l,m-decomposed DOS with grouped split degeneracies

[FPrith]

When evaluating l,m-decomposed DOS plots for systems containing transition metals (e.g., Cr₂O₃, ScN, etc.), is it possible to group several d bands together, such as by separating the t₂g and eg bands, and visualize them in DOS plots?
If there is already an Issue related to this topic or a description on how to do that in your manual, I apologize for having missed it.

[ajjackson]

I don't believe it is. Your best bet is probably to write all the orbitals to a text file with the --orbitals option and then create a custom plot which combines things as desired. If using Matplotlib, you could point to (or copy) the relevant style files from Sumo to get a consistent appearance:
https://matplotlib.org/stable/users/explain/customizing.html#customizing-with-style-sheets
sumo_base.mplstyle
sumo_dos.mplstyle

I think this falls under the wider problem that

• any imaginable grouping of PDOS contributions will be useful for somebody's research
• it is not practical to support them all as special cases of the existing options/syntax.

We really need a more general solution to this problem, where some domain-specific language can be used to specify arbitrary combinations and then the common cases are just "shortcuts" to this system.

[FPrith]

Thank you very much for your quick response!

I found your suggestion "to write all the orbitals to a text file with the --orbitals option and then create a custom plot which combines things as desired" to be very reasonable.
I have confirmed that this approach is indeed feasible using a dat file.
I will proceed with my analysis using this process.

[ajjackson]

Great! Good luck with your research. As I say, this is something which a better design could improve on. Hopefully one day there will be a bit more resource on Sumo again and we can do something about it!