Talk:Photon/Archive 2

From Wikipedia, the free encyclopedia

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Depiction of photons

On about Aug, 9 and Aug, 11 User:Bkalafut deleted the images to this article, stating that:

Artist's rendition of photons.
The photon can be seen to behave as a wave or a particle, depending on how it is measured

Conceptual structure of a linearly polarized photon: a self-propagating transverse oscillating wave of electric and magnetic fields.

“This schematic is of a classical wave solution to Maxwell's equations and has nothing to do with quantum electrodynamics or photons” (right)

“This image is just of colored squiggles and doesn't appear to have anything to do with a quantized light field.” (left)

I disagree with Bkalfut, not only on logical basis but also on fact he is removing important parts without first using the talk page, and I will re-add a discussion with sourcing from Nobelist Martinus Veltman using his 2003 book Facts and Mysteries in Elementary Particle Physics. If anyone has comments or knows of better images please discuss here. Thanks: ---Sadi Carnot 15:55, 26 August 2006 (UTC)

I added a new section; it still needs a little cleaning due to the partial merge with the nomenclature section; and I can't figure out how to up-size the images using the wikipedia:gallery markup text? If someone can help, please do so. Thanks:--Sadi Carnot 16:55, 26 August 2006 (UTC)

--- I think it needs work, but I like the idea. Too bad all the images have waves in them. Is there a way to depict a photon without waves? What is the basis for the particle in "Thomas Young showed that light existed in both particle and wave form"??? Dicklyon 17:06, 26 August 2006 (UTC)

I am not sure I like the use of any of those pictures, mostly because they are the exact same photos used in the article on light and we want this article to focus on the quantum nature of light. If we use any photos, perhaps it should be diagrams of the quantum harmonic oscillator, since that is very helpful description of photons.

Also, you wrote:

"Historically, the nature of the photon in terms of a concept of its visual structure, as well the depiction of its internal structure"

I am not sure what this means, but I feel like it is implying the photon has internal structure (i.e. it is a composite particle), which it is not. I would suggest removing the photos and this intro. sentence and starting with "Historically, several scientists...", and rewording that paragraph.Waxigloo 17:21, 26 August 2006 (UTC)

OK...seriously: All of those images have nothing to do with the article itself. Does anyone else agree that the photos should be removed? I don't want to upset Sadi Carnot, who added them all, by removing them but if we could get some other opinions on this.Waxigloo 18:01, 26 August 2006 (UTC)

Illustrations are a good idea even they are imperfect. I think they should be improved. For example, the linear polarization one should be changed to circular polarization, since that seems more appropriate for photons. Dicklyon 18:40, 26 August 2006 (UTC)

I added all the images I could think of relating to the photon; we must remember that most people so not think abstractly, but rather visually. You can use all the math you want to try to explain something, but if you can’t draw a working model of it, no one is going to understand it. We still need one more good image to complete the set (so to fit the frame-work of eight images). If you know of one please add it. Thanks:--Sadi Carnot 19:02, 26 August 2006 (UTC)

Perhaps we can intersperse them throughout the article to illustrate various points? I've been missing some images, but I agree, they don't seem very helpful to illustrate photons per se, but they might help in illustrating other concepts that we mention in the text, e.g., interference, electromagnetic mode, etc. If peppered through the text, they wouldn't interrupt the flow so much, either. If that seems OK, I might try tonight when all my visitors are asleep! ;) Willow 21:30, 26 August 2006 (UTC)

I agree images are nice -- I like the idea of having useful images throughout the article to help explain things. I don't think a group of pictures that have to do with light.Waxigloo 22:56, 26 August 2006 (UTC)

Since the particle nature of light is manifested primarily when it interacts with other things, the appropriate images for this article would be Feynman diagrams of photons interacting with other things, e.g. Bremsstrahlung, Compton scattering. JRSpriggs 09:20, 27 August 2006 (UTC)

Willow: Well done with figures -- I think it looks much better now.Waxigloo 20:35, 27 August 2006 (UTC)

Wave number question

The lead says "has three continuous parameters, the components of its wave vector." Doesn't this imply a particular state with no uncertainty of energy or momentum? And what about phase and helicity parameters? Isn't this statement too narrow to be true in general and to be part of the lead? Dicklyon 18:40, 26 August 2006 (UTC)

The sentence in question is "A photon can have two different helicities (polarization states) and has three continuous parameters, the components of its wave vector.". As you see, it already mentions helicity (the spin projection in the direction of motion, i.e. polarization). Individual photons do not have a definable phase (see the section on "Wave–particle duality"). The closest one gets to that is the probability amplitude that the photon is in that basis state. And the alternative to talking about a "particular state" (i.e. a pure basis state) is to talk about non-basis states or mixtures of states which is too difficult for the introduction. JRSpriggs 09:32, 27 August 2006 (UTC)

Momentum's time component and the units of the Metric tensor

I tried to finesse these issues by rewriting the section "Energy, momentum, angular momentum and mass" to avoid mentioning the time component of momentum. I hope that it is satisfactory to you-all. JRSpriggs 10:05, 27 August 2006 (UTC)

Time for Peer review?

