ABINIT Discussion Forums

Dear all,
I am having some difficulties using Abinit 6.2 to calculate the dielectric function of Sb2Te3. I have followed the optics tutorial which explains the procedure for calculating the dielectric function of GaAs but I do not achieve sensible numbers when I apply this same procedure to Sb2Te3 (typically, when I calculate the refractive index, the values are 10 times greater than measured values and those calculated using castep). I am certain that I am making a fundamental mistake. I would be grateful for any of your help and suggestions.

Here is an outline of my procedure:
Relax the 5 atom primitive cell (from 15 atom unit cell with spgroup=166 [R-3m])
Set the spgroup=1 (p1)
[DS1] Calculate the self consistent wave functions for the IBZ with minimum number of bands
[DS2] Do a NSC calculation with a large number of bands
[DS3 - 5] calculate the response functions for the e-field aligned along different axes.

I have attached my input file for the optical calculation and listed below some warnings that appear multiple tims in the log file.

-P-0000 hdr_check: WARNING -
-P-0000 k pt num 1 input nband= 50 not equal disk file nband= 28

This error message appears for all kpts.



-P-0000 hdr_check: WARNING -
-P-0000 Restart of self-consistent calculation need translated wavefunctions.
-P-0000 Indeed, critical differences between current calculation and
-P-0000 restart file have been detected in:
-P-0000 * the number, position, or weight of k-points
-P-0000 * the format of wavefunctions (istwfk)
-P-0000 =======================================================================


Total localisation tensor (bohr^2) in cartesian coordinates
WARNING : still subject to testing - especially symmetries.
WARNING : Localization tensor calculation (this does not apply to other propert
ies).
Not all d/dk perturbations were computed. So the localization tensor in recipr
ocal space is incomplete,
and transformation to cartesian coordinates may be wrong.

respfn : d/dk was computed, but no 2DTE, so no DDB output.
respfn : exiting-P-0000
-P-0000 ========================================================================
========


I also have few question about the optics tutorial (toptic_1.in)- for dataset 2, we are doing a NSC calculation so why is nstep not 1?

The abinit version should be 6.8.1 and not 6.2 which I wrote in the original message.

Thanks.

Rob

In the non-scf case nstep is the number of non-scf steps. You should leave it much higher than 1, and that's what is giving you garbage.