Fixing all internal motion

[Axel]

On Jun 17, 3:35 am, yimanu <duu... at gmail.com> wrote:
> Hello everyone

hello yimanu,

> Does anyone know a trick to fix all bonds, angles and
> dihedrals(allowing for just translation of the molecule) of a
> particular molecule
> in an md simulation apart from using many G3X3s?

yes. the trick is to treat that molecule as a rigid
body and propagating it via its center of mass for
translation and quaternions (or euler angles) for
rotation. see, e.g, http://lammps.sandia.gov/doc/fix_rigid.html

to the best of my knowledge, this is not (yet) possible in cp2k,
but i may be wrong (there is a _lot_ of code in cp2k...)

using a large number of individual constraints will become
quite time consuming and may get you into convergence problems.

cheers,
   axel.


>
> thanks
>
> yimanu