The ABINIT Users Mailing List ( CLOSED )

[ilija.dukovski@mearscorp.com]

I'm sorry to repeat a subject from a week ago,
but I'm interested in obtaining gkk-s not only for the
Fermi surface. I plan to write a code to do that 
so I need some help regarding the format of the 
gkk files. 


First question is how are the matrix elements in the gkk
file (telphon_3o_GKK.bin in the example) ordered. 

is the inner loop n or nprime if the matrix element is

<k+q,nprime|H1|k,n>

is it:
do n=1,nband
   do nprime=1,nband
      
or 
do nprime=1,nband
   do n=1,nband ?

This order is important if one is not interested in integrated 
quantities.

Also anaddb with input 

elphflag 1

produces the file "gkkp_file_ascii" with contents such as the
following:

 qp=   0.000000000000000E&#43000;  0.000000000000000E&#43000;
0.000000000000000E&#43000;
  0.000000000000000E&#43000;  0.000000000000000E&#43000;  
0.000000000000000E&#43000;
   729.256434564191        729.256698690758        729.256724057270
 kp=   0.000000000000000E&#43000;  0.000000000000000E&#43000;
0.000000000000000E&#43000;
           2           2           1  0.000000000000000E&#43000;
           2           2           2  1.250711611564023E-034
           2           2           3  2.612430267913443E-034
           2           2           4  2.312964634635740E-018
           2           2           5  2.312964634635741E-018
           2           2           6  2.312964634635742E-018


I assume the line: 
2           2           1  0.000000000000000E&#43000;
means

n=2, nprime=2, j=1 and |g_(kp_nprime_n)^(qp,j)|^2=0.000000000000000E&#43000;

where n and nprime are electronic bands and j is the phonon branch,

Is my assumption correct? Which one is n and which is nprime?

Is there a way to obtain the full set of gkk elements for bands other than FS
without coding? :)

Thank you,

Ilija Dukovski