[lammps-users] graphene atoms getting knocked during equilibration

I am trying to run a simulation where i have two graphene sheets and placed 10 water molecules between them. But when I am running the simulation, I see then that water molecules are knocking out the carbon atoms from the graphene sheet after some time and disintegrates it entirely. Please tell me where I am going wrong.
given below is the input script and attached with this email are the data files.

atom_style full
boundary p p p
dimension 3
newton on
units real

read_data water10mol.data
read_data double_graph.data add append

pair_style hybrid/overlay lj/cut/coul/long 9.0 airebo 3.0 0 1
pair_coeff * * airebo CH.airebo C C C
pair_coeff 1 1 lj/cut/coul/long 0.0 0.0
#pair_coeff 2 2 lj/cut/coul/long 0.066 3.550
pair_coeff 3 3 lj/cut/coul/long 0.15535 3.166
pair_modify mix arithmetic

kspace_style pppm 1.0e-4

group graphene type 2

bond_style mm3
bond_coeff 1 450.0 1.0
angle_style mm3
angle_coeff 1 100.0 109.47

neighbor 2.0 bin
run_style verlet
neigh_modify delay 0 every 1 check yes
thermo 100
timestep 0.01
fix 1 all nve
run 100000
unfix 1

Eagerly waiting for your reply
Harkishan Dua
Dept. of Physics
Assam University Silchar

double_graph.data (4.48 KB)

water10mol.data (2.11 KB)

Your force field assignments seem incorrect. You should be using pair style hybrid, not hybrid/overlay (otherwise there will be no mixing of LJ parameters) and you are adding the parameters for carbon to all atoms. For more details please see the documentation.

axel.

I have made the changes to the hybrid/overlay.
but,
when i write

2 2 airebo CH.airebo C C
(as per the data file where 2 means carbon)
instead of

    • airebo CH.airebo C C,

I get the message, incorrect args for pair coefficients.

Actually I tried running the simulation with only lj style by specifying the parameters for all different atoms.
but the graphene sheet just disintegrated on running the simulation.
that’s why i thought of taking the airebo only for carbon and lj for the remainder of atoms in a hybrid potential set.

sir what should i do to prevent the graphene sheet from breaking?

I have made the changes to the hybrid/overlay.

which changes?

but,
when i write

2 2 airebo CH.airebo C C
(as per the data file where 2 means carbon)

that is not valid. the documentation for pair style airebo states that explicitly. it also seems inconsistent with your data file which defines an atom type for carbon and one for hydrogen (to be represented with airebo, i presume). please keep in mind that in classical models atom types are assigned based on the environment, not chemical elements.

instead of

    • airebo CH.airebo C C,

I get the message, incorrect args for pair coefficients.

yes, deservedly so. Please study the documentation.

Actually I tried running the simulation with only lj style by specifying the parameters for all different atoms.
but the graphene sheet just disintegrated on running the simulation.
that’s why i thought of taking the airebo only for carbon and lj for the remainder of atoms in a hybrid potential set.

but that is not what your input did. bad input → bad forces → bad simulation.

sir what should i do to prevent the graphene sheet from breaking?

you need proper and correctly working parameters for each component (and test them separately first).
you also need to make certain that atom types are correctly assigned and used (you have overlaps there).
you need to learn more about force fields in general and how to correctly apply them in complex situations. this is beyond the scope of this mailing list. consult with your adviser to get proper training (either directly or with a sufficiently competent tutor).

the manual can help you with the syntax but not with understanding what a suitable model would be and how to get the corresponding parameters.

axel.