higher density problem

LAMMPS LAMMPS Mailing List Mirror
1

Dear all
I am simulating a binary Lennard Jones system with 1024 particle and 1.2 density in reduced unit.input file is like that
3d Lennard-Jones melt
variable x index 1
variable y index 1
variable z index 1

variable xx equal 8*$x
variable yy equal 8*$y
variable zz equal 8*$z

units lj
atom_style atomic

lattice bcc 1.2
region box block 0 {xx} 0 {yy} 0 ${zz}
create_box 1 box
create_atoms 1 box

#read_data initdat.dat
#read_restart restart4.dat
mass 1 1

velocity all create 1.2 11152 mom yes rot yes dist gaussian

pair_style lj/cut 2.5
pair_coeff 1 1 1.0 1.0 2.5
#pair_coeff 1 2 1.5 0.8 2.5
#pair_coeff 2 2 0.5 0.88 2.5

neighbor 0.3 bin
neigh_modify delay 0 every 20 check no
fix 1 all nvt temp 1.2 1.2 100.0

thermo_style custom step temp press pe ke etotal vol enthalpy
thermo 100
thermo_modify flush yes

dump 2 all custom 10 dump.xyz id type xu yu zu
dump_modify 2 sort 1

restart 1000 restart5.dat restart6.dat
run 200000

when density is 0.8442 i am getting a correct dump file from where i am getting correct msd & vanhove correlation function.But in case of density 1.2 i am getting wrong result from simulation.

Would you have any clue regarding this?

2

Dear all
I am simulating a binary Lennard Jones system with 1024 particle and 1.2
density in reduced unit.input file is like that
3d Lennard-Jones melt

[...]

when density is 0.8442 i am getting a correct dump file from where i am
getting correct msd & vanhove correlation function.But in case of density
1.2 i am getting wrong result from simulation.

what do you mean by "wrong"?

on some level classical MD is *always* "wrong".
question is, whether this degree of wrong is
acceptable for the information that you are
looking for. you also always have to consider
that in MD you have two types of intrinsic
errors: 1) the (systematic) error from the choice
of model (e.g. pairwise additive potential) and
2) the (statistical) error from incomplete sampling
due to too short trajectories and finite size effects.

Would you have any clue regarding this?

not unless you clue us in as to where *your* problem lies.

axel.