Re: [ESPResSo-users] LANGEVIN_PER_PARTICLE

[Lorenzo Isella]

Hello,
I implemented by hand a similar feature (by some modification of the  
expresso code) some years ago for a paper.
I modelled the dynamics of combustion-generated nanoparticles (larger than  
the air molecules), consisting of several spherules bound together by a  
short-ranged potential.
The Langevin noise was used to model the gas-to-spherule collisions, which  
is also the mechanism giving rise to the friction.
Now, the inner spherules, deep inside the aggregate, are more difficult to  
reach from the air molecules (hence they are shielded) than those on the  
surface and as a consequence they experience a different friction.
Have a look at

http://dx.doi.org/10.1016/j.jcis.2011.01.072

Cheers

Lorenzo


On Wed, 07 May 2014 09:27:15 +0200, Marcello Sega  
<address@hidden> wrote:

> Hi guys!
>
> For the record: it does indeed make sense to have different frictions.
> Think about a simple electrolyte. Anions and cations have different
> effective friction coefficients because of two effects, namely 1) the
> Stokes friction itself (ions and cations have different sizes) and 2)
> the different way they interact with water. This translates to
> different specific conductivities of the two type of ions, and in a
> coarse-grained simulation setting a separate friction coefficient is a
> straightforward way to model this (combined with using different
> particle masses).
>
>
> M.
>
>
>
> On Wed, May 7, 2014 at 9:19 AM, Tristan Bereau <address@hidden>  
> wrote:
> > Dear Olaf and Stefan,
> >
> > Thanks for the quick replies.
> >
> > The meaning is indeed very clear--I wasn't pointing at any lack of
> > documentation. I was merely curious as to whether any scientific study
> > had used such a scheme. I can actually think of several applications
> > where this could be of interest.
> >
> > Best,
> > Tristan
> >
> > On Wed, May 7, 2014 at 9:01 AM, Olaf Lenz <address@hidden>  
> > wrote:
> > > Hi!
> > >
> > > I admit that the docs are very compact, but on the other hand the  
> > > meaning is
> > > relatively clear, isn't it? However, it would be a good idea to also  
> > > hint to
> > > the feature in the Langevin description.
> > >
> > > I am not aware of any real physical application. We use it mainly for
> > > demonstration purposes, see Stefan's mailing.
> > >
> > > Olaf
> > >
> > >
> > >
> > > 2014-05-07 8:55 GMT+02:00 Tristan Bereau <address@hidden>:
> > >
> > > > Dear all,
> > > >
> > > > I've just stumbled across this wonderful LANGEVIN_PER_PARTICLE feature
> > > > in ESPResSo, and was wondering if anyone could point me to some
> > > > relevant literature, especially of applications. I couldn't find any
> > > > reference from the user manual.
> > > >
> > > > Many thanks.
> > > >
> > > > Best,
> > > > Tristan
> > >
> > >
> > >
> > > --
> > > Dr. rer. nat. Olaf Lenz
> > > Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart
> > > Phone: +49-711-685-63607