MD using LAMMPS of CVD of graphene graphene growth on a Ni (100) facet

LAMMPS LAMMPS General Discussion
,
1

I am trying to mimic the results of the paper Molecular dynamics simulation of graphene growth on Ni(100) facet by chemical vapor deposition | Journal of Applied Physics | AIP Publishing. I have written the input script ( see below) but unable to produce results. Kindly take a look and help.

units metal

boundary p p p

dimension 3

atom_style atomic

read_data ni100

dump 1 all atom 100 dump.ni100dep

timestep 0.002

run_style verlet

thermo_style custom step temp pe lx ly lz press

thermo 100

neighbor 2.0 bin

neigh_modify every 1 delay 0 check yes

pair_style hybrid lj/cut 8.0 eam airebo 3.4

pair_coeff * * airebo CH.airebo NULL C

pair_coeff 1 2 lj/cut 0.0486 3.0665

pair_coeff 1 1 eam Ni_u3.eam

velocity all create 300 123456

minimize 1e-6 1e-8 1000 10000

fix 1 all temp/berendsen 300 300 100

run 1000

unfix 1

fix 1 all temp/berendsen 300 1000 100

run 100000

unfix 1

region depo block 0 9.96992 0 9.96992 21.1494 21.1494

create_atoms 2 random 32 1234 depo

group gdepo region depo

variable ct equal temp

print “Current temp equal ${ct}”

fix 1 all temp/berendsen ${ct} 1000 100

run 10000

2

The paper specifies an integration timestep of 0.2 fs, while your timestep setting makes that 2 fs. Beyond that, unless you tell us what you were expecting to get and what you got instead, it’s not straightforward to tell what is wrong or right with your script.

1 Like
3

Please study your output. It should show a warning that must not be ignored for your purposes and exhibits a fundamental flaw in your input:

WARNING: No fixes with time integration, atoms won't move (src/verlet.cpp:60)
4

My type 1 atom is Ni and type 2 is C. And want to apply Airebo, Embedded atom model, and Lenard Jones potentials between C-C, Ni-Ni, and Ni-C atoms, respectively.
Is the below code is right ?

pair_style hybrid lj/cut 8.0 eam airebo 3.4
pair_coeff * * airebo CH.airebo C C
pair_coeff 1 2 lj/cut 0.0486 3.0665
pair_coeff 1 1 eam Ni_u3.eam

5

Why do you think there would be a problem with it?

6

I am first applying airebo to all , then defining lj and eam to specific types, so will they over ride airebo and at last my airebo will only be executed to remaining (which is 2 2 i.e C - C) ?

7

That does not answer my question.

8

I just want to know whether the applied code is exactly replicating the description provided (Airebo, Embedded atom model, and Lenard Jones potentials between C-C, Ni-Ni, and Ni-C atoms) above or NOT.

9

You won’t be any wiser if I would say “yes” or “no” to this.
How can you know whether I am correct? What if I am a person that likes to pull tricks on people and given them deliberately the wrong answer?
What are you going to write in your report or paper? “I had asked some random person to confirm whether my simulation input was correct”??

This is a decision that ultimately you have to take responsibility for and if you don’t trust yourself sufficiently, you have to apply the scientific method™ to confirm your choices.

1 Like