Langevin + NPH

[Youri Ran]

Dear all,

I'm trying to run a simulation with a Langevin thermostat and NPH barostat. Running both methods on the same particles lead to double integration per timestep. I found in the documentation that the intersection of the filters in methods must be Null.

Is there possibility of running a simulation with a Langevin thermostat and NPH barostat?

Kind regards,

Youri Ran

[Michael Howard]

Hi Youri,

I was thinking about this recently myself for a project, and I do not think it is currently possible out of the box in HOOMD. The reason is that the Langevin thermostat is actually an integration method (rather than a set of forces that are used by another integration method like velocity Verlet). NPH is also an integration method, and as you note, integration methods must operate on disjoint sets of particles.

Regards,

Mike