letif... at gmail.com
Thu Apr 18 15:10:07 UTC 2013
I am interested in the shear viscosity of some semiempirical (AM1 and PM3) water models, that requires the calculation of the stress tensor during MD simulations.
So I ran some dynamics with 64 water molecules using periodic boundary conditions and calculated the stress analytically. However, it led to a very strange behaviour of the autocorrelation function of the off diagonal elements, so I did some debug runs and realised that the virial test did not pass (the force test was OK).
Is it possible that it happens because the analytical stress calculation is not implemented yet for these SE methods?
Thanks a lot,
More information about the CP2K-user