Splitted DOS

Message

kambiz · #1 Post by **kambiz** » Thu Mar 15, 2012 9:37 am

Hi,

I did a DOS calculation for a the adsorption of Oxygen on Co. The DOS profile of Oxygen shows the significant contribution of d-orbital to the total DOS (s+p+d).

I was just wondering what does it mean, as oxygen has no d orbital. Should I ignore the d orbital and consider s+d as the total?
I d really appreciate your help.

Cheers
Kambiz

Danny · #2 Post by **Danny** » Thu Mar 15, 2012 11:10 am

make sure that your wigner-seitz radius is not to big (i.e. try to have none overlapping spheres in your system) If you have overlapping spheres you might actually be seeing d-states of Co

kambiz · #3 Post by **kambiz** » Thu Mar 15, 2012 1:48 pm

Thanks Danny,
Yeah that s the reason. I should try their covalant radius.

Dr_Nick · #4 Post by **Dr_Nick** » Fri Mar 16, 2012 2:49 am

Be aware that the DOSCAR section in the manual is outdated...

The ordering in the DOSCAR (You will have 9 columns for a non spin-polarized calculation) is explained here:

http://cms.mpi.univie.ac.at/vasp-forum/ ... php?4.3477

So maybe you might have mistaken d states for p states of O.

kambiz · #5 Post by **kambiz** » Fri Mar 16, 2012 3:02 am

Thanks Dr_Nick

I knew that

I m using Vasp TST Tools (http://theory.cm.utexas.edu/vtsttools/). It can correctly read DOSCAR files.

kambiz · #6 Post by **kambiz** » Fri Mar 16, 2012 7:41 am

Hi Danny,

Is it possible that because of the large wigner-seitz radius, 3d-Co state affects also the 2p-O??

Cheers

Danny · #7 Post by **Danny** » Sat Mar 17, 2012 1:09 am

If they overlap in energy I think it could be possible that they get projected there too. The best way to find out is by
1) comparing the different attomic dos's
2) play with the WZ radius If they disappear (for a reasonable value), this shows they should not be there.

Danny