Talk:Ferromagnetism/Archive 1

Good article
Just wanted to say this is becoming a good article. I like that the intro makes clear that ferromagnetism is the 'ordinary' magnetism everyone is familiar with. The explanation is remarkably clear considering ferromagnetism is such a complicated, technical, subtle phenomenon. Great work. -- Chetvorno TALK 22:49, 9 October 2008 (UTC)


 * Thanks. A few more pictures might be nice, though.  Pictures that come to mind: a picture of a familiar magnet of some sort, a picture of magnetic domains, a phase diagram, a hysteresis curve for the magnetization vs. applied field.... —Steven G. Johnson (talk) 00:33, 10 October 2008 (UTC)

lol —Preceding unsigned comment added by 75.110.41.144 (talk) 01:06, 8 January 2010 (UTC)

Unanwered questions not mentioned in introduction
The fact that phycisists no little about how ferromagnetism works on the atomic level should be prominently discussed in the article and similarly mentioned in the lede. I'm reading this which seems like relevant for this article. __meco (talk) 19:35, 4 September 2010 (UTC)

Strongest?
From the introduction: "In physics, several different types of magnetism are distinguished. Ferromagnetism is the strongest type; it is the only type that can produce forces strong enough to be felt under normal conditions..."

Surely electromagnets are the strongest type, or am I missing something? Either way, the phrasing is confusing. — Preceding unsigned comment added by 81.231.157.21 (talk) 14:19, 17 July 2011 (UTC)


 * Electromagnets use ferromagnetism. I wrote the sentence you mention; it was meant to refer to paramagnetism and diamagnetism.  Many substances which are normally considered nonmagnetic (nonferromagnetic) like aluminum and liquid oxygen are slightly attracted to magnets, this is called paramagnetism.  Many other nonferromagnetic substances like water, plastic, and copper are slightly repelled by magnets, this is called diamagnetism.  These forces are very weak, and can only be detected by sensitive instruments in a laboratory, but are of interest to scientists.  Only a few substances, like iron and nickel, are strongly enough attracted to magnets so the attraction can be felt by the hand; these are the ones commonly considered "magnetic", and the attraction is caused by a separate effect called ferromagnetism.  You're right, the introduction is confusing and needs to be rewritten.  -- Chetvorno TALK 18:57, 17 July 2011 (UTC)

Lithium gas
Lithium gas

In 2009, a team of MIT physicists demonstrated that a lithium gas cooled to less than one Kelvin can exhibit ferromagnetism.[6] The team cooled fermionic lithium-6 to less than 150 billionths of one Kelvin above absolute zero using infrared laser cooling. This demonstration is the first time that ferromagnetism has been demonstrated in a gas.

And? And what is the significance of that event? Stevenmitchell (talk) 16:39, 31 July 2011 (UTC)

Terminology Soapbox
I know, this is not the proper place for these comments but whatever. There's three linguistic problems I have with the terms 'ferromagnetism' and 'ferrimagnetism'.


 * ferrOmagnetism sometimes includes ferrImagnetism, and sometimes doesn't. Therefore, it's one word with two closely related definitions.  Hence all this confusing wording like
 * Ferromagnetism (including ferrimagnetism) is ...
 * ...are either ferromagnetic or ferrimagnetic
 * Only a few substances are ferromagnetic....
 * "ferromagnetic" in this narrower sense


 * their spelling differs only by one letter buried deep inside. Not only are these easy to confuse when reading and writing, probably everybody has to remember to say 'phaRAOH magnetic' and 'faiRY magnetic' instead of the way they'd casually say them, 'fairmagnetic'.  (Same goes for sulfate/sulfite/sulfide but whatever).


 * It ain't just Iron anymore. So why do they still start with ferr?

