spin-orbit coupling

[8813204602]

Dear all,

I have already done the GW calculations without considering the spin-orbit coupling for a two-dimensional material. Now I want to include the spin-orbit effect as well:

1) Should I just import SOC in espresso input files or should I make changes to Yambo inputs as well?

2) Do I have to do all the convergence tests for GW calculations again from the beginning or can I use the same convergent parameters obtained from the GW calculations without spin-orbit effect?

Thanks & Regards ,
Mitra Helmi

[Daniele Varsano]

Dear Mitra,

1) Should I just import SOC in espresso input files or should I make changes to Yambo inputs as well?

You just need to perform a QE calculation including SOC, then p2y will recognise it and all the calculation will be done with spinorial wfs using the same inputs.

2) Do I have to do all the convergence tests for GW calculations again from the beginning or can I use the same convergent parameters obtained from the GW calculations without spin-orbit effect?

You can keep probably most of the parameters you obtained without considering SOC, but paying attention to the number of bands (both in X, G, BSE) as the number of occupied states in the presence of SOC is doubled.

Best,
Daniele

[8813204602]

Dear Daniele,

When I use Yambo 5.0.3 to perform BSE calculations on GW calculations with SOC, it writes in the kernel loop section of the report file and in the output:

[SOC] Perturbative SOC : no

Given that the results obtained are different from the case without SOC, does this mean that SOC is not included in the BSE calculations or is it just a typo?

The report file of BSE calculation with SOC is attached.

Best,
Mitra Helmi

[Daniele Varsano]

Dear Mitra,

the SOC is included, the message tells you that the SOC term has been included explicitly and not in a perturbative way.

Best,
Daniele