read_restart for two different structure

LAMMPS LAMMPS Mailing List Mirror
1

Hello,

Hello,

I am trying to attach DLC film on BCC iron structure.

pair_coeff arguments in below input script worked when I tried to create C and Fe in one input script without using read_restart.

However, I am getting error as “incorrect args for pair coefficients” when I use read_restart to bring out DLC structure.

Here’s what I did in the input script,

  1. Prepare DLC structure with restart file (restart.DLCall_end.1310721)

  2. Read restart file and create iron structure below DLC box

  3. Equilibrate

I am using airebo for C-C bond, and eam/fs for Fe-Fe and Fe-C bonds.

I am not sure if I can generate another structure attached to the structure recalled by read_restart, because I couldn’t define atom types on atoms brought by read_restart.

What if I want to use read_restart twice so that I can bring two different structure with different shape/potential in one input and simulate it? Is it possible? For example, I have semispherical body and box to make sliding contact between them, i.e., restart.sphere and restart.box will be read in the input and simulate sliding contact.

— input ------------------------------------------------------------------

initialization

units metal

dimension 3

atom_style atomic

read_restart restart.DLCall_end.1310721

#Fe

lattice bcc 2.87

change_box all z final -17.85 17.85 boundary p p f units box

region

region sub block 0.0 17.85 0.0 17.85 -17.85 0 units box

region box block 0 17.85 0 17.85 -17.85 17.85 units box

region DLC block 0 17.85 0 17.85 0 17.85 units box

#iron

mass 1 55.845

create_atoms

create_atoms 1 region sub

group

group box region box

group iron type 1

group sub region sub

group DLC region DLC

pair_style

pair_style hybrid airebo 3.0 eam/fs

pair_coeff

pair_coeff * * eam/fs Fe-C_Hepburn_Ackland.eam.fs Fe C

pair_coeff * * airebo CH.airebo NULL C

2

How many atom types were defined in your restart file?
If it is not 2, then the pair_coeff commands do not match
the # of atom types.

Steve

3

There is only one atom type(carbon) in restart file.

I want to make two different lattice structure in one simulation. But they are far different in potential, shape, position, process of preparation.

Going back to the first question,

Is it possible to use read_restart twice so that I can bring two different structure with different shape/potential in one input and simulate it? For example, I have a semispherical body and a box to make sliding contact between them, i.e., restart.sphere and restart.box will be read in the input and simulate sliding contact.

Thank you,

Sungae Lee

4

So b/c you have 1 atom type in the restart file, that
is why you are getting your error.

You cannot read 2 different restart files into one
simulation script. You can convert both of them
to data files, then merge the data files (by hand)
into one data file and read that in to perform a simulation.

Since one of the restart files has a single atom, that should
be trivial to do.

Steve

5

Thank you for the help, Steve.

I want to make sure one thing here.

So it is impossible to create two lattice in same input script, i.e., one from read_restart and the other from create_atoms.

In other words, to create atoms, I should use either read_restart(data) or create_atoms, but both.

Am I correct?

Thank you again.

Sungae

6

You can read_restart followed by creating some more
atoms. But some of the settings in the restart file

(like # of atom types) cannot be changed once the
restart file defines them. If you need to that, I suggest

turning the restart file into a data file, which you can
easily edit.

Steve

7

Thank you for the help, Steve.

I want to make sure one thing here.

So it is impossible to create two lattice in same input script, i.e., one
from read_restart and the other from create_atoms.

In other words, to create atoms, I should use either read_restart(data) or
create_atoms, but both.

Am I correct?

not really.

the decisive step that can be done only once is the creation of a
simulation box, as that also implies locking in the atom_style, the
numbers of various types and some other global settings (like newton).

both read_restart and read_data include this, create_atoms does not.

in you case, rather than using read_restart, i would recommend to
convert the restart to a data file and then edit the data file to
adjust the number of types as needed to make room for the atoms you
want to create later on. restarts cannot be changed, since they are
binary.

you can then use create_atoms to your liking for as long as you do not
exceed the boundaries defined in the data file.

axel.