Toluene interface with PP

Dear lammps-users,

I’m doing simulations about a PP surface interaction with toluene environment. The initial structure is one layer of toluene (density ~1.2) above another layer of amorphous PP in z direction. I want to relax the system by getting toluene molecules closer to PP surface and let them occupy the top room above PP uniformly. I keep getting a error after 1 timestep of npt running :ERROR on proc 0: Particle on or inside fix wall surface.
Several things I’ve tried:

  1. change the wall style: lj/harmonic to lj/93
  2. change the box size to be bigger

Could anyone give me more suggestions on this? Thanks!

Lu

Here is my input file:

Dear lammps-users,

I'm doing simulations about a PP surface interaction with toluene
environment. The initial structure is one layer of toluene (density ~1.2)
above another layer of amorphous PP in z direction. I want to relax the
system by getting toluene molecules closer to PP surface and let them occupy
the top room above PP uniformly. I keep getting a error after 1 timestep of
npt running :ERROR on proc 0: Particle on or inside fix wall surface.
Several things I've tried:
1. change the wall style: lj/harmonic to lj/93
2. change the box size to be bigger

Could anyone give me more suggestions on this? Thanks!

- reduce the time step
- run a minimization first
- check your paramters
- run with nvt before switching to npt.

axel.

Dear Axel and lammps-users,

Thanks for your help. As you suggested, I’ve tried:

  • reduce the time step --> the simulation would have atoms missing at earlier stage with timestep of 0.5 and 0.1.
  • run a minimization first --> there was a minimization at the beginning already:)
  • check your paramters --> I wonder how should I adjust the wall/lj93 parameters?
  • run with nvt before switching to npt --> I’ve added in the input file before npt running, but the job is killed at the 2nd timestep of nvt already…

Here I attach the output of simulation, one thing is the potential energy becomes very high at the end of minimization, what do you think may cause this?

Thanks again!

Lu

Dear Axel and lammps-users,

Thanks for your help. As you suggested, I've tried:
- reduce the time step --> the simulation would have atoms missing at
earlier stage with timestep of 0.5 and 0.1.
- run a minimization first --> there was a minimization at the beginning
already:)
- check your paramters --> I wonder how should I adjust the wall/lj93
parameters?

- run with nvt before switching to npt --> I've added in the input file
before npt running, but the job is killed at the 2nd timestep of nvt
already...

Here I attach the output of simulation, one thing is the potential energy
becomes very high at the end of minimization, what do you think may cause
this?

what i think is:

- you have a quite outdated executable
- your simulation goes crazy already during
  the part where you run nve/limit
  find you what is happening, when your
  potential energy goes up so much.

the typical way to debug this is to write out
your trajectory to a dump file very frequently
and have a close look.

the simulation goes on for a bit, but after that
it is so messed up that it is not worth looking
at that part. the problem is earlier.

axel.

Dear Axel,

Thanks for the quick response. I would look closer to it now first :slight_smile:

Really appreciate your time.

Lu