I've been reading through my son's physics textbook (junior high) and got puzzled by the chapter on magnetism. I do understand that this is addressing young teens but still, I find the logic presented somewhat confusing. It goes like this:

  1. Considering rod-shaped magnets, they have a North pole and a South pole. If you put two such magnets together, opposite poles attract and when they snap together, they basically form a longer rod-shaped magnet.
  2. Thus we might consider doing the opposite: cutting or breaking a rod-shaped magnet in half, which gives us two smaller rod-shaped magnets that, again, have a North and a South pole.
  3. If we imagine that we keep on doing that (cutting a magnet in two over and over again), eventually we'll end up with some smallest building blocks which we call "elementary magnets".
  4. Even an iron nail consists of such "elementary magnets" on the inside but the difference between such a nail and a magnet is that the elementary magnets are all oriented identically inside the magnet while they are randomly oriented in the nail.
  5. If you rub a magnet over a nail, your aligning some of the latter's elementary magnets so that it also (temporarily) behaves like a magnet.

I find this model leaves to be desired.

For one, this whole idea of point 3. seems questionable to a lay person like me: what exactly is meant by "smallest building blocks"? Atoms? Or is this purely an abstract thought?

But more importantly, point 5 seems to suggest that these "elementary magnets" can either move or rotate - so how come they don't auto-align themselves all the time, given that they and their surroundings (i.e., other elementary magnets) are magnetic? As in, if there was one elementary magnet located next to another one but oriented in the opposite direction, shouldn't their poles make it so that they want to rotate away from each other in order to face the same direction?

Even if they just happened to be in some kind of attraction/repulsion equilibrium that's different from all elementary magnets being oriented the same way, why would a temporarily magnetized nail ever lose its magnetic property again?

  • $\begingroup$ (3) The "elementary magnets" are just electrons, either from their spin or their orbital angular momentum. (5) In most materials, the interaction between the electrons just isn't strong enough. Ferromagnets have a special interaction that aligns the "elementary magnets", which is stronger than the interaction in most materials. $\endgroup$
    – knzhou
    Sep 13, 2023 at 22:56
  • $\begingroup$ You might find this useful - MAGNETS: How Do They Work? $\endgroup$
    – mmesser314
    Sep 14, 2023 at 1:17
  • $\begingroup$ I reckon that niels nielsens answer explains well the situation for ferromagnets. Also, I believe that the authors of the book tried to explain a general concept, while avoiding to oversimplify magnetism. After all, magnetism is a complicated phenomenon. E.g. in antiferromagnets the magnetic moments of neighbouring atoms are anti-parallel. Thus, antiferromagnets is not explainable, if we only use magnetic interactions between the atoms. The magnetic interaction is one component, but the full model possesses multiple components. $\endgroup$
    – Semoi
    Sep 14, 2023 at 20:12
  • $\begingroup$ For a quantitative but simpleified approach you could look into the Heisenberg model to which you apply an external field. $\endgroup$
    – LPZ
    Sep 15, 2023 at 14:59

1 Answer 1


Here are the answers.

the elemental building block of a large magnet is indeed the individual iron atom, which is itself a magnet.

Note that within a chunk of iron, there exist tiny domains in which all the nearest neighbor atoms are all aligned in the same direction- but where the orientation of the domains is random, so all the magnetic fields of the domains cancel each other out.

When you are magnetizing a chunk of iron, you are actually turning the domain alignments so they are all pointing in the same direction. Now note that when a domain is in any alignment, it wants to remain in that particular alignment, which means it takes work to change the domain alignments.

How to minimize that work? There is a temperature above which the work needed to change the alignment is furnished by the random vibration of the iron atoms in their atomic lattice positions. Above this temperature, the magnetism of the sample melts away to nothing. So to make a nice strong magnet, you can heat the chunk of iron above that critical temperature, bring another magnet close by, and then cool the chunk down below that temperature. The aligned domains are thereby locked into position and you have a strong magnet.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.