I suggest ferrOmagnetic for the general concept (that's the only one I was taught in physics 101), and then more specific names for the more specific varieties. Specific names that DON'T start with 'fer'. You could replace Antiferromagnetic too while you're at it; doesn't really seem like the exact opposite. OsamaBinLogin (talk) 16:11, 10 November 2011 (UTC)
 * The problem is, that is how the terms are used in the scientific literature. It's not up to us to change the way these terms are used - we just need to ensure that the context always makes the meaning clear. RockMagnetist (talk) 16:20, 10 November 2011 (UTC)

domain wall terminology
An anonymous editor disagrees with me on the definition of Néel and Bloch walls. I suspect the disagreement is in the interpretation of "in the plane of the wall". In this image, the plane of the wall is represented by either of the two dashed lines and all of the magnetization directions are in this plane. The direction of change is perpendicular to the wall. RockMagnetist (talk) 21:21, 29 November 2011 (UTC)

Physical origin
The physical origin section needs some serious reworking. It might be best to discuss the (quantum) origins of magnetism starting from the classical current/loop analogy then discuss spin pairing/hund's rule. No need to define or simplify the concepts of spin/electron/orbital/paired/etc. Just link to those pages. Also what's with the pauli exclusion stuff? That likely doesn't need to be there either. -para —Preceding unsigned comment added by 75.69.135.142 (talk) 05:49, 11 January 2010 (UTC)

Older talk
Magnetization is spelled two different ways in the first paragraph, "magnetization", the correct spelling, and "magnetisation", which I do not believe is an acceptable spelling. There is no edit link for the introduction section so I could not change this myself. Perhaps someone who knows how to do this could change it. Thank you.

-Oscar

"Magnetisation" is the UK spelling which, though rarer, should still be acceptable... -Joao —Preceding unsigned comment added by 85.68.58.58 (talk) 19:52, 20 June 2009 (UTC)

Yea, it the British spelling. You might want to read a few English (the language we are referring to here) science papers first... Stevenmitchell (talk) 16:34, 31 July 2011 (UTC)

There are a large number of pages that were gone over by someone last night, I think the one who signs himself "~ender", where links to orbitals were replaced by orbitals. While I might prefer that the 'E' not be capitalized, I'm not going to make a big deal about it ;), but that article is much more relevant to discussion of orbitals than the orbit article is. -- John Owens 23:58 Apr 15, 2003 (UTC)


 * The word "orbital" here is used as an adjective, not a noun. In this sense it refers to the circumferential motion of the electron around the nucleus. It is not a synonym for electronic configuration. It therefore seems misleading to write orbitals . Although the orbit article is mostly about astronomy, it gives the appropriate definition for the term "orbital" as used in this case: "an orbit is the path that an object makes around another object under the influence of some force". -- Tim Starling 01:27 Apr 16, 2003 (UTC)


 * Oh yeah, I guess so, ~ender may have painted with a bit too wide a brush. Given its relevance, though, we should probably find somewhere else to link it (if it isn't already). I'll work on that, if you haven't already. -- John Owens 01:34 Apr 16, 2003 (UTC)
 * OK, electron shell is already linked in there, which redirects to electron configuration. That'll do just fine for me. -- John Owens 01:39 Apr 16, 2003 (UTC)

Can we maybe split the table into two pieces so it isn't too long? - Omegatron 22:57, Mar 23, 2004 (UTC)


 * How about converting it to a vertical table which the text can flow around? -- Tim Starling 00:28, Mar 24, 2004 (UTC)


 * Here's a vertical table...it's actually my original table, but I rotated it so that it would take less space (it was overwhelming the article). I didn't think to let the text flow around it; I'm not sure how to do that, actually, so knock yourself out. Steven G. Johnson 00:44, 24 Mar 2004 (UTC)


 * I tried. It is not perfect, but it works for now. - Omegatron 17:24, Mar 24, 2004 (UTC)


 * Thanks, it looks better. If someone could link the element names, that would be nice too.  Steven G. Johnson 20:53, 24 Mar 2004 (UTC)

Bloch walls are only one type of domain boundary - Neel walls would be a second example. Perhaps this term could be changed to "Domain Wall" and Bloch/Neel walls listed as example?

