Fix rigid and Fix temp/berendsen, press/berendsen

The answer to your first question is that it probably doesn't make sense to apply fix press/berendsen or fix temp/berendsen to groups of atoms that are in rigid bodies (i.e. where fix rigid has been applied). Even though these are not time-integration fixes, these berendsen fixes are designed to be applied on a per atom basis, so it doesn't make sense to apply them to atoms that are part of rigid bodies.

The answer to your second question is yes, it should be fine to apply fix rigid to a big (1000 - 10000 atom) nanoparticle embedded in a fluid that does not have fix rigid applied to it. Although I've not done a simulation like this, I imagine it should be possible to do it as you suggest. Should be no problem with loss of interaction between the particle and matrix, as long as you leave those interactions on, and they are defined correctly. You'll probably want to apply only fix rigid to the nanoparticle, and not apply fix rigid to the matrix fluid but instead apply a regular time integration fix to the matrix fluid.

Paul

A bit more info. On your first Q, the fix rigid doc page message
simply means that fix rigid itself does time integration, so
you should not also time integrate rigid atoms with another
fix, like fix nve or fix nvt.

Doing NPT on a system with rigid bodies is discussed on the
fix rigid doc page. See the section starting with:

If you with to perform NPT or NPH dynamics (barostatting), you cannot
use both “fix npt”_fix_nh.html and fix rigid/npt (or the nph
variants). This is because there can only be one fix which monitors
the global pressure and changes the simulation box dimensions. So you
have 3 choices …

It mentions fix press/berendsen as one of the 3 options.

Steve

Thanks Paul and Steve for yours informations.