symmetry of phonon modes

Moderators: mverstra, aromero

Post Reply
fdatchi
Posts: 9
Joined: Wed Jul 21, 2010 3:22 pm

symmetry of phonon modes

Post by fdatchi » Thu Aug 19, 2010 2:54 pm

Hi all,

After reporting the eigendisplacements of phonon modes in the output of anaddb, there is an "Analysis of degeneracies and characters", where is listed the symmetry characters of each vibration mode in the form of a list of numbers as in the example below:

Analysis of degeneracies and characters (maximum tolerance=1.00E-06 a.u.)
Symmetry characters of vibration mode # 1
degenerate with vibration modes # 2 to 3
3.0 -1.0 -1.0 -1.0 1.0 -1.0 1.0 -1.0 0.0 -0.0 -0.0 -0.0 1.0 -1.0 -1.0 1.0
0.0 -0.0 0.0 0.0 1.0 1.0 -1.0 -1.0

I couldn't find any information about what this list of numbers correspond to. I imagine these are the characters of the symmetry operations of the symmetry point group but how to know which symmetry operation ? Furthermore, is the program able to determine the irreducible representations corresponding to the phonon modes ?

Thanks,
Frederic

ilukacevic
Posts: 271
Joined: Sat Jan 16, 2010 12:05 pm
Location: Dept. of Physics, University J. J. Strossmayer, Osijek, Croatia
Contact:

Re: symmetry of phonon modes

Post by ilukacevic » Fri Aug 20, 2010 10:25 am

Hi!

There was a lecture on this by G.-M. Rignanese in Lausanne this May. I have the pdf, but I cannot attach it. Maybe you could download it from cecam website if you have an account (if not, create one if it's possible). If you do, go to files section of the "Linear and non-linear responses of solids with the ABINIT software : phonons, electric fields, and other perturbations", April 26, 2010 to April 30, 2010, Location : CECAM-HQ-EPFL, Lausanne, Switzerland Tutorial and you'll find it there.
I hope this will help.

Cheers!

Igor

fdatchi
Posts: 9
Joined: Wed Jul 21, 2010 3:22 pm

Re: symmetry of phonon modes

Post by fdatchi » Fri Aug 20, 2010 12:05 pm

Thanks Igor, I could get the file.
I understand from this tutorial that the list of character numbers are given in the same order as the list of symmetry operations given in the log file.
I wonder why the symmetry operations are only given in the log file, in my opinion this is important enough information to figure in the out file...
Just a suggestion !

Frederic

roginovicci
Posts: 75
Joined: Thu Dec 02, 2010 10:36 pm

Re: symmetry of phonon modes

Post by roginovicci » Mon Sep 15, 2014 1:52 pm

I understand from this tutorial that the list of character numbers are given in the same order as the list of symmetry operations given in the log file.
I wonder why the symmetry operations are only given in the log file, in my opinion this is important enough information to figure in the out file...


The symmetry operations not even printed in log since abinit 7.6 version (and maybe earlier). While list of symmetry operations are generated in 6.8 version. Is there some additional variable I should put in input file or is there other way how to find which symmetry operation is devoted to selected symmetry character from whole Symmetry character list.

sheng
Posts: 64
Joined: Fri Apr 11, 2014 3:44 pm

Re: symmetry of phonon modes

Post by sheng » Wed Jan 28, 2015 2:19 pm

The symmetry operations not even printed in log since abinit 7.6 version


Sorry to bump this old thread, but the similar situation still persists in version 7.10.2. Has anyone know how to distinguish the symmetry characters in anaddb output with those in bilbao yet?
It is possible to guess some some obvious modes but it is not practical in general.

I also noticed that the symmetry character output is intended for the phonon computed without considering non-analyticity only.
This had been voiced out by Yaraslav in this forum before (see this thread http://forum.abinit.org/viewtopic.php?f=10&t=2739 ).

Can someone please clarify?

mverstra
Posts: 641
Joined: Wed Aug 19, 2009 12:01 pm

Re: symmetry of phonon modes

Post by mverstra » Tue May 05, 2015 3:37 pm

Hello,

the symmetry operations are in the same order as in the main abinit output, and in the DDB file. I agree it is easier to have them on hand, and will add their printing to the anaddb log file for abinit v8 (to be released in a few months or so).