---

Physical origins?
The Physical Origins section seems to state that ferromagnetism is caused by the "spin" momentum only, and clearly states that the effect of orbital angular momentum is to manifest diamagnetism.

&mdash;DÅ‚ugosz


 * You're right...the article has suffered from a common shorthand in the physics literature (including in the Kittel reference cited). In Ashcroft and Mermin (Solid State Physics), they write: It is the widespread practice to refer to to the operators in the Heisenberg Hamiltonian [for ferromagnetism] as spin operators, even though the spin operator for each ion here represents its total angular momentum which, in general, has both a spin and an orbital part.  It is still true that both ferromagnetism and paramagnetism require partially filled electron shells, of course, in order that their total moment be non-zero.  I've fixed the article now, I think. &mdash;Steven G. Johnson 22:27, Apr 19, 2005 (UTC)

Ferromagnetism vs. Ferrimagnetism
Ferrites and Garnets are not ferromagnetic materials, but ferrimagnetic materials. Hence Y3Fe5O12,NiOFe2O3 etc should go to the ferrimagnetism section. The definition for ferromagnetism is not clearly mentioned. There are many flaws in the article which may confuse the reader. Justin


 * That table comes from Kittel, where it is labelled "ferromagnetic crystals". I think that there is some confusion here because the term "ferromagnetism" seems to be used in two ways.  First, it continues to be used in the older sense of any phenomenon by which a material exhibits a spontaneous magnetization.  Second, as described in Ashcroft and Mermin (Solid State Physics, p. 695):
 * The term "ferromagnetic" is also used in a more restrictive sense, when one distinguishes among the varieties of ferromagnetic states that can occur when there are many (not necessarily identical) magnetic ions per primitive cell. In such contexts the term "ferromagnetic" is often reserved for those magnetic structures in which all the local moments have a positive component along the direction of spontaneous magnetization. Solids possessing a spontaneous magnetization that fail to satisfy this criterion are called ferrimagnets.
 * I've noticed that some people like to insist that the older, more general sense is now "wrong", but that doesn't seem to be supported by two well-regarded recent textbooks (Kittel and Ashcroft/Mermin). However, I agree that the article should discuss both the narrower and the broader meanings, and should distinguish ferrimagnetic materials in the table.  —Steven G. Johnson 01:27, 16 December 2005 (UTC)

Well, with the inclusion of some comments, the article may look well for a general reader. But as a scientist working in the field of magnetism, I would rather prefer to use more specific terms with the advances in research. It will be nice if we could refer to magnetism books by Cullity, Chikazumi or Jiles for more specific explanation about ferro, ferri or paramagnetic materials. Anyway I am happy that such discussions will help in the clear understanding of science and warm wishes for the contributors. Justin

Wrong numbers (and units) for magnetization
In this article it states that the magnetization of iron is 170.7 mT (1707 gauss). This is wrong. The problem originates in the choice of Gauss as a unit for magnetization.

To be exact, the SI unit for magnetization is A/m and the cgs unit for magnetization is emu/cm3. Often we will convert this magnetization to an equivalent magnetic flux density (the term magnetic induction is also used) for which the units are Tesla (T) in SI and Gauss (G) in cgs. This creates a problem when converting between the unit system. The conversion factor between magnetization and magnetic flux density is the permeability of free space which is μ0 = 4π•10-7 Tm/A in SI and which is dimensionless and equal to unity in cgs. This means that we can, in cgs units, say that the magnetization is e.g. 1700 emu/cm3 or 1700 G, but in SI these two values would be different. It is true that converting between Gauss and Tesla is done by dividing by 104, however this is only true when we are talking about a true magnetic flux density - not when we are talking about a magnetization! If we are expressing a magnetization in Gauss we should first convert to emu/cm3 by multiplying by 1, then to A/m by multiplying by 103, and finally to T by multiplying by 4π•10-7. In total we must thus multiply the value in Gauss by 4π•10-4 to obtain the value in Tesla. It is the factor 4π that is missing from the values quoted in the text.

