Interaction potential-Regarding

Hi all,
I am using lammps for simulating an ionic system (a complex system of perovskite structure) for thermal treatment using morse (for short range interactions) and coulomb potential (for long-range). I have two questions

  1. Is that the right combination of potential to use?

  2. And do we need to include ewald term (kspace_style) for including the long term interactions. I have read somewhere we need to include ewald summation for ionic systems. But If I include that, my system (2 spherical particles kept at some distance apart) are moving further apart when I heat the system to 500K. How to overcome these?

Kindly clarify my doubts.

Thanks in advance.

Hi all,
         I am using lammps for simulating an ionic system (a complex
system of perovskite structure) for thermal treatment using morse (for
short range interactions) and coulomb potential (for long-range). I have
two questions

​please note, that your questions are not really LAMMPS questions, but
general questions about force fields and MD ​simulations. those are issues
that you should discuss with your adviser and experienced colleagues as
well as consult text books and relevant publications about. a mailing list
makes a poor replacement for competent advising.

1) Is that the right combination of potential to use?

​this is difficult​ answer because it is not a good question. there is no
such thing as "the right potential". there can only be "good enough for a
specific purpose". empirical potentials and numerical simulations are by
their very nature approximate and thus one has to validate one's choices.
that is why people prepare and publish potential parameters and force
fields for specific purposes and document their details as well as how well
they are applicable to specific purposes and conditions and what kind of
accuracy you can expect. you cannot just arbitrarily add terms to those
parameterized interactions.
in short: you should have located the specific parameters in some paper and
that very paper should tell you exactly what LAMMPS styles you need to use
and how "right" (= accurate under certain conditions) it is.

2) And do we need to include ewald term (kspace_style) for including the
long term interactions. I have read somewhere we need to include ewald
summation for ionic systems. But If I include that, my system (2 spherical
particles kept at some distance apart) are moving further apart when I heat
the system to 500K. How to overcome these?

​the simulation will follow the physics of the model you feed it. as a
computer program, LAMMPS is extremely unintelligent and will simply do what
you ask it. when unexpected things happen, you need to stop and try to
understand what is causing this rather than worrying about how to stop the
unexpected behavior.
after all, you are doing research here and are not trying to create an
animated movie, right?

​as before, this is not a suitable topic for a mailing list, but an
indication that you need to discuss in person with people competent in this
issue. and the first candidate for that would be your adviser, who is
tasked with providing you with the suitable guidance for your research.​

​axel.​