How Does Spin-Dependent Chemical Potential Generate a Spin Current?

[amjad-sh]

Hello
In a paper I am working we have a two metal connected to each other,
Within the ballistic approximation we adopt the Landauer approach and generate a pure spin current by
applying a spin bias that is modeled as difference of spin-dependent chemical potentials across the interface.
Can somebody who has knowledge about spintronics explain to me the how difference of spin dependent chemical potential across the interface can generate a spin current?
THANKS

This is the paper I'm working on

 

[Telemachus]

I don't know nothing about spintronics, but the chemical potential comes from thermodynamics, and as every potential, it says basically that at equilibrium, you would minimize this potential, and if you have a species with a higher chemical potential, for example, two different gases separated by a wall, then, if you remove the barrier, you would have a flow, and a mix of particles in order to minimize the chemical potential. For spins, I suppose it is analogous. I imagine, that, for example you have some ferromagnetic material, so the domains interact, and the spins points up or down, in order to minimize some sort of potential. But its only speculation (I haven't read the attachment neither).