Yambo Community Forum

Dear Yambo developers and users,

I am doing GW+BSE calculation for metall materials.
But after running yambo, the out file reported "The system is a metal but Drude term not included".
Is this value needed for both GW and BSE calculatins?
If so how do I calculate this value(Drude team)?

Best regards.
Yilong Luo

Dear Yilong,

please sign your post with your name and affiliation. This is a rule of the forum, and you can do once for all by filling the signature in the user profile.

The Drude term can be inserted empirically by considering plasmon frequency and lifetime from literature. Otherwise, you can estimate it at independent particle level.

You can have a look at this thread for a discussion on the topic:
viewtopic.php?t=2347

Best,
Daniele

Dear Daniele,

Thanks for your reply.I will read the papers.

Dear Daniele,

I have experimentally obtained a plasma frequency, can it be directly used as the real part of the Drude term?
I read your book on The magneto-optical Kerr effect(https://wiki.yambo-code.eu/wiki/index.p ... err_effect) in detail ,and I wonder how you get the real and imaginary parts os Drude terms?
If I do not add the Drude term to the GW calculation, but add Drude to the BSE ,will I get the correct value?

Best regards.

Dear Yilong,

the Drude terms are empirical, you should get from the literature (e.g. experiments) and they are respectively the Drude Plasmon and its width (related to the lifetime).
Please note, that finite q contribution of intraband transition are correctly taken into account, and you can calculate the RPA response at finite small q and see the plasmon peak. Increasing the k points sampling and looking at plasmon frequency at the smaller q you can extrapolate the q=0 contribution and use this value in the Drude term.

In principle, you can also ignore it and converge the calculation wrt the k points as finite q contribution are anyway taken into account. The
long wavelength limit will be not considered but its weight it is reduced when more and more k points are considered. In some case, anyway,
a very large k point sampling is needed to reach convergence.

If you add it in the BSE this will add the Drude plasmon in the absorption spectrum, but not in the screening term.

Best,
Daniele