NPT Ensemble - Temperature and Pressure Damping Coefficients

LAMMPS LAMMPS Mailing List Mirror
1

Dear Lammps Users,

I am trying to implement NPT ensemble on my model. In my model, i am using a core-shell model to represent the anions. I am using 0.1 fs as my time step.

In the manual it is mentioned that in many models it is appropriate to use around 1000 time steps for the pressure and around 100 time steps for the temperature.

In the input file in.coreshell.thermostats (attached) he is using very small values in his NPT implementation; 0.04 for the temperature and 0.4 for the pressure.

This small values are also giving me the most reasonable behavior for my systems.

What are the mathematical/physical values for this coefficients? And what are the appropriate values for a system with a core-shell?

Best regards
Ahmed

in.coreshell.thermostats (2.3 KB)

2

If you want to know how the nose-hoover chain thermostat and barostat works, you have to study the original publications mentioned in the manual.

The numbers given in the manual are for typical Molecular systems with typical thermal coupling and compressibility. You change system properties and the required setting have to change, too. There is some degree of trial and error involved as well.

An ideal thermostat and barostat only couples weakly and provides that fluctuations that allow to approach the desired statistical mechanical ensemble.

Axel

3

I suggest you email the author of the CORESHELL package.
Hendrik’s email is at the top of the various source files
in src/CORESHELL.

Steve