Re: [ESPResSo-users] weird force between charged rods

[Axel Arnold]

On 21.03.14 09:43, Fabien Paillusson wrote:

By the way, I was thinking about Axel's answer about adding by hand the rod-rod force. That is a priori fine with me but wouldn't I need to take into account the force owing to the periodic images of the rods? And then, I think it stops to be simple...

Technically not, since the constraint is also not periodically replicated. Espresso only adds the constraint force for the primary box, no periodic images of constraints. It would be demanding to compute the particle-rod interaction for an infinite array. Probably one can deduce a formula, but we don't have that yet. Also, the rod was in fact meant to be used with MMM1D, which doesn't have periodic boundaries perpendicular to the rod.

Axel

2014-03-20 19:22 GMT+01:00 Stefan Kesselheim <address@hidden>:

Dear Fabien,
I'm not sure if force the rod constraint was checked recently. It might well be possible that it is counted, for example, with the wrong sign. I believe I wrote the constraint force stuff for repulsive surfaces, but I'm not very sure about the forces of the rod constraint. Instead of the rod constraint I have always been using a line of particles. This could be an easy check.
On the other hand: What are your system parameters? If you have DNA and trivalent ions at room temperature, like-charged attraction is, for example, expected.
Cheers
Stefan

On Mar 20, 2014, at 5:56 PM, Fabien Paillusson <address@hidden> wrote:

> Hi,
>
> I am simulating an apparently simple system with two infinite charged and similar  rods (as constraints) embedded in two cylinders (as constraints) neutralized by counterions in a 3d periodic box.
> When I run this system with P3M and look at density maps for instance, they look totally fine and fit with my intuition of the physics going on in there.
> However, when I try to compute the force on the rods (with the force constraint command), I always get an attractive result while it should be repulsive!
>
> I have tried to change many things including accuracy of the P3M method, size of the box to prevent periodic images to interact with one another or rod charge density so that I get below the Manning condensation threshold and then I have naked charged rods interacting within a uniform background of opposite charges and I always get the same attractive force whatever I do.
>
> Any idea on what could be going on?
>
> Thank you very much for any kind of help you could provide.
>
> Best,
>
> Fabien Paillusson
>
> Departament de fisica fonamental
> Universitat de Barcelona
> Marti i Franques, planta 3 i 4,
> ES-08028, Barcelona
> Personal webpage

-- 
JP Dr. Axel Arnold
ICP, Universität Stuttgart
Allmandring 3
70569 Stuttgart, Germany
Email: address@hidden
Tel: +49 711 685 67609