In data 23 agosto 2009 alle ore 18:46:26, Bipul Rakshit <bipulrr at gmail.com> ha scritto: > I have a question about phonon calculation. > I am doing phonon of K2S (antifluorite structure), using 10x10x10 > k-points mesh. In that calculations i find -ve frequencies. > > But then on increasing the mesh 14x14x14 and also the kinetic energy > cutoff from 32 to 38 Ry, now i am getting all the +ve freq.
Dear Bipul, unluckily convergence of physical quantities is not variational with the number of k-points; i.e. if you increase the number of k-points you will go toward convergence, but not necessarily in a monotone way. Notwithstanding that, if I had to bet, I would say that the problem was the cutoff too low: if the Potassium pseudopotential includes semi-core electrons it may be quite hard. Without knowing the detail of your calculation there is little more I can honestly say. You could try to perform two more calculation: one increasing only the number of k-points and one only the cutoff. best regards -- Lorenzo Paulatto SISSA & DEMOCRITOS (Trieste) phone: +39 040 3787 511 skype: paulatz www: http://people.sissa.it/~paulatto/ *** save italian brains *** http://saveitalianbrains.wordpress.com/
