[lammps-users] meam parameter input

LAMMPS LAMMPS Mailing List Mirror
1

Hi Lammps users,

I have a question regarding use of the MEAM potential. How do I put the
parameters from the papers into the parameter files provided with LAMMPS?

The documentation doesn't explain the parameter files and neither do the
reference papers cited in the documentation. I tried to match the
parameters but it didn't work out (see attached). There's something wrong
with this even before I add in the LJ components to the potential.

Thanks for your help,
John

the specifics:
I want to model Graphene using the potential parameters from B-J Lee, J.W.
Lee Calphad 29 (2005) 7-16. How do I put these parameters into the
library.meam and the C.meam files?

He gives these parameters in the paper:
Ec 7.37 re 1.54 B 4.446 A 1.18
B0 4.25 B1 2.8 B2 2.0 B3 5.0
t1 3.2 t2 1.44 t3 -4.48
Cmax 2.8 Cmin 1.41 d 0.0

Now the problem is that the library.meam file has a different format.

# meam data from vax files fcc,bcc,dia 11/4/92
# elt lat z ielement atwt
# alpha b0 b1 b2 b3 alat esub asub
# t0 t1 t2 t3 rozero ibar
'C' 'dia' 4. 6 12.0111
4.38 4.10 4.200 5.00 3.00 3.567 7.37 1.000
1.0 5.0 9.34 -1.00 2.25 1

And the additional parameter file, for example the one from SiC.meam,
also has a different format:
lattce(1,2) = 'dia'
Ec(1,2) = 6.4325
alpha(1,2) = 4.37
re(1,2) = 1.8878
rho0(2) = 2.25
rc = 4.0
delr = 0.1
Cmax(1,2,1) = 4.0
Cmax(1,2,2) = 4.0
Cmax(2,2,1) = 4.0
Cmax(1,1,2) = 4.0

data.graphene (13.2 KB)

lib.meam (421 Bytes)

C.meam (369 Bytes)

graph.out (61.2 KB)

graph.lammpstrj (205 KB)

graph.in (405 Bytes)

2

This is a Q for Greg Wagner (CCd).

Steve

3

Hi John,

Have you solved this problem to use the meam potential in B-J Lee,
J.W. Lee' paper (Calphad 29 (2005) 7-16)?

Recently, I'm also trying to use the MEAM potential to simulate the
graphene. After reading the papers (Calphad 29 (2005) 7-16 and PRB,
64,184, 102(2001)) and the code of meam potential, I think the setup
in the potential files may be in the below way:

lib.meam

4

Hi Dongshan,

Thanks for your help! I tried out the parameters as you suggest, but I
found that they cause the graphene sheet to compress unphysically in a NVE
simulation. The simulation quickly becomes unstable and collapses.

For just 2 carbon atoms, the bond average bond length is 1.16948 (separate
simulation 500 frames NVE).

This was using the LAMMPS (24 Sep 2010) version.

I used exactly these parameters (Re: Lee, et. al. Calphad 29 (2005)):
# meam data from vax files fcc,bcc,dia 11/4/92
# elt lat z ielement atwt
# alpha b0 b1 b2 b3 alat esub asub
# t0 t1 t2 t3 rozero ibar

'C' 'dia' 4. 6 12.0111
4.38 4.25 2.80 2.00 5.00 3.567 7.37 1.18
1.0 3.2 1.44 -4.48 1.0 3

C.meam:
re(1,1) = 1.42
delr = 0.0
nn2(1,1) = 1
Cmin(1,1,1) = 1.41

Thanks again for your advice! I don't know if I am missing something
obvious. It is not clear to me how to get the graphene to work. Did you
manage to get graphene to run using these parameters? What version of
LAMMPS are you using?

Best,
John

data.graph (6.55 KB)

graph.out (6.21 KB)

graph.in (371 Bytes)

graph.lammpstrj (33 KB)

library.meam (421 Bytes)

C.meam (58 Bytes)

5

Hi John,

Yes, the graphene with the present MEAM potential input will collapses
in MD simulation after tens of steps. I'm sorry that I gave your an
inconsiderate answer before careful testing. I just tested the input
for one step running.

I also tried to use the other 5 parameter sets for C element in
library.meam and found the graphene still collapsed. I guess there
must be wrong about the graphene simualtion using MEAM in lammps. But
now I don't have time to look into the code. Hope somebody can make it
clear. If you solve this problem, please tell me.

Thanks!

Dongshan