[Pw_forum] symmetrized projection in atomic projected DOS

[Paolo Giannozzi]

On Feb 28, 2008, at 5:01 , alan chen wrote:

> Dear PWSCF users,
>      I have a question about the flag 'lsym' in the atomic  
> projected DOS (projwfc.f90). The flag 'lsym' is set .true. by  
> default. I ran it both with 'lsym' = .true. or .false. and found  
> that if the Block state is nondegenerate, the decomposition into  
> atomic orbitals is the same no matter whether 'lsym'=.true.  
> or .false. However for those degenerate states, the coefficients  
> are different. For example, if there are three degenerate Block  
> states and 'lsym'=.true., the coefficients in the decomposition  
> (i.e. modulous square of the overlap) are the same for these three  
> states. If 'lsym' = false, the coefficients are different in the  
> three states.
>      So why are the results different if 'lsym' = true and false?  
> And which one is correct?
>      What is symmetrized projection?

hint:
- your third question should be the first one (and the only one)
- your second question doesn't make any sense: both are correct
(hopefully)
- your first question doesn't make any sense either: they differ because
they are (slightly) different quantities
- the answer is contained in your observations

Disclaimer: I didn't write that part of the code

Paolo
---
Paolo Giannozzi, Dept of Physics, University of Udine
via delle Scienze 208, 33100 Udine, Italy
Phone +39-0432-558216, fax +39-0432-558222