[Pw_forum] PDOS

Gabriele Sclauzero sclauzer at sissa.it
Thu Dec 3 17:21:54 CET 2009


Dear Olga,

Olga Sedelnikova wrote:
> Dear PWscf Users,
>  
> I 've used projwfc.x post processing code to find the partial DOS.  The 
> second collunm of pdos_tot (dos(E)) agrees with the DOS 
> obtained by dos.x PP , the third collumn (pdos(E)) is the sum of all 
> ldos (from files like f.pdos_atm_4(C)_wfc_2(p)). I was confused to find 
> discreapency of this values.

Why do you expect that the sum of all PDOS should coincide with the DOS? They are not the 
same quantity. I expect them to be very different well above the Fermi energy.

> What is the reason of such problem?

It is not really a problem: it depends on the fact that PWs become a complete basis set in 
the solid as you increase the cutoff, while a set of plain atomic orbitals (as computed 
from an atomic calculation) aren't.

>  
> I want to separate the contribution of pi and sigma from the DOS of my 
> structure.
> What value of the total DOS should I use for comparison?

I think you should use the summed PDOS, since you need the individual PDOSs to separate 
the pi and sigma contributions. As a drawback you will find that there will be some charge 
missing (try to sum the PDOS of the atoms in an isolated molecule for instance: the sum of 
all PDOSs may not give a integer number of electrons).
Another method could be to separate KS eigenstates by symmetry (if this can distinguish pi 
from sigma: in an isolated molecule it could) and integrate the charges separately.
This however requires some coding or simply some more hassle if you can manage to do it by 
using postprocessing subroutines (bands.x to identify symmetry, pp.x to plot charges 
densities and the sum up those with the same symmetry,...), and it seems feasible only for 
simple geometries.

Regards


GS

>  
> Thank you,
> O.Sedelnikova
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-- 


o ------------------------------------------------ o
| Gabriele Sclauzero, PhD Student                  |
| c/o:   SISSA & CNR-INFM Democritos,              |
|        via Beirut 2-4, 34014 Trieste (Italy)     |
| email: sclauzer at sissa.it                         |
| phone: +39 040 3787 511                          |
| skype: gurlonotturno                             |
o ------------------------------------------------ o


More information about the Pw_forum mailing list