Re: [ESPResSo-users] suitable r_cut for DPD

[Narges Nikoofard]

Dear all,

I wonder if I can ask a more general question about r_cut for dpd, if possible.

In the user-guide, it is written: "When using a Lennard-Jones interaction, r_cut = 2^(1/6) σ is a good value to choose, so that the thermostat acts on the relative velocities between nearest neighbor particles. Larger cutoffs including next nearest neighbors or even more are unphysical."

My questions are:
(1) Does the suggested cut-off depends on the cut-off radius of the Lennard-Jones potential? In other words, the Lennard-Jones potential should be truncated at the minimum?

(2) Why larger cut-offs are unphysical? Does this result from the physical imagination that only neighboring spheres have friction? Or, the shape of potentials are important in this reasoning?

Please apologize me for interrupting again.

Many thanks in advance,

Narges Nikoofard

On Mon, Jan 25, 2016 at 3:30 PM, Narges Nikoofard <address@hidden> wrote:

Dear all,

I am doing a coarse-grained simulation on the self-assembly of specific peptides in explicit solvent.

I have used an existing model for coarse-graining amino acids and to find the Lennard Jones parameters for the CG beads. In this model, the cut-off radius for the LJ potential is defined to be 15 angstroms (four times the smallest bead).

Considering that in the simulations, the cut-off radius for the LJ is larger than 2^(1/6)*sigma, is it still suggested to use r_cut = 2^(1/6)*sigma for DPD?

I would be grateful if someone could kindly help me in this issue.

Many thanks in advance,
Narges Nikoofard