I was talking with a friend about a design for a perpetual motion machine that used magnets, but he said that magnets eventually lose their magnetism. I don't want anyone to go into how Thermodynamics disproves perpetual motion, I just want someone to answer this question: Do magnets ever lose their magnetism, and if so how long does it take?


This is an interesting question. It would make some sense that a magnet would operate like a battery and eventually die, but it doesn't work like that.

Colloquially speaking, a magnet has its characteristic properties because of a certain alignment of its atoms.

Hence speaking generally about permanent magnets, the way it works is that the magnet loses its power is when the atoms come out of alignment.

There are various ways the atoms can be forced to disturb their alignment and hence leading to the magnets losing their magnetism such as via heating them or giving it a jolt or shock or by putting it in a demagnetizing magnetic field etc.

For temporary magnets, well, it depends on process to process and they can have varied life times.

Hope it helps!

P.S I tried to keep this answer as simple as possible.

  • $\begingroup$ Thank you ritvik1512. I understood your answer with no trouble at all. It answered my question and was simple. $\endgroup$ – DragonSlayer3 Nov 5 '15 at 0:25

The answer depends on the magnet. A temporary magnet can lose its magnetization in less than 1 hour. Neodymium magnets lose less than 1% of their strength over 10 years. Permanent magnets such as sintered Nd-Fe-B magnets remain magnetized indefinitely.

  • $\begingroup$ Iron magnets need a keeper to complete the circuit and preserve the magnetism. $\endgroup$ – Martin Beckett Nov 4 '15 at 5:00
  • $\begingroup$ What exactly is a 'keeper'? $\endgroup$ – DIYser Nov 4 '15 at 5:32
  • $\begingroup$ keeper wiki "A magnet keeper, or armature is a ferromagnetic bar made from soft iron or steel which is placed across the poles of a permanent magnet to help preserve the strength of the magnet by completing the magnetic circuit; it is particularly useful for magnets which have a low magnetic coercivity, such as Alnico magnets. " $\endgroup$ – anna v Nov 4 '15 at 5:56
  • $\begingroup$ Thanks, anna v. It seems more like a band-aid for an inferior magnet chemistry. $\endgroup$ – DIYser Nov 4 '15 at 14:29
  • $\begingroup$ Aren't "neodymium magnets" and "sintered Nd-Fe-B magnets" the same thing? $\endgroup$ – Thomas May 14 '16 at 4:55

Do magnets ever lose their magnetism, and if so how long does it take?

The answer by DIYser is incomplete. If magnets are not interacting the time table he gives are reasonable. If they interact in any way , they lose magnetization.

One can use a permanent magnet to magnetize iron for example. The energy needed for the ordering of magnetic dipoles in the unmagnetized material is taken from the potential energy of the oriented dipoles, demagnetizing them slowly. So how long it takes to demagnetize a permanent magnet depends on the process that induced demagnetization.

Look at the answer here . Permanent magnets are not really permanent.

As for energy extraction, think of a magnet as similar to an electric battery. One can get useful work from an electric battery, but it finally discharges. Analogously, the lined up magnetic domains in a permanent magnet were either "charged" by the creations of the earth energies, or by using electric currents , i.e. storing potential energy in the magnetic material. One might think an ingenious way of getting kinetic energy from the magnet's potential one, but eventually the domains will become randomized. In the electric battery chemical bonds that were storing energy are destroyed by energy extraction, in the magnet orientation is destroyed.

  • $\begingroup$ There are undoubtedly many things that can reduce the strength of a magnet and even some things that will nullify a magnetic field. I've even seen a video of a guy reverse the magnetic field of a magnet by magnetization. If what you typed is true, how long would a neodymium magnet hang from the underside of an iron beam before losing enough 'energy' that it could not maintain enough of an attraction to remain in contact with the beam (i.e. before it will fall)? I've seen speculations that it would be over 400 yrs--for most practical purposes--heat is the biggest threat to demagnetization. $\endgroup$ – DIYser Nov 4 '15 at 5:39
  • $\begingroup$ I cannot give an estimate, it would depend on the heat conditions ( heat also demagnetizes). As it is hanging it is interacting with gravity so some magnetic potential energy will be lost when it first attaches to the iron beam. If it is motionless I would think only the heat will demagnetize further, (randomness induced by thermodynamic processes) and stray magnetic fields . $\endgroup$ – anna v Nov 4 '15 at 5:53
  • $\begingroup$ Ok.. fair enough. If a machine moved the magnet back & forth to make & break its connection to the iron beam, how many make-break events would it take to make the magnet so weak that it wouldn't have the force to hold to the underside of the iron beam even one more time? $\endgroup$ – DIYser Nov 6 '15 at 2:00
  • $\begingroup$ have a look at adamsmagnetic.com/blogs/2012/… . $\endgroup$ – anna v Nov 6 '15 at 5:34
  • $\begingroup$ As far as using a magnet to extract more energy than put in, it is no go. The processes that demagnetize magnets, like heat and shocks are slow because not much energy is expended on demagnitization. If you have a process that aims at extracting appreciable energy the demagnetization will be much faster. $\endgroup$ – anna v Nov 6 '15 at 7:11

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