negative fermi energy- surface calculations

Message

kamala · #1 Post by **kamala** » Fri May 21, 2010 7:45 am

Hello,
I am trying surface calculations for semiconductors. The slab is 11 layer thick with equal vacuum length. I have applied dipole correction along z.

When I calculate DOS (tetrahedron method, prec=accurate), I get negative values for fermi energy. I did see some discussion on this. But I am unable to get rid of this problem. I am using GGA_PBE

Can somebody tell me how to go about this problem
Thanks

forsdan · #2 Post by **forsdan** » Sun May 23, 2010 4:12 pm

Can you explain why you think this is a problem? Since there in general is no absolute reference for the energy in the calculation, a negative or positive value for the fermi-energy doesn't say anything at all...

If you want to you can in your case align your energy levels with respect to the vacuum level by evaluating the planar average electrostatic potential, if you now want an absolute reference.

Cheers,
/Dan

<span class='smallblacktext'>[ Edited Mon May 24 2010, 12:30PM ]</span>

kamala · #3 Post by **kamala** » Mon May 24, 2010 6:27 am

Dan,
OK,
I am trying to get band offsets. So I need a reference

Related question as you have mentioned the average ES pot....

I have calculated the LOCPOT. I am trying to understand the format of this file. How do I plot E vs Z? that is the planar average Electrostatic potential?

Please help
Thanks
Kamala

forsdan · #4 Post by **forsdan** » Mon May 24, 2010 10:29 am

The format of the LOCPOT file is similar to the format of the CHGCAR file, so you can understand it from the manual entry

http://cms.mpi.univie.ac.at/vasp/vasp/CHGCAR_file.html

There are also a numerous threads that discuss the format of the CHGCAR file.

You can then write a script that evaluates the planar average electrostatic potential at each z-coordinate from the data structure. This is quite straightforward once you have understood the structure.

/Dan

kamala · #5 Post by **kamala** » Wed May 26, 2010 8:46 am

Thanks much Dan,
One more question.. where do I get the information of grid along z direction in real space?
Kamala

forsdan · #6 Post by **forsdan** » Wed May 26, 2010 5:22 pm

The size of the real space grids are given by NGXF, NGYF, NGZF. You can find them in the OUTCAR file. Their values are also written in the headers of the LOCPOT, CHGCAR files before the data points.

/Dan
<span class='smallblacktext'>[ Edited Wed May 26 2010, 07:23PM ]</span>