Issue with MEAM potential availability and fix atom/swap for Al–Si–Cu–Mg–Fe and Al-Mg-Zrsystem

I am working on an MD simulation using LAMMPS for a multicomponent alloy system consisting of Al–Si–Cu–Mg–Fe . I am using the MEAM potential framework.

While checking the official MEAM potential repository (https://www.ctcms.nist.gov/potentials/MEAM/), I could not find a single combined MEAM parameter file that supports all five elements (Al, Si, Cu, Mg, and Fe) together. and i want this comination also Al-Mg-Zr .

Because of this, I am facing difficulties when using fix atom/swap , as it requires a consistent interatomic potential definition for all involved atom types . The simulation either fails during initialization or gives errors related to missing potential parameters.

If a complete Al–Si–Cu–Mg–Fe and Al-Mg-Zr MEAM potential is not available, what is the recommended approach?

There is no such thing as an “official MEAM potential repository”.

The kind folks at NIST have put in a significant effort to build a database of potential files, but it is not the only database (there is one by the OpenKIM project, too) and neither is by no means complete. In both cases the database maintainers depend on individual people submitting their potentials for inclusion into their database. Not everybody does that.

You have multiple options:

  1. search the published literature to see, if there is a MEAM potential parameterization that fits your needs
  2. search the published literature to see, if there is some other kind of potential that fits your needs
  3. learn how to parameterize MEAM potentials and create a suitable set of parameters by yourself (and publish it and submit it to the NIST database, of course :wink:)
  4. find a collaborator that knows how to create such parameterizations and work on creating a suitable parameter set together
  5. these days there are also many options to use machine learning to somewhat automate the process of parameterizing a model. such ML potentials come many different “flavors”: some simpler and faster, some highly accurate and slower. Again, a thorough survey of the published literature is the obvious thing to do
  6. MEAM potentials come in two parts: the MEAM library file with basic parameters and the potential file with adjustments for the specific combination of elements and proportions and thermodynamic conditions. If the elements you desire are all present in the first file, but you don’t adjustment settings for all elements in the second file, you can still run calculations. But of course those will have significantly reduced accuracy, since the parameter set you are using has not been adjusted for your specific compound.

Regardless of which path you follow, please keep in mind that regardless of whether there is a published potential for the elements you are looking for, it does not automatically mean that that potential is suitable for the specific conditions and the makeup of your compound. If you are studying something that the parameter set was not optimized for, you may get poor results. So it is always critical that you validate the potentials and settings. This is extremely important for the option 6.