lost atoms and particles not depositing properly

LAMMPS LAMMPS Mailing List Mirror
1

Hi, lammps user.

I am doing a CVD of graphene growth on Ni and Cu substrate. But I am facing a lot of problems. (I am using reactive force field)
Problems:

  1. Whenever I do charge equilibrium few atoms get lost

Including charge equilibrium (Putting thermo modify ignore )
– Whenever I try to deposit atoms, it does not deposit. Warning of deposition failure is shown
Including charge equilibrium and including NVE
– The atom deposits but does not interact with the particles, diffuses and leaves the simulation box

Excluding charge equilibrium

  1. The atom deposits but does not interact with the particles, diffuses and leaves the simulation box

I am not understanding the reasons. I have added the files including the potential files. Can anyone please help me. If you can provide me with a sample script for CVD, It will be really helpful for me.

N.B. I have also tried using the deposition script in lammps example. The same scenario is happening

Thanks in advance

ffield.reax.CHNi (5.85 KB)

inNi.lmp (2.25 KB)

ffield.reax.CCu (15 KB)

inCu.lmp (1.84 KB)

2

your input files have multiple problems:

  • with reaxff you usually need a shorter timestep than what is the default for real units
  • in the Cu example you use fix temp/berendsen and fix nvt on the same group of atoms. a very bad idea
  • in the Ni example there is no time integration in the part where fix deposit is used

also, you are running at a fairly high temperature. have the force field parameters you are using parameterized for that? keep in mind that ReaxFF parameters are not very transferable.

I would suggest you build your input in stages:

  • first simulate the metal by itself and find the simulation settings you need for good energy conservation and reasonable behavior of the system.
  • then add a few C atoms (close to their expected final adsorption site) using create_atoms single and repeat the process of adjusting the simulation settings for a stable and meaningful simulation
  • then you can switch to using fix deposit.

You should also keep in mind that when depositing atoms on a hot surface, it is not automatic that they get adsorbed instantly, but rather that you get an equilibrium between a gas and the surface.

Axel.

3

Thankyou Axel for the suggession
I have tried changing it as you have said. Can you please check the file and provide me some suggestion. because The C atoms are still not interacting with the Ni atoms and passes through the simulation box

ffield.reax.CHNi (5.85 KB)

in.relax.drop.lmp (3.28 KB)