[lammps-users] Create 2 atoms types using one potential?

I have been simulating the thermal properties of bulk crystalline metals with the EAM/FS potential. Everything is in order and working well. Here is the data.file I’m using for a small bulk just to test the functioning:

LAMMPS 2010-09-24 data file for Cu_3

108 atoms
1 atom types
0.0000000 10.8441000 xlo xhi
0.0000000 10.8441000 ylo yhi
0.0000000 10.8441000 zlo zhi

Masses

1 63.55 # Cu

Atoms

1 1 0.99982226 0.50285089 1.00016952
2 1 0.99982226 2.31020093 2.80751968
3 1 2.80717230 0.50285090 2.80751968
4 etc…

and here is the control file Im using:

units metal
boundary p p p

atom_style atomic

read_data data.Cu_3

pair_style eam/fs
pair_coeff * * /home/jtooker/data/Cu_EAM_FS/3/Cu.eam.fs Cu

neighbor 0.3 bin
neigh_modify delay 10

fix 1 all npt temp 300 300 0.01 iso 100000 100000 1
thermo 100
timestep 0.001

dump 1 all atom 10 dump.Cu_3
dump_modify 1 scale no

run 1000

Currently all atoms are of atom type #1. My inclination is to reassign all atoms on one face of my cubic system to atom type #2 and then use the fix command to assign a velocity to this plane of atoms. I change the data.file to say there are two atom types and I edit the masses line to show both atoms (which are identical and should interact with the same potential Im already using.) I manually go through the list of atoms and change the atom type for the appropriate atoms I get the following error from running LAMMPS.

< jtooker @localhost /home/jtooker/Desktop/3/Shock> lmp_linux < in.control
LAMMPS (9 Aug 2010)
Reading data file …
orthogonal box = (0 0 0) to (10.8441 10.8441 10.8441)
1 by 1 by 1 processor grid
108 atoms
ERROR: Incorrect args for pair coefficients

Here is the new data.file that gives me this error:

LAMMPS 2010-09-24 data file for Cu_3

108 atoms
2 atom types
0.0000000 10.8441000 xlo xhi
0.0000000 10.8441000 ylo yhi
0.0000000 10.8441000 zlo zhi

Masses

1 63.55 # Cu
2 63.55 # Cu

Atoms

1 2 0.99982226 0.50285089 1.00016952
2 2 0.99982226 2.31020093 2.80751968
3 1 2.80717230 0.50285090 2.80751968
4 1 2.80717230 2.31020093 1.00016952
5 1 4.61452246 0.50285090 1.00016952
6 1 4.61452246 2.31020094 2.80751969
7 1 6.42187214 0.50285090 2.80751969
8 1 6.42187214 2.31020094 1.00016953
9 1 8.22922230 0.50285090 1.00016953
10 1 8.22922230 2.31020094 2.80751969
11 1 10.03657246 0.50285091 2.80751969
12 1 10.03657246 2.31020095 1.00016954
13 2 0.99982226 4.11755085 1.00016952
14 2 0.99982226 5.92490101 2.80751969
15 1 etc…

What Im not understanding is how arguments for pair coefficients are even coming into the reading data.file part. Do I need to say something about pair coefficients in data.file or do I have to replace the asterisks Im using for the pair coeffs in my control file (included above)?

Thanks for any input!
jt

Hey Jonathan,

pair_coeff * * /home/jtooker/data/Cu_EAM_FS/3/Cu.eam.fs Cu
Did you make sure to add the second copper atom type to the type mapping? I.e., did you change the line to this:
pair_coeff * * /home/jtooker/data/Cu_EAM_FS/3/Cu.eam.fs Cu Cu
If not, that’d definitely cause the error.
Tyler

Hi Tyler,

Yes, thanks that was the problem. Sorry so long for the gratitude, thanks a lot!

jt