Problems setting up NPT simulation

Dear lammps user

I am a first time user of lammps and I’d like to do a NPT MD of Au following the procedure of Heinz et al (J. Phys. Chem. C 2008, 112, 17281–17290). However, I fail already at the first step, computing the equilibrium density. Using the LJ parameters from the article, the Au atoms form some kind trimers. At step 0 everything is correct, atom positions and cell are as expected. But only after a few steps the mess begins. Quite weird. Something is wrong, probably with the fix command, and I’m not sure where to start with. It would be very kind if you could take a look at my input files to see if there is something obviously wrong.

Thank you very much!

Roger

The input file:

clear

------------------------ INITIALIZATION ----------------------------

units real
dimension 3
atom_style full
boundary p p p

---------- Create Atoms ---------------------

read_data input.lmpdat

replicate 5 5 5

------------------------ FORCE FIELDS ------------------------------

pair_style lj/cut 12.0
pair_coeff 1 1 5.29 2.951

------------------------- SETTINGS ---------------------------------

fix 1 all npt temp 298.15 298.15 5.0 &
tri -1.0 -1.0 100.0
thermo 100
thermo_style custom temp press etotal ke pe ebond eangle edihed eimp evdwl ecoul elong

dump dumpdata all atom 10 density.lammpstrj

timestep 1.0
run_style verlet

run 100000
minimize 1.0e-4 1.0e-6 100 1000

The file input.lmpdat

LAMMPS data file generated by OpenBabel
3 atoms
2 bonds
1 angles
0 dihedrals
0 impropers
1 atom types
2 bond types
1 angle types
0 dihedral types
0 improper types
0.0 2.8793 xlo xhi
0.0 2.4935 ylo yhi
0.0 7.0529 zlo zhi
-1.43960 0.0 0.0 xy xz yz
Masses

1 196.967 # Au

Atoms

1 1 1 0.00000 0.00000 0.00000 0.00000 # Au
2 1 1 0.00000 1.43967 0.83119 2.35097 # Au
3 1 1 0.00000 -0.00000 1.66239 4.70194 # Au

Bonds

1 1 1 3 # Au:Au
2 1 3 2 # Au:Au
Angles

1 1 2 3 1 # Au:Au:Au

Dear lammps user

I am a first time user of lammps and I'd like to do a NPT MD of Au
following the procedure of Heinz et al (J. Phys. Chem. C 2008, 112,
17281–17290). However, I fail already at the first step, computing the
equilibrium density. Using the LJ parameters from the article, the Au atoms
form some kind trimers. At step 0 everything is correct, atom positions and
cell are as expected. But only after a few steps the mess begins. Quite
weird. Something is wrong, probably with the fix command, and I'm not sure
where to start with. It would be very kind if you could take a look at my
input files to see if there is something obviously wrong.

​sorry, we cannot debug all input files that don't give the expected
behavior. your input seems inconsistent. you define bonds and angles, but
i there are no potentials or corresponding parameters defined for it.
please also see the thermodynamics output in your log file.

it also looks as if you are trying to do something quite complicated as the
first step. it is highly recommended to start with simpler simulations to
gain experience, best in collaboration or under the supervision of a more
experienced person.

finally, it is rarely a good idea to start a freshly constructed system
with a npt simulation and *then* run a minimization. usually the procedure
is the other way around, i.e. minimization, fixed volume simulation (fix
nvt) and only if that works well switch to fix npt.

axel.

“the Au atoms form some kind trimers.” Yes, that is because you used OpenBabel to build a system consisting of exactly one trimer. I don’t think that is going to give you a good model for Au. Try building an FCC crystal using the LAMMPS lattice command.