Bug report: Implememtation of Spin orbit coupling in CSC for quantum ESPRESSO

[wk-0720]

I would like to calculate the 4f orbitals of Ce-based materials using the CSC method with Spin-Orbit Coupling (SOC). However, when I ran the calculation, I encountered the following error in Quantum ESPRESSO:

 Error in routine iosys (1):

 DMFT update not implemented with nspin > 1

I am using solid_dmft version 3.3.X, TRIQS, and Quantum ESPRESSO version 7.5
Is it not possible to perform CSC calculations with SOC in the current version?

[the-hampel]

Hi @wk-0720 ,

Yes I fear that we just did not have the developer capacity to test and work on this. I do not think that there is a fundamental barrier here, but it is not tested in QE. Let's say I would assume that updates for magnetic calculations (no SOC) should be very easy to implement in QE. SOC calculations maybe less trivial but are doable. Anything you would like to add @phibeck @alberto-carta that I do not have on my radar?

Let's say if you are interested in this feature and are willing to implement and test this someone will probably be able to help you here.

On the VASP interface on the unstable branch we just added this features in case that is an option.

Best,
Alex

[wk-0720]

Thank you for your reply.

I plan to attempt to modify the source code myself to address this issue.I would strictly appreciate it if you could share any useful information or advice to guide me.

Best regards,

[alberto-carta]

Hi wk, sorry that I answer only now. Me and @phibeck are the maintainers for the QE interface right now.

I currently do not have the bandwidth to work on this on my own, but you can count on my full technical support for the implementation.

I think the main work is less on the solid_dmft side, and more on the Quantum Espresso side.

The main file that handles the DMFT charge update logic is in PW/src/sum_band.f90, around line 500.

I am extremely happy to provide information on exactly how the logic is implemented. If this becomes too lengthy for a github issue, we might want to bring this on a PM.

Cordially,

Alberto

[wk-0720]

Thank you for the reply.

I plan to read the relevant papers on CSC and review the source code myself to clarify the issue before attempting the modification. I may reach out to you in this thread again if I need further advice. I would appreciate your support at that time.

Best regards,