So Fe80B20 really has a "magnetization" of 127.5 mT × 4π = 1.602 T (assuming the value in Gauss is correct) and pure iron has a "magnetization" of 2.145 T.

If you want to be 100 % correct you should always state magnetization in A/m or emu/cm3, but that would be very unhelpful for the reader since most scientific articles and books will state magnetization in Tesla or Gauss. —Preceding unsigned comment added by Jhekkonen (talk • contribs)
 * intensity, not density -lysdexia 00:35, 25 September 2006 (UTC)


 * I think you are right. The original number of 1707 gauss comes from Kittel and should be correct.  The incorrect conversion was introduced by this edit.  I'll fix the missing 4&pi;, thanks.  —Steven G. Johnson 01:11, 25 September 2006 (UTC)


 * No, that change was not correct. I'm removing it until it is straightened out.  Gene Nygaard 07:59, 16 December 2006 (UTC)

Here is the text that was currently there:
 * "One example of such an amorphous alloy is Fe80B20 (Metglas 2605) which has a Curie temperature of 647 K and a room-temperature (300 K) saturation magnetization of 1.58 teslas (1257 gauss), compared with 1043 K and 2.15 T (1707 gauss) for pure iron from above. The melting point, or more precisely the glass transition temperature, is only 714 K for the alloy versus a melting point of 1811 K for pure iron."

As Steven pointed out, the source used had the 1707 gauss figure, not Jhekkonen's "emu/cm³".

If you are going to have numbers that represent "magnetization" (the "magnetic field intensity" measured in A/m in SI) as a quantity distinct from "magnetic flux density" (a proper term for the quantity measured in teslas in SI; lysdexia's comment is a failure to recognize the shift between quantities being measured) you need to make sure that the distinction is maintained in the units used.

I have enough understanding of the differences between a non-rationalized cgs system and the rationalized SI system to understand that the 4π factor can enter into calculations, but this rewriting isn't right. Gene Nygaard 08:17, 16 December 2006 (UTC)


 * My guess is that a big part of the problem here is that User:Jhekkonen is getting the gauss and the oersted mixed up; a somewhat understandable confusion, since their meanings were originally switched from what they are now, but that change took place back in the 1930s or earlier and isn't likely to have been the meaning of gauss in the referenced book. Gene Nygaard 08:32, 16 December 2006 (UTC)

I don't think I am getting anything mixed up. If we use a source which states the saturation magnetization of iron in SI units (Elementary solid state physics by Omar) we find a value 1.74•106 A/m. We convert to Tesla by multiplying by 4π•10-7 and arrives at the value 2.19 T, which is comparable to the result I obtain if I assume that Kittel is "mistaken". My point is that people use Gauss, Oersted and (less often) emu/cm³ interchangeably and Kittel misuses Gauss as a unit for magnetization. The use of Gauss as a unit for "magnetic field" is very common but I don't think it is corrent (neither does wikipedia).

In short, the magnetization of iron is around 1.7•106 A/m or 1700 Oersted (or emu/cm³). However, the magnetization is often expressed as an equivalent magnetic flux density and then we obtain 2.1 T or 1700 Gauss. Jhekkonen 18:35, 14 February 2007 (UTC) The term 'Magnetization' (or spelled with an 's') represents a vector field folks, not a single scalar quantity! Does the article mean remanance, coercivity?.... If so 2T is too high for iron...Should be more like 0.1T. —Preceding unsigned comment added by 78.148.187.172 (talk) 22:06, 6 September 2010 (UTC)

Magnetisation in that sense means "saturation magnetisation", Ms, and is a scalar: it's the maximum magnetic moment per volume of a given material. For iron it is indeed about 2.2T (it varies with temperature, obv). The unit can be quite confusing, as some unit systems give you a 4π factor between magnetisation and field and others don't. In good SI units, Ms for iron is 1750 kA/m. Joaosampaio (talk) 10:50, 12 September 2010 (UTC)

Exchange Coupling for the Lay Reader
I would love to have an elaboration of this paragraph in the article, since it is the very heart of the explanation of ferromagnetism. Here it is:


 * According to classical electromagnetism, two nearby magnetic dipoles will tend to align in opposite directions (which would create an antiferromagnetic material). In a ferromagnet, however, they tend to align in the same direction because of the Pauli principle: two electrons with the same spin state cannot lie at the same position, and thus feel an effective additional repulsion that lowers their electrostatic energy. This difference in energy is called the exchange energy and induces nearby electrons to align.

