Page 1 of 2

The output of GW

Posted: Mon May 24, 2010 1:38 pm
by sdwang
Dear all:
What are the "Eo E-Eo Sc(Eo)" in the output file of GW?

Thanks!
SDWang

Re: The output of GW

Posted: Mon May 24, 2010 4:41 pm
by myrta gruning
sdwang wrote:Dear all:
What are the "Eo E-Eo Sc(Eo)" in the output file of GW?

Thanks!
SDWang
Hallo Shudong Wang

Eo bare energy (coming out from the DFT calculation you input to yambo)
E-Eo difference between quasiparticle energy E and the bare energy Eo, so that is the QP correction
Sc(Eo) Value of the \Sigma_c at Eo

Regards
m

Re: The output of GW

Posted: Mon May 24, 2010 4:52 pm
by sdwang
Thank you for your reply.
That is to say Yambo can calcualte the bare band structurres of DFT (E0)according to scf and nscf of pwscf??( my interface is pwscf)

Re: The output of GW

Posted: Mon May 24, 2010 5:03 pm
by Daniele Varsano
Yes, the DFT band structure is reported, you can easily check it.

Cheers,

Daniele

Re: The output of GW

Posted: Mon May 24, 2010 5:30 pm
by myrta gruning
Hallo
Just a remark.
Yambo is not calculating the DFT band structure, is reporting the DFT energies calculated in the DFT run by (in your case) pwscf in the k mesh you have chosen and that you provide as an input.
As Daniele said, in the report file you find all DFT eigenenergies at each k-point of the mesh. These are just an input of Yambo, not an output.

Regards,
m

Re: The output of GW

Posted: Tue May 25, 2010 3:11 am
by sdwang
myrta gruning wrote:Hallo
Just a remark.
Yambo is not calculating the DFT band structure, is reporting the DFT energies calculated in the DFT run by (in your case) pwscf in the k mesh you have chosen and that you provide as an input.
As Daniele said, in the report file you find all DFT eigenenergies at each k-point of the mesh. These are just an input of Yambo, not an output.

Regards,
m
Thank you!
In the output of GW,there are:
[02.05] Energies [ev] & Occupations
===================================

Fermi Energy[ev] - T[ev/K] : 5.951745 0.000000 0.000000
States summary : Full Metallic Empty
0001-0004 0005-0050
Indirect Gaps [ev]: 1.117994 3.154146
Direct Gaps [ev]: 2.794973 5.387020


For " Indirect Gaps [ev]: 1.117994 3.154146", the 1.117994 is the data of DFT ,and the 3.154146 is the GW data, is it right??

S D Wang

Re: The output of GW

Posted: Tue May 25, 2010 9:14 am
by myrta gruning
[02.05] Energies [ev] & Occupations
===================================

Fermi Energy[ev] - T[ev/K] : 5.951745 0.000000 0.000000
States summary : Full Metallic Empty
0001-0004 0005-0050
Indirect Gaps [ev]: 1.117994 3.154146
Direct Gaps [ev]: 2.794973 5.387020

For " Indirect Gaps [ev]: 1.117994 3.154146", the 1.117994 is the data of DFT ,and the 3.154146 is the GW data, is it right??
Hallo,

No, both are DFT gaps (relative to the k-grid you have in you calculation).

Note that the report file is organized in sections and subsections.
What you reported is section 2.05
all section 2 is about:

Code: Select all

[02] Input variables setup
The report on the QP part is under

Code: Select all

 [06] Dyson equation: Newton solver
in particular under

Code: Select all

[06.02] Q(uasi)P(article) properties
There you find a part similar to that:

Code: Select all

  QP [eV] @ K [1] (iku): 0.000     0.000     0.000   
   B=19 Eo= -1.56 E= -3.74 E-Eo= -2.18 Z=0.86 So=  2.07418 xx=-18.57731 Vxc=-13.96801
   B=20 Eo= -1.56 E= -3.76 E-Eo= -2.19 Z=0.86 So=  2.06385 xx=-18.57655 Vxc=-13.96766
   B=21 Eo= -1.22 E= -3.31 E-Eo= -2.08 Z=0.86 So=  2.09297 xx=-18.70781 Vxc=-14.19493
   B=22 Eo= -1.22 E= -3.07 E-Eo= -1.85 Z=0.85 So=  1.96251 xx=-17.81735 Vxc=-13.67904
   B=23 Eo=  0.00 E= -1.18 E-Eo= -1.18 Z=0.86 So= 1.177    xx=-16.05    Vxc=-13.50
   B=24 Eo=  0.00 E= -1.18 E-Eo= -1.18 Z=0.86 So=  1.17296 xx=-16.04541 Vxc=-13.50312
   B=25 Eo=  4.39 E=  7.96 E-Eo=  3.57 Z=0.88 So= -0.63854 xx= -7.97938 Vxc=-12.69211
   B=26 Eo=  4.39 E=  7.96 E-Eo=  3.57 Z=0.88 So= -0.63949 xx= -7.97820 Vxc=-12.69144
   B=27 Eo=  5.02 E=  8.66 E-Eo=  3.64 Z=0.89 So= -1.14389 xx= -7.11480 Vxc=-12.33968
where Z is the renormalization factor, xx the Sigma_x matrix element, Vxc the Vxc matrix element and So the Sigma_c matrix element for band B @K[#]

Note as well that the r_* are report files. The output of the calculation (when is human readable) is reported in the o.* files. In this case o.qp.

Regards,
m

Re: The output of GW

Posted: Tue May 25, 2010 9:38 am
by sdwang
"No, both are DFT gaps (relative to the k-grid you have in you calculation)."
What does this mean?? Why there are two different indirect band gap? I am confused. I hope you can explain it in detail.
Thanks!

Re: The output of GW

Posted: Tue May 25, 2010 9:49 am
by myrta gruning
Hallo,

Maybe you are confused by the word gap, thinking just about a minimum gap.
Think about Gap as energy difference between a valence and conduction band at a certain k. Then Yambo lists the two lowest gap, both direct and indirect.

Looking in the list of DFT energies at the various k point you have below the

Code: Select all

ndirect Gaps [ev]: 1.117994 3.154146
Direct Gaps [ev]: 2.794973 5.387020
you should be able to find out to which transition these gaps corresponds.

m

Re: The output of GW

Posted: Tue May 25, 2010 10:30 am
by sdwang
myrta gruning wrote:Hallo,

Maybe you are confused by the word gap, thinking just about a minimum gap.
Think about Gap as energy difference between a valence and conduction band at a certain k. Then Yambo lists the two lowest gap, both direct and indirect.

Looking in the list of DFT energies at the various k point you have below the

Code: Select all

ndirect Gaps [ev]: 1.117994 3.154146
Direct Gaps [ev]: 2.794973 5.387020
you should be able to find out to which transition these gaps corresponds.

m
Take the direct gap as a example, is the 2.794973 eV denotes the samllest direct gap ,and the 5.387020 denoyes the most wide direct gap?