Diffusion, molecules will flow from high concentration to low

[kbm]

This might be a stupid question, but it has me kind of confused. In diffusion, molecules will flow from high concentration to low concentration, and the explanation typically given is that in the laws of thermodynamics there is a net increase in entropy over time.

But I am trying to think of it from a perspective of physical forces acting on the molecules...

So say there is an aquarium type thing with a mesh screen separating it into two halves, and a bunch of some solute, say salt, is dumped into one side of it. By diffusion, after some time, there will be net movement of the solute molecules from the side they were dumped on into the other side. What I am confused about is, if the ions are in an area of relatively little motion, then to move to a completely different area would require some outside force acting on the ions themselves to propel them to the new area, wouldn't it?

 

[Borek]

No. Ions always move at random. It happens that as effect of these random motions they tend to be uniformly distributed, as such distributions are much more probable.

Go here:

http://www.chembuddy.com/?left=all&right=download

Scroll down and download diffusion program. Ions are always added on top, and just by random motions they reach the bottom.

 

[Kracatoan]

And if you were interested, the random motion is Brownian motion, which, like most science, no-one really understands.

 

[Borek]

And if you were interested, the random motion is Brownian motion, which, like most science, no-one really understands.

Care to elaborate? As far as I know Brownian motion is understood quite well.

 

[Kracatoan]

Because we can only model it in 1D.

 

[Kracatoan]

although said model works fairly well in 3D scenarios.

 

[Borek]

Because we can only model it in 1D.

This is simply not true, there is problem with 3d models.