[anaddb] lowest mode true negative  [SOLVED]

MULTIBINIT, aTDEP, ANADDB, optics, cut3d, mrg* postprocessors

Moderators: MMNSchmitt, gonze

Locked
kduffelmann
Posts: 3
Joined: Wed May 22, 2019 5:18 pm

[anaddb] lowest mode true negative  [SOLVED]

Post by kduffelmann » Wed May 22, 2019 5:29 pm

Hello,
I'm trying to simulate a small Silicon-Cristall, to get the Raman-Signal.
Following the tutorial of NLO, The calculation comes to an end in the anaddb run.
The errorcode is

Code: Select all

Electronic dielectric tensor
   4072.93389323  -3711.12116148      3.55711257
  -3711.12116148   4072.93389323      3.55711257
      3.57613481      3.57613481      1.78547640

--- !ERROR
src_file: m_ddb_diel.F90
src_line: 368
mpi_rank: 0
message: |
    The lowest mode appears to be a "true" negative mode,
    and not an acoustic mode. This precludes the computation
    of the frequency-dependent dielectric tensor.
    Action : likely there is no action to be taken, although you,
    could try to raise your convergence parameters (ecut and k-points).
    For your information, here are the four lowest frequencies :
     -1.08523540E-03 -8.27428547E-04 -7.94477183E-04 -5.45799253E-04
...

If I check with a run, where the silicon is in full structure it is quite to the end.
Is there a way to remove the problematic modes from the calculation?

Attached is the out file.
analyseddb.out
Output from anaddb
(38.16 KiB) Downloaded 386 times


Best Regards,
Konrad

ebousquet
Posts: 469
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: [anaddb] lowest mode true negative

Post by ebousquet » Tue Jun 04, 2019 11:04 am

Dear Konrad,
I think there is something wrong in your calculation because in the phonon frequencies you have many unstable modes, which should not be there for Si:

Code: Select all

 Phonon frequencies in cm-1    :
- -2.380835E+02 -1.815993E+02 -1.723104E+02 -1.197845E+02 -2.434362E-01
- -2.289132E-01 -2.088297E-02  6.349271E+00  4.785288E+01  6.673645E+01
-  8.355107E+01  8.519651E+01  9.529332E+01  1.011378E+02  1.248023E+02

Either the structure you used is not OK or the convergence is not OK, etc.
Best wishes,
Eric

kduffelmann
Posts: 3
Joined: Wed May 22, 2019 5:18 pm

Re: [anaddb] lowest mode true negative

Post by kduffelmann » Mon Jul 29, 2019 4:13 pm

Hey Eric,
sorry for my late reply.
I tried to change the way of describing Silicon and lost the group in the process.
Thanks for your suggestions.
Best regards
Konrad

Locked