# Do Permanent Magnets Dissipate?

Discussion in 'Electronic Basics' started by Juno, Jan 5, 2006.

1. ### JunoGuest

Hello all,

The subject title pretty much describes my question. I was discussing
various topics, such as electricity, theory, and forces of friction and
magnetism with a friend of mine when he asked me if permanent magnets
slowly lose their magnetism over time. I honestly didn't know the
answer, and searching Google didn't help me either. I know that
traditionally, there are only four ways to demagnitize a magnet:

* Heat. Heating a magnet past its Curie point will destroy the long
range ordering.
* Contact. Stroking one magnet with another in random fashion will
demagnetize the magnet being stroked, in some cases; some materials
have a very high coercive field and cannot be demagnetized with other
permanent magnets.
* Hammering or jarring. Such activity will destroy the long range
ordering within the magnet.
* Being placed in a solenoid which has an alternating current being
passed through it. The alternating current will disrupt the long range
ordering, in much the same way that direct current can cause ordering.

(Found at Wikipedia:
http://en.wikipedia.org/wiki/Permanent_magnet#How_to_demagnetize_materials)

However, I don't know if permanent magnets will simply lose their
forces of magnetism over time. Personally, I don't see why they would,
as I can't see any force(outside of the four listed) that would work on
them to lose their magnetic abilites. Does anyone know the answer?
Thanks in advance for any help.

Cheers,
Juno

2. ### Pooh BearGuest

AlNiCos certainly do

Graham

3. ### JunoGuest

Thanks for the quick reply! In response to your answer, maybe I should
rephrase my question then: Are there any permenent magnets that do not
lose their magnetism over time? Such as neodymium-iron-boron magnets,
or maybe something else? The strength of magnetism does not necessarily
matter; I just need to know if permenent magnets that don't lose their
magnetism exist.

4. ### John PopelishGuest

temperature is a statical sort of thing. Even at temperature well
below the curie temperature, there is a non zero chance that a given
domain will momentarily get hit with an effective thermal energy that
approaches what it would see at the curie temperature. So there is
not a perfect magnetic stability, even below the curie temperature.
For some materials, like hard steel, magnetism decays with a half life
short enough for the effect to be directly observable. With high
coercivity materials, samarium cobalt or neodymium, the decay rate is
very much slower. I think the term that describes this effect is
magnetic disaccomodation.

There is an analogous effect for dielectrics that have been polarized
with an electric field (electrets and piezo transducers).

5. ### Bob MyersGuest

In a practical sense (depending on what time span you
have in mind, of course), the answer is "yes," meaning only
that there are "permanent" magnets whose magnetization will
not appreciably decrease over the expected useful life of
whatever product they're used in. Over the long term, of
course, the answer is "no" - eventually, any "permanent"
magnet will slowly lose magnetization unless there is some
mechanism provided for refreshing it. Of course, "long term"
can mean some pretty lengthy periods.

Bob M.

6. ### Pooh BearGuest

Ceramic magnets are pretty permanent I believe.

Graham

7. ### Mark FergersonGuest

You meant "statistical"? Yes.
The general idea is that magnetic domains don't want to be in the
long-range order that we call "magnetized" because it's a high-energy
situation for them relative to the unmagnetized, disordered state we
usually find magnetizable materials in. When we magnetize something we
add energy to it by ordering the domains.

To get a feel for it build an analogy. Take two small bar magnets and
let them join naturally, all poles together like a four-way handshake.
Notice their external field is greatly reduced; the magnets are analogs
of the domains in a magnetizable material that isn't magnetized. Imagine
stacking more magnets to the sides, building up a "magnetic crystal" so
that each pair of magnets attracts each nearby magnet. It doesn't quite
work in threespace with real magnets but you can get pretty close.

Now pull the first two magnets apart and stack them north-south so
that one pole of each is free. Their external fields add just like
ordered domains. Now stack more to the sides so they _don't_ attract
each nearby magnet and notice that the field gets stronger and stronger.

(Of course this is very hard to do without using say duct tape; the
analogy doesn't account for the fact that the magnetic forces between
domains are much weaker than the chemical bonds making up the lattices
that support the domains).

This is basically what happens when a piece of magnetizable material
is magnetized; the domains are rotated so that the fields add, but it's
an unstable state, and incoming energy (like heat energy shaking the
lattices) can allow some of the domains to "fall" back into the old,
familiar, comfortable four-way handshake. Notice that all four of Juno's
examples do just that; excite the lattices so that the domains have some
freedom to rotate. John, you're right; even at temperatures well below
the Curie point all magnets will slowly demagnetize due to the random
local domain rotations caused by low-energy phonons exciting the lattices.

Now for the last question; the only true "permanent" magnet I know of
offhand would be a superconducting loop; it only has one domain in a
manner of speaking.

Mark L. Fergerson

8. ### Charles SchulerGuest

It is called entropy and the answer is yes.

9. ### John PopelishGuest

Damn spell checker.
(snip)
To quote Doctor Memory (from Firesign's "I think We're All Bozos on
This Bus"), "The system is less energetic if domains of opposite
polarity alternate." And then he excuses himself for a nanosecond to
flush the toilets and set off the fireworks.

10. ### JunoGuest

Handy explanation of the processes involved, and why permenet magnets
eventually will lose their polarity. I suppose even at cryogenic
tempuratures, it is possible for magnets to lose their strength?
Interesting comment you made at the end of your reply though:
Now this might be more what I'm looking for. I'm designing a
theoretical machine that requires parts capable of remaining (truly)
permanently magnetized. As you say, these superconducting loops only
have one domain; in that case, would it be possible to arrange them in
a way to function identically as their traditional magnet counterparts?