There are a number of tools that analyze the irreps (e.g. in the WURM database), but it is easy to do with the bilbao web apps. The irreps are usually associated with a mode at Gamma, irrespective of the direction you choose to look at the LO TO splitting in.

Best
Matthieu Verstraete
University of Liege, Belgium

sheng
Posts: 64
Joined: Fri Apr 11, 2014 3:44 pm

Re: symmetry of phonon modes

Post by sheng » Wed Jun 03, 2015 2:21 pm

Thanks for the information, now I can assign the operation according the symrel variable.

The irreps are usually associated with a mode at Gamma, irrespective of the direction you choose to look at the LO TO splitting in.


I notice there is a mismatch of order of irreps mode when LO TO splitting is accounted for. This is due to the phonon frequencies are printed in ascending order of magnitude instead of the irrep modes.

For clarity I provide some examples below.
Without LO TO splitting:

Code: Select all

  Phonon wavevector (reduced coordinates) :  0.00000  0.00000  0.00000
 Phonon energies in Hartree :
  -1.116640E-03 -1.116640E-03 -1.116640E-03  0.000000E+00  0.000000E+00
   0.000000E+00  7.561680E-04  7.561680E-04  7.561680E-04  1.268015E-03
   1.268015E-03  1.268015E-03  2.050070E-03  2.050070E-03  2.050070E-03
 Phonon frequencies in cm-1    :
- -2.450741E+02 -2.450741E+02 -2.450741E+02  0.000000E+00  0.000000E+00
-  0.000000E+00  1.659597E+02  1.659597E+02  1.659597E+02  2.782972E+02
-  2.782972E+02  2.782972E+02  4.499383E+02  4.499383E+02  4.499383E+02

and the irreps is as follow:

Code: Select all

 Analysis of degeneracies and characters (maximum tolerance=1.00E-06 a.u.)
 Symmetry characters of vibration mode #   1
       degenerate with vibration modes #   2 to    3
  3.0 -3.0 -1.0  1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0
 -0.0  0.0 -0.0  0.0  0.0 -0.0  0.0 -0.0  1.0 -1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0
 -0.0  0.0  0.0 -0.0  0.0 -0.0 -0.0  0.0  1.0 -1.0  1.0 -1.0 -1.0  1.0 -1.0  1.0
 Symmetry characters of vibration mode #   4
       degenerate with vibration modes #   5 to    6
  3.0 -3.0 -1.0  1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0
 -0.0  0.0 -0.0  0.0  0.0 -0.0  0.0 -0.0  1.0 -1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0
 -0.0  0.0  0.0 -0.0  0.0 -0.0 -0.0  0.0  1.0 -1.0  1.0 -1.0 -1.0  1.0 -1.0  1.0
 Symmetry characters of vibration mode #   7
       degenerate with vibration modes #   8 to    9
  3.0 -3.0 -1.0  1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0
  0.0 -0.0  0.0 -0.0 -0.0  0.0 -0.0  0.0  1.0 -1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0
  0.0 -0.0 -0.0  0.0 -0.0  0.0  0.0 -0.0  1.0 -1.0  1.0 -1.0 -1.0  1.0 -1.0  1.0
 Symmetry characters of vibration mode #  10
       degenerate with vibration modes #  11 to   12
  3.0 -3.0 -1.0  1.0 -1.0  1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0  1.0 -1.0
 -0.0  0.0 -0.0  0.0  0.0 -0.0  0.0 -0.0 -1.0  1.0  1.0 -1.0  1.0 -1.0 -1.0  1.0
 -0.0  0.0  0.0 -0.0  0.0 -0.0 -0.0  0.0 -1.0  1.0 -1.0  1.0  1.0 -1.0  1.0 -1.0
 Symmetry characters of vibration mode #  13
       degenerate with vibration modes #  14 to   15
  3.0 -3.0 -1.0  1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0  1.0 -1.0 -1.0  1.0
  0.0 -0.0  0.0 -0.0 -0.0  0.0 -0.0  0.0  1.0 -1.0 -1.0  1.0 -1.0  1.0  1.0 -1.0
  0.0 -0.0 -0.0  0.0 -0.0  0.0  0.0 -0.0  1.0 -1.0  1.0 -1.0 -1.0  1.0 -1.0  1.0


In this case it can be seen clearly that vibrational modes 4-6 are acoustic modes.

