Respected,
This is my first post in this forum.
I have been trying to simulate a slag-like model to understand the viscosity of Ca2O-Al2O3-Na2O, but the problem lies with the fact that there are no resources to find the value of the interatomic parameters of Ca-Na for Buck potential calculation, now doing it using LJ potential but still require accurate values.
Also dont wish to divert to DFT calculations.
Unfortunately, you are posting to a forum where we discuss the LAMMPS simulation software, but not how to get potential parameters. They are merely input data.
The canonical place to look for existing potential parameters is the published literature. If you can’t find anything there, you may need to learn how to parameterize such potentials from the publications that provide the parameters you are already using.
Please also note that in classical models, there are not a single set of parameters for a given element, but a single set only applies for certain compounds or local environment (e.g. there are 10s of parameter sets for carbon atoms in most biologically relevant force fields), and on top of that there is only a limited range of temperature and pressure/density where those parameters are suitable.
If you want accurate results these days, the most popular approach is to train a machine learning potential.