[QE-users] igk_k , davcio, evc and wfck2r.f90

[Aritz Leonardo]

Dear all:

I am having a little trouble understanding how to read and manipulate 
the wavefunction of a nscf calculation. (My final goal is to build 
matrix elements of type <k+q|exp^(+i(q+G)r)|k> )

I have followed the program wfck2r.f90 in which a calculation is loaded 
and the wavefunction is transformed into real space and saved to the 
disk, but I do not understand fully the the ordering of C_k(G) elements 
and maybe somebody could help me.

at some point of the code after loading the calculation the wavefunction 
is read by:

DO ik = first_k, last_k

      npw = ngk(ik)
      CALL davcio (evc, 2*nwordwfc, iunwfc, ik, - 1)

If I print on the screen for a check , the values of evc and evc(igk_k) 
corresponding to the first 4 planewaves, (1st band )

    WRITE(*,*) evc(1,1), evc(igk_k(1,ik),1)
    WRITE(*,*) evc(2,1), evc(igk_k(2,ik),1)
    WRITE(*,*) evc(3,1), evc(igk_k(3,ik),1)
    WRITE(*,*) evc(4,1), evc(igk_k(4,ik),1)

it turns out that the ordering that I get, does not match the order of 
the file K000"ik"/evc.dat file except for the gamma point which of 
course, always agrees.

In other words, after converting to xml using iotk the file 
K00002/evc.dat, although I can identify the values they are disordered 
with respect to evc or evc(igk_k). For instance ik=3

Screen ik=3

*            evc (ig,1)                    evc igk_k ((1,3),1)*
  (-0.259592093457288,-0.577862862809762) 
(-0.259592093457288,-0.577862862809762)
  (-0.623388775544561,0.112694178864960) 
(0.130227173344509,-7.741477644355699E-002)
  (0.130227182144171,-7.741477133293871E-002) 
(-0.623388775544561,0.112694178864960)
  (0.130227173344509,-7.741477644355699E-002) 
(9.798387850941684E-003,0.151182528291742)

*K00003/evc.xml*

<?xml version="1.0"?>
<?iotk version="1.2.0"?>
<?iotk file_version="1.0"?>
<?iotk binary="F"?>
<?iotk qe_syntax="F"?>
<WFC>
   <INFO ngw="631" igwx="464" gamma_only="F" nbnd="20" ik="3" nk="16" 
ispin="1" nspin="1" scale_factor="1.000000000000000E+000"/>
   <evc.1 type="complex" size="464">
-2.595920934572880E-001,-5.778628628097616E-001
  1.302271821441705E-001,-7.741477133293871E-002
-6.551644050809464E-002, 4.102286385954088E-003
-6.233887755445614E-001, 1.126941788649597E-001

Thank you very much in advance for any help!

Aritz

Univ. of the Basque country


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20190904/be917f77/attachment.html>