Adding DrudeWBS term for metallic systems

[Zafer Kandemir]

Dear Developers,

I calculated the optical spectrum of metallic systems with GW+BSE level. I plotted EELS graph. How to decide the Drude variables for this system? "DrudeWBS (real part, imaginary part) # eV" Is it right?
"The imaginary part of Drude corresponds to the width of the peak appearing in the EELS": In my figure, is the value of about 0.2 corresponding to this expression? Or 1.9?

Best regards.
Zafer

[Daniele Varsano]

Dear Zafer,

How to decide the Drude variables for this system?

This is an empirical parameter, you can choose it by extracting from the literature (either theo or experiments).
Another way is to calculate the eels for very small q value (compatible with your q point grid) and extrapolating to the lim q->0

DrudeWBS (real part, imaginary part) # eV Is it right?

Yes, right

"The imaginary part of Drude corresponds to the width of the peak appearing in the EELS": In my figure, is the value of about 0.2 corresponding to this expression? Or 1.9?

It seems to be 0.2eV, this is indeed the width of the Lorentzian peaked on the plasmon energy.

Best,
Daniele

[Zafer Kandemir]

Dear Daniele,

Thank you for your reply. Can we call the real and imaginary part of the Drude: the plasma frequency w_p (in eV) and the collision frequency (in eV)?

Best regards.
Zafer

[Daniele Varsano]

Dear Zafer,
it is not clear to me what you mean by collision frequency, anyway the imaginary part it is related with the lifetime of the excitation.
Best,
Daniele