Page 1 of 1

equivalent positions and partial DOS

Posted: Fri Sep 11, 2009 3:45 pm
by panda
I have specified the primitive unit cell for my system, and in this cell there are three ions of type X, six ions of type Y, and six ions of type Z. The ions in the unit cell, when projected into space, form an extended solid. My question is, if these 15 ions are required to construct the extended solid what can I expect to see in the partial DOS? Will the partial DOS be slightly different for each ion, based upon the interactions of the other ions in the lattice, or are all of the ions "equivalent" and so will have exactly the same partial DOS? It is my understanding from crystallography that if the ions were "equivalent" then we would not need three ions of type X, but rather only 1 ion of type X, as all other positions could be reproduced from this equivalent position, which in this system is not the case.

Any input is appreciated, Thanks

equivalent positions and partial DOS

Posted: Mon Oct 12, 2009 12:53 pm
by admin
from symmetry reasons, the local DOS of equivalent atoms have to be the same. However, as vasp does no additional "post-production" symmetrization, the local DOS of equivalent atoms may differ slightly.
please either symmetrize it yourself, or simply take teh averaged DOS of the equivalent atoms.

equivalent positions and partial DOS

Posted: Mon Nov 09, 2009 5:12 am
by panda
Thank you for the response, thankfully I have since resolved this issue and the partially decomposed bands came out equivalent.