Did not find all elements in MEAM library file

I have a cluster of atoms that were all 1 atom type. However, now I have separated them into two groups (surface and core). I am changing my input file to compute msd for these two different groups. I have followed the instructions for the pair_coeff command (pair_style meam) to change it appropriately. However I get the error “Did not find all elements in MEAM library file”. I never got this error when I was working with just one atom type. Attached is the input file. Note that formerly I had “pair_coeff * * library.meam Ga Ga.meam Ga”, which worked just fine when I had one atom type.
Thanks,
Anne

in.galliumsim (537 Bytes)

And don't you think you need to adjust your pair potential settings?

How should lammps know that both atom types are of thensame kind?

Axel.

Sorry but could you be more specific please? The pair style meam is the same regardless of the number of atom types. If you look at http://lammps.sandia.gov/doc/pair_meam.html you can see that the potential is written for n different atom types, specified by the pair_coeff command. Look at the Si C example, the pair_coeff command tells lammps 3 of the atom types are the same. Similarly the part of my input "pair_coeff * * library.meam Ga Ga Ga.meam Ga Ga" tells it both atom types are gallium.
I'm not quite sure what you are suggesting...

Sorry but could you be more specific please? The pair style meam is the same regardless of the number of atom types. If you look at LAMMPS Molecular Dynamics Simulator you can see that the potential is written for n different atom types, specified by the pair_coeff command. Look at the Si C example, the pair_coeff command tells lammps 3 of the atom types are the same. Similarly the part of my input "pair_coeff * * library.meam Ga Ga Ga.meam Ga Ga" tells it both atom types are gallium.
I'm not quite sure what you are suggesting...

your e-mail quoted "pair_coeff * * library.meam Ga Ga.meam Ga"

which is obviously wrong and what i was referring to, but

"pair_coeff * * library.meam Ga Ga Ga.meam Ga Ga"

isn't correct either, if you look at the documentation again.

axel.

I think there is a misunderstanding. You say "pair_coeff * * library.meam Ga Ga.meam Ga" is "obviously wrong". Your statement is incorrect- this has worked just fine in my input file. My question was about a new pair_coeff line for my new input file with 2 atom types.
Anyway, in case anyone has had the same problem "pair_coeff * * library.meam Ga Ga.meam Ga Ga" is correct. The problem was that for multiple atom types of the same element, you only need to specify that element once in between the library and .meam arguments. Note that you still must repeat the element for the number of atom types after the .meam argument.

I think there is a misunderstanding. You say "pair_coeff * * library.meam Ga Ga.meam Ga" is "obviously wrong". Your

it is obviously wrong for an input with two atom types like you were
describing and asking help for.

statement is incorrect- this has worked just fine in my input file. My question was about a new pair_coeff line for my new i

yes, but it only works for system with one atom type, not two.

i suggest that in the future you do not quote parts of an input that
you don't ask about and don't expect people to *have* to look it up in
an attachment.

as always: the easier you make it to help you, the more likely you are
to get good help.

axel.