Permanent Magnets: How Energy Is Stored in Magnetic Fields

[jostpuur]

So we can think permanent magnets to be composed of large number of tiny magnetic dipoles? You can hear this in very elementary introductions. I would guess this is some kind of approximation to the spins of the atoms, or something...

There is energy stored in the magnetic field, so if these magnetic dipoles would rearrange themselves so that their magnetic fields would cancel, the system would reach lower potential. Why does this not happen with permanent magnets?

 

[marcusl]

This does happen in a newly prepared sample of magnetic material, or in one that has been demagnetized ("degaussed"). However the microscopic view is complicated. Atomic moments are not just randomly distributed; instead it is energetically favorable to form small pockets or domains that have the moments all aligned inside. It is the directions and volumes of the domains that are randomized and, on average, sum to zero. When a ferromagnet is magnetized, domains along the magnetization direction grow at the expense of those pointing in other directions.

Having described that, we can address your question. There is an energy cost to rotate the atomic moments or, equivalently, grow the domains. In essence there is friction. This is reflected macroscopically as the energy lost in traversing the B-H hysteresis loop.

 

[jostpuur]

How big these "small magnetic dipoles" actually are? For example, I have difficulty believing that individual atomic nucleus in the lattice would suffer any friction if they rearranged their spin orientation. Usually only macroscopic objects experience friction. This "friction" is probably something very different from mechanical friction. What is the physical mechanism behind it?

 

[marcusl]

The energy dissipated in the hysteresis loop shows up as heat, prompting the analogy to friction. The actual mechanisms involve coupling of magnetic moments to vibrational modes, effects of crystal imperfections and impurities, magnetostriction (where changes in magnetism cause mechanical strains in the crystal), and others.