Equilibration Procedure

Dear All,

I am a PhD student and I am relatively new to MD and LAMMPS. I am trying to simulate the following equilibration:

“All MD simulations of graphene and diamond used the same equilibration procedure. Each system was first heated from 0 to 300 K over 5 ps using a Langevin thermostat. A Berendsen thermostat and barostat were then used to maintain the system at 300 K and zero pressure for 16 ps. The barostat was then removed and the system maintained at 300 K for 2 ps. A time step of 0.2 fs was used in all simulations.”

I have used this script:

------------------------ INITIALIZATION ----------------------------

units metal
dimension 3
boundary p p p
atom_style atomic
newton on

----------------------- ATOM DEFINITION ----------------------------

read_data data.graphene

mass 1 12 # carbon

------------------------ FORCE FIELDS ------------------------------

pair_style airebo 3.0
pair_coeff * * CH.airebo C

------------------------- SETTINGS ---------------------------------

Unless you used a command-line option to not output
a log file, you should get everything (and more) that
prints to the screen in log.lammps.

Try it with one of the provided scripts if yours
is not working.

Steve

Dear All,

I am a PhD student and I am relatively new to MD and LAMMPS. I am trying to simulate the following equilibration:

“All MD simulations of graphene and diamond used the same equilibration procedure. Each system was first heated from 0 to 300 K over 5 ps using a Langevin thermostat. A Berendsen thermostat and barostat were then used to maintain the system at 300 K and zero pressure for 16 ps. The barostat was then removed and the system maintained at 300 K for 2 ps. A time step of 0.2 fs was used in all simulations.”

Keep in mind that the Berendsen thermostat generates an ensemble of conformations which are inconsistent with the Boltzmann distribution. These conformations are not the same ones you would get at your target temperature. Practically speaking, this means you won’t be able to apply statistical reweighting (or use enhanced sampling such as replica exchange) if you choose this thermostat. Berendsen has fallen out of favor for these reasons. Nobody should be using these old thermostats in this era. I’m on sort of a crusade to get people to abandon them. (It hurt us. It caused our group to waste months of time trying to figure out why our reaampled data did not make sense.)

Hi.

Thank you, I used Nose-Hoover for other cases. I was just trying to emulate the results in the paper that Berendsen thermostat. Thank you for pointing out the disadvantages of using Berendsen, which I didn’t know.

Thanks,
Deepak Behera