quantum physics and magnetism?

BigMacnFries

I read on some physics website somewhere (I forgot which one) that a magnetic field around a bar magnet exists because there is a tiny probability that some electrons exist outside the bar magnet and when they do they emit virtual photons that can interact with other things to create the magnetic force. Is there any truth in this? If not, can you tell me a quantum explantion of magnetism, or point me to a good webpage, Thanks.

dextercioby

Yes,there is a quantum explanation of magnetism within the realms of quantum mechanics and quantum statistical mechanics.Any solid book on solid state physics should have a chapter on magnetism.

Auerbach,"Interacting electrons and Quantum Magnetism",Springer Verlag,1994.

Daniel.

ZapperZ

Er... some electrons exist OUTSIDE the bar magnet? And this is what causes virtual photons?!!!! They must have omitted this when I studied quantum magnetism. You certainly do not find this in Auerbach's book!

I'm not going to comment anymore till I see this "physics website".

Zz.

Gokul43201

It is quackery - period.

References for the quantum picture of magnetism :

1. Look through the Magnetism chapters of any of the standard cond matter/solid state physics texts. I recommend Kittel (who did a lot of work on magnetism) and Marder.

For a deeper understanding :

2. Magnetism - vol I, Rado and Suhl

3. Molecular Paramagnetism, Boudreaux and Mulay

4. The Theory of Magnetism, Matthis

Rough outline of study :

Hund's Rules and Crystal fields - magnetism in transition metal and rare earth salts

Energetics, spin waves, symmetry breaking

Exchange in insulators : direct, super, double exchange, the Hubbard Model, spin Hamiltonia - the Heisenberg model, Ferro and Antiferromagnetic ground states in insulators

Ferromagnetism in metals - exchange in metals (Ruderman-Kittel), role of free electrons and the Stoner Model (this part is real tricky and is the least well understood - ie: "why is a chunk of iron magnetic ?")

Jonny_trigonometry

Magnetic flux quantization within a superconducting ring...

Is this caused by a condition such that only an integral number of electron de brogile wavelengths can exist in the ring? Much like how only integral wavelengths are allowed to orbit a nucleus thus giving rise to n-states? Is the electron cloud of an atom thought of as a superconductor in this way? Was this the reasoning behind an experiment to detect magnetic flux quantization, or was it observed before the reasoning was established?

Gokul43201

This is a different question and should be asked in a separate thread.

No, it is not.

No, it is not.

No, it was not.

The quantization of the flux [Deaver and Fairbank, PRL, 7, 43 (1961), Doll and Nabauer, PRL, 7, 51 (1961)] was observed over a decade after it was predicted by London. The only similarity with the atomic model is that flux quantization in a ring results from essentially the same thing as angular momentum quantization in an atom does : that the wavefunction must be single-valued upon going around the loop an integer number of times.

Dr.Brain

There are some special materials, formed in the earth's crust in such a way that they show properties of magnetism.These materials exhibit magnetism in the same way the current wire exhibits it due to moving charges.Electrons at atomic level mover around atoms in circular paths.These electrons move in symmetric paths such that if you take a small area inside the material, the currents cancel but the ones at the corners of the bar donot.Its something similar to a solenoid such that there is no field inside but there is just outside it.

Gokul43201

Can you point me to a reference or a link or a material name. I have not heard of a material whose magnetization vanishes away from the surface (into the bulk)...nor does your explanation make sense to me.

Perhaps I can learn something new here.

Dr.Brain

Ok i saw that i messed it up a bit , the correct explaination to my best knowledge is :

Actually this is the explaination for 'what causes the magnetic field in magnets?'

Actually the atoms in these materials are responsible.The electron moving around atoms constitute current at atomic level in these materials.These currents can be thought of being equivalent to current carrying circular loops.In magnets , these loops are arranged parallel to each other and have currents in the same sense.At any pt. inside magnet, net current is zero because the currents from adjascent loops cancel each other.However there is current along the surface where currents donot cancel.This is just like solenoids, where there is no current inside the solenoid as per amperian rule.

Gokul43201

That's part analogy, part explanation, and part omission. And it certainly is not a quantum picture of magnetism.

ZapperZ

I think you are missing the point of the issue of "quantum magnetism". Everyone knows about the origin of the magnetic dipole moment in atoms. That isn't the problem. But what causes a material to be diamagnetic, ferromagnetic, antiferromagnetic, paramagnetic, etc... is a MANY-BODY problem that is very difficult to solve and is still an on-going research field. This is especially true when you find these individual dipoles to lower dimensions. (2D plane and 1D chain). Why do these individual magnetic moment line up with their neighbors in one instant while they allign in opposite direction in another instant in a different material? This is where you have everything from mean-field approximation, to the t-J model, etc.

The issue of magnetism in matter isn't just the individual magnetic moment, but the LONG-RANGE pattern of the distribution and orientiation of these moments. Such long-range order is what determines if it is a ferromagnetic, antiferromagnetic, etc....

Zz.

Dr.Brain

I agree with you.As far as my textual knowledge goes,in the paramagnetic materials the atomic dipole align themselves in the direction of applied magnetic field , while in ferromagnetic materials , the atomic dipoles can align themseleves without enough external magnetic field being applied , while in diamagnetic materials , there are no atomic dipoles , that is all in all they cancel each other, therefore dipole moment has to be induced by the applied magnetic field, and thus as per Lenz Law, the atomic dipoles oppose the applied external field.

Though it is clear from above that we certainly know a lot about how materials behave but certainly not 'why they behave like this'...

Gokul43201

If there are no "atomic dipoles", then what are the things that cancel each other ?

Sorry, but this is at best extremely incoherent.

You do not know why they behave like this. That is not to say that the Condensed Matter Physics community does not know.

A whole lot is known about why they behave like this, and my first post in this thread talks about the steps to take towards gaining this understanding. Again, you are missing the point that the quantum theory of magnetism explains a lot more than you seem to be aware of (as you persist with classical explanations).

Meir Achuz

"Quote:
therefore dipole moment has to be induced by the applied magnetic field, and thus as per Lenz Law, the atomic dipoles oppose the applied external field.

Sorry, but this is at best extremely incoherent"

I will just clarify this part of the thread. Diamagnetism arises for molecules that have no permanent mag dipole moment. By Lenz's law (or RH rules in Faraday's law) applying a magnetic field B will induce a dipole moment in the molecule that reduces the total B field. This called "diamagnetism". The same result comes in either a classical or quantum calculation. Diamagnetism and paramagnetism can be explained by treating just the individual molecules either classically or in QM. They are not manyh body effects. Ferromagnetism is a many body quantum effect that cannot be explained ("why") classically, although its effects can be treated classically.

Gokul43201

Meir,

I'll have to take exception to this line :
How does one determine the magnetic moment [itex]\mathbf{\mu} = -g \mu_B \mathbf{J}[/itex] of an individual ion classically ? What is the classical value of the Lande' g-factor ? What is the classical analog of J, and how does one determine this without Hund's Rules ? Assuming even, that the magnetic moment of a single ion can be determined classically, the act of performing a thermal averaging classically, will introduce an error - given by the difference between the Langevin function [itex]L(x) = coth(x) - 1/x [/itex] and the Brillouin function, [itex]B_J(x) = \frac{2J+1}{2J}coth \left( \frac{(2J+1)x}{2J} \right) - \frac{1}{2J}coth \left( \frac{x}{2J} \right) [/itex].

In addition, how would one account for things like spin-orbit coupling, crystal fields and quenching of the orbital angular momentum ? Just because there is no long range order in dia/para-magnetism does not mean they can be adequately described classically.