Hi, I was thinking of putting us up for peer review. We might get some good insights that would help us get to Featured article status. What do you all think — is the article ready for that kind of attention? Willow 14:52, 28 August 2006 (UTC)

I fully support the idea. I am a physicist working within the Wikipedia Physics Project and I would gladly support a peer-review of this article. I'll put in the template and get the ball rolling... Astrobayes 19:41, 28 August 2006 (UTC)

Willow, the peer review is now complete and can be accessed through the template links at the top of this page. Additional comments will be made about the historical dates, if warranted, but thus far all appropriate review comments can be found on the now-archived review page. Please let me know if I can be of any further assistance. I'll keep an eye on this article. Cheers, Astrobayes 20:46, 1 September 2006 (UTC)

Virtual photons and polarization

Willow asked about this on Wikipedia Physics Project talk page. My understanding is that virtual photons have four polarization states. See, for example, Sakurai's Advanced Quantum Mechanics, Section 4.6, pages 254-256 on the Covariant photon propagator, where there are four photon states summed over and where he explicitly discussed the four polarization states of a virtual photon. Salsb 14:19, 31 August 2006 (UTC)

I've mulled over this for some time, and had discussions with some of my coworkers and there seem to be two camps on this one: those who are in favor of the three-state idea and those in favor of the four-state idea. I also suggest Sakurai's quantum text and also D. Griffiths's Elementary Particles textbook as well. I'm still working on ideas for the article and I'll post them in the peer review page soon. Cheers, Astrobayes 17:35, 31 August 2006 (UTC)

I seem to recall hearing about a two-state option as well. It's some alternate method of constructing Feynman diagrams where the photons have two degrees of freedom, and once you know how to use it, it's supposed to be more efficient. Probably "chiral" something. But I could be completely wrong. Melchoir 17:44, 31 August 2006 (UTC)

The argument based on having four components of the vector potential is (as I said before) undercut by the fact that one's choice of gauge determines one of the four components in terms of the other three. In fact, in the Weyl gauge one of the components (the voltage) is eliminated completely. JRSpriggs 07:32, 1 September 2006 (UTC)

This was my understanding as well. It looks like Sakurai rediscovers the gauge condition whenever photons are real, having disregarded it in the derivation. The experimentalist in me says that's not a terribly concrete reason to say they have different polarizations. — Laura Scudder ☎ 15:24, 1 September 2006 (UTC)

I really appreciate everyone taking to time to figure this out! As you all seem to be saying, it may very well reduce to the choice of gauge. My own cursory reading suggests that, in manifestly covariant gauges such as the Lorentz gauge $\partial_{\mu} A^{\mu} = 0$ , you have to sum over four polarizations, whereas in others such as the Coulomb gauge $\boldsymbol\nabla \cdot \mathbf{A} = 0$ , you can get away with summing over three, or even two. The basic approach seems to be that you Fourier-transform the four components of

A μ

and convert the 8 Fourier coefficients per wave-vector (four components x positive/negative frequency) into four pairs of creation/annihilation operators. There seems to be a practical reason for going to a manifestly covariant gauge — aside from not accidentally dropping a term needed for relativistic covariance in a calculation ;) — but the whole thing is mysterious to me still.

For completeness, here are a few other references that seem to use four polarizations for virtual photons:

Heitler W. (1952) The Quantum Theory of Radiation, Dover, pp. 87-103.

Bogoliubov NN and Shirkov DV (1959) Introduction to the Theory of Quantized Fields, Interscience, pp. 55-58.

Ryder LH (1985) Quantum field theory, Cambridge University Press, pp. 143-153.

Thanks for everyone's help in making Photon better! Willow 16:05, 1 September 2006 (UTC)

Gauge of course does have a lot to do with the answer to the question. I have read on this and asked my colleagues and I support the idea that virtual photons have three polarizations (and of course 'real' photons we know can have two). Here are some sources that I found particularly illuminating:

Bjorrken, J.D. (1970) Inelastic Scattering of Polarized Leptons from Polarized Nucleons, Physical Review D, v1.#5

Bintinger, D. et.al. (1985) Measurement of the Total Hadronic Cross Section in Virtual Photon-Photon Interactions, Physical Review D, v54.#8

Brodsky, Hautmann, & Soper (1997) Virtual photon scattering at high energies as a probe of the short distance Pomeron, Physical Review D, v56.#11

And in addition to these, you will find some great discussions on Gauge in texts like Lewis Ryder's Introduction to Quantum Field Theory, mentioned above, as well as the J.D. Griffiths Introduction to Elementary Particles and his undergraduate electrodynamics text. But I must say I'm disappointed in J.J.Sakurai's discussions on virtual photon polarization - they are not as illuminating as the other texts mentioned here. As for peer review comments other than this virtual photon polarization question, they will be posted on the SPR page for this article today, before the weekend as promised. Cheers, Astrobayes 19:21, 1 September 2006 (UTC)

I actually don't understand this discussion here. Virtual photons are just a calculational tool. You cannot do any experiment to decide the properties of virtual photons. It's a meaningless question to ask whether they have 4 or less polarizations. You could say that it's four if you do a calculation with 4 and three if you use three.

In field theories, if you sum the entire perturbative expansion then the result doesn't converge. It isn't even Borel resummable, as 't Hooft has shown. So, the picture of virtual photons buzzing around causing the perturbations is not only not "what is really going on", it is even a bad approximate picture. Count Iblis 22:42, 1 September 2006 (UTC)

From your point of view, I see why you state what you do. Virtual photons aren't "real" in the sense that they are always on the "inside" of the Feynman diagrams, a calculational tool to represent interactions that take us from the initial particle states to the final particle states in the "real world." But this discussion led us to a very important result that the virtual polarization depends upon mathematical convention, and so the article can now correctly reflect that fact to readers who want a complete picture of where all of the concepts surrounding the photon can lead. There is value in the discussion then, just as there is value in answering the question "why is the sky blue?" or "why is there no photon frequency corresponding to 'purple'?" when we know very well that the concept of "color" is really a matter of perception and classification. These questions lead to valuable discussions of the subject which can be used to improve the article. That's my opinion anyway. Cheers, Astrobayes 22:58, 1 September 2006 (UTC)

Yes, I agree with that :) Count Iblis 01:15, 2 September 2006 (UTC)

