposed that Limited space parameters is a way to select sphere-shaped regions around expected atom positions. For example I load 307 limited space positions. That lead to robust calculation and better RMS factor. But now it take five minutes to load xml file with real space. And if I try to increase real space size from 10A to 20A with fixed step 0.1A, my computer run out of memory, even if the radiuses of limited spaces are equal to 1 voxel, so 307 voxels in total.

RE: Welcome to Open Discussion

T. Matsushita (sp8matusita) — Thu, 04 Aug 2011 00:53:52 GMT

Thank you for the information of the format. I just accepted you to our project. Please login to our project and create a ticket of the issue. You can post the result of the test and so on. I want to support your research. Thank you.

RE: Welcome to Open Discussion

ragekeeper (ragekeeper) — Wed, 03 Aug 2011 06:22:52 GMT

In Igor Pro isosurface is appended to Gizmo window by coomand AppendToGizmo /N=GizmoWindow isosurface=isosurfaceWave Where isosurfaceWave is a 3 dimension wave function (array, matrix) whicj can be loaded from *.ITX file. For example: IGOR WAVES/D/N=(2,2,3) wave0 BEGIN 0 1 2 3 1 2 4 8 00 01 10 11 END Note that empty lines between layers are needed. But if you do not like that text files weight too much you can use binary format. From Igor Pro help: "Some Igor users have written custom programs that write Igor Binary files which they load into an experiment. Igor Technical Note #003, "Igor Binary Format", provides the details that a programmer needs to do this. See also Igor Pro Technical Note PTN003." I found this note in folder WaveMetrics\Igor Pro Folder\Technical Notes\Igor Tech Notes\TN003 Igor binary format. There are also source code on Visual C++ v5 as example. I am confused about how to report bugs. Should I create new ticket, mail to member of the project or post here?

RE: Welcome to Open Discussion

T. Matsushita (sp8matusita) — Sat, 30 Jul 2011 05:44:10 GMT

OK. Now, it is difficult to satisfy your offer because I don't know the detail of the Igor format of 3D image. But, if the format is clear, I can create it. If you wish, please join in our project, and please provide the informations.

RE: Welcome to Open Discussion

ragekeeper (ragekeeper) — Fri, 29 Jul 2011 09:21:00 GMT

Question: How to save real space in Igor Pro file? I need this because for me it is easier to create isosurface in Igor Pro than in Ligthwave 3d. As for now I try to save each section as *.itf file and then stack all layers together in one 3d array but it took too much time and efforts. So some convertor is needed.

Welcome to Open Discussion

Administrator (admin) — Thu, 02 Apr 2009 07:50:39 GMT

Welcome to Open Discussion