abinit_kss

[WEI Wei]

Dear All,
I use ABINIT to generate the *.KSS file for graphene, and then "a2y" to get the SAVE file. The problem is that I read a band gap in the "r_setup" file
--------------------------------------------------
Indirect Gaps [ev]: 0.492424 2.842075
Direct Gaps [ev]: 0.492424 4.991697
--------------------------------------------------

The input file for *.KSS file generation is :

-----------------------------------------------------------------------------------------------------------------------------------------
ndtset 2

#Dataset 1 : usual self-consistent calculation
kptopt1 1
nshiftk1 1
shiftk1 0.0 0.0 0.0
ngkpt1 14 14 1
prtden1 1
toldfe1 1.0d-12
nband1 10
occopt1 1
iscf1 7
#istwfk1 *1
#Dataset 2 : the band structure
iscf2 -2
getden2 1
kptopt2 1
ngkpt2 14 14 1
nband2 350
nbandkss2 350
kssform2 3
tolwfr2 1.0d-14
enunit2 1
symmorphi2 0
istwfk2 *1
nshiftk2 1
shiftk2 0.0 0.0 0.0
#Definition of the unit cell
acell 5.3672 5.3672 33.4333
rprim -0.866 -0.5 0.0
0.866 -0.5 0.0
0.0 0.0 1.0

#Definition of the atom types
ntypat 1
znucl 14

#Definition of the atoms
natom 2
typat 1 1
xred 0.6666666666666666 0.3333333333333334 -0.4871936633664358
0.3333333333333334 0.6666666666666667 0.4871936633664359
#Definition of the planewave basis set
ecut 30.0

#Definition of the SCF procedure
nstep 1000
diemac 12
#chksymbreak 0
-----------------------------------------------------------------------------------------------------------------------------------------

So, could you please tell me from where the problem originate? Thanks a lot!

[myrta gruning]

Dear Wei

What is the problem is not really clear to me. Can you be more explicit?
Is there any discrepancy with abinit? Yambo read all the info's about the dft calculations from abinit.
Note that the fact that yambo reports a gap does not imply that this is the gap @K. It is the smaller gap it finds a among the kpts you have in your grid. Further in the report you can also check kpts by kpts the KS energies. Those are, apart from shifting the Fermi enervy, the ones read from abinit.

M

[WEI Wei]

Dear Myrta,

Thank you for your swift reply!

I mean there is a gap at the Dirac (K) point reading from the "o.qp", which is the same as that in "r_setup". But, the band structure from ABINIT is correct.

Thanks.

[myrta gruning]

Dear Wei

you should check the abinit output related to the KSS file. Do you get there a gap @ K? Then you need to review that calculation.
Anyway I noticed that you have silicon and not carbon

#Definition of the atom types
ntypat 1
znucl 14

if this is really your input, please check, since maybe you are not calculating what you think.

m

[WEI Wei]

Dear Myrta,

That is fine, I used "znucl 6" in my calculation. I posted another input file because I can not login our cluster yestoday.
I also checked the results from ESPRESSO, and I read from the file "o.qp"
--------------------------------------------------------------------
59.00000 4.00000 0.00000 -0.28635 -0.28804
59.00000 5.00000 1.21766 0.21483 -2.49709
--------------------------------------------------------------------

from the file "r_setup"

--------------------------------------------------------------------
Indirect Gaps [ev]: 1.217657 8.284184
Direct Gaps [ev]: 1.21766 13.82861
--------------------------------------------------------------------

I checked vervy k point in the file "o.qp", and the value at k point of 59 seems the K point.

The input file for GW calculation useing YAMBO:
----------------------------------------------------------------------------------------------------------------------------------
gw0 # [R GW] GoWo Quasiparticle energy levels
ppa # [R Xp] Plasmon Pole Approximation
HF_and_locXC # [R XX] Hartree-Fock Self-energy and Vxc
em1d # [R Xd] Dynamical Inverse Dielectric Matrix
EXXRLvcs= 4543 RL # [XX] Exchange RL components
% QpntsRXp
1 | 73 | # [Xp] Transferred momenta
%
% BndsRnXp
1 | 300 | # [Xp] Polarization function bands
%
NGsBlkXp= 450 RL # [Xp] Response block size
% LongDrXp
1.000000 | 0.000000 | 0.000000 | # [Xp] [cc] Electric Field
%
PPAPntXp= 27.21138 eV # [Xp] PPA imaginary energy
% GbndRnge
1 | 300 | # [GW] G[W] bands range
%
GDamping= 0.100000 eV # [GW] G[W] damping
dScStep= 0.100000 eV # [GW] Energy step to evalute Z factors
DysSolver= "n" # [GW] Dyson Equation solver (`n`,`s`,`g`)
%QPkrange # [GW] QP generalized Kpoint/Band indices
1| 73| 1|10|
%
%QPerange # [GW] QP generalized Kpoint/Energy indices
1| 73| 0.0|-1.0|
%
----------------------------------------------------------------------------------------------------------------------------------

Thanks a lot!

[Daniele Varsano]

Dear Wei Wei,
take in mind that what you read from the o.qp is the GW result, while in the report
file, it is reported the gap extracted form the KS eigenvalues, read from the previous
DFT calculation (abinit or QE). Now, what it is strange is that the abinit result and what
you find in r_setup is different. Can you post the abinit output file and the r_setup file?

Cheers,

Daniele

[WEI Wei]

Dear Daniele,

Thank you very much for your respones.

The data for graphene within ABINIT is now not available. However, I post the results from ESPRESSO, which has the same problem with the results from ABINIT.

Attached please find the DFT input files ("gra_scf.in" and "gra_nscf.in"), output ("gra_nscf.out" and "gra_scf.out"), "r_setup" files, "yambo.in" and "o.qp"

Thanks again.

[myrta gruning]

Dear Wei

there is no inconsistency between what yambo report and the output of pwscf you provided
for the kpt 59

k = 0.2347-0.1524 0.0000 band energies (ev):
-19.4927 -18.5596 -17.0233 -5.9872 -4.7695 6.7579 7.6553 7.9295

you see there is a gap, the same reported by yambo. The only difference is a shift in the KS values, since Yambo put the top of the valence at 0 eV.

Let me stress this again, Yambo does not perform any DFT, but relies on the KS energies and wavefunctions calculated by either Abinit and PWscf. Except (not very probable) errors in the interface, what it is in Yambo about DFT is what you got from QE or ABINIT

Cheers,
m

[WEI Wei]

Dear Myrta,

Thank you for your reply. I definitely understand what do you mean.
Anyway, I can get the correct band structure. It seems that I missed something in the input file for generating *.KSS file. I will check them all.

Thanks again.