Unfortunately, it is not yet possible to combine the -hf and -so options
simultaneously, but probably this will be possible in the next WIEN2k
release. At the moment, the best is maybe to do what you suggest, but
you need to follow the steps below

1) run_lapw -hf ...
2) cp case.klist_fbz case.klist
3) cp case.kgen_fbz case.kgen
4) cp case.vectorhf case.vector
5) cp case.energyhf case.energy
6) initso_lapw (do NOT add any relativistic LO)
7) x lapwso (WITHOUT -p)
8) x lapw2 -fermi -so (WITHOUT -p)
9) cp $WIENROOT/SRC_templates/case.inop case.inop (same for injoint and inkram)
10) modify case.in/op/joint/kram according to your needs.
11) x optic -so (WITHOUT -p)
12) x joint
13) x kram

Before "run_lapw -hf ..." choose emax in case.in1(c) and
nband in case.inhf, which are large enough also for SO and optic,
but don't specify an excessively large nband in order to avoid
hybrid calculations that are too expensive. This is nband which will
determine the range of energy for plotting the optic spectra.

F. tran

On Monday 2016-08-22 20:00, Luis Ogando wrote:

Date: Mon, 22 Aug 2016 20:00:50
From: Luis Ogando <lcoda...@gmail.com>
Reply-To: A Mailing list for WIEN2k users <wien@zeus.theochem.tuwien.ac.at>
To: A Mailing list for WIEN2k users <wien@zeus.theochem.tuwien.ac.at>
Subject: [Wien] Dielectric function , HSE (YS-PBE0) and spin-orbit coupling

Dear Wien2k community,

   I would like to calculate the dielectric function of GaP in the wurtzite
symmetry but I only get a reasonable band structure using HSE (YS-PBE0).
   My problem is that spin-orbit coupling (SOC) has a significant
contribution to the electronic properties of this system and I was thinking
about some way to, at least, estimate the SOC effect on "epsilon".
   Considering that SOC is relevant but less important than HSE, my first
trial would be:
1) run a regular SCF cycle with -hf option ( run_lapw -hf ... )
2) after that, I would run " x lapwso " to get a rough idea of the SOC
3) calculate the dielectric function as explained in the Wien user guide for
the SOC case, but including -hf with the appropriate executables.
   As I do not in deep know the programs, I am not sure if this would work
or give rise to a crash.
   Do you believe that this procedure would give me the approximate SOC
effects on the dielectric function ?
   Any other suggestion would be highly appreciated.
   Thanks in advance.
   All the best,

Wien mailing list

Reply via email to