LAMMPS WWW Site - LAMMPS Documentation - LAMMPS Mailing List Archives
Re: [lammps-users] regarding my simulation
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [lammps-users] regarding my simulation

From: Axel Kohlmeyer <akohlmey@...24...>
Date: Wed, 25 Apr 2018 06:09:24 -0400

On Wed, Apr 25, 2018 at 4:33 AM, nikhilesh jaladi <jaladinikhilesh@...24...> wrote:
hello sir,
i am trying to simulate argon channel between two fixed gold plates, for my simulation. i am trying to use eam potential for gold for some reason the file is not recognized by the lammps.

​that is because your second pair_style command will wipe out the effect of the first (same for pair_coeff)​. if you want to have two pair styles active at the same time, you have to use pair style hybrid.

there are lots of other, rather obvious issues, too:
- you are using "real" units, but most EAM potential files require using "metal" units (including all of those bundled with LAMMPS)
- using a variable cell time integrator fix (i.e. fix npt) makes little sense here, especially when coupled isotropically
​- for a system like this, you must not ignore lost atoms. those are a sign of a simulation gone bad. with all periodic boundaries, there should be no atoms lost.
- ​the sketch in the image suggest, that you have non-periodic boundaries in the direction of the gold slabs
​- you need a way to immobilize those gold plates. they should be thermalized differently, while the argon part should not be thermalized at all​
- your box definition doesn't look like it is commensurate with the gold lattice spacing and thus you won't have proper continuation across periodic boundaries

​i strongly suggest to first figure out how to do a bulk argon system simulation correctly (and in metal units!)​, i.e. setup, equilibration and production with checking the results (e.g. g(r), self-diffusion) against published data. then do the same with the gold setup, then only the gold plates, and only if all of those were done successfully, i would try to set up and run the combined system.


i am posting the code here.
log             data.85K_heating.log
units        real
boundary    p p p
atom_style    atomic
region          box block 0 150 0 6 0 6
create_box      2 box
region          1 block   0 1 0 6 0 6
region          2 block   149 150 0 6 0 6
region          3 block   2 4 0 4.7 0 4.7
region          4 block   147 149 0 4.7 0 4.7

# Create Au atoms
lattice         fcc 4.08
create_atoms    1  region 1
create_atoms    1  region 2
mass        1 196.966
pair_style    eam
pair_coeff    * * Au_u3.eam   
# Create ar atoms
lattice         fcc 5.4
create_atoms    2  region 3
create_atoms    2  region 4
mass        2 39.948
pair_style    lj/cut 8.5
pair_coeff    * * .2381 3.41
velocity    all create 85.0 1 dist gaussian
neighbor    2.5 bin
neigh_modify    every 2 delay 10
fix        1 all npt temp 85.0 85.0 25.0 iso 20.0 20.0 100000.0
thermo_modify   lost ignore flush yes
minimize     1.0e-4 1.0e-6 1000 10000
timestep    5
run_style    verlet
thermo         500
dump         85Kheat all custom 5000 heating_sim_85K.dim.*lammpstrj x y z
thermo_style     custom step pe press temp vol ke etotal density
thermo_modify   lost ignore flush yes
run        100000

the above picture is the simulation result i am trying to replicate i need help.

thanking you,
nikhilesh jaladi

Check out the vibrant tech community on one of the world's most
engaging tech sites,!
lammps-users mailing list

Dr. Axel Kohlmeyer  akohlmey@...12...24...
College of Science & Technology, Temple University, Philadelphia PA, USA
International Centre for Theoretical Physics, Trieste. Italy.