Two-polarization treatment in QED

Melchoir, Willow, and even Astrobayes (in the peer review), all mention that there is a "2 polarization" usage of virtual photons as well. I have never heard of this (and the article currently doesn't include this as a possibility), and the gauge condition seems like it could only eliminate one polarization (although the peer review seems to state otherwise). So how does this work? And how can the Coulomb interaction work with only 2 polarizations, as it seems to suggest virtual photons can only have "real/physical" polarizations? I would very much be interested in hearing more about this. -- Gregory9 21:26, 4 September 2006 (UTC)

I'm no expert, but here's my (limited) understanding. In basic textbooks about quantum electrodynamics (there's an oxymoron! ;), they seem to adopt both the Lorentz $\partial_{\mu} A^{\mu} = 0$ and Coulomb $\boldsymbol\nabla \cdot \mathbf{A} = 0$ gauge conditions for quantizing the pure electromagnetic field (i.e., no charges are present), which is possible because photons are massless. These two conditions reduce the four components of

A μ

to the two independent polarizations observed physically; the electrostatic potential

φ

is zero (no charges present) and the Coulomb gauge condition makes the polarizations transverse since its Fourier transform is $\mathbf{k} \cdot \tilde{\mathbf{A}} = 0$ . The coefficients of the Fourier-transformed vector potential $\tilde{\mathbf{A}}$ are plane-wave solutions of the corresponding wave-vector $\mathbf{k}$ , and the Hamiltonian for the pure EM field can be written as a set of simple harmonic oscillators, one for each possible wave-vector. Upon quantization, these coefficients become creation and annihilation operators for a photon with that wave-vector. Treating the charge-field interaction term in the Hamiltonian as a perturbation, one can derive Einstein's A coefficients and first-order corrections for radiative processes. I guess, strictly speaking, these processes involve real photons, not virtual photons. However, higher-order corrections can involve processes in which virtual photons temporarily have nonzero 4-momentum magnitude (invariant mass, gasp!), as Melchoir pointed out. This is where my understanding begins to dissolve into a grey fog peppered with Feynman diagrams. ;) Hoping this helps, and hoping that more knowledgable people will step in, Willow 11:31, 5 September 2006 (UTC)

Next steps to Featured article status

Wow! Thank you very much, Astrobayes, for your hard work and very thorough review. I think you'll find the Lewis and Bose references at the bottom, but I forgot to include the Hertz and Maxwell refs :(

I agree with Astrobayes' suggestions for re-ordering and making the Semiclassical section less daunting to lay-people; how does everyone else feel, though? I'm sure that we can finesse the virtual photon polarization thing.

To reach Featured article status, what are the next steps before we submit? The criteria are here. We might need to improve our referencing format and eliminate any overlooked weasel words; the article should also be "stable", i.e., undergo no significant changes for, I dunno, maybe a week? Astrobayes' excellent review seems to cover all the bases, but maybe we should request a general peer review before submitting Photon as a FA candidate — what do you all think? Willow 22:01, 1 September 2006 (UTC)

I agree completely. Now that the article has had an SPR, a general peer review by non-specialists would provide the remaining impetus for finishing the path to featured article status. As I've stated on the SPR Photon page, my next contribution to this article will be verifying in detail the historical accuracy. After that, I'll help in any way I can. Thanks for getting the ball rolling, Willow! Cheers, Astrobayes 22:12, 1 September 2006 (UTC)

Definitely do a peer review first. If I came across the article on FAC, here's what I'd say:

Inline citations! They're the only way to be sure that everything is verified.
Structure: The logical relationship between sections is unclear. Among other effects, this makes the Table of Contents hard to read. There are too many top-level sections; a hierarchy of subsections would help out here.
Comprehensiveness: "Experimental limits on mass" is great, but how about charge?
The "See also" section is too long, and its purpose is unclear. Links that have been prominently mentioned above don't need to be here. Any concepts that are particularly relevant to photons should be worked into the article. Concepts that are only vaguely relevant need their connection to be explained.

Melchoir 22:30, 1 September 2006 (UTC)

Partial rewrite of photon mass section

I've tried to explain the physics behind some experiments that have been done to establish the limits on the photon mass. The reference list at the end of the article should be moved to the ref. section at the end of the article. But it would be good to first incorporate the results of these articles. E.g. in the article by Roderic Lakes a limit of 2 10^(-16) eV is derived using the torque exerted on a magnetized ring in the presense of the galactic vector potential. This is less controversial than the limit of 10^(-27) eV. Count Iblis 01:12, 2 September 2006 (UTC)

Rather than continue to lump all the references together, we should be going in the direction of citing them to specific facts (and by "we" I mean not I, since I haven't read them). Melchoir 17:07, 2 September 2006 (UTC)

References

The article currently seems to have a mix of reference systems - some as footnotes, most only in the References section. I would recommend that some sort of inline references are used throughout the article, with possibly the exception of general references used throughout the whole article. I would recommend either using footnotes throughout, or using Harvard referencing with footnotes reserved for notes rather than references. Thoughts/comments? Mike Peel 08:42, 3 September 2006 (UTC)

I think that the in-line references with <ref> </ref> are very hard to read when editing. Therefor, I think that it would be better to avoid them. I have to make an effort to see where the normal text ends and begins again in some cases. JRSpriggs 09:51, 3 September 2006 (UTC)

Hi, Mike, thanks for all your work on the references! I also favor Harvard referencing, although I haven't always been consistent about it. I noticed that other FA's sub-section their References by subject — should we do likewise? Willow 09:57, 3 September 2006 (UTC)

