[QE-users] turbo_eels.x results

[Elio Physics]

Dear Iurri,

Thank you very much for  your detailed response.

Elie
________________________________
From: users <users-bounces at lists.quantum-espresso.org> on behalf of Iurii TIMROV via users <users at lists.quantum-espresso.org>
Sent: Friday, May 20, 2022 6:16 AM
To: Quantum Espresso users Forum <users at lists.quantum-espresso.org>
Subject: Re: [QE-users] turbo_eels.x results


> My question is  How would I compare these results to those outputted by the epsilon.x toolset?


epsilon.x is based on the independent particle approximation (IPA).


turbo_eels.x has the input keyword "approximation":

- 'TDDFT' : use the adiabatic LDA/GGA exchange-correlation kernel

- 'IPA' : use the independent particle approximation

- 'RPA_with_CLFE': Random Phase Approximation (RPA) with Crystal Local Field Effects (CLFE)


So if you want to compare turbo_eels.x  with epsilon.x you need to use approximation='IPA'


> epsilon.x outputs the different tensor components (xx,yy,zz), whereas turbo_spectrum.x just outputs one column for the real part of the dielectric function and another column for the imaginary part.


In turbo_eels.x the perturbation is a plane wave exp(iq*r). In the limit when q->0 we have exp(iq*r) ~ iq*r + ...

Given that q=(qx,qy,qz) we have iq*r = iqx * x + iqy * y + iqz * z

In your case you have chosen q=(qx,0,0), hence iq*r = iqx * x, so you have the x component of the perturbation.

In turboEELS the susceptibility is a scalar \chi(q,w). One has to modify the code to have instead three components of the response: x,y,z.

So the total response should be \chi_alpha_beta, where alpha=x,y,z and beta=x,y,z.

The bottom line: turboEELS was not designed to compute absorption spectra but to compute electron energy loss spectra. If one wants to use turboEELS to compute the full tensor \chi_alpha_beta for the absorption spectroscopy, some modifications of the code are needed. The best solution would be to generalize turbo_lanczos.x to work for solids including SOC, but this requires some work.


Note that Yambo also has TDDFT, and it works for solids (though I do not know if SOC is implemented). So I would give it a try.


HTH


Greetings,

Iurii


--
Dr. Iurii TIMROV
Senior Research Scientist
Theory and Simulation of Materials (THEOS)
Swiss Federal Institute of Technology Lausanne (EPFL)
CH-1015 Lausanne, Switzerland
+41 21 69 34 881
http://people.epfl.ch/265334
________________________________
From: users <users-bounces at lists.quantum-espresso.org> on behalf of Elio Physics <Elio-Physics at live.com>
Sent: Thursday, May 19, 2022 7:48:43 PM
To: Quantum Espresso users Forum
Subject: [QE-users] turbo_eels.x results

Dear QE users,

I am performing some absorption calculations using the turbo_eels.x and turbo_spectrum.x executables at q1=0.0001, q2=0.0, q3=0.0, as is advised by the developers of these toolsets.
The reason I am not using the tubo_lanczos.x  is because I am not using a supercell and I have included the spin-orbit coupling (SOC).
I have got some interesting results. My question is  How would I compare these results to those outputted by the epsilon.x toolset?
epsilon.x outputs the different tensor components (xx,yy,zz), whereas turbo_spectrum.x just outputs one column for the real part of the dielectric function and another column for the imaginary part.

Would it be reasonable to compare these results to the average epsilon_1 and epsilon_2 given by epsilon.x?

Regards

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20220520/c69c9992/attachment.html>