[Q-e-developers] About charge density

[Jiqiang Li]

Dear developers,
      I have puzzle in following aspects,
      (1)The subroutine symrho is marked as "symmetrize rho(G)", what does the "symmetrize charge density in G space' mean? and what's its consequence?

      (2)For charge density A and B: If C(r)=A(r)+B(r), i.e., we sum the two terms in real space, e.g. A(r)= \sum_i(occupation) {<i|i>}, B(r) = <m|n>. For Gamma point only calculation, <m|n> is still a real array, as the ordinary charge density is.
           2a) then can we obatain that C(G) = A(G) + B(G)?  and how about the oppsite process? where A(G) <-FFT-> A(r), B(G) <- FFT -> B(r),  C(G) <- FFT-> C(r), respectively. 
           2b) If not, does the symmetrizing process affect it?  
      
      (3) Actually I want to realize these process:  First I run a whole ordinary SCF calculation and at the end output the  wavefunction on the disk. Then I run another new SCF calculation, but in this new calucation  a new charge density, rho_new = rho + <m|n> = \sum_i(occupation) { <i|i>} + <m|n>, is treated as input charge density in SCF loop . The term <m|n> is fixed as a constant array  read from disk in every SCF loop. So Where should I add the code to realize rho_new=rho + <m|n> ? In subroutine sum_band_gamma  or other place ? Any advice would be gratefully appreciated!
 
Best regards!
 
Jiqiang Li
 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/developers/attachments/20150611/fc27e306/attachment.html>