However by considering non-analyticity, the phonon frequencies become:

Code: Select all

  Phonon at Gamma, with non-analyticity in the
  direction (cartesian coordinates)  1.00000  0.00000  0.00000
 Phonon energies in Hartree :
  -1.116640E-03 -1.116640E-03  0.000000E+00  0.000000E+00  0.000000E+00
   7.154374E-04  7.561680E-04  7.561680E-04  1.268015E-03  1.268015E-03
   1.268015E-03  2.008212E-03  2.050070E-03  2.050070E-03  2.866569E-03
 Phonon frequencies in cm-1    :
- -2.450741E+02 -2.450741E+02  0.000000E+00  0.000000E+00  0.000000E+00
-  1.570204E+02  1.659597E+02  1.659597E+02  2.782972E+02  2.782972E+02
-  2.782972E+02  4.407516E+02  4.499383E+02  4.499383E+02  6.291393E+02

The acoustic modes in this case become vibrational modes 3-5.

The phonon frequencies and the eigendisplacements are not ordered according to the irreps , making the assignment of symmetry modes difficult.


Another issue is that the anaddb utility is unable to correctly assign the symmetry of acoustic modes if acoustic sum rule is imposed.
For example in my calculation the acoustic modes are supposed to be a doubly degenerate state E and a single state A1, with minute differences in frequencies without acoustic sum rule but the code is able to give the correct symmetry modes.
But when acoustic sum rule is imposed so that the 3 acoustic modes are strictly zero, the code reports that the acoustic modes are triply degenerate.

I will be grateful for any clarification.

jerkov
Posts: 4
Joined: Tue Nov 26, 2019 6:02 pm

Re: symmetry of phonon modes

Post by jerkov » Fri Jan 10, 2020 2:44 am

How are you using symrel to get the irrep of your modes?

I have this output after running anaddb:
Analysis of degeneracies and characters (maximum tolerance=1.00E-06 a.u.)
Symmetry characters of vibration mode # 1
degenerate with vibration modes # 2 to 3
3.0 1.0
Symmetry characters of vibration mode # 4
1.0 -1.0
Symmetry characters of vibration mode # 5
1.0 1.0
and so on....

Others in this thread have mentioned using the order of the symmetry relations are listed in a particular order in the output file, but I can't find them.

So my question is how to go from the numbers printed below each mode (3.0 1.0 for example) to a irrep character like A2 or B1.

Cheers

ebousquet
Posts: 318
Joined: Tue Apr 19, 2011 11:13 am
Location: University of Liege, Belgium

Re: symmetry of phonon modes

Post by ebousquet » Thu Jan 16, 2020 10:12 pm

Dear jerkov,
The matrix of each symmetry operations can be found in the output file after the flag symrel, which is given in the order used to print the character of phonon modes.
To these calculated characters and the crystallographic ones can be made by comparing the characters given by anaddb for each symmetry (+1, -1, etc) to the ones tabulated in crystallo tables (for example in Bilbao web site: https://www.cryst.ehu.es/rep/point.html). However, to do so, you have to order the symmetry list given by Abinit with the ones of the crystallo tables (if you have a mx, my and mz in crystallo tables, you have to be sure you follow the order when comparing with anaddb output, I mean you have to compare mx in tables with mx from Abinit).
A more indirect technique could be through mode condensation (using freeze_displ flag https://docs.abinit.org/variables/anaddb/#alphon) and symmetry adapted mode analysis from Bilbao (AMPLIMODE module) or ISOTROPY (ISODISTORT module), but this takes some extra steps that I can detail if interested.
Best wishes,
Eric

Post Reply