11. ### Mark FergersonGuest

Sure, just much slower than at shirtsleeve temperatures.
I don't see why not (as long as the temperature is kept low enough
that they don't quench). A clearer answer depends on a clearer question;
IOW what do you want them to do?

Mark L. Fergerson

12. ### JunoGuest

Well, essentially I'm trying to create a levitating rod inside of a
drum, and have the rod rotate inside in mid-air, levitated by magnets
located on the inside of the drum. It would look something like this:

_________________________________
| |
| ************************************* |
| ************************************* |
|________________________________|

I know that there already exist devices like that, but they all depend
on magnets, which as we've discussed previously, dissipate over time.
I'm trying to create a version that won't lose it's magnetic
properties, and be able to keep rod properly levitated. Remember, this
is an entirely theoretical machine, but I'm just trying to figure out
if it is at least physically possible. With your proposal of using a
superconducting coil as a replacement for the magnets, that would solve
the problem of the drum and rod losing magnetic properties. By
replacing the exterior surface of the rod and interior surface of the
drum with superconducting coils, one would be able to levitate the rod
without any eventual magnetic loss.

If temperature is a concern for superconducting coils, then perhaps
placing the whole apparatus in a cryogenic chamber would keep
everything at a proper temperature. Obviously, as no room-temperature
superconducting material yet exists, I suppose I will need to modify my
plans to include a cryogenic encasement. In your opinion, do you think
the use of superconducting coils in a cold enough environment be able
to levitate the rod without any eventual loss of magnetic ability?

13. ### Jasen BettsGuest

superconductors repel magnets.

if you want levitation put a normal magnet in a superconducting
bowl...

or in your case possibly a superconducting tube-magnet inside a superconducting
vessel.

Bye.
Jasen

14. ### JunoGuest

That looks like the best solution at the moment. As I can't use a
normal magnet(since it will dissipate over time, which it what I'm
trying to avoid), superconducting coils seem to be the best option for
*truly* permanent magnetism. Your suggestion, "a superconducting
tube-magnet inside a superconducting vessel" is exaclty what I plan to
use. But, as I am not an expert on superconducting materials and their
properties, I've got a question. Would a superconducting coil require a
current of electricity to operate as a magnet, or can it perform this
function without the application of electricty?

15. ### Mark FergersonGuest

Once you start a supercurrent (what a superconductor, er,
superconducts) going in a closed loop, it keeps going "indefinitely".

That's in quotes to indicate that it depends on certain conditions
not changing; the temp stays low enough, the loop isn't broken, no huge
magnetic field is imposed, etc.

IOW it doesn't need any more power to keep going because it has no
way to dissipate the power it took to start the supercurrent, as opposed
to ordinary resistive circuits which do, you see.

BTW it sounds like you're trying to reinvent the magnetic bearing

Mark L. Fergerson

16. ### Jasen BettsGuest

yes they are "charged" by beeing cooled to superconducting temperature in
the presence of a magnetic field (often from an electromagnet)

after they are superconducting the external magnetic field is removed
and this induces the current that turns them into superconducting magnets.

Bye.
Jasen

17. ### JunoGuest

Thanks for explaining that for me. The conditions required for a
properly charged superconducting coil to have a singular domain and
remain that way seem to be pretty easy to take care of, save for one:
The temperature. Although a superconductor has very little resistance,
wouldn't there be a non-zero chance that at any temperature above
absolute zero, there would be resistance, and that the resistance would
slowly eat away at the energy of the closed loop? So that, over a very
long period of time, the electrical charge running through the loop
would be released as heat energy, then the superconducting coil would
fail to have a singular domain, then the levitating rod would fail
because the magnetic properties that kept it suspended have failed? If
I am missing something, or if you're simply referring to temperatures
going beyond the Curie temps, then perhaps I am wrong, and the coil
will retain it's singular domain properties forever, which is what I'm
trying to achieve. I'm not trying to reinvent magnetic bearings,
although they do bear a close resemblance to some of what my project is
concerned with. Thanks for your help.

18. ### Mark FergersonGuest

_Zero_ resistance.
No. This is a quantum-mechanical effect; the resistance curve has a
singularity (goes to zero) at Tc.
The short answer is that since this is a quantum-mechanical effect,
the statistics say that while there indeed is a non-zero possibility for
a phonon (quantum of heat) to quench a superconductor while it's well
below Tc, it's about as likely as all the neutrons in it decaying at once.

The long answer gets very mathematically messy, but the bottom line
is that as far as failure modes go, put that one way down on the list.

Mark L. Fergerson

19. ### Rich GriseGuest

Nope, it's exactly zero:
http://en.wikipedia.org/wiki/Superconductor
"...Superconductors are also able to maintain a current with no applied
voltage whatsoever, a property exploited in superconducting
electromagnets such as those found in MRI machines. Experiments have
demonstrated that currents in superconducting coils can persist for years
without any measurable degradation. Experimental evidence points to a
current lifetime of at least 100,000 years, and theoretical estimates for
the lifetime of persistent current exceed the lifetime of the universe...."

Cheers!
Rich

20. ### Rich GriseGuest

I saw a way cool thing on some science show - they took a disk of
superconductor and put it in the bottom of a beaker, dropped a little
magnet on it, and poured in some liquid helium. When the superconductor
got down to temp, the magnet simply rose. It was almost spooky!

Cheers!
Rich