Now I have a solid high school physics background and a little exposure to basic quantum mechanics. I find the sentence "two electrons with the same spin state cannot lie at the same position, and thus feel an effective additional repulsion that lowers their electrostatic energy" a bit too brief. After thinking about this, I wonder if first it ought to be explicitly mentioned that the electrons will tend to choose their spin states in a manner minimizing the total energy of the system. Then the quoted sentence above might be rephrased so that it is clearly telling us which choice of spins has a lower energy state.

After that, I am left asking the following question, which may either be totally off-base or have an answer obvious to people with more expertise: While electrostatic repulsion causes electrons to have a higher potential energy when closer together, how is it possible that the Pauli exclusion principle gives rise to a kind of repulsion having no corresponding potential energy? Perhaps an additional sentence or two will make this clearer. —Neoprote 19:19, 29 September 2006 (UTC)

Pauli principle
"In a ferromagnet, however, they tend to align in the same direction because of the Pauli principle: two electrons with the same spin state cannot lie at the same position, and thus feel an effective additional repulsion that lowers their electrostatic energy. This difference in energy is called the exchange energy and induces nearby electrons ...."

I think the above quoted text is not clear about why the spin state align in the same direction. My understanding is : " The aligned spin state can have "exchange energy", but anti-aligned ones do not have the benefit of lowering energy from exchange energy." But I donot see why the Pauli principle is relevent here. —Preceding unsigned comment added by 129.252.244.96 (talk • contribs) 02:37, 13 October 2006


 * I also find the paragraph to be correct but confusing (see also previous Neoprote topic ). I will try to reformulate the paragraph, mentioning that state with parallel spin has (because of the Pauli principle) anisimmetric wave function (it's space part actually) and thus has lower electrostatic energy. Which is explained in detail in Pauli principle and Exchange interaction. - Kopovoi 13:29, 7 March 2007 (UTC)

And it is not Pauli principle but spin-statistics theorem which tells, that thouse state are antisymmetric. Kopovoi 11:47, 11 March 2007 (UTC)


 * The Pauli principle is a consequence of the spin-statistics theorem, and is effectively a re-statement of it but in simpler terms that many people are much more familiar with. (Everyone learns the Pauli principle in high school, but the spin-statistics theorem is only presented when you learn quantum mechanics in more detail in terms of wavefunctions.) In an introductory article like this, we need to state things in terms that are as accessible and intuitive as possible, for as broad an audience as possible, without being incorrect, and refer to other articles for more precise mathematical statements.  —Steven G. Johnson 15:48, 11 March 2007 (UTC)

Very little, or nearly all?
Surely ferromagnetism is responsible for nearly all of the magnetic behavior encountered in everyday life.

Normal, Everyday Magnetism
Yes, MementoVivere, it may be subjective. But here's the thing. Why would someone be looking up ferromagnetism on Wikikpedia? Answer: because they don't know what it is. They read the word in an article or news story or whatever and said "What's that?"

There is a very easy way to tell them in one sentence, extremely clearly, what ferromagnetism is. The first sentence, or certainly the first paragraph, of this article should do so. Your changes, while great and accurate, leave the layperson profoundly unenlightened about his query. He will read the article and potentially miss the point that "Hey, you know horseshoe magnets, fridge magnets, and every other magnet you've ever consciously encountered in your life? Those were ferromagnets." Novel-Technology 11:57, 8 March 2007 (UTC)

Diagram request
It would be informative to illustrate ferromagnetic atoms and domains, to show how the phenomenon arises. Also possibly how it is different from ferrimagnetism and antiferromagnetism. -- Beland 04:07, 1 April 2007 (UTC)