RuntimeError when trying to project power spectra on phonon mode about dynaphopy HOT 6 CLOSED

Apparently there is some problem with the primitive cell definition.
I will need more detailed information to help you.
Can you post the input file, structure, etc.. you are using?
Also what version of phonopy are you using?

from dynaphopy.

Hello Mr. Carreras,

This is my POSCAR file of structure I used to generate mentioned supercells:

Ge Te
1.00000000000000000000
4.3809171844060309e+00 0.0000000000000000e+00 0.0000000000000000e+00
2.3360289524527250e+00 3.7061306115675215e+00 0.0000000000000000e+00
2.3360289524527250e+00 1.2889233044168964e+00 3.4747778094906849e+00
1 1
Direct
0.0000000000E+00 0.0000000000E+00 0.0000000000E+00
0.53046975575816 0.53046975575816 0.53046975575816

My input file for Dynaphopy is:

STRUCTURE FILE POSCAR
POSCAR

FORCE CONSTANTS
FORCE_CONSTANTS

#FORCE SETS
#FORCE_SETS

SUPERCELL MATRIX PHONOPY
5 0 0
0 5 0
0 0 5

PRIMITIVE MATRIX
4.3809171844060309e+00 0.0000000000000000e+00 0.0000000000000000e+00
2.3360289524527250e+00 3.7061306115675215e+00 0.0000000000000000e+00
2.3360289524527250e+00 1.2889233044168964e+00 3.4747778094906849e+00

I am not sure why there is a problem because everything I try to do with phonopy works. I generated FORCE_CONSTANTS file, calculated band structure with phonopy.

from dynaphopy.

Now I see the problem. Your definition of the primitive matrix is incorrect (maybe the name is misleading). Primitive matrix has to be written in fractional coordinates (respect to unit cell),
equivalent to primitive axis described in phonopy manual:

http://atztogo.github.io/phonopy/setting-tags.html?highlight=tag#primitive-axes-or-primitive-axis

from dynaphopy.

Thank you Mr. Carreras, i corrected my input file and now it works.

However I still get non zero power spectra at high frequencies for acoustic modes at Gamma point and it is not single peaked but rather have multiple peaks. I have to fix my structure to run NVT to get phonons, could it be because my structure is not relaxed? Also should I be getting power spectra non zero at frequencies higher than Debye? I

from dynaphopy.

I'm not sure I understand you well, but in any case you definitely need to relax your structure to calculate the phonon force constants. You should check with phonopy that you have (at least) no imaginary harmonic frequencies at gamma. If the force constants (and therefore the phonon eigenvectors) are not properly calculated then you may get an unexpected behavior in the projections of the MD trajectory into the phonon modes.

from dynaphopy.

Thank you Mr. Carreras for your advice and patience. I am closing this issue now.

from dynaphopy.