ERROR: MEAM library error

LAMMPS LAMMPS Mailing List Mirror
1

DearAll,
**I have a trouble to use "**pair_style hybrid/overlay“. There are 3 elements in my simulation box: Fe N C. I use the potential as follows:
pair_style hybrid/overlay meam tersoff tersoff meam
pair_coeff * * meam 1 library.meam Fe C FeC.meam Fe NULL C # Fe-Fe, Fe-C
pair_coeff * * tersoff 1 SiC.tersoff NULL NULL C # C-C
pair_coeff * * tersoff 2 BNC.tersoff NULL N C # N-N N-C
pair_coeff * * meam 2 library.meam Fe N FeN.meam Fe N NULL # Fe-N

To use meam potential for Fe-N, I add parameters in library.meam as follows:
‘N’ ‘fcc’ 12. 7 14.006
5.96 2.75 4.0 4.0 4.0 5.00 4.88 1.80
1.0 0.05 1.00 0.0 1.0 0

But there is an ERROR: MEAM library error 2 (…/pair_meam.cpp:726), and the details are as follows:
pair_style hybrid/overlay meam tersoff tersoff meam
pair_coeff * * meam 1 library.meam Fe C FeC.meam Fe NULL C
pair_coeff * * tersoff 1 SiC.tersoff NULL NULL C
pair_coeff * * tersoff 2 BNC.tersoff NULL N C
pair_coeff * * meam 2 library.meam Fe N FeN.meam Fe N NULL
ERROR: MEAM library error 2 (…/pair_meam.cpp:726)

I don’t know how to resolve this problem. Could you please tell me how to do?

Thanks very much!
Best wishes!

BNC.tersoff (5.37 KB)

FeC.meam (797 Bytes)

FeN.meam (573 Bytes)

in.FeNC (2.08 KB)

log.lammps (1.81 KB)

library.meam (9.7 KB)

SiC.tersoff (1.53 KB)

2

DearAll,
I have a trouble to use "pair_style hybrid/overlay“. There are 3 elements in
my simulation box: Fe N C. I use the potential as follows:

i already answered this.

a) you cannot use meam twice in hybrid pair styles
b) why do you use hybrid/overlay which *adds* potentials, and not
plain hybrid, which seems more suitable to what you want.
c) you are trying to treat many-body potentials as if they are
pairwise additive. that cannot work. you are computing multiple
interactions multiple times.
c) and i have serious doubts that you can mix and match different
models potentials like this at all.

you need to discuss with your adviser and/or more experienced
colleagues, as your question is showing that you need to first gain
some more fundamental understanding of the models that you are using.
this cannot be solved by a simple "do this, not that" kind of answer
from a mailing list. ...and a mailing list cannot replace proper
training and advising.

axel.