Personally, I'd lean towards not subsectioning - I would expect that at least some of the references will overlap into several sections, which would cause problems. If Harvard referencing is used, then inline tags should be used to show where the references apply, for example (Einstein 1616b) would be put near/after the text where Einstein's second paper from 1916 has been cited. Mike Peel 10:05, 3 September 2006 (UTC)

Having now tried the inline Harvard references, I'm beginning to think that we should change to the footnotes system throughout. The Harvard approach makes me either be redundant (e.g., "In 1909, Einstein showed...(Einstein 1909).") or stilt my writing uncomfortably. I agree with JRSpriggs that the footnotes are difficult for editing, but they seem now to be better than the alternatives. Thoughts from the other editors here? Willow 16:07, 5 September 2006 (UTC)

I think we should use footnotes for the moment. But I would suggest that wikipedia should work on automatically numbered inline refs, similary to current footnotes. Footnotes are not refs, most journals distinguish between the two and footnotes appear on the bottom of the page while the refs end up in the references section. I also agree with JRSpriggs about the markup text in the main text. This should be improved. The markup text should mostly be in the References section. Another thing that is missing in wikipedia are equation numbers. Count Iblis 17:14, 5 September 2006 (UTC)

Spin tutorial

The angular momentum of ANY system in quantum mechanics is restricted to:

$L = \hbar \, \sqrt{s (s+1)},$

Where s is 1/2, 1, 3/2, 2, etc.

See spin quantum number and so on. An electron with "spin" of 1/2 does not have a total angular momentum of 1/2* $\hbar$ . Rather this is merely the component of its spin in any given direction. But the total angular momentum of an electron is [sqrt(3)/2] $\hbar$ . When we say electrons have a spin of 1/2 we are merely following the convention of giving the component in any given direction, not the total. The same is true for a photon. Photons are spin-1 particles, but this does NOT mean that a photon carries spin angular momenta of $\hbar$ , which is what the article now says. Rather, photons carry total angular momenta of sqrt(2) * $\hbar$ . I tried to fix this in the article and was reverted. Hey, folks! If you don't know even this much elementary QM, you shouldn't touch this article. Okay? And for those of you who do know better, shame on you. S B H arris 20:37, 5 September 2006 (UTC)

See Pauli matrices. For example, if s=1/2 (an electron, say), then the operator for the i^th component of its spin angular momentum is $\frac{\hbar}{2} \sigma_i$ . So if we use the Pythagorean theorem, we get $L = \frac{\hbar}{2} \sqrt{\sigma_1^2 + \sigma_2^2 + \sigma_3^2} = \frac{\sqrt{3}\,\hbar}{2}$ . JRSpriggs 09:40, 6 September 2006 (UTC)

Exactly. A number which can also be recovered from the h-bar * sqrt [s(s+1)] formula, and is good for any massive spin-1/2 particle. For photons you need h-bar times the total L, which has only 2 independent components not 3 (since it's traveling at c in one direction there can't be a third spacial spin component in that direction) so you get h-bar times sqrt (1+1) = sqrt(2) h-bar. My question is what about neutrinos? The Dirac neutrino was supposed to be massless with an intrinsic helicity. I would think that would require only two spin components like the photon, and thus give spin of sqrt(2)/2, but the standard 3 are used, giving the same total as for the Dirac electron. As we now think neutrinos have a rest mass and are Majorana not Dirac particles, so each is its own antiparticle (you just have to manage to go faster than the particle to convert neutrino to antineutrino), I suppose the neutrino spin is rightousouly sqrt(3)/2 total, just as for the electron? Yes? S B H arris 20:34, 9 September 2006 (UTC)

You write: Hey, folks! If you don't know even this much elementary QM, you shouldn't touch this article. Okay? And for those of you who do know better, shame on you.

I find it incredibly insulting that you would be so arrogant and belittling towards not only some random editor, but also any other editors that may come by later (that last sentence in particular seems to be yelling at anyone who may edit). And on top of that, of the two sides, I think it is yours that is wrong on this topic.

I think the square root equation you show in the first post is true for massive particles only. Secondly, I believe your comment "which has only 2 independent components not 3 (since it's traveling at c in one direction there can't be a third spacial spin component in that direction)" is also incorrect. The angular momentum of a photon is parallel/antiparallel to its direction of motion (its spin states can also be considered its helicity state). Furthermore, the photon has no "transverse" angular momentum states as you claim, so I agree with the previous edit and disagree with yours. The total angular momentum of a photon is 1 (in units of hbar).

Of course I don't know everything, and heck I could be wrong. I am confident about my statements above but not absolutely so, and there should be no absolutes in such things. If you aren't even willing to entertain the possibility you are wrong, no real discussion can happen, and you will just scare away editors. If you are now willing to entertain the possibility you are wrong as well, and still disagree with me, I am interested and willing to have a discussion. -- Gregory9 12:06, 10 September 2006 (UTC)

It doesn't matter if particles are massive or not. Consider a massive particle in the |l=1,m=1> state. This state is orthogonal to the |l=1,m=0> state, so you can't find the particle in this state. Yet the state |l=1,m=1> is an eigenstate of L^2 with eigenvalue

l(l+1).

If quantum theory would be different and somehow the total angular momentum for the photon were different from a massive spin 1 particle then you could experimentally rule out a massive photon, however tiny its mass. That's clearly not the case. In the early days of QM it had been argued that since a massive photon has 3 polarization states instead of two, black bodies should radiate 3/2 times the energy if the photon were massive and a massive photon was thus ruled out. Bohr later showed that this is not true. If the photon has a tiny mass then its longitudinal component will have almost no interactions with matter. Black bodies will thus not radiate longitudnal photons. Count Iblis 16:16, 12 September 2006 (UTC)

That is an interesting tidbit of history regarding Bohr. Thank you for sharing it, and if you know the reference, I would be interested in looking it up.

