Page 1 of 1
arbitrary specification of site for local pDOS.
Posted: Mon Sep 11, 2006 7:51 pm
by rwixom
Is it possible to specify a location and sphere (RWIGS) for calculations of site projected pDOS. More specifically I would like to look at the pDOS for a vacancy. Since there is no ion in this location, it is not included in the calculation of the DOS.
Any advice?
Thanks.
arbitrary specification of site for local pDOS.
Posted: Wed Sep 13, 2006 12:04 pm
by admin
yes, please include an empty sphere as atom in POSCAR AFTER
the position of the last atom : simply add
Empty (spheres)
1
0.0000000000000000 0.000000000000000 0.0000000000000000
(the cell has to be centered around the vacancy sphere)
and choose LORBIT=1 or 2 in INCAR
the pDOS at the vacancy is written in DOSCAR like the other partial DOSes
arbitrary specification of site for local pDOS.
Posted: Tue Oct 02, 2012 2:49 pm
by KurtG
good!
arbitrary specification of site for local pDOS.
Posted: Fri Oct 12, 2012 7:08 am
by sondon
That is great~! Thank you for the tip!
I just have tried the recipes, and some numbers anyways.
But, I was wondering:
1.
What is the exact meaning of lm-decomposition in the DOS of this empty site?
For example, if my empty site has some px and py component,
does this mean the wavefunctions around the empty site are just lm-decomposed,
or the state is composed of px or py components of the surroundings?
2. (the cell has to be centered around the vacancy sphere) <- I could not get the phrase clearly.
Does this mean I should translate the rest of the ionic positions accordingly
to set the vacancy position to be the unweighted center of mass of the cell
or set the vacancy to be always at (0,0,0)?
In any case, even though we can place multiple vacancies simultaneously,
should I place each vacancy at the center and translate all the other accordingly to calculate the corresponding LDOS?
Many thanks in advance for any comment.
<span class='smallblacktext'>[ Edited Fri Oct 12 2012, 07:16AM ]</span>