Fields in ferromagnetism and proper range of atom size

[Plastic Photon]

When iron, cobalt and nickel are placed in weak fields they assume a large magnetic polarization. I was wondering if this weak field needs to be either diagmagnetic or paramagnetic field. My guess is that it is paramagnetic but not sure which paramagnets could be used.

'Weak field' is used in Linus Pauling's General Chemistry but in my textbook at school just 'magnetic field' is used and it also states that the size of the atom must be within the proper range. I don't know whether size is meant as atomic weight or atomic radii. Wonder if you could help me out with that one as well.

 

[inha]

There is no such thing as a paramagnetic or diamagnetic magnetic field. Diamagnetism and paramagnetism are properities of matter that indicate how the matter responds to external magnetic fields.

 

[nbo10]

I suggest that you find another textbook. Fe, Co and Ni do not need an external magnetic field to become ferromagnetic you just need to lower the temperature below the materials transition temperature. And the size of the atom has nothing to do with magnetism.

 

[Gokul43201]

When iron, cobalt and nickel are placed in weak fields they assume a large magnetic polarization. I was wondering if this weak field needs to be either diagmagnetic or paramagnetic field. My guess is that it is paramagnetic but not sure which paramagnets could be used.
'Weak field' is used in Linus Pauling's General Chemistry but in my textbook at school just 'magnetic field' is used and it also states that the size of the atom must be within the proper range. I don't know whether size is meant as atomic weight or atomic radii. Wonder if you could help me out with that one as well.

Some of your questions have been answered above, but it's still not clear from your post what your objective is.

You do not use a diamagnet or paramagnet to make a weak field. If you want a weak field, all you have to do is nothing (there's already the Earth's magnetic field).

If however, you want to generate something like a hysteresis plot for a ferromagnetic material, you want to be able to apply continuously varying fields. The easiest way to do this is with an electromagnet. The field in a solenoid can be made large or small by changing the current through it - all it takes is a variable resistor (potentiometer).

That's the easy bit. The non-trivial task is to measure the magnetization. The only home-riggable measurement scheme I can think of is that used in a Vibrating sample magnetometer, and that's far from easily implemented.

Don't forget that you can also make very weak fields by using very strong permanent ferromagnets, simply by increasing the distance from the sample.

 

[Gokul43201]

I suggest that you find another textbook. Fe, Co and Ni do not need an external magnetic field to become ferromagnetic you just need to lower the temperature below the materials transition temperature.

Plastic Photon (PP) didn't say you neded an external field to induce ferromagnetism, but rather, to induce a non-zero net magnetization.

And the size of the atom has nothing to do with magnetism.

Not true. What the textbook (I have Pauling lying around somewhere; I'll take a look later if you, PP, post the location of this passage in the book) was referring to was likely the Neel-Slater curve. The "size" in this case, is the interatomic separation, which is related to the atomic radius and the crystal geometry.

 

[nbo10]

But you don't need an external field to induce a non-zero net magnetization. Otherwise plan old magnets would never be magnets.

I've never read Pauling but if you look at any theory of magnetism atomic radius never enters into the picture.

 

[Gokul43201]

But you don't need an external field to induce a non-zero net magnetization.

That's not right. If you cool a macroscopically sized paramagnet down below the Curie temperature, with H=0, you will get a ferromagnet with virtually no net magnetization.

Otherwise plan old magnets would never be magnets.

Plain old magnets are made by orienting their domains in a strong, (>2T) saturating field.

I've never read Pauling but if you look at any theory of magnetism atomic radius never enters into the picture.

In ferromagnetism, it's almost the first thing you'll come across when you do the Heitler-London calculation for the exchange parameter. Did you even try to look up the Neel-Slater curve before responding ? Do you think it's just a coincidence that the three ferromagnets with the highest Curie temperatures (Fe, Co, Ni) are within a couple of percent of each other's atomic sizes (and within less than half a percent of each other's interatomic separations) ?

In para- and dia-magnetism, atomic sizes and separations are less important, but still manifest in effects like crystal field splitting and nuclear hyperfine splitting.

 

[Igor_S]

That's not right. If you cool a macroscopically sized paramagnet down below the Curie temperature, with H=0, you will get a ferromagnet with virtually no net magnetization.

That's because ferromagnetic domains will (on average) cancel each other, right ? So weak external field is needed to turn some more domains in the direction of the field, so then we will have some net magnetization.

Ofcourse, inside each domain, there is some (nonzero) magnetization, so I think this is what is actually meant by saying that ferromagnets have some magnetization even in the absence of external field.

 

[nbo10]

Did you even try to look up the Neel-Slater curve before responding ?

I've looked through 5 CMP books and 2 books on magnetism and can't even find this term.

 

[Gokul43201]

That's because ferromagnetic domains will (on average) cancel each other, right ? So weak external field is needed to turn some more domains in the direction of the field, so then we will have some net magnetization.
Ofcourse, inside each domain, there is some (nonzero) magnetization, so I think this is what is actually meant by saying that ferromagnets have some magnetization even in the absence of external field.

Yes, that's right. There's local non-zero magnetization (long range order) over length-scales much greater than lattice parameters. But in the absence of an external field the domain magnetizations cancel off each other in order to minimize free energy. Of course, you can turn on a field and then turn it off, and retain some large magnetization. This is nothing but the remanence which results from domain wall pinning. Given enough time (age of universe ++), the remanence of any ferromagnet will go to zero. Domain wall thermodynamics is only pseudo-equilibrium behavior.

 

[Gokul43201]

I've looked through 5 CMP books and 2 books on magnetism and can't even find this term.

My books are not with me now, so I'll give you references later, but a Google search did precious little. There was just one hit : http://ej.iop.org/links/q29/a0Z8Jl1HvonqdTIhhhOrgA/jfv16i5p651.pdf

That's only proof that such a thing exists and it involves interatomic separations (ie: that I'm not giving you BS).

I'll dig up better references later, but an important idea is that the distinguishing characteristic between ferro- and antiferromagnetism is the value of the exchange integral. Different values of J give you different ground states, as you find out from solving the Heisenberg model in stat mech. This J is, crudely speaking, a measure of wavefunction overlap between near-neighbor atoms and has everything to do with their sizes and separations.

 

[inha]

404 not found. The idea is clear but I haven't seen that term anywhere either.

 

[Gokul43201]

404 not found. The idea is clear but I haven't seen that term anywhere either.

Aargh ! It's a vast right-wing conspiracy, I tell ye !

This is the best I can do now :

http://www.google.com/search?hl=en&q="neel-slater"&btnG=Google+Search



Perhaps there's a more popular name for the same thing...

 

[inha]

Do you know of an alternative name for it? It seems to be for real (unlike the moon landing and other government conspired things) but rather unknown atleast by that name.

edit: damnit I didn't read your last sentence for some reason. Never mind this post.
edit2: searching for "Néél-Slater" gives a few more results.

 

[fadafan]

I think it is better than you see "the Bethe-slater" curve which is found faster than "Neel-Slater" curve.

 

[Gokul43201]

I think it is better than you see "the Bethe-slater" curve which is found faster than "Neel-Slater" curve.

Yes, that's the one. Thanks, fadafan. I had no idea it was also known as hte Bethe-Slater curve...in fact, I find it surprising. While Bethe occasionally dabbled in condensed matter, he mostly stuck to high energy and cosmology.