Interesting, but the Pop Sci article is typical in missing the mark while attempting to appeal to readers. In the writer's defense, he did link to a much better article here:
Graphene Under Strain Creates Gigantic Pseudo-Magnetic Fields « Berkeley Lab News Center
Unpaired electrons, and the alignment of one their spins (similar to the spin of the Earth on its axis as it rotates around the sun) are considered to be the source of magnetic fields. The electrons, being unpaired, wander in a cloud. Get a few of them spinning with similar orientation of spins while wandering, and a sensible magnetism appears.
This article shows a similar effect can be observed within a tightly bound area of a stressed lattice, with only a minimal number of electrons involved.
I've always been fascinated by magnetism, and repeatedly drawn back to wondering about it (sorry about the pun). I think I finally have an idea of what is going on with it. The classical idea is that unpaired electron spin (as in how the Earth spins on its axis as opposed to how is spins around the sun) is the cause. If the spins line up, then there is obvious magnetism, if they don't, the spins cancel each other and there is no apparent magnetism.
Next, consider that magnetic "lines of force" can be modified and channeled, exhibit resiliency and the materials only exhibit magnetism up to a certain temperature, at which point the alignment falls apart and magnetism disappears.
We know that magnetism is not a ray or packet of energy like a photon, since we cannot "project" magnetism by itself. We know that it doesn't quite follow the inverse square law, since the forces in a long bar magnet only sorta follow the bar and inverse square law. We know that magnetic materials exhibit some crystal-like properties, in requiring alignment, ceasing to exhibit at certain temperatures, etc.
One way this all makes sense is that the magnetic material does NOT only physically exist as a bar of iron, but that each atom of that bar of iron is actually a cloud of sub-atomic particles that has a central locus within the bar, but is present in some form outside of what we perceive as the bar. The mass of the particles that are present outside of the bar is either non-existent or minuscule, but then electrons have little mass.
The electron cloud would have a periodicity based on harmonics of the speed and shape of rotation around the nucleus. The resiliency exhibited is the exact same centripetal momentum exhibited by a gyroscope, only on an electron level. Since electrons have far fewer collisions with air molecules than gyroscopic toys, the decay rate of the momentum into entropic movement is slow, but IIRC, magnets DO lose their properties over time and have to be "recharged."
What becomes even more interesting is the idea that all materials likely have an electron cloud, and when those atoms and electrons are in alignment, they exhibit more complex properties, intrinsic to the alignment and frequency of the harmonics.
That could present answers as to WHY homeopathy can work, even in double blind studies, little nagging WTF experiences, like why I have different reactions to just being around quantities of various metals, and a host of other effects.