Crystal dynamics in LAMMPS

Dear LAMMPS community,

I am doing molecular dynamics simulations of fluids, but I am also interested in studying the simulation of crystal dynamics. For example, I want to simulate the vibrations of the atoms or molecules around their equilibrium positions for some crystals. Honestly, I am at the beginning stage of this field and have not yet done any simulations on the crystal. I am a little bit confused about where to find a reference explaining the simulations of crystal dynamics.

  1. Could you guide me to some reference on the crystal dynamics simulations?
  2. Is there any software by which someone can obtain a file containing the the coordinates of the basis atoms or molecules in some crystal ? Like a file ready to be inputted in lammps using the read_data command.

Much thanks
Mohammed

Hi Mohammad,
For your first question, you can check Mark Tschopp LAMMPS tutorials Link.
For you second question you can use Atomsk.
Good luck

I am not asking about LAMMPS tutorials. I have a very good knowledge of using LAMMPS itself. I have been simulating fluids for a while using LAMMPS.

More precisely, there is a book by Allen and Tildesley on computer simulations of liquids. What would be the correspondent one for crystals?

Atomsk does not support creating a molecular system as far as i know. Is there an alternative software for creating molecular crystals?

For question number 2, I think you can maybe use some crystal structure database. There are many. For example: https://www.ccdc.cam.ac.uk/
I once wanted a data file for a specific structure of ZIF-8 and I found it there. I then used a software to convert the cif to lammps data file. You can build the connectivity and even propagate the system using in-house codes or lammps commands if you wish.

For question number 1, what do you want to know exactly?

I would not so easily dismiss well intended advice.

But not so much about crystal systems, so studying a tutorial that gives you an introduction in how to create crystal geometries is not such a bad suggestion.

This is off-topic for a forum about LAMMPS. In general, most of the “dynamics” are the same in solids and liquids. You will probably be better off studying a book about crystallography first.

If you are looking for crystal structures you will often find them stored as atomic systems in CIF files. ATOMSK can read and parse such files and convert them to LAMMPS data files. People in crystallography are not much interested in bonds since those are more a concept of quantum mechanics or empirical studies for molecular compounds. But if you have the information about the atom positions, it is not too difficult to add bonds (and angles/dihedrals) to the topology after the fact.

That said, since you have very little experience in crystallography, it is probably a very good idea to practice with atomic systems in the first place and get some experience with simple systems, before moving to more complex system. Bypassing the steps of gradually building skills and instead just jumping to the final stage is often taking more time and more effort and littered with mistakes and frustrations.