rheology with DPD

LAMMPS LAMMPS Mailing List Mirror
1

Hi Axel,

Thanks for the reply. I already did some troubleshooting actually and the method produces reasonable results. I was just curious if anyone else has gone this route. My test cases included shearing a periodic domain with water (phase lag between stress and strain for this case should be ~90 degrees). I also measured the viscosity and it gives the same values as the viscosity I got using different methods (Hagen-Poiseuille flow and Couette flow). Interestingly, if I use just fix deform without sllod then I did not get as good of an agreement for the phase lag, but that may just be from the particular setup I was using. For fix deform I do remap the velocities, I forgot to mention that earlier.

2

The DPD thermostat works on the relative velocities
of the pair of atoms, so should be fine if the system
is flowing. However I don’t think you want to use
two thermostats. So you could skip nvt/sllod and check
if the system is flowing with a velocity profile
that matches the deformation rate. It might take a while
to get there. If not, you could use nvt/sllod, but I don’t
think you should enable the DPD thermostat as well.
I.e. not pair dpd/tstat.

Someone with more experience with NEMD for DPD
may have different suggestions.

Steve