Water diffusion in MOFs (verbosity level of output file)

Dear Axel,
Many thanks for your reply.

2- In the attached input file, I have defined cross Lennard-Jones potentials explicitly. They are defined between Oxygen and Hydrogen atoms of TIP4P water molecules with 6 other atom types belonging to the adsorbent framework (i.e. MOF). At this stage, I am unsure whether I need to define a similar set of pairwise interactions for coulombic potentials (something similar to lines 68 and 69 of the input file which are currently commented out) or the kspace_style takes care of all the cross coulombic interactions between water and framework atoms?

pair style lj/cut/tip4p/long already contains coulomb interactions (with special treatment of the location of the point M for TIP4P water molecules, which is computed on the fly while the Oxygen atom is skipped as it has no charge). this is the short range part. the long range part is computed in pppm/tip4p

Right! But how does LAMMPS know for which pair of atoms it should calculate Coulombic potential, if we don’t define them explicitly? Does lj/cut/tip4p/long automatically include Coulombic interactions of water with any other atom type available in the system? Or we should somehow define pairwise interaction for Coulombic potential as we do it for LJ?

Thank you in advance for your clarification.

Best regards,

Amir

Dear Axel,
Many thanks for your reply.

2- In the attached input file, I have defined cross Lennard-Jones potentials explicitly. They are defined between Oxygen and Hydrogen atoms of TIP4P water molecules with 6 other atom types belonging to the adsorbent framework (i.e. MOF). At this stage, I am unsure whether I need to define a similar set of pairwise interactions for coulombic potentials (something similar to lines 68 and 69 of the input file which are currently commented out) or the kspace_style takes care of all the cross coulombic interactions between water and framework atoms?

pair style lj/cut/tip4p/long already contains coulomb interactions (with special treatment of the location of the point M for TIP4P water molecules, which is computed on the fly while the Oxygen atom is skipped as it has no charge). this is the short range part. the long range part is computed in pppm/tip4p

Right! But how does LAMMPS know for which pair of atoms it should calculate Coulombic potential, if we don’t define them explicitly? Does lj/cut/tip4p/long automatically include Coulombic interactions of water with any other atom type available in the system? Or we should somehow define pairwise interaction for Coulombic potential as we do it for LJ?

coulomb interactions, especially when using long-range electrostatics, are always computed between all atoms and the “parameters” are implicit due to having to assign (partial) charges to all atoms (or set them to zero). as mentioned in my previous response, that computation is already included in the pair style.

axel.