Is the Spin of Electrons Aligned Opposite to Net Magnetization in Ferromagnets?

[eintagsfliege]

Hi everybody

The bands of up- and down-spins are split in the bandstructur of a ferromagnet. Does the part with the higher amount result in a net magnetization?
If yes, doesn't that mean, that electrons passing a ferromagnet, they align their spin in the opposite direction compared to the magnetization? Since the minority band is at the fermi-level, the spins should be in opposite direction to the net magnetization.

Thanks for any help...
[URL]http://www.sciencemag.org/cgi/content/full/282/5394/1660/F1[/URL]

 

[kanato]

Yes to your first question: you add up the total number of electrons for both spins and the difference gives you the net magnetization.

No to your second question: a ferromagnet usually has a Fermi surface for both spins, so it can support conduction for either. Conduction is proportionate to the density of states at the Fermi level, which can be larger of smaller in either spin channel. There are materials called half-metals which have a gap in one spin channel and not the other; these materials will only support conduction of one type of spin (usually in the majority channel I think).

 

[eintagsfliege]

hm...how comes then, that the electrons passing a uniform magnetization tend to align its spin to the magnetization?

 

[kanato]

The presence of a magnetic field will change the Hamiltonian and thus the energy levels. Roughly speaking, this will add a potential energy term mu*B to the Hamiltonian and split the energy levels for spin up and down electrons, so that spins aligned with the field become more populated.

 

[eintagsfliege]

But even when no magnetic field is applied, the magnetization polarizes the charger carriers passing by.
I don't understand it, since it seems that the minority band has a higher density of states at the fermi level.

 

[kanato]

Huh? I don't understand your question. You said "no magnetic field is applied, the magnetization polarizes the charger carriers passing by..." charge carries are passing by what?

The spin minority/majority has nothing to do with the density of states at the Fermi level, it's about the integrated density of states up to the Fermi level.