How to set attractive force between dispersed solid particle or atoms and frozen atoms in the fixed wall for dpd simulation

Dear all,

I am using dpd method to simulate dispersed solid particles transport and deposition on the wall in gas domain with a 2D channel. The fluid is gas and the dispersed phase is solid particle.

Now I have couple of questions about the simulation:

  1. Can I use dpd method to simulate gas flow and treat gas as a series of dpd particles? I saw majority of scholars used dpd method to simulate liquid dpd particles and simulate dispersed solid particles transport in liquid dpd domain. But I hardly saw some researches treat dpd particles as gas dpd particles.

  2. How can I set attractive force between dispersed solid particles and frozen wall atoms in the fixed wall, the attractive force can make dispersed solid particles deposit on the specific wall? I saw some commands including fix wall/lj93, fix wall/harmonic and others. Which one can be used to set attractive force? If none of them can not be used to set attractive force, how can I set the attractive force?

  3. Is it necessary to use “fix rigid” for dispersed solid particles if I only care about dispersed solid particles deposition on the specific wall instead of impaction among solid particles?

Can anyone give me advice to solve these problems?

Really appreciate any suggestions.

Best Regards,

Michael Zhao

Most of these questions are not within the scope of the LAMMPS mailing list, and no one is going to tell you how to do your research. Deposition of attractive (order Micron particles) in a Gas-solid flow is challenging for a number of reasons and until you understand why, no advice about what models have attraction are going to be of any aid. If you don’t understand the limitations imposed by different time-scales in this problem (set by the attractive well for you particles/liquid bridge, Reynolds numbers on DPD, contact time/stiffness for granules-depends on adhesion model e.g. Derjaguin-Muller-Toporov or Johnson-Kendall-Roberts, ballistic time-scales set by characteristic velocities of the granules and the solids volume fraction), you will become adept at producing digital garbage.

Please speak with your advisor or read the literature about this. You will find you have a lot of thinking to do.

Dear Eric,

Thank you for your nice advices. I understand what you have said.

Actually I want to simplify my problem and I want to use a series of dpd atoms to simulate solid particle, only different atom type from dpd fluid atoms. I only care about solid dpd atoms deposition on the wall, not caring about atom impact. So I just want to set the attractive between the flow solid dpd atoms and frozen wall atoms in the fixed wall. Which kind of command can be used to set the attractive between atoms? Then I can use to experiment results to adjust the parameters for the attractive until the simulation results agree with the experiment results. SO would you like to tell me which command, something like pair_sytle or fix command, can do that in LAMMPS?

Appreciate your reply.


Michael Zhao

If you solid and wall particles are different types, then
you can use pair_style hybrid to set different pair styles
for different I,J interactions. DPD is not attractive, but
you could use a simple LJ potential with a cutoff
at 2.5 sigma. The bigger epsilon is, the deeper the
well that will trap a solid particle when it moves close
to a wall particle.