However, I do not agree with your arguement. It appears you are saying that the photon has the same states as a massive spin 1 particle, but that we can just never measure it to be in the other states because they are not created in those states, nor will they ever interact if they are in that state. This sounds like claiming there is an Aether but that we can never detect it. If you are going to claim something exists, please give an example with an experimental difference.

L and L_z commute. So even if we consider for a moment that there are some "ghost states" that can't be occupied, I can still ask what the total angular momentum is while restricting us to +/-1 helicity states. Is there an experiment that measures L and not just L_z of a photon? If so, this would settle this debate. -- Gregory9 05:06, 13 September 2006 (UTC)

Well, to answer my own question, in theory you could measure the total angular momentum of a hydrogen in it's ground state (ie zero). Absorb a photon, and measure again. Obviously the answer will indeed be Sqrt[2]hbar. So it only makes sense to call the total angular momentum of a photon to be Sqrt[2]hbar as Count Iblis and SBHarris claim. So I was wrong here about that value.

However I still object to SBHarris's claim that the photon doesn't have a spin component parallel to its direction of motion. And I am still confused about these transverse spin components of a massless particle, so if someone could comment further it would be appreciated. If I measure the S_x to be 1, then because S_z cannot be zero, I now know that S_y = 0. This violates the uncertainty princple as I know one transverse spin component by measuring the other. Does this mean that the transverse spin components cannot be measured? If it can never be measured, then it seems like they are useless quantities and all that matters is the helicity and total angular momentum. I see Count Ibis already changed the wording in the article from SBHarris's edit to remove reference to other components, so I guess we are done now. -- Gregory9 06:20, 13 September 2006 (UTC)

Subpages?

I've been thinking about how to offer different levels of presentation to our readers, e.g., to let interested readers actually see the derivation or formula from which some assertion in the main text follows. Perhaps we could do that with subpages such as [[Photon/Semiclassical radiation theory]]? One could even have subpages of subpages, for readers who really wanted to delve deeply. I suspect that the idea won't work, but I thought I'd at least float the idea. What do you all think? Willow 12:22, 9 September 2006 (UTC)

P.S. What does everyone think of the inline footnotes — OK?

That's a good idea! Count Iblis 13:21, 9 September 2006 (UTC)

It would probably be better to create new pages, rather than subpages, as that seems to be the way people do things on Wikipedia. For example, use Derivation of the Semiclassical radiation theory for photons (or something snappier). I would only do this if you can justify a wikipedia article on the derivation, though - i.e. it's a notable derivation, and not something that can simply be referenced. Mike Peel 08:41, 16 September 2006 (UTC)

Subpages in article space are contrary to policy (not sure where to find it at the moment; take a look at WP:ASR, though). The idea is that, since each article can be freely copied by mirrors under the GFDL, it should stand on its own, and not be dependent on other WP articles. Of course that doesn't mean you shouldn't use wikilinks to other articles, but it does mean that the article shouldn't say something that's wrong or misleading to someone who can't find the link, and it's also interpreted to militate against any hierarchical structure for articles. --Trovatore 20:21, 16 September 2006 (UTC)

The people at the Village Pump seemed to feel the same way. My only motivation for subpages was the occasional case where one would like to include an extended footnote that can't really stand on its own, e.g., showing that the electromagnetic field can be decomposed into independent simple harmonic oscillators, or stuff like that. I definitely see the rationale for the "no subages" policy, though. Oh well, it was just a thought! Willow 20:38, 16 September 2006 (UTC)

Symbol use

Why are both $ν$ and $f$ used as symbols for the photon's frequency? --Sadi Carnot 15:18, 11 September 2006 (UTC)

Because different authors have chosen different symbols. People who choose f probably do it because it's easier to type and you don't have to remind people that it's not a v. There's no deep reason. --Strait 20:03, 11 September 2006 (UTC)

Or did you mean in this article? I notice it isn't internally consistent. That should be fixed. --Strait 20:11, 11 September 2006 (UTC)

Is this really true? "In chemistry and optical engineering, photons are usually symbolized by hν, the amount of energy of each photon." Could someone give a context for this sort of usage? --Strait 20:11, 11 September 2006 (UTC)

I added the bit about engineering to the statement about chemistry. In diagrams of image sensors, for example, a squiggly line with hv on it indicates an incoming photon. A just did a quick book search to see what I can find, and the one I found has hv in a sort of chemical equation [1] here's another like that: [2]

Hi, Strait, welcome to Photon! Chemists write "hν" over a reaction arrow to indicate that the reaction is initiated by the absorption of a photon. I think an example is the halogenation of alkanes; ultraviolet light is used to form the halide radical, which then reacts with the alkane. I personally don't have any experience with optical engineering, though. (Oops, I see that someone was here ahead of me!)

I'm surprised that you say our usage is internally inconsistent — I couldn't find any f or other inconsistency. Are we overlooking something? Thanks for your help! Willow 20:57, 11 September 2006 (UTC)

Bohratommodel.png uses f. --Strait 20:59, 11 September 2006 (UTC)

Thanks for catching that, Strait! I also noticed that the orbitals weren't scaled correctly, so I made a new figure. It uses no English, so that other Wiki's can use it, too. In its original form at the Commons, it's scaled so that 1 inch = 1 Angstrom. Strictly speaking, the 3→2 photon should be ~656 nm; maybe I should make it red, instead of green? What do you think? Looking forward to other things you catch, with many thanks, Willow 22:04, 11 September 2006 (UTC)

P.S. I decided the Figure looked better in gold and red, anyway, so I went ahead and changed it — hope you like it! :) Willow 22:55, 11 September 2006 (UTC)

Well, more to the point, yesterday I was looking up photon in:

Q is for Quantum – An Encyclopedia of Particle Physics (2000)
Oxford Dictionary of Chemistry (2004)
The Essential Dictionary of Science (2004)

They all use E = hf as the formula for the energy of a photon, where f = frequency; also the Wikipedia frequency article use “f” as well. So the inconsistency between this page and these other pages puzzled me. Today, however, I looked up frequency in Britannica (2002), it says: “the symbols most often used for frequency are f and the Greek letters nu (ν) and omega (ω). Nu is used more often when specifying electromagnetic waves, such as light, and gamma rays; omega is mostly used by electrical engineers in referring to alternating current.” Maybe we should add these clarification notes to both the photon and the frequency article? Talk later: --Sadi Carnot 22:57, 11 September 2006 (UTC)

Ready for Non-scientific Peer Review?

