Talk:Quantum reflection

Efficient quantum reflection

I dislike this section. It would be good to offer the deduction, not only the final fit. dima 05:56, 10 September 2007 (UTC)

Definition

The section titled "Definition" is written poorly and does not cite sources. Also it is unclear what colloquium is begin talked about, when it happened, etc.

Comment

There should be a link to the WKB approximation and something could be said about complex classical turning points (i.e. an analytic continuation of the position in the argument of the wave-function). 05:11, July 14, 2008 User:TimothyRias

Is this real?

I have serious reservations about this article. It does not mention the classic evanescent waves solutions first analyzed by Slater, and standard reflection in diffraction. I would need convincing that this is anything more than the equivalent of RHEED with atoms. Ldm1954 (talk) 23:22, 15 May 2023 (UTC)

To be valid, this article _should_ be RHEED with atoms... and with small to indefinitely large molecules, neutrons (Feynman Lectures), and other particles. So, given the breadth of what _could_ be reflected, shouldn't RHEED instead be a special case of this more inclusive article on the far broader category of coherent reflection of pretty much any EM-aware entity capable of detectable quantum wave behavior?

Regarding Slater's decades of foundational work, might you have a good John C. Slater reference you could include and describe? This one might help but, alas, it's behind a paywall:

John C. Slater, Interaction of Waves in Crystals, Rev. Mod. Phys. 30, 197 (1958). Terry Bollinger (talk) 03:05, 16 May 2023 (UTC)

I suggest taking the existing Grazing incidence diffraction stub and generalizing it, something like:

• Basics
• Theory
• Reflection high-energy electron diffraction
• Surface X-ray Diffraction (SXRD)
• Helium diffraction
• Ion diffraction
• Optical
• Cold atom (this article)

A couple of paragraphs or so on each, with a "Main article" cross-ref. The first two topics exist. I think there is something on He scattering, but I did not find it immediately.

I will probably do this in a day or so -- just a paragraph cross-referencing at the top for the moment.

I will dig up the relevant Slater ref, and read the papers herein to see if I believe this is truly different, as they seem to claim. Ldm1954 (talk) 09:12, 16 May 2023 (UTC)

Nice points. Two thoughts are below.

(Also: I don't recall seeing this article before yesterday. If I did something 15 years that got me email-flagged into this discussion, I don't recall it, so I'm not vested in the text of this article. That's true even if at some forgotten point I helped write it. :)

(1) The grazing incidence article you mentioned addresses only the special case of crystalline reflection. Where's amorphous? E.g., where's Feynman's nicely lab-verified low-angle liquid mercury neutron reflection example? Leaving that case out would be a big omission by incorrectly implying that _only_ crystalline lattices reflect.

(2) The purpose of this particular article seems to be Feynman-ish. That is, it's trying to inform as broad a range of readers as possible of surprisingly general quantum principle that then guides the _selection_ of the best calculation model — or, in some cases, the _lack_ of a sufficiently apt math model.

That general principle is that as long as (a) the surface is approximately smooth at the momentum wavelength of the incoming object, and (b) the surface "interacts" with that surface (not necessarily in crystalline order) via a known force (not necessarily EM), then any type of quantum wave (the Maxwell EM wave quantized on detection for photons, the Scrödinger wave for object containing fermions), the object may coherently reflect.

Evanescent EM waves _do not_ fully cover all such cases. In particular, slow neutrons reflecting from amorphous mercury surfaces interact primarily with the collective _nuclei_ of the mercury surface, not its evanescent electron structure. Terry Bollinger (talk) 11:53, 16 May 2023 (UTC)

Minor clarification. If z is the normal out from the surface, an evanescent wave (x-ray/electron/atom/neutron) has a form $exp(qz+ik.r)$, i.e. it decays inside, bring a standing wave (no current flow into the sample). Ldm1954 (talk) 12:05, 16 May 2023 (UTC)

It is not that one. The closest I can find to a Slater paper is Physical Review 1937 Vol. 51 Issue 10 Pages 0840-0846 DOI: 10.1103/PhysRev.51.840, just take the limit with the imaginary (adsorbative) part to zero. Actually evanescent waves are in Bethe's paper (like almost everything), Annalen Der Physik 1928 Vol. 87 Issue 17 Pages 55-129 DOI: 10.1002/andp.19283921704 Ldm1954 (talk) 19:10, 17 May 2023 (UTC)

