Differences in the Magnetic Moments of PAW_GGA and GGA

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


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
gferguson
Newbie
Newbie
Posts: 2
Joined: Tue Mar 09, 2010 9:35 pm

Differences in the Magnetic Moments of PAW_GGA and GGA

#1 Post by gferguson » Wed Apr 28, 2010 2:22 pm

While performing relaxations for small Co clusters I noticed something confusing. The GGA pseudopotential gives a magnetic moment of 10 while the PAW_GGA pseudopotential gives a magnetic moment of 6. All other parameters are identical. Does anyone know why the magnetic moment is so sensitive to the choice of pseudopotential?
Last edited by gferguson on Wed Apr 28, 2010 2:22 pm, edited 1 time in total.

forsdan
Sr. Member
Sr. Member
Posts: 339
Joined: Mon Apr 24, 2006 9:07 am
License Nr.: 173
Location: Gothenburg, Sweden

Differences in the Magnetic Moments of PAW_GGA and GGA

#2 Post by forsdan » Sun May 02, 2010 6:34 pm

How sure are you that the two calculations have converged to the groundstate? Depending on the initial magnetic moment you can end up in a local minima. Have you explored the configuration space by using NUPDOWN to fix the magnetization?

Also perform an investigation of an isolated atom to check the groundstate of the pseudopotentials. See the manual for more details

http://cms.mpi.univie.ac.at/vasp/vasp/node230.html
http://cms.mpi.univie.ac.at/vasp/vasp/node231.html

First with that information, it is possible to discuss the difference (if any).

Cheers,
/Dan
<span class='smallblacktext'>[ Edited Sun May 02 2010, 08:34PM ]</span>
Last edited by forsdan on Sun May 02, 2010 6:34 pm, edited 1 time in total.

Post Reply