Simulate an equilibrium reaction

Dear all,

I am now considering an equilibrium reaction in classical MD simulation. I just noticed that there is a fix bond/react which can model a chemical reaction, but it seems to me that the reaction there can only be happening in a single direction. Does the current Lammps support the MD simulation of an equilibrium reaction?

Best,
Jiaxing

Dear all,

I am now considering an equilibrium reaction in classical MD simulation. I just noticed that there is a fix bond/react which can model a chemical reaction, but it seems to me that the reaction there can only be happening in a single direction. Does the current Lammps support the MD simulation of an equilibrium reaction?

jake (cc'd) is the author of fix bond/react and actively supports it,
so he is the best person to comment on its capabilities and
limitations.

beyond that LAMMPS does support *multiple* reactive force fields, most
prominently ReaxFF, for which a wide gamut of parameter sets for
diverse systems and thermodynamic conditions exist.

besides, what do you mean by "equilibrium reaction"?

axel.

Dear all,

Thank you. When saying equilibrium reaction, I refer to a chemical reaction which can happen both in the forward or backward direction, for example association/disassociation of titration groups.

Best,
Jiaxing

Hi Jiaxing,

You can define reverse reactions the same way you define forward reactions. For simple reaction steps, you can simply switch the pre- and post-reaction templates (see bond/react doc page).

As far as I know, you would be among the first to use bond/react in such a capacity; you should be sure to test the sensivity of your system to parameters such as Rmin/Rmax (distance to trigger the reaction), and the ‘prob’ keyword (probability of an eligible reaction actually occurring).

Jake