Can I ask one general question about how to choose the potential for the molecular dynamics simulation with LAMMPS?

Suppose one wants to simulate the reaction between two different materials/systems and there many different potentials to describe the interaction between these two systems. Different potentials give completely opposite simulation results. One set of potential makes two systems react; while, the other set of potential makes two systems intact with each other.

The point is that one does not know what these two systems react in reality and wants to know what would happen in reality through the MD simulation. There is no reference discussing the reaction between these two systems in reality as well.

In such case, how should one choose the potential for the simulation? Is there any criterion to choose the potential?

Thank you in advance.


  • $\begingroup$ If experimental view are unavailable, first principle results should be considered as next reference. Nowadays, one can easily benchmark potential against small dataset $\endgroup$ Commented May 6 at 9:29


You must log in to answer this question.

Browse other questions tagged .