Are we ready for a non-scientific peer review? Most of the outstanding issues seem to have been addressed, although some may have been overlooked. Perhaps Mike Peel will be so good as to put all the new references into that {{cite}} format? If we're all agreed, we can go for peer review, and afterwards FA status. Something to look forward to! :) Willow 18:16, 11 September 2006 (UTC)

A couple of comments from my skim over the article. Is there anyway to get rid of this self reference in the lead? (Throughout this article, the term light refers to all forms of electromagnetic radiation, not just to light visible to the human eye.) For an article of this length the lead is far to short, see WP:LEAD, I would try and make the lead a bit more "average reader" friendly, mabye a bit more like the EB article since after the lead the math will frighten most people off. Otherwise its very good, but you might want to ask someone who hasn't worked on the article to copyedit the text.--Peta 01:13, 13 September 2006 (UTC)

Yes, easy. Delete that sentence and change the first one to: "In modern physics, the photon is the elementary particle responsible for electromagnetic interactions and light (that is, electromagnetic radiation of all wavelengths, not just those visible to humans)." And then a paragraph break before the properties. I tried, but my editing is messed up, maybe due to the article length or something. Dicklyon 03:05, 13 September 2006 (UTC)

Those are all excellent suggestions -- thanks for taking the time, Peta and Dick! Until you and Opabinia commented on it, I never realized how dense and difficult the lead would be for a non-physicist. We'll all work on improving it. :) Willow 16:41, 13 September 2006 (UTC)

Article is too long (55 kb)

Willow, looking at the history section, I note that you (primarily) seem to have added over 25 kb of new material to this article within the last three weeks. The limit to article size, for numerous reasons, see article size, is 32 kilobytes. I would suggest that you break up this article, down to a reasonable size (below 32 kb), and partition off many parts to their own page, using only mini sub-articles here, with "see main" links attached. --Sadi Carnot 23:48, 11 September 2006 (UTC)

Hi Sadi,

Time flies when you're having fun, doesn't it? It's hard to believe that I started working on this article four weeks ago today. I hope that you all like my contributions, and that I haven't been too abrasive in pursuing a good Photon article. If so, I really hope you forgive me.

The article is long, but not too long. The main prose has ~4500 words and <11 pages, which are below the limits set by our policy. (The 55k number includes all the references, tables, etc.) However, the article may indeed be difficult for lay-people to absorb in one sitting (20 minutes), since it may present too many novel concepts; in that case, the interested reader would presumably have to digest the article piecemeal, in multiple readings. It's good that we put the more accessible historical material early in the article, to give such readers a secure haven from which to venture out.

Speaking for myself, the breadth of our coverage seems good. In general, the depth seems reasonably OK, too, although perhaps the article could be made more shallow in a few places. We already have "See main" links in most sections. The flow and terseness of the writing seem decent as well, although again they could stand some improvement. The peer review (and, eventually, the FA candidacy) seem likely to give us a more objective assessment and perhaps helpful guidance on writing for lay-people. What do other people here think? Willow 11:05, 12 September 2006 (UTC)

Willow, your efforts seen enthusastic and well intentioned; there are, however, limitations to online reading which apply even more so for technical articles. From article size, "articles longer than 12 to 15 printed pages (more than 30 to 35 KB of readable text) take longer to read than the upper limit of the average adult's attention span—20 minutes. An important consideration is that attention span is lower for children, adults of below-average intelligence, and all those with attention deficit disorders (groups we would like to serve as well)."

The photon article is 14 pages (printed). Wikipedia is a process of evolution. More people are going to come and add to this article when you are done with it. It will always grow. Wikipedia has unlimited storage space. A good rule of thumb, for science articles, is to begin to chop up the article after it goes past 10 pages (printed). There is no reason to bulk up articles to the point that nobody is going to want to read them. --Sadi Carnot 12:31, 12 September 2006 (UTC)

Thanks for your kind words! I worked to reduce the text today. Right now, the PostScript form of the "Printable version" is 8 full pages, if we don't count the Table of Contents and the References. Is that OK with everyone? I'll add the "Experimental determination of photon mass" into a separate article about the mass of the photon, and may do likewise for the semiclassical theory of electromagnetic radiation. Talk to you all soon, Willow 22:42, 12 September 2006 (UTC)

I would suggest that we start a new page: Photon (historical development), and move a lot of the content of this page, e.g. historical development, double-slits, early objections, early quantum views, Bose-Einstein, etc., there. In this manner, we could have a small paragraph here with a "see main" link attached.--Sadi Carnot 22:08, 13 September 2006 (UTC)

Limits on photon charge make no sense

see here and here. The reason is that there is no well defined theory of a charged photon, so you cannot translate experimental/observational results and derive limits on the photon charge. What you can say is that all experiments are consistent with the standard model in which the photon has no charge. So, I think that this article should at least say that limits on photon charge are disputed.

