Buckingham potential

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

I’m currently trying to module PVDF and obtain a working potential for further studies. I’m trying to utilize potentials that are stated in literature to compare results and in some of the references they use a buckingham potential for their pair style. When I utilize this function with their constants lammps produces

an error.

When I utilize an npt it gives me this error:

ERROR: Non-numeric pressure - simulation unstable (…/fix_nh.cpp:1029)

When I use a nvt, it gives me this error:

ERROR on proc 2: Bond atoms 2208 2211 missing on proc 2 at step 10 (…/ntopo_bond_all.cpp:63)

Here is the first part of my script for lammps:

#Settings

VARIABLES

variable T equal 463.15
variable V equal vol
variable p equal 400 # correlation length
variable s equal 5 # sample interval
variable d equal 2000 # dump interval
variable dt equal 1.0 # time step

#Converting Variables

variable kb equal 1.3806504e-23 #[J/K/** Boltzman Constant]
variable atm2Pa equal 101325.0
variable A2m equal 1.0e-10
variable fs2s equal 1.0e-15
variable convert equal {atm2Pa}*{atm2Pa}{fs2s}*{A2m}{A2m}*{A2m}

variable fname index lammps_data_mod
variable simname index lammps_data_mod

Initialization

units real
boundary p p p
atom_style full
read_data lammps_data_mod

Potential Information

neighbor 0.4 bin
neigh_modify every 10 one 10000
bond_style harmonic
bond_coeff 1 327.6 1.085
bond_coeff 2 499.3 1.357
bond_coeff 3 308.9 1.534
angle_style harmonic
angle_coeff 1 42.9 108.45
angle_coeff 2 90.0 107.74
angle_coeff 3 38.5 109.27
angle_coeff 4 120.0 105.27
angle_coeff 5 62.0 107.1
angle_coeff 6 80.3 118.4
angle_coeff 7 80.3 118.4
dihedral_style fourier
dihedral_coeff 1 6 0.395 1 180 0.72 2 180 -0.38 3 180 -0.205 4 180 0.425 5 180 -0.025 6 180
dihedral_coeff 2 3 0 1 180 0 2 180 0 3 180
dihedral_coeff 3 3 0 1 180 0 2 180 0 3 180
dihedral_coeff 4 3 0 1 180 0 2 180 0 3 180
dihedral_coeff 5 6 0.355 1 180 0.345 2 180 -0.38 3 180 0.14 4 180 0.145 5 180 -0.025 6 180
dihedral_coeff 6 3 0 1 180 0 2 180 0 3 180
pair_style hybrid buck/coul/cut 10.5 lj/cut 10.5
dielectric 1.0
pair_coeff 1 1 buck/coul/cut 14976.0 0.323624595469 640.8 10.5
pair_coeff 1 2 buck/coul/cut 4320.0 0.292825768668 138.24 10.5
pair_coeff 1 3 buck/coul/cut 45094.0 0.26191037427 260.77 10.5
pair_coeff 1 4 buck/coul/cut 14976.0 0.323624595469 640.8 10.5
pair_coeff 2 2 buck/coul/cut 2649.6 0.267379679144 27.36 10.5
pair_coeff 2 3 buck/coul/cut 12300.0 0.241365161353 53.88 10.5
pair_coeff 2 4 buck/coul/cut 4320.0 0.292825768668 138.24 10.5
pair_coeff 3 3 buck/coul/cut 135782.0 0.221126417973 106.12 10.5
pair_coeff 3 4 buck/coul/cut 45094.0 0.26191037427 260.77 10.5
pair_coeff 4 4 buck/coul/cut 14976.0 0.323624595469 640.8 10.5
pair_coeff 4 5 lj/cut 0.173 3.600 10.5
pair_coeff 4 6 lj/cut 1.165 3.310 10.5
pair_coeff 1 5 lj/cut 0.173 3.600 10.5
pair_coeff 1 6 lj/cut 1.165 3.310 10.5
pair_coeff 2 5 lj/cut 0.063 2.600 10.5
pair_coeff 2 6 lj/cut 0.425 2.310 10.5
pair_coeff 3 5 lj/cut 0.126 3.087 10.5
pair_coeff 3 6 lj/cut 0.849 2.797 10.5
pair_coeff 5 5 lj/cut 0.200 3.200 10.5
pair_coeff 5 6 lj/cut 1.345 2.910 10.5
pair_coeff 6 6 lj/cut 9.052 2.620 10.5

2

There are more settings than just the force field parameters that determine the stability of an MD run: time step, neighbor list settings, thermostat parameters, initial geometry, choice of additional fixes.

Axel.

3

Maybe it is due to poor initial conditions. Have you visualized your initial system?

4

@Stephan Paquay: Yes I have visualized my assembly; what should I be looking for to see if its unstable? Also, when using all the same parameter but instead using a Lennard-Jones potential, I don’t get this error.

Thanks,

5

If the simulation is unstable you will observe some atoms acquiring an unrealistically high velocity or fly out of the simulation box. The fact that it works for a Lennard-Jones potential however seems to imply that the initial conditions should be OK. Does the simulation work for a smaller time step size?