Time integration scheme when using "fix brownian"

Hi all,

I am using the “fix brownian” style to perform simulations of colloidal particles. According to the documentation, activating this fix will make LAMMPS implement the Euler-Maruyama time integration scheme. However, during the final setup stage of the simulation, the message “Setting up Verlet run ...” appears on the terminal. Of course, using Verlet time integration is undesirable in this case. I wonder if anyone knows whether or not this terminal message means that I am actually running my simulations with a Verlet scheme instead of the Euler-Maruyama one.

Thank you,
Christos

That message is referring to the run_style command — LAMMPS documentation

In LAMMPS you only have the choice between a Velocity Verlet or r-RESPA timestepping (which will collapse to a Velocity Verlet in case there is only one level). So it will show either Verlet or r-RESPA in the run setup message. This determines the flow of control as outlined here: 4.6. How a timestep works — LAMMPS documentation

You can have other integration schemes, but they must be implemented within the framework that LAMMPS offers.