Limits on the photon mass are also disputed but that's because there are different models for the photon mass, see here. Count Iblis 15:42, 12 September 2006 (UTC)

As always, I advocate going with what the Review of Particle Physics says, which is <5×10^-30e with some comments, but no serious reservations. See [3], page 2. --Strait 16:36, 12 September 2006 (UTC)

This article is too long to go into details, so I guess we need to stick with the current "official" limits. It would be a good idea to start new articles on both the photon mass and the photon charge though. There is already an article on Proca theory on massive photon, but this is not general enough. Count Iblis 17:59, 12 September 2006 (UTC)

Photons in matter

I have a hard time interpreting this section. How can a photon travel slower than c? How can a scattered photon have differnet energy than it started with? The way I've seen these handled before is to admit that the matter involved is absorbing and re-radiating, simultaneously. In that case, it all makes sense. But for particles to change their identity or their speed this way is somewhat nonsensical, isn't it? How can the photon with a different energy be the same photon that came in? Dicklyon 00:06, 13 September 2006 (UTC)

Each photon is indistinguishable from any other photon in the same state. So if there is some amplitude for an incoming photon to be absorbed and re-emitted and some amplitude for it to pass thru unalterred, then the two cases can interfere with each other to produce the outgoing photon. One cannot say whether the outgoing photon is the same photon or not. But its phase (the complex argument of its probability amplitude) will be shifted and the photon thus delayed (slowed). If it is reflected or deflected, then its frequency and energy can change due to the Doppler shift caused by the motion of the entity which reflected or deflected it. JRSpriggs 09:38, 13 September 2006 (UTC)

Suggestions from a non-physicist

Since Willow poked some of us working on the protein article to have a look at this one, here's a few suggestions from a scientist who hasn't thought about photons in years:

For FAC purposes, the lead will almost certainly need to be expanded - I'd suggest putting a more general/non-technical explanation in the first paragraph and putting the current information in a second paragraph. (FAC seems to also have a strange obsession with putting footnote numbers after punctuation, even though as far as I know it's nonstandard.)
This is stretching the limits of dumbed-down-ness, but I'm sure I'm not the only lazy one who writes "hv" (not ν) - worth mentioning this as an alternative?
There's quite a lot of historical information, which is very well-written, but the chronological organization of the article is potentially confusing without an initial summary of the modern consensus view (otherwise the reader gets a sense of repeatedly being told "close but wrong" information). I thought of this when reading the "energy, momentum, angular momentum and mass" section, which occurs late in the article but contains most of the information that a reader generally acquainted with the subject would be familiar with.
It should be made clearer that the "gamma ray microscope" is a thought experiment, not an actual device. (Also that the required "collision" isn't really.)
There's a couple of quotes that appear somewhat arbitrarily without direct citations - they stand out because the rest is very well cited. I'm thinking mostly of 'honorable funeral' and 'all the world is quantized' - are they actual quoted words, or just turns of phrase?
The section on the Bose-Einstein model is a bit confusing - it reads like it used to be longer and got chopped too much. "In 1926, Paul Dirac showed that the number of photons is not conserved." - this sentence seems to come out of nowhere. Not conserved in the formation of a Bose-Einstein condensate? Similarly, the subsequent discussion of whether photons are bosons or fermions doesn't have an obvious relationship with the preceding paragraph.
I don't understand what is meant by "the density of radiation at frequency ν within the cavity". Is it assumed that the density is not time-dependent? Is there any significance to the decomposition of Bij into its two component rates?
"...photons automatically obey Bose–Einstein statistics." - somehow automatically seems like an odd word choice here. Something more like inherently? (Also, because I didn't understand the statement about Dirac's photon-conservation result earlier, I'm not clear on how photons can obey B-E statistics, and yet shouldn't they by definition, because they're bosons?)
"The unification of the photon with W and Z gauge bosons..." - the concept of "unification" hasn't been explicitly defined. The gauge boson section also mentions that "photons are predicted to be massless and chargeless", but these are prerequisites for falling under gauge theory, so is this a prediction or an assumption?
The "invariant mass" section mentions the necessity of counting virtual photons in QED, but it's not previously explained whether virtual photons add to invariant mass.
The mass of citations (hurr, bad pun) for the masslessness of the photon are nice to have, but kind of distracting in a big string. I don't know how this would look formatting-wise, but maybe just one number to link all the references would work better.

Erm, I think that's it. Nice job, everyone. Opabinia regalis 05:34, 13 September 2006 (UTC)

Non-scientific Peer Review

Thanks to the efforts of everyone here and our friends from biochemistry, I think we might be ready for non-scientific peer review. I'm going to look at a few other encyclopedias to see whether we missed anything, but the article is respectable, don't you think? If we're agreed, I'll add the peer-review template late tonight or tomorrow morning. The present length of the printed article is ~8.5 pages. Willow 16:01, 14 September 2006 (UTC)

I don't know how you are doing your counting, but the article is 14 "printed pages"; I'm sure it will hit the 20-page mark soon. --Sadi Carnot 13:11, 20 September 2006 (UTC)

Hi, Sadi, I hit the "Printable version" link in the left-hand column and printed it out on my little printer at home. The printed article was ~8.5 pages, if you count the main prose and neglect the Table of Contents, References, See also's, etc. as Wikipedia recommends doing in judging the length of an article.

So far, no one has complained about the length in the peer reviews. The article is "a tad technical", though, so we may well lose some "children, adults of below-average intelligence, and all those with attention deficit disorders" for that reason. Quantum field theory can be demanding even for people of above-average intelligence. ;) What do other people here think about the length and coverage?

