Re: [ESPResSo] viscosity

[Axel Arnold]

On Thursday 01 April 2010 17:51:07 Guillaume Larsen wrote:
> mea culpa...
>
> About the problems i mentioned below, i have just found the answer. The
> problem came from the omission of bond type when defining the polymers. So,
> fortunately the viscosity is positive.
> However, concerning the second problem (increasing the number of particles
> and polymers), i still have problems (but not the same). ESPResSo returns
> that he 'failed to place the polymer for n times'. Modifying the values of
> polymers bond length or the shield has no effect on the result, still the
> same problem.
>
> Somebody has a clue?

Hi,

the polymer command tries to place polymers without overlap (at least if you 
ask for a  SAW). So, if you pack too many into a box, it will very quickly 
take extremely long until a polymer can be placed, and if it takes too long, 
Espresso gives the "failed to place" error. You could try to increase the 
number of maximal tries per polymer, but since the fail probability grows 
exponentially with the number of beads, this won't get you very far.

One way out of this is normally to simply set up a melt of random walks, and 
then gradually turn on the LJ-interaction using LJ force capping, so that the 
chains eventually find their own way to distribute as a SAW.

Cheers,
Axel

-- 
Dr. Axel Arnold Tel: +49 711 685 63593
ICP, Universität Stuttgart      Email: address@hidden
Pfaffenwaldring 27
70569 Stuttgart, Germany