[SOLVED] Adding the l-decomposed DOS
Posted: Fri Sep 30, 2011 8:11 am
Dear all,
I need help in trying to understand the DOS.
I'm preparing the exercises for the students and got to the part where we want to see the atomic contributions to the DOS. I used prtdos 3, and got the l-decomposed DOS for both atoms (there a 2 atoms in the unit cell). Then I thought of adding the local DOSs for l=0,...,4 in order to get the total DOS belonging to the first atom. And same for the second. But when I added these two, I saw numerically that they do not add to the total DOS obtained with prtdos 2. For one energy point, for example, they differ for about 40 units in the DOS.
Now I would like to know the reason(s) for this. Is it a numerical error (tetrahedron method)? Or because of the badly chosen ratsph (I've checked that the atomic spheres do not overlap)? Am I thinking it all wrong?
Thanks for all the answers!
Yours,
Igor Lukacevic
I need help in trying to understand the DOS.
I'm preparing the exercises for the students and got to the part where we want to see the atomic contributions to the DOS. I used prtdos 3, and got the l-decomposed DOS for both atoms (there a 2 atoms in the unit cell). Then I thought of adding the local DOSs for l=0,...,4 in order to get the total DOS belonging to the first atom. And same for the second. But when I added these two, I saw numerically that they do not add to the total DOS obtained with prtdos 2. For one energy point, for example, they differ for about 40 units in the DOS.
Now I would like to know the reason(s) for this. Is it a numerical error (tetrahedron method)? Or because of the badly chosen ratsph (I've checked that the atomic spheres do not overlap)? Am I thinking it all wrong?
Thanks for all the answers!
Yours,
Igor Lukacevic