[CPMD-list] Kpoints Help!

[Yogesh Tiwary]

Hi All,
  
  I am a beginner in DFT calculations and VASP. I am calculating total  energy of a supercell containing (110) surface with 6x5  (in-channelxcross-channel) surface atoms and 6 atomic layers with top 3  relaxed. I am using a 7 atomic layer vacuum spacing and replicating the  cell in x and y directions. I am using the following KPOINTS setting:
  
  K-POINTS
  0
  Monskhorst-pack
  4 4 1
  0 0 0
  
  I believe that i should get the exact same value for energy if i use  3x10 surface atoms with 6 atomic layers and change the KPOINTS to (8 2  1) to get the same kpoints density in the brillouin zone and avoid  kpoints error. But, i am getting a difference in energy of about 0.01  eV which is much more than what i want. Please let me know if anyone  can tell me where i am going wrong.
  I am using the following INCAR file:
  
  PREC=accurate
  LREAL=.TRUE.
  ISTART=1
   electronic relaxation
  ISMEAR=-1
  SIGMA=0.2
  GGA=91
  
   ionic relaxation
  EDIFFG=-0.01
  IBRION=2
  NSW=10
  
  
  Thanks,
  Yogesh

 		
---------------------------------
Talk is cheap. Use Yahoo! Messenger to make PC-to-Phone calls.  Great rates starting at 1¢/min.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://cpmd.org/pipermail/cpmd-list/attachments/20060627/0fd4ba58/attachment.html