[lammps-users] Question about dipole

LAMMPS LAMMPS Mailing List Mirror
1

Dear Steve,

I am a Lammps Code user. I found a serious problem about Lammps. I am simulationg a carbon nanotube with water molecules inside. I use Airebo potential for carbon atoms and SPC model for water molecules. And I set up a dipole moment for each carbon atom of carbon tube. The directions of dipole moments are radial pointed to the axis of carbon tube. The interaction between carbon atoms with dipole moments and water molecules with charges is described by pair_style dipole/cut command. Changing only the magnitude of dipole moments of carbon atoms (all other parameters are same), I surprisingly found that the force of each atom is not changed. That is to say, I obtained same results with different dipole moment of carbon atom. I have also done many tests and obtained same results. I think maybe this is a bug of Lammps. I hope you can check this problem. I also attach my input script and data file with small computational expense. I am looking forward to your reply. If you need more information, please contact me.

Thank you very much.

JS

in.ble (1.65 KB)

data.ble (128 KB)

2

The SPC model does not have point dipoles on water molecules. They
are already dipoles due to point charge separation. If you want to
use pair_style dipole/cut then you should debug it without all the
other complications in your input script. E.g. use only that potential
and dump out the forces and see if they are what you expect. Note that
you are not using the torques produced by dipole/dipole interactions, as
you are using them on point particles and you are not integrating their
rotation with something like fix nve/sphere. So I think you have an odd
model.

Steve