Is it possible to know if there are any formation of bonds after running a simulation? I used “atom_style: charge” for Silicon and Hydrogen. In OVITO software it seems they’re creating bonds after a certain time steps, but I want to know if there are any numerical ways to confirm the bonds.
While it may be very deceptive, but what most visualization tools show is not necessarily representative of “bond formation”. The usually employ some simple distance criterion and between any pair of atoms, that that are closer than that distance, a bond is drawn.
Most classical potentials require bonds to be entered explicitly via a bond topology. There the bond is an input parameter (and the visualization tool may or may not show it, depending on whether you provide the topology information) as well as its (average) length. A Lennard-Jones potential has the same strength in every direction so that is most certainly not forming a bond.
Simple manybody potentials like Stillinger-Weber or Tersoff may promote some preferred angles between groups of three atoms or augment the equilibrium distance of a bond depending on how close a third atom is.
You have to go to rather complex (reactive) potentials like AIREBO or ReaxFF which compute a “bond order” or “bond strength” and then have a dynamic bond topology. For ReaxFF this bond order and a list of bonds where the order is larger than a given cutoff can be output with a fix (and from that information grouping into molecular species computed).
Visualization tools do not use that information.
As should be clear from the explanations above, the answer depends strongly on the model you are using in LAMMPS and whether that bond order information, if computed, is made accessible. As far as I can recall, this is currently only possible for ReaxFF, but I may be wrong, so you need to study the documentation about (and associated publications) the model in your calculations.