ABINIT Discussion Forums

Dear abinit users,
I tried to calculate the band structure and DOS of graphene bilayer but I could not even have the non-degenerate bands near the Fermi level.
I would be very grateful if you can show me what are the parameters who I can modify to draw the curve of good band structure of graphene bilayer.
some parameters used in my input file,

ngkpt 16 16 16
kptbounds
0.000 0.000 0.000
0.000 0.500 0.000
-0.333 0.667 0.000
0.000 0.000 0.000
nband 8
natom 4
typat 4*1
xangst
0 0 0
0.71 -1.23 0
0 0 3.4
-0.71 -1.23 3.4

acell 4.6 4.6 30
angdeg 90 90 120

What is your ecut?

If you're using PAW, what is pawecutdg?

Also, the best way to do a band structure is first to calculate the density and then calculate the band structure in another dataset. Have a look at tutorial 3.5 for an example.

Regards,

Simon

Dear Simon,
this is my complet input file:
ndtset 2
chkprim 0
kptopt1 1
ngkpt1 4 4 4
prtden1 1
toldfe1 1.0d-11
iscf2 -2
getden2 1
kptopt2 -3
nband2 8
ndivk2 30 30 30
kptbounds2
0.000 0.000 0.000
0.000 0.500 0.000
-0.333 0.667 0.000
0.000 0.000 0.000

tolwfr2 1.0d-18
enunit2 1
prteig2 1

ixc 1

nstep 1000
occopt 7

tsmear 0.01 eV
iprcel 45

acell 4.6 4.6 30
angdeg 90 90 120

ntypat 1
znucl 6

natom 4
typat 4*1
xangst
0 0 0
0.71 -1.23 0
0 0 3.4
-0.71 -1.23 3.4

ecut 30

Regards.

Your k-grid looks a bit low for the density calculation. Other than that, is Abinit finding the correct symmetry?

Instead of using xangst, you could use
xred 0 0 0
1/3 1/3 0
0 0 0.21
2/3 1/3 0.21 (not sure about this one)

FInally, it might be an effect of the relaxation. Have you relaxed the atomic structure?

Simon

Hello,
I tried with reduced coordinates
xred
0 0 0
1/3 1/3 0
0 0 0.21
2/3 1/3 0.21
and ngkpt 16 16 4 but still no good result for the band structure.
 I would be grateful if you could help me to get the good coordinates to plot the band structure of graphene bilayer.
Best regards

Maybe increasing ecut could help? Is it tightly converged?

I think the problem lies with nband. I assume you have 4 electrons per atom (it's carbon) so you need 2 bands per atom therefore 8. However, you should always include some conduction band. Maybe increasing nband to 12 could solve your problem.

hi ,
i think you can use Abinit tutorial for the band structure of Si
& then use xred instead of xangst
the xred for one layer graphene is
xred
1.3333333333333335 1.6666666666666665 0.5000000000000000
-1.3333333333333335 -1.6666666666666663 0.5000000000000000
& if you want to use spgroup191 I think .
xred
0 1/3 0.5
at the end i want to say that xred is really better than xangst & i advice U to use xred for your calculation.

sincerely

hamideh.

This discussion is in the wrong forum - please do not post randomly!