[Pw_forum] Details of partial density of states calculation

[Stefano de Gironcoli]

Dear Natalie, Dear Lorenzo

   the code projwfc calculate the projection of the crystal  
wavefunctions on the reference atomic wavefunctions of each atom in  
the cell (after these have bee orthogonalized). as such it strongly  
depends on which atomic wavefunctions are included in the reference  
computation in the atom.

   especially in the case these are extended orbitals (like the s-wave  
valence states of transition metals) the projections may not be very  
useful as a way to decompose the dos in atomic contributions.

   I think what you suggest, if i understand it correctly, can be a  
useful alternative:
   computing the amount of charge inside the sphere that can be  
assigned to the different angular momentum channels using the PAW  
reconstruction. the piece of information needed, in addition to the  
becp=<beta_a|psi> are the integrals of the corresponding atomic pseudo  
wavefunctions <phi_a|phi_a>_R or <phi_tilte_a|phi_tilde_a>_R+Q_aa  
where phi_a and phi_tilde_a are the AE and PS atomic wavefunctions  
respectively and Q_aa is the integral of the augmentation charge and  
the integrals are limited to the core region inside a radius R.
    something like that...

stefano


Quoting "Holzwarth, Natalie" <natalie at wfu.edu>:

> Dear Lorenzo,
>
>         Thanks for your answer.   I can understand that projwfc may not
> know about PAW, but  the code apparently calls calbec which seems to
> calculate the radial integral about an atom <beta|psi> where presumably
> beta is obtained from the pseudopotential file???    Perhaps this internal
> code notation is not consistent with the UPF file?   On the other hand, if
> the notation is consistent, then  the beta term in the UPF file  is the
> projector for PAW  (I think -- or at least that is true for ATOMPAW).
> This is the reason for my assumption that <p^a_i|\psi> is calculated to
> determine the partial density of states.   I am trying to write a paper
> with some pdos plots and wanted explain what the code does.    It seems
> like my suggestion to provide another option for the code would be a bit of
> work so perhaps can wait.  In any case, it is always good to know what has
> been calculated....      Thanks in advance for any further
> suggestions/discussion.    Natalie
>
> N. A. W. Holzwarth                                       email:
> natalie at wfu.edu
> Department of Physics                                  web:
> http://www.wfu.edu/~natalie
> Wake Forest University                                 phone:
> 1-336-758-5510
> Winston-Salem, NC 27109 USA                     office: Rm. 300 Olin
> Physical Lab
>
>
> On Mon, May 13, 2013 at 5:14 PM, Lorenzo Paulatto <
> lorenzo.paulatto at impmc.upmc.fr> wrote:
>
>> On 05/08/2013 06:43 PM, Holzwarth, Natalie wrote:
>>
>> > It slightly worries me that the pdos defined in this way is fairly
>> > sensitive to the PAW dataset used.    I am having a debate with myself
>> > about whether it might be good to append a piece of code I wrote long
>> > ago to your code to calculate <p^a_i|\psi> and calculate a better
>> > estimate of the charge inside the augmentation sphere. (Assuming it
>> > would be possible to figure out the details of that code.)   I welcome
>> > your advice and commentary.
>> >
>> Dear Natalie,
>> if I remember correctly, the PDOS code is not PAW-aware, i.e. it
>> projects on pseudo atomic wavefunctions that it can find in the
>> pseudpotential, completely ignoring the PAW projectors and stuff. It
>> should depends on the dataset only if the reference configuration is
>> changed, but it is true that nobody probably checked it properly.
>>
>> On the other hand, there is a code that reconstructs the PAW charge
>> (it's pawplot.f90 in the PP/src directory), I do not know if it has
>> anything in common with what you did. I'll have a look at it if you
>> wish, if it improves prowfc it is of course welcome.
>>
>> bests
>>
>>
>>
>> --
>> Dr. Lorenzo Paulatto
>> IdR @ IMPMC -- CNRS & Université Paris 6
>> phone:+33 (0)1 44275 084 / skype: paulatz
>> www:  http://www-int.impmc.upmc.fr/~paulatto/
>> mail: 23-24/4é16 Boîte courrier 115, 4 place Jussieu 75252 Paris Cédex 5
>>
>> _______________________________________________
>> Pw_forum mailing list
>> Pw_forum at pwscf.org
>> http://pwscf.org/mailman/listinfo/pw_forum
>>