[Pw_forum] Projection of dos on to molecular orbitals

Guido Fratesi fratesi at mater.unimib.it
Tue Jun 16 12:18:07 CEST 2015

Dear Vasudevan,

the tool is aimed at what you are willing to do with:
"Full system" the molecule adsorbed on a surface.
"Part" the molecule in gas phase.

Statement 2) below refers to the atomic orbitals, not to the 
eigenstates. This tool passes through an atomic-basis-set 
representation, and the two basis sets should correspond (they describe 
the same Hilbert space) so that you can make a dot product of two 
vectors describing 1 the eigenstate of the full system and 2 the 
eigenstate of the part.


On 15/06/2015 12:26, vasudevan m.v wrote:
> Dear QE users,
> I want to project the density of states of complete system (molecule
> adsorbed on a surface) on to molecular orbitals of adsorbate and find
> population of different molecular orbitals. If I understood correctly
> there is an post-processing tool molecularpdos.x in quantum ESPRESSO v
> 5.1.2 which does this job. If I am mistaken please correct me.
> (http://www.quantum-espresso.org/wp-content/uploads/Doc/pp_user_guide/node8.html)
> Statement 1)
>  From ..../espresso-5.1.2/PP/Doc/INPUT_molecularpdos.html
>   " /Then the eigenvectors of the full system are projected onto the
> ones of the part. For example, to decompose the PDOS of an adsorbed
> molecule into its molecular orbital, as determined by a gas-phase
> calculation/"
> I don't understand what is meant by decomposition of pdos while in first
> part it says eigenvectors of full system are used.
> Statement 2)
> "/The atomic wavefunctions identified by the ranges
>   i_atmwfc_beg_full:i_atmwfc_end_full (full system) and
>   i_atmwfc_beg_part:i_atmwfc_end_part (molecular part)  should
> correspond to the same atomic states. See the  header of the output of
> projwfc.x for more information/."
> I feel the above statement slightly confusing as in statement 1 it is
> said that "/Then the eigenvectors of the full system are projected onto
> the ones of the part/"
> Hope that things are conveyed well.
> With regards
> Vasudevan M V
> Bangalore

Guido Fratesi

Dipartimento di Fisica
Universita` degli Studi di Milano
Via Celoria 16, 20133 Milano, Italy

More information about the users mailing list