Strange magnet behavior

I was reminded this morning of an “unsolved mystery” from years ago, and I thought maybe some of the Dopers might be able to shed some light.

My company had a booth at a trade show, and they gave us all magnetic name badges with our name and the company logo. You put the badge on the outside of your shirt pocket, and the other metal piece inside the pocket. The magnetic force holds the badge in place without making any holes in your shirt. So far, so good.

Here is the mystery. The two pieces of the badge had a very strong magnetic attraction to each other, just like any other magnetic material. But neither of the pieces had any magnetic attraction to any other metallic material. Put it on the fridge, it falls off. Put it on the metal part of your cubicle, it falls off. Try to pick up paper clips, no dice.

Has anyone else seen magnets(?) like these? How can magnetism be selective like this?

I’d have to guess that these badges used a technique similar to those flat AlNiCo refrigerator magnets, carried to the extreme. These refrigerator magnets are magnetized in alternating polarity domains, each one about 1/10[sup]th[/sup] of an inch wide. The result of this is that they will stick to a magnetic surface when placed right up against it, but won’t stick when more than a few sheets of paper are inserted, because the alternating magnetic domains begin to cancel out beyond the surface of the magnet, attaining full cancellation at about the spacing of the domains: .1". If we carry this to the extreme, I can easily imagine the magnet not having the strength to stick significantly to any surface UNLESS that surface also had identically-spaced alternating magnetic domains. In that case, the oppositely aligned domains would tend to attract each other fairly strongly. Put two flat refrigerator magnets back-to-back and try to slide one across the other and you’lll see what I mean.

Wouldn’t that sort of set up be helpful for magnetic items used in an office setting? Suppose you have your magnetic name-badge on your shirtpocket and you absentmindedly drop a floppy disk (like those exist…) in your pocket. I would think that a magnet of limited range would be very valuable. Or you set your name-badge on your desktop computer, or toss it in your laptop briefcase, or whatever. I too have a magnetic namebadge and I’ve worried I might damage some of my high-tech toys. Never noticed that it wouldn’t stick to other things. I’ll have to try that when I get home.

Not all of those magnetic name badges are like that. In fact, I’ve never personally encountered one - the OP was the first I’d heard of them, in fact. The ones I had when I worked for Radio Shack would stick strongly to anything ferromagentic.

But yes, you’re correct that such a badge would be safe to have around magnetic media, assuming they do work the way I outlined above.

While that’s a very clever idea, I wonder if anyone has seen a device that is KNOWN to have that construction.

I ask because the alternating domains aren’t so much reacting to the dipole orientation in an unmagnetized ferromagnetic substrate as inducing an opposing ferromagnetic dipole.

I can see the suggested mechanism working, especially if the magnets weren’t very strong in the first place (e.g. flexible magnetic material) especially with the field lines oriented perpendicular to the surface of the “striped” sheet: when the sheet is alone, the lines of manetic force would go directly between the adjacent stripes wirh very little “side leakage” (field lobes) to induce a dipole in an unmagnetized ferromagnetic substrate, and the confounding effects of the random domain orientation in the unmagnetized bulk material would have reduce the effective field even more (one reason magnetic fields penetrate magnetic materials poorly) However, when applied to a matching striped dipole, the lines would flow almost as evenly to the “partner” sheet as to the adjacent stripe in the same sheet,a d since the partner dipole is premagnetized, rather han weakly indiuced, the attraction could be quite strong.

I seem to recall working through all this as an undergrad, doing Hallidane and Resnick. My question is – has anyone specifically seen this in action or done the math recently in a Physics problem set?

Also, has anyone done the math for other combinations? E.g. If we allow two opposing orientations in the three axes defined by the edges of the stripes, how much preference would a pattern of stripes show for its partner vs. a nonmatching pattern? Would this only work effectively with alternating stripes?

In re: office use
Stray fixed magnetic fields actually pose a negligible risk to data on magnetic media. Since I can’t find a link to one of these experiments, I urge you to NOT take my word for it. Go ahead and try to induce a data error in a floppy with a permanent magnet. Dust or grease is far more likely to cause errors

The real danger, if any, comes from rapidly changing, and especially alternating magnetic fields. e.g. Bulk demagnetizers create an alternating field NOT a strong fixed field; unshielded speakers or the degaussing coil on your monitor can do more damage withtheir rapidly changing flux than even a fairly strong slow field, like the ringer on an old style telephone.

The deliberate magnetization of recording uses a strong field to orient a sizeable (but still fairly small) minority of the domains. Almost any standard college text will list the theoretical magnetism of plain iron if the domains were 100% oriented; it’s stunningly beond anything we’ve achieved with any high-tech materials.

However data corruption tends to operate on the “give me an excuse” principle. Since the vast majority of domains in magnetic media are NOT aligned, rapidly changing fields can ‘jiggle’ susceptible domains, which may, int turn, affect nearby domains somewhat. The more dense the data in a recording, the fewer domains would suffice to flip a bit. Remember: 90%+ were never contributing anyway.

I would actually worry that such small scale striped field might be more effective at demagnetizing or scrambling magnetized media than the fixed fields it’s meant to replace. Neither would pose any real danger at a distance of several centimeters, but while a kitchen magnet will do little damage if you slide on the back of a floppy, the “striped kitchen magnet” might look like a rapidly alternating field.

Well, I just did a quick experiment to try to confirm my hypothesis. I took a flexible refrigerator magnet and cut it exactly in half - this way ensures that the domains line up properly. One half will hold up to two 3x5 cards on a steel surface, but adding a third causes it to slide off. However, placing the two halves back-to-back will hold up to eight 3x5 cards between them. Rather than merely doubling the number of cards a single magnet could hold, that number was cubed. Interesting.