[Pw_forum] Not converging scf calculation

[李晓川]

your ecut may be too large!and your atomic position may be in wrong array.Besides,the Kpoints is a little smaller,(pbe and pw91)pseudopoteneial is usuarly used .

At 2013-04-28 16:59:47,"Ker Park" <kerpark at hotmail.com> wrote:

Hello all,


I am doing scf calculation for bulk MoS2, but it never reaches convergence.
I tried up to 300 Ry for 'ecut' though, the total energies were around 250 Ry and fluctuated randomly (energies were even positive). I am copying my 'scf.in' file below. Please generously provide any suggestions. The crystal structure is Hexagonal structure with a basis of 6 atoms (I think). I checked a lot of times, but my structure setup could be wrong. I am also copying the addresses where the crystal structure is shown. Please let me know if my structure could be simpler.


Many thanks,
Kerr


http://www.drilube.co.jp/english/product/molybdenum.html


http://nsfafresh.org/wiki/index.php?title=MoS2


http://www.machinerylubrication.com/Read/861/solid-film-lubricants




------------------------------------------------------------------------------------------------------ 
 !MoS2 scf.in
 &control
    calculation='scf'
    restart_mode='from_scratch',
    !pseudo_dir='directory where pseudopotentials are stored/',
    !outdir='directory where large files are written/'
    pseudo_dir='../../pseudo',
    outdir='./output100'
    prefix='PH',
 /
 &system    
    ibrav=4, celldm(1)=5.9715, celldm(3)=3.8905,
    nat=6, ntyp=2, ecutwfc =300
 /
 &electrons
    conv_thr =  1.0d-8
    mixing_beta = 0.7
 /
ATOMIC_SPECIES
 Mo  95.94    Mo.pw-mt_fhi.UPF
 S   32.065   S.pw-mt_fhi.UPF
ATOMIC_POSITIONS alat
 Mo 0      0      0
 S  2.9858 1.7238 2.9971 
 S  0      0      8.6190
 Mo 2.9858 1.7238 11.6161 
 S  0      0      14.6133
 S  1.5800 0.9122 20.2352
K_POINTS automatic
 4 4 4 0 0 0
------------------------------------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20130428/fc344134/attachment.html>