From hands-on session: 2_8_fccNi_dos: don't understand DOSCAR file

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


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
lahaye
Jr. Member
Jr. Member
Posts: 98
Joined: Fri Apr 14, 2006 5:08 am
Location: Suwon - Korea

From hands-on session: 2_8_fccNi_dos: don't understand DOSCAR file

#1 Post by lahaye » Thu Jun 01, 2006 7:27 am

Hello,

From the hands-on session (provided on the VASP homepage), there
an example in Handson Session 2, directory "2_8_fccNi_dos".

It calculates the DOS of a Ni bulk, with this INCAR file:

Code: Select all

  SYSTEM = fcc Ni
  ISTART = 0 ; ICHARG=2
  ENCUT  =    270
  ISMEAR =   -5 
  LORBIT=11
  ISPIN=2
  MAGMOM = 1
The DOSCAR is explained at
http://cms.mpi.univie.ac.at/vasp/vasp/node71.html

The DOSCAR file has roughly three sections:
1. Header (I understand that)
2. DOS & Intergrated DOS, with 5 columns (I understand that too).
3. The so-called "lm- and site-projected DOS", with 19 columns (I don't understand this one).

Can somebody explain what these 19 columns per line represent?
I tried several possibilities, but neither counted up to 19 columns....

If I remove the ISPIN and MAGMOM line in the INCAR file, then
that corresponding part 3 of the DOSCAR file has 10 columns,
yet another misterious number of columns.

I would appreciate very much if someone could explain to me what
these columns in the DOSCAR file are about.

I used the Ni-fcc handson example, as it might be a well known example
among the vasp users. I hope through this example, to understand
better the general layout of the DOSCAR file.


Thanks so much,
Rob.
Last edited by lahaye on Thu Jun 01, 2006 7:27 am, 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

From hands-on session: 2_8_fccNi_dos: don't understand DOSCAR file

#2 Post by forsdan » Thu Jun 01, 2006 1:20 pm

As far I understand the 10 columns are

E followed by lm-DOS for each l going from -m,...,0,...,+m with the lm-ordering: s, p_y, p_z, p_x, d_xy, d_yz, d_z2-r2, d_xz, d_x2-y2. If spin-polarized calculations are performed (ISPIN = 2), the spin-down component of each respective partial DOS will be printed after the spin-up component, so the ordering will be: s(up), s(down), p_y(up), p_y(down) etc. and therefore you will have 19 columns. Please correct me if I'm wrong.

Best regards
Dan Fors
Last edited by forsdan on Thu Jun 01, 2006 1:20 pm, edited 1 time in total.

Post Reply