Dear Clarence Leung,

> But I can find the input description of turbo_EELS .


The description of the input parameters for turboEELS is here: 
TDDFPT/Doc/INPUT_EELS.def


> What is the meaning of q ?


It is a transferred momentum. Have a look at these papers:


"Electron energy loss and inelastic x-ray scattering cross sections from 
time-dependent density-functional perturbation theory"
Authors: I. Timrov, N. Vast, R. Gebauer, and S. Baroni
Source: PHYSICAL REVIEW B   Volume: 88   Article Number: 064301   Published: 
AUG 2 2013
See also: PHYSICAL REVIEW B 91, 139901(E) (2015)

"turboEELS - A code for the simulation of the electron energy loss and 
inelastic X-ray scattering spectra using the Liouville-Lanczos approach to 
time-dependent density-functional perturbation theory"
Authors: I. Timrov, N. Vast, R. Gebauer, and S. Baroni
Source: Computer Physics Communications  Volume: 196  Article Number: 460 
Published: 4 JUNE 2015


> How can we get (x y z) 3 Polarization direction ? use the command ipol = 4 ? 
> like the turbo_lanczos?


No, "ipol" variable is not present in turboEELS.


As I wrote above, q is the transferred momentum, it is a vector. So depending 
on the direction of q you can get different behavior of the loss function 
-Im[1/eps] and Im[eps] (it depends also on the modulus of q). Have a look at 
this paper: I. Timrov et al., Phys. Rev. B 95, 094301 (2017).

HTH


Iurii



Dr. Iurii Timrov
Postdoctoral Researcher
Swiss Federal Institute of Technology Lausanne (EPFL)
Laboratory of Theory and Simulation of Materials (THEOS)
CH-1015 Lausanne, Switzerland



________________________________
?件人: [email protected] <[email protected]> 代表 Timrov Iurii 
<[email protected]>
?送??: 2017年3月6日 22:43:42
收件人: [email protected]
主?: Re: [Pw_forum] Error in Turbo_lanczos.x to calculate the adsorption 
coefficiency


Dear Clarence,


When posting to pw_forum please do not forget to write your affiliation!


> As far as I know, the turboTDDFT is not extended to metals,


Yes, you are right.


> So we should use fixed occupation in the scf calculation, right?


Yes


> However, if I use fixed occupation in scf, the error is
> ' charge is wrong: smearing is needed  '. Because the system considered is 
> odd electron.

Then you have to use smearing. I think there is no workaround with turboTDDFT 
for metals: one has to extend it to metals.

> What should I do to calculate the adsorption coefficiency of system with odd 
> electrons.


aBsorption, not aDsorption


Probably you may try to use turboEELS, which is extended to metals. Namely, use 
very small q (but not exactly zero) and plot the imaginary part of the 
dielectric function.


HTH


Iurii


--
Dr. Iurii Timrov
Postdoctoral Researcher
Swiss Federal Institute of Technology Lausanne (EPFL)
Laboratory of Theory and Simulation of Materials (THEOS)
CH-1015 Lausanne, Switzerland

________________________________
From: [email protected] <[email protected]> on behalf of 
LEUNG Clarence <[email protected]>
Sent: Monday, March 6, 2017 9:51 AM
To: [email protected]
Subject: [Pw_forum] Error in Turbo_lanczos.x to calculate the adsorption 
coefficiency


Dear PWscf users,


As far as I know, the turboTDDFT is not extended to metals, So we should use 
fixed occupation in the scf calculation, right?


However, if I use fixed occupation in scf, the error is

' charge is wrong: smearing is needed  '. Because the system considered is odd 
electron.

What should I do to calculate the adsorption coefficiency of system with odd 
electrons.

Many thanks,

Clarence

_______________________________________________
Pw_forum mailing list
[email protected]
http://pwscf.org/mailman/listinfo/pw_forum

Reply via email to