Thank you, these look helpful! My Slater refer was, at best, a shot in the dark. I'll look these up and try to respond soon. Good refs are always helpful. Perhaps you could add in what you found to the article? Terry Bollinger (talk) 19:49, 17 May 2023 (UTC)

I am still in two minds about the whole article. I made some tweaks, but I am not enthusiastic about it. Phrases such as "it is the extended quantum wave packet of the particle, rather than the particle itself" don't make sense. Is it really any different from standard reflective diffraction or matter waves? Ldm1954 (talk) 20:07, 17 May 2023 (UTC)

I think the article has a place -- see my earlier comments -- but I agree that sentences like that one that reflect a particular quantum interpretation are problematic. My reading of that sentence is it's a pilot-wave perspective (e.g., John Bell's position), treating "particle" and "wave" as two _entirely separate_ entities. Bell was a very smart fellow, but given that QED reproduces the wave shape at any moment in time by _nothing more_ than assessing all possible particle paths, I don't find separating the two plausible.

And I agree that whatever the intent was of that sentence was, it _must_ be equivalent to diffraction of matter waves. But surely some matter wave methods also use wave packets? It's pretty limiting experimentally if you only look at cases where the wave has traveled long enough to approximate a plane wave over the entire reflective surface. Terry Bollinger (talk) 21:09, 17 May 2023 (UTC)

I cannot answer for all matter waves methods. One I known is in Scanning transmission electron microscopy where you simulate with a 2D wavepacket at the top of a sample, then numerically solve a relativistcally corrected (effective mass) Schrodinger's equation. A few people have included mutual coherence function, which is similar to a density matrix. I am not sure if anyone has done this for reflection, as AFM ended interest in reflection electron microscopy in the late 1980s.

Maybe in the He diffraction literature, which is significantly larger than this somewhat niche area.

A good discussion, but I am not sure how important this article is so it is not at the top of my priority 🫠 Ldm1954 (talk) 12:54, 19 May 2023 (UTC)

Thanks. This page is low-priority for me, too, but I must admit that's because I don't see much need to change it. This article is _inherently_ uncomfortable -- and probably should be -- because it hits on points like the experimentally well-verified reflection of low-angle slow neutrons off mercury surfaces that are easy to describe in a _casual_ way with Schrodinger wave packets, but then get a bit weird if you start thinking too hard about the details.

For example, the neutron wave functions must "see" and interact with strong force in all of those individual mercury nuclei, and on a _per nucleus_ basis, you could likely use something like an evanescent wave model. But on the other hand, all of those nuclei are separated and utterly isolated by astronomically larger distances than, say, anything in the microwave domain. So how does _that_ work? Neutron-nucleus interactions occur at gamma range frequencies, but the _reflection_ works only at the astronomically lower-energy momentum frequencies of the neutrons. I'm sure there are plenty of papers on such issues, but I also know that in the cases where I've looked up papers on such topics, I ended up a lot less impressed than I thought I would be. Hand waving is hand waving, even if it's done with lots of equations to make it look less hand-wavy.

Similar remarks might be said about the Mossbauer effect, which to me is something _no one_ would have predicted purely from theory. Once it was observed, sure, you can get some decent models together, and they all fit within the math parameters of quantum theory. But _predicting_ Mossbauer? I don't think that would have happened.

All of this is also why I think that, as long as they solidly reflect actual lab results, uncomfortable pages like this are a good thing, not a bad thing. They remind us that despite huge modeling progress over the past century, there is still a surprising amount of serious debate (and disagreement) about the deeper physical nature of wave functions.

My preference would be to let this particular sleeping dog stay sleeping. Terry Bollinger (talk) 13:49, 19 May 2023 (UTC)

Agreed.

N.B, if you switch to reciprocal space everything is simple. Ldm1954 (talk) 13:55, 19 May 2023 (UTC)

I love reciprocal space! Though I'd say its own mysteries, such as _exactly_ what happens to a point particle translated into reciprocal.

In any case, please have a fantastic day and weekend. Terry Bollinger (talk) 14:11, 19 May 2023 (UTC)