Re: [ESPResSo-users] Velocity in nanochannel

[Axel Arnold]

Hi!

LB walls introduce no slip boundary conditions, so that you should get a 
Poiseuille flow profile. If you see acceleration near the walls, your 
constraints apply only to the particles. Maybe,  you didn't add corresponding 
LB walls, or you put them too close to the periodic boundaries, so that no 
nodes can be marked as boundary. In both cases, David's argument holds and you 
get acceleration.

Axel 

> On 05.01.2015, at 07:27, David Schwörer <address@hidden> wrote:
> 
> Hi,
> 
> The constraint is a 'flat' surface.
> The direction of the interaction has only a component orthogonal to the
> wall - because this is what a lj does. therefore if the particle moves
> along the wall, it only feels a force preventing it from entering the wall.
> There is no friction between wall and particle.
> 
> Due to the increased orderering close to the wall, the particles in
> front of the wall collide less often.
> Therefore the particles on the wall are faster.
> 
> I think with transversal dpd you can add some friction, or by fixing
> particles on the surface.
> 
> Best
> David
> 
>> On 01/04/2015 11:10 PM, roya moghaddasi wrote:
>>> Hello Espresso users;
>>> 
>>> I have a nanochannel in my simulation, I also have lattice boltzmann
>>> inside the nanochannel, the problem is that when I calculate the
>>> velocity near the nanochannel, it is not close to zero, instead it is
>>> greater than the center of the channel, which is, in my view
>>> unrealistic. The case is the same when I eliminate LB and use langevin
>>> dynamic.
>>> 
>>> Can anyone help me to solve this problem?
>>> 
>>> Best,
>>> Roya
>> 
>> 
>> 
> 
>> 
>> 
> 

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