I am wondering if Parrinello-Rahman scheme can be implemented in lammps. The doc page for fix npt references the Parinello/Rahman paper, but it is not very clear if this scheme can be implemeted.
Could someone give an insight on it?
The Parrinello-Rahman (PR) equations of motion are a special case of a more general set of extended-Lagrangian methods that are implemented in LAMMPS and are invoked using the fix npt command. The understanding of extended-Lagrangian equations of motion and how to integrate them has progressed quite a bit since PR, and so the default settings of fix npt do not give you PR, but something that is better than PR. If you insist on reverting to PR, you can do that, by using keywords to change the default settings, but it is not generally a good idea, unless you really know what you are doing.