Skip to main content

Timeline for d orbitals PDOS analysis

Current License: CC BY-SA 4.0

16 events
when toggle format what by license comment
May 1, 2024 at 19:25 comment added Chengcheng Xiao In its essence, yes. The LORBIT=12 part is what does the magic.
May 1, 2024 at 17:16 comment added Kratos1611 It's fantasting @ChengchengXiao. Thank you so much. I'm sure this will work. I will let you know after applying on my system. I think we can connect through email. I have an one more question, that to plot just PDOS , I should perform again SCF with high NEDOS value and with LORBIT = 12, right? then I can continue the postprocessing calculation using your code.
May 1, 2024 at 13:19 history edited Chengcheng Xiao CC BY-SA 4.0
added automatic search of the best Eular angle example to GitHub repo.
Apr 29, 2024 at 20:44 comment added Chengcheng Xiao Let us continue this discussion in chat.
Apr 29, 2024 at 20:23 comment added Kratos1611 Thank you so much Xiao for your efforts. I will apply it on my system and will let you know. One think it is also applicable for $m_l$ projected DOS right?
Apr 29, 2024 at 19:41 comment added Chengcheng Xiao Yeah, the example is now live at github.com/Chengcheng-Xiao/RotSph/tree/master/example/BTO.
Apr 29, 2024 at 19:37 history edited Chengcheng Xiao CC BY-SA 4.0
Added link to example. Slight editing to make the texts easier to understand.
Apr 29, 2024 at 18:37 comment added Kratos1611 Ok, I think you meant LORBIT = 12. Yes add an example it will be very helpful.
Apr 29, 2024 at 17:31 comment added Chengcheng Xiao Update: I got IORBIT=12 working, will add an example to the GitHub repository.
Apr 29, 2024 at 14:37 comment added Chengcheng Xiao On second thought, this will not work for IORBIT=11 as it only outputs the square of the projection coefficients. While IORBIT=12 might allow this to work, I need more time to investigate. In the meantime, I think you'll have better luck with: vasp.at/wiki/index.php/LOCPROJ.
Apr 29, 2024 at 9:44 comment added Kratos1611 @ChengchengXiao So to use your code from where should I start, I mean which files should I use according to VASP files DOSCAR, PROCAR or POSCAR ? Shoul I use your mathametical notebook or rotsph.py file?
Apr 26, 2024 at 20:41 comment added Chengcheng Xiao Okay, in short, you need to construct a rotation matrix (3x3) that transforms to your desired frame, then use my code to generate the transformation matrix (for d-orbitals, a 5x5 matrix). At each energy, you have 5 PDOS data (note: order matters! for d-orbitals: d_xy d_yz d_z2 d_xz d_x2-y2), dot that with the obtained transformation matrix and you get a new set of 5 optimized PDOS data, voila.
Apr 26, 2024 at 20:32 comment added Chengcheng Xiao I'm sorry that repository was a spur-of-the-moment thing for me and I've not tried it out. But how it was supposed to work should be recorded in the Mathematica notebook.
Apr 26, 2024 at 20:22 comment added Kratos1611 Exactly, you get my issues correctly. Can you share some example how to use your code? It's great and if I provide you VASP's POSCAR file, can you show me how to transform coordinates writing the code?
Apr 26, 2024 at 20:12 comment added Andrey Poletayev From the follow-on post by OP the structure is a layered material with sodium, most likely a battery cathode ;). These things are usually sufficiently covalent to have a lot of oxygen $p$ mixing with metal $d$ at all energies plotted here. See for example a review: iopscience.iop.org/article/10.1088/1361-648X/aacb05/meta
Apr 26, 2024 at 19:58 history answered Chengcheng Xiao CC BY-SA 4.0