Relation between partial DOS and total DOS

Message

icymist · #1 Post by **icymist** » Thu Feb 12, 2009 4:15 pm

Hi all,

I would like to know what is the relation between the partial DOS and the total DOS that we have in the DOSCAR.

I am assuming that total DOS (approx. =) sum of partial DOS of all atoms? Am I wrong in assuming this?

I have done a spin polarized calculation of an Al atom in a primitive cell in the fcc lattice. I have set LORBIT = 11. I computed the sum of all the partial DOS (sup, sdown, pxup, pxdown, etc.) of the Al atom (from the second section of DOSCAR) and plotted it along with the toal DOS (first section of DOSCAR). The plot is at http://picasaweb.google.com/icymist/Wor ... 0577306242

I would like to know why the sum of the partial DOS does not match with that of the partial DOS. Can someone help? BTW, if I multiply all the partial DOS with 4, they more or less match up with the total DOS.

job · #2 Post by **job** » Sat Feb 14, 2009 8:14 am

Partial DOS is calculated in spheres around each atom with radius specified by RWIGS. You cannot fill space with non-overlapping spheres. This is all explained in the manual.