Atoms lost in units:real

Thank you Dr.Axel. I am using lammps for my simulation. I have run Argon simulation in an nve ensemble. As I said previously this works in lj unit system but not in real units. I ran another example which is mentioned on your webite as tutorial 2 for topotools. The system is methane on a lattice (64 methane molecules) and uses real units for the simulation. I ran this system in the nve ensemble with the exact data file and the input file provided on the webpage. I did not carry out minimization or equilbration ( I deleted the lines after the run 100 command in your input script file in.step2a, and ran the system in the nve ensemble for 1000 time steps). But the energy is not conserved. I am assuming that this configuration is physical and that the integrator used by lammps is accurate and the pair coefficients which you provided in the input script file are consistent with the units real used in the simulation.

I do not presently know why the energy is not conserved for this system. I have attached the input file and the output from my simulation.

input script for topotools tutorial step 2a

units lj
boundary p p p
atom_style full

read_data data.step2a

interaction styles

pair_style lj/cut/coul/cut 2.5
bond_style harmonic
angle_style harmonic
#dihedral_style opls
pair_modify mix geometric tail yes

OPLS considers 1-4 interactions with 50%.

special_bonds lj/coul 0.0 0.0 0.5

force field parameters

pair_coeff * * 1 1 # CT

the remaining parameters are inferred from mixing.

bond_coeff 1 340.0 1.09 # CT-HC
angle_coeff 1 33.0 107.8 # HC-CT-HC

initialize random velocities and run a few steps

of MD to break symmetries.

velocity all create 100.0 43244 dist gaussian mom no rot no
fix 1 all nve
thermo 100
thermo_style custom pe ke etotal
run 1000

PotEng KinEng TotEng
236.82259 95.087841 331.91043
169.81252 118.50126 288.31378
-12.051649 148.73705 136.6854
-47.48619 135.26383 87.77764
-10.760042 132.74911 121.98907
15.423638 132.36293 147.78657
-33.64636 121.6198 87.973439
-15.355468 144.60607 129.2506
2.5017449 136.58239 139.08414
38.394641 118.31986 156.7145
-42.575488 147.23727 104.66178

I will look into this more carefully. If everything goes right then the energy should be conserved in the NVE ensemble right ?.

Thanks
Arun

Thank you Dr.Axel. I am using lammps for my simulation. I have run Argon
simulation in an nve ensemble. As I said previously this works in lj unit
system but not in real units. I ran another example which is mentioned on
your webite as tutorial 2 for topotools. The system is methane on a lattice
(64 methane molecules) and uses real units for the simulation. I ran this
system in the nve ensemble with the exact data file and the input file
provided on the webpage. I did not carry out minimization or equilbration (
I deleted the lines after the run 100 command in your input script file
in.step2a, and ran the system in the nve ensemble for 1000 time steps). But
the energy is not conserved. I am assuming that this configuration is
physical and that the integrator used by lammps is accurate and the pair
coefficients which you provided in the input script file are consistent with
the units real used in the simulation.

I do not presently know why the energy is not conserved for this system. I

i already gave you a list of possible reasons. i am convinced that it
is due to bad input parameters. i also told you, that you need to get
more training in MD fundamentals or else this will not end well.

axel.

this works in lj unit system but not in real units.

Then your real script is not doing the same thing
as your LJ script. If you choose the params in your
real script correctly you can make it do exactly

the same thing as your LJ script. The only difference
will be small round-offs so that the 2 trajectories
drift apart (slowly) over 100s or 1000s of steps.

It’s really up to you to figure out what is different
about your 2 scripts. No one on the list is likely
to figure that out for you.

Steve