Your interpretation is correct.

The k-points refer of course to the k-points in case.klist (or case.klist_band if you used x lapw1 -band).
So you can create your case.klist_band using xcrysden, then run
x lapw1 -band
x optic
and the k-points are as you specified.

The "band index" refers of course to the eigenvalues. So you have to look at case.output1 and "count" the eigenvalues to know which band index refers to which eigenvalue. There is no way to specify "10 bands around EF", but it should be fairly trivial to find out which bands you are interested in.

On 11/02/2016 08:05 AM, Yong Woo Kim wrote:
Dear Wien2k users,

Hello, I am running wien version 14.2 on linux compiled with gfortran.

My purpose is to calculate the momentum matrix elements between
conduction and valence bands near the band gap of corundum
sapphire(Al2O3), along a certain path in the Brillouin zone.

I followed the user guide and managed to get the .mommat2 output file as

   Complex momentum matrix elements: (Re_x,Im_x),(Re_y,Im_y),(Re_z,Im_z)

   KP:     1 NEMIN NEMAX :     1   86 dE: -5.0  3.0 K:         1

      1   1-0.747795E-14-0.139254E-17 0.676976E-14-0.162488E-19
0.307244E-03-0.374322E-10   0.00000000
      1   2-0.991857E-15-0.163311E-16 0.318893E-14
0.454409E-18-0.183492E-10 0.244818E-16   0.00066966
      1   3 0.287701E-03
0.469102E-14-0.889069E-03-0.145126E-13-0.111407E-13 0.903536E-17
      2  10 0.323187E-05 0.109539E-14 0.234143E-04 0.790507E-14
0.108066E-14 0.169248E-16   0.00134548

... and so on

However, I am having difficulty in interpreting this output file. What I
understood about the output is that the program calculated the momentum
matrix elements for 86 bands at the k point 1, and moves on to k point 2
and so on. (Looking at the end of the file KP extends up to 146)

If my interpretation is right(please correct me if I'm wrong), the
problem is that I do not know where KP=1 is in the BZ and which band
bands 1-86 refer to.

So my question is, is there a way to obtain the momentum matrix elements
for certain bands, lets say about 10 bands that are closest to the band
gap? Also I would like to know if it is possible to obtain the matrix
elements along a certain path in the BZ. (Such as the path that I took
to draw the band structure using 'bandstructure')

It would be really helpful if anyone could help me on this matter.
Thank you very much in advance.

Best regards,
Yong Woo Kim

Wien mailing list


Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-165300             FAX: +43-1-58801-165982
Email:    WIEN2k:
Wien mailing list

Reply via email to