The non-scientific peer review seems to be going well, and has definitely improved the article. Maybe after another few days to allow other more comments, we can nominate it as a Featured Article. Keep your fingers crossed! Willow 14:44, 20 September 2006 (UTC)

I've followed this process from the beginning and it's going great. As far as my last commitment to check the historical accuracy of the various sections, my delayed but final response is this: they are well-written, the dates and players match well with the sources I've consulted, and the only thing that remains is, per comments above in previous sections and in my SPR of the article: the chronology and organizational layout might benefit from a little restructuring. The SPR page just linked contains my suggestion for the layout based upon the content of the individual sections at the time. WillowW, I think it's great the initiative you've taken here and I've just been glad to have been one of the grains of sand riding this wave. I look forward to the FA! Cheers, Astrobayes 17:13, 22 September 2006 (UTC)

Edits of 20-21 September 2006

Sadi Carnot has added some excellent elements to our article which were sorely missing, such as the retinal example and mentioning the wave-particle duality in the lead.

However, the new lead section needs a little work, since it no longer conforms to Wikipedia's guidelines for the lead, which the experienced editor Peta recently called to our attention on this page.

The lead section should provide a clear and concise introduction to an article's topic, establish context, and characterize the terms. It should contain several paragraphs, depending on the length of the article, and should provide an overview of the main points the article will make, summarizing the primary reasons the subject matter is interesting or notable, and including a mention of its notable controversies, if there are any. The lead should be capable of standing alone as a concise overview of the article, should be written in a clear and accessible style, should be carefully sourced like the rest of the text, and should encourage the reader to read more.

A table later in the guideline article suggests that we should have 3-4 paragraphs in the lead. Are there any objections to returning the two "Overview" paragraphs back into the lead?

Another issue is that, as written, some of the new material in the lead might be worded confusingly. For example, the expression "force carrier for all forms of electromagnetic radiation" might confuse readers between two different aspects of photons, their conveyance of light energy/momentum and their mediation of electromagnetic interactions. Also, the sentence "The photon, essentially, is a 'quantum' or packet of electromagnetic radiation measured in proportion to its ability to move an electron between atomic orbitals in atoms and molecules." seems somewhat vague and, at the same time, overly specialized to absorption/electronic transitions in atoms and molecules. Finally, readers might become confused when we say in the lead unqualifiedly that photons move at the speed of light; perhaps, we should add in vacuum or in empty space?

By the way, we haven't received any more criticisms of our article at Wikipedia:Peer review so, if we're all agreed, I'll submit the FA candidacy tomorrow. Willow 17:16, 21 September 2006 (UTC)

I think it's ready for FA. -- SCZenz 22:05, 21 September 2006 (UTC)

We're now a Featured Article Candidate

Hi all,

The FA candidacy seems to be going well; three Supports (four, if we count Count Iblis) and no Opposes so far. Please add your Support votes to help make this article a FA — thanks! :D Willow 22:51, 23 September 2006 (UTC)

OK, we're at 5 Supports (Thanks, Count! :), 1 "nice" Comment and no Opposes. We'll get at least one more review from a physicist, since Laura Scudder agreed to review the article for us. She always has good insights and seems fair so, either way, it's a win for us and for the article. Keep the fingers crossed! :D Willow 15:22, 25 September 2006 (UTC)

Photoelectric effect in CCDs?

Willow, once before I corrected the idea that the photoelectric effect is involved in CCDs or other image sensors. It's an analogous effect, but it doesn't involve kicking electrons out of a material, just kicking them from valence band to conduction band. As far as I can see from definitions of photoelectric effect, it does not apply; I don't know if there's a special term for it. Since you've put it back, I think you should correct it. Dicklyon 03:06, 25 September 2006 (UTC)

OK, thanks very much for catching that. I'd like to avoid having to explain semiconductors to the lay audience, though; maybe we can finesse it? I'll do my best, and then crash. This whole "technological applications" section has been hard to write. :( On a happier note, we seem to be doing OK at the FAC. Willow 03:30, 25 September 2006 (UTC)

P.S. Please let me know if I flub the fix — thanks!

Background of info table

Hi, the info table at the start of the article has a dark grey background. Could it be changed to something lighter? I'd fix it myself, but I don't know if the colour was chosen on purpose, to match other articles on particles. --Kjoonlee 06:15, 26 September 2006 (UTC)

Hi, Kjoon, I agree, it was too dark to read easily. I lightened it just now, but we may have to Talk to the other particle pages, to see if we all can be consistent — they're still in the dark ;). Thanks for pointing that out! :) Willow 06:40, 26 September 2006 (UTC)

Thank you. :) I edited one boson article, and it seems fermions need to be edited next. (Unless someone objects, that is.) --Kjoonlee 06:50, 26 September 2006 (UTC)

Done. :) List of articles I edited is at Wikipedia talk:WikiProject Physics#Background colour of info tables. --Kjoonlee 08:27, 26 September 2006 (UTC)

{{main}} links

At the moment, most sections have {{main}} links at the top of them. I would suggest that {{main}} is only used where the article being linked to covers the material in the section it's being linked from in more depth - for example, the main article for the "Physical properties of the photon" section would be something that discusses the physical properties of the photon in more detail than the section does. If the page being linked to covers something related, but off-topic, then use {{see also}}, or if it gives a broader scope to a topic then use {{further}}. I'd make the change myself, but I thought it might be better if someone more familiar with the details of the topic made the change. Mike Peel 07:31, 26 September 2006 (UTC)

Thanks, Mike! I hadn't realized that there were different types of templates for those sorts of links. I think it's fixed up now. If you'd like to weigh in at the FA candidacy, that'd be great. :) Willow 14:28, 26 September 2006 (UTC)

Retrieved from "http://en.wikipedia.org../../../p/h/o/Talk%7EPhoton_Archive_2_bc62.html"

Category: Talk archives