Dear all,
I used fix nve along with fix langevin and fix press/berendsen to model the Al liquid in my submitted manuscript, the corresponding commands in the "in.Al" file to perform thermostat, barostat and time integration was like this:
fix 3 all langevin 1500 300 0.5 920332 zero yes
fix 2 all press/berendsen iso 0 0 1000
fix 1 all nve
run 100000
unfix 1
unfix 2
unfix 3
And in the manuscript I described the above methods like this:
"the simulations were carried out in an NVE ensemble with constant particle number (N), volume (V), energy (E), accompanied by the Langevin thermostat method to maintain a constant T and the Berendsen barostat method to maintain a constant P. During the simulation, pressure P was kept zero."
However, the editor and reviewers thought that "temperature and/or pressure of the system cannot be controlled in the NVE ensemble", and recommended me to describe the method correctly.
Now I'm wondering what's wrong with my description as provided above, and how to improve it. I know that the use of fix nve together with fix langevin and fix press/berendsen are allowed according to LAMMPS manual, however, is it suitable to say that my simulated system is in an nve ensemble? Or should I say that it's in an npt ensemble since the temperature and pressure are controled during the simulation? Could anyone help me with a more correct description of the above methods in my revision?
Dear all,
I used fix nve along with fix langevin and fix press/berendsen to model the Al liquid in my submitted manuscript, the corresponding commands in the "in.Al" file to perform thermostat, barostat and time integration was like this:
fix 3 all langevin 1500 300 0.5 920332 zero yes
fix 2 all press/berendsen iso 0 0 1000
fix 1 all nve
run 100000
unfix 1
unfix 2
unfix 3
And in the manuscript I described the above methods like this:
"the simulations were carried out in an NVE ensemble with constant particle number (N), volume (V), energy (E), accompanied by the Langevin thermostat method to maintain a constant T and the Berendsen barostat method to maintain a constant P. During the simulation, pressure P was kept zero."
However, the editor and reviewers thought that "temperature and/or pressure of the system cannot be controlled in the NVE ensemble", and recommended me to describe the method correctly.
Now I'm wondering what's wrong with my description as provided above, and how to improve it. I know that the use of fix nve together with fix langevin and fix press/berendsen are allowed according to LAMMPS manual, however, is it suitable to say that my simulated system is in an nve ensemble? Or should I say that it's in an npt ensemble since the temperature and pressure are controled during the simulation? Could anyone help me with a more correct description of the above methods in my revision?
the reviewers and the editor are correct. you should inform yourself
about statistical mechanical ensembles before making statements like
this. it has been explained *very* many times that using fix nve does
not necessarily mean you are running a simulation in NVE ensemble.
this is only the case if there is *nothing else* that modifies either
the number of particles, the volume or the total energy.
with fix press/berendsen you modify the second and with fix langevin
you modify the last.
calling this NPT ensemble is also not correct. at best it is some kind
of approximation. what you do is: time integration with a velocity
verlet algorithm, temperature control with langevin dynamics (quote
parameters) and pressure control with a berendsen barostat (quote
relaxation time). plain and simple. both langevin and berendsen add a
bias to your sampling that makes it different from a proper NPT
ensemble. whether this has a significant impact on your results or not
is a different story.