Problem with the bands calculation

Queries about input and output files, running specific calculations, etc.


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
PSorokin
Newbie
Newbie
Posts: 3
Joined: Thu Sep 07, 2006 6:48 am

Problem with the bands calculation

#1 Post by PSorokin » Fri Apr 24, 2009 9:44 am

Hello!

I believe that this is just a technical question. I tried to calculate the electronic structure of graphane (graphene with adsorbed hydrogen).
I relaxed a structure (got good values of atomic geometry)
I calculated the DOS (got the good value of band gap: 3.4 eV)
I calculated the bands and...got the band gap much bigger than from the DOS. Why? Moreover I increased the density of k-points from the 16x16x1 to 24x24x1 and the band gap from the bands increased from the 4.83 eV to 6.6 eV. The band gap from the DOS remained constant.

I did the relaxation of the structure with the following INCAR:

PREC = Accurate
NSW = 99
IBRION = 1
ISIF = 4
POTIM = 0.5
ISMEAR=-5
ALGO = VeryFast
LREAL = False

with the following KPOINTS
16x16x16
0
GMonkhorst
16 16 1
0 0 0

and with the following POSCAR

Graphane single layer
1
2.514699 0.000000 0.000000
1.257350 2.177793 0.000000
0.000000 0.000000 25.076865
2 2
Cartezian
-0.001155 -0.000667 0.227902
1.256669 0.725539 -0.227964
-0.000910 -0.000526 1.356322
1.256845 0.725640 -1.356383

(first atom type is the carbon, second one is the hydrogen. XC=91)

I calculated the DOS with the same KPOINTS and the following INCAR:
PREC = Accurate
ISTART = 1
ICHARGE = 11
NSW = 0
IBRION = -1
ISIF = 4
POTIM = 0.5

SIGMA = 0.5
ISMEAR=-5
EMIN = -13.917
EMAX = 6.15

ALGO = VeryFast
LREAL = False

The bands were calculated with the same INCAR (I changed only ISMEAR value from -5 to 0) and with the following KPOINTS
Bands
100
Line-mode
Direct
0.5 0.5 0.0 ! M
0.0 0.0 0.0 ! G

0.0 0.0 0.0 ! G
0.666667 0.33333 0.0 ! K

0.666667 0.33333 0.0 ! K
0.5 0.5 0.0 ! M


Thanks!

Sincerely,
Pavel
Last edited by PSorokin on Fri Apr 24, 2009 9:44 am, edited 1 time in total.

PSorokin
Newbie
Newbie
Posts: 3
Joined: Thu Sep 07, 2006 6:48 am

Problem with the bands calculation

#2 Post by PSorokin » Mon Apr 27, 2009 9:55 am

I've found the origin of my problem. I've calculated the band structure from the CHGCAR obtained during the calculation of the DOS. The number of k-points between two runs is different and therefore I've got the wrong Fermi level and the bands.
<span class='smallblacktext'>[ Edited Mon Apr 27 2009, 12:01PM ]</span>
Last edited by PSorokin on Mon Apr 27, 2009 9:55 am, edited 1 time in total.

Post Reply