Definition of polarizability alpha in 1D

[S_Grillo]

Dear Yambonauts,

I have an issue with the definiton of the polarizability alpha in the 1D case.
I do not understand how Yambo calculates it and how I should relate it to the known formula.
To me, it looks like the values I get from the BSE are too big, say, if I turn off the cutoff and convert the eps2 to alpha.
I am missing something.

Find attached the optical spectra calculated at the IPA and BSE/G0W0 level, the former with and without the coulomb cutoff included.
Let me know if you need other files.

Best

Simone

[Daniele Varsano]

Dear Simone,

I do not know what to you refer with the "known" formula in 1D. Can you please indicate some reference?
What yambo calculate is alpha=-X/q^2*eps_2_alpha
In 1D eps_2_alpha = Vol/L, here L is the non-periodic length.
and the q-->0 limit should not have problems.
Actually, the BSE results seem reasonable, or at least are not crazy, but I do not know what is your reference.
In contrast, the IPA value of alpha seems nonsense and note that the "truncated" potential here does not enter.

Anyway, it is clear that something is not working in the calculations, and it should be inspected.

I will try to reproduce your problem in a 1D system and if needed I will ask you an example to reproduce it.

Best,

Daniele

[Daniele Varsano]

Dear Simone,

just noticed you are using a very old version of the code.
Many things changed since then. Please update your yambo release and let us know if it fix the problem.

Daniele