A problem about the effective charge functions.

Dear lammps users

When I read through the eam potential manual, I noticed that for eam potential it need three functions. But for eam/alloy it only need two functions without the effective charge function. Then I read one simple eam potential provided by the lammps: Cu_mishin1.eam.alloy. I noticed that although the manual said that the alloy should contain only two functions, it actually contain three functions. My question is: is the last function to be the effective charge function? Also is the effective charge function treated the same as simple eam potential which is:

r*phi = 27.2 * 0.529 * Zi * Zj

Thanks very much for your help.
Cheers!
Hao

You should not need to worry about how EAM potential files are formatted in different formats (funcfl vs. setfl). The rule is: use “pair_style eam” with potential files that end with “.eam” and use “pair_style eam/alloy” with files that end with “.eam.alloy”.

It appears you used pair_style eam but with a .eam.alloy potential file.

Ray

Yes, I just want to use Cu_mishin1.eam.alloy for my simulation. I only want to use eam potential since in my simulation there is only copper. Thanks very much!

If anyone have the eam format file for eam potential from mishin, that would be perfect. Thanks very much!

The EAM potential can be invoked by pair_style eam and eam/alloy. Both eam and eam/alloy pair styles can be applied to pure elements or alloys. They are exactly the same in terms of potential formalism and LAMMPS’ implementation, and the only difference is the potential file format. Therefore, you can safely use eam/alloy for your pure Cu simulation, and you should use “eam/alloy” if you were to use that particular Mishin parameterization. As far as I know, Mishin only published EAM potentials in the setfl format (therefore you have to use eam/alloy in LAMMPS).

Hope this clarifies.

Ray

The EAM potential can be invoked by pair_style eam and eam/alloy. Both eam
and eam/alloy pair styles can be applied to pure elements or alloys. They
are exactly the same in terms of potential formalism and LAMMPS’
implementation, and the only difference is the potential file format.

in fact, the eam/alloy pair style and the eam pair style use the
*identical function* for computing all interactions. you can see it in
the source code. the eam/alloy class (pair_eam_alloy.cpp) is derived
from eam and has no "compute()" function of its own. it uses the one
from the eam class (pair_eam.cpp).

axel.

Thanks all.