Dear GAO Zhe, Thanks for your suggestion. Of cause one can calculate the vibrational free energy directly from the frequencies. I did not think about that. :(
Regards Torstein Fjermestad University of Oslo Norway On Wed, 11 Jul 2012 17:54:42 +0800 (CST), "GAO Zhe" <flux_ray12 at 163.com> wrote: > According to my experience, the general process is right, but, this > general process is normally working for crystal rather than > amorphous, > liquid and gas. Since the contribution from vibration is not that > much > in those matter, for example, transition contributes much more in gas > phase. Therefore, Monte Carlo or Molecular Dynamic could be better > choice in these cases. > Technically, in your calculation, H2O molecule may exist in a box, > therefore only the Gamma phonon calculation is meaningful. Since q2r > means fitting the finite samplings of dynamic matrics to infinite > interatomic force constants, only one point is not that reliable to > find good aproach of IFCs. Actually, since vibrational free energy is > defined with the term of hv, you can derectly obtain Fvib from your > first step. > > -- > GAO Zhe > CMC Lab, Materials Science & Engineering Department, > Seoul National University, South Korea >