🚀 go-pugleaf

>columbus@strident.think.com (Michael Weiss)
>Not a compelling mystery, but I was wondering what the current status is of
>explanations of ferromagnetism from "first principles". I remember reading
>in Feynmann that Ising-model type calculations at the time were giving the
>wrong sign for iron.  However, that was nearly 30 years ago.  What's the
>current scoop?

The ferromagnetism of Fe is mostly caused by the conduction electrons,
not by isolated ions, so the Ising model is not appropriate.  The
failure of the Ising model does not mean that more appropriate models
don't exist.  There is still active research to quantitatively model
the ferromagnetism of the transition metals because they are a hard
case for band structure theories (which is also in part why HiTc
theories are hard).  The reason for this difficulty is the large part
that many-body interactions play.

The $0.05 answer: The fundamental underlying cause of conduction
electron "itinerant" ferromagnetism is the Pauli exclusion principle.
In some sense the exclusion principle causes a repulsive force between
electrons of the same energy and spin.  (This is really an indirect
manifestation of the Coulomb repulsion between like charges.)  The
converse of the preceding statement is that if you have a low-density
electron gas, there is an effective net force aligning the spins.
Thus, a large-lattice-constant metal will have a tendency towards
ferromagnetism.  This is especially true in a degenerate system
because the high density of states at the Fermi level means a big
energetic win for opening up an energy gap.  3d transition metals have
large lattice constants and flat, degenerate d bands => ferromagnets.

-- 
Alison Chaiken 				alison@wsrcc.com
(510) 422-7129 [daytime] 		or chaiken@cmsgee.llnl.gov
Look if you like, but you will have to leap.