Dear Saikat
We agree with Roberto, MC=150 Ry is pretty low if you need to get good
phonon properties for graphene like materials.
Another delicate point would be the kpoint sampling of your system. I
did not see the size of your box, so Kgridcutoff is meaningless.
Remember, good phonons need good forces, good forces need good energies!
MeshCutoff and kpoint sampling is critical.
Regards
Ricardo
El 15/10/2012 09:12, R.C.Pasianot escribió:
Hi Saikat,
Assuming your structure is stable, negative frequencies generally
mean poor convergence of either, the optimized structure or the
following FC run.
I have no experience with CNT, though the parameters you're using
seem pretty ok to me, but for one: it is "Meshcutoff" not "Messcutoff"
as you type.
In any case, 150 Ry might be on the low side (depending on box's size).
It is a particularly critical parameter, check the manual for the
"eggbox" effect.
Also, you might want to check the *.FC file that Siesta outputs. Namely,
the forces for + and - displacements should be about the same,
particularly on the moving atom (file structure has been explained
in a recent post to the list).
Good luck,
Roberto
On Mon, 15 Oct 2012, [email protected] wrote:
Dear siesta user,
I am doing phonon calculation of carbon nanotube using siesta. But I am
getting negative frequencies for more than half of the calculated
frequencies. I wanted to know reason for getting negative frequencies
of my calculation.
I have optimized the structure using
MD.MaxForceTol 0.001 eV/Ang
And condition for scf are as follows
DM.Tolerance 0.5d-4
DM.Require.Energy.Convergence true
DM.Energy.Tolerance .5d-4 eV
My input file for type of run- FC ----
SystemName Phonon calculation # Descriptive name of the system
SystemLabel Phonon_fc # Short name for naming files
# Output options
LongOutput false
# Species and atoms
%block ChemicalSpeciesLabel
1 6 C
%endblock ChemicalSpeciesLabel
NumberOfSpecies 1
WriteKpoints true
PAO.Basis<basis.fdf
LatticeConstant 1 Ang
Messcutoff 150 Ry
Kgridcutoff 10.0 Ang
xc.functional GGA
xc.authors PBE
DM.UseSaveDM true
MaxSCFIterations 500
DM.MixingWeight 0.25
DM.NumberPulay 4
DM.Tolerance 0.4d-4
DM.Require.Energy.Convergence true
DM.Energy.Tolerance .5d-4 eV
#Atomic coordinates
%include PhononFC.fdf
AtomicCoordinatesFormat Fractional
Thanking You.
__________________
Saikat Paik
IIT Kanpur
UP,India
--
-------------------------------------------------------------------------
Dr. Ricardo Faccio
Prof. Adjunto de Física
Mail: Cryssmat-Lab., Cátedra de Física, DETEMA
Facultad de Química, Universidad de la República
Av. Gral. Flores 2124, C.C. 1157
C.P. 11800, Montevideo, Uruguay.
E-mail: [email protected]
Phone: (+598) 2924 9859
(+598) 2929 0648
Fax: (+598) 2924 1906
Web: http://cryssmat.fq.edu.uy/ricardo/ricardo.htm
