Dear lammpsusers,

I am simulating a 4nm diameter nanoparticle with fixed Boundary conditions on three directions. I have used vacuum surrounding the nanoparticle by placing the nanoparticle at the geometric centre of 8nm cube (simulation box). I use NVT to perform equilibration at 300 K. On equilibration, I get pressure oscillating around 0 bar which I expect as vacuum is there.

As lammps uses the entire volume of simulation box for pressure calculations, I am wondering if this is the correct way to get pressure as our crystal doesn’t occupy the simulation box completely. Can we get the dynamic volume of crystal at each timestep and use for pressure calculations?

Any help is greatly appreciated.

Thanks

Nikhil

Dear lammpsusers,

I am simulating a 4nm diameter nanoparticle with fixed Boundary conditions

on three directions. I have used vacuum surrounding the nanoparticle by

placing the nanoparticle at the geometric centre of 8nm cube (simulation

box). I use NVT to perform equilibration at 300 K. On equilibration, I get

pressure oscillating around 0 bar which I expect as vacuum is there.

As lammps uses the entire volume of simulation box for pressure

calculations, I am wondering if this is the correct way to get pressure as

our crystal doesn't occupy the simulation box completely. Can we get the

dynamic volume of crystal at each timestep and use for pressure

calculations?

what is the point? pressure has no meaning for an isolated particle in

vacuum. there is nothing it is interacting with but itself and this will

relax into the state, that is defined by the force field you are using.

besides, the volume of such a nano-particle is not well defined in the

first place, so you can only approximate it. ...and even if you get a good

approximation, the outcome is just a scaling factor.

axel.

For calculating any quantity in MD, you should get its average on the corresponding ensemble (NVT in your case) which is equal to the average over time (the ergodic hypothesis). As some properties like temperature have small oscillations during time, their instantaneous values and their ensemble average are very close so you will be happy that your temperature is correct. In contrast, some properties like pressure of solids and liquids have aggressive oscillations during time which makes you upset of not getting the right pressure (and even not getting any reportable value). But, don’t worry. You can average pressure during a long period of time. If you did the other things correctly, your average value should be close to the desired one (zero in your simulation).

As Axel has told, while you are not imposing nay external force to the crystal, the pressure should be zero.

By the way, as solids and liquids are almost incompressible, only a small difference in the density will result in considerable changes in the pressure. Thus, don’t be so sensitive to the exact value of the pressure.

Sent with Mailtrack