[DFTB-Plus-User] Vibrational spectra saga continued

[Ben Hourahine]

Hello Andrei,

full k-points are supported for the electronic structure (see the manual 
section
on KPointsAndWeights). The problem could equally well be the 
initialisation of
the MD. Are you using an optimised structure and allowing for burn in 
time before
taking data? Also are you thermostatting during the run (the Anderson 
thermostat
destroys velocity autocorrelation)?

Cheers

Ben

On 14/12/10 01:01, Andrei Buin wrote:
> Thanks B?lint, it helped.  Now, i have the following problem. I'm trying
> to get vibrational spectra by fourier transforming  velocity-velocity
> auto-correlation function, and i get noise. It is interesting, if i
> increase the unit cell size i get different noise pattern. I was
> wondering does DFTB+ does only Gamma point MD or K-sampling is
> included ? If i do the same with dl_poly it gives reasonable phonon DOS.
> Sampling time is at least 10 ps. I'm doing bulk TiO2 system, with 6
> atoms in the unit cell with 8x8x8 sampling and 72 atoms with 2x2x2
> sampling. I don't think there should be problem with SK parametrisation
> since these parameters has been tested at least for the bulk.
>
> Any ideas ?
>
> Thanks.
>
> With Best Regards, Andrei Buin.
>
>
> _______________________________________________
> DFTB-Plus-User mailing list
> DFTB-Plus-User at dftb-plus.info
> http://www.dftb-plus.info/mailman/listinfo/dftb-plus-user


-- 
      Dr. B. Hourahine, SUPA, Department of Physics,
    University of Strathclyde, John Anderson Building,
           107 Rottenrow, Glasgow G4 0NG, UK.
     +44 141 548 2325, benjamin.hourahine at strath.ac.uk
    The University of Strathclyde is a charitable body,
registered in Scotland, with registration number SC015263