Dear all,
i am trying to calculate the self diffusion coefficient in a periodic box. i read that the compute msd can calculate the mean square displacement. The manual says ****"If an atom is part of a rigid body (see the fix rigid command), it's periodic image flags are altered, and its contribution to the MSD may not reflect its true contribution"*** (a) If someone uses the fix shake it happens the same thing?
(b) is there any way to compute the msd in a periodic box which has the fix shake with lammps? or only via a writtend by "me " postprocessing tool?
Atoms/molecules that are SHAKEn do not alter the image flags, just fix rigid.
So you could, for example, compute MSD on a single atom in a molecule
(the O in water),
and get a good answer for the molecule. See also the compute